Abstract The ancestral makeup of Hispanic women in the United States is complex; they form an admixed population including European (EUR), Amerindian (AI), and African (AFR) ancestry. Studies have demonstrated differences in the distribution of molecular subtypes of breast cancer between non-Hispanic white women and Hispanic women. To better understand the relationship of breast cancer biomarkers/molecular subtypes and ancestry in a U.S. Hispanic population, we assessed histologic characteristics, tumor cell protein expression by immunohistochemistry (IHC), gene expression using the Affymetrix Human HTA 2.0 array, and ancestry by ancestry informative markers (AIMs) in a cohort of 331 Hispanic women diagnosed with breast cancer in either Washington or New Mexico. To assess individual ancestry, we genotyped a panel of 215 AIMs in 260 women with available DNA. We ran ancestry analyses using one to ten populations in the model; three populations gave the lowest cross-validation error (0.53). In this cohort, we observed the largest contribution from EUR ancestry (mean=58.6%; range=8.3-97.6%), followed by AI ancestry (mean=33.6%; range=0-89.5%) and AFR ancestry (mean=7.9%; range=0-83.9%). We did not see significant associations between ancestry proportion and individual IHC markers ER, PR, HER2, CK5/6, EGFR, or tumor grade. However, three markers (tumor lymphocyte response, p53 positivity, high Ki67 positivity) were associated with ancestry: greater AFR ancestry with higher lymphocyte response (OR=1.045 per 1% increase in AFR ancestry, 95% CI: 1.004-1.087, p=0.03) and an increasing AI ancestry with high Ki67 positivity (OR=1.032 per 1% increase in AI ancestry, 95% CI: 1.005-1.061, p=0.02) and high p53 expression (OR=1.035 per 1% increase in AI ancestry, 95% CI: 1.008-1.062, p=0.01). We compared St. Gallen IHC subtypes with ancestry and observed a positive association between the ER-/PR-/HER2+ subtype and AI ancestry. Specifically, for every 1% increase in AI ancestry, we observed a 4.2% (95% CI: 0.2%-8.3%) increased risk for ER-/PR-/HER2+ disease compared to patients with Luminal A (ER+/PR+/HER2-) disease (p=0.04). Gene expression intrinsic subtyping by PAM50 was also suggestive of this association: an increasing proportion of AI ancestry was associated with increased odds of having the HER2-enriched subtype compared with Luminal A subtype (OR=1.030, 95% CI: 0.995-1.067, p=0.09). In conclusion, ancestry in this population of Hispanic women with breast cancer was mixed. Greater AI ancestry was associated with increased odds of ER-/PR-/HER2+ disease and of two markers of aggressive breast cancer, high tumor cell proliferation and p53 expression. There was no association identified between ancestry and triple negative breast cancer. Other studies have identified that HER2+ breast cancer rates in Hispanic women are higher than those in non-Hispanic white women. These data highlight the ancestral heterogeneity of the Hispanic population and the need to fully understand which subgroups of Hispanic women might be at an increased risk for specific breast cancer types, especially those that are aggressive and associated with poor outcomes. Citation Format: Jamie Guenthoer, Xiaoyu Chai, Karen W. Makar, Chris S. Carlson, Rania Bakkar, Irena B. King, Huining Kang, Beti Thompson, Christopher I. Li, Li Hsu, Linda S. Cook, Peggy L. Porter. Relationship of ancestry and breast cancer subtypes in Hispanic women from New Mexico and Washington State. [abstract]. In: Proceedings of the Eighth AACR Conference on The Science of Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; Nov 13-16, 2015; Atlanta, GA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2016;25(3 Suppl):Abstract nr C56.