Improving Access to Clinical Trial Data

THE CONCERN ABOUT transparency is all around us, from claims made about the foods we eat to how a person’s data is shared among entities. Similarly, transparency is also a concern with the growing data sharing movement among clinical trial shareholders, including sponsors, clinicians, participants and other interested parties who may be able to utilize the information earlier and more easily to further scientific understandings and initiatives in healthcare. Specifically, when sharing data during the clinical trial process, protecting patient privacy and confidential information about the drug or device being evaluated must be balanced with concerns about a lack of transparency and timely reporting that plagues the ability to effect change, thus stifling innovation.

Data can be used as a mechanism for progress when it is appropriately gathered, analyzed, reported and, in turn, reallocated to benefit future studies, saving time and dollars and reducing duplication of efforts, thus speeding the pace of research. On the other hand, there are valid concerns about how much and what type of data is shared, as well as how sharing might negatively impact confidentiality of critical information and privacy. Significant issues are on the line as the community works together to find pathways forward, respectful of the interests of all parties, to benefit the greater good.

So, how do we get there from here? Is the problem with data transparency determined purely from the investigators’ standpoint? Is a siloed data sharing infrastructure unable to facilitate innovation? Or, more inclusively, is there a collective concern, whereby improvements in any of these limitations will lead to forward steps resulting in greater progress? From manufacturers to investigators, patients and those administering study oversight and review, better collaboration will fit the puzzle pieces together and foster proactive and positive change.

FDAAA and ClinicalTrials.gov

The registration of study trials and subsequent data reporting enable the U.S. Food and Drug Administration (FDA) to oversee the safety of study design protocols and outcomes. When studies are not registered, however, FDA loses that ability, and patients, providers, fellow investigators and other interested parties are left with an incomplete picture of the trials, as well as their effects on patient populations.

As a matter of record, the 2007 FDA Amendments Act (FDAAA) mandates sponsors of applicable clinical trials register and report summary results at the congressionally authorized clinicaltrials.gov registry created to provide information about publicly and privately supported clinical trials on the National Institutes of Health’s (NIH) database. As part of that mandate, FDAAA also requires sponsors to report trial results within one year after the completion of data collection for the prespecified primary outcome or within one year after the date of early termination, unless there are legally acceptable reasons for the delay.

Unfortunately, reviews and comparisons of trials listed on clinicaltrials.gov have found compliance with these mandates has been historically poor regardless of the funding source, trial phase and whether the trial is under FDA oversight. Interestingly, one study identified trials funded by industry tended to have greater reporting results than those funded by NIH or other sources. And, earlier phase trials tend to report fewer results compared to Phase IV trials, with the presumption earlier trials are more proof-of-concept and may contain more proprietary information. However, even within a five-year period, only 80 percent of industry-funded studies reported their results, compared to 50 percent for NIH-funded studies and 42 percent of trials funded by academic or other sources. In addition, studies conducted without FDA oversight were also less likely to report on time.¹

Recommendations for Change

In 2015, in an attempt to improve the entire process and encourage shareholder participation, the Institute of Medicine (IOM) released a report titled “Sharing Clinical Trial Data — Maximizing Benefits, Minimizing Risk,” which included a wide range of recommendations for addressing the problems of three types of study data:¹

• Metadata (or data about the data, including protocols),
• Raw data (data collected from participants that has been “cleaned, abstracted, coded and transcribed to become the analyzable data”) and

• Analysis data (such as summaries, lay summaries, etc.).²

The report recommends a culture of greater data transparency by all parties to improve the bottom line. With greater transparency, data sharing would become an expected part of a study trial process in which funders contribute to the development of an infrastructure that allows for data sharing in accordance with specified grant terms and per regulations. In addition, it recommends funders consider past data sharing as part of their future considerations for funding requests and, as part of that funding, include enforcement requirements for data sharing as part of grant terms.

The IOM also recommended regulatory authorities around the world work toward harmonizing clinical study report templates that avoid revealing stockholder and participant confidential information. By doing this, future clinical trials could be designed with the expectation of data sharing among stakeholders and use of common data elements as part of the study design unless there is a compelling reason not to do so.

The release of study data would then provide an opportunity for those involved in clinical trials to publish their results prior to allowing access to secondary investigators in a manner in which commercial interests of sponsors working to gain regulatory approval is protected. It is the IOM’s recommendation that any adverse event summaries, summary level results and lay summaries be provided to trial participants and be publicly available no later than 12 months after study completion. In addition, IOM recommends the full data package be shared no later than 18 months after study completion (barring any pending regulatory applications).²

When deciding to release data, privacy considerations for protecting patient data cannot be understated. Patients must understand not only the study protocols for the trial they are considering, but how and what data might be shared and what protections for that data are in place. For instance, there are many ways to protect data confidentiality, including de-identification, data use agreements and independent review panels.

Change via Social Contracts

Building upon the IOM 2015 report in support of clinical trial data sharing, FasterCures (an action tank that works to speed and improve the medical research system) suggests a social contract with implied rights and responsibilities for each stakeholder. Dubbed Health Citizenship, “The path to better health and the advancement of science,” says FasterCures, “begin[s] and end[s] with engaged patients.”³

According to FasterCures, lackluster data reporting, including registration of study trials, is a lost opportunity for cascading positive impacts on the costs of research and the speed of scientific discovery. Reporting both positive and adverse outcomes in a timely manner equates to improved efficiencies in attempts to reproduce results, as well as reduces redundancies in costs and time while saving precious research dollars. It goes without saying, there is no larger benefit when negative results go unreported because it eliminates the ability of providers and patients to benefit from the research.

It is the hope of many thought leaders that as data collection, sharing and transparency continue to evolve, including first-person sourcing such as wearable devices and the mining of de-identified electronic health records, the barriers currently posed with traditional methods of research and regulatory requirements are reanalyzed fostering a forward trajectory and increasing the speed of progress.

Heeding the Call

FDA is mandated by the FDAAA to maintain a website that provides safety information, including safety alerts, warning letters and links to clinical trial information. In addition, it has begun to incorporate broader transparency policies designed to bridge the gap of some discrepancies between the comprehensive information within its possession and other incomplete information released to the public,⁴ which could have negative outcomes for public health.

FDA Commissioner Scott Gottlieb issued a statement in January 2018 on the agency’s new steps to enhance the transparency of clinical trial information to support innovation and scientific inquiry related to new drugs. This includes a pilot program to repackage information released in clinical study reports (CSRs) so the information is easier to access on the agency’s drug approval database drugs@FDA. It is FDA’s view that releasing more premarket and postmarket clinical trial data will enable greater extraction of detailed clinical evidence supporting FDA’s approval decisions. CSRs contain bottomline detailed scientific summaries, including the clinical trial’s methods, efficacy and safety results. As part of the pilot, FDA will release, as applicable, the study report body, including protocols and amendments, as well as the statistical analysis plan for each of the participating product’s pivotal studies, while protecting patient privacy, trade secrets and confidential commercial information contained in the CSRs.

In addition, FDA’s Digital Health Software Precertification Pilot Program is evaluating nine companies designing medical device software against quality and excellence standards, including product quality, patient safety, clinical responsibility, cybersecurity responsibility and proactive culture. The agency will then work with the developers to collect and interpret real-world information about their medical device software to assess the product’s safety and effectiveness and to address any emerging risks.⁵

Practical and Pragmatic Steps Forward

As the opportunity for meaningful data collection accelerates, study sponsors, clinicians and their patients all play an active role in ensuring change progresses in a thoughtful manner. However the details of these initiatives play out, efforts for improving transparency and encouraging active participation for all parties will contribute to moving healthcare forward toward improved patient outcomes.