(P 289) IDENTIFICATION OF EFFECTIVE THERAPEUTIC COMBINATIONS WITH THE NUCLEAR EXPORT PROTEIN INHIBITOR, SELINEXOR, IN PEDIATRIC SOLID TUMORS

Objective: Malignant rhabdoid tumor (MRT) and Wilms tumor (WT) comprise more than 5% of all pediatric cancers. Despite intensive multimodality therapy, outcomes remain dismal for a subset of patients with aggressive or high-risk molecular features. Characteristic of most pediatric cancers, MRT and WT demonstrate relatively low frequencies of somatic mutations compared to adult tumors and generally lack therapeutically targetable genetic alterations. Additionally, conventional monotherapy is often insufficient and more susceptible to the development of resistance in these aggressive tumors. Therefore, our group applies a novel systems biology approach to identify and evaluate non-genetically encoded vulnerabilities that can be targeted in combination in MRT, WT, and other pediatric solid tumors.

Methods: MetaVIPER analysis, an extension of the Virtual Inference of Protein-activity by Enriched Regulon (VIPER) algorithm, was performed to computationally infer protein activity from MRT and WT whole transcriptomic data available in the TARGET database. Expanded metaVIPER analysis demonstrated exportin-1 (XPO1) as having consistently high inferred activity. In vitro cell viability assays utilizing the selective inhibitor of XPO1, selinexor, in combination with other agents targeting proteins identified by metaVIPER were performed in solid tumor cell lines to evaluate the effects on proliferation. Analysis of drug combinations was performed using SynergyFinder 3.0. A malignant rhabdoid tumor cell line was treated sequentially with selinexor and the DNA-damaging agent, doxorubicin, and DNA damage recovery was analyzed by immunofluorescence using anti-γH2A.X antibody.

Results: MetaVIPER analysis identified consistent high inferred activity of XPO1 in MRT and WT compared to other tumor types. Additional targets, such as topoisomerases (TOP1, TOP2), aurora kinases (AURKA, AURKB), histone deacetylase (HDAC2), and cyclin-dependent kinases (CDK2, CDK4), were also identified as having relatively high inferred activity in MRT and WT (Figure 1). Synergy analyses demonstrated primarily additive responses when selinexor was used in combination with anti-tumor agents in MRT, WT, and other pediatric solid tumor cell lines (Figure 2). The combination of selinexor with palbociclib represents a promising combination in MRT. No combination with selinexor was antagonistic in either the MRT or WT cell lines. The sequential combination of doxorubicin followed by selinexor resulted in delayed recovery from damage caused by the DNA-damaging agent (Figure 3). Selinexor alone did not induce DNA damage. In vivo studies validating sequential and additive/synergistic combinations are ongoing.

Conclusion: XPO1 represents a non-genetically encoded vulnerability in MRT and WT. The sequential combination of a DNA-damaging agent followed by the XPO1 inhibitor, selinexor, results in delayed recovery from DNA damage, suggesting selinexor interferes with the cell’s ability to repair DNA damage and drug sequence may be an important consideration for combination therapy with this agent. Furthermore, targeting multiple non-genetically encoded vulnerabilities detected by metaVIPER represents a promising strategy for identifying therapeutic drug combinations. Additional in vivo validation studies are needed to support the development of a pediatric combination clinical trial.