Researchers investigating the level of interaction between the US Food and Drug Administration and pharmaceutical companies found that no meetings to discuss study design took place in 20% of recent new drug approvals.
The study, published in JAMA, also reports that when meetings did occur, a quarter of the recommendations made by the US Food and Drug Administration (FDA), for the improvement of study design or primary outcomes, were not heeded.
Federal regulations encourage meetings between the FDA and pharmaceutical companies in order to improve studies that aim to test the efficacy and safety of new drugs.
These meetings can often lead to FDA recommendations for improving the research. However, not only are such meetings not mandatory, but also pharmaceutical companies are not bound to follow any recommendations that might be made.
The research was conducted by Dr. Steven Voloshin, of the Dartmouth Institute for Health Policy and Clinical Practice in Lebanon, NH, and colleagues. The team conducted a review of around 200 FDA documents pertaining to 35 new drugs, approved between February 1st, 2011, and February 29th, 2012.
The documents included filing checklists, medical reviews, memos, minutes from meetings, statistical reviews and summary reviews.
All comments from the FDA in the documents were analyzed, along with recommendations made about the design of drug studies – controls, doses and study durations, for example – and their primary outcomes. How these recommendations would affect the quality of the studies was also characterized and recorded.
Pharmaceutical companies met with the FDA to discuss the studies for 28 of the 35 newly approved drugs. A total of 53 recommendations were made by the FDA regarding study design or primary outcome for 21 of the drug approvals.
The researchers deemed that 51 of these recommendations would increase the quality of the drug studies, through measures such as adding controls or lengthening the study’s duration. The other two recommendations were considered to have an uncertain effect.
Of the 53 recommendations made by the FDA, pharmaceutical companies were found to comply with only 40 (75%). Examples of noncompliance included choosing progression-free ahead of overall survival as a primary study outcome, and conducting uncontrolled studies when randomized trials of the drugs brentuximab and crizotinib were requested.
Pharmaceutical companies also have the opportunity to request that the FDA review key trial protocols. The FDA agrees to not object to any study design issues when reviewing drugs for approval if they have previously endorsed trial protocols.
In the study, the researchers found that protocol reviews were requested in 21 of the 35 new drug approval cases. In these 21 reviews, the FDA only endorsed protocol in 12 cases.
The authors acknowledge that their study was limited by only analyzing drugs that had been given approval by the FDA. However, they believe that rejected drugs may have lower compliance with recommendations and fewer special protocol assessment endorsements.
In order to improve the quality of drug approval studies, the authors suggest increasing the powers of the FDA as a potential approach:
“One approach for enhancing quality of drug approval studies would be to institute mandatory FDA review of pivotal trial protocols with the power to issue binding recommendations, which may be even more important with increasingly flexible approval pathways.”
A previous report commissioned by the FDA suggested that stronger early FDA involvement could improve the approval process, through speeding up the approval of effective drugs and identifying more clearly ineffective or harmful ones, report the authors.
A previous study published in JAMA found that two thirds of dietary supplements recalled by the FDA for containing banned pharmaceutical ingredients still contained the offending substances at least 6 months after their recall.