Abstract Background: The importance of mental health to improve the quality of cancer survivorship and outcomes is gaining increased attention. Major depressive disorder (MDD) has been associated with higher breast cancer (BC) incidence, BC-specific and all-cause mortality. The effect of MDD on BC recurrence remains under-studied. The Veterans Health Administration (VHA) provides a unique opportunity to examine this relationship as it mandates annual MDD screening in the primary care setting and reports national screening rates of 96%. Objective: To examine the relationship between pre-existing MDD and BC recurrence among a diverse cohort of women in the United States. Methods: We established a retrospective cohort of women (age ≥ 18 years) diagnosed with early-stage incident BC from 2010 through 2019 using the VHA’s electronic medical record database. We used a two-year window to identify women with clinically diagnosed MDD prior to their BC diagnosis. Our outcome of interest was BC recurrence (local recurrence/second primary and distant recurrence). We used multivariable proportional hazards regression to examine the association between MDD and BC recurrence, accounting for competing risk of death. Our analyses were adjusted for age, race, ethnicity, marital status, priority group rating (proxy for socioeconomic status), rurality of primary residence, geographical region, cancer stage, year of BC diagnosis, and comorbidity burden. Priority group rating is a metric that incorporates patients’ income, receipt of VHA assistance benefits, capacity for gainful employment, and severity of service-connected conditions. We conducted subgroup analyses by tumor subtype. RESULTS: Our cohort consists of 6,045 women with BC, of whom 1,750 (29.0%) had a pre-existing MDD diagnosis. The median length of follow-up from BC diagnosis was 4.5 years (IQR=4.8 years). The average age at BC diagnosis was 57 years (SD=11 years) overall and 56 years (SD=10 years) among women with MDD. Sixty-one percent of women were white, 29% were black, 27% were married, and 28% were rural dwelling. Forty-four percent of invasive carcinomas were localized and 21% were regional. Thirty-five percent of patients had ductal carcinoma in situ. Women with MDD had a higher average Charlson Comorbidity Index score (mean=0.75, SD=1.13 vs. mean=0.58, SD=1.03) and proportion with a priority group rating between 1-4 (40% vs. 32%). The distribution of other described patient characteristics among women with MDD were similar to the overall cohort. On multivariable analysis, women with MDD had a 31% (95% CI: 1.11, 1.56) higher risk of BC recurrence compared to women without MDD. MDD was associated with a 29% (95% CI: 1.01, 1.66) and 33% (95% CI: 1.05, 1.67) higher risk of local recurrence/second primary and distant recurrence, respectively. The relationship between MDD and BC recurrence remained significant after adjusting for tumor subtype (HR=1.40, 95% CI: 1.07, 1.83). In subgroup analyses stratified by tumor subtype, MDD was associated with a higher risk of BC recurrence among women with estrogen receptor positive versus negative tumors (HR=1.65, 95% CI: 1.27, 2.14 vs. HR=0.71, 95% CI: 0.34, 1.46) and HER2/neu negative versus positive tumors (HR=1.53, 95% CI: 1.16, 2.00 vs. HR=1.24, 95% CI: 0.51, 2.96). The interaction between estrogen receptor status and MDD was statistically significant (p-interaction=0.02). CONCLUSION: Women with incident BC should be screened for MDD early in the cancer care continuum given the prevalence of MDD and association with increased risk of local recurrence/second primary and distant BC recurrence. Further research is needed to understand the interplay between these two diseases and their treatments. Citation Format: Maya Aboumrad, Brian Shiner, Avonne Connor, Bradley V. Watts, Kala Visvanathan. Depression and breast cancer recurrence among female veterans in the United States: a retrospective cohort study [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P3-03-08.