P4-06-03: Multiplex Gene Expression of Disseminated Tumor Cells in the Bone Marrow of Breast Cancer Patients Identifies Novel Therapeutic Targets.

Abstract

Abstract Background: Disseminated tumor cells (DTCs) are detected in the bone marrow (BM) of up to 40% of breast cancer patients at the time of diagnosis and are an independent prognostic factor for recurrent disease. Present techniques for detection of DTC are often laborious, and insensitive due to the molecular heterogeneity of the DTCs. We have previously optimized and validated a novel, multiplexed gene expression technology platform, Nanostring nCounter™ (NC) which counts single molecules of RNA, for the multi-marker detection of DTCs in BM at a sensitivity of 1 cancer cell per 1 million nucleated BM cells. We now validate a 36 gene panel for the detection and molecular characterization of DTCs in BM. Methods: Hybridization probes for 36 genes whose expression are associated with breast cancer, metastasis, and/or the cancer stem cell phenotype, and which exhibit no or low expression levels in normal bone marrow by qRT-PCR were developed for the NC assay. Total RNA was isolated from whole BM collected from the right and left iliac crest from breast cancer patients and healthy volunteers. 5 ug of RNA was analyzed, in duplicate with the NC assay. BM was scored positive for expression of an individual gene if expression in duplicate samples was 2 standard deviations above mean expression in a set of 11 independent normal BM samples. Results: Bilateral BM samples were analyzed prior to any therapy from 20 patients: 8 developed metastatic disease within 2–48 months (mean of 23 months) after diagnosis, and 12 had no evidence of metastatic disease with 3–5 years follow-up. Overall, expression of at least one gene in the 36-gene multi-marker panel was detected in 17 patients (85%). There was excellent correlation between individual gene expression in both the right and left iliac crest samples from the same patient. Six of the 8 patients (75%) who developed metastatic disease had detectable expression of 1–3 genes. Two genes were commonly associated with metastatic disease development. 50% (3 of 6) of the patients who had detectable expression of EBB2 in their BM developed metastatic disease, although this did not correlate with expression in the corresponding primary tumor from the same patient. 80% (4 of5) of the patients who expressed the hedgehog pathway gene, Ptch1, in their BM developed metastatic disease. Conclusions: Our data demonstrate the feasibility of using a 36-plex NC assay to detect gene expression associated with BM DTCs in breast cancer patients. We found expression of 2 targetable genes associated with the development of metastatic disease, ERBB2 and Ptch1. ERBB2 expression in BM did not correlate with expression in the primary tumor. The molecular diversity of gene expression observed underscores the need for a multiplexed gene expression panel. Ongoing studies are evaluating the clinical utility of this assay to detect DTCs relative to existing techniques, for predicting relapse-free survival, molecular classification, and selecting appropriate targeted therapeutics based on BM DTC profiles in breast cancer patients. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P4-06-03.