Is there a shared genetic basis or causal relationship between polycystic ovary syndrome (PCOS) and a range of psychiatric disorders?
Genome-wide genetic correlation analysis and bidirectional Mendelian randomisation (MR) analysis suggest no shared genetic basis or causal relationship of PCOS with psychiatric disorders including depression, anxiety, schizophrenia and bipolar disorder.
The comorbidity of PCOS with a range of psychiatric disorders has been recognised by epidemiological investigations yet a causal relationship remains unclear. Understanding of how genetic variations contribute to the susceptibility to PCOS and psychiatry disorders could provide meaningful insights into disease mechanisms.
We incorporated summary statistics from the hitherto largest genome-wide association studies (GWAS) conducted in subjects with PCOS (Ncase = 9322) or four common psychiatric disorders (depression, anxiety, schizophrenia and bipolar disorder) (Ncase ranges between 20 352 and 246 363), all of European ancestry.
We quantified pairwise genetic correlation to understand the shared genetic predisposition using genome-wide genetic variants. We performed a two-sample bidirectional Mendelian randomisation analysis to make causal inferences, using GWAS-identified 102 depression-associated genetic instruments, 6 anxiety-associated instruments, 179 schizophrenia-associated instruments, 30 bipolar disorder-associated instruments and 14 PCOS-associated instruments. We performed several important sensitivity analyses examining sex hormones and utilising different MR approaches.
We did not find significant genetic correlations (rg) for PCOS with psychiatric disorders (depression (rg = 0.09, P = 0.06), anxiety (rg = 0.15, P = 0.06), schizophrenia (rg = 0.02, P = 0.59), bipolar disorder (rg = 0.08, P = 0.19)). Genetic predisposition to PCOS was associated with depression in some of our MR approaches, without any evidence of pleiotropy (PMR-Egger intercept = 0.60). However, this weak PCOS-depression causal association attenuated to null after adjusting for BMI (1.00 (0.99-1.02), P = 0.28). On the contrary, we did not observe any statistically significant association between genetically instrumented PCOS with other psychiatric disorders (anxiety 1.01 (0.93-1.08), P = 0.89; schizophrenia 1.03 (0.97-1.10), P = 0.37; bipolar disorder 0.96 (0.90-1.03), P = 0.26). Bidirectional MR did not reveal an effect by which mental health conditions influenced PCOS risk.
Despite our study being the largest in sample size of its kind, the overall negligible causal relationship between PCOS and psychiatric outcomes may reflect a true null association but may also be due to a true effect too modest to be detected. We were not able to investigate PCOS subtypes and used an overall heterogenous PCOS sample due to limited availability of data.
Our comprehensive analysis does not identify a shared genetic basis of PCOS with psychiatric diseases. Although genetically instrumented PCOS appears to correlate with depression, such an effect is likely mediated by BMI, suggesting a role of non-genetic exposures underlying the observed comorbidity.
The work was supported by the Swedish Medical Research Council 2018-02435 (to E.S.V.), Novo Nordisk Foundation NNF19OC0056647 (to E.S.V.), the Adlerbert Research Foundation (to E.S.V.), the SRP in Diabetes at Karolinska Institutet (to E.S.V.) and the Swedish Research Council VR 2018-02247 (to X.J.). The funders had no influence on the data collection, analyses or conclusions of the study. No conflict of interests to declare.

© The Author(s) 2021. Published by Oxford University Press on behalf of European Society of Human Reproduction and Embryology. All rights reserved. For permissions, please email: