285 Background: The optimal treatment strategy following resistance to epidermal growth factor receptor–tyrosine kinase inhibitors (EGFR-TKIs) in patients with EGFR-mutated non–small cell lung cancer (NSCLC) remains controversial. Although the combination of atezolizumab, bevacizumab, carboplatin, and paclitaxel (ABCP) has demonstrated clinical efficacy, real-world comparative data versus carboplatin, pemetrexed, and bevacizumab (CPemB) are limited. Methods: We conducted a multicenter retrospective observational study across eight institutions in Japan between April 2012 and June 2025. Patients with EGFR-mutated NSCLC who developed resistance to EGFR-TKIs and subsequently received either CPemB or ABCP were included. Treatment efficacy, survival outcomes, and adverse events were assessed. To account for baseline imbalances, inverse probability of treatment weighting (IPTW) based on propensity scores was applied. Results: A total of 119 patients were included. The CPemB group comprised 56 patients and the ABCP group comprised 63 patients. The objective response rate was significantly lower in the CPemB group than in the ABCP group (32.1% vs. 57.1%; p = 0.0095). In the unadjusted analysis, median progression-free survival (PFS) was 7.62 months with CPemB and 6.93 months with ABCP, whereas median overall survival (OS) was 17.5 months with CPemB and 14.55 months with ABCP. After IPTW adjustment, no significant differences in PFS or OS were observed, consistent with the unadjusted findings. Distinct safety profiles were identified: thrombocytopenia and hepatic dysfunction occurred more frequently with CPemB, whereas febrile neutropenia, rash, and peripheral neuropathy were more common with ABCP. Conclusions: In patients with EGFR-mutated NSCLC after EGFR-TKI resistance, ABCP achieved a higher objective response rate than CPemB. Survival outcomes were comparable between the two regimens. Given their distinct toxicity profiles, treatment selection should be individualized according to patient characteristics and preferences. Larger prospective studies are warranted to validate these findings.
Nishimura et al. (Tue,) studied this question.