Disease severity varies widely in patients with severe sepsis. Previous trials (IL-1RA, TNF-sR p55, antithrombin, and drotrecogin alfa activated (DAA)) suggest that more severely ill patients benefit most from treatment.
The aim of this study was to evaluate the efficacy of eritoran for interaction effects with baseline illness severity.
Prospective covariates from a randomized, double-blind, placebo-controlled, phase 2 trial were analyzed for treatment interaction measured by 28-day mortality. Breslow-Day and multiple logistic regression (LR) were used to assess categorical (CAT) and continuous (CONT) treatment by severity-of-illness interactions.
Modified intent-to-treat population (n = 292) all-cause 28-day mortality was: placebo, 33.3% (32/96); total eritoran 45 mg/105 mg, 29.6% (58/196). LR analysis identified Acute Physiology and Chronic Health Evaluation (APACHE) II scores, Predicted Risk of Mortality (PROM) scores, IL-6, age, sex, race, and eritoran as associated with survival outcomes. Significant treatment interactions were observed (eritoran vs. placebo) for baseline covariates: APACHE II (CAT, P = 0.059; CONT, P = 0.035); PROM scores (CAT, P = 0.028; CONT, P = 0.008); number of organ failures (CAT, P = 0.079); international normalized ratio (CAT, P = 0.05); and acute physiology score (CONT, P = 0.039). No significant treatment interactions were observed with age, sex, shock, DAA use, infection site, microorganism type, platelets, IL-6, or endotoxin levels. Interaction results were similar for eritoran 105 mg only versus placebo.
Potential survival benefits of eritoran in severe sepsis patients may be associated with high severity of illness. Treatment by disease severity interaction will be further explored in a phase 3 trial.