Should the public registration of clinical trials be voluntaryor mandatory, as bills introduced earlier this month in Congresswould require? What trials should be included? Proposals formandatory registration have gained support because of concernsabout the effects on medical practice of concealing negativedata, concerns about the protection of research subjects, particularlyin studies of investigational products, and concerns about theavailability of information to researchers, physicians, peopleseeking to enroll as subjects in trials, and the public.1,2Critics argue that mandatory registration could reveal informationthat manufacturers consider proprietary, such as the resultsof small or exploratory studies, or impede the development ofdrugs or devices by forcing the release of research strategiesto competitors. Proponents argue that these risks must be weighedagainst the benefits of registration and that registration willonly be meaningful if all trials that should be registered are.They also doubt that voluntary efforts will work.
Pharmaceutical companies or other commercial entities may havebusiness reasons to conceal certain trials or to disclose findingsselectively. Recent proposals, however, require the registrationof trials at inception, before subjects are enrolled, so thatthe existence and design of every trial is immediately partof the public record.2 A clinical-trials registry may eventuallyinclude the results of a trial or a link to a report of thetrial in a peer-reviewed journal. Posting on the Internet asummary of the results, as some companies have begun to do,is not registration. Despite the inherent interest in results,they are only meaningful in the context of the design and conductof a study and the analysis of the data.
A comprehensive public trials registry, www.clinicaltrials.gov,already provides a mechanism for the voluntary registrationof clinical trials. The registry, developed by the Food andDrug Administration (FDA) and the National Institutes of Health(NIH), through the National Library of Medicine, includes studiesof drugs for serious or life-threatening conditions, as requiredby the FDA Modernization Act of 1997. However, it accepts allresearch studies in human volunteers that are designed to answerspecific health questions, regardless of study design or whetherthe studies are of drugs or other interventions. As a resultof recent changes, it can also accept international trials.As of mid-October, www.clinicaltrials.gov listed 11,844 trials,including 3651 that have been completed and 4190 that are recruitingsubjects. In September, more trials with commercial sponsorswere registered than in prior months (see Figure).
Figure. Trials Registered with www.clinicaltrials.gov between January and September 2004.
Data are from the National Library of Medicine.
In June, clinical-trial registration gained important support.The American Medical Association (AMA) endorsed the idea, andNew York State Attorney General Eliot Spitzer sued GlaxoSmithKlinefor concealing negative information about the antidepressantmedication paroxetine. In August, as part of the settlementof that lawsuit, GlaxoSmithKline agreed to post on its corporateWeb site a summary of clinical-study reports for every company-sponsoredtrial of its medications completed after December 27, 2000.There is no requirement, however, that the studies be registeredat inception. In September, as part of a separate agreementwith Spitzer, Forest Laboratories, which manufactures the antidepressantmedications citalopram and escitalopram, agreed to post on itscorporate Web site summaries of the results of clinical studiesof marketed drugs completed after January 1, 2000. Forest alsoagreed to post the number, title, starting date, and key objectivesof each phase 3 and phase 4 trial when it is initiated.
In response to the concerns about concealed data, the PharmaceuticalResearch and Manufacturers of America (PhRMA), the leading industrytrade group, has established an electronic database of clinical-studyresults (www.clinicalstudyresults.org), which, according toa press release, "provides doctors and patients unprecedentedaccess to clinical study information for marketed medicines."The database, which became available to the public on October1, is a voluntary repository of the results of controlled trialsof drug products that are marketed in the United States andthat are sponsored by member companies. It is not a clinical-trialsregistry. As of October 12, the database listed studies fromseven companies involving 10 medications; the numbers are expectedto increase.
The AMA, the International Committee of Medical Journal Editors(ICMJE), and some Democratic lawmakers in Congress advocatemandatory registration of clinical trials. The AMA recommendedthat the federal government "establish a comprehensive registryfor all clinical trials conducted in the United States."1,3In September, the ICMJE, which represents 11 general medicaljournals, including the New England Journal of Medicine, announcedthat "member journals will require, as condition of considerationfor publication, registration in a public trials registry. Trialsmust register at or before the onset of patient enrollment."2The policy takes effect in the summer of 2005. Although notadvocating a particular registry, the ICMJE established specificrequirements, which are currently met by www.clinicaltrials.govand which may eventually be met by additional registries. Althoughother journals have voiced support for this approach, it isnot known how many will actually adopt a similar policy.
In early October, Senators Christopher Dodd (D-Conn.), EdwardKennedy (D-Mass.), Tim Johnson (D-S.D.), and Ron Wyden (D-Oreg.)introduced the Fair Access to Clinical Trials Act in the Senate,and Congressmen Edward Markey (D-Mass.) and Henry Waxman (D-Calif.)introduced a companion bill in the House.4,5 The bills, whichare similar but not identical, would build on www.clinicaltrials.govto create a mandatory public electronic database of clinicaltrials that would be searchable without charge. It would beadministered by the federal government and would meet the minimumcriteria for a trial registry set by the ICMJE (see Table).The national registry would include "all publicly and privatelyfunded clinical trials involving drugs, biological products,or devices regardless of the outcome of the trial."5 An exceptionwould be made for most trials conducted to test solely the safetyof an unapproved product or to investigate pharmacokinetics.
Table. Proposed Information in the Clinical Trials Data Bank under the Fair Access to Clinical Trials Act.
The legislation would ensure the accuracy of the informationsupplied, through audits and the ability to correct inaccurateor false or misleading information. Compliance would be ensuredby overlapping mechanisms. Registration would be required beforeclinical trials could be approved by institutional review boards,a condition that is also supported by the AMA.3 Registrationwould also be required before the FDA could authorize the testingof an investigational product.4 For federally supported trials,it would be required as a condition of an individual or institution'sreceiving a federal grant, contract, or other award. The Departmentof Health and Human Services (DHHS) could impose fines of upto $10,000 per day for noncompliance. In a related area, theSenate legislation would ban contracts involving researchersor their institutions that prohibit, limit, or unreasonablydelay the discussion of the results of a clinical trial at ascientific meeting or the publication of such results.5
Despite the support for public registration of clinical trials,the prospects for mandatory registration are uncertain. Individualinstitutional review boards could decide on their own initiativethat registration is ethically required as a condition of theirapproval of trials, but they are unlikely to do so without nationalguidance. The DHHS, whose agencies include the FDA and the NIH,has not taken a position on mandatory registration, nor haveother governments. The effective dates of the ICMJE proposalare months away. Although the Fair Access to Clinical TrialsAct would mandate registration, it has no Republican cosponsorsin either the House or the Senate, and it was introduced asCongress was about to adjourn. The legislative outcome in theUnited States may depend on the results of next week's elections.
References
Steinbrook R. Public registration of clinical trials. N Engl J Med 2004;351:315-317. [Free Full Text]
De Angelis C, Drazen JM, Frizelle FA, et al. Clinical trial registration: a statement from the International Committee of Medical Journal Editors. N Engl J Med 2004;351:1250-1251. [Free Full Text]
Council of Scientific Affairs. Featured CSA report: influence of funding source on outcome, validity, and reliability of pharmaceutical research (A-04). Chicago: American Medical Association, June 2004.
Fair Access to Clinical Trials Act of 2004, H.R. 5252, 108th Cong., 2d Sess.
Fair Access to Clinical Trials Act of 2004, S. 2933, 108th Cong., 2d Sess.
Lott, J. P., Drazen, J. M.
(2007). Trial Registration. NEJM
356: 1581-1581
[Full Text]
Santoro, E., Rossi, V., Pandolfini, C., Bonati, M.
(2006). DEC-net: the development of the European register of clinical trials on medicines for children. Clin Trials
3: 366-375
[Abstract]
Popp, R. L., Lorell, B. H., Stone, G. W., Laskey, W., Smith, J. J., Kaplan, A. V.
(2006). An Outline for Public Registration of Clinical Trials Evaluating Medical Devices. J Am Coll Cardiol
47: 1518-1521
[Abstract][Full Text]
Zarin, D. A., Tse, T., Ide, N. C.
(2005). Trial Registration at ClinicalTrials.gov between May and October 2005. NEJM
353: 2779-2787
[Abstract][Full Text]
Hayward, R. A., Kent, D. M., Vijan, S., Hofer, T. P.
(2005). Reporting Clinical Trial Results To Inform Providers, Payers, And Consumers. Health Aff (Millwood)
24: 1571-1581
[Abstract][Full Text]
Norris, J. W.
(2005). Antiplatelet Agents in Secondary Prevention of Stroke: A Perspective. Stroke
36: 2034-2036
[Abstract][Full Text]
Abraham, E.
(2005). The AJRCCM in 2005. Am. J. Respir. Crit. Care Med.
171: 1-2
[Full Text]
Previous
