Abstract
Uterine leiomyomata or fibroids are highly heritable, common, and benign tumors of the uterus with poorly understood etiology. Previous GWAS have reported 72 associated genes but included limited numbers of non-European individuals. Here, we identify 11 novel genes associated with fibroids across multi-ancestry and ancestry-stratified GWAS analyses. We replicate a known fibroid GWAS gene in African ancestry individuals and estimate the SNP-based heritability of fibroids in African ancestry populations as 15.9%. Using genetically predicted gene expression and colocalization analyses, we identify 46 novel genes associated with fibroids. These genes are significantly enriched in cancer, cell death and survival, reproductive system disease, and cellular growth and proliferation networks. We also find that increased predicted expression of HEATR3 in uterine tissue is associated with fibroids across ancestry strata. Overall, we report genetic variants associated with fibroids coupled with functional and gene pathway enrichment analyses.
Similar content being viewed by others
Introduction
Uterine fibroids, or leiomyomata, are benign monoclonal elliptical growths of smooth muscle and connective tissue in the myometrium of the uterus. They are the most common benign tumor affecting people with a uterus. The cumulative incidence of fibroids by age 50 is estimated to be almost 70% among White individuals and over 80% in Black individuals1. Common symptoms of fibroids include heavy menstrual bleeding, abnormal menstrual bleeding, and pelvic pain2. People with fibroids report quality-of-life scores similar to or lower than patients with other chronic diseases such as, diabetes, breast cancer, and cardiovascular disease3. Fibroids are the leading cause of hysterectomy, comprising 40% of indications4, and the estimated annual costs for the United States due to fibroids are between $5.9 to $34.4 billion5. Established risk factors for fibroids include early menarche, obesity, premenopausal age, Black race (self-selected), and family history6. A protective effect has been seen with parity1,7,8,9. Additional factors that have been associated with fibroids include, and are not limited to, vitamin D deficiency10,11, hypertension12,13, stressful life events14,15,16, and experiences of racial discrimination17.
Genetic factors also contribute to risk, with estimated twin-based heritability ranging from 26 to 63%18,19. There have been 13 fibroid genome-wide association studies (GWAS) that collectively report 72 genes with genome-wide significant variants20,21,22,23,24,25,26,27,28,29,30,31,32. Investigations of fibroid genetics in diverse ancestry groups have been limited in African ancestry populations20,24.
Here we report a large, multi-ancestry GWAS meta-analysis of uterine fibroids including 74,294 cases (27.7% non-European descent) and 465,810 controls (18.3% non-European descent) from a combination of publicly available summary statistics and newly available data. We identified multiple novel genes associated with fibroids in both multi-ancestry and ancestry-specific analyses. We report functional interpretation of significant loci with genetically predicted gene expression. Genes were also characterized into functional pathways and enrichment sets, and pathways relating to cancer and cell cycle were observed.
Results
Data summary
The data used in these meta-analyses come from a combination of publicly-available summary statistics and newly run GWAS (Fig. 1; Supplementary Data 1)20,24,27,33. We conducted four meta-analyses: 1) The European ancestry meta-analysis comprised of 53,711 cases and 380,441 controls. 2) The East Asian/Central South Asian ancestry meta-analysis data comprised of 14,905 cases and 69,609 controls. 3) The African ancestry meta-analysis data comprised of 5678 cases and 15,760 controls. 4) The multi-ancestry meta-analysis comprised of 74,294 cases and 465,810 controls. Variants described below were mapped by nearest distance to transcription start sites.
AFR African Ancestry, EUR European Ancestry, EAS East Asian Ancestry, CSA Central South Asian Ancestry, Multi Multi Ancestry.
Multi-ancestry and ancestry stratified GWAS meta-analyses SNP-level associations
The results of the multi-ancestry meta-analysis are summarized in Fig. 2 and Supplementary Data 2a, 3a. There were 372 sentinel single nucleotide polymorphisms (SNPs) associated with fibroids in the multi-ancestry meta-analysis (independent significant SNPs with r2 > 0.1) (Supplementary Data 2a). The overall most significant SNPs were rs78378222 (odds ratio [OR] 0.53, 95% confidence interval [CI] 0.50–0.56, p = 2.57 × 10−132) which maps to gene TP53 (3 prime untranslated region [UTR]), and rs58415480 (OR 0.82, 95% CI 0.81–0.84, p = 5.58 × 10−115) which maps to SYNE1 (intron), both well-established associations with fibroids21,24,27,28. After conditional analyses, we identified eight fibroid-associated genes with at least one significantly associated variant that have not been previously published (Fig. 2, Supplementary Data 3a, Supplementary Fig. 6a–f). In our discussion of results, novel associations refer to genes that have never been reported to be associated with fibroids, while “previously unpublished” gene associations include those that have been reported in biobank databases but have not been described in literature. The two novel gene associations include rs74582999 (VIP, intron) and rs761779 (FOXO3, regulatory region). The other six previously unpublished genes include rs149261442 (TEKT1, intron), rs184210518 (SLC16A11, downstream), rs184918809 (RPEL1, upstream), rs555566736 (SLC12A7, intron), rs76382168 (TTC28, intron) and rs78899396 (POLR2A, intron). Of note, rs76382168 has a heterogeneity p-value less than 0.01. We identified an additional 22 significant secondary signals after conditioning on the 372 sentinel SNPs (Supplementary Data 2a). We validated a total of 190 variants in the GWAS Catalog from previous fibroid GWAS.
a Manhattan plot, gold labels = novel, black labels = previously unpublished. b Sentinel variant at novel gene results, asterisk = novel gene association. Logistic regression statistical tests; multiple testing correction p-value threshold used (5 × 10−8).
The genomic inflation factor λGC was 1.09 (Supplementary Fig. 1a), and the Linkage Disequilibrium Score Regression (LDSC) intercept was 1.06 (standard error [SE] 0.01; Supplementary Data 4). The SumHer Linkage Disequilibrium Adjusted Kinships (LDAK) model SNP-based heritability method was used to estimate heritability for each meta-analysis34. The multi-ancestry meta-analyses heritability was estimated using the GBR HapMap3 annotated tags, and heritability was estimated at 0.05 (SE 0.002; Supplementary Data 4).
The results of the European ancestry meta-analysis are summarized in Fig. 3 and Supplementary Data 2b, 3b. In the European ancestry analysis, there were 216 sentinel SNPs associated with fibroids (Supplementary Data 2b). The overall most significant SNPs were the same as the multi-ancestry, rs78378222 (OR 0.53, 95% CI 0.50–0.56, p = 2.57 × 10−132, TP53, 3 prime UTR), and rs58415480 (OR 0.82, 95% CI 0.80–0.83, p = 2.67 × 10−113, SYNE1, intron). The effect size and p-value for rs78378222 are the same as the multi-ancestry because only the European data cohorts contributed to this variant in the multi-ancestry analysis. After conditional analyses, we identified four fibroid-associated genes with at least one significantly associated variant that have not been described in prior studies (Fig. 3, Supplementary Data 3b, Supplementary Fig. 6g–j). The two novel gene associations included rs74582999 (VIP, intron) and rs76798800 (DCST2, intron). The two previously unpublished genes included rs149261442 (TEKT1, intron) and rs35447002 (C11orf87, intergenic). Of note, rs149261442 and rs35447002 have a heterogeneity p-value less than 0.01. We identified an additional 11 significant secondary signals after conditioning on the 216 sentinel SNPs (Supplementary Data 2b). We validated a total of 178 variants in the GWAS Catalog from previous fibroid GWAS. For the European analysis the λGC was 1.17 (Supplementary Fig. 1b), and the LDSC intercept was 1.07 (SE 0.011; Supplementary Data 4)35. Using GBR HapMap3 annotated tags, heritability was estimated to be 0.07 (SE 0.003).
a Manhattan plot, gold labels = novel gene associations, black labels = previously unpublished genes. b Sentinel variant at novel gene results, asterisk = novel gene association. Logistic regression statistical tests; multiple testing correction p-value threshold used (5 × 10−8).
The results of the East Asian/Central South Asian ancestry meta-analysis are summarized in Fig. 4 and Supplementary Data 2c, 3c. In the East Asian/Central South Asian ancestry analysis there were 108 sentinel SNPs (Supplementary Data 2c). The overall most significant SNPs were rs73392700 (OR 0.77, 95% CI 0.74–0.79, p = 1.09 × 10−50) in SIRT3 (intron) and rs141244868 (OR 0.77, 95% CI 0.74–0.79, p = 1.30 × 10−50) in PSMD13 (intron), and both genes have been previously associated with fibroids23,24,27,30. We identified an additional two significant variants after conditioning on the 108 sentinel SNPs (Supplementary Data 2c). No novel gene associations were identified, but we did identify secondary signals in previously associated genes (Supplementary Data 3c). We validated a total of 110 variants in the GWAS Catalog. Within East Asian/Central South Asian ancestry analysis the λGC was 1.07 (Supplementary Fig. 1c), and the LDSC intercept was 1.01 (SE 0.009; Supplementary Data 4). Using SumHer and the EAS HapMap3 tags, heritability was estimated to be 0.115 (SE 0.007).
Significant genes were replications of previously identified associations. Logistic regression statistical tests; multiple testing correction p-value threshold used (5 × 10−8).
The results of the African ancestry meta-analysis are summarized in Fig. 5 and Supplementary Data 2d, 3d. In the African ancestry analysis, there were two sentinel SNPs and a total of six statistically significant SNPs. The most significant variant was rs56897532 (OR 0.78, 95% CI 0.72–0.85, p = 5.39 × 10−9) in COL22A1 (intergenic) and is a novel gene association with fibroids (Fig. 5, Supplementary Data 3d, Supplementary Fig. 6k). Variants rs12280757, rs7106353, rs12285041, rs12285028, and rs61889170 map to WT1 (all intronic) and were the next most significant. WT1 has been previously associated with fibroids but not in this population. These specific variants are different from the previously identified variants in WT121,24,27,28. The association with the variant in COL22A1 was only observed in the African ancestry analysis. The λGC was 1.01 (Supplementary Fig. 1d), and the LDSC intercept was 1.01 (SE 0.006, Supplementary Data 4). Using AFR HapMap3 tags, heritability was estimated to be 0.159 (SE 0.040).
a Manhattan plot, gold labels = novel gene associations. b Sentinel variant at novel gene result, asterisk = novel gene association. Logistic regression statistical tests; multiple testing correction p-value threshold used (5 × 10−8).
A detailed overview of effect sizes and effect allele frequency plots between the ancestry groups are shown in Supplementary Fig. 2. The effect size correlations of all significant independent variants from any of the four meta-analyses are plotted. Between the African ancestry and the European ancestry analyses, the effect size correlation (R) was 0.44. The correlation between the East Asian/Central South Asian ancestry and the European ancestry meta-analyses was 0.92. The effect allele frequency correlation for both the African and European ancestries analyses (R = 0.62) and the East Asian/Central South Asian and European ancestries analyses (R = 0.95) were positive.
Functional annotation and gene set analysis
We used the Functional Mapping and Annotation (FUMA) web tool36 to conduct various post-GWAS analyses (Supplementary Data 5, 6; Supplementary Fig. 3). In the multi-ancestry, European ancestry, and East Asian/Central South Asian ancestry analyses, we found that associated variants were primarily intronic with significant enrichment in this variant category, and also statistically significant depletion in intergenic and non-coding RNA regions, consistent with previous research (Supplementary Fig. 3a–c)31. The associated variants within African ancestry analysis were significantly enriched in intergenic regions and non-coding RNA intronic regions (Supplementary Fig. 3d).
A gene-set enrichment test was also performed by FUMA for 476 input genes. For the multi-ancestry analysis, the top five significantly enriched Gene Ontology biological processes gene sets were replicative senescence (adjusted p = 4.74 × 10−4), DNA damage response signal transduction by p53 class mediator resulting in cell cycle arrest (adjusted p = 4.74 × 10−4), mitotic G1 S transition checkpoint signaling (adjusted p = 6.08 × 10−4), regulation of response to DNA damage stimulus (adjusted p = 6.38 × 10−4), and negative regulation of telomere maintenance (adjusted p = 1.87 × 10−3). There were 52 other significantly enriched Gene Ontology biological processes gene sets, with many related to cell cycle control, DNA damage response, and p53 signaling (Supplementary Data 5a). There were 110 GWAS catalog phenotypes with significant gene set enrichment as well. Uterine fibroids were the most significant (adjusted p = 2.10 × 10−55). The next four most significantly enriched gene-sets were refractive error (adjusted p = 6.23 × 10−17), waist-to-hip ratio adjusted for BMI (adjusted p = 6.46 × 10−14), prostate cancer (adjusted p = 1.41 × 10−13), and Crohn’s disease (adjusted p = 2.07 × 10−10) (Supplementary Data 6a).
The enrichment tests with 378 input genes in the European ancestry group had 10 significant Gene Ontology biological processes gene sets, with the top five being replicative senescence (adjusted p = 4.87 × 10−4), post-transcriptional regulation of gene expression (adjusted p = 1.49 × 10−2), cell cycle G1 S phase transition (adjusted p = 1.49 × 10−2), signal transduction involved in cell cycle checkpoint (adjusted p = 1.58 × 10−2), and mitotic cell cycle (adjusted p = 1.58 × 10−2) (Supplementary Data 5b). There were 81 GWAS catalog reported gene-sets that had statistically significant enrichment. Again, uterine fibroids were the most significant (adjusted p = 2.00 × 10−44) followed by prostate cancer (adjusted p = 2.16 × 10−22), Crohn’s disease (adjusted p = 1.22 × 10−10), and pancreatic cancer (adjusted p = 8.72 × 10−9) (Supplementary Data 6b).
In the East Asian/Central South Asian ancestry analysis, with 124 input genes, there were 12 significant Gene Ontology biological processes enrichment sets. Similar to the previous two, they were DNA damage response signal transduction by p53 class mediator resulting in cell cycle arrest (adjusted p = 6.50 × 10−3), hepoxilin metabolic process (adjusted p = 2.19 × 10−2), replicative senescence (adjusted p = 2.19 × 10−2), DNA damage response signal transduction by p53 class mediator (adjusted p = 2.19 × 10−2), and mitotic G1 S transition checkpoint signaling (adjusted p = 2.31 × 10−2) (Supplementary Data 5c). There were 47 significant GWAS catalog gene sets enriched, where the top five were uterine fibroids (adjusted p = 3.02 × 10−48), waist-to-hip ratio adjusted for BMI (adjusted p = 4.39 × 10−8), waist-hip index (adjusted p = 2.01 × 10−7), platelet count (adjusted p = 2.10 × 10−6), and gynecologic disease (adjusted p = 2.10 × 10−6) (Supplementary Data 6c).
In the multi-ancestry analysis, there was significant enrichment in upregulated differentially expressed gene sets for uterine tissue (Supplementary Fig. 3a). For the European ancestry analyses, there was significant enrichment in upregulated differentially expressed gene sets for uterine tissue and cervical uterine tissue (Supplementary Fig. 3b). East Asian/Central South Asian ancestry analyses had no significant enrichment (Supplementary Fig. 3c). We were underpowered to apply these gene-based tests to the African ancestry meta-analysis results.
Transcriptome-wide association analysis with genetically predicted gene expression
We used S-PrediXcan with 49 tissues from the GTEx project to interpret the associations between genetically predicted gene expression (GPGE) and uterine fibroid risk (Supplementary Fig. 4a). In the multi-ancestry analysis, we identified 1223 significant gene expression-tissue pairs (p-value < 9.4 × 10−8) associated with fibroids across all tissues, of which 240 had a colocalized SNP (H4 posterior probability > 0.8, H3 posterior probability < 0.3) (Supplementary Data 7a). Of these, there were 60 unique genes and 40 of those genes have not been previously associated with fibroids in the GWAS catalog. The most significant colocalized GPGE association was with ACTRT3 expression in the pituitary tissue (OR 5.54, 95% CI 4.19–7.33, p = 3.43 × 10−33). HEATR3 predicted gene expression had a significant and colocalized association where increased predicted expression consistently associated with increased fibroid risk.
In ancestry-specific analyses, we identified multiple significant predicted gene expression associations (Supplementary Data 7b–d, Supplementary Fig. 4b–d), some of which were unique significant gene-tissue colocalized pairs not observed in the multi-ancestry analysis (Table 1). In the European analysis, we identified 31 significant novel and colocalized unique gene associations with fibroids (Supplementary Data 7b), four of which were not significant and colocalized in other analysis strata. There were 16 significant novel colocalized unique gene associations with fibroids in the East Asian/Central South Asian ancestry analysis (Supplementary Data 7b). The most significant association was WNT4 in lung tissue (OR 0.43, 95% CI 0.37–0.49, p = 6.01 × 10−32).
In the African ancestry analysis, we identified 15 significant gene expression-tissue pairs associated with fibroids across all tissues, but none had a colocalized gene (Supplementary Fig. 4d). Within these 15 pairs, there were four unique genes (NQO1, AC004889.1, TXNL4B, and GGT1), and all four of those genes have not been previously associated with fibroids in the GWAS catalog and were significant only in the African ancestry analysis (Supplementary Data 7d). The most significant GPGE association was with NQO1 in esophagus mucosa (OR 0.67, 95% CI 0.60–0.76, p = 3.39 × 10−10). In sum, we identified 46 novel significant and colocalized genes that have not previously been associated with fibroids across all four analyses.
Within specifically uterine tissue for the multi-ancestry analysis, there were 13 significant GPGE associations, and the only colocalized one was at HEATR3 (Table 1, Supplementary Fig. 5a). HEATR3 has been previously associated in a GPGE analysis, but in skeletal tissue only24. Seven of the 13 are novel gene associations with fibroids. Within uterine tissue for the European ancestry analysis, there were 15 significant GPGE associations, and the one that colocalized was also at HEATR3 (Table 1, Supplementary Fig. 5b). 11 of these 15 associations are novel gene associations with fibroids. The 11 novel significant gene associations observed in the European analysis included SULT1E1, LRRC34, ATP1B2, LYG1, MIR1915HG, NPAT, UIMC1, LINC01126, AC010883.1, UTP3, and C12orf65. For the East Asian/Central South Asian ancestry uterine tissue analysis, there were six significant GPGE associations (Supplementary Fig. 5c). Two of these are novel gene associations with fibroids, CDK2AP1 and C12orf65, and again HEATR3 was the only gene with a colocalized result (Table 1). CDK2AP1 predicted expression in uterine tissue was only significantly associated in the East Asian/Central South Asian analysis. There were no significant predicted gene expression associations in uterine tissue in the African ancestry analysis (Supplementary Fig. 5d).
Ingenuity Pathway Analysis
We ran Ingenuity Pathway Analysis using all colocalized significant GPGE results across the four meta-analyses and incorporating multiple open access gene map and pathway databases. The top three networks that had significant enrichment of overlapping molecules included (1) Cancer, Organismal Injury and Abnormalities, Reproductive System Disease; (2) Cancer, Neurological Disease, Organismal Injury and Abnormalities; and (3) Cell Death and Survival, Cellular Movement, Organismal Injury and Abnormalities (Supplementary Data 8). The molecular connectivity of the first network is illustrated in Fig. 6, with primary molecules including 16 focus molecules such as BET1L forming the foundation of the network. Out of 400 pathways tested, 75 were significant. The top enriched pathways included Ovarian Cancer Signaling (−log(p) = 3.05), HOTAIR Regulatory Pathways (−log(p) = 3.45), and SUMOylation of transcription factors (−log(p) = 3.13) (Supplementary Data 9). The two most significant upstream regulators, calculated from experimentally observed relationships between regulators and dataset genes to predict upstream transcriptional regulators, were TRRAP and TFAP2A. (Supplementary Data 10).
Hypergeometric distribution testing with right-tailed Fisher’s Exact Test; p-value threshold used (0.05).
Discussion
We present results from multi-ancestry genome-wide meta-analyses of uterine fibroids that leverages a larger sample size relative to prior studies, inclusion of diverse populations, and assessment of GPGE and pathways analysis of results. Within this study, we identified a total of 11 previously unpublished fibroid-associated genes. Our SNP-based heritability estimates from all four analyses were in the range of previously reported heritability (2.8–33%)28,31,37, and we observed high heritability in the African ancestry analysis (15.9%). We used an LDAK approach for heritability estimation (SumHer) due to availability of reference annotations tags for each analysis and improved performance in estimates of complex traits34,38. We also identified 46 novel genes associated with fibroids from GPGE and colocalization analyses.
We observed correlated effect allele frequencies between both the East Asian/Central South Asian and European ancestry analyses, and the African and European ancestry analyses. The effect directions of most variants were generally consistent between European and East Asian/Central South Asian ancestry populations. There was less consistency with the African ancestry group as evidenced by the correlation R of 0.44 for the European ancestry and African ancestry effect size comparison, which could be attributable to the smaller sample size of the African ancestry analyses (Supplementary Fig. 1a, b).
In our multi-ancestry meta-analysis, we identified six genes associated with fibroids only in publicly available biobank browsers and two novel genes, VIP (vasoactive intestinal peptide), and FOXO3 (forkhead box O3). VIP has been shown to inhibit uterine smooth muscle activity and act as a neurotransmitter in the female genital tract39. Additionally, VIP infusion has been shown to reverse fibrosis in myocardial cells in rats40 and VIP is known to have effects on blood pressure, which as mentioned previously is a fibroid risk factor41. FOXO3 is predicted to be involved in neoplastic cell transformation with tumor progression and angiogenesis, and there is conflicting evidence on whether it improves or harms prognosis. FOXO3 has a conserved domain that mediates its interactions between estrogen related receptor α (ERRα) and tumor protein p5342. Recently published data suggests that FOXO3a expression is increased in uterine smooth muscle tumors compared to normal myometrium43.
The DCST2 association was significant only in the European ancestry analysis. The index SNP (rs76798800) has been associated with 25-hydroxyvitamin D concentrations with the effect allele associated with decreased vitamin D levels and increased risk of fibroids44. Vitamin D deficiency is a risk factor for fibroids with evidence of overlapping genetic risk loci, and supplementation, and higher concentrations of vitamin D has been shown to decrease the size of fibroids11,45,46.
A majority of the previously unpublished genes have been associated with tumorigeneses and tumor-suppressor pathways, primarily in cancer phenotypes, including: SLC16A11, POLR2A, C11orf87, and SLC12A7 47,48,49,50,51,52. Some of these genes are associated with other risk factors for fibroids like hypertension, such as SLC12A712,53. TTC28 has been shown to be significantly down-regulated in leiomyomas with certain chromosomal deletions, but not in normal myometrium tissue54.
African ancestry analyses resulted in two significant independent loci, with variants uniquely significant in this analysis. One of the genes, WT1 (Wilms Tumor 1) is the first replication of a known fibroid-associated gene in African ancestry individuals24,28. The other gene, (Collagen Type XXII Alpha 1 Chain, COL22A1), is a novel fibroid association. Differentially expressed gene analysis has identified COL22A1 upregulation in fibroid tumors of all sizes55. COL22A1 is a gene that encodes collagen that structurally belongs to the FACIT protein family (fibril-associated collagens with interrupted triple helices).
We identified many significant associations of gene expression in specific tissues with fibroids through GPGE analyses. One notable association with a high odds ratio was ACTRT3 (actin related protein T3) in pituitary tissue (OR 5.54, 95% CI 4.19–7.33, p = 3.43 × 10−33). ACTRT3 expression has been observed to be increased in bladder tumors56. We identified one significant and colocalized gene in uterine tissue, HEATR3 (HEAT repeat protein 3), with an increased gene expression associated with risk of fibroids. This gene has been identified in previous S-PrediXcan studies but was only significantly associated with predicted expression in skeletal tissue24. HEATR3 has an important role in ribosomal protein transport and 5S ribonucleoprotein particle assembly57. Other studies have identified this as a risk gene associated with esophageal cancer and have observed an increased expression of HEATR3 in bladder cancer58,59. Further research into the role of increased HEATR3 gene expression in fibroid pathogenesis may illuminate future therapeutic directions.
In the multi-ancestry and European ancestry FUMA pathway enrichment analyses, we observed significant associations for Gene Ontology biological pathways of DNA damage response, cell cycle signaling, and checkpoint signaling. Our East Asian/Central South Asian enriched pathways included multiple pathways relating to gonad development and sex determination, and the others relating to smooth muscle cell migration and proliferation. However, all three analysis groups had the most significant GWAS catalog enrichment in uterine fibroids. These analyses provide a way to interpret the differentially powered tests. The collective and combination of results may be more informative rather than comparing the results, as differentially enriched pathways across ancestry groups is likely a function of power, and less of differing biology.
We used use Ingenuity Pathway Analysis to characterize our many significant predicted gene expression results across all the analysis groups. Shared central molecules of some of the most significant pathways included CD44, WNT4, or TP53. The significant networks, similar to the patterns observed in the FUMA pathway enrichment analyses, centered around cancer and cell cycle, with the most significant network being related to reproductive system disease.
Despite our study representing a large increase in both sample size and diversity of data used to analyze associations with uterine fibroids, the African ancestry analysis was underpowered for certain analyses like pathway enrichment. However, we were able to detect both shared and unique risk genes. We used a combination of both race and genetically informed ancestry to conduct our population stratification due to limitations in the datasets available. We affirm that race and ancestry are not interchangeable, and race is not the ideal population descriptor in most genetic studies as it is a social construct60. Future data collection methods of large genomic studies should incorporate the use of genetic ancestry or other population descriptors like genetic similarity when stratifying data.
The GTEx samples used in our genetically predicted gene expression analyses come from European ancestry individuals, and while many genetic factors are shared across population groups, this limits the application to non-European ancestry analyses. There were only 129 samples included in the uterine tissue model61, which affects our power to detect associations. Significant results in tissues such as testis and prostate can provide insight as the predicted gene expression may indicate relationships in reproductive and gonadal tissues. Increased representation of female-reproductive organs in tissue models would improve future discovery and validation. However, we were still able to identify at least one colocalized and significant predicted gene expression association in uterine tissue. Current studies are underway to add resources for gene expression of fibroids and the uterus, which can be utilized in future TWAS studies62,63.
Heritability analysis can be subject to over-estimation when using SumHer64, and we were limited in available annotation tags for the European and multi-ancestry analyses to the GBR HapMap3 annotation tags. However, as mentioned, our estimates were similar to previous reports that also used this approach and within the range of SNP-based heritability using other methods31,34,38. Our quality control LDSC analysis for the cross ancestry analysis was also limited in using European LD scores instead of a mixed ancestry option. Because European ancestry is the largest proportion of our multi-ancestry data set, we elected to use the European ancestry tags for this analysis. We were limited on further extrapolation of the gene-set enrichment tests as they only provide summarized correlation information, but the results could be utilized in future gene investigations.
In conclusion, this large-scale analysis identified several new shared and ancestry-specific genetic variations and genes of interest expressed within uterine tissue associated with fibroids. We detected known and novel genes in the under-researched African ancestry population and estimated a SNP-based heritability of 15.9%. This demonstrates the value of increasing and diversifying participation in GWAS. We have expanded the catalog of genetic associations with fibroids. Many of these genes have roles in DNA damage, cell cycle regulation, or are associated with fibroid risk factors like hypertension and vitamin D levels. These findings provide evidence for potential candidates for future research on fibroid etiology and targeted therapeutics.
Methods
Ethics
This research complies with all relevant ethical regulations, and was designated exempt non-human subjects research by the Vanderbilt University Medical Center Institutional Review Board as all study data was de-identified.
Study population
We conducted a meta-analysis of summary statistics of female participants greater than 18 years old from previous GWAS of uterine fibroids (BioVU, eMERGE, Coronary Artery Risk Development in Young Adults [CARDIA], Black Women’s Health Study [BWHS], United Kingdom Biobank [UKB], FinnGen, and Biobank Japan [BBJ])20,24,27,33, and newly acquired results (BioVU, eMERGE, All of Us Research Program [All of Us]), described below. Each cohort uses one of several strategies to specify ancestral or demographic groups that we use as strata in our meta-analyses. The population descriptors of each data source and analysis filters are presented in Supplementary Data 1. Some of the phenotyping was based on survey response. Previous literature has shown that self-report can still be a useful approach, with one study validating 96% of self-reported diagnoses65. Genetic summary statistics from each meta-analysis are available in the GWAS Catalog (see Data Availability).
BioVU
Our laboratory performed a genome-wide association study (GWAS) analysis using BioVU, Vanderbilt University’s biorepository of DNA extracted from discarded blood collected during routine clinical testing and de-identified medical records in the Synthetic Derivative as previously described in ref. 66. Vanderbilt University Medical Center patients may sign the BioVU Consent Form to indicate whether they volunteer to donate their blood samples to this genome bank for studies on genomic and bioinformatics research on gene phenotypes and gene by environment interactions. As of January 2022, more than 275,000 DNA samples were recorded. BioVU fibroid phenotypes were defined by previously validated electronic medical record algorithm67.
Newly acquired BioVU participant DNA samples were genotyped on a custom Illumina Multi-Ethnic Genotyping Array (MEGA-ex; Illumina Inc., San Diego, CA, USA). Quality control involved excluding samples or variants with missingness rates above 2%. Samples were also excluded if consent had been revoked, sample was duplicated, or sex concordance checks failed. Imputation was performed on the Michigan Imputation Server (MIS) v1.2.4 using Minimac4 and the Haplotype Reference Consortium (HRC) panel v1.1. The GWAS was performed using logistic regression implemented in SNPTEST (v2.5.4) in a race-stratified approach (non-Hispanic White, non-Hispanic Black) with covariates of age at fibroid diagnosis and first ten principal components (PCs).
eMERGE
The eMERGE Network is a consortium of several EHR-linked biorepositories formed with the goal of developing approaches for the use of the EHR in genomic research. Consortium membership has evolved over eMERGE’s 11-year history, with many sites contributing data including Group Health/University of Washington, Marshfield Clinic, Mayo Clinic, Northwestern University, Vanderbilt University, Children’s Hospital of Philadelphia (CHOP), Boston Children’s Hospital (BCH), Cincinnati Children’s Hospital Medical Center (CCHMC), Geisinger Health System, Mount Sinai School of Medicine, Harvard University, and Columbia University. The eMERGE study was approved by the Institutional Review Board at each site and all methods were performed in accordance with the relevant guidelines and regulations. Cases were identified by presence of 2 or more ICD-9 (International Classification of Diseases) or ICD-10 codes for fibroids. A detailed description of the eMERGE network has been previously published68. Participants in the eMERGE network were genotyped separately, then imputed and merged. A detailed description of the genotyping, imputation, and quality control of the eMERGE phase III array dataset has been previously reported69. Summary statistics from eMERGE phases I-II were previously published and used our validated fibroid phenotype algorithm. Additional samples through phase III were phenotyped using phecodes, and race-stratified GWAS were performed with this data using logistic regression implemented in PLINK2 and adjusted for ten principal components of ancestry and study site.
All of Us
Limiting all participants in CDR V6 (C2022Q2R7) to female sex at birth, we used the SNOMED (Systematized Medical Nomenclature for Medicine -- Clinical Terminology) uterine fibroid code to find all females with condition codes for uterine fibroids. Age was determined by age at first condition. Females with two or more condition codes are considered a case. Females with one code had to be confirmed with at least one CPT (Current Procedural Terminology) code for pelvic imaging to be considered a case. We also used the Personal Medical History Survey for self-report. Females who selected “fibroids” to the following question “Has a doctor or health care provider ever told you that you have…? (select all that apply)” were also considered a case. Age for females with self-report data only was calculated as age at survey. Controls had to have at least two CPT codes for pelvic imaging with no history of fibroids, fibroid treatment, fibroid surgery, or self-reported fibroids. Age for controls was determined by age at 1st pelvic imaging. We filtered this cohort for females with whole genome sequencing. Lastly, we separated cases and controls into predicted ancestry groups in All of Us. Details about ancestry predictions can be found in the All of Us V6 WGS Quality Report. Quality control measures included minor allele frequency greater than 0.1% and call rate greater than 95%. The regression was adjusted for the top seven principal components and age as described above and was run using Hail Version-0.2.130. All uterine fibroid research was conducted in the All of Us research workbench. Details of the All of Us Research Program are previously reported70,71.
UK Biobank
The UKB is a large-scale biomedical database that aims to improve public health by enabling scientific discoveries. African, European, East Asian, and Central South Asian individuals were subset from the UKB. Fibroid status was defined using a combination of presence of 2 or more ICD-9 or ICD-10 codes or reporting of a fibroid diagnosis from survey response as defined previously24. We restricted to variants minor allele frequency greater than 0.5% and missingness less than 5%. Genetic analysis of imputed array data used logistic regression implemented in PLINK2 and adjusted for the first ten principal components.
FinnGen
FinnGen is a large public-private partnership comprised of DNA and health data from up to 500,000 Finnish biobank participants33. We included published summary statistics from FinnGen release 8 for uterine fibroids which used ICD-10 diagnostic codes. Their methods are available on their online documentation, https://finngen.gitbook.io/. In brief, their association tests are run using regenie, adjusted for age, ten principal components, and genotyping batch as covariates. They used the approximate Firth test for variants with an initial p-value less than 0.0133.
Biobank Japan
Biobank Japan is a multi-institutional hospital-based registry comprised of DNA and medical records from individuals of Japanese ancestry. We utilized published summary statistics for uterine fibroids identified with diagnostic codes phecode 218.1. These methods are publicly available, and the binary trait association test included adjustments for age, age2, and the top 20 principal components using a generalized linear mixed model in SAIGE (v.0.37)27.
Black Women’s Health Study
Black Women’s Health Study (BWHS) is a U.S. prospective cohort study of 59,000 United States Black women who enrolled in the study in 1995 and have been followed by biennial questionnaires since then. Uterine fibroid diagnoses were ascertained by self-report from premenopausal participants, with medical record validation among a subset of cases65,72,73. GWAS of fibroids in BWHS was previously described. Briefly, the summary statistics were generated from a SNPTEST2 single-variant association analysis adjusted for the first ten principal components20.
CARDIA
CARDIA is a prospective multi-center study with 5115 adult White and Black participants of the age group 18 to 30 years, recruited from four centers. Details of the CARDIA study design have been previously published74. The CARDIA Women’s Study performed standardized study ultrasounds on women from CARDIA to detect the presence or absence of fibroids75. The fibroids GWAS has been previously described; briefly GWAS was performed using SNPTEST2 and adjusted for the first five principal components20.
Meta-analysis
Four fixed-effect inverse variance-weighted meta-analyses were performed using METAL software v(2011-3-25)76: the groups being genome-wide summary statistics of European ancestry, African ancestry, East Asian/Central South Asian ancestry, and all data combined (multi-ancestry). Stratification into analysis groups was based on either genetically determined ancestry or self-reported/electronic health record race due to data availability limitations (Supplementary Data 1).
Statistical significance was at the standard genome wide significance level of p < 5 × 10−8 using Bonferroni correction. The genomic inflation factor was calculated for each meta-analysis. The LDSC intercept was estimated using LD score regression (v1.0.1) for each meta-analysis group with common variants from HapMap335. Each variant had to be present in at least two data sets in each meta-analysis to be included in subsequent analyses. Variants were mapped to genes based on location and OpenTarget Genetics77. The correlation of effect sizes between meta-analyses was performed using R function corr.
SNP-based heritability
SNP-based heritability (h2SNP) of fibroids was also estimated using SumHer34 version 5.2. We used the pre-computed UK Biobank based BLD-LDAK model (4 models). We set the check-sums flag to NO to ignore the fact that some markers were not included in the annotated tag list, but only 0.14% were at most not overlapped. Common variants restricted to HapMap3 SNPs were used to estimate heritability. Insertions and deletions were excluded from SumHer as SumHer only accepts single nucleotide variations.
FUMA
Functional gene annotation was done using the online Functional Mapping and Annotation (FUMA)36 tool for all four meta-analyses summary statistics (accessed August 2023). FUMA identified sentinel variants determined as variants in low LD (r2 < 0.1). ANNOVAR annotations are performed for all variants, and their Gene2Func application provided differentially expressed gene and gene-set enrichment analyses36. Variants are mapped to genes and those are used for enrichment tests like differentially expressed gene sets and Gene Ontology sets. Regional locus-zoom plots were generated using FUMA.
Conditional analysis and variant categorization
We used Genome-wide Complex Trait Analysis (GCTA)78 version 1.93.0 to perform conditional-joint association testing for secondary signals. We used our FUMA-determined sentinel SNPs to identify secondary signals. Then we used the GWAS-catalog SNPs associated with fibroids to identify secondary signals to previously discovered variants. We used public databases like OpenTarget Genetics79 and the GWAS catalog to determine if these were novel gene associations, previously unpublished gene associations, or secondary signals in known gene associations.
Genetically predicted gene expression
We used S-PrediXcan (v0.7.1) with JTI models to conduct genetically predicted gene expression80,81. S-PrediXcan conducts association tests between phenotypes and gene expression levels predicted by the genetic variants provided as summary statistics based on a library of tissues from the Genotype-Tissue Expression (GTEx) project. We used covariance matrices built for all 49 GTEx v8 tissues. European genetic based covariance matrices were used for all four analysis groups. We conducted S-PrediXcan with each of the four meta-analysis summary statistics. Significance cut off was at 9.4 × 10−8 which was determined on the number of gene-tissue pairs tested. Further filtering of results was conducted based on colocalization analyses.
Colocalization
We performed colocalization analysis between the summary statistics from the meta-analyses and gene expression in GTEx v8 models in 49 tissues. The Approximate Bayes Factor (ABF) analyses were calculated using the R library package coloc (v5.2.2) and the method coloc.abf with the default priors. Significant colocalized genes were identified as those with a hypothesis 4 (H4) probability greater than 0.8, hypothesis 3 (H3) probability less than 0.3, and GPGE p-value less than the Bonferroni corrected significance level of 9.4 × 10−8. H3 is the hypothesis that both traits are associated and have different single causal variants. H4 is the hypothesis that both traits are associated and share the same single causal variant82.
Gene pathway analysis
The combined list of significant and colocalized gene-tissue results from S-PrediXcan (results from all four meta-analysis) were analyzed using the core analysis function in Ingenuity Pathway Analysis (IPA) software (Qiagen, accessed March 2024). IPA software algorithms employed have been described previously. All analyses including canonical signaling pathways, diseases and functions, and regulatory networks were ordered by enrichment p-value83.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
Genetic association summary statistics for each meta-analysis have been deposited in the GWAS Catalog under accession codes GCST90461957 (multi-ancestry), GCST90461958 (European ancestry), GCST90461959 (East Asian/Central South Asian ancestry), GCST90461960 (African ancestry). Study-specific summary statistics for BioBank Japan are available on https://www.pheweb.jp/pheno/UF, and those for FinnGen are available on https://r7.finngen.fi/pheno/CD2_BENIGN_LEIOMYOMA_UTERI. UKBB data access can be requested from https://www.ukbiobank.ac.uk/enable-your-research/apply-for-access. BioVU [https://victr.vumc.org/how-to-use-biovu/] and eMERGE [https://emerge-network.org/collaborate/] data also require approved access, which can be requested at their respective links. The All of Us [https://www.researchallofus.org/] data is accessible on the Researcher Workbench with registered access. The data from CARDIA have been deposited at dbGaP accession phs000285.v3.p2 [https://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/study.cgi?study_id=phs000285.v3.p2] and the BWHS data at dbGaP accession phs001409.v2 [https://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/study.cgi?study_id=phs001409].The predicted expression models used are publicly available on https://predictdb.org/post/2021/07/21/gtex-v8-models-on-eqtl-and-sqtl/. The other data generated in this study are provided in the Supplementary Data files.
References
Baird, D. D., Dunson, D. B., Hill, M. C., Cousins, D. & Schectman, J. M. High cumulative incidence of uterine leiomyoma in black and white women: ultrasound evidence. Am. J. Obstet. Gynecol. 188, 100–107 (2003).
Zimmermann, A., Bernuit, D., Gerlinger, C., Schaefers, M. & Geppert, K. Prevalence, symptoms and management of uterine fibroids: an international internet-based survey of 21,746 women. BMC women’s. health 12, 1–11 (2012).
Zhang, J., Go, V.-A., Blanck, J. F. & Singh, B. A Systematic Review of Minimally Invasive Treatments for Uterine Fibroid-Related Bleeding. Reprod. Sci. 29, 2786–2809 (2021).
Whiteman, M. K. et al. Inpatient hysterectomy surveillance in the United States, 2000-2004. Am. J. Obstet. Gynecol. 198, 34. e1–e7. (2008).
Cardozo, E. R. et al. The estimated annual cost of uterine leiomyomata in the United States. Am. J. Obstet. Gynecol. 206, 211. e1–211. e9 (2012).
Flake, G. P., Andersen, J. & Dixon, D. Etiology and pathogenesis of uterine leiomyomas: a review. Environ. health Perspect. 111, 1037–1054 (2003).
Laughlin, S. K. et al. Pregnancy-related fibroid reduction. Fertil. Steril. 94, 2421–2423 (2010).
Laughlin, S. K., Hartmann, K. E. & Baird, D. D. Postpartum factors and natural fibroid regression. Am. J. Obstet. Gynecol. 204, 496. e1–496. e6 (2011).
Wise, L. A. et al. Reproductive factors, hormonal contraception, and risk of uterine leiomyomata in African-American women: a prospective study. Am. J. Epidemiol. 159, 113–123 (2004).
Xu, F. et al. Vitamin D as a risk factor for the presence of asymptomatic uterine fibroids in premenopausal Han Chinese women. Fertil. Steril. 115, 1288–1293 (2021).
Harmon, Q. E. et al. Vitamin D and uterine fibroid growth, incidence, and loss: a prospective ultrasound study. Fertil. Steril. 118, 1127–1136 (2022).
Boynton-Jarrett, R., Rich-Edwards, J., Malspeis, S., Missmer, S. A. & Wright, R. A prospective study of hypertension and risk of uterine leiomyomata. Am. J. Epidemiol. 161, 628–638 (2005).
Haan, Y. C. et al. The risk of hypertension and cardiovascular disease in women with uterine fibroids. J. Clin. Hypertension 20, 718–726 (2018).
Salehi, A. M., Jenabi, E., Farashi, S., Aghababaei, S. & Salimi, Z. The environmental risk factors related to uterine leiomyoma: An umbrella review. J. Gynecol. Obstet. Hum. Reprod. 52, 102517 (2023).
Vines, A. I., Nguyen, T. T. X., Ta, M., Esserman, D. & Baird, D. D. Self‐reported daily stress, squelching of anger and the management of daily stress and the prevalence of uterine leiomyomata: the ultrasound screening study. Stress Health 27, e188–e194 (2011).
Boynton-Jarrett, R., Rich-Edwards, J. W., Jun, H.-J., Hibert, E. N. & Wright, R. J. Abuse in childhood and risk of uterine leiomyoma: the role of emotional support in biologic resilience. Epidemiology 22, 6–14 (2011).
Wise, L. A. et al. Perceived racial discrimination and risk of uterine leiomyomata. Epidemiology 18, 747–757 (2007).
Luoto, R. et al. Heritability and risk factors of uterine fibroids—the Finnish Twin Cohort study. Maturitas 37, 15–26 (2000).
Snieder, H., MacGregor, A. J. & Spector, T. D. Genes control the cessation of a woman’s reproductive life: a twin study of hysterectomy and age at menopause. J. Clin. Endocrinol. Metab. 83, 1875–1880 (1998).
Hellwege, J. N. et al. A multi-stage genome-wide association study of uterine fibroids in African Americans. Hum. Genet. 136, 1363–1373 (2017).
Rafnar, T. et al. Variants associating with uterine leiomyoma highlight genetic background shared by various cancers and hormone-related traits. Nat. Commun. 9, 3636 (2018).
Välimäki, N. et al. Genetic predisposition to uterine leiomyoma is determined by loci for genitourinary development and genome stability. Elife 7, e37110 (2018).
Cha, P.-C. et al. A genome-wide association study identifies three loci associated with susceptibility to uterine fibroids. Nat. Genet. 43, 447–450 (2011).
Edwards, T. L. et al. A Trans-Ethnic Genome-Wide Association Study of Uterine Fibroids. Front Genet 10, 511 (2019).
Sakai, K. et al. Identification of a novel uterine leiomyoma GWAS locus in a Japanese population. Sci. Rep. 10, 1197 (2020).
Eggert, S. L. et al. Genome-wide linkage and association analyses implicate FASN in predisposition to uterine leiomyomata. Am. J. Hum. Genet. 91, 621–628 (2012).
Sakaue, S. et al. A cross-population atlas of genetic associations for 220 human phenotypes. Nat. Genet. 53, 1415–1424 (2021).
Gallagher, C. et al. Genome-wide association and epidemiological analyses reveal common genetic origins between uterine leiomyomata and endometriosis. Nat. Commun. 10, 4857 (2019).
Ishigaki, K. et al. Large-scale genome-wide association study in a Japanese population identifies novel susceptibility loci across different diseases. Nat. Genet. 52, 669–679 (2020).
Masuda, T. et al. GWAS of five gynecologic diseases and cross-trait analysis in Japanese. Eur. J. Hum. Genet. 28, 95–107 (2020).
Sliz, E. et al. Evidence of a causal effect of genetic tendency to gain muscle mass on uterine leiomyomata. Nat. Commun. 14, 542 (2023).
Tai, A.-S. et al. Genome-wide causal mediation analysis identifies genetic loci associated with uterine fibroids mediated by age at menarche. Hum. Reprod. 37, 2197–2212 (2022).
Kurki, M. I. et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature 613, 508–518 (2023).
Speed, D. & Balding, D. J. SumHer better estimates the SNP heritability of complex traits from summary statistics. Nat. Genet. 51, 277–284 (2019).
Bulik-Sullivan, B. K. et al. LD Score regression distinguishes confounding from polygenicity in genome-wide association studies. Nat. Genet. 47, 291–295 (2015).
Watanabe, K., Taskesen, E., Van Bochoven, A. & Posthuma, D. Functional mapping and annotation of genetic associations with FUMA. Nat. Commun. 8, 1826 (2017).
Bray, M. J. et al. Estimating uterine fibroid SNP-based heritability in European American women with imaging-confirmed fibroids. Hum. heredity 84, 73–81 (2019).
Speed, D., Kaphle, A. & Balding, D. J. SNP‐based heritability and selection analyses: Improved models and new results. Bioessays 44, 2100170 (2022).
OTTESEN, B. et al. Vasoactive intestinal polypeptide (VIP) increases vaginal blood flow and inhibits uterine smooth muscle activity in women. Eur. J. Clin. Investig. 13, 321–324 (1983).
Duggan, K. A., Hodge, G., Chen, J. & Hunter, T. Vasoactive intestinal peptide infusion reverses existing myocardial fibrosis in the rat. Eur. J. Pharmacol. 862, 172629 (2019).
Henning, R. J. & Sawmiller, D. R. Vasoactive intestinal peptide: cardiovascular effects. Cardiovasc. Res. 49, 27–37 (2001).
Song, S. S. et al. High expression of FOXO3 is associated with poor prognosis in patients with hepatocellular carcinoma. Oncol. Lett. 19, 3181–3188 (2020).
de Almeida, T. G. et al. FOXO3a deregulation in uterine smooth muscle tumors. Clinics 79, 100350 (2024).
Hyppönen, E., Vimaleswaran, K. S. & Zhou, A. Genetic determinants of 25-hydroxyvitamin D concentrations and their relevance to public health. Nutrients 14, 4408 (2022).
Hajhashemi, M., Ansari, M., Haghollahi, F. & Eslami, B. The effect of vitamin D supplementation on the size of uterine leiomyoma in women with vitamin D deficiency. Casp. J. Intern. Med. 10, 125 (2019).
Wise, L. A., Ruiz-Narváez, E. A., Haddad, S. A., Rosenberg, L. & Palmer, J. R. Polymorphisms in vitamin D–related genes and risk of uterine leiomyomata. Fertil. Steril. 102, 503–510. e1 (2014).
Li, J. et al. A pan-cancer analysis revealed the role of the SLC16 family in cancer. Channels 15, 528–540 (2021).
Pontén, F., Jirström, K. & Uhlen, M. The Human Protein Atlas—a tool for pathology. J. Pathol.: A J. Pathological Soc. Gt. Br. Irel. 216, 387–393 (2008).
Lyu, X., Wang, P., Qiao, Q. & Jiang, Y. Genomic stratification based on microenvironment immune types and PD-L1 for tailoring therapeutic strategies in bladder cancer. BMC Cancer 21, 646 (2021).
Xu, J. et al. Precise targeting of POLR2A as a therapeutic strategy for human triple negative breast cancer. Nat. Nanotechnol. 14, 388–397 (2019).
Tran, M. T. et al. Methylomic analysis identifies C11orf87 as a novel epigenetic biomarker for GI cancers. Plos one 16, e0250499 (2021).
Chen, Y.-F., Chou, C.-Y., Ellory, J. C. & Shen, M.-R. The emerging role of KCl cotransport in tumor biology. Am. J. Transl. Res. 2, 345 (2010).
Tremblay, A. M., Dufour, C. R., Ghahremani, M., Reudelhuber, T. L. & Giguère, V. Physiological genomics identifies estrogen-related receptor α as a regulator of renal sodium and potassium homeostasis and the renin-angiotensin pathway. Mol. Endocrinol. 24, 22–32 (2010).
Mehine, M. et al. Integrated data analysis reveals uterine leiomyoma subtypes with distinct driver pathways and biomarkers. Proc. Natl Acad. Sci. 113, 1315–1320 (2016).
Grandhi, A., Guo, W. & Peddada, S. D. A multiple testing procedure for multi-dimensional pairwise comparisons with application to gene expression studies. BMC Bioinforma. 17, 1–12 (2016).
Figueroa, J. D. et al. Genome-wide association study identifies multiple loci associated with bladder cancer risk. Hum. Mol. Genet. 23, 1387–1398 (2014).
Calviño, F. R. et al. Symportin 1 chaperones 5S RNP assembly during ribosome biogenesis by occupying an essential rRNA-binding site. Nat. Commun. 6, 6510 (2015).
Wu, C. et al. Genome-wide association analyses of esophageal squamous cell carcinoma in Chinese identify multiple susceptibility loci and gene-environment interactions. Nat. Genet. 44, 1090–1097 (2012).
Dou, X. et al. HEATR3 involved in the cell proliferation, metastasis and cell cycle development of bladder cancer acts as a tumor suppressor. Mol. Genet. Genomics 298, 1353–1364 (2023).
Feero, W. G. et al. Guidance on use of race, ethnicity, and geographic origin as proxies for genetic ancestry groups in biomedical publications. Am. J. Hum. Genet. 111, 621–623 (2024).
Consortium, G. The GTEx Consortium atlas of genetic regulatory effects across human tissues. Science 369, 1318–1330 (2020).
Edwards, T. L. et al. Challenges and Opportunities for Data Science in Women’s Health. Annu. Rev. Biomed. Data Sci. 6, 23–45 (2023).
Piekos, J. A. et al. RNA Sequencing of Uterine Leiomyoma Tumors and Control Myometrium Detects Relationships with Cancer Genes and Pathways. Available at SSRN 4343748.
Srivastava, A. K., Williams, S. M. & Zhang, G. Heritability estimation approaches utilizing genome‐wide data. Curr. Protoc. 3, e734 (2023).
Wise, L. A., Palmer, J. R., Stewart, E. A. & Rosenberg, L. Age-specific incidence rates for self-reported uterine leiomyomata in the Black Women’s Health Study. Obstet. Gynecol. 105, 563–568 (2005).
Roden, D. M. et al. Development of a large‐scale de‐identified DNA biobank to enable personalized medicine. Clin. Pharmacol. Therapeutics 84, 362–369 (2008).
Feingold-Link, L., Edwards, T. L., Jones, S., Hartmann, K. E. & Velez Edwards, D. R. Enhancing uterine fibroid research through utilization of biorepositories linked to electronic medical record data. J. women’s. health 23, 1027–1032 (2014).
McCarty, C. A. et al. The eMERGE Network: a consortium of biorepositories linked to electronic medical records data for conducting genomic studies. BMC Med. genomics 4, 1–11 (2011).
Stanaway, I. B. et al. The eMERGE genotype set of 83,717 subjects imputed to~ 40 million variants genome wide and association with the herpes zoster medical record phenotype. Genet. Epidemiol. 43, 63–81 (2019).
Investigators, A.o.U.R.P. The “All of Us” research program. N. Engl. J. Med. 381, 668–676 (2019).
Investigators, T. Ao. U. R. P. G. Genomic data in the All of Us Research Program. Nature 627, 340–346 (2024).
Wise, L. A. et al. Risk of uterine leiomyomata in relation to tobacco, alcohol and caffeine consumption in the Black Women’s Health Study. Hum. Reprod. 19, 1746–1754 (2004).
Wise, L. A. et al. Influence of body size and body fat distribution on risk of uterine leiomyomata in US black women. Epidemiology 16, 346–354 (2005).
Friedman, G. D. et al. CARDIA: study design, recruitment, and some characteristics of the examined subjects. J. Clin. Epidemiol. 41, 1105–1116 (1988).
Wellons, M. F. et al. Racial differences in self-reported infertility and risk factors for infertility in a cohort of black and white women: the CARDIA Women’s Study. Fertil. Steril. 90, 1640–1648 (2008).
Willer, C. J., Li, Y. & Abecasis, G. R. METAL: fast and efficient meta-analysis of genomewide association scans. Bioinformatics 26, 2190–2191 (2010).
Koscielny, G. et al. Open Targets: a platform for therapeutic target identification and validation. Nucleic acids Res. 45, D985–D994 (2017).
Yang, J. et al. Conditional and joint multiple-SNP analysis of GWAS summary statistics identifies additional variants influencing complex traits. Nat. Genet. 44, 369–375 (2012).
Mountjoy, E. et al. An open approach to systematically prioritize causal variants and genes at all published human GWAS trait-associated loci. Nat. Genet. 53, 1527–1533 (2021).
Barbeira, A. N. et al. Exploring the phenotypic consequences of tissue specific gene expression variation inferred from GWAS summary statistics. Nat. Commun. 9, 1825 (2018).
Zhou, D. et al. A unified framework for joint-tissue transcriptome-wide association and Mendelian randomization analysis. Nat. Genet. 52, 1239–1246 (2020).
Wallace, C. A more accurate method for colocalisation analysis allowing for multiple causal variants. PLoS Genet. 17, e1009440 (2021).
Krämer, A., Green, J., Pollard, J. Jr & Tugendreich, S. Causal analysis approaches in ingenuity pathway analysis. Bioinformatics 30, 523–530 (2014).
Acknowledgements
This work was supported in by UL1TR000445 (JK), K12AR084232 (JNH, EAJ, MMS: PI: DRVE), R01HD074711(DRVE), R03HD078567(DRVE), R01HD093671 (DRVE/TLE), and R01HD112169 (DRVE/TLE). Additional support was provided by R01GM139891 (WW), K25DK128563 (AK), ZIAHG200417 (AW, JMK: PI: JCD), U01CA164974 (JRP). Uterine fibroid-related work from the Black Women’s Health Study (BWHS) was supported in part by R01HD057966 (LAW), R01CA098663 (JRP). Vanderbilt University Medical Center’s BioVU projects are supported by numerous sources: institutional funding, private agencies, and federal grants. These include NIH funded Shared Instrumentation Grant S10OD017985, S10RR025141, and S10OD025092; CTSA grants UL1TR002243, UL1TR000445, and UL1RR024975. Genomic data are also supported by investigator-led projects that include U01HG004798, R01NS032830, RC2GM092618, P50GM115305, U01HG006378, U19HL065962, R01HD074711; and additional funding sources listed here: https://victr.vumc.org/biovu-funding/. The research has been conducted using the UK Biobank Resource (Application #13869) http://www.ukbiobank.ac.uk. We want to acknowledge the participants and investigators of the FinnGen study. This work is supported by multiple eMERGE phases. Phase 1: The eMERGE Network was initiated and funded by NHGRI, in conjunction with additional funding from NIGMS through the following grants: U01HG004610 (Group Health Cooperative/University of Washington); U01HG004608 (Marshfield Clinic Research Foundation and Vanderbilt University Medical Center); U01HG04599 (Mayo Clinic); U01HG004609 (Northwestern University); U01HG04603 (Vanderbilt University Medical Center, also serving as the Administrative Coordinating Center); U01HG004438 (CIDR) and U01HG004424 (the Broad Institute) serving as Genotyping Centers. Phase 2: The eMERGE Network was initiated and funded by NHGRI through the following grants: U01HG006389 (Essentia Institute of Rural Health, Marshfield Clinic Research Foundation and Pennsylvania State University); U01HG006382 (Geisinger Clinic); U01HG006375 (Group Health Cooperative/University of Washington); U01HG006379 (Mayo Clinic); U01HG006380 (Icahn School of Medicine at Mount Sinai); U01HG006388 (Northwestern University); U01HG006378 (Vanderbilt University Medical Center); and U01HG006385 (Vanderbilt University Medical Center serving as the Coordinating Center); U01HG004438 (CIDR) and U01HG004424 (the Broad Institute) serving as Genotyping Centers. Phase 3: This phase of the eMERGE Network was initiated and funded by the NHGRI through the following grants: U01HG008657 (Group Health Cooperative/University of Washington); U01HG008685 (Brigham and Women’s Hospital); U01HG008672 (Vanderbilt University Medical Center); U01HG008666 (Cincinnati Children’s Hospital Medical Center); U01HG006379 (Mayo Clinic); U01HG008679 (Geisinger Clinic); U01HG008680 (Columbia University Health Sciences); U01HG008684 (Children’s Hospital of Philadelphia); U01HG008673 (Northwestern University); U01HG008701 (Vanderbilt University Medical Center serving as the Coordinating Center); U01HG008676 (Partners Healthcare/Broad Institute); U01HG008664 (Baylor College of Medicine); and U54MD007593 (Meharry Medical College).
Author information
Authors and Affiliations
Contributions
Conceptualization: J.K., T.L.E., D.R.V.E., J.N.H.; Data curation: J.K., A.W., S.H.J., L.A.W., T.L.E., D.R.V.E., J.N.H.; Formal analysis: J.K., A.W., H.N., J.J., M.M.S.; Funding acquisition T.L.E., D.R.V.E.; Investigation: J.K., H.N., M.M.S.; Project administration: S.H.J., T.L.E., D.R.V.E., J.N.H.; Resources: J.M.K., M.A.A., J.R.P., L.J.R.T., L.C.K., E.A.R.N., L.A.W., J.C., J.C.D., W.Q.W.; Supervision: T.L.E., D.R.V.E., J.N.H.; Visualization: J.K., H.N.; Writing – original draft: J.K., T.L.E., D.R.V.E., J.N.H.; Writing – review and editing: J.K., A.W., M.M.S., L.A.W., A.K., E.A.J., E.A.R.N., L.J.R.T., L.C.K., J.R.P., M.A.A., J.C.D., P.M., K.K., W.Q.W.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks Danielle Rasooly, C. Scott Gallagher and the other, anonymous, reviewers for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Kim, J., Williams, A., Noh, H. et al. Genome-wide meta-analysis identifies novel risk loci for uterine fibroids within and across multiple ancestry groups. Nat Commun 16, 2273 (2025). https://doi.org/10.1038/s41467-025-57483-5
Received:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s41467-025-57483-5
