HOME HELP FEEDBACK HOW TO CITE ABSTRACTS ARCHIVE CME INFORMATION SEARCH Cancer Research Clinical Cancer Research Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics Molecular Cancer Research Cancer Prevention Research Cancer Prevention Journals Portal Cancer Reviews Online Annual Meeting Education Book Meeting Abstracts Online		QUICK SEARCH:   [advanced] Author: Keyword(s):

Services Similar articles in this journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Knudsen, B. S. Articles by Kristal, A. Search for Related Content PubMed Articles by Knudsen, B. S. Articles by Kristal, A.

Diagnostic Technologies and Molecular and Cellular Profiling: Biomarker Assay Development, Validation, and Qualification

Branched chain DNA assay for accurate measurement of RNA in formalin-fixed tissues

Fred Hutchinson Cancer Research Ctr., Seattle, WA, University of Washington, Seattle, WA, Panomics, Fremont, CA

Abstract

A14

Background: The large resource of well-annotated formalin-fixed tissues is of limited use for prognostic and epidemiological studies because the quantification of RNA expression in archival tissues is problematic. The difficulty in measuring RNA expression arises from the extensive fragmentation and the crosslinking of RNA after formalin fixation. The branched chain DNA assay (bDNA assay) and the Quantigene® procedure are designed to overcome the adverse effects of formalin fixation. In the bDNA assay, a set of capture probes binds the RNA of interest and sequential hybridization to a strand-specific probe set and a branched DNA non-enzymatic amplifier are used for quantification of bound RNA. Here we validate the bDNA assay by comparison to qPCR.Methods: Duplicate samples from 10 different xenografts were snap frozen or fixed in formalin. Samples were prepared either by RNA isolation (pRNA) or by solubilization in a tissue homogenization buffer (thRNA). Probe sets for capture of formalin-fixed (FF) RNA were specifically designed for short RNA fragments. We measured the expression of six genes in 20 samples in parallel by qPCR and bDNA assay. For comparison of reliability between the bDNA and qPCR assays, we determined the intraclass correlation coefficient (ICC). To analyze the accuracy of the bDNA assay, we used qPCR measurements of frozen tissues as a reference and obtained correlation coefficients () directly from the variance components model. The sensitivity of bDNA assay was compared to qPCR for pRNA and thRNA sample preparations.Results: Measurements using the bDNA assay for RNA preparations from FF tissues were highly reproducible, with ICCs of six genes ranging between 0.84 and 0.97. In contrast a parallel analysis by qPCR revealed ICCs between 0.00 and 0.67. ICCs for frozen tissues ranged between 0.25 and 0.99 for bDNA and 0.64 to 0.99 for the qPCR assay. The correlation coefficients for measurements of gene expression by bDNA assay in FF and frozen tissues compared to qPCR measurements of frozen tissues were between 0.53 and 0.93 and between 0.22 and 0.93, respectively. In contrast, the correlation coefficients for measurements in frozen and FF tissue by qPCR were between -0.33 and 0.72. The sensitivity of the bDNA for quantification of pRNA in FF tissues was comparable to qPCR. However, an increased sensitivity of the bDNA assay was observed when measuring RNA in tissue homogenates.Conclusion: The bDNA permits measurements of RNA in tissue homogenates without RNA isolation and provides better reproducibility, accuracy and sensitivity than PCR-based methods.