Preclinical Testing Of a Novel Axl-Kinase Inhibitor In Chronic Lymphocytic Leukemia

The receptor tyrosine kinase Axl belongs to the TAM (Tyro-3, Axl and Mer) family and is involved in the progression of several human malignancies including chronic lymphocytic leukemia (CLL), where it is has been found to be overexpressed in comparison to normal B-cells. An increasing body of evidence suggests that Axl acts as an oncogene which increases the survival, proliferation, metastatic potential and chemotherapy resistance of tumor cells. Hence, it has been recently identified as a potential therapeutic target in a wide range of tumor entities with deregulated Axl expression including prostate cancer, glioma, lung cancer and CLL. Here, we investigated two different Axl inhibitors for their ability to inhibit the migratory capacity and survival of leukemic cells in preclinical CLL models.

In vitro studies: We measured soluble Axl plasma concentrations by enzyme-linked immunosorbent assay (ELISA) in 71 CLL patients and 24 healthy donors. Soluble Axl levels were not significantly higher in CLL patients compared to healthy donors (p=0. 11). However, in CLL patients high sAxl plasma concentrations were differentially expressed with some patients exhibiting normal and others elevated plasma concentrations. The latter showed an association with shorter time to first treatment (p=0.0005) and several poor prognostic markers (e.g. CD38, FISH cytogentics, Binet stage).

Freshly isolated PBMC (>90% CD5+CD19+) from CLL patients were incubated in serum free medium for 48h containing concentrations series of 2 different Axl inhibitors: BMS777607, a previously published inhibitor of the MET kinase family, and LDC2636, a novel inhibitor of the TAM receptor tyrosine kinase (RTK) family with high affinity to Axl. Viability of CLL cells was assessed by trypan blue staining and flow cytometry employing annexin V staining. Cellular polarization was analyzed by time-lapse microscopy. We detected cytotoxic effects in a patient-dependent manner that were more prevalent in LDC2636 as compared to BMS777607 treated cells (IC₅₀= 0.21 µM vs. 2.88 µM, p<0.05, n=5). The cellular polarization of the remaining viable cells was significantly reduced in a dose dependent fashion in comparison to vehicle only controls (LDC2636 IC₅₀ = 7.2 µM, p<0.00001; BMS777607: IC₅₀=6.2µM; p=0.0004). Of note, both Axl inhibitors exhibited significantly weaker effects on both, the viability and polarization of normal PBMC over the whole concentration range tested (p<0.05, n=5).

In vivo studies To verify our hypothesis that reduced cell polarization results in decreased homing of leukemic cells in vivo we employed a recently developed adoptive transfer model of CLL. In this model NOD/SCID/IL2Rgc^null(NSG) mice were pre-treated with a single intraperitoneal (i.p.) bolus of LDC2636 or BMS777607 (20 mg/kg) and subsequently transplanted with primary CLL cells. Both Axl inhibitors significantly reduced the homing capacity of CLL cells to the BM of NSG mice by 46% and 59%, respectively, compared to vehicle treated controls (LDC2636: p=0.0063, BMS777607 p=0.0007; n=4).

To evaluate if LDC2636 also exhibits effects in a disease-relevant CLL model, we applied a MEC-1 xenograft model which causes a lethal leukemia in NSG mice. We pretreated the mice with 40mg/kg Axl inhibitor or vehicle-only control i.p. and subsequently transplanted the CLL cell line MEC-1 intravenously. The following four days the mice were injected again with 40mg/kg LDC2636 or vehicle-only i.p. We evaluated the survival time and found that mice treated with LDC2636 lived significantly longer than vehicle-only controls (24 vs. 18 days median survival, p=0.0016, n=15). Mice that received only LDC2636 and no Mec-1 cells did not show any effect.

These data demonstrate that Axl inhibitors exert potent in vitro and in vivo activity against human CLL cells, which is caused at least in part by the suppression of CLL homing to their supportive stromal niches.