Abstract Background: Mounting evidence indicates that tumor-specific host immunity is associated with favorable prognosis in a variety of cancer settings. However, the role of immunity in breast cancer remains controversial. The purpose of this study was to determine the significance of tumor infiltrating leukocytes (TIL) with respect to pathological features and patient survival in estrogen receptor alpha (ER) negative breast cancer. Patients and Methods: Tissue microarrays and immunohistochemistry were used to assess and categorize TIL in a cohort of 255 ER-negative invasive ductal breast carcinomas. Cases were registered by the Manitoba Breast Tumor Bank during the years 1988 to 2000 and had a minimum follow-up of 36 months. Stromal and intraepithelial TIL expressing the markers CD3, CD8, CD4, TIA1, CD25, FOXP3, CD20, CD68, and myeloperoxidase (MPO) were quantified by an experienced pathologist (PHW) and evaluated for associations with pathological parameters and survival. The median value for each marker was used as a cut-point to define high versus low cases. Results: TIL expressing CD4, FOXP3, TIA1, CD20, or MPO were significantly associated with improved overall and disease-free survival (DFS). A combination of high TIA1+ and FOXP3+ TIL, termed IR (immune response)-high, was superior to any single TIL marker and was significantly associated with DFS in multivariate analysis involving the following covariates: nodal status, patient age, tumor size, tumor grade, Her2/PR expression, and adjuvant treatment (HR = 0.478, 95% CI 0.246–0.932; P = 0.030). IR associated with DFS principally in patients with axillary lymph node metastasis (P = 0.0019 versus P = 0.2659 for node-negative patients) and in those who did not receive adjuvant radiation or chemotherapy (P < 0.0001 versus P = 0.0489 for adjuvant-treated patients). Similarly, nodal status and adjuvant therapy had prognostic value only for patients with low IR. Indeed, the outcomes of IR-high cases with or without lymph node metastasis were nearly identical (10 year DFS rates of 65% and 69%, respectively; P = 0.6503), as were those of IR-high patients who did or did not receive adjuvant therapy (10 year DFS rates of 68% and 67%, respectively; P = 0.7997). Unlike standard prognostic features such as nodal status (P = 0.7941), IR correlated with DFS at times greater than five years after diagnosis (P = 0.0084). Conclusions: High levels of TIL are indicative of favorable prognosis in ER-negative breast cancer. The prognostic effects of TIL are particularly relevant for node-positive patients and those not treated with adjuvant radiation or chemotherapy. These findings suggest that in ER-negative disease, the survival benefits of adjuvant therapy may be restricted to patients with poor tumor-specific immune responses. Assessment of TIL may improve overall prognostic prediction for ER-negative breast cancer and may be useful in identifying patients who are likely to respond to adjuvant regimens. Furthermore, our data suggest that ER-negative breast cancer may be a strong candidate for the design and deployment of novel immunotherapies. Citation Information: Clin Cancer Res 2010;16(7 Suppl):B13