Preclinical Studies of Salinomycin In Multiple Myeloma (MM) Models: Targeting of Side Population (SP) Cells In the Context of Tumor – Microenvironment Interactions.

Abstract 1574

Cancer cells with stem cell-like features are a topic of intense research because their resistance to existing drugs is considered a culprit for relapses, even in patients with complete remission defined by clinical, biochemical and imaging parameters or by sensitive molecular techniques. Salinomycin, an antibacterial and coccidiodostatic ionophore, is reported (Cell 2009;138(4):645-59) to be >100-fold more potent against breast cancer cells with stem cell-like phenotype after mesenchymal transdifferentiation due to stable transfection with shRNA against CDH1 than against the parental cells. We evaluated whether salinomycin could also exhibit a similar activity against stem cell-like cells in multiple myeloma (MM). To establish a comparative reference for such potential activity, we first tested salinomycin (0-10 uM for up to 72hrs) against a panel of 15 MM cell lines and observed IC₅₀ values <1 uM in 10/15 cell lines tested, including >80% reduction of tumor cell viability in 6/15 cell lines tested at 0.5 uM, i.e. levels lower than the IC50 values for in vitro activity of salinomycin against breast cancer cells with (HMLE-shCDH1, IC₅₀ ∼1 uM) or without (HMLE-shControl, IC50 >>10 uM) stem cell-like features. CD138+ purified primary tumor cells from 3 MM patients responded to salinomycin with IC₅₀ values (105, 332 and 750 nM, respectively) in the same range as MM cell lines. In vitro combinations with bortezomib, doxorubicin, melphalan, and dexamethasone showed overall no antagonism, while evidence of additive or even synergistic effect could be identified in certain dose ranges. Because MM cell lines and primary tumor cells responded concordantly to salinomycin and with higher sensitivity than breast cancer stem cell-like cells, we hypothesized that MM cells may in general be more responsive to salinomycin than other tumors. Since tumor-stromal interactions can increase the expression of transcriptional signatures of “stemness” in MM cells, we embarked on characterizing the anti-MM properties of salinomycin using compartment-specific bioluminescence imaging (CSBLI) assays. These showed that co-culture with stromal cells did not confer resistance to salinomycin in 5 MM cell lines (MM.1S, OCI-My5, KMS-11, KMS-18, NCI-H929) and in fact enhanced its activity against 4 of them. Side population (SP) cells, defined by their ability to efflux Hoechst stain, represent a stem cell-like population which was identified in MM cell lines and could represent the functional equivalent of the mesenchymally transdifferentiated breast cancer stem cell-like cells. We observed that salinomycin reduces the SP fraction of MM cell lines at doses >20 times lower than those required for in vitro effect against the bulk <<main population>> of the respective cell lines. Interestingly, the anti-SP effect of salinomycin was more pronounced in the presence of stroma, similarly to the CSBLI studies on the entire MM cell population and consistent with our prior observation that tumor-stroma interaction enhances transcriptional signatures of ≪stemness≫ in the tumor compartment. However, when we tested the in vivo anti-MM activity of salinomycin in an orthotopic model of i.v. injected Luc+ MM cells, no anti-MM activity (in terms of tumor burden decrease or overall survival prolongation) was observed at the maximum tolerated dose (1 mg/kg i.p. daily, which is consistent with most studies reported thus far in the literature). Ex vivo treatment of KMS-11 cells with salinomycin doses (100 nM for 72 hrs) selectively targeting SP cells was followed by s.c. injection of these cells or vehicle-treated controls in sublethallly irradiated SCID/NOD mice, but no statistically significant improvement in tumor burden or overall survival was observed. Our in vitro results indicate that salinomycin exhibits intriguing in vitro anti-MM activity, not only against SP cells but also against the bulk ≪main≫ MM cell population, even in the presence of stromal support. In contrast, the in vivo activity of salinomycin is compromised by side effects in the orthotopic model of MM lesions, while short term ex vivo exposure of tumor cells is conceivably insufficient to eradicate clonogenic cells and lead to appreciable delay in tumor growth in vivo. Our studies point to intriguing features as well as notable challenges that have to overcome before salinomycin or other more selective agents of this class can be safely tested in clinical trials in MM.