19P - Single-cell transcriptomic and cell-cell communication profiles in breast cancer responders to chemotherapy or chemo-immunotherapy

Background

For most patients (pts) with early-stage aggressive breast cancers (BC), neoadjuvant therapy (NAT) with chemotherapy (CT) with/without HER2 or immune checkpoint blockade (ICB) is standard. However, around half of these pts do not respond and face higher recurrence risks. This study uses single-cell RNA sequencing (scRNA-seq) to explore cellular and molecular heterogeneity and potential mechanisms driving response/resistance to NAT.

Methods

Tissue samples were prospectively collected pre- and post-NAT, yielding high-quality scRNA-seq data from 30 pts (18 responders), totaling 286,346 cells. The cohort included 4 Luminal B-like (LumB), 5 HER2-positive (HER2+), and 21 triple-negative breast cancer (TNBC) pts (14 treated with CT, 7 with CT-ICB). Cells were processed using the 10x Genomics Chromium Single Cell 5’ platform for gene expression and T-cell/B-cell repertoire analysis.

Results

BC subtypes showed distinct tumor cell origins: TNBC had a basal/luminal progenitor origin, while LumB and HER2+ had a luminal hormone-responsive origin. Also, LumB had fewer immune cells at baseline compared to TNBC and HER2+, particularly CXCL13+ and effector T cells. Responders to NAT had higher proportions of G2/M proliferating T cells (p<0.01) pre-NAT and an increase in the CD8 effector/Treg ratio post-NAT (p<0.001), while non-responders had higher levels of T cells in the G1 phase. A detailed analysis of the T cell compartment revealed that response to CT in TNBC was associated with pre-NAT clonotype expansion of CXCL13+ CD8+ T cell and cell-cell communications (CCCs) between tumor cells and CXCL13+ and effector cells. Interestingly, response to CT-ICB was characterized by specific pre-NAT CCCs compared to the response to CT, primarily among CXCL13+ CD4+, CXCL13+ CD8+ T cells, and Tregs via CXCL13/CXCR5, OX40/OX40L, and CCL3-5/CCR5 ligand-receptor pairs. These findings suggest that immune cell interactions forming tertiary lymphoid structures might serve as specific biomarkers for response to ICB.

Conclusions

Our findings provide insight into cellular heterogeneity and dynamics across BC subtypes, revealing potential predictive biomarkers for NAT response in breast cancer.

Editorial acknowledgement

Clinical trial identification

Legal entity responsible for the study

The authors.

Funding

Institut Jules Bordet, Breast Cancer Research Foundation, Le Fonds de la Recherche Scientifique. This study was performed during the author's ESMO Translational Research Fellowship.

Disclosure

L. Buisseret: research funding to the Institution from AstraZeneca/MedImmune, advisory role for Domain Therapeutics, iTeos Therapeutics, AstraZeneca, salary partly covered by the Fondation contre le Cancer (Belgium), member of the IMMUcan Consortium (EORTC), travel grants from Gilead, AstraZeneca, Roche; M. Ignatiadis: consultancy roles for Seattle Genetics, Daiichi Sankyo, AstraZeneca, Menarini/Stemline, Gilead Sciences, Rejuveron Senescence Therapeutics, and Novartis, grant/research support (to institution) from Pfizer, Roche, Inivata Inc., and Natera Inc., and uncompensated roles for EORTC (EORTC Board of Directors 2018-2021, Chair EORTC Breast Cancer Group 2021-2024). C. Sotiriou: Royalties for Signatur Biosciences and advisory role for Exact Sciences, Gilead. All other authors have declared no conflicts of interest.