Admixture mapping of uterine fibroid size and number in African American women

Abstract

Objective: To evaluate the relationship between genetic ancestry and uterine fibroid characteristics.

Design: Cross-sectional study.

Setting: Not applicable.

Patient(s): A total of 609 African American participants with image- or surgery-confirmed fibroids in a biorepository at Vanderbilt University electronic health record biorepository and the Coronary Artery Risk Development in Young Adults studies were included.

Intervention(s): None.

Main outcome measure(s): Outcome measures include fibroid number (single vs. multiple), volume of largest fibroid, and largest fibroid dimension of all fibroid measurements.

Result(s): Global ancestry meta-analyses revealed a significant inverse association between percentage of European ancestry and risk of multiple fibroids (odds ratio: 0.78; 95% confidence interval 0.66, 0.93; P=6.05 × 10^-3). Local ancestry meta-analyses revealed five suggestive (P<4.80 × 10^-3) admixture mapping peaks in 2q14.3-2q21.1, 3p14.2-3p14.1, 7q32.2-7q33, 10q21.1, 14q24.2-14q24.3, for number of fibroids and one suggestive admixture mapping peak (P<1.97 × 10^-3) in 10q24.1-10q24.32 for volume of largest fibroid. Single variant association meta-analyses of the strongest associated region from admixture mapping of fibroid number (10q21.1) revealed a strong association at single nucleotide polymorphism variant rs12219990 (odds ratio: 0.41; 95% confidence interval 0.28, 0.60; P=3.82 × 10^-6) that was significant after correction for multiple testing.

Conclusion(s): Increasing African ancestry is associated with multiple fibroids but not with fibroid size. Local ancestry analyses identified several novel genomic regions not previously associated with fibroid number and increasing volume. Future studies are needed to explore the genetic impact that ancestry plays into the development of fibroid characteristics.

Keywords: Fibroids; admixture mapping; global ancestry; leiomyomata; local ancestry.

Figure 1. Admixture mapping analysis of chromosome 10 with overlapping single SNP association analysis results

The X-axis indicates the genomic position along chromosome 10 in Mb. The top of the Y-axis indicates -log₁₀(p value) from the meta-analysis between BioVU and CARDIA AA from logistic regression of admixture mapping values that were generated from LAMP-ANC (solid black line – not conditioned for rs12219990; dashed green line – conditioned on rs12219990) with fibroid number (single vs multiple) as the outcome. The admixture mapping analysis was adjusted for age, BMI, 5 PCs. The solid green line represents the significance threshold. The solid blue line represents the suggestive threshold. The bottom portion of the Y-axis indicates the log₁₀(p value) for the single SNP association analyses (blue circles indicates genotyped SNPs; red circles indicates imputed SNPs) with fibroid number (single vs multiple) as the outcome. The imputed region was found by taking 1 –log down from the most significant mapping peak above the suggest threshold. The imputed region encompassed 56,258,202 to 59,854,651 bp. The single SNP association analysis was adjusted for age, BMI, and 5 PCs. The SNP, rs12219990, was imputed (BioVU info score – 0.921; CARDIA info score – 0.922). The dotted red line represents the significance threshold.