Project 4: an observational study in parallel to a gene therapy trial in Canavan

Children'S Hosp Of Philadelphia, Philadelphia PA

Investigators

Linked publications, trials & patents

Abstract

Summary/Abstract for Project 4: Canavan disease as a template for clinical trial readiness in ultrarare leukodystrophies The last decade has revolutionized leukodystrophy (LD) therapeutics, including trials of adeno-associated virus 9 (AAV9) for Canavan Disease (CD; ongoing) and multiple sulfatase deficiency (MSD; anticipated in 2025). We anticipate that innovative approaches with modified adeno-associated virus (AAV) vectors with improved cerebral tropism will further offer opportunities for trials benefitting the ultrarare leukodystrophies. Ultrarare LD trial design is complicated by the inherent clinical heterogeneity and difficulty predicting subtype in young or minimally symptomatic populations. A combination of biomarkers, genotype, and early clinical features collectively influence the clinical determination of LD subtypes. This spectrum and the inability to definitively predict course will make trial design and analysis challenging. To address these gaps, especially for the ultrarare LDs, we develop innovative approaches using CD as an example of the feasibility of predictive modeling with ultrasmall cohorts (~40 individuals) (Aim 1). We will develop two complementary models for risk stratification: a staged model and a machine-learning approach. The ability of these models to accurately assign disease subtype will be compared across two independent test and validation cohorts. Our team includes experienced biostatisticians and data engineers able to define databases and models. The anticipated outcome and metric of success is creation of dynamic disease prediction tools using approaches generalizable to other ultrarare LDs. Additionally complicating trials, these disorders currently have poorly understood systemic phenotypes,23-26 noted by families to strongly impact quality of life. In the context of a clinical trial, understanding systemic phenotypes is crucial to permit attribution of adverse events (AEs). For instance, AAV-based trials require surveillance for pre-existing conditions pre-existing conditions that may increase risk or severity of established AAV toxicities, some of which have associated mortality. Thus, in Aim 2, we will create capture real-world LD comorbidities. This regulatory-ready database will help anticipate events, define safety criteria, and better guide our families. The anticipated outcome is an understanding of pre-existing conditions that may otherwise be misattributed during a clinical trial. Finally, in Aim 3, we will establish real-world protocols to capture AAV toxicities and permit longitudinal safety monitoring. The anticipated outcome is shareable tools for safety monitoring. Using ongoing clinical development in CD as a guide, we anticipate developing a tool kit of platforms for clinical trials in LDs with very small research cohorts. This will prepare the LD community for the advent of clinical trials in ultrarare LDs, taking into consideration their complex phenotypes.