BRAF/MEK Pathway is Associated With Breast Cancer in ER-dependent Mode and Improves ER Status-based Cancer Recurrence Prediction

Abstract

BackgroundAberrant BRAF/MEK signaling was found in nearly 50% of human malignancies and proved to play a critical role in the tumorigenesis of multiple cancers. However, this pathway was relatively seldom studied in breast cancer, and the role of this pathway in the pathogenesis of breast cancer is still controversial. Materials and MethodsBreast cancer gene expression data from The Cancer Genome Atlas (TCGA) and 43 Affymetrix microarray datasets were analyzed. The BRAF/MEK pathway activity was presented with phosphorylated ERK level (for the TCGA dataset) or computed by a gene signature-based algorithm (for Affymetrix datasets). Aberrant activation of BRAF/MEK pathway in breast cancer was assessed in matched normal/tumor tissues. The associations of the BRAF/MEK pathway with clinical outcome in patients with breast cancer were analyzed by logistic regression, Cox regression, and Kaplan-Meier methods. ResultsDown-regulation of the BRAF/MEK pathway was observed in atypical ductal hyperplasia, ductal carcinoma in situ, and invasive breast cancers, with the exception of human epidermal growth factor receptor 2-positive and triple-negative breast cancers. Higher BRAF/MEK pathway activities were associated with better survival in estrogen receptor (ER)-positive (overall hazard ratio [HR], 0.85; P = 5.47E−5; n = 3128) or progesterone receptor-positive (overall HR, 0.85; P = 4.19E−3; n = 1537) breast cancers, but with worse survival in ER-negative (overall HR, 1.13; P = .01; n = 1107) or progesterone receptor-negative (overall HR, 1.13; P = .01; n = 1219) breast cancers. Combination with BRAF/MEK pathway activities could improve ER status-based recurrence prediction for breast cancer. ConclusionBRAF/MEK pathway was associated with the recurrence risk of breast cancer in an ER status-dependent mode. Combination with BRAF/MEK pathway activities could improve the ER status-based recurrence prediction in breast cancer.