Medical consequences of pathogenic CNVs in adults: Analysis of the UK Biobank

Background Genomic copy number variants (CNVs) increase risk for early-onset neurodevelopmental disorders but their impact on medical outcomes in later life is poorly understood. The UK Biobank, with half a million well-phenotyped adults, presents an opportunity to study the medical consequences of CNV in middle and old age. Methods We called 54 CNVs associated with clinical phenotypes or genomic disorders, including their reciprocal deletions or duplications, in all Biobank participants. We used logistic regression analysis to test CNVs for associations with 58 common medical phenotypes. Findings CNV carriers had an increased risk of developing 37 of the 58 phenotypes at nominal levels of statistical significance, with 19 associations surviving Bonferroni correction (p<8·6×10−4). Tests of each of the 54 CNVs for association with each of the 58 phenotypes identified 18 associations that survived Bonferroni correction (p<1·6×10−5) and a further 57 that were associated at a false discovery rate (FDR) threshold of 0·1. Thirteen CNVs had three or more significant associations at FDR=0·1, with the largest number of phenotypes (N=15) found for deletions at 16p11·2. The most common CNVs (frequency 0·5-0·7%) have no or minimal impact on medical outcomes in adults. Interpretation Some of the 54 tested CNVs have profound effects on physical health, even in people who have largely escaped early neurodevelopmental outcomes. Our work provides clinicians with a morbidity map of potential outcomes among carriers of these CNVs. Funding MRC UK, Wellcome Trust UK