Background: The EndoPredict (EP) score is an RNA-based multigene test to predict the likelihood of distant recurrence in ER-positive (ER+), HER2-negative (HER2−) breast cancer (BC) patients treated with adjuvant endocrine therapy. Results from two large randomized phase III trials involving endocrine therapy only (n > 1700) demonstrated a prognostic power of the EP score beyond what can be achieved by combining the commonly used clinicopathological parameters (Filipits M, 2011). The performance of the EP in chemotherapy-treated patients has not been evaluated yet. Here, we analyzed the EndoPredict score in node-positive ER+/HER2− BC patients from the GEICAM-9906 trial, treated with adjuvant chemotherapy followed by hormonal therapy.

Methods: Patients included in this study participated in the GEICAM/9906 trial and were either treated with fluorouracil, epirubicin, and cyclophosphamide (FEC) or with FEC followed by weekly paclitaxel (FEC-P) (Martin M, 2008). ESR1 and ERBB2 gene expression were assessed by qRT-PCR in 800 formalin-fixed paraffin embedded (FFPE) tumor samples out of 1246 patients included in the GEICAM/9906 trial. The EndoPredict score (including eight prognostic genes) was successfully determined in 555 out of the 566 ER+/HER2− patients. Patients were assigned into two categories (high/low), according to the predefined EP cut-off value (Filipits M, 2011). The primary endpoint for the analysis was distant metastasis. Metastasis rates were estimated using the Kaplan–Meier method. Multivariate analysis was performed using Cox regression. Interaction between treatment effects and EP was tested as well.

Results: Twenty-five percent of patients (n = 141) were classified as EP-low-risk. Kaplan Meier analysis demonstrated that the metastasis-free survival (MFS) was 92% in the EP-low risk vs. 69% in the EP-high-risk group (absolute difference of 23%, HR 4.4 (2.3–8.4) p < 0.0001). Multivariate analysis showed that EP is an independent prognostic parameter after adjustment for age, grade, lymph node status and tumor size. EP was found to be prognostic in pre- (p = 0.0002, HR = 5.5 (2.2–13.6)) and postmenopausal (p = 0.0129, HR = 3.3 (1.3–8.2)) BC patients. There were not statistically significant differences in MFS between treatment arms (FEC vs. FEC-P) neither in the high nor in the low-risk groups. Interaction test between chemotherapy arm and EP score was not significant.

Conclusions: The results of this study shows that the EndoPredict score is an independent prognostic parameter in node-positive, ER+/HER2− BC patients treated with adjuvant chemotherapy followed by hormonal therapy. EndoPredict was not found to be predictive of weekly paclitaxel efficacy. Novel predictive biomarkers are needed to identify the small subset of patients with ER+/HER2− tumors that actually benefit from weekly paclitaxel-containing regimens.

Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P2-10-11.