Genomic landscape of clinically acquired resistance alterations in patients treated with KRASG12C inhibitors.

Abstract

Background: Mutant-selective inhibitors of KRAS^G12C (KRAS^G12Ci) have demonstrated efficacy in KRAS^G12C cancers. However, resistance invariably develops, resulting in short-lived responses. We aimed to define the genomic landscape of acquired resistance to KRAS^G12Ci and to elucidate whether novel classes of KRAS inhibitors can overcome these resistance mechanisms.

Methods: To assess clinical frequencies of acquired resistance alterations, we evaluated genomic sequencing data from post-progression cell-free DNA samples in patients treated with KRAS^G12Ci at two US cancer centers, alongside data from six previously published studies. Cell viability assays using engineered cell models were employed to functionally validate candidate resistance drivers and to evaluate novel classes of KRAS inhibitors.

Results: A total of 143 patients were analyzed. Most patients had non-small cell lung cancer (NSCLC, n=68) or colorectal cancer (CRC, n=58) and were treated with single-agent KRAS^G12Ci (n=109) or combined with anti-EGFR antibodies (n=30). RAS/MAPK alterations emerged in 46% of patients (n=66), with 39% developing ≥1 new KRAS alteration (n=56) and 23% (n=33) showing multiple concurrent alterations. The genomic landscape of acquired alterations included KRAS activating mutations (25% of patients), KRAS amplifications (22%), RAF/MAPK mutations/fusions (21%), KRAS switch-II pocket mutations (14%) and NRAS/HRAS mutations (8%). Notably, the proportion of patients with ≥1 acquired RAS/MAPK alteration was significantly higher in CRC compared to NSCLC (69% vs. 26%, p<0.001). Functional studies confirmed most alterations as resistance drivers. Sotorasib, adagrasib, and divarasib demonstrated distinct activity against KRAS switch-II pocket mutations, yet all were responsive to the RAS(ON) G12C-selective tri-complex inhibitor RM-018. The KRAS-selective inhibitor Pan-KRAS-In-1 effectively targeted KRAS activating mutations, and the RAS(ON) multi-selective tri-complex inhibitor RMC-7797 demonstrated high potency across all RAS alterations.

Conclusion: Acquired RAS/MAPK alterations are recurrent drivers of resistance to KRAS^G12Ci detected in CRC and, less frequently, in NSCLC. Preclinical data suggest that novel (K)RAS inhibitors may overcome many of these resistance alterations.