FDA taking more cautious approach to non-inferiority trials

29-Aug-2010 - Last updated on 30-Aug-2010 at 09:25 GMT

The FDA is taking a more conservative approach to data from non-inferiority trials when reviewing NDAs, according to a GAO report.

Applicants’ data is now more rigorously reviewed by the US Food and Drug Administration (FDA). This requires companies to adopt superior trial design and more thorough analysis, such as calculation of or justification of non-inferiority margins.

Non-inferiority trials are used to show a treatment is as effective as a currently approved therapy. Approval of a drug that is as effective, or slightly less effective, than the existing standard could be granted based on superior safety, fewer drug-to-drug interactions, convenience or lower cost.

FDA new drug application (NDA) regulations have included provisions for non-inferiority trials since 1985. However, there are still some questions and concerns about their use.

To address these issues a bipartisan group of US politicians asked the Government Accountability Office (GAO) to review the FDA’s use of evidence from non-inferiority trials.

“Our review of FDA’s indication-specific guidance showed that the agency has become more conservative in allowing evidence from non-inferiority trials to demonstrate the effectiveness of new drugs”, wrote the GAO in its review.

This more conservative approach is detailed in 17 indication-specific guidance documents issued by the FDA from January 2002 through June 2010. For eight indications, including cancer and HIV, the FDA states non-inferiority trials may be able to demonstrate effectiveness.

A further nine documents state data from non-inferiority trials may be insufficient. For instance, in 2007 and 2008 the FDA published guidance for several less severe infections, such as bacterial sinusitis, which state non-inferiority trials may be unable to demonstrate effectiveness.

Tougher indication-specific guidance is supported by a March 2010 document focused on non-inferiority trials. In the March draft guidance the FDA covers in greater detail the key principles involved in using non-inferiority trials to establish a drug’s effectiveness.

Findings and response

The GAO reviewed 175 new molecular entity (NME) submissions processed by the FDA from fiscal 2002 to 2009. Of these, 43 included evidence from at least one non-inferiority trial. About half of submissions using data from non-inferiority trials were for antimicrobial drugs.

Approval was granted for 29 of the 43 NDA’s which included data from non-inferiority trials. Data from non-inferiority trials was pivotal in 18 of the approved applications. In the other 11 cases approval was granted based on data from other trial types, such as superiority versus placebo.

The findings of the review were cautiously welcomed by Senator Chuck Grassley and Representatives Henry Waxman, John Dingell, Bart Stupak, and Edward Markey, who collectively requested the review.

In general the politicians welcomed the tougher approach adopted by the FDA but called for further work, in collaboration with industry, to continue progressing understanding of non-inferiority trials.

“Although the pipeline for antibiotics is critically bare, no one benefits from ineffective drugs—and so I encourage FDA and industry to continue to collaborate to do everything possible to ensure that safe and effective antibiotics are developed”, said Waxman.