Disease trajectories of rare genomic variants contributing to lesional brain disorders

A growing number of disease-causing genes underlying childhood epilepsies have been discovered, many of which can be targeted with precision medicines. Genetic etiologies are implicated in a wide range of childhood epilepsies, including lesional brain disorders. We have previously generated genome sequencing data in children with lesional epilepsies through the Kids First X01 mechanism. We now propose to jointly analyze genomic sequencing data with longitudinal clinical data utilizing the Fast Healthcare Interoperability Resources (FHIR) data model, which is currently implemented within the Gabriella Miller Kids First Pediatric Research Program. While the Kids First data resource offers foundational clinical information, a more extensive array of longitudinal clinical data from Electronic Medical Records (EMR) is vital to deeply comprehend the influence of genetic variation on the clinical trajectory of childhood epilepsies. Our current initiative advocates for the integration of genomic data from two Kids First Cohorts, specifically targeting epilepsy and brain tumors, with longitudinal EMR data. This strategy is designed to analyze the phenotypic trajectories associated with rare genomic variants and assess the combined predictive strength of clinical and genomic data in anticipating disease-relevant outcomes. First, we aim to outline disease trajectories and treatment landscapes linked to rare genomic variants (Aim #1). We will establish longitudinal disease histories across 13,983 patient years, including diagnosis codes, procedures, and medications using the Human Phenotype Ontology (HPO). We will assess phenotypic associations, treatments, and procedures in monthly increments for established and candidate genetic etiologies and outline clinical trajectories and treatment landscapes of individuals carrying rare variants. This analysis will provide the most comprehensive picture of underlying genotype-phenotype association to date. In addition, we will assess the predictive power of joint clinical and genomic data for disease-relevant outcomes (Aim #2). A combination of genomic information and limited EMR data is sufficient to predict severe childhood epilepsies early in the disease course. We will define 25 EMR-based outcomes and develop decision-tree and Random Forest models to identify combinations with high predictive power for each outcome, followed by assessing age- based feature importance. This analysis will provide an overview of how clinical features and rare genetic variants can jointly predict clinical outcomes in a broad cohort of children with brain lesions. At the end of this R03 project, we will have assessed both the phenotypic consequences of rare genomic variation in lesional brain disorders and the predictive power of joint clinical and genomic information. Our project will serve as a pilot, assessing how FHIR-derived, deidentified EMR data deposited in the Kids First Data Resource can be leveraged to gain insight into the trajectories of lesional brain disorders and predict clinical outcomes. Project Number: 1R03HD122148-01 | Fiscal Year: 2026 | NIH Institute/Center: Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) | Principal Investigator: Ingo Helbig (+1 co-PI) | Institution: CHILDREN'S HOSP OF PHILADELPHIA, PHILADELPHIA, PA | Award Amount: $356,000 | Activity Code: R03 | Study Section: Special Emphasis Panel[ZRG1 MGG-R (90)] View on NIH RePORTER: https://reporter.nih.gov/project-details/11354302