Abstract P6-10-04: Landscape of immune-cell signatures in early high-risk breast cancer (BC) reveals clinically-relevant enrichment of immune subpopulations

Abstract

Abstract Background: Immunosurveillance suppression, evasion, or avoidance, has emerged as a key targetable hallmark of cancer, driven by e.g. checkpoint expression, T-cell exhaustion, and immunosuppressive tumor microenvironment (TME). Many of these processes generate defined combinations of immune-cell infiltrates at the tumor site, which can be detected by immunohistochemistry (IHC), CyTOF, or more recently can be inferred from gene-expression deconvolution. While significant work has been done to study gene signatures in the TME, the clinical relevance of such immune-cell gene signature on therapy has not been studied to a great extent. We investigate the hypothesis that the individual patterns of immune-cell signatures determine the clinical behavior of breast cancer (BC), in particular response to neoadjuvant chemotherapy. Methods: We performed a retrospective-prospective analysis of a subset of the GeparSepto study (NCT01583426) in which women with primary invasive BC were randomized to either nab-paclitaxel or solvent-based paclitaxel followed by EC. This study was limited to 279 HER2- patients with sufficient quality sample remaining to perform whole-transcriptome RNAseq (~200 × 106 reads per tumor). Immune activity in the TME was inferred by comparing expression of 23 immune-cell-specific gene signatures derived by Bindea et al. (Immunity, 2013) to those from a background population of 1467 similarly-profiled unselected tumor samples from the NantOmics database. Results: Within this cohort the most predominant high immune-cell signatures were for natural killer (NK) cells (71%), and regulatory T-cells (70%). Stimulatory T-cell signatures were high in approximately half of the population including Th2 (53%), effector-memory (53%), follicular helper (51%), Th1 (41%), and Gamma-delta T-cells (39%). While cytotoxic CD8+ T-cell signature was high in only 19.0% of patients, the signature for the CD56dim cytolytic subset of NK cells was high in 48.0% of patients. The most infrequently detected gene signatures were for innate response cell-types: Mast cells (7.5%), Macrophages (10.8%), immature dendritic cells (11.5%) and neutrophils (11.8%). Of the 23 immune signatures, 17 were significantly differentially activated in TNBC compared to hormone-receptor positive (HR+) patients (p<0.05 respectively): 8/10 signatures more active in TNBC are associated with adaptive immune response (e.g. T-cell and B-cell signatures), whereas 5/7 signatures associated with HR+ are related to innate immune response (e.g. eosinophils and dendritic cells). Select adaptive immune signatures were associated with aggressive tumors: Elevated NK CD56dim, Th1, and activated dendritic cell (aDC), signatures were associated with grade 3 tumors as well as with elevated levels of Ki67 (p<0.0001 respectively). The most predictive TME signature for paclitaxel-based therapy was T follicular helper cells (TFH) with DFS and OS hazard ratio of 0.62 (95% CI: 0.47-0.81; p=0.0004) and 0.55 (95% CI: 0.39-0.77; p=0.0005) respectively, as well as a 1.63 (95% CI: 1.12-2.36; p=0.0107) odds-ratio for achieving pCR. Conclusion: Whole-transcriptome sequencing in breast cancer FFPE core biopsies from clinical cohorts can be used to identify immune-cell signatures. Specifically, adaptive immunity through NK rather than T-cell response appears prevalent in high-risk TNBC. The patterns of these immune signatures, in particular the presence of T follicular helper cells, reflect the clinical behavior of breast cancer and might be used to identify tumors with an increased response rate to neoadjuvant chemotherapy. Citation Format: Christopher Szeto, Carsten Denkert, Peter A Fasching, Stephen Benz, Karsten E Weber, Jan Budczies, Andreas Schneeweiss, Elmar Stickeler, Sabine Schmatloch, Christian Jackisch, Thomas Karn, Hans P Sinn, Mathias Warm, Marion van Mackelenbergh, Shahrooz Rabizadeh, Christian Schem, Ernst Heinmöller, Volkmar Müller, Frederik Marmé, Patrick Soon-Shiong, Valentina Nekljudova, Sibylle Loibl, Michael Untch. Landscape of immune-cell signatures in early high-risk breast cancer (BC) reveals clinically-relevant enrichment of immune subpopulations [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 P6-10-04.