Dynamic Plasma KRAS ctDNA Profiling Identifies Treatment Response and Nascent Resistance in Metastatic Non-Small-Cell Lung Cancer: A Pilot Study

Background: KRAS mutations are common oncogenic drivers in metastatic lung cancer (mLC) and may have prognostic and therapeutic implications. Circulating tumor DNA (ctDNA) analysis offers a minimally invasive approach for detecting tumor-specific alterations and monitoring tumor dynamics. This pilot study investigated the concordance between tissue and plasma KRAS mutational status and evaluated its clinical significance. Methods: Forty patients with mLC (25 males, 15 females; mean age 70.8 years) were included. KRAS mutations were analyzed in formalin-fixed paraffin-embedded tumor tissue using next-generation sequencing and in baseline plasma ctDNA using the Biocartis Idylla™ system. Concordance was assessed using Cohen’s kappa statistics. Associations with clinicopathological characteristics and survival outcomes (overall survival (OS), progression-free survival (PFS), and time to progression (TTP)) were examined. Longitudinal ctDNA monitoring was performed in a subset of patients. Results: KRAS mutations were detected in 14/36 (38.89%) of tumor tissues and 12/40 (30.00%) of plasma samples, predominantly involving codon 12, with G12C as the most frequent variant. Tissue–plasma concordance was assessed in the 36 patients with matched samples and was substantial (κ = 0.64), while mutation subtype showed almost perfect agreement among concordant cases (percentage of agreement: 88.9%, k = 0.80, p = 0.002, 95% CI: 0.4400–1.0000). No significant associations were observed with clinicopathological variables. Longitudinal analysis revealed dynamic ctDNA changes, with molecular progression occasionally preceding clinical progression. Patients with KRAS-mutated ctDNA had shorter PFS and OS compared to those with wild-type ctDNA (median PFS: 134 vs. 146 days; median OS: 163 vs. 193 days). Conclusions: Plasma ctDNA detection of KRAS mutations identifies a poor-prognosis subgroup of mLC patients and provides dynamic insights into tumor evolution. Serial ctDNA monitoring may anticipate disease progression and support real-time assessment of treatment response, highlighting its potential role in translational research and precision oncology.