8605 Background: MK-1084 is an oral, next-generation, selective KRAS G12C-GDP covalent inhibitor. The phase 1 KANDLELIT-001 study (NCT05067283) showed manageable safety and antitumor activity for MK-1084 monotherapy in KRAS G12C-mut solid tumors and MK-1084 + pembrolizumab (pembro) in KRAS G12C-mut mNSCLC. We report additional NSCLC and preliminary ctDNA data from this study. Methods: Pts had confirmed KRAS G12C-mut, RECIST-measurable disease and ECOG PS 0-1. Pts with any advanced solid tumor and ≥1 prior systemic therapy received MK-1084 monotherapy 25-800 mg/d in arms 1 and 3. Pts with previously untreated mNSCLC and PD-L1 TPS ≥1% received MK-1084 25-400 mg/d + pembro 200 mg Q3W in arm 2 dose escalation and expansion cohorts. Pts with previously untreated nonsquamous mNSCLC received MK-1084 50-200 mg/d + pembro 200 mg, carboplatin, and pemetrexed Q3W in arm 4. Dose-limiting toxicities (DLTs), AEs, and AEs leading to discontinuation were the primary endpoints; ORR, DCR, and PFS per RECIST v1.1 by investigator review were secondary. KRAS G12C variant allele fraction (VAF) and maximum somatic allele frequency (MSAF) in ctDNA were assessed in serial blood samples collected from 23 pts in arm 1 using the Guardant Health OMNI panel. Results: There were 99 pts in arms 1+3 (21 with NSCLC), 34 in arm 2 escalation cohorts, 26 in arm 2 expansion cohorts, and 24 in arm 4 as of the 12 Aug 2024 data cutoff. Median study follow-up was 14.8 mo, 16.2 mo, 2.5 mo, and 4.1 mo, respectively. DLTs occurred in 1 pt in arm 2 (gr 3 ALT and AST increase) and 1 pt in arm 4 (gr 3 diarrhea). Drug-related AEs occurred in 62% of pts in arms 1+3, 88% of pts in arm 2, and 96% of pts in arm 4, were gr ≥3 in 9%, 33%, and 58%, and led to discontinuation of any drug in 1%, 20%, and 17%. There was 1 drug-related death (myelosuppression and platelet count decrease in arm 2). Rates of drug-related ALT increase (any/gr ≥3) were 16%/3% in arm 1, 33%/10% in arm 2, and 33%/4% in arm 4. Rates of drug-related AST increase (any/gr ≥3) were 17%/3%, 30%/8%, and 25%/4%. Efficacy is shown in the Table. Median KRAS G12C VAF was 14.0% at baseline and 0.9% at week 6; median MSAF was 26.0% and 2.2%, respectively. Conclusions: In pts with KRAS G12C-mut mNSCLC, MK-1084 shows manageable safety and antitumor activity as monotherapy for previously treated disease and in combination with pembro ± chemo as first-line (1L) therapy. The >90% decrease from baseline in KRAS G12C VAF in ctDNA confirms MK-1084 target engagement. The phase 3 KANDLELIT-004 study is evaluating MK-1084 + pembro as 1L therapy for KRAS G12C-mut mNSCLC with PD-L1 TPS ≥50%. Clinical trial information: NCT05067283 . Arm ORR DCR PFS n a % (95% CI) n a % (95% CI) n Med (95% CI), mo 1+3 NSCLC(MK-1084 alone) 21 38 (18-62) 21 76 (53-92) 21 8 (4-NR) 2 escalation(MK-1084 + pembro) 34 74 (56-87) 34 91 (76-98) 34 25 (9-NR) 2 expansion(MK-1084 + pembro) 20 40 (19-64) 20 80 (56-94) 26 NR (NR-NR) 4(MK-1084 + pembro + chemo) 22 41 (21-64) 22 82 (60-95) 24 NR (5-NR) a Pts with ≥1 MK-1084 dose ≥5 wk before data cutoff.