EHA Library - The official digital education library of European Hematology Association (EHA)

In chronic lymphocytic leukemia (CLL), NOTCH1 is constitutively expressed promoting leukemia cell survival and resistance to apoptosis (Rosati et al, Blood 2009). Additionally, NOTCH1 PEST mutations have emerged as one of the most frequent somatic alterations in CLL, affecting up to 20% of patients (Sportoletti et al, Leukemia 2014). Thus, inhibiting NOTCH1 activity represents a potential therapeutic opportunity for this disease and the incorporation of NOTCH1 pathway antagonists may improve standard CLL treatment. Targeting NOTCH1 has been a therapeutic strategy of interest in many cancers. However, the use of gamma secretase inhibitors (GSI) evaluated in clinical trials showed on target toxicities suggesting the need for the discovery of more selective NOTCH1 pathway antagonists that preferentially target NOTCH1 versus NOTCH2 or that target mutated receptors compared to wild-type. Recently, an expression-based screen identified several calcium modulators as a potential strategy to target NOTCH1 (Roti et al, Cancer Cell 2013). Among numerous ion flux modulators validated to induce a NOTCH1 off signature, one of the top hits was the clinically relevant calcium channel blocker, bepridil, used to treat patients with cardiac disease. Bepridil demonstrated anti NOTCH1 modulating activity in T-ALL by a mechanism unique from GSI (Roti et al, ASH 2009).

The goal of the study was to evaluate whether bepridil exerts antitumor activity in primary CLL cells in vitro and in a xenotransplant model and whether these effects are associated with NOTCH1 inhibition.

In vitro, we evaluated NOTCH1 expression in primary CLL cells after 24 hours of bepridil treatment using western blot and flow cytometric analysis. We also measured apoptosis using annexin V/propidium iodide staining and by assessing PARP, MCL-1 and NOXA expression. In vivo, CLL cells were transplanted into NSG mice and engraftment was evaluated after 28 days of bepridil treatment.

Bepridil treatment reduced the viability of primary CLL cells at a 2.5 μM concentration. In treated CLL cells, viability significantly decreased to 23.8±20.2% compared to 41.5±21.8% of the vehicle control (N=40, p <0.0001). Conversely, bepridil did not affect viability of normal T cells from CLL patients nor the viability of B and T cells from healthy donors, demonstrating that bepridil selectively impairs the viability of B neoplastic cells compared to normal hematopoietic cells. Bepridil significantly increased the percentage of annexin V/propidium iodide positive apoptotic CLL cells compared to vehicle treated cells (49.2±20.9% vs 29.3±15.3% N=34   p<0.0001). These apoptotic effects were supported by the detection of increased PARP degradation (p<0.05), significant reduction of MCL-1 protein expression (p<0.001) and a 5-fold up-regulation of NOXA transcript levels. Induction of apoptosis with bepridil treatment in cultured CLL cells was not correlated with ZAP-70 expression, IGVH rearrangement or NOTCH1 mutation status. However, the NOTCH1 pathway was inhibited at concentrations of bepridil that induced apoptosis. Specifically, flow cytometric analysis of cultured CLL cells (N= 15) demonstrated that bepridil significantly reduced the surface expression of NOTCH1 compared to vehicle (32.8±19.2% vs. 50.6±18.9% respectively, p<0.05). As previously reported in T-ALL bepridil treatment lead to down regulation of the trans-membrane bound portion of NOTCH1. Notably, NOTCH2 protein level remained unchanged in bepridil treated CLL samples compared to vehicle (N=3), supporting a preferential effect of bepridil on NOTCH1. Interestingly, we observed that CLL cells co-cultured with different stromal layers (including primary mesenchymal cells, HS5 and OP9 cell lines) fail to apoptose upon bepridil treatment. The addition of a CXCR4 antagonist restored bepridil efficacy, suggesting a synergistic effects against the survival stimuli of the stroma. Finally, we established CLL NSG primagrafts and tested bepridil in this leukemia model. Strikingly, flow cytometry analysis revealed a statistically significant decrease in human CD45+CD19+CD5+ percentage in the spleen of bepridil-treated mice compared to vehicle (1.9±1% vs. 10.8±10%, N=13, p<0.05). As anticipated by our results in vitro bepridil treatment did not have a major effect in the neoplastic bone marrow cell population of treated animals.

In conclusion, we showed that bepridil reduced viability and increased apoptosis in primary CLL cells in vitro. This antileukemic effect is associated with the inhibition of the NOTCH1 signal. Moreover, this clinically relevant drug demonstrated efficacy in controlling splenic disease in a mouse model of human CLL, suggesting a potential for rapid translation to clinical testing.