Abstract 3390: Validating the RNAscope for molecular profiling of key biomarkers associated with gemcitabine resistance

Introduction: Acelarin® is the first anti-cancer ProTide to enter the clinic and has achieved a very high rate of disease control in a Phase I study (ProGem1) of patients with advanced progressive solid tumours. It is a first-in-class nucleotide analogue specifically designed to bypass key cancer resistance pathways associated with gemcitabine, including altered expression of human equilibrative nucleoside transporter 1 (hENT1), deoxycytidine kinase (dCK) and cytidine deaminase (CDA). We have used a cytological RNA-based assay, RNAscope® and quantitative Reverse Transcriptase Polymerase Chain Reaction (qPCR) analysis to quantify these three biomarkers in seven cancer cell lines, and reviewed and tested a panel of commercially available antibodies against each protein. The aim was to select and validate a method suitable for stratifying tumours according to their molecular signature and allow the identification of patients unlikely to derive clinical benefit from gemcitabine. Methods: Cells from seven cancer lines (BxPC-3, MiaPaCa-2, PANC-1, SK-MES-1, NCI-H1703, OVCAR-3 and OVCAR-8) were prepared in formalin-fixed paraffin-embedded blocks. In addition, transiently transfected MiaPaCa-2 cells overexpressing hENT1, dCK and CDA were prepared. Expression of mRNA was detected using RNAscope and quantified by SpotStudio as well as by qPCR. Antibodies were examined by Western-blot analysis, reverse phase protein arrays (RPPA) and immunohistochemistry. Protein expression by immunofluorescence was quantified using Aqua HistoRx technology. Results: RNAscope detected mRNA for CDA at low levels in all cell lines. dCK and hENT1 were detected at low (in BxPC-3) to high (in NCI-H1703) expression levels across the cell lines. The qPCR method quantified precisely the amount of RNA, which demonstrated a positive and significant correlation with RNAscope data (Spearman correlation rho = 0.76, p = 0.037 for CDA, rho = 0.67, p = 0.08 for dCK, rho = 0.67, p = 0.08). Most antibodies were rejected because of low specificity as observed by Westerns, RPPA and immunohistochemistry. In addition, both techniques identified the gene over-expression in the transfected cell lines where the expression was up to 100 fold higher in transfected cells compared to their wild-type pairs. Conclusions: Selection of antibodies against hENT1, CDA and DCK for clinical use remains challenging. The RNAscope technology was validated, allowing reliable detection of biomarkers thought to be associated wit