To evaluate feasibility and performance of motor clinical outcome assessments (COAs) in the Alexander disease (AxD) for application in future natural history studies and clinical trials.

Alexander disease (AxD) is a rare, progressive leukodystrophy caused by variants in the GFAP gene. The heterogeneous AxD phenotype, characterized by progressive neurodegeneration, complicates the selection of clinical outcome assessments (COAs) for potential clinical trials. Given recent advancements in disease-modifying therapies for AxD, there is an urgent unmet need for feasible COAs that can capture gross motor function in this population.

Gross motor function in 75 AxD participants was assessed using the Gross Motor Function Measure-88 (GMFM-88), 10-Meter Walk/Run Test (10MWT), 6-Minute Walk Test (6MWT), Pediatric Balance Scale (PBS), and Berg Balance Scale (BBS). Participants were categorized using Yoshida’s classification to identify cerebral, intermediate and bulbospinal forms. Descriptive statistics, Mann-Whitney, Kruskal-Wallis tests, and Spearman correlation analysis were used to evaluate performance and identify correlations between assessments.

The GMFM-88 captured a wide range of motor function in cerebral AxD (N=75, median 64.6%, IQR 59.36), demonstrating significant difference between cerebral, intermediate, and bulbospinal subtypes (p= p=0.0026 and p=0.01, respectively). The 10MWT (N=33) and 6MWT (N=33) were able stratify performance within intermediate and bulbospinal subjects, with the majority of cerebral-type participants were unable to complete testing due to lack of independent ambulation. Balance assessments (PBS N=26, BBS N=27) effectively stratified performance between subtypes. Strong direct and inverse correlations were found between all COAs (ρ =-0.88 to -0.7 and 0.63 to 0.97), demonstrating convergent validity.

Considering disease progression (loss of or inability to independently ambulate) and the variable clinical phenotype, our findings suggest the use of multiple COAs in necessary to comprehensively assess gross motor function across AxD subtypes longitudinally, successfully stratifying the cohort. This study provides a foundation for future AxD natural history studies and clinical trials.