Gain-of-function mutations in FMS-like tyrosine kinase 3 (FLT3) are detected in approximately 30 % of newly diagnosed acute myeloid leukemia (AML) cases [1,2]. Because of its high frequency and association with poor prognosis, mutant FLT3 is an attractive target for AML therapy [3].

FLT3 is composed of four domains: the extracellular portion that binds FLT3 ligand, the transmembrane (TM) domain, the juxtamembrane (JM) domain, and a highly conserved intracellular kinase domain (Fig. 1, Panel A). The 3D structure of the FLT3 kinase domain includes a smaller N-terminal lobe, a larger C-terminal lobe and a hinge region that connects the two lobes. The key structures related to kinase activity include the JM domain, which serves an autoinhibitory function, the activation loop (A-loop) and the catalytic loop (Fig. 1, Panel B). The JM domain contains several key tyrosine residues as phosphorylation sites. The A-loop also has 1–3 tyrosine residues. The A-loop is structurally flexible and has two conformations: DFG-in or DFG-out. The DFG motif refers to three highly conserved amino acid residues at the base of the A-loop (Asp829-Phe830-Gly831 in FLT3). In the DFG-in (active) conformation, the A-loop is in an open and extended state, characterized by the Asp829 residue pointing at the ATP-binding pocket and the Phe830 residue pointing to the back cleft. In this conformation, the Asp829 residue interacts with ATP and promotes kinase activation. In the DFG-out (inactive) conformation, the A-loop is in a closed state, precluding binding of ATP and substrates. Thus, the kinase domain adopts an autoinhibited conformation, characterized by the Asp829 residue pointing to the back cleft and the Phe830 residue pointing to the ATP-binding pocket [4,5]. Notably, the DFG-out conformation generates a hydrophobic pocket adjacent to the ATP-binding pocket, behind the “gatekeeper” residue F691 [6].

The native or “wild type” (WT) FLT3 is kept inactive monometrically. The JM domain binds in a strategic area of FLT3 and interacts with multiple key structural components, thus working as a wedge that stabilizes the inactive state of FLT3 [7]. The A-loop, blocked by the JM domain, and stabilized by multiple residues’ interactions, is kept in the autoinhibited DFG-out conformation (Fig. 1, Panel B). Upon binding to FLT3 ligand (FL), FLT3 dimerizes and brings the two intracellular domains into close proximity. The subsequent transphosphorylation of tyrosine residues at the JM domains forces them to wind back, exposing the A-loops; while the following phosphorylation of A-loops forces them to adopt the DFG-in conformation, promoting the further activation of FLT3 [8]. FLT3 activating mutations have two subtypes: internal tandem duplication at the juxtamembrane domain (“FLT3-ITD”), and point mutations in the activation loop of the tyrosine kinase domain (“FLT3-TKD”) (Fig. 1, Panel A). FLT3-ITD disrupts the inhibitory function of the JM domain, thus promoting FLT3 activation in the absence of FLT3 ligand; FLT3-TKD mutations also enhance unregulated FLT3 activation through stabilization of the active, DFG-in conformation.

Based on their binding mode with the target kinase, inhibitors are categorized as type I or type II. Type I inhibitors bind to the ATP-binding pocket while type II inhibitors bind to both the ATP-binding pocket and the hydrophobic pocket adjacent to the ATP-binding pocket, which is only accessible when the A-loop is in the inactive DFG-out conformation. Two generations of FLT3 inhibitors have been discovered and studied in clinical trials, with three FLT3 inhibitors currently approved for clinical use (Fig. 2, Panel A). Midostaurin is a type I, first-generation FLT3 inhibitor with minimal activity as monotherapy. Midostaurin lacks selectivity and is only approved by the USFDA for use in combination with cytotoxic chemotherapy agents [[9], [10], [11]]. As a type II [12], second-generation FLT3 inhibitor, quizartinib [[13], [14], [15], [16], [17], [18], [19]] has substantially improved specificity and selectivity, and was approved in Japan as monotherapy for AML cases with FLT3-ITD mutation. Quizartinib displays substantial initial clinical efficacy, but resistant mutations rapidly emerge at residues D835 and F691 [[20], [21], [22]]. The D835Y mutation in the activation loop breaks the hydrogen bond between Asp835 and Ser838 which helps stabilize the DFG-out conformation [23]. Because the type II inhibitor quizartinib only binds to a DFG-out conformation, FLT3-D835Y is highly quizartinib-resistant. The F691L mutation at the “gatekeeper” residue is bulky and leads to steric clash with quizartinib, preventing its binding to the hydrophobic pocket behind this position [[24], [25], [26]].

Gilteritinib is a second-generation FLT3 inhibitor that was recently approved by USFDA as monotherapy for the treatment of relapsed or refractory AML with FLT3 mutations [[27], [28], [29]]. As a type I inhibitor, gilteritinib is not impacted by the D835Y mutation. However, its inhibitory effects are diminished by steric clash caused by the F691L mutation [30]. In order to overcome drug resistance, new compounds with balanced inhibition against FLT3-ITD and FLT3-ITD with secondary mutations are needed, especially FLT3-ITD/F691L which is gilteritinib resistant.

Imidazo [1,2-a]pyridine derivatives were previously identified to potentially have promising FLT3 inhibitory effects [31]. Our studies identified 7-(4-(methylsulfonyl)phenyl)-3-(5-(pyridin-3-yl)thiophen-2-yl)imidazo [1,2-a]pyridine (Ling-5o) as a balanced FLT3-ITD and FLT3-ITD secondary mutants inhibitor (Fig. 2, Panel B) [32]. However, compared to quizartinib and gilteritinib, the inhibitory potency of Ling-5o is low, with sub-micromolar IC50s on MOLM14 cells and MOLM14 cells with drug resistant secondary mutations. Therefore, new FLT3 inhibitors with balanced inhibitory effects on FLT3-ITD and FLT3-ITD secondary mutations, that are more potent than Ling-5o, are needed. Here, we report that through structural modifications, 6-(7-(1-methyl-1H-pyrazol-4-yl)imidazo [1,2-a]pyridin-3-yl)-N-(4-(methylsulfonyl)phenyl)pyridin-2-amine (24) and its analogues were identified as balanced FLT3-ITD and FLT3-ITD secondary mutants inhibitors with improved potency.