Successful Development of the First Community Guidelines for Variant Function Classification to Standardize Clinical Trial Enrollment
Abstract number :
2.054
Submission category :
12. Genetics / 12A. Human Studies
Year :
2025
Submission ID :
735
Source :
www.aesnet.org
Presentation date :
12/7/2025 12:00:00 AM
Published date :
Authors :
Presenting Author: Tobias Brünger, PhD – Department of Neurology, McGovern Medical School, The University of Texas Health Science Center at Houston, Houston, TX, USA
SCN2A-FCG Consortium, Working Group – SCN2A Functional Classification Initiative
Dennis Lal, PhD – Department of Neurology, McGovern Medical School, The University of Texas Health Science Center at Houston, Houston, TX, USA
Rationale: Variants in epilepsy-associated genes often lead to a broad spectrum of neurodevelopmental phenotypes, ranging from early-onset epileptic encephalopathies to intellectual disability and autism. Emerging precision therapies target gene specific disease mechanisms such as gain- or loss-of-function (GoF/LoF) at the molecular level, making accurate functional classification of variants essential for safe and effective clinical trial design. Yet no standardized, scalable framework currently exists to assign functional labels in a systematic, evidence-based manner. SCN2A, one of the most clinically and functionally diverse epilepsy genes, illustrates these challenges and served as the pilot gene for developing a standardized, community based and data-driven classification framework.
Methods: We developed the SCN2A Functional Classification Guidelines (SCN2A-FCG), a structured, Bayesian point-based system for assigning functional labels—GoF, LoF, or uncertain function (VUF). Evidence codes span four domains: clinical phenotypes, functional assays, in silico predictions, and paralogous variant data. Likelihood ratios derived from curated datasets informed initial quantitative weights. Clinical evidence weights were calculated using data from Simons Searchlight (n=214) and the Danish Epilepsy Center (n=981). Molecular and computational weights were derived from 100 functionally tested SCN2A variants and 478 variants across other sodium channels. All weights and thresholds were refined through expert panel review to ensure consistency, interpretability, and clinical relevance. Evidence grouping and caps were applied to avoid overweighting correlated data. Final classification thresholds were set at >6 for likely GoF, < –6 for likely LoF, and –5 to +5 for VUF.
Genetics