Concordance of genotype for polymorphisms in DNA isolated from peripheral blood and colorectal cancer tumor samples

Background & aim: Results from different pharmacogenetic association studies in colorectal cancer are often conflicting. Both peripheral blood and formalin-fixed, paraffin-embedded (FFPE) tissue are routinely used as DNA source. This could cause bias due to somatic alterations in tumor tissue, such as loss of heterozygosity. We therefore compared genotypes in DNA from peripheral blood and FFPE colorectal tumor samples for SNPs with putative influence on the cytotoxicity of chemotherapy. Materials & methods: Eleven SNPs in nine genes involved in anticancer drug metabolism or efficacy were determined in matched samples from blood and FFPE tissue of colorectal tumors by pyrosequencing and TaqMan((R)) techniques. The -statistic was calculated to assess concordance. Results: A total of 149 paired FFPE tissue and EDTA blood DNA samples were available for comparison. Overall, 20 out of 1418 genotypes were discordant (1.4%); in ten cases, loss of heterozygosity could not be ruled out. Only GSTP1 showed significant discordance between FFPE tissue and blood genotype ( = 0.947; 95% CI: 0.896-0.998). Conclusion: FFPE tissue-derived DNA can be used as a valid proxy for germline DNA for a selection of SNPs in (retrospective) pharmacogenetic association studies in colorectal cancer. However, for future studies, genotyping of blood-derived DNA is preferred. Original submitted 29 May 2013; Revision submitted 23 August 2013