Activity of adagrasib (MRTX849) in patients with KRAS^G12C-mutated NSCLC and active, untreated CNS metastases in the KRYSTAL-1 trial.

LBA9009 Background: CNS metastases (mets) occur in 27–42% of NSCLC harboring KRAS G12C mutations and are associated with poor prognosis. Adagrasib (ada), an investigational agent, is a KRAS G12C inhibitor that irreversibly and selectively binds KRAS G12C , locking it in its inactive state, and is optimized for favorable PK properties, including a long half-life (̃24 h) and dose-dependent PK. Ada has demonstrated dose-dependent CNS penetration and intracranial (IC) tumor regression in preclinical models of CNS mets. Methods: KRYSTAL-1 (NCT03785249) is a multicohort Phase 1/2 study evaluating ada as monotherapy or in combination with selected agents in patients (pts) with advanced solid tumors harboring a KRAS G12C mutation. In a Phase 1b cohort, pts with KRAS G12C -mutant NSCLC and active, untreated CNS mets were treated with ada 600 mg BID. The objectives for this Phase 1b cohort were to evaluate safety and clinical activity, including systemic and IC objective response rate (ORR) by blinded independent central review (BICR), duration of response (DOR), progression-free survival (PFS), and overall survival (OS). Systemic responses were assessed by RECIST 1.1; IC responses were assessed by modified RANO criteria and IC RECIST. Cerebrospinal fluid (CSF) was collected when feasible and was used to measure ada concentrations for optional evaluation. Results: As of 31 December 2021, 25 pts with NSCLC were enrolled and treated. Median follow-up was 6.6 months at the cutoff date. Median age was 66 years, 52% were female, 28%/72% had ECOG PS 0/1, and median lines of prior systemic therapy was one (range 0–4+). IC ORR per modified RANO criteria by BICR was 31.6% (6/19; 3 CRs, 2 PRs, 1 unconfirmed PR); IC disease control rate (DCR) was 84.2% (16/19, including 10 SDs). Median IC DOR was not reached (95% CI 4.1–NE); median IC PFS was 4.2 months (95% CI 3.8–NE). CSF samples were obtained for two pts for whom regression of CNS mets was observed and ada CSF concentrations exceeded the CNS penetrance partition coefficients observed for other therapies with demonstrated CNS penetration and CNS antitumor activity (K p,uu 0.47). Systemic ORR by BICR was 35.0% (7/20), DCR was 80.0% (16/20) and median DOR was 9.6 months (95% CI 2.7–9.6); median PFS was 5.6 months (95% CI 3.8–11.0). For all pts enrolled, OS was immature, and the median had not been reached at the time of analysis. Safety was consistent with that previously reported with ada; any grade TRAEs occurred in 96% of pts, grade 3 TRAEs in 36%, and there were no grade 4/5 TRAEs. Further data describing IC activity assessed by IC RECIST will be presented. Conclusions: Ada was well tolerated, and these preliminary data demonstrate CNS penetration and encouraging IC activity in pretreated pts with NSCLC harboring a KRAS G12C mutation and active, untreated CNS mets. These are the first clinical data demonstrating CNS-specific activity of a KRAS G12C inhibitor in this population. Clinical trial information: NCT03785249.