Adaptive immune microenvironment associated with prolonged disease-free interval in estrogen receptor positive breast cancer

Abstract Adaptive immune response is associated with improved breast cancer outcomes, especially in the early interval after estrogen receptor negative (ER-) diagnoses. Long-term outcomes have seldom been evaluated. We characterized immune patterns based on predefined 48 immune genes, including adaptive and innate immune pathways, among 1899 Carolina Breast Cancer Study (CBCS) participants with long-term disease-free interval (DFI) data (with 10+ years follow-up). Latent Class Analysis, modeling both expression patterns and magnitude, identified Adaptive, Innate, and Quiet immune microenvironments. Cox proportional hazards estimated associations between immune classes and DFI in early (≤5 years) and late windows (5–10 years). Quiet immune class was significantly associated with a higher hazard of breast cancer recurrence in ER+ in both 5-year [Hazard Ratio (HR): 2.03, 95% Confidence Interval (CI): (1.16, 3.56)] and 5-to-10-year [HR: 4.93, 95% CI: (1.76, 13.77)] timeframes compared to Adaptive class. Quiet showed a similar but higher association in ER- tumors compared to Adaptive [5-year HR: 2.67 95% CI: (1.65, 4.33)]. In SCAN-B, long-term prognostic associations among ER+ breast cancers were directionally consistent with those observed in the CBCS, although effect sizes were attenuated. Classification of breast cancer microenvironments may have value for predicting long-term breast cancer prognosis.