Phosphorylated ERalpha, HIF-1alpha, and MAPK signaling as predictors of primary endocrine treatment response and resistance in patients with breast cancer.
Generali D., Buffa FM., Berruti A., Brizzi MP., Campo L., Bonardi S., Bersiga A., Allevi G., Milani M., Aguggini S., Papotti M., Dogliotti L., Bottini A., Harris AL., Fox SB.
PURPOSE: We aimed to identify signaling pathways involved in the response and resistance to aromatase inhibitor therapy in patients with breast cancer. PATIENTS AND METHODS: One hundred fourteen women with T2-4 N0-1, estrogen receptor (ER) alpha-positive tumors were randomly assigned to neoadjuvant letrozole or letrozole plus metronomic cyclophosphamide. Twenty-four tumor proteins involved in apoptosis, cell survival, hypoxia, angiogenesis, growth factor, and hormone signaling were assessed by immunohistochemistry in pretreatment samples (eg, caspase 3, phospho- mammalian target of rapamycin, hypoxia-inducible factor 1alpha [HIF-1alpha], vascular endothelial growth factor, mitogen-activated protein kinase [MAPK], phosphorylated epidermal growth factor receptor, phosphorylated ERalpha [pERalpha]). A multivariate generalized linear regression approach was applied using a penalized least-square minimization to perform variable selection and regularization. Ten-fold cross-validation and iterative leave-one-out were employed to validate and test the model, respectively. Tumor size, nodal status, age, tumor grade, histological type, and treatment were included in the analysis. RESULTS: Ninety-one patients (81%) attained a disease response, 48 achieved a complete clinical response (43%) whereas 22 did not respond (19%). Increased pERalpha and decreased p44/42 MAPK were significant factors for complete response to treatment in all leave-one-out iterations. Increased p44/42 MAPK and HIF-1alpha were significant factors for treatment resistance in all leave-one-out iterations. There was no significant interaction between these variables and treatment. CONCLUSION: Activated ERalpha form was an independent factor for sensitivity to chemoendocrine treatment, whereas HIF-1alpha and p44/42 MAPK were independent factors for resistance. Although further confirmatory analyses are needed, these findings have clear potential implications for future strategies in the management of clinical trials with aromatase inhibitors in the breast cancer.