Validation of genotyping of gastrointestinal stromal tumor in Japan

Abstract

e21502 Background: Most gastrointestinal stromal tumors (GIST) have activating mutations in the KIT or PDGFRA gene. Genotyping of GIST is important in Dx and Tx of GIST. Methods of genotyping using genomic DNA extracted from paraffin-embedded specimens are diverse and not standardized. We did validation study of genotyping using special reference to sequencing data obtained from cDNA from fresh GIST samples. Methods: Three DNA extraction methods (QIAamp, DEXPAT, or original) and four PCR methods (Ex Taq, AmpliTaq condition-1, AmpliTaq condition-2, or QIAGEN Tag) were compared using 20 paraffin-embedded specimens with special reference to sequencing data obtained from cDNA from corresponding 20 fresh GIST samples. After DNA extraction, KIT exon 9, 11, 13 and 17, and PDGFRA exon 12 and 18 were amplified by each PCR method using specific primers and directly sequenced. Results: In evaluation of PCR method, the protocol with Ex Taq showed 100% amplication of DNA and sequence agreement, the protocol with QIAGEN Tag 99%, and the protocol with AmpliTaq condition-2 86% agreement, and the protocol with AmpliTaq condition-1 showed much less amplication and higher disagreement. For the DNA extraction, the protocol with QIAamp showed best DNA extraction and its DNA sequence data were consistent with reference sequence in 98%, DNA sequence obtained using DEXPAT showed 33% consistency, and 89% of DNA sequence data obtained from an original method was agreed with reference data. Some modifications improved DNA amplication but inconsistent sequence data also increased probably due to miss-PCR. Conclusions: Each DNA extraction method had different quantity of DNA and four PCR methods showed different quality. Using this validation study, a standard genotyping method in Japan was established. No significant financial relationships to disclose.