Trial Registration Report Card
http://www.100md.com
《新英格兰医药杂志》
One measure of medical progress is new treatments. The discovery of a novel therapy takes time and money, but more important, it requires the mutual effort of groups that, while they share the common goal of improved treatment, often have fundamentally competing interests. These interests intersect at the clinical trial. Patients who are looking for more effective and safer treatment agree to take part in a clinical trial in the hope that they will benefit from such treatment or that others with similar conditions will benefit later. The company developing the new therapy shares the hope that the trial will be successful, because it wants to market the tested therapy exclusively and profitably for as long as possible before its competitors can launch a similar therapy into the marketplace. These goals, though overlapping, are inevitably in conflict and will generate tension. Such tension has been thrown into sharp relief over the past 15 months by the push for clinical trial registration.
The academic establishment and patients have argued that when patients, motivated by altruism, participate (or even consider participating) in a clinical trial, they are entitled to understand fully all the options available to them in the various trials that are currently recruiting subjects. In addition, their participation in a clinical trial should result in generalizable knowledge that will be available to future patients and investigators to improve patient care. This can happen only when appropriate details of the clinical trial are made available to the public in a timely fashion. The Internet and public registries have made this possible.
Some in industry have argued that to open their portfolio of clinical trials to public scrutiny, particularly the scrutiny of other drug companies, would put them at such a competitive disadvantage that they would be unable to bring new products to market. Congress, however, decided to encourage openness by enacting on November 21, 1997, Section 113 of the Food and Drug Administration Modernization Act (FDAMA 113).1 Section 113 ultimately created ClinicalTrials.gov as an Internet-based public resource for information on studies of drugs, including biologic drug products, that are conducted under the FDA's investigational-new-drug regulations2 and involve the treatment of serious or life-threatening diseases and conditions.
In September 2004, the International Committee of Medical Journal Editors (ICMJE) announced that its journals would not publish the results of any ongoing trial that had not been appropriately registered in ClinicalTrials.gov or another qualified public registry by September 13, 2005.3,4 In this issue of the Journal, Zarin et al.5 provide a report card on compliance with these legislative and ICMJE requirements and the quality of the reporting that occurred before and after the ICMJE deadline for clinical trial registration. This report card examines whether the data fields required by FDAMA 113 have been completed in a meaningful fashion, with details about the drug or other intervention being studied and the prespecified measure of the trial's primary outcome. Without such critical data, registration becomes meaningless.
Zarin et al. show that there was a dramatic change in the number of trials registered during the summer of 2005. There can be no doubt that this spike was related to the ICMJE statement and deadline, because the rate of registration fell (though to a rate higher than that before the statement) after the deadline for registration passed. In addition, they show that the Intervention Name field was universally completed in a meaningful fashion when the trials were sponsored by academic institutions or the National Institutes of Health. In contrast, among trials registered by commercial sponsors, compliance with this field was variable. Here the message is more nuanced.
The vast majority of commercial entities provided meaningful data in most of their entries before the ICMJE statement and continued to do so during the summer of 2005. However, in the spring, some companies, such as Merck, GlaxoSmithKline, and Pfizer, provided meaningful entries in the Intervention Name field in an astonishingly low number of registrations. During the summer, Merck amended most of its meaningless entries to include clinically useful information in this field; by October they were in compliance in more than 99 percent of their registrations. GlaxoSmithKline and Pfizer are still using meaningless entries in the Intervention Name field in 21 percent and 11 percent of entries, respectively. This is puzzling, since most other companies are able to comply fully with the requirements of FDAMA 113.
The second critical measure examined by Zarin et al. was the number of records with the Primary Outcome field completed. Here the data are less reassuring, and the performance of some companies remains abysmal. Of note, by October 2005, Novartis had completed this field only 3 percent of the time, and Merck only 20 percent of the time. Again, many of their competitors were in virtually full compliance, undercutting any argument that this failure reflects a commercial imperative.
The ICMJE requirement that clinical trials be registered if they are to be considered for publication has been a resounding success. But the report cards for some companies would read "improved but could do better." We demand complete compliance, because trial registration makes moral sense. When patients put themselves at risk to participate in clinical trials, they do so with the tacit understanding that their risk is part of the public record, not merely the secret record of the sponsor.
In our opinion, it is unacceptable for a trial sponsor not to register its trial in a complete, meaningful, and timely fashion. We call for all clinical investigators and patients to participate only in fully registered trials. This call has recently been echoed by the major organization representing academic medical centers in the United States — the Association of American Medical Colleges.6 If a company continues to register trials using meaningless data, with no respect for the registration process and the patients who participate in those trials, investigators and patients should refuse to participate. If a company realizes that secrecy and failure to register mean that it cannot meet its recruitment goals, it will quickly recognize that the consequence of that secrecy is commercial failure, not competitive success. We must monitor the companies that are currently noncompliant to ensure that all live up to the spirit of the law as reflected in meaningful clinical trial registration.
References
Public Health Service Act, 42 U.S.C. 282(j).
Investigational New Drug Application, 21 C.F.R. 312.
De Angelis CD, Drazen JM, Frizelle FA, et al. Clinical trial registration: a statement from the International Committee of Medical Journal Editors. N Engl J Med 2004;352:1250-1251.
De Angelis CD, Drazen JM, Frizelle FA, et al. Is this clinical trial fully registered? A statement from the International Committee of Medical Journal Editors. N Engl J Med 2005;353:2436-2438.
Zarin DA, Tse T, Ide NC. Trial registration at ClinicalTrials.gov between May and October 2005. N Engl J Med 2005;353:2779-2787.
Assocation of American Medical Colleges. Principles for protecting integrity in the conduct and reporting of clinical trials. (Accessed December 8, 2005, at http://www.aamc.org/research/clinicaltrialsreporting/start.htm.)(Jeffrey M. Drazen, M.D., )
The academic establishment and patients have argued that when patients, motivated by altruism, participate (or even consider participating) in a clinical trial, they are entitled to understand fully all the options available to them in the various trials that are currently recruiting subjects. In addition, their participation in a clinical trial should result in generalizable knowledge that will be available to future patients and investigators to improve patient care. This can happen only when appropriate details of the clinical trial are made available to the public in a timely fashion. The Internet and public registries have made this possible.
Some in industry have argued that to open their portfolio of clinical trials to public scrutiny, particularly the scrutiny of other drug companies, would put them at such a competitive disadvantage that they would be unable to bring new products to market. Congress, however, decided to encourage openness by enacting on November 21, 1997, Section 113 of the Food and Drug Administration Modernization Act (FDAMA 113).1 Section 113 ultimately created ClinicalTrials.gov as an Internet-based public resource for information on studies of drugs, including biologic drug products, that are conducted under the FDA's investigational-new-drug regulations2 and involve the treatment of serious or life-threatening diseases and conditions.
In September 2004, the International Committee of Medical Journal Editors (ICMJE) announced that its journals would not publish the results of any ongoing trial that had not been appropriately registered in ClinicalTrials.gov or another qualified public registry by September 13, 2005.3,4 In this issue of the Journal, Zarin et al.5 provide a report card on compliance with these legislative and ICMJE requirements and the quality of the reporting that occurred before and after the ICMJE deadline for clinical trial registration. This report card examines whether the data fields required by FDAMA 113 have been completed in a meaningful fashion, with details about the drug or other intervention being studied and the prespecified measure of the trial's primary outcome. Without such critical data, registration becomes meaningless.
Zarin et al. show that there was a dramatic change in the number of trials registered during the summer of 2005. There can be no doubt that this spike was related to the ICMJE statement and deadline, because the rate of registration fell (though to a rate higher than that before the statement) after the deadline for registration passed. In addition, they show that the Intervention Name field was universally completed in a meaningful fashion when the trials were sponsored by academic institutions or the National Institutes of Health. In contrast, among trials registered by commercial sponsors, compliance with this field was variable. Here the message is more nuanced.
The vast majority of commercial entities provided meaningful data in most of their entries before the ICMJE statement and continued to do so during the summer of 2005. However, in the spring, some companies, such as Merck, GlaxoSmithKline, and Pfizer, provided meaningful entries in the Intervention Name field in an astonishingly low number of registrations. During the summer, Merck amended most of its meaningless entries to include clinically useful information in this field; by October they were in compliance in more than 99 percent of their registrations. GlaxoSmithKline and Pfizer are still using meaningless entries in the Intervention Name field in 21 percent and 11 percent of entries, respectively. This is puzzling, since most other companies are able to comply fully with the requirements of FDAMA 113.
The second critical measure examined by Zarin et al. was the number of records with the Primary Outcome field completed. Here the data are less reassuring, and the performance of some companies remains abysmal. Of note, by October 2005, Novartis had completed this field only 3 percent of the time, and Merck only 20 percent of the time. Again, many of their competitors were in virtually full compliance, undercutting any argument that this failure reflects a commercial imperative.
The ICMJE requirement that clinical trials be registered if they are to be considered for publication has been a resounding success. But the report cards for some companies would read "improved but could do better." We demand complete compliance, because trial registration makes moral sense. When patients put themselves at risk to participate in clinical trials, they do so with the tacit understanding that their risk is part of the public record, not merely the secret record of the sponsor.
In our opinion, it is unacceptable for a trial sponsor not to register its trial in a complete, meaningful, and timely fashion. We call for all clinical investigators and patients to participate only in fully registered trials. This call has recently been echoed by the major organization representing academic medical centers in the United States — the Association of American Medical Colleges.6 If a company continues to register trials using meaningless data, with no respect for the registration process and the patients who participate in those trials, investigators and patients should refuse to participate. If a company realizes that secrecy and failure to register mean that it cannot meet its recruitment goals, it will quickly recognize that the consequence of that secrecy is commercial failure, not competitive success. We must monitor the companies that are currently noncompliant to ensure that all live up to the spirit of the law as reflected in meaningful clinical trial registration.
References
Public Health Service Act, 42 U.S.C. 282(j).
Investigational New Drug Application, 21 C.F.R. 312.
De Angelis CD, Drazen JM, Frizelle FA, et al. Clinical trial registration: a statement from the International Committee of Medical Journal Editors. N Engl J Med 2004;352:1250-1251.
De Angelis CD, Drazen JM, Frizelle FA, et al. Is this clinical trial fully registered? A statement from the International Committee of Medical Journal Editors. N Engl J Med 2005;353:2436-2438.
Zarin DA, Tse T, Ide NC. Trial registration at ClinicalTrials.gov between May and October 2005. N Engl J Med 2005;353:2779-2787.
Assocation of American Medical Colleges. Principles for protecting integrity in the conduct and reporting of clinical trials. (Accessed December 8, 2005, at http://www.aamc.org/research/clinicaltrialsreporting/start.htm.)(Jeffrey M. Drazen, M.D., )