An Enrichment Strategy For Sepsis Clinical Trials.

INTRODUCTION: Selecting participants for research based on their risk is an enrichment strategy with potential for enhancing clinical trials in sepsis. Adult Septic Shock Information and Stratification (ASSIST) is a tool for estimating mortality risk that incorporates a panel of biomarkers, age, lactate, and chronic health status. We assessed the utility of ASSIST as an enrichment strategy in a clinical trial testing the efficacy of a polyclonal antitumor necrosis factor-α fragment antibody (AZD9773) in adults with severe sepsis or septic shock. We hypothesized that the effects of AZD9773 are dependent on baseline mortality risk, as estimated by ASSIST. METHODS: This was a post hoc analysis of a recently completed trial of 300 subjects, randomized to placebo, low-dose AZD9773, or high-dose AZD9773. The study's primary endpoint was number of ventilator-free days. There were 286 subjects with available plasma samples from study entry. We measured the ASSIST biomarkers and assigned each subject to low, intermediate, and high-risk strata based on their ASSIST mortality probability. To mirror the study's original statistical plan, we estimated the least squares mean ventilator-free days for each study arm within risk strata. RESULTS: The effect of study arm within ASSIST-based risk strata was significant (P = 0.017). Within the low-risk group, there was an increase in ventilator-free days for both drug arms. Within the intermediate-risk group, there was an increase in ventilator-free days among those in the low-dose arm, but a decrease in the high-dose arm. Among high-risk patients, there was a decrease of ventilator-free days in both drug arms. Analogous associations were observed when modeling 28-day mortality. CONCLUSIONS: In this study, a beneficial effect of AZD9773 might have been observed if the trial selected low to intermediate-risk patients. ASSIST has the potential to serve as an enrichment tool for sepsis clinical trials.

Duke Scholars

Author Christopher Lindsell Biostatistics & Bioinformatics, Division of Biostatistics