Beyond Reward: Genetic Enrichment in Pruning Pathways Redefines Substance Use Disorder as a Disorder of Synaptic Maturation

Emerging evidence indicates that the genetics of substance-use disorders (SUDs) extend beyond canonical dopaminergic and metabolic pathways. To evaluate neurodevelopmental contributors, we re-examined summary statistics from a multivariate genome-wide association study of 1,025,550 individuals of European ancestry. Gene-based analysis with MAGMA, partitioned heritability with stratified LD-score regression, and transcriptome-wide association with S-PrediXcan were applied to bespoke gene sets: glutamatergic signalling, three synaptic-pruning variants (core, expanded, and pruning-exclusive sets), and negative controls. MAGMA identified a significant competitive enrichment for the expanded pruning set (Bonferroni-corrected p = 0.036) that remained after removing genes overlapping glutamatergic pathways (false-discovery-rate p = 0.033). Stratified LD-score regression corroborated this signal, showing a 1.06-fold enrichment of heritability (Bonferroni-corrected p ≈ 0.002). S-PrediXcan produced a concordant, albeit weaker, pattern; the top association was RHOA, a cytoskeletal regulator of pruning (p ≈ 3 × 10⁻⁹). Directionality of effects implied that risk alleles favour excessive pruning. We propose that such variants exaggerate adolescent synaptic elimination within reward circuitry, leaving networks hyper-responsive to dopaminergic and alcohol-related stimuli. This neurodevelopmental perspective aligns SUD onset during adolescence with its frequent comorbidity patterns, positioning synaptic pruning as a new, actionable target for prevention and intervention.