Abstract 1350: <i>ESR1</i> expression as a predictive biomarker for immune checkpoint inhibitor efficacy in estrogen receptor-positive, HER2-negative breast cancer

Abstract Introduction: Immune checkpoint inhibitors (ICIs) hold promise for preventing late recurrence, a major cause of mortality in estrogen receptor-positive/HER2-negative (ER+/HER2-) subtype, which constitutes the majority of breast cancer (BC) cases. The response to ICIs is worse in this subtype compared to triple-negative breast cancer (TNBC). Recent clinical trials, CheckMate 7FL and KEYNOTE-756, have uniformly demonstrated that ER+/HER2- BC with less Estrogen Receptor (ER)-α staining in cancer cells shows a better response to ICIs. Considering the relatively precise quantification achievable through transcriptomic analysis, we hypothesized that ER-α gene (ESR1) expression level is a negative predictive biomarker of ER+/HER2- BC response to ICIs. Methods: Total of 4138 BC patients from ten cohorts (GSE271080, I-SPY2 [GSE173839], TCGA, METABRIC, SCAN-B [GSE96058], GSE246613, GSE255107/GSE248288, GSE161529, GSE167036) were analyzed. High expression of ESR1 was defined as the top two-thirds of expression levels for each gene. Results: In the ER+/HER2- subtype, high ESR1 expression was associated with significantly less intratumoral-Tumor Infiltrating Lymphocytes in both TCGA and METABRIC cohorts (p&amp;lt;0.001). High ESR1 expression was also associated with less infiltrations of CD8+ T cells, dendritic cells (DCs), cancer-associated fibroblasts (CAFs), a lower CD8/Treg ratio, and a lower cytolytic activity consistently in TCGA, METABRIC, SCAN-B cohorts (all p&amp;lt;0.01). Further, the low ESR1 group significantly enriched multiple immune-related gene sets, including IL2 signaling, TNF-α signaling, and interferon-γ response. In the I-SPY2 trial, high ESR1 expression correlated with lower pathological complete response rates after neoadjuvant immunotherapy (n=50, p&amp;lt;0.001). The predictive accuracy of ESR1 expression (AUC=0.88) exceeded that of PDL1 (AUC=0.60) in HR+/HER2- BC. Single-cell analysis (GSE246613) revealed negative associations between ICI response and ESR1 expression in cancer/normal epithelial cells, endothelial cells, and plasmablasts, as well as activation of estrogen signaling in these cell types. Notably, CAFs showed a positive association between ESR1 expression, estrogen signaling activation, and ICI response. Conclusion: ESR1 expression is a promising predictive biomarker for ICI response in ER+/HER2- BC. Additionally, CAF infiltration and estrogen signaling activation in CAFs may enhance ICI response, highlighting their potential as therapeutic targets. Citation Format: Jun Arima, Kohei Chida, Rongrong Wu, Kohei Taniguchi, Mitsuhiko Iwamoto, Takashi Ishikawa, Kenichi Hakamada, Sang-Woong Lee, Kazuaki Takabe. ESR1 expression as a predictive biomarker for immune checkpoint inhibitor efficacy in estrogen receptor-positive, HER2-negative breast cancer [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 1350.