Cerebellar volume as imaging outcome in progressive multiple sclerosis

BACKGROUND AND PURPOSE To assess whether cerebellar volumes changes could represent a sensitive outcome measure in primary-progressive MS. MATERIAL AND METHODS Changes in cerebellar volumes over one-year follow-up, estimated in 26 primary-progressive MS patients and 20 controls with Freesurfer longitudinal pipeline, were assessed using Wilcoxon test and tested for their correlation with disability worsening by a logistic regression. Clinical worsening was defined as EDSS score increase or change of >20% for 25-foot walk test or 9-hole peg test scores at follow-up. Sample sizes for given treatment effects and power were calculated. The findings were validated in an independent cohort of 20 primary-progressive MS patients. RESULTS Significant changes were detected in brain T1 lesion volume (p<0.01), cerebellar T2 and T1 lesion volume (p<0.01 and p<0.05), cerebellar volume, cerebellar cortex volume, and cerebellar WM volume (p<0.001). Only cerebellar volume and cerebellar cortex volume percentage change were significantly reduced in clinically progressed patients when compared to patients who did not progress (p<0.01; respectively AUC of 0.91 and 0.96). Cerebellar volume percentage changes were consistent in the exploration and validation cohorts (cerebellar volume -1.90±1.11% vs -1.47±2.30%; cerebellar cortex volume -1.68±1.41% vs -1.56±2.23%). Based on our results the numbers of patients required to detect a 30% effect are 81 per arm for cerebellar volume and 162 per arm for cerebellar cortex volume (90% power, type 1 error alpha = 0.05). CONCLUSIONS Our results suggest a role for cerebellar cortex volume and cerebellar volume as potential short-term imaging metrics to monitor treatment effect in primary-progressive MS clinical trials.