Abstract Background: Most genome-wide association studies (GWAS) have been carried out in European ancestry populations; no risk variants for breast cancer have been identified solely from African ancestry GWAS data. Few GWAS hits have replicated in African ancestry populations. Methods: In a nested case–control study of breast cancer in the Black Women's Health Study (1,199 cases/1,948 controls), we evaluated index single-nucleotide polymorphisms (SNP) in 21 loci from GWAS of European or Asian ancestry populations, overall, in subtypes defined by estrogen receptor (ER) and progesterone receptor (PR) status (ER+/PR+, n = 336; ER−/PR−, n = 229), and in triple-negative breast cancer (TNBC, N = 81). To evaluate the contribution of genetic factors to population differences in breast cancer subtype, we also examined global percent African ancestry. Results: Index SNPs in five loci were replicated, including three associated with ER−/PR− breast cancer (TERT rs10069690 in 5p15.33, rs704010 in 10q22.3, and rs8170 in 19p13.11): per allele ORs were 1.29 [95% confidence interval (CI) 1.04–1.59], P = 0.02, 1.52 (95% CI 1.12–2.08), P = 0.01, and 1.30 (95% CI 1.01–1.68), P = 0.04, respectively. Stronger associations were observed for TNBC. Furthermore, cases in the highest quintile of percent African ancestry were three times more likely to have TNBC than ER+/PR+ cancer. Conclusions: These findings provide the first confirmation of the TNBC SNP rs8170 in an African ancestry population, and independent confirmation of the TERT ER− SNP. Furthermore, the risk of developing ER− breast cancer, particularly TNBC, increased with increasing proportion of global African ancestry. Impact: The findings illustrate the importance of genetic factors in the disproportionately high occurrence of TNBC in African American women. Cancer Epidemiol Biomarkers Prev; 22(1); 127–34. ©2012 AACR.