Liquid biopsy as tool to monitor and predict clinical benefit from chemotherapy (CT) and immunotherapy (IT) in advanced non-small cell lung cancer (aNSCLC): A prospective study

Abstract

Background: Detecting tumor-specific genetic alterations in plasma has the potential to monitor biological effects of treatment. Aim of the study is to validate the method in EGFR-ALK-ROS1 wild-type (wt) aNSCLC and explore the correlation of variation of mutations in plasma with outcome. Method(s): aNSCLC patients (pts) undergoing systemic anti-cancer treatment were prospectively enrolled in mono-institutional trial MAGIC-1. Tumor DNA was screened through Mass Array (Sequenom) or Next-generation sequencing (NGS). Plasma samples were collected at baseline (T1), after the first cycle of treatment (T2), at first radiological restaging (T3) and at radiological progression (PD) (T4). Plasma DNA was analyzed with Digital droplet PCR (ddPCR) for KRAS-mutated pts (KRAS-m) and NGS for KRAS-wt pts. Semiquantitative index of fractional abundancy (FA) of mutated allele was used. Result(s): During 2017 we enrolled 161 pts. Here we depict the KRAS-m cohort (n=43), including 11 pts treated with IT. At baseline, KRAS mutation (mut) in cfDNA was detected in 21 samples (49%; 95% CI 33.3-64.5). The predictive value of KRAS mut presence in plasma increased from T1 to T3 and was statistically significant for PD (detected in 18 pts) at T3 with an AUC 0.73 (95% CI 0.55-0.91, p-value=0.0132); sensitivity and specificity for PD at cut-off value of 0.013 FA were 58% (95% CI 28-85) and 89% (95% CI 67-99). Any FA reduction (T1-T3) showed an ability to discriminate between groups (non-PD vs PD) of 85% (95% CI: 72-97, p-value<0.0001) with a sensitivity for detection of non-PD of 67% (95% CI 35-90) and a specificity of 89% (95% CI 67-99). Among pts treated with IT, the predictive value of FA reduction (T1-T3) was 86% (95% CI 58-100, p-value=0.0124), with sensitivity of 100% (95% CI 16-100) and specificity of 86% (95% CI 42-100). The only case treated with IT experiencing death within 12 weeks had FA increase from 0 to 10% at T2. Conclusion(s): Dynamic variations of KRAS mut in plasma significantly correlate with radiological disease control. Early variation of FA of mutated allele potentially individuates pts experiencing worse outcome during IT. Plasma NGS analyses and expansion cohort in pts treated with IT are ongoing.