Presence of PCR artifacts in Sanger sequencing in Formalin-fixed paraffin-embedded (FFPE) tissue – experience in a collective of 990 advanced NSCLC

Abstract

Background: Despite the rapid development of new molecular techniques such as Next Generation Sequencing (NGS), Sanger sequencing has been thus far the gold standard for mutation analysis. It is constantly used for daily routine diagnostics because it represents a quick and comprehensive available method for mutation analyses. Although Sanger sequencing is a good validated method, PCR artifacts may occur in formalin fixed paraffin embedded (FFPE) material. This constitutes a serious source of error.
Aims: To assess the prevalence of typical and atypical EGFR mutations in exon 19 and 21 in a collective of 990 advanced non-small cell lung cancer (NSCLC) patients, focusing especially on methodological issues and challenges concerning mutation analysis, particularly PCR artifacts.
Material and Methods: We examined 990 NSCLC (FFPE material) by Sanger sequencing for exon 19 and 21 of the EGFR gene. Four cases dropped out because of insufficient DNA quality (n =986). Results: Beside 101 typical exon 19 and 21 mutations (99 cases, two double mutations) we found 45 additional cases with distinct peaks at atypical positions in exon 19 and 21 in our first analysis. This would have implied a mutation rate of 14.6 %. Only six of these putative atypical mutations (all exon 21 and none of the exon 19 mutations) could be validated by repeated mutation analysis. All other peaks were not reproducible, therefore considered as PCR artifacts and consequently as wild type. Altogether we found 105 cases (107 mutations, 10.6 % of cases) with typical/atypical mutations in exon 19 and 21 of the EGFR gene.
Conclusion: In our opinion it is in general important to detect and report all mutations even at atypical sites to discover their possible clinical relevance. However, one must always be aware of the possibility, reasons and prevention of PCR artifacts in FFPE tissue. Therefore, prior to reporting mutations at uncommon sites these must be validated by repeated analyses.