Abstract GS5-06: InteractPrint predicts clinically meaningful interactions between cancer epithelial cells and immune cells: Lessons from a single-cell breast cancer atlas

Abstract BACKGROUND: While immunotherapy has revolutionized the treatment of many solid tumors, the efficacy of immunotherapy regimens is comparatively lower in breast cancer. Immunotherapy efficacy is often negatively correlated with intratumor heterogeneity. Novel immunotherapy approaches in breast cancer should leverage how cancer epithelial cell heterogeneity affects immune cells in the tumor microenvironment. However, current definitions of cancer epithelial cell heterogeneity in breast cancer have limited resolution. Single cell RNA-seq (scRNA-seq) provides an unprecedented opportunity to further define cancer epithelial cell heterogeneity and identify how heterogeneity influences interactions with immune cells. METHODS: We generated a novel scRNA-seq dataset of 236,363 cells from 119 primary breast tumors biopsied from 88 patients taken from 8 publicly available datasets, currently the largest published scRNA-seq dataset in breast cancer. To define cancer epithelial cell heterogeneity, we performed unsupervised clustering and supervised clustering based on molecular subtype and expression of clinical target genes on all cancer epithelial cells. This identified 11 gene elements (GEs), which reflect key molecular features that vary between cancer epithelial cells. Receptor-ligand pairing analysis allowed us to determine how cells that highly express each GE interact with various immune cells. We developed InteractPrint, a score to predict the predominant tumor-interacting immune cells, based on the GE composition of an individual patient tumor. RESULTS: In our dataset, 17% of samples were HER2+, 41% were HR+, and 42% were TNBC. This dataset was statistically powered to characterize cancer epithelial cell heterogeneity. For each of the 11 GEs, we predicted interactions with immune cells. Experimentally, GEs with predicted NK cell interactions showed sensitivity to NK cell cytotoxicity. In a spatially resolved transcriptomics dataset, GEs with predicted T cell interactions demonstrated colocalization with CD8+ T cells, while those with limited predicted T cell interactions did not. To infer GE-immune interactions at the patient level (GEs define cell-level interactions), we developed InteractPrint. To validate InteractPrint, we assessed the accuracy of the T cell InteractPrint in predicting response to anti-PD-1 therapy. Across two trials and all breast cancer subtypes, T cell InteractPrint demonstrated significant improvement over PD-L1 in predicting response to anti-PD-1 therapy. In an scRNA-seq dataset of samples from patients treated with pembrolizumab, we observed AUC of 85% (p < 0.005) for T cell InteractPrint vs. 61% (p > 0.05) for PD-L1 in predicting response to anti-PD-1 therapy. In patients treated with paclitaxel + pembrolizumab in the I-SPY 2 trial, we observed AUC of 81% (p < 0.00001) for T cell InteractPrint versus 72% (p = 0.001) for PD-L1. CONCLUSIONS: Our results demonstrate considerable cancer epithelial cell heterogeneity across primary breast tumor samples and clinical subtypes. We defined this heterogeneity and leveraged it to predict immune cell interactions within a patient’s tumor. We developed T cell InteractPrint to capture heterogeneous interactions between cancer epithelial cells and CD8+ T cells. T cell InteractPrint is predictive of response to anti-PD-1 immune checkpoint inhibition at higher AUC than PD-L1. This provides a path forward for the interpretation of cancer epithelial cell heterogeneity in a clinically meaningful way. Citation Format: Lily Xu, Kaitlyn Saunders, Hildur Knutsdottir, Kenian Chen, Julia Maues, Christine Hodgdon, Evanthia T. Roussos Torres, Sangeetha Reddy, Lin Xu, Isaac Chan. InteractPrint predicts clinically meaningful interactions between cancer epithelial cells and immune cells: Lessons from a single-cell breast cancer atlas [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr GS5-06.