Abstract 1391: Gene expression profiling for cancer classification in circulating tumor cells

Background: Enumeration and molecular characterization of circulating tumor cells (CTCs) offer a non-invasive method for tumor analysis, particularly in cases where the biopsy is difficult to obtain or has been exhausted by conventional diagnostic methods. The 92-gene assay (CancerTYPE ID®, bioTheranostics, Inc.) is a clinically-validated cancer classifier based on the collective expression of 87 informative genes, many of which are involved in lineage commitment and signal transduction. The objective of this study was to determine the feasibility of a blood-based application using differential gene expression. Methods: Technical feasibility studies for cancer cell enrichment and RNA viability were conducted by purifying cancer cell lines that were spiked into donor blood using a microfluidic chip (Celsee PREP™100 system, Celsee Diagnostics) followed by immunodepletion of leukocytes. Quantitative RT-PCR and immunoflourescence were performed on cells harvested at different steps of the purification workflow. Quantitative RT-PCR on individual genes specific to each lineage compartment was used to estimate cell types and cell numbers that were co-purified by the workflow. Gene expression was evaluated by the 92-gene assay. Results: An integrated workflow was established that allows for the molecular characterization of cancer cells purified from blood. Cell enrichment on the microfluidic chip followed by purification using immunodepletion resulted in efficient recovery (> 80% of input cells) and viable RNA from cancer cells spiked into whole blood. A key result of this study was the demonstration that the level of cancer-specific gene expression in purified samples was similar to the level of expression from an equivalent number of unspiked cancer cells suggests that significant cell loss does not occur during the purification workflow. A comparison of gene expression profiles from purified and mixed cell populations allowed for the identification of subsets of genes that are differentially expressed in cancer cells compared with leukocytes. Conclusion: A functional workflow with viable RNA and efficient recovery of cancer cells that are within the technical specifications of the 92-gene assay was established. Findings from this feasibility study are fundamental to the potential for CTC-based cancer classification. Citation Format: Harris S. Soifer, Ranelle C. Salunga, Tristan G. Harris, Jose Ramirez, Priyadarshini Gogoi, Yixin Wang, Saedeh Sepehri, Kalyan Ha