Abstract PS5-02-23: Aretha trial: Safety run-in results of eribulin + evexomostat in metastatic triple-negative breast cancer with metabolic dysfunction

Abstract Background: The prognosis for metastatic triple negative breast cancer (metTNBC) remains poor and novel therapeutic strategies are urgently needed. Insulin resistance & obesity are independent predictors of worse survival in metTNBC. Moreover, Homeostatic Model Assessment for Insulin Resistance (HOMA-IR) and related metabolic markers often worsen significantly during chemotherapy. Methionine aminopeptidase 2 (MetAP2/p67) is an enzyme involved in angiogenesis and regulation of metabolic hormones that is overexpressed in breast cancer. In a phase I study of evexomostat, a MetAP2 inhibitor, 80% of patients with advanced solid tumors had stable disease and improvements in fasting insulin, adiponectin, and VEGF-C. The primary objective of this phase 2 study is to assess the efficacy of evexomostat on insulin resistance in combination with standard of care eribulin in metTNBC. Here, we report results from the safety run-in phase and exploratory biomarker changes. Methods: This is a multicenter phase 2 randomized controlled trial of evexomostat + eribulin (NCT05570253). Key eligibility include metTNBC, baseline HbA1c >5.5% and/or BMI ≥30, and ≤2 prior treatment lines in the metastatic setting. In the safety run-in period, 15 patients received evexomostat 49 mg/m2 Q2 weeks + eribulin 1.4 mg/m2 on D1&8 of a 21-day cycle. The safety threshold was defined as ≥4/15 patients experiencing serious adverse events (SAEs) not resolving within 14 days. The primary endpoint of the randomized phase is change in HOMA-IR at 12 weeks. Secondary endpoints include objective response rate (ORR) and safety and tolerability. Patients were considered evaluable if they received ≥1 dose of evexomostat. Change in HOMA-IR and other markers was analyzed using Wilcoxon signed rank test. Results: Among the 15 patients in the safety run-in phase, 2/15 experienced SAEs, both of which were grade 3 anemia. The most common any-grade AE was anemia (9/15, 60%). AEs of grade ≥3 were reported in 11 patients; the most frequent grade ≥3 AE was decreased neutrophil count. Accordingly, the safety stopping rule was not triggered and the trial advanced to the randomization phase. Exploratory analyses from baseline to week 12 demonstrated significant changes in LDL cholesterol, adiponectin, leptin, resistin, VEGF-D, and sFRP-5 (p=0.016, 0.009, 0.021, 0.016, 0.027, and 0.012 respectively; see Table). There was no increase in HOMA-IR (p>0.9). Among 12 patients evaluable for response; ORR was 17% and clinical benefit rate was 67%. Conclusions: Evexomostat + eribulin safety was confirmed and the trial advanced to the randomization phase including a placebo-controlled arm with eribulin. Preliminary biomarker results indicate that insulin resistance was unchanged, and favorable changes were seen in other metabolic markers. Citation Format: S. Shen, V. Solomon, D. Williams, C. White, Y. Chen, M. Meem, I. Warren, P. Drullinsky, S. Schweber, R. Wang, M. Gorsky, A. Gucalp, J. Shen, R. Sanford, A. Widman, M. Robson, L. Norton, T. Traina, D. Graham, R. Mahtani, N. M. Iyengar. Aretha trial: Safety run-in results of eribulin + evexomostat in metastatic triple-negative breast cancer with metabolic dysfunction [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 PS5-02-23.