PHARMACEUTICAL COMBINATION OF WNT SIGNALING AND MACC1 INHIBITORS

Abstract

A pharmaceutical combination, includes an inhibitor of the Wnt/β-catenin signaling pathway and an inhibitor of MACC1. One combination includes an inhibitor of S100A4 as a Wnt-signaling inhibitor, preferably niclosamide, and a statin or MEK1 inhibitor as an inhibitor of MACC1. A pharmaceutical composition can include the combination. The combination or composition can be used in the treatment of a tumor disease, such as a solid tumor, and/or for the treatment and/or prophylaxis of tumor metastasis.

Claims

1. A pharmaceutical combination, comprising a. an inhibitor of the Wnt/β-catenin signaling pathway, and b. an inhibitor of MACC1.

2. The pharmaceutical combination according to claim 1, wherein the inhibitor of the Wnt/β-catenin signaling pathway is an inhibitor of S100A4.

3. The pharmaceutical combination according to claim 2, wherein the inhibitor of S100A4 is niclosamide or derivative thereof, sulindac, calcimycin, ICG001, FH535, LF3, or a phenothiazine.

4. The pharmaceutical combination according to claim 1, wherein the inhibitor of MACC1 is a statin.

5. The pharmaceutical combination according to claim 1, wherein the inhibitor of MACC1 is a MEK1 inhibitor.

6. The pharmaceutical combination according to claim 1, comprising a. niclosamide, and b. a statin and/or a MEK1 inhibitor.

7. The pharmaceutical combination according to claim 1, wherein the inhibitor of MACC1 is a statin selected from the group consisting of atorvastatin, lovastatin, fluvastatin, pitarvastatin, pravastatin, rosuvastatin and/or simvastatin.

8. The pharmaceutical combination according to claim 1, wherein the inhibitor of MACC1 is a statin selected from atorvastatin, lovastatin, fluvastatin, pitarvastatin, pravastatin, rosuvastatin and/or simvastatin, and the Wnt/β-catenin signaling pathway is niclosamide or derivative thereof.

9. The pharmaceutical combination according to claim 1, wherein the inhibitor of MACC1 is a MEK1 inhibitor selected from the group consisting of AZD6244 (selumetinib), GSK1120212 (trametinib) and cobimetinib.

10. The pharmaceutical combination according to claim 1, wherein the combination is selected from the group consisting of niclosamide and atorvastatin, niclosamide and lovastatin, niclosamide and fluvastatin, niclosamide and AZD6244 (selumetinib), and niclosamide and GSK1120212 (trametinib).

11. The pharmaceutical combination according to claim 1, wherein (a.) the inhibitor of the Wnt/β-catenin signaling pathway and (b.) the inhibitor of MACC1 have relative amounts of 10000:1 to 1:10000 by weight.

12. The pharmaceutical combination according to claim 1, wherein the inhibitor of the Wnt/β-catenin signaling pathway is in a pharmaceutical composition in admixture with a pharmaceutically acceptable carrier, and the inhibitor of MACC1 is in a separate pharmaceutical composition in admixture with a pharmaceutically acceptable carrier, or the inhibitor of the Wnt/β-catenin signaling pathway and the inhibitor of MACC1 are present in a kit, in spatial proximity but in separate containers and/or compositions, or the inhibitor of the Wnt/β-catenin signaling pathway and the inhibitor of MACC1 are combined in a single pharmaceutical composition in admixture with a pharmaceutically acceptable carrier.

13. A method of treating a tumor disease in a subject in need thereof, comprising administering the pharmaceutical combination according to claim 1 to the subject.

14. The method according to claim 13, wherein the tumor disease is a solid tumor, or selected from the group consisting of gastrointestinal, colorectal, gastric, esophageal, pancreatic, hepatocellular, biliary, lung, nasopharyngeal, renal, bladder, ovarian, brain, bone, head and neck, prostate, melanoma and breast cancer.

15. The method according to claim 13 for treating and/or reducing the risk of tumor metastasis.

16. The method according to claim 13, wherein the tumor cells to be treated exhibit increased expression and/or activity of MACC1 and S100A4 compared to a health control.

17. The method according to claim 13, wherein the subject of treatment exhibits stage 0, I, II, III or IV colorectal cancer, and/or wherein the subject of treatment will undergo and/or has undergone surgery to remove a solid tumor.

18-19. (canceled)

20. The pharmaceutical combination according to claim 7, wherein the statin is selected from the group consisting of atorvastatin, lovastatin and fluvastatin.

21. The pharmaceutical combination according to claim 11, wherein (a.) the inhibitor of the Wnt/β-catenin signaling pathway is niclosamide and (b.) the inhibitor of MACC1 is a statin, and (a.) and (b.) have relative amounts of 1000:1 to 1:1.

22. The pharmaceutical combination according to claim 11, wherein (a.) the inhibitor of the Wnt/β-catenin signaling pathway is niclosamide and (b.) the inhibitor of MACC1 is a MEK1 inhibitor, and (a.) and (b.) have relative amounts of 5000:1 to 1:1.

23. The pharmaceutical combination according to claim 11, wherein (a.) the inhibitor of the Wnt/β-catenin signaling pathway is niclosamide and (b.) the MEK1 inhibitor is GSK1120212 (trametinib), and (a.) and (b.) have relative amounts of 2000:1 to 500:1.

24. The pharmaceutical combination according to claim 11, wherein (a.) the inhibitor of the Wnt/β-catenin signaling pathway is niclosamide and (b.) the MEK1 inhibitor is AZD6244 (selumetinib) or cobimetinib, and (a.) and (b.) have relative amounts of 1000:1 to 1:1.

Description

FIGURES

[0246] The invention is further described by the figures. These are not intended to limit the scope of the invention.

[0247] FIG. 1: Statins inhibit MACC1 expression

[0248] HCT116 cells were treated with increasing amounts (1-10 μM) of different, clinically relevant statins dissolved in DMSO. Following 24 h of treatment cells were harvested for RNA and protein isolation. MACC1 mRNA expression was analyzed by qRT-PCR and is expressed relative to DMSO treated control samples. Protein amounts were analyzed by western blot. All tested statins were able to significantly reduce MACC1 mRNA expression and MACC1 protein in a dose-dependent fashion.

[0249] FIG. 2: Migration induced by supernatants from MACC1 overexpressing cells.

[0250] Cell culture supernatants from SW480/vector (low MACC1 expression), SW480/MACC1 (different clones, ectopically MACC1 overexpressing) and SW620 (endogenously high MACC1 expression) were collected to treat different tumor cell lines in vitro. SW480 cells (low MACC1 expression) show increased migration in the boyden chamber assay when treated with cell culture supernatants of MACC1 positive cells compared to controls treated with supernatants of MACC1 low SW480 cells. SW620 cells (high endogenous MACC1 expression) show no increase in migration under these conditions. When MACC1 is depleted by shRNA in these cells, motility is decreased and can be rescued by cell culture supernatants of MACC1 positive cells. These results were confirmed in another human tumor cell line of another entity (MiaPaCa, pancreas carcinoma). Therefore, supernatants collected from MACC1 overexpressing cells induce migration on different wild type cells without ectopic MACC1 overexpression.

[0251] FIG. 3: SILAC-Mass-spec analysis

[0252] Cell cultures of SW480/vector (low MACC1 expression) and SW480/MACC1 (ectopically high MACC1 expression) were treated with light and heavy media to label newly synthesized proteins secreted in the cell culture supernatant. After protein isolation and LC-MS/MS analysis the MACC1 induced secretome was identified. The MACC1 specific supernatants harbor newly secreted S100A4. SW480/vector vs SW480/MACC1: MACC1 overexpression induces S100 proteins, in particular S100A4 secretion.

[0253] FIG. 4: Biomarker combination of MACC1 AND S100A4 identifies best colorectal cancer patients at high risk

[0254] Tumor (A) and plasma (B) samples were collected from colorectal cancer patients and analyzed for MACC1 and S100A4 gene expression by qRT-PCR. Patients with low gene expression of both biomarkers show the best metastasis-free and overall survival. Patients with high gene expression of one biomarker show reduced metastasis-free and overall survival. The worst metastasis-free and overall survival was evident for patients with high gene expression of both biomarkers. Therefore, MACC1 and S100A4 can identify patients with increased risk for metastasis formation compared to patients with low expression. This prognosis is further improved if both biomarkers are highly expressed: (A) in primary CRC tumors; (B) in CRC patient plasma.

[0255] FIG. 5: MACC1 enhances Wnt signaling.

[0256] MACC1 overexpression in SW480 (A) and HCT116 (B) cells increases, while MACC1 depletion in SW620 (C) cells decreases, TCF signaling activity, indicative for positive regulation of Wnt signaling by MACC1. Immunoblot of MACC1 overexpression in SW480 and knockdown in SW620 (D). β-catenin knockdown decreases MACC1-induced TCF promoter activity (E) and MACC1-induced migration (F), indicating that MACC1 induces Wnt signaling through β-catenin in colorectal cancer cells.

[0257] FIG. 6: MACC1 induces S100A4 expression through Wnt signaling.

[0258] Ectopic overexpression of MACC1 in HCT116 cells (A) increases TCF promoter activity (B), subsequently upregulating S100A4-promoter activity (C) and S100A4 gene expression at mRNA (D) as well as protein (E) level. Knockout of MACC1 in SW620 cells (F) decreases TCF promoter activity (G), subsequently downregulating S100A4-promoter activity (H) and S100A4 gene expression at mRNA (I) as well as protein (J) level. In a colorectal cancer cohort of 54 patients MACC1 and S100A4 mRNA expression level were positively correlated (K).

[0259] FIG. 7: The effect of statins and niclosamide applied as single agents on cellular motility and proliferation in vitro.

[0260] HCT116 cells were treated with three concentrations of atorvastatin, lovastatin, fluvastatin and niclosamide for 48 h. Wound closure was monitored every second hour using the IncuCyte real-time system. Wound closure was measured relative to the initial wound and expressed as fold change compared to DMSO treated samples. A pre-established image collection of HCT116 was used to teach the software detection of cells and wound. The experiments show, that there is a dose dependent reduction of wound closure ability for all drugs tested. Untreated and DMSO treated cells served as controls. Data represents mean ±SEM of at least 3 independent experiments in triplicate.

[0261] FIG. 8: The effect of statins and niclosamide applied as combined agents on cellular motility and proliferation in vitro.

[0262] HCT116 cells were treated with three concentrations of atorvastatin, lovastatin, fluvastatin in combination with three concentrations of niclosamide for 48 h. As a representative example the combination of a 50% reduced dose is shown. For each drug combination the synergy matrix for all concentrations is depicted. Positive values demonstrate the presence of synergy between the two agents. A negative value would indicate antagonism between the two agents. A value of 0 indicates an additive effect of two agents.

[0263] Wound closure was monitored every second hour using the IncuCyte real-time system. Wound closure was measured relative to the initial wound and expressed as fold change compared to DMSO treated samples. A pre-established image collection of HCT116 was used to teach the software detection of cells and wound. The experiments show that there is a synergistic reduction of wound closure when a statin is combined with niclosamide. Untreated and DMSO treated cells served as controls. Data represents mean ±SEM of at least 3 independent experiments in triplicate.

[0264] FIG. 9: The effect of MEK1 inhibitors AZD6244 and GSK1120212 and niclosamide applied as combined agents on cellular motility and proliferation in vitro.

[0265] HCT116 cells were treated with three concentrations of GSK1120212 in combination with three concentrations of niclosamide. For each drug combination the synergy matrix for all concentrations is depicted. Positive values demonstrate the presence of synergy between the two agents. A negative value would indicate antagonism between the two agents. A value of 0 indicates an additive effect of two agents.

[0266] A synergistic example of the combination of 0.1 μM GSK1120212 and 0.25 μM niclosamide is shown. For treatments the synergy matrix for all concentrations is depicted. Wound closure was monitored every second hour using the IncuCyte real-time system. Wound closure was measured relative to the initial wound and expressed as fold change compared to DMSO treated samples. A pre-established image collection of HCT116 was used to teach the software detection of cells and wound. The experiments show that there is a synergistic reduction of wound closure when the MEK1 inhibitor GSK1120212 (trametinib) is combined with niclosamide. Untreated and DMSO treated cells served as controls.

[0267] FIG. 10: The effect of atorvastatin, fluvastatin and niclosamide in an in vivo mouse model at the equivalent of maximum human dose.

[0268] Severe combined immunodeficiency (SCID)-beige mice (n=10 per group) were intrasplenically transplanted with HCT116-CMVp-Luc cells and treated daily with solvent, 13 mg/kg atorvastatin or fluvastatin (p.o.), 328 mg/kg niclosamide (p.o.) or the combination of niclosamide with one of the statins, at the indicated doses. These doses correspond to the maximum human doses established for the statins and niclosamide. Metastasis to the liver was analyzed with bioluminescence imaging after an intraperitoneal application of 150 mg/kg D-Luciferin at the experimental endpoint. Following whole animal imaging livers were removed and analyzed as isolated organs. Metastasis formation was significantly reduced when atorvastatin, fluvastatin and niclosamide were applied as single drugs. This metastasis inhibition was also evident in the combinatorial treatment.

[0269] FIG. 11: The effect of atorvastatin, fluvastatin and niclosamide in an in vivo mouse model at reduced dose.

[0270] Severe combined immunodeficiency (SCID)-beige mice (n=10 per group) were intrasplenically transplanted with HCT116-CMVp-Luc cells and treated daily with solvent, 1.5 mg/kg atorvastatin or fluvastatin (p.o.), 164 mg/kg niclosamide (p.o.) or the combination of niclosamide with one of the statins, at the indicated doses. These doses therefore correspond to 12.5% of the maximum human dose for the statins and 50% of the maximum human dose for niclosamide. Metastasis to the liver was analyzed with bioluminescence imaging after an intraperitoneal application of 150 mg/kg D-Luciferin at the experimental endpoint. Following whole animal imaging livers were removed and analyzed as isolated organs. No reduction in metastasis formation was observed when atorvastatin, fluvastatin and niclosamide were applied as single drugs. However, a reduction in metastasis was observed in the combinatorial treatments, further supporting that the combination of active agents leads to a synergistic effect.

EXAMPLES

[0271] The invention is further described by the following examples. These are not intended to limit the scope of the invention.

Materials and Methods

Cell Lines and Drug Treatment

[0272] HCT116, SW620 and SW480 colorectal carcinoma cells were purchased from ATCC. HCT116 cells were stably transduced with GFP/ MACC1-GFP plasmids to produce MACC1 overexpressing cell lines and SW620 cells were stably transduced by sh-control/sh-MACC1 plasmids to produce MACC1 knockdown cell lines. SW480 stable MACC1 overexpressing cell lines were created using pCDNA3.1 MACC1 plasmids and the control cells were produced using pCDNA3.1 empty plasmids followed clone selection using Gentamycin (Gibco). Knock-out of MACC1 in SW620 cells was achieved by CRISPR-Cas9 based gene editing followed by Puromycin (Gibco) based selection and subsequently the clones were picked using single cell sorting. Cells were cultured in RPMI supplemented with 10% FCS in 5% CO.sub.2 humidified atmosphere and sub passaged 2-3 times a week to maintain cultures subconfluent. Drugs were solubilized in DMSO, stock concentrations were 2 mM for niclosamide (Sigma-Aldrich, St Louis, Mo., USA), 2 mM for GSK1120212 (trametinib, Selleck Chemicals, Munich, Germany) and selumetinib (AZD6244, Selleck Chemicals, Munich, Germany) and 10 mM for statins (atorvastatin, lovastatin, fluvastatin, TCI Deutschland GmbH, Eschborn, Germany). All drugs were pre-diluted to 1000× final concentration in DMSO. Control cells were treated with equal amounts of DMSO.

Reporter Assays

[0273] Luciferase reporter assay was used to assess Wnt signaling and S100A4-promoter activity. Wnt signaling activity was measured using TOP-flash reporter construct containing 6×TCF promoter consensus sequence cloned ahead of firefly luciferase in a pGL4.23 reporter plasmid (Promega) (Kindly provided by Dr. Giridhar Mudduluru). S100A4 promoter activity was measured using S100A4 promoter luciferase system harboring core S100A4 promoter sequence cloned ahead of firefly luciferase reporter in a pGL1.4 reporter plasmid (Invitrogen). Briefly, 2×10.sup.5 cells were seeded into 24-well plates. 24 h post incubation cells were transfected with 500 ng of the reporter luciferase construct along with 25 ng of renilla luciferase plasmid to normalize for transfection efficiency. Transfection was carried out using Lipofectamine 2000 (Life Technologies) transfection reagent according to the manufacturers instructions. After 48 h of transfection luciferase activity was measured using the Dual Luciferase Assay Kit (Promega). The firefly luciferase activity values were normalized using renilla luciferase values and data were represented as Mean±S.E.M of three independent experiments.

RNA Interference

[0274] Silencer Select pre-designed siRNA against β-catenin (#16704) and negative control (#4390844) were purchased from Ambion. Transfection of the cells was carried out using Lipofectamine RNAiMAX (Life Technologies) transfection reagent according to the manufacturers instructions. After 48h, cells were re-seeded for further treatments and/or analysis.

Boyden Chamber Migration Assay

[0275] Cells were starved in FCS-free medium overnight. Transwell Boyden chamber inserts (Corning) were equilibrated in medium containing 10% FCS for 4 h at 37° C. prior to cell seeding. 1×10.sup.5 serum-starved cells were seeded into each insert in medium containing 2% FCS. Medium containing 10% FCS was added to the bottom well as a chemoattractant and chambers were incubated at 37° C. in humidified incubator at 5% CO.sub.2. After 16 h cells on the bottom end of the transwell membrane were harvested by trypsinization. The migrated cells were quantified using Cell-Titer-Glo reagent (Promega) according to manufacturer's instructions.

RNA Isolation and qRT-PCR

[0276] RNA was isolated using the Gene Matrix Universal RNA Purification Kit (EURx, Poland) according to the manufacturer's instructions. RNA was quantified and 50 ng of RNA was reverse transcribed using a reaction mix containing MuLV reverse transcriptase, 10 mM MgCl2, 1× PCR-Buffer, 250 μM pooled dNTPs, 1U RNAse inhibitor and random hexamers (all from Applied Biosystems). The reaction was performed at 42° C. for 15 min, 99° C. for 5 min and subsequent cooling at 4° C. for 5 min. The cDNA was amplified using SYBR Green chemistry (Promega) using the LightCycler 480 II system (Roche Diagnostics) at the following PCR conditions: 95° C. for 2 min followed by 45 cycles of 95° C. for 7 s, 60° C. for 10 s and 72° C. for 20 s. The primers used for quantification are as follows: MACC1 Fwd 5′-TTCTTTTGATTCCTCCGGTGA-3′ and Rev 5′-ACTCTGATGGGCATGTGCTG-3′; S100A4 Fwd 5′-TGTGATGGTGTCCACCTTCC-3′ and Rev 5′- CCTGTTGCTGTCCAAGTTGC-3′. Data analysis was performed with LightCycler 480 Software release 1.5.0 SP3 (Roche Diagnostics). Mean values was calculated from RT-qPCR duplicates. Each mean value of the expressed gene was normalized to the respective RP-II cDNA. Data is represented as mean ±S.E.M of three independent experiments.

Protein Extraction and Immunoblotting

[0277] Whole cell protein extraction was performed on 5×10.sup.5 HCT116 and SW620 cells. Cells were lysed using RIPA buffer (50 mM Tris-HCT pH 7.5, 150 mM NaCl, 1% Nonidet P-40) supplemented with cOmplete Protease inhibitor (Roche) for 30 min at 4° C. The lysate was clarified and the protein content was measured using the BCA Kit (Pierce). 20 μg protein lysate were denatured with LDS-containing NuPage Loading buffer (invitrogen) and DTT for 10 min.

[0278] Proteins were then resolved in 10% SDS-PAGE followed by protein transfer onto PVDF membranes (BioRAD). The membrane was blocked with 5% w/v skim milk powder and 1% w/v BSA in TBST (10 mM Tris-HCL pH 8, 150 mM NaCl and 0.1% Tween 20) for 60 min. The membrane was incubated overnight with the respective primary antibodies (anti-MACC1, #HPA020081, Sigma Aldrich; S100A4, #5114, Dako and β-Actin, #31430, Invitrogen; Lamin B1, #12586s, Cell Signalling). The membranes were washed and further incubated with corresponding horseradish peroxidase-tagged secondary antibodies (anti-mouse IgG HRP-conjugated, #31430, Invitrogen and anti-rabbit IgG HRP-conjugated, #W4018, Promega) for 60 min. The membranes were further washed and developed using chemiluminescent reagent WesternBright ECL kit (Advansta) and subsequent exposure to Fuji medical X-ray film SuperRX (Fujifilm).

Wound Healing Assay and Calculation of Synergy

[0279] For wound healing assays the HCT116 human colon cancer cell line was used. HCT116 cells are endogenously positive for MACC1 and S100A4 expression at the mRNA and protein level. In addition, they show decreased motility when either marker is inhibited.

[0280] For wound healing (scratch) experiments cells were passaged to a density of 6×10.sup.5 per ml and cultured for 48 h. Cells were then harvested and counted. For wound healing assays 1.2×10.sup.5 cells were seeded in 100 μl RPMI supplemented with 10% FCS in 96 well Image Lock Plates (Essen Bioscience, Hertfordshire, UK). Cells were allowed to adhere for 6 hours. Wounds in the monolayer were applied using the IncuCyte WoundMaker (Essen Bioscience). After applying the wounds 100 μl of 2-fold concentrated drug solutions were added. DMSO treated and untreated samples served as controls. For each drug three different (niclosamide 1 μM, 0.5 μM and 0.25 μM, each statin 5 μM, 2.5 μM and 1.25 μM, MEK1 inhibitors 1 μM, 0.1 μM and 0.01 μM) concentrations and combinations thereof were applied, each in triplicate. Of all statins, we tested lovastatin, fluvastatin and atorvastatin. Wound closure was monitored every second hour in the IncuCyte system (Essen Bioscience). A pre-established image collection of HCT116 was used to teach the software detection of cells and wound. Wound confluency was expressed relative to DMSO treated controls.

[0281] Synergy was analyzed using combenefit 2.021. DMSO treated samples were set to 100%. To calculate synergy, we used the Loewe isobole equation model. For the calculations dose-response curves for three single concentrations and the nine combinations thereof were used.

[0282] The Loewe model was applied since there is a certain degree of target interaction already shown: first, active Wnt-signaling is important for MACC1 driving tumor progression and invasiveness (Lemos C, Clin Cancer Res, 2016) and second, S100A4 is a Wnt-signaling target gene and is found in the MACC1 induced secretome. The three single concentrations and nine combinations thereof were expressed relative to the control.

Animals and Drug Treatment in vivo

[0283] All experiments were performed in accordance with the United Kingdom Coordinated Committee on Cancer Research (UKCCCR) guidelines and approved by the responsible local authorities (State Office of Health and Social Affairs, Berlin, Germany). For in vivo drug testing a xenograft mouse model was used as described earlier (Stein 2009 Nat Med, Sack 2011 J Natl Cancer Inst,

[0284] Juneja & Kobelt 2017 Plos Biol). In brief, HCT116-CMVp-LUC cells (5×10.sup.5 cells per mouse, resuspended in 50 μL PBS) were intrasplenically transplanted into 6-week-old female SCID beige (SCID bg/bg) mice. The animals were assigned randomly into treatment groups.

[0285] Drugs were administered as a suspension using a gavage tube. Both niclosamide (either 328 mg/kg or 164 mg/kg) and a statin (atorvastatin or fluvastatin, at 13 mg/kg or 1.5 mg/kg) were administered daily, orally. Control mice were treated with the appropriate volume of solvent solution. The in vivo experimentation and luminescence imaging were conducted as described below.

[0286] The in vivo experiments were terminated when the animals in the control group showed signs of increased suffering due to tumor/metastasis burden and liver damage like swollen abdomen (ascites formation), reduced activity, and reduced food intake (ethical/humane endpoint).

[0287] For luminescence imaging, mice were anesthetized with 5% Isoflurane and received intraperitoneally 150 mg/kg D-luciferin (Biosynth, Staad, Switzerland) dissolved in sterile PBS. Anesthesia was maintained with 2% isoflurane. Imaging was performed with the NightOWL LB 981 system (Berthold Technologies, Bad Wildbad, Germany). ImageJ version 1.48v (NIH, Bethesda, Md.) was used for color coding of signal intensity (presenting a 256 grayscale) and overlay pictures.

Results

[0288] MACC1 was reported to enhance WNT signaling and its target gene expression leading to increased cancer cell migration and invasion in vitro and tumor formation as well as metastasis in vivo (Stein et al. 2009 Nat Med. Jan;15(1):59-67; Zhen et al. 2014, Oncotarget. Jun 15;5(11):3756-69; Lemos et al. 2016, Clin Cancer Res. Jun 1;22(11):2812-24). Ectopic overexpression of MACC1 increased β-catenin nuclear localization, thereby increasing WNT target genes. Conversely, MACC1 knock-down decreased β-catenin nuclear localization and reduced their expression. To inhibit MACC1 gene expression, different clinically relevant statins were tested for their ability to reduce MACC1 mRNA and protein expression. All tested statins reduced MACC1 gene expression significantly (FIG. 1).

[0289] Furthermore, ectopic overexpression of MACC1 induces in vitro the release of factors into the medium, which mediate increased cell motility and invasion in cells of different origins (FIG. 2).

[0290] These MACC1 specific supernatants harbor newly secreted S100A4, as determined by SILAC analyses (FIG. 3).

[0291] This was further substantiated by MACC1 and S100A4 expression correlation analysis on a publicly available microarray dataset of CRC patients from NCBI GEO database (Tsuji et al. 2012). MACC1 and S100A4 gene expression were positively and significantly correlated regarding metastasis free and overall survival in the CRC patient cohort (p=0.4115, p=0.002, n=54) (FIG. 4).

[0292] We identified that MACC1 overexpression increased WNT signaling (TCF promoter) activity and subsequently inducing S100A4 promoter activity in colorectal cancer cells (FIG. 5). This increase in S100A4 promoter activity was further evident in the S100A4 gene and protein expression in these cells.

[0293] We also identified that the increase in S100A4 expression in MACC1 ectopically overexpressing cells results from an increased WNT signaling activity in these cells. Conversely, MACC1 knock-down decreased S100A4 promoter activity and gene expression (FIG. 6).

[0294] Both MACC1 and S100A4 have been shown to induce migration and metastasis, independently. Our recent work has unraveled a novel mechanism whereby MACC1 induces migration and metastasis via a Wnt/S100A4 axis. Therefore, direct inhibition of Wnt/β-catenin signaling along with MACC1 inhibition represents a viable therapeutic strategy.

[0295] In line with this, we tested three different statins in combination with niclosamide. This targets two distinct pathways. All tested statins were able to reduce MACC1 mRNA expression and MACC1 protein (FIG. 1). The expression of MACC1 is inhibited by the statins, thereby MACC1 induced effects are reduced, and niclosamide interferes in Wnt/β-catenin signaling leading to reduced S100A4 expression. First, we confirmed that the all the compounds are able to reduce cellular motility when applied as monotherapy in the wound healing assay. We found a dose dependent reduction of wound closure over time. Cells either non- or DMSO treated closed the wound in the monolayer within 48 h. Compared to this, wound closure of cells treated with monotherapy was delayed for all drugs (FIG. 7).

[0296] Next, we tested, if the statins can act synergistically when applied together with niclosamide to reduce cellular motility (FIGS. 8A-C). Here we aim to reduce the drug amount needed to inhibit motility. Compared to the single treatments, inhibition of wound closure was increased when two drugs, niclosamide with one of the statin, were combined. Exemplified for a reduced concentration (2.5 μM for a statin and 0.5 μM for niclosamide) compared to our initial description for HCT116 (5 μM for a statin and 1 μM for niclosamide) we found a synergistically reduced ability to close the wound applied to the HCT116 monolayer.

[0297] We have shown that MACC1 is phosphorylated by MEK1 leading to induction of the MACC1 mediated effects. Therefore, we tested, if the MEK1 inhibitors GSK1120212 (trametinib, approved for melanoma treatment) and AZD6244 (selumetinib) act synergistically on cellular motility when combined with niclosamide. Similarly, we applied three different concentrations of trametinib, selumetinib and niclosamide to HCT116 cells and monitored wound closure over time. Here a synergistic activity is detectable, but at a lower level compared to the combination of statins with niclosamide (FIG. 9A-B).

[0298] To translate our results to a clinical application, drug combinations composed of niclosamide with a statin were tested in vivo. The combinations niclosamide and fluvastatin, and niclosamide and atorvastatin, were assessed in an in vivo model (mice intrasplenically transplanted with HCT116-CMVp-Luc cells) using oral administration, as described in the methods section above and in FIG. 10-11.

[0299] In a first set of experiments, the niclosamide, atorvastatin and fluvastatin were administered both individually and in combination. The statins were administered at 13 mg/kg, and niclosamide at 328 mg/kg, which are doses equivalent to the maximum human dose established for these agents (80 mg daily for the statins, 2 g daily for niclosamide).

[0300] As can be seen from FIG. 10, metastasis formation (as determined by luminescence measurement) was reduced when atorvastatin, fluvastatin and niclosamide were applied as single drugs, at these high doses. This metastasis inhibition was also evident in the combinatorial treatment.

[0301] This experiment was repeated using reduced doses of niclosamide, atorvastatin and fluvastatin. The statins were administered at 1.5 mg/kg, and niclosamide at 164 mg/kg. These doses correspond to 12.5% of the maximum human dose for the statins and 50% of the maximum human dose for niclosamide (9.2 mg daily for the statins, 1 g daily for niclosamide).

[0302] As can be seen from FIG. 11, no reduction in metastasis formation as determined by luminescence measurement was observed when atorvastatin, fluvastatin and niclosamide were applied as single drugs. However, a reduction in metastasis was observed in the combinatorial treatments, further supporting that the combination of active agents leads to an unexpected synergistic effect in reducing metastasis.

[0303] Experiments using a combination of lovastatin and niclosamide have also been conducted and show similar results.

[0304] In summary, the combinations of a statin with niclosamide, and the MEK1 inhibitor GSK1120212 (trametinib) combined with niclosamide, are able to synergistically reduce cellular motility in the in vitro wound healing assay. Additionally, the combination of lovastatin, fluvastatin or atorvastatin, combined with niclosamide, is superior and synergistic to monotherapy in reducing metastasis formation in the xenograft mouse model.

Conclusion of the Examples

[0305] The inventors show that the close association of S100A4 and MACC1 and their overexpression is associated with poor prognosis of affected CRC patients. The inventors also show the link of these underexplored biomarkers to Wnt-signaling. These findings prompted the inventors to exploit these targets by combined intervention of tumor progression and metastasis formation using repositioned small molecule inhibitors such as niclosamide, statins and MEK1 tyrosine kinase inhibitors. This intervention strategy using drug combinations showed an unexpected synergistic efficacy for metastasis inhibition. This is a strong indication, that targeting of these key pathways and molecules causally involved in cancer metastasis can lead to efficient intervention particularly for patients, who are at high risk for metastasis due to S100A4 and MACC1 overexpression. Thus, treatment of those patients stratified for S100A4 and MACC1 overexpression will be beneficial for these patients and holds enormous promise for clinical use.