To disclose or not to disclose? Shakespearean allusions aside, there is little question that drugmakers and researchers are under increasing pressure to comply with demands to publicly disclose clinical trial results. The failure to do so, of course, raises concerns that undisclosed results can lead to biased medical literature, especially in meta-analyses, and inappropriate treatment decisions.
Now, another study finds that clinical trial results have been kept under wraps. The latest evidence appears in the Journal of Clinical Oncology and reports that nearly half of cancer drug trials reviewed had no results publicly available at ClinicalTrials.gov or in journals three years after completion.
Moreover, compliance with the Food and Drug Administration Amendments Act of 2007, which mandates reporting of summary clinical trial results on the ClinicalTrials.gov within one year of completion, was poor - only 13 percent were published. The researchers reviewed 646 Phase II, III and IV cancer drug trials, including 209 randomized controlled trials, or RCTs, which were registered with ClinicalTrials.gov or published in PubMed between December 2007 and May 2010 (here is the abstract).
The study reveals “an incomplete picture of cancer drug reporting that has potentially profound effects on patient management and counseling, as timely and complete dissemination of results can refine patient treatments, outcomes and safety,” according to an accompanying editorial written by a pair of professors at the University of South Carolina.
This is hardly the first time such trends have been noticed. A series of papers published in BMJ last year found a remarkable amount of clinical trial data regularly went missing. One notable example - just 22 percent of studies examined had adhered to the FDAAA mandate (back story).
But the JCO review attempts to focus attention on a specific disease state in a way that underscores the extent of the overall problem. Cancer, after all, is an insidious malady that claims so many victims every day in so many different ways. And the lack of timely information – in the form of clinical trial results – only compounds the difficulties in attacking the disease.
Specifically, the JCO study found that one year after completion, results for 9 percent were posted at ClinicalTrials.gov, 12 percent were published in journals and 20 percent were available from both sources. For RCTs one year following completion, results for 12 percent were posted at ClinicalTrials.gov, 5 percent were published in journals and 17 percent were available in both places.
At three years, for all clinical trials and RCTs, respectively, the results for 31 and 38 percent were posted online at ClinicalTrials.gov, 35 and 32 percent were published in journals, and 55 and 56 percent were available from both sources. For phase III trials, the results were publicly available for 15 percent of the studies at 12 months, 39 percent at 24 months and 64 percent at 36 months.
Interestingly, compliance with the FDAAA was higher for trials funded only by industry compared with mixed funding or only academic funding – 20 percent vs. 11 percent vs. 6 percent, respectively. What is not clear, the editorial notes, is whether selective incomplete reporting was due to differences in studies with statistically non-significant outcomes or corporate priorities thwarted early dissemination.
And here are a few further details concerning the composition of the trials – 85 percent were Phase II; 71 percent indicated patient recruitment had been completed and RCTs represented 32 percent of the total. Overall, 201 trials or 31 percent had only academic funding; 156, or 24 percent, had both industry and academic funding and 289, or 45 percent, had industry funding.
What should be done? Sandip Prasad and Charles Bennett of the University of South Carolina suggest in their editorial that individual researchers who receive federal funding should be required to report results to ClinicalTrials.gov as a condition for receiving additional funds and the FDA needs to get tough with recalcitrant drugmakers. They also endorse an idea proposed by Harlan Krumholz, who heads the Yale Open Data Access project, to create an independent entity to provide independent data analysis.
STORY ENDS HERE