The largest source of diagnostic and well_annotated patient tissues consists of formalin-fixed and paraffin embedded (FFPE) tissue blocks. Because the measurement of RNA in FFPE tissues is unreliable, an immense resource of molecular information is not fully accessible. Therefore, we evaluated the branched-chain DNA (bDNA) assay (Quantigene®) as a promising novel technology for quantification of RNA in FFPE tissues. The bDNA assay relies on the capture of RNA from a tissue homogenate with probe sets specifically adjusted for small RNA fragments and does not require RNA extraction or enzymatic amplification of target RNA. The reproducibility (intra-method reliability) and validity (inter-method reliability) were determined by comparison of 20 frozen and 20 FFPE samples from 10 prostate cancer xenografts. The measurement error (technical variability attributed to the assay) for FFPE tissues was < 0.7% for the bDNA assay and 5.0 - 17.6% for the qRT-PCR assay. Comparing measurements of 5 genes in FFPE tissues to measurements in frozen tissues by qRT-PCR as a reference method, we observed an increase in the range of correlation coefficients from 0.38 - 0.95 for qRT-PCR to 0.6 - 0.94 for the bDNA assay. In addition, the limit of quantification in FFPE tissues was increased by 10-fold in samples of tissue homogenate compared to those of purified RNA. An application of the bDNA assay to macrodissected prostate cancer and normal tissues revealed the expected overexpression of a a panel of 10 cancer genes. In conclusion, the bDNA assay significantly improved the assay reliability compared to qRT-PCR when measuring a diagnostic, prognostic or treatment-related gene expression panel in FFPE tissues.

Second AACR International Conference on Molecular Diagnostics in Cancer Therapeutic Development-- Sep 17-20, 2007; Atlanta, GA