Exome chip evaluation of genetic variants for association with uterine fibroids

Abstract

Uterine fibroid(UFs) affect up to 77% of women by menopause with large racial disparities and account for $9.4 billion in yearly healthcare costs. Although UFs are heritable, genetic risk is poorly understood. The first genome-wide association study(GWAS) of UFs was performed in 2011 in a Japanese population, however, to-date few large-scale genetic studies have been performed in US populations.(1) The objective of this study is to conduct a whole exome association study of UF risk in European Americans(EA) and African Americans(AA). This is case-control genetic association study of UFs defined using pelvic imaging data obtained from an electronic medical record biorepository (BioVU DNA Repository). Logistic regression adjusted for ancestry and age was used in 728 EA and AA DNA samples (EA: 246 cases, 243 controls; AA: 121 cases, 118 controls) to evaluate SNPs for association, stratified by race. Meta-analyses of results with fixed effect models were performed to obtain combined evidence for associations across racial groups. Our strongest association within EAs was within t-complex 11, testis-specific like 1 (TCP11L1, rs1064005, OR 1.79, 95% CI 1.36-2.32, p 3.94x10-5) and within multiple SNPs within the HLA region (smallest p 9.28x10-5). It is of note that the exome chip has enhanced coverage of the HLA region compared to GWAS arrays. Among AAs strong associations were observed within glipican 4 (GC4, OR = 0.46, 95% CI 0.31-0.68, p = 8.73x10-5). Meta-analysis across EA and AA further strengthened the associations observed at TCP11L1 (smallest meta-p 4.09x-7, with effect sizes consistent across races). Prior gene expression studies of TCP11L have shown increased expression in tumor tissues compared to normal tissues across several different cancers and variants within HLA genes have been associated with UF risk in prior studies. These pilot data suggest common variants in exonic regions increase risk for UF in both EA and AA populations. However, further validation of our study findings is needed to confirm our results. We are currently genotyping >4,000 samples for GWAS and exome arrays to further evaluate the relationship between gene variants and UF risk.