Background: Our previously developed method, which efficiently integrates multi-omic data and sequentially clusters samples using the Wasserstein distance and hierarchical clustering, identified a distinct subtype of breast cancer with poor prognosis within The Cancer Genome Atlas (TCGA) breast cancer cohort [1]. Methods and Materials: To further characterize the biological features associated with this subtype, we analyzed multi-modal data, including DNA mutation, hypoxia signatures, and tumor infiltrating lymphocytes (TILs) estimated on pathology images [2]. In the TCGA breast cancer cohort (N = 726), the poor prognosis group was defined as a high-risk group (N = 53) and the remaining simples (N = 673) were categorized as a low-risk group. Results: In Kaplan-Meier analysis, a significant difference in survival rates was found between the high- and low-risk groups with a log-rank p of 0.0006. A higher rate of African American individuals (37.4%) belonged to the basal subtype relative to Asian (8.1%) and White (12.9%) individuals (p = 5.6E-10). A difference in DNA mutation rates was assessed between the high- and low-risk groups. The low-risk group exhibited a higher mutation rate in PIK3CA (p = 0.0002), while the high-risk group displayed a higher mutation rate in TP53 (p = 0.0038). For hypoxia signatures, the high-risk group showed significantly higher hypoxia scores than the low-risk group (p < 0.0001). African American individuals exhibited significantly higher hypoxia scores compared to other races (p < 0.0001). In tumor immune microenvironment analysis, the low-risk group showed a higher leukocyte fraction than the high-risk group (p = 1.3E-05). The spatial fraction of TILs estimated on histology images was higher in the low-risk group compared to the high-risk group in the basal subtype (p = 0.0362). Conclusion: These findings indicate that breast cancers with poor prognosis are associated with higher hypoxia levels and altered tumor immune microenvironment characteristics, including lower leukocyte fraction and lower TILs.