Modeling Disease Progression and Placebo Response in Huntington Disease: Insights From Enroll-HD and GENERATION HD1 Cohorts.

Background and objectives: Huntington disease is a rare neurodegenerative disorder with no disease-modifying therapies. This study aimed to quantify longitudinal changes in Unified Huntington's Disease Rating Scale (UHDRS) scores and evaluate their susceptibility to placebo response, enhancing our understanding of disease progression and ability to optimize future trials.

Methods: We used data from the Enroll-HD natural history study (NCT01574053) and the GENERATION HD1 phase 3 clinical trial (NCT03761849) to model disease progression and placebo response for UHDRS scores, which are commonly used to evaluate disease progression in clinical trials.

Results: We included 8,071 Enroll-HD participants (mean baseline age: 51.4 years, 51.5% female) to develop a natural history progression model using baseline characteristics as predictive covariates. This model was then used to predict natural history progression of 260 participants from the GENERATION HD1 placebo arm (mean baseline age: 48.7 years, 43.5% female).The progression measured by Total Functional Capacity (TFC) in the GENERATION HD1 placebo arm aligned with predicted natural history (within 95% CI), indicating no significant placebo response. However, significant improvements (outside the 95% CI of the model) were observed after baseline for Total Motor Score (TMS), Symbol Digit Modalities Test (SDMT) score, and Stroop Word Reading (SWR) score. The improvement in TMS persisted until the end of the dosing period (week 69), converging to natural history predictions at subsequent follow-up visits (weeks 85 and 101), suggesting a placebo effect. By contrast, cognitive scores (SDMT and SWR) showed sustained significant improvement (outside the 95% CI) up to the final follow-up visit at week 101, likely due to practice effects from the more frequent testing schedule in GENERATION HD1 compared with the annual assessments in Enroll-HD. Consequently, the composite UHDRS (cUHDRS) score, a linear combination of TFC, TMS, SDMT, and SWR, is influenced by both placebo and practice effects.

Discussion: Our results suggest that clinical scores in Huntington disease trials are susceptible to long-term placebo responses. These effects should be considered in future trial designs, especially when comparing trial data with natural history studies. Although based on the largest Huntington's trial, our results rely on limited placebo data and may not generalize to other trials with different populations, treatments, and designs.