Clinical characteristic and outcomes of KRAS G13 mutant non-small cell lung cancers.

Abstract

e21017 Background: KRAS is the most commonly mutated oncogene in lung adenocarcinoma. Different KRAS mutations (mts) have unique prognostic and therapeutic implications. New drugs to target KRAS G12C have raised interest in studying the unique clinical and molecular characteristics of the different KRAS subtypes. We sought to evaluate the clinical outcomes of KRAS G13C/D (*) mutant (mut) non-small cell lung cancers (NSCLC) patients (pts), compared to those with G12C and other non-G12C mutations. Methods: We identified 255 KRAS mt NSCLC pts at Brown University-Lifespan Cancer Institute that were diagnosed between 06/2016 to 12/2021. Demographics, therapies and outcomes were retrospectively recorded. Progression-free (PFS) and overall survival (OS) was compared using log-rank test. For clinical relevance, we divided pts into G12C, G13*, and other non-G12C groups. Treatments (tt) received were divided into chemotherapy alone and immune-checkpoint inhibitor (IO) containing regimens. Results: Across the KRAS cohort (n = 255), median age was 71 years and 62% were females. Most pts had advanced disease (85%) at diagnosis (dx) and > 90% were smokers. Most common KRAS mt was G12C (42%), and 7% of pts had G13*. Compared to other KRAS mut groups, G13 * had worse PFS counted from dx (median 4 vs 12 m, p = .037). When comparing outcomes of any tt among pts with G13*, G12C and other non-G12C KRAS mut, G13 group had the worst PFS (2.1 vs 10.3 vs 6.4 m, p = .004) and OS (4.1 vs 27.8 vs 18.1 m, p = .027). When treated with IO-based regimens, pts with G13 had also worse outcomes than those with other KRAS mts: PFS 1.8 vs 7.3 m, respectively, (p < 0.001) and OS 1.8 vs 16.5 m, respectively (p = .006). Within the G13 group, pts who received IO-based regimens had worse PFS (1.8 vs 7.9 m, p = .037) and OS (1.8 vs 16.1 m, p = .093), compared to chemotherapy alone. While response to all systemic therapies was poor in G13* group (disease control rate [DCR] 10%, p = .004), DCR was particularly low with IO-based tt in G13* (0%) compared to other groups (p = .023). Conclusions: Pts with KRAS G13C/D had poor survival and response to therapies in comparison with G12C and other non-G12C KRAS mts, particularly with IO-containing regimens. Recently (Judd et al, MCT-21-0201), STK11 and KEAP1 were reported to be frequently co-mutated with KRAS G13* (37.3 and 12.6%) compared to other KRAS mts, and this genotype may confer primary resistance to IO. We hypothesize that the poor outcomes and immune resistance in G13* mut may be secondary to higher prevalence of STK11 and KEAP1 co-mutations. Larger data sets should confirm our observations and correlate them with PD-L1 scores and co-mutations.[Table: see text]