Abstract PS2-07-16: Clinicogenomic characterization and ctDNA detection of ESR1 fusion positive metastatic breast cancer

Abstract Background: ESR1 fusions are an emerging mechanism of acquired resistance to endocrine therapy (ET) in hormone-receptor positive (HR+) metastatic breast cancer (MBC). These fusions drive estrogen-independence and eliminate binding sites for all approved ER-targeting agents, posing a significant clinical challenge. Non-invasive detection of ESR1 fusions is limited by diverse gene partners and variable DNA breakpoints, which complicates assembling meaningfully large cohorts. By integrating high quality RNA-sequencing (RNA-seq), a sensitive, methylation-informed circulating tumor DNA (ctDNA), and robust clinical data, we set out to comprehensively characterize ESR1 fusion-positive disease. Methods: Patients with HR+ MBC harboring ESR1 fusions were retrospectively identified in a single academic center using three approaches: (1) whole transcriptome RNA-seq on fresh-frozen biopsies, (2) transcriptome profiling on archival tissue as part of standard clinical testing, and (3) a next-generation methylation-informed ctDNA assay. For fusion-positive cases, comprehensive clinical and outcome data were collected. For cases with RNA-seq, PAM50 assignments and Gene Set Enrichment Analyses (GSEA) were performed. Results: We identified 63 patients with an ESR1 fusion, with a median age at metastatic diagnosis of 56 years (range 30 - 83). Fusions were detected in 4.5% (9 of 202) of patients with luminal metastatic disease via tumor RNA-seq, in 9.9% (45 of 454) of patients with HR+ MBC via ctDNA testing, and in 11 additional cases as part of routine clinical testing; with 2 cases confirmed by both RNA-seq and ctDNA. In patients with longitudinal sampling, 76% (32 of 42) had a prior fusion-negative test, suggesting fusions are predominately acquired events. Fusions were detected a median of 2.6 years (range 0 - 10.9) from metastatic diagnosis and emerged after a median of 4 prior lines of therapy (range 1 - 14). All patients received ET prior to fusion detection (58% SERM, 83% AI, 60% SERD). Post-fusion detection, the longest interval of disease control with combination ET was 10.1 months whereas single agent ET was only 2.8 months. Four patients received an oral selective estrogen receptor degrader (SERD) after fusion detection-with the longest time to next treatment being 3.9 months. Genomically, the most recurrent 3’ partners were: AKAP12 (in 7 cases), CCDC170, PLEKHG1, VTA1 (each in 5 cases), and IMPG1 (in 4 cases), with the majority of partners being promiscuous or n-of-1 events. All ESR1 fusion positive tumors with RNA-seq were PAM50 classified as Luminal B, and GSEA of ESR1-fusion positive cases versus other Luminal B tumors revealed highly significant enrichment in proliferation pathways-including G2M checkpoint, E2F and MYC targets (FDR q-val < 0.25). The median overall survival after fusion detection was 5.4 months and 44 of 63 (70%) patients had passed at the time of this analysis; with prospective analyses ongoing. Additional genomic correlates, detailed clinical outcomes, and clinical case vignettes will be presented. Conclusions: This comprehensive clinicogenomic study of ESR1 fusion-positive MBC further establishes ER fusions as relatively common drivers of ET resistance. Critically, these fusions require more sensitive detection assays and serial testing-making prior estimates of prevalence likely conservative. ESR1 fusions appear acquired in ET resistant disease, are a marker of poor prognosis with a hyperproliferative biology, and confer broad resistance to subsequent ET-including oral SERDs. As next-generation SERDs more effectively target ESR1 point mutations, we hypothesize ESR1 fusions will become an increasingly significant clinical problem. Collectively, these findings highlight an urgent need to characterize ER fusion-positive MBC to establish more effective therapeutic options. Citation Format: K. Santos, A. Lebrón-Torres, C. Weipert, S. Morganti, K. Smith, A. Patel, C. Snow, M. Hughes, S. Ravikumar, G. Suggs, J. Bsat, S. Oesterreich, A. V. Lee, E. P. Winer, H. J. Burstein, D. L. Abravanel, H. A. Parsons, S. M. Tolaney, R. M. Jeselsohn, N. U. Lin4, N. Priedigkeit. Clinicogenomic characterization and ctDNA detection of ESR1 fusion positive metastatic breast cancer [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS2-07-16.