Contrasting Clinical Implications of NRAS Vs. KRAS Mutations in AML – Only KRAS Mutations Are Associated with Adverse Outcome

Introduction:

Acute myeloid leukemia (AML) is a genetically and clinically heterogenous malignancy with highly prevalent structural alterations and less common sequence variants in children and adolescents. NRAS and KRAS mutations are the most prevalent mutations in pediatric AML with variable impacts on outcome. Iyoda et al recently demonstrated an association of KRAS mutation with adverse outcome in KMT2A-r AML (Leukemia 2024). They further suggested that only mutations involving the G12, but not G13 codon are associated with outcome. In this study, we provide a comprehensive evaluation of NRAS and KRAS mutations in a large cohort of genomically well-studied children and young adults with AML and specifically evaluate the role of the involved codon in outcome determination, especially in KMT2A-r AML.

Methods:

A combination of whole genome, targeted capture, transcriptome, and Archer panel sequencing were performed on samples from patients enrolled in the Children’s Oncology Group (COG) studies for pediatric AML. A retrospective analysis was performed to determine the prognostic significance of NRAS and KRAS mutations based on their codon in patients with KMT2A-r AML (n=371) and other AML (n=1670).

Results:

Of the 2041 patients tested, 548 patients (26.8%) had NRAS mutations and 240 patients had KRAS mutations (11.8%) with a cumulative incidence of 35.3% for all RAS mutations. Patients with and without NRAS mutations, regardless of the involved codon, had similar outcomes with no differences in event-free survival (5-year EFS; 46% vs. 45%, p=0.55) or overall survival (OS; 66% vs. 63%, p=0.20). There was a similar prevalence of NRAS mutations in patients with KMT2A-r compared to those without (26.3% vs. 24.4%, p=0.486). KMT2A-r AML patients with and without NRAS mutations had similar EFS (31% vs. 34%, p=0.65) and OS (54% vs. 53%; p=0.84).

In contrast to NRAS mutations, KRAS mutations were enriched in patients with KMT2A-r AML compared to those without (19.8% vs. 6.3%, <0.0001). KRAS mutations were further categorized by codon and amino acid alterations (G12, G13 and Q61 variants) and outcomes were compared.

Patients with KRAS mutations have 5-year EFS of 40% and OS of 60%. Patients with G12 and G13 mutations had significantly worse EFS (20%, p=0.019 and 25% (p=0.007, respectively) and OS (30%, p=0.0077 and 40% p=0.021, respectively), whereas patients with Q61 had no significant difference in EFS (45%) or OS (70%).

To mitigate possible bias driven by high risk MLL fusion partners, we excluded KMT2A-r patients with known high-risk fusions (KMT2A-AFF1, KMT2A-MLLT4, KMT2A-ABI1, KMT2A-MLLT10 and KMT2A-MLLT1) and reanalysed impact of KRAS mutations only the standard risk KMT2A-r AML cohort. Of the 198 patients with standard risk KMT2A-r AML, 23 (12%) had KRAS G12 (n=12) or G13 (n=11) mutations. In this setting, the G12 and G13 cohort was combined given the small sample size. In this standard risk fusion cohort, patients with KRAS G12 or G13 mutations had significantly inferior outcomes compared to KRAS WT with EFS of 20% (0.005) and OS of 35% (0.0004), demonstrating that KRAS can help identify a cohort of high risk KMT2A-r cases that are considered standard risk by current classification schema.

Discussion:

In this study we clarify prognostic implications of NRAS vs. KRAS mutations in AML. We establish a lack of association of NRAS mutations with outcome. More importantly we show high enrichment of KRAS mutations in KMT2A-r AML and an association with worse outcomes in this high-risk population. We further delineate the critical role of KRAS mutation codon and ability of KRAS G12/13 (not Q61) to define a high-risk population within the standard-risk KMT2A-r AML cohort. This data will help refine risk stratification for therapeutic purposes.