TREATMENT OF CANCER USING A CEA CD3 BISPECIFIC ANTIBODY AND A WNT SIGNALING INHIBITOR

Abstract

The present invention relates to the treatment of cancer, in particular to the treatment of cancer using a CEA CD3 bispecific antibody and a Wnt signaling inhibitor.

Claims

1. A method for treating cancer in an individual comprising administering to the individual a CEA CD3 bispecific antibody and a Wnt signaling inhibitor.

2. The method of claim 1, wherein the CEA CD3 bispecific antibody comprises: (i) a first antigen binding moiety that specifically binds to CD3 and comprises a heavy chain variable region comprising the heavy chain CDR (HCDR) 1 of SEQ ID NO: 1, the HCDR2 of SEQ ID NO: 2, and the HCDR3 of SEQ ID NO: 3; and a light chain variable region comprising the light chain CDR (LCDR) 1 of SEQ ID NO: 4, the LCDR2 of SEQ ID NO: 5 and the LCDR3 of SEQ ID NO: 6; and (ii) a second antigen binding moiety that specifically binds to CEA and comprises (i) a heavy chain variable region comprising the heavy chain CDR (HCDR) 1 of SEQ ID NO: 9, the HCDR2 of SEQ ID NO: 10, and the HCDR3 of SEQ ID NO: 11; and a light chain variable region comprising the light chain CDR (LCDR) 1 of SEQ ID NO: 12, the LCDR2 of SEQ ID NO: 13 and the LCDR3 of SEQ ID NO: 14; or (ii) a heavy chain variable region comprising the heavy chain CDR (HCDR) 1 of SEQ ID NO: 17, the HCDR2 of SEQ ID NO: 18, and the HCDR3 of SEQ ID NO: 19; and a light chain variable region comprising the light chain CDR (LCDR) 1 of SEQ ID NO: 20, the LCDR2 of SEQ ID NO: 21 and the LCDR3 of SEQ ID NO: 22.

3. The method of claim 2, wherein the CEA CD3 bispecific antibody further comprises a third antigen binding moiety that specifically binds to CEA.

4. The method of claim 2, wherein the CEA CD3 bispecific antibody comprises an Fc domain composed of a first and a second subunit.

5. The method of claim 4, wherein the Fc domain of the CEA CD3 bispecific antibody comprises a modification promoting the association of the first and the second subunit of the Fc domain.

6. The method of claim 4, wherein the Fc domain comprises one or more amino acid substitution that reduces binding to an Fc receptor or effector function.

7. The method of claim 3, wherein the CEA CD3 bispecific antibody comprises (i) a first antigen binding moiety that specifically binds to CD3, comprising a heavy chain variable region comprising the heavy chain CDR (HCDR) 1 of SEQ ID NO: 1, the HCDR2 of SEQ ID NO: 2, and the HCDR3 of SEQ ID NO: 3; and a light chain variable region comprising the light chain CDR (LCDR) 1 of SEQ ID NO: 4, the LCDR2 of SEQ ID NO: 5 and the LCDR3 of SEQ ID NO: 6, wherein the first antigen binding moiety is a crossover Fab molecule wherein either the variable or the constant regions of the Fab light chain and the Fab heavy chain are exchanged; (ii) a second and a third antigen binding moiety that specifically bind to CEA, comprising (i) a heavy chain variable region comprising the heavy chain CDR (HCDR) 1 of SEQ ID NO: 9, the HCDR2 of SEQ ID NO: 10, and the HCDR3 of SEQ ID NO: 11; and a light chain variable region comprising the light chain CDR (LCDR) 1 of SEQ ID NO: 12, the LCDR2 of SEQ ID NO: 13 and the LCDR3 of SEQ ID NO: 14; or (ii) a heavy chain variable region comprising the heavy chain CDR (HCDR) 1 of SEQ ID NO: 17, the HCDR2 of SEQ ID NO: 18, and the HCDR3 of SEQ ID NO: 19; and a light chain variable region comprising the light chain CDR (LCDR) 1 of SEQ ID NO: 20, the LCDR2 of SEQ ID NO: 21 and the LCDR3 of SEQ ID NO: 22, wherein the second and third antigen binding moiety are each a Fab molecule; (iii) an Fc domain composed of a first and a second subunit, wherein the second antigen binding moiety is fused at the C-terminus of the Fab heavy chain to the N-terminus of the Fab heavy chain of the first antigen binding moiety, and the first antigen binding moiety is fused at the C-terminus of the Fab heavy chain to the N-terminus of the first subunit of the Fc domain, and wherein the third antigen binding moiety is fused at the C-terminus of the Fab heavy chain to the N-terminus of the second subunit of the Fc domain.

8. The method of claim 7, wherein the second and third antigen binding moiety are each a conventional Fab molecule.

9. The method of claim 2, wherein the first antigen binding moiety of the CEA CD3 bispecific antibody comprises a heavy chain variable region sequence that is at least about 95%, 96%, 97%, 98%, 99% or 100% identical to the amino acid sequence of SEQ ID NO: 7 and a light chain variable region sequence that is at least about 95%, 96%, 97%, 98%, 99% or 100% identical to the amino acid sequence of SEQ ID NO: 8, and the second antigen binding moiety of the CEA CD3 bispecific antibody comprises (i) a heavy chain variable region sequence that is at least about 95%, 96%, 97%, 98%, 99% or 100% identical to the amino acid sequence of SEQ ID NO: 15 and a light chain variable region sequence that is at least about 95%, 96%, 97%, 98%, 99% or 100% identical to the amino acid sequence of SEQ ID NO: 16, or (ii) a heavy chain variable region sequence that is at least about 95%, 96%, 97%, 98%, 99% or 100% identical to the amino acid sequence of SEQ ID NO: 23 and a light chain variable region sequence that is at least about 95%, 96%, 97%, 98%, 99% or 100% identical to the amino acid sequence of SEQ ID NO: 24.

10. The method of claim 1, wherein the CEA CD3 bispecific antibody is cibisatamab.

11. The method of claim 1, wherein the Wnt signaling inhibitor is a Wnt/β-catenin signaling inhibitor.

12. The method of claim 1, wherein the Wnt signaling inhibitor targets a component of the Wnt signaling pathway selected from the group consisting of Frizzled (Fz), Disheveled (DVL), Porcupine, tankyrase, and glycogen synthase kinase 3 (3 (GSK-3(3).

13. The method of claim 12, wherein the Wnt signaling inhibitor is a tankyrase inhibitor.

14. The method of claim 13, wherein the Wnt signaling inhibitor is Compound 21: ##STR00004## wherein R.sup.1 is Me and R.sup.2 is CH2-N-(4-NMe.sub.2)-piperidine.

15. The method of claim 1, wherein the treatment further comprises administration of a PD-L1 binding antagonist.

16. The method of claim 15, wherein the PD-L1 antagonist is atezolizumab.

17. The method of claim 1, wherein the cancer is a CEA-positive cancer.

18. The method of claim 17, wherein the cancer is selected from the group consisting of colorectal cancer, lung cancer, pancreatic cancer, breast cancer, and gastric cancer.

19. The method of claim 18, wherein the cancer is colorectal cancer.

20. A kit comprising a first medicament comprising a CEA CD3 bispecific antibody, a second medicament comprising a Wnt signaling inhibitor, and a package insert comprising instructions for administration of the first medicament in combination with the second medicament for treating cancer in an individual.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0164] FIG. 1. Growth curves for eight patient derived colorectal cancer organoid (PDO) lines with various cell surface CEA expression levels, that were treated with cibisatamab or an untargeted control antibody during 10 days of co-culture. Each PDO was cultured with T cells from three different allogeneic donors at an effector-to-target (E:T) ratio of 2:1 and means are shown.

[0165] FIG. 2. (A) Comparison of the fraction of CEA′ cells in each of eight PDO with the growth reduction achieved with cibisatamab versus the untargeted control antibody at the assay endpoint in FIG. 1. (B) Correlation analysis of growth reduction and the fraction of CEA.sub.hi cells for all PDOs. A linear regression line and the Pearson correlation coefficient and p value of the significance test are shown.

[0166] FIG. 3. Cell surface CEA expression of two colorectal cancer organoid lines with and without tankyrase-inhibitor treatment.

[0167] FIG. 4. Growth of colorectal cancer organoid lines over 7 days when cultured in the presence of CD8 T cells and cibisatamab or the untargeted control antibody, with or without 2 μM tankyrase-inhibitor (provided either as pre-treatment or continuous treatment).

[0168] FIG. 5. Growth of colorectal cancer organoid lines over 7 days when cultured in the presence of CD8 T cells and cibisatamab or the untargeted control antibody, with or without 10 μM tankyrase-inhibitor (provided as pre-treatment).

EXAMPLES

[0169] The following are examples of methods and compositions of the invention. It is understood that various other aspects may be practiced, given the general description provided above.

Example 1. Cibisatamab Sensitivity of PDOs in an Allogeneic T Cell Co-Culture Assay

[0170] Recently developed protocols allow the expansion and long term propagation of cancer cells as so called patient derived organoids (PDOs) from CRC biopsies (Sato et al., Gastroenterology. 2011; 141(5):1762-72). PDOs have been suggested to better represent the biological characteristics of patient tumors than cancer cell lines which were often established decades ago and have undergone changes during long term culture on plastic. The ability to rapidly generate model systems from patients furthermore enables matching of disease stage and prior treatment history to those of patients in whom novel drugs are clinically tested.

[0171] In order to assess the sensitivity of patient-derived colorectal cancer organoids (PDOs) to cibisatamab immunotherapy, eight patient derived colorectal cancer organoid lines with various cell surface CEA expression levels were generated and treated with cibisatamab (20 nM) or an untargeted control antibody (20 nM) during 10 days of co-culture with allogeneic CD8 T cells. CD8 T cells were isolated from allogeneic healthy donor peripheral blood mononuclear cells (PBMCs) by magnetic bead sorting and expanded in vitro with IL2 and CD3/CD28 beads for 7-14 days. GFP-tagged CRC PDO cells were then seeded in 96 well plates, T cells were added the following day and the co-cultures were imaged every 2-3 days on an automated 96 well plate fluorescence microscope. Effector to target (E:T) ratios of 2:1 and 5:1 were tested and an E:T of 2:1 was chosen for subsequent experiments as it effectively suppressed growth of the CEA.sub.hi PDO CRC-01 and showed no activity in the presence of the untargeted antibody (DP47-TCB) which was used as a negative control. Co-culture with CD8 T cells without any antibody was included as a further control to enable the identification of alloreactive donor T cells. Co-cultures in which T cells showed alloreactivity (observed in less than one in ten experiments) were excluded from the analysis and assays were repeated until each PDO line was tested with CD8 T cells from 3 independent allogeneic donors.

[0172] Eight PDO lines were tested: three CEA.sub.hi PDOs (i.e. containing predominantly CEA.sub.hi cells; CRC-01, CRC-05, CRC-07), four PDOs with mixed CEA expression (i.e. containing large subpopulations of both CEA.sub.hi and CEA.sub.lo cells; CRC-02, CRC-03, CRC-04, CRC-08), and one CEA.sub.lo PDO (i.e. containing predominantly CEA.sub.lo cells; CRC-06).

[0173] All three CEA.sub.hi PDOs were highly sensitive to treatment with CD8 T cells and cibisatamab whereas the predominantly CEA.sub.lo PDO CRC-06 showed resistance under these experimental conditions, as anticipated (FIG. 1). Our assay assessed the impact of cibisatamab over a period of 7-10 days and showed 89-100% growth inhibition in CEA.sub.hi PDOs. This confirms the high efficacy of cibisatamab to re-direct T cells to antigen positive cells in this assay.

[0174] We next tested the four PDOs with mixed CEA expression. Each of these continued to proliferate despite treatment with cibisatamab and T cells, with only a moderate reduction of the cancer cell growth rate compared to controls (FIGS. 1 and 2A). Thus, PDOs with mixed CEA expression only showed a partial response to this CEA targeting immunotherapy. Pearson correlation analysis of the achieved growth reduction with the fraction of CEA.sub.hi cancer cells in each PDO showed a strong and significant correlation of the percentage of cells showing high CEA expression within an organoid line and its sensitivity to cibisatamab (r=0.9152, 95% CI: 0.593 to 0.9848; p=0.0014; FIG. 2B).

[0175] This demonstrates that organoids which express uniformly high levels of CEA on the cell surface are sensitive to cibisatamab, organoids with predominantly CEA low cells are resistant and organoids with bimodal/mixed CEA expression only show limited sensitivity.

Example 2. Cell Surface CEA Expression of Two Colorectal Cancer Organoid Lines with and without Wnt Signalling Inhibitor Treatment

[0176] We flow sorted CEA.sub.hi and CEA.sub.lo cells from 3 PDOs and performed RNA expression analysis to investigate the mechanisms that regulate CEA expression and generate heterogeneity. We applied gene set enrichment analysis (GSEA) (Subramanian et al., Proc Natl Acad Sci USA. 2005 Sep. 30. 2005 Oct. 25; 102(43):15545-50) to identify potential molecular pathways which associate with CEA gene expression levels. WNT/β-catenin signalling was a significantly enriched signature following multiple testing correction and was upregulated in the CEA.sub.lo populations (data not shown). The WNT/β-catenin signalling pathway is genetically activated in the majority of CRCs, most frequently through mutations and loss of heterozygosity of the APC tumor suppressor gene and less commonly through mutations of other regulators of WNT signalling such as RNF43 or in β-catenin/CTNNB1 itself (Network CGA, Nature. 2012 Jul. 18; 487(7407):330-7; Giannakis et al., Nat Genet. 2014 December; 46(12):1264-6). High WNT/β-catenin pathway activity and absence of CEA expression are features of the intestinal crypt bottom where intestinal stem cells reside (Jothy et al., Am J Pathol. 1993 July; 143(1):250-7; Barker et al., Nat Rev Mol Cell Biol. 2013 Dec. 11; 15:19). Moreover, high WNT/β-catenin pathway activity is also a characteristic of colon cancer stem cells (de Sousa et al., Clin Cancer Res. 2011 Feb. 15; 17(4):647 LP-653).

[0177] We investigated if pharmacological inhibition of the WNT/β-catenin pathway enhances CEA expression as predicted by these data. Two PDO lines with mixed CEA expression were treated with an inhibitor of WNT signalling: the tankyrase inhibitor compound 21 which inhibits the downstream WNT/β-catenin pathway by stabilizing the β-catenin destruction complex (Elliott et al., Med Chem Comm. 2015; 6(9):1687-92; Mariotti et al., Br J Pharmacol. 2017; 174(24):4611-36). The Wnt signaling inhibitor increased CEA expression and the CEA.sub.hi subpopulation in both PDOs (FIG. 3).

[0178] An increase in CEA expression and the CEA.sub.hi subpopulation in PDOs with mixed CEA expression was also seen with another inhibitor of Wnt signaling, the porcupine inhibitor LGK-974 which prevents WNT ligand secretion and hence autocrine and paracrine WNT-receptor activation (results not shown).

[0179] These results confirmed a role of WNT/β-catenin signalling as a regulator of CEA expression in CRC PDOs.

Example 3. Combination Therapy of Cibisatamab and Tankyrase Inhibitor

[0180] We investigated growth of two PDO lines with mixed CEA expression (CRC-08, and CRC-06 after prolonged culture as compared to Example 1 above) when treated with the combination of cibisatamab and the tankyrase inhibitor Compound 21.

[0181] PDOs were cultured over 7 days in the presence of CD8 T cells and 20 nM of cibisatamab or the untargeted control antibody (DP47-TCB). Co-cultures were either performed a) without tankyrase-inhibitor, b) following 48 hours of pre-treatment with tankyrase-inhibitor which was removed at when T cells were added, or c) following 48 hours pre-treatment with tankyrase-inhibitor which was replenished at the time T cells were added for continuous tankyrase-inhibitor exposure. FIG. 4 shows the results for a 2 μM concentration of the tankyrase inhibitor, and FIG. 5 shows the results for a 10 μM concentration of tankyrase-inhibitor (continuous administration of 10 μM tankyrase-inhibitor over the entire assay duration was toxic to cancer cells and the data is not shown). The confluence of the GFP-labelled colorectal cancer organoid cultures was tracked by microscopy for 7 days following addition of T cells and antibody. Growth from the seeding density in the presence of non-targeting control to day 7 was defined as 100%. CD8 T cells had been generated from allogeneic healthy donor cells by extracting peripheral blood mononuclear cells followed by stimulation with IL-2 and CD3-beads and expansion in vitro. Experiments were performed in triplicates and the results shown are averages. Error bars represent one standard deviation. These data demonstrate that tankyrase-inhibitor treatment increases the sensitivity of colorectal cancer spheroid cultures to cibisatamab in a dose- and time-dependent fashion.

Example 4. Material and Methods

Human Samples and Cell Lines

[0182] Imaging-guided core biopsies from metastatic colorectal cancers that had been treated with at least two prior lines of chemotherapy were obtained from the Prospect C and Prospect R trials (Chief investigator: D. Cunningham, UK national ethics committee approval numbers: 12/L0/0914 and 14/LO/1812, respectively). One endoscopic biopsy from a treatment naïve primary colorectal cancer was obtained from the FOrMAT trial (Chief investigator: N. Starling, UK national ethics committee approval number 13/LO/1274). Trials were run at the Royal Marsden Hospital and all patients provided written informed consent before trial inclusion. Anonymized buffy coats from healthy donors were obtained from the local blood bank (National ethics committee approval number 06/Q1206/106) or through the Improving Outcomes in Cancer biobanking protocol at the Barts Cancer Institute (Chief investigator: T. Powles, UK national ethics committee approval number: 13/EM/0327) from individuals providing written informed consent. DLD-1 and MKN-45 cell lines were obtained from the American Type Culture Collection and were maintained in RPMI 1640 medium supplemented with 10% FBS, 1× Glutamax and 100 units/ml penicillin/streptomicin (Thermo Fisher).

Generation of Patient Derived Organoids

[0183] PDO cultures from CRC-01, CRC-02 and CRC-06 were established directly from core biopsies by rough chopping followed by embedding in growth factor reduced Matrigel (Corning). Very small biopsy fragments were available from CRC-03, CRC-04, CRC-05, CRC-07 and CRC-08 and these were first grafted subcutaneously or under the kidney capsule of female CD1 nude mice by the Tumour Profiling Unit at the Institute of Cancer Research (Home office license number PD498FF8D). Mice were culled once tumors had grown and tumors were removed and dissociated in a gentleMAX Octo dissociator using the Human Tumour Dissociation Kit (Miltenyi Biotec). Mouse cells were magnetically removed using the Mouse Cell Depletion Kit (Miltenyi Biotec), and purified human tumour cells were embedded into growth factor reduced Matrigel. PDOs were expanded in Matrigel as described (Sato et al., Gastroenterology. 2011; 141(5):1762-72) using Advanced DMEM/F12 media supplemented with 1× Glutamax, 100 units/ml penicillin/streptomycin, 1×B27, 1×N2, 10 mM HEPES (all Thermo Fisher), 1 mM N-acetyl cysteine, 10 mM nicotinamide, 10 μM SB202190, 10 nM gastrin, 10 μM Y27632 (Sigma Aldrich), 10 nM prostaglandin E2, 500 nM A-83-01, 100 ng/ml Wnt3a (Biotechne), 50 ng/ml EGF (Merck), 1 μg/ml R-Spondin, 100 ng/ml Noggin, and 100 ng/ml FGF10 (Peprotech). After at least 2 months of continuous growth in the matrigel matrix (minimum of 12 passages), the PDOs were first eGFP tagged (see below) and then adapted to grow in DMEM/F12 (Sigma Aldrich) with 20% fetal bovine serum (FBS), 1× Glutamax, 100 units/ml penicillin/streptomycin containing 2% Matrigel. PDO cultures were maintained in these conditions and used as required for T cell co-culture assays and FACS analysis. Genetic analyses of colon cancer driver genes were performed on each PDO line and these were identical to the mutations that had been identified in the matched tumor biopsies.

Labelling of PDOs with Nuclear eGFP

[0184] The nuclei of PDOs were labelled by introducing an eGFP tagged histone 2B construct (pLKO.1-LV-H2B-GFP) (Beronja et al., Nat Med. 2010 July; 16(7):821-7) to enable cell quantification by automated microscopy. For virus generation, HEK-293T cells were cultured in DMEM supplemented with 10% FBS, 1× Glutamax and 100 units/ml penicillin/streptomycin. Lentiviral particles were produced by overnight transfection with a plasmid mixture containing 9 μg of pLKO.1-LV-H2B-GFP, 2.25 pg of psPAX2 packaging plasmid (gift from Didier Trono; Addgene plasmid #12260; http://n2t.net/addgene:12260; RRID: Addgene_12260) and 0.75 ug of pMD2.G envelope plasmid (gift from Didier Trono; Addgene plasmid #12259; http://n2t.net/addgene:12259; RRID: Addgene_12259) using TransIT-293 transfection reagent (Mirus). The cells were media changed the following day, virus harvested after 24 hours and passed through a 0.45 uM filter before use. For lentiviral transduction PDOs were harvested from the cultures in Matrigel and dissociated to single cells using TrypLE Express (Thermo Fisher), and pelleted. The pellets were resuspended in media with the addition of virus and 1 nM polybrene (Sigma Aldrich) and centrifuged at 300 g for 1 hour. The samples were resuspended and plated in culture for between 6 hours and overnight, before replacing the media. Following recovery and expansion, eGFP positive cells were sorted by flow cytometry and further expanded before use.

Surface CEA Expression Analysis by Flow Cytometry

[0185] Cell lines were harvested using enzyme-free Cell Dissociation Buffer (Thermo Fisher) and PDOs with TrypLE Express (Gibco). 2×10.sup.5 cells were stained with 20 nM of the human anti-human CEA antibody CH1A1A (Roche) and 25 ug/ml of the R-Phycoerythrin conjugated secondary antibody AffiniPure F(ab′)2 Fragment Goat Anti-Human IgG, Fcγ Fragment Specific (Stratech). DRAQ7 (Biostatus) staining was included for dead cell exclusion. CEA expression was analysed on a Sony SH800 flow cytometer. Gate boundaries were set at the trough between high and low CEA populations in PDOs with mixed CEA expression and identical gates were used across all samples. The percentage of CEA.sub.hi and CEA.sub.lo populations and mean fluorescence intensities (MFI) were calculated for each PDO.

CD8 T Cells Expansion from Peripheral Blood Mononuclear Cells

[0186] Peripheral Blood Mononuclear Cells (PBMCs) were isolated from buffy coats with Ficoll-Paque according to the manufacturer's protocol (GE Healthcare). CD8 T cells were isolated from PBMCs with Human CD8 Dynabeads FlowComp (Thermo Fisher). The purity of CD8 T cells was assessed by flow cytometry (Alexa Fluor 488 anti-human CD8, Sony Biotechnology) and only populations with at least 90% CD8 positive cells were used for expansion with the CD3/CD28 Dynabeads Human T-Activator kit (Thermo Fisher) in RPMI 1640 supplemented with 10% FBS (Biosera), 1× Glutamax, 100 units penicillin/streptomycin and 30 U/mL IL-2 (Sigma Aldrich) following the manufacturer's protocol.

Co-Culture of PDOs and CD8 T Cells

[0187] PDOs were harvested with TrypLE Express and neutralised with DMEM/F12 Ham medium (Sigma Aldrich) with 10% FBS. Cells were filtered through a 70 μm filter, counted and resuspended in phenol-red free RPMI medium (Thermo Fisher) supplemented with 10% FBS (Biosera), 1× Glutamax and 100 units penicillin-streptomycin. On day 0, 5000 tumor cells per well of a 96 well-plate (Corning Special Optics Microplate) were plated. CD8 T cells were added on day 1 at the indicated effector to target (E:T) ratios with 20 nM of cibisatamab or 20 nM of the untargeted negative control antibody DP47-TCB (both provided by Roche). Tumor cells without CD8 T cells and without antibody were also included as controls. All conditions were plated in triplicates and at least 3 different healthy donors were tested on each of the 8 PDOs.

Cancer Cell Growth Assessment by Immunofluorescence Microscopy

[0188] The GFP confluence was quantified every 48h-72h over a 10-day period using the GFP confluence application on the Celigo Imaging Cytometer (Nexcelom Bioscience). GFP confluence analysis was able to track the growth of GFP positive PDO cells over multiple timepoints without erroneously counting the T cells in the co-culture. Confluence analysis was furthermore superior to the counting of cell nuclei which generated inaccurate results in areas of high cancer cell density such as the PDO centre. The main advantage of confluence analysis over measuring spheroid diameters is the ability to track even the growth of PDOs showing highly variable shapes. Growth curves were generated with CD8 T cells from three different healthy blood donors. The percentage growth reduction was calculated from readings taken between days 7 to 9, before PDOs showed growth retardation, likely due to exhaustion of the growth media. In order to calculate the percentage of growth reduction, confluence at day 1 was subtracted and the confluence in wells treated with the DP47-TCB control antibody at the endpoint was set to 100%.

Wnt/β-Catenin Pathway Inhibition Assay

[0189] 10.sup.5 PDO cells/well were seeded in 12 well plates and allowed to attach overnight. Media were changed and cells were treated with DMSO control or with 10 μM tankyrase inhibitor (Compound 21) (Elliott et al., Med Chem Comm. 2015; 6(9):1687-92) or 10 μM porcupine inhibitor (LGK-974, SelleckChem) for 3 days. Cells were harvested using TrypLE Express, stained for CEA with the CH1A1A primary antibody and the R-Phycoerythrin conjugated secondary antibody and analyzed by FACS as described above.

Statistical Analyses

[0190] Pearson correlation analysis and the paired t-tests were performed with the GraphPad Prism software. All p values are two tailed. Gene set enrichment analysis was performed with the GSEA software V3.0 using 5000 gene set permutations and the Hallmarks V6.2 gene set collection (Subramanian et al., Proc Natl Acad Sci USA. 2005 Sep. 30. 2005 Oct. 25; 102(43):15545-50).

[0191] Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, the descriptions and examples should not be construed as limiting the scope of the invention. The disclosures of all patent and scientific literature cited herein are expressly incorporated in their entirety by reference.