Page 130 - 2018_11-Haematologica-web

K. Seipel et al.
est survival rates. Survival rates are higher in NK-AML patients with high allelic ratio of FLT3-ITD and NPM1mut or low allelic ratio of FLT3-ITD and NPM1 wild type or FLT3wt and NPM1mut which have all been classified as intermediate risk.16 The leukemic cells of all these AML subsets have substantially elevated levels of cellular p53 antagonists and reduced p53 activity17 which identifies them as targets for treatments aiming to restore p53 func- tion including conventional chemotherapy with genotoxic compounds and non-genotoxic treatments with p53 reac- tivating compounds.
MDM2 is an established cellular p53 antagonist fre- quently overexpressed in AML cells. A variety of MDM2 inhibitors have been developed and tested in AML cell lines. Compounds currently in clinical trials for the treat- ment of AML include idasanutlin (RG7388), NVP- CGM097 and NVP-HDM201.12,18,19 We have studied the MDM2 inhibitor idasanutlin in combination with the MEK inhibitor cobimetinib in AML cell lines and patient samples and found this combination to be effective only in AML cells expressing high levels of FLT3 and MDM2 protein.20 The activity of NVP-CGM097 was investigated in AML cell lines and primary AML cells expressing wild type and mutant p53, alone and in combination with the FLT3 inhibitor PKC412 (midostaurin) or the MEK inhibitor AZD 6244.19 Synergy was observed when NVP- CGM097 was combined with FLT3 inhibition against oncogenic FLT3 expressing cells, as well as when com- bined with MEK inhibition in cells with activated MAPK signalling. In addition to reactivating p53 in AML cells by specific MDM2 inhibition the FLT3 receptor can be directly targeted by (more or less) specific tyrosine kinase inhibitors.5,21
In the present study, we tested a variety of FLT3 and MDM2 inhibitors. The effects on cell survival of FLT3- ITD AML cells were consistent for the FLT3 inhibitors midostaurin (PKC412), quizartinib (ACC220) and gilteri- tinib (ASP2215), but varied for the MDM2 inhibitors idasanutlin (RG7388), NVP-CGM097 and NVP-HDM201. The most potent MDM2 inhibitor NVP-HDM201 exhibit- ed superior combinatory effects on cell viability of FLT3- ITD AML cells together with midostaurin, and moderate combinatory effects together with quizartinib and gilteri- tinib. The different combinatory potentials may be related to the target specificity of the three FLT3 inhibitors. While quizartinib (ACC220) inhibits FLT3 and PDGFR kinases,22 gilteritinib (ASP2215) inhibits FLT3, LTK, ALK, and AXL kinases,8 and midostaurin (PKC412) inhibits FLT3, KIT, PKC, PPK, VEGFR-2, PDGFR, and SYK kinases.23 PDGFR and VEGFR-2 are expressed in the bone marrow of AML patients24,25 and, like FLT3, signal via PI3K/AKT and MDM2 to inhibit p53.26,27 The stem cell growth factor receptor KIT is expressed in the bone marrow and, like FLT3-ITD, signals via PI3K/AKT and MDM2 to inhibit p53, and via JAK2 and STAT5.28
MOLM-13 and MV4-11 AML cells were most suscepti- ble to the FLT3 inhibitor midostaurin. Both cell lines have a high allelic ratio of FLT3-ITD and harbor MLL rearrange- ments, created by t(9;11) and t(4;11), encoding MLL-AF9 and MLL-ENL, respectively. FLT3 and MLL cooperate in AML29 and leukemic cells with mutations in FLT3 and MLL are known to be susceptible to midostaurin.30,31 In the absence of MLL mutations, FLT3-ITD AML cells were less susceptible to midostaurin, but more susceptible to NVP- HDM201 indicating that the presence of MLL fusion pro-
teins may change the susceptibility of AML cells to FLT3 and MDM2 inhibitors.
NK-AML cells with FLT3-ITD/TP53wt/NPM1wt were particularly susceptible to the combination of midostaurin with conventional induction therapy or with NVP- HDM201. In contrast, NK-AML cells with mutated TP53 were rather resistant to midostaurin and NVP-HDM201, and they have previously been shown to be resistant to chemotherapy with genotoxic compounds32 and to other MDM2 inhibitors.33 As both the conventional induction therapy and MDM2 inhibition induce cell cycle exit and apoptosis via p53 activation, these treatments can only be effective in a TP53 wild type context.9 The presence of one mutated TP53 allele may be sufficient to suppress the function of the remaining p53 wild type protein. Hence, restoring wild-type p53 activity may be a promising option for treatment of AML with mutated TP53.34 HL-60 cells with deleted TP53 were resistant to midostaurin and HDM201, indicating that these two compounds specifi- cally target AML cells with functional p53 protein. This is in apparent contrast to the FLT3- inhibitor sorafenib and the MDM2 inhibitor nutlin-3 which promoted synergistic cytotoxicity irrespectively of FLT3 and p53 status via induction of the pro-apoptotic Bcl-2 family members Bax and Bak in p53 wild type and p53 deleted cells.35
It seems remarkable that FLT3wt/TP53wt/NPM1mut AML cells turned out to be as refractory to midostaurin and NVP-HDM201 as TP53mut cells. This suggests that the presence of mutated NPM1 protein is sufficient to pre- vent induction of p53 target genes by midostaurin or NPV- HDM201 in these cells. The mutated NPM1 protein (NPM1c) reduces the susceptibility of FLT3-ITD AML cells to both midostaurin and NVP-HDM201, indicating that the same molecular mechanisms may be involved in FLT3wt and FLT3-ITD cells. NPM1c is exported to the cytoplasm36 where it inhibits the tumor suppressor protein p53 by cytoplasmic retention. This leads to reduction of p53 protein in the nucleus and decreases its stability and transcriptional activation function.37 It appears that this loss of p53 activity by NPM1c cannot be compensated by inhibition of MDM2 or FLT3, at least not with NVP- HDM201 or midostaurin. In contrast, nutlin-3 can induce apoptosis and cell death in FLT3wt/TP53wt/NPM1mut cells,17 indicating a fundamental difference in the activities of the two MDM2 inhibitors.
The molecular mechanisms involved in p53 activation appear to be different in the conventional induction treat- ment. Here, FLT3wt/TP53wt/NPM1mut AML cells were more susceptible to conventional induction treatment than TP53mut cells, indicating that the presence of NPM1c is not sufficient to prevent p53 activation by geno- toxic substances. This may be reflected by the usually favorable outcome of FLT3wt/TP53wt/NPM1mut AML patients after standard induction treatment.38,39 Midostaurin can enhance effects of conventional induc- tion therapy in FLT3-ITD cells in vitro and in vivo. As eval- uated in a randomized, double-blind, phase III interna- tional study (clinicaltrials.gov identifier 00651261) in high- risk patients with newly diagnosed, FLT3-mutant AML (CALGB10603/RATIFY trial), the multikinase inhibitor midostaurin plus standard chemotherapy improved sur- vival compared with placebo plus chemotherapy.40,41 On April 28, 2017, the U.S. Food and Drug Administration approved midostaurin (RYDAPT, Novartis Pharmaceuticals Corp.) for the treatment of adult patients
1870
haematologica | 2018; 103(11)