IntroductionSubjects with amyotrophic lateral sclerosis (ALS) treated with Radicava® (edaravone) IV (intravenous; Mitsubishi Tanabe Pharma America [MTPA], hereafter “MTPA IV edaravone”) in Study MCI186–19 had a significantly slower physical functional decline vs placebo-treated subjects as measured by the revised ALS Functional Rating Scale (ALSFRS-R) and analyzed by the linear mixed model for repeated measures (MMRM). This Study 19 post hoc analysis of MTPA IV edaravone–treated and placebo-treated subjects evaluated linear and nonlinear latent class mixed models defining trajectories based on identifying the model with the lowest Bayesian information criterion. The best model differentiated 4 nonlinear trajectories in ALS subjects. ALSFRS-R total score in MTPA IV edaravone–treated and placebo-treated subjects was evaluated for these 4 nonlinear latent class trajectory groups. MethodsDisease trajectories of MCI186–19 MTPA IV edaravone–treated or placebo-treated ALS subjects who completed the double-blind period were investigated using latent class analysis (LCA) statistical models to identify potential unique nonlinear ALSFRS-R disease trajectories. ResultsALSFRS-R trajectories revealed 4 unique nonlinear trajectory latent classes per treatment group in MTPA IV edaravone–treated and placebo-treated ALS subjects completing the MCI186–19 double-blind period. Latent classes 2–4 had statistically significant slowing of ALSFRS-R total score decline in the predicted nonlinear trajectories of MTPA IV edaravone–treated vs placebo-treated ALS subjects. ConclusionsThis post hoc analysis suggests MTPA IV edaravone treatment results in slower ALSFRS-R decline vs placebo in most predicted nonlinear trajectories. LCA is a novel approach that may benefit future trial analyses.