Statistical design of Path to Prevention platform trial in early neuronal alpha-synuclein disease
In the authors' words
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.
In plain words AI summary
It uses a platform design where multiple treatments can be tested at once, with 125 participants getting each active treatment and a varying number getting placebo. The trial measures success using two main endpoints: a brain imaging biomarker and a clinical movement scale score. Computer simulations showed that each treatment group has enough participants to detect a 40% slowing of disease progression. Adding data from non-randomized participants increased the statistical power by up to 4% for each endpoint.
Written by Kimi K2.5 using only the abstract above. Jev checked that every sentence is supported by the abstract sentences it cites, and removed any that were not. Point at a sentence to highlight its source. 1 sentence was removed for not being supported by the abstract.
Appeared: Saturday, November 7. medRxiv. Preprint, not yet peer-reviewed.