Clonal dominance of EGFR and efficacy of EGFR-tyrosine kinase inhibitors (EGFR-TKIs) in NSCLC.

Abstract

e21501 Background: Though EGFR mutations in lung cancer are identified most often on the trunks of tumor phylogenetic trees in early stage, whether EGFR would always be the dominant clone in the advanced stage is unknown, as cancer evolution often follows a branched trajectory, with divergent subclones evolving simultaneously. The impact of clonal dominance of EGFR on outcomes with targeted therapies has not been explored. Methods: Paired tumor and plasma samples at diagnosis were obtained from systemic treatment naïve patients with advanced NSCLC in the clinical trial (NCT03059641). cfDNA and tumor DNA were sequenced by target-capture deep sequencing of 1021 genes related to solid tumors, with blood cells as the germline control. Clonal dominance analysis was performed on the basis of the CCF determined for SNVs and clustered in plasma or tumor sequencing data using a modified version of Pyclone. PFS was estimated using Kaplan-Meier method and compared using log-rank test. Results: From February 2017 to December 2019, 300 advanced NSCLC patients were enrolled prospectively from 14 centers cross China. One hundred and fourteen EGFR mutant patients treated with EGFR-TKI were followed until disease progression (PD). The medium follow-up time was 10 month (1-27 months) and 92 (80.7%) patients have reached PD, with the ORR of 74.6% (85/114), mPFS of 10.5 months. Clonal dominance analysis of EGFR showed 76 patients had EGFR as the dominant clone according to tissue NGS results, and 66 patients as the dominant clone according to plasma cfDNA NGS results (p = 0.04). The ORR was significantly higher for patients with EGFR as dominant clone according to plasma cfDNA NGS results (84.8% vs 60.9%, p = 0.016), and PFS was significantly longer (12 vs 8 months, HR = 2.58). There was no difference when using tissue NGS results to analyze EGFR clonal dominance. However, if comparing patients who were defined as EGFR dominant clone by both tissue and plasma (n = 44) with those defined as EGFR nondominant (n = 9), EGFR dominance was associated with higher ORR (84.1% vs 44.4%, p = 0.01), and longer PFS (11 vs 6 months, HR = 9.88) significantly. Moreover, multivariate Cox proportional hazard ratio analysis demonstrated it as an independent prognostic indicator of EGFR-TKIs. Conclusions: Clonal dominance of EGFR in the pretreatment plasma cfDNA is associated with the efficacy of EGFR-TKIs in NSCLC. This study indicated the importance of evaluating clonal dominance in the current clinical practice and future trial designs. Clinical trial information: NCT03059641.