[Senate Hearing 109-75]
[From the U.S. Government Publishing Office]



                                                         S. Hrg. 109-75

            ENSURING DRUG SAFETY: WHERE DO WE GO FROM HERE?

=======================================================================

                                HEARING

                                 OF THE

                    COMMITTEE ON HEALTH, EDUCATION,
                          LABOR, AND PENSIONS
                          UNITED STATES SENATE

                       ONE HUNDRED NINTH CONGRESS

                             FIRST SESSION

                                   ON

EXAMINING THE FOOD AND DRUG ADIMINISTRATION'S PROCESS OF ENSURING DRUG 
                                 SAFETY

                               __________

                             MARCH 3, 2005

                               __________

 Printed for the use of the Committee on Health, Education, Labor, and 
                                Pensions

                    U.S. GOVERNMENT PRINTING OFFICE
20-047                      WASHINGTON : 2005
_____________________________________________________________________________
For Sale by the Superintendent of Documents, U.S. Government Printing Office
Internet: bookstore.gpo.gov  Phone: toll free (866) 512-1800; (202) 512ï¿½091800  
Fax: (202) 512ï¿½092250 Mail: Stop SSOP, Washington, DC 20402ï¿½090001


          COMMITTEE ON HEALTH, EDUCATION, LABOR, AND PENSIONS

                   MICHAEL B. ENZI, Wyoming, Chairman

JUDD GREGG, New Hampshire            EDWARD M. KENNEDY, Massachusetts
BILL FRIST, Tennessee                CHRISTOPHER J. DODD, Connecticut
LAMAR ALEXANDER, Tennessee           TOM HARKIN, Iowa
RICHARD BURR, North Carolina         BARBARA A. MIKULSKI, Maryland
JOHNNY ISAKSON, Georgia              JAMES M. JEFFORDS (I), Vermont
MIKE DeWINE, Ohio                    JEFF BINGAMAN, New Mexico
JOHN ENSIGN, Nevada                  PATTY MURRAY, Washington
ORRIN G. HATCH, Utah                 JACK REED, Rhode Island
JEFF SESSIONS, Alabama               HILLARY RODHAM CLINTON, New York
PAT ROBERTS, Kansas

               Katherine Brunett McGuire, Staff Director

      J. Michael Myers, Minority Staff Director and Chief Counsel

                                  (ii)

  




                            C O N T E N T S

                               __________

                               STATEMENTS

                        THURSDAY, MARCH 5, 2005

                                                                   Page
Enzi, Hon. Michael B., Chairman, Committee on Health, Education, 
  Labor, and Pensions, opening statement.........................     1
Kennedy, Hon. Edward M., a U.S. Senator from the State of 
  Massachusetts, opening statement...............................     3
Woodcock, Janet, M.D., Acting Deputy Commissioner for Operations, 
  U.S. Food and Drug Administration..............................     4
    Woodcock, Dr. Janet, and Dr. Kweder, joint prepared statement    17
Wilson, Cecil B., M.D., member, American Medical Association 
  Board of Trustees, Winter Park, FL; Keith L. Carson, chairman, 
  the Williamsburg Bioprocessing Foundation, Virginia Beach, VA; 
  Raymond Woosley, M.D., PhD, president, the Critical Path 
  Institute, professor of medicine and pharmacology, University 
  of Arizona, Tucson, AZ; and Bruce M. Psaty, M.D., PhD, 
  professor of medicine, Epidemology and Health Services, 
  codirector, Cardiovascular Health Research Unit, University of 
  Washington, Seattle, WA........................................    25
    Prepared statement of Dr. Wilson.............................    26
    Prepared statement of Mr. Carson.............................    32
    Prepared statement of Dr. Woosley............................    35
    Prepared statement of Dr. Psaty..............................    42

                          ADDITIONAL MATERIAL

Statements, articles, publications, letters, etc.:
    Clinton, Hon. Hillary Rodham, a U.S. Senator from the State 
      of New York, prepared statement............................    54
    Questions of Senator Clinton for Janet Woodcock, M.D.........    54
    American Medical Association (Docket No. 2004D-0188).........    55
    American Medical Association (Docket No. 02N-0528)...........    59
    American Society of Health-System Pharmacists, prepared 
      statement..................................................    64
    Grassley, Hon. Charles E., a U.S. Senator from the State of 
      Iowa, prepared statement...................................    67
    Response to questions of Senator Enzi by Janet Woodcock, M.D.    69
    Response to questions of Senator Hatch by Janet Woodcock, 
      M.D........................................................    71
    Response to questions of Senator Gregg by Janet Woodcock, 
      M.D........................................................    72
    Response to questions of Senator Kennedy by Janet Woodcock, 
      M.D........................................................    74
    Response to questions of the HELP Committee by Dr. Wilson:
        Response to questions of Senator Enzi....................    79
        Response to questions of Senator Hatch...................    81
        Response to questions of Senator Kennedy.................    83
    Response to questions of Senator Enzi by Keith L. Carson.....    83
    Response to questions of Senator Hatch by Keith L. Carson....    85
    Response to questions of the HELP Committee by Raymond L. 
      Woosley, M.D.:
        Response to questions of Senator Enzi....................    85
        Response to questions of Senator Hatch...................    86
        Response to questions of Senator Kennedy.................    90
    Response to questions of Senator Enzi by Bruce M. Psaty, M.D.    92
    Response to questions of Senator Hatch by Bruce M. Psaty, 
      M.D........................................................    94
    Response to questions of Senator Kennedy by Bruce M. Psaty, 
      M.D........................................................    96

                                 (iii)

  

 
            ENSURING DRUG SAFETY: WHERE DO WE GO FROM HERE?

                              ----------                              


                        THURSDAY, MARCH 3, 2005

                                       U.S. Senate,
       Committee on Health, Education, Labor, and Pensions,
                                                    Washington, DC.
    The committee met, pursuant to notice, at 10:02 a.m., in 
room 106, Dirksen Senate Office Building, Senator Enzi, 
chairman of the committee, presiding.
    Present: Senators Enzi, Burr, Isakson, Kennedy, Murray, and 
Reed.

                   Opening Statement of Senator Enzi

    The Chairman. I will call the hearing to order. Good 
morning, and welcome to the second in a series on prescription 
drug safety.
    On Tuesday we received a number of recommendations from the 
Food and Drug Administration and from outside experts on ways 
that the FDA can improve its process for weighing the benefits 
and risks of prescription drugs.
    Today we are going to look to the future and consider the 
implications for drug safety, development and regulation. As I 
mentioned Tuesday, we have made major changes to the drug 
approval process over the past dozen years. Congress has passed 
a series of bills with overwhelming bipartisan support to bring 
more consistency, transparency and accountability to the drug 
approval process, but we have reached a critical juncture in 
the long history of the FDA. We need new and better ways to 
predict the safety and efficacy of the new drugs before they 
enter widespread use. We also need new and better ways to 
communicate with patients and physicians regarding the benefits 
and risks of new drugs. And we need the FDA, industry, 
physicians and patients to be vigilant and to work together to 
ensure the continued safety of prescription drugs once they are 
approved and on the market.
    Doing nothing to address the current controversies is not 
an option. However, overreacting to recent events could be just 
as dangerous as doing nothing. With all due respect to the 
press, we should not be making policy to make headlines, or in 
response to the headlines. This issue is too important for 
that. We must take extraordinary care to find the right 
approach.
    We have had and want to have the world's best example of 
drug safety. I have convened these hearings because I am 
concerned about the FDA and drug safety. All of us here today 
are aware of the recent controversies that have raised 
questions about the safety of our prescription drugs and 
whether the FDA's process for reviewing and approving drugs is 
working.
    Tuesday we heard about some of the immediate steps the FDA 
is taking to maintain the public's confidence in the Agency. We 
also heard about some steps that Congress may need to take. But 
as we consider how to deal with the questions raised by these 
recent controversies, we must also be focused on the future. As 
I said Tuesday, we should not sacrifice safety in order to 
speed drugs to the market, but we do not want to return to the 
days of the drug lag when desperate American patients waited 
for drugs that were available for months or even years in 
Europe, especially now that we stand on the cusp of tremendous 
medical advances based on new scientific insights.
    Just in the past decade we have sequenced the human genome 
and doubled the NIH's budget. This has resulted in an explosion 
of basic scientific knowledge and the promise of an avalanche 
of new therapies. But despite that increase in scientific know-
how many threatening diseases and conditions still lack 
effective treatment. More than a million Americans will have a 
heart attack this year. Some 600,000 Americans will suffer a 
stroke, and close to 500,000 women worldwide will die from 
breast cancer. The advances in biomedical knowledge have not 
been advanced by advances in new therapeutics. In fact the 
number of submissions to the FDA for new drugs and biologics 
each year has actually decreased since the completion of the 
human genome sequence.
    Part of the problem is that science is racing far ahead of 
our current regulatory regime. For instance, most of the tools 
we use to test the toxicity of a drug are decades old. One-
third of all drugs fail during preclinical or clinical trials 
due to the toxic nature of the compounds being tested. If we 
were better able to predict these failures before trials even 
began, we could both improve drug safety and save billions of 
dollars spent on research and development that leads nowhere. 
We need to match our investment in biomedical research through 
the NIH with new thinking, new methods and new resources to 
improve and streamline the regulatory process at the FDA.
    But that raises three questions. First, how do we use new 
scientific knowledge to improve the medical product development 
process? Second, can we also use this knowledge to improve drug 
safety? And finally, how do we communicate new information 
about safety and risk more effectively within the Agency and to 
patients and to physicians?
    I look forward to hearing the testimony of our witnesses. I 
trust they will be able to help us to begin to answer these 
questions. I know I speak for myself and Senator Kennedy when I 
say that the Nation is looking to us, the members of this 
committee, to ensure that drugs approved for use by the FDA are 
safe when used as intended, and do not pose an unreasonable 
risk to the public. And when we act to improve our drug 
approval system and to get the FDA ready for the future, we 
will act through this committee in a bipartisan and 
comprehensive fashion.
    After Senator Kennedy makes his opening statement, we will 
hear from Dr. Janet Woodcock, the Acting Deputy Commissioner 
for Operations at FDA.
    I will recognize Senator Kennedy for his statement.

                  Opening Statement of Senator Kennedy

    Senator Kennedy. Thank you very much, Mr. Chairman, and I 
want to thank you and commend you for calling a second hearing 
on the safety of prescription drugs, and look forward to 
working with you to pass the legislation needed to correct the 
glaring drug safety problems identified in these hearings.
    We learned a great deal on Tuesday's hearing about the 
problem. FDA obviously needs the authority from Congress to 
order drugs on the market to be relabeled when clear safety 
concerns arise. It is disgraceful it took nearly 2 years to 
relabel Vioxx after the so-called VIGOR trial. And Dr. Kweder 
of the FDA said such authority over relabeling would have 
helped, and that should be part of any legislation we propose.
    We also know from Tuesday's hearing that FDA needs 
additional funds to monitor the safety of drugs on the market, 
and Congress needs to make room in the budget for this 
important priority.
    Another major defect in the current law is that FDA lacks 
the authority to require manufacturers to conduct a further 
drug safety study after the drug has been approved and goes on 
the market. It is essential for FDA to be able to order such 
trials. Such authority may be the only way to ensure that the 
safety problems discovered after approval are studied it 
effectively in clinical trials.
    The problems of Vioxx and Celebrex were discovered in 
voluntary clinical trials to study potential new uses for the 
drugs after they first came on the market. Those trials showed 
conclusively that these drugs could cause heart attack or 
stroke in some patients. If Merck and Pfizer had not 
voluntarily conducted these studies millions might still be at 
risk from using these drugs. Most of the witnesses, including 
Dr. Kweder of FDA, said this additional authority would help 
the Agency respond to drug safety problems.
    We have concerns about how FDA acts on drug safety issues. 
Although a warning was belatedly added to the Vioxx label as to 
cardiovascular risk, Dr. Kweder told us that the warning had 
little effect in encouraging safer use of drugs. We need to 
understand why.
    Dr. Alistair Wood, who chaired the FDA's Advisory Committee 
meeting on Vioxx and related drugs, has said that the 
precaution was essentially meaningless. Some experts question 
whether relabeling is ever effective. FDA may well need better 
ways to encourage or even require safe use of drugs.
    Still another issue is direct to consumer advertising, 
which encourages wide use of new drugs before we know enough 
about their safety. Vioxx and Celebrex were heavily promoted 
through direct consumer advertising. We have all seen the TV 
ads for these drugs with people skating and playing golf. These 
drugs were developed in the hope that they would reduce stomach 
bleeding, a risk faced by perhaps 5 percent of those who needed 
pain medications. Yet these ads did not say if you are at risk 
for stomach bleeding, this drug may be for you; ask your 
physician. Instead they peddled the drugs to everybody. A 
recent study found that large numbers of patients who used 
these drugs were not at risk for stomach bleeding. These ads 
almost certainly encouraged the unnecessary use of these drugs. 
Patients saw an ad, asked their doctor about them, and started 
taking them even though another drug might have been just as 
effective. How many of these patients suffered a stroke or 
heart attack or died because of it?
    Those drugs increase the risk of heart attack and stroke, 
perhaps doubling the risk. They were used widely by as many as 
50 million people for more than 5 years before the risk was 
finally discovered. We will probably never know how many 
thousands suffered because of it. All prescription drugs carry 
some risk, and we are clearly not doing enough to minimize 
them. FDA needs better ways to do so, and I look forward to 
working with our chairman to enact more effective drug 
legislation.
    Again I commend our chairman for his leadership, and I 
welcome our witnesses today and look forward to their 
testimony.
    The Chairman. Thank you for your comment and insights. You 
add a lot of history to the committee.
    I welcome our first witness which would be Janet Woodcock. 
Dr. Woodcock, the Acting Deputy Commissioner for Operations at 
the U.S. Food and Drug Administration. She has served as the 
Director of the Center for Drug Evaluation and Research at the 
FDA. The center is responsible for regulating prescription, 
over-the-counter and generic drugs. She has had close 
interactions with diverse constituencies including the clinical 
and scientific community, members of Congress, the 
administration, national media, patient and consumer advocacy 
groups, the international drug regulatory community, the 
pharmaceutical industry, and representatives of Federal and 
State agencies.
    She was selected as the Director for CDER in 1994 and under 
her leadership the regulatory decisionmaking has been made more 
open and transparent to the public, and we congratulate you on 
your Acting Deputy Commissioner position now and look forward 
to your testimony, Dr. Woodcock.

 STATEMENT OF JANET WOODCOCK, M.D., ACTING DEPUTY COMMISSIONER 
       FOR OPERATIONS, U.S. FOOD AND DRUG ADMINISTRATION

    Dr. Woodcock. Thank you. Mr. Chairman and distinguished 
members of the committee, I thank you for the opportunity to 
testify on the important issue of drug safety.
    Five years ago this committee held hearings on adverse drug 
reactions. At that time the discussion focused on the estimated 
100,000 deaths per year, as well as hundreds of thousands of 
hospitalizations and economic losses estimated in the billions 
of dollars. This is a longstanding and serious problem that I 
have worked on over many years of my professional life at FDA.
    Since that hearing FDA has taken many important steps to 
enhance drug safety, but while drugs bring profound benefit to 
our population, we must continue to work together to mitigate 
their harm. Safety findings with antidepressants and the COX-2 
inhibitors again illustrate how much more work needs to be 
done. FDA has taken additional steps as discussed at today's 
hearing to improve management and transparency around major 
drug safety issues.
    Additional ideas for identifying drug side effects have 
been put forth by many parties. While these ideas are part of 
the solution, you should know that they are not the whole 
answer. Why not? Because all drugs have side effects. Radically 
restricting drug availability would clearly reduce the number 
of side effects, but would also greatly diminish the treatments 
available to doctors and to patients.
    Our long-term goal cannot simply be detection and 
restriction. It must focus on prevention and good management, 
and unlike the situation in 2000 there is now hope we can do 
just that, hope from new science and new technology.
    What is this new science? Right now at the time a drug is 
approved, we know it works in some people and we know the 
safety results for those people. What we do not know is who it 
works in and who is at risk for a side effect, and we rarely 
know how to monitor people taking the drug to check if a side 
effect is developing and to intervene to prevent it.
    Why is this? Not for lack of trying. By the time a drug is 
approved usually hundreds of millions of dollars have been 
spent on development and animal and human testing, but until 
recently we lacked the scientific tools to predict individual 
response to therapy. Today that is changing. New science such 
as pharmacogenomics, proteomics, advanced imaging technologies 
and computer modeling are making prediction possible.
    FDA's Critical Path Initiative, launched last spring, is 
focused on modernizing the process for drug development to 
rapidly incorporate new scientific methods. These methods will 
help individualize therapy, select patients who will respond to 
treatment, and avoid people at high risk for side effects. This 
is not a futuristic dream. As I speak today, tests are being 
done in real patients in real medical practice to target 
therapy and avoid side effects.
    In December FDA approved the first commercial 
pharmacogenetic test that allows patients and doctors to 
predict whether a person will be overdosed or underdosed by 
many common drugs. Once this test is done it is valid for a 
lifetime. So your relative, someone you know who may carry a 
gene that makes them metabolize drugs slowly and causes them to 
have many more adverse events from drugs, will know that they 
should start on lower doses.
    And in the treatment of cancer, some targeted therapies are 
currently making a difference in people's lives. Many more are 
on the way. Targeting helps select people who have a good 
chance of a positive response to therapy. This improves safety 
because people who cannot respond do not get the drug.
    As part of the Critical Path Initiative FDA will work to 
incorporate these new tools into drug development as rapidly as 
possible. For example, we will issue our final pharmacogenomic 
guidance very soon. This establishes processes for moving new 
scientific techniques into the drug development and regulatory 
process. Modernization will take time and scientific effort, 
but at the end of the day, having new tools to decide who 
should take a drug, who should not take a drug, and to monitor 
for side effects will bring about a new era in drug safety.
    At the same time, new technology, information technology 
holds great promise for improving drug safety. With the help of 
electronic data linking medication use with health outcomes we 
can find rare side effects that were not seen before approval. 
At the same time there are new ways to rapidly inform patients 
and prescribers about emerging safety data using electronic 
media. And equally important, computer-based interventions to 
provide support to prescribers has been repeatedly shown to 
reduce side effects and reduce hospitalization.
    In summary, of course we must develop better ways to detect 
drug side effects, but detection is not enough. We must improve 
the science of therapeutics and make sure that that new 
information is in the hands of those who need it. These are the 
advances that will radically improve drug safety in this 
country.
    Thank you.
    The Chairman. Thank you. We always appreciate the 
additional information, and I want to assure you and all of the 
other people who will be testifying today that their entire 
statement will be made a part of the record.
    In the way of questions, we heard the suggestion Tuesday 
that the FDA needs greater authority to require drug labeling 
changes. After hearing that I asked my staff if they would look 
to see what the present authority is. It says that if the 
labeling of such drug is false or misleading in any particular 
and it was not corrected within a reasonable time after receipt 
of written notice, that it can be pulled off the market by the 
FDA. That seemed to me to be quite a bit of authority. Is it 
sufficient authority?
    I noticed with Vioxx that the timeline evidently began with 
discussions in October of 2001 and there was continued analysis 
and data from various studies, and then the FDA and Merck 
agreed to a new labeling on April of 2002. That is a six-month 
time span. Is that the amount of time that is usually required 
to make major labeling changes?
    Dr. Woodcock. No, that is an unusually long amount of time. 
And actually the data from the VIGOR trial that we are talking 
about the safety data on Vioxx was available earlier than that. 
And currently with the recent changes that FDA has made in 
information dissemination we are now, as we announced several 
weeks ago, planning to put information directly out to the 
public on emerging safety issues. This will reach directly to 
the prescribers and the patients.
    Part of the problem with label changes, even if a label 
change is made, that is a paper document. The old labels are in 
distribution and take a while to be changed over and so forth. 
So this new mechanism that FDA is proposing really should deal 
with this problem very effectively which is getting the 
information directly out to the people who need it in a timely 
manner.
    The Chairman. Will there be negotiation before this data is 
put out to the public or is this just basic information, and 
then the labeling would change?
    Dr. Woodcock. That is correct. What is proposed--and there 
is going to be time for public discussion of how this actually 
occurs--what is proposed is that FDA will put out the factual 
information about the finding before the label is actually 
changed.
    The Chairman. Thank you. Tuesday we also talked about drug 
ads and how they need to be made clear, consistent and honest. 
Patients and physicians deserve to have as much information as 
possible about medical conditions and how to treat them, but 
the information also needs to be of high quality. Do you 
consider the current direct to consumer advertising to be high 
quality? Are there changes Congress or the FDA needs to 
consider making regarding regulation of these ads?
    Dr. Woodcock. Clearly direct to consumer advertising is a 
double-edged sword. It has benefits in informing people about 
treatments that they may not have been aware for their 
condition. Our surveys show that doctors find that it improves 
and increases the amount of conversation they have with their 
patients about treatment options. On the other side though, it 
may, direct to consumer advertising in some cases may increase 
awareness of a drug, such as in the Vioxx situation, and people 
may not have a full understanding of the risks of that 
medication.
    The direct to consumer advertising regulation takes into 
account the fact that patients must go to a physician to obtain 
a prescription for a drug, and so it is intended simply to 
allow that certain information be provided to the patient, but 
not complete information. The direct to consumer advertising is 
supposed to be balanced and not misleading. Those are the 
standards.
    The Chairman. Thank you. Some have suggested that 
accelerating the drug approval through PDUFA has caused the FDA 
to decrease its focus on drug safety. Could you comment on this 
assessment?
    Dr. Woodcock. The pre-market review, which is what is done 
under the PDUFA program, has about 50 percent of its focus on 
drug safety, and the use fee program enhanced that. To use a 
simple analogy, FDA is like the building inspectors. Someone 
else builds the building according to code. FDA does the code, 
writes the code, and then FDA comes in at the end and that is 
the review time under the user fee program, and we inspect the 
results and make sure they are up to code. What PDUFA did is 
help us add civil engineers or mechanical engineers. In our 
case it helped us add more scientific experts so that we could 
do that assessment quickly, and we could make sure that the 
standards for drug approval are kept up to date in current 
scientific standards.
    So we feel there has been a tremendous focus on safety 
during the user fee years, and people have to recognize that 
the development of drugs, the clinical development, is done 
before the application is sent to FDA, and the PDUFA clock has 
to do with how long the FDA takes to review that information.
    The Chairman. I want to thank you for your concise answers 
and great examples. My time has expired.
    Senator Kennedy.
    Senator Kennedy. Thank you very much.
    Just on the advertising issue, currently the manufacturer 
is allowed to flood the airways with ads for a drug on the day 
it is approved, and they can send salespersons to every 
doctor's office in the country to sway them to prescribe the 
new pill. It is hard to deny that the massive ad campaign led 
millions of patients who did not need them to take the COX-2 
drugs, with tragic results.
    I suppose the question that some of us would ask is, could 
this be happening with other drugs? Would it not be better to 
take a more cautious approach with advertising with newly 
approved drugs, and do you think people are taking the 
medicines that are not suitable for them as a result of the ad 
campaign?
    Dr. Woodcock. We have looked into how physicians treat 
direct to consumer advertising, their attitudes toward direct 
to consumer advertising. We asked them if they prescribe drugs 
that they would not have otherwise prescribed because the 
patients came in and asked for them. And in some cases that 
does happen. There is no doubt about it.
    The construct that we are working under is that the 
physicians will be learned intermediaries and will decide after 
a conversation with the patient whether or not a particular 
drug is appropriate for that condition and that patient.
    Senator Kennedy. Have we gotten rid of these incentives? 
You mentioned the hearings that we had 5 years ago. I remember 
them, where we had all of these very incredible incentives to 
both the doctors and the ad people in terms of trips, tickets, 
just about everything under the sun. I was just wondering, just 
quickly, have those practices come back now or are we pretty 
free from all of those?
    Dr. Woodcock. No, those practices are still going on, and I 
think most of us in the medical profession believe that the 
detailing and, for example, physicians on average have six 
visits per month of detail persons to their office. We believe 
that that has more influence on prescribing patters than direct 
to consumer advertising, and certainly the budget is much 
larger.
    Senator Kennedy. I was thinking of the incentives that were 
given both to the doctors for the use of these drugs. You 
probably remember those hearings. As I say, I do not want to 
take a lot of time, but most of the abuses that we identified 
during that period of time, have they been dealt with pretty 
effectively or are some of them creeping back in. You can 
answer later on, whatever you want to do.
    Dr. Woodcock. I will be happy to answer more fully later 
on, but I believe they still continue.
    Senator Kennedy. I would be interested if you could let us 
know, but I thank you.
    On Tuesday Dr. Kweder acknowledged several FDA lapses in 
relabeling and informing physicians about the risks of Vioxx. 
In your view what were the other mistakes or missed 
opportunities that resulted in the Vioxx drug disaster? Can you 
assure the American people that these kind of lapses will not 
happen again?
    Dr. Woodcock. The lapses that we identified were the fact 
that a communication about the findings of the VIGOR trial were 
not effective in reaching the prescribers and the patients 
despite a label change that was made. For example, most 
prescribers really did not know about this finding, and so we 
are making much greater efforts now to communicate, as I said 
earlier, directly to prescribers and to patients so that they 
are aware of these findings. Everyone needs to recognize that 
today we know more about drug safety than ever before. One of 
the consequences of that is that we find out things that are 
bad. We find out more about adverse events of drugs.
    A good example is the estrogens. A very large trial was 
done at NIH, and it was found that postmenopausal use of 
estrogens increase several serious conditions including heart 
disease and breast cancer. This drug had been on the market for 
45 years. It is not a cause for dismay that we are learning 
these things. It means that we are making medical progress and 
we finally have the means to understand these problems. What we 
need to do is move forward in that area and continue to learn 
more about drug side effects and how to avoid them.
    Senator Kennedy. I think it is a good thing to get that 
information out to the public as quickly as possible. At the 
Tuesday hearing we heard the FDA lacked the ability to require, 
not just to request, labeling changes to drugs already on the 
market. I am sure much of the public simply assumed that FDA 
had the basic authority to assure the safety of the medicines 
that they take. Maybe equally shocking for the public is to 
realize the FDA cannot require manufacturers to conduct follow-
up safety studies on drugs already on the market. The Chairman 
pointed out that you have the power, you can withdraw it, so 
you have the heavy hand. But withdrawing a drug disregards the 
value that a particular drug, even I think the COX-2 drugs, 
would have for some patients on it, and withdrawal is a very 
dramatic step. Do you not think that relabeling authority would 
be helpful in avoiding safety problems and also do you not 
think FDA should be able to require safety studies of approved 
drugs if there is a public health need to do so?
    Dr. Woodcock. I think there are tradeoffs there and I think 
that is something that has been debated a long time and that 
Congress will debate. I do believe the steps we have taken to 
have information directly from FDA to the public, to the 
practitioners and to the patients will help deal with some of 
these problems. If there is a need for an additional study to 
be done it is going to become quite apparent as we communicate 
this safety information.
    Senator Kennedy. My time is up, Mr. Chairman. Thank you.
    The Chairman. Thank you.
    Senator Burr.
    Senator Burr. Thank you, Mr. Chairman.
    Welcome, Dr. Woodcock. It is great to see you. Thank you 
for what you do. Let me just ask you very plainly, can 
manufacturers go to market on a new drug that has been approved 
where the FDA has not agreed to the labeling on that original 
package?
    Dr. Woodcock. No.
    Senator Burr. So no product that is approved by the FDA can 
be placed on the market unless a manufacturer has had a sign-
off by the FDA on the original labeling?
    Dr. Woodcock. That is correct.
    Senator Burr. I think individuals have suggested that 
either somebody hid something or the FDA's process is in fact 
broken. Do you believe that the FDA is approving dangerous 
drugs to go on the market today?
    Dr. Woodcock. I believe on the basis of my professional 
experience, which is very extensive, and that my knowledge of 
approvals in past decades, we have the strongest and most 
detailed scientific evaluation of drugs that has ever occurred 
anywhere in the world today.
    Senator Burr. So our approval process is not broken?
    Dr. Woodcock. No. It is stronger than it has ever been.
    Senator Burr. Can we all agree that it could get better?
    Dr. Woodcock. It has got to get better.
    Senator Burr. Can you tell me how many labels Vioxx has 
had?
    Dr. Woodcock. No. [Laughter.]
    Senator Burr. It was approved in May of 1999 I believe. I 
think with material changes, it is either two or three. I would 
ask the chairman for unanimous consent that the committee allow 
the labels that have been available for Vioxx to be included as 
part of the record, if I could get the chairman's attention for 
one second. [Laughter.] Mr. Chairman? Mr. Chairman, if I could 
get your attention for 1 minute, I would ask unanimous consent 
that the labels that Vioxx has carried on its packaging since 
May of 1999 be included as part of the record.
    The Chairman. Without objection.
    Senator Burr. Thank you.
    [The Vioxx labels follow:]

    [Editors Note-Due to the high cost of printing, previously 
published materials submitted by witnesses may be found in the files of 
the committee.]

    Senator Burr. I think it is important because it is my 
understanding that the original labeling for Vioxx included 
under adverse events heart attack, stroke, congestive heart 
failure, high blood pressure. It was not black-boxed, it was 
not something that was emphasized, but it was an indication 
that had been discovered in the clinical trials. Is that your 
understanding?
    Dr. Woodcock. I do not know about myocardial infarction of 
heart attack. These other ones that you mention are true for 
all the anti-inflammatory agents.
    Senator Burr. So there is a common thread that runs through 
these and one would expect physicians to be fairly well aware 
of that.
    Dr. Woodcock. Most are.
    Senator Burr. Let me ask you, there are some that suggest, 
I think all suggest we need a more robust postapproval 
surveillance process. Some suggest that that has to be outside 
of the FDA. Can I ask you for your professional opinion on 
whether it needs to be inside the FDA or outside the FDA first?
    Dr. Woodcock. There are two issues. One issue is a robust 
surveillance system, and that means more active surveillance, 
better access to the data that is out there about the use of 
drugs in this country. That needs to be enhanced. It does not 
need I think to be one place or another. It needs to be 
enhanced.
    Senator Burr. Let me stop you there if I could because I 
want you to clarify another thing for the committee. Some have 
suggested that the clinical data should be made available not 
just to the physician world, but to the general public. In your 
professional opinion do you believe that the general public can 
disseminate clinical data in a way that it would be useful to 
their decision process?
    Dr. Woodcock. Can you clarify which clinical data you are 
referring to?
    Senator Burr. It would be the clinical trials.
    Dr. Woodcock. We have highly-trained scientists. We have to 
train--and they have all gone to medical school or gotten their 
PhD's--we have to train them additionally once they arrive at 
the FDA to analyze the raw data on clinical trials, and we are 
extremely expert at that, but we can always improve.
    That said, it is not something that the general public 
really could evaluate. Before we had electronic data we would 
get this in a tractor-trailer coming to the FDA. It is a very 
large amount of data that is involved. We have already said we 
believe summaries of information and results should be made 
available to the public.
    Senator Burr. If you could finish the answer that I 
interrupted you on.
    Dr. Woodcock. Yes. As far as who should make decisions 
around drug safety, that is the second part. First we need a 
strong surveillance system in this country. Then the question 
is who makes the decisions. Some people have said they believe 
that members of the FDA who were involved in the decisionmaking 
have some sort of bias, an intellectual bias or whatever 
against removing the drug. For that reason FDA has proposed 
that we put together a board composed of qualified experts from 
both within the FDA and external, none of whom would have been 
involved in the decision to approve that drug, to address that 
specific issue.
    But in general, in my professional opinion, it is very 
important to have people involved in decisionmaking who 
actually treat patients with that condition, because if you 
only look at risk we would not have any drugs. It would not be 
sensible to have any drugs because they all cause harm. Even 
acetaminophen very common drug, over-the-counter, number one 
cause of drug-induced liver failure in the United States. If 
you took that fact in isolation what you would say is that 
should not be on the market. So you have to have the benefit 
side right in front of you when you are evaluating the safety 
problems of drugs.
    Senator Burr. My time has run out, but you do uphold the 
FDA's proposal that that be housed within the FDA but with 
outside individuals, outside experts, outside docs coming in 
and participating in the review process?
    Dr. Woodcock. That is correct.
    Senator Burr. Thank you.
    Thank you, Mr. Chairman.
    The Chairman. Thank you, Senator Burr.
    Senator Murray.
    Senator Murray. I believe Senator Reed was ahead of me.
    The Chairman. Oh, I am sorry. Senator Reed?
    Senator Reed. Thank you, Mr. Chairman.
    Thank you, doctor, for your testimony. Let me continue a 
point that Senator Burr raised, the issue of two standards that 
you have to apply when reviewing a drug: safety and efficacy, 
and also the standard of risk benefit analysis. Those standards 
seem in some cases contradictory or at least have to be 
balanced. Can you elaborate on how you are doing that in a 
practical situation?
    Dr. Woodcock. The FDA is charged with making sure that 
drugs on the market are safe and effective. We interpret that 
safety to mean that the benefits of the drug in the intended 
population outweigh its risk, because there is really no way 
that we can approve drugs that are absolutely without risk. So 
when we say drugs are effective and safe, we mean that their 
proven demonstrated benefits from the clinical trials outweigh 
the risks of the drug for the people that it is indicated for.
    Senator Reed. Has that always been the regulatory standard 
or is this something that had evolved over the last several 
years. Has there been any recent change that we should note 
here?
    Dr. Woodcock. That has always been the regulatory standard. 
The change over the past decades is now drugs are studied in 
more people before they are put on the market, and we know more 
about them.
    Senator Reed. Let me turn to another topic, clinical 
trials. We acknowledge that every drug goes on the market with 
a clinical trial, but I think we also acknowledge that all 
clinical trials are not equal. Some are testing the best 
hypotheses, some might not have the best hypotheses. Some might 
have a long longitudinal survey. Some might be very quick. 
Practically speaking, I think you would acknowledge that there 
are differences in clinical trials. How do you deal with 
different kinds of trials in terms of making a decision about 
putting a drug on the market?
    Dr. Woodcock. That is a good question, and I used earlier 
the analogy of the building inspectors and the building code. 
Not only does FDA do the building inspection, we look at the 
results at the end, we write the code, so we say what the 
standards are for showing safety and what the standards in the 
clinical trials would be for showing effectiveness. So we would 
say, ``You have to test these patients in this manner for 3 
months, 6 months, whatever we say, and show these endpoints for 
effectiveness, and you have to do it twice. You have to 
replicate the results in another trial.'' So we set those 
standards and then when we are doing the review, we are 
reviewing against the standards that have been established.
    Senator Reed. And there is a constant process of evaluating 
how good you are in setting these standards internally.
    Dr. Woodcock. Yes. We try to revise our guidance as we 
learn from things that have gone wrong, from things that have 
gone right in the clinical trials and post marketing.
    Senator Reed. Without elaborating in detail on the latest 
situation with Vioxx, have you reevaluated the standards you 
set and the trial dynamics that you put in place?
    Dr. Woodcock. Yes. When we discussed with our advisory 
committee a couple weeks ago, one of the issues that was raised 
is that drugs that are intended to treat symptomatic 
conditions, say arthritis, typically if they are just for 
symptoms, those drugs have not been studied for a long time in 
the clinical trials, 3 months, 5 months, something like that. 
That has increased over time. In other words we require much 
more study than we used to 10 years ago, 15 years ago. 
Nevertheless it was clear that there probably, even though 
Vioxx and Celebrex had more patients than any other anti-
inflammatory drug had had before they got on the market, it 
still was not enough people and enough time to detect these 
particular problems. So we have to continue to evaluate how to 
deal with that.
    Senator Reed. A final question since I have very little 
time left. That is, we have passed the PDUFA bill. We have 
passed the FDA Modernization Act. Part of that is to streamline 
the delivery of drugs to the marketplace, but inherent in that 
is at least the issue of whether or not the streamlining has 
curtailed clinical trials, curtailed judgment about putting 
drugs in the marketplace. Very, very briefly, what is your 
conclusion?
    Dr. Woodcock. Well, again, I think that you have to 
separate the time and the standards for doing the clinical 
trials which occurs before they are sent to FDA from what is 
the subject of the user fee legislation which has to do with 
how fast the FDA reviews those results once they are in house. 
What the user fee program has also given us is more experts who 
can work on setting the standards during the clinical trials, 
and although we have not advanced as much as I would like in 
the clinical outcome area, we have made tremendous advances in 
clinical pharmacology in many of the underlying specialties, 
and these have improved drug safety quite a bit.
    Senator Reed. Is it fair to say that these legislative 
initiatives have basically changed the response of FDA, but not 
the requirements of the investigators to pursue these 
investigations, and has not materially in your view affected 
the standards for clinical trials or for evaluation?
    Dr. Woodcock. No. I think the clinical standards have 
remained the same. There have been many additional safety 
standards added over the past decade as we have learned more. 
You can ask the pharmaceutical companies about this because 
they are constantly saying, look, you now have to test 
electrophysiologic parameters for the heart and certain other 
types of drug metabolism and so forth. So these things have 
been added. The overall standards have not been lowered, but we 
can review drugs faster because we have more experts.
    Senator Reed. Thank you very much.
    Thank you, Mr. Chairman.
    The Chairman. Senator Murray.
    Senator Murray. Thank you, Mr. Chairman. I really 
appreciate your having these hearings. I think it is really 
giving us some good insight and information and will help us 
move forward to some possible legislative solutions.
    I do want to recognize a witness from the second panel who 
is with us from the University of Washington, Dr. Bruce Psaty. 
He is a professor of medicine, epidemiology and health services 
at the University of Washington, and I appreciate him traveling 
all the way out here to participate.
    Let me just begin, Dr. Woodcock, with we have heard a lot 
about the FDA approval process and the potential conflict of 
interest that may be there because of the PDUFA fee. Can you 
tell us if you can the number of new drug applications and the 
number of new drugs that have been approved versus how many 
have been disapproved?
    Dr. Woodcock. I cannot tell you the numbers out of my head.
    Senator Murray. Percentage perhaps?
    Dr. Woodcock. Traditionally I think about 25 to 30 percent 
are not approved. I do not believe that number has changed much 
over time, over the decades.
    Senator Murray. So even with the PDUFA fee, the number 
disapproved is about the same percentage.
    Dr. Woodcock. I would like to say though that although that 
means perhaps we have not decreased our standards, it is a 
failure of the enterprise. Those 20 percent of drugs that are 
turned down, many patients were exposed to those drugs during 
development and gave their time and effort to trying to see 
whether they were safe and effective. We need to do better on 
prediction. That is not a good number really.
    Senator Murray. What is the average length of time for new 
drug approval today?
    Dr. Woodcock. For priority drugs that are felt to bring a 
public health advance or an advance in therapy of some kind, it 
is about 6 months for FDA to review those, and for the standard 
drugs it is about 13.
    Senator Murray. About 13 months, okay. I want to follow up 
on what Senator Reed was asking you about in terms of clinical 
trials and working with the manufacturers. Do you work with 
them to make sure that they are not just focusing on whether a 
drug is effective or whether or not there are risks attached to 
that drug?
    Dr. Woodcock. One of the innovations in the user fee 
program, which I support highly, is greater interaction of the 
regulators with the drug companies while they are doing the 
clinical, the human experimentation phase. What we do, we tell 
them we are going to expect, say, you to have 500 patients 
exposed for this amount of time, and we would expect you to do 
these kind of safety tests during the clinical trials. So there 
is an opportunity for the FDA to intervene and explain its 
expectations on safety and effectiveness during the trials.
    Senator Murray. So you are not just looking at whether the 
drug is effective, you are working with them to make sure that 
we are looking at risks?
    Dr. Woodcock. We just did an internal survey of how much 
effort we spend on safety versus effectiveness, and we found 
for the whole center we spend 50 percent of our time on safety.
    Senator Murray. You were also asked by Senator Burr about 
this independent office on drug safety. I think you answered 
him that you did not think it was necessary. I would like to 
ask a few questions about that. Are the same FDA employees 
reviewing the new drug applications and the postmarket adverse 
events data, do you know?
    Dr. Woodcock. Not the data. The way the structure is set 
up, the Office of New Drugs continues to follow a drug 
throughout its life cycle until it becomes generic, in which 
case the Officer of Generic Drugs takes over some of that. The 
postmarketing surveillance system which gets our MedWatch 
information, which is called our Ayres database, is operated by 
the Office of Drug Safety. They run the surveillance system, 
and so they are looking at that data as it comes in.
    Senator Murray. So the manufacturer is not working with the 
same employee throughout the entire----
    Dr. Woodcock. Not necessarily on the MedWatch data, but 
they are working with the same division on their label all the 
way through the life cycle of the drug.
    Senator Murray. What kind of processes do you have in place 
to protect against any conflict of interest?
    Dr. Woodcock. We have a team approach to review. We have a 
large number of individuals who are involved in a hierarchical 
approach to review so that supervisors also look at the work of 
the primary reviewers, and that is all brought together in a 
team process.
    Senator Murray. Is there a concern about conflict of 
interest or do you feel that you have enough employees?
    Dr. Woodcock. The individuals reviewing all this 
information are physicians. They are at the Agency rather than 
out in private practice or industry because of their public 
health orientation. And there may be a concern about this 
intellectual bias because people were involved in approving a 
drug, and that is what I have heard.
    Whenever we have a serious drug safety issue, more people 
are involved in looking at that, people who had no involvement 
in the pre-market decision. And the new procedures that have 
been put in place will even open up that more. So that should 
take care, in my mind, of the issue of is there some bias by 
those who were actually involved in the beginning.
    Senator Murray. What would the drawbacks be of having an 
independent office?
    Dr. Woodcock. Can you define ``independent?''
    Senator Murray. As I am hearing it defined by people who 
are suggesting it, separate from FDA looking at the post 
market.
    Dr. Woodcock. I think people do not understand totally the 
drug life cycle. Medications are often approved for very narrow 
indications. The FDA remains involved as new trials are done. 
Say, cancer drugs, they are often approved for the most 
desperate cases, and then additional trials are done to see 
what the real use of that cancer drug would be in the wider 
population. Pediatric trials are done and so forth. So all the 
time after the drug has been approved, there is a very active 
process going on with new indications being added, new 
warnings. It is really a continuum before the pre-market and 
the postmarket. It is by no means that we sort of forget this 
drug after it is approved.
    So you need the input of all the experts on the drug and 
the tremendous amount of expertise they have gained about that 
drug and maybe that class of drugs before marketing, in the 
postmarket period as well.
    Senator Murray. I appreciate that very much. Thank you.
    Thank you, Mr. Chairman.
    The Chairman. Thank you.
    Senator Isakson.
    Senator Isakson. Thank you, Mr. Chairman.
    I apologize that I am late and did not hear your testimony 
and might be redundant in my questions, so be very brief or 
tell me I am redundant and to look it up in the book.
    I think following up on the previous question, your opinion 
on this external review, just real quickly, what is your take 
on that or the need for it?
    Dr. Woodcock. I think of course the FDA is open to various 
approaches and suggestions, but people need to have the 
correct--say in medicine you have to have the correct diagnosis 
before you move into treatment. I think people really need to 
think about what problem are they trying to address before 
thinking about what treatment to apply.
    As I was just saying, the postapproval phase for a drug is 
a very active one where more clinical trials are going on, the 
label is being changed, the drug is being studied in additional 
populations. So the involvement of the medical team in the drug 
also continues throughout that time, and generally these are 
subspecialists. We have neurosurgeons, neurologists, 
gastroenterologists and so forth all across the medical 
subspecialties. They have very deep and profound knowledge of 
their particular area and the use of therapeutics in that area.
    So it would be difficult to imagine moving that process to 
some other unit, but perhaps people mean something else by what 
they are saying with the independent safety board.
    Senator Isakson. So within FDA following the approval of a 
drug there is an ongoing tracking of the effects of that--or 
any effects that are reported by physicians prescribing that 
drug, and that comes to FDA?
    Dr. Woodcock. Yes. That part comes to the Office of Drug 
Safety, but all the clinical trials and oversight of those 
clinical trials and the safety of the subjects and how the 
trials are designed and so forth, those are done by the medical 
divisions.
    Senator Isakson. How does that come back to FDA, that 
information, on paper and pencil or----
    Dr. Woodcock. It is normally electronic, and these 
additional studies have to be done under an IND because human 
subjects are being exposed in experimental conditions. So the 
FDA has a great deal of interaction and oversight into those 
ongoing trials.
    Senator Isakson. Are you one that subscribes to a belief 
that I am coming to believe, and that is, one of the greatest 
things we could do in health care overall is get technology 
involved so this information flows seamlessly and accurately 
from physician and pharmacist and patient, and there is an 
integration of that information so it can be pulled out 
quickly?
    Dr. Woodcock. You are reflecting my testimony. I totally 
agree with that, and we have been working very hard at the FDA 
to do our part to make that happen.
    Senator Isakson. Are you familiar with a company by the 
name of Greenway?
    Dr. Woodcock. No.
    Senator Isakson. After the hearing I will give you a note. 
There is some development on that end that is now actually out 
in practice in my home State of Georgia that is showing great 
promise in terms of information technology, patient, doctor, 
pharmacist.
    Last I just have a comment. You all take a lot of knocks 
lately and there has been a lot of criticism. I would like to 
just tell a story for the sake of the chairman and the 
committee. On the 12th of September in 2001 a pharmaceutical 
company in my district contacted me, Solvay. They had a burn 
treatment known as Flamazine, which was in its final--I do not 
know if I am using the right word but I remember they kept 
using the word ``protocol''--before issuance. New York City had 
run out of the only other approved ointment of this type, and 
obviously we had a serious, tragic--I just want everybody to 
know that FDA's response that day and the ability to get those 
badly-needed medicines to New York City in a time of great 
crisis was nothing short of unbelievable, which I do not think 
portends that you only do timely work in an emergency, I might 
add, but you do good work all the time. I appreciate what you 
do.
    That is all of my questions and comments, Mr. Chairman.
    The Chairman. Thank you. That reminds me that I refer to 
this hearing room as the reassurance room. Following September 
11th, while most of the Senators were up in New York looking at 
Ground Zero, those of us in the Banking Committee were in this 
room holding hearings with the stock market people to reassure 
America that the stock market was still working, that it would 
open on the following Monday, and that everything would be 
fine, and that there was plenty of backup.
    Part of what we are doing with these hearings is giving 
some reassurance on the condition of FDA in getting their 
insight and others' insights into the kinds of things we can do 
to do the job even better.
    I thank you very much for your testimony. We will keep the 
record open for another 10 days. That will give you a chance if 
you want to expand on anything that you have answered today, 
and Senator Kennedy had one particularly difficult one to 
answer just on the spur of the moment in the timeframe that we 
have, and also give members of the panel an opportunity to 
address additional questions that may be more technical than 
the general public might be interested in, but that would be 
helpful to our decisions. Thank you very much for testifying 
today.
    [The joint prepared statement of Dr. Woodcock and Dr. 
Kweder follows:]

        Joint Prepared Statement of Dr. Woodcock and Dr. Kweder

                              INTRODUCTION

    Mr. Chairman and members of the committee, I am Dr. Sandra Kweder, 
Deputy Director of the Office of New Drugs at the Center for Drug 
Evaluation and Research (CDER), United States Food and Drug 
Administration (FDA or the Agency). I am pleased to be here today to 
discuss drug safety and the drug approval process.
    Because of the importance of these issues, you are holding two 
hearings over the course of 3 days. Dr. Janet Woodcock, FDA's Acting 
Deputy Commissioner for Operations, will appear at your hearing on 
March 3. We have one written statement to address both hearings.

                       SAFETY IS A HIGH PRIORITY

    Let me begin with a few words about safety, and I will return to 
this issue throughout our written testimony. Modern drugs provide 
unmistakable and significant health benefits. FDA's drug review process 
is recognized worldwide as a gold standard. Indeed, we believe that FDA 
maintains the highest standards for drug approval. There have been 
significant additions to those standards during the last several 
decades, in response to advances in medical science. Currently, FDA 
approves drugs after they are studied in many more patients and undergo 
more detailed safety evaluation than ever before. FDA grants approval 
to drugs after a sponsor demonstrates that their benefits outweigh 
their risks for a specific population and a specific use, and that the 
drug meets the statutory standard for safety and efficacy. However, no 
amount of study before marketing will ever elucidate all the 
information about effectiveness or all the risks of a new drug. 
Therefore, post-marketing surveillance is extremely important.
    Adverse effects that are not detected during clinical trials are 
identified after approval through post-marketing clinical trials, 
spontaneous reporting of adverse events, or observational studies based 
on more widespread use of the product following approval. That is why 
Congress has supported and FDA has created a post-market drug safety 
program designed to collect and assess adverse events identified after 
approval for all drugs we regulate.
    This program serves as a complement to the pre-market safety 
reviews required for approval of prescription drugs in the U.S. FDA 
also evaluates and responds to adverse events identified in ongoing, 
post-market clinical trials that test approved drugs for other 
indications. We also evaluate and respond to events reported by 
physicians, their patients, or drug manufacturers. With this 
information, we make label changes and take other regulatory action as 
needed.
    It is important to emphasize that all approved drugs pose some 
level of risk, such as the risks identified in clinical trials and 
listed on the labeling of the product. Unless a new drug's demonstrated 
benefit outweighs its known risks for its intended population, FDA will 
not approve the drug. However, we cannot anticipate all possible 
effects of a drug based on data from the clinical trials that precede 
approval.

             NEW FDA INITIATIVES TO STRENGTHEN DRUG SAFETY

November 2004 Five-Step Plan

    At FDA, we are constantly striving to improve our processes and 
methods, and thereby better serve the public health. Recent 
developments have prompted us to refocus our drug safety efforts and 
take additional steps to identify drugs that may have unacceptable risk 
profiles.
    On November 5, 2004, Acting Commissioner Crawford announced a five-
step plan to strengthen FDA's drug safety program. First, it called for 
FDA to sponsor an Institute of Medicine (IOM) study to evaluate the 
current drug safety system. An IOM committee will study the 
effectiveness of the U.S. drug safety system, with an emphasis on the 
post-marketing phase, and assess what additional steps FDA could take 
to learn more about the side effects of drugs as they are actually 
used. We will ask IOM to examine FDA's role within the health care 
delivery system and recommend measures to enhance the confidence of 
Americans in the safety and effectiveness of their drugs.
    Second, Dr. Crawford announced that CDER would implement a program 
for addressing differences of professional opinion. I am pleased to 
report that CDER recently put this program into effect. Currently, in 
most cases, free and open discussion of scientific issues among review 
teams and with supervisors, managers and external advisers, leads to an 
agreed course of action. Sometimes, however, a consensus decision 
cannot be reached, and an employee may feel that his or her opinion was 
not adequately considered. Such disagreements can have a potentially 
significant public health impact.
    In an effort to improve the current process, CDER has formalized a 
program to help ensure that the opinions of dissenting scientific 
reviewers are formally addressed and transparent in its decision-making 
process. An ad hoc panel, including FDA staff and outside experts not 
directly involved in disputed decisions, will have 30 days to review 
all relevant materials and recommend to the Center Director an 
appropriate course of action.
    Third, CDER will conduct a national search to fill the currently 
vacant position of Director of the Office of Drug Safety (ODS), which 
is responsible for overseeing the post-marketing safety program for all 
drugs. CDER is seeking a candidate who is a nationally recognized drug 
safety expert with knowledge of the basic science of drug development 
and surveillance, and a strong commitment to protecting the public 
health. CDER is working with the Office of Personnel Management on this 
search.
    Fourth, in the coming year CDER will conduct additional workshops 
and advisory committee meetings to discuss complex drug safety and risk 
management issues. Most recently, for example, the Agency conducted a 3 
day Advisory Committee meeting that examined COX-2 selective non-
steroidal anti-inflammatory drugs and related medicines. The committee 
held its meeting on February 16-18, 2005, and heard presentations from 
more than 25 experts. At the end of the meeting, the Advisory Committee 
issued recommendations that the Agency is promptly and carefully 
reviewing before taking further action.
    Finally, FDA intends to publish final versions of three guidances 
that the Agency developed to help pharmaceutical firms manage risks 
involving drugs and biological products. These guidances should assist 
pharmaceutical firms identify and assess potential safety risks not 
only before a drug reaches the market but also after a drug is already 
on the market. FDA expects to publish the final guidances in the second 
quarter of 2005.

February 2005 Drug Safety Announcement

    On February 15, 2005, HHS Secretary Leavitt and Acting Commissioner 
Crawford unveiled a new, emboldened vision for FDA that will promote a 
culture of openness and enhanced oversight within the Agency. As part 
of this vision, FDA will create a new independent Drug Safety Oversight 
Board (DSB) to oversee the management of drug safety issues, and will 
improve transparency by providing emerging information to health 
providers and patients about the risks and benefits of medicines.
    Under this proposal, FDA will enhance the independence of internal 
deliberations and decisions regarding risk/benefit analyses and 
consumer safety by creating an independent DSB. The DSB will oversee 
the management of important drug safety issues within CDER. The DSB 
will be comprised of individuals from FDA who were not involved in the 
initial review of the drug, as well as medical experts from other HHS 
agencies and government departments (e.g., the National Institutes of 
Health and Department of Veterans Affairs). CDER's Deputy Director will 
serve as the Chair of the DSB. The DSB also will consult with other 
medical experts and representatives of patient and consumer groups.
    FDA will also increase the transparency of the Agency's decision-
making process by establishing new and expanding existing communication 
channels to provide drug safety information to the public. These 
channels will help ensure that established and emerging drug safety 
data are quickly available in an easily accessible form. The increased 
openness will enable patients and their health care professionals to 
make better-informed decisions about individual treatment options. The 
Agency is also proposing a new Drug Watch web page that will include 
emerging information about possible serious side effects or other 
safety risks for previously and newly approved drugs. This resource 
will contain valuable information that may alter the benefit/risk 
analysis for a drug or affect patient selection or monitoring 
decisions. The web resource may also contain information about measures 
that patients and practitioners can take to prevent or mitigate harm. 
This information resource will significantly enhance public knowledge 
and understanding of safety issues by discussing emerging or potential 
safety problems even before FDA has reached a conclusion that would 
prompt a regulatory action. As always, FDA is committed to maintaining 
patient privacy as it implements these measures.
    As FDA develops these communication formats, the Agency will 
solicit public input on how FDA should manage potential concerns 
associated with disseminating emerging information prior to regulatory 
action. The Agency will also issue draft guidance on procedures and 
criteria we will use to identify drugs and information that will appear 
on the Drug Watch web page. In addition, FDA will actively seek 
feedback from health care professionals, patients and consumers on how 
best to make this information available to them.

Increased Funding for the Office of Drug Safety

    FDA has a longstanding commitment to provide a strong resource base 
for ODS. As the graph set forth below demonstrates, we have steadily 
increased the financial and human resources dedicated to post-market 
drug safety over the past decade.
    The budget for fiscal year 2006 continues this commitment. The 
President has proposed a 24 percent increase for FDA's post-market 
safety program to help further ensure that America's drug product 
supply is safe and effective, and of the highest quality. Under this 
proposal, CDER's ODS would receive increased funding to expand the 
Agency's ability to rapidly survey, identify and respond to potential 
safety concerns for drugs on the market. ODS will hire additional staff 
to manage and lead safety reviews, will increase the number of staff 
with expertise in critical areas such as risk management, risk 
communication and epidemiology, and will increase access to a wide 
range of clinical, pharmacy and administrative databases. The 
Administration's proposed budget for ODS will increase by $6.5 million, 
including $1.5 million in user fees, for a total fiscal year 2006 ODS 
funding level of $33.4 million. PDUFA resources will represent nearly 
one-third of the ODS budget for the coming year. Our commitment to 
increase resources available for post-market safety will enhance the 
structural changes we are proposing to advance drug safety.



                       THE DRUG APPROVAL PROCESS

Pre-Approval Focus on Safety

    FDA's focus on safety begins at the earliest stages of drug 
development, when we review a product under an investigational new drug 
(IND) application. During the IND period, products must complete three 
phases of clinical (human) trials. Phase I studies involve the initial 
introduction of an IND drug into humans to assess the most common acute 
adverse effects and examine the size of doses that patients can take 
safely without a high incidence of side effects. However, before 
beginning human trials, the sponsor must perform extensive animal 
toxicity studies. Researchers closely monitor these studies. They may 
conduct Phase I trials in patients, but often rely on healthy volunteer 
subjects. In general, these studies yield initial safety data and 
useful information to establish the appropriate dose of the drug.
    Phase II includes the early controlled clinical studies conducted 
to obtain additional information on appropriate dosing, as well as 
preliminary data on the effectiveness of the drug for a specific 
indication in patients with the disease or condition. This phase of 
testing also helps identify short-term side effects and risks possibly 
associated with the drug. Phase II studies are typically well 
controlled, closely monitored and conducted in studies that usually 
involve several hundred patients. In these studies, researchers compare 
results of patients receiving the drug with those who receive a 
placebo, a different dose of the test drug, and/or another active drug. 
At the conclusion of these studies, FDA and the sponsor meet to 
determine if the drug's development should advance to Phase III and how 
to design and conduct further trials.
    Finally, researchers design Phase III trials for a larger number of 
patients and build on the data gained from the first two phases of 
trials. These studies provide the additional information about safety 
and effectiveness needed to evaluate the overall benefit-risk 
relationship of the drug. Phase III studies also provide the basis for 
extrapolating the results to the general population and provide 
essential information for the package labeling. Once the results of all 
the clinical trials are available, the sponsor of the application 
(usually the manufacturer of the product) analyzes all the data and 
submits a new drug application (NDA) or biologics license application 
to FDA for review.

Post-Approval Risk Assessment

    Once FDA approves a drug, the post-marketing monitoring stage 
begins. The sponsor (typically the manufacturer) is required to submit 
periodic safety updates to FDA on their drug. Also during this period, 
we continuously receive adverse event reports through our MedWatch 
system from other sources such as health care providers and patients. 
Safety experts review and analyze the reports to establish the 
frequency and seriousness of the adverse events. Our response to 
information from this ongoing surveillance depends on an evaluation of 
the aggregate public health benefit of the product compared to its 
evolving risk profile. FDA carefully considers the seriousness and the 
frequency of reported adverse events as well as the estimated number of 
patients who benefit from the drug. The occurrence of a rare event, 
even a serious event, may or may not, by itself, be sufficient to take 
a drug product off the market. Adverse event reports do not solely 
provide all the data necessary to identify any potential risks that may 
be associated with a specific product or class of products; however, 
over time, they provide us with another piece to a complex puzzle.
    If the public health benefit of the product outweighs its known 
risks for the intended population and intended use, FDA allows the 
continued marketing of the drug. Often, as more becomes known about the 
potential risks or benefits of a product, its label will be revised so 
that it better reflects information on appropriate use. For example, 
FDA may ask the manufacturer to revise the labeling to add information 
on adverse reactions not previously listed, to add new warnings 
describing conditions under which the drug should not be used, or to 
add new precautions advising doctors of measures to minimize risk. FDA 
often issues Public Health Advisories and information sheets for health 
care providers and patients that discuss the new safety information. In 
the event of reports of death or life-threatening injury, FDA and the 
sponsor may consider restricting the distribution of the product or 
removing it from the market. Our action will depend on the frequency of 
the reports, the seriousness of the diseases or conditions for which 
the drug provides a benefit, the availability of alternative therapy, 
and the consequences of not treating the disease.
    The issue of how to detect and limit adverse reactions can be 
challenging. How to weigh the impact of these adverse drug reactions 
against the benefits of these products on individual patients and the 
public health is multifaceted and complex, and involves scientific as 
well as public health issues.

               STATUTORY CHANGES TO DRUG APPROVAL AT FDA

    FDA was founded in response to concerns about safety, and attention 
to safety pervades everything that we do. In the Federal Food, Drug and 
Cosmetic Act of 1938, Congress gave FDA the authority to review the 
evidence that a drug was safe for its intended use. In 1962, Congress 
added a requirement that drug sponsors also demonstrate that a drug is 
effective, using adequate and well controlled studies. Thus, drug 
safety means that the demonstrated benefits of a drug outweigh its 
known and potential risks for the intended population and use. In 
recent years, Congress has enacted legislation that provides 
significant additional tools to improve our focus on safety: the 
Prescription Drug User Fee Act (PDUFA) and the Food and Drug 
Administration Modernization Act (FDAMA).
    In 1992, Congress enacted PDUFA. This landmark legislation provided 
significant resources for FDA to hire more medical and scientific 
reviewers to conduct pre-market reviews, to hire support personnel and 
field investigators to speed the application review process for human 
drug and biological products, and to acquire critical information 
technology infrastructure to support our review process.
    In 1997, following the success of PDUFA I, Congress reauthorized 
the program for an additional 5 years when it enacted FDAMA of 1997. 
With PDUFA II came higher expectations for product reviews and 
additional goals designed to reduce drug development times.
    In 2002, Congress reauthorized PDUFA for a third time. PDUFA III 
places great emphasis on ensuring that user fees provide a sound 
financial footing for FDA's new drug and biologic review process and, 
for the first time, gives FDA authority to expend PDUFA resources on 
risk management and drug safety activities during the approval process 
and during the first 2 to 3 years following drug approval. Mr. 
Chairman, your Committee played a significant role in creating and 
reauthorizing PDUFA, and on behalf of my colleagues at FDA and 
countless patients throughout America who benefit from the therapies 
approved under the PDUFA process, I thank you for your efforts.
    One of the primary goals of PDUFA was to address the significant 
delay in U.S. patients' access to new medicines. The objective was to 
increase benefits to patients, without increasing risks. Before PDUFA, 
drug lag was a serious concern for U.S. patients and practitioners. 
Life-saving drugs were available to patients in other countries months 
and sometimes years before they were available in the U.S. Because of 
the additional resources and process improvements implemented since 
PDUFA I became law, the average FDA drug review time has declined by 
more than 12 months.
    It is important to emphasize that an recent study by Berndt, et al. 
of the National Bureau of Economic Research found no significant 
differences in the rates of safety withdrawals for drugs approved 
before PDUFA compared to drugs approved during the PDUFA era. This 
research confirms FDA's analysis on the same subject. In addition, we 
are now adding black box warnings sooner than we did before PDUFA. This 
indicates that PDUFA has been successful in both speeding access and 
preserving safety.
    In general, PDUFA authorizes FDA to collect fees from companies 
that produce certain human drug and biological products. When a sponsor 
seeks FDA approval for a new drug or biologic product, it must submit 
an application accompanied by a fee to support the review process. In 
addition, companies pay annual fees for each manufacturing 
establishment and for each prescription drug product marketed. Before 
PDUFA, taxpayers alone paid for product reviews through budgets 
provided by Congress. Under the PDUFA approach, industry provides 
additional funding in return for FDA's efforts to meet drug-review 
performance goals that emphasize timelines but do not alter or 
compromise our commitment to ensuring that drugs are safe and effective 
before they are approved for marketing.

               PDUFA III--GREATER EMPHASIS ON DRUG SAFETY

    PDUFA fees are essential to our efforts to improve drug safety. Our 
trained health professionals work to help ensure and improve drug 
safety using a process of scientific review, monitoring, and analysis 
throughout the life cycle of the drugs we approve for marketing. A 
focus on safety initiates during the pre-marketing phase, when the 
earliest work on drug discovery begins. As the drug development process 
continues, we evaluate the safety of the therapeutic compound over a 
number of years during pre-clinical testing, clinical trials involving 
humans and eventually, with the submission of an NDA for FDA review. 
Thanks to PDUFA, we are able to commit far greater resources to our 
important safety responsibilities.
    Under PDUFA III, Congress granted authority for FDA to expend user 
fees for post-market safety review. FDA made this a top priority during 
our PDUFA negotiations. Beginning with PDUFA III, for drugs approved 
after October 1, 2002, we can spend PDUFA resources on ``collecting, 
developing, and reviewing safety information on drugs, including 
adverse event reports'' for up to 3 years after the date of approval. 
The initiative to address drug safety for PDUFA III products helps FDA 
better understand a drug's risk profile, provide risk feedback to the 
sponsors and provide essential safety information to patients and 
health practitioners.
    From October 1, 2002, through December 31, 2004, FDA reviewed 63 
risk management plans for drug and biologic products. Twenty-eight of 
these related to applications submitted after PDUFA III took effect. We 
also conducted pre-approval safety conferences, risk management plan 
reviews, drug safety meetings, and meetings with sponsors to discuss 
proposed drug supplements.
    In response to PDUFA III, FDA held a public meeting in April 2003 
to discuss risk assessment, risk management, and pharmacovigilance 
practices. On May 5, 2004, based on the valuable information generated 
through the meeting process, we published three draft guidances on 
these important drug safety topics. FDA received extensive comments on 
these documents, and we expect to publish all three final guidances in 
the second quarter of 2005.

                        SAFETY ADVANCES IN FDAMA

    Enacted in 1997, FDAMA has been an important addition to FDA's 
legal framework. FDAMA passed following a thorough Congressional 
examination of the Agency's policies and programs. It instituted a 
number of comprehensive changes, reaffirmed the Agency's vital role in 
protecting the public health and served as the vehicle for enacting 
PDUFA II.

Pediatric Exclusivity and Safer Use of Drugs in Children

    For decades, children were prescribed medications that had not been 
studied for safety and efficacy in pediatric populations. As a 
component of FDAMA, Congress provided incentives to sponsors to conduct 
pediatric clinical trials. Section 111 of FDAMA authorized FDA to grant 
an additional 6 months of marketing exclusivity (known as pediatric 
exclusivity) to pharmaceutical manufacturers that conduct studies of 
certain drugs in pediatric populations. The objective of section 111 
was to promote pediatric safety and efficacy studies of drugs. With the 
valuable information generated by these studies, the product labeling 
can then be updated to include appropriate information on use of the 
drug in the pediatric population. To qualify for pediatric exclusivity, 
sponsors must conduct pediatric studies according to the terms of a 
Written Request issued by FDA and submit the results of those studies 
in an NDA or supplement.
    In 2002, Congress renewed this authority when it enacted the Best 
Pharmaceuticals for Children Act (BPCA). BPCA also mandates that FDA 
report to the Pediatric Advisory Committee, in a public forum, any 
safety concerns during the 1 year period after we grant pediatric 
exclusivity. To date, we have reported safety concerns on 34 drugs at 
six separate public advisory meetings.
    Finally, BPCA contains important, new disclosure requirements. 
Outside of BPCA, the Agency generally may not publicly disclose 
information contained in an IND, unapproved NDA, or unapproved 
supplemental NDA. Once FDA approves an NDA or supplemental NDA, the 
Agency can make public certain summary information regarding the safety 
and effectiveness of the product for the approved indication.
    However, section 9 of BPCA gives FDA important new disclosure 
authority. BPCA requires that, no later than 180 days after the 
submission of studies conducted in response to a Written Request, the 
Agency must publish a summary of FDA's medical and clinical 
pharmacology reviews of those studies. Moreover, we must publish this 
information regardless of whether our action on the pediatric 
application is an approval, approvable, or not-approvable action. Thus 
under FDAMA, information on pediatric studies conducted in response to 
Written Requests was not available until after the supplemental 
application was approved. In contrast, under BPCA, a summary of FDA's 
medical and clinical pharmacology reviews of pediatric studies is 
publicly available regardless of the action taken on the application. 
Since 2002, FDA has posted the summaries of these reviews for 41 
products submitted in response to a Written Request on FDA's website 
at: (http://www.fda.gov/cder/pediatric/Summaryreview.htm). This 
information provides a rich source of valuable safety information to 
allow pediatricians to make more informed decisions about whether and 
how to use these drugs in their patients.

Post-Marketing Safety Studies

    On April 30, 2001, FDA's regulations implementing section 130 of 
FDAMA, which requires sponsors of approved drugs and biologics to 
report annually on the status of post-marketing commitments, became 
effective. These regulations modified existing reporting requirements 
for NDA drug studies and created a new reporting requirement for 
biologic products.
    FDA may request that the sponsor conduct post-marketing studies to 
provide additional important information on how a drug works in 
expanded patient populations or to identify safety issues that occur at 
very low frequency or in special patient populations. The post-
marketing safety study obligations in section 130 are of keen interest 
to patient and consumer advocates who track the completion of post-
marketing commitments and FDA's efforts to review study results and 
modify drug labeling. The regulations implementing section 130 provide 
FDA with a mechanism to monitor study progress through the annual 
submission of study status reports. FDA posts the status of post-
marketing studies on its public website and publishes an annual summary 
of industry's progress in fulfilling post-marketing commitments in the 
Federal Register.

                             CRITICAL PATH

    On March 16, 2004, FDA released a report addressing the recent 
slowdown in innovative medical therapies submitted to FDA for approval: 
``Innovation/Stagnation: Challenge and Opportunity on the Critical Path 
to New Medical Products.'' The report describes options to modernize 
the medical product development process to try to make it more 
predictable and less costly. The report focuses on ways that FDA could 
collaborate with academic researchers, product developers, patient 
groups, and other stakeholders to make the critical path much faster, 
predictable, and less costly.

Enhancing the Safety of Medical Products

    During drug development, safety issues should be detected as early 
as possible. However, because of limitations of current methods, safety 
problems are often uncovered only during clinical trials or, 
occasionally, after marketing. Despite efforts to develop better 
methods, some tools used for toxicology and human safety testing are 
outdated. Clinical testing, even if extensive, often fails to detect 
important safety problems, either because they are uncommon or because 
the tested population was not representative of eventual recipients. 
Conversely, some models create worrisome signals that may not be 
predictive of a human safety problem.
    There are opportunities for developing tools that can more reliably 
and efficiently determine the safety of a new medical product. To meet 
this challenge, FDA has called for a new focus on modernizing the tools 
that applied biomedical researchers and product developers use to 
assess the safety and effectiveness of potential new products. Many of 
these tools--diagnostics such as pharmacogenomic tests and imaging 
techniques--would also be used after marketing to monitor safety in the 
real world clinical setting. The Critical Path report describes 
opportunities for FDA, working with academia, patient groups, industry, 
and other government agencies, to embark on a collaborative research 
effort. The goal is to create new performance standards and predictive 
tools that will provide better answers about the safety and 
effectiveness of investigational products, to do this faster and with 
more certainty, and to enhance the safety of these products in the 
clinic.
    In addition to improved safety tools, Critical Path also focuses on 
tools that will help individualize therapy. We enhance safety when the 
target population does not include individuals who cannot benefit from 
the treatment. For these individuals, drug exposure is all risk. Better 
tools for individualized therapy will help to identify patients who 
will respond to therapy. New science has provided the basic knowledge 
to make these tools a reality.
    Critical Path is not a fundamental departure for FDA, but rather 
builds on the Agency's proven ``best practices'' for expediting the 
availability of promising medical technologies. While the report 
touches on all aspects of medical product development, identifying new 
tools to address drug safety challenges would represent a giant step 
down the Critical Path.

                               CONCLUSION

    At FDA, providing the American public with safe and effective 
medical products is our core mission. We base decisions to approve a 
drug or to keep it on the market if new safety findings surface on a 
careful balancing of risk and benefit to patients. This is a 
multifaceted and complex decision process, involving scientific and 
public health issues. The recent initiatives we have announced will 
improve our current system to assess drug safety. Moreover, as we 
strive for continuous improvement, we will continue to evaluate new 
approaches to advance drug safety. As always, we value input from 
Congress, patients and the medical community as we develop and refine 
these drug safety initiatives.
    Once again, thank you for the opportunity to testify before the 
Committee today. I am happy to respond to questions.

    The Chairman. As the next panel takes their place, I will 
go ahead with introductions. On the next panel we have Dr. 
Cecil B. Wilson, who is an internist from Winter Park, FL and a 
member of the AMA Board of Trustees since 2002. Dr. Wilson has 
been in private practice of internal medicine in Central 
Florida for 30 years. He is board certified in internal 
medicine and a member of the American College of Physicians. 
Dr. Wilson will discuss the impact on prescribers of changes in 
the way FDA communicates.
    Also I have Mr. Keith L. Carson, Chairman of the 
Williamsburg BioProcessing Foundation. Mr. Carson started the 
foundation in 1994 and currently serves as the organization's 
chairman. He edits BioProcessing Journal, a print journal that 
features articles adapted from selected presentation given at 
the foundation's conferences. Mr. Carson will discuss the 
impact of new technologies on drug safety and how FDA can 
improve its processes, including through the developing of 
reference materials.
    Dr. Raymond Woosley is a pharmacologist whose research has 
been published in over 250 publications, and has investigated 
the basic and clinical pharmacology of drugs for the drug 
treatment of arrhythmias and the cardiac toxicity of drugs. In 
January of 2005 he assumed the position of the President of the 
Critical Path Institute, C-Path, a nonprofit corporation formed 
by the Food and Drug Administration, SRI International and the 
University of Arizona to accelerate the developing of safe 
innovative medicines. Dr. Woosley will discuss his ideas on how 
to increase the industry capabilities to develop innovative 
methods for accelerated drug discovery and development and how 
the FDA may have to change to adapt to these new methods.
    We have Dr. Bruce Psaty. Dr. Bruce Psaty is a Professor of 
Medicine and Epidemiology and Co-Director of the Cardiovascular 
Health Research Unit at the University of Washington in 
Seattle. A practicing general internist at Harbor View Medical 
Center, he is a cardiovascular disease epidemiologist with 
interest and expertise in pharmacoepidemiology, 
pharmacogenetics and drug safety. He will discuss his 
recommendations for improving FDA's drug safety process.
    I thank the panel and we will begin then with the testimony 
of Dr. Wilson.

 STATEMENTS OF CECIL B. WILSON, M.D., MEMBER, AMERICAN MEDICAL 
   ASSOCIATION BOARD OF TRUSTEES, WINTER PARK, FL; KEITH L. 
 CARSON, CHAIRMAN, THE WILLIAMSBURG BIOPROCESSING FOUNDATION, 
VIRGINIA BEACH, VA; RAYMOND WOOSLEY, M.D., PhD, PRESIDENT, THE 
      CRITICAL PATH INSTITUTE, PROFESSOR OF MEDICINE AND 
 PHARMACOLOGY, UNIVERSITY OF ARIZONA, TUCSON, AZ; AND BRUCE M. 
   PSATY, M.D., PhD, PROFESSOR OF MEDICINE, EPIDEMIOLOGY AND 
 HEALTH SERVICES, CO-DIRECTOR, CARDIOVASCULAR HEALTH RESEARCH 
           UNIT, UNIVERSITY OF WASHINGTON, SEATTLE WA

    Dr. Wilson. Good morning, Chairman Enzi and members of the 
committee. My name, as you have heard, is Cecil Wilson. I am a 
member of the American Medical Association's Board of Trustees 
and a practicing internist in Winter Park, FL.
    On behalf of the AMA I would like to thank you for holding 
today's hearings and for inviting us to participate. Today I 
will discuss how FDA decisions regarding drug approval 
postmarketing surveillance, product labeling, off-label use and 
risk management impact practicing physicians like myself. I 
will also discuss the AMA's recommendations to improve drug 
safety and minimize the impact on physicians' ability to 
practice medicine.
    Our recommendations include more active approaches to 
postmarketing surveillance, a final FDA rule on package 
inserts, the preservation of off-label prescribing, and 
continued collaboration between the FDA, the pharmaceutical 
industry and physicians to develop better risk communication 
tools. Approving a prescription drug or biologic for marketing 
is a primary way in which the FDA affects physician practice.
    Since PDUFA was passed in 1992 new drugs are getting to 
market faster and importantly, and as you heard earlier, 
studies have shown that this has been accomplished without 
increasing the number of drug withdrawals. The AMA hopes that 
any new efforts to enhance drug safety can be accomplished 
without reversing this trend. As more drugs become available, 
the AMA recognizes the need to improve postmarketing 
surveillance. Such efforts would enhance our ability to 
identify rare but potentially serious adverse events in a 
timely fashion. So the AMA supports active approaches to 
postmarketing surveillance. For example, well-designed studies 
on newly marketed drugs would help to quickly assess the risk 
of these drugs once they are in actual clinical use.
    Another primary way in which FDA decisions affect 
physicians is through product labeling, especially the package 
insert. The package insert is designed and intended to inform 
physicians about risk and benefits of a drug. Unfortunately, 
today's package insert has become a long and complicated really 
legal document rather than a useful resource for physicians.
    In my own practice when a patient presents who is on a new 
drug from another physician or when I simply want to check or 
double check the dosage of a particular drug, I, as well as 
other physicians, look up a drug's package insert. The problem 
we encounter is that the package insert contains so much 
information that it makes it difficult to find what we really 
need. So information such as dosage, contraindications, major 
risk and potential drug interactions are often varied within 
the highly technical text of the document.
    In the year 2000 the FDA issued a proposed rule to make the 
package insert more user friendly for physicians. The AMA 
supports this effort and urges the FDA to finalize this rule. 
Further, the FDA should ensure that physicians' ability to 
prescribe drugs off label not be impeded. In some instances 
prescribing a product off label is the most appropriate therapy 
based on the latest and best scientific evidence, and for some 
patient populations it may be the only treatment.
    Finally, Mr. Chair, over the past few years the FDA has 
opposed a number of risk management tools to enhance drug 
safety. Some of these tools may be useful, but some could lead 
to unintended consequences such as decrease patients' access to 
valuable medical treatments. For the vast majority of 
prescription drugs the patient insert, combined with effective 
postmarketing surveillance should constitute the risk 
management plan. Additional risk management tools such as 
patient agreements, enrollment programs or tools that create 
special rules for prescribing should be used only as a last 
resort to keep products with unique and important benefits on 
the market.
    Thank you, Mr. Chairman.
    [The prepared statement of Dr. Wilson follows:]

              Prepared Statement of Cecil B. Wilson, M.D.

    The American Medical Association (AMA) appreciates the opportunity 
to present its views on ways to ensure drug safety in this country and 
the implications for practicing physicians. We commend the Chairman and 
Members of this Committee for holding this important hearing. The AMA 
shares a common goal with Congress and the Food and Drug Administration 
(FDA) to optimize the benefit/risk balance of drug therapy and minimize 
the risks of prescription drug and biologic products.
    As our Nation's drug regulatory agency, the FDA ensures that 
beneficial drug products are made available to the public with labels 
that contain adequate information about the product's risks and 
benefits, and protects the public from false claims. While the FDA's 
approval process is considered the ``gold standard'' around the world, 
the FDA's determination that a product is safe and effective is not 
meant to signal an absence of risk. Drug and biologic products, by 
their very nature, carry with them certain risks, some of which are 
discovered after approval. Pharmaceutical manufacturers, the FDA, 
physicians and patients all play essential roles in minimizing those 
risks and enhancing the benefits of prescription drugs and biologics.
    The AMA supports the FDA's proposals to improve the format and 
content of the package insert, which is the portion of a drug product's 
labeling directed primarily to physicians. We have also been a 
proponent of more widespread use of the MedWatch program (FDA's adverse 
event reporting system) by encouraging physicians to participate. More 
recently, the AMA provided testimony and commentary on specific FDA 
initiatives related to the risk management of prescription drugs, 
including their concept paper on ``Risk Management Programs'' and their 
draft guidance for industry on the ``Development and Use of Risk 
Minimization Action Plans'' (Attachments 1 & 2).
    This statement will focus on how FDA decisions impact practicing 
physicians through the drug approval process; postmarketing 
surveillance efforts; product labeling developed to guide physicians in 
the appropriate use of medications; policies on unlabeled uses; and 
risk management. In addition, we make recommendations to improve drug 
safety and minimize the impact on physicians' ability to practice 
medicine, including: more active approaches to post marketing 
surveillance; final FDA rules on the format and content of package 
inserts; the preservation of physicians' ability to prescribe 
medications for unlabeled uses; and collaboration between the FDA, 
pharmaceutical industry, and physicians to develop better risk 
communication tools.

               FDA DECISIONS AFFECTING PHYSICIAN PRACTICE

Drug Approval

    The FDA's decision to approve a prescription drug or biologic 
product for marketing moves that product from an investigational status 
to an approved product available for widespread use. Approving a 
prescription drug or biologic for marketing is the primary way in which 
the FDA affects physician practice. Over the years, the FDA approval 
process has resulted in access to a wide array of prescription drug and 
biologic products for use by physicians in the care of their patients.
    Fifteen years ago, a primary complaint about the FDA was that the 
drug approval process was too slow. The problem was referred to as a 
``drug lag'' because at the time, the United States stood well behind 
other industrialized countries in getting needed drugs to market. After 
numerous complaints, the government began to focus its attention on 
improving FDA drug review timelines. In 1992, Congress passed the 
Prescription Drug User Fee Act (PDUFA), which authorized the FDA to 
collect user fees from companies that produce drug and biologic 
products. Under PDUFA, these fees were provided in exchange for an FDA 
agreement to meet drug-review performance goals, which emphasized 
timeliness. PDUFA was reauthorized by the Food and Drug Administration 
Modernization Act (FDAMA) of 1997 (PDUFA II) and again by the Public 
Health Security and Bioterrorism Preparedness and Response Act of 2002 
(PDUFA III).
    These acts required the FDA to: (1) speed agency review of New Drug 
Applications (NDAs) and Biologic Licensing Applications (BLAs); (2) 
improve the efficiency of drug development before submission of new 
drug or biologic applications; and (3) further improve the quality and 
efficiency of drug development, review, and risk management for newly 
approved products--all without compromising safety. According to the 
FDA, before PDUFA, the agency approved about 40 percent of the new 
drugs introduced on the world market either first or within 1 year of 
their introduction in another country. After PDUFA and through 2002, 
this percentage had nearly doubled. Additionally, the median total 
review time for new drugs and biologics decreased from approximately 23 
months to 12 months, with even shorter median approval times for drugs 
designated for priority review.
    Concern has been expressed about the number of drugs approved under 
PDUFA that have been withdrawn for safety reasons. However, an FDA 
analysis showed the rate of withdrawal for safety reasons of ``new 
molecular entities'' pre-PDUFA was 2.7 percent, while the rate post-
PDUFA was 2.5 percent. This is not a significant difference. Thus, it 
appears as if the FDA has met its obligations under PDUFA to increase 
the efficiency of the drug review process without compromising the 
safety of approved drug products. Therefore, the AMA and its physician 
members hope that any new efforts to improve drug safety can be 
accomplished without reversing the improvements that have occurred in 
the drug approval process.

Postmarketing Surveillance

    If formal postmarketing studies are not conducted by manufacturers 
or clinical investigators to obtain safety information, observational 
data collected by physicians, other health professionals, and patients 
are the cornerstone for evaluating and characterizing a drug's risk 
profile in actual clinical use. Currently, the FDA maintains an adverse 
event reporting system termed MedWatch, which incorporates both a 
mandatory adverse event reporting system for manufacturers subject to 
the Agency's postmarketing safety reporting regulations, and a 
voluntary, adverse event reporting system for health care 
professionals, consumers, and patients. MedWatch can be an effective 
tool for detecting signals suggesting that a drug may be associated 
with a rare, but serious, adverse event.
    However, the MedWatch program is a passive system and it is limited 
by its reliance on voluntary reporting, which inevitably leads to under 
reporting. Under reporting and uncertainty about the actual extent of 
drug exposure, make it difficult to estimate true rates of occurrence 
of drug-induced adverse events. Because of their observational nature, 
spontaneous reports also are limited in their ability to establish 
causality. Given the limitations of spontaneous reporting systems, 
concerns have been raised about the FDA's ability to detect serious 
adverse events that occur during the postmarketing phase of a drug 
product's life cycle. Thus, as efforts are devoted to improving drug 
safety, attention should be directed toward enhancing postmarketing 
surveillance by using more active approaches. For example, well 
designed pharmacoepidemiologic studies on newly marketed drugs could 
enhance our ability to more accurately determine a drug's adverse event 
profile in a timely manner.
    Recently, the FDA announced its intent to create an independent 
Drug Safety Oversight Board comprised of FDA staff as well as medical 
experts from other Department of Health and Human Services agencies and 
other government departments to oversee the management of important 
drug safety issues. The AMA has not taken a position on this issue.
    In addition, the FDA pledged to ``expand existing communication 
channels and create new ones to ensure that established and emerging 
drug safety data are quickly available to the public (and physicians) 
in an easily accessible form with the intent of enabling patients and 
their health care professional to make better-informed decisions about 
individual treatment options.'' One of these proposed channels would be 
a new ``Drug Watch'' Web page for emerging data and risk information, 
and the AMA applauds these efforts to enhance transparency. However, 
the FDA must provide clear advice when it disseminates emerging or 
preliminary information prior to taking regulatory action.

Product Labeling

    Product labeling decisions are made by the FDA in collaboration 
with the drug sponsor, usually the manufacturer. The product labeling 
includes the materials and language that comprise the product's 
packaging, label and package insert. The package insert is that portion 
of the approved labeling that is directed primarily to physicians to 
inform them about a product's risks and benefits, and to provide 
guidance on the conditions of appropriate use. However, today's package 
insert has become a barrier to effective risk communication, serving 
more as a legal document rather than a resource of useful information 
for practicing physicians. The FDA has recognized this problem and in 
December 2000, it issued a proposed rule to modify the format and 
content of the package insert with the goal of making the information 
more useful and user-friendly to physicians. Their recommendations 
included a more simplified, ``Highlights of Prescribing Information'' 
section within the package insert. The AMA continues to strongly 
support FDA efforts to make package inserts more useful and user-
friendly for physicians and encourages the FDA to issue a final rule to 
that effect.

Unlabeled/Off-Label Uses

    In an effort to strengthen drug safety, the FDA recently announced 
its commitment to sponsoring an Institute of Medicine study on drug 
safety systems with an emphasis on the postmarketing phase, including 
the study of unlabeled (also known as off-label) use. Unlabeled uses 
are defined as the use of a drug product for indications or in patient 
populations, doses, or routes of administration that are not included 
in FDA-approved labeling. Under the Federal Food, Drug, and Cosmetic 
(FD&C) Act, a drug approved by the FDA for marketing may be labeled, 
promoted, and advertised by a manufacturer for only those uses for 
which the drug's safety and efficacy have been established. The 
manufacturer submits data to the FDA demonstrating substantial evidence 
of efficacy and safety for each labeled indication. Even though PDUFA 
has reduced the review time for efficacy supplements (i.e., 
Supplemental New Drug Applications or SNDAs), manufacturers are not 
required to and may choose not to seek FDA approval for all useful 
indications. One major reason for not submitting an SNDA is because the 
expense of regulatory compliance may be greater than the eventual 
revenues expected (e.g., if patent protection for the drug product has 
expired, or if the patient population affected by the new use is very 
small). A sponsor also may not seek FDA approval because of 
difficulties in conducting controlled clinical trials (e.g., for 
ethical reasons, or due to the inability to recruit patients).
    A physician may choose to prescribe a drug for uses, in treatment 
regimens, or in patient populations that are not part of the FDA-
approved labeling. The decision to prescribe a drug for an unlabeled 
use is made by the physician in light of all information available and 
in the best interest of the individual patient. Prescribing for an 
unlabeled use requires the physician to use the same judgment and 
prudence as exercised in medical practice for it to conform to accepted 
professional standards. Given the prevalence of unlabeled uses and the 
fact that in many clinical situations such use may represent the most 
appropriate treatment, the prescribing of FDA-approved drugs for 
unlabeled uses is often necessary for optimal patient care. Therefore, 
the AMA has had longstanding policy:
    ``That a physician may lawfully use an FDA approved drug product 
for an unlabeled indication when such use is based upon sound 
scientific evidence and sound medical opinion (Policy 120.988, AMA 
Policy Compendium).''
    The position of the FDA on physician prescribing of unlabeled uses 
supports that of the AMA. The FDA's published statement that addresses 
the appropriateness and legality of prescribing FDA-approved drugs for 
unlabeled uses includes the following:
    ``The Food, Drug and Cosmetic Act does not limit the manner in 
which a physician may use an approved drug. Once a product has been 
approved for marketing, a physician may prescribe it for uses or in 
treatment regimens or patient populations that are not included in 
approved labeling. Such ``unapproved'' or, more precisely, 
``unlabeled'' uses may be appropriate and rational in certain 
circumstances, and may, in fact, reflect approaches to drug therapy 
that have been extensively reported in medical literature (FDA Drug 
Bulletin. 1982; 12:4--5).''
    It is important to emphasize that the AMA strongly supports the 
SNDA process to add new uses for drugs to FDA-approved labeling. 
However, given the disparity between the actual submission of SNDAs and 
the evolution of evidence-based medical practice, physician prescribing 
for unlabeled uses should not be impeded by any actions taken to 
improve drug safety.

Risk Management of Prescription Drug Products

    In 1999, an FDA Task Force published ``Managing the Risks from 
Medical Product Use.'' Subsequently, in the context of PDUFA III, the 
FDA agreed to provide guidance for the regulated industry on risk 
management activities for drug and biological products. In addition to 
conducting a Part 15 Hearing on risk management in 2002, the FDA issued 
three Concept Papers (``Premarketing Risk Assessment,'' ``Risk 
Management Programs,'' and ``Risk Assessment of Observational Data.'') 
for comment in 2003, and then released three ``draft`` Guidances for 
Industry, (``Premarketing Risk Assessment,'' ``Development and Use of 
Risk Minimization Action Plans [RiskMAPs],'' and ``Good 
Pharmacovigilance Practices and Pharmacoepidemiologic Assessment'') in 
2004. A RiskMAP is a strategic safety program designed to meet specific 
goals and objectives in minimizing known risks of a product while 
preserving its benefits.
    Routine risk minimization measures include use and revision of the 
package insert, combined with postmarketing surveillance. These 
measures should constitute the risk management plan for the vast 
majority of drug and biologic products. The draft guidance on RiskMAPs 
identified several additional tools that could be considered in 
designing risk minimization plans when reliance on the package insert 
as the primary tool may be inadequate. These tools can generally be 
placed under the following three categories:
     Targeted education and outreach (e.g., physician letters; 
training programs for physicians or patients; medication guides);
     Reminder system, processes or forms (e.g., patient 
agreements or acknowledgement forms; certification programs for 
physicians; enrollment of physicians and/or patients in special 
educational programs; specialized systems or records that attest to 
safety measures having been satisfied); and
     Performance-linked access systems (e.g., prescription can 
be ordered only by specially certified physicians; use of compulsory 
fulfillment systems; product dispensing only to patients with evidence 
of lab tests results or other documentation).
    Implications for Physicians. In government's efforts to improve 
drug safety, there may be a desire to use, more routinely, those risk 
minimization tools that extend beyond targeted education and outreach 
to include a more pervasive use of tools associated with reminder 
systems and/or performance-linked access systems. A number of these 
approaches would directly manage or restrict physician prescribing and 
may have unintended consequences.
    These unintended consequences include:
    (1) preventing some patients (who would benefit from higher risk 
drugs) from having access to them because of added burdens on the 
prescriber;
    (2) prescribing of less effective, less studied, and even less safe 
alternative drugs that are not subject to restrictions because they are 
simply much easier to use;
    (3) employing multiple and complex risk management tools that may 
be confusing to both physician and patient and, potentially result in 
unintended medication errors;
    (4) creating administrative burdens for physicians that would 
likely result in the drug not being prescribed at all (unless the 
restricted drug is truly innovative); and
    (5) possibly adversely impacting pharmaceutical company research 
and development in promising areas where restrictive risk management of 
drugs is anticipated.
    Rather than focus on restrictions, the AMA believes that the FDA, 
the pharmaceutical industry, and physician organizations must 
collaborate and identify innovative ways to communicate new risk 
information about a drug or biological product to physicians so they 
will be aware of it, remember it and act on it when prescribing a drug. 
The AMA previously proposed potential ways to improve risk 
communication about drugs to physicians in its comment letters to FDA 
on risk management (see Attachments 1 & 2).
    High level risk minimization tools, such as performance-linked 
access systems and some reminder systems, should be used only as a last 
resort to keep high-risk drug products with unique and important 
benefits on the market. The AMA encourages the FDA and the product 
sponsor to work with relevant physician organizations to assure that 
the minimum number and least intrusive RiskMAP tools are selected to 
achieve the risk minimization objective.

Recommendations

    The AMA is pleased to offer the following recommendations to the 
Committee. We believe these recommendations will both improve drug 
safety and not adversely impact how physicians practice medicine. The 
recommendations are as follows:
    1. Improved postmarketing surveillance for potential adverse events 
can be achieved without slowing down the premarket drug approval 
process. The AMA supports the use of more active approaches to enhance 
postmarketing surveillance.
    2. The FDA should issue a final rule, as soon as possible, 
implementing modifications to the format and content of the package 
insert with the goal of making the information more useful and user-
friendly to physicians.
    3. Physician prescribing for unlabeled uses should not be impeded 
because prescribing of FDA-approved drugs for unlabeled uses is often 
necessary for optimal patient care.
    4. The package insert, combined with effective postmarketing 
surveillance, should constitute the risk management plan for the vast 
majority of drug and biologic products. When this is insufficient to 
ensure an appropriate level of drug safety, then effective risk 
communication to physicians should be the primary means to reduce risks 
of drugs. The AMA urges the FDA and the pharmaceutical industry to 
collaborate with physician organizations to develop better risk 
communication vehicles and approaches. High level risk minimization 
tools, such as performance-linked access systems, should be used only 
as a last resort to keep high-risk products with unique and important 
benefits on the market.
    The AMA once again, commends the Committee for holding today's 
hearing, and we thank the chairman for the opportunity to present our 
views. We look forward to working together on this important issue.

Attachments

    1. AMA Letter 7/6/04 to FDA RE: Draft Guidance for Industry on 
``Development and Use of Risk Minimization Action Plans'' [Docket No. 
2004D-0188]
    2. AMA Letter 4/29/03 to FDA RE: Risk Management [Docket No. 02N-
0528]

    The Chairman. Thank you very much, and I very much 
appreciate your concise testimony. Your entire testimony will 
be a part of the record, and that gives us more time for 
questions too.
    Mr. Keith Carson.
    Mr. Carson. Mr. Chairman, members of the committee, I want 
to thank you for inviting me to come talk to you today.
    I am a chemical engineer, different than a number of your 
panelists in my discipline and background, actually trained as 
a process engineer, and have been working for over 25 years in 
helping companies scale up and produce biological products at 
large scale, including vaccines, antibodies, recombinant 
proteins, and even some of the newer products that are now 
coming out. So I will give you I hope a slightly different 
perspective.
    In my organization, again, I started 10 years ago, we find 
ourselves in a fairly unique position I believe, and at times 
being a neutral party that can work with both FDA and with 
industry since we really have no affiliation with either, and 
that is a position we have been in several times in helping to 
develop what are called reference materials.
    Biological products are very, very difficult to not only 
manufacture but to characterize and understand exactly what 
they are or to compare one product to another. This is where 
reference materials come in. By being able to establish a well-
characterized reference material, then the manufacturers can 
compare their products to this one standard or this reference 
material.
    So the most successful project to date is one where an 
adenovirus was produced and is now currently being stored at 
ATCC. It is being used throughout the world as a reference 
material for validating internal reference materials and their 
assays.
    I cannot tell you enough how difficult some of these 
products are not only to manufacture but to regulate. They are 
all different. Most of the discussions here are about drugs or 
chemically formulated products, but when you get into biologics 
it becomes incredibly more complex. We have a group that are 
now known or being called ``well-characterized biologics,'' 
including the monoclonal antibodies, recombinant antibodies, 
recombinant proteins. But beyond that then you have viral 
products, viral vaccines, bacterial vaccines. And then on top 
of that the cellular products, the cell therapy products are 
incredibly difficult to product and to produce on a consistent 
basis. The lot to lot variability is very, very difficult to 
maintain.
    One reason I am bringing this up is the people that we have 
there at FDA are not only receiving more and more submissions, 
they are having to work on more and more complex products all 
the time, products that they are just now trying to get their 
arms around and figure out the best way to try to regulate 
them, and even figure out what questions to ask.
    The U.S. is considered to be the world leader in this 
technology. We certainly lost our edge in many other 
manufacturing and technology areas, but throughout the world we 
are considered the technology source for biotechnology and for 
biological processing. Our meetings that we hold in Europe and 
in Asia, they still want us to bring 70 percent of our speakers 
from North America. So we have a leadership role in the world 
here and we are highly respected, not only from a processing 
standpoint and technology standpoint, but from a regulatory 
standpoint. These other countries harmonize their regulations 
to a great extent around what has been done here and developed 
here by the FDA.
    All of these products have risks, as Dr. Woodcock 
mentioned, even Tylenol as she mentioned. She did not use the 
generic name, but--or the marketed name, but all of these 
products have risk. The important thing is to try to define the 
patients that can tolerate these products the best and give 
them these products and then identify the patients that would 
have adverse events, and that is where I think we will talk 
about some of these new technologies that might help us move in 
that direction.
    Thank you.
    [The prepared statement of Mr. Carson follows:]

                 Prepared Statement of Keith L. Carson

    I am a chemical engineer with over 25 years experience in the 
biopharmaceutical industry. My training is as a process engineer, and 
my focus has been on the large-scale production of biological products 
including viral vaccines, antibodies, recombinant proteins, viral gene 
vectors, and cellular therapies.
    I also received an MBA in marketing from George Washington 
University in Washington, DC, and lived on Capitol Hill from 1981 to 
1993.
    I helped found the Virginia Biotechnology Association and have 
served as a board member for 7 years, plus one term as the 
association's president. I currently serve as an advisor to the board, 
and provide advice on biotech business development for the State and 
Hampton Road area.
    In 1994, I founded the Williamsburg BioProcessing Foundation, or 
``WilBio,'' in Virginia Beach, Virginia. WilBio is a biotech 
information company that publishes the BioProcessing JournalTM, and 
organizes 12 international conferences on the development and 
production of biological products for human health care.
    Our mission is to help develop safe and effective biological 
products, and our objectives are to make product development less 
costly, provide a trained workforce for the biotech industry, and 
improve communication between FDA and industry, and the academic 
processing centers.
    Since 2000, WilBio has signed several FDA Co-Sponsorship Agreements 
for the development of viral reference materials. Our role has been to 
serve as a facilitator and coordinator for Working Groups, which manage 
the development of these materials and are made up of representatives 
from FDA, industry, and academia. In 2002, the first project resulted 
in the production of a well-characterized adenovirus, which is now used 
by product sponsors throughout the world to validate assays and 
internal reference materials.
    Also through a Co-Sponsorship agreement, this is the 3rd year that 
WilBio has helped organize and manage the Annual FDA Science Forum, 
which is held at the DC Convention Center in May. With approximately 
2,000 attendees from government, industry, and academia; this unique 
meeting offers the best opportunity to learn about the scientific 
interests and activities at FDA, and how science is used to help 
formulate policy and regulate products.
    I have just completed a lengthy analysis of FDA's Critical Path 
Initiative, and have written a review article for our publication. A 
copy of this article has been submitted to the committee, and 
additional copies are available upon request. As you will note, key 
concepts for this initiative include: well-characterized reference 
materials, standardized analytical methods, shared characterization and 
clinical data, and improved regulatory guidelines. I have proposed that 
working groups be formed to tackle these issues in a fashion similar to 
the one taken for the highly successful adenoviral reference material 
project.
    Today, I am here to testify about Drug Safety, and specifically the 
impact that new technologies could have on drug discovery, development, 
and approval. While these technologies appear to offer tremendous 
potential, their use and implementation are still in a very early 
stage; and a number of technical, logistical, and ethical issues must 
be resolved.
    Technological advances in sensors, analytical methods, 
instrumentation, and computing power are happening so quickly that it 
is very difficult to know how they can best be applied, or understand 
what the information they generate actually means. Some breakthroughs 
are occurring, but many years will be required to devise the 
correlations needed to make the data useful.
    To keep up with these technological advances and use them in the 
regulatory process, FDA must be properly funded to staff and equip 
their labs, plus train their personnel. Product reviewers must be 
familiar with these technologies and how to apply them, as well as 
comprehend the sponsor data that is submitted.
    The Agency also needs a permanent Commissioner who can provide 
consistent leadership, policy, and direction.

                           Executive Summary

    I am here to testify about Drug Safety, and specifically the impact 
that new technologies could have on drug discovery, development, and 
approval. While these technologies appear to offer tremendous 
potential, their use and implementation are still in a very early 
stage; and a number of technical, logistical, and ethical issues must 
be resolved. The technologies receiving the most attention include: 
microarray technology, in silico models, datamining, and proteomics.
    Technological advances in sensors, analytical methods, 
instrumentation, and computing power are happening so quickly that it 
is very difficult to know how they can best be applied, or understand 
what the information they generate actually means. Some breakthroughs 
are occurring, but years will be required to devise the correlations 
needed to make most of this data useful.
    To utilize the technologies, patient samples are analyzed for 
differences in genetic markers, or in the application of proteomics, 
differences in expressed proteins that can be linked with genetic 
markers. In particular, researchers are looking for genetic markers 
that could be associated with the uptake of a particular drug, or even 
its metabolism. In other applications, specific genetic markers could 
be related to potential adverse events.
    Personalized medicine is based on matching drug treatments with the 
patient-specific genetic markers that could predict a more favorable 
outcome. With this approach, a higher chance of success could be 
predicted with certain treatments, and adverse events could hopefully 
be avoided.
    However, there are a number of technical, logistical, and ethical 
issues involved with the use of these techniques. First, a patient must 
submit to having his genetic profile determined and analyzed, and many 
individuals are concerned about how this data might be used. Certain 
genes have already been linked to a predisposition for various 
diseases, and the patient could be barred from insurance or certain 
types of employment if this information were accessible.
    In addition, very large computing power will be needed to hold the 
vast amount of data that will be generated. Then when numerous product 
sponsors want to share their data, the hardware requirements become 
even greater, and patient confidentiality could be further compromised. 
And if correlations are developed, the reasons for any relationships 
may not be known for many years.
    To keep up with these technological advances and use them in the 
regulatory process, FDA must be properly funded to staff and equip 
their labs, plus train their personnel. Product reviewers must be 
familiar with these technologies and how to apply them, as well as 
comprehend the sponsor data that is submitted.

    The Chairman. Thank you.
    Dr. Woosley.
    Dr.  Woosley. Chairman Enzi and members of the committee, 
as I was introduced, I am the President of a newly formed 
nonprofit organization created to facilitate innovations in 
drug development. Our goal is to create a forum for drug 
development that does not exist where scientists from the FDA, 
academia and the drug industry can bring innovations into drug 
development, innovations that will give patients the earliest 
possible access to the safest possible medications.
    We believe that such a forum can spawn innovation from 
scientific interchange by serving as a neutral territory, free 
of the regulatory environment that limits most interactions of 
the FDA with the industry.
    I am also Director of the Center for Education and Research 
on Therapeutics called CERT, and interdisciplinary center 
funded by AHRQ. Our CERT is one of 7 centers authorized by 
Congress in FDMA, the Food and Drug Modernization Act, to work 
with the FDA to improve the medical outcomes from therapeutics. 
Drug safety is included in that. In my testimony today I will 
recommend changes in drug development, drug regulation and drug 
surveillance that address a crisis, and I do not use that word 
lightly. I took it out of my testimony probably three times, 
but I want to emphasize that I have thought about this a great 
deal and I do believe there is a crisis and I hope I can 
convince that that word is not hyperbole.
    I think this crisis in drug development translates into a 
crisis in the future in public health. My concern is for the 
long-term viability of the pharmaceutical industry and the 
likelihood that without change patients will have even less 
access to new medical therapies in the future. There are 
numerous signals of this crisis and they are listed in my 
written testimony. The following are a few that I hope will 
convince you that I am not over-using that word.
    The pharmaceutical industry now spends an average of 12 to 
15 years and almost a billion dollars for each drug that it 
successfully develops. Yet with all of that investment of time 
and money, too often it fails to detect serious adverse effects 
of its products until they have been on the market usually for 
years and millions of patients have been exposed to harm. Also 
these unnecessarily long billion dollar development programs 
leave only 2 to 5 years of market time before generic 
competition begins. This short amount of time to make back a 
billion dollars plus large profits and lack of competition from 
other products results in unacceptably high drug prices, prices 
so high that Americans are going to other countries to buy 
their medicines.
    Over the last 10 years what has happened? The 
pharmaceutical industry has increased its investment in R&D by 
250 percent. But what has happened from that 250 percent? The 
number of products, significant new products submitted to the 
FDA for review has fallen by 50 percent. Drug failures during 
testing have doubled, tens of billions of dollars in 
development dollars were lost when 17 drugs were removed from 
the market. Many patients suffered serious injury, and hundreds 
of class action lawsuits now threaten the very existence of 
some of our Nation's most successful companies.
    So while the issue of drug safety must be addressed--and I 
am very pleased you are having these hearings--I encourage you 
to do so in the context of this broader crisis in which the 
wrong action, as you, Chairman Enzi, did mention earlier, could 
really have unintended consequences and could further threaten 
the future viability of the U.S. pharmaceutical industry, and 
therefore, the availability of vital new medicines.
    For example, adding a requirement for Phase IV studies to 
our broken system without other important changes could be 
catastrophic. Any changes to the system must really 
simultaneously correct the flaws that are in our current system 
that leads to 12 to 15 years of development and billions of 
dollars expended, and it must create an efficient, effective 
postmarketing drug surveillance system.
    Other recommendations are in my testimony, but the 
following are my two major points. The FDA needs more options 
for regulatory action. I recommend creating an optional new 
process of stages approval of drugs, and it is described in 
this article which I would like, if you would, to add to the 
record. The details are there, and I would be glad to answer 
questions about that.
    The Chairman. Without objection.
    [The article follows:]

    [Editors Note-Due to the high cost of printing, previously 
published materials submitted by witnesses may be found in the files of 
the committee.]

    Dr.  Woosley. This track would not result in earlier 
marketing, but more close surveillance of all drugs.
    My second recommendation is that the FDA should be 
adequately funded to carry out its mission. The budget should 
include funds for FDA's Critical Path Initiative that was 
mentioned by Dr. Woodcock. That is a very important addition to 
this because it will improve the process to accelerate the 
development of safe drugs.
    Because a major part of drug safety is safe use of drugs, 
FDA and AHRQ should be given adequate funds to create together 
a comprehensive multifaceted safety surveillance system. We 
have to create something new that does not exist, and in my 
written testimony I describe the characteristics of an 
important community-based safety network that is needed for the 
early detection and quantification of drug adverse events, 
something that is not going to be available by data dredging 
and data mining of the currently available databases.
    In summary, drug safety problems are only a symptom of a 
flawed system of drug development. In the Critical Path 
Initiative the FDA has offered to be part of the solution to 
this serious problem. The Critical Path Initiative is a great 
investment because it has the potential to improve drug safety, 
facilitate drug development and to substantially reduce the 
future costs of medications, especially the $720 billion 
estimated for a Medicare prescription drug benefit.
    Thank you.
    [The prepared statement of Dr. Woosley follows:]

          Prepared Statement of Raymond L. Woosley, M.D., Ph.D

    Senator Enzi and members of the committee, I am Dr. Raymond 
Woosley, President of The Critical Path Institute, a non-profit 
organization created to facilitate innovations in drug development. Our 
goal is to create a forum for drug development scientists from the FDA, 
academia and the pharmaceutical industry to evaluate innovations in 
drug development; innovations that will give patients the earliest 
possible access to the safest possible medications. We believe that 
such a forum, i.e. a neutral territory, is essential to bring about 
needed changes in the ways drugs are developed. The Institute is 
working closely with the FDA Commissioner's office and other scientists 
at the FDA, The University of Arizona and SRI International (formerly 
Stanford Research Institute) to develop a formal arrangement for this 
collaboration. I am also the Director of the Center for Education and 
Research on Therapeutics (CERT), a Center at the University of Arizona 
funded by the Agency for Healthcare Research and Quality. CERT is one 
of seven centers in the Nation authorized by Congress to improve the 
medical outcomes from therapeutics. After 30 years of research and 
teaching in medical schools at Vanderbilt University, Georgetown 
University and most recently the University of Arizona, I will leave my 
academic position in July to lead the CERT and The Critical Path 
Institute. This will enable me to focus my efforts on what I believe is 
a crisis in pharmaceutical development.

                      A CRISIS IN DRUG DEVELOPMENT

    This crisis can best be appreciated by looking at recent events and 
the following data:
    1. The pharmaceutical industry spends 12-15 years and almost a 
billion dollars for each drug that is successfully developed. Yet, in 
spite of such an investment in time and dollars, this process still 
fails to detect serious adverse effects of products until they are on 
the market, often for years, and millions of Americans have been 
exposed to potential harm.
    2. Pharmaceuticals have been one of our Nation's most successful 
industries. However, over the last 10 years, the industry increased its 
investment in research and development by 250 percent but the number of 
new products submitted for FDA review has fallen by 50 percent.
    3. The proportion of drugs that fail during development has doubled 
in the last 10 years.
    4. In the last 8 years, over half of the 15 drugs removed from the 
market because of safety concerns were, in fact, safe when used as 
directed in the labeling. Warning labels did not prevent drugs from 
being used in ways that resulted in harm to patients. Tens of billions 
of research and development dollars were wasted and many patients 
suffered serious injury.
    5. Personalized medicines, the promise of human genome research, 
are only rarely being developed because of the high cost of drug 
development relative to the potential market size.
    6. The protracted and costly development of drugs, combined with 
the limited time in the market before generic competition begins, 
results in unacceptably high drug costs and drug re-importation from 
countries that employ price controls.
    7. Skyrocketing estimates for the cost of a Medicare prescription 
drug benefit have prompted consideration of policies and pricing 
negotiations that would limit access to new medicines and threaten 
future research and development of medicines that are needed by 
patients with chronic and debilitating diseases.
    8. In addition to concern for the patients harmed by drugs, there 
is another societal concern. The removal of drugs from the market has 
resulted in hundreds of class action law suits that threaten the very 
existence of some of our Nation's most successful companies.
    So, while the issue of drug safety must be addressed, we must do so 
in the context of a broader crisis in which the wrong action(s) could 
further threaten the future viability of the pharmaceutical industry 
and the availability of vital new medicines. For example, adding a 
requirement for phase IV monitoring to our broken system without other 
changes would be catastrophic. At the same time, the absence of an 
effective drug safety program is one of the major contributors to the 
delays in drug development that adds to high costs and delayed access 
to important new medicines. Two-thirds of the FDA medical reviewers 
recently surveyed expressed concern that the post-marketing 
surveillance system at the FDA was inadequate. I have no doubt that 
this concern must have a negative influence the reviewers' willingness 
to assist the industry in accelerated development of even the most 
important new medicines. Therefore, it is essential and timely that we 
discuss how to improve the development of drugs and assure their safe 
use.

              BASIC ``FACTS OF LIFE'' FOR PHARMACEUTICALS

    A critical first principle is that there is no such thing as a 
``safe drug''. Even the title of these hearings, ``Ensuring drug 
safety'' is an impossible goal. No one can ensure drug safety; we can 
only expect the FDA to identify drugs with an acceptable risk/benefit 
ratio, inform the public, and develop methods to maximize benefit and 
minimize harm. FDA approval will never mean that a drug is ``safe.'' 
Instead it signifies that the available evidence indicates that a drug 
should be ``relatively safe when used as directed.'' All medicines that 
have pharmacologic effects must be assumed to have the potential for 
harm. This is a message that must be better appreciated by the public 
so that they are not surprised when newly marketed drugs are found to 
have adverse effects.

     THE FDA MUST BE GIVEN ADEQUATE NUMBERS OF PEOPLE AND RESOURCES

    Over the last 20 years I have served as a frequent advisor to the 
FDA, usually on issues of drug safety. In this capacity, I learned 
first hand the limitations that exist in the FDA's legal authority as 
well as the FDA's limited resources. It doesn't appear in their budgets 
but information technology and computer allocations have been slashed 
in recent years. The agency that handles some of the most complex and 
vital data in the world relies upon information handling systems that 
were discarded decades ago in most corporations. Only 109 scientists 
monitor the safety data from over 3,000 prescription drugs. Where a 
complete system of drug safety surveillance is needed, the FDA is 
forced to rely on its voluntary reporting system for adverse events.

  THE FDA LACKS ADEQUATE LEGAL AUTHORITY TO EFFECTIVELY REGULATE DRUGS

    Once a drug is marketed, the FDA has no control over the way it is 
used in clinical practice. Relatively safe drugs are often used in 
unsafe ways (e.g. in combination with other interacting drugs or in 
excessive dosage or duration). As is the case in Canada and other 
countries, the FDA should be given the authority to restrict or suspend 
access to drugs when serious questions arise about their safety.
    The FDA also lacks any authority to demand further research on 
marketed drugs. Warning labels, though commonly required by the FDA, 
are known to be ineffective. The only effective tools that the FDA has 
to protect the public are, (1) to keep a drug off the market or, (2) 
once on the market, try to take it off. Because of its limited 
resources, the FDA rarely attempts legal action to remove drugs from 
the market. In almost every case, drugs are voluntarily removed by the 
manufacturer because of pressure from the FDA and not deliberate legal 
action by the FDA.

                           A BETTER TOOL BOX

    The FDA needs more options for action. The FDA could better perform 
its responsibility if it had a broader range of options with which it 
can respond to the ever broadening spectrum of drug information that is 
generated over the pharmaceutical life of a drug.

              A PROPOSAL FOR STAGED APPROVAL OF NEW DRUGS

    Because more information than ever before is being generated about 
the value and risks of new drugs and because time is required for this 
information to be assimilated into the practice of medicine, there is a 
need for earlier approval followed by tightly controlled and more 
gradually increasing usage of new medications. Figure 1 demonstrates an 
alternative path for new drugs that I believe should be considered, 
debated and evaluated. It proposes an earlier approval but more gradual 
growth in use of a prescription drug combined with a comprehensive 
safety assessment in the marketplace. As can be seen, there is an 
earlier and more gradual rise in the number of patients treated in this 
model. This allows time for more complete safety testing and 
assimilation of the drug into the practice of medicine before millions 
are exposed to the drugs.
    The first change suggested is in phase II, which would be expanded 
to include more complete characterization of the drug's dose-response 
relationship in the intended population and sub-populations (e.g. the 
very elderly, those with renal insufficiency, co-morbid conditions, 
etc) and for completion of any necessary targeted drug interaction 
studies. These latter studies should be those based on in vitro 
predictions, e.g. cytochrome P450 or drug transporter interaction 
studies. Phase II should include modern computing techniques such as in 
silico simulation of trials, enrichment using biomarkers, adaptive 
trial design and others suggested in the FDA's Critical Path 
Initiative.
    Market-I: At the end of a more comprehensive and informative phase 
II requiring approximately 4 years, the drug could be approved for 
marketing to a carefully defined population of patients (Market-I in 
figure 1). This is very similar to the way AIDS drugs were developed in 
2-4 years without taking dangerous shortcuts.
    A Safety System: To make the early release of a drug feasible and 
rationale, it will be essential to have an intensive plan for post-
marketing safety assessment and risk management. Academic programs such 
as the Centers for Education and Research on Therapeutics 
(competitively funded by the Agency for Healthcare Research and Quality 
to improve outcomes from medical therapies) can help develop risk 
management programs and conduct outcomes research on large databases 
and registries to confirm the efficacy and safety predicted from phase 
II. As they evaluate the safety of the drug, they can also use similar 
methods to confirm efficacy for initial indications and evaluate the 
potential efficacy of the drug in new indications. In most cases, the 
new drug should initially be given under observed conditions, using a 
system like the yellow card system in the U.K. in which physicians 
report the outcome of therapy in each patient receiving a specific drug 
on a ``yellow card.'' Modern electronic medical record systems make it 
possible to have a system like the U.K.'s General Practitioner's 
Network which tracks the outcome of every patient they treat with a new 
drug. Also, modern electronic registries can detect adverse event 
signals earlier and compare the safety of new and older drugs in a 
class. The CERTs could play a role similar to the pharmacovigilance 
centers in France and monitor drug outcomes in the community. The FDA 
and the pharmaceutical sponsor would have to agree to the use of 
measures to assure that the drug is used as directed in labeling. 
Sponsors could be encouraged to follow the lead of at least one 
innovative company that paid commissions to sales representatives based 
upon how well doctors in their region used the company's drug instead 
of how often the drug was prescribed. Effective risk management 
programs have been successfully developed in the past for drugs with 
the potential for serious toxicity, e.g. clozapine. Because this 
antipsychotic drug can cause fatal bone marrow toxicity in 1 percent 
patients per year of treatment, proof of monitoring of white blood 
count is required before the drug can be dispensed. This has reduced 
the incidence of fatal toxicity by 60 percent.
    A Novel System: In Arizona, The Critical Path Institute and the 
CERT are exploring the feasibility of developing an innovative 
community based safety surveillance system. This system would resemble 
programs in the UK and France in that it would prospectively gather 
data on the outcomes of new medicines and submit it directly to the 
FDA. I believe that such a system must be developed de novo because the 
information needed to address drug safety cannot be gleaned from 
currently available databases. Data mining only works when the 
information you need is somewhere in the system. For the same reason, 
linking databases will never give adequate information. The system must 
be relatively inexpensive, should not interfere with the practice of 
medicine or pharmacy and should be flexible enough to detect suspected 
and unsuspected adverse events of any newly marketed drug. It should be 
able to quantify the rate of adverse event occurrences and even answer 
questions of relative safety by comparing the outcomes with selected 
comparator drugs. It should provide positive feedback to physicians in 
order to prevent future adverse events and improve drug outcomes. If 
the system were effective, even drugs with the potential for serious 
adverse events might be able to remain on the market. For this or any 
program to be successful, the FDA must be given the staff and resources 
to participate in the design and implementation of this system and then 
to monitor the data that are gathered from this system.
    The staged approval model would allow a pharmaceutical company to 
begin marketing its product earlier with a lower total capital 
investment and at a time when much more of the patent life is still in 
effect. It should also make it possible to detect any serious life-
threatening problems earlier before millions have been exposed, 
reducing the frequency of litigation and class action law suits. Also, 
for companies using this track, serious consideration should be given 
to offering indemnification from law suits filed for adverse events in 
return for the sponsor paying for any medical expenses resulting from 
such adverse reactions. This would provide patients and the drug 
sponsor some protection from the potential harm from a new drug.
    Market II: If after a period of careful observation on the market, 
the drug appears safe and effective, it could be given approval for an 
expanded market with fewer or no restrictions to its use (Market II on 
the diagram). Market II is effectively the same as the current market 
in which any licensed physician can prescribe a marketed drug for any 
indication, as long as the physician has evidence that such use has a 
scientific basis.
    Pharmacist Assisted Care (PAC) and OTC: If a marketed prescription 
drug is found to be relatively safe and used for a condition that can 
be self-diagnosed by the patient, it has been customary for it to be 
given non-prescription status, often called ``over-the-counter'' or 
OTC. This may or may not be attractive to the pharmaceutical sponsor 
depending upon many economic and market factors. In some cases the 
sponsor would like to expand the market by having the drug available 
OTC. However in many cases such as the statin drugs for lowering 
cholesterol, some aspect of a drug's use requires medical supervision 
and the FDA is reluctant to approve its use without medical 
supervision. In these cases, there is no alternative now available but 
to deny approval of OTC status. However, in Canada and many other 
countries there is another option. The drug can be given ``behind the 
counter'' status. ``Behind the counter'' means that the drug is 
available in pharmacies for patients who ask for the medication but 
only after consultation with a pharmacist. The pharmacist can perform 
any pre-screening or counseling that could make it more likely that the 
drug will be used safely. This additional step, ``Pharmacist Assisted 
Care'' (PAC), could widen the therapeutic benefit to patients, better 
utilize the important role of pharmacists and minimize the risk of 
therapy. After a period of safe use in the PAC category, a drug may be 
recommended for full OTC status when justified.

      THE NEED FOR INNOVATIONS IN THE PROCESS OF DRUG DEVELOPMENT

    To address the increasing delays and failures in drug development, 
the leadership of the FDA has proposed the ``Critical Path 
Initiative''. This proposal includes efforts to optimize drug 
development and identify new ways to test medicines that will give 
greater assurance of safety and effectiveness than we have now. 
However, this plan will require new partners and new resources for the 
FDA. A recent report from former Secretary of Health and Human Services 
(HHS), Tommy Thompson, pointed out the need for partnering between FDA 
and the NIH, CMS and CDC. But to do this collaborative work, it must 
have resources that are not provided in the current budget. Also, the 
fact that the FDA budget is under the Department of Agriculture and not 
under the full control of the Secretary of HHS is an impediment to 
forming these partnerships.
    The FDA's Critical Path Initiative calls for academic partnerships 
to develop innovations that improve drug development. Forming these 
will also require that the Agency have the staff and resources to 
participate. Just as the Moffet Center in Illinois was established by 
the FDA to address food safety, academic sites can be ``neutral 
ground'' where scientists can share their knowledge and expertise in 
drug development and drug safety without commercial conflicts of 
interest. These public/private partnerships can enable scientists from 
the FDA, academia and industry to develop methods to increase the 
efficiency and informativeness of the drug development process.
    The Critical Path Institute that I lead was created for this 
purpose. Out of serious concern over the relative safety and 
availability of new medicines, the citizens of Tucson and Southern 
Arizona have committed over $9 million to seed the initial work of the 
Institute.
    We believe that an investment that enables the FDA to facilitate 
the development of safe drugs is a good investment, especially at this 
time. Medicare estimates that its prescription drug benefit will cost 
over $720 billion in its first 10 years. That estimate surely assumes 
that new drug costs will continue to rise at its current rapid rate. If 
we are ever going to have less expensive new drugs, we must shorten the 
development time, increase the number of drugs successfully developed 
in order to stimulate competition in the marketplace and improve the 
safety information about these drugs. Increased numbers of drugs for a 
specific disease enable competition to yield lower drug prices. Larger 
numbers of drugs with different actions will better meet the needs of 
our biologically diverse population. ``One size'' does not ``fit all.'' 
Furthermore, adequately studied drugs and the safe use of drugs can 
result in lower healthcare costs and improved health.
    Lowering drugs costs by accelerating the development of safer 
medications is a far better alternative than ``re-importation'' of 
drugs which is just an indirect means to use foreign price controls to 
lower our consumers' drug costs. It would be preferable to give the FDA 
the resources it needs to help improve the process of developing drugs.

                       SUMMARY OF RECOMMENDATIONS

    1. Permanent experienced leadership for the FDA is essential. 
Acting Commissioner Lester Crawford, Acting Deputy Commissioner Janet 
Woodcock and many others working with them are experienced leaders who 
can, with the proper resources, lead positive change at the FDA.
    2. The FDA should be adequately funded to carry out its mission. 
This support should include funds for the Critical Path Initiative and 
an effective safety surveillance system.
    3. The ``user fee'' system should be replaced with a system in 
which industry support is not directly linked to the FDA's work and 
performance.
    4. Determination of drug safety requires an assessment of both risk 
and benefit and should remain the purview and responsibility of the 
FDA, and not of a separate agency as I and others have previously 
suggested. However, the on-going safety evaluations of marketed drugs 
should be made by FDA scientists who were not responsible for the 
original approval recommendation.
    5. The FDA should develop a comprehensive post-marketing assessment 
program for drugs using inter-agency collaborative programs and public-
private partnerships.
    6. Just as the National Transportation Safety Board is responsible 
for investigating all accidents and then makes recommendations for 
safety, there should be an analogous independent body to conduct in-
depth review of the process used to detect serious issues/events in 
drug development and the response to those events by the FDA and 
industry. This body could assess the roles played by consumers, 
healthcare providers, health professions educators, the FDA, the 
pharmaceutical industry, the press and even Congress.
    7. The FDA should be given the authority to release drugs in stages 
that are appropriate for the drugs' level of development and the 
information that is known at the time.
    I am extremely grateful for the opportunity to provide testimony at 
this hearing. I hope you find my perspective of value as you review the 
FDA's drug safety system.




    The Chairman. Thank you very much.
    Dr. Psaty. Thank you very much. Thank you for the 
opportunity to testify. My name is Bruce Psaty and, as you 
indicated, I am a general internist and a cardiovascular 
disease epidemiologist. I have broad interests in public health 
and drug safety.
    The COX-2 inhibitors, a new class of nonsteroidal anti-
inflammatory drugs, were supposed to have fewer serious side 
effects than other available nonsteroidals. After more than 5 
years on the market, an increased risk of heart attack, stroke 
were confirmed for Vioxx, Celebrex and Bextra. Some of the 20 
million Vioxx users and 27 million Celebrex users were injured. 
Indeed, the integrity of the American Drug Safety System has 
been called into question. How can this problem be prevented in 
the future?
    Recommendations:
    No. 1. Give balanced attention to risks and benefits in the 
FDA decisions. To use a drug wisely, patients and physicians 
need to know about both risks and benefits. The design of the 
preapproval trials of the COX-2 inhibitors minimized the 
possibility of uncovering evidence of cardiovascular harm. Some 
of the trials with unfavorable results went unpublished. If 
manufacturers do not address the potential risks, as well as 
benefits, with equal scientific rigor, then, in the interests 
of public health, the FDA must insist that they do so, both 
before and after approval.
    No 2. Require large long-term trials. The limited 
preapproval of the evaluation of the COX-2 inhibitors was not 
adequate. Medicines that will be used by millions of Americans 
for long periods of time are best evaluated in large, long-term 
clinical trials that are started as early as possible in the 
approval process. These trials need not delay the approval. 
This approach used for the lipid-lowering statin drugs has 
benefited patients, physicians and the pharmaceutical industry.
    No. 3. Create an independent Center for Drug Safety within 
the FDA to oversee drugs after marketing. In a commentary, 
entitled, ``Postmarketing surveillance--lack of vigilance, lack 
of trust,'' the editors of JAMA write, ``It is unreasonable to 
expect that the same agency that was responsible for approval 
of drug licensing and labeling would also be committed to 
actively seek to prove itself wrong.'' Other scientists and 
former FDA officials have also advocated an independent Center 
for Drug Safety.
    No. 4. Invest the Center for Drug Safety with new authority 
to regulate drugs that are on the market. Revisions to the 
Vioxx product label took 2 years. The Center for Drug Safety 
must be able to compel manufacturers in a timely fashion to 
revise labels, to conduct patient education or physician 
education, to limit advertising, to complete promised studies, 
to conduct new studies, to suspend sales or to withdraw drugs. 
The Center for Drug Safety should be responsible for 
postmarketing evaluations, including the determinations of the 
risks and benefits for drugs that are on the market.
    No. 5. Provide the Center for Drug Safety with new 
resources. America has become the drug safety testing ground 
for new medications, such as COX-2 inhibitors. According to Dr. 
David Kessler, former head of the FDA, ``PDUFA should have had 
funding on the safety side from the beginning, but industry 
refused to accept that. We wanted it. The industry said, no.''
    Senator Enzi, you referred to the drug lag. And your 
committee did terrific work in solving the drug lag. We now 
have a safety lag, and I would encourage you to address the 
safety lag. In the Office of New Drugs, more than 1,000 
employees work to review a few dozen new drugs each year. In 
the Office of Drug Safety, 109 employees work to evaluate the 
safety of thousands of drugs that are currently on the market. 
The FDA needs an independent center whose mission, vision and 
values are geared toward evaluating and monitoring drugs that 
are on the market.
    No. 6. Strengthen postmarketing safety systems. The FDA's 
MedWatch system, which has been characterized as a 
fundamentally 1950s approach, lacks many of the features of 
high-quality epidemiologic studies, including the ability to 
validate events by standard criteria, the ability to identify 
controls and so forth. The State of this system stands in stark 
contrast to the enormous expansion of the pharmaceutical 
industry during the past several decades. In 2004, the COX-2 
inhibitors had combined sales of more than $6 billion. Several 
new mechanisms to conduct postmarketing surveillance rapidly 
and efficiently merit support.
    In summary, regardless of the speed of approval, toxic 
molecules occasionally make it to market as drugs. To protect 
the health of the public, the most important recommendation is 
an independent Center for Drug Safety with new authority and 
funding. Ongoing congressional oversight of the FDA, of CDER, 
and the new Center for Drug Safety would afford an important 
opportunity for the public discussion of drug safety.
    Thank you.
    [The prepared statement of Dr. Psaty follows:]

           Prepared Statement of Bruce M. Psaty, M.D., Ph.D.

    Mr Chairman and members of the committee, thank you for the 
opportunity to testify. My name is Bruce Psaty. As a practicing general 
internist and cardiovascular-disease epidemiologist, I have broad 
interests in public health and drug safety.
    The COX-2 inhibitors, a new class of non-steroidal anti-
inflammatory drugs, were supposed to have fewer serious side effects 
than other available non-steroidals. After more than 5 years on the 
market, an increased risk of heart attack and stroke has been confirmed 
for Vioxx, Celebrex, and Bextra (1-5). Some of the 20 million Vioxx 
users and 27 million Celebrex users have been injured. Indeed, the 
integrity of the American drug-safety system itself has been 
questioned. How can this problem be prevented in the future?

                            RECOMMENDATIONS

    1. Give balanced attention to risks and benefits in FDA decisions 
(6-8). To use a drug wisely, patients and physicians need to know about 
both risks and benefits. The design of the pre-approval trials of the 
COX-2 inhibitors minimized the possibility of uncovering evidence of 
cardiovascular harm. If manufacturers do not address the potential 
risks and benefits with equal scientific rigor, then in the interests 
of public health, the FDA must insist that they do so, both before and 
after approval.
    2. Require large long-term trials (9). The limited pre-approval 
evaluation of the COX-2 inhibitors was not adequate. Medicines that 
will be used by millions of Americans for long periods of time are best 
evaluated in large long-term clinical trials that are started as early 
as possible in the drug approval process. These trials need not delay 
approval. This approach, used for the lipid-lowering statin drugs, has 
benefited patients, physicians and the pharmaceutical industry.
    3. Create an independent Center for Drug Safety within the FDA to 
oversee drugs after marketing (10-14). In a commentary entitled, 
``Postmarketing surveillance--lack of vigilance, lack of trust,'' the 
editors of JAMA, write: ``It is unreasonable to expect that the same 
agency that was responsible for approval of drug licensing and labeling 
would also be committed to actively seek evidence to prove itself 
wrong.'' Other scientists and former FDA officials have also advocated 
a truly independent Center for Drug Safety.
    4. Invest the Center for Drug Safety with new authority to regulate 
drugs that are on the market (4,15). Revisions to the Vioxx product 
label in 2002 took more than 2 years to negotiate. The CDS must be able 
to compel manufacturers, in a timely fashion, to revise product labels, 
to conduct patient or physician education, to limit advertising, to 
complete promised studies, to conduct new studies, to suspend sales and 
to withdraw drugs. The Center for Drug Safety should be responsible for 
post-marketing evaluations, including determinations of the balance of 
risks and benefits for drugs that are on the market.
    5. Provide the Center for Drug Safety with new resources 
(14,16,17). America has become the drug-safety testing ground for new 
medications, such as the COX-2 inhibitors. According to Dr. David 
Kessler, former head of the FDA, ``PDUFA should have had funding on the 
safety side from the beginning, but the industry refused to accept that 
. . . . We wanted it. The industry said no.'' Since 1992, FDA resources 
for drug safety have dwindled. In the Office of New Drugs, more than 
1,000 employees work to review a few dozen new drugs per year. In the 
Office of Drug Safety, 109 employees work to evaluate the safety of 
thousands of drugs currently on the market.
    6. Strengthen US post-marketing safety systems (18-21). The FDA's 
MedWatch system, which has been characterized as ``fundamentally a 
1950s-era approach,'' lacks many of the features of high-quality 
epidemiologic studies, including validation of events by standard 
criteria, complete ascertainment of cases, population-based controls, 
comparable assessment of drug use and risk factors, and so forth. The 
state of this system stands in stark contrast to the enormous expansion 
of the pharmaceutical industry during the past several decades. In 
2004, the three COX-2 inhibitors alone had combined sales more than $6 
billion dollars. Several new mechanisms to conduct post-marketing 
surveillance rapidly and efficiently merit support.
    Regardless of the speed of approval, toxic molecules occasionally 
make it to market as drugs. To protect the health of the public, the 
most important recommendation is an independent Center for Drug Safety 
with new authority and funding. On-going congressional oversight of the 
FDA, CDER, and the new Center for Drug Safety would afford an important 
forum for the public discussion of drug safety. Thank you.

                       SUPPLEMENTARY INFORMATION

    Post-marketing surveillance. When drugs are approved as ``safe and 
effective'' for their intended use, the known benefits appear to 
outweigh the known risks (20). At the time of regulatory approval for 
most drugs, a number of issues remain unknown--the occurrence of rare 
but serious adverse drug events, drug interactions, late events during 
treatment or after the discontinuation of treatment, effects in 
pregnancy or differential effects in subgroups that may be defined by 
age, sex, race, or other factors. In contrast to the highly structured 
pre-marketing evaluation, post-marketing surveillance has little 
structure. According to Gale, ``the regulatory process creates an 
evidence-free zone at the time of launch of new drugs'' (20). 
Pharmaceutical companies often promise post-marketing clinical trials 
as a condition of approval. In practice, however, more than half of 
these promised studies, according to an FDA report, have not been 
started (15). The FDA lacks authority to insist that these promised 
studies be completed or to compel new post-marketing studies. The FDA 
post-marketing regulations require only that pharmaceutical companies 
collect, review and report to the FDA all suspected adverse drug 
reactions (ADRs) thought to be associated with the drug (22,23). While 
both companies and the FDA can analyze the ADR data and recommend 
actions such as label changes, additional warnings, or new studies, the 
FDA regulations largely focus on reporting procedures and thus leave 
unclear who is required to initiate these actions.
    U.S. post-marketing surveillance system. MedWatch, the FDA safety 
information system and adverse event reporting program, encourages 
physicians to report ADRs on a voluntary basis (18). Although the FDA 
received 286,755 ADR reports in 2001 (24), these data have major well-
known limitations. The MedWatch ADR data are suitable only to identify 
rare serious adverse drug events that occur early in treatment and that 
are unrelated to the indication of the drug. For example, the lipid-
lowering statin drug, Baycol (cerivastatin), was withdrawn from the 
market in 2001 because it was associated with high rates of 
rhabdomyolysis, a breakdown of muscle cells that causes pain, kidney 
failure and sometimes death (19,21,25). The MedWatch ADR data lack many 
of the features of high-quality epidemiologic studies, including 
validation of events by standard criteria, complete ascertainment of 
cases, population-based controls, comparable assessment of drug use and 
risk factors, and so forth. It would not have been possible to use the 
MedWatch system to detect reliably, for instance, the increased risk of 
cardiovascular events associated with the COX-2 inhibitors. One recent 
commentator characterized the MedWatch system as ``fundamentally a 
1950s-era approach'' (26).
    Growth of drug sales. The lack of development in post-marketing 
surveillance systems stands in stark contrast to the enormous expansion 
of the pharmaceutical industry during the past several decades. 
Although the costs of drug development are high, spending on 
prescription drugs between 1997 and 2001 increased by about 18 percent 
per year; and in 2001, the total prescription drug expenditures in the 
U.S. reached $154.5 billion dollars (27). In 2004, despite the 
withdrawal of Vioxx in September, the three COX-2 inhibitors alone--
Vioxx, Celebrex, and Bextra--had combined sales more than $6 billion 
dollars, or an average of about $16 million per day (28). The recent 
growth of the pharmaceutical industry has outstripped the safety 
systems that were developed when the industry was young.
    Epidemiologic studies and new opportunities. In the past, data 
sources used to conduct high-quality observational studies of the risks 
and benefits of drugs have included existing cohort studies (29), 
administrative data from health maintenance organizations (30,31), 
Medicaid data (32,33), Medicare data linked to cancer registries (34), 
and international databases with drug data (35,36). In addition to 
AHRQ-funded Centers for Education and Research in Therapeutics (26), 
the FDA has had cooperative agreements with several institutions to 
investigate drug safety, but the available funds have diminished in 
recent years. Several new opportunities are on the horizon. First, data 
from new Medicare drug benefit can be linked with hospital and 
ambulatory care data to create a new resource for the study of drugs in 
older adults. With appropriate protections for privacy, these data 
should be available to the FDA and independent scientists interested in 
drug safety. Secondly, as part of the NIH Roadmap Project, the HMO-
Research Network--Coordinated Clinical Studies Network will create an 
infrastructure for conducting studies on substantial numbers of the 
U.S. population, and the movement toward an EPICcare based electronic 
record among the network members should soon provide the opportunity to 
conduct post-marketing surveillance rapidly and efficiently.
    Post-marketing clinical trials. The pharmaceutical industry 
supports a number of post-marketing clinical trials, often for new 
indications. The cardiovascular harm associated with the COX-2 
inhibitors became apparent in studies that were conducted for new 
indications such as the prevention of non-cancerous tumors in the colon 
(1-3). For the lipid-lowering statin drugs, for instance, the large 
long-term clinical trials have provided robust evidence about their 
health benefits in preventing cardiovascular complications of high 
levels of cholesterol (37-40). On the basis of this evidence, the 
indications for the statin drugs have expanded, statin drug sales have 
increased, and the health of the public has improved. Rapid publication 
and widespread dissemination of favorable findings is standard 
practice.
    Failure to publish trials with unfavorable results. Unfavorable 
results tend not to get published. In the manufacturer's trial of 1.6 
mg of Baycol, about 12 percent of patients developed signs and symptoms 
compatible with rhabdomyolysis (25). The high rate of adverse effects, 
with a dose that was only twice as high as the approved dose of 0.8 mg, 
``led to a consensus by the [company's communications] committee not to 
publish the results of this study'' (25). Similarly, in 2000, Pfizer 
completed a randomized trial of celecoxib in Alzheimer's patients, but 
never published the unfavorable cardiovascular results and only made 
them publicly available in January 2005 (41). The results of this 
Alzheimer's study were not submitted to the FDA until June 2001, 
several months after a safety review that established labeling for 
Celebrex. Human subjects participate in studies to contribute to 
science and public health. Failure to publish findings not only 
violates their trust, but it also misrepresents the evidence about 
risks and benefits for patients and physicians. Federal action to 
assure that all clinical trials are registered and reported in a timely 
fashion is important.
    Prescription Drug Users Fee Act (PDUFA) of 1992. In the late 1980s 
and early 1990s, the pressure from companies and patients alike was not 
for additional safety evaluations, but for shorter approval times (42). 
In response to the criticism that the FDA approval times were too long, 
Congress introduced user fees in 1992. Pharmaceutical companies seeking 
drug approvals paid fees that enabled the FDA to hire additional staff, 
and the FDA was expected to meet new requirements for the timelines of 
new-drug approvals (16). According to Dr. David Kessler, head of the 
FDA from 1990 to 1997, ``PDUFA should have had funding on the safety 
side from the beginning, but the industry refused to accept that . . . 
. We wanted it. The industry said no'' (17). The 1992 user fee act and 
its reauthorization in 1997 prohibited the agency from spending users 
fees ``on post-marketing surveillance or other drug-safety programs'' 
(14). The reauthorization in 2003 included some provisions for safety. 
During the period 1992 to 2003, this approach--more and faster new 
approvals without additional funds for safety surveillance--relied 
increasingly on the honesty, trustworthiness, and integrity of the 
pharmaceutical industry in the conduct of its own post-marketing safety 
evaluations.
    PDUFA, review times, and funding for safety. The PDUFA act in 1992 
and its reauthorizations in 1997 and 2003 reduced the time required for 
review of a new drug application by the FDA from 33 months in 1992 down 
to about 13 or 14 months in 2001 (17). As a result, the proportion of 
new molecular entities that are first introduced in the U.S. has 
increased from 2 to 3 percent in the early 1980s up to 60 percent in 
1998 (43). New medicines are now indeed available to Americans more 
quickly. At the same time, U.S. patients also became the first to 
receive new medications, some of which, such as COX-2 inhibitors, are 
subsequently discovered to have serious adverse effects. The Office of 
Inspector General 2003 Report on the FDA's Review Process for New Drug 
Applications has assessed the impact of the new review process at the 
FDA (16). Funding for safety has also been affected. In 1992, 53 
percent of the budget of the FDA Center for Drug Evaluation went to new 
drug reviews, and the rest went to surveillance, laboratories and other 
safety efforts. In 2003, 79 percent went to new drug reviews. Resources 
available for safety have dwindled (44). Drug recalls following 
approval increased from 1.56 percent in 1993-1996 up to 5.35 percent 
for 1997-2001 (10).
    Calls for an independent Center or Office of Drug Safety. In a 
recent commentary, the JAMA editors advocate an independent center or 
office of drug safety: ``It is unreasonable to expect the same agency 
that was responsible for approval of drug licensing and labeling would 
also be committed to actively seek evidence to prove itself wrong (ie, 
that the decision to approve the product was subsequently shown to be 
incorrect)'' (10). Other recent commentaries in JAMA (13) and the New 
England Journal of Medicine have recommended the creation of an 
independent drug-safety board ``to monitor drug safety, investigate 
reports of drug toxicity, and recommend actions to minimize the risks 
of drug therapy'' (14). The new Advisory Board on Drug Safety announced 
by Michael O. Leavitt, secretary of Health and Human Services on 
February 15 is not adequate. According to Dr. William Schultz, FDA 
deputy commissioner for policy from 1994 to 1998, ``The FDA should 
separate the monitoring of drugs after they have been approved from the 
drug review function'' (12).
    Need for additional authority in the Center for Drug Safety. In 
March 2000, Merck was aware that compared with naproxen, Vioxx 
increased the risk of heart attacks (45). In February 2001, an FDA 
Advisory Committee reviewed the safety data, but revisions to the 
``Precautions'' section of the VIOXX product label were delayed until 
April 2002. The public health rationale for the 2 year delay in 
revising the product label remains unclear. Although the FDA can call 
Advisory Committee meetings or issue press releases, talk papers, 
guidances, and requests to manufacturers, these powers are not adequate 
to regulate drugs that are on the market. For an approved drug, the FDA 
currently engages in protracted negotiations with manufacturers rather 
than mandating manufacturers: to change a product label, to conduct 
patient or physician education, to limit advertising to patients or 
physicians, to modify approved indications, to restrict use to selected 
patients, to complete post-marketing studies agreed upon at the time of 
approval, to conduct additional post-marketing studies or trials, to 
suspend marketing or withdraw a drug. At least one pharmaceutical 
executive has advocated providing the FDA with additional authority to 
mandate studies after drugs are approved (46). Moreover, provisional 
approval for the first 2 or 3 years would provide an opportunity to re-
review the balance of risk and benefit.
    Elements required to protect the health of the public. The failure 
to pose a question often precludes the possibility of obtaining an 
answer. Pharmaceutical companies generally lack enthusiasm for 
aggressively pursuing questions about the safety of their drugs. In 
science, only those questions that are investigated with well-designed 
studies have a decent chance of producing a solid answer. If the 
pharmaceutical industry does not pose critical questions about drug 
safety, the FDA must do so in an effort to protect the health of the 
public. Key elements related to the study of drug safety include: (1) 
the generation of ideas about a drug's risks as well as its benefits; 
(2) a sustained effort to investigate or document risks as well as 
benefits; (3) the availability of high-quality surveillance systems or 
the conduct of specifically designed studies to assess risks as well as 
benefits; and (4) the willingness to publish findings about risks as 
well as benefits. If manufacturers do not provide support for a 
vigorous and balanced scientific evaluation of safety signals for drugs 
that are already on the market, the Center for Drug Safety must do so 
to protect the health of the public.
    Activities of the Center for Drug Safety. At the time of approval 
for each new drug and on the basis of information available in the NDA 
and other studies, the Center for Drug Safety needs to identify a set 
of studies required to address the key unanswered questions, 
particularly the pursuit of potential safety ``signals'' or ``plausible 
biologic hypotheses'' on behalf of the health of the public. Depending 
on the drug, the indication and the known safety profile, the studies 
may include Phase IV trials, epidemiologic studies, pharmacokinetic-
pharmacodynamic studies, close surveillance of ADR reports, or a 
combination of several approaches. Specific post-marketing trials or 
studies should be designed, conducted and completed in a timely 
fashion. The Center for Drug Safety should be responsible for assessing 
the balance of risk and benefit of drugs that are on the market.

References

    1. Nussmeier NA, Whelton AA, Brown MT, et al. Complications of the 
COX-2 inhibitors parecoxib and valdecoxib after cardiac surgery. N Engl 
J Med 2005; 352; Epub.
    2. Solomon SD, McMurray JJV, Pfeffer MA, et al. for the Adenoma 
Prevention with Celecoxib (APC) Study Investigators. Cardiovascular 
risk associated with celecoxib in a clinical trial for colorectal 
adenoma prevention. N Engl J Med 2005; 352: Epub.
    3. Bresalier RS, Sandler RS, Quan H, et al. for the APPROVE Trial 
Investigators. Cardiovascular events associated with rofecoxib in a 3-
year randomized colorectal adenoma chemoprevention trial. N Engl J Med 
2005; 352: Epub.
    4. Psaty BM, Furberg CD. COX-2 inhibitors--lessons in drug safety. 
N Engl J Med 2005; Epub February 15, 2005.
    5. Drazen JM. COX-2 inhibitors: a lesson in unexpected problems. N 
Engl J Med 2005; Epub Feb 15, 2005.
    6. McAdam BF, Catella-Lawson F, Mardini IA, Kapoor S, Lawson JA, 
FitzGerald GA. Systemic biosynthesis of prostacyclin by cyclooxygenase 
(COX)-2: the human pharmacology of a selective inhibitor of COX-2. Proc 
Natl Acad Sci 1999;96:272-277.
    7. Villalba ML. FDA Medical Officer Review of VIOXX (rofecoxib), 
NDA 21-042 (capsules) and NDA 21-052 (oral solution). Http://
www.fda.gov/cder/foi/nda/index.htm.
    8. Writing Group for the Women's Health Initative. Risks and 
benefits of estrogen plus progestin in healthy postmenopausal women: 
principal results from the Women's Health Initiative randomized 
controlled trial. JAMA 2002;288:321-333.
    9. Psaty BM, Weiss NS, Furberg CD, et al. Surrogate endpoints, 
health outcomes, and the drug approval process for the treatment of 
risk factors for cardiovascular disease. JAMA 1999;282:786-790.
    10. Fontanarosa PB, Rennie D, DeAngelis CD. Postmarketing 
surveillance--lack of vigilance, lack of trust [editorial]. JAMA 
2004;292:2647-2650.
    11. Harris G. FDA to create advisory board on drug safety. New York 
Times, February 16, 2005.
    12. Schultz WB. How to improve drug safety. Washington Post 
(washingtonpost.com), December 2, 2004, page A35.
    13. Moore TJ, Psaty BM, Furberg CD. Time to act on drug safety. 
JAMA 1998;279:1571-1573.
    14. Wood AJJ, Stein CM, Woosley R. Making medicines safer--the need 
for an independent drug safety board. N Engl J Med 1998;339:1851-1854.
    15. Steinman M. FDA reports more than half of post-approval 
prescription drug studies not begun. California Healthline, May 23, 
2003.
    16. Department of Health and Human Services Office of Inspector 
General. FDA's review process for new drug applications: a management 
review. Department of Health and Human Services; OEI-01-01-00590; March 
2003.
    17. Graham J, James F. Flaws in drug agency put consumer at risk. 
The Chicago Tribune, February 20, 2005.
    18. Kessler DA. Introducing MEDWatch: a new approach to reporting 
medication and device adverse effects and product problems. JAMA 
1993;269:2765-2767.
    19. Graham DJ, Staffa JA, Shatin D, Andrade SE, Schech SD, La 
Grenade L, et al. Incidence of hospitalized rhabdomyolysis in patients 
treated with lipid-lowering drugs. JAMA 2004;292:2585-2590.
    20. Gale EA. Lessons from the glitazones: a story of drug 
development. Lancet 2001;357:1870-1875.
    21. Pauley III WH. Memorandum and order in re: Bayer AG securities 
litigation. Southern District of New York, No 03 Civ 1546, Sept 30, 
2004: http://www.nysd.uscourts.gov/courtweb/public.htm.
    22. Food and Drug Administration, Department of Health and Human 
Services. Expedited safety reporting requirements for human drug and 
biologic products, 21 CRF Parts 20, 310, 312, 314 and 600, final rule. 
Federal Register Oct 7, 1997;62:52237-52253.
    23. Food and Drug Administration, Department of Health and Human 
Services. Safety reporting requirements for human drug and biological 
products, 21 CFR Parts 310, 312, 314, 600, 601, and 606, proposed rule. 
Federal Register March 14, 2003;68:12406-12497.
    24. Food and Drug Administration. CDER 2001 report to the nation: 
improving public health through human drugs. Rockville, Maryland, 2002. 
http://www.fda.gov/cder/reports/rtn/2001/rtn2001.pdf.
    25. Psaty BM, Furberg CD, Ray WA, Weiss NS. Potential for conflict 
of interest in the evaluation of suspected adverse drug reactions: use 
of cerivastatin and risk of rhabdomyolysis. JAMA 2004;292:2622-2631.
    26. Strom BL. Potential for conflict of interest in the evaluation 
of suspected adverse drug reactions: a counterpoint. JAMA 
2004;292:2643-2646.
    27. The National Institute for Health Care Management Research and 
Education Foundation. Prescrption drug expenditures in 2001: another 
year of escalating costs. NIHCM Foundation, Washington, DC, April 2002; 
http://www.nihcm.org/index2.html.
    28. Berenson A, Feder B. A reminder that no drug is risk-free. New 
York Times, February 19, 2005.
    29. Psaty BM, Lee M, Savage PJ, Rutan GH, German PS, Lyles M. 
Assessing the use of medications in the elderly: method and initial 
results in the Cardiovascular Health Study. J Clin Epidemiol 
1992;45:683-692.
    30. Psaty BM, Koepsell TD, Siscovick D, Wagner EH, Wahl P, LoGerfo 
JP, Inui TS. An approach to several problems in the use of large 
databases for population-based case-control studies of the therapeutic 
efficacy and safety of anti-hypertensive medicines. Stat Med 
1991;10:653-662.
    31. Graham DJ, Campem D, Hui R, et al. Risk of acute myocardial 
infarction and sudden cardiac death in patients treated with cyclo-
oxygenase 2 selective and non-selective non-steroidal anti-inflammatory 
drugs: nested case-control study. Lancet 2005; Epub Jan 25, 2005.
    32. Ray WA, Griffin MR. Use of medicaid data for 
pharmacoepidemiology. Am J Epidemiol 1989;129:837-84.
    33. Ray WA, Stein CM, Daugherty JR, Hall K, Arbogast PG, Griffin 
MR. COX-2 selective non-steroidal anti-flammatory drugs and risk of 
serious coronary heart disease. Lancet 2002;360:1071-1073.
    34. Shahinian VB, Kuo YF, Freeman JL, Goodwin JS. Risk of fracture 
after androgen deprivation for prostate cancer. N Engl J Med 
2005;352:154-164.
    35. Jick H, Jick SS, Derby LE. Validation of information recorded 
on general practitioner based computerized data resources in the United 
Kingdom. BMJ 1991;302:766-8.
    36. Herings RMC. Pharmo: a record linkage system for postmarking 
surveillance of prescription drugs in The Netherlands [Thesis, 
University of Utrecht, The Netherlands]. Elinkwijk BV, Utrecht: 1993.
    37. Scandinavian Simvastatin Survival Study Group. Randomised trial 
of cholesterol lowering in 4444 patients with coronary heart disease: 
the Scandinavian Simvastatin Survival Study (4S). Lancet 1994;344:1383-
1389.
    38. Shepherd J, Cobbe SM, Isles CG, et al for the West of Scotland 
Coronary Prevention Study Group. Prevention of coronary heart disease 
with pravastatin in men with hypercholesterolemia. N Engl J Med 
1995;333:1301-1307.
    39. Downs JR, Clearfield M, Weis S, et al. Primary prevention of 
acute coronary events with lovastatin in men and women with average 
cholesterol levels: results of AFCAPS/TexCAPS. JAMA 1998;279:1615-1622.
    40. Sever PS, Dahlof B, Poulter NR, et al. Prevention of coronary 
and stroke events with atorvastatin in hypertensive patients who have 
average or lower-than-average cholesterol concentrations in the Anglo-
Scandinavian Cardiac Outcomes Trial--Lipid Lowering Arm (ASCOT-LLA). 
Lancet 2003;361:1149-1158.
    41. Pfizer Inc. A double-blind randomized placebo-controlled 
comparative study of celecoxib (SC-58635) for the inhibition of 
progression of Alzheimer's Disease, protocol IQ5-97-02-001. Http://
www.clinicalstudyresults.org/documents/company-study	76	0.pdf; last 
accessed Jan 29, 2005.
    42. Meadows M. Why drugs get pulled off the market. FDA Consumer 
2002; Jan/Feb, 11-17.
    43. Friedman MA, Woodcock J, Lumpkin MM, Shuren JE, Hass AE, 
Thompson LJ. The safety of newly approved medicines: do recent market 
removals mean there is a problem? JAMA 1999;281:1728-1734.
    44. Harris G. Regulation redefined: At FDA, strong drug ties and 
less monitoring. New York Times, December 6, 2004.
    45. Bombardier C, Laine L, Reicin A, et al for the VIGOR Study 
Group. Comparison of upper gastrointestinal toxicity of Rofecoxib and 
Naproxen in patients with rheumatoid arthritis. N Engl J Med 
2000;343:1520-1528.
    46. Anonymous. FDA needs more teeth, says GSK [GlaxoSmithKline] 
head Jean-Pierre Garnier. The Daily Telegraph [United Kingdom 
newspaper], December 5, 2004.

    The Chairman. Thank you very much. I appreciate the 
testimony of all of the witnesses. I have a huge number of 
questions. I know I am not going to be able to get through all 
of them. But as I mentioned before, we will be submitting some 
in writing, particularly the more detailed ones.
    I will begin with Dr. Wilson. One of our witnesses on 
Tuesday said that patients are going to doctors and getting 
drugs they should not because of direct-to-consumer 
advertising. How would you respond to that? Could you comment 
on the role of direct-to-consumer advertising in educating 
patients and in the doctors' prescribing decisions?
    Dr. Wilson. Yes, thank you, Mr. Chairman. The AMA supports 
patients' increased access to drug information and, clearly, 
direct-to-consumer advertising provides that. But let me just 
say, as a practicing physician, I would echo the remarks that 
you heard earlier this morning that it, also, exerts an 
enormous pressure on the physician. And so our concern is the 
impact direct-to-consumer advertising has on the physician-
patient relationship. It, obviously, is advertising that is 
designed to sell a product, and we would suggest that it does 
not present in the ads the kind of accurate and objective 
information that patients need.
    In 1993, the AMA published some guidelines or agreed to 
some guidelines in consultation with the FDA which deal with 
direct-to-consumer advertising. So we would urge the 
pharmaceutical industry to use those guidelines.
    The Chairman. Thank you.
    Mr. Carson, you mentioned data mining in your testimony. 
What are the potentials and limitations for data mining to 
identify safety signals early? What are the hurdles to 
developing a comprehensive database for use with validated 
mining tools? Does the FDA have the computational resources it 
would need to truly be able to use those databases?
    Mr. Carson. Probably not, as far as the last part of your 
question. The amount of data that is involved is phenomenal and 
is rapidly growing. One of the biggest problems is in having 
the data available or making it available, getting the 
companies to make the data available unless they are required 
to. There are some aspects where it is voluntary, some where it 
is required, some where it is just not asked for at all or it 
is not involved.
    I think that industry needs to see the advantage of making 
the data available and sharing it amongst themselves. And then 
there is a lot of data available at the FDA that other 
companies do not see, but a lot of it is proprietary. They have 
to keep it confidential for the companies that have supplied 
the data. So there would have to be agreement that this data 
could be made available.
    In addition to that, of course, is the confidentiality 
issues surrounding the patient. There, of course, is a lot of 
concern, and you see reports of this of people will be, many 
people will be very reluctant to have these genetic screenings 
done and to have that information in a database where they do 
not know if it is controllable or not. Of course, there was the 
story the other day in the paper about one of the major banks 
losing a lot of information on their customers. Well, I think 
that in general the public does not trust these large databanks 
or people that control this data.
    So I think getting the data is probably the biggest 
problem. There would be ways to provide the computing 
capability of the cooperation were there. That is probably the 
easiest part, but does FDA have the resources right now? I 
would probably think they do not. They probably need additional 
funding for that.
    The Chairman. Thank you.
    Dr. Woosley, you indicate in your written testimony that 
you no longer support the idea of a separate drug safety agency 
and that you now believe that the determination of drug safety 
requires an assessment of both risk and benefit and should 
remain in FDA. What made you change your mind?
    Dr. Woosley. Further information, and the kind of points 
that were made by Dr. Woodcock this morning were compelling. I 
think she, clearly, described the need for an ongoing 
evaluation of the benefits and the safety. And that really does 
need some memory, some corporate memory, of what the drug has 
done.
    But the other points she made is, also, very important, and 
that is what the other drugs in that category might have done 
because often the safety assessment of one drug really needs to 
include the safety of all the other drugs around it and in that 
class. Because predicting drug safety is often knowing what the 
others did.
    In fact, I went back and read our New England Journal 
article a few years ago that we wrote, and it really, what we 
were calling for was an oversight, like the NTSB, not that 
actually does the drug safety analysis, but looks at drug use, 
in general, and are the systems that we have in place 
appropriate? Are drugs being developed, regulated, and are they 
being used in the appropriate way, and are there places outside 
the FDA that need to get involved?
    For example, I have taught medical students for 30 years, 
and I should be taking some blame when doctors misprescribe. 
Only 15 percent of the medical schools in this country have 
required courses in clinical pharmacology and therapeutics. We, 
as medical educators, really need to be part of solving this 
problem, also. AHRQ needs to be there, to be looking at the 
drug use. They have the drug safety programs, the medical 
errors programs, but it is not just the FDA that is involved. 
They cannot control drug use.
    So, again, to answer your question more specifically,i do 
not support a separate, independent safety agency outside of 
the FDA. I think the FDA needs to be involved in the changes 
that they are making to bring the decisionmaking away from the 
people who approved it into an environment where there will be 
greater input, but still have the efficacy there is going to 
make me very happy with the changes.
    The Chairman. Thank you. And I think probably local 
pharmacists appreciate your comments because it emphasizes the 
role that they play in the whole process, too.
    My time is expired.
    Senator Isakson.
    Senator Isakson. Thank you, Mr. Chairman.
    Following up on that question, Dr. Psaty, I take it you 
disagree with that.
    Dr. Psaty. No. Actually, I have not advocated necessarily 
that the independent office be outside the FDA. I think, in 
many instances, the FDA has done terrific work. I would 
question some of the decisions they have made. But the Office 
of New Drugs currently dominates CDER, the drug review section, 
and the current structure at the FDA is just what the industry 
desires--a powerful engine to approve drugs and a weak effort 
to investigate safety in the postmarketing setting.
    What the American public needs and deserves, in addition to 
the rapid approval of drugs, is a center whose mission is 
devoted to postmarketing safety evaluations.
    Senator Isakson. So your reference to independence was 
independence of the original testing not independence of the 
Agency.
    Dr. Psaty. Yes, it was. And I said in my testimony an 
independent center within the FDA.
    Senator Isakson. I just wanted to make sure I understood 
that.
    Thank you.
    Dr. Psaty. Yes, sir.
    Senator Isakson. Dr. Wilson, is the insert that you have 
difficulty reading the same one I am supposed to read in my 
medicine bottle?
    Dr. Wilson. Senator, I would assume that it is very 
similar. [Laughter.] As a matter of fact, in that regard, I 
frequently will move from the part I am supposed to read to the 
part the patient is supposed to read, and I find them equally 
unclear.
    Senator Isakson. The only thing harder to read than the 
insert in a medicine bottle is the doctor's signature on a 
prescription. [Laughter.]
    Dr. Wilson. And the better the doctor, the worse their 
writing is.
    Senator Isakson. And I would acknowledge that I am sure 
legal liability plays a large role in what has to go in that 
information, and I understand that, having been a businessman 
for 33 years. And I do not like for the Government to get in 
the business of starting to say how big letters ought to be, 
and how short words ought to be and things like that, but I 
would say that, from a standpoint of a patient's, that 
information regards to warnings to a patient, which certainly 
is what we ought to think about, giving what we are talking 
about today, and that information a doctor needs with regard to 
what they need to know vis-a-vis warnings the patient might 
get, if it was a little larger or at least at the beginning of 
that insert, it would sure be a help to me, and I take it, it 
would be a help to you. I do not know if that is AMA's 
recommendation, but that was just an independent advertisement.
    Dr. Woosley, you read, and I am not sure I heard you say 
it, but you read in your recommendations I read that the user 
fee system should be replaced with a system in which industry 
support is not directly linked to the FDA's work and 
performance. Would you elaborate on that for a second.
    Dr. Woosley. I applaud the willingness of the 
pharmaceutical industry to help the Agency do its job in 
reviewing drugs. But when you tie it, like piecework, to the 
review of the drug, it really misses the real need that the 
Agency has for something more than just reviewing that product. 
When a new drug goes on the market, there are drug interactions 
with other drugs that can affect the safety and the use of 
other drugs. That is not paid for when you just pay for the 
review. So, by having more drugs on the market, you create more 
work for everybody and more surveillance.
    So what I would suggest is that we come up with a way to 
fund the Agency, ideally, without user fees. In a perfect 
world, that would be the goal, but that may not be the reality. 
So a compromise would be to find a way that the industry can 
support the FDA's role in assisting them in the regulation of 
their products. Now, that does not mean just reviewing the NDA. 
It means all of the work for the FDA that the FDA needs to do 
to improve the drug development process, to improve the drug 
surveillance system. How that is done, I mean, that is what you 
gentlemen and ladies are expert at, but I think the basic 
principle of tying it like piecework to the product creates the 
wrong environment.
    It, also, because of the large amount of money coming into 
the FDA for that part of it, and the lack of additional money 
for the rest of the mission, I made the analogy to, if you open 
up a restaurant and you sold all sorts of sandwiches, but 
everybody bought the roast beef, then you become a roast beef 
store. And that is what the FDA has become because there is so 
much money now going for the review process and not for the 
rest of its work, people think that it is beholding to the 
industry. It is not beholding to the industry. It is just that 
that part of it is well-funded. The rest of its mission needs 
to be funded, also.
    Senator Isakson. My time is almost up, but I want to make 
sure I understand. You talked about the staged approval 
process. Did that mean that there would be a stage at which 
certain patients might get a drug before total approval took 
place? Is that what that meant?
    Dr. Woosley. That is what it means. It means that the drugs 
would be used in people who have been tested, that it would 
move into the community in a staged fashion, and as you learn 
more about it, then it would be expanded into the broader 
population.
    Senator Isakson. Just to comment, Mr. Chairman. I think her 
name, Mrs. Washington Ines, I think was her name, that 
testified yesterday or the day before on behalf of cancer 
patients. That is a particular census of patients where it 
appears to me that would be a process well worth looking at. Is 
that the kind of thing you are talking about?
    Dr. Woosley. Exactly.
    Senator Isakson. Thank you, Doctor. Thank you, Mr. 
Chairman.
    The Chairman. Thank you.
    Senator Burr.
    Senator Burr. Thank you, Mr. Chairman.
    I thank all of you for your willingness to be here and for 
the expertise that you bring.
    Dr. Wilson, I have got to ask you about one thing. One, you 
need to know I do not hold much confidence, and have not held 
much confidence, in MedWatch. I think that the marketplace has 
changed. I think it is an antiquated program. It worked at one 
time, and the fact is that we cannot count too much on it 
regardless of how it is structured, so I am not going to ask 
you how to restructure it or your confidence in it.
    I am concerned that in your written testimony you talked 
about the FDA needing to spoon-feed physicians, and I think 
that in the system that we have got, and certainly with the 
ability for physicians to make off-label decisions about 
prescriptions that they write, it is really incumbent on 
physicians not necessarily to wait for pharmaceutical reps or 
for the FDA or for some outside entity to share with them 
either the original information of a new drug or the ongoing 
revisions that might be learned in a postsurveillance process 
or in the practice of medicine.
    So I am going to ask you to elaborate, if you will, exactly 
what you meant because I do not think I read it in the same 
context.
    Dr. Wilson. Thank you, Senator. And I hope I did not say 
``spoon feed'' in the testimony.
    Physicians have a significant responsibility to recognize 
untoward effects and to take those known untoward effects of 
drugs into account when they are counseling with patients, be 
sure the patients understand both the risks and the benefits. 
Having gone through a college and medical school, I believe 
strongly that communication and education can change behavior. 
So I think my observation about present methods of 
communication, whether it is MedWatch or whether it is the drug 
insert, which we have discussed, I would suggest that if it is 
not working, it is not because communication and education 
cannot work, it is because it is because it is not good 
communication and education.
    So I think we, as physicians, would suggest that we need to 
look at ways to communicate, and people who know how to do that 
do it very well. But I would not discard that because the 
challenge in discarding it is to put regulations which 
prescribe who can prescribe the medicine and under what 
circumstance, which put requirements on patients in terms of 
reporting and testing. Efforts which may well decrease adverse 
side effects, I would suggest, very well will decrease access 
to medication. So we are wedded to better communication, better 
surveillance, better education.
    Senator Burr. I agree with you wholeheartedly, and I think 
communication is at the root of it. And just since 1997, when 
FDMA was passed, technology has changed in such a way that you 
no longer have a pharmaceutical reference book that you go to. 
First, you go to your computer. It is a much faster, easier, 
more complete analysis that one can receive.
    There is only one problem. It has to be initiated either by 
an individual or by a medical professional. They have to be 
willing to go there and to look for the answer, but I think 
that the answers are available. They are certainly not 
available in the things that we do not know, and I encourage 
everybody within the system to begin to look at how we not 
mandate it, but continue to make it easier for individuals to 
want to do that.
    Dr. Woosley, welcome. It is great to see you again. I have 
got to ask you, because I have got great pride in CERT, how is 
it going? Can you give us an update?
    Dr.  Woosley. It is going great, and I really thank you, as 
former Congressman Burr, for being one of the champions and the 
visionaries that helped us create this CERT. There are seven of 
them now. They are competing for creating four more.
    I must say, though, the biggest limitation--remember, you 
remember this I am sure--CERTS were created to work with the 
FDA. The biggest failing of the CERTs is the lack of ability to 
work with the FDA to the level that we need to. They do not 
have the budgets to send people. They send one person to each 
of our meetings every quarter. But day-to-day interactions with 
the FDA needs to occur for us to be successful in our CERTs.
    That is what I was saying earlier is that because safe 
drugs means safe use of drugs, we need the people, like AHRQ 
and CERT, to be part of the solution in drug safety.
    Senator Burr. I think we have heard, in a pretty 
coordinated fashion, both several days ago and then from this 
panel, that the focus should not be on how we create something 
new. What we have is a system that can work if you properly 
fund it. If the emphasis is on making sure that you fill in the 
holes and the gaps that exist, and I hope that, in fact, we are 
going to do that, my hope, looking at the changes that have 
just happened again since 1997, the creation of CERT, the 
frustrations that I think more people share than just me with 
MedWatch, are that we can design something that is better. It 
can be within FDA. It can take the talents of professionals 
that we have there. It can use the expertise in the outside 
world of physicians and researchers across the country.
    In addition to that, we now have the human genome mapping 
complete, and our ability now to take this to another level, 
whether that is within the FDA or within NIH or within HHS or 
within academia in this country, to be able to look at the 
current compounds that we have from a human genome standpoint 
and begin to target where we might not have picked up something 
that we might want to go back and look at or that raises a red 
flag is available, but we have got to have the initiative to go 
there.
    So it may be that this conversation is not just about where 
do we go on this particular incident, but where do we go down 
the road based on more technology and based upon where our 
knowledge allows us to go. And it is refreshing to think that 
we could be at a point where we are not tasked with dealing 
with a crisis, where we are actually a visionary committee, and 
I think it is because of the chairman's willingness to do these 
hearings that we are actually focused now and, hopefully, will 
focus out a little bit further than just tomorrow.
    Mr. Chairman, thank you.
    The Chairman. Thank you.
    I wish we had more time for going into this. I have about 
another 20 questions that I came with, and I have got another 
dozen questions that I have developed as a result of the things 
that you said, that we will need more clarification on, but we 
will get to those and would appreciate a response.
    We have had some excellent testimony. I really appreciate 
your interest, the information, your attendance. The hearings 
that we have had on the FDA have made a great contribution I 
think to the debate on drug safety. The witnesses have given us 
the benefit of their vast and varied experience, and we have 
had a tremendously varied panel in all instances. They have 
brought us some serious and some innovative proposals to 
improve the system and given us a lot to think about.
    I look forward to working with my colleagues on both sides 
of the aisle to develop a comprehensive response to the issues 
that have been raised in these hearings. As I mentioned, there 
will be further questions. Your full testimony will be in the 
record. You will have an opportunity to expand on any of your 
remarks or any of the questions that others were given as well, 
and those will become a part of the record. The record will be 
open for 10 days for that. So I thank you for your 
participation, and this hearing is now adjourned.
    [Additional Information follows:]

                         Additional Information

          Prepared Statement of Senator Hillary Rodham Clinton

    I would like to thank Senator Enzi and Senator Kennedy for 
convening today's hearing, and I look forward to learning more 
about the ways in which the FDA and private industry are using 
technology to further the agency's mission of ensuring the 
safety of medications for all Americans.
    I am extremely interested in the opportunity to learn more 
about the FDA's Critical Path Initiative, through which the 
agency will identify and prioritize the most pressing drug 
development problems facing our public health system.
    This initiative will expedite the development of new 
technologies that can improve the assessment of new products, 
allow for the use of more targeted therapies, and improve post-
marketing safety monitoring.
    The innovations that could result from such an initiative 
would be invaluable to our health system. Not only could they 
be used to improve the health status of Americans, but they 
would improve the quality of care provided to consumers, and 
could reduce the impact of medical errors upon the healthcare 
system.
    While First Lady, I began my work with the FDA on 
developing the Pediatric Rule, which ensures that drugs 
marketed to pediatric populations have first been tested on 
children. And I've continued to work on these issues with my 
colleagues, including Senators DeWine and Dodd, during my time 
in the Senate. I know that my colleagues and I are interested 
in the ways that the technologies we discuss today could be 
used to further deliver clinically appropriate treatments to 
pediatric populations.
    For example, genetic markers may someday allow physicians 
to determine how a specific patient will react to a given 
medication, thus removing many of the dosing uncertainties that 
exist in modern medicine.
    I am pleased to learn that the FDA has already identified 
the data collected through the Best Pharmaceuticals for 
Children Act and the Pediatric Rule as the foundation for 
development of medications that are increasingly safe and 
efficacious for use in pediatric populations.
    And I believe the lessons we learn from analyzing this 
pediatric data will benefit all patient populations with 
specific dosing and treatment needs--in short, all Americans.
    Again, I would like to thank Senators Enzi and Kennedy for 
convening today's hearing, and I look forward to working with 
my colleagues on the HELP committee to developing some 
practical, bipartisan solutions to encourage the development of 
innovative and safe medications for American consumers.

         Questions of Senator Clinton For Janet Woodcock, M.D.
    Question 1. In the FDA's ``Innovation or Stagnation'' report, the 
agency mentions the research possibilities associated with the body of 
data collected by the FDA as a result of the Best Pharmaceuticals for 
Children Act and the Pediatric Rule. Could you please elaborate on the 
ways that the Critical Path Initiative might utilize this data to 
perform a comprehensive analysis of pediatric pharmacology, safety and 
efficacy?
    Question 2. I was interested to learn that the technology 
development initiatives supported by the FDA might allow clinicians to 
target medications to very specific patient populations. Could you 
please comment on the ways in which the targeting of medications might 
enhance the delivery of clinical appropriate treatments to children 
beyond the scope of what we are currently able to achieve with the 
Pediatric Rule and Best Pharmaceuticals for Children Act?

                                 ______
                                 
                       American Medical Association
                                          Chicago, Il 60610
                                                       July 6, 2004
Division of Dockets Management (HFA-305)
Food and Drug Administration
5630 Fishers Lane, Room 1061
Rockville, MD 20852

RE: Draft Guidance for Industry on ``Development and Use of Risk 
        Minimization Action Plans'' [Docket No. 2004D-0188]

    The American Medical Association (AMA) is pleased to offer its 
comments on the Food and Drug Administration's (FDA) May 2004 Draft 
Guidance for Industry on ``Development and Use of Risk Minimization 
Action Plans'' [Fed Reg. 2004;69:25130-25132]. The AMA's comments focus 
principally on Sections II-V of the Draft Guidance, and generally are 
consistent with AMA's previous comments of April 29, 2003 on the FDA's 
Concept Paper, ``Risk Management Programs'' [Docket No. 02N-0528], and 
of May 22, 2002 in testimony at FDA's Public Meeting on the Risk 
Management of Prescription Drugs.

GENERAL COMMENTS ABOUT THE DRAFT GUIDANCE, DEVELOPMENT AND USE OF RISK 
                       MINIMIZATION ACTION PLANS

    The AMA shares a common goal with the FDA to optimize the benefit/
risk balance of drug therapy and to minimize the risks of drug and 
biological products. However, the AMA remains concerned about the 
number of Risk Minimization Action Plan (RiskMAP) tools described in 
the Draft Guidance that would directly manage or restrict physician 
prescribing. If these tools are expanded to more pharmaceutical 
products, the potential for unintended consequences such as reduced 
patient access to necessary drugs or reduced manufacturer investments 
in innovative therapies is significant. Thus, the AMA continues to 
recommend that higher level risk minimization tools, such as 
performance-linked access systems and some reminder systems, should be 
used only as a last resort to keep high-risk products with unique and 
important benefits on the market.
    On the other hand, the AMA commends the FDA for incorporating 
changes into the Draft Guidance that respond to some of our criticisms 
of the Agency's 2003 Concept Paper on this subject. In particular, the 
AMA is pleased that the FDA is encouraging drug sponsors to:
     Develop RiskMAPs only for products that pose an unusual 
type or level of risk;
     Use RiskMAPs judiciously to minimize risks without 
encumbering drug availability or otherwise interfering with the 
delivery of product benefits to patients;
     Seek the input of other stakeholders, including 
physicians, when planning risk minimization activities and when 
selecting specific RiskMAP tools;
     Apply objective criteria when determining whether a 
RiskMAP is necessary for a particular product;
     Select the minimum number of RiskMAP tools necessary to 
minimize the risk, select tools based on available evidence of 
effectiveness, and objectively evaluate the effectiveness of RiskMAPs 
and their tools using evidence-based performance measures;
     Adopt tools that facilitate the central role of the health 
care practitioner in controlling the risks of medical product use; and
     Consider unintended consequences of a RiskMAP, such as 
reduced access, as part of the sponsor's Evaluation Plan.
    The AMA offers the following comments on individual Sections II-V 
of the Draft Guidance.

                         SECTION II: BACKGROUND

    The AMA agrees with the FDA that ``when planning risk assessment 
and risk minimization activities, sponsors should consider stakeholder 
input (e.g., from consumers, pharmacists, physicians, third-party 
payers).'' However, the AMA believes the FDA needs to put greater 
emphasis on this important point in a Final Guidance.
    The AMA continues to urge open communication and collaboration 
among the FDA, the pharmaceutical industry, and national physician 
organizations on the subject of risk management. Such communication and 
collaboration is needed at the macro level so that the FDA's overall 
risk management initiative achieves an appropriate balance between the 
need to protect patients from harm and the need to avoid heavy-handed 
regulations that interfere with medical practice. Furthermore, 
collaboration among the FDA, a product sponsor, and relevant physician 
organizations also is recommended for individual product RiskMAPs, as 
described in the Draft Guidance, to ensure that the RiskMAP is 
effective, feasible and acceptable in usual health care practices.
    Furthermore, the FDA also may wish to consider establishing a 
permanent advisory council of practicing physicians, representing a 
large number of national medical specialty societies, that could advise 
the Agency on issues like RiskMAPs on an ongoing basis.

    SECTION III: THE ROLE OF RISK MINIMIZATION AND RISKMAPS IN RISK 
                               MANAGEMENT

    Determining an Appropriate Risk Minimization Approach. The AMA 
strongly agrees with the FDA that the FDA-approved professional 
labeling (Package Insert [PI]), updated from time-to-time to 
incorporate information from routine postmarketing surveillance, is 
sufficient to be the routine risk minimization plan for the vast 
majority of drug and biological products. The information provided in 
the PI, along with other information about a product (e.g., published 
clinical trials), should remain the standard method of providing 
benefit and risk information to physicians about the use of a drug or 
biological product.
    However, as previously communicated to FDA, the AMA believes that 
the current PI for prescription drugs is a barrier to effective risk 
communication because it has become a legal document rather than a 
resource of useful information for busy practicing physicians. In 
December 2000, the FDA issued a Proposed Rule to modify the format and 
content of the PI with the goal of making the information more useful 
and user-friendly to physicians. The AMA has supported this effort, 
especially the proposed ``Highlights of Prescribing Information.'' The 
AMA urges the FDA to issue a Final Rule implementing these changes to 
the PI as soon as possible.
    Furthermore, the FDA should promptly develop and make readily 
available (e.g., via the Internet) a computerized database of the most 
up-to-date prescription drug labeling for all products. Such a database 
could have prominently placed safety alerts for new risk information on 
selected drugs. Physicians need to be trained to use this database for 
their professional labeling needs in lieu of the hard-copy Physicians 
Desk Reference (PDR) that is both cumbersome and dated for certain 
products.
    Definition of Risk Minimization Action Plan (RiskMAP). The AMA 
accepts the FDA's definition of a RiskMAP as ``a strategic safety 
program designed to meet specific goals and objectives in minimizing 
known risks of a product while preserving its benefits.'' Moreover, the 
AMA agrees with the FDA that tools used to meet RiskMAP goals and 
objectives do not apply to routine risk minimization plans, i.e., FDA-
approved professional labeling.
    Determining When a RiskMAP Should be Considered. The AMA agrees 
with the FDA that the decision to develop a RiskMAP needs to be 
determined on a case-by-case basis. Moreover, the AMA supports the 
FDA's recommendation to use objective criteria, such as type of risk, 
magnitude of risk, frequency of risk, populations at greatest risk and/
or those likely to derive the most benefit, existence of alternative 
treatments, reversibility of adverse events observed, preventability of 
the adverse event, and probability of benefit, when considering whether 
a RiskMAP is necessary. As previously discussed, the AMA encourages the 
FDA and the product sponsor to seek the input of relevant physician 
organizations in determining whether a RiskMAP is needed. This will 
give further assurance to physicians that the process is equitable and 
driven by good science.

      SECTION IV: TOOLS FOR ACHIEVING RISKMAP GOALS AND OBJECTIVES

    Relationship of RiskMAP Tools to Objectives and Goals. The AMA has 
no specific comments on this section.
    Categories of RiskMAP Tools. The AMA accepts the FDA's three 
categories of RiskMAP tools, i.e., targeted education and outreach, 
reminder systems, and performance-linked access systems.
    Description of RiskMAP Tools. The AMA supports the establishment of 
a RiskMAP Web site by FDA. At a minimum, this Web site should contain a 
description of RiskMAP tools that have been used and all available 
evidence on the effectiveness of each tool in achieving a risk 
minimization objective and/or goal. The AMA believes this is necessary 
to convince health care practitioners that a potentially burdensome 
RiskMAP tool can effectively improve the benefit/risk balance for a 
drug product.
    Selecting and Developing the Best Tools. This is an especially 
important section of the Draft Guidance, and the AMA commends the FDA 
for its recommendations to product sponsors, that when selecting 
RiskMAP tools, to:
     Maintain the widest possible access to the product with 
the least burden to the health care system that is compatible with 
adequate risk minimization;
     Identify the key stakeholders (e.g., physicians) who have 
the capacity to minimize the product's risks and to define their roles;
     Seek input from these stakeholders, including physicians, 
on the feasibility of implementing and accepting a particular RiskMAP 
tool in usual health care practices;
     Use RiskMAP tools with the least burdensome effect on 
physician-patient relationships;
     Select tools based on available evidence of effectiveness 
in achieving the specified objective; and
     Consider, and seek to avoid, unintended consequences of 
tool implementation that obstruct risk minimization and product 
benefit.
    The AMA also appreciates the FDA's recognition that physicians are 
the most important managers of product risks once a drug is marketed 
and, furthermore, that the FDA does not have the authority to control 
prescribing decisions made by physicians for their patients. The AMA 
strongly agrees with the FDA's view that product sponsors should 
recognize this central role played by physicians in controlling the 
risks of medical product use and should adopt tools that facilitate 
this role.
    Only time and experience will answer the question as to whether 
drug product sponsors are implementing RiskMAPs that are consistent 
with the recommendations put forth by the FDA in this section of the 
Draft Guidance. The AMA is hopeful that this will be the case. When 
RiskMAPs are considered necessary, the AMA encourages the FDA and the 
product sponsor to work with relevant physician organizations to assure 
that the minimum number and least intrusive RiskMAP tools are selected 
to achieve the risk minimization objective. Whenever possible, targeted 
education and outreach should be the RiskMAP tools selected, and the 
AMA refers the FDA to our letter of April 29, 2003 to Docket No. 02N-
0528 for detailed comments on how risk communication to physicians can 
be improved.
    As stated earlier in this letter, the AMA continues to believe that 
higher level risk minimization tools, such as performance-linked access 
systems and some reminder systems, should be used only as a last resort 
to keep high-risk products with unique and important benefits on the 
market. As discussed in detail in our earlier letter of April 29, 2003, 
a number of potential unintended consequences, including reduced access 
to necessary therapies, substitution of less effective therapies that 
are not subject to RiskMAPs, multiple burdensome and confusing RiskMAPs 
that can lead to errors, and adverse effects on pharmaceutical 
innovation, may result if RiskMAPs with high level risk minimization 
tools are more commonly employed.
    Mechanisms Available to the FDA to Minimize Risks. The AMA has no 
specific comments on this section.

SECTION V: RISKMAP EVALUATION: ASSESSING THE EFFECTIVENESS OF TOOLS AND 
                                THE PLAN

    Rationale for RiskMAP Evaluation. The AMA is in strong agreement 
with the FDA regarding the need for well-designed studies to 
periodically evaluate the effectiveness of a RiskMAP. The AMA concurs 
that the most important evaluation is of the overall performance of a 
RiskMAP in achieving its targeted health outcomes and goals. However, 
the AMA also agrees that separate assessments should be done for 
individual tool performance and for acceptability of RiskMAP tools by 
physicians.
    Considerations in Designing a RiskMAP Evaluation Plan. The AMA is 
in general agreement with the FDA on the details of this section. In 
particular, the AMA supports the following FDA recommendations:
     When possible, drug product sponsors should select well-
defined, evidence-based, and objective performance measures tailored to 
the particular RiskMAP to determine whether the RiskMAP's goals or 
objectives are being achieved.
     Whenever feasible, drug product sponsors should design 
evaluation plans to include at least two different, quantitative, 
representative, and minimally biased evaluation methods for each 
critical RiskMAP goal to compensate for the limitations of the other.
     Drug product sponsors should periodically evaluate each 
RiskMAP tool to ensure it is materially contributing to the achievement 
of RiskMAP objectives and goals to eliminate ineffective tools and 
concentrate resources on useful tools.
     Drug product sponsors should evaluate RiskMAP tools prior 
to implementation; this should include pilot testing to assess 
comprehension, acceptance, feasibility, and other factors to determine 
how readily RiskMAP tools will fit into everyday physician practices.
     Formal evaluation plans are unnecessary for routine risk 
minimization plans, i.e., FDA-approved professional labeling.
    FDA Assessment of RiskMAP Evaluation Results. The AMA generally 
supports this section on how the product sponsor reports a RiskMAP 
evaluation to the FDA, and that FDA will perform its own assessment of 
RiskMAP effectiveness.
    Making Information from RiskMAP Evaluation Available to the Public. 
As stated earlier in this letter, the AMA supports the establishment of 
a RiskMAP Web site by FDA that would include descriptions of RiskMAP 
tools and all available evidence on the effectiveness of these tools. 
The AMA also believes that this Web site should contain results of 
evaluations of RiskMAPs that have been previously implemented to inform 
physicians and the public about the effectiveness of the program in 
meeting its risk minimization objectives and goals. While the AMA 
understands that some product sponsor information will remain 
proprietary, we believe it is in the sponsor's and FDA's best interests 
to be as transparent as possible about the effectiveness of a RiskMAP. 
Such transparency will provide credible evidence to physicians and the 
public that a particular RiskMAP either did or did not effectively 
improve the benefit/risk balance for a drug product.

                           ADDITIONAL COMMENTS

    In our letter of April 29, 2003, the AMA offered two additional 
comments that have not been adequately addressed by the FDA in the 
Draft Guidance. First, concern has been expressed by physicians and 
pharmacists that it is difficult to remember the various risk 
management programs (now called RiskMAPs), and especially the multiple 
risk management (RiskMAP) tools, currently employed for various drug 
products. This is because each risk management program has been 
uniquely developed for a specific drug product and, therefore, all of 
the current programs are different in their requirements. However, in 
Section IV(D) of the Draft Guidance, FDA continues to suggest that the 
best RiskMAP tool or tools be selected on a case-by-case basis.
    To address this concern, the AMA encourages the FDA, in 
collaboration with the pharmaceutical industry and other stakeholders 
(e.g., physician organizations), to take a more systems-based approach 
to RiskMAPs. Appropriate tools should be prospectively developed based 
on evidence of effectiveness, and a standard set of tools for each 
level of risk should be part of a standard ``toolbox'' of RiskMAP 
tools. When a product meets the criteria for a RiskMAP at a certain 
level, to the extent possible, a standard set of tools should be 
employed in that product's RiskMAP. At a minimum, any given tool should 
be consistent across products.
    The AMA's other comment that was not addressed in the FDA's Draft 
Guidance regards the incorporation of RiskMAPs for drug products into 
more global quality assurance programs. The AMA believes that the FDA, 
the pharmaceutical industry, physician organizations, and other 
stakeholders need to consider the incorporation of risk management 
(RiskMAPs) for drug and biological products into more global quality 
assurance programs. As electronic health records (EHRs) and E-
prescribing become more common and they are electronically linked to 
other aspects of care (e.g. lab test results), it should be possible to 
effectively incorporate RiskMAPs, as part of overall quality assurance, 
into the normal routine of physician practice. As an analogy, the 
Physician Consortium for Performance Measurement, convened by the AMA, 
is currently developing physician performance measures derived from 
evidence-based practice guidelines. The AMA is working with physician 
group practices that have EHRs to incorporate the performance measures 
into their systems so that satisfying the performance criteria becomes 
a routine part of medical practice.

                               CONCLUSION

    In conclusion, the AMA appreciates the opportunity to comment on 
the FDA's Draft Guidance for Industry on ``Development and Use of Risk 
Minimization Action Plans.'' We hope that our insight into the issues 
discussed in the Draft Guidance proves helpful for the FDA as it moves 
to finalize this Guidance. We look forward to working with the Agency 
as it continues its activities in this area.
            Sincerely,
                                Michael D. Maves, M.D., MBA

                                 ______
                                 
                       American Medical Association
                                    Chicago, Illinois 60610
                                                     April 29, 2003
Dockets Management Branch (HFA-305)
Food and Drug Administration
5630 Fishers Lane, Room 1061
Rockville, MD 20852

RE: Risk Management [Docket No. 02N-0528]

    The American Medical Association (AMA) is pleased to offer its 
comments on the Food and Drug Administration's (FDA) Notice on risk 
management activities for drug and biological products that was issued 
in the March 7, 2003 Federal Register. The AMA's comments focus 
principally on Sections II-V of the FDA Concept Paper, ``Risk 
Management Programs,'' and generally are consistent with the AMA's 
testimony at the FDA's Public Meeting on the Risk Management of 
Prescription Drugs on May 22, 2002. We further address the FDA's 
question about improving the quality of spontaneously reported case 
reports [of adverse events], which was part of the FDA Concept Paper, 
`` Risk Assessment of Observational Data: Good Pharmacovigilance 
Practices and Pharmacoepidemiologic Assessment.''

   GENERAL COMMENTS ABOUT THE CONCEPT PAPER, RISK MANAGEMENT PROGRAMS

    The AMA has had a longstanding commitment both to improving the 
quality of medical care delivered by physicians to patients and to 
promoting efforts to improve patient safety. In furtherance of this 
goal, the AMA established the National Patient Safety Foundation in 
1997 and has participated in a number of initiatives on clinical 
quality improvement. The AMA also has been a partner and strong 
supporter of MedWatch, the FDA's adverse event reporting program. As 
such, the AMA shares a common goal with the FDA to optimize the 
benefit/risk balance of drug therapy and to minimize the risks of drug 
and biological products.
    However, a number of the risk management tools described in the 
FDA's Concept Paper would directly manage or restrict physician 
prescribing. The AMA has serious concerns about the potential 
unintended consequences if these tools were expanded to more 
pharmaceutical products. We are particularly concerned that the use of 
these risk management tools could prevent some patients who would 
benefit from higher-risk drugs from having access to them, or that 
potential restrictions on prescribing could serve as a deterrent to 
manufacturer investments in innovative therapies. As expressed in our 
testimony last year, the AMA is also concerned that the FDA, and drug 
sponsors, may be attempting to regulate the practice of medicine 
through some of these risk management tools in ways that exceed the 
FDA's statutory authority.
    Other than the AMA's testimony at the FDA Public Meeting in May 
2002, we are unaware of any input from national medical specialty 
societies on the FDA's risk management initiatives. The AMA believes it 
is essential that there be open communication and collaboration among 
the FDA, the pharmaceutical industry, and national physician 
organizations on this subject. Such communication and collaboration is 
needed at the macro level so that the FDA's overall risk management 
initiative achieves an appropriate balance between the need to protect 
patients from harm and the need to avoid heavy-handed regulations that 
interfere with medical practice. Furthermore, collaboration among the 
FDA, a product sponsor, and relevant physician organizations also is 
recommended when a risk management program, as described in the Concept 
Paper, is being contemplated for a specific drug or biological product.

             SECTION II: IMPORTANT RISK MANAGEMENT CONCEPTS

    The AMA strongly agrees with the FDA that the Package Insert (PI), 
as defined in this section of the Concept Paper, combined with routine 
postmarketing surveillance should constitute the risk management plan 
for the vast majority of drug and biological products. The information 
provided in the PI, along with other information about a product (e.g., 
published clinical trials), should remain the standard method of 
providing benefit and risk information to physicians about the use of a 
drug or biological product.
    However, the AMA believes that the current PI for prescription 
drugs is a barrier to effective risk communication because it has 
become a legal document rather than a resource of useful information 
for busy practicing physicians. In December 2000, the FDA issued a 
proposed rule to modify the format and content of the PI with the goal 
of making the information more useful and user-friendly to physicians. 
The AMA has supported this effort, especially the proposed ``Highlights 
of Prescribing Information.'' The AMA urges the FDA to issue a final 
rule implementing these changes to the PI as soon as possible.
    Furthermore, the FDA should promptly develop and make readily 
available (e.g., via the Internet) a computerized database of the most 
up-to-date prescription drug labeling for all products. Such a database 
could have prominently placed safety alerts for new risk information on 
selected drugs. Physicians need to be trained to use this database for 
their professional labeling needs in lieu of the hard-copy Physicians 
Desk Reference (PDR) that is both cumbersome and dated for certain 
products.

      SECTION III: WHEN WOULD AN RMP BEYOND THE PACKAGE INSERT BE 
                              APPROPRIATE?

    The AMA accepts the FDA's definition of a risk management program 
(RMP) as ``a strategic safety program designed to decrease product risk 
by using one or more interventions or tools beyond the package insert'' 
(see Section II of the Concept Paper). Thus, the remainder of the AMA's 
comments will assume a drug or biological product requires a Level 2, 
3, or 4 RMP, as defined in Section IV of the Concept Paper.
    The AMA agrees with the FDA that the decision to develop an RMP for 
a particular product, and the level of the RMP, needs to be determined 
on a case-by-case basis. This will depend on the severity of the risks 
when compared to the magnitude of the benefits for a drug or biological 
product, and the likelihood that an RMP would lower the risks without 
adversely affecting the benefits. As discussed above, the input of 
relevant physician organizations in this decision-making should help 
the FDA and the product's sponsor select the most appropriate RMP for 
the product.
    To help determine whether any drug or biological product needs an 
RMP, as well as the level of the RMP, the AMA believes it would be 
useful for the FDA, the pharmaceutical industry, and physician 
organizations to collaborate on the development of objective criteria 
for making this determination. Severity of risk, frequency of risk, 
reversibility of risk by an effective RMP, importance of product 
benefit to patient outcome, and availability and relative benefit/risk 
of alternative therapies are among the factors that should be 
considered in developing criteria for determining whether an RMP is 
needed. This collaborative development of objective criteria to 
determine the need for an RMP would give some assurance to all 
stakeholders that the process is equitable and driven by good science.

   SECTION IV: WHAT INTERVENTIONS OR TOOLS ARE AVAILABLE FOR USE IN 
                  ACHIEVING RMP GOALS AND OBJECTIVES?

    The AMA has a number of comments on this section of the Concept 
Paper. In making our comments, we have assumed the Level 1--4 
categorization scheme, as proposed by the FDA under Section IV(D), is 
applicable.
    Physician education (Level 2) should be the risk management tool 
used for most drug and biological products that need an RMP.
    The AMA believes that the FDA should promote physician education 
through improved risk communication as the tool that should be used for 
most drug and biological products that need an RMP. Level 3 and Level 4 
RMPs should be used only as a last resort to keep high-risk products 
with unique and important benefits on the market.
    Based on our experience at the May 2002 Public Hearing, the AMA is 
concerned that the FDA has a predetermined view that risk communication 
to physicians is ineffective in modifying prescribing behavior to 
minimize risk. For example, the FDA considers the effectiveness of 
traditional ``Dear Doctor'' letters that are mailed to physicians when 
new and important risks are discovered to be questionable. While this 
may be true, it is an indication that more innovative and effective 
approaches to physician education about risk need to be developed, not 
an indication that Level 3 and 4 RMPs should be more frequently 
employed. The AMA urges the FDA to work with all stakeholders to make 
physician education through improved risk communication an effective--
and the preferred--RMP for most products.
    The AMA believes that the FDA, the pharmaceutical industry, and 
physician organizations must collaborate and identify innovative ways 
to communicate new risk information about a drug or biological product 
to physicians so they will be aware of it, remember it, accept it, and 
act on it when prescribing a drug. At the May 2002 Public Meeting, the 
AMA presented a number of potential ways to accomplish this goal. Most 
of these options could be implemented immediately, including:
     The FDA, the pharmaceutical industry, and physician 
organizations should undertake a major CME initiative on risk 
communication. Physicians need to be aware of labeling changes that 
identify serious adverse events and that, in some cases, these serious 
adverse events can be minimized by modifications in prescribing. The 
AMA's recommendations that the FDA publish its final rule on the PI and 
create a computerized database of up-to-date PIs, as discussed above, 
should be implemented as part of this education initiative.
     The FDA, in collaboration with physician organizations, 
should work with major medical journals and medical society web site 
editors to identify standard places for the dissemination of important 
new risk information about drugs and biological products.
     ``Dear Doctor'' letters should be disseminated by 
mechanisms other than hard-copy mail. Alternative mechanisms should 
include publication in medical journals (possibly as paid 
advertisements), placement on medical society web sites, and 
transmission to individual physicians by blast fax, blast email, and 
direct daily downloads to personal digital assistants (PDAs). Unlike 
letters, electronic transmission is inexpensive, timely, and 
repeatable. Thus, important risk information can be reinforced by more 
than one transmission.
     The content and format of ``Dear Doctor'' letters should 
be changed to emphasize the need for action by the prescribing 
physician. For example, a ``Dear Doctor'' letter should contain a bold-
faced opening paragraph that emphasizes the possible severe outcome 
(e.g., permanent harm or death) to patients from the new adverse event, 
that the adverse event is probably preventable if the drug is used 
appropriately, and what necessary steps the physician must take to 
prescribe the drug appropriately.
     Pharmaceutical companies should be obliged to train and 
send their sales forces to physicians to educate them on important new 
risk information about company products. The company should provide 
incentives to sales representatives to do this because the highest 
priority of any company should be to prevent harm to patients who use 
their products. The effectiveness of the 80,000 pharmaceutical sales 
representatives in the United States in promoting the benefits of their 
companies' products is well documented, and they could have similar 
success in educating physicians about important product risks.
     New information technologies, such as computerized 
physician order entry (CPOE), offer enormous opportunities to 
communicate important risk information about drug and biological 
products. CPOE systems with well-designed decision support programs 
potentially could communicate important new risk information to 
physicians at the point of prescribing, i.e., at a time when the 
information is most needed. As these new information technologies 
become integrated into physician practice, the FDA, the pharmaceutical 
industry, and physician organizations should work with database 
providers and software vendors to incorporate the appropriate risk 
information into these electronic systems.
     The AMA encourages the FDA and the pharmaceutical industry 
to work with physician organizations to optimize physician education 
about the risks of drug and biological products through identification 
and implementation of effective methods of risk communication. The AMA 
also recommends that the Centers for Education and Research on 
Therapeutics (CERTs) program be charged with developing a research 
agenda in risk communication to help identify new and effective 
educational strategies.
    Level 3 and Level 4 RMPs should be used only as a last resort to 
keep high-risk products with unique and important benefits on the 
market.
    The AMA has concerns about many of the tools that the FDA has 
proposed under Level 3 and Level 4 RMPs including:
     prescribing only by registered physicians (restricted 
distribution);
     certification programs for physicians;
     enrollment of physicians in a safety program;
     specialized systems or records that attest to safety 
measures having been satisfied (e.g., stickers, physician attestation 
of capabilities);
     dispensing only to patients with evidence or other 
documentation of safe use conditions (e.g., lab test results) 
(restricted distribution); and
     patient agreements/informed consent.
    As discussed above, the AMA has general concerns about the FDA and 
product sponsors managing or restricting physician prescribing. There 
also are a number of other reasons why the AMA believes that Level 3 
and Level 4 RMPs should be used only as a last resort to ensure that 
high-risk products with unique and important benefits remain on the 
market. These reasons include:
     While Level 3 and Level 4 RMPs may reduce risk, such 
programs most likely will also reduce access. Some patients who would 
benefit from a product subject to a high-level RMP may not be 
prescribed that product because of the added burdens on the prescriber.
     A less effective, less studied, and even less safe 
alternative drug or biological product not subject to a high-level RMP 
may be prescribed instead of a product with a Level 3 or Level 4 RMP. 
There is some anecdotal information to suggest that this may be 
happening with drugs used to treat cardiac arrhythmias. Sotalol and 
quinidine, neither subject to an RMP, may be prescribed instead of 
dofetilide, which is subject to a high-level RMP, when dofetilide is 
actually the preferred drug.
     Level 3 and Level 4 RMPs that employ multiple tools are 
complex and may be confusing to both the physician and patient. This 
could result in unintended medication errors unrelated to adherence to 
the RMP. This could be magnified in patients with multiple diseases who 
are on multiple drug products with multiple high-level RMPs, all of 
which could be different.
     Many of the tools for Level 3 and Level 4 RMPs are 
administrative burdens for physicians. Therefore, unless the product 
provides a truly innovative therapy for a particular disease or for a 
specific subset of patients with a disease, it is unlikely that 
physicians will take the necessary time to prescribe the product.
     It is unclear what the impact of Level 3 and Level 4 RMPs 
will have on pharmaceutical company research and development plans. It 
is possible that a company could cease development of a promising drug 
because of the likelihood of a high-level RMP. High-level RMPs could 
have an adverse effect on pharmaceutical innovation, which would 
ultimately limit new drug discoveries.
     For certain drugs subject to Level 3 and Level 4 RMPs, 
patients may seek these products from alternative sources, such as 
illegal foreign Internet sites. For example, if a patient knows about 
the product but cannot find it easy to obtain in the United States, 
then the patient may take direct action and purchase the drug 
illegally. Also, a patient may be concerned about his or her privacy 
and want to avoid a high-level RMP that mandates patient registration 
with a pharmaceutical company.
    For all of these reasons, the AMA believes the FDA must be highly 
discriminating in requiring a drug or biological product to have a 
Level 3 or Level 4 RMP. The serious nature of the risk must clearly be 
validated. As discussed earlier, objective criteria, agreed to by all 
stakeholders, should be developed to determine the need for such a 
high-level RMP. In addition, the FDA and the company must take great 
care in selecting the tools that will be employed in the RMP. Only the 
minimum number of tools needed to effectively reduce the risk should be 
employed in the RMP. Only those tools that have been shown to be 
effective in reducing the risk should be used, and the tools should be 
acceptable to other stakeholders (e.g., physicians).
    An Integrated, Systems-Based Approach to Risk Management of Drug 
and Biological Products is Preferred to Product-Specific RMPs.
    As discussed above, the decision to develop a RMP for a particular 
product, and the level of the RMP, needs to be determined on a case-by-
case basis using objective criteria. On the other hand, the RMPs for 
any given level of risk should be as uniform as possible across 
products. This is especially the case for Level 3 and Level 4 RMPs.
    Currently, the FDA uses a product-by-product approach in developing 
an RMP. Thus, every product has its unique RMP. For high-level RMPs, 
which often employ multiple tools, this results in a number of complex, 
administratively burdensome, and, in some cases, conflicting RMPs. As 
discussed above, this can be confusing to both physicians and patients 
and potentially could result in unintended medication errors.
    Furthermore, it is unclear to the AMA whether any of the different 
Level 3 or Level 4 RMPs for currently marketed drug products, or the 
tools used in these high-level RMPs, have been thoroughly evaluated for 
effectiveness. The AMA requests the FDA to be forthcoming with any 
information about the effectiveness of current RMPs. The AMA also 
questions the impact on patient care of certain tools, such as 
requiring stickers to be placed on handwritten prescriptions, when 
physicians or hospitals no longer use paper prescriptions.
    The AMA encourages the FDA, in collaboration with the 
pharmaceutical industry and other stakeholders (e.g., physician 
organizations), to take a more systems-based approach to risk 
management programs. Appropriate tools should be prospectively 
developed based on evidence of effectiveness, and a standard set of 
tools for each level of risk should be part of a standard ``toolbox'' 
of risk management tools. When a product meets the criteria for a RMP 
at a certain level, to the extent possible, a standard set of tools 
should be employed in that product's RMP. At a minimum, any given tool 
should be consistent across products.
    The AMA also believes that the FDA, the CERTs program, the 
pharmaceutical industry, physician organizations, and other 
stakeholders need to consider the incorporation of risk management for 
drug and biological products into more global quality assurance 
programs. As electronic medical records (EMRs) and CPOE become more 
common and they are electronically linked to other aspects of care 
(e.g. lab test results), it should be possible to effectively 
incorporate drug risk management, as part of overall quality assurance, 
into the normal routine of physician practice. As an analogy, the 
Physician Consortium for Performance Measurement, convened by the AMA, 
is currently developing physician performance measures derived from 
evidence-based practice guidelines. The AMA is working with physician 
group practices that have EMRs to incorporate the performance measures 
into their systems so that satisfying the performance criteria becomes 
a routine part of medical practice.

   SECTION V: HOW AND WHEN CAN RISK MANAGEMENT PROGRAMS BE EVALUATED?

    The AMA strongly supports the evaluation of RMPs for effectiveness. 
In particular, we support the FDA's intent to require risk management 
tools to be pretested prior to their implementation in an RMP. As part 
of this pretesting, the FDA and the sponsor should seek the input of 
physicians and other affected stakeholders to see if the particular 
tool is acceptable. The AMA strongly concurs with the FDA that any RMP 
also must be evaluated after implementation to determine whether the 
program has met its desired objectives. As an important first step and 
as discussed above, the AMA believes that the FDA and the relevant 
sponsors of drug products with high-level RMPs currently should 
evaluate those RMPs, and the tools used in the RMPs, for effectiveness.
    The AMA concurs with the FDA's view that metrics which capture 
actual health outcome data are preferred to those that measure a 
surrogate event or a process. Metrics, preferably quantitative, should 
be well-defined and validated. The AMA agrees with the FDA that two 
different and complementary evaluation methods should be used for key 
RMP goals or objectives. The AMA shares the FDA's view that spontaneous 
adverse event data should not be used as an outcome measure for RMP 
evaluation. The AMA also agrees with the FDA about the limitations of 
administrative claims data for evaluation of RMPs.

      SUMMARY COMMENTS ON CONCEPT PAPER, RISK MANAGEMENT PROGRAMS

    In summary, the AMA shares a common goal with the FDA to optimize 
the benefit/risk balance of drug therapy and to minimize the risks of 
drug and biological products. The AMA concurs with the FDA that the PI, 
combined with postmarketing surveillance, should constitute the risk 
management plan for the vast majority of drug and biological products. 
The AMA urges the FDA to publish its final rule on the PI and to 
develop a computerized database of PIs that is publicly available.
    The need for a RMP, and the level of the RMP, should be made on a 
case-by-case basis using objective criteria that need to be developed 
by the FDA, in collaboration with the pharmaceutical industry and 
physician organizations. The AMA believes that the vast majority of 
drug or biological products that require an RMP should fall into Level 
2. Again, the AMA supports a collaborative effort among the FDA, the 
pharmaceutical industry, and physician organizations to optimize 
physician education about the risks of drug and biological products 
through identification and implementation of effective methods of risk 
communication.
    The AMA has a number of concerns about Level 3 and Level 4 RMPs and 
recommends that these high-level RMPs be used only as a last resort to 
keep high-risk products with unique and important benefits on the 
market. There needs to be a clear documented need for a high-level RMP 
that is based on objective criteria. Furthermore, the FDA is encouraged 
to use an integrated, systems-based approach to these high-level risk 
management programs to make them more uniform and less intrusive to 
physicians. While evaluation of the effectiveness of RMPs, and of their 
risk management tools, is recommended for all levels of RMPs, this is 
especially important for Level 3 and Level 4 RMPs.
    The AMA also is concerned that the FDA and drug sponsors may be 
attempting to regulate the practice of medicine through some of the 
tools proposed for these high-level risk management programs. It has 
been long established that the FDA is not authorized to control the 
practice of medicine. American Pharmaceutical Association vs. 
Weinberger 377 F. Supp. 824, 829 n. 9 (D.D.C. 1974), aff'd sub nom. 
APhA v. Mathews 530 F.2d 1054 (D.C. Cir 1976).

HOW CAN THE QUALITY OF SPONTANEOUSLY REPORTED CASE REPORTS BE IMPROVED? 
 (from fda concept paper, risk assessment of observational data: good 
    pharmacovigilance practices and pharmacoepidemiologic assessment
    Spontaneous adverse event reports serve an important purpose in 
generating signals about serious adverse events that may be caused by 
drug and biological products. Because physicians are the group best 
able to observe and communicate information about adverse events, the 
AMA has had longstanding policy that physicians have an obligation to 
inform the FDA or product sponsors about potential serious adverse 
events associated with drug and biological products.
    For the above reasons, the AMA has been a proactive MedWatch 
partner since the program's inception. For example, the AMA was a co-
sponsor of one of the first public meetings on MedWatch. Over the 
years, the AMA has also worked with the FDA to educate physicians about 
the importance of voluntary reporting, on what to report, about how to 
make a meaningful report, and how to cooperate fully with follow-up 
calls from sponsors or the FDA. The AMA reaffirms its commitment to the 
MedWatch program and stands ready to work with the FDA and the 
pharmaceutical industry to continue to educate physicians about the 
importance of spontaneous reporting.

                               CONCLUSION

    In conclusion, the AMA appreciates the opportunity to comment on 
the FDA's risk management activities. We hope that our insight into the 
issues discussed in the Concept Papers proves helpful for the FDA, and 
we look forward to working with the Agency as it moves forward in this 
area.
            Sincerely,
                                Michael D. Maves, M.D., MBA

                                 ______
                                 
     Statement of The American Society of Health-System Pharmacists

    The American Society of Health-System Pharmacists (ASHP) is pleased 
to present the United States Senate Health, Education, Labor and 
Pensions Committee with comments on an issue of grave importance--the 
safety of our drug supply. It is essential that the American public 
have confidence in our Nation's drug approval and monitoring systems' 
ability to maintain the integrity of our drug supply and protect 
patients' health.
    For more than 60 years, ASHP has helped pharmacists who practice in 
hospitals and health systems improve medication use and enhance patient 
safety. The Society's 30,000 members include pharmacists and pharmacy 
technicians who practice in inpatient, outpatient, home-care, and long-
term-care settings, as well as pharmacy students.
    ASHP has long taken a leadership role in efforts to improve 
medication safety. Pharmacists are the health care professionals best 
educated and positioned to monitor the safe and appropriate use of 
medications, often serving as the final safety check before medications 
reach the patient. ASHP keeps health-system pharmacists informed of 
drug safety issues and helps them respond appropriately. ASHP publishes 
drug information both for health professionals and consumers, promotes 
evidence based medication use programs, disseminates FDA's MedWatch 
notices, and maintains a drug shortage Web site that informs and offers 
guidance to help pharmacists manage shortages.
    In recent years, drug safety and the FDA approval and monitoring 
process have come under intense scrutiny. With many important new drugs 
entering the market each year, some of which have been fast-tracked 
through the approval process, FDA's ability to monitor safety has been 
questioned. The short duration and small number of participants in the 
clinical drug trials required for FDA-approval dictates that the 
toxicity of new products cannot be fully understood when a drug is 
approved and initially marketed. FDA's postmarketing surveillance, 
therefore, needs to be modernized and strengthened to provide ongoing 
assessment of products on the market. Moreover, the FDA needs 
sufficient resources to fully implement the depth of programs necessary 
to prevent injury and save lives.
    The FDA faces a difficult challenge--establishing a system of drug 
approval and monitoring that maintains a balance between the benefits 
of bringing a new, potentially life-saving drug to market quickly, and 
the risks associated with widespread use of a new drug.
    This testimony will walk through the drug approval and monitoring 
process, examining opportunities to improve safety, while maintaining 
this important balance.

            ISSUES RELATED TO THE FDA DRUG APPROVAL PROCESS

    Under the Federal Food, Drug, and Cosmetic (FD&C) Act, the FDA is 
responsible for ensuring that all new drugs are safe and effective. 
Before any drug is approved for marketing in the United States, the FDA 
must make a determination that the drug is safe and effective for the 
conditions of use in the drug's labeling and that the benefits of 
approval outweigh the drug's risks.
    According to a report by the Inspector General, new drug reviewers 
in the FDA's Center for Drug Evaluation and Research (CDER) have 
experienced shorter approval times and increased pressures to recommend 
approval of a drug even if they have reservations about the drug's 
safety or efficacy.\1\ This pressure is illustrative of the need to 
better educate the American public about both the risks and the 
benefits of new drugs. It also highlights the potential conflict that 
arises by having FDA funding for the drug approval process relying 
heavily on user fees.
---------------------------------------------------------------------------
    \1\ FDA's Review Process for New Drug Applications: A Management 
Review, Inspector General Report March 2003. (OEI-01-01-00590)
---------------------------------------------------------------------------
              ISSUES RELATED TO POSTMARKETING SURVEILLANCE

    Regardless of the rigor of the premarket drug approval process, 
postmarketing surveillance is essential to ensuring drug safety. The 
more widespread, longer-term use of a product in the real world detects 
adverse effects that often go undetected during clinical trials.
    ASHP is pleased that on February 15, 2005, the FDA announced a plan 
to improve the way the FDA manages drug safety information to make 
FDA's review and decision-making processes more independent and 
transparent. However, it is important to review FDA's authority to 
monitor and examine drugs once they are on the market to ensure the FDA 
has sufficient authority to develop an enhanced postmarketing 
surveillance system necessary to meet today's needs.
    Adverse Event Reporting Must be Encouraged. FDA's MedWatch program, 
which provides for the reporting of adverse drug events, is essential 
to detecting enhanced risk associated with medications. ASHP, through 
its Web site, coordinates an effort to provide this information in a 
timely manner to the pharmacy community. However, the MedWatch program 
must be strengthened to encourage more reporting. The information 
gathered from the MedWatch program must be acted upon in a timely 
manner, by someone separate from the team that initially approved the 
drug for marketing. Information must also be made available to patients 
and providers in a timely manner. This may require the FDA being 
granted additional authority to make labeling changes.
    Authority Needed to Require Certain Postmarketing Safety Studies. 
The FDA's ability to measure the ultimate safety of a drug once it has 
entered the market is limited by the fact that the FDA cannot conduct 
independent clinical trials, and it is unclear whether the FDA can 
require manufacturers to conduct such studies. In order for the FDA to 
fully understand side effects of an approved drug that may not have 
surfaced in the limited premarket test group, it is essential that the 
FDA be able to require these studies under certain circumstances.
    Funding Needed for Postmarketing Clinical Effectiveness Studies. 
There is also a significant need for studies comparing the clinical 
effectiveness of medications on the market. Pharmacists, other members 
of the health care team, patients, and private and public payers need 
objective, authoritative, and reliable evidence in order to make the 
best treatment decisions. Such research will contribute to the practice 
of evidence-based patient care, good clinical decision-making, and 
rational drug use. Since prescription drugs represent a significant 
portion of health care costs, the need for such research is 
increasingly important. Only the Federal Government has the ability to 
support such independent comparative research, provide oversight to 
safeguard the integrity of the research process, and disseminate the 
findings.
    We encourage committee members to support expanded funding for the 
Agency for Healthcare Research and Quality (AHRQ) to sponsor this type 
of research. Impartial private sector entities could supplement the 
efforts of AHRQ, but the Federal Government needs to take the lead.
    Clinical Trials Should be Disclosed in National Registry. While 
expanded FDA authority to require postmarketing clinical trials is an 
important start to building a stronger drug monitoring system, it will 
have limited impact if this information is not made available to the 
FDA and the public in some way. Disclosure is essential to creating a 
system of transparency and accountability necessary to promote consumer 
confidence.
    ASHP supports the establishment of a mandatory registry established 
and administered by the Department of Health and Human Services. This 
registry should build upon the existing registry administered by the 
National Institutes of Health for clinical trials dealing with the 
effectiveness of treatments for serious and life-threatening 
conditions, and it should cover all publicly and privately funded 
clinical trials.
    All clinical trials undertaken, but not yet completed, should be 
added to the registry and, upon completion, the results should be 
posted as quickly as possible after FDA approval but before marketing 
commences. Strong enforcement mechanisms are necessary to ensure 
compliance.
    Any opposition raised regarding the disclosure of a company's 
research action plan is outweighed by the public and individual 
patient's right to know and critically examine all available studies 
and their results.
    Additional FDA Funding Needed for Postmarketing Surveillance. 
Additional funding is needed by the FDA, particularly in the area of 
postmarketing surveillance. Due to the fact that the Prescription Drug 
User Fee Act (PDUFA) requires manufacturers to pay a user fee when they 
submit a drug approval application, more resources are available for 
drug approval review than for postmarketing monitoring.
    Such funding is necessary in order for the FDA to conduct 
postmarketing surveillance and establish a national clinical trial 
registry.

          OTHER INITIATIVES ESSENTIAL TO IMPROVING DRUG SAFETY

    Legislation Needed to Encourage Medical Errors Reporting. 
Legislation is needed to help create a culture of safety that would 
entice individuals to report medical errors and ``near misses.'' The 
Senate passed legislation last year, the Patient Safety and Quality 
Improvement Act (S. 720), that would establish a system for reporting 
and analyzing errors reports and establish peer review protections for 
individuals reporting to the system. ASHP is encouraged that the 
committee plans to move that legislation forward this year.
    Congress Should Consider a New Category of Drugs to Help Balance 
Safety and Access Concerns. There is a significant push to make more 
products available over-the-counter. In order to balance access with 
safety concerns, the Congress should consider making the appropriate 
changes in Federal statutes and regulations to establish an 
intermediate category of drug products that do not require a 
prescription but are available only from pharmacists and licensed 
health care professionals who are authorized to prescribe medications.
    Pharmacists, who have the education, training, and expertise to 
help patients make appropriate therapeutic decisions, would be able to 
provide drugs in this new category directly to patients without a 
prescription, on the basis of appropriate assessment and professional 
consultation. This would enhance patient access, while addressing 
safety concerns that prevent drugs from being dispensed over the 
counter.
    Current Drug Importation Laws Should be Enforced Vigorously Until A 
System Can be Established to Maintain Current FDA Assurance of the 
Safety and Authenticity. Working outside of the FDA's regulatory 
framework to import drugs from other countries is counterproductive to 
efforts to strengthen FDA's drug approval and monitoring processes. 
Current importation efforts increase the risk that Americans will 
receive drugs that do not meet the FDA's standards and are harmed as a 
result.
    ASHP believes that current laws and regulations related to 
importation should be upheld and vigorously enforced until in order to 
(1) maintain the integrity of the pharmaceutical supply chain to avoid 
the introduction of counterfeit products into the United States; (2) 
provide for continued patient access to pharmacist review of all 
medications and preserves the patient-pharmacist-prescriber 
relationship; and (3) provide adequate patient counseling and 
education, particularly to patients taking multiple high-risk 
medications.
    Before any consideration is given to opening the United States 
market to medications from abroad, systems should be put in place to 
guarantee the integrity of any new distribution networks. Related to 
this point, ASHP encourages stronger authority for the FDA and others 
to control the prescribing and dispensing of medications via the 
Internet. ASHP supports efforts that require pharmacy World Wide Web 
sites to list the States in which the pharmacy and pharmacists are 
licensed, and, if prescribing services are offered, requires that the 
sites (1) ensure that a legitimate patient-prescriber relationship 
exists (consistent with professional practice standards) and (2) list 
the States in which the prescribers are licensed.
    FDA Should Be Given Broader Authority to Notify Providers of Drug 
Product Shortages. ASHP members and other health care providers have 
increasingly experienced drug shortages. These shortages not only 
affect access to care but, also due to the limited notice providers 
receive of impending shortages, increase the cost of alternative care 
and the likelihood of medication-related complications. ASHP strongly 
believes that the Congress and the FDA should consider expanding the 
definition of ``medically necessary'' drug products to enhance FDA's 
authority to require pharmaceutical manufacturers to notify the 
appropriate government body well in advance of voluntarily 
discontinuing a product and put in place effective sanctions for 
manufacturers that do not comply with this mandate.
    Direct-to-Consumer Advertising of Specific Drugs Should be 
Prohibited. Direct-to-Consumer (DTC) advertising has more than doubled 
over the last 5 years. Despite this dramatic expansion in advertising, 
FDA enforcement actions against ads that are in violation of FDA 
standards have dropped. While the FDA can require DTC ads to be 
scientifically accurate and provide a fair balance of risks and 
benefits, the FDA lacks the necessary resources to assure that 
companies comply.
    ASHP is concerned about the impact DTC advertising for specific 
drug products has had on fostering inappropriate prescribing and 
supports a ban on such advertisements.
    Drug Samples Should be More Carefully Regulated. ASHP also believes 
that addressing drug safety is incomplete without considering safety 
concerns that arise due to manufacturer samples that go through 
distribution channels that (1) do not foster pharmacist oversight of 
therapy, (2) result in poor drug control, allowing patients to receive 
improperly labeled and packaged, deteriorated, out-dated, and 
unrecorded drugs, (3) provide access to prescription drugs by 
unauthorized, untrained personnel, (4) may encourage inappropriate 
prescribing habits, or (5) may increase the cost of treatment for all 
patients.
    FDA should be encouraged to provide the additional guidance 
necessary to ensure drug samples are distributed through channels 
meeting these criteria.
    Manufacturers Should be Required to Make Available Unit Dose 
Packaging of Medications Commonly Dispensed in Hospital and Ambulatory 
Health Care Settings. ASHP urges the FDA to require manufacturers to 
provide all medications commonly used in hospitals and other ambulatory 
health care settings in ready-to-use unit dose packaging. The FDA 
issued a final regulation in February 2004 that requires pharmaceutical 
manufacturers to apply bar codes to ``most prescription drugs'' and 
``certain over-the-counter drugs that are commonly used in hospitals 
and dispensed pursuant to'' a medication order. Currently, many drugs 
are not available from manufacturers packaged as a single dose ready 
for dispensing. In order for bar coding to have the greatest effect on 
patient safety, bar coded packaging of medications must be available at 
the unit-of-use level. Lack of availability of appropriate dosage forms 
packaged for unit dose dispensing places the burden on hospital 
pharmacy departments, who are not well situated to take on this role. 
In the interest of patient safety, manufacturers should be required to 
provide medications in dosage packaging commonly used in hospitals and 
ambulatory health system settings.
    Concluding Comments. ASHP is encouraged by the reasoned approach 
the committee has taken to addressing drug safety concerns. The U.S. 
has a solid record for drug safety that should not be overlooked. The 
system, however, will benefit from a careful examination and 
enhancements to address current weaknesses. We look forward to working 
with the committee to develop balanced solutions to enhancing safe 
medication use and moving any necessary legislation forward.
                                 ______
                                 
                 Prepared Statement of Senator Grassley

    Chairman Enzi, I congratulate you on becoming chairman of the 
Health, Education, Labor, and Pensions (HELP) Committee, and thank you 
for your leadership in holding hearings on the Food and Drug 
Administration (FDA) and drug safety. As you know, these issues have 
been a central concern of mine during the past year. My staff on the 
Finance Committee has been investigating serious allegations raised by 
whistleblowers that call into question whether the FDA is fulfilling 
its mission to protect the health and safety of Americans.
    Indeed, the Food and Drug Administration has failed to put patient 
safety first with respect to both SSRIs and COX-2 drugs. Last November, 
the Finance Committee held an oversight hearing based on an 
investigation of the withdrawal of Vioxx, Merck's blockbuster COX-2 
drug. Red flags had been raised about the safety risks of Vioxx before 
and after the drug had been approved by the FDA. The Vioxx hearing shed 
some much needed light on how the Food and Drug Administration 
regulated or rather failed to regulate Vioxx effectively.
    The Finance Committee has a responsibility to more than 80 million 
Americans who receive health care coverage, including prescription 
drugs, under the Medicare and Medicaid programs. As Chairman of the 
Finance Committee, Merck's withdrawal of Vioxx was of particular 
interest because Medicaid paid over $1 billion for Vioxx while it was 
on the market. Medicare and Medicaid beneficiaries rely on medicines 
paid for with federal funds. The Finance Committee has a responsibility 
to make sure that every federal dollar paid for drugs is not misspent 
on unsafe drugs and drug companies who profit at the expense of 
consumer safety.
    Historically, the Food and Drug Administration has met its charge 
to protect the health and safety of the American public. Those who work 
at the agency are, by and large, committed to doing no harm. Even so, 
the FDA has also stood watch over failures when it comes to drug safety 
this past year. Consequently, public confidence has been shaken.
    When the FDA approves a drug, it's considered a Good Housekeeping 
seal of approval. Consumers should not have to second guess the safety 
of what's in their medicine cabinets. When the FDA approves a drug, 
Americans should be able to bank on its benefits outweighing its risks. 
If a drug presents an unacceptable risk, the FDA should take it off the 
market. This risk-benefit analysis should be non-negotiable. When it 
comesto putting patient safety first there is no room at the table for 
drug companies. The FDA should not be sitting down to negotiate with 
drug companies whose priorities too often appear to lie with their 
stockholders.A vital and pressing concern today is post-marketing 
surveillance and the way FDA monitors the safety of prescription drugs. 
Dr. Raymond Woolsey, a witness at today's hearing, stated in a 
Frontline interview in November 2003 that ``the number of people hired 
at the [FDA] to protect, to analyze data and drug safety, is criminal. 
. . The teams that are needed to do drug safety are infinitely more 
than what they've got right now. We don't have a safety system in this 
country.''
    One of my concerns is that the FDA has a relationship with drug 
companies that is far too cozy. That's exactly the opposite of what it 
should be. Despite findings from a Merck study that heart attacks were 
five times higher for Vioxx patients than for patients on another drug, 
nearly two years passed before label changes were made by the FDA. 
Consumers and doctors remained largely unaware of the cardiovascular 
risks while Merck continued to aggressively market Vioxx during that 
time. The overriding concern of the FDA should have been the health and 
safety of the American people.
    Evaluating drug safety, of course, involves balancing the risks 
against the benefits of each drug, and Vioxx is no different, but we 
need to know what the risks are in order to make those risk-benefit 
calculations. And in the case of Vioxx, doctors and patients did not 
have that opportunity. What I also find disturbing is that Merck 
negotiated with the FDA to place information about the cardiovascular 
risks of Vioxx in the "Precautions" section of the label rather than 
prominently displaying it as a "Warning."
    Several witnesses at the Finance Committee's November hearing 
believed the FDA should have required a black box warning for Vioxx, 
the strongest label warning the FDA requires. FDA's own advisory 
committees recently agreed. Less than two weeks ago, the FDA joint 
advisory committee recommended black box warnings on all COX-2 labels.
    FDA has also disregarded and downplayed important concerns and 
warnings from its own best scientists. We saw evidence of that in the 
way FDA treated Dr. Andrew Mosholder's findings on SSRIs and Dr. David 
Graham's findings on Vioxx. The FDA even attempted to undermine the 
publication of Dr. Graham's findings in The Lancet. According to Dr. 
Graham's peer reviewed and published findings, an estimated 88,000-
140,000 excess cases of serious coronary heart disease are attributable 
to Vioxx, with about half these cases being fatal. That means there may 
be as many deaths attributable to Vioxx as the number of soldiers who 
died during the Vietnam War. Information about the Vioxx disaster needs 
to be shared with the public and shared in a timely manner.
    This week I introduced the Fair Access to Clinical Trials Act of 
2005 with Senator Dodd. I introduced this legislation as part of a 
sustained effort to restore public confidence in the federal 
government's food and drug safety agency. The FACT Act would expand 
www.clinicaltrials.gov to create a publicly accessible national data 
bank of clinical trial information comprised of a clinical trial 
registry and a clinical trial results database.Enactment of this bill 
would be a meaningful step toward greater transparency and 
accountability in clinical trials and the scientific process.
    In addition, Senator Dodd and I are working on a bill to establish 
an independent office for drug safety within the FDA. The independence 
of the office would not exist solely on an organizational chart. The 
office would have an independent director and regulatory authority. 
When it comes to drug safety, intra-agency community is indeed 
essential. However, the Office of New Drugs is hampered by real and 
perceived conflicts of interest. An independent drug safety office 
would more effectively regulate drugs post market. It doesn't make 
sense, from an accountability standpoint, to have the office that 
reviews the safety of drugs that are already on the market to be under 
the thumb of the office that puts the drugs on the market in the first 
place.
    As I continue with my Constitutional duties to conduct oversight as 
Chairman of the Finance Committee, I look forward to working closely 
with you, Mr. Chairman, to ensure that critical changes are made within 
the FDA to keep the agency focused on its mission to protect public 
health and safety. I look to your leadership and seek your support with 
the legislation that is necessary to help get FDA back on the right 
track. Again, I commend you for holding these hearings and look forward 
to working with you.

                                 ______
                                 
     Response to Questions of Senator Enzi by Janet Woodcock, M.D.
Hon. Mike Enzi,
Chairman,
Committee on Health, Education, Labor, and Pensions,
U.S. Senate,
Washington, D.C. 20510-6300.

    Dear Chairman Enzi: Thank you for the facsimiles dated March 21, 
2005, including questions for the record related to the Committee's 
recent hearings, March 1 and 3, 2005, entitled, ``FDA's Drug Approval 
Process: Up to the Challenge?'' We have repeated your questions below, 
followed by the Food and Drug Administration's (FDA or the Agency) 
response.

    Question 1. Some witnesses at our Tuesday hearing emphasized that 
clinical trials, because of their size and length, cannot always 
predict fully the potential side effects of a drug. Would you explain 
this in more detail?
    Answer 1. The most recent actions concerning the drug Vioxx 
(rofecoxib) illustrate the vital importance of the ongoing assessment 
of the safety of a product once it is in widespread use. FDA grants 
approval to drugs after a sponsor demonstrates that they are safe and 
effective for a given indication. Experience has shown that the full 
magnitude of some potential risks do not always emerge during the 
mandatory clinical trials conducted before approval to evaluate these 
products for safety and effectiveness. A new drug application (NDA) 
typically includes safety data on several hundred to several thousand 
patients. If a serious adverse event occurs in 1 in 5,000 or even 1 in 
1,000 users, it might not occur in clinical trials of this size. When 
the drug is used by many times that number of patients, that event 
could show up as a serious risk. Occasionally, serious adverse effects 
are identified after approval either in post-marketing clinical trials 
or through spontaneous reporting of adverse events. That is why 
Congress has supported, and FDA has created, a strong post-market drug 
safety program designed to assess adverse events identified after 
approval for all of the medical products it regulates as a complement 
to the pre-market safety reviews required for approval of prescription 
drugs in the United States.

    Question 2. What technologies and processes are currently available 
to predict potential adverse events during drug development and/or 
identify post-marketing safety issues?
    Answer 2. During the Pre-Market Phase, ODS works with OND at pre-
NDA and pre-Biologics Licensing Applications (BLA) meetings with 
industry to review safety information and to discuss proposed risk 
management plans and the need for any post-approval risk management 
studies. During NDA and BLA review, ODS and OND work together in the 
development and review of risk management programs. ODS provides 
expertise in the review of proposed proprietary drug names, labeling, 
and packaging to minimize medication errors, patient labeling and 
Medication Guides, and Phase 4 safety studies. ODS staff are involved 
in the preparation for and may present information at advisory 
committee meetings involving safety issues and risk management for 
pending applications and when post-marketing safety information is 
available for similar products.
    Under Title 21, Code of Federal Regulations (CFR) 314.80, Post-
Marketing Reporting of Adverse Drug Experiences, subsections (b) 
(review of adverse drug experiences) and (c) (reporting requirements), 
manufacturers are required to review and report to FDA all adverse drug 
experience information obtained or otherwise received by the applicant 
from any source, foreign or domestic. This includes information derived 
from commercial marketing experience, post-marketing clinical 
investigations, post-marketing epidemiological/ surveillance studies, 
and reports in the scientific literature and unpublished scientific 
papers. There is comparable language in 21 CFR 600.80 for biologics.
    FDA recently published a proposed rule that would require drug 
manufacturers to submit reports electronically. The rule, if finalized, 
would help harmonize worldwide reporting of post-marketing safety 
information and expedite detection of safety problems for marketed 
drugs. The Agency is also expanding the reporting requirements for 
manufacturers of biological products to include adverse event reports 
from unlicensed blood banks and transfusion services.
    In concert with industry's reporting requirements, the Food and 
Drug Administration Modernization Act of 1997 (FDAMA) requires sponsors 
of approved drugs and biological products report to FDA annually on the 
progress of their post-marketing study commitments, both those that are 
required and those that are agreed upon in writing. Under FDAMA, FDA is 
obligated to track the progress of post-marketing study commitments, 
make certain information about commitments available to the public, and 
to report annually in the Federal Register on the performance of post-
marketing study commitments (PMC). This tracking and reporting allows 
for FDA to monitor compliance of PMCs. The status of PMCs is published 
on the CDER website at http://www.fda.gov/cder/pmc/default.htm.
    During the Post-Market Phase, one of ODS' primary roles is to 
provide expertise in the review of post-marketing safety data and to 
maintain and coordinate CDER's post-marketing surveillance and risk 
assessment program. This program includes the Adverse Event Reporting 
System (AERS), a computerized information database designed to support 
the FDA's post-marketing safety surveillance program for all approved 
drug and therapeutic biologic products. Information in AERS comes from 
required reporting by companies and through voluntary reports submitted 
directly to FDA's MedWatch program by consumers and health 
professionals, which together total more than 350,000 reports per year. 
ODS review of AERS data may provide signals of safety issues that did 
not appear during the drug development process as well as those that 
appear more frequently or with a greater degree of severity after 
approval than was seen in clinical trials. ODS review of AERS reports 
may also detect product quality problems that are referred to the 
Office of Compliance.
    Individual reports may trigger further evaluation of similar 
reports in the AERS database and could signal important safety concerns 
prompting regulatory actions both in the U.S. and abroad. To further 
investigate safety signals, ODS safety evaluators, epidemiologists and 
drug utilization specialists perform research to define drug use, 
background rates of the event in the treated population, and 
epidemiological trends.
    In addition to AERS, ODS staff are responsible for the acquisition, 
analysis, and interpretation of information from contracted databases 
on drug use in various populations, including in-patients, children, 
and patients over time that help place safety signals into context and 
inform regulatory decision-making. For newly approved products with 
important safety concerns, ODS independently evaluates product 
utilization to evaluate whether these products are being used in a safe 
manner and works collaboratively and pro-actively with OND and industry 
on related issues.
    ODS' cooperative agreement program in pharmacoepidemiology provides 
CDER with access to external experts with access to population-based 
databases for the purpose of studying important post-marketing drug 
safety questions. CDER works collaboratively with the cooperative 
agreement partners to identify research areas and to design and conduct 
studies to investigate suspected associations between specific drug 
exposures and specific adverse events and to estimate risk. FDA is 
revising this program to use contracts, rather than cooperative 
agreements, to help focus on drug safety issues that are of the highest 
priority and urgency to the Agency.
    ODS also uses additional data sources as needed, such as the 
National Electronic Injury Surveillance System: All Injury Program 
(NEISS-AIP), an ongoing active surveillance system for the purpose of 
collecting data on all injuries presented to a probability sample of 
U.S. hospital emergency departments; the Drug Abuse Warning Network, a 
public health surveillance system that monitors drug-related visits to 
hospital emergency departments and drug-related deaths investigated by 
medical examiners and coroners.
    ODS reviews reports of medication errors that have occurred with 
marketed products and recommends changes to product names, labeling, 
and/or packaging to prevent future errors. ODS works with OND and the 
Office of Generic Drugs to review risk management programs for approved 
products to assess their implementation and effectiveness. ODS reviews 
patient labeling and Medication Guides that are put in place to address 
serious and significant public health concerns both pre-marketing and 
when issues arise after a product is marketed. ODS assists the Office 
of Training and Communication in the development of Consumer Drug 
Information Sheets by assessing readability.
    ODS' MedWatch program is an important tool in both acquiring and 
disseminating safety information. In addition to receiving direct 
reports of serious adverse events and problems related to drugs and 
other medical products regulated by FDA from consumers and health 
professionals, the MedWatch program also provides important and timely 
clinical information on safety issues involving medical products, 
including prescription and over-the-counter drugs, biologics, medical 
and radiation-emitting devices, and special nutritional products (e.g., 
medical foods, dietary supplements and infant formulas). Medical 
product safety alerts, recalls, withdrawals, and important labeling 
changes that may affect the health of all Americans are disseminated to 
the medical community and the general public via the MedWatch website 
and the MedWatch E-list that provides e-mail updates to over 46,000 
subscribers. MedWatch Partners are over 150 health care professional 
organizations, consumer groups, and web-media groups that work with FDA 
to help keep their members informed about medical product safety 
information and reporting. Partners are encouraged to play an active 
role in post-marketing surveillance.
    FDA created an independent Drug Safety Oversight Board (DSB or the 
Board) to enhance the independence of internal deliberations and 
decisions regarding risk/benefit analyses and consumer safety. The DSB 
will oversee the management of important drug safety issues within the 
CDER. The DSB will be comprised of members from FDA and medical experts 
from other HHS agencies and government departments (e.g., Department of 
Veterans Affairs). The Board also will consult with other medical 
experts and representatives of patient and consumer groups. CDER's 
Manual of Policies and Procedures has been updated to reflect the 
organization of the DSB; you may view this document by visiting: http:/
/www.fda.gov/cder/mapp/4151-3.pdf.
    FDA recently asked the Institute of Medicine (IOM) to look at the 
structure of our post-marketing surveillance program and to give FDA 
their expert advice on whether additional changes are needed to the 
Agency's approach to drug safety, which could include recommendations 
for changes in the CDER's organizational structure.
                                 ______
                                 
     Response to Questions of Senator Hatch by Janet Woodcock, M.D.

    Question 1. I believe it is important for the FDA to look for ways 
to address differences of opinion within the agency. I think that 
[when] FDA officials have conflicting messages about the safety of 
specific drugs, it is extremely confusing to the general public. How 
would this program address disagreements among FDA scientists regarding 
the safety of a specific drug? Would this program have the FDA speak 
with one voice or would the public be given the opportunity to review 
the concerns of all FDA scientists? How would the new independent Drug 
Safety Oversight Board interact with the individuals involved with this 
program?
    Answer 1. In November 2004, Dr. Crawford announced that CDER was 
establishing a new ``Differing Professional Opinions and Dispute 
Resolution'' Program. For more information about this program, please 
visit: http://www.fda.gov/bbs/topics/news/2004/new01131.html. CDER has 
developed a Manual of Policies and Procedures describing how this 
dispute resolution process will be managed. This process is available 
to any individuals in the Center who wish to express their differing 
professional opinions (DPOs) concerning any regulatory actions or 
policy decisions with significant public health impacts in instances 
when the normal procedures for resolving internal disputes are not 
sufficient.
    In most cases, free and open discussion of scientific issues among 
review teams, and with supervisors, managers, and external advisors 
leads to an agreed course of action. Sometimes, however, a consensus 
decision cannot be reached, and an employee may feel his or her opinion 
was not adequately considered. Such disagreements can have a 
potentially significant public health impact. That is why CDER's new 
program provides for a review of the involved differing professional 
opinions by FDA and outside experts. An ad hoc panel, whose members 
were not directly involved in the disputed decisions, will have 30 days 
to review all relevant materials and recommend to the Center Director 
an appropriate course of action.
    In addition to this program for resolving individual disputes 
regarding any regulatory matters or policies, CDER is establishing the 
Drug Safety Oversight Board (DSB) to specifically address drug safety 
issues and to assist in the resolution of disputes and differing 
professional opinions between staff from the Office of New Drugs and 
the Office of Drug Safety within the Center for Drugs. The membership 
of the Board has been carefully balanced with equal representation from 
both the Office of New Drugs and the Office of Drug Safety, along with 
representatives from other offices within CDER, the Center for Biologic 
Evaluation and Research, the Center for Devices and Radiological 
Health, the National Cancer Institute and the Department of Veterans 
Affairs. These organizational disputes will be addressed during 
meetings of the Board, with a final recommendation reached through 
achievement of consensus or through voting of the Board if no consensus 
can be reached.

    Question 2. I commend the FDA on its willingness to increase 
communication with the public on drug safety. I believe that will make 
a significant difference for both physicians and their patients when 
they are making treatment decisions. Besides the internet, how will 
this information be provided to the public so they may be able to 
access information about pharmaceutical products?
    Answer 2. As part of the vision announced on February 15, 2005, by 
the Department of Health and Human Service's Secretary Leavitt and 
FDA's Acting Commissioner Crawford, FDA will create a new independent 
DSB to oversee the management of drug safety issues. The Agency plans 
to improve transparency by providing emerging information to health 
providers and patients about the risks and benefits of medicines.
    We understand that many Americans still do not have access to the 
Internet and that Internet communication is just one form of 
communication. In addition to using the Internet as a communication 
medium, FDA also uses trade and other press as well as other forms of 
external communication to convey safety information. We routinely use 
print media (periodicals) and coordinate with the press to publicize 
information. We will continue to use magazine advertisements, press 
releases, and other campaigns to inform the public.
    FDA uses a very important means of communication, FDA-approved 
patient labeling, which includes:
    Patient Package Inserts. For some prescription medicines, FDA 
approves special patient materials to instruct patients about the safe 
use of the product. These materials may be given to patients by their 
health care provider or pharmacist, and are considered part of FDA-
regulated product labeling.
    Medication Guides. FDA may require distribution of Medication 
Guides, FDA-approved patient information, for selected prescription 
drugs that pose a serious and significant public health concern. 
Medication Guides will be required if FDA determines that one or more 
of the following circumstances exist:
    Patient labeling could help prevent serious adverse effects;
    The drug product has serious risk(s) (relative to benefits) of 
which patients should be made aware because information concerning the 
risk(s) could affect patients' decision to use, or to continue to use, 
the product;
    The drug product is important to health and patient adherence to 
directions for use is crucial to the drug's effectiveness.
    In addition, FDA will be creating a Drug Watch Web Page, which will 
include emerging information for both previously and newly approved 
drugs about possible serious side effects or other safety risks that 
have the potential to alter the benefit/risk analysis of a drug, affect 
patient selection or monitoring decisions, or that can be avoided 
through measures taken to prevent or mitigate harm. FDA has recently 
issued a draft guidance entitled; ``FDA's `Drug Watch' for Emerging 
Drug Safety Information,'' which articulates the Agency's current 
thinking on the topic. This draft guidance is open for public comment 
and may be viewed by visiting: http://www.fda.gov/cder/guidance/
6657dft.pdf. Other new communication channels will also include:
    Health care Professional Information Sheets. One-page information 
sheets for health care professionals for all drugs on FDA's Drug Watch 
and all drugs with Medication Guides (FDA-approved patient labeling) 
containing the most important new information for safe and effective 
product use, such as known and potential safety issues based on reports 
of adverse events, new information that may affect prescribing of the 
drug, and the approved indications and benefits of the drug.
    Patient Information Sheets. One-page information sheets for 
patients containing new safety information as well as basic information 
about how to use the drug in a consumer friendly format for all 
products on Drug Watch.
    All of this information will be provided on the Internet.
                                 ______
                                 
     Response to Questions of Senator Gregg by Janet Woodcock, M.D.

    Question 1. During the hearings, when questioned about whether the 
FDA needed additional authority to require label or labeling changes on 
prescription drug products it appeared that Dr. Sandra Kweder and Dr. 
Janet Woodcock--the FDA witnesses at each of the hearings--responded 
differently to the question. Does FDA have adequate authority for the 
drug approval and postmarket surveillance processes? Does FDA need any 
additional authority to require label or labeling changes on drug 
products, to require phase IV clinical trials, or to withdraw marketed 
drug products? Does FDA need any additional authority to ensure the 
safety and efficacy of new and marketed drugs?
    Answer 1. We do not believe new statutory authority is needed. We 
will use all existing regulatory authority and enforcement powers when 
negotiating label changes with drug companies or when monitoring or 
managing drug safety issues. In most cases, FDA and the sponsor are 
able to reach agreement on the labeling text fairly quickly (a few 
weeks). As Dr. Janet Woodcock testified on March 3, 2005, a key factor 
in labeling changes is that once a label change is made, old labels in 
paper form are still in distribution and it takes time to get newer 
labels into circulation. Dr. Woodcock testified that the new strategy 
of posting drug safety information sooner using the Drug Watch 
mechanism will help alleviate that factor because it will enable FDA to 
get information directly to the people who need it in a timely manner.
    In addition to the Drug Watch web page, our February 15, 2005, 
announcement included plans to create a new Drug Safety Oversight Board 
(DSB) to provide independent oversight and advice on the management of 
important drug safety issues and to manage the dissemination of certain 
safety information through FDA's website to health care professionals 
and patients. For more information on this initiative, please visit: 
http://www.fda.gov/oc/factsheets/drugsafety.html. Also, FDA is 
intensifying our current efforts to provide the public with the most 
important information for the safe and effective use of drugs in 
patient-friendly language. Two tools, Patient Information Sheets and 
Health care Information Sheets, will allow FDA to deliver emerging 
safety information to patients and health care providers.
    To carry out these enhancements, the Agency's fiscal year 2006 
budget request includes an increase of $5 million for the Office of 
Drug Safety, bringing total funding to $22.9 million (a nearly 25 
percent increase).

    Question 2. Concerning the recent withdrawal of marketed drugs, I 
understand that 12 of the 17 drugs taken off the market were used in 
ways that were unsafe. What drugs were they? How were they used? Was 
this use off-label? Did the companies involved withdraw the drug 
products from the market or did FDA require the withdrawals?
    Answer 2. We are providing a list below of safety-based drug 
withdrawals associated with unsafe labeled or off-label use of the drug 
product. All of these products were voluntarily withdrawn by the 
manufacturer. Those drugs for which the company made the withdrawal 
decision independent of FDA advice are indicated with an asterisk. In 
all other cases, the decision was made jointly by FDA and the company 
or by the company upon FDA recommendation.
    Propulsid (cisapride), a drug to treat gastroesophageal symptoms, 
was withdrawn in 2000. It was withdrawn due to adverse events 
associated with labeled, contraindicated usage (concomitant use of 
medications, which inhibit a certain metabolic pathway), which 
continued despite the fact that the label included a boxed warning 
about this interaction and several ``Dear Healthcare Provider'' letters 
were sent out to practicing physicians as reminders.
    Lotronex (alosetron), a drug to treat Irritable Bowel Syndrome in 
women, was withdrawn in 2000 and reintroduced in 2002 under a 
restricted distribution program. Lotronex was withdrawn because of the 
occurrence of ischemic colitis in patients taking the drug, many of 
whom were prescribed the drug for unapproved indications.
    Duract (bromphenac), a drug to treat acute pain, was withdrawn in 
1998. The drug was found to have a very high rate of liver toxicity 
when used for longer than a few weeks. Despite several label changes, 
including a boxed warning and several ``Dear Health Care Provider'' 
letters advising use of the drug for no more than two weeks, cases of 
severe liver toxicity continued to be reported in patients taking the 
drug for prolonged periods.
    Seldane (terfenadine), a drug to treat seasonal allergic rhinitis, 
withdrawn in 1998. Seldane caused an abnormal prolongation of heart 
electrical pathways when used in combination with other drugs that 
inhibited certain metabolic pathways. Despite a boxed warning and other 
measures to educate prescribers, reports of serious cardiovascular 
events, including death, continued due to Seldane and contraindicated 
concomitant medicines.
    *Hismanal (astemizole), a drug to treat seasonal allergic rhinitis, 
was withdrawn in 1999. This drug was withdrawn under the same set of 
circumstances that Seldane had been several months before.
    Mibefradil, a drug to lower blood pressure, was withdrawn in 1998 
because it caused a potentially fatal heart rhythm disturbance when 
used in combination with some other drugs. Warnings and other 
notifications were not successful in mitigating such use.
    Voluntarily withdrawn from the market by the manufacturer, but not 
due to use inconsistent with the label:
    Bextra (valdecoxib), a drug to treat acute and chronic pain, was 
withdrawn in 2005. It was withdrawn due to post-marketing data 
regarding risk of CV safety associated with it and other anti-
inflammatory drugs, and a very high rate of serious skin reactions that 
is not shared by other similar drugs.
    *Vioxx (rofecoxib), a drug to treat acute and chronic pain, was 
withdrawn in 2004. It was withdrawn due to concerns of an increased 
risk of CV events.
    Baycol (cerivastatin), a drug to treat high cholesterol, was 
withdrawn in 2001. It was withdrawn due to reports of rhabdomyolysis 
(severe and potentially serious muscle toxicity) that was found to be 
far more common with Baycol than other similar drugs.
    *Raplon (rapacuronium), a drug used to induce anesthesia, was 
withdrawn in 2001 due to reports of serious bronchospasm associated 
with its use.
    Phenylpropanolamine, a nasal decongestant, was withdrawn in 2000. 
Never approved by FDA, this very old drug was voluntarily removed from 
numerous prescription and non-prescription products by manufacturers, 
due to concerns that it might cause strokes.
    Rezulin (troglitazone), a drug to treat diabetes, was withdrawn in 
2000 at the time of introduction of newly approved products that could 
serve as less toxic, adequate substitutes. Rezulin included labeling 
warning of liver toxicity and the need for routine monitoring of 
patients. Several ``Dear Healthcare Provider'' letters about this 
toxicity and need for monitoring were issued.
    *Rexar (grepafloxacin), an antibiotic, was withdrawn in 1999. It 
was withdrawn by the company after the manufacturer observed a small 
number of severe CV events.

    Question 3. For several years, FDA has been implementing an 
initiative to make product labels easier to use by consumers. For 
example, FDA issued regulations to require that nonprescription drugs 
carry clear, simple and readable labeling. FDA took this action to make 
it easier for consumers to understand information about OTC drug 
products, including the benefits and risks, and how the drugs should be 
used most effectively. Does FDA intend to review prescription drug 
labeling to see if it is possible to make it easier for consumers and 
caregivers to find and understand important information about the 
products?
    Answer 3. FDA agrees that the current package insert format is 
inadequate. Therefore, we have embarked on a major initiative to 
improve it. In recent years, there has been an increase in the length, 
detail and complexity of prescription drug labeling, making it harder 
for health care practitioners to find specific information and to 
discern the most critical information in product labeling. In the 
Federal Register of December 22, 2000, (65 FR 81082) FDA issued a 
proposed rule to revise its regulations governing the content and 
format of labeling for human prescription drug products. Prior to 
issuing the proposal, the Agency evaluated the usefulness of 
prescription drug labeling for its principal audience to determine 
whether, and how, its content and format could be improved. The Agency 
used focus groups, a national physician survey, a public meeting and 
written comments to develop multiple prototypes and to ascertain how 
prescription drug labeling is used by health care practitioners, what 
labeling information practitioners consider most important, and how 
practitioners believed labeling could be improved. The Agency developed 
a prototype based on this accumulated information as the model for the 
proposed rule. FDA received many comments on the proposed rule and is 
working to finalize it in the near future. Publication of this rule 
will be accompanied by publication of four implementing guidance 
documents.
                                 ______
                                 
    Response to Questions of Senator Kennedy by Janet Woodcock, M.D.

    Question 1. In your testimony, you and Dr. Kweder described how the 
agency approves drugs under the statutory ``safe and effective'' 
standard, which the agency applies by comparing the known benefits of 
the drug against the known risks and assessing whether the benefits 
exceed the risks. Dr. Kweder and you explained, for example, how the 
agency assessed benefit with respect to the disease or condition to be 
treated, so that treating a tension headache, which goes away on its 
own and is in no sense life-threatening, is considered to offer 
considerably less benefit than treating a life-threatening cancer, and 
the risks that may be tolerable are considerably higher.
    Under section 515 of the statute, the agency approves pre-market 
approval applications for medical devices if there is reasonable 
assurance of safety and effectiveness. Does application of this 
standard also involve comparing benefit and risk? If so, please explain 
how the application is similar to, and differs from, the risk-benefit 
assessment that the agency performs for a drug? If not, why not? Please 
explain in detail, and with examples, how the agency applies the 
standard of reasonable assurance of safety and effectiveness.
    Answer 1. FDA wants to emphasize that all drugs and devices, 
regardless of their approval mechanism, must have benefits that 
outweigh risks.
    For devices going through the pre-market approval application 
process, there is a reasonable assurance that a device is safe when it 
can be determined, based on valid scientific evidence, that the 
probable benefits to health from use of the device for its intended 
uses and conditions of use, when accompanied by adequate directions and 
warnings against unsafe use, outweigh any probable risks (21 CFR 
860.7(d)(1)). There is a reasonable assurance that a device is 
effective when it can be determined, based on valid scientific 
evidence, that in a significant portion of the target population, the 
use of the device for its intended uses and conditions of use, when 
accompanied by adequate directions for use and warnings against unsafe 
use, will provide clinically significant results (21 860.7(e)(1)).
    For devices, the mechanisms of assessing the benefits and the risks 
are analogous to drugs. Assurance of safety is based upon 
investigations using laboratory animals, investigations involving human 
subjects, and non-clinical investigations including in vitro studies 
(21 CFR 860.7(d)(2)). The valid scientific evidence to determine a 
reasonable assurance of effectiveness of a device shall consist 
principally of well controlled investigations (21 CFR 860.7(e)(2)). The 
device regulations also allow for the consideration of other types of 
studies as well (21 CFR 860.7(e)(2)).
    The Center for Devices and Radiological Health (CDRH) similarly 
assesses benefit with respect to the disease or condition to be 
treated. For example, a device that uses shockwave therapy to treat a 
pain associated with tendonitis, which may be treated with physical 
therapy and is in no sense life-threatening, is considered to offer 
less benefit than devices used to treat a life-threatening condition, 
such as replacing a faulty heart valve. Further, the risks that would 
be considered acceptable will be higher in the latter case.
    FDA's analysis of VIOXX was based on our assessment of the 
available data, the recommendations from the advisory committee, and 
our best judgment of the potential benefits of the drugs compared to 
the potential risks of the drug when used according to the 
recommendations included in the revised labeling.

    Question 2a. In the VIGOR trial, compared with naproxen, Vioxx at 
50 mg increased the risk of heart attacks by a factor of 5. At that 
dose, the approved indication was the short-term treatment of acute 
pain. Do these risk-benefit analyses still seem reasonable to you?
    Answer 2a. FDA carefully considered all the CV findings from the 
VIGOR study, and available data from other trials, in assessing the 
impact of the VIGOR data on the continued safe use of the drug. FDA 
presented the results of the VIGOR study at a public Arthritis Advisory 
Committee meeting on February 8, 2001. The GI, CV, and general safety 
results of the VIGOR study were presented and discussed extensively. 
The expert panel, which included two cardiologists, recommended that 
both the positive GI information (reduced risk of serious 
gastrointestinal bleeding versus naproxen) as well as the potential 
increased risk of CV events (compared to naproxen) be included in the 
label. The panel did not recommend withdrawal of the 50 mg dose from 
the market.
    Vioxx 50 mg was originally approved only for the short-term 
management of acute pain. Based on the data available to FDA at that 
time, we concluded that the potential risk of short-term use of VIOXX 
50 mg did not outweigh the potential benefits. FDA did however, 
implement changes to the VIOXX labeling that specifically stated that 
the prolonged use of the 50 mg dose was not recommended and that the 
maximum recommended dose for prolonged use in osteoarthritis and 
rheumatoid arthritis was 25 mg daily.
    FDA's analysis of VIOXX was based on our assessment of the 
available data, the recommendations from the advisory committee, and 
our best judgment of the potential benefits of the drugs compared to 
the potential risks of the drug when used according to the 
recommendations included in the revised labeling.
    As you know, FDA convened a joint meeting of the Arthritis Advisory 
Committee and the Drug Safety and Risk Management Advisory Committees 
in February 2005 to discuss overall benefit to risk considerations, 
including CV and GI safety concerns for COX-2 selective non-steroidal 
anti-inflammatory drugs. The Advisory Committees analyzed all available 
information from recent studies of Vioxx, Celebrex, Bextra, naproxen, 
and other data for non-selective NSAIDs and COX-2 selective products.
    Following the joint meeting, FDA scientists conducted a thorough 
internal review of the available data regarding CV safety issued for 
COX-2 selective and non-selective NSAIDs. It was determined that Bextra 
was associated with an approximately two-fold increased risk of serious 
CV events compared to placebo. On April 6, 2005, FDA completed a 
``Decision Memo--Analysis and Recommendations for Agency Action--COX-2 
Selective and Non-Selective NSAIDs'' based upon the internal review. 
The Decision Memo stated that, based upon detailed conclusions, the 
Agency should ask the manufacturer of Bextra, Pfizer Pharmaceuticals, 
to voluntarily remove Bextra from the U.S. market. Pfizer agreed to do 
so.
    On April 7, 2005, FDA issued the enclosed Public Health Advisory, 
indicating that the Agency had asked Pfizer to voluntarily remove 
Bextra from the U.S. market. The Agency is also asking manufacturers of 
all marketed prescription NSAIDs, including Celebrex (celecoxib) to 
revise the labeling (package insert) for their products to include a 
boxed warning, highlighting the potential for increased risk of CV 
events and the well described, serious, potential life-threatening GI 
bleeding associated with their use.
    Further, manufacturers of non-prescription (over-the-counter) 
NSAIDs are also being asked to revise their labeling to provide more 
specific information about the potential CV and GI risks of their 
individual products and remind patients of the limited dose and 
duration of treatment of these products in accordance with the package 
instructions. In addition, FDA advised the public to contact their 
health care providers to see if other marketed NSAIDs may be helpful in 
treating their pain. For more information on FDA's recent actions, 
please visit: http://www.fda.gov/cder/drug/infopage/COX2/default.htm.

    Question 2b. Some have attributed tens of thousands of deaths to 
the use of Vioxx. Dr. Kweder has described these deaths as 
``theoretical.'' Dr. Galson said they were based on ``junk science.'' 
Does the agency stand by these statements?
    Answer 2b. Dr. Kweder's description of deaths as ``theoretical'' is 
a reference to epidemiological studies that project the numbers of 
deaths from adverse drug reactions. Epidemiology takes certain 
information that is received and attempts to apply that information to 
the entire population. Conclusions from this data are not conclusive, 
but are merely estimates. Any epidemiological study that says a million 
patients suffered adverse events is a theoretical estimation of the 
total adverse drug reactions that is derived by extrapolating the 
actual numbers observed to the population as a whole. Therefore, these 
can be considered ``theoretical'' deaths. The accuracy of projection is 
highly dependant on the accuracy of the information, methodology, and 
assumptions used in the study. While theoretical deaths are estimates 
and must be distinguished from confirmed deaths, this nevertheless 
provides valuable information about potential outcomes, and FDA takes 
this information very seriously in determining whether to revise the 
risk/benefit profile for a particular drug.
    The ``junk science'' quote comes from a conversation with a 
Washington Post reporter that was taken out of context. The FDA 
official interviewed was asked to respond to the reporter's question 
about whether it was accurate to state that a certain number of 
individuals in particular Congressional districts could be said to have 
died of cardiovascular events because of Vioxx. It is not possible to 
derive from risk estimates the cause of a particular individual's 
death. This concept is complicated to explain, and FDA's official 
reacted to it by characterizing the particular statement regarding 
deaths of individuals in a Congressional district as ``junk science.''

    Question 2c. Does the agency believe that Vioxx increases the risk 
of heart attack and stroke? If not, why not?
    Answer 2c. Following the February 16-18, 2005, joint meeting of 
FDA's Arthritis Advisory Committee and the Drug Safety and Risk 
Management Advisory Committee, FDA scientists conducted an internal 
review, taking into consideration the recommendations of the Advisory 
Committees, of the available data regarding CV safety for COX-2 
selective and non-selective NSAIDs. On April 6, 2005, FDA completed a 
document entitled, ``Decision Memo--Analysis and Recommendations for 
Agency Action--COX-2 Selective and Non-Selective NSAIDs'' (located at: 
http://www.fda.gov/cder/drug/infopage/COX2/NSAIDdecisionMemo.pdf). The 
Decision Memo reflects the Agency's evaluation of the risk/benefit 
profile of Vioxx, Celebrex, and Bextra, among others. The memo 
concluded that ``these three approved COX-2 selective drugs are 
associated with an increased risk of serious CV events, at least at 
some dose, with reasonably prolonged use.''

    Question 2d. If so, does the agency believe that some of the 20 
million people who used Vioxx actually experienced heart attack or 
stroke because of Vioxx? If not, why not? If so, how many heart attacks 
and strokes does the agency believe Vioxx caused?
    Answer 2d. Based on evidence from clinical trials, it is possible 
that some of the patients who were treated with Vioxx may have 
experienced a heart attack or stroke related to the use of Vioxx. We 
cannot reliably state how many heart attacks and strokes were caused by 
Vioxx alone, in part because the estimates vary depending on how you 
assume Vioxx was used and at what time point you believe the CV risks 
of Vioxx start.
    In his presentation prepared for FDA's Advisory Committee meeting 
on NSAIDs in February 2005, Dr. Robert O'Neill described the many 
challenges and assumptions involved in making projections of harm 
associated with Vioxx to the general population that might have been 
exposed to Vioxx. To illustrate these challenges, he considered the 
types of assumptions, often unverifiable, that have to be made to make 
any projection. He illustrated an approach that emphasized the 
importance of when during exposure the risk occurs, what is its time 
pattern, and how many people are estimated to be chronically exposed 
for various durations. The clinical trials available were used for some 
estimates. The approach estimated the cumulative risk of confirmed 
thrombotic events, myocardial infarction (MI) and sudden death for 
Vioxx from two different clinical trials, VIGOR and APPROVe, trial 
designs of different duration, with different patient populations, with 
different doses, and with different control groups. He then estimated 
and projected to the population sizes that would typically have been 
exposed for different durations of chronic usage. For MI and sudden CV 
death, the projections for the 25mg and 50mg combined were between 
32,418 to 33,093 people, depending upon assumptions. For confirmed 
thrombotic events occurring with 14 days of last study dose the 
projection were between 47,710 and 49,440. Taking all the known and 
unknown sources of variability as well as the statistical uncertainty 
in the data, the estimates could be larger or smaller.

    Question 3a. Please explain how the agency's response to the VIGOR 
trial would have been different if Secretary Leavitt's recently 
announced independent board had been in place at the time? What actions 
would have occurred?
    Answer 3a. FDA's experience with Vioxx and other recent drug safety 
events are the impetus behind our forming the Drug Safety Oversight 
Board. Had the Drug Safety Board existed at the time FDA became aware 
of the results of the VIGOR trial, we believe that CDER staff would 
have identified the trial results and concerns as an issue that would 
have warranted deliberation by the Board. It is difficult to speculate 
exactly what would have happened, but we feel that the Board as it is 
currently conceptually proposed would have made its best scientific 
judgments and recommendations based on all of the safety information 
available.

    Question 3b. Is there reason to think its analysis of the VIGOR 
data and recommendations stemming from it would have been different 
from those of the FDA reviewers at the time? What authority would they 
have had to take corrective actions?
    Answer 3b. Again, it is difficult to speculate what would have 
happened in this case had the Board existed at the time the VIGOR trial 
results became known. However, whatever recommendations the Board may 
have made would have been presented to the Center Director who would 
have made the final determination on how to proceed. It then would have 
been the responsibility of the appropriate CDER program offices to 
carry out the decision of the Center Director.

    Question 3c. Does the agency believe that use of the drug would 
have been different? In particular, would significantly fewer people 
have experienced heart attack and stroke? If not, please explain why 
the recent FDA proposals are an adequate response to the Vioxx 
disaster.
    Answer 3c. Again, it is difficult to speculate with confidence the 
level of influence the Drug Safety Oversight Board's involvement would 
have made in the VIOXX case. We are confident, however, that the 
actions that have been announced recently including the formation of 
the DSB, the introduction of the ``Drug Watch,'' increased involvement 
of outside expertise, recent publication of risk management guidance, 
and other activities are an appropriate proactive response to the 
recent drug safety concerns. But we do not assume that these actions 
alone are adequate. We are continuing to proactively discuss and seek 
advice and feedback from internal and external sources on possible 
additional actions we can take. For example, we look forward to the 
Institute of Medicine's report on the drug safety system in the U.S.
    While we would not characterize the Vioxx situation as a 
``disaster,'' we have openly conceded the fact that there is room for 
improvement in our post-marketing drug surveillance program and that is 
why we have acted quickly to improve our drug safety processes and our 
internal and external communication of drug safety issues. We agree 
that discussions with the company to change the labeling for Vioxx were 
lengthy. However, in the future, we expect that the Drug Watch page 
will provide a forum for informing the public sooner about emerging 
safety issues.

    Question 4. In January 2005, the data from an early clinical trial 
of Celebrex indicating an increased risk of cardiovascular problems 
were posted on a clinical registry database. The results from this 
trial had never before been released to the public. Were these results 
given to the FDA? When? In what form? Please explain in detail how the 
agency assessed these data. Were they ever reflected in the labeling of 
the drug? Did the agency do any sort of follow-up? Did this study play 
any role in the agency's assessment of the heightened cardiovascular 
risk from Vioxx shown by the VIGOR trial?
    Answer 4. The sponsor gave these results to FDA. The report, 
submitted on June 7, 2001, indicated that a number of patients (n=13) 
who participated in the study had adverse events, which were listed 
using the WHO preferred term ``cerebrovascular disorder.'' The Division 
of Neuropharmacological Drug Products (DNDP) requested narratives be 
submitted for all 13 patients. Upon review of the narratives all 13 
appear to have unequivocally sustained either a stroke or transient 
ischemic attack during or within 28 days of the study's termination. 
The incidence of serious adverse events (including deaths), adverse 
event discontinuations and all adverse events, when evaluating 
individual adverse events as coded by specific preferred terms, did not 
raise any safety concerns which warranted further action.
    In October 2004, the Division again looked at the CV adverse event 
data contained in the June 7, 2001, study report, focusing on 
myocardial infarction, angina, and cerebrovascular disorder (since 
myocardial infarction and stroke were the events of greater concern 
with rofecoxib) and did not include atrial fibrillation, cardiac 
failure and pulmonary edema. DNDP felt that no action was needed based 
on these results, except to further evaluate the events that were 
subsumed under the term ``cerebrovascular disorder.''
    On October 22, 2004, after assessing the overall adverse event 
profile for this study, DNDP asked the sponsor to provide clinical 
narratives for patients who participated in the study and had adverse 
events that were listed using the WHO preferred term ``cerebrovascular 
disorder.'' The objective of this request was to clarify whether the 
patients actually had strokes or transient ischemic attacks.
    Based on the narratives provided and reviewed by DNDP, the deaths 
in the study seem not to be primarily related to thrombotic vascular 
events (many different presumed causes), and the difference between 
drug and placebo in the incidence of vascular serious adverse events 
(or non-serious adverse events in general) does not clearly indict the 
drug.
    Our conclusions that no action was needed was based on the fact 
that the events were small in number in both treatment groups, not 
uncommon in older individuals, not strikingly different in incidence 
between the drug and placebo groups, and in some instances more common, 
in fact, in the placebo group.
    This study did not play role in the Agency's assessment of the 
heightened CV risk from Vioxx shown by the VIGOR trial as it was an 
Alzheimer's study and was not designed nor did it focus on those 
endpoints.

    Question 5. You mentioned in the Q&A that drug sales 
representatives are still using egregious marketing techniques-such as 
detailing, or shadowing of doctors in their offices, offering expensive 
gifts and samples and honoraria for prescribing physicians-and that 
those practices have a large effect on prescribing habits. Can FDA 
encourage drug companies to use better educational techniques, such as 
peer-reviewed communication of results, to get the word out about new 
products? Would it help the FDA promote rational prescribing if the FDA 
had the authority to limit such marketing for recently approved drugs, 
whose risk-benefit profiles are still not fully understood? If not, why 
not? How in your view should the concern about promotion to physicians 
be addressed?
    Answer 5. FDA does not regulate practices such as shadowing of 
doctors in their offices or offering expensive gifts or honoraria for 
prescribing physicians, and FDA does not regulate the practice of 
medicine.
    FDA's regulations do require that all promotional materials be 
submitted to FDA on Form 2253 at the same time as they are used to 
promote to healthcare professionals or to consumers. In other words, 
FDA's review of promotional materials is generally intended to occur 
post hoc. If FDA finds the materials to be inaccurate or unbalanced, 
FDA may take enforcement actions requesting that sponsors stop using 
violative materials. In some cases, FDA may request that sponsors run 
corrective advertisements or issue corrective letters to correct 
product misimpressions created by false, misleading, or unbalanced 
materials.
    CDER's Division of Drug Marketing, Advertising, and Communications 
(DDMAC) is responsible for regulating prescription drug promotion. 
DDMAC's mission is to protect the public health by helping to ensure 
that prescription drug information is truthful, balanced, and 
accurately communicated. This is accomplished through a comprehensive 
surveillance, enforcement and education program, and by fostering 
optimal communication of labeling and promotional information to both 
health care professionals and consumers.
    Promotional programs and materials performed and disseminated by 
companies are subject to the labeling and advertising provisions of the 
FD&C Act. The FD&C Act and regulations do not distinguish between 
promotion to professional or consumer audiences. Section 502(n) of the 
FD&C Act specifies that prescription drug advertisements must contain 
``a true statement of . . . information in brief summary relating to 
side effects, contraindications, and effectiveness'' of the advertised 
product. The implementing regulations specify that prescription drug 
advertisements cannot be false or misleading, cannot omit material 
facts, and must present a fair balance between effectiveness and risk 
information.
    FDA regulates advertisements and other promotional material, called 
``promotional labeling,'' disseminated by or on behalf of the 
advertised product's manufacturer, packer or distributor. According to 
the October 2002 GAO report entitled, Prescription Drugs: FDA Oversight 
of Direct-to-Consumer Advertising Has Limitations, ``Promotion to 
physicians accounted for more than 80 percent of all promotional 
spending by pharmaceutical companies in 2001.'' Therefore, the bulk of 
the Agency's time spent reviewing promotional material, is spent 
reviewing materials produced for promotion to health care 
professionals, such as detail aids used by manufacturer 
representatives, convention displays, file cards, booklets, and 
videotapes, which is distinct from advertising directed toward 
consumers.
    Thank you again for the opportunity to testify before the committee 
on this important topic and also for the opportunity to submit these 
answers for the hearing record. If there are further questions, please 
let us know.
            Sincerely,
                                             Patrick Ronan,
                            Assistant Commissioner for Legislation.
                                 ______
                                 
  Response to Questions of the HELP Committee by Cecil B. Wilson, M.D.

                 RESPONSE TO QUESTIONS OF SENATOR ENZI

    Question 1. What does FDA do to get important new information about 
drugs to doctors and patients, short of an actual label change? How 
might FDA improve their communications?
    Answer 1. The primary way by which the FDA communicates information 
about a drug's risks and benefits to physicians is through the package 
insert. However, when changing a product's labeling is not appropriate, 
the FDA relies primarily on electronic means to communicate important 
new drug or safety information, including the posting of public health 
advisories directly on the FDA Web site. Approximately 46,000 
individuals receive direct e-mail notices of safety alerts that are 
posted on the MedWatch homepage. Additionally, approximately 160 
MedWatch partners, including the AMA, assist in disseminating safety 
alert information.
    The FDA, the pharmaceutical industry, and physician organizations 
must collaborate and identify innovative ways to communicate new risk 
information about a drug or biological product to physicians so they 
will be aware of it, remember it and act on it when prescribing a drug. 
Potential collaborative activities include:
     Undertaking a major CME initiative on risk communication;
     Working with major medical journals and medical society 
web site editors to identify standard places for the dissemination of 
important new risk information about drugs and biological products;
     Using alternative mechanisms to transmit ``Dear Doctor'' 
letters, which disseminate important prescribing information from the 
pharmaceutical companies to physicians (e.g., publication in medical 
journals, possibly as paid advertisements; placement on medical society 
web sites; and transmission to individual physicians by blast fax, 
blast e-mail, or direct daily downloads to personal digital assistants 
[PDAs]);
     Changing the content and format of ``Dear Doctor'' letters 
to emphasize the need for action by the prescribing physician; and
     Encouraging pharmaceutical companies to train and send 
their sales forces to physicians to educate them on important new risk 
information about company products.
    Additionally, the FDA issued a proposed rule in December 2000 to 
modify the format and content of the package insert with the goal of 
making the information more useful and user-friendly to physicians. 
Their recommendations included a more simplified, ``Highlights of 
Prescribing Information'' section within the package insert. The AMA 
continues to strongly support FDA efforts to make package inserts more 
useful and user-friendly for physicians and encourages the FDA to issue 
a final rule to that effect.

    Question 2. We have heard that the voluntary and passive nature of 
the adverse event reporting system may result in under-reporting of 
safety issues. What can you suggest to this Committee as to how we 
could improve that reporting? How might new technologies, such as 
Electronic Medical Records be applied to the Adverse Event Reporting 
System and what are the benefits one could expect to achieve?
    Answer 2. As mentioned in our written testimony, if formal 
postmarketing studies are not conducted by manufacturers or clinical 
investigators to obtain safety information, observational data 
collected by physicians, other health professionals, and patients are 
the keys to evaluating and characterizing a drug's risk profile in 
actual clinical use. Currently, the FDA maintains an adverse event 
reporting system termed MedWatch, which incorporates both a mandatory 
adverse event reporting system for manufacturers subject to the 
Agency's postmarketing safety reporting regulations, and a voluntary, 
adverse event reporting system for health care professionals, 
consumers, and patients. MedWatch can be an effective tool for 
detecting signals suggesting that a drug may be associated with a rare, 
but serious, adverse event.
    However, the MedWatch program is a passive system and it is limited 
by its reliance on voluntary reporting, which inevitably leads to under 
reporting. Under reporting and uncertainty about the actual extent of 
drug exposure, make it difficult to estimate true rates of occurrence 
of drug-induced adverse events. Because of their observational nature, 
spontaneous reports also are limited in their ability to establish 
causality. In order to enhance this program, better educational efforts 
are needed to inform physicians and other health professionals on how, 
when, and where to report suspected serious adverse events.
    Additionally, attention should be directed toward enhancing 
postmarketing surveillance by using more active approaches. For 
example, well designed pharmacoepidemiologic studies on newly marketed 
drugs could substantially enhance our ability to more accurately 
determine a drug's adverse event profile in a timely manner.

    Question 3. I know that there are questions about both quantity and 
quality of adverse event reports. FDA already receives hundreds of 
thousands of reports a year, and perhaps should be receiving more. How 
do they separate the truly important events from the rest? What can 
treating physicians do to improve the quality of reporting that you do 
to the FDA?
    Answer 3. Based on our understanding, the FDA defines adverse drug 
events as those occurring: (1) in the course of use of a drug product 
in professional practice; (2) from drug overdose, whether accidental or 
intentional; (3) from drug abuse; (4) from drug withdrawal; and (5) 
from any ``failure of expected pharmacological action.'' According to 
the FDA, MedWatch is especially interested in receiving: (1) reports of 
serious adverse event reports that are novel or not currently included 
in the drug's labeling; (2) all serious events associated with new 
drugs during their first 3 years on the market; and (3) previously 
reported reactions if they are serious and occur in clusters.\1\
---------------------------------------------------------------------------
    \1\ A serious adverse drug reaction is defined as one that results 
in or prolongs hospitalization, is life-threatening, contributes to 
significant disability, or results in the death of the patient.
---------------------------------------------------------------------------
    To improve the quality of reporting to MedWatch, better educational 
efforts are needed to inform physicians and other health professionals 
on how, when, and where to report suspected serious adverse events.

    Question 4. FDA evaluates the risk/benefit ratio of a drug for a 
population, but doctors and patients evaluate it on an individual 
basis. Could you comment on the value and limitations of off-label 
prescribing? Would Federal restrictions on off-label use negatively 
interfere with the doctor-patient relationship?
    Answer 4. Unlabeled (off-label) uses are defined as the use of a 
drug product for indications or in patient populations, doses, or 
routes of administration that are not included in FDA-approved 
labeling. Under the Federal Food, Drug, and Cosmetic Act, a drug 
approved by the FDA for marketing may be labeled, promoted, and 
advertised by a manufacturer for only FDA approved uses. Even though 
the Prescription Drug User Fee Act (PDUFA) has reduced the review time 
for Supplemental New Drug Applications or SNDAs, manufacturers are not 
required to and may not choose to seek FDA approval for all useful 
indications. This occurs because the expense of regulatory compliance 
may be greater than the eventual revenues expected. A sponsor may also 
not seek FDA approval because of difficulties in conducting controlled 
clinical trials (e.g., for ethical reasons, or due to the inability to 
recruit patients).
    A physician may choose to prescribe a drug for uses, in treatment 
regimens, or in patient populations that are not part of the FDA-
approved labeling. The decision to prescribe a drug for an unlabeled 
use is made by the physician in light of all information available and 
in the best interest of the individual patient. Prescribing for an 
unlabeled use requires the physician to use the same judgment and 
prudence as exercised in medical practice for it to conform to accepted 
professional standards. Given the prevalence of unlabeled uses and the 
fact that in many clinical situations such use may represent the most 
appropriate treatment (and in some cases the only treatment), the 
prescribing of FDA-approved drugs for unlabeled uses is often necessary 
for optimal patient care.
    The AMA also strongly supports the SNDA process. However, given the 
disparity between the actual submission of SNDAs and the evolution of 
evidence-based medical practice, physician prescribing for unlabeled 
uses should not be impeded by any actions taken to improve drug safety.

    Question 5. Could you comment on the value of FDA's new Web site, 
announced by HHS Secretary Leavitt and Acting Commissioner Crawford, as 
a step to make sure patients and doctors have the latest and best 
information about the drugs they are using?
    Answer 5. The AMA applauds HHS and FDA efforts to enhance the 
transparency of the drug surveillance and risk communication processes 
with the creation of the ``Drug Watch'' web page. However, the FDA must 
provide clear advice when it disseminates emerging or preliminary 
information prior to taking regulatory action.

                 RESPONSE TO QUESTIONS OF SENATOR HATCH

    Question 1. I understood from your remarks that the FDA had over 
the last several years increased the speed of drug approvals without 
any decrements in drug safety. The observational data collected by 
clinicians and patients that is used to evaluate and characterize a 
drug's risk profile in actual clinical use has received a fair amount 
of criticism lately. Would you please comment on the limitations and 
benefits of the current system and any changes you would recommend?
    Answer 1. As mentioned in our written testimony, if formal 
postmarketing studies are not conducted by manufacturers or clinical 
investigators to obtain safety information, observational data 
collected by physicians, other health professionals, and patients are 
the keys to evaluating and characterizing a drug's risk profile in 
actual clinical use. Currently, the FDA maintains an adverse event 
reporting system termed MedWatch, which incorporates both a mandatory 
adverse event reporting system for manufacturers subject to the 
Agency's postmarketing safety reporting regulations, and a voluntary, 
adverse event reporting system for health care professionals, 
consumers, and patients. MedWatch can be an effective tool for 
detecting signals suggesting that a drug may be associated with a rare, 
but serious, adverse event.
    However, the MedWatch program is a passive system and it is limited 
by its reliance on voluntary reporting, which inevitably leads to under 
reporting. Under reporting and uncertainty about the actual extent of 
drug exposure, make it difficult to estimate true rates of occurrence 
of drug-induced adverse events. Because of their observational nature, 
spontaneous reports also are limited in their ability to establish 
causality. In order to enhance this program, better educational efforts 
are needed to inform physicians and other health professionals on how, 
when, and where to report suspected serious adverse events.
    Additionally, attention should be directed toward enhancing 
postmarketing surveillance by using more active approaches. For 
example, well designed pharmacoepidemiologic studies on newly marketed 
drugs could substantially enhance our ability to more accurately 
determine a drug's adverse event profile in a timely manner.

    Question 2. How would you recommend we enhance postmarketing 
surveillance? What more active approaches do you see as promising? How 
is it best to notify clinicians of changes in practice or new findings? 
Passive means, such as having them log onto a Web site, are likely to 
be less effective than more active methods--but what specifically do 
you see as the best practices?
    Answer 2. Well designed pharmacoepidemiologic studies on newly 
marketed drugs could substantially enhance our ability to more 
accurately determine a drug's adverse event profile in a timely manner.
    Furthermore, the FDA, the pharmaceutical industry, and physician 
organizations must collaborate and identify innovative ways to 
communicate new risk information about a drug or biological product to 
physicians so they will be aware of it, remember it and act on it when 
prescribing a drug. Potential collaborative activities include:
     Undertaking a major CME initiative on risk communication;
     Working with major medical journals and medical society 
web site editors to identify standard places for the dissemination of 
important new risk information about drugs and biological products;
     Using alternative mechanisms to transmit ``Dear Doctor'' 
letters, which disseminate important prescribing information from the 
pharmaceutical companies to physicians (e.g., publication in medical 
journals, possibly as paid advertisements; placement on medical society 
web sites; and transmission to individual physicians by blast fax, 
blast e-mail, or direct daily downloads to personal digital assistants 
[PDAs]);
     Changing the content and format of ``Dear Doctor'' letters 
to emphasize the need for action by the prescribing physician; and
     Encouraging pharmaceutical companies to train and send 
their sales forces to physicians to educate them on important new risk 
information about company products.

    Question 3. I appreciate your educating us on the reasons why so 
many patients are being properly treated by off-label uses of drugs. Do 
you see the decision by a physician or other clinician to use a drug in 
this way as an exercise in clinical judgment, in the same way that the 
physician decides what diagnostic test to use, what diagnosis is the 
most likely, whether the treatment plan is successful, and so on?
    Answer 3. Unlabeled (off-label) uses are defined as the use of a 
drug product for indications or in patient populations, doses, or 
routes of administration that are not included in FDA-approved 
labeling. Under the Federal Food, Drug, and Cosmetic Act, a drug 
approved by the FDA for marketing may be labeled, promoted, and 
advertised by a manufacturer for only FDA approved uses. Even though 
the Prescription Drug User Fee Act (PDUFA) has reduced the review time 
for Supplemental New Drug Applications or SNDAs, manufacturers are not 
required to and may not choose to seek FDA approval for all useful 
indications. This occurs because the expense of regulatory compliance 
may be greater than the eventual revenues expected. A sponsor may also 
not seek FDA approval because of difficulties in conducting controlled 
clinical trials (e.g., for ethical reasons, or due to the inability to 
recruit patients).
    The decision to prescribe a drug for an unlabeled use is made by 
the physician in light of all information available and in the best 
interest of the individual patient. Prescribing for an unlabeled use 
requires the physician to use the same judgment and prudence as 
exercised in medical practice for it to conform to accepted 
professional standards. In some instances, prescribing a product off-
label is the most appropriate therapy based on the latest, best 
scientific evidence. In some patient populations, it may be the only 
treatment option.

    Question 4. We have heard several people recommend that the FDA be 
able to ``control'' how drugs are used. Do you agree? Should the 
government, through the FDA, decide how patients should be treated, or 
is that a matter for the clinician who is caring for that individual 
patient? Should we instead focus on improving educational outreach 
programs and surveillance notifications to clinicians?
    Answer 4. As stated in our testimony, FDA-approved drug product 
labeling (i.e., the package insert) should be the primary means by 
which the FDA communicates risks about drug products to physicians for 
the vast majority of drugs. Higher level risk communication and risk 
minimization tools that extend beyond the package insert, such as 
performance-linked access systems and some reminder systems, should be 
used only as a last resort to keep high-risk drug products with truly 
unique and important benefits on the market.
    In the government's efforts to improve drug safety, there may be a 
desire to use, more routinely, risk minimization tools that extend 
beyond not only the package insert, but also beyond targeted education 
and outreach in an effort to improve drug safety. A number of these 
approaches would directly manage or restrict physician prescribing and 
may lead to unintended consequences.
    Rather than focus on restrictions, the AMA believes that the FDA, 
the pharmaceutical industry, and physician organizations must 
collaborate and identify innovative ways to communicate new risk 
information about a drug or biological product to physicians so they 
will be aware of it, remember it and act on it when prescribing a drug. 
The AMA encourages the FDA and the product sponsor to work with 
relevant physician organizations to assure that the minimum number and 
least intrusive tools are selected to achieve the risk minimization 
objective.
    The AMA believes that individual States should regulate the 
practice of medicine. AMA policy provides that, ``the AMA and 
interested physicians will continue to work with the Food and Drug 
Administration to prevent the unnecessary intrusion of the government 
and other regulatory bodies into the doctor-patient relationship, 
especially as it concerns the prescription of medication.'' (AMA Policy 
H-100.971).

                RESPONSE TO QUESTIONS OF SENATOR KENNEDY

    Question 1. Why do you believe that direct-to-consumer ads that 
encourage patients already in treatment to ask their doctor about 
prescribing a particular drug or switching them to a new drug are so 
effective? Why are drug companies spending so much time and money 
shadowing doctors, and showering them with gifts and honoraria and 
samples, as part of their marketing? I assume that all of these 
practices work in getting physicians to prescribe drugs they wouldn't 
have otherwise. Is that beneficial?
    Answer 1. The AMA supports patients' increased access to drug 
information, but is concerned about the impact direct-to-consumer (DTC) 
advertisements have on the physician-patient relationship. In 
consultation with the FDA, the AMA developed guidelines (AMA Policy H-
105.988) in 1993 for acceptable DTC advertisements. These 
advertisements should promote accurate, balanced information that can 
provide educational benefit for consumers. The AMA policy also urges 
the FDA and the pharmaceutical industry to conduct or fund 
``independent'' research to study the effects of the DTC ads on the 
physician-patient relationship, health outcomes and costs.
    Regarding gifts given to physicians by pharmaceutical companies, 
some gifts that reflect customary practices of the industry, may not be 
consistent with the AMA Principles of Medical Ethics (AMA Code of 
Medial Ethics E-8.061). These Principles were developed by the AMA's 
Council on Ethical and Judicial Affairs and are designed to guide 
physicians on the inappropriateness of accepting gifts from the 
pharmaceutical industry.

    Question 2. Besides asking for FDA help in getting better quality 
information to doctors and patients, what is the American Medical 
Association doing to ensure that doctor's prescribing habits are based 
on peer-reviewed evidence, rather than on clever marketing to doctors 
and patients? Are those strategies working? Does the FDA need to step 
in and regulate communication between the drug industry and the public 
or the promotional goods and services offered by the drug industry to 
prescribing doctors?
    Answer 2. The AMA supports activities designed to foster the 
development and implementation of evidence-based, physician-level 
clinical quality improvement efforts. The AMA convened the Physician 
Consortium for Performance Improvement to identify and develop 
performance measurement resources for physicians. The Consortium is 
comprised of clinical content experts from more than 60 State and 
medical specialty societies, methodological experts, the Agency for 
Health Research Quality (AHRQ) and the Center for Medicare and Medicaid 
Services (CMS).
    Through publication of the Journal of the American Medical 
Association (JAMA) and the Archive specialty journals, the AMA is the 
world's leading medical organization in publishing peer-reviewed 
articles intended to inform and guide evidence-based clinical practice. 
Additionally, the AMA is a leader in providing Continuing Medical 
Education (CME) activities. In 2004, the AMA sponsored more than 320 
CME activities serving more than 43,000 physicians.
    The AMA believes that individual States should regulate the 
practice of medicine. AMA policy provides that, ``the AMA and 
interested physicians will continue to work with the Food and Drug 
Administration to prevent the unnecessary intrusion of the government 
and other regulatory bodies into the doctor-patient relationship, 
especially as it concerns the prescription of medication.'' (AMA Policy 
H-100.971).
                                 ______
                                 
        Response to Questions of Senator Enzi by Keith L. Carson

    Question 1. Personalized medicine intrigues me. Each product is a 
small market but the field overall has huge potential. How do we make 
the drug development process efficient enough to make these products 
worth pursuing?
    Answer 1. Personalized Medicine will allow doctors and drug 
companies to identify individuals who are genetically susceptible to 
certain diseases, as well as those who have a propensity to either 
respond well to, or experience an adverse event from, certain drugs. 
With this information, doctors will have a better chance to prescribe 
drugs to which a patient will respond well, an avoid drugs to which a 
patient will have problems. The product manufacturers will also be able 
to use this information in clinical trials to select patients who would 
probably do well in the trial, and avoid patients who could have 
adverse reactions.
    Genetic mapping could therefore revolutionize medicine by making 
drugs work better while avoiding undesirable reactions. Clinical trials 
could be done with smaller patient populations and result in fewer 
adverse events. The product development cost savings could be 
tremendous, plus the welfare of the clinical trial participants would 
be greatly improved.
    Personalized Medicine would not necessarily result in small markets 
for each product, but would improve the performance of products while 
preventing harmful side-effects. If patients were identified as having 
a higher probability of bad side effects with a product, no one would 
want them to take it. The use of this technology could prevent a 
tremendous amount of litigation costs and settlements that now result 
from drug related adverse events and deaths.
    The potential savings from smaller clinical trial populations and 
fewer patient law suits are such that product developers 
(biopharmaceutical companies) will eagerly adopt this new technology 
and use it extensively.
    Unfortunately, there are significant problems associated with 
Personalized Medicine. Many patients will be reluctant to have their 
genes mapped, since they don't know how the information will be used, 
or by whom. In addition, very large databases will have to be built and 
managed to share enough information for Personalize Medicine to be 
effective. Doctors will have to do a far better job of sharing patient 
response data, and product developers will have to share what they now 
consider proprietary clinical data with the public and other companies.

    Question 2. What other emerging technologies might be useful in 
identifying at-risk patients or populations and/or predicting potential 
adverse events? Are there processes in place for the use of these 
technologies in regulatory decision-making? What are some of the 
hurdles to validation, regulatory acceptance and broad application of 
these technologies?
    Answer 2. I'm not aware of technologies other than pharmacogenomics 
(gene-mapping) that have this potential. However, gene-mapping 
technology currently relies on the use of microarray technology, plus 
datamining and in silico technology to establish the correlations 
needed for it to be useful in Personalized Medicine.
    I know that the FDA is trying to build up their technical 
capabilities in this area, so that they can better understand the 
technology and apply it. However, I'm sure they will need more funding 
to do this properly.
    FDA is the only entity that has access to all of the clinical trial 
data that is submitted. They are in a unique position to manage this 
data and make it available for correlations to patient gene-mapping. 
However, such an endeavor will take massive data storage and computing 
capability.
    I suggest that Congress provide adequate funding so FDA can have 
outside firms provide the necessary storage and analysis services. Such 
services are becoming a commodity and prices are very competitive. FDA 
doesn't have the internal expertise to build and maintain a state-of-
the-art capability, and they certainly don't have the hardware. They 
should rely on outside contractors as much as possible.
    Once a large enough database has been assembled, then FDA could use 
the data to make comparisons between similar products, and help make 
better regulatory decisions concerning the design and size of clinical 
trials, plus the selection of specific patients for these trials. As 
more clinical data became available for a particular product, the 
Agency could even decide that the product should be licensed for 
specific patient profiles, while other studies continued.
    To validate or accept this technology, sufficient data will be 
needed. Product response data is required to show which patients did 
well or had problems with a certain drug. Adverse events will have to 
be experienced to establish the correlations needed to predict which 
patients could be susceptible to them. To get this data, product 
developers must be more willing to share clinical data with the public 
and other companies, and doctors must do a better job of reporting how 
patients respond to new drugs.
    In addition, patients will have to become more comfortable with 
having their genes mapped. Many people are concerned about how this 
information will be used. They are especially fearful of what insurance 
companies and employers could do with the data. Systems must be devised 
that provide adequate data encryption and controlled access.

    Question 3. What one action could Congress take that would most 
dramatically improve drug safety in the U.S.?
    Answer 3. Congress could increase funding so that FDA can fully 
utilize this new technology. In addition, Congress could work with AMA 
to implement better electronic systems through which doctors can report 
drug responses--good and bad.
                                 ______
                                 
       Response to Questions of Senator Hatch by Keith L. Carson

    Question 1. Do you believe that the lack of a modern IT 
infrastructure at the FDA impairs the agency's ability to do its job 
and hire the best people?
    Answer 1. I believe that the lack of modern IT infrastructure will 
severely impede FDA's ability to utilize new technologies such as 
advanced data management, datamining, and in silico technology. FDA is 
the only entity that has access to all the clinical trial data that is 
submitted, and has a unique opportunity to make this data available for 
use in Personalized Medicine.
    To acquire adequate infrastructure, I suggest that much of it 
should be obtained through services from outside contractors. However, 
significant upgrades will still be needed for internal data management 
capabilities that should not be outsourced.
    I don't know if a lack of IT infrastructure is keeping FDA from 
hiring the best people, but it could certainly prevent top-notch IT 
people from joining the Agency.
    However, FDA budget restrictions provide a far bigger impediment to 
capturing the best people. The Agency is known for having a very tight 
budget, where everything is difficult to justify and buy. Top technical 
people will go where the resources are available for them to do the 
best work.

    Question 2. Is the FDA competitive when it comes to attracting new 
staff? What do you believe that the FDA should do to attract the best 
and brightest scientists?
    Answer 2. I seriously doubt that FDA is competitive when it comes 
to attracting new staff. FDA is constantly losing good people to 
industry positions that pay far more money and provide vastly superior 
resources. In addition, these industry positions don't involve the 
political and bureaucratic hassles that an Agency job is known for. At 
FDA, every decision must be scrutinized as to ethical and legal 
implications, plus pass through multiple levels of management. Such an 
environment stifles innovation and sound decision making.
    To attract the best and brightest scientists, FDA must have the 
resources to equip, supply, and staff their laboratories. Without 
proper funding, good people would never put up with the bureaucratic 
hassles and red tape that the Agency is known for.
    The FDA does have some very talented and brilliant people, who are 
truly dedicated public servants. I could not work under the pressures, 
internal politics, and bureaucratic nightmares these folks endure on a 
daily basis. Then, when their programs are drastically underfunded, I 
don't know how they can take it. We should all be proud to have the 
folks that are there, and Congress should do all it can to adequately 
fund FDA's laboratories.

                                 ______
                                 
Response to Questions of the HELP Committee by Raymond L. Woosley, MD, 
                                 Ph.D.

                 RESPONSE TO QUESTIONS OF SENATOR ENZI

    Question 1. In your testimony, you describe some initiatives for 
tracking outcomes of treatment with new drugs, particularly the UK's 
``yellow card'' system. The UK has a national health care system, 
unlike the US. How do you envision setting up a physician-based 
national drug safety evaluation system in the absence of a national 
health care system?
    Answer 1. I don't think such a system is the best approach for the 
United States. The cost of reimbursing physicians would make it too 
expensive to implement and it would compete with physician's time to 
allocate for patient care. That is why I have suggested a pharmacy-
based system for drug safety. The expertise of pharmacy technicians, 
pharmacists and clinical pharmacists are under-utilized and would cost 
far less than a physician-driven system. The proposal described in my 
written testimony would be community-based and focused on outpatient 
healthcare. We would also need to have a hospital-based pharmacy 
network. The principles would be the same.
    We would give pharmacy technicians, pharmacists and clinical 
pharmacists special training in drug safety surveillance using a 
curriculum designed by the AHRQ-funded Centers for Education and 
Research on Therapeutics (CERTs). They could receive the training 
through distance learning (such as ``telemedicine''). The surveys 
(questionnaires developed by the FDA and CERTs collaborators) would be 
performed by pharmacy technicians. Specially trained pharmacists (for 
outpatients) and clinical hospital pharmacists (for inpatients) could 
address the interpretation of medical information and contact 
physicians when necessary to obtain more detailed information.
    The Quality Improvement Organizations (QIOs) funded by the Centers 
for Medicare and Medicaid (CMS) could assist by abstracting medical 
records, a function they now do under contract for CMS. The information 
gained would not only improve outcomes (which both increases safety and 
lowers cost to CMS), it would provide data for the FDA Office of Drug 
Safety that is not now available.
    The data would also be rapidly available so that the number of 
people exposed to risks should be minimized. For the 15 drugs removed 
from the market since 1997, it took an average of 5.9 years before the 
harm was detected and final action taken. I envision a rapid response 
system that saves lives and therefore limits liability for companies in 
subsequent litigation.

    Question 2. I am intrigued by your suggestion of a staged approval 
process for new drugs. However, I'm concerned that there would be great 
demand for a new drug as soon as it is approved. Patients who are 
desperately ill want and need access to new treatments. Could you tell 
me more about how you would restrict access to these drugs?
    Answer 2. I hesitate to use the word ``restrict'' because I share 
your concern for patients who need new medicines. I think we need a 
process that makes the drug available as early as possible but only for 
those patients who are representative of those for whom a new drug has 
shown benefit. If a drug has only been tested in patients with a 
certain type of cancer and patients with renal or liver disease have 
never been studied, we should prohibit or at least strongly discourage 
patients with renal disease from receiving the drug. Likewise, the very 
elderly should be discouraged from taking a new drug if it has only 
been studied in younger people. Perhaps, they could only get the drug 
under a registry system so that we would know what happens when the 
drug is used off label.
    Every drug will be different and we need a system that is flexible 
enough to rapidly response to patient needs. Once in place throughout 
the country, the Community Based Pharmacy Safety Network described 
above could perform this type of function.
    While some may argue this system would limit the ability of doctors 
to prescribe approved drugs for any use them deem appropriate, I 
suggest it would provide a proper balance between the rights of medical 
practitioners and the rights of patients to receive safe and effective 
drugs for their particular medical issues.

    Question 3. What one action could Congress take that would most 
dramatically improve drug safety in the U.S.?
    Answer 3. I agree with Acting Commissioner Lester Crawford in his 
recent testimony that (and I paraphrase) the best way to address drug 
safety is to enable the FDA to pursue the Critical Path Initiative 
described in the white paper: ``Innovation or Stagnation, Challenge and 
Opportunity on the Critical Path to New Medical Products.'' I recommend 
that Congress provide the FDA with $50 million in funding for the 
Critical Path Initiative to utilize the currently available mechanisms 
to both enhance drug safety and enable the development and approval of 
new drugs for serious medical conditions.
    The drug safety problem cannot be solved by the FDA alone. These 
funds would enable the FDA to work with its sister agencies (CMS, AHRQ, 
CDC and NIH), the academic community and the industry to address drug 
safety and the interrelated problems that result from the lack of 
innovation in the process of drug development.

                 RESPONSE TO QUESTIONS OF SENATOR HATCH

    Question 1a. You mention that the pharmaceutical industry spends 
12-15 years and nearly a billion dollars on each drug that is 
successfully developed. You also describe that the proportion of drugs 
that fail during development has doubled in the last 10 years. Why is 
that?
    Answer 1a. The most complete answers to this question can be found 
in the FDA's white paper that was released in March of 2004: 
``Innovation or Stagnation, Challenge and Opportunity on the Critical 
Path to New Medical Products.'' First, I should expand on my testimony: 
Estimates range from $850 million to $1.7 Billion for the amount that 
the industry must spend overall in order to get one drug approved. This 
includes the cost of many drug failures. The amount spent on any one 
drug that succeeds is actually less, probably about $400 million. 
However, the cost of failures is a real cost and one that must be 
accounted for. The rising number of drug failures, both before and 
after marketing, adds tremendously to the cost of pharmaceutical 
research and development. Since 1997, 17 major drug products were 
removed from the market. The costs of their development were in the 
billions of dollars and should be considered a loss from the standpoint 
of ``opportunity costs.'' Removing these drugs from the market also is 
associated with billions of dollars in losses due to litigation and 
personal injury claims.
    A short answer to your question is the following: Over the last 
twenty years we have had a revolution in science that has created new 
opportunities to learn more about the medicines that are being 
developed. The FDA has required that drug companies add research 
projects to the development process in order to know more about the 
drugs and how to use them. For example, the FDA now asks companies to 
identify which enzymes break down the drug because we know that some 
people, due to their individual genetics, can fail to break down the 
drug, which could build up in their bodies to cause harm.
    We now know that two drugs can interact when taken together, so we 
ask companies to do specific studies to test for drug interactions 
before marketing. We now know that some drugs from every possible class 
can have effects on the heart that result in potentially lethal heart 
rhythm abnormalities. We now ask companies to conduct studies to screen 
for this problem.
    These are but a few of the many important new requirements that 
have improved our understanding of how to use medicines more safely. 
However, these requirements add time and expense to the development 
process. On the other hand, there are opportunities to remove certain 
requirements from the traditional process. For example, we still 
require that all companies conduct studies to determine in two rodent 
species (usually rats and mice) the dose of the new drug that kills 
half of the animals. Often rodents don't have the same proteins as 
humans so the studies may never be relevant. However, there is no 
ongoing process for the FDA to meet with other scientists in drug 
development and reach a consensus on what work should no longer be 
required.
    It will take a very special process that brings the very best 
science and a willingness of all involved to share data and experience. 
That does not now exist, but it is the kind of partnership called for 
in the Critical Path Initiative (``Innovation or Stagnation, Challenge 
and Opportunity on the Critical Path to New Medical Products'').

    Question 1b. Do you know how many drugs are not brought to market? 
In other words, how much money and time does the pharmaceutical 
industry invest in research that does not result in a new marketable 
drug?
    Answer 1b. Estimates are that only 11 percent of drugs that enter 
clinical testing are ever approved. Therefore, for every drug approved, 
nine others fail. Since only about 30 new drugs are approved each year, 
approximately 270 fail using these numbers for projection. The figure 
below is taken from Nature Reviews: Drug Discovery, 3 (8): 711, 2004. 
It shows the percentage of successful drugs in each class.
    It is difficult to know how much the industry loses on drugs that 
fail but it could be as high as 55 percent of their $40 billion annual 
investment in research and development. I base this on current 
estimates that a company must plan to spend about $400 million on a 
successful drug and the fact that $875 million is the final cost per 
drug including failures. Unfortunately our ability to predict failure 
is getting worse, not better. Again the work proposed in the Critical 
Path Initiative would help by developing ``biomarkers'' for many 
indications that could more accurately predict the safety and 
effectiveness of drugs.




    Question 2. You testified that industry investment has increased by 
250 percent over the last 10 years, but the number of new products 
submitted for FDA review has fallen by 50 percent. Why is that?
    Answer 2. I believe the rising cost and protracted development 
times have deterred companies from taking more new drugs into 
development. The consolidation of the pharmaceutical industry has also 
been a major factor. Since 1980, 48 companies have collapsed into six 
and with every merger or purchase a large number of drugs have been 
taken out of development. Also, since 1995, the number of failures 
during development has risen and only half as many drugs now succeed to 
reach market.

    Question 3a. You mention that over half of the 15 drugs that were 
removed from the market for safety concerns over the last 8 years were 
in fact safe when used as directed. Should these drugs not have been 
removed?
    Answer 3a. If we had an effective means to assure their safe use 
(i.e., in which use is limited to those conditions where safety and 
efficacy have been proven), these drugs could have remained on the 
market. This is why the FDA must partner with CMS and AHRQ and others 
to find ways to better inform healthcare providers in ways that result 
in safe medication use. The Centers for Education and Research on 
Therapeutics (CERTs) were authorized by Congress with this task. 
However, the CERTs have been too small in number and inadequately 
funded to accomplish this task. Also the FDA has not had sufficient 
staff or resources to work on this with AHRQ and the CERTs. CMS and 
other insurers are the financial benefactors if this can be 
accomplished.
    When terfenadine was removed from the market, the generic form of 
the drug had just been approved by the FDA. However, when it was 
removed, it left only the very expensive brand named non-sedating 
antihistamines. Billions of dollars spent on these antihistamines 
(Claritin, Allegra and Zyrtec) could have been saved if generic 
terfenadine could have remained on the market and used safely, i.e. in 
a way in which it was not taken with drugs such as erythromycin that 
interacted and made its use dangerous.

    Question 3b. If they were effective when used according to their 
indications, did not their removal harm those patients who had been 
using them properly and receiving benefit from them?
    Answer 3b. Indeed. For some of these drugs, there was no 
alternative therapy and patients and doctors lost the benefit of 
effective therapy. For some, there were alternatives but, as mentioned 
above for terfenadine, the alternatives were very expensive and caused 
financial hardship for patients. Vioxx is another example where it was 
intended for patients at risk for gastrointestinal bleeding. Now, they 
do not have access to this safer drug and may be harmed by the 
available agents. By removing Vioxx from the market, the underlying 
issues of why Vioxx and other Cox 2 inhibitors were developed and 
approved have not gone away.

    Question 4. Is the problem in drug development the lengthy, 
extensive, and yet inadequate approval process which costs so much, or 
is it the limited time the company is allowed to recoup its R&D costs 
before generic competition? Or is it the litigious environment in which 
we live that leads to the industry's heightened fears about labeling or 
lawsuits?
    Answer 4. These are all contributors to the problem. However, I 
would not favor extending the patent coverage. This would only 
encourage and reward further inefficiency. We should focus on improving 
(i.e., shortening) the process. AIDS drugs were developed in an average 
of 3 years without risk to patients. That should be the model for all 
drugs.

    Question 5a. You mention that ``once a drug is marketed, the FDA 
has no control over the way it is used in clinical practice.'' Should 
the FDA control this?
    Answer 5a. No, I do not think the FDA should control the practice 
of medicine. This should be controlled by professional societies. 
However, we must encourage better prescribing in every way possible. 
One way that is being addressed by Dr. Mark McClellan, the head of CMS, 
which involves the way we reimburse caregivers. He believes that payors 
like CMS should reward evidenced-based practices. We do not do that 
today. We pay the same for a visit in which a drug was prescribed 
inappropriately as we do for a visit in which the drug was prescribed 
in ways proven to be safe and effective. By rewarding evidenced-based 
practices, CMS can help FDA be sure the drugs they approve are used 
safely.

    Question 5b. Or should clinicians? Isn't there a partnership 
between the FDA and practitioners?
    Answer 5b. I don't think there is such a partnership. The FDA 
writes labels for drugs that do not result in safe practice. The FDA 
has to resort to removing drugs from the market because its ``Dear 
Doctor'' letters are too often ignored, both by doctors and by 
pharmacists.

    Question 5c. Isn't it the role of continuing medical education--and 
in a worst case scenario our malpractice system--to ensure that 
practitioners are using drugs appropriately?
    Answer 5c. Research has shown that conventional continuing medical 
education (CME) programs do not effectively improve the practice of 
medicine. CME when provided by pharmaceutical companies yields 
increased use of a medicine but that is not always the best practice of 
medicine. Vioxx is again a great example, where it was promoted through 
CME programs that resulted in its overuse, i.e., use in patients 
without risk of gastrointestinal bleeding.
    Yes, physicians fear malpractice claims and that fear affects the 
way they practice medicine but not always in ways that we would prefer. 
For example, physicians order unnecessary lab tests to guard against 
claims that are unlikely to occur. They often prescribe unnecessary 
medicines because they are afraid they will be accused of doing nothing 
to help the patient. Healthcare providers want to prescribe the very 
best medicines for their patients. When they fail to do so, it is often 
the ``system'' that has failed them. For example, many times drug 
interactions occur because the physician is unaware of medicines 
prescribed by other doctors.

    Question 6a. If the FDA receives the authority to demand further 
research on marketed drugs, does this also imply they will have the 
responsibility to fund the research as well?
    Answer 6a. I would not recommend giving the FDA the authority to 
require companies to do research after a drug is marketed. Instead, I 
suggest that the FDA, NIH and AHRQ work with the drug company sponsor 
to agree upon what research is needed. The FDA and NIH or AHRQ should 
conduct the agreed upon studies with funding from user fees paid by the 
sponsor. Since the studies would be conducted by NIH and/or AHRQ, there 
would be less concern about the validity of the research. While such 
research is being conducted, the FDA should be given the authority to 
change the way the drug is marketed and distributed. It should be 
allowed to suspend direct to consumer advertising. It could work with 
AHRQ and insurers to prohibit payment for off label prescription if the 
use is possibly dangerous or of unproven value. For example, if a drug 
is considered potentially unsafe for very elderly patients and safety 
studies have not been conducted in this population, CMS and other 
insurers should not provide payment for the drug in those patients. If 
these limitations are inadequate to protect the public, the FDA should 
also be given the authority to suspend marketing. During this 
suspension, every effort should be made to make the drug available to 
those for whom there are no alternative therapies of serious illnesses.

    Question 6b. If they are not going to fund it, will the companies? 
And what will happen to those patients who are benefiting from the drug 
if the company decides that it is more cost effective to remove the 
drug from the market rather than do additional studies?
    Answer 6b. The drug companies will resist changes in post marketing 
research and surveillance because they will want to retain control over 
the research on their products. It is unlikely that they would conduct 
research projects unless they agree upon the methods to be used. It 
will be difficult for NIH or AHRQ to conduct research on drugs unless 
the manufacturer is not in agreement. The company will have to provide 
the medications for the study and they often control the manufacturing 
process. Therefore, I think it is important that this research be 
mutually agreed upon. The FDA will need leverage in order to see that 
the research is done in the best way to protect the public. For this 
reason, the FDA must have the authority to suspend direct to consumer 
ads, suspend sales, etc.
    Compared to the potential profits from a marketed drug, the cost of 
most studies is only a small fraction. I do not think a company will 
choose to remove a drug from the market because of the need for more 
studies. The incentive of having a larger market or less legal 
liability should encourage a company to collaborate in the necessary 
research.

    Question 7. How should we replace the ``user fee'' system so as to 
avoid linking industry support to the FDA's work and performance while 
simultaneously ensuring that the FDA has the funds it needs to carry 
out its mission?
    Answer 7. The funds that are needed to support the FDA should be 
provided by all companies and based on a fraction of sales. The funds 
should pay for all of the activities of the FDA, not just review of new 
drugs and/or devices. I recognize that these charges could lead to 
higher prices for consumers (although market forces can limit pricing, 
especially in a world where more efficient and faster safe drug 
development is practiced). However, even if some prices are higher, 
patients and the public health will be better served by having the FDA 
adequately funded to perform its mission.

                RESPONSE TO QUESTIONS OF SENATOR KENNEDY

    Question 1. Could you please elaborate on why the post approval 
drug safety surveillance system needs to be enhanced? Please describe 
the components of a robust system for post approval safety monitoring, 
including your community pharmacy safety net project?
    Answer 1. Although some tout the current safety system for its 
ability to detect rare adverse events, it is dangerously inefficient 
and slow to respond. The 15 drugs removed from the market over the last 
seven years were on the market for an average of 5.9 years before they 
were detected by the spontaneous reporting system and removed from the 
market. One of the major limitations of the current system is the lack 
of ability to detect rare events quickly and to estimate the frequency. 
Another weakness is the inability to conduct comparative safety 
analyses.
    Therefore, the ideal system would have the following 
characteristics:
    1. Ability to detect rare adverse events very early after 
marketing.
    2. Ability to accurately quantify the incidence of rare adverse 
events.
    3. Ability to compare the safety of alternative therapies.
    4. Ability to detect adverse events that are restricted to special 
populations, e.g., children, the very elderly, patients with renal or 
liver disease, etc.
    5. Ability to detect adverse events under special conditions of 
use, e.g., nursing homes, hospitals, hospices, etc.
    6. Ability to identify risk factors associated with the adverse 
event, e.g., presence of other illnesses, biological sex, concomitant 
medicines, etc. (This information will be essential in the design of 
methods to prevent or minimize future occurrences of the adverse 
event.)
    7. Ability to focus on potential adverse events identified by FDA 
medical reviewers during review of a new drug application.
    8. Ability to address specific questions that arise after the drug 
is on the market.
    The community pharmacy safety net would address many but not all of 
these requirements. The following is a plan that The C-Path Institute 
is developing for a pilot project to be conducted in community 
pharmacies:
              The Community Pharmacy Safety Network (CPSN)
         A Community-Based National Medications Safety Program
        The most effective means of maximizing the benefit and 
        minimizing the risk of harm from medicines is to have every 
        member of the healthcare team, including patients and their 
        families, fully informed in how to use medicines safely. The 
        current healthcare delivery system fails to adequately inform 
        patients how to use medications optimally and fails to provide 
        the U.S. Food and Drug Administration (FDA) with adequate 
        information about the outcomes of medication usage. This is a 
        proposal for the creation of a novel community-based network of 
        individuals trained to help patients maximize the benefits of 
        their medications and also trained to obtain reliable data for 
        the FDA on the beneficial and adverse outcomes of medicines. 
        This is information not currently available in any database.

        A pilot demonstration program for the nation will be conducted 
        in Southern Arizona by the CPSN. The Critical Path Institute 
        and the Arizona Center for Education and Research on 
        Therapeutics (AzCERT) will develop two special curricula for 
        certificate programs in the Pharmacy Technician training 
        program at Pima Community College and the College of Pharmacy 
        at the UA. This curriculum will enable Certified Pharmacy 
        Technicians (CPTs) and certified pharmacists to aid patients in 
        the safe use of new medications and medications in general. 
        CPTs will also maintain a log of patients receiving pre-
        specified medicines and obtain baseline and follow-up 
        information from those patients to determine the outcome of 
        their therapies. Scientists in the Arizona CERT will 
        participate in the development of the curriculum and the 
        evaluation of the effectiveness of the program. Scientists from 
        CERT, C-Path and the FDA will collaborate to identify medicines 
        to be monitored, define comparator cohorts and prepare survey 
        questions to define outcome parameters for each cohort of 
        patients. The CPTs will enter the survey results into a web-
        based database for analysis by the Office of Drug Safety at the 
        FDA. When complex medical events are detected by the CPT 
        survey, a Certified Pharmacist with special training in drug 
        safety will gather information and submit a report to the FDA 
        using standardized MedDRA terminology. The Pharmacist will also 
        report to the prescribing physicians informing them how their 
        patients responded to their therapies and provide summary 
        information from the FDA so they can benchmark their practice 
        to the experience of other physicians and patient populations.

        The CERT will evaluate the impact of the CPSN on patient safety 
        by comparing the outcomes of patients who were given special 
        training in the safe use of their medicines by CPTs to the 
        outcomes in a matched control group receiving routine 
        conventional care. If this pilot is successful, the overall 
        cost of medical therapies should be reduced justifying future 
        payment by insurers for CPSN services.

        When operational, this system will be able to conduct 
        prospective post-marketing surveillance for drugs. Unlike 
        current systems, it will be able to determine the denominator 
        (number exposed) for adverse events and, when comparator drugs 
        are available, comparative safety of drugs will be determined. 
        The CPSN will be able to determine how drugs are being 
        prescribed and if there is evidence of their effectiveness in 
        new diseases or uses. Because CPSN will provide greater 
        assurance that drug safety will be effectively monitored and 
        adverse events detected earlier, it will be possible for the 
        FDA to accelerate the approval of new drugs without 
        compromising the public safety. The CPSN will also enable 
        patients to play an active role in the management of their 
        therapies reducing the risk of preventable adverse events and 
        hastening the detection of unanticipated adverse reactions.

    Question 2. Please describe how our responses to bioterrorism and 
to drug safety might overlap and complement one another?
    Answer 2. Because of the almost unlimited number of biologicals 
that can be used in bioterrorism, it will be necessary to have the 
capacity to develop a broad range of preventative or therapeutic drugs 
quickly. It is unlikely that there will be adequate time to test the 
drugs completely, so it will be essential that we have a work force 
prepared and trained in drug safety analysis and surveillance to 
participate in the evaluation of any new agents employed in the 
response to bioterrorist attacks. The Community Based Pharmacy Network 
and other needed programs can be essential elements in a plan for being 
prepared for safety surveillance. The same curriculum used to certify 
pharmacists and pharmacy technicians can be used to train members of 
the military to conduct safety assessment of any new drugs being 
developed for bioterrorism.

                                 ______
                                 
    Response to Questions of Senator Enzi From Bruce M. Psaty, M.D.

    Question 1. You support creating an independent Office of Drug 
Safety. This could result in an office that only looks at problems. 
Isn't it more appropriate for risks and benefits to be considered 
together?
    Answer 1. It is appropriate, even essential, for the new 
independent Center for Marketed-Drug Evaluation and Research (CMDER) to 
consider both risks and benefits. The FDA needs to be reorganized to 
achieve this new division of labor: (1) the Office of New Drugs (OND) 
in the Center for Drug Evaluation and Research (CDER) reviews New Drug 
Applications (NDAs) or supplemental NDAs (sNDAs) and assesses the risks 
and benefits of drugs before approval; and (2) the new CMDER reviews 
and assesses the risks and benefits of prescription drugs after 
approval. Consultation and coordination between CDER and CMDER around 
the time of drug approval will be important. At the time of approval, 
authority to regulate the new drug will pass from CDER to CMDER. 
Reorganization of the FDA to achieve this new division of labor--pre-
marketing evaluations by CDER and post-marketing evaluations by CMDER--
will eliminate the conflict of interest that OND currently has in 
evaluating the post-marketing safety of drugs that OND had approved in 
the first place.
    CDER's OND is effective at reviewing and approving new drugs, and 
many of the FDA reviewers there are skilled in conducting the pre-
approval reviews. But when drugs go on the market, they are used by 
large numbers of people, many of whom would not have been eligible for 
the pre-approval trials. What the American public needs and deserves, 
in addition to the rapid approval of drugs, is a new FDA CMDER whose 
primary mission, vision and values are geared toward monitoring and 
assessing the safety and efficacy of drugs that are on the market. 
Post-marketing surveillance is a different mission, and needs to be a 
separate but equal partner with CDER's pre-approval evaluation. The new 
CMDER will need the skills of public health scientists, epidemiologists 
and ethicists to evaluate the risks and benefits to the population as a 
whole. Provided with new authorities, the independent CMDER can pursue 
aggressively key safety questions that industry would sometimes prefer 
to ignore and protect the health of the public by considering not only 
the risks but also the benefits in the populations that actually use 
the new drugs. Regular congressional oversight can help to assure the 
quality of FDA efforts at both CDER and CMDER and provide an important 
forum for the discussion of drug efficacy and safety.

    Question 2. You have indicated that the Office of Drug Safety is 
underfunded. Could you comment on the adequacy of the President's 
proposed fiscal year 2006 budget for FDA, which includes a 24 percent 
increase for FDA's post-market safety program? If this figure is not 
adequate, how much funding would it take?
    Answer 2.The proposed increase of $6.5 dollars (24 percent 
increase) for the FDA's post-marketing safety program in fiscal year 
2006 is not adequate. The first 10 years of PDUFA, between 1992 and 
2002, did not permit user fees to be used for drug safety activities. 
According to the General Accounting Office report on the effects of 
user fees: \1\ ``The implementation of PUDFA has been successful in 
bringing new drugs and biologics to the U.S. market more rapidly than 
before. However, maintaining adequate funding for approving new drugs 
and biologics has had the unintended effect of reducing the share of 
funding and staffing for other activities'' including drug-safety-
related activities. ``According to FDA officials,'' the GAO report 
continues, \1\ ``safety problems not detected in clinical trials are 
more likely to be found first among U.S. patients because they are 
increasingly first to have access to new drugs.'' In 1992, when America 
began to serve as a major drug-safety testing ground, there was little 
or no attention to enhancing drug safety efforts. According to Dr David 
Kessler, \2\ former head of the FDA, ``PDUFA should have had funding on 
the safety side from the beginning, but the industry refused to accept 
that . . . . We wanted it. The industry said no.'' The development of 
infra-structure, described in response to Senator Hatch's first 
question, remains a major unmet need and will require considerable 
investment.
    In the preparation of the March 2003 report, ``FDA's Review Process 
for New Drug Applications,'' \3\ the Office of the Inspector General 
conducted a survey of 401 CDER reviewers. Fully ``66 percent of FDA 
respondents indicated on our survey that they were somewhat or not at 
all confident that the FDA adequately monitors the safety of 
prescription drugs once they are on the market.'' \3\ The lack of trust 
and confidence in the FDA post-marketing surveillance system, expressed 
by FDA reviewers, is shared by many independent scientists. The FDA's 
meager post-marketing surveillance system is further stressed by the 
mounting challenges. In the 8-year period between 1995 and 2003, the 
number of post-marketing adverse event reports received by the FDA 
MedWatch adverse-event reporting system increased by 137 percent, from 
156,477 in 1995 to 370,887 in 2003.\4\ \5\ In the 4 year period between 
1997 and 2001, retail spending on prescription drugs almost doubled, 
increasing from $78.9 billion in 1997 to $154.5 billion in 2001.\6\ 
Over the same time period, the 50 percent increase in the FTE at the 
Office of Drug Safety lagged far behind the huge increase in the 
exposure of the U.S. population to prescription drugs.
    The current needs of the new Center for Marketed Drug Evaluation 
and Research represent not just ``current needs'' but also an array of 
previously unmet needs, some of which have accumulated for more than a 
decade. A serious effort to improve drug safety in America will require 
a substantial investment. Infrastructure and collaborations for safety 
studies that have been ignored for a decade need to be developed. In 
order to protect the health of the public, the budget for the new CMDER 
will need to move, over the next several years, close to the level of 
the budget for the OND. Required funding levels will depend in part on 
the existence of new authorities. If the CMDER cannot compel the 
pharmaceutical industry to conduct key post-marketing studies and if 
CMDER has to fund the conduct of independent studies, the funding 
requirements for drug safety would be especially large.

    Question 3. You support a mandatory Federal registry of clinical 
trials. How would you set up a results database so that the information 
is actually useful for patients and providers?
    Answer 3. The International Committee of Medical Journal Editors 
has called for clinical trial registration and for ``full transparency 
with respect to the performance and reporting of clinical trials.'' \7\ 
The rationale is important. To use drugs wisely, patients and 
physicians need to know about risks and benefits. Risk-benefit analyses 
need complete and accurate information about all relevant studies. The 
primary purpose of a Federal clinical trials registry is to assure that 
all randomized trials are fully reported. In the absence of a mandatory 
registry of clinical trials, the pharmaceutical industry occasionally 
finds it difficult to resist the marketing instinct to conceal studies 
with unfavorable findings. In a study of Alzheimer's patients, for 
instance, Pfizer's Celebrex (celecoxib) increased the risk of 
cardiovascular events.\8\ Although the study was completed in 1999, its 
results were not submitted to the FDA until June 2001, several months 
after a safety review that established new labeling for Celebrex. The 
findings, which were never published, were finally posted on an 
industry website in January 2005.\8\ This selective-non-publication 
approach creates a distorted knowledge base so that risk-benefit 
analyses of Celebrex cannot take into account the cardiovascular harm 
detected in this study.
    For these reasons, the results database of the Federal clinical 
trials registry will be most useful to scientists who want to conduct 
reviews and meta-analyses of all the existing studies of a particular 
drug. The available data and the data structures must be adequate for 
and useful to these scientists. To make each entry in the database 
useful for patients and providers, the trial results should include a 
clear statement of the original hypotheses, its study population, the 
primary findings in quantitative terms, secondary analyses and safety 
analyses. The findings from one trial, however, are not adequate to 
make informed treatment decisions. Instead, treatment decisions should 
be based on the totality of the evidence about a drug's safety and 
efficacy. In other words, patients and providers would do well to seek 
out independent reviews and meta-analyses of the studies conducted on 
the drug of interest. Journals often make some of these major high-
quality reviews available on their websites without cost to non 
subscribers. In December 2004, moreover, Consumer's Union launched 
``Consumer's Reports Best Buy Drugs,'' which is ``a major new public 
education program that is designed to provide unbiased information 
about the comparative effectiveness and cost-effectiveness of 
prescription drugs.'' \9\ This service is ``provided free to 
consumers.'' The effort to integrate information on the safety and 
efficacy of new drugs in an independent and unbiased fashion is also an 
important role for the new CMDER. It is these integrated independent 
reviews, and not individual entries in the clinical-trials database, 
that will be most useful to providers and patients.

    Question 4. It seems to me that achieving the right balance between 
the benefit and risk is the key challenge both of approving a new drug 
and of deciding whether to keep it on the market. What would you 
suggest could (or should) be done differently so that benefit-risk 
decision can be made in the best possible way?
    Answer 4. The best possible way to make risk-benefit decisions is 
to have available good scientific evidence from randomized clinical 
trials that assess the risks and benefits with equal scientific rigor. 
The pharmaceutical industry expends enormous energy in the effort to 
demonstrate a drug's efficacy. For the purpose of evaluating efficacy, 
industry studies are generally well designed and adequately powered. 
But the industry efforts to identify and quantify risks of a drug are 
modest or less. To an independent scientist, for instance, the known 
biologic effects of the COX-2 inhibitors would suggest two different 
hypotheses--the possibility of benefit to the stomach and the 
possibility of injury to the heart. While many Vioxx (rofecoxib) 
studies were well designed to assess the potential benefits of pain 
relief or protection of the stomach, these same studies often minimized 
the chances of finding any cardiovascular harm. In general, they were 
small, lasted a few weeks or months, excluded patients with heart 
disease or patients who used aspirin, and paid little specific 
attention to cardiovascular events. The evidence provided by this 
efficacy-driven marketing-focused approach to science--well-studied 
benefits and ill-defined risks--undermines the validity of a genuine 
risk-benefit analysis.
    The FDA has two major opportunities to help industry address these 
deficiencies. In the pre-approval stage, the OND can insist that 
clinical trials examine adequately risks as well as benefits. In 
addition to the traditional collection of adverse event reports, 
specific safety outcomes can be defined in advance and assessed in a 
standardized fashion. Large long-term trials of chronic-disease 
medicines should be started as early as possible in the drug-approval 
process. The OND can also insist that patients included in the pre-
approval studies are representative of those who will actually use the 
new drugs. In the post-marketing setting, the new CMDER can address the 
key questions that remained unanswered at the time of approval. 
Although 6-week studies may be adequate to demonstrate that drugs such 
as Vioxx reduce the pain of arthritis in the knee or hip, pain 
medicines do not cure osteoarthritis, and large numbers of patients 
with arthritis will take these drugs for many years. Osteoarthritis is 
common in older adults, many of whom have cardiovascular disease and 
take aspirin. Under these circumstances, for instance, CMDER can insist 
on the conduct of post-marketing studies or clinical trials that 
evaluate the new drug in the patients who actually take them and in the 
way that these patients actually take the new drugs. In the post-
marketing setting, CMDER needs at once to assure that the evaluations 
of marketed drugs match the way that the marketed drug is used in the 
U.S. population and, at the same time, to work with physicians and 
patients to limit the use of the marketed drug to the approved 
indications and populations. In short, the FDA needs to take an pro-
active role in assuring the quantity and quality of the data for risk-
benefit analyses, both before and after approval.
                                 ______
                                 
     Response to Questions of Senator Hatch by Bruce M. Psaty, M.D.

    Question 1. Do you think that the current post-marketing 
evaluations are adequate? Could you suggest some specific ways in which 
they should be improved?
    Answer 1. The current MedWatch program is adequate for only one 
minor, though important, drug-safety effort.\10\ This voluntary 
reporting system is suitable to identify rare serious adverse drug 
events that occur early in treatment and that are unrelated to the 
indication of the drug. For example, the lipid-lowering statin drug, 
Baycol (cerivastatin), was withdrawn from the market in 2001 because it 
was associated with high rates of rhabdomyolysis, a breakdown of muscle 
cells that causes pain, kidney failure and sometimes death.\11\ \12\ 
\13\ It would not have been possible to use the MedWatch system to 
detect reliably, for instance, the increased risk of cardiovascular 
events associated with the COX-2 inhibitors.
    The MedWatch adverse drug-reaction data--recently characterized as 
``fundamentally a 1950s-era approach'' \14\ lack many of the features 
of high-quality epidemiologic studies, including validation of events 
by standard criteria, complete ascertainment of cases, population-based 
controls, comparable assessment of drug use and risk factors, and so 
forth. In short, the current post-marketing surveillance systems are 
not adequate to meet the new surveillance challenges posed by the rapid 
approval of new drugs. Several opportunities deserve to be cultivated. 
First, data from new Medicare drug benefit can be linked with hospital 
and ambulatory care data to create a new resource for the study of 
drugs in older adults. With appropriate protections for privacy, these 
data should be available to the FDA and independent scientists 
interested in drug safety. Secondly, as part of the NIH Roadmap 
Project, the HMO-Research Network--Coordinated Clinical Studies Network 
will create an infrastructure for conducting studies on substantial 
numbers of the U.S. population, and the movement toward an EPICcare 
based electronic record among the network members should soon provide 
the opportunity to conduct post-marketing surveillance rapidly and 
efficiently.
    The development of these ``research resources'' needs to be 
complemented by a new public-health vision and sense of mission at the 
FDA. As indicated in the response to Senator Enzi's questions, 
pharmaceutical companies sometimes lack enthusiasm for pursuing 
questions about the safety of their drugs. Although the risk-benefit 
analysis may appear favorable in the well persons typically recruited 
to the pre-approval trials, the populations who take new drugs include 
many groups such as the elderly and those with other serious health 
conditions. As a result, the risk-benefit assessments may differ 
markedly in the post-marketing setting from those of the pre-approval-
trials setting.
    At time of approval, the new independent CMDER should review the 
NDA and all available data, published and unpublished, to identify the 
set of studies required to address the key unanswered questions, 
particularly the pursuit of potential safety ``signals'' or ``plausible 
biologic hypotheses'' on behalf of the health of the public. The 
current FDA risk-management approach places on the manufacturer the 
responsibility for defining the safety questions of interest and 
designing the studies to answer those questions. Rather than serve as 
simply a reviewer of industry's proposals, the FDA's CMDER should 
actively select the questions, the studies, and the designs that merit 
attention. CMDER's work in this regard should be peer-reviewed by 
independent scientists to assure that the key questions are properly 
identified and that the studies required to address them are properly 
designed. Depending on the drug, the indication and the known safety 
profile, the studies may include Phase IV trials, epidemiologic 
studies, pharmacokinetic-pharmacodynamic studies, close surveillance of 
ADR reports, or a combination of several approaches. Specific post-
marketing trials or studies should be designed, conducted and completed 
in a timely fashion. Provisional approval may be another useful 
approach. The CMDER should be responsible for assessing the balance of 
risk and benefit of drugs that are on the market.

    Question 2. In your testimony, you state that since 1992 funding 
for drug safety has dwindled. Yet in her testimony, Dr. Sandra Kweder 
of FDA stated that resources devoted to drug safety, both human and 
financial, have steadily increased over the past decade. For instance, 
the ODS budget increased from around $7 million in fiscal year 1996 to 
a proposal of more than $33 million in fiscal year 2006. Likewise, ODS 
employment has increased during that time from 52 FTEs in fiscal year 
1996 to 137 FTEs in fiscal year 2006. Moreover, PDUFA III authorized 
FDA to use some of its user fee money for risk management activities, 
and PDUFA resources will represent nearly one third of the ODS budget 
for the coming fiscal year. Doesn't this increased funding--both from 
PDUFA and otherwise--evidence a strong and continuing commitment to 
drug safety?
    Answer 2. As I have indicated in responses to the questions from 
Senator Enzi, the efforts since 1992 to increase the speed of approval 
for new drugs have not been accompanied by commensurate improvements on 
the drug safety side. According to the General Accounting Office 
report, ``FDA reduced staffing levels for non-PUDFA activities each 
year, leaving the agency fewer resources to perform its other 
responsibilities,'' including safety-related activities.\1\ The old 
system not only produced a ``drug lag'' that kept new medicines from 
U.S. patients, but it also provided a ``safety buffer'' that protected 
U.S. patients. New drugs would first come to market in Europe; safety 
problems would be identified there; and the problematic drugs--
practolol and thalidomide are examples--would never come to market in 
the U.S. . Now, with America as the new drug-safety testing ground, the 
``safety buffer'' has been replaced by a ``safety lag.'' Today, U.S. 
patients remain at risk of large-scale injuries such as the tens of 
thousands of heart attacks and strokes caused by Vioxx. The U.S. drug 
safety systems were not originally upgraded to meet the new challenges 
of PUDFA in 1992; infra-structure and collaborations have not been 
developed to conduct new studies rapidly or efficiently; and in recent 
years, the spending on prescription drugs and the reporting of adverse 
events have expanded much faster than the resources provided to ODS. 
The budget increases thus remain inadequate.

    Question 3. In the supplementary information provided with your 
testimony, you suggest that pharmaceutical companies are not complying 
with their obligations to conduct Phase IV studies. Yet in a recently 
published report on the performance of drug and biologic firms in 
conducting post-marketing (Phase IV) studies, FDA finds that, of the 
studies concluded between October 1, 2003 and September 20, 2004, no 
studies were identified where the commitment was not met. Likewise, the 
study indicates that only 1 percent of the pending studies for NDAs and 
ANDAs were delayed. Doesn't this report indicate that pharmaceutical 
firms are doing a good job meeting their post-marketing study 
commitments in a timely manner?
    Answer 3. Under the Food and Drug Modernization Act (FDAMA) of 
1997, the FDA is required to report annually in the Federal Register on 
the status of postmarketing study commitments made by manufacturers of 
approved drug and biological products. The most recent report includes 
information available through September 30, 2004.\15\ The following 
table summarizes the status of the open post-marketing commitments for 
New Drug Applications (NDA) and Abbreviated NDAs (ANDA).

                 Table. Open Post-Marketing Commitments
------------------------------------------------------------------------
                         Status                              NDA/ANDA
------------------------------------------------------------------------
Pending (study not yet initiated, not delayed).........        812 (68%)
Ongoing (proceeding on schedule).......................        219 (18%)
Delayed (behind schedule)..............................          15 (1%)
Terminated (study ended, no final report submitted)....           2 (<1)
Submitted (study ended, final report submitted)........        143 (12%)
Total number of open commitments.......................      1191 (100%)
------------------------------------------------------------------------

    The number of open post-marketing commitments, currently at 1191, 
remains large (Table). Although 143 (12 percent) were submitted and 
another 234 (19 percent) were ongoing or delayed, a total of 812 (68 
percent) have not yet been initiated. It appears that if these 
``pending studies'' had no ``original schedule'' at the time of the 
original agreement to conduct them, they can never be delayed, and some 
may remain in the ``pending'' category in perpetuity. The large number 
of ``pending'' studies is a concern.
    The tardiness with post-marketing commitments can be most clearly 
illustrated by the accelerated approval mechanism, where the post-
marketing studies are essential to defining what, if any, clinical 
benefit may derive from drugs approved on the basis of surrogate end 
points. In testimony before the Senate Health, Education, Labor and 
Pensions Hearing on the ``FDA's Drug Approval Process: Up to the 
Challenge?'' on March 1, 2005 and in a recent article, \16\ Dr Fleming 
raised concerns about the quality and the timing of post-marketing 
commitments associated with the accelerated approval (subpart H) 
regulatory process. ``One of the more disturbing facts revealed in that 
meeting (ODAA, March 2003),'' writes Fleming, ``was that the average 
time between the granting of marketing through AA [accelerated 
approval] and the completion of on-going validation trials for these 
eight products was projected to be 10 years . . . . Furthermore, after 
receiving authorization to market the product, the sponsor often has a 
loss of the sense of urgency that in the premarketing settings is a 
powerful driving force for the sponsor to obtain timely evaluation of 
the benefit-to-risk profile of the intervention.'' \16\ Dagher and 
colleagues discuss some of the same accelerated-approval cancer drugs 
in their review.\17\ For the 15 cancer drugs that were granted 
accelerated approval between 1992 and 2004, postmarketing studies that 
used a clinical outcome were required for full approval, and to date, 
only 6 (40 percent) of the 15 have met their post-marketing 
requirements.\17\ This leisurely approach to completing post-marketing 
commitments is not adequate.
                                 ______
                                 
    Response to Questions of Senator Kennedy by Bruce M. Psaty, M.D.

    Question 1. Pharmaceutical manufacturers must pay attention to risk 
and benefit when designing drugs and deciding what to take to market. 
Is it appropriate to give them the responsibility for drug safety after 
marketing?
    Answer 1. This approach assigns the fox to guard the hen house. In 
the case of Baycol (cerivastatin), a lipid-lowering statin drug, the 
manufacturer was unable to overcome the inherent conflict of interest 
in interpreting and responding to the adverse event reports that 
appeared soon after the drug was launched.\13\ Several opportunities 
were missed to protect patients and prevent injuries. The manufacturer 
was slow to respond to signals, failed to conduct key safety studies in 
a timely fashion, decided not to publish a high-dose study with 
unfavorable findings, and ignored the safety recommendations of its own 
scientists and epidemiologists.
    Soon after the Baycol was on the market, Bayer scientists conducted 
excellent analyses of the FDA's MedWatch data. They identified 
rhabdomyolysis as a major adverse effect of Baycol, used either alone 
or in combination with Lopid (gemfibrozil). According to a memo from 
Bayer scientists in March 2000, ``The findings [of internal analyses] 
indicate that in patients receiving monotherapy, cerivastatin 
substantially elevates risk for rhabdomyolysis compared with other 
statins. In combination with gemfibrozil, cerivastatin patients were 
also found to be at a remarkable disadvantage compared with patients 
receiving gemfibrozil with another statin.'' \13\ This information was 
not, however, communicated to patients, physicians or the FDA.
    These safety findings were reported to David Ebsworth, who headed 
Bayer AG's Pharmaceuticals Business Group. In a memorandum and order 
regarding the Bayer AG Securities Litigation, District Judge William H. 
Pauley III summarizes the response to these safety concerns: ``On 
August 2, 2000, senior members of Bayer's Global Drug Safety team and 
consultants met with Plischke to discuss the accumulation of adverse 
event reports. A consensus emerged that the data concerning Baycol's 
dangers `was putting the brand at risk.' When that conclusion was 
communicated to Ebsworth, he dismissed the reservations of the safety 
experts and instructed his marketing team `to promote the hell out of 
this product.''' \12\ Baycol remained on the market for another year. 
By 2000, annual drug sales for Baycol were supposed to be several 
hundred million dollars and were soon expected to reach block-buster 
proportions of $1 billion per year. Ebsworth's one-sided attention to 
drug sales rather than drug safety delayed the withdrawal of the Baycol 
and harmed patients. Hence, the need for a strong Center for Marketed-
Drug Evaluation and Research to protect the health of the public.

    Question 2. You have advocated a separate Center for Drug Safety, 
within the FDA, for post-market assessments. Why do you feel a separate 
center is needed? Some seem concerned that a separate safety center 
would make decisions strictly based on determinations of safety, 
without considering the benefits of drugs. Is that a valid concern? How 
might we ensure that all decisions made are both unbiased and based on 
a weighing of risk against benefit?
    Answer 2. The Office of New Drugs dominates CDER. In the last year, 
safety concerns about antidepressants and COX-2 inhibitors did not find 
an open forum for expression at the FDA. The report by the Office of 
the Inspector general points to the fact that ``21 percent of FDA 
respondents indicated that the work environment allowed for the 
expression of differing scientific opinions to a small or no extent.'' 
\3\ The current structure and culture at the FDA is just what industry 
desires--a powerful engine to approve drugs and a weak effort to 
investigate safety in the post-marketing setting. What the American 
public needs and deserves, in addition to the rapid approval of drugs, 
is a Center whose mission is devoted to post-marketing evaluations of 
the safety and efficacy of marketed drugs.
    In a recent commentary, the JAMA editors point out the inherent 
conflict of interest when the OND reviews its own approval decisions: 
``It is unreasonable to expect the same agency that was responsible for 
approval of drug licensing and labeling would also be committed to 
actively seek evidence to prove itself wrong (ie, that the decision to 
approve the product was subsequently shown to be incorrect).'' \18\ 
Although a new CMDER would still be within the FDA, the separation of 
the CMDER from CDER is necessary to assure the independent review of 
drug safety questions. William Schultz, the FDA's deputy commissioner 
for policy from 1994 to 1998, agrees: ``FDA should separate the 
monitoring of drugs after they have been approved from the drug review 
function . . . . The post-market function should be separated from the 
drug review function.'' \19\
    Several key FDA leadership positions have been vacant in recent 
years. In public statements, some FDA officials have occasionally 
seemed to lack a public-health perspective on the balance of risks and 
benefits. For instance, two senior CDER officials called epidemiologic 
estimates of Vioxx injuries ``junk science'' \20\ and claimed that the 
deaths caused by Vioxx were ``not real deaths.'' \20\ As the FDA 
Advisory Committee indicated in mid February, the COX-2 inhibitors do 
increase the risk of heart attack and stroke. Tens of thousands of 
patients were injured, and many died. The CDER leadership has yet to 
offer an explanation for why the cardiovascular risks of the COX-2 
inhibitors remained undetected for so many years. The studies designed 
by Merck for the approval of Vioxx were individually sound. But as a 
group, they minimized the possibility of finding cardiovascular harm, 
which was biologically plausible on the basis of the known actions of 
Vioxx. The FDA missed an opportunity to protect the health of the 
public by failing to insist at several stages that the company attend 
carefully to potential risks as well as benefits. Without an 
appreciation for the opportunities for prevention that may have been 
missed, it is not clear that the CDER leadership can avoid future 
Vioxx-like drug disasters. To protect the health of the public, the FDA 
needs an independent CMDER whose primary mission, vision and values are 
geared toward monitoring and assessing the safety and efficacy of drugs 
that are on the market. Passing the responsibility from CDER to CMDER 
at the time of approval also avoids the potential conflict of interest 
mentioned by the JAMA editors.

    Question 3. Has the FDA done a good job in assessing risk versus 
benefit for new drugs? How about for approved drugs in the post-
marketing setting? Can you explain any similarities or differences?
    Answer 3. In the pre-approval setting, the FDA often does a good 
job of assessing what is reported by the company about the risks and 
benefits of new drugs. On the basis of the pre-approval studies, 
however, the FDA often has limited data on the risks and benefits of 
drugs that will be used long term for chronic conditions such as 
arthritis. People use arthritis pain medications for many years, but 
before Vioxx was approved, several thousand patients received the drug 
usually in small trials that were designed to assess pain relief over 
the course of a few weeks or months.\21\ Only about 750 patients 
received usual doses of Vioxx, 12.5 mg or 25 mg per day, for a year or 
longer. Even though the FDA medical officer observed a three-fold 
increase in the risk of ``thromboembolic events'' in short-term 
studies, \22\ Vioxx was approved and eventually used by many patients--
those with heart disease and those taking aspirin--who had often been 
excluded from the pre-approval trials. Large long-term clinical trials 
of drugs that will be used by millions of Americans for many years are 
best started as early as possible in the drug-approval process. This 
public-health approach, which need not slow the drug approval process, 
recognizes and anticipates how the drugs will be used in the population 
once they are approved.
    In the post-marketing setting, some OND decisions seem more 
industry-friendly than public-health friendly. For instance, in 
clinical trials of the 0.8 mg supplemental NDA for Baycol, the findings 
suggested a high likelihood of harm to about 7 percent of thin elderly 
women.\23\ Although the manufacturer needed this dose to compete with 
Lipitor (atorvastatin), the public-health rationale for approving the 
0.8 mg dose, given the known risks, was not clear. Within a year, 
Baycol was withdrawn from the market because of rhabdomyolysis, a 
serious adverse event that was especially common at high doses. A 
second example is Rezulin (troglitazone), a diabetes drug, that was 
associated with episodes of liver failure and death soon after 
marketing. The drug was withdrawn from the market rapidly in the United 
Kingdom.\24\ In the United States, the FDA and the manufacturer 
experimented with a series of recommendations and label-revisions that 
did not work.\25\ \26\ Rezulin was eventually withdrawn from the U.S. 
market in 2000.\1\ Finally, in the VIGOR trial, \27\ 50 mg dose of 
Vioxx was associated with a five-fold higher risk of heart attack than 
naproxen. With the large number of effective low-risk pain medications 
then available, it is not clear how the FDA's formal risk-benefit 
analysis could have favored keeping the 50 mg dose of Vioxx on the 
market for the indication of acute pain relief.
    The issue is not just weighing risk versus benefit but also the 
quantity and quality of evidence about risk and benefit. In the post-
marketing setting, industry tends to do studies that generate good news 
for the marketing departments. Those studies, which focus on benefits, 
are often cleverly designed to generate little useful information about 
risk. This asymmetry in evidence--well-studied benefits and ill-defined 
risks--undermines the validity of the knowledge base. In the post-
marketing setting, the FDA needs a strong and independent CMDER to 
pursue aggressively key safety questions that industry would sometimes 
prefer to ignore. I have not seen this sense of mission from CDER 
leadership in recent years.

References

    1. United Sates General Accounting Office. Report to the Chairman, 
Committee on Health, Education, Labor, and Pensions, U.S. Senate: Food 
and Drug Administration: Effect of User Fees on Drug Approval Times, 
Withdrawals, and Other Agency Activities. GAO-02-959, September 2002. 
Http: //www.gao.gov/new.items/d02958.pdf.
    2. Graham J, James F. Flaws in drug agency put consumer at risk. 
The Chicago Tribune, February 20, 2005.
    3. Department of Health and Human Services Office of Inspector 
General. FDA's review process for new drug applications: a management 
review. Department of Health and Human Services; OEI-01-01-00590; March 
2003.
    4. Food and Drug Administration. CDER 2001 report to the nation: 
improving public health through human drugs. Rockville, Maryland. 
http://www.fda.gov/cder/reports/rtn/2001/rtn2001.pdf.
    5. Food and Drug Administration. CDER 2003 report to the nation: 
improving public health through human drugs. Rockville, Maryland.
    6. The National Institute for Health Care Management Research and 
Education Foundation. Prescription drug expenditures in 2001: another 
year of escalating costs. NIHCM Foundation, Washington, DC, April 2002; 
http://www.nihcm.org/index2.html.
    7. DeAngelis C, Drazen JM, Frizelle FA, et al. Clinical trial 
registration: a statement from the International Committee of Medical 
Journal Editors. Hypertension 2005;45:631-632.
    8. Pfizer Inc. A double-blind randomized placebo-controlled 
comparative study of celecoxib (SC-58635) for the inhibition of 
progression of Alzheimer's Disease, protocol IQ5-97-02-001. Http://
www.clinicalstudyresults.org/documents/company-study	76	0.pdf; last 
accessed Jan 29, 2005.
    9. Consumers Union. Statement of Consumers Union of U.S., Inc, on 
``Ensuring Drug Safety: Where do We Go from Here?'' Senate Committee on 
Health, Education, Labor, and Pensions, March 3, 2005.
    10. Kessler DA. Introducing MEDWatch: a new approach to reporting 
medication and device adverse effects and product problems. JAMA 
1993;269:2765-2767.
    11. Graham DJ, Staffa JA, Shatin D, Andrade SE, Schech SD, La 
Grenade L, et al. Incidence of hospitalized rhabdomyolysis in patients 
treated with lipid-lowering drugs. JAMA 2004;292:2585-2590.
    12. Pauley III WH. Memorandum and order in re: Bayer AG securities 
litigation. Southern District of New York, No 03 Civ 1546, Sept 30, 
2004: http://www.nysd.uscourts.gov/courtweb/public.htm.
    13. Psaty BM, Furberg CD, Ray WA, Weiss NS. Potential for conflict 
of interest in the evaluation of suspected adverse drug reactions: use 
of cerivastatin and risk of rhabdomyolysis. JAMA 2004;292:2622-2631.
    14. Strom BL. Potential for conflict of interest in the evaluation 
of suspected adverse drug reactions: a counterpoint. JAMA 
2004;292:2643-2646.
    15. Food and Drug Administration. Report on the performance of drug 
and biologics firms in conducting postmarketing commitment studies; 
availability. Federal Register, February 18, 2005, Vol 70, No 33, pages 
8379-8381.
    16. Fleming TR. Surrogate endpoints and FDA's accelerated approval 
process. Health Affairs 2005;24:67-79.
    17. Dagher R, Johnson J, Williams G, Keegan P, Pazdur R. 
Accelerated approval of oncology products: a decade of experience. JNCI 
2004;96:1500-1509.
    18. Fontanarosa PB, Rennie D, DeAngelis CD. Postmarketing 
surveillance--lack of vigilance, lack of trust [editorial]. JAMA 
2004;292:2647-2650.
    19. Schultz WB. How to improve drug safety. Washington Post 
(washingtonpost.com), December 2, 2004, page A35.
    20. Harris G. FDA failing in drug safety, official asserts. New 
York Times, November 19, 2004.
    21. Psaty BM, Furberg CD. COX-2 inhibitors--lessons in drug safety. 
N Engl J Med 2005;352:1133-1135. Epub February 15, 2005.
    22. Villalba ML. FDA Medical Officer Review of VIOXX (rofecoxib), 
NDA 21-042 (capsules) and NDA 21-052 (oral solution). Http://
www.fda.gov/cder/foi/nda/index.htm.
    23. Food and Drug Administration Medical Review. Center for Drug 
Evaluation and Research Application Number: 20-740/S008/S013 
(cerivastatin). Http://www.fda.gov/cder/foi/nda/2000/20-740S008	Baycol	
medr.pdf.
    24. Mitchell P. Shock as troglitazone withdrawn in UK. Lancet 1997; 
350:1685.
    25. Graham DJ, Green L. Epidemiology of heptotoxicity with 
troglitazone. Office of Postmarketing Drug and Risk Assessment, Center 
for Drug Evaluation and Research, FDA, FDA hearing, March 26, 1999.
    26. Gale EA. Lessons from the glitazones: a story of drug 
development. Lancet 2001;357:1870-1875.
    27. Bombardier C, Laine L, Reicin A, et al for the VIGOR Study 
Group. Comparison of upper gastrointestinal toxicity of rofecoxib and 
naproxen in patients with rheumatoid arthritis. N Engl J Med 
2000;343:1520-1528.

    [Whereupon, at 11:44 a.m., the committee adjourned.]

