Background
Epidermal growth factor receptor (EGFR) mutations (EGFRm) represent one of the most common genomic alterations identified among patients with non-small cell lung cancer (NSCLC). Several targeted agents for patients with EGFRm have been proven safe and effective, including the third-generation tyrosine kinase inhibitor (TKI) osimertinib. Nonetheless, some patients will present with or develop EGFR-TKI resistance mechanisms.
Conclusion
EGFR-independent alterations are common among patients with primary resistance to osimertinib and significantly impact PFS and OS. Our results suggest that among Hispanic patients, other variables associated with intrinsic resistance include the number of commutations, high levels AXL mRNA, and low levels of BIM mRNA, T790M de novo, EGFR p.L858R presence, and a high tumoral mutational burden.
Methods
An observational longitudinal cohort study was conducted with two groups of patients, those with intrinsic resistance (cohort A) and those with long-term survival (cohort B). All patients were treated and followed between January 2018 and May 2022. All patients were assessed for Programmed Cell Death Ligand 1 (PD-L1) expression and Bcl-2-like protein 11 (BIM)/AXL mRNA expression before starting TKI. After 8 weeks of treatment, a liquid biopsy was performed to determine the presence of circulating free DNA (cfDNA), and next-generation sequencing (NGS) was used to identify mutations at the time of progression. In both cohorts, overall response rate (ORR), progression-free survival (PFS), and overall survival (OS) were evaluated.
Objective
We characterized the genomic landscape of primary resistance to osimertinib among Hispanic patients with EGFR-mutant NSCLC.
Results
We found a homogeneous distribution of EGFR-sensitizing mutations in both cohorts. For cohort A, exon 21 mutations were more common than exon 19 deletions (ex19dels) for cohort B (P = 0.0001). The reported ORR for osimertinib was 6.3% and 100% for cohorts A and B, respectively (P = 0.0001). PFS was significantly higher in cohort B (27.4 months vs. 3.1 months; P = 0.0001) and ex19del patients versus L858R (24.5 months, 95% confidence interval [CI] 18.2-NR), vs. 7.6 months, 95% CI 4.8-21.1; P = 0.001). OS was considerably lower for cohort A (20.1 months vs. 36.0 months; P = 0.0001) and was better for patients with ex19del, no brain metastasis, and low tumor mutation burden. At the time of progression, more mutations were found in cohort A, identifying off-target alterations more frequently, including TP53, RAS, and RB1.