Over the last 50 years, the number of cancer related deaths has decreased by only 2%. One of the most promising approaches to reduce breast cancer mortality is to develop tools for early detection and early intervention of breast cancers. Normal mammary gland homeostasis requires the coordinated regulation of signaling networks; whereas, dysregulation of signaling networks occurs during breast cancer initiation. Reverse-phase protein microarray (RPPM) is a high-throughput proteomic tool, developed to test for dysregulation of protein signaling networks in human biopsy specimens. For example, ErbB2/HER2/neu, Erb3, EGFR, Akt, Erk1/2 receptor tyrosine kinases all play a role in cancer cell growth and survival, and developing targeted therapies against these pathways is a promising tailored approach to breast cancer treatment.

To test which tyrosine kinases are activated in early mammary carcinogenesis, reverse phase protein microarray analysis (RPPM) was performed on 31 random periareolar fine needle aspirates (RPFNA) obtained from a cohort of asymptomatic high-risk women, testing 59 antibody endpoints related to cell growth and survival pathways, including ER, EGFR, Her2, Erb3, Akt and Erk1/2, to test for patterns of differential protein expression. RPFNA allows for serial sampling of breast cytology and subsequent histology from asymptomatic women at high risk for developing breast cancer and also provides the ability to monitor response to preventative treatments. Epithelial cell clusters from RPFNA samples were isolated using AutoPix automated laser capture microdissection. RPPM uses denatured lysate so antigen retrieval, which is a limitation for tissue arrays, is not a problem. In addition, each sample is printed in serial dilution, providing an internal standard, and RPPM does not require direct labeling of the sample, thus improving reproducibility, sensitivity, and robustness. An interrogation of phosphorylated versus non-phosphorylated state of proteins was performed using a Wilcoxon rank sum test. Spearman rank correlation coefficients were used to estimate the association between RPPM total- and phospho-protein expression as a function of cytology; a 2-sided p value < 0.05 was used to determine if a significant positive or negative association exists.

Four distinct clusters of phospho-proteins were identified amongst the 31 RPFNA samples. Varying cell lines, including 15hTert, MDA231, T47D, were then treated with inhibitors based on the protein pathways identified in these clusters. Fluoxetine, a common antidepressant, was shown to affect cancer cell viability as well as the phosphorylation of Erk1/2 pathway in a concentration and time dependent manner.

Diagnosing cancers based on proteomic signatures in addition to histopathology will allow for individualized selection of therapeutic combinations that can best target the entire disease-specific errant protein network of a patient. This proteomic signature has great potential for daily clinical practice considering the RPPM technology is commercially available through Theranostics Health and could be used every time a woman undergoes a routine biopsy. In addition, understanding a key protein pathway and how it is wired could have a profound effect on both functional biology and on the understanding of cancer mechanisms as a whole.

Citation Information: Clin Cancer Res 2010;16(14 Suppl):A15.