Abstract
Altered cortisol regulation is implicated in Attention-Deficit/Hyperactivity Disorder (ADHD), but causality remains debated. While meta-analyses suggest that lower basal cortisol levels, especially in the morning, correlate with ADHD, study heterogeneity prompts further inquiry. Leveraging post-genome-wide association approaches, we examined morning cortisol levels (n = 25,314) and ADHD (n = 225,543). We employed seven Mendelian Randomization methods for causal inference, assessed global and local genetic correlations, and applied the conjFDR method. Additionally, we conducted polygenic score (PGS) analysis in an independent ADHD sample (n = 1660). We found evidence of regional pleiotropy rather than causality, indicating shared genomic architecture. Morning cortisol levels and ADHD showed local genetic correlations in two regions-on chromosomes 5 and 22-one with variants correlating positively and the other negatively. These regions harbor genes associated with psychiatric disorders, including two previously linked to ADHD (RASGRF2 and TRIOBP). ConjFDR revealed one independent hit (rs28406364) jointly associated with ADHD and cortisol levels. Our PGS analysis linked cortisol PGS to externalizing behavior comorbidity only in the European ancestry group. Adjusting for psychiatric disorders, we found an association with ADHD, with cases exhibiting lower cortisol PGS. Our findings suggest ADHD and cortisol levels reflect a complex interplay involving arousal regulation rather than a simple stress-response mechanism. The data align with broader physiological models, supporting a U-shaped relationship between cortisol levels and ADHD traits. This conceptual shift has significant implications for understanding ADHD as systemic dysregulation rather than solely a cognitive or attentional disorder.