Doing More Good than Harm: Need for a Cease Fire
Article Outline
On May 21, 2007, a meta-analysis of 42 randomized trials of 26,011 patients with 158 myocardial infarctions was posted prepublication, which concluded that rosiglitazone produced a statistically significant relative risk of 1.43.1 The lower and upper bounds of the 95% confidence interval of 1.03 and 1.98 indicate that this meta-analysis was unable to distinguish reliably whether rosiglitazone conferred no hazard or a substantial hazard. Among numerous methodological limitations, trials that did not accrue any clinical cardiovascular disease endpoints but contributed person-years of exposure to drug safety, were excluded, producing a biased overestimate of harm.2 The chief value of this meta-analysis should have been to formulate a hypothesis about one possible hazard of rosiglitazone that may offset any potential benefits.
Hypothesis formulation should lead to hypothesis testing in another randomized trial of adequate size and duration. In fact, the Rosiglitazone Evaluated for Cardiovascular Outcomes (RECORD) trial was already in progress, but the adverse publicity from the meta-analysis claiming clear evidence of harm led patients to withdraw from the trial and the Institutional Review Board to require their re-consent. Finally, the United States Food and Drug Administration requested unblinded interim data. The investigators decided that their only rational choice was to publish their interim findings, which were posted prepublication on June 5, 2007.3 The interim findings from RECORD, based on 419 adjudicated cardiovascular events, showed a relative risk of 1.08, with a 95% confidence interval from 0.89 to 1.31. Thus, RECORD was also inconclusive because it could not continue until either its scheduled end or a sufficient totality of evidence emerged for the independent Data and Safety Monitoring Board (DSMB) to act.
The adverse publicity from a meta-analysis that formulated a hypothesis precluded the only means available to test that hypothesis. In a peer-reviewed publication, 3 Swedish health authorities later wrote that the DSMB of RECORD “should consider the overall pattern and promote earlier stopping if warranted.”4 Such pronouncements by regulatory authorities are likely to erode the confidence and trust in DSMBs whose independent judgments about safety and efficacy in emerging data are crucial to protection of human subjects and integrity of randomized trials.
These events raise the alarming specter that academicians, prestigious medical journals, and regulatory and other government authorities are overstating concerns about drug safety before a sufficient totality of evidence has emerged. With respect to academicians, the legitimate concerns about conflicts of interest in publications in peer-reviewed journals or presentations should not be limited to financial but should also include intellectual conflicts. A well-designed and -conducted randomized trial to detect the most plausible small-to-moderate effects provides a degree of reliability that is never possible to achieve from any meta-analysis or observational study. Such concerns are only increased in large prescription-based cohort and case-control studies that are increasingly performed because computerized prescription records enhance their feasibility.
To do more good than harm, a ceasefire from academicians, prestigious medical journals, regulatory and other government authorities, who should report but not make the news, seems prudent. In addition, seeking détente with an industry that should be assumed to be trying to prove their drug has a net benefit rather than trying to determine whether their drug has a net benefit, seems necessary but not sufficient. The guiding principle about medical interventions should remain that rational decisions for individual patients and the health of the general public should be based on a sufficient totality of evidence, which, for small to moderate effects, must include large-scale randomized trials.
References
- . Effect of Rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. N Engl J Med. 2007;356:2457–2471
- . Epidemiology in Medicine. Boston: Little, Brown and Company; 1987;
- Rosiglitazone evaluated for cardiovascular outcomes—an interim analysis. N Engl J Med. 2007;357:28–38
- . The conscientious judgement of a DSMB—statistical stopping rules re-examined. Eur J Clin Pharmacol. 2008;64:69–72
Funding: See last page of article.
Conflict of Interest: None.
Authorship: The authors declare that each of them has participated in the drafting (CHH) and revision (CHH, DD, NBM, SB, JB) of the manuscript and that we have seen and approved the final version.
Funding: The financial affiliations with all organizations, including any with a financial interest, direct or indirect, in the subject matter or materials discussed in the article that may affect the conduct or reporting of the work submitted are listed below.Dr. Hennekens is funded by the Charles E. Schmidt College of Biomedical Science, Department of Clinical Science and Medical Education & Center of Excellence at Florida Atlantic University (FAU) as Principal Investigator (PI) on 2 investigator initiated research grants funded to FAU by Bayer. He serves as an independent scientist in an advisory role to investigators and sponsors, including as Chair or Member on Data and Safety Monitoring Boards (DSMBs), for Actelion, Amgen, AstraZeneca, Bristol-Myers Squibb, Food and Drug Administration (FDA), Hughes Hubbard & Reed LLP, National Association for Continuing Education, National Institutes of Health (NIH), Pfizer, PriMed, Sidley Austin LLP, and UpToDate. He serves on speakers' bureaus for the International Atherosclerosis Society and AstraZeneca, concerning lipids and heart failure, as well as Pfizer, concerning lipids. He receives royalties for authorship or editorship of 3 textbooks. He receives royalties as coinventor on patents concerning inflammatory markers and cardiovascular disease, which are held by Brigham and Women's Hospital. He has an investment management relationship with SunTrust Bank, who has sole discretionary investment authority.Dr. DeMets is supported 40% for teaching and administrative duties as Chair. 15% from an NIH grant to the University of Wisconsin-Madison (UW-M). 45% as PI on UW-M contracts with industry as an independent statistical data analysis center. Current sponsors are Astra-Zeneca, Amgen and Bristol-Myers Squib. He receives royalties as author or editor of 3 textbooks and has Tax Sheltered Accounts which are retirement accounts in mutual funds with Fidelity and UBS. He holds small amounts of stocks with Sun and Intell. He serves as consultant to NIH, FDA and the pharmaceutical or device industry. He serves as an independent biostatistician in an advisory role to investigators and sponsors as a member of DSMBs for Actelion, Amgen, Astra-Zeneca, Bayer, Boehringer-Ingelheim, Bristol-Myers Squibb, CVRx, Genentech, Glaxo-Smith-Kline, Novartis, Otsuka, Pfizer, Roche, Sanofi-Aventis, Takeda, as well as Duke, Harvard and Hamilton Clinical Research Institutes.Dr. Bairey-Merz is funded by Cedars-Sinai Medical Center, the Women's Guild Endowed Chair in Women's Health, the Barbra Streisand Endowment for Women's Heart Research and Education Program, and PI on three investigator initiated research grants funded by NIH. She serves in an advisory role to investigators and sponsors, including as Chair of a DSMB for NIH and on advisory boards for the American Heart Association and American College of Cardiology. She received honoraria for lecturing at Stanford University, Association of Black Cardiologists, Berlin Heart Institute, Madera Medical Society, Argentine Society of Cardiology, American College of Physicians, MedReviews/CVT, Mayo Clinic, University of New Mexico, Northwestern University, MedCon, University of California, Davis, and Abbott Laboratories. She serves as consultant to Novartis, CMP Media, Rodale Press, University of Pittsburgh, Centers for Disease Control, Adventist Health, Bayer, Remillard and Associates, Navigant, Hunt and Associates, Pfizer, NIH, Karolinska Institute, Strategy Group, Biological Systems Processing-BSP. She holds stock in Eli Lilly, Medtronic, Johnson and Johnson, as well as small amounts of stocks with Boston Scientific, Teva Pharmaceuticals, and ATS Medical.Dr. Borer serves on DSMBs or cardiac event adjudication committees for Salix. Celladon, NovaCardia, Pfizer, Takeda, Gilead and the NIH. He serves as consultant to Servier, BioMarin, ARMGO, Johnson and Johnson, Savient, Indigo, Hoffmann-La Roche, and Sanofi.
PII: S0002-9343(08)01062-0
doi:10.1016/j.amjmed.2008.10.021
© 2009 Published by Elsevier Inc.