Serial cfDNA assessment of response and resistance to EGFR-TKI for patients with EGFR-L858R mutant lung cancer from a prospective clinical trial

Qing Zhou, Jin-Ji Yang, Zhi-Hong Chen, Xu-Chao Zhang, Hong-Hong Yan, Chong-Rui Xu, Jian Su, Hua-Jun Chen, Hai-Yan Tu, Wen-Zhao Zhong, Xue-Ning Yang, Yi-Long Wu

Detecting epidermal growth factor receptor (EGFR) activating mutations in plasma could guide EGFR-tyrosine kinase inhibitor (EGFR-TKI) treatment for advanced non-small cell lung cancer (NSCLC). However, dynamic quantitative changes of plasma EGFR mutations during the whole course of EGFR-TKI treatment and its correlation with clinical outcomes were not determined. The aim of this study was to measure changes of plasma EGFR L858R mutation during EGFR-TKI treatment and to determine its correlation with the response and resistance to EGFR-TKI. This study was a pre-planned exploratory analysis of a randomized phase III trial conducted from 2009 to 2014 comparing erlotinib with gefitinib in advanced NSCLC harboring EGFR mutations in tumor (CTONG0901). Totally, 256 patients were enrolled in CTONG0901 and randomized to receive erlotinib or gefitinib. One hundred and eight patients harbored L858R mutation in their tumors and 80 patients provided serial blood samples as pre-planned scheduled. Serial plasma L858R was detected using quantitative polymerase chain reaction. Dynamic types of plasma L858R were analyzed using Ward's hierarchical clustering method. Progression-free survival (PFS) and overall survival (OS) were compared between different types. As a whole, the quantity of L858R decreased and reached the lowest level at the time of best response to EGFR-TKI. After the analysis of Ward's hierarchical clustering method, two dynamic types were found. In 61 patients, L858R increased to its highest level when disease progressed (ascend type), while in 19 patients, L858R maintained a stable level when disease progressed (stable type). Median PFS was 11.1 months (95 % CI, 6.6-15.6) and 7.5 months (95 % CI, 1.4-13.6) in patients with ascend and stable types, respectively (P = 0.023). Median OS was 19.7 months (95 % CI, 16.5-22.9) and 16.0 months (95 % CI, 13.4-18.5), respectively (P = 0.050). This is the first report finding two different dynamic types of plasma L858R mutation during EGFR-TKI treatment based on a prospective randomized study. Different dynamic types were correlated with benefits from EGFR-TKI. The impact of plasma L858R levels at disease progression on subsequent treatment strategy needs further exploration. ClinicalTrials.gov, NCT01024413.