First-line (1L) divarasib plus pembrolizumab (pembro) in advanced or metastatic KRAS G12C+ non–small cell lung cancer (NSCLC): Results from the Krascendo-170 study.

8510 Background: Krascendo 170 (NCT05789082) is an open-label, phase Ib/II, dose-finding and -expansion study of divarasib, an oral next-generation KRAS G12C inhibitor, with/without other anti-cancer therapies in 1L KRAS G12C + NSCLC. We report the primary analysis for pts receiving divarasib plus pembro in the programmed death-ligand 1 (PD-L1)-positive cohort (tumor cell TC ≥1%; cohort A1) and key preliminary data for the PD-L1-negative cohort (TC <1%; cohort A2). Methods: Eligible pts were aged ≥18 years with untreated advanced or metastatic KRAS G12C + NSCLC and ECOG PS ≤1. All pts received oral divarasib once daily plus intravenous pembro 200 mg once every 3 weeks until loss of clinical benefit, disease progression, study withdrawal or unacceptable toxicity. In cohort A1, pts were randomized 1:1 to receive divarasib 200 or 400 mg. In cohort A2, all pts received divarasib 400 mg. We present data only from pts receiving divarasib 400 mg. Primary endpoints were safety and tolerability. Secondary endpoints included objective response rate (ORR), progression-free survival (PFS) and duration of response (DOR; all investigator-assessed per RECIST v1.1). Results: At data cutoff (Oct 29, 2025), 59 pts were enrolled to receive divarasib 400 mg in cohort A1 and 23 pts in cohort A2. In cohort A1 (median duration of divarasib treatment 10.6 months range 0.3–24.3), treatment-related AEs (TRAEs; divarasib- or pembro-related) were reported in 98% of pts; 65% of pts had grade (gr) 3/4 TRAEs and 0% had gr 5 TRAEs (Table). The most common TRAEs were diarrhea (75%; gr 3/4 16%), nausea (64%; gr 3/4 2%), vomiting (49%; gr 3/4 2%), ALT increase (49%; gr 3/4 20%) and AST increase (47%; gr 3/4 18%). In cohort A1, confirmed ORR was 73%, no patients had progressive disease as their best response, median DOR was not reached and median PFS was 19.3 months (95% CI 12.4–not estimable NE). In cohort A2 (median duration of divarasib treatment 2.8 months range 0.5–8.8), TRAEs were reported in 100% of pts (65% gr 3/4; 0% gr 5) and unconfirmed ORR was 70%. Conclusions: In Krascendo 170, divarasib 400 mg once daily plus pembro had a manageable safety profile. Promising efficacy was shown in pts with advanced KRAS G12C + NSCLC in both PD-L1- positive and -negative cohorts. A global, phase III trial investigating divarasib 400 mg in 1L advanced KRAS G12C + NSCLC (Krascendo 2; NCT06793215) is currently enrolling. Clinical trial information: NCT05789082 . Cohort A1, n=59* ,† Cohort A2, n=23 Median follow-up, months 12.2 3.4 Safety, n (%)Any gr TRAEGr 3/4 TRAE / serious TRAETRAE leading to Dose reduction / interruption Discontinuation 54 (98)36 (65) / 19 (35)28 (51) / 40 (73)14 (25) 23 (100)15 (65) / 3 (13)13 (57) / 14 (61)3 (13) ORR, % (95% CI) 73 (60–84) 70 ‡ (47–87) Median PFS, months 95% CI 19.3 12.4–NE – *3 pts did not receive treatment, 1 pt received divarasib 200 mg; † safety analysis comprised all pts treated with divarasib 400 mg (n=55); ‡ unconfirmed.