FLT3 in AML: the inhibitor landscape

FLT3 is the most frequently mutated gene in acute myeloid leukemia, altered in roughly a third of patients, and for two decades it was a target everyone agreed mattered and nobody could drug well. Three inhibitors are now FDA approved, but treating them as interchangeable is a mistake. They bind different conformations of the kinase, hit different resistance mutations, and slot into different points in the treatment course. Here is the landscape as it actually stands.

Two kinds of FLT3 mutation

Before the drugs, the mutations. FLT3 lesions in AML come in two flavors, and the distinction drives everything downstream.

• Internal tandem duplications (FLT3-ITD) - in-frame duplications in the juxtamembrane domain, found in about 25% of AML. They relieve the autoinhibitory clamp and drive constitutive, ligand-independent signaling. ITD carries a high relapse rate and a high allelic ratio predicts worse outcomes.
• Tyrosine kinase domain point mutations (FLT3-TKD) - most commonly D835 substitutions in the activation loop, found in about 7-10%. These also activate the kinase, but they matter clinically because they confer resistance to a whole class of FLT3 inhibitors, as we will see.

Type I versus type II: the conformation that decides everything

FLT3 inhibitors split into two structural classes by which kinase conformation they bind. This is the single most useful concept for making sense of the clinical data.

Type II inhibitors bind the inactive, DFG-out conformation, reaching into the back hydrophobic pocket that opens when the activation loop swings away. The catch: the D835 activation- loop mutations stabilize the active, DFG-in state, so the DFG-out pocket the type II drug needs never forms. Type II inhibitors hit ITD beautifully and fail against D835.

Type I inhibitors bind the active, DFG-in conformation in the ATP pocket itself. Because they do not depend on the activation loop being displaced, they retain activity against both ITD and the D835 TKD mutants. That breadth is why the type I agents became the relapsed/refractory workhorses.

The three approved drugs

• Midostaurin (PKC412, Rydapt) - type I, but a broad multikinase inhibitor rather than a clean FLT3 agent. FDA approved April 2017 in combination with intensive induction and consolidation chemotherapy for newly diagnosed FLT3-mutated AML. The pivotal RATIFY trial (Stone et al., NEJM 2017) showed a significant overall and event-free survival benefit from adding midostaurin to 7+3 chemotherapy. It is a frontline add-on, not a single agent.
• Gilteritinib (ASP2215, Xospata) - type I, potent and selective for FLT3, active against both ITD and D835 TKD mutants. FDA approved November 2018 for relapsed or refractory FLT3-mutated AML on the strength of the ADMIRAL trial (Perl et al., NEJM 2019), which beat salvage chemotherapy on overall survival (median 9.3 vs 5.6 months) as a single oral agent. This is the relapsed/refractory standard of care.
• Quizartinib (AC220, Vanflyta) - type II, highly ITD-selective. FDA approved July 2023 in combination with induction and consolidation chemotherapy, and as continuation monotherapy, for newly diagnosed FLT3-ITD-positive AML. The QuANTUM-First trial (Erba et al., Lancet 2023) reported a near-doubling of median overall survival (31.9 vs 15.1 months) versus chemotherapy alone. Note the label: ITD-positive only. Because it is type II, it does not reliably cover D835 TKD disease.

How resistance emerges

The resistance patterns track the conformation logic. Under type II pressure (quizartinib, or off-label sorafenib), tumors acquire activation-loop D835 mutations and the F691L gatekeeper mutation, both of which block the DFG-out pocket. Switching to a type I agent like gilteritinib can recover activity against D835, though F691L is harder for everyone because it sits at the gatekeeper position that controls access to the ATP pocket itself. Beyond on-target mutations, off-target escape through RAS/MAPK reactivation and FLT3-independent clones is common, which is why durable single-agent control is rare and combination strategies dominate the trial pipeline.

Where the field is going

Three threads are worth watching. Combinations with venetoclax and hypomethylating agents are extending FLT3 inhibition to older, unfit patients who cannot take intensive chemotherapy. Maintenance after allogeneic transplant, where FLT3 inhibitors suppress residual ITD clones, is increasingly standard. And next-generation agents and rational sequencing aim at the F691L gatekeeper and at the polyclonal resistance that limits every current drug.

Try the docking yourself

The type I / type II distinction is exactly the kind of thing molecular docking makes tangible. Open Studio and pick FLT3 from the target catalog with the D835Y mutation chip, then dock a type II scaffold (quizartinib-like) and a type I scaffold (gilteritinib-like) against the same receptor. The type II compound should lose binding affinity against the activation-loop mutant while the type I compound holds, reproducing the clinical resistance story in silico. Liganx is molecular docking online: free, browser-based, and built for precisely this kind of mutation-aware comparison. If you want to try molecular docking on a FLT3 resistance mutation without a local install, that is the fastest path.

Primary sources

• Stone RM, et al. Midostaurin plus Chemotherapy for Acute Myeloid Leukemia with a FLT3 Mutation. NEJM 377, 454-464 (2017). doi:10.1056/NEJMoa1614359
• Perl AE, et al. Gilteritinib or Chemotherapy for Relapsed or Refractory FLT3-Mutated AML. NEJM 381, 1728-1740 (2019). doi:10.1056/NEJMoa1902688
• Erba HP, et al. Quizartinib plus chemotherapy in newly diagnosed patients with FLT3-internal-tandem-duplication-positive acute myeloid leukaemia (QuANTUM-First): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet 401, 1571-1583 (2023). doi:10.1016/S0140-6736(23)00464-6