Despite a 2007 law requiring sponsors of clinical trials to make final results publicly available within one year, fewer than 10 percent of cancer trials met the deadline between 2007 and 2010, and more than half failed to report findings within three years, according a new analysis by French researchers (JCO 2013;31:2998–3003).
The Food and Drug Amendments Act (FDAA) of 2007 requires results to be posted at the FDA's ClinicalTrials.gov website within 12 months, yet the review of 646 late-phase trials, including 209 randomized controlled trials (RCTs), found that only 13 were reported in that time at the website or in peer-reviewed publications listed on PubMed.
Lead author Agnes Dechartres, MD, Assistant Professor of Clinical Epidemiology at Hôpital Hôtel-Dieu and the Centre d'Epidémiologie Clinique, INSERM, in Paris, said that several issues might explain the delays: “It could be due to a lack of knowledge about the law, lack of time for posting results, or complications with trials involving drugs seeking initial approval or approval for a new use,” she said. “In the latter two cases, it would be good to have [data] posted on time for transparency and validation of results. It is also likely that some authors wait for publication in journals before posting their results at ClinicalTrials.gov, whereas posting should be independent from publication in journals.”
One year after completion of the trials, the cumulative percentages of studies with results posted at ClinicalTrials.gov was nine percent, in peer-reviewed journals was 12 percent, or in both were 20 percent, respectively, the researchers found. For RCTs, the percentages were 12, five, and 17 percent, the researchers found.
At 36 months, the rates were 31, 35, and 55 percent, respectively, and for RCTs, 38, 32, and 56 percent. Public availability of phase III trials was five percent at 12 months, 39 percent at 24 months, and 64 percent at 36 months.
On September 5, 2012, the U.S. Department of Health and Human Services (DHHS) transferred authority for monitoring and enforcing mandatory clinical trial reporting to the FDA, making the agency responsible for determining whether any data has not been submitted, or was submitted but is false or misleading in any way. If information is found lacking, the agency has the authority to notify the responsible party and give them the opportunity to submit the required data within 30 days.
Systemic Publishing Problem?
But, publishing trial results, especially in more prestigious journals, is part of the problem, according to Dechartres: “The entire system contributes to delays. “Investigators try to publish their results in journals with the highest impact factor, and publishing results in [such] journals typically takes a long time, in part because of multiple submissions and rejections. Some studies have also linked time to publication with whether or not study results are positive, with positive results being published more often and more quickly than negative ones,” she noted.
Studies have shown that the median timeline for a study to be published in a peer-reviewed journal is between three and five and a half years.
Section 801 Established Penalties
Section 801 establishes penalties for failing to post completed trial results within one year, including civil monetary penalties and, for federally funded studies, withholding of grant funds.
“Although these penalties can exhort sponsors to post results in a timely manner, improved monitoring compliance and contacting the responsible party when results are not posted might improve compliance before penalties are levied,” she said. “Again, certification should be available at ClinicalTrials.gov for complete transparency in order to explain why the public cannot access the results.”
The consequences of selective and/or delayed reporting are significant for both cancer patients and oncologists, note the authors of an accompanying editorial, Sandip M. Prasad, MD, and Charles L. Bennett, MD, of the University of South Carolina College of Pharmacy (JCO 2013;31:2981–2982).
Incomplete reporting “has potentially profound effects on patient management and counseling, because timely and complete dissemination of results can refine patient treatments, outcomes, and safety.”
Not reporting study results, especially those from clinical trials with statistically non-significant findings or where significant toxicities are identified, can have potentially far-reaching effects, they said, noting that meta-analyses of both safety and efficacy outcomes are affected by missing data due to publication bias in favor of positive results.
Not publishing studies with non-significant findings, the editorial said, “typically leads to overestimation of treatment effects by the inclusion of published (and therefore more likely positive) studies, and underestimation of safety effects by omission of non-published safety findings.”
‘Alice in Wonderland Feel’
Several studies published in BMJ last year reported numerous instances where clinical trial reporting fell short. In one, researchers looked at unpublished evidence of existing meta-analyses of nine drugs approved by the FDA in 2001 and 2002 and found identical estimates of drug efficacy in only three out of 41 cases. In another paper, researchers at the University of Nottingham, UK, reported that only 22 percent of trials conducted in 2009 met the 2007 one-year FDA reporting mandate for logging their results on the ClinicalTrials.gov website.
Commenting on the findings, Elizabeth Loder, a contributing editor at BMJ, and Richard Lehman, a consultant psychologist at Oxford University, said there is “an Alice in Wonderland” feel to the Nottingham investigators' efforts: “Acting on the public's behalf, [they searched] over hill and dale and among the paperwork of regulatory bodies and drug companies to put together pieces of data that should have been freely available in the first place,” they wrote in their accompanying editorial.
One month after the study was published, on February 13, 2012, Henry Waxman (D-Calif) and several other lawmakers sent a letter to NIH Commissioner Margaret Hamburg, seeking an explanation. Jeanne Ireland, FDA Assistant Commissioner for Legislation, responded that the analysis included some trials that were completed before the new law took effect, and that the agency was actively “pursuing responsible parties” in 15 clinical trials where results had not been posted by the FDAA deadline.
‘No Excuse for Delays’
Andrew P. Prayle, PhD, lead author of the University of Nottingham study, told OT that patients, especially those who participate in clinical trials, are owed more timely and accurate reporting: “Clinical trials could not be conducted without patients willing to participate in them, and they deserve to know the results,” he said. “The finding that results were not available for half of the examined trials three years after completion leaves little room for doubt that things need to improve. There is no excuse for not registering trial results—how else can we be certain that any statistically significant results were not cherry-picked out of a large number of measured variables?”
Nonetheless, he said he did consider the current FDAA timeline to be somewhat tight: “One year is difficult, but it is achievable. After obtaining all data on the primary completion of a study, it can take a lot of work to perform essential data integrity checking, even in small trials, run analyses, and then organize meetings with the investigators prior to disseminating the results.”
Prayle said that he believes that the NIH and FDA have done a good job in disseminating the message to both journals and researchers that the deadline for reporting results can and should be met.
“Key leaders like Dr. Deborah Zarin, at the NIH Intramural Research Program, have spent a considerable amount of time on this, including contacting trial investigators behind the scenes to encourage regular updates at the ClinicalTrials.gov website.”