Treatment intensity may alter the prognostic impact of baseline co-mutations in newly diagnosed FLT3-ITD AML treated with FLT3 inhibitor-based therapy

Abstract Introduction: FLT3 inhibitors (FLT3i) have improved outcomes in newly diagnosed (ND) FLT3-ITD mutated (mut) AML. A recent analysis of the QuANTUM-1 trial (Levis et al., EHA 2025) identified a subgroup of ND FLT3-ITD mut AML patients (pts) with co-occurring NPM1 mut +/- DNMT3A mut +/- epigenetic regulator mut as being particularly sensitive to intensive chemotherapy (IC) + Quizartinib. We sought to broadly assess the prognostic impact of baseline co-mutations (co-mut) in our cohort of ND FLT3-ITD mut AML. Methods Retrospective study conducted at MDACC, Houston including adults with ND FLT3-ITDmut AML who received frontline FLT3i-based IC or low intensity therapy (LIT) regimen from 09/2013 - 07/2024. FLT3-TKD mut were excluded. Baseline genetic mut were grouped for analysis as follows: Epigenetic regulator (DNMT3A/TET2/IDH1/IDH2/WT1), myelodysplasia related (MR; ASXL1, RUNX1, EZH2), and RAS pathway (KRAS, NRAS, PTPN11). Outcomes were stratified by baseline co-mut across IC/LIT regimens. Results A total of 213 pts were included; 83 received IC + FLT3i [sorafenib in 56 (68%), midostaurin in 1 (1%), gilteritinib in 23 (28%), and quizartinib in 3 (4%)] while 130 received LIT + FLT3i [including sorafenib in 32 (25%), midostaurin in 1 (1%), gilteritinib in 48 (37%), and quizartinib in 49 (38%)]. LIT regimens were venetoclax (VEN) based in 69 (53%) and non-VEN based in 61 (47%). In the IC+FLT3i group, median (med) age was 52 years (range 22 - 72). NPM1 co-mut was noted in 46 pts (55%). Other co-mut included epigenetic regulators (56, 68%), RAS pathway (17, 21%), and MR genes (12/82, 15%). NPM1 co-occurred with epigenetic regulator mut in 38 pts (46%). Composite complete remission (CRc = CR + CRi) was attained in 67 pts (81%) for the full IC + FLT3i cohort: CRc rates were higher among NPM1 co-mut (98% vs 60%, p<0.01), and lower among RAS pathway co-mut (59% vs 86%, p=0.02) or MR gene co-mut pts (42% vs 89%, p<0.01). CRc rates were also higher in the NPM1 + epigenetic regulator co-mut group when compared to those without mut in either NPM1 or epigenetic regulators (97% vs 74%, p= 0.01). A total of 47 pts (57%) underwent allogeneic stem cell transplant (SCT) in CR1. After a med follow up of 71.6 months (mos), med overall survival (OS) for the full IC cohort was not reached (NR); 3-year OS 58%. Pts with NPM1 co-mut had superior OS (med, NR vs 24 mos, p=0.05) while those with co-mut in MR genes (med, 13.3 mos vs NR, p<0.01) and RAS pathway (med, 10.2 mos vs NR, p=0.06) had lower OS. Pts with NPM1 + epigenetic regulator co-mut had a numerically higher med OS compared to those without mut in either (NR vs 66.3 mos, p=0.2). OS was similar for pts with DNMT3A vs no DNMT3A co-mut (med, 45.2 vs NR, p=0.7) and NPM1 with DNMT3A vs NPM1 without DNMT3A (med, NR vs NR, p=0.5). In the LIT+FLT3i group (n=130), med age was 72 years (range 23 - 91). NPM1 co-mut was noted in 54 pts (42%). Other co-mut included epigenetic regulators (84/128, 66%), MR genes (36/127, 28%), and RAS pathway (22, 17%). VEN was used (as HMA + VEN + FLT3i “triplet”) in 69 pts (53%). CRc was attained in 92 pts (71%). CRc rates were lower in RAS pathway co-mut (50% vs 95%, p=0.04), while there was no difference in CRc rates when stratified by NPM1 (72% vs 70%, p=0.85) or MR gene co-mut (69% vs 73%, p=0.83). CRc rates were similar in the NPM1 + epigenetic regulator co-mut group compared to those without either (77% vs 64%, p=0.31). CRc rates were higher with triplet compared to LIT + FLT3i (no VEN) (87% vs 53%, p<0.01). 29 pts (22%) underwent SCT in CR1. With a med follow-up of 35.8 mos, med OS for the full LIT + FLT3i cohort was 14.1 mos. There was a trend to lower OS in RAS pathway mut pts (med, 13.6 vs 14.1 mos, p=0.08), but no difference when stratified by NPM1 (med, 11.6 vs 15.6 mos, p=0.7) or MR gene co-mut (med, 15.8 vs 12 mos, p=0.5). Pts with NPM1 + epigenetic regulator co-mut had similar OS compared to those without either (14 vs 14 mos, p=0.7). Conclusions The impact of baseline co-mut on frontline FLT3i-based therapy in ND FLT3-ITD mut AML may vary by the nature of the induction regimen. The presence of NPM1 mut +/- epigenetic regulator mut was associated with improved outcomes with IC + FLT3i, consistent with findings from QuANTUM-1, but importantly, we noted that co-mut in RAS pathway and MR genes portended a poorer prognosis. In contrast, co-mut in NPM1 and MR genes did not influence outcomes with LIT + FLT3i regimens (with or without VEN).