The role of rare copy number variants in early-onset depression

Depression is a highly heterogeneous condition. Depression with an onset in childhood and early adolescence has a worse clinical course, is more heritable, and shows a lower genetic correlation with other depression subtypes, than does later-onset depression. It is also more strongly associated with neurodevelopmental (ND) comorbidities and genetic liability to attention-deficit hyperactivity disorder. Thus, we hypothesised that early-onset depression represents a distinctive 'neurodevelopmental' depression subtype associated with an increased burden of rare copy number variants (CNVs) that are enriched in ND conditions. We tested this hypothesis using four population cohorts across the UK, Norway, and Sweden.

Participants were ascertained from four population cohorts across the UK, Norway, and Sweden. Early-onset depression was defined as a score >11 on the self-reported Short Mood and Feelings Questionnaire between ages 10 and 14 years (cases n = 5994 vs. controls n = 26,388) and, for secondary analyses, using ICD-10 criteria for major depressive disorder (MDD) with onset  $\le$ 14 years (cases n = 856 vs. controls n = 96,769). Carriers of large, rare (>500 kb, <1% frequency) CNVs and known ND CNVs were identified. Primary analyses tested associations between early-onset depression and (i) large, rare CNVs, and (ii) ND CNVs. Secondary analyses investigated parent-reported measures of early-onset depression.

Meta-analysis did not identify any robust associations between early-onset depression (SMFQ-defined) and large, rare CNVs (OR = 0.92 [95% CI = 0.84-1.02], p = 0.12) or ND CNVs (OR = 1.06 [0.85-1.31], p = 0.60). No robust associations were observed between early-onset depression, defined using ICD-10 MDD criteria, and large rare CNVs (OR = 1.08 [0.86-1.36], p = 0.49) or ND CNVs (OR = 0.69 [0.34-1.39], p = 0.30).

Our findings did not support the hypothesis that individuals with early-onset depression show enrichment for large, rare or known ND CNVs.