Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jun 1.
Published in final edited form as: Mol Cancer Ther. 2021 Sep 28:10.1158/1535-7163.MCT-21-0370. doi: 10.1158/1535-7163.MCT-21-0370

Rational Combination of CRM1 Inhibitor Selinexor and Olaparib Shows Synergy in Ovarian Cancer Cell Lines and Mouse Models

Katelyn F Handley 1,2,3, Cristian Rodriguez-Aguayo 4, Shaolin Ma 1, Elaine Stur 1, Robiya Joseph 1, Emine Bayraktar 1, Santosh K Dasari 1, Nghi Nguyen 5, Reid T Powell 5, Mary Sobieski 5, Cristina Ivan 4, Mark Kim 1, Sujanitha Umamaheswaran 1, Deanna Glassman 1, Yunfei Wen 1, Paola Amero 4, Clifford Stephan 5, Robert L Coleman 6, Yosef Landesman 7, Shannon N Westin 1, Prahlad T Ram 8, Anil K Sood 1
PMCID: PMC8643313  NIHMSID: NIHMS1745855  PMID: 34583979

Abstract

CRM1 inhibitors have demonstrated antitumor effects in ovarian and other cancers; however, rational combinations are largely unexplored. We performed a high-throughput drug library screen to identify drugs that might combine well with selinexor in ovarian cancer. Next, we tested the combination of selinexor with the top hit from the drug screen in vitro and in vivo. Finally, we assessed for mechanisms underlying the identified synergy using reverse phase protein arrays (RPPA). The drug library screen assessing 688 drugs identified olaparib (a PARP inhibitor) as the most synergistic combination with selinexor. Synergy was further demonstrated by MTT assays. In the A2780luc ip1 mouse model, the combination of selinexor and olaparib yielded significantly lower tumor weight and fewer tumor nodules compared with the control group (P < 0.04 and P < 0.03). In the OVCAR5 mouse model, the combination yielded significantly fewer nodules (P = 0.006) and markedly lower tumor weight compared with the control group (P = 0.059). RPPA analysis indicated decreased expression of DNA damage repair proteins and increased expression of tumor suppressor proteins in the combination treatment group. Collectively, our preclinical findings indicate that combination with selinexor to expand the utility and efficacy of PARP inhibitors in ovarian cancer warrants further exploration.

Introduction

Ovarian cancer is a very challenging disease to manage given its late stage at diagnosis and predilection for recurrence, eventually leading to resistant disease for which there are limited treatment options (1). The search continues for more effective drug combinations and chemotherapy adjuncts to battle this disease.

The overexpression of chromosome region maintenance 1 (CRM1/XPO1) and its correlation with poor prognosis have been demonstrated in ovarian cancer and other solid malignancies such as pancreatic cancer, glioma, and osteosarcoma (2–5). CRM1 inhibitors are selective inhibitors of nuclear export (SINE). They prevent the nuclear export of tumor suppressor proteins, such as TP53, RB1, and FOXO proteins; eIF4E, which facilitates the export and translation of proto-oncogene mRNA (6–7); and IGF2BP1, which binds eIF5A in the cytoplasm, preventing its transport to the mitochondria where it would elicit an apoptosis cascade (8–14). We have previously demonstrated the antitumor effects of selinexor, a CRM1 inhibitor, in ovarian cancer mouse models (8). A phase II trial of selinexor in patients with recurrent ovarian cancer achieved a 30% disease control rate (15), and selinexor is currently being tested further in clinical trials. However, rational combinations with selinexor have yet to be determined. Therefore, we aimed to identify combinations with selinexor to optimize its use in ovarian cancer.

Materials and Methods

Cell lines and culture conditions

Human ovarian cancer cell lines including A2780 (RRID: CVCL_0134) variants, HeyA8 (RRID: CVCL_8878), and SKOV3ip1 (RRID: CVCL_0C84) were obtained from The University of Texas MD Anderson Cancer Center Cytogenetics and Cell Authentication Core Facility; OVCAR3 (RRID: CVCL_0465) was obtained from Dr. Gabriel Lopez-Berestein (originally obtained from the NIH); OVCA432 (RRID: CVCL_3769) was obtained from Dr. Ronny Drapkin at Dana Farber Cancer Institute; COV362 (RRID: CVCL_2420) was obtained from The European Collection of Authenticated Cell Cultures via Sigma; Kuramochi (RRID: CVCL_1345) was obtained from the Japanese Collection of Research Bioresources Cell Bank; OVCAR4 (RRID: CVCL_1627), OVCAR5 (RRID: CVCL_1628), and OVCAR8 (RRID: CVCL_1629) were obtained from the NIH/NCI; and Caov3 was obtained from ATCC. The immortalized human vascular endothelial cell line RF24 (RRID: CVCL_AX74) was obtained from Dr. Lee Ellis. A2780 variants, HeyA8, SKOV3ip1, OVCAR3, OVCAR4, OVCAR8, OVCA432, and Kuramochi cells were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS) and 1% gentamycin. OVCAR5, Caov3, and COV362 cells were cultured in DMEM medium supplemented with 10% FBS and 1% gentamycin. RF24 cells were cultured in MEM medium supplemented with 10% FBS, sodium pyruvate, amino acids, and 1% gentamycin. All cells were cultured at 37°C with 5% CO2. All cell lines were authenticated by the core facility by short tandem repeat profiling and were routinely tested for Mycoplasma by PCR. Cells were used within 20 passages from thaw for in vitro experiments and 10 passages from thaw for in vivo experiments.

Both A2780 variants and OVCAR5 are BRCA wild-type and homologous recombination-proficient (HRP, 16–19).

Western blot

Total cell extracts were prepared in lysis buffer [25 mmol/L Tris (pH 7.5), 150 mmol/L NaCl, 0.1% SDS, 0.5% sodium deoxycholate, 1% Triton-X] and Halt Protease Inhibitor Cocktail (Cat. 87786, Thermo Scientific) and Phosphatase Inhibitor Cocktail (Cat. 78420, Thermo Scientific). Five micrograms of cell lysate were boiled at 100°C for 5 minutes and loaded onto a 4-15% gradient polyacrylamide gel. After separation, proteins were transferred onto nitrocellulose membranes. Membranes were blocked with 5% milk in TBST for 30 minutes, then incubated with anti-human CRM1 antibody (1:1000, Cat. ab24189, Abcam, RRID: AB_2257217), RAD51 antibody (1:4000, Cat.ABE257, Millipore, RRID: AB_10850319), or CHEK1 antibody (1:1000, Cat. 2360, Cell Signaling Technology, RRID: AB_2080320) at 4°C overnight. After washing with TBST, the membranes were incubated with horseradish peroxidase-conjugated donkey anti-rabbit IgG (1:3000, Cat. NA934, GE Healthcare), anti-rabbit IgG (1:4000, Cat. NA934V, GE Healthcare), or anti-mouse IgG (1:2000, Cat. NA931V, GE Healthcare) for 1 hour at room temperature. Anti-human β-actin antibody (1:2000, Cat. A5441, Sigma, RRID: AB_476744) or vinculin antibody (1:3000, Cat. V9131, Sigma) was used as a loading control. Protein band intensity was quantified using ImageJ software (National Institutes of Health).

High-throughput screening

Initial high-throughput screening

High-throughput screens were performed by the Gulf Coast Consortia’s (GCC) Combinatorial Drug Discovery Program at Texas A&M Health Science Center: Institute of Bioscience and Technology. Cell lines were screened against a collection of three drug libraries, namely The Broad Informer Set, NCI Approved Oncology Drugs (AOD) Set VIII, and an in-house collection of clinically relevant drugs referred to as Custom Clinical. A detailed description of the contents of these libraries can be found at https://ibt.tamu.edu/cores/high-throughput/approved-drugs-2.pdf.

For screening assays, a total of 500 cells per well suspended in 50 μL of media were seeded into Greiner Black 384-well μClear plates (Cat. 781091) using a Multidrop Combi liquid dispenser (ThermoFisher Scientific). The plates were kept at room temperature after seeding for 40-60 minutes before placing them into a cell culture incubator. The cells were allowed to form a monolayer overnight at 37°C in a humidified chamber (> 95% relative humidity) with 5% CO2. After recovery, 50 nL of drugs were transferred into each well using an Echo 550 acoustic dispensing platform (Labcyte). A non-treated plate was fixed and nuclei stained with 4′,6-diamidino-2-phenylindole (DAPI) at the time of drug treatment (Day 0) to provide the number of cells present at the time of treatment. In the primary screen, three concentrations were tested (1 μM, 0.1 μM, and 0.01 μM) with a fixed volume of DMSO (0.1% v/v) and two biological replicates. Each assay plate contained a fixed concentration of the drugs in addition to a negative control (0.1% DMSO), two positive controls (10 μM doxorubicin and selinexor), and an 8-point dose-response curve of the selinexor. After 72 hours of incubation in the presence of drug, plates were fixed with 0.4% paraformaldehyde and nuclei stained with DAPI using an integrated Hydrospeed plate washer (Tecan Life Sciences) and Multidrop Combi dispenser. Plates were imaged on an IN Cell Analyzer 6000 laser-based confocal imaging platform (General Electric Healthcare Bio-Sciences) and nuclei counted using the algorithms developed using the IN Cell Developer Toolbox software (version 1.6).

Statistical analysis for high-throughput screening

Statistical analysis of assay performance was done in accordance with the National Center for Advancing Translational Sciences Assay Guidance Manual (20). In brief, a running statistical evaluation was performed on each plate throughout the course of the screening campaign to evaluate the consistency of results. Metrics evaluated included the rate of growth of the negative controls, the coefficient of variance of the positive and negative controls, and assay robustness determined from the Z’ statistic. Assay reproducibility and experimental drift were determined using the minimum significance ratio calculated from the standard deviation of half-maximal inhibitory concentration (IC50) values of the on-plate positive control curves. Pharmacologic data were normalized using the growth index proposed by Hafner et al (21). This method effectively minimizes the influence of the rate of growth on measurements of drug susceptibility, thus providing a robust metric to make comparisons between cell lines while also providing the ability to differentiate cytotoxic from cytostatic effects. The data were fitted against a cascade of nonlinear regression models, each with different initialization criteria, to identify the best fit using a combination of R and Pipeline Pilot (Dassault Systemes/Biovia) software platforms. Finally, dose-response data were summarized as an area under the curve value calculated by numerically integrating fitted concentration-response curves.

High-throughput screening combination of anchor and probes

We performed a high-throughput combination screen using an anchor-probe strategy on two cell lines (A2780ip2 and OVCAR5). The anchor, selinexor, was combined with 34 probe compounds identified in the single-agent high-throughput screen. In total, eight concentrations of probe molecules were tested in pairwise combination with five concentrations of the anchor (100 nM, 55 nM, 30 nM, 20 nM, and 10 nM). By testing varying stoichiometries of anchors and probes, we effectively minimized the false-negative rate and maximized the chance of detecting synergistic interactions.

Drug interactions were binned into three classes (antagonistic, additive, or synergistic) by comparing the experimentally observed dose-response surface to a theoretical surface calculated from the independent single-agent activity using a Bliss synergy model. To fulfill the statistical requirements of this model, raw cell count data were normalized from 0 (inactive) to 1 (full-active). To minimize the detection of false positives due to assay artifacts or the presence of outliers, we fitted the three-dimensional combination surface via a support vector regression-based approach using the e1071 package in R (22, 23). The overall interaction was then determined by subtracting the volume between the fitted Bliss and experimentally observed surfaces.

The assay was highly robust, with a mean Robust Z’ of 0.84 and 0.82 for the A2780ip2 and OVCAR5 respectively. There was also high inter-plate reproducibility of dose response information, which was determined using the minimum significant ratio (MSR) calculation from on-plate doxorubicin control curves. Those data show MSR values of 1.23 and 1.76 for the A2780ip2 and OVCAR5 respectively. For this statistic, a value less than 3 is generally considered acceptable.

Orthogonal cell viability assay

To evaluate the cytotoxicity of selinexor and olaparib, alone and in combination, cells were plated in a 96-well plate with an initial density of 2000 cells per well for A2780 or 3000 cells per well for OVCAR5. After 24 hours, once cells were attached, culture media was removed and replaced with 100 μL culture media containing serial dilutions of selinexor and olaparib. After 72 hours incubation, cells were incubated with 0.05% MTT solution for 90 minutes. The supernatant was removed and the MTT formazan dissolved in 100 μL dimethyl sulfoxide. The absorbance in each well was determined at 570 nm by a BioTek uQuant microplate spectrophotometer. Triplicate biological replicates were performed. Dose-response curves were plotted in GraphPad Prism 8.0.0, the IC50 and Combination Index (CI) determined by CompuSyn software (combosyn.com) (24), and synergy further assessed using the Bliss model in SynergyFinder (https://synergyfinder.fimm.fi) (25). The Bliss synergy scores in this platform indicate synergy if greater than 10, additivity if less than or equal to 10 but greater than or equal to −10, and antagonism if less than −10. Selinexor was provided by Karyopharm Therapeutics. Olaparib was purchased from LC Laboratories (Cat. O-9201).

In vivo models of ovarian cancer

All in vivo protocols were approved by MD Anderson Cancer Center’s Institutional Animal Care and Use Committee. Female athymic nude mice were purchased from Taconic Biosciences. Mice were injected intraperitoneally with 1 × 106 A2780luc ip1 or OVCAR5luc cells in 200 μL Hanks’ Balanced Salt Solution (Cat. 21-021-CV, Corning). In vivo luciferase imaging was conducted as described previously to observe the mice for tumor uptake (26). Imaging started 6 days after cell injection using the IVIS Spectrum in vivo imaging system coupled to Living Image software (Xenogen) for image analysis. Treatments were initiated 10 days after cell injection, once IVIS demonstrated tumor uptake. The mice were randomly divided into four treatment groups of 12 to 13 mice each: vehicle control, selinexor treatment alone, olaparib treatment alone, and combination treatment with selinexor and olaparib. Selinexor was dissolved in 0.6% Plasdone PVP K29/32/0.6% Poloxamer Pluronic F-68 in sterile water and administered via oral gavage (7.5 mg/kg three times weekly). Olaparib was reconstituted in 10% dimethyl sulfoxide and 10% (2-Hydroxypropyl)-beta-cyclodextrin (Cat. H107-100G, Sigma) in phosphate-buffered saline and administered via oral gavage (50 mg/kg daily). Treatment continued for 16 to 23 days. Once any group became moribund, all mice were euthanized. Mouse weight, tumor weight, and number and location of tumor nodules were recorded. Tumor specimens were formalin-fixed and paraffin-embedded, embedded and frozen in optimal cutting temperature compound (Mercedes Scientific), or snap-frozen.

Reverse phase protein array and NetWalker analysis

A2780 cells were treated with vehicle control, selinexor (0.1 μM), olaparib (1.2 μM), or the combination of selinexor and olaparib for 24 hours before collection of cell lysates. Cells were lysed in 150μL of reverse phase protein array (RPPA) lysis buffer (1% Triton X-100, 50 mM HEPES, pH 7.4, 150 mM NaCl, 1.5 mM MgCl2, 1 mM EGTA, 100 mM NaF, 10 mM Na pyrophosphate, 1 mM Na3VO4, 10% glycerol, containing protease and phosphatase inhibitors from Roche Applied Science, Cat. 05056489001 and 04906837001, respectively). The lysates were centrifuged for 10 minutes at maximum speed before collection of the supernatant. The protein concentration was adjusted to 1.5 μg/mL, and the cell lysate was mixed with 4XSDS sample buffer (40% glycerol, 8% SDS, 0.25 M Tris-HCL, pH 6.8 with 10% beta-mercaptoethanol) in a ratio of three to one. Otherwise, the RPPA was performed by MD Anderson Cancer Center’s Functional Proteomics RPPA Core Facility as previously described (27). Data were analyzed using NetWalker software (28).

Data availability

Key data and code produced in this manuscript is available at https://zenodo.org/badge/latestdoi/381717015.

Statistical analysis

The Shapiro-Wilk test was applied to determine whether data were normally distributed. For normally distributed data, a Student t-test was used to compare two groups. For data with a non-parametric distribution, a Mann-Whitney-Wilcoxon test was used. A P value less than 0.05 was considered statistically significant. All statistical tests were two-sided. Box-and-whisker plots were used to visualize the data. Analyses were carried out in the R statistical environment (version 4.0.3) (https://www.r-project.org/).

Results

CRM-1 expression in ovarian cancer cells

Western blot demonstrated baseline CRM1 protein expression in 12 of 13 ovarian cancer cell lines tested, as well as in the endothelial cell line RF24 (Supplementary Figure 1). The Broad Institute Cancer Cell Line Encyclopedia RNA sequencing data were also queried, and CRM1 mRNA expression was demonstrated in all 47 ovarian cancer cell lines tested (Supplementary Table 1).

High-throughput drug library screening identifies rational combinations with selinexor

A systematic high-throughput drug screen was carried out to test the effect of selinexor and three drug libraries containing a total of 688 drugs on cell growth assays (Figure 1A). The three drug libraries were comprised of both FDA-approved agents and clinical candidates. Thirty-two compounds with distinct mechanisms of action demonstrated substantial growth inhibition in A2780 cells when tested as single agents. To identify drugs synergistic with selinexor, these 32 drugs, as well as the three FDA-approved poly (ADP-ribose) polymerase (PARP) inhibitors as standard of care drugs, were then tested in combination with selinexor (Figure 1B). Four of these compounds showed a synergistic effect when tested in combination with selinexor; among these were PARP inhibitors olaparib, rucaparib, and niraparib, with olaparib demonstrating the greatest synergistic interaction score (Figure 1C, D). In a second cell line, the high-grade serous ovarian cancer cell line OVCAR5, the three analyzed PARP inhibitors demonstrated at least additive effects in combination with selinexor (Supplementary Figure 2).

Figure 1.

Figure 1.

Open in a new tab

A) Flowchart illustrating the process used for high-throughput drug screening to identify synergistic drug combinations. B) Waterfall plot of the synergy scores for drug combinations with selinexor from most synergistic to most antagonistic. Green indicates synergy (> 1), yellow additivity (≤ 1 and ≥ −1), and red antagonism (< −1). C and D) The combinatorial analysis outcome panels including dose-response matrix (C) and difference from predicted Bliss Independence surface model (D) for the top four most synergistic drug combinations with selinexor, including olaparib, dactinomycin, rucaparib, and niraparib, in A2780 ovarian cancer cells. Log molar concentrations of each drug are indicated. In panel (C), the colors indicate raw cell count data normalized from 0 (inactive) to 1 (full-active). In panel (D), green regions represent synergy, red regions antagonism, and black additivity.

The combination of selinexor and olaparib increases cytotoxicity in ovarian cancer cell lines

Given the utility of PARP inhibitors for ovarian cancer and the observed effects from our screen, we verified the in vitro synergistic effect between selinexor and olaparib, the top “hit” from high-throughput drug screening, on A2780 cells with an MTT assay. The IC50 levels of selinexor and olaparib alone in A2780 were 0.1 μM and 1.1μM, respectively, after 72 hours of treatment (Figure 2A). The combination of selinexor and olaparib demonstrated a synergistic effect on A2780 cells, indicated by a combination index less than 1 (Figure 2B). Next, we applied the Bliss synergy model to calculate synergy scores using SynergyFinder. The summary synergy score indicated additivity across all doses tested, and indicated synergy as the dose ranges were fine-tuned (Figure 2C). The dose ranges with strongest synergy are highlighted in Figure 2C. The maximum synergistic response occurred between 0.1 to 0.2 μM selinexor and 0.4 to 1.6 μM olaparib.

Figure 2.

Figure 2.

Open in a new tab

Effects of single-agents selinexor and olaparib and their combination on cell viability in A2780 ovarian cancer cells in vitro. A) Mean half-maximal inhibitory concentration (IC50) of selinexor and olaparib in A2780 cells after 72 hours of treatment from three independent experiments. B) A2780 cells were treated with selinexor and olaparib alone and in combination at the indicated concentrations for 72 hours followed by MTT analysis to determine the percent cell viability. Combination index (CI) values were mostly less than one, indicating synergy. CI values indicate additivity when equal to one and antagonism when greater than one. Data are representative of three independent experiments. C) Visualization of the calculated 2D synergy maps from the SynergyFinder Bliss Independence model combinatorial analysis of the MTT cell viability assay in a larger dose range (left) and a more fine-tuned dose range (right). Red regions represent synergy and green regions represent antagonism. White rectangles designate the region of maximum synergy. Synergy scores were calculated using SynergyFinder. Data are representative of at least five experiments.

Selinexor and olaparib suppress tumor growth in ovarian cancer xenograft models

To investigate the antitumor effects of selinexor and olaparib in vivo, we created an ovarian cancer mouse model by intraperitoneal injection of A2780luc ip1 cells (Figure 3A). At the conclusion of the experiment, mice treated with selinexor or olaparib monotherapy demonstrated lower median tumor weight and fewer tumor nodules compared with the control group, although the differences did not achieve statistical significance. In contrast, the combination of selinexor and olaparib yielded significantly lower tumor weight and fewer nodules than the control group (P < 0.04 and P < 0.03, respectively; Figure 3B and C, Supplementary Table 2). Additionally, the body weights of the mice in all four groups did not differ significantly (Supplementary Figure 3).

Figure 3.

Figure 3.

Open in a new tab

The antitumor effect of selinexor combined with olaparib in ovarian cancer xenograft models. A) Schematic of in vivo experimental timeline. B) Effect of olaparib (50 mg/kg oral gavage daily), selinexor (7.5 mg/kg oral gavage twice weekly), and the combination (olaparib 50 mg/kg oral gavage daily and selinexor 7.5 mg/kg oral gavage twice weekly) on tumor weight in mice bearing A2780luc ip1 or OVCAR5luc tumors. C) Effect of olaparib, selinexor, and the combination on the number of tumor nodules in mice bearing A2780luc ip1 or OVCAR5luc tumors. Student’s t-test and Mann-Whitney-Wilcoxon test were used as indicated. All statistical tests were two-sided. Box-and-whisker plots are shown, with the box representing the first (lower) and third (upper) quartiles and whiskers representing 1.5 times the interquartile range. *P < 0.05, **P<0.01.

To extend our investigation into a high-grade serous ovarian cancer model, we next evaluated selinexor and olaparib in the OVCAR5luc tumor model with 11 to 12 mice per group. We observed lower median tumor weight and lower median number of tumor nodules in all three treatment groups compared with the control group (Figure 3B and C). The combination of selinexor and olaparib yielded significantly fewer nodules (P = 0.006) and markedly lower tumor weight compared with the control group (P = 0.059, Figure 3B and C). In this model, there was an observed difference in the body weights of the treated mice compared with the control mice at the time of necropsy; however, the control group also had a higher mean body weight before treatment, and more than half of the difference in body weights can be accounted for in the difference in tumor weights (Supplementary Figure 3A and B).

Selinexor combined with olaparib leads to decreased activity of DNA damage repair pathways and increased expression of tumor suppressors

To identify possible mechanisms behind the synergy between selinexor and olaparib, we performed RPPA analysis. We treated A2780 cells with either vehicle control, selinexor, olaparib, or the combination prior to analysis in order to assess alterations in the downstream pathways between treatment groups.

The vast interconnections between the CRM1 (known target of selinexor) network and the network of analyzed proteins downregulated by selinexor, as well as the interconnections between various parts of the PARP1 (known target of olaparib) network and the network of analyzed proteins downregulated by olaparib, validated that the RPPA analyzed a meaningful sample of proteins from each network (Supplementary Figure 4A, B). Broadly, NetWalker analysis of the proteins downregulated by the combination and their network connections to CRM1 and PARP1 demonstrated a greatly magnified number of interactions compared to either drug alone (Supplementary Figure 4A, B, C). The many interactions between the networks and pathways of the two drugs’ targets may play a role in their synergy (Supplementary Figure 4C). In contrast, the interactions between the networks and pathways of CRM1 and GPX4 (known target of ML162, the most antagonistic drug paired with selinexor in the drug screen) were much fewer (Supplementary Figure 4D).

As expected, we identified decreased expression of DNA damage repair proteins in both the selinexor arm and the combination arm (Figure 4). These included proteins involved in homologous recombination (e.g. RAD51, BRCA1, CHEK1, and phosphorylated CHEK1 (29, 30)) and mismatch repair (e.g. MSH2, MSH6), as well as FOXM1 (31), the “master regulator” of DNA damage repair (Figure 4A, B). The combination decreased the expression of RAD51, BRCA1, CHEK1, and FOXM1 by at least 20% compared with olaparib alone (Figure 4E). The decrease in RAD51 and CHEK1 levels was confirmed by Western blot (Supplementary Figure 5A, B). NetWalker identified downregulated pathways after treatment with the combination which include positive regulation of the cell cycle, FOXM1 transcription factor network, G1/S transition of the mitotic cell cycle, G2/M transition of the mitotic cell cycle, and DNA damage checkpoint. The analysis of the homologous recombination repair network in NetWalker demonstrated that all analyzed proteins were downregulated (blue) or unchanged (white), with the exception of TP53BP1, which is antagonistic to the homologous recombination repair pathway (32, 33) (Supplementary Figure 6A). The analysis of the mismatch repair network in NetWalker also demonstrated that all analyzed proteins were downregulated (blue) (Supplementary Figure 6B).

Figure 4.

Figure 4.

Open in a new tab

Reverse phase protein array results demonstrating decreased expression of DNA damage repair proteins and increased expression of tumor suppressor proteins with the combination of selinexor and olaparib. A and B) The network of proteins with decreased expression (by at least 20%) after treatment with the combination of selinexor and olaparib (A) and olaparib alone (B) relative to vehicle control. C and D) The network of proteins with increased expression (by at least 20%) after treatment with the combination of selinexor and olaparib (C) and olaparib alone (D) relative to vehicle control. E) The network of proteins with decreased expression (by at least 20%) after treatment with the combination of selinexor and olaparib relative to olaparib alone. F) The network of proteins with increased expression (by at least 20%) after treatment with the combination of selinexor and olaparib relative to olaparib alone. Networks determined using NetWalker analysis.

Interestingly, the combination also increased the expression of tumor suppressor proteins (TP53, CDKN1A (34)) to a greater extent than either monotherapy alone (Figure 4C, D, F). NetWalker identified upregulated pathways after treatment with the combination, including direct p53 effectors and the apoptosis pathway. Of note, CRM1 inhibitors are known to cause nuclear accumulation of TP53 and CDKN1A (6, 7, 35, 36); thus, these higher levels of proteins should be localized to the nucleus, where they can perform their function as tumor suppressors – assuming they are not structurally altered or misfolded.

Discussion

The key findings of our study include the synergy of olaparib and selinexor in vitro and the enhanced antitumor efficacy of the combination in vivo.

Since 2009, the rapidly evolving landscape of novel drugs in ovarian cancer has included PARP inhibitors, such as olaparib (37), which have become routinely used in the past few years. PARP inhibitors are clinically effective as monotherapy in patients with advanced recurrent BRCA-mutant or homologous recombination-deficient ovarian cancer, as maintenance treatment in patients with stage II-IV BRCA-mutant ovarian cancer following complete or partial response to frontline debulking surgery and chemotherapy, and as maintenance treatment in recurrent ovarian cancer following complete or partial response to a platinum-based chemotherapy regimen (38, 39, 40). PARP inhibitors function by inhibiting DNA single-strand break repair, promoting the probability of a double-strand break which can lead to cell death when homologous recombination repair machinery is absent or compromised - as well as through PARP trapping. Early studies demonstrated a therapeutic benefit from PARP inhibitors in BRCA-mutant ovarian cancers. More recently, their effect has been demonstrated in sporadic ovarian cancers with other deficiencies in the homologous recombination pathway (41) as well as those with defects in the mismatch repair pathway or nucleotide excision repair pathway (42).

As confirmed in this study, selinexor reduces the expression of DNA damage repair proteins, including those involved in homologous recombination and in mismatch repair (43), and it has been previously established that tumors which exhibit a defect in another DNA damage repair pathway—and more so two DNA damage repair pathways—will be more susceptible to PARP inhibitors (42). Selinexor can be exploited to artificially instill a defect in the DNA damage repair pathway in BRCA wild-type, HRP tumor cells (e.g., A2780 and OVCAR5), sensitizing them to olaparib. The combination with selinexor could expand the clinical utility of PARP inhibitors to those patients with BRCA wild-type, HRP tumors.

The robust in vitro and in vivo effects of the combination of selinexor and olaparib support further preclinical research and advancement of clinical trials assessing this all-oral combination. This combination may be developed to 1) enhance PARP inhibitor activity in biomarker negative cases, 2) enhance PARP inhibitor activity in biomarker positive cases, where PARP inhibitors already work as monotherapy, and 3) re-establish efficacy in tumors in which PARP inhibitors initially worked prior to the development of resistance. Our in vivo studies suggest that this combination is well tolerated; however, some overlapping side effect profiles between the two drugs, including anemia and neutropenia (15, 44–46), warrant a close assessment of the combination’s tolerability in clinical trials. Further study of PARP inhibitors with the new second-generation SINE, eltanexor, may also prove beneficial, as it was reported to be more tolerable than selinexor in preclinical studies (47–49).

Conclusion

Selinexor and olaparib are synergistic in HRP preclinical ovarian cancer cell models in vitro and in vivo. The combination with SINE compounds to expand the utility and efficacy of PARP inhibitors in ovarian cancer warrants further exploration.

Supplementary Material

1

Acknowledgements:

Editorial support was provided by Bryan Tutt, Scientific Editor, Research Medical Library. Pankaj Singh was instrumental in the conceptualization and development of the Texas A&M drug synergy analysis pipeline. The study drug, selinexor, was supplied by Karyopharm Therapeutics. K. F. Handley is supported by a training fellowship from the Gulf Coast Consortia, on the Computational Cancer Biology Training Program (CPRIT Grant No. RP170593). C. Rodriguez-Aguayo was funded by the NIH through the Ovarian SPORE Career Enhancement Program and NCI grant P50CA217685. S. Ma is supported by the Foundation for Women’s Cancer research grant (Sponsor Award number FP00009883). E. Stur is supported by Ovarian Cancer Research Alliance (OCRA number FP00006137). Y. Wen is supported by the Department of Defense Ovarian Cancer Research Program (W81XWH-20-1-0335), the National Institutes of Health Uterine Cancer SPORE P50CA098258, and the National Comprehensive Cancer Network. S. N. Westin is supported by P50 CA217685 and the GOG Foundation Scholar Investigator Award. A. K. Sood is supported by P50 CA217685, the American Cancer Society Research Professor Award, and the Frank McGraw Memorial Chair in Cancer Research. This research was supported by the NIH/NCI under award number P30CA016672 and used the Cytogenetics and Cell Authentication and Reverse Phase Protein Array (RPPA) core facilities. The screening work was supported by a CPRIT Core Facilities Support Award, RP150578, and the Gulf Coast Consortia High Throughput Screening Program Core grant, The Combinatorial Drug Discovery Program, #RP150578.

Conflicts of Interest:

SNW reports personal fees for consulting from Agenus, AstraZeneca, Circulogene, Clovis Oncology, Merck, Novartis, Pfizer, Roche/Genentech, GSK/Tesaro, Eisai, and Zentalis, and research funding to the institution from ArQule, AstraZeneca, Bayer, Bio-Path, Clovis Oncology, Cotinga Pharmaceuticals, Novartis, Roche/Genentech, and GSK/Tesaro. RLC reports consulting for AstraZeneca as well as participation in the trial steering committee for AstraZeneca. YL reports salary from Karyopharm and a stock option program from Karyopharm. AKS reports consulting for Astra Zeneca, Kiyatec, and Merck as well as being a shareholder in BioPath. All other authors declare no potential conflicts of interest.

Abbreviations List:

RPPA

reverse phase protein arrays

CRM1/XPO1

chromosome region maintenance 1

SINE

selective inhibitors of nuclear export

FBS

fetal bovine serum

HRP

homologous recombination-proficient

GCC

Gulf Coast Consortia

AOD

Approved Oncology Drugs

DAPI

4′,6-diamidino-2-phenylindole

IC50

half-maximal inhibitory concentration

MSR

minimum significant ratio

CI

Combination Index

PARP

poly (ADP-ribose) polymerase

References

  • 1.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin 2019;69(1):7–34. [DOI] [PubMed] [Google Scholar]
  • 2.Noske A, Weichert W, Niesporek S, Roske A, Buckendahl A, Koch I, et al. Expression of the nuclear export protein chromosomal region maintenance/exportin 1/Xpo1 is a prognostic factor in human ovarian cancer. Cancer 2008; 112(8): 1733–43. [DOI] [PubMed] [Google Scholar]
  • 3.Huang WY, Yue L, Qiu WS, Wang LW, Zhou XH, Sun YJ. Prognostic value of CRM1 in pancreas cancer. Clin Invest Med 2009; 32: e315. [PubMed] [Google Scholar]
  • 4.Shen A, Wang Y, Zhao Y, Zou L, Sun L, Cheng C. Expression of CRM1 in human gliomas and its significance in p27 expression and clinical prognosis. Neurosurgery 2009; 65: 153–9. [DOI] [PubMed] [Google Scholar]
  • 5.Yao Y, Dong Y, Lin F, Zhao H, Shen Z, Chen P, et al. The expression of CRM1 is associated with prognosis in human osteosarcoma. Oncol Rep 2009; 21: 229–35. [PubMed] [Google Scholar]
  • 6.Culjkovic-Kraljacic B, Fernando TM, Marullo R, Calvo-Vidal N, Verma A, Yang S, et al. Combinatorial targeting of nuclear export and translation of RNA inhibits aggressive B-cell lymphoma. Blood 2015; 127(7): 858–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Fischer MA, Friedlander SY, Arrate MR, Chang H, Gorska AE, Fuller LD, et al. Ventoclax response is enhanced by selective inhibitor of nuclear export compounds in hematologic malignancies. Blood Adv 2020; 4(3): 586–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Miyake T, Pradeep S, Wu SY, Rupaimoole R, Zand B, Wen Y, et al. XPO1/CRM1 inhibition causes antitumor effects by mitochondrial accumulation of eIF5A. Clin Cancer Res 2015; 21(14): 3286–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Culjkovic B, Topisirovic I, Skrabanek L, Ruiz-Gutierrez M, Borden KLB. eIF4E is a central node of an RNA regulon that governs cellular proliferation. J Cell Biol 2006; 175(3): 415–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Pourdehnad M, Truitt ML, Siddiqi IN, Ducker GS, Shokat KM, Ruggero D. Myc and mTOR converge on a common node in protein synthesis control that confers synthetic lethality in Myc-driven cancers. PNAS 2013; 110(29): 11988–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Culjkovic B, Topisirovic I, Skrabanek L, Ruiz-Gutierrez M, Borden KLB. eIF4E promotes nuclear export of cyclin D1 mRNA via an element in the 3’UTR. J Cell Biol 2005; 169(2): 245–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Kudo N, Khochbin S, Nishi K, Kitano K, Yanagida M, Yoshida M, et al. Molecular cloning and cell cycle-dependent expression of mammalian CRM1, a protein involved in nuclear export of proteins. J of Biol Chem 1997; 272: 29742–51. [DOI] [PubMed] [Google Scholar]
  • 13.Stade K, Ford CS, Guthrie C, Weis K. Exportin 1 (CRM1p) is an essential nuclear export factor. Cell 1997; 90: 1041–1050. [DOI] [PubMed] [Google Scholar]
  • 14.Gravina GL, Senapedis W, McCauley D, Baloglu E, Shacham S, Festuccia C. Nucleo-cytoplasmic transport as a therapeutic target of cancer. J Hematol Oncol. 2014; 7: 85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Vergote IB, Lund B, Peen U, Umajuridze Z, Mau-Sorensen M, Kranich A, et al. Phase 2 study of the exportin 1 inhibitor selinexor in patients with recurrent gynecological malignancies. Gynecol Oncol 2020; 156 (2): 208–314. [DOI] [PubMed] [Google Scholar]
  • 16.Bradbury A, O’Donnell R, Drew Y, Curti NJ, Saha SS. Characterization of ovarian cancer cell line NIH-OVCAR3 and implications of genomic, transcriptomic, proteomic and functional DNA damage response biomarkers for therapeutic targeting. Cancers 2020; 12(7): 1939. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Cerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, et al. The cBio cancer genomics portal: An open platform for exploring multidimensional cancer genomics data. Cancer Discov 2012; 2(5): 401–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Choi YE, Battelli C, Watson J, Liu J, Curtis J, Morse AN, et al. Sublethal concentrations of 17-AAG suppress homologous recombination DNA repair and enhance sensitivity to carboplatin and olaparib in HR proficient ovarian cancer cells. Oncotarget 2014; 5(9): 2678–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Ma Y, Chen P, Drisko JA, Khabele D, Godwin AK, Chen Q. Pharmacological ascorbate induces ‘BRCAness’ and enhances the effects of Poly(ADP-Ribose)polymerase inhibitors against BRCA1/2 wild-type ovarian cancer. Oncol Lett 2020; 19(4): 2629–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Markossian S, Grossman A, Brimacombe K, Arkin M, Auld D, Austin CP, et al. , eds. Assay guidance manual. Bethesda, Maryland: Eli Lilly & Company, National Center for Advancing Translational Sciences; 2004. Accessed March 5, 2021. https://www.ncbi.nlm.nih.gov/books/NBK53196/ [PubMed] [Google Scholar]
  • 21.Hafner M, Niepel M, Chung M, Sorger PK. Growth rate inhibition metrics correct for confounders in measuring sensitivity to cancer drugs. Nat Methods. 2016. June; 13(6):521–7. doi: 10.1038/nmeth.3853. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Meyer D, Dimitriadou E, Hornik K, Weingessel A, Leisch F (2020). E1071: Misc Functions of the Department of Statistics, Probability Theory Group (Formerly: E1071), TU Wien. R package version 1.7-4. Package ‘e1071’. https://CRAN.R-project.org/package=e1071. [Google Scholar]
  • 23.Qin L, Liu S, Liu H, Zhang Y. Support vector regression and least squares support vector regression for hermetic dose-response curves fitting. Chemosphere 2010; 78(3): 327–34. [DOI] [PubMed] [Google Scholar]
  • 24.Chou TC, Talalay P. Quantitative analysis of dose-effect relationships: the combined effects of multiple drugs or enzyme inhibitors. Adv Enzyme Regul 1984; 22:27–55. [DOI] [PubMed] [Google Scholar]
  • 25.Ianevski A, Giri AK, Aittokallio T. SynergyFinder 2.0: visual analytics of multi-drug combination synergies. Nucleic Acids Res 2020; 48(W1): W488–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Gharpure KM, Pradeep S, Sans M, Rupaimoole R, Ivan C, Wu SY, et al. FABP4 as a key determinant of metastatic potential of ovarian cancer. Nat Commun 2018; 9(1): 2923. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Tibes R, Qiu Y, Lu Y, Hennessy B, Andreeff M, Mills GB, et al. Reverse phase protein array: validation of a novel proteomic technology and utility for analysis of primary leukemia specimens and hematopoietic stem cells. Mol Cancer Ther 2006; 5(10): 2512–21. [DOI] [PubMed] [Google Scholar]
  • 28.Komurov K, Dursun S, Erdin S, Ram PT. NetWalker: a contextual network analysis tool for functional genomics. BMC Genomics 2012; 13: 282. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Sorensen CS, Hansen LT, Dziegielewski J, Syljuasen RG, Lundin C, Bartek J, et al. The cell-cycle checkpoint kinase Chk1 is required for mammalian homologous recombination repair. Nat Cell Biol 2005; 7: 195–201. [DOI] [PubMed] [Google Scholar]
  • 30.Kasahara K, Goto H, Enomoto M, Tomono Y, Kiyono T, Inagaki M. 14–3-3gamma mediates Cdc25A proteolysis to block premature mitotic entry after DNA damage. EMBO J 2010; 29(16): 2802–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Zona S, Bella L, Burton MJ, Nestal de Moraes G, Lam EWF. FOXM1: An emerging master regulator of DNA damage response and genotoxic agent resistance. Biochim Biophys Acta 2014; 1839(11): 1316–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Isono M, Niimi A, Oike T, Hagiwara Y, Sato H, Sekine R, et al. BRCA1 directs the repair pathway to homologous recombination by promoting 53BP1 dephosphorylation. Cell Rep. 2017; 18(2): 520–32. [DOI] [PubMed] [Google Scholar]
  • 33.Hurley RM, Hendrickson AEW, Visscher DW, Ansell P, Harrell MI, Wagner JM, et al. 53BP1 as a potential predictor of response tin PARP inhibitor-treated homologous recombination-deficient ovarian cancer. Gynecol Oncol. 2019; 153(1): 127–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Gravina GL, Senapedis W, McCauley D, Baloglu E, Shacham S, Festuccia C. Nucleo-cytoplasmic transport as a therapeutic target of cancer. J Hematol Oncol 2014; 7: 85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Gounder MM, Zer A, Tap WD, Salah S, Dickson MA, Gupta AA, et al. Phase IB study of selinexor , a first-in-class inhibitor of nuclear export, in patients with advanced refractory bone or soft tissue sarcoma. J Clin Oncol. 2016; 34(26):3166–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Subhash VV, Yeo MS, Wang L, Tan SH, Wong FY, Thuya WL, et al. Anti-tumor efficacy of selinexor (KPT-330) in gastric cancer is dependent on nuclear accumulation of p53 tumor suppressor. Sci Rep. 2018; 8: 12248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Fong PC, Boss DS, Yap TA, Tutt A, Wu P, Mergui-Roelvink M, et al. Inhibition of Poly(ADP-Ribose) Polymerase in tumors from BRCA mutation carriers. N Engl J Med 2009; 361:123–34. [DOI] [PubMed] [Google Scholar]
  • 38.Walsh C Targeted therapy for ovarian cancer: the rapidly evolving landscape of PARP inhibitor use. Minerva Ginecologica 2018; 70(2): 150–70. [DOI] [PubMed] [Google Scholar]
  • 39.Armstrong DK, Alvarez RD, Bakkum-Gamez JN, Barroilhet L, Behbakht K, Berchuck A, et al. NCCN Guidelines Insights: Ovarian Cancer, Version 1.2019. J Natl Compr Canc Netw 2019; 17(8): 896–909. [DOI] [PubMed] [Google Scholar]
  • 40.Moore KN, Secord AA, Geller MA, Miller DS, Cloven N, Fleming GF, et al. Niraparib monotherapy for late-line treatment of ovarian cancer (QUADRA): a multicenter, open-label, single-arm, phase 2 trial. Lancet Oncol. 2019; 20(5): 636–48. [DOI] [PubMed] [Google Scholar]
  • 41.Ihnen M, zu Eulenburg C, Kolarova T, Qi JW, Manivong K, Chalukya M, et al. Therapeutic potential of the Poly(ADP-ribose) polymerase inhibitor rucaparib for the treatment of sporadic human ovarian cancer. Mol Cancer Ther 2013; 12(6): 1002–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Fleury H, Carmona E, Morin VG, Meunier L, Masson JY, Tonin PN, et al. Cumulative defects in DNA repair pathways drive the PARP inhibitor response in high-grade serous epithelial ovarian cancer cell lines. Oncotarget 2017; 8: 40152–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Kashyap T, Argueta C, Unger T, Klebanov B, Debler S, Senapedis W, et al. Selinexor reduces the expression of DNA damage repair proteins and sensitizes cancer cells to DNA damaging agents. Oncotarget 2018; 9(56): 30773–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Razak ARA, Mau-Soerensen M, Gabrail NY, Gerecitano JF, Shields AF, Unger TJ, et al. First-in-class, first-in-human phase I study of selinexor, a selective inhibitor of nuclear export, in patients with advanced solid tumors. J Clin Oncol 2016; 34(34): 4142–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Moore K, Colombo N, Scambia G, Kim BG, Oaknin A, Friedlander M, et al. Maintenance olaparib in patients with newly diagnosed advanced ovarian cancer. N Engl J Med 2018; 379: 2495–505. [DOI] [PubMed] [Google Scholar]
  • 46.Penson RT, Valencia RV, Cibula D, Colombo N, Leath CA 3rd, Bidzinski M, et al. Olaparib versus nonplatinum chemotherapy in patients with platinum-sensitive relapsed ovarian cancer and a germline BRCA1/2 mutation (SOLO3): A randomized phase III trial. J Clin Oncol 2020; 38(11): 1164–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Hing ZA, Fung HYJ, Ranganathan P, Mitchell S, El-Gamal D, Woyach JA, et al. Next-generation XPO1 inhibitors shows improved efficacy and in vivo tolerability in hematological malignancies. Leukemia 2016; 30: 2364–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Vercruysse T, De Bie J, Neggers JE, Jacquemyn M, Vanstreels E, Schmid-Burgk JL, et al. The second-generation exportin-1 inhibitor KPT-8602 demonstrates potent activity against acute lymphoblastic leukemia. Clin Cancer Res 2017; 23(10): 2528–41. [DOI] [PubMed] [Google Scholar]
  • 49.Etchin J, Berezovskaya A, Conway AS, Galinsky IA, Stone RM, Baloglu E, et al. KPT-8602, a second-generation inhibitor of XPO1-mediated nuclear export, is well tolerated and highly active against AML blasts and leukemia-initiating cells. Leukemia 2017; 31: 143–50. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1745855-supplement-1.docx (2.9MB, docx)

Data Availability Statement

Key data and code produced in this manuscript is available at https://zenodo.org/badge/latestdoi/381717015.

RESOURCES