Automated extraction of DNA and RNA from a single formalin-fixed paraffin-embedded tissue section for analysis of both single-nucleotide polymorphisms and mRNA expression

Abstract

Background: There is an increasing need for the identification of both DNA and RNA biomarkers from pathodiagnostic formalin-fixed paraffin-embedded (FFPE) tissue samples for the exploration of individualized therapy strategies in cancer. We investigated a fully automated, xylene-free nucleic acid extraction method for the simultaneous analysis of RNA and DNA biomarkers related to breast cancer.

Methods: We copurified both RNA and DNA from a single 10-μm section of 210 paired samples of FFPE tumor and adjacent normal tissues (1-25 years of archival time) using a fully automated extraction method. Half of the eluate was DNase I digested for mRNA expression analysis performed by using reverse-transcription quantitative PCR for the genes estrogen receptor 1 (ESR1), progesterone receptor (PGR), v-erb-b2 erythroblastic leukemia viral oncogene homolog 2, neuro/glioblastoma derived oncogene homolog (avian) (ERBB2), epoxide hydrolase 1 (EPHX1), baculoviral IAP repeat-containing 5 (BIRC5), matrix metallopeptidase 7 (MMP7), vascular endothelial growth factor A (VEGFA), and topoisomerase (DNA) II alpha 170kDa (TOP2A). The remaining undigested aliquot was used for the analysis of 7 single-nucleotide polymorphisms (SNPs) by MALDI-TOF mass spectrometry.

Results: In 208 of 210 samples (99.0%) the protocol yielded robust quantification-cycle values for both RNA and DNA normalization. Expression of the 8 breast cancer genes was detected in 81%-100% of tumor tissues and 21%-100% of normal tissues. The 7 SNPs were successfully genotyped in 91%-97% of tumor and 94%-97% of normal tissues. Allele concordance between tumor and normal tissue was 98.9%-99.5%.

Conclusions: This fully automated process allowed an efficient simultaneous extraction of both RNA and DNA from a single FFPE section and subsequent dual analysis of selected genes. High gene expression and genotyping detection rates demonstrate the feasibility of molecular profiling from limited archival patient samples.