Digital PCR detection of EGFR somatic mutations in non-small-cell lung cancer formalin fixed paraffin embedded samples

Abstract

BackgroundDigital PCR (dPCR) is proposed to replace real time PCR and Sanger sequencing for detection and quantification of rare mutations, frequently unnoticed in the mass of tumoral cells. Screening of endothelial growth factor receptor (EGFR) mutations is mandatory before treatment with EGFR-targeted therapy with small-molecule tyrosine kinase inhibitors, which has been approved for the treatment of advanced non-small-cell lung cancer (NSCLC). ObjectiveIn order to establish a cost-effective method for detection of mutations, we optimized dPCR identification of EGFR mutations in exons 18–21, and determined dPCR sensitivity, limits of detection (LoD) and quantification (LoQ). MethodsFor clinical validation, we compared the performance of dPCR and castPCR in 57 NSCL formalin fixed paraffin embedded samples and 10 lung cancer-free formalin fixed paraffin embedded samples. ResultsEGFR mutations DEL19, p.L858R, p.G719X, p.L861Q and p.T790 M were detected by dPCR in 27 samples versus 11 detected by castPCR (p = 0.014). LoD was determined as 100 molecules of DNA/uL and LoQ as 1%. Most of the samples (87%) identified by competitive Allele-Specific TaqMan (castPCR) as wild-type and by dPCR as mutated, presented less than 10% mutated DNA molecules (mean 4.57%). Accuracy of dPCR was 94.44%, as measured with the assay recommended by the College of American Pathologists. ConclusionThese results indicated higher sensibility and specificity of dPCR for screening EGFR mutations in NSCLC biopsies, compared to castPCR.