Abstract
BackgroundThe comorbidity between polycystic ovary syndrome (PCOS) and obesity has long been observed in clinical settings, but their shared genetic basis remains unclear.MethodsLeveraging summary statistics of large-scale GWAS(s) conducted in European-ancestry populations on body mass index (adult BMI, Nfemale=434,794; childhood BMI, N=39,620), waist-to-hip ratio (WHR, Nfemale=381,152), WHR adjusted for BMI (WHRadjBMI, Nfemale=379,501), and PCOS (Ncase=10,074, Ncontrol=103,164), we performed a large-scale genome-wide cross-trait analysis to quantify overall and local genetic correlation, to identify shared loci, and to infer causal relationship.ResultsWe found positive genetic correlations between PCOS and adult BMI (rg=0.47, P=2.19×10−16), childhood BMI (rg=0.31, P=6.72×10−5), and WHR (rg=0.32, P=1.34×10−10), all withstanding Bonferroni correction. A suggestive significant genetic correlation was found between PCOS and WHRadjBMI (rg=0.09, P=0.04). Partitioning the whole genome into 1703 nearly independent regions, we observed a significant local genetic correlation for adult BMI and PCOS at chromosome 18: 57630483–59020751. We identified 16 shared loci underlying PCOS and obesity-related traits via cross-trait meta-analysis including 9 loci shared between BMI and PCOS (adult BMI and PCOS: 5 loci; childhood BMI and PCOS: 4 loci), 6 loci shared between WHR and PCOS, and 5 loci shared between WHRadjBMI and PCOS. Mendelian randomization (MR) supported the causal roles of both adult BMI (OR=2.92, 95% CI=2.33–3.67) and childhood BMI (OR=2.76, 95% CI=2.09–3.66) in PCOS, but not WHR (OR=1.19, 95% CI=0.93–1.52) or WHRadjBMI (OR=1.03, 95% CI=0.87–1.22). Genetic predisposition to PCOS did not seem to influence the risk of obesity-related traits.ConclusionsOur cross-trait analysis suggests a shared genetic basis underlying obesity and PCOS and provides novel insights into the biological mechanisms underlying these complex traits. Our work informs public health intervention by confirming the important role of weight management in PCOS prevention.
Highlights
The comorbidity between polycystic ovary syndrome (PCOS) and obesity has long been observed in clinical settings, but their shared genetic basis remains unclear
Despite three Mendelian randomizations (MR) studies [7, 9, 10] have been conducted to explore the role of adult body mass index (BMI) in PCOS, these studies used a small number of index SNPs (< 100 instruments vs. ~ 300 female-specific BMI instruments currently identified by genome-wide associations studies (GWAS) [11]); lacked sensitivity analyses to verify model assumptions; and lacked sex-specific analysis, i.e., using genetic data derived from a sex-mixed population instead of using female-specific data for a gynecological outcome PCOS
We examined the role of BMI (childhood (N=39,620) [14] and adult [11]), waistto-hip ratio [15] (WHR, adult), and waist-to-hip ratio adjusted for BMI [15] (WHRadjBMI, adult) in the development of PCOS (NPCOS=10,074; Ncontrol=103,164) [7], performing analyses to quantify overall and local genetic correlations, to identify shared loci and to infer causal relationships
Summary
The comorbidity between polycystic ovary syndrome (PCOS) and obesity has long been observed in clinical settings, but their shared genetic basis remains unclear. The development of obesity and PCOS involves strong genetic components evidenced by recent discoveries from large-scale genome-wide associations studies (GWAS) These genetic data enable the utilization of a compiled analytical strategy—genome-wide cross-trait analysis—to determine shared and distinct genetic architecture which can provide better understandings and novel insights into disease mechanisms [6]. Such analysis features several analytic aspects: genetic correlation analysis to estimate overall and local genetic correlation, cross-trait meta-analysis to identify shared loci, and Mendelian randomizations (MR) to make causal inferences. Despite three MR studies [7, 9, 10] have been conducted to explore the role of adult BMI in PCOS, these studies used a small number of index SNPs (< 100 instruments vs. ~ 300 female-specific BMI instruments currently identified by GWAS [11]); lacked sensitivity analyses to verify model assumptions; and lacked sex-specific analysis, i.e., using genetic data derived from a sex-mixed population instead of using female-specific data for a gynecological outcome PCOS
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
