RSS Feeds
-
Building the Evidence Base Beyond Regulatory Approval
Submit responseAn August 19, 2008 Annals of Internal Medicine article, The ADVANTAGE Seeding Trial: A Review of Internal Documents"[1] by Kevin Hill and colleagues raises questions about research at pharmaceutical companies extending well beyond their appropriate concern regarding “seeding” clinical experience through the guise of a clinical trial. Most notably, Hill et al question the scientific merit of all research after a medicine’s approval beyond that required by the FDA. In doing so, they ignore some nearly universal gaps in knowledge regarding the safety and effectiveness of new medicines at the time of approval in the US.
Why is additional research needed post approval? Approval of medicines in the US typically requires evidence of efficacy and safety in carefully diagnosed populations treated according to well defined and fixed treatment protocols in comparison to placebo. This evidence leaves many critical questions unanswered about most new medicines: how does it compare to other treatments? How does it perform in a setting where patients have a variety of concomitant illnesses and treatments or do not always comply with treatment? These questions are increasingly echoed by insurers, public health administrators, and federal legislators who now call for studies involving representative populations treated in typical practice settings, and relevant comparator treatments.[2,3] The Medicare Prescription Drug Act allowed up to $50 million for research on “comparative effectiveness”, although clearly this will not be sufficient to address the need. To the question posed by Drs Sox and Rennie [4] in an accompanying editorial, “Why would a drug company go to the expense and bother of conducting a trial involving hundreds of practitioners – each recruiting a few patients – when a study based at a few large medical centers could accomplish the same scientific purposes much more efficiently? – we would respond “so that doctors have medical evidence appropriate to community-practice patients and so that healthcare systems have the evidence needed to determine access to medical innovation”.
Ultimately, we agree with the conclusions reached by Sox that the 'goodness’ of any trial is inherent in whether it addresses an important, new question in a well designed study. Accordingly, the decision to conduct a trial is a serious responsibility and one that should be reached through input from a diverse group of experts in and outside of pharmaceutical companies.
Ignoring or discounting appropriately designed and conducted trials denies physicians important evidence on which to base treatment decisions.
REFERENCES:
1. The ADVANTAGE seeding trial: a review of internal documents. Hill KP, Ross JS, Egilman DS, Krumholz HM. Ann Intern Med 2008; 149:251-258.
2. Tunis, S.R., D.B. Stryer, and C.M. Clancy, Practical clinical trials: increasing the value of clinical research for decision making in clinical and health policy. JAMA, 2003. 290(12): p. 1624-32.
3. Wilensky, G.R., Developing a center for comparative effectiveness information. Health Aff (Millwood), 2006. 25(6): p. w572-85.
4. Seeding trials: just say "no". Sox HC and Rennie D. Ann Intern Med 2008; 149:279-180.
Conflict of Interest:
Authors are employees of Pfizer
-
Re: Seeding bias vs. clinical trials aimed at improving health
Submit responseEthical clinical trials should be aimed first at improving patient or public health by learning the truth(s) about new drugs, and comparing them with established alternative treatments. If this is true, then ethics review boards which approve experiments designed primarily to benefit third parties might themselves be considered "unethical".
Why are relatively few non-conflicted doctors involved in the enrolment and assessment of patients for controlled trials? An obvious reason is that pharmaceutical companies orient their experiments, as well as their largesse, to doctors perceived to be "friendly" opinion leaders. They are less likely to seek out principal or co-investigators known for their independent judgement or scepticism, let alone physicians who might criticise a new drug were it to prove inferior.
The solution is to return to the roots of ethical medical science. We need to design more high quality trials and fund them publicly, so that we base therapeutic strategies on high quality science rather than "junk science". Ultimately this strategy would be far more beneficial for individual patients and for society, and much less costly for those who pay for health services.
Many physicians, both academic and non-academic, might welcome the chance to enrol patients in truly scientific trials. In recent years, UKPDS (diabetes) , HOPE (cardiovascular disease), or AD2000 (dementia), each provided widely useful insights not available from the typical marketing trials.
It's time for our professional societies and the leaders of clinical trial design to mobilize the rest of us. Why not reclaim control of the research agenda? After all, we doctors will one day be patients and have the same interest as anyone else in knowing objectively what are the best treatments.
Conflict of Interest:
"Dr. Perry has been paid for expert opinion as a consultant to plaintiff in litigation against Pfizer Inc. in regard to alleged off-label marketing of gabapentin for pain."
-
Seeding bias
Submit responseMulticentre trials arose in the 1960s in response to two needs. Firstly the need for power, to accumulate sufficient subjects in a timely manner to meet the aims of the study. Secondly the need to minimise bias due to investigator or site specific factors not otherwise identifiable. Current study designs meet the first need, but sadly fail to meet the second need.
Site or investigator bias can be detected by testing that the estimated effects are consistent across sites and or investigators. Power to achieve this is dependent on numbers of events & subjects at each site and is greatest with many subjects from a few sites. However most "modern" multicentre trials recruit small numbers of subjects at many sites. With many sites investigator bias is more likely to occur and less likely to be detected than with fewer sites.
Multicentre trials are thus usually 1. unable to detect site or investigator bias; 2. more likely to be affected by site or investigator bias. Such multisite study designs are also more expensive. Therefore why are they done?
One answer is improved validity through wider subject representation and ability to detect e.g. interethnic differences. However, a more plausible answer is marketing.
Any multicentre study with large numbers of investigators and sites has marketing advantages and scientific disadvantages. In particular seeds of bias from sites or investigators are likely to occur and unlikely to be detected.
Conflict of Interest:
None declared
-
Advantage Seeding Trial: A Review of Internal Documents
Submit responseTo the Editor:
In light of the recent publication by Hill, et. al.(1), I feel that it is appropriate for me to clarify my role in the ADVANTAGE trial (2). The trial was performed in the late 1990s. I was the principal investigator for this trial at my site at the University of Texas Medical Branch. The study was approved by our Institutional Review Board before we participated in the study.
After the initial draft of the paper, preparation for submission and publication was a collaborative effort among the authors, including myself, other co-investigators, and collaborating authors at Merck.
As noted in a letter by Braunstein and Polis (3), the decision to report cardiovascular outcomes was trusted to “external blinded panels of medical specialists.” The numbers of events reported in the original article published in the Annals of Internal Medicine were correct to the best of my knowledge when the article was published. Other information from other sites that I was not privy to and any other conclusion was solely the responsibility and determination of Merck.
I participated in the article’s preparation and publication because I thought and still believe ADVANTAGE addressed a valid scientific question. The editors and external reviewers of the Annals agreed with this by publishing the article as did at two internationally attended scientific meetings that accepted scientific abstracts for presentation based on data from the study (4,5).
References
1. Hill, K.P., Ross, J.S., Egliman, D.S., Krumholz, H.M. The ADVANTAGE Seeding Trial: A Review of Internal Documents. Annals Int. Med. 2008;149:251-8.
2. Lisse, J.R., Perlman, M., Johansson, G., Shoemaker, J.R., Schechman, J., Skalky, C.S., Dixon, M.E., Polis, A.B., Mollen, A.J., Geba, G.P. Gastrointestinal Tolerability and Effectiveness of Rofexcoxib, versus Naproxyn in the Treatment of Osteoarthritis. Annals. Int. Med. 2003;139:539-546.
3. Braunstein, N., Polis, A. Report of Specific Cardiovascular the Outcomes ADVANTAGE Trial. Annals Int. Med. 2005;143:158-159.
4. Geba GP, Lisse JR, Perlmam M, Polis AB, Dixon ME, et al; Gastrointestinal tolerability in primary care patients treated with naproxen or rofecoxib for osteoarthritis: the advantage trial; Annals of Rheumatology Disease; July 2001; Vol. 60, (Suppl.) 1, SAT0096.
Conflict of Interest:
None declared
-
Re: ADVANTAGE: Science first, marketing second
Submit responseIN RESPONSE:
The evidence that Merck pursued a seeding trial to support sales of Vioxx is clear and is derived from their own internal documents.
Dr. Edelman states that ADVANTAGE was developed by a group that was separate from the Marketing department. David Anstice, the former President of Merck’s Human Health organization, which had overarching responsibility for the study, affirmed in court, under oath, that ADVANTAGE was initiated and conducted by doctors in marketing (1).
ADVANTAGE is not a case, as suggested by Dr. Edelman, in which marketing is leveraging a study that originates within a company’s research division – ADVANTAGE is a study that was developed by marketing to boost sales. In the published Merck memo about the Merck Marketing Annual Awards, the objectives of ADVANTAGE were described as “to provide product trial among a key physician group to accelerate uptake of Vioxx as the second entrant in a highly competitive new class and gather data important to this customer group. The trial was designed and executed in the spirit of Merck marketing principles…” (2).
Dr. Edelman claims that the purpose of ADVANTAGE was to answer unaddressed scientific questions. The stated objective was “to assess the tolerability of rofecoxib compared with naproxen for treatment of osteoarthritis” even as the drug had already been compared with other non- steroidal anti-inflammatory drugs. The claim of the scientific value of the study is countered by the assessment of Dr. Edward Scolnick, former President of Merck’s Research division, who called the study “intellectually redundant” (3).
Dr. Edelman states that Merck’s business interests were understood by physician-investigators, participants and institutional review boards. Although these groups would know that Merck is a for-profit company, we doubt that they understood the extent to which marketing played a role in the design and execution of the study. Moreover, the physician-ivestigators likely did not know that they were the subjects of a study of the effect of ADVANTAGE on the prescribing habits of the physician- investigators.
Dr. Edelman states that ADVANTAGE does not meet the definition of a seeding trial. We believe otherwise and find it very consistent with the definition by Kessler and colleagues (4). ADVANTAGE has many of the key attributes of a seeding trial. It included a scientific objective of limited value, the recruitment of investigators because they are frequent prescribers, and the sponsorship of the studies by the company's sales and marketing division rather than its research department.
Finally, as a clarification, the date of the Marketing memo represented in Figure 1 was 1/4/99, and a revision in 12/1999 is noted in the bottom right corner, indicating that the original memo was created before the start of ADVANTAGE (4).
References
1. Trial testimony of David Anstice; Hermans v. Merck; 27 January 2007; page 620, lines 2 to 16.
2. McKines C, Sherwood L. Memo to Merck Marketing. Best Physician Program Award. 4 January 1999. Bates No. MRK-ACJ0000495. Accessed at http://dida.library.ucsf.edu/tid/vio21x10 on 15 September 2008.
3. Scolnick E. E-mail to Greene D. ADVANTAGE CV event tables. 7 April 2001. Bates No. MRK-ACR0009150. Accessed at http://dida.library.ucsf.edu/tid/vio22x10 on 15 September 2008.
4. Kessler DA, Rose JL, Temple RJ, Schapiro R, Griffin JP (1994) Therapeutic- class wars--- drug promotion in a competitive marketplace. N Engl J Med 331: 1350-1353.
Conflict of Interest:
All authors were compensated for participation in litigation against Merck at the request of plaintiffs.
-
So What's New?
Submit responseFor the ten years, I have participated in ethical review of research studies, both as an ordinary member of research ethics committees (institutional review bodies) and as chairman. In all of that time we have had the practice of describing certain studies as a "Marketing-study." Such studies are very easy to identify because they usually compare new wonder drug against the currently popular teatment and usually are sufficient to prove statistically that the treatment being tested is "not inferior" to the old treatment. If drug companies did not fund such projects however, there would be little research carried out into the efficacay of drugs and therefore it is illogical to ban the carrying out of "marketing" studies.
The solution does not lie in preventing information from being created; it lies in ensuring that the people who read studies have sufficient education to be able to identify when a study is clearly biased in favor of a particular viewpoint. Usually, this is very simple - all that is necessary is to look at the funding body listed in the acknowledgements. If it is the company whose product whose virtues are extolled, any good results should be discounted by at least 50%.
Conflict of Interest:
None declared
-
A Proposal for Funding Ethical Clinical Trials
Submit responseTo The Editor
Many Phase III (pre-FDA approval) and Phase IV (post-marketing) clinical drug trials are conducted by the pharmaceutical industry, whose goal is to (rightfully) maximize their profit. When a clinical trial demonstrates that a drug is not effective or has side effects, we have seen pharmaceutical companies suppress/delay the release of this information, as they (literally) have billions of dollars at risk. We now learn(1) that the inherent design of pharmaceutical trials is done as an extension of the pharmaceutical company's marketing division. In the end, the public's health is ill-served by a clinical trial that is designed, funded and conducted by a pharmaceutical company which has a vested interest in the outcome of the trial.
One solution to this dilemma would be for the FDA, and the major medical journals, to refuse to accept Phase III and Phase IV trial results that were funded by, or under the guidance of, any entity that has a financial interest in the outcome of the trial. Instead, pharmaceutical companies, upon the completion of their Phase II clinical trials, should be required to pay a "clinical trial fee" to the FDA. The "clinical trial fee" would be used to fund all Phase III and Phase IV trials. These FDA funded trials would then be conducted by academic clinician-scientists, who have no financial interest in the outcome of the trial.
This proposal would help to ensure that clinical trials are properly designed, provide financial support to the FDA and academic clinician-scientists, while also ensuring that the public's health interest is the first priority. And, this solution should not be any more expensive for the pharmaceutical industry than the current situation.
1) Kevin P. Hill, MD, MHS; Joseph S. Ross, MD, MHS; David S. Egilman, MD, MPH; and Harlan M. Krumholz, MD, SM; The ADVANTAGE Seeding Trial: A Review of Internal Documents. Ann Int Med 2008 ; 149: 251
Conflict of Interest:
None declared
-
ADVANTAGE: Science first, marketing second
Submit responseMerck would like to correct numerous inaccuracies in the latest article authored by paid consultants to plaintiffs' lawyers in the VIOXX litigation against Merck and respond to this biased attack on Merck's scientific integrity.1The scientific purpose of the ADVANTAGE trial was to answer unaddressed questions about VIOXX. The study, a double blind, randomized, controlled trial, was designed, analyzed and published by the Clinical Development (CDP) unit of Merck's US Human Health (USHH) organization, in conjunction with investigators. CDP was part of the Medical and Scientific Affairs department of USHH and was separate from the Marketing department within USHH contrary to the assertion of Hill et al. ADVANTAGE does not meet the definition of a seeding study in any of the references cited by the authors in their paper.2,3
The scientific objectives of the ADVANTAGE study are clearly detailed in the protocol, which was reviewed and approved in February, 1999, through the routine channels including sign off by senior Merck Research Laboratories (MRL) officials.4 These documents were produced to plaintiffs in the VIOXX litigation and available to Hill et al for their analysis and contained keywords from their purported search, despite their assertion that they could find no evidence of MRL involvement in the study. The scientific purpose of ADVANTAGE was properly disclosed to physicians-investigators, participants and institutional review boards and Merck's business interests were understood. The editors of Annals documented in their letter of acceptance for publication of the study in 2003, that physicians would be interested in the type of results ADVANTAGE produced.5
As is always the case, a scientifically sound, rigorously conducted study that shows the benefits of a drug for patients and doctors would also benefit the company that produced the drug. Accordingly, Marketing would be expected to evaluate the commercial possibilities of ADVANTAGE.* This does not change the fact that the primary intent of the study was to answer scientific questions of importance to primary care physicians.
Merck believes there is great value in conducting scientifically based clinical studies to address scientific questions. We believe we acted appropriately in the ADVANTAGE trial, and stand behind our strong beliefs in the principles of scientific integrity. We regret that neither the authors nor editors chose to contact Merck to verify their information or to discuss their conclusions. We encourage readers to read Merck's open letter at: http://www.merck.com/newsroom/press_releases/corporate/. * Note that the correct date of the Marketing memo represented in Figure 1 in the paper by Hill et al. was January 4, 2000, when the study was substantially complete, not 1999.
References: 1. Hill KP, Ross JS, Egilman DS, Krumholz HM. The ADVANTAGE seeding trial: a review of internal documents. Ann Intern Med. 2008;149:251-258. 2. Kessler DA, Rose JL, Temple RJ, Schapiro R, Griffin JP. Therapeutic-class wars – drug promotion in a competitive marketplace. N Engl J Med. 1994;331:1350-3. 3. Psaty BAM, Rennie D. Clinical trial investigators and their prescribing patterns: another dimension to the relationship between physician investigators and the pharmaceutical industry [Editorial]. JAMA. 2006;295:2787-90. 4. Carbone KK. ADVANTAGE protocol approval memo. Bates No. LEH804576 to LEH804584. Accessed at http://www.merck.com/newsroom/vioxx/pdf/protocol_approvals.pdf on 19 August 2008. 5. Mulrow CD. ADVANTAGE acceptance letter. Bates No. MRK-APQ0006943 to MRK-APQ0006948. Accessed at http://www.merck.com/newsroom/vioxx/pdf/annals.pdf on 19 August 2008.
Conflict of Interest:
Executive Director, Global Center for Scientific Affairs, Merck Research Laboratories