Background: Epidermal growth factor receptor (EGFR) mutation is predictive for the efficacy of EGFR tyrosine kinase inhibitors in advanced non-small-cell lung cancer (NSCLC) treatment. We evaluated the performance, sensitivity, and concordance between five EGFR tests.
Materials and methods: DNA admixtures (n = 34; 1%-50% mutant plasmid DNA) and samples from NSCLC patients [116 formalin-fixed paraffin-embedded (FFPE) tissue, 29 matched bronchofiberscopic brushing (BB) cytology, and 20 additional pleural effusion (PE) cytology samples] were analyzed. EGFR mutation tests were PCR-Invader, peptide nucleic acid-locked nucleic acid PCR clamp, direct sequencing, Cycleave, and Scorpion Amplification Refractory Mutation System (ARMS). Analysis success, mutation status, and concordance rates were assessed.
Results: All tests except direct sequencing detected four mutation types at ≥1% mutant DNA. Analysis success rates were 91.4%-100% (FFPE) and 100% (BB and PE cytology), respectively. Inter-assay concordance rates of successfully analyzed samples were 94.3%-100% (FFPE; kappa coefficients: 0.88-1.00), 93.1%-100% (BB cytology; 0.86-1.00), and 85.0%-100% (PE cytology; 0.70-1.00), and 93.1%-96.6% (0.86-0.93) between BB cytology and matched FFPE.
Conclusions: All EGFR assays carried out comparably in the analysis of FFPE and cytology samples. Cytology-derived DNA is a viable alternative to FFPE samples for analyzing EGFR mutations.