The prevalence and prognostic significance of <i>KRAS G12C</i> mutation in metastatic colorectal cancer.

Abstract

289 Background: Kristen rat sarcoma virus ( KRAS ) mutations are reported in 40-45% of colorectal cancer (CRC), and they are associated with poor prognosis. KRAS G12C mutation is reported in 3-4% of CRC patients and may be associated with worse prognosis compared to RAS non-G12C mutations. Here, we explored the prevalence and prognostic significance of KRAS G12C mutation compared to RAS non-G12C and BRAF V600E in a cohort of CRC patients from a tertiary cancer center in the United States (US). Methods: The cBioCancer Genomics Portal was used to obtain genomic data. The prevalence of KRAS G12C mutation, RAS non-G12C ( KRAS G12D, KRAS G13D, KRAS G12V, KRAS A146T, KRAS G12A, KRAS G12S, KRAS G13C, KRAS V14I, KRAS Q61H, KRAS G12R, KRAS Q61R, KRAS K117N, NRAS Q61K , NRAS Q61R , NRAS Q61L , NRAS Q61H , NRAS G13D , NRAS G13C NRAS G13R NRAS G12D , NRAS G12C , NRAS G12V, NRAS G12V and NRAS G60V ) mutations and BRAF V600E mutations was calculated. The median overall survival (mOS) was explored in CRC (MSK, Cancer Cell 2018 cohort). Patients in this cohort were not treated with KRAS G12C targeted therapy. Survival analysis was performed using the Kaplan-Meier method with log-rank test comparisons. Results: In this cohort of patients with stage IV CRC (N=1134), the prevalence of KRAS G12C mutation in patients with stage IV CRC was 3.1%. In patients with RAS non-G12C mutations, BRAF V600E mutation, and RAS/BRAF V600E wild-type (WT), the prevalence was 41.7%, 7.5%, and 47.7% respectively. The mOS in KRAS G12C mutation, RAS non-G12C mutations, BRAF V600E mutation, and RAS/BRAF V600E WT was 65 months (m), 55m, 19m and 105m respectively. Compared to RAS non-G12C , the mOS of CRC patients with KRAS G12C was (65m vs 55m, p=0.54) . Compared to patients with BRAF V600E mutation, the mOS was better in patients with KRAS G12C (65m vs. 19m, p &lt;0.0001) and in patients with RAS non-G12C (55m vs. 19m, p &lt;0.0001). Conclusions: The prevalence of KRAS G12C mutation in this cohort was relatively similar to other reports at 3.1%. Consistent with several other reports, there was no statistically significant difference in the mOS between patients with KRAS G12C mutation and patients with RAS non-G12C mutations. Patients with BRAF V600E mutations had worse mOS compared to patients with KRAS G12C mutation and RAS non-G12C mutations. The reported high mOS in this cohort is possibly due to selection bias (patients able to afford and receive treatment at this tertiary center).