Five Technologies for Detecting the EGFR T790M Mutation in the Circulating Cell-Free DNA of Patients With Non-small Cell Lung Cancer: A Comparison

Abstract

Third-generation tyrosine kinase inhibitors (TKIs) were developed to overcome T790M-mediated resistance to earlier generations of epidermal growth factor receptor <i>(EGFR)</i>-targeted TKIs. We compared four well-established and one in-house method for the analysis of the <i>EGFR</i> T790M mutation in plasma cell-free DNA (cfDNA), in hope to find a better way to select non-small cell lung cancer (NSCLC) patients appropriate for 3rd-generation TKI therapy. For sensitivity levels of each method, plasmid DNA with <i>EGFR</i> T790M mutations was serially diluted with cfDNA from healthy controls with wild type <i>EGFR</i>. The clinical performance was analyzed in a clinical cohort of <i>EGFR</i> mutation-positive NSCLC patients with acquired EGFR TKI resistance (<i>n</i> = 40). All methods except the therascreen kit (Qiagen) had a sensitivity level of 10 copies of T790M plasmid DNA in the spiked specimen. The detection rates of the <i>EGFR</i> T790M mutation in plasma cfDNA from the clinical cohort were 42.5, 35, 32.5, 22.5, and 17.5% for the in-house ARMS method, Bio-Rad droplet digital PCR, PANAMutyper, <i>Therascreen EGFR Plasma</i> RGQ PCR Kit and Cobas EGFR Mutation kit (with suboptimal template amounts), respectively. Osimertinib was given to 17 of 20 patients with <i>EGFR</i> T790M mutations. The best treatment responses, based on the RECIST criteria, included 6 partial responses (PR) and 7 stable diseases (SD). The PANAMutyper and the Bio-Rad droplet digital PCR were comparable, the Cobas EGFR Mutation kit required significantly more template for testing. The best combination would be the in-house ARMS method plus the PANAMutyper or Bio-Rad droplet digital PCR, which would have a detection rate of 50% (20/40) and a disease control rate of 76% (13/17).