Background
Systematic review and Bayesian network meta-analysis of 14 RCTs (N=3,177) comparing 8 regimens for advanced EGFR-mutated NSCLC post-TKI progression: CT+ivonescimab, CT+amivantamab+lazertinib, CT+immunotherapy+bevacizumab, CT+amivantamab, CT+bevacizumab, CT+immunotherapy, CT alone, immunotherapy alone.
Interventions and follow up
Treatment: Network meta-analysis of randomized trials in EGFR-mutated NSCLC after progression on first-line EGFR-TKI therapy
Primary endpoint: PFS and OS comparisons across post-TKI regimens
mFollow up: Variable across included trials
Primary endpoint: PFS and OS comparisons across post-TKI regimens
mFollow up: Variable across included trials
Results
Analysis: Surface under cumulative ranking curve (SUCRA) methodology
PFS ranking: 1st CT+ivonescimab; 2nd CT+amivantamab+lazertinib; 3rd CT+immunotherapy+bevacizumab; 4th CT+amivantamab
OS ranking: 1st CT+amivantamab; 2nd CT+ivonescimab (similar); comparisons not statistically significant (immature data)
ORR/DCR: best with CT+amivantamab, then CT+amivantamab+lazertinib
Grade ≥3 AEs: highest with CT+amivantamab+lazertinib, then CT+amivantamab
PFS ranking: 1st CT+ivonescimab; 2nd CT+amivantamab+lazertinib; 3rd CT+immunotherapy+bevacizumab; 4th CT+amivantamab
OS ranking: 1st CT+amivantamab; 2nd CT+ivonescimab (similar); comparisons not statistically significant (immature data)
ORR/DCR: best with CT+amivantamab, then CT+amivantamab+lazertinib
Grade ≥3 AEs: highest with CT+amivantamab+lazertinib, then CT+amivantamab
Adverse events
CT+amivantamab+lazertinib: elevated toxicity (diarrhea, mucositis)
CT+amivantamab: moderate toxicity
CT+immunotherapy+bevacizumab: lower toxicity
CT+ivonescimab: moderate-to-low toxicity
CT+amivantamab: moderate toxicity
CT+immunotherapy+bevacizumab: lower toxicity
CT+ivonescimab: moderate-to-low toxicity
Conclusions
CT+ivonescimab is optimal for PFS with acceptable safety; CT+amivantamab+lazertinib has the highest ORR/DCR but elevated grade ≥3 AEs; CT+amivantamab and CT+immunotherapy+bevacizumab are viable alternatives balancing efficacy and safety.
Key Limitations
Indirect cross-trial comparisons (no head-to-head data for most pairs); immature OS data with non-significant rankings; heterogeneity in trial populations, lines of therapy, and endpoints; ivonescimab data largely from Chinese trials; SUCRA rankings can overstate small differences. Hypothesis-generating for sequencing.
Clinical Context
Post-TKI progression in EGFR+ NSCLC remains an area without a single dominant standard. Per ASCO and ESMO, options after osimertinib include platinum-doublet chemotherapy with or without amivantamab (MARIPOSA-2) or antiangiogenic/immunotherapy combinations. This NMA informs sequencing but cannot replace randomized head-to-head trials; ivonescimab combinations are emerging, predominantly from Asian data.