VP Clinical Pharmacology Vigil Neuroscience, Inc, United States
Disclosure(s):
Francois Gaudreault, PhD: No relevant disclosure to display
Objectives: Adult-onset leukoencephalopathy with axonal spheroids and pigmented glia (ALSP) is a rare, autosomal dominant, neurodegenerative disorder caused by a loss-of-function mutation in the colony-stimulating factor 1 receptor gene (CSF1R) resulting in disruption of intracellular signaling and primary microglial dysfunction in the central nervous system (CNS). Iluzanebart, a fully human monoclonal antibody agonist of the triggering receptor expressed on myeloid cells 2 (TREM2), is currently being investigated as a potential treatment for ALSP. TREM2 agonism by iluzanebart is hypothesized to compensate for CSF1R loss-of-function, resulting in a potential slowing of clinical decline in patients with ALSP. The objective of this analysis is to demonstrate proof-of-concept for a model-informed drug development (MIDD) approach to quantify treatment effect in a Phase 2 study (P2, NCT05677659) utilizing data from a prospective natural history study (NHS, NCT05020743) as an external control.
Methods: An integrated disease progression model was constructed to simultaneously describe the longitudinal time-course of biomarker and key clinical measurements using data from the NHS: soluble CSF1R (sCSF1R), neurofilament light chain (NfL), ventricle volume (as an indicator of brain volume loss), Cortical Basal Ganglia Functional Scale (CBFS), Montreal Cognitive Assessment (MoCA). Clinical trial simulations were performed to assess study precision using the model to determine the power to detect potentially relevant treatment effects using a model-based approach. Five hundred (500) replicate P2 studies were simulated for each sample size from 5 to 50 in increments of 5. The simulated open-label treatment data were pooled with the observed data from the NHS and a model parameter was estimated for treatment effect. Precision was assessed as the proportion of trial replicates in which the 95% Confidence Interval (CI) of the parameter estimate for treatment effect fell within 60% to 140% of the true effect size.
Results: A joint indirect response disease progression model was fit to all endpoints simultaneously with correlation in random effects across responses. Standard model diagnostics and visual predictive checks demonstrated that the model adequately described each endpoint. Using the MIDD approach, precise estimates of treatment effect could be achieved with the sample size in the P2 study.
Conclusions: The MIDD approach evaluated in this simulation study demonstrated proof-of-concept for use as a method of quantifying iluzanebart treatment effect using NHS as an external control. Such a MIDD approach may afford greater power than traditional biostatistical methods by leveraging all available data, including NHS, to detect treatment effect, quantify the contributions of demographic and clinical factors on disease and response, and assess the correlation of markers of disease to relevant clinical outcomes in neurological rare diseases such as ALSP.
