Extensive analysis of signaling pathway molecules in breast cancer: association with clinicopathological characteristics.

Background: The aim of this study was to extensively analyze the signaling pathway molecules in breast cancer and to explore candidate biomarkers for clinicopathological relevance.
Methods: We assessed the expression of key factors in cell signaling, namely p-AKT, cyclin D1, P27, p-p70S6 K, p-4EBP1, and p-MAPK/ERK, within 338 invasive breast cancer patients. These factors were immunohistochemically examined in tumor tissues and assessed by staining score. Staining scores were analyzed by a clustering method to devise a new classification based on pathway activity. We investigated the relationships among staining scores, the clustering classification, and patient characteristics.
Results: The proportion of patients displaying high expression levels were as follows: p-AKT, 75%; cyclin D1, 12%; P27, 53%; p-p70S6 K, 37%; p-4EBP1, 19%; and p-MAPK/ERK, 3%. Patients were classified into two groups on the basis of staining scores. Group 1 (39%) included more positive cases for p-4EBP1, p-MAPK/ERK, and p-p70S6 K and fewer positive cases for P27 and cyclin D1 than Group 2 (61%). The clustering classification was significantly related to subgrouping by hormone receptor and HER2 (P < 0.001), nuclear grade (P < 0.001) and histological subtype (P = 0.034). A strong positive correlation was identified between p-AKT and P27, cyclin D1 and P27, p-p70S6 K and p-4EBP1, p-p70S6 K and p-MAPK/ERK, and between p-4EBP1 and p-MAPK/ERK. Levels of p-p70S6 K were significantly related to recurrence in both univariate (RR = 0.75, P < 0.001) and multivariate (RR = 0.71, P = 0.049) analyses.
Conclusions: The present study helps us to understand the characteristics of signaling pathway status in breast cancers. Moreover, p-p70S6 K expression may be of use in predicting clinical outcome.