Genome-wide association study identifies age-related genetic effects on childhood refractive error and myopia risk

Objective To characterize age-related genetic effects on childhood refractive error and to examine whether a polygenic risk score (PRS) has age-related patterns in myopia risk stratification. Methods We performed a meta-analysis of genome-wide association studies (meta-GWAS) using three pediatric cohorts: the Hong Kong Children Eye Study (HKCES‑1), the Low‑Concentration Atropine for Myopia Progression (LAMP‑1) study, and the Guangzhou Twin Eye Study (GTES), including 1,903 children. Replication was conducted in independent HKCES‑2 ( N = 2,800). Age‑stratified analyses were performed for significant loci. A PRS was constructed and used to estimate myopia risk ratios across age groups 5–6, 7–8, and 9–10 years. Results Two reported loci, MIR4275 rs2131464 (β = − 0.63, P = 6.75 × 10⁻⁹) and TENM3 rs17074027 (β = − 0.38, P = 9.52 × 10⁻⁹), were confirmed, and a novel locus, TMEM132D rs4759965 (β = 0.34, P = 5.30 × 10⁻⁸), was identified. TMEM132D rs4759965 and FAM135B rs4609227 were associated with refraction only at ages 7–8 years, whereas MIR4275 rs2131464 and TENM3 rs17074027 were associated at ages 9–10 years. In the replication cohort, an 8-SNP PRS showed an age-related pattern of risk stratification: among children aged 7–8 years, those in the highest PRS group had a 2.37-fold higher risk of myopia than those in the lowest PRS group (P FDR = 0.031). Conclusions Genetic influences on childhood myopia are age-dependent, with distinct loci associated with refractive error at different ages. The 8-SNP PRS stratified myopia risk in an age-related manner, with the strongest stratification at ages 7–8 years, suggesting a potential developmental window for genetic risk. Further validation in additional independent cohorts is warranted.