- Defining Clinical Trial Transparency
- Clinical Trial Spin and Suppression
- Benefits and Costs of Transparency
- Legislative Strategies to Encourage Transparency
- Bottom-Up Strategies for Clinical Trial Transparency
- Conclusions: Towards Greater Transparency
- Clinical Trial Transparency References
Transparency in clinical trial results is not a panacea. New laws and policy measures to improve transparency must carefully balance the interests of the various stakeholders, including the industrial trial sponsors, investors in those companies, patients and trial subjects, doctors, and governmental agencies including the FDA and the NIH. However, the numerous safety and reporting problems that have occurred suggest that our current bias should be towards greater transparency, rather than greater protections for industry. Given the complexity and numerous constituencies involved in performing clinical trials, it unlikely that there is a single law or policy measure that could guarantee publication of trial data. In fact, a strict law could potentially backfire and result either in a chilling of research altogether, the movement of clinical research to different locales, or design of trials specifically to avoid reporting requirements.
Instead, a broad range of measures should be applied, both by gradual extensions of FDAAA requirements and pressure from the front lines of clinical trials, with the goal of gradually introducing a culture of transparency and availability of primary results.
Phase I Trials also be added to the reporting requirements of FDAAA, rather than being optional. The preponderance of evidence from pharmaceutical company behaviour over the last decade seems to be that they cannot be relied upon to naturally do the right thing. Given that Phase I trials are intended to evaluate safety, they are highly relevant to protecting patients from risks, and there is potentially much to be learned even from a drug whose development is terminated due to safety issues. In the current environment, data from such a situation might still never become available, and future trials with similar interventions could be performed and put subjects at unnecessary risks. In addition to the importance of publishing outcomes of even Phase I trials, protocols should also be made available as early as possible, so that experts outside of the FDA can review them. For example, the tragedy involving the phase I trial of TGN1412 could have been prevented if the Phase I protocol had been subjected to more outside review, which could have predicted the possible risks and proposed a safer protocol in which subjects were not all dosed simultaneously. The risk to the company of having competitors know about the existence and timing of the study would have been minimal, and avoiding this mishap would have been much better for the company overall, which failed along with the TGN1412.
FDA funding should be increased, so that the agency can operate with greater flexibility and independence. At present, a significant portion of the FDA budget comes from user fees that are collected from companies submitting trials for evaluation. Although there is no evidence that these fees directly influence the outcomes of evaluations, it is likely that the FDA would be able to operate with greater independence and confidence if its funding was more secure and if more inspections could be done.
Finally, the grassroots measures described in the previous section could be implemented on a limited scale. Institutions, such as a university or hospital should set standards for clinical investigator agreements that require all data (including Phase I trials) to be made publicly available in ClinicalTrials.gov at the conclusion of a trial. Imposing such a standard would cost very little, but could be implemented quickly and will not require a protracted legislative process. Individual physicians could also refuse to participate in trials that did not make data available. For example, in the Actonel case described in the first section of this paper, the initial controversy arose because the clinical investigator had agreed contractually to cede publication rights to the company and to keep data confidential. Stronger guidelines at the outset of a trial would prevent obfuscation and withholding of data. Lastly, patients should also consider refusing to sign away access to their own clinical data. Given the difficulties in finding and enrolling patients in clinical trials, patients are in a strong position to negotiate on some of the terms of their participation. In conjunction with the full implementation of FDAAA and the possible extensions discussed, grassroots efforts to insist on greater clinical trial transparency could help ensure faster detection of safety risks, faster disclosure of drug development realities for investors, and better data with which doctors and the medical community can deliver evidence-based care.