Characteristics of Progressive Supranuclear Palsy Progression – a Longitudinal Cohort Study with Neuropathology (124)

Objective: To determine whether longitudinal changes in the PSP Rating Scale (PSPRS) reported in recent clinical trials match those seen in a regional outpatient healthcare setting. Background: The PSPRS is a multi-domain clinical rating scale used to monitor disease severity and progression in PSP. The annualised rate of change in clinical trials is typically 10–12 points/year. Understanding how PSPRS changes over time, and for different groups of patients, will improve both trial design for disease-modifying therapies and clarification of natural disease course. Design/Methods: Longitudinal PSPRS data collected over 1 to 12 years was obtained from a natural history cohort based in the East of England National Health Service (NHS) specialist PSP clinic in Cambridge, UK. We focus on patients with a clinical diagnosis of PSP during their lifetime, including Richardson syndrome and ‘variant’ PSP syndromes. Neuropathological examination was undertaken at the Cambridge Brain Bank for a subset of patients. Results: 219 patients (male=58%, mean age 72.14±7.02 years) with a clinical diagnosis of probable or possible PSP were identified. Average PSPRS score at first assessment was 40 (95% CI 38–42). 144 patients underwent more than one assessment with a median follow-up 18 months (95% CI 15–21 months). A linear mixed-effects model estimated progression of 6.3 points/year (95% CI 5.8–6.9). Among the pathologically-proven cases (n=40, male=52%, mean age 70.38±7.23 years), average score at first assessment was 41 (95% CI 37–45) with a median follow-up 22 months (95% CI 15–29 months) and estimated progression 6.4 points/year (95% CI 5.0–7.7). Richardson syndrome cases with pathology confirmation (n=28) progressed at 7.9 points/year (95% CI 6.3–9.3). Conclusions: Compared to clinical trial data, our PSP cohort progressed at a slower rate, including cases with neuropathological confirmation. This has implications for participant selection procedures in trials, the need for placebo control over historical control data, and generalisation from trial cohorts to healthcare settings. Disclosure: Dr. Street has nothing to disclose. Dr. Rowe has nothing to disclose.