Abstract P3-07-07: Functional hotness score of cytotoxic T-lymphocytes infiltration to predicts outcome of triple-negative breast cancer patients

Abstract

Abstract Introduction: Cytotoxic T-lymphocytes (CTLs) infiltration into tumor has been shown to predict better prognosis in breast cancer. Triple-negative breast cancer (TNBC) has been shown to have higher immunogenicity and enhanced CTL attraction. CD8A is a surface marker of CTLs and Granzyme B (GZMB) is a serine protease, secreted by CTLs to mediate apoptosis of the target cells. CXCL10 is a chemokine, which is known to attract CTLs into tumors in some types of cancers. However, the impact of CXLC10 expression and its role in the CTL attraction to TNBC remain unknown. Therefore, we hypothesized that Functional Hotness Score (FHS) that incorporate expressions of CXLC10 and CTL markers associates with anti-cancer immune cell infiltration and better survival of TNBC independent of mutation burden. Methods: Based on CD8A, GZMB and CXCL10 expression levels of the tumor, we established Functional Hotness Score (FHS) utilizing The Cancer Genome Atlas (TCGA) breast cancer cohort. As a validation cohort, METABRIC cohort was utilized. As neoadjuvant chemotherapy cohort, GSE20271, GSE20194, GSE22093, GSE23988, GSE25066 and GSE28844 from Gene Expression Omnibus (GEO) were utilized. As metastatic breast cancer dataset, GSE110590 from GEO was utilized. Gene Set Enrichment Analysis (GSEA) was conducted utilizing hallmark gene sets. Infiltrated immune cell fraction was estimated by CIBERSROT algorithm. Results: Out of 6 neoadjuvant cohorts, 5 cohorts showed higher CXCL10 expression levels in pathological completely response (pCR) tumors compared to residual tumors (RD). On the other hand, 1 and 3 cohorts showed higher expression levels of CD8A and GZMB in pCR tumors, respectively. Utilizing these 3 gene expression levels and hazard ratio of overall survival (OS) of the patients, prognostic Functional Hotness Score (FHS) was developed. As we expected, patients with High FHS tumors showed significantly better OS in TCGA cohort (p=0.006). This finding was validated in METABRIC cohort. In the metastatic breast cancer cohort, metastatic tumors showed lower FHS compared to primary tumors, such as liver (p=0.003) and lung metastases (p=0.031). Whereas, most of the brain metastases tissues showed similar level of FHS compared to primary tumors (p=0.068). Subgroup analysis based on subtype revealed that TNBC showed highest FHS among each subtype with wide variation. TNBC was the only subtype that high FHS showed better OS (p=0.001). High FHS TNBC significantly enriched immune response related gene sets, such as allograft rejection (p<0.001), complement (p<0.001), IFN-γ response (p<0.001), IL2/STAT5 signaling (p<0.001), inflammatory response (p=0.002), IL6/JAK/STAT3 signaling (p<0.001) and IFN-α response (p<0.001). High FHS TNBC showed higher anti-cancer immune cell infiltration, such as CD8+ T lymphocytes, activated memory CD4+ T lymphocytes, γδ T lymphocytes, memory B lymphocytes and M1 macrophage, and lower resting or pro-cancer immune cells such as resting CD4+ T lymphocytes, regulatory T lymphocytes, naïve B lymphocytes and M2 macrophages. There was no correlation between FSH and mutation count (r=0.443). To investigate if the prognostications of FHS are associated with mutation burden, we stratified TNBC into high and low mutation burden group. Interestingly, higher FHS showed better OS regardless mutation burden (high mutation; p=0.019, low mutation; p=0.049) in TNBC. Conclusion: In conclusion, TNBC with higher FHS based on the expression levels of CD8A, GZMB and CXCL10 showed improved prognosis with higher immune cell infiltration and immune reaction regardless mutation burden. Citation Format: Eriko Katsuta, Li Yan, Mateusz Opyrchal, Pawel Kalinski, Kazuaki Takabe. Functional hotness score of cytotoxic T-lymphocytes infiltration to predicts outcome of triple-negative breast cancer patients [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P3-07-07.