Sotorasib (KRAS G12C) Improves CNS PFS vs Docetaxel in Pretreated NSCLC

CodeBreaK 200 (phase 3; global; NCT04303780) randomized previously treated KRAS G12C-mutated advanced NSCLC (~10–13% of non-squamous NSCLC) to Amgen's sotorasib 960 mg daily vs docetaxel 75 mg/m² q3w; this exploratory post-hoc analysis assessed patients with treated, stable brain metastases at baseline (n=69: 40 vs 29) by blinded independent central review using modified RANO-BM. With 20.0-month median follow-up, CNS PFS favored sotorasib 9.6 vs 4.5 months (HR 0.43, 95% CI 0.20–0.92; p=0.02); among baseline lesions ≥10 mm, ≥30% intracranial shrinkage occurred in 33.3% vs 15.4%. Safety in this subgroup mirrored the overall trial (grade≥3 TRAEs 33% vs 40%), and the CNS finding complements the global CodeBreaK 200 PFS advantage (5.6 vs 4.5 months; HR 0.66) over docetaxel; for context, BMS’s adagrasib has prospective intracranial activity in untreated CNS mets (IC ORR 42%, mPFS 5.4 months), though populations differ.