LAUSR

Background: The microenvironment of triple-negative breast cancer (TNBC) can be divided into three clusters based on bioinformatics-based immunogenomic analysis: the “immune-desert” cluster, the “innate immune-inactivated” cluster, and the “immune-inflamed” cluster. The immune-inflamed cluster is considered as “hot tumor” while the other two are considered as “cold tumor”. Methods: To investigate the prognostic effect of microenvironment phenotypes on TNBC, we compared relapse-free survival (RFS) of different phenotypes in 100 patients with RNA sequencing-based expression data from the PATTERN trial (NCT01216111, published in JAMA Oncol 2020), which indicated a superior efficacy of adjuvant paclitaxel-plus-carboplatin regimen compared to the regimen of cyclophosphamide/epirubicin/fluorouracil followed by docetaxel for TNBC. We also analyzed the efficacy of the two regimens for different immune phenotypes to explore potential treatment strategies. Results: No significant difference in RFS was observed between the “hot tumor” and the “cold tumor” (hazard ratio [HR] = 0.68, 95% confidence interval [CI] 0.28–1.66, P = 0.40). However, the “hot tumor” subtype was associated with significantly longer RFS in node-positive patients (HR = 0.27, 95%CI 0.07–0.97, P = 0.03). Consistently, a similar trend to improved RFS of the “hot tumor” phenotype was detected in patients with stage pT2-3 tumors (HR = 0.29, 95%CI 0.06–1.30, P = 0.08). Furthermore, no significant difference in RFS between the two treatment arms was observed in patients with “hot tumor” (HR = 0.39, 95% CI 0.08–2.01, P = 0.24) or “cold tumor” (HR = 1.05, 95% CI 0.39–2.82, P = 0.92). Conclusion: The microenvironment phenotype in TNBC might have prognostic significance to patients with a high risk of recurrence. The association of the microenvironment phenotypes with the efficacy of adjuvant chemotherapy for TNBC remains to be further studied.