Impact of post-resistance molecular retesting strategies on treatment timeliness and clinical outcomes in EGFR-mutant advanced NSCLC: a real-world study

Abstract

To evaluate the association of ctDNA-based molecular retesting with treatment timeliness and clinical outcomes in patients with advanced EGFR-mutant NSCLC after first-line EGFR-TKI resistance. We retrospectively analyzed patients with advanced EGFR-mutant NSCLC who had radiologically confirmed progression after first-line EGFR-TKIs between 2019 and 2024. Patients were classified as no retesting (n=86), tissue NGS (n=94), or ctDNA testing (n=49). Primary endpoints were second-line progression-free survival (2L-PFS) and overall survival from initiation of second-line therapy (2L-OS). Secondary endpoints included time to treatment initiation (TTI) and treatment patterns. Survival was analyzed using Kaplan–Meier and log-rank methods; Cox models were used for exploratory subgroup analyses. A 1:1:1 propensity score–matched analysis and sensitivity analyses restricted to patients who initiated second-line therapy within 28 or 30 days after progression were performed. Among 229 patients, median TTI was shorter with ctDNA than tissue NGS (11.0 vs. 20.5 days, P < 0.001). Subsequent targeted therapy was most frequently used in the ctDNA group (95.9%), followed by the tissue NGS (89.4%) and no-retesting groups (81.4%; P = 0.039). After matching, median 2L-PFS was 5.0, 9.0, and 9.0 months in the no-retesting, tissue NGS, and ctDNA groups, respectively (overall P = 0.005). Median 2L-OS was 12.0, 20.0, and 19.0 months, respectively (overall P = 0.074). No significant differences in 2L-PFS or 2L-OS were observed between ctDNA and tissue NGS, and findings were consistent in 28- and 30-day sensitivity analyses. Compared with tissue NGS, ctDNA testing was associated with shorter TTI, with no significant differences in 2L-PFS or 2L-OS. The ctDNA group had the highest observed targeted therapy use.