Microarray gene expression patterns recapitulated on a clinically tractable diagnostic platform using breast and lung cancer formalin-fixed paraffin-embedded tissues.

Abstract

e21027 Background: In Vitro Diagnostic Multivariate Index Assays (IVDMIA) are diagnostic tools that produce a diagnostic, prognostic and predictive index. The development of a gene expression (GE)-based IVDMIA require investigation of devices that integrate accurate, rapid and cost-effective multi-GE analysis with conventional protocols for tissue processing. We evaluated the Panomics’ QuantiGene Plex 2.0 expression analysis system (QGS). The QGS is a nucleic acid hybridization-based assay that allows direct quantification of mRNA transcripts from formalin-fixed, paraffin-embedded (FFPE) tissue sections. In this study, we evaluate this method to investigate its utility in diagnostic GE analysis. Methods: We analyzed 16 FFPE breast and 24 FFPE non-small cell lung cancer (NSCLC) samples using the QGS. There was pathologic confirmation of tumor cellularity in all specimens, estrogen receptor (ER) status and proliferative capacity in breast cancer, and histology of NSCLC tumors. In the breast cancer tumors, we focused on genes discovered by microarray analysis that distinguish ER+ and ER- breast tumors as well as tumors with low and high proliferation rates (the “Genetic Grade Signature”, GGS). In the NSCLC tumor set, we focused on genes that correlated with pathologic histology (the “A/S” panel). QuantiGene probes for each of the target genes were designed against corresponding Affymetrix probe set target sequences. Resultant GE profiles were normalized to housekeeping genes and evaluated for correlations. Results: Each of the 8 genes of the ER status panel was correlated with tumor ER status (p<0.05). In unsupervised cluster analysis, the GE profiles of these genes accurately segregated ER+ and ER- tumors. The GGS genes, transformed to a single metagene, showed a robust association with IHC-derived KI67 scores, consistent with involvement in a larger proliferation signature. Finally, in the NSCLC tumors, the “A/S” signature genes showed a marked ability to distinguish NSCLC adenocarcinomas from squamous cell carcinomas. Conclusions: The QGS method may be a used as a diagnostic platform for multi-gene diagnostics.