Hospitalizations as an outcome measure in COURAGE-ALS

Abstract

Objective: To describe the development of a methodology to characterize hospitalizations and their relationship to amyotrophic lateral sclerosis (ALS) and provide results using this process in a phase 3 trial of reldesemtiv in ALS. Methods: ALS clinical trialists assisted in developing a classification system to determine if a hospitalization was related to ALS (HR-ALS), unrelated (HU-ALS), or if the relationship was indeterminate (HI-ALS) and this was applied by the investigators to hospitalizations in COURAGE-ALS. Time to first hospitalization and number of hospitalizations were compared between those assigned reldesemtiv or placebo for up to 48 weeks. Demographic and clinical features were evaluated for prediction of hospitalization risk; this analysis was limited to those participants who completed the first 24-week double-blind placebo-controlled portion of the trial. Results: COURAGE-ALS terminated early due to futility. Time to first hospitalization was similar in the reldesemtiv compared to placebo arms as was the incidence, with 86 of the participants (17.6% of those originally assigned placebo and 18.0% originally on reldesemtiv) experiencing an event. The largest percentage of events was classified as HR-ALS for both placebo (64%, 18/28) and reldesemtiv (76%, 44/58). In a multivariate model, only bulbar or respiratory onset disease was a significant risk factor for hospitalization. Conclusion: While most hospitalizations in COURAGE-ALS were HR-ALS, HU-ALS and HI-ALS also occurred. When using hospitalization as an endpoint in an ALS clinical trial, recording its relationship to ALS provides additional details to characterize disease burden and clinical meaningfulness of the endpoint.