Characteristics and Outcomes of Patients (pts) with Acute Myeloid Leukemia (AML) and FLT3-Tyrosine Kinase Domain (TKD) Mutations

Background:

FLT3-TKD mutations are present in 7-10% of newly diagnosed AML. However, its prognostic impact remains poorly defined, especially in the current treatment landscape, wherein FLT3-inhibitors (FLT3i) and venetoclax (VEN) are often used in frontline regimens. In one of the largest contemporary series, we describe the characteristics and outcomes of frontline FLT3-TKD AML, including impact of frontline treatment and allogeneic stem cell transplant (allo SCT).

Methods:

Retrospective analysis including all pts with AML who received frontline AML therapy at MDACC, Houston from 01/2012 – 12/2021. Pts were included in 2 cohorts per baseline mutational status: FLT3-TKD mutated (m), and NPM1m TKD wild type (WT). FLT3-ITDm AML was excluded from both cohorts. Survival outcomes were estimated using the Kaplan Meir method.

Results:

In the FLT3-TKDm cohort (n=98), median (med) age was 64 yrs (25-89). Induction therapy was intensive chemotherapy (IC) in 46 pts and low intensity (LIT) in 52 pts.

In the IC group (n=46), med age was 55 yrs (25-80), 21 (46%) female. AML was secondary/therapy-related (S/T) in 4 pts (9%), adverse risk ELN 2022 in 13 pts (28%), and NPM1m in 22 pts (48%). VEN was used in 11 (24%), type 1 FLT3i in 8 (17%), both VEN and FLT3i in 2 (4%) and neither in 25 (54%). Of the 44 evaluable pts, composite CR (CRc = CR + CRi) was attained in 38 pts (83%), including 10/11 (91%) receiving VEN, 5/8 (63%) receiving FLT3i, 2/2 receiving both, and 21/23 (91%) receiving neither, and med OS in these groups was NR (3yr OS 70%), NR (3 yr OS 63%), NR (3 yr OS 100%), and 17.5mo respectively.

With med follow up (f/u) of 31.5 mo, med OS was 41.7 mo (95% CI: 12 mo - NR). 19 pts (41%) had allo SCT. On landmark analysis at 4.1 mo (med time to SCT), allo SCT improved med OS compared with non SCT pts (NR vs 16.1 mo, HR 0.3, 95% CI: 0.1 - 0.9). On stratifying by NPM1 status, CRc was attained in 22/22 (100%) with NPM1m, and 16/22 (73%) with NPM1 WT. Pts with NPM1 mut had a longer med OS than NPM1 WT (NR, 3-yr OS: 71% vs 13.8 mo, HR 0.4, 95% CI 0.2 - 0.9, p=0.03) Allo SCT was done in 12/22 (55%) with NPM1m and 7/24 (29%) with NPM1 WT. On landmark analysis, Allo SCT did not improve med OS in NPM1m (NR, 3-yr OS 75% vs 37.5 mo, HR 0.7, 95% CI 0.1 – 3, p= 0.6), but improved it in NPM1 WT pts (NR, 3-yr OS 80% vs 9.5 mo, HR 0.2, 95% CI 0.04-0.7, p= 0.05).

In the LIT group (n=52), med age was 70 yrs (34-89), 25 (48%) female. AML was S/T in 21 pts (40%), adverse risk in 29 pts (56%), and NPM1m in 23 pts (44%). VEN was used in 17 (33%), both VEN and type 1 FLT3 inhibitor (FLT3i) in 9 (17%), and neither in 26 (50%). CRc was attained in 33 (64%) overall, including 14 (82%) receiving VEN, 8 (89%) receiving both VEN and FLT3i, and 11 (42%) receiving neither. With med f/u of 35.4 mo, med OS for the LIT group was 6.3 mo (95% CI: 2.8 mo - 16 mo). 5 pts (10%) got allo SCT. On landmark analysis at 3.2 mo (med time to SCT), alloSCT pts had longer med OS than non SCT pts (NR vs 13 mo, HR 0.2, 95% CI: 0.1 - 0.7). On stratifying by NPM1 status, CRc was attained in 19 (83%) with NPM1m, and 14 (48%) with NPM1 WT. Pts with NPM1m had longer med OS compared to NPM1 WT (22.9 mo vs 4.4 mo, HR 0.5, 95% CI: 0.2 - 0.9, p=0.01).

In the NPM1m FLT3-TKD WT cohort (n=143), med age was 67 yrs (17-84). Induction therapy was IC in 49 (34%) and LIT in 94 (66%). For the IC group, med age was 53 yrs (17-69), S/T AML in 3 (6%), and adverse risk in 8 (16%). CRc was attained in 47/48 (98%) evaluable pts. With a med f/u of 41.2 mos, med OS was NR (95% CI: 21.9 mos - NR); 3-yr OS 63%. 20 (41%) underwent allo SCT. On landmark analysis at 5.4 mo (med time to SCT), allo SCT did not have longer med OS than non SCT pts (NR vs NR, HR 0.9, 95% CI: 0.3-2, p= 0.8). For the LIT group, med age was 72 yrs (60-84), S/T AML in 19 (20%), and adverse ELN 2022 in 29 pts (31%). CRc was attained in 78/93 pts (84%). With a med f/u of 35 mo, med OS was 20.6 mo (95% CI: 4 mo – 38 mo). 15 (16%) got allo SCT. On landmark analysis at 4.1 mo, allo SCT pts did not have longer med OS than non-SCT pts (NR vs 32.8 mo, HR 0.7, 95% CI 0.3 - 1.5, p=0.4).

Conclusion:

In our cohort of frontline AML pts, FLT3-TKDm without NPM1 co-mutation was associated with lower response rates to both IC and LIT and med OS of 13.8mo and 4.4 mo respectively: alloSCT was associated with improved OS and should be considered in first remission. The presence of an NPM1 co-mutation in the FLT3-TKDm cohort improved response and OS to both IC and LIT, with outcomes similar to NPM1m without FLT3-TKD: OS was not clearly improved with alloSCT and the need for alloSCT in first remission warrants further prospective evaluation.