Prevalence and Clinical Implications of N-RAS Mutations in Childhood AML – A Report From the Children's Oncology Group.

Abstract 3115

Poster Board III-52

Mutations in the RAS proto-oncogenes are frequent in acute myeloid leukemia (AML) and serve as prototypic Class I lesions, initiating key downstream hyper-proliferative signal transduction pathways. N-RAS mutations (N-RASmut) are common in AML, occurring in 10-20% of adult and pediatric AML patients; however their prognostic significance in both adults and children remains disputed. Due to a frequent association with a normal karyotype, delineating the impact of these mutations on outcome may enable appropriate risk-adapted therapeutic approaches. Here, we report on the incidence and prognostic significance of N-RASmut in a cohort of 825 pediatric AML patients treated on 2 recent co-operative group studies, CCG-2961 and COG AAML03P1. In total, of the 825 children with de novo AML who underwent N-RAS mutational analysis, 86 (10%) were positive. Gender, age, race, presence of hepatosplenomegaly and FAB subtype were comparable between patients with and without N-RASmut. There was no statistically significant difference between those with and without N-RASmut with respect to specific chromosome class. FLT3/Internal tandem duplications (FLT3/ITD) were less common in those with N-RASmut (2% vs. 9%, p=0.03). In contrast, nucleophosmin (NPM) mutations were more common in those with N-RASmut (13% vs. 5%, p=0.02). Overall, N-RASmut showed a significant correlation with low risk AML (CBF, CEBPαa, or NPM, p=0.04) and an inverse relationship with high risk disease (-5/5q- or -7, FLT3/ITD/high allelic ratio, p=0.007). Patients with N-RASmut had similar complete remission (CR) rates following two courses of induction chemotherapy compared with non-mutated patients (79% vs. 79%, p=0.92). Those in CR had a similar relapse rate regardless of the presence of N-RASmut (RR 39% vs. 36%, p=0.97). Five year event free survival (EFS) and overall survival (OS) from study entry were also comparable, however N-RASmut patients demonstrated a marked increase in overall treatment related mortality (TRM) (21% vs. 14%, p=0.03), which was maintained when high and low risk patients were excluded (22% vs. 11%, p=0.05). Evaluation of TRM in CCG-2961 compared with COG AAML03P1 demonstrated that increased TRM in N-RASmut was only seen in patients treated on CCG-2961 and TRM was primarily due to invasive fungal and gram negative infections. Interestingly, for patients treated on CCG-2961, who received interval compressed intensive induction therapy, this increase in TRM correlated with a decrease in both OS and disease free survival from CR. Remission specimens from patients with N-RASmut were negative for the mutation, demonstrating that an increased infectious risk could not be attributed to a host polymorphism. We found N-RASmut did not contribute to increased disease recurrence in pediatric AML. Our findings indicate that N-RAS mutations do not identify a high risk population, however, as such mutations lead to downstream activation of signal transduction pathways, they may identify a target for directed therapy.