Real-world comparative effectiveness and safety of sotorasib versus adagrasib in previously treated KRAS G12C-mutant non-small cell lung cancer: A propensity score-matched cohort study.

282

Background: Sotorasib and adagrasib are selective KRAS G12C inhibitors approved for previously treated advanced NSCLC. No head-to-head trial exists, and indirect comparisons from pivotal trials (CodeBreaK 200, KRYSTAL-12) are limited by cross-trial heterogeneity. Real-world comparative effectiveness and safety data are needed to inform treatment selection, particularly given distinct toxicity profiles (e.g., QTc prolongation with adagrasib, hepatotoxicity with sotorasib). Methods: Using the TriNetX Global Collaborative Network (171 healthcare organizations), we identified adults with NSCLC (ICD-10: C34.x) who received sotorasib (Cohort A) or adagrasib (Cohort B) as index therapy. The primary outcome was overall survival (OS). Secondary outcomes included hepatotoxicity, diarrhea, nausea/vomiting, QTc prolongation/arrhythmia, fatigue, intracranial progression, and ILD/pneumonitis. Propensity score matching (1:1) adjusted for age, sex, race/ethnicity, brain/bone/liver/visceral metastases, Charlson comorbidity components, and baseline labs (creatinine, albumin, hemoglobin, LDH, ALP). Kaplan-Meier analysis with log-rank testing and Cox proportional hazards regression were performed. Results: After matching, 461 patients per cohort were analyzed. Median follow-up was 289 days (sotorasib) and 162 days (adagrasib). Median OS was comparable: 392 vs 406 days (HR 0.983; 95% CI, 0.812–1.190; P = 0.859). Safety outcomes are shown in Table 1. Adagrasib was associated with significantly higher risk of QTc prolongation/arrhythmia (HR 0.537; 95% CI, 0.346–0.833; P = 0.005), while hepatotoxicity, GI toxicity, ILD, and intracranial progression rates were similar. Conclusions: In this large real-world propensity score-matched analysis, sotorasib and adagrasib demonstrated comparable OS in previously treated KRAS G12C-mutant NSCLC. Adagrasib was associated with significantly higher QTc prolongation/arrhythmia risk, consistent with known pharmacologic differences. These findings support both agents as viable options with treatment selection guided by patient-specific cardiac risk profiles and comorbidities.

Outcomes after propensity score matching (n=461 per cohort).