Treatment prioritization of chemotherapy-anchored regimens after EGFR-TKI failure in EGFR-mutant NSCLC: a systematic review, pairwise meta-analysis and Bayesian network meta-analysis with probabilistic multi-criteria decision analysis.

In advanced EGFR-mutant non-small cell lung cancer (NSCLC) after EGFR-TKI failure, chemotherapy alone offers limited benefit, while heterogeneous efficacy and safety across combination regimens complicate treatment selection. We systematically compared chemotherapy-anchored regimens in this setting to establish an evidence-informed treatment prioritization framework. Following PRISMA 2020 guidelines, we searched PubMed, Embase, Cochrane Library, and oncology conference for randomized controlled trials evaluating patients with advanced EGFR-mutant NSCLC progressing after EGFR-TKI therapy, comparing experimental regimens (with or without chemotherapy) against platinum-based doublet chemotherapy (chemotherapy-anchored, chemo-anchored) for efficacy and safety outcomes. Primary outcomes were progression-free survival (PFS) and overall survival (OS), analyzed using hazard ratios (HRs). Fixed-effects pairwise meta-analyses and Bayesian random-effects network meta-analyses were performed to evaluate 8 treatment strategies: chemotherapy alone (Chemo, platinum-based doublet chemotherapy); chemotherapy plus immune checkpoint inhibitors (Chemo_ICI); chemotherapy plus anti-angiogenic therapy (Chemo_Antiangio); chemotherapy plus immune checkpoint inhibitors and anti-angiogenic therapy (Chemo_ICI_Antiangio); chemotherapy plus ivonescimab (Chemo_Ivo); sacTMT (sacituzumab tirumotecan); chemotherapy plus amivantamab (Chemo_Ami); and chemotherapy plus amivantamab and lazertinib (Chemo_Ami_Laz). Probabilistic multi-criteria decision analysis (pMCDA) integrated multi-criteria outcomes under three clinically informed weighting schemes (survival-focused, balanced, and safety-focused). All analyses were conducted using R software. Eleven RCTs (n=3,606) were analyzed. In pairwise analyses versus Chemo, pooled results showed significant PFS benefit for Chemo_Ivo (HR 0.50, 95% CI 0.41-0.60), Chemo_ICI_Antiangio (HR 0.55, 95% CI 0.45-0.68), and Chemo_ICI (HR 0.77, 95% CI 0.67-0.88), while Chemo_Ivo also improved OS (HR 0.76, 95% CI 0.64-0.91). Under random-effects models, HR point estimates remained identical; significance was maintained in overall analyses but lost in several subgroups as confidence intervals encompassed 1. In network meta-analysis, Chemo_Ami_Laz (HR 0.44, 95% CrI 0.32-0.61), Chemo_Ami (HR 0.48, 95% CrI 0.33-0.70), sac-TMT (HR 0.49, 95% CrI 0.35-0.68), and Chemo_Ivo (HR 0.49, 95% CrI 0.38-0.64) showed the greatest PFS benefit versus Chemo. For OS, only sac-TMT (HR 0.60, 95% CrI 0.41-0.90) and Chemo_Ivo (HR 0.76, 95% CrI 0.60-0.98) were associated with a significant survival advantage. Chemo_Ami_Laz (OR 12.06, 95% CrI 3.64-53.50) and Chemo_Ami (OR 3.71, 95% CrI 1.12-12.19) were associated with a higher risk of grade ≥3 TRAEs. In pMCDA integrating efficacy and safety, sac-TMT, Chemo_Ami, and Chemo_Ivo consistently ranked as the top three strategies across all weighting schemes. Integrating efficacy and safety, sac-TMT, Chemo_Ami, and Chemo_Ivo appear to be the most favorable treatment options after EGFR-TKI failure in EGFR-mutant NSCLC. https://www.crd.york.ac.uk/PROSPERO/, identifier CRD420251268510.