Abstract 2279: DNA methylation-based classifier predicts SERD benefit in ESR1 wild-type HR+/HER2- breast cancer

Abstract Background: Resistance to aromatase inhibitors (AIs) in hormone receptor-positive (HR+) metastatic breast cancer (mBC) is inevitable and frequently driven by ESR1 mutations. Selective estrogen receptor degraders (SERDs) target estrogen receptors for degradation and provide clinical benefit in this setting. ESR1 mutations are established biomarkers guiding transition from AI to SERD-based endocrine therapy. However, many patients lack detectable ESR1 mutations, and no validated biomarker exists to guide SERD use in ESR1 wild-type (ESR1-wt) patients. We investigated whether a blood-based, DNA methylation-derived hormone receptor (mHR) activity score could predict SERD benefit in ESR1-wt patients. Methods: An mHR activity score was developed using a logistic regression model integrating over 400 most informative regions from genome-wide methylation capture of Guardant360 Liquid test. mHR-high/-low thresholds were defined using 2,000 clonal ESR1-mutant mBC samples. Validation was performed in a real-world ESR1-wt cohort (N=765) with post-AI progression plasma samples from the GuardantINFORM database. Clinical outcomes defined by time to treatment discontinuation (TTD) of SERD or chemotherapy were evaluated across mHR-high/-low groups using Kaplan-Meier methods and multivariable Cox models, adjusting for tumor fraction, age, and treatment type. Results: Across 12,000 breast cancer samples tested with Guardant360 Liquid, the mHR activity score was positively correlated with ESR1 mutation status and higher in tumors with clonal versus subclonal ESR1 mutations. In the ESR1-wt validation cohort, mHR-high patients showed significantly longer median TTD with SERD treatment (Fulvestrant mono or combination) compared to mHR-low patients (adjusted hazard ratio: 0.44, 95%CI: 0.21-0.89; p = 0.02; median TTD 107 vs. 89 days). Moreover, mHR status predicted relative benefit from SERD versus chemotherapy: among mHR-high patients, SERD treatment yielded significantly longer TTD than chemotherapy (adjusted hazard ratio: 0.41; 95%CI: 0.21-0.81; p = 0.01; median TTD 107 vs. 87 days), while mHR-low patients exhibited a trend towards shorter TTD on SERD versus chemotherapy (adjusted hazard ratio: 1.83; 95% CI: 0.87-3.87; p = 0.11; median TTD 89 vs. 108 days). Conclusions: A blood-based DNA methylation-derived mHR score captures hormone receptor-driven epigenetic activity and predicts benefit from SERD therapy in ESR1-wt HR+ mBC. This classifier enables real-time, noninvasive stratification of ESR1-wt patients, identifying a molecularly defined subset more likely to derive benefit from SERD-based therapy. These findings support the potential clinical utility of methylation-based biomarkers to extend precision endocrine therapy beyond ESR1 mutation profiling. Citation Format: Shile Zhang, Nicole Zhang, Vishnu Ramani, Tingting Jiang, Marisa Juntilla, Amar Das, Martina Lefterova, Justin Odegaard, Darya Chudova, Matthew Ellis, . DNA methylation-based classifier predicts SERD benefit in ESR1 wild-type HR+/HER2- breast cancer abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 2279.