Common and rare variant contributions to discontinuation of stimulant treatment in ADHD

Stimulants are the first-line pharmacological treatment for ADHD and generally effective, yet 35-61% of individuals discontinue treatment within a year. We investigated the contribution of common and rare genetic variants to early stimulant discontinuation using data from 18,362 individuals with ADHD (31% female) initiating stimulants in iPSYCH, a Danish population-based case-cohort linked to national registers. Discontinuation was defined as a ≥ 180-day gap between dispensations within one year of initiation. We examined genetic differences by age groups, estimated SNP-heritability (h²_SNP), conducted genome-wide association studies (GWAS), polygenic score (PGS) analyses, and assessed associations with protein truncating variants (PTV). Within one year, 7102 individuals (39%) had discontinued stimulants. Age-stratified analyses (cut-off: age 16) revealed low genetic correlation (r₉ = 0.23, 95% CI: -0.37, 0.83) between children and adolescents/adults. The h²_snp for discontinuation was 0.06 (95% CI: 0.02, 0.11) overall, 0.08 (95% CI: 0.02, 0.14) in children, and 0.14 (95% CI: 0.02, 0.27) in adolescents/adults. No genome-wide significant loci were identified overall or in adolescents/adults; however, one locus (SLC5A12, chromosome 11) reached genome-wide significance in children. Ten of 36 PGSs were associated with discontinuation, with higher psychiatric risk PGSs predicting increased discontinuation, while educational attainment and BMI PGSs showed divergent effects by age. Reduced burden of dopamine-related PTVs was nominally associated with discontinuation, particularly in adolescents/adults. These findings suggest modest contributions of both common and rare variants to stimulant discontinuation in ADHD and point to potential developmental differences in genetic architecture.