Clinical Trial Results and Clinical Trial Protocol Mismatch
[Posted on: Thursday, May 19, 2016]

Clinical trial transparency has been a concern for a long time. An analysis by researchers at the University of Oxford found that more than 40% of the trial results published in the top 5 medical journals corresponded to new endpoints not included in the clinical protocol. In some case, none of the outcomes listed in the clinical protocol were discussed in the article published upon completion of the trial. Only about 13% of the published articles honestly reported all the outcomes listed in the original clinical trial protocol. The authors of the survey compare the outcomes listed in the protocol to those reported in the published article. The goal of this survey was to verify all the trial objectives and expected outcomes either listed in the original clinical trial protocol or its official amendment. Any new endpoint, not listed in the protocol but discussed in the publication was deemed new. Protocol amendments are common and must be approved by the IRBs and notified to the regulatory agencies before implementation. Till the data lock for a given trial, sponsors can make changes to a given clinical trial protocol. Per the current laws, sponsors are required to update clinical trial registries such as clinicaltrials.gov website each time changes are made. However, it is well established that sponsors are not diligent about updating all public listings. This could explain the mismatch between published results and clinical protocol. Also, once a trial is completed, the data analysis and mining may lead to newer more interesting outcomes leading to a given sponsor publishing only these outcomes. The published article is not the same as a formal clinical study report. Journals have severe limitations on space available and hence the published results can never contain all results from a given trial. While the researchers at Oxford may have noble goal to increase trial transparency, it seems they may be highlighting the limitations of publishing trial results in medical journals.