Axl inhibition: a potential road to a novel acute myeloid leukemia therapy?

Abstract

Novel treatment options in acute myeloid leukemia (AML) are urgently needed; treatment has not changed significantly over the past decades and survival is still dismal, especially in elderly patients. Axl, a member of the Tyro3, Axl, Mer (TAM) receptor family, mediates proliferation and survival of AML cells and is upregulated upon cytostatic treatment. In addition, AML cells induce expression of the Axl ligand growth arrest-specific gene 6 (Gas6) in bone marrow stroma cells, which further amplifies their growth and therapy resistance. Interruption of Axl signaling by pharmacological approaches, including the small molecule Axl inhibitor BGB324, decreased disease burden and prolonged survival of AML mice. The Gas6-Axl pathway has translational relevance because Axl is expressed by approximately 50% of AML patients and Axl-targeting approaches can block growth of primary human AML cells. Thus, Axl represents a potential novel target in AML and BGB324 is now in clinical development.

Keywords:

• acute myeloid leukemia
• Axl
• BGB342
• Gas6
• stroma

Acute myeloid leukemia (AML) is a genetically heterogeneous malignancy characterized by inhibition of myeloid differentiation with subsequent accumulation of immature blasts leading to reduced production of healthy hematopoietic cells. AML is the second most common form of leukemia and its incidence increases with age, with the majority of cases being diagnosed at the age >65 years [1,2].

In the past decade, major advancements have been made in identifying different molecular and genetic subgroups and elucidating the molecular pathways of AML pathogenesis [2,3]. Partly, this information is now used to guide treatment decisions [2], but mostly has not (yet) resulted in approval of novel therapies. Consequently, AML treatment has not changed significantly over the past 40 years. Outcome has improved in younger patients mainly due to better supportive care (i.e., use of broad spectrum antibiotics, antivirals, new antifungals and antiemetics) and increased use of allogeneic stem cell transplantation. However, 5-year survival is approximately 50% for patients <60 years and 10–20% for older patients [1,2].

The standard cytarabine and daunorubicin induction regimen (‘7+3’) is used to induce remission, followed by a consolidation therapy comprising bone marrow stem cell transplantation or intensive, cytarabine-based chemotherapy according to individual prognostic risk factors. For many elderly patients, intensive chemotherapy is not feasible due to reduced performance status and comorbidities. Furthermore, many elderly patients present with poor prognostic factors, which limit the success of chemotherapy [1,2].

It is evident therefore that new therapy regimes are urgently needed and emerging data suggest that inhibition of the receptor tyrosine kinase Axl represents a promising novel therapeutic strategy [4–6].

Axl was initially identified as a transforming gene in chronic myeloid leukemia. Axl belongs to the Tyro3, Axl, Mer (TAM) subfamily of receptor tyrosine kinases, comprising TAM [7,8]. There are four putative TAMR ligands, growth arrest-specific gene 6 (Gas6), Protein S, tubby and tubby like protein 1 (tulp1) [9]. Gas6 has subnanomolar affinity for Axl and is the only activating ligand for Axl. Mer has reduced binding affinity for Gas6 relative to Axl and Tyro, while Protein S binds preferentially to Tyro and Mer. Tubby binds exclusively to Mer and tulp1 interacts with all TAMs. Very little is known about the biological function of tubby and tulp1 [9,10].

Since their discovery more than 20 years ago, TAMR have been shown to be overexpressed in many solid tumors including breast, lung and brain, and to induce proliferation, survival and resistance to apoptosis. Axl is also known to play a role in mediating migration and invasiveness of cancer cells [9]. In the 1990s, Axl was first linked to AML, when it was shown that Axl could be detected primarily in cells derived from myeloid versus lymphoid malignancies [11,12].

Two recent publications have independently identified Axl as a potential new therapeutic target in AML by demonstrating that Axl inhibition, achieved through shRNA or the small molecule Axl inhibitor BGB324 (formerly R428), increased apoptosis and inhibited proliferation of FLT3-ITD and FLT3 wild-type AML cell lines and primary AML cells in vitro [4,5]. Furthermore, Axl inhibition reduced tumor burden and prolonged survival in different mouse models, including the systemic HoxA9/Meis1 model [4,5]. BGB324 is an orally available highly specific Axl inhibitor, which inhibits Axl with a 10- to 100-fold higher affinity than other kinases, with the exception of Tie-2 (threefold), FLT4 (5.5-fold), and FLT1, Ret and Abl (eight- to ninefold) [4,13]. Our data indicate that Axl inhibition decreased pAkt, pErk and Bcl-2 levels in vitro and in vivo, which could explain the observed reduced cell proliferation and increased apoptosis [4]. However, it is feasible that Axl signaling exerts its pro-AML effects via additional signaling pathways, as seen in B-cell chronic lymphocytic leukemia cells where Axl is constitutively activated and acts as a docking site for intracellular kinases, including Syk [14].

Furthermore, we could demonstrate that Axl is involved in chemoresistance as BGB324 treatment renders AML cell lines more sensitive to chemotherapy [4]. AML cells also stimulate bone marrow-derived stroma cells to upregulate Gas6, in a paracrine manner, which further increases the chemoresistance of AML cells. As anticipated, an additive effect was observed when BGB324 was used in combination with cytarabine in primary AML cells. In vitro experiments using Gas6-deficient HL60 cells showed that these cells were resistant against BGB324 while they were sensitive in vivo when Gas6 could be delivered by the mouse stroma. These findings indicate that presence of Gas6 might be required for therapeutic efficacy of the drug [4].

Immunohistochemical analysis of bone marrow biopsies from AML patients revealed that Gas6 was minimally expressed by AML cells compared to stroma cells. In cell types with fibroblastic/mesenchymal morphology (referred to as bone marrow-derived stroma cells), Gas6 expression was higher in samples from AML patients compared to healthy controls, while osteoclast expression was lower in AML samples with osteoblasts and endothelial cells exhibiting similar Gas6 expression levels. Overall, Gas6 expression level was increased in AML bone marrow plasma compared to healthy controls [4].

These observations support the importance of Gas6-mediated signaling in human disease.

Our data indicate that Axl mRNA is expressed by 57% of AML patients [4]. Furthermore, they corroborate the hypothesis that Axl mRNA expression is important in the response to chemotherapy, as expression above the mean represents an independent adverse prognostic factor in n = 64 cytogenetically normal AML patients treated with chemotherapy [4]. Similar data obtained in another study show that Axl mRNA expression was associated with worse prognosis in multivariate Cox regression models adjusted for age, Auer rods and leukocyte counts (progression-free survival, p = 0.015; overall survival, p = 0.05). Axl expression was also associated with bcl-2 mRNA expression (n = 54, r = 0.32; p = 0.02) and CD34 expression (n = 38, r = 0.42; p = 0.008) [11].

In our cohort of AML patient samples Gas6, Mer and Tyro3 mRNA expression lacked prognostic impact, while Whitmann et al. have recently shown that Gas6 mRNA expression was associated with shorter overall survival and disease-free survival in their set of 270 patients with cytogenetically normal AML. Furthermore, their data failed to reveal a prognostic impact for Axl or any other of the TAMR [6]. These conflicting data might be due to differences in the patient cohort and potentially to methodological differences as gene expression array data was used in their investigations while we used PCR-based methods. Collectively, these data do corroborate the involvement of Gas6-induced signaling in AML biology.

Using a variety of approaches, including a soluble Axl–Fc fusion protein, which binds Gas6 and Protein S, thereby preventing interaction with TAMR, an Axl-targeting siRNA and foretinib, a non-specific tyrosine kinase inhibitor, Park et al. demonstrated that Axl was phosphorylated in FLT3-ITD and FLT3 wild-type AML cells [5]. They further demonstrated that inhibition of Axl in primary FLT3-ITD AML cells inhibited activation of Erk, STAT5, Akt and FLT3 and induced the differentiation of AML blasts [5]. The authors did not observe any additive therapeutic effect with FLT3 inhibition using PKC412, when combined with Gas6 blockade in vivo, suggesting that inhibiting Axl signaling attenuates FLT3 downstream signaling pathways [5]. Gas6 blockade by Axl–Fc reduced tumor growth in the subcutaneous MV4;11 (FLT3-ITD) xenograft mouse model and leukemic burden in a severe combined immunodeficiency mouse model using engrafted FLT3-ITD AML blasts from patients [5]. Altogether, data from different groups indicate importance of Gas6-Axl signaling in AML pathobiology [4–6,11].

In addition to Axl, Mer has been identified as a potential therapeutic target in AML [15]. Of note, Axl is absent or only minimally expressed by malignant plasma cells in multiple myeloma (MM), while Mer is the TAMR through which Gas6 exerts its pro-proliferative and pro-survival effects [16]. Gas6 is secreted via an autocrine mechanism by malignant plasma cells in MM in contrast to the paracrine mechanism observed in AML. Axl has recently been identified as a possible new therapeutic target in chronic lymphocytic leukemia [17]. Thus, expression and biological relevance of different TAMRs are non-redundant and vary in a cell-type and context-specific way in hematologic cancer.

Axl expression is also involved in the development of resistance against chemotherapy [18] and EGF(R)-targeting agents in lung cancer [19] and colon cancer [20].

TAMR are expressed by macrophages and dendritic cells and concerns about the validity of selecting TAMR as novel drug targets were raised when it was discovered that knockout mice lacking both Axl and Mer were susceptible to developing colon cancer [21], while Mer-deficient mice were blind [22]. BGB324 binds preferentially to Axl, while the affinity for Mer is 16- and 50-fold lower in biochemical and cell-based assays, respectively [13] [Lorens J, Unpublished Data]. Furthermore, as the Axl/Gas6 axis seems to be activated in AML patients compared to healthy subjects, the existence of a workable therapeutic window is very likely. In a recently conducted Phase I trial, BGB324 was administered to healthy volunteers with minimal toxicity [23]. Based on emerging data on the role of the Gas6-Axl axis in AML and solid tumors further clinical trials are warranted to define the maximum tolerated dose and establish the side-effect profile of BGB324.

Furthermore, anti-Axl antibodies could be useful for therapeutic targeting of Axl [24,25]. An interesting alternative approach for specific targeting of Gas6 could be novel high-affinity engineered Axl ‘decoy receptor’ capable to neutralize Gas6 in the femtomolar range [26].

Financial & competing interests disclosure

S Loges is supported by the Max-Eder group leader program from Deutsche Krebshilfe, the Deutsche Forschungsgemeinschaft (Grant #LO1863/3-1), the Jose Carreras Stiftung (DJCLS R14/06), the Roggenbuck Stiftung, the Hamburger Krebsgesellschaft, the Medical Faculty of the University of Hamburg (FFM program) and the Hamburger Exzellenzinitiative (LEXI program). M Janning is supported by a Hubertus-Wald fellowship. M Janning received Advisory Board Honoraria from Genzyme. S Loges received Advisory Board Honoraria, Travel Support and Research Support from BerGenBio. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.