Abstract Background: There is limited literature regarding Haitian women and breast cancer. What exists is based on screening with no literature on clinical or pathologic subtypes of breast cancer. We examined patient and tumor characteristics in Haitian born women with invasive breast cancer in our ethnically diverse tertiary care hospital. Methods: We expanded a pre-existing database to include all women diagnosed with breast cancer from 1998–2011. From the electronic medical record we documented age, body mass index (BMI), place of birth, self-identified race/ethnicity, date of diagnosis, T, N, M stage, grade, ER, PR, HER2 expression, recurrence and treatment data including neoadjuvant, surgical, adjuvant and radiation therapy and adequacy of treatment. We determined associations using contingency tables and multivariate logistic regression. Results: 480 cases identified. Patients were racially diverse: 35% white (W), 41% black (B), 14% Hispanic and 10% other with 8% of patients self-identified as Haitian (Ha). Ha patients were younger at diagnosis, with mean age 55.1 vs. 59.4 overall p=0.04. Mean age W: 62 B: 58. Mean BMI was not significant. Ha women had an odds ratio (OR) of triple negative (Tneg) breast cancer: 3.078 (CI: 1.608−5.892, p=0.0007), B: 3.392 (CI: 2.113 - 5.441, p=<0.0001), W: 0.339 (CI: 0.192−0.599, p=0.0002) when controlling for age and BMI. Ha OR of ER+: 0.459 (CI: 0.246−0.859, p=0.0149), B: 0.384 (CI: 0.257 - 0.575, p<0.0001), W: 0.8771 (CI: 1.525 - 3.792 p=0.0002). Ha OR of PR+: 0.535 (CI: 0.288−0.995, p=0.0481) when controlling for age and BMI. Her2+ expression was not significantly different. OR of recurrence in Ha: 2.110 (CI: 1.068−4.171, p= 0.0317) controlling for age and BMI. When grade was added to the multiple regression model, the odds of recurrence in Ha women was no longer significant OR 1.845 (CI: 0.923−3.690, p=0.0833); however, recurrence appeared to be associated with grade OR 1.858 (CI: 1.273−2.711, p=0.0013). Higher grade was significantly associated with Ha status, controlling for categorical age and BMI (mean score differences p = 0.0086). Crude rates of each grade in Ha women were 4% grade 1, 38% grade 2, 58% grade 3 and the rates in non-Ha women were 14% grade 1, 48% grade 2, 38% grade 3. Exploring recurrence further, treatment records were reviewed and while a minority of patients did not receive adequate primary therapy, the majority of women did receive adequate therapy. Conclusions: Ha women were younger at diagnosis, had a higher risk of triple negative, ER negative and PR negative tumors and had more grade 3 tumors when controlling for age and BMI. Rates of recurrence appeared higher in Ha women, but the association was no longer significant when controlling for grade in addition to age and BMI. Analysis is ongoing, and we expect to report data on survival and treatment adequacy when compared to other groups in our dataset. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P2-14-06.