|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 26, No 4 (February 1), 2008: pp. 680-682 © 2008 American Society of Clinical Oncology. DOI: 10.1200/JCO.2007.15.5234
Secrecy and Integrity in Clinical Trials
From the Division of Pediatrics, Children's Cancer Hospital at The University of Texas M.D. Anderson Cancer Center, Houston, TX Corresponding author: Robert J. Wells, MD, Pediatrics, The University of Texas M.D. Anderson Cancer Center, Unit 87, 1515 Holcombe Blvd, Houston, TX 77030; e-mail: rjwells{at}mdanderson.org INTRODUCTION Mrs Jones was a woman with resected node-positive, HER-2/neu–positive breast cancer who was considering her options for adjuvant systemic therapy in September 2004. She was aware of the literature regarding the use of a newer drug treatment for breast cancer that had been demonstrated to be effective for some women with metastatic breast cancer and that appeared to be tolerable in the adjuvant setting. She was offered participation in a large randomized clinical trial where she would be randomly assigned to get standard systemic therapy versus the same standard systemic therapy plus the new agent that was under investigation. The patient had been told that over 2,000 women had been entered on this trial over the preceding 3 years, and she asked for preliminary information regarding how these women were faring so it could help her decide whether she wanted to participate in this clinical trial. Confidentiality (secrecy) of preliminary data sets in clinical trials sponsored by both the federal government and industry has been standard policy in oncology in the United States since the mid-1990s.1 Recent attention has been drawn to this policy of secrecy by individuals leaking preliminary data to the investment community and by industry selectively releasing only favorable final data.2-5 Missing from the recent controversy has been any discussion of the conflict inherent with this policy of secrecy with other regulations, policies and standards of behavior that govern the conduct of research, physician-patient relationships, and individual and organizational integrity. As someone who has spent the last 20 years in academic medicine, I find concerns about maintaining secrecy difficult to understand. Scientific tradition (which albeit is not always honored) is that open discussion and exchange of information and data are of benefit to all. In research laboratories, preliminary data are presented at open lab meetings attended by most of the lab, colleagues from down the hall, and often visiting scientists from other institutions. Hopefully, the resulting discussion of the data results in a synthesis of ideas that improves the overall research quality and leads to advancement of knowledge. In clinical areas, patient rounds, grand rounds, and clinical conferences are almost always open to all, including students, house officers, visiting professors, fellows, and sometimes patients and/or their advocates. Preliminary ideas and data from reports of various types are presented and discussed. Clinical conferences and rounds are intended to be wide ranging, intellectually stimulating, and educational. It is hoped that these sessions will lead to better outcomes for our patients. As an educator, one of my jobs is to see that dissemination of knowledge takes place and that students understand the strengths and weaknesses of both "preliminary" and "final" data. Given this, I have more concerns regarding restricting information, than disseminating it. Federal regulations governing research require open exchanges of information (nonsecrecy). For example, Code of Federal Regulations 45.46 requires that all information that might effect a subject's decision to participate in a clinical trial be disclosed to the subject or potential subject.6 This regulation also requires that, if new information becomes available that might influence continued participation of the subject, this information must also be disclosed. Finally, this regulation requires that the subjects be informed of this commitment. Interim or preliminary data from ongoing, large double-blinded trials may well influence a subject's willingness to enter clinical trials, yet are not disclosed to either the physician responsible for the patient/subject, the physician investigator, or the patient subject. The reason for nondisclosure is that in the past when this information was released, some patients and/or their physicians decided that entry into, or continuation of, the trial was not in the patient/subject's best interest. Anecdotally, accrual to the trial sometimes either slowed or stopped, and the trial ended prematurely, in the view of "the scientific community" and the trial investigators.7 So, interim data for a particular trial may exist that may affect a subject's willingness to participate in a clinical trial but are not disclosed, despite the commitment to the research subject to do so, because the data have been classified as "confidential" or "secret." One of the interesting points of investment community "leaks" is that the investment community is determining the "market" value of preliminary results of clinical trials. Presumably, this information is of value to the investment community because it allows them to anticipate the end results of studies well enough to make money for their investors. Market value is clear evidence that preliminary data are significant data and might also be of use to many others as well who have an interest in the outcome of the trial. For example, if the physician investigators treating patients with an "innovative" drug that they hoped would be of benefit see no evidence of response or benefit with the first six to 10 patients and become less hopeful, might not this information be useful to the seventh or 11th potential patient research subject in deciding whether or not they wanted to enter that clinical trial as a subject? Cfr 45.46 also requires that institutional review boards (IRBs) be established, that these boards review clinical trials at least annually, and that the institutional review board evaluate the risks and benefits to the subjects in these trials. While IRBs are inundated with toxicity reports, they are usually not given efficacy data, particularly for large, prospective, randomized clinical trials. Without efficacy data, only risk assessments can be done and not risk/benefit analyses. Many times, only the data safety monitoring committee (DSMC) has both efficacy and toxicity data. However, they analyze the data set infrequently (sometimes less often than annually) at predetermined points of time, because frequent analyses of data make the final interpretation of the trial data more difficult. In addition, the decision process the DSMC uses is conservative, with the purpose of being able to convince a sometimes skeptical scientific community and clinicians that the study's conclusion warrants a change in practice.8 The IRB, whose role is subject protection, as well as the patient/subjects who are directly at risk, might reasonably choose to use a more liberal (somewhat less certain) decision point. For example, a DSMC generally only closes a randomized trial if one regimen is demonstrated to be clearly superior. A patient, their physician, or an IRB, which has the primary role of protecting the research subject from harm, probably would, or should, recommend closing a randomized trial when one arm is shown to be either better than or equal to the alternative, which is many times apparent long before superiority is demonstrated. The classification of the preliminary data set as "secret" or "confidential" prevents the IRB from fulfilling their role of patient protection, despite the commitment to the subjects that they will be, in fact, protected. Traditional international ethical standards set the responsibility for patient safety on the investigator, who is supposed to terminate any trial if/when he/she comes to believe the experiment has an unfavorable risk/benefit ratio for the subject.9 Since trials are based on a scientific hypothesis or a speculation (new drug/treatment A will be better than previous standard B) that "seems reasonable," real data can shed light on whether or not the speculation is accurate. Sometimes this becomes apparent much sooner than the study statistician calculates, since their calculations are based on prestudy speculations as well. Blinding the investigator and having no one look at the accruing data and performing risk-benefit evaluations to see if prestudy speculations were close, or not for periods of time that are in some instances well over a year, jeopardizes patient safety. It also prevents investigators from fulfilling their responsibilities for ethical research conduct because they have inadequate and incomplete information that prevents them from coming to any rational conclusion about whether to continue or terminate the study. Finally, secrecy leads to a reversal of the priority of interests. Usually advancing science/knowledge is regarded as a benefit to society, which in some cases could be in conflict with the best interests of the individual research subject.9 International standards of medical ethics, beginning with the Hippocratic tradition, instruct physicians to place the interests of their patients above the physician's own interests.10 A modern variant of this guidance adds that physicians must serve the interests of their patients, and that this service must not be compromised by market forces, societal pressures, or administrative exigencies.11 This modern variant stresses the importance of honesty in dealing with patients. International research standards also stress the importance of the interests of the research subject. They specifically state that if there is a conflict between the research subject's interests and the interests of society, the research subject's interests should prevail. It is the duty of both the research investigator and the physician responsible for medical care to assure that this priority is maintained.9 Those responsible for current research policies and administration have taken a different ethical view of the situation and have stressed the interests of society over those of the research subject.12 Hiding preliminary information that has discouraged ongoing patient subject participation in clinical trials in the past, misleading the patient about "full" disclosure, and misrepresenting review policies favors society's interests to advance knowledge over individual research subject's interests to avoid harm. Secrecy is both evidence of the reversal of this priority and the means to effect the change in priority. The dubious ethical nature of this practice, as well as the lack of evidence that such practices are really necessary to advance science, have been noted by others but have been generally ignored.13 From my perspective, the big problem with secrecy in clinical trials is that the medical and scientific community is being dishonest with patients and research subjects. We are compromising our integrity and the safety of research subjects, while engaging in unethical research practices and undermining ethical standards of research. Our integrity would be best maintained by the elimination of "secret preliminary data sets" from clinical trials. All data sets, preliminary and final, should always be openly available, particularly to our patient/subject and to the physicians responsible for their care.
Editor's Note AUTHOR'S DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST The author(s) indicated no potential conflicts of interest. NOTES Author's disclosures of potential conflicts of interest and author contributions are found at the end of this article. REFERENCES 1. Smith MA, Ungerleider RS, Korn EL, et al: Role of the independent data monitoring committee in randomized clinical trials sponsored by the National Cancer Institute. J Clin Oncol 15:2736-2743, 1998 2. Topol EJ, Blumenthal D: Physicians and the investment industry. JAMA 293:2654-2657, 2005 3. Steinbrook R: Wall Street and clinical trials. N Engl J Med 353:1091-1093, 2005 4. American Society of Clinical Oncology: Interactions with the investment industry: Practical and ethical implications. J Clin Oncol 25:338-340, 2007 5. Whittington CJ, Kendall T, Fonagy P, et al: Selective serotonin reuptake inhibitors in childhood depression: Systematic review of published versus unpublished data. Lancet 363:1341-1345, 2004[CrossRef][Medline] 6. OPRR Reports Protection of Human Subjects Title 45 Code of Federal Regulations part 46, Department of Health and Human Services. March 15, 1994 7. Ellingberg SS, Fleming TR, DeMets DL: Confidentiality issues relating to the data monitoring committee, in: Data Monitoring Committees in Clinical Trials: A Practical Perspective. John Wiley & Sons, Chicester, England, 2003, pp 69-87 8. Ellingberg SS, Fleming TR, DeMets DL: Statistical, philosophical and ethical issues in data monitoring, in: Data Monitoring Committees in Clinical Trials: A Practical Perspective. John Wiley & Sons, Chicester, England, 2003, pp 119-152 9. Jansen AR, Veatch RM, Walters L: The Nuremberg Code and the World Medical Association Declaration of Helsinki, in: Sourcebook in Bioethics: A Documentary History. Georgetown University Press, Washington, DC, 1998, pp 11-15 10. Von Staden H: In a pure and holy way: Personal and professional conduct in the Hippocratic Oath. J History Med Allied Sci 51:406-408, 1996 11. ABIM, ACP-ASIM, EFIM: Medical professionalism in the new millennium: A physician charter. Ann Intern Med 136:243-246, 2002 12. Pocock SJ: Statistical and ethical issues in monitoring clinical trials. Stat Med 12:1459-1469, 1993[Medline] 13. Lilford RJ, Braunholtz D, Edwards S, et al: Monitoring of clinical trials: Interim data should be publically available. BMJ 323:441-442, 2001 Submitted November 26, 2007; accepted December 3, 2007. Related Article
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2008 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
