Next-generation sequencing reveals novel resistance mechanisms and molecular heterogeneity in EGFR-mutant non-small cell lung cancer with acquired resistance to EGFR-TKIs

Abstract

OBJECTIVES:

Despite initial responses to epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) in EGFR mutant non-small cell lung cancer, patients invariably develop acquired resistance. In this study, we performed next-generation sequencing in pre- and post-EGFR-TKI tumor samples to identify novel resistance mechanisms to EGFR-TKIs.

MATERIAL AND METHODS:

We collected tumor tissues before EGFR-TKI treatment and after progression from 19 NSCLC patients to analyze genomic alterations in 409 cancer related genes. Bioinformatics analyses were used to identify mutations in which the allele frequencies are significantly changed, or newly appeared after progression.

RESULTS:

Overall, mutation rates and compositions were similar between pre- and post-EGFR-TKI tumors. We identified EGFR T790M as the most common mechanism of acquired resistance (63.2%). No pre-EGFR-TKI tumor had a preexisting T790M mutation, suggesting that tumors acquired T790M mutations following progression on EGFR-TKIs. Compared to T790M-positive tumors, T790M-negative tumors showed relatively high tumor mutation burden and shorter survival, suggesting T790M-negative patients as a potential candidate for immune checkpoint inhibitors. TP53 mutation was also significantly enriched in the T790M-negative tumors. Finally, we described here for the first time a novel missense mutation (T263P), which occurred concurrently with an activating G719A mutation, in the extracellular domain II of EGFR in a patient with poor response to erlotinib. Ba/F3 cells harboring EGFR T263P/G719A mutation showed higher sensitivity to afatinib, compared to gefitinib due to inhibition of EGFR/HER2 heterodimerization.

CONCLUSION:

Comprehensive genomic analysis of post-EGFR-TKI tumors can provide novel insight into the complex molecular mechanisms of acquired resistance to EGFR-TKIs.