Edaravone efficacy in amyotrophic lateral sclerosis with reduced forced vital capacity: Post-hoc analysis of Study 19 (MCI186-19) [clinical trial NCT01492686]

Abstract

Background Edaravone slowed the rate of functional decline in subjects with amyotrophic lateral sclerosis (ALS) in phase 3 study MCI186-19 (Study 19). One of the Study 19 inclusion criteria was forced vital capacity (FVC) ≥80% of predicted (≥80%p). Therefore, the study provided no information on edaravone efficacy in subjects with FVC <80%p. In Study 19, 24-week, double-blind treatment was followed by open-label treatment where all subjects received edaravone. At 24 weeks, some subjects had FVC <80%p (FVC24 <80%p). This allowed for post-hoc assessment of the effects of edaravone in subgroups of subjects with FVC24 ≥80%p vs <80%p. Objective To address the question of the efficacy of edaravone in ALS patients with FVC <80%p. Methods Post-hoc analysis of Study 19 comparing edaravone efficacy at week 48 in subjects with FVC24 ≥80%p vs <80%p. Results With edaravone treatment, subjects in both the FVC24 ≥80%p and the FVC24 <80%p subgroups experienced a reduction in ALS Functional Rating Scale-Revised (ALSFRS-R) score loss vs placebo subjects through week 48. For the FVC24 ≥80%p subgroup, the changes in ALSFRS-R scores from baseline to week 48 were −7.63 for edaravone-edaravone vs −9.69 for placebo-edaravone, a difference of 2.05 (P = .034; 95% CI: 0.16, 3.94). For the FVC24 <80%p subgroup, the changes in ALSFRS-R scores from baseline to week 48 were −10.26 for edaravone-edaravone vs −15.20 for placebo-edaravone, a difference of 4.94 (P = .0038; 95% CI: 1.64, 8.25). Linear regression analysis indicated that, in the FVC24 <80%p subgroup, there was a notable change in the slope of the ALSFRS-R score-vs-time graph after the start of edaravone treatment. Conclusion ALS subjects in the Study 19 placebo arm had a slowing in disease progression, even when edaravone was added with an FVC of <80%p prior to starting edaravone. A randomized, placebo-controlled study is needed to validate these post-hoc findings.