Divarasib Extends Survival by 5.7 Months Over Approved KRAS Inhibitors in Phase III NSCLC Trial

The open-label trial randomized 338 patients across 26 countries to divarasib or the two approved KRAS G12C inhibitors after progression on prior therapy. Median progression-free survival reached 9.8 months with divarasib versus 5.7 months in controls, while overall survival improved to 17.7 versus 12 months at 14.3 months median follow-up. Objective response rate doubled to 58.4% with responses lasting 11.9 months versus 6.9 months, and grade 3 or higher adverse events occurred in 25% versus 35% despite longer exposure. Prior first-generation inhibitors like sotorasib showed modest benefit in CodeBreaK 100 with frequent resistance via secondary KRAS alterations; divarasib's tighter binding appears to delay such escape. The head-to-head design addresses a gap left by single-arm approvals, yet the open-label structure and exclusion of untreated brain metastases introduce potential bias and limit generalizability to real-world populations with CNS involvement. Context from earlier observational cohorts indicates KRAS G12C occurs in roughly 14% of NSCLC, where sequential targeted therapy remains rare. Divarasib's tolerability edge may support longer treatment duration, but whether the survival gain persists in patients with co-mutations such as STK11 or KEAP1 requires subgroup reporting not yet detailed. Next steps include regulatory review and trials combining divarasib with checkpoint inhibitors or chemotherapy; confirmatory blinded studies with broader CNS eligibility will determine if the observed benefit holds in less selected populations.