Clinical validation of the EndoPredict test in node-positive, chemotherapy-treated ER+/HER2− breast cancer patients: results from the GEICAM 9906 trial

Abstract

Introduction: EndoPredict (EP) is an RNA-based multigene test that predicts the likelihood of distant recurrence in patients with estrogen receptor-positive (ER+), human epidermal growth factor receptor 2–negative (HER2−) breast cancer (BC) who are being treated with adjuvant endocrine therapy. Herein we report the prospective-retrospective clinical validation of EP in the node-positive, chemotherapy-treated, ER+/HER2− BC patients in the GEICAM 9906 trial. Methods: The patients (N = 1,246) were treated either with six cycles of fluorouracil, epirubicin and cyclophosphamide (FEC) or with four cycles of FEC followed by eight weekly courses of paclitaxel (FEC-P), as well as with endocrine therapy if they had hormone receptor–positive disease. The patients were assigned to EP risk categories (low or high) according to prespecified cutoff levels. The primary endpoint in the clinical validation of EP was distant metastasis-free survival (MFS). Metastasis rates were estimated using the Kaplan-Meier method, and multivariate analysis was performed using Cox regression. Results: The molecular EP score and the combined molecular and clinical EPclin score were successfully determined in 555 ER+/HER2− tumors from the 800 available samples in the GEICAM 9906 trial. On the basis of the EP, 25% of patients (n = 141) were classified as low risk. MFS was 93% in the low-risk group and 70% in the high-risk group (absolute risk reduction = 23%, hazard ratio (HR) = 4.8, 95% confidence interval (CI) = 2.5 to 9.5; P < 0.0001). Multivariate analysis showed that, in this ER+/HER2− cohort, EP results are an independent prognostic parameter after adjustment for age, grade, lymph node status, tumor size, treatment arm, ER and progesterone receptor (PR) status and proliferation index (Ki67). Using the predefined EPclin score, 13% of patients (n = 74) were assigned to the low-risk group, who had excellent outcomes and no distant recurrence events (absolute risk reduction vs high-risk group = 28%; P < 0.0001). Furthermore, EP was prognostic in premenopausal patients (HR = 6.7, 95% CI = 2.4 to 18.3; P = 0.0002) and postmenopausal patients (HR = 3.3, 95% CI = 1.3 to 8.5; P = 0.0109). There were no statistically significant differences in MFS between treatment arms (FEC vs FEC-P) in either the high- or low-risk groups. The interaction test results between the chemotherapy arm and the EP score were not significant. Conclusions: EP is an independent prognostic parameter in node-positive, ER+/HER2− BC patients treated with adjuvant chemotherapy followed by hormone therapy. EP did not predict a greater efficacy of FEC-P compared to FEC alone.