RAS mutations negate the favorable impact of NPM1 in older patients with newly diagnosed Acute Myeloid Leukemia treated with ven/HMA Free

Background: Activating RAS mutations play an important role in the pathogenesis of acute myeloid leukemia (AML) and occur in approximately 15-20% of patients with newly diagnosed (ND) AML. Although RAS mutations do not independently impact survival, the 4-gene molecular prognostic risk signature (mPRS) that is used to risk-stratify clinical outcomes in patients treated with venetoclax (VEN) and azacitidine and led to the updated ELN 2024, considered RAS mutations intermediate-risk. Furthermore, NPM1 was no longer favorable prognostic when co-occurring with signaling gene (SG) mutations (i.e., FLT3, NRAS, KRAS). However, given the small number of patients with a co-occurring NPM1and RAS mutation (n=8) or FLT3-ITD (n=16), it is unclear whether these patients truly have worse outcome compared with those without SG mutations. We aimed to investigate the prognostic impact of NPM1 alone and in the co-presence of RAS and/or FLT3-ITD mutations, among patients with favorable- or intermediate-risk AML per ELN 2024, in two independent cohorts of older patients (≥60 years) treated with VEN in combination with a hypomethylating agent (HMA).

Methods: This retrospective study included data from two different cohorts of patients aged ≥60 years with ND AML who were treated with first-line (1L) VEN+HMA and were enrolled in the Beat AML trial (NCT03013998) or who were included in the US-based Flatiron Health research database. Mutation analysis (next-generation sequencing or polymerase chain reaction) was performed at diagnosis. Overall survival (OS) was estimated using the Kaplan-Meier method. OS was defined as the date of consent (Beat AML) or from start of 1L (Flatiron) to the date of death, allogeneic cell transplantation, or last follow-up.

Results: A total of 187 patients enrolled on the Beat AML trial with favorable- (n=128) or intermediate-risk AML (n=59) per ELN 2024 were included. The median age was 74 years (range, 61-89). Forty (21%) patients were NPM1^mut, 41 (21%) were RAS^mut, and 21 (11%) patients were FLT3-ITD^pos. The median OS was 48.6 weeks with a 2-yr OS of 40%. NPM1^mutwas associated with a trend toward increased OS compared to NPM1^wt (2-yr OS 54% vs 37%, p=0.12). NeitherRAS^mut (p=0.2), nor FLT3-ITD (p=0.84) impacted outcome. NPM1^mutwithout RAS^mut (n=29) had a non-significant superior 2-yr OS of 58% compared to NPM1^mutwith RAS^mut(n=11), NPM1^wtwith RAS^mut(n=30) and NPM1^wtwithout RAS(n=117) patients(30%, 38%, 39%, respectively; p=0.22). NPM1^mutwithout SG (n=20) was also associated with non-significant improved 2-yr OS of 59% compared to NPM1^mutwith SG^mut(n=20), NPM1^wtSG^mut(n=39), NPM1^wtSG^wt(n=108) patients(45%, 29% and 40%, respectively; p=0.27).

Due to the small number of patients, a similar analysis was performed in a larger real-world cohort of ND AML patients (n=803). The median age was 76 years (range, 60-84), 645 (80%) patients were favorable- and 158 (20%) were intermediate-risk per ELN 2024. One hundred twenty-six patients had NPM1^mut, 155 had RAS^mut (113 NRAS^mut, 64 KRAS^mut) and 104 patients were FLT3-ITD^pos. The median OS was 40 weeks (range, 0-390). NPM1^mutwas associated with significantly longer OS compared to NPM1^wt (p=0.0078) and RAS^mut was associated with a shorter OS compared to RAS^wt (p=0.0001), irrespective of KRAS (p=0.055) or NRAS (p=0.0001). FLT3-ITD did not impact outcome (p=0.12). Stratified by the presence of RAS mutation, 2-yr OS of NPM1^mutRAS^mut was inferior compared to NPM1^mutRAS^wt(2-yr OS 27% vs 51%, p=0.001). NPM1^mut in the co-presence of SG^mutwas also associated with inferior 2-yr OS compared to NPM1^mut in the absence of SG^mut(2-yr OS 35% vs 49%, p=0.036), however, when we compared NPM1^mut with or without FLT3-ITD, outcome was not impacted (p=0.29).

Conclusions: SG mutations are common in older patients with AML, often with NPM1 co-mutations. We demonstrated that while NPM1^mutwere prognostically favorable, the presence of co-occurring SG mutations negates the favorable effect of NPM1^mut in ND older adults with AML, particularly RAS^mutrather than FLT3-ITD. These findings warrant investigation of new therapies, such as the addition of menin inhibitors to VEN/AZA, in the subset of NPM1^mutRAS^mutFLT3-ITD^neg patients. Phase 3 trials are currently underway, and benefit to this patient subset should be analyzed.