Accuracy of the cobas EGFR Mutation Assay in Non–small-cell Lung Cancer Compared With Three Laboratory-developed Tests

Abstract

BackgroundThe reliability of the cobas EGFR assay to detect epidermal growth factor receptor (EGFR) mutations in non–small-cell lung cancer (NSCLC) as an in vitro diagnostic test was compared with 3 laboratory-developed tests (LDTs). Materials and MethodsAfter screening for EGFR mutations using formalin-fixed-paraffin-embedded NSCLC tissue sections using the cobas EGFR assay, 151 samples were further tested with 3 LDTs; the peptide nucleic acid-locked nucleic acid polymerase chain reaction (PCR) clamp, PCR invader, and Cycleave assays. The cobas EGFR assay performance was evaluated by determining the concordance rate and κ-coefficient between the assays. In samples exhibiting discrepancies in the EGFR mutation status in the 4 assays, next-generation sequencing was performed to confirm mutated sequences. ResultsConcordance rates and κ-coefficients between the cobas EGFR assay and the other tests were 96.0% and 0.921 for the peptide nucleic acid-locked nucleic acid PCR clamp assay, 94.0% and 0.881 for the PCR invader assay, and 96.7% and 0.934 for the Cycleave assay, respectively. Data showed very good agreement with the other assays. Precise mutated sequences or exons in the EGFR gene matched in 137 samples (90.7%). Different results were obtained in 4 samples (2.6%), owing to systemic limitations of the assay. Next-generation sequencing of 10 (6.6%) samples with discordant results exhibited a concordance rate of 60% to 80% in each assay. ConclusionsThe cobas EGFR assay showed high concordance rates and κ-coefficients between the 3 compared LDTs and can be used to select patients who would benefit from EGFR-tyrosine kinase inhibitors.