Boutron I, Dechartres A, Baron G, Li J, Ravaud P. Sharing of Data From Industry-Funded Registered Clinical Trials. JAMA. 2016;315(24):2729-2730. doi:10.1001/jama.2016.6310
Access to individual patient-level data from clinical trials could be an important step forward in clinical research.1- 4 Some pharmaceutical companies have committed to share such data. The largest repository is the Clinical Study Data Request (CSDR) website, wherein companies voluntarily list studies for which data can be requested.5 To evaluate the completeness of data sharing on CSDR, we investigated the proportion of randomized clinical trials (RCTs) registered at ClinicalTrials.gov that were listed at CSDR.
All drugs other than vaccines listed on CSDR by all sponsors actively involved in data sharing, defined as listing at least 100 studies in June 2014, were studied. One company listing less than 100 studies and 3 companies listing no studies were excluded.
ClinicalTrials.gov was searched by using the sponsor name in the “search field” on January 15, 2016. Studies were selected that were “completed” or “terminated” with a completion date of at least 18 months before the search date (to allow time for studies to be analyzed, published, and listed on CSDR); were recorded as “interventional” and “randomized”; and reported the drug name or any synonym (identified on the Drug Information Portal of the US National Library of Medicine) in the “intervention” field and the appropriate sponsor in the “sponsors” field. Single-group designs were excluded.
All available data were downloaded, including trial registry identification number for all studies listed at CSDR for the sponsors, and the ClinicalTrials.gov and CSDR databases were merged by trial registry number. All sponsors registered trials only at ClinicalTrials.gov.
For the 61 targeted drugs from 4 sponsors (drugs: Roche, 13; Lilly, 3; Boehringer Ingelheim, 5; GlaxoSmithKline [GSK], 40), 966 RCTs (462 751 participants) registered at ClinicalTrials.gov were identified; 512 RCTs (53%) (342 271 participants; ie, 74% of the participants involved in these studies) were listed at CSDR. The availability of each document at CSDR (ie, raw data set, annotated case report form, protocol, analysis-ready data set, reporting and analysis plan, clinical study report) varied from 83% to 99%; records for 385 RCTs (40%) reported that all documents were available. The proportion of registered trials listed on CSDR varied from 33% for Roche to 66% for GSK and trials with all information available from 24% for Boehringer Ingelheim to 58% for GSK (Table).
Also, 440 RCTs (175 177 participants) listed at CSDR but not registered at ClinicalTrials.gov were identified. These studies were sponsored by Roche (34 studies; 18 420 participants), Lilly (2 studies; 512 participants), and GSK (404 studies; 156 245 participants). For 359 of these RCTs (82%), the study start date recorded at CSDR was before 2005 (when registration was first required).
Despite a delay of 18 months since the completion of drug trials by the company sponsor, only 53% of the RCTs from the 4 sponsors registered at ClinicalTrials.gov were listed at CSDR, with differences between sponsors. Data were available for a large number of participants, but an equally large amount of data was not available.
Limitations in this study need to be acknowledged. We did not evaluate whether data were actually made available. For this, a research proposal should be submitted, reviewed, and approved by an independent review panel, a data-sharing agreement signed, and the data obtained through the website. However, the number of submitted proposals is low and thus does not give a complete picture of the willingness of the companies to share data.6 Furthermore, all drugs manufactured by the sponsor and the results of trials were not considered, and therefore selective listing on the website cannot be excluded. Only studies that were listed at CSDR were considered, but data from studies that are not listed can be requested. The sponsor policy related to data sharing was not taken into account. Some sponsors agree to share only selected studies. However, the purpose of the study was to evaluate the amount of available data, regardless of whether a trial was not listed because of company policy or for another reason. In addition, only 1 repository was evaluated.
Corresponding Author: Isabelle Boutron, MD, PhD, Centre d'Epidémiologie Clinique, Hôpital Hôtel Dieu, Aile A2 1er étage, 1, Place du parvis Notre Dame, 75181 Paris Cedex 4 (firstname.lastname@example.org).
Author Contributions: Drs Boutron and Baron had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Boutron, Dechartres, Ravaud.
Acquisition, analysis, or interpretation of data: Boutron, Dechartres, Baron, Li.
Drafting of the manuscript: Boutron.
Critical revision of the manuscript for important intellectual content: Dechartres, Baron, Li, Ravaud.
Statistical analysis: Baron.
Obtained funding: Ravaud.
Administrative, technical, or material support: Li.
Study supervision: Ravaud.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.