Two small but significant steps forward were taken yesterday in the direction of open access to clinical trial data.
A committee of the Institute of Medicine (IOM) yesterday called for the sharing of supporting data for clinical trials results within six months after publication, with a full analyzable data set shared no later than 18 months after study completion or 30 days after regulatory approval.
Also yesterday, Johnson & Johnson (J&J) joined the Yale Open Data Access (YODA) Project in agreeing to share data from clinical trials for medical devices and diagnostics. The YODA Project said in a statement that the agreement “establishes a fully independent intermediary to manage requests and promote data use,” as J&J has done with pharmaceutical clinical trial data since last year.
Under the agreement, the YODA Project will approve or deny requests from investigators for de-identified patient data associated with the pharmaceutical, medical device, and diagnostic clinical trials conducted by J&J companies.
“This action will benefit society and represents a major step forward in the effort to promote data sharing, as Johnson & Johnson’s leadership in this area now extends from sharing its drug data to sharing its device and diagnostics data,” said Dr. Harlan Krumholz, professor of medicine and leader of the YODA Project. “We hope this action serves as a catalyst to others to join the momentum on open science.”
The two new developments join what the IOM report acknowledged was progress toward data sharing in recent years. Eleven drug developers have committed to sharing clinical trial data through clinicalstudydatarequest.com and allow an independent review panel to decide data requests: Astellas, Bayer, Boehringer Ingelheim, GlaxoSmithKline, Lilly, Novartis, Roche, Sanofi, Takeda, UCB, and ViiV Healthcare.
Three industry groups— Pharmaceutical Research and Manufacturers of America (PhRMA), the European Federation of Pharmaceutical Industries and Associations (EFPIA), and the Biotechnology Industry Organization (BIO)—signaled support for sharing clinical trial data beyond recent industry norms, but not the open-access sought by Europe’s chief drug regulator and GlaxoSmithKline. The European Medicines Agency in October issued a more expansive data-sharing policy last October, though not the full open-access sought by advocates.
The IOM’s Committee on Strategies for Responsible Sharing of Clinical Trial Data similarly said its recommendations were intended to balance data sharing’s benefits—maximizing knowledge, stimulating new research ideas, and avoiding duplicative trials—with its challenges.
The panel defined those challenges as protecting privacy and consent for trial patients; providing researchers adequate time to analyze and publish their data, safeguarding commercially confidential information, and assuring research institutes that requirements for sharing clinical data will not be “unfunded mandates.”
Also included among challenges was “guarding against invalid secondary analyses, which could undermine trust in clinical trials or otherwise harm public health”—whether through inadvertent errors in data analysis or, in the U.S., the prospect of monetary gain through qui tam lawsuits,” the report stated.
The panel expressed fear that such invalid analyses “will lead to claims of risk that are not scientifically valid, which may in turn lead to lawsuits for negligence that, while without merit, are expensive to respond to and defend against.”
IOM’s committee cited criticism from outside researchers over the PLATO trial, whose data led to approval of AstraZeneca’s Brilinta (ticagrelor). Questions about geographic discrepancies of outcomes, study-site monitoring, and the timing of deaths prompted the U.S. Department of Justice to investigate the trial. DoJ closed its investigation last year without carrying out any action.
“The sharing of clinical trial data needs to be carried out in a way that maintains incentives for sponsors and researchers to develop new clinical trials and therapies and that sustains patients’ willingness to participate in them,” said committee chair Bernard Lo, president of the Greenwall Foundation, a funder of bioethics research. “Our recommendations attempt to balance the interests of different stakeholders with the public interest in having the best information possible regarding the effectiveness and safety of therapies.”
The committee said clinical trial “stakeholders” should mitigate the risks and enhance the benefits of data sharing by employing “data use” agreements; creating independent review panels with members of the public to decide sharing requests, and maintaining transparency in data-access policies.
The panel defined stakeholders to include funders and sponsors, as well as disease-focused patient advocacy groups, regulators, and overseers such as institutional review boards, research ethics committees, investigators, their research institutions or universities, journals, and membership and professional societies.
“Stakeholders in clinical trials should foster a culture in which data sharing is the expected norm, and should commit to responsible strategies aimed at maximizing the benefits, minimizing the risks, and overcoming the challenges of sharing clinical trial data for all parties,” the IOM panel stated.
IOM’s committee said that data-use agreements should include provisions aimed at protecting clinical trial participants, advancing the goal of producing scientifically valid secondary analyses, giving credit to the investigators who collected the clinical trial data, protecting the intellectual property interests of sponsors, and ultimately improving patient care.
“Although it is unclear whether and how data use agreements will be enforced, the committee believes these agreements have significant normative, symbolic, and deterrent value, setting professional expectations and standards for responsible behavior,” the committee concluded.
The 13-member committee included experts in key scientific and research-related domains, including academia, industry, funding bodies, regulatory activities, scientific publications, clinicians, and patients, the IOM said.