Abstract B13: Automated multiplex assay for diagnostic application of contextual microRNA signatures in breast cancer

MicroRNAs (miRNAs) are rapidly emerging as a novel class of biomarkers for theranostic applications in cancer. Since solid tumor lesions are complex and heterogeneous mixtures of different cell types, we previously developed a multiplex tissue slide-based assay to measure miRNA expression levels at single cell resolution and to extract the contextual information embedded within both cancer and non-cancer cell compartments of the tumor microenvironment. In this on-going clinical study, we are characterizing the cell type-specific altered expression of a panel of breast cancer-associated miRNAs on large and independent cohorts of breast cancer patients. To maximize the number of independent patient cases per slide and to minimize the number of independent experiments and slides for data analysis, we are interrogating miRNA expression on high-density tissue microarrays (TMAs). TMAs designed by NCI Cooperative Breast Cancer Tissue Resources to find statistically significant associations between an investigative biomarker and disease progression and/or disease outcome will serve as discovery set. Our institutional TMAs will serve as validation set and will also enable us to assess the potential utility of these miRNA contextual signatures to predict treatment response in different adjuvant and neoadjuvant therapy settings using recurrence-free survival and complete pathological response, respectively, as the clinical endpoint marker. Our preliminary results indicate that inflammation and desmoplasia present as co-diseases or co-morbidities in breast cancer, and they are both major confounding factors in the correct interpretation of expression profiling experiments and the assignment of etiological roles to leukocyte-expressed miRNAs such as miR-155 and reactive fibroblast-expressed miRNAs such as miR-21. Intriguingly, we have identified a subgroup of HER2-overexpressing cases in which miR-21 expression is mainly upregulated with cancer cells, while miR-21 expression predominantly accumulates within reactive fibroblasts in the rest of HER2-overexpresssing cases and the immense majority of cases from other breast cancer subtypes. The clinical implications of cell source and expression levels of miR-21 and other miRNAs under study will be presented at this meeting. We will also discuss innovative strategies that we are implementing for integration of translational miRNA and well-established protein biomarkers to increase molecular characterization of breast tumors and for pipeline automatization from assay performance to whole slide image acquisition and computer-assisted analysis to enhance diagnostic power of current clinical assays. Citation Format: Wendy Wells, Sung-hee Choi, Gary Schwartz, Lorenzo F. Sempere. Automated multiplex assay for diagnostic application of contextual microRNA signatures in breast cancer [abstract]. In: Proceedings of the AACR Special Conference on Noncoding RNAs and Cancer; 2012 Jan 8-11; Miami Beach, FL. Philadelphia (PA): AACR; Cancer Res 2012;72(2 Suppl):Abstract nr B13.