Abstract
INTRODUCTION: Cognitive impairment is common but often overlooked due to motor symptoms in progressive supranuclear palsy-Richardson syndrome (PSP-RS). This study investigates whether cognitive deficits predict disease progression in PSP-RS. METHODS: A total of 146 PSP-RS from the Tilavonemab trial were evaluated at baseline and over 52 weeks using the PSP-Rating Scale (PSPRS), the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS), and the Unified Parkinson's Disease Rating Scale Part-II (UPDRS-II). Multiple linear regression analyses were performed between the RBANS, UPDRS-II, and the PSPRS change. Clinical scores, gray matter volumes, and neurofilament-light chain (NfL) were compared using analyses of covariance (ANCOVAs) and linear mixed models between language score-groups. RESULTS: Lower RBANS-language at baseline predicted greater PSPRS worsening over time. The low language-score group showed poorer cognitive performance, elevated NfL, and reduced gray matter volume in language-related areas. DISCUSSION: Speech/language deficits predict worse prognosis in PSP-RS, emphasizing the value of including language scores in clinical trials. HIGHLIGHTS: Speech and language deficits predict a worse prognosis in progressive supranuclear palsy-Richardson syndrome (PSP-RS). Lower language scores are associated with worse cognitive performance over time. Lower language scores related to higher neurofilament-light chain (NfL) at baseline. The low language-score group presented greater atrophy in language-related brain areas. Stratifying PSP-RS cases using language scores may improve clinical trials.