2253P - RAS-precision medicine trans-atlantic partnership: Comparative analysis of KRAS codon 12 and 13 mutations in non-small cell lung cancer

Background

RAS is the most commonly mutated oncogene in cancer, with KRAS mutated in ∼30% of non-small cell lung cancer (NSCLC). The most common KRAS mutation in lung-adenocarcinoma is KRAS codon 12 followed by 13. Newly developed drugs to target KRAS-G12C have highlighted the importance of studying the unique molecular and functional mechanisms of individual allelic subtypes.

Methods

We developed a doxycycline inducible panel of isogenic cell lines (MLE-12: alveolar type II cells) expressing flag-tagged KRAS WT or KRAS mutant: G12C, G12D, G13C and G13D. Cells were characterised in 3D using proliferation, viability, transformation and signalling assays. A panel of “RASless” mouse embryonic fibroblasts (MEFs) expressing KRAS WT, G12C, G12D and G13D were also evaluated. For clinical validation, we developed a clinical database (RAS-PM), to interrogate co-mutational frequencies and pathogenicity.

Results

Expression of KRAS WT, 12C and 12D in MLE-12 cells resulted in an increase in proliferation, compared to a decrease or plateau observed with KRAS 13C or 13D expression. Furthermore, there was increased viability in cells expressing KRAS codon 12 mutations compared to WT or codon 13 mutations. All KRAS mutant MLE-12 cells expressed markers of transformation (EPCAM and CD44), including codon 13 mutants. In addition, there were differences in activation of AKT signalling between KRAS G12D and G13D expressing cells. Therapeutic vulnerabilities to a novel active-state selective RAS (‘ON’) inhibitor were assessed, where the most and least sensitive cells were those expressing G12C or G12D, respectively. RAS-PM identified that KRAS G13D was more prevalent in NF1 mutant cancers vs. NF1 wildtype (NF1 MT: 6/24, 25%; vs. NF1 WT: 4/281, 1.4%; p<0.0001). Pathogenic NF1-mutations were identified in 5/6 (83%) of G13D/NF1 co-mutant cases.

Conclusions

Our findings highlight molecular and functional differences across the panel of KRAS mutations evaluated demonstrating differences in proliferation, viability, and signalling. These preclinical results indicate unique therapeutic vulnerabilities across allelic subgroups. RNA-sequencing and kinome profiling will allow us to further characterise these mutations in NSCLC.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Revolution medicines.

Disclosure

C.R. Lindsay: Financial Interests, Institutional, Advisory Role: Amgen; Financial Interests, Institutional, Research Funding: Revolution Medicines. All other authors have declared no conflicts of interest.