Statistical design of Path to Prevention platform trial in early neuronal alpha-synuclein disease

The Path to Prevention (P2P) platform trial is a first-in-class, phase 2A, randomized, double-blind, placebo-controlled learning study. Its design expedites the assessment of investigational therapies for the treatment of early-stage Neuronal a-Synuclein Disease (NSD), a biological disease continuum that encompasses and often precedes clinically manifest Parkinson's Disease. P2P is embedded within the Parkinson's Precision Medicine Initiative (PPMI), a robust observational longitudinal cohort study on Parkinson's disease progression and a trial-facilitating enrollment framework. A regimen, comprised of an active investigational therapy and a matched placebo control, evaluates a therapeutic candidate within the P2P perpetual platform framework. Multiple regimens can be enrolled, enabling efficient testing of multiple therapeutics. A two-stage randomization scheme determines regimen sample size with 125 participants assigned to a regimen's active therapy and a variable number to placebo control. The treated follow up period is a minimum of 96 weeks and a maximum of 144 weeks, with the last randomized participant followed for 96 weeks. As a pioneering interventional trial in early-stage NSD, P2P endpoint selection was critically informed by modeling and analysis of available PPMI data. In P2P, regimens will be evaluated for efficacy using two primary endpoints: 1) a biomarker outcome of Dopamine Transporter Single-Photon Emission Computed Tomography (DAT SPECT) imaging measured by the bilateral putamen Specific Binding Ratio (SBR) and 2) a clinical outcome measured by the Movement Disorder Society-Sponsored Revision of the Unified Parkinson's Disease Rating Scale (MDS-UPDRS) Part III score. Each endpoint will be assessed separately using a longitudinal Bayesian mixed effects model, which compares the active treatment group to the shared placebo control group. The model uses regimen-specific controls at full weight and dynamically discounts controls from other regimens. Additionally, placebo control estimates are supplemented by non-randomized control data from the PPMI cohort. Clinical trial simulations, informed by longitudinal data from 275 P2P-eligible PPMI study participants (Age 60-81, 35% Female, 65% Male), were conducted to evaluate the design. Each regimen is well-powered to detect a 40% slowing in progression for both the clinical and biomarker primary endpoints when comparing active treatment to randomized control data. Supplementing the control estimation with non-randomized, P2P-eligible registry participant data increases power in the explored scenarios up to 4% for each endpoint. The embedded P2P platform trial is a powerful learning study that leverages statistical efficiencies from the platform trial design and established PPMI infrastructure and is poised to advance treatment development in early-stage NSD.