Non-inferiority design: Issues and challenges in anti-infective drug trials for severe bacterial infections
View Presentation View Presentation
*Thamban Valappil, FDA 

Keywords: Non-inferiority trial, antimicrobial, mortality endpoint, prior anti-biotic use, HABP/VABP, non-inferiority margin, constancy assumption

Non-inferiority (NI) trial designs are appropriate for severe or life-threatening diseases where the magnitude of active control treatment effect over placebo is large, reliable, and based on survival benefit to patients. There have been a number of public discussions in recent years regarding the use of active-controlled trials designed to show non-inferiority as a basis for approval of antimicrobial drug products. These discussions have considered sensitive and well-defined efficacy endpoints that are clinically meaningful. The time point of assessment of these endpoint(s) and disease severity are important factors in the determination of the NI margin. For serious infections such as hospital-acquired /ventilator-associated bacterial pneumonia (HABP/VABP), an investigator assessment of cure or ‘clinical response’ as the primary efficacy endpoint is not seem possible as it includes biomarkers that are not a direct measure of how a patient feels, functions or survives; and may not be on the causal pathway of the disease. Historical evidence of treatment effect for HABP/VABP is available only for 28-day all-cause mortality as the primary endpoint and there is no data to justify a NI margin for clinical response outcome, although it could be of some interest. The discussion will include the design and analysis considerations in current non-inferiority trials, impact of prior anti-biotic therapy use and its potential to bias the non-inferiority findings and other challenges.