Abstract Background Recent molecular profiling advances have transformed breast cancer treatment. These breakthroughs have largely benefited patients with hormone receptor-positive and triple-negative breast cancers, while HER2-positive (HER2+) breast cancer remains less understood molecularly. We evaluated the tumor immune microenvironment of HER2+ breast cancer, assessing immune cell infiltration, TMB, PD-L1 status, and HLA allele prevalence, to uncover biomarkers for treatment guidance. Methods We used Tempus Lens (Tempus AI, Inc., Chicago, IL) to identify 455 female patients with HER2+ breast cancer and xT and xR testing. Patients were classified as having localized (stage 1-3, N = 124) or de novo metastatic disease (stage 4, N = 331). Wilcoxon rank sum, Fisher’s exact, and Pearson’s Chi-squared tests were used to compare groups. Immune cell proportions were estimated using quanTIseq. TMB, PD-L1, HLA, MSI, and MMR were also analyzed. Real-world progression-free survival (rwPFS) was time from therapy (neoadjuvant or first-line) start to progression. Real-world overall survival (rwOS) was time from first-line therapy start to death or loss to follow up. Median rwOS and rwPFS were estimated with Kaplan-Meier (KM) and hazard ratios (HR) using Cox proportional hazard models (CoxPh). Results In patients with localized disease, the median proportion of tumor-infiltrating B cells was significantly higher than in those with metastatic disease (p 0.001). A similar pattern was observed for natural killer (NK) cells, CD8+ T cells, and regulatory T cells (Tregs), with enrichment in localized stage patients (p0.001, p0.001, and p 0.001, respectively). TMB high status (10 mutations/Mb) was observed in 6.7% of patients with localized disease and 10% of patients with de novo metastatic disease. 29% of patients with localized disease were PD-L1 IHC positive, as were 11% of patients with de novo metastatic disease. Interestingly, in metastatic patients who received standard chemotherapy and/or anti-HER2 therapy as first line treatment, positive PD-L1 status and high TMB were significantly associated with worse median rwOS (p0.021 and p0.001, respectively). Across all HER2+ breast cancer patients with patient-reported race information available, the HLA-A *02:01 allele exhibited the highest overall frequency (43%). Stratified by race, this allele was most prevalent in White patients (48%), with significantly lower frequencies observed in Asian (25%) and Black (24%) patients (p=0.018), consistent with previous studies. In contrast, the HLA-A *33:03 allele demonstrated a markedly higher frequency in Asian patients (33%), while its prevalence was substantially lower in Black (17%) and White (1.4%) patients (p0.001). Similarly, the HLA-A *23:01 allele was more frequent in Black patients (36%) compared to White (5.2%) and Asian (0%) patients (p0.001). Conclusions Our findings suggest a potential therapeutic benefit of incorporating immunotherapy into the treatment paradigm for HER2+ breast cancer, in both localized and metastatic disease settings. Moreover, the ethnically enriched HLA polymorphisms within our patient cohort may help elucidate observed disparities in clinical outcomes and offer valuable insights for the development of population-tailored immunotherapeutic strategies. Citation Format: Y. Gao, M. Weitz, S. Fragkogianni, K. Layng, H. Mai, J. Chang. The Molecular and Immune Landscape of HER2 Positive Breast Cancer abstract. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS4-05-21.
Gao et al. (2026) studied this question.