Abstract 3129: Single-cell RNA sequencing of early-stage to refractory metastatic ER+ breast cancers reveal phenotypic states underlying resistance to multiple lines of therapy

Abstract Breast cancer is a leading cause of global cancer-related mortality. About 75% of breast tumors are estrogen receptor-positive (ER+). Nearly 30-50% of ER+ breast cancers eventually become refractory to multiple lines of therapy and progress to an uncurable metastatic stage despite positive 5-year results on primary endocrine therapy. To uncover the malignant phenotypes underlying the development of therapeutic resistance, we performed single-cell RNA sequencing (scRNA-seq) of 219,532 cells from 80 ER+ breast cancer patients. This included 28 solid tumors and 52 metastatic fluid samples (pleural effusions and ascites). We grouped samples in three groups to study the evolution of biological phenotypes based on the number of lines of therapy administered to a patient: 1. Early (non-metastatic), 2. Advanced (metastatic with less than 3 lines of therapy), and Refractory (metastatic with 3 or more lines of therapy). We used Pareto task inference analysis to model evolutionary task trade-offs across malignant cell populations from the early, advanced, and refractory tumors. Our model revealed 5 unique phenotypic states or “archetypes” that existed across ER+ cancer cells. A multitask learning model based on group lasso revealed the specialized biological tasks associated with each archetype, including 1. Metabolism and self-renewal, 2. EGF/STAT3 signaling, 3. Estrogen signaling, 4. Apoptosis, and 5. MYC activation with rapid proliferation. We found the estrogen signaling archetype was progressively lost in advanced and refractory tumors (P=7.2x10-110), while the MYC archetype was progressively gained in the advanced and refractory tumors (P=2.45x10-9). The EGF/STAT3 archetype was unique to refractory tumors (P= 1.42x10-37). We further investigated the evolution of resistant traits by modeling the pathway phenotypes in malignant cells against the number of lines of therapy received by the patients. Patients refractive to multiple lines of therapy progressively gained mTOR signaling pathway activity concurrent with loss of estrogen-signaling. Multiple pathways that crosstalk with the mTOR pathway were activated in the tumors receiving multiple lines of therapy, including IGFR, and EGF/ERBB signaling. In comparison to early and advanced tumors, refractory tumors also gained JNK/p38 signaling (P=0.006) and activated glycolysis (P=5x10-6). This study reveals how metastatic ER+ breast cancer cells acquire capabilities for resistance to multiple lines of therapy, thus serving as a unique resource for understanding refractory ER+ breast cancer biology. Citation Format: Aritro Nath, Patrick Cosgrove, Jason Griffiths, Jeffrey Chang, Adam Cohen, Andrea Bild. Single-cell RNA sequencing of early-stage to refractory metastatic ER+ breast cancers reveal phenotypic states underlying resistance to multiple lines of therapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3129.