Abstract Background: Endocrine therapy resistance (ETR) is a frequent challenge in metastatic ER+ HER2- breast cancer (MBC) and is driven by different molecular mechanisms. Since the first observation of a transcriptionally active ESR1 fusion involving exons 1-6 (e6) of ESR1 fused to YAP1 (Li et al., Cell Reports, 2013), e6 ESR1 gene fusions are increasingly recognized as a cause of ETR. However, subsequent studies have found that 3’ fusion partners are highly heterogeneous, leading to questions regarding clinical actionability. Here, we report an ESR1 fusion classification system with real world evidence (RWE) support. Methods: ESR1 fusions were detected in circulating tumor DNA (ctDNA) by sequencing the intron between exons 6 and 7 to identify fusion gene breakpoints. Based on the study by Gou et al. (Cancer Research, 2021), a 4-group classification system was devised: Group (G)1: known active (lab tested); G2: likely active (not lab tested but transcription factor TF or transcriptional co-activator CoA partner and at least 2 instances in GuardantINFORM™ RWE database of 1,276 cases); G3: possibly active (not TF or CoA partner but at least 3 recurrent cases); and G4: inactive fusion (all other fusions i.e., ESR1 fusions not involving exon 6, out of frame, chromosome 6 partner in frame, exon to exon, or ESR1 to ESR1). Propensity matching was used to account for different levels of ctDNA, age, and lines of therapy. Results: The most prevalent partners by fusion group were YAP1 (n = 21), ARNT2 (n = 9), LPP (n = 2) for G1; SS18 (n = 4, previously reported sarcoma fusion partner), TAZ (n = 4), ESRRG (n = 4), NFAT5 (n = 4) for G2; PXN (n = 9), TFG (n = 5, previously reported breast cancer fusion partner), PROSER1 (n = 4) for G3; and in frame examples of SYNE1 (n = 62), PLEKHG1 (n = 57), SNAP91 (n = 29), CCDC170 (n = 44, latter case previously reported to be functionally inactive Lei et al., Cell Reports, 2018) for G4. Since G1, 2 and 3 had similar adverse prognostic effects, outcome analysis involved comparisons to G5, defined as “no fusion detected” and fusion groups 1, 2 and 3 in combination. In a multivariate analysis, patients with G1-3 ESR1 fusions receiving ET monotherapy as second-line treatment had a rwOS HR of 3.0 (95% CI 1.7-5.5, P = 0.0002) when compared to G5 (no fusion). Supporting the inactive classification, G4 had an indistinguishable clinical course from G5. Patients with G1-3 ESR1 fusions receiving heterogeneous second-line ET combinations (including chemotherapy and antibody drug conjugates) also had an increased risk of death compared to G5 based on a multivariate analysis (HR 2.0, 95% CI 1.2-3.3, P = 0.01). Conclusions: Despite low prevalence, estimated at 2.1%, in ER+ HER2- MBC cases post AI therapy, G1-3 ESR1 fusions were associated with an increased risk of early death when managed with second-line ET monotherapy or heterogeneous second-line ET combinations. These results highlight the need for novel therapeutic options for patients with G1-3 ESR1 fusions. Citation Format: S. A. Wander, D. G. Stover, J. T. Lei, N. Zhang, A. Hardin, M. J. Ellis. Prevalence and clinical significance of exon 6 ESR1 gene fusions in advanced breast cancer after disease progression on aromatase inhibitors 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 PS2-08-17.
Wander et al. (Tue,) studied this question.