Abstract PS2-08-01: A predictive gene signature for benefit from dose-dense adjuvant chemotherapy in patients with high-risk early breast cancer: results from the PANTHER phase III trial

Abstract Introduction: Dose dense adjuvant chemotherapy (DD-ACT) for high-risk breast cancer (BC) improves outcomes compared to standard interval treatment (SD-ACT) regardless of any clinicopathologic factor. SET2,3 is the only gene signature that predicted benefit from DD-ACT in the CALGB C9741 trial for patients with ER-positive BC with low endocrine activity, compared though with a suboptimal control of paclitaxel every three weeks. Based on our previous results, we hypothesized that proliferation and immune gene expression may identify patients benefiting from treatment escalation. Material and Methods: PANTHER (NCT00798070) is an international phase III trial which compared adjuvant sequential epirubicin/cyclophosphamide and docetaxel administered either every 2 (DD-ACT) or every 3 weeks (SD-ACT), with tailored dosing at the dose-dense schedule according to hematologic toxicity for patients with resected node positive or high risk node negative BC. Bulk RNA was extracted from tumor-rich FFPE surgical biospecimens from patients enrolled at Swedish sites and sequenced in the NovaSeq X Plus system. Mean expression was quantified for three predefined gene modules: a mitosis kinase score (MKS), an immune kinase score (IKS), and an immune activation score (IAS). MKS was independently combined with each of the two immune-related scores (IKS and IAS) to stratify patients into low and high activity groups. Primary endpoint of the trial was BC recurrence-free survival (BCRFS) and secondary endpoints were event-free (EFS), distant disease-free (DDFS) and overall survival (OS). The median follow-up time was 10.2 years. Results: RNA sequencing was conducted on 506 FFPE tumor samples, of which 481 were retained for downstream analysis following quality control. Multivariable analysis, adjusted for tumor size, lymph node status, grade, and BC subtype, identified a subgroup of patients (n=157, 32.6% of the cohort) who derived greater benefit from DD-ACT compared to SD-ACT (HR=0.29, 95% CI 0.14-0.59, p=0.001), while the remaining patients (n=322) did not benefit from DD-ACT (HR=1.22, 95% CI 0.74-2.01, p=0.443; pinteraction = 0.02). Results were consistent for all secondary endpoints. This DD-ACT sensitive group was characterized by tumors with low expression of both mitotic and immune kinase signaling genes, or by tumors with high mitotic activity and a cold immune microenvironment. Conclusions: We describe a predictive gene signature that identifies a subgroup of patients who derive greater benefit from DD-ACT compared to SD-ACT across all survival endpoints and BC subtypes, independent of clinicopathologic factors, when compared with an optimal control of docetaxel every three weeks. Finally, we aim to validate our findings in independent patient cohorts. Citation Format: D. Salgkamis, M. Sarafidis, E. Sifakis, I. Zerdes, S. Agartz, J. Hartman, M. Hellström, T. Foukakis, J. Bergh, A. Matikas. A predictive gene signature for benefit from dose-dense adjuvant chemotherapy in patients with high-risk early breast cancer: results from the PANTHER phase III trial [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 PS2-08-01.