Introduction: Prior work has shown infertile men are more likely to suffer poor health outcomes, including diabetes, heart disease, and cancer. A genetic link between health and fertility has been hypothesized but a lack of genetic data in most studies prevented examination. Finding a genetic link would permit screening to identify susceptible men. Azoospermia factor region C microdeletions (AZFc-MD) are common genetic variants in non-obstructive azoospermia. This study aimed to identify men with AZFc-MD in the UK Biobank and evaluate associations between AZFc-MD and poor health outcomes.
Methods: Unrelated white men in the UK Biobank with Y chromosome single nucleotide polymorphism (SNP) data were included. AZFc intensity ratios were calculated by dividing the mean log 2 SNP intensity ratios in the AZFc region by the mean Y chromosome SNP intensity ratios. Men with AZFc intensity ratios = 3 standard deviations below the population median were labeled likely AZFc-MD. Association tests adjusted for age, genotype chip, and the first 10 genotype principal components were ran between AZFc-MD status and 5 predetermined infertility linked phenotypes and 15 health outcomes.
Results: 181,978 men were included of whom 122 (~1 per 1,500) had likely AZFc-MD. AZFc-MD was associated with markers of fertility, including decreased testosterone levels, fewer number of children fathered, and diagnosis of testicular dysfunction; however, only a negative association with vasectomies was significant (OR: 0.13, 95% CI: [0.01 - 0.57], p = 0.04). For the health outcomes, AZFc-MD was associated with a younger age at death (Coefficient: -0.41, 95% CI: [-0.75 - -0.07], p = 0.017), Hodgkin’s lymphoma (OR: 66.9, 95% CI: [3.69 - 332.14], p < 0.0001), follicular lymphoma (OR: 10.5, 95% CI: [0.59 - 47.82], p = 0.02), and unspecified renal failure (OR: 4.6, 95% CI: [1.61 - 10.20], p = 0.001). There were trends towards increased likelihood for obesity, myocardial infarction, heart failure, and younger age at cancer diagnosis; however, these were not significant.
Conclusions: In this Biobank-scale study, likely AZFc-MD had a similar incidence to previously reported data and were associated with infertility phenotypes, increasing confidence in our methods. Likely AZFc-MD was linked to several health outcomes, including age of death, lymphoma, and renal failure, complementing prior work in this area. The power of this study to detect significant associations may be limited by the number of AZFc-MD cases. Future work will refine methods for calling AZFc-MD and expand the number of phenotypes investigated.