Clinical Data Sharing: The Case for Full Transparency
Under a new proposal for clinical research transparency, scientists, clinicians, and study sponsors will be required to share clinical trial data before manuscripts are considered for publication.
Clinical trials are a critical part of the development process for new medical products and require significant investment from several stakeholders, including patient volunteers, sponsors, researchers, regulatory agencies, and publishers. Sharing of clinical data through publication of results serves many purposes. At an intellectual level, published data provide a conduit for advancement of scientific discovery, minimize duplication of efforts, and foster collaboration and shared problem solving.
The true value of published clinical trial data perhaps lies on a humanistic level—patients volunteer to test the safety and efficacy of an investigational drug or to compare the benefit of one drug over another. Data transparency therefore carries both ethical and moral obligations, because the potential harm that can arise from selective publication, delayed publication, or no publication is too significant.
Data sharing is not a new concept; it has existed since the dawn of scientific discovery. The optimal scenario is that all clinical research data are published and accessible to all. The reality, however, is that a significant amount of data generated through clinical research is not published.
Saito and Gill reported that among 400 randomly selected studies entered as “completed” in the ClinicalTrials.gov database between January 1, 2008, and December 31, 2008, only approximately 30% remained published up to 4 years after study completion.1 Moreover, the study revealed publication bias where trials in early stages of clinical development (phases 1 and 2 vs phases 3 and 4), trials with sample size ≤50 subjects, and trials funded by the pharmaceutical industry were less likely to be published.
The importance of scientific transparency through sharing of clinical trial data, particularly data generated with federal funding, was recognized by the federal government through passage of the Food and Drug Administration Modernization Act (FDAMA) of 1997,2 which provided a mandate to the National Institutes of Health (NIH) to establish the ClinicalTrials.gov (CTG) online registry in 2000. The FDAMA required clinical trials to register on CTG within 21 days of first patient enrollment but did not require posting of clinical trial results.
A second Act, the FDA Amendments Act (FDAAA), was passed in 2007 to close the gap inherent in the FDAMA by requiring “applicable clinical trials” to post basic study results to CTG within one year of completion, irrespective of the funding organization.3 “Applicable clinical trials” primarily referred to data relevant to product registration; prelicensure studies of drugs and biologic products and small feasibility studies of devices were exempted. The impact is evident because the definition of “applicable clinical trials” was open to interpretation, and exemptions possibly contributed to the discretionary publication of clinical trial data revealed in the study by Saito and Gill. Consequently, while the intentions of FDAMA and FDAAA were good, they failed to achieve the goal for all clinical research data to be shared and publicly available.