FDA guidance on non-inferiority trials

FDA guidance on non-inferiority trials
10th February 2017 Sue Carr

As if justifying one wasn’t tough enough…now Sponsor have two non-inferiority margins to deal with?! Dealing with FDA’s new guidance on non-inferiority trial design and analysis

In November 2016 FDA issued their final guidance on non-inferiority trials to establish drug effectives, coming 6 years after the initial draft guidance – click here to view.

Active-controlled non-inferiority trials are sometimes necessary in drug development being where a new drug is compared to a proven control with the general aim being to show that the new drug is acceptably close in efficacy to the control.

Non-inferiority trials are typically tough to design, involving complex notions such as ‘assay sensitivity’ and ‘constancy’, and setting and justifying the non-inferiority margin be a difficult and challenging exercise that requires statistical experience and insight. This is made even more complex by the formal introducing of two margins, the so-called M1 and M2 margins. The former M1 margin is to be constructed such that, in meeting it, evidence is provided that the new drug would have beaten placebo if placebo could have been included; and the latter M2 margin is a clinically defined margin, more stringent than M1, and defined such that, in meeting it, evidence is provided that the new drug retains some desired fraction of the efficacy of the active control drug.

Fortunately, KJC Statistics has considerable experience with non-inferiority trial design and analysis, having published multiple times in the area, and can help you navigate this complex and sometimes confusing new FDA guidance. Contact us for more information or guidance.