Cord Blood 3-epi-25-Hydroxyvitamin D₃ and Cognitive Autism-Related Traits at Age 10 Years: An Exploratory Secondary Analysis of the KLOTHO Birth Cohort

Background: The C3 epimer of 25-hydroxyvitamin D₃ [3-epi-25(OH)D₃] is measurable in maternal and neonatal circulation, but its developmental relevance is unknown. Objective: To evaluate, using prospectively collected data from the KLOTHO birth cohort, whether cord blood 3-epi-25-hydroxyvitamin D₃ [3-epi-25(OH)D₃] is associated with autism-related trait domains at age 10 years. Methods: This exploratory secondary analysis used prospectively collected data from the KLOTHO birth cohort. Vitamin D metabolites were quantified by liquid chromatography–tandem mass spectrometry at delivery. Parent-reported autism-related traits were assessed with the Australian Scale for Asperger’s Syndrome. Record-level integrity checks preceded analysis. Spearman correlations, sex-stratified analyses, and supportive ordinary least-squares regression models were used. False-discovery-rate adjustment was applied to 13 unique bivariate tests. Results: A total of 87 offspring, including 41 boys and 46 girls, had available cord 3-epi-25(OH)D₃ and cognitive-domain data at age 10 years. Higher cord 3-epi-25(OH)D₃ was associated with a lower cognitive-domain score (Spearman ρ = −0.412; p < 0.001; Benjamini–Hochberg q = 0.001). The crude linear estimate was also inverse (B = −0.285; SE = 0.087; 95% CI, −0.458 to −0.112; p = 0.002), but it attenuated after adjustment, and the adjusted confidence intervals included the null. The rank-correlation estimate was larger in boys (n = 41; ρ = −0.611; p < 0.001) than in girls (n = 46; ρ = −0.304; p = 0.040), although the between-sex difference was not statistically significant (Fisher z = −1.78; p = 0.075). The stricter record-linkage sensitivity analysis remained significant (n = 84; ρ = −0.350; p = 0.001). Conclusions: Cord 3-epi-25(OH)D₃ showed an inverse rank association with cognitive autism-related traits at age 10 years that remained significant after multiplicity correction and in the stricter record-linkage sensitivity analysis. The crude linear estimate was directionally concordant, but the adjusted estimates included the null. The overall domain pattern was not uniform, and the formal between-sex comparison was not statistically significant. These findings remain exploratory and do not support causal or sex-specific conclusions.