The data liberation movement: regulation of clinical trial data sharing in the European Union and the United States.

• Author: Westergren, Amy

1. INTRODUCTION II. BACKGROUND III. RISKS AND BENEFITS OF SHARING NON-SUMMARY DATA IV. THE PATH TO NON-SUMMARY DATA SHARING A. Introduction B. Transparency Laws C. EMA Policy on Reactive Non-Summary Data Sharing D. The EMA and Proactive Non-Summary Data Sharing E. The FDA Policies and Reactive Non-Summary Data Sharing F. FDA and Proactive Non-Summary Data Sharing V. CONCLUSION I. INTRODUCTION

   There are currently over 200,000 studies registered on clinicaltrials.gov, (1) and 4,000 new studies take place in the European economic area each year. (2) Clinical trials yield thousands of pages of valuable safety and efficacy data. (3) The trial's sponsor submits that data to a regulatory body, like the U.S. Food and Drug Administration (FDA) in the united States or the European Medicines Agency (EMA) in the European Union, to obtain approval to market a drug, biologic, vaccine, or other therapeutic health product to the public. (4) Clinical trial data submitted to support a marketing application could be pooled and re-analyzed to benefit the public, (5) but only if the data are made available for those uses. Historically, pharmaceutical manufacturers were unwilling to share the raw data they generated in clinical trials, and the policies of the United States and the European Union enabled that secrecy.

   However, right now, clinical trial data are the subject of an international discussion about government transparency, public health, and scientific innovation that threatens the data exclusivity that the drug industry has typically enjoyed. (6) Industry is under increasing pressure to provide outside scientists access to detailed, patient-level clinical trial data, as well as the protocols followed to collect data and the analytic tools used to interpret its meaning, for the sake of innovation and for the good of public health. (7)

   This comment will focus on the role that regulatory bodies in the European Union and United States are playing in making data sharing a compulsory resultant of the marketing approval process. Each has taken action to increase the amount of data available to the public through governmental agencies. (8) The discussion will explore and compare the paths to increased data sharing taken by the European Union, through the EMA, and the United States, through the FDA. It will show that the evolution of the EMA's early reactive data sharing rules into proactive, patient-level data sharing policies was founded on a public health argument that is not available to the FDA.

   Part II will follow the evolution of data sharing regimes. Part III will put the data-sharing discussion into the context of its real world implications through an evaluation of some justifications for increased sharing as well as potential risks. Finally, Part IV compares the transparency laws governing release of data in the European Union and the United States to discern their similarities and critical differences. The purpose of this comment is to show that the FDA is constrained by U.S. law in a way that the EMA is unconstrained and to discuss the distribution of risk in the data sharing systems that are emerging in the European Union and the United States.

2. BACKGROUND

   Between 1997 and 2008, a global push for greater disclosure of information relating to clinical trials led to policy changes at the World Health Organization (WHO), the World Medical Association (WMA), the EMA, and the FDA and to the creation of three clinical trial registries. As a result of these efforts, the public gained access to basic information about a study. (9)

   In the United States, the National Institutes of Health (NIH) established the country's first clinical trial database, clinicaltrials.gov, in February of 2000, under authority granted to it by the U.S. Congress in the Food and Drug Administration Modernization Act (FDAMA). (10) The FDAMA requires reporting of "a description of the purpose of each experimental drug, patient eligibility criteria for participation in the trial, a description of the location of clinical trial sites, and a point of contact for those wanting to enroll in the trial." (11) In addition, in 2002, the FDA published a guidance document suggesting reporting of four categories of data: (1) Descriptive Information; (2) Recruitment Information; (3) Location and Contact Information; and (4) Administrative Data. (12)

   These categories encompass descriptive information about the study itself, rather than data produced by the study. "Descriptive information" includes general information about the study, its design, the intervention being tested, and the condition or disease it is meant to treat. (13) "Recruitment information" includes the status of participant recruitment, e.g. "recruiting" or "no longer recruiting," along with study enrollment eligibility criteria. (14)

   Similarly, the WHO worked to make general clinical trial information available on a global scale. Participants in a 2004 Ministerial Summit on Health Research called for the WHO to facilitate "a platform linking a network of international clinical trial registers to ensure a single point of access and the unambiguous identification of trials." (15) A 2005 World Health Assembly Resolution expanded on that idea. (16) Participant nations called for better access to clinical trial information for patients, families, patient groups, and others. (17) In 2005, the WHO established the International Clinical Trial Registry Platform (ICTRP). (18) The data stored in the ICTRP is similar to the trial information found on clinicaltrials.gov; it includes generalized information about the purpose of the trial and the sponsor. (19)

   In the European Union, the EudraCT and European Clinical Trials Databases were developed to collect clinical trial information for use by regulators and for release to the public. (20) Regulation (EC) No. 726/2004 created the EMA and called for a publicly accessible database to contain certain information about approved drugs. (21) Additionally, the database was to "include references to data on clinical trials currently being carried out or already completed." (22) In 2009, the European Commission released a list of data fields from the EudraCT database that would be made public. (23) As in the United States, the European Union disclosed general categories of information relating to the sponsor, the investigational product, and the clinical trial. (24)

   The need for greater transparency was recognized outside of government. In 2008, the Fifty-Ninth WMA conference amended the Declaration of Helsinki (Declaration) to include two new principles that related to clinical trial information. (25) The WMA is an independent international organization representing physicians. (26) Its Declaration is a statement of ethical principles for medical research involving human subjects. (27) After 2008, the Declaration included, in Section 19, an expression of the WMA's perspective that registration of clinical trials is an ethical imperative: "Every clinical trial must be registered in a publicly accessible database before recruitment of the first subject." (28)

   Eventually, publication of general information about clinical trials gave way to publication of summary results information. At the time the European Commission's 2009 guidance was published, the EudraCT system did not contain results information. (29) However, the commission made it clear that summary results data would be released to the public once that information actually became available in EudraCT. (30) In 2012, the European Commission provided for the release of some summary results information, (31) and, as of July 2014, EudraCT's programming has been updated to fully implement summary results sharing. (32) EudraCT's summary results fields are mostly identical to those found in clinicaltrials.gov. (33)

   The Food Drug Administration Amendments Act (FDAAA), passed in 2007, mandated the publication of summary results information on clinicaltrials.gov. (34) It expanded the types of trials that must be reported in the database and increased the scope of information that trial sponsors are required to submit. (35) The FDAAA requires the FDA to adopt rules implementing the expanded results sharing requirements by 2010. However, the FDA did not propose a rule under the FDAAA until 2014, (36) and, as of this writing, no final rule has been promulgated. The NIH, the body that maintains the database, modified clinicaltrials.gov in 2008 to enable the implementation of the FDAAA. (37) In spite of a regulatory void, some sponsors began submitting results data in 2008. (38)

   This movement to share summary results data is now giving way to a movement to share raw, participant-level data. There is a call to share not only raw data but the study plans and methods of analysis that constitute the context in which those data were collected and evaluated (Non-Summary Data). (39)

3. RISKS AND BENEFITS OF SHARING NON-SUMMARY DATA

   Proponents for Non-Summary Data sharing cite principled and practical advantages that are grounded in government transparency and public health. Independent scientists can use raw clinical trial data to discover earlier endpoints for drug testing, (40) conduct patient-level meta-analyses for safety and efficacy, (41) and act as a check on regulators. (42) All of these uses have the potential to make the public safer. (43)

   The risks involved in clinical trial participation could be reduced. Locating indicators of a drug's effectiveness earlier in the clinical trial process could shorten the length of trials and reduce the time it takes to get treatment options to patients. (44) If the answer to a research question is apparent from existing data, drug manufacturers could forego a clinical trial, (45) and prevent the repetition of doomed trials on new groups of participants. (46)

   Therapies could be tailored for effectiveness and safety. A single clinical trial utilizes a small groups of...