In 2005, the International Committee of Medical Journal Editors (ICMJE) required registration of trials in a public registry at or before the time of first patient enrollment as a condition for publication in their member journals. This policy significantly stimulated trial transparency as noted by sharp increases in clinicaltrials.gov trial counts and development of company trial registries (which kept us very busy at Citeline). Nearly seven years ago, the Food and Drug Administration Amendments Act (FDAAA) was enacted and mandated registration of certain clinical trials and their results to the US NIH clinicaltrials.gov web site, specifically Phase II-IV studies subject to FDA regulation (i.e. trials with at least one US site or are conducted under an IND or IDE).
We decided to take a quick snapshot of trial result reporting by two sources to gauge how compliant the industry has been with respect to FDAAA. According to clinicaltrials.gov, a total of 10,793 interventional phase III trials sponsored by the NIH, another US Federal Agency, or industry have been completed. Of these, just under 3,000 or 27.5% have results available. Trialtrove identifies a total of 11,596 completed industry-sponsored Phase III trials with 66% reporting results (either as a full publication or in medical conference abstracts). For Phase II trials, clinicaltrials.gov reveals 18% of all interventional trials disclosed results, while 63% of trials have had some report on endpoints or trial outcome per Trialtrove. So, depending on the source, we’d give the industry either a gentleman’s C or B+!
A related issue to compliance is the perceived bias in reporting negative clinical trial outcomes. Trialtrove reviewed over six years of results from clinical trials sponsored by industry or large cooperative groups, be they journal articles, conference abstracts, press releases, or other sponsor-issued documents. For Phase II trials, we have to give the industry an A for transparency in reporting negative primary outcomes. In this quick analysis, we’ve summed up negative to include actual negative primary outcomes with reports where the primary endpoint result was either unknown or not discernible as either positive or negative – erring on the side of not meeting the primary endpoint(s). Despite this bias on our part, we are inclined to grade the Phase III trials as a solid “B” to industry.
Industry sponsors appear to be exceling in terms of transparency for phase II trial results, but are the rest of their marks satisfactory enough? Let us know what you think, please.