RESUMEN
Purpose@#We aimed to evaluate whether the addition of pemetrexed is effective in improving progression-free survival (PFS) in epidermal growth factor receptor (EGFR)–mutated patients with or without concomitant alterations. @*Materials and Methods@#This multicenter clinical trial was conducted in China from June 15, 2018, to May 31, 2019. A total of 92 non–small cell lung cancer (NSCLC) patients harboring EGFR-sensitive mutations were included and divided into concomitant and non-concomitant groups. Patients in each group were randomly treated with EGFR–tyrosine kinase inhibitor (TKI) monotherapy or EGFR-TKI combined with pemetrexed in a ratio of 1:1. PFS was recorded as the primary endpoint. @*Results@#The overall median PFS of this cohort was 10.1 months. There were no significant differences in PFS between patients with and without concomitant and between patients received TKI monotherapy and TKI combined with pemetrexed (p=0.210 and p=0.085, respectively). Stratification analysis indicated that patients received TKI monotherapy had a significantly longer PFS in non-concomitant group than that in concomitant group (p=0.002). In concomitant group, patients received TKI combined with pemetrexed had a significantly longer PFS than patients received TKI monotherapy (p=0.013). Molecular dynamic analysis showed rapidly emerging EGFR T790M in patients received TKI monotherapy. EGFR mutation abundance decreased in patients received TKI combined chemotherapy, which supports better efficacy for a TKI combined chemotherapy as compared to TKI monotherapy. A good correlation between therapeutic efficacy and a change in circulating tumor DNA (ctDNA) status was found in 66% of patients, supporting the guiding role of ctDNA minimal residual disease (MRD) in NSCLC treatment. @*Conclusion@#EGFR-TKI monotherapy is applicable to EGFR-sensitive patients without concomitant alterations, while a TKI combined chemotherapy is applicable to EGFR-sensitive patients with concomitant alterations. CtDNA MRD may be a potential biomarker for predicting therapeutic efficacy.