Role of Circulating Tumor DNA Profiling in Patients with Non-Small Cell Lung Cancer Treated with EGFR Inhibitor

<b><i>Objectives:</i></b> During targeted therapy, tumor heterogeneity can drive the evolution of multiple tumor subclones harboring unique resistance mechanisms. Sequential profiling of plasma cell-free DNA (cfDNA) provides a noninvasive method for early detection of patient progression. We investigated whether the genetic dynamics detected in cfDNA during treatment can act as a predictive or prognostic marker of outcome. <b><i>Methods:</i></b> Patients with advanced <i>EGFR</i>-mutated non-small cell lung cancer (NSCLC) were included for consecutive blood sampling during EGFR-tyrosine kinase inhibitor (TKI) treatment. Blood samples were serially collected from patients at baseline, first follow-up, and progression. Extracted cfDNA was analyzed with next-generation sequencing. <b><i>Results:</i></b> Serial plasma samples (<i>n</i> = 187) from 63 patients were analyzed, and 44 patients showed circulating tumor DNA (ctDNA). <i>EGFR</i> mutations were detected in 36 of the 44 patients at baseline (81.8%). <i>EGFR</i> mutations were no longer detected in 19 of 36 shedders (52.8%) at 2 months after EGFR-TKI treatment and rebounded with resistant <i>EGFR</i> mutations (T790M or C797S) at progression. Other driver mutations such as <i>KRAS</i> G12D and <i>BRAF</i> V600E were found at baseline regardless of tissue <i>EGFR</i> status, suggesting tumor heterogeneity. Detection of ctDNA (shedder) at baseline associated with poor overall survival (<i>p</i> = 0.04) compared to nonshedder. Furthermore, in patients showing <i>EGFR</i> mutations in plasma at baseline, the clearing rate of those during the first 8 weeks of treatment served as a positive predictor for clinical outcome. <b><i>Conclusion:</i></b> Longitudinal liquid biopsies capture spatial and temporal heterogeneity underlining resistance to EGFR-TKIs in NSCLC. Thus, ctDNA monitoring during EGFR-TKI treatment is useful for detecting resistance mutations or predicting response. Dense serial monitoring using blood enables early prediction of treatment failure and provides a window of opportunity for well-timed intervention.