Abstract 1256: BCL11B predetermines an immature drug-resistant cell state in breast cancer that is vulnerable to TNF-α treatment

Abstract Background: The development of chemoresistance in many cancer patients poses a significant clinical challenge, while the underlying causes for heterogeneous drug responses remain largely elusive. Increasing evidence suggests that tumors exhibit extensive heterogeneity and harbor cancer subpopulations in less differentiated states that are linked to treatment resistance. Therefore, dissecting immature cancer cell states pre-existing in tumors is essential for identifying biomarkers associated with chemoresistance and establishing targeted therapeutic strategies, particularly in breast cancer, where specific biomarkers are largely lacking. Methods: We utilized CytoTRACE, a robust computational framework for predicting the cellular differentiation status from single-cell RNA sequencing data, to unveil the immature cancer cell state and its associated markers in single-cell atlases of human breast cancer. The association between the identified immature cell state and patient survival was determined by inferring the proportions of epithelial cancer cells expressing the marker gene in bulk gene expression datasets. The MMTV-PyMT mouse breast cancer model was used to experimentally investigate the role of the immature cancer cell state in conferring chemoresistance. Single-cell profiling was conducted on tumor samples before and after treatment to examine the inherent characteristics of immature cancer cells and their response to drug pressure. The results obtained from mouse tumors were subsequently confirmed using human breast cancer models. Results and Conclusion: We revealed that BCL11B, a zinc finger transcription factor, defines a distinct immature cancer cell state in human breast cancer that displays basal cell features and demonstrates a heterogeneous expression pattern across breast tumors. Breast cancer patients with higher levels of BCL11B+ cancer cells face a significantly increased risk of relapse following chemotherapy. This pre-existing tumor subpopulation exhibits intrinsic fitness advantages and demonstrates a remarkable ability to evolve into a drug-resistant state through various BCL11B-dependent pre-existing and transcriptional adaptation mechanisms. Notably, TNF-α was pinpointed as a natural inhibitor of BCL11B, and treatment with TNF-α significantly increases chemotherapy efficacy by inducing the exit of the drug-resistant cell state. In conclusion, our findings identify BCL11B as a valuable biomarker for predicting chemotherapy resistance and suggest TNF-α combination treatment for breast cancer patients at high risk of developing chemoresistance. Citation Format: Zhen Qi,1 Gunsagar S. Gulati,2 Angera H. Kuo,1 Shaheen S. Sikandar,3 William Hai Dang Ho,1 Dalong Qian,1 Frederick M. Dirbas,1 Aaron M. Newman,1 Shang Cai,4 Michael F. Clarke1. BCL11B predetermines an immature drug-resistant cell state in breast cancer that is vulnerable to TNF-α treatment [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1256.