Acute Myeloid Leukemia (AML) is a genetically heterogeneous disease characterized by clonal expansion of immature myeloid progenitor cells in the bone marrow (BM). Mutations of the FMS-like tyrosine kinase 3 (FLT3) gene occur in approximately 30% of AML cases, with Internal Tandem Duplications (ITD) being the most common type of mutation. Several FLT3 specific inhibitors (TKI) have been developed such as quizartinib, crenolanib and midostaurin (recently approved for clinical use). Nevertheless FLT3-ITD is associated with unfavorable prognosis and patients develop drug resistance with the underlying mechanisms remaining largely unexplained. Recently, changes within the actin cytoskeleton were associated with drug resistance development in various cancers. FLT3-ITD mutations are associated with RAC1 activation. RAC1 belongs to the family of RHO GTPases and enhances the actin polymerization by inducing the expression of N-WASP or WAVE2 and ARP2/3 complex. Therefore, we investigated actin cytoskeleton rearrangements through RAC1 activation as a potential mechanism contributing to Midostaurin resistance in AML.

Material and methods

First, we developed two Midostaurin resistant AML cell lines (MID-RES, MV4-11 and MOLM-13). Single cell measurements of Cell Stiffnes, cell adhesion forces between tumor and HS5 stroma cells and Actin filaments were performed by Atomic Force Microscopy (FluidFM®) and SIM microscopy, respectively. RAC1 activation was measured by RAC1 activation kit provided by Cytoskeleton. FLT3 surface and intracellular expression was measured by Flow cytometry and western blot, respectively. Cell death was analyzed by Annexin/PI staining in flow cytometry.


The MID-RES cell lines MV4-11/MOLM-13 showed higher FLT3 surface and intracellular expression compared to their MID sensitive parental cells. In line with our expectations, we observed RAC1 activation, as well as an up-regulation of actin polymerization positive regulators such as N-WASP, WAVE2, PFN1 and ARP2/3 complex and the inhibition of actin polymerization negative regulator P-ser3 CFL1 in MID-RES cells. FLT3 receptor knock down by siRNAs reversed the MID resistance and reduced RAC1 activation and actin polymerization inducers expression. Likewise, bioinformatic analysis from publicly available microarray expression data (E-MTAB-3444), confirmed positive correlation between actin polymerization inducers and FLT3 signaling expression in 178 FLT3-ITD (r=0,67) and 461 FLT3 WT(r= 0,57) de novoAML patients.

RAC1 induced Actin polymerization positively correlates with actin filaments growth and cell stiffness, which was observed in our MID-RES cells, higher load of actin filaments and increased cell stiffness.

The combination between RAC1 specific inhibitor, EHT1864 and Midostaurin synergistically induces cell death in MID-RES cells by arresting cell cycle in G0/G1 phase and activating apoptosis. Beside, this combination reduced the adhesion forces to stroma cells, decreased the expression of PFN1/N-WASP/ARP2 and consequently reduced drastically the number of actin filaments and cell stiffness in MID-RES cells. EHT1864 and Midostaurin (alone and in combination) were not toxic in PBMCs obtained from healthy donors. Interestingly, this combination increase >45 % cell death in cells obtained from refractory FLT3-mutated AML patient (this patient was relapsed (≥ 50% residual blasts in the bone marrow)under Chemotherapy+Midostaurin combination).The specific knock down of PFN1/N-WASP/ARP2 with siRNAs equally reversed the resistance to Midostaurin.

Of note, RAC1 regulates the anti-apoptotic BCL2. Indeed, EHT1864 in combination with Midostaurin reduced anti-apoptotic family BCL2/MCL1 expression and increases the pro-apoptotic proteins BAX/PUMA. As expected, our MID-RES cells showed higher sensitivity to BCL2 inhibitor Venetoclax, than their parental cells. The combinations EHT1864+venetoclax, venetoclax+midostaurin and venetoclax+Midostaurin+EHT1864 synergistically induced cell death in MID-RES cells.


Actin polymerization inducers N-WASP, ARP2/3 complex and PFN1 may provide a valuable approach to overcome Midostaurin resistance in AML. Our data further suggest that the addition of BCL2 inhibition through EHT1864 and venetoclax could represent an interesting strategy to potentiate the activity of Midostaurin in FLT3 mutated AML.


Duell:Regeneron Pharmaceuticals, Inc.: Research Funding. Rosenwald:MorphoSys: Consultancy.

Author notes


Asterisk with author names denotes non-ASH members.