Development of a Pharmacodynamic Assay for XPO1 Occupancy Using Fluorescence Cross Correlation Spectroscopy (FCCS)

Introduction:

Selinexor, a Selective Inhibitor of Nuclear Export (SINE) compound, is an anti-cancer drug that is currently being evaluated in Phase I and II clinical trials for the treatment of solid and hematological malignancies (clinicaltrials.gov). XPO1 (exportin 1/CRM1), the target of selinexor, is commonly overexpressed in cancer. XPO1 is the main nuclear export protein with over 200 different protein cargos which include tumor suppressor and growth regulatory proteins. Binding of selinexor to XPO1 inhibits the nuclear export of TSPs and GRPs and induces cell cycle arrest followed by cancer cell apoptosis.Certain cancers are more sensitive to selinexor than others. We hypothesized that part of the inherent differences in sensitivities are attributed to different levels of target inactivation upon selinexor binding. To predict patient response to treatment, we developed a binding assay based on Fluorescence Cross Correlation Spectroscopy (FCCS) to measure XPO1 target occupancy in cancer cells.

Methods:

Cells were treated in culture for 4 hours with 0-10 µM selinexor. Cells were harvested, lysed, and cleared by ultracentrifugation. In vivo, nude mice were inoculated in both flanks with Z-138 tumor cells. Once tumors reached approximately 250 mm³, mice were administered a single dose of 0, 5, 10, 15, or 20 mg/kg selinexor. Tumors were harvested 6 hours post-dose, lysed, and cleared by ultracentrifugation. Cell or tumor supernatants were incubated with a fluorescently labeled antibody (Ab) against XPO1 and fluorescently labeled Leptomycin B (LMB). To allow binding of labeled Ab and LMB to XPO1, samples were measured by FCCS after 15, 120, and 1200 minutes. FCCS measures signal fluctuations induced by fluorescent molecules diffusing through an illuminated microscopic detection spot. The approach allows simultaneous monitoring of 2 different fluorescently labeled molecules and extracts information on particle concentration, size, binding state, and molecular brightness. Interacting particles are scored by codiffusion. FCCS measurements were performed to quantify the concentration of dually labeled XPO1 molecules in prepared lysates. In the absence of selinexor, complexes of XPO1, the labeled antibody (aXPO1AB_ATTO488), and the red labeled LMB (LMB_DY647) are formed and indicate low target occupancy. XPO1 molecules occupied by selinexor cannot bind to the labeled tracer and decrease the number of dually labeled particles. Accordingly, loss of complex formation is a measure of target occupancy in cells and tumor samples.

Results:

Comparison of cytotoxicity and XPO1 binding in hematological cell lines treated with increasing concentrations of selinexor in vitro indicated that XPO1 occupancy by selinexor occurred to the same extent regardless of drug sensitivity (Table 1). However, there were two observations that distinguished the selinexor-resistant cell line THP-1 from the other 3 sensitive cell lines (Table 1): 1. Untreated THP-1 cells had the highest XPO1 occupancy of fluorescently labeled LMB compared to lower values in more sensitive cells, and 2. Treatment of THP-1 cells with very low dose selinexor (10 nM) demonstrated lower saturation of XPO1 protein, while treatment of MV-4-11, Z-138, and MM.1S with 10 nM selinexor resulted in marked XPO1 occupancy. These results suggest that high levels of XPO1 proteins might lead to resistance to XPO1 inhibition. We are currently testing whether these two observations are general predictors of resistance in other cell lines. Mouse studies showed that XPO1 target occupancy could be measured in tumors and that there was a dose-dependent effect of selinexor on complex formation with >90% target saturation at 10 mg/kg selinexor (estimated C_max ~2400 ng/ml).

Conclusions: For the first time, the FCCS method permits the direct measurement XPO1 occupancy by selinexor in cells and tumors following treatment. Here, we demonstrate XPO1 occupancy saturation at 10 mg/kg (30 mg/m²) in mice, a dose active against xenografts, and consistent with doses of selinexor that show anti-cancer activity in patients with heavily pretreated hematologic and solid tumors.