Clinical trial data should be shared as broadly as possible to help spur scientific innovation and answer questions of importance to public health.
"The question is not whether, but how, these data should be broadly shared," wrote Michelle M. Mello and her colleagues from Harvard University, Boston, the Pharmaceutical Research and Manufacturers of America, and several consulting companies, in an article published online Oct. 21 in the New England Journal of Medicine.
At a minimum, data-sharing should be available for trials of all approved prescription drugs, devices, and biologics in any country that has adequate intellectual property protection. A system has to ensure responsible use of data, protect privacy of research participants, and treat "all qualified data requesters and trial sponsors evenhandedly," requiring both generators and requesters to work according to the same rigorous scientific principles (N. Engl. J. Med. 2013 Oct. 21 [doi: 10.1056/NEJMhle1309073]).
The demand for more data from and about clinical trials – including protocol designs, results summaries, and more recently, raw input data – has grown over the past 15 years. Some medical journals have pushed for disclosure of more data upon request. The Food and Drug Administration increasingly has been requiring disclosure, and pharmaceutical manufacturers in the United States and Europe have made commitments to making more information public. Beginning in March 2014, the European Medicines Agency will require disclosure of some raw data, individual case report forms, and other data.
The authors envision at least four potential models for sharing data. With purely open access, everything would be available for download for free. This is the riskiest model, they said, since it would provide the least accountability.
Another model: The data generator would keep the data but answer very specific requests. A third model would have the clinical trial sponsor review data requests and decide whether and how to release the data.
In the last model, an independent review board would determine whether the data should be released. The board would collect the data from the sponsor and issue it to the requester, on a limited, need-to-know basis.
This model would likely best balance all of the competing needs, according to the authors. The independent board promises "to ensure accountability on the part of data generators and users and allow trial sponsors a voice while precluding them from denying access to data for reasons the public would not consider legitimate," they wrote.
An independent board also would help protect research participants and make sure that the playing field is level among all stakeholders, they added.
The authors noted many benefits to allowing wider access to patient-level data, such as independent analyses of safety and effectiveness, new lines of inquiry that could expose product flaws or trial design flaws, and the potential to answer questions that might affect public health but that weren’t explored in the original study.
The biggest downside is that individual participants’ privacy could be compromised, according to the authors. The risk of exposure "raises critical questions about how to ensure that participants understand the potential ramifications of data sharing," they wrote.
Mandatory disclosure could also discourage investment in research and development if manufacturers believe that the data could be used by competitors. Wider data-sharing could also lead to second-guessing of approvals by regulatory agencies.
Clinical trial data-sharing received attention from the World Medical Association in its most recent update of the Declaration of Helsinki, published online Oct. 19 in JAMA (doi: 10.1001/jama.2013.281053). Three general principles addressed data sharing:
• Number 9: It is the duty of physicians who are involved in medical research to protect the life, health, dignity, integrity, right to self-determination, privacy, and confidentiality of personal information of research subjects. The responsibility for the protection of research subjects must always rest with the physician or other health care professionals and never with the research subjects, even though they have given consent.
• Number 24. Every precaution must be taken to protect the privacy of research subjects and the confidentiality of their personal information.
• Number 32. For medical research using identifiable human material or data, such as research on material or data contained in biobanks or similar repositories, physicians must seek informed consent for its collection, storage and/or reuse. There may be exceptional situations where consent would be impossible or impracticable to obtain for such research. In such situations the research may be done only after consideration and approval of a research ethics committee.
Several of the authors of the paper published in the New England Journal of Medicine disclosed that they work for consulting companies that receive fees from various manufacturers and from academic medical centers. The working group was convened through the Multi-Regional Clinical Trials Center at Harvard University, which receives funds from pharmaceutical companies and not-for-profit entities.