Anti-CD137 antibodies and uses thereof

Abstract

The present invention relates to antibodies (and fragments, variants, fusions and derivatives thereof) with binding specificity for domain 2 of human CD137 which are capable of inhibiting the binding of a reference antibody to human CD137. The antibodies and fragments have utility in the treatment of diseases such as cancer. The invention also relates to pharmaceutical compositions, uses, methods and kits comprising such antibodies.

Claims

1. An antibody or an antigen-binding fragment thereof with binding specificity for domain 2 of human CD137 wherein the antibody or antigen-binding fragment is a CD137 agonist and is capable of inhibiting the binding of reference antibody ‘1630/1631’ to human CD137, and wherein the antibody or antigen-binding fragment comprises: i) a) a heavy chain CDR1 sequence with the consensus sequence G, F, T/N, F, G, Y, S, Y (SEQ ID NO: 31); b) a heavy chain CDR2 sequence with the consensus sequence I, G, S, G/T, S, S, Y/H, T (SEQ ID NO: 32); and c) a heavy chain CDR3 sequence with the sequence ARVYSSPGIDY (SEQ ID NO: 5); and ii) a) a light chain CDR1 sequence with the consensus sequence Q, S, I, S/G, S, Y/T (SEQ ID NO: 33); b) a light chain CDR2 sequence with the consensus sequence A/G, A, S (SEQ ID NO: 34); and c) a light chain CDR3 sequence with the sequence QQYYTWVPFT (SEQ ID NO: 8).

2. An antibody or an antigen-binding fragment thereof according to claim 1, wherein the antibody or antigen-binding fragment is capable of inhibiting the binding of reference antibody ‘2674/2675’ to human CD137.

3. An antibody or antigen-binding fragment thereof according to claim 1 wherein the antibody or antigen-binding fragment exhibits one or more of the following properties: a) the ability to stimulate CD137 and activate T cells and other immune cells via a cross-linking dependent mechanism; and/or b) cross-reactivity with cyno-CD137 antibodies; and/or c) is capable of binding an Fc receptor, optionally, wherein the ability of the antibody to activate T cells is dependent upon binding to both CD137 and Fc receptors; and/or d) is capable of inducing tumor immunity.

4. An antibody or antigen-binding fragment thereof according to claim 1 wherein the antibody or antigen-binding fragment is incapable of inducing the following upon binding to cells expressing CD137: a) antibody-dependent cellular cytotoxicity (ADCC); b) antibody-dependent cellular phagocytosis (ADCP); and/or c) complement-dependent cytotoxicity (CDC).

5. An antibody or antigen-binding fragment thereof according to claim 1 wherein the antibody or antigen-binding fragment is capable of binding to an epitope on the extracellular domain of CD137 which overlaps, at least in part, with the epitope on CD137 to which reference antibody 1630/1631 is capable of binding, optionally, wherein the antibody or antigen-binding fragment is capable of binding to an epitope on the extracellular domain of CD137 which overlaps, at least in part, with the epitope on CD137 to which reference antibody 2674/2675 is capable of binding, further optionally wherein the epitope is located at or within amino acids 66 to 107 of human CD137.

6. An antibody or antigen-binding fragment thereof according to claim 1 comprising: a) an intact antibody; or b) an antigen-binding fragment selected from the group consisting of Fv fragments, and Fab-like fragments.

7. An antibody or antigen-binding fragment according to claim 1 wherein: a) the antibody or antigen-binding fragment thereof is a recombinant polypeptide; and/or b) the antibody or antigen-binding fragment thereof is monoclonal; and/or c) the antibody or antigen-binding fragment thereof is human or humanised.

8. An antibody or antigen-binding fragment thereof according to claim 1 comprising: a) a heavy chain variable region comprising the CDRs of SEQ ID NOs 3, 4 and 5, optionally wherein the antibody or antigen-binding fragment thereof comprises a heavy chain variable region having the amino acid sequence of SEQ ID NO: 1 or an amino acid sequence having at least 60% sequence identity therewith; and/or b) a light chain variable region comprising the CDRs of SEQ ID NOs: 6, 7 and 8, optionally wherein the antibody or antigen-binding fragment thereof comprises a light chain variable region having the amino acid sequence of SEQ ID NO: 2 or an amino acid sequence having at least 60% sequence identity therewith; and/or c) the antibody or antigen-binding fragment thereof comprises the CDRs of SEQ ID NOs: 3, 4, 5, 6, 7 and 8, optionally wherein the antibody or antigen-binding fragment thereof comprises a heavy chain variable region which comprises or consists of the amino acid sequence of SEQ ID NO: 1 and a light chain variable region which comprises or consists of the amino acid sequence of SEQ ID NO: 2.

9. An antibody or antigen-binding fragment thereof according to claim 1 comprising: a) a heavy chain variable region comprising the CDRs of SEQ ID NOs 21, 22 and 23, optionally wherein the antibody or antigen-binding fragment thereof comprises a heavy chain variable region having the amino acid sequence of SEQ ID NO 19: or an amino acid sequence having at least 60% sequence identity therewith; and/or b) a light chain variable region comprising the CDRs of SEQ ID NOs: 24, 25 and 26, optionally wherein the antibody or antigen-binding fragment thereof comprises a light chain variable region having the amino acid sequence of SEQ ID NO: 20 or an amino acid sequence having at least 60% sequence identity therewith; and/or c) the antibody or antigen-binding fragment thereof comprises the CDRs of SEQ ID NOs: 21, 22, 23, 24, 25 and 26, optionally wherein the antibody or antigen-binding fragment thereof comprises a heavy chain variable region which comprises or consists of the amino acid sequence of SEQ ID NO: 19 and a light chain variable region which comprises or consists of the amino acid sequence of SEQ ID NO: 20.

10. An antibody or antigen-binding fragment thereof according to claim 1 comprising: a) a heavy chain constant region, or part thereof, optionally wherein the heavy chain constant region is of an immunoglobulin subtype selected from the group consisting of IgG1, IgG2, IgG3 and IgG4, further optionally wherein the heavy chain constant region comprises or consists of an amino acid sequence selected from the group consisting of SEQ ID NO: 12, 13, 14 and 15; and/or b) a light chain constant region, or part thereof, optionally wherein the light chain constant region is of a kappa or lambda light chain, further optionally wherein the light chain constant region comprises or consists of an amino acid sequence of SEQ ID NO: 16.

11. An antibody or antigen-binding fragment thereof according to claim 1 comprising an Fc region, optionally wherein a) the Fc region is naturally occurring; or b) the Fc region is non-naturally occurring, further optionally wherein the Fc region comprises mutations to shorten the half-life of the antibody or antigen binding fragment.

12. An antibody or antigen-binding fragment thereof according to claim 1 comprising: i) a) a heavy chain comprising a variable region of SEQ ID NO: 1 together with a constant region of SEQ ID NO: 13; and b) a light chain comprising a variable region of SEQ ID NO: 2 together with a constant region of SEQ ID NO: 16; and/or ii) a) a heavy chain comprising a variable region of SEQ ID NO: 19 together with a constant region of SEQ ID NO: 13; and b) a light chain comprising a variable region of SEQ ID NO: 20 together with a constant region of SEQ ID NO: 16.

13. An antibody or antigen-binding fragment thereof according to claim 1 wherein: a) the antibody is an intact IgG4 molecule comprising or consisting of two heavy chains having an amino acid sequence of SEQ ID NO: 17 and two light chains having an amino acid sequence of SEQ ID NO: 18; and/or b) the antibody is an intact IgG4 molecule comprising or consisting of two heavy chains having an amino acid sequence of SEQ ID NO: 29 and two light chains having an amino acid sequence of SEQ ID NO: 30.

14. An antibody or antigen-binding fragment thereof according to claim 1 further comprising: i) a cytotoxic moiety, optionally wherein the cytotoxic moiety comprises or consists of a) a radioisotope or b) cytotoxic drug; and/or ii) a detectable moiety, optionally wherein the detectable moiety comprises or consists of a radioisotope, and optionally wherein the cytotoxic moiety and/or detectable moiety is joined to the antibody or antigen-binding fragment thereof indirectly, via a linking moiety, optionally wherein the linking moiety is a chelator, further optionally wherein the chelator is selected from the group consisting of derivatives of 1,4,7,10-tetraazacyclododecane-1,4,7,10,tetraacetic acid (DOTA), deferoxamine (DFO), derivatives of diethylenetriaminepentaacetic acid (DTPA), derivatives of S-2-(4-Isothiocyanatobenzyl)-1,4,7-triazacyclononane-1,4,7-triacetic acid (NOTA) and derivatives of 1,4,8,11-tetraazacyclodocedan-1,4,8,11-tetraacetic acid (TETA).

15. A pharmaceutical composition comprising an effective amount of an antibody or antigen-binding fragment thereof according to claim 1 and a pharmaceutically-acceptable diluent, carrier or excipient.

16. An antibody or antigen-binding fragment thereof according to claim 6, wherein said intact antibody is an IgG1, IgG2, IgG3 or IgG4 antibody; wherein said Fv fragment is a single chain Fv or disulphide-bonded Fv; wherein said Fab-like fragment is a Fab fragment, Fab′ fragment, or F(ab)2 fragment.

17. An antibody or antigen-binding fragment thereof according to claim 8, wherein a) the antibody or antigen-binding fragment thereof comprises a heavy chain variable region having an amino acid sequence having at least 70%, 80%, or 90% sequence identity with SEQ ID NO: 1; and/or b) the antibody or antigen-binding fragment thereof comprises a light chain variable region having an amino acid sequence having at least 70%, 80%, or 90% sequence identity with SEQ ID NO: 2.

18. An antibody or antigen-binding fragment thereof according to claim 9, wherein a) the antibody or antigen-binding fragment thereof comprises a heavy chain variable region having an amino acid sequence having at least 70%, 80%, or 90% sequence identity with SEQ ID NO: 19; and/or b) the antibody or antigen-binding fragment thereof comprises a light chain variable region having an amino acid sequence having at least 70%, 80%, or 90% sequence identity with SEQ ID NO: 20.

Description

(1) Preferred, non-limiting examples which embody certain aspects of the invention will now be described, with reference to the following figures:

(2) FIG. 1 shows binding to CD137 human and cynomolgus CD137. Data from separate two experiments included.

(3) FIG. 2 shows the CD137 variants in Example 6.

(4) FIG. 3 shows a summary of two experiments of CD137 mAb competition with CD137L binding to CHO-huCD137 cells when titrated (from left to right 25 μg/ml).

(5) FIG. 4 shows the stimulation index of clones normalized to reference REF1.

(6) FIG. 5 shows the induction of NF-κB mediated signaling by the antibodies with and without cross-linking.

(7) FIG. 6 shows the effect of treatment with the 1630/1631 antibody on tumour volume in a mouse tumour model.

(8) FIG. 7 shows determined patch size and Aggscore for 2674/2675 as well as the parental clone 1630/1631 clone via Schrödinger analysis.

(9) FIG. 8 shows binding of 2674/2675 to human and cynomolgus CD137.

(10) FIG. 9 shows crosslinking of 2674/2675 and parental clone 1630/1631 with FcγRI transfected CHO cells in the CD137 reporter assay.

(11) FIG. 10 shows crosslinking of 2674/2675 and parental clone 1630/1631 with FcγRIIa R131 and FcγRIIb in the CD137 reporter assay.

(12) FIG. 11 shows CHO cells transfected with empty vector (pcDNA3.1), used to determine crosslinking independent activation in the reporter cell line.

(13) FIG. 12 shows IFN-γ production of CD8+ T cells after stimulation with CD137 mAbs when crosslinked with FcγRI transfected CHO cells. Summary of IFN-γ response normalized against 2674/2675 agonist response in a CD8.sup.+ T cell agonist assay (n=5).

(14) FIG. 13 shows Dot plots showing the correlation between the mean expression values of Fcγ receptor (X axes) and TNFRSF9 (CD137, Y axes) for various human cancers. Cancers with an above average expression (mean expression level ≥10) of both Fcγ receptor and CD137 have been highlighted as clear symbols.

EXAMPLES

Example 1—Selection of CD137 Antibodies from Alligator GOLD

(15) Phage display selections were performed using a human antibody (scFv) library, Alligator

(16) GOLD. Selections towards recombinant CD137 in soluble form, coated onto the surface of beads or tubes, or expressed on the surface of CD137-transfected cells were performed. CTLA4-Fc and an irrelevant His-tagged protein were used as non-targets included in excess in the selections. Prior to each selection round, the phage stocks were pre-selected towards non-target proteins, beads or CD137 negative cells to remove unspecific binders.

(17) To identify specific binders from the phage selection, approximately 4500 individual clones were screened in phage format using ELISA coated with either recombinant target (CD137-Fc) or non-target protein, followed by confirmation as soluble scFv for some clones. Clones exhibiting specific binding to CD137 were sequenced and unique clones were produced as IgG for further characterization.

Example 2—Binding to Human CD137 Measured by ELISA

(18) Aim

(19) The aim was to determine binding potency of the CD137 antibody.

(20) Material and Methods

(21) Binding of CD137 antibodies to recombinant human CD137 was determined by sandwich ELISA. Briefly, ELISA plates (Greiner #655074) coated with recombinant human CD137-Fc (R&D #838-4B) were incubated with serial dilutions of the various CD137 antibodies to be investigated. CD137 antibodies were detected using HRP-conjugated goat-anti-human kappa light chain (AbD Serotec #STAR127P) and developed with SuperSignal ELISA Pico Chemiluminescent substrate (Pierce #37069). EC50 values of the various antibodies were determined in 2-6 separate experiments.

(22) Two different reference antibodies with specificity for CD137, synthesized from published amino acid sequence information, were used in this study (designated “REF1” and “REF2”).

(23) The other reference antibodies used, namely REF3, REF4 and REF5, are human CD137-specific monospecific IgG antibodies obtained from the Alligator GOLD library. They are agonistic and stimulate T cells upon binding to CD137. The binding epitopes of the reference antibodies has been established as outlined in Examples 6-8 (see below).

(24) The reference antibodies were selected because they have previously undergone at least some clinical testing and so represent the benchmark against which new anti-CD137 antibodies can be judged for improved properties and/or function.

(25) Results and Conclusion

(26) Exemplary antibody 1630/1631 exhibits EC50 values in a similar range as those of the reference antibodies, i.e. sub nM. Data is summarized in Table 1.

(27) TABLE-US-00020 TABLE 1 EC50 values (nM) of CD137 antibodies determined by ELISA for human CD137. Antibody Mean SD n REF1 0.75 0.137 8 REF2 0.33 0.069 5 REF3 0.39 0.037 3 REF4 0.41 0.050 4 REF8 0.38 0.137 2 1630/1631 0.27 0.078 4 n = number of data points.

Example 3—Binding to Human and Cynomolgus CD137 Measured by Flow Cytometry

(28) Aim

(29) The aim of this study was to determine the binding to human and cynomolgus (Macaca fascicularis) CD137.

(30) Material and Methods

(31) Binding and EC50 was determined using flow cytometric of CHO cells transfected with human CD137, cynomolgus CD137 or empty vector. The extracellular part of human or cynomolgus CD137 was fused to the transmembrane and intracellular part of human CD40 and cloned into pcDNA3.1. The vector was subsequently stably transfected into CHO cells. Expression of CD137 was confirmed by flow cytometry using CD137 antibody (human CD137-PE, BD Biosciences #555956) for 30 min at 4° C. CD137-transfected and empty vector-transfected cells were incubated with CD137 antibodies for at least 1 h at 4° C. to saturate the binding. In order to minimize antibody internalization, 0.05% sodium azide was used in the incubation buffer and all work was performed on ice. The CD137 antibodies were detected using PE-conjugated anti-hIgG antibody (109-115-098, Jackson Immunoresearch laboratories), incubated for 30 min at 4° C. Directly after staining the cells were fixed with a paraformaldehyde solution (10× concentrate BD CellFIX, BD biosciences #340181). Cells were analyzed by flow cytometry using FACSVerse (BD Biosciences). The median fluorescence intensity (MFI) for each sample was determined and the dose response data was analysed using Graph Pad Prism.

(32) MFI data was normalized for each antibody, where 0% is defined as the lowest value and 100% is the highest value in the dose titration for each antibody. EC50 and 95% confidence interval were calculated with Graph Pad Prism based on data from the two experiments (non-linear regression (curve fit), constraints set to 0 and 100).

(33) Results and Conclusion

(34) Binding to CHO-huCD137, CHO-cyCD137 and CHO-pcDNA was confirmed in two separate experiments (FIG. 1). 1630/1631 binds to human CD137 with EC50 comparable with the two reference antibodies REF1 and REF2. 1630/1631 binds well to cynomolgus CD137. Reference antibody REF1 and REF8 (FIG. 1) binds very weakly or not at all to cynomolgus CD137. REF8 exhibits weak binding and does not reach a complete saturation.

(35) The EC50 determination is presented as 95% confidence intervals for each CD137 antibody tested in order to include the inter and intra assay variations (Table 2).

(36) TABLE-US-00021 TABLE 2 95% confidence intervals for the EC50 of each CD137 antibody determined as an average from two experiments of normalized data. Binding to human Binding to cyno Ratio, Antibody CD137, EC50 (μg/mL) CD137, EC50 (μg/mL) cyno:human REF1 1.00-1.99 Nd Nd REF2 0.21-0.31 0.13-0.24 0.69 REF3 0.20-0.36 Nd Nd REF4 0.16-0.27 0.11-0.17 0.67 REF8 0.20-0.42 >3 >14 1630/1631 0.17-0.26 0.12-0.16 0.63 Nd: Not detectable

Example 4—Affinity Measured by Biacore

(37) Aim

(38) The aim was to estimate the affinity, on rate and off rate of the different CD137 antibodies.

(39) Material and Methods

(40) Human CD137 (R&D systems) was immobilized to the Biacore™ sensor chip, CMS, using conventional amine coupling. The tested antibody and control (serially diluted 1/2 10-0.63 nM) were analyzed for binding in HBS-P (GE, #BR-1003-68) at a flow rate of 30 μl/ml. The association was followed for 5 minutes and the dissociation for 15 minutes. Regeneration was performed twice using 10 mM Glycine pH 1.7 for 30 seconds. The kinetic parameters and the affinity constants were calculated using 1:1 Langmuir model.

(41) Results and Conclusion

(42) The affinities of the antibodies were in the nanomolar to sub-nanomolar range (Table 3) measured using bivalent antibodies flowed over CD137 coated on the chip surface.

(43) TABLE-US-00022 TABLE 3 Kinetic parameters measured by surface plasmon resonance Antibody ka (1/Ms) kd (1/s) KD (M) REF4 6.76E+05 6.60E−04 9.76E−10 REF8 3.92E+05 5.19E−04 1.32E−09 1630/1631 1.85E+06 1.18E−03 6.41E−10 REF2 1.05E+06 4.45E−04 4.24E−10

Example 5—Target Specificity of the CD137 Antibodies

(44) Aim

(45) The aim with this study was to evaluate the risk that any of the CD137 antibody binds targets other than CD137.

(46) Material and Methods

(47) Binding to TNFR superfamily members for which ELISA methods had already been established (CD40 and OX40) was evaluated to detect potential propensity to cross react to non-target proteins. In addition, a BLAST search was performed identifying TNFRSF21 as the most similar sequence (34% sequence identity). Since this sequence similarity is rather low, determination of non-target binding to OX40 and CD40 was considered sufficient.

(48) ELISA plates (Greiner #655074) were coated with 50 μl/well of recombinant human OX40 (R&D #1493-CD), CD40-Fc (Ancell #504-820) or CD137 (R&D #838-4B) diluted to a final concentration of 0.5 μg/ml in PBS for 1 h at 37° C. or overnight at 4° C. Plates were washed with PBS+0.05% TWEEN20 (PBST), followed by block with PBST+1% bovine serum albumin (BSA). Antibody samples were prepared as serial 1/10 dilutions from 10-0.01 μg/ml in PBST+1% BSA and incubated for 1 h in room temperature, followed by detection using a horse radish peroxidase-conjugated anti-human kappa light chain antibody (AbD Serotec #STAR127P) and developed using SuperSignal ELISA Pico Chemiluminescent substrate (Pierce ThermoScientific #37069).

(49) Results and Conclusion

(50) TABLE-US-00023 TABLE 4 Summary of CD137 antibody unspecific binding to OX40 and CD40 Antibody Binding to OX40 and CD40 EC50 CD137 REF3 No REF4 Weak; EC50 >6 μg/ml (40 nM) 0.4 nM REF8 No 1630/1631 No 0.4 nM 2674/2675 No 0.3 nM

(51) The results from the two experiments were similar. One antibody (REF4) exhibited weak binding to OX40 and CD40, whereas none of the remaining antibody showed any detectable binding to either OX40 or CD40. An overview of antibodies analyzed, and results from the two experiments is shown in Table 4.

(52) Further, binding to primary PBL from multiple blood donors was tested. The binding of 1630/1631 and 2674/2675 to PBL was similar to Reference antibodies. No relevant unspecific binding to non-target proteins was detected.

Example 6—Domain Mapping of Antibodies Binding to CD137

(53) Aim

(54) The aim was to define distinct classes of epitope specificity, and compare to the properties of reference antibody.

(55) Material and Methods

(56) The ability of each antibody to bind to a panel of human/mouse CD137 chimeras expressed on the surface of transfected cells was analyzed by flow cytometry.

(57) The chimeras were designed by exchanging domains or modules of the human CD137 with the corresponding mouse domain. Genes of CD137 human/mouse chimeras were synthesized (GenScript) and constructs cloned into pcDNA3.1 vector (Invitrogen) and transiently transfected into FreeStyle 293-F cells (Invitrogen). The transfected cells were incubated with CD137 antibodies and control antibodies, followed by incubation with anti-human IgG-PE (Jackson Immunoresearch) for detection and analyzed with FACS Verse (BD Biosciences). Binding to the different chimeric constructs was calculated as relative MFI compared to the binding of the isotype control, followed by normalization to the full-length human CD137 construct to minimize the effect of affinity differences between individual antibodies.

(58) Results and Conclusion

(59) Three binding patterns was observed as described below FIG. 2. Data is summarized in Table 5.

(60) Pattern A

(61) Antibody REF1 depends on domain 1 for binding to human CD137.

(62) Pattern B

(63) Antibodies REF3, REF4, 2674/2675 and 1630/1631 are mainly dependent on domain 2 for binding to human CD137.

(64) Pattern C

(65) Antibodies REF2 (Reference antibody) and REF8 appear to be mainly dependent on domains 3B-4A for binding to human CD137.

(66) TABLE-US-00024 TABLE 5 Median fluorescence intensity (MFI) for antibody sample/isotype control, normalized to full-length human CD137. 2 Domain 1 1630/ 2674/ 3B-4A Clone Description REF1 1631 2675 REF3 REF4 REF2 REF8 1550 Human CD137 with mouse 0.12 0.11 0.13 0.05 0.05 0.22 0.17 domains 1, 2A and 2B (aa 24-86) 1551 Human CD137 with mouse 0.41 0.10 0.15 0.04 0.05 0.37 0.33 domains 2A, 2B and 3A (aa 47-96) 1552 Human CD137 with mouse 0.76 0.25 0.13 0.05 0.06 0.19 0.18 domains 2B, 3A and 3B (aa 64-118) 1553 Human CD137 with mouse 1.07 0.91 1.08 0.65 0.65 0.17 0.17 domains 3A, 3B and 4A (aa 87-133) 1554 Human CD137 with mouse 0.82 0.85 0.88 0.84 0.51 0.16 0.17 domains 3B, 4A and 4B (aa 97-159) 1555 Human CD137 with mouse 0.11 0.35 0.38 0.24 0.26 0.26 0.32 domains 1 and 4B and region of unknown function (aa 24-46 and aa 139-186)  1030* Human full length CD137 1 1 1 1 1 1 1

Example 7—CD137 Ligand Blocking

(67) Aim and Background

(68) The aim was to determine if the CD137 antibodies block the CD137 ligand binding.

(69) If the CD137 antibodies bind to epitopes close to the ligand binding region, binding to the antigen can lead to partly or total block of ligand biding. Binding close to the CD137 ligand binding epitope may also affect the ligand binding due to steric hindrance or conformational changes of the CD137 ligand binding epitope. All CD137 antibodies were titrated against a fixed concentration of CD137L for evaluation of ligand blocking properties.

(70) Material and Method

(71) CHO-cells transfected with human CD137 were used for the ligand competition. The extracellular part of human CD137 was fused to the transmembrane and intracellular part of hCD40 and cloned into pcDNA3.1. The vector was subsequently stably transfected into CHO cells. The expression of CD137 was confirmed by staining with commercial antibody targeting CD137.

(72) The CHO-huCD137 were pre-incubated with CD137 monoclonal antibodies, titrating down from a predetermined saturating concentration (0.25 μg/ml), for 1 h at +4 C before the addition of CD137 ligand at a concentration at EC50. After co-incubation for another 30 min at +4 C, the cells were washed and bound CD137 ligand was detected with anti-FLAG-APC (Cell signaling technology). Before analyzation the cells were fixed with paraformaldehyde (10× concentrate BD CellFIX, BD biosciences). Analyzation was performed with FACSverse and the MFI (Median Fluorescence Intensity) was calculated with FlowJo software.

(73) Results and Conclusion

(74) It can be concluded that all CD137 mAbs tested were not blocking the CD137 ligand binding (Table 6, FIG. 3). CD137 mAbs belonging to group B and C, binding to domain 2B-4A, block the CD137L (including 2674/2675 and 1630/1631). REF1 belonging to group A which bound to domain 1, did not block CD137 ligand. REF1 increased the binding of the CD137L.

(75) TABLE-US-00025 TABLE 6 Maximal CD137 ligand competition of the CD137 antibodies. Group CD137L, (domain CD137 max mapping) mAb inhib. A REF1 −167%  B 1630 69% B 2674/2675 66% C REF2 −26% 

Example 8—Competition ELISA

(76) Aim and Background

(77) By competing each CD137 antibody with each another, it is possible to determine antibodies binding to similar epitopes based on their blocking pattern. The competition ELISA is performed by co-incubating biotinylated CD137 antibodies with non-biotinylated CD137 antibodies when binding to coated CD137-Fc. Competition is defined as loss of signal from the biotinylated CD137 antibody. Low competition values could either be due to no competition between the antibodies or binding kinetics of the antibodies. Binding of one antibody could also lead to steric hindrance or conformational changes when binding the antigen which affects the binding of the other CD137 antibody.

(78) Material and Method

(79) CD137 antibodies were biotinylated (EZ-link NHS-LC-Biotin, ThermoFisher) and intact binding properties to CD137-Fc was verified with ELISA by comparing EC50 between biotinylated and non-biotinylated anti-CD137 mAbs. Non-biotinylated anti-CD137 (anti-CD137-bio) was pre-incubated to CD137-Fc at concentrations 30 times higher than the determined EC50 for 0.5 h. Without washing, anti-CD137-bio was added and co-incubated for another 1 h. The binding of anti-CD137-bio was detected with Streptavidin-HRP (Pierce). Competition was calculated as the relative number by dividing the binding measured to other antibodies relative to its maximum competition (competing with itself). The relative values obtained were normalized against the maximum blocking capacity (Table 7).

(80) TABLE-US-00026 TABLE 7 Summary of CD137 antibody competition ELISA from two experiments. REF1 REF4 1630/1631 REF2 REF8 REF1 100 7 5 5 4 REF2 15 41 70 94 61 REF4 18 58 91 63 50 REF8 4 49 91 100 82 1630/1631 14 31 56 23 16

(81) Result and Conclusion

(82) The competition ELISA was repeated two times. In both experiments, several of the CD137 mAbs did not fully compete with itself (Table 7). The antibody REF1 that belongs to domain mapping group A, displayed a unique pattern in the competition ELISA. The other CD137 antibodies that were analyzed displayed similar blocking patterns. Differences in binding kinetics between those antibodies may explain some of the minor variations in the binding patterns among these antibodies, although it cannot be excluded that the small variations within groups reflects actual differences in the binding epitope.

Example 9—In Vitro Efficacy of CD137 Antibodies

(83) Aim

(84) The aim was to identify CD137 antibodies with agonistic activity.

(85) Material and Methods

(86) Agonistic activity of CD137 antibodies was evaluated in a T cell assay based on primary human CD8+ T cells. Briefly, CD8+ T cells were separated from human peripheral blood mononuclear cells by MACS separation (Miltenyi #130-096-495) according to the manufacturer's protocol. Cells were incubated in 96-well microtiter plates (NuncThermo Scientific #268200), pre-coated with anti-CD3 antibody (clone OKT3, Affymetrix eBioscience #16-0037) and titrated concentrations of the CD137 antibody to be tested. Following 72 or 96-hour incubation, culture medium was harvested and IFN-γ levels were determined by ELISA (BD #555142).

(87) Each clone was analyzed in at least 6 donors and compared to the reference CD137 antibody REF1 and the negative control antibody.

(88) Due to large intra-donor variations the stimulation index (SI, fold induction by antibody compared to negative control) was determined for each sample and normalized to the stimulation index for the reference antibody REF1.

(89) Results and Conclusion

(90) Several clones with efficacy comparable to the reference REF1 were identified. Data is summarized in FIG. 4.

(91) Table 8 Table 8 indicates the absolute IFN-γ levels induced by CD137 stimulation. However, all antibodies were not analyzed head-to-head in all donors, and the normalized SI is more relevant for comparison of the efficacy.

(92) TABLE-US-00027 TABLE 8 IFN-γ production levels induced by the various antibody. Mean IFN-γ Min IFN-γ Max IFN-γ Clone name (pg/ml) (pg/ml) (pg/ml) n Ctrl IgG 2502 337 8526 13 REF1 42268 2256 136802 12 REF4 26749 11952 51832 8 REF8 52448 7727 123127 8 1630/1631 51236 3361 145055 8

Example 10—In Vitro NFkB Reporter Assay

(93) 293T cells (30 million) were transfected with plasmids which encode for human CD137, firefly luciferase under NF-κB promoter and renilla. After 5 hours of transfection, the antibodies were added at three different concentrations. 18 hours later, cells were harvested and luciferase reporter assay (Promega) was performed. The cells were cultured with soluble antibodies without crosslinking, as well as with cross-linking, at 5 μg/ml with crosslinking using anti-IgG antibody.

(94) Results

(95) 1630/1631 stimulates CD137 inducing NF-κB mediated signaling when cross linked but not in the absence of a cross linking agent. In contrast, REF1 induce CD137 signaling also in the absence of a cross linking agent.

Example 11—In Vivo Anti-Tumour Effect in HT-29 Colon Cancer Model

(96) Summary

(97) The anti-tumour effect of 1630/1631 was investigated using hPBMC humanized immunodeficient mice and subcutaneous tumour models of HT-29 colon carcinoma.

(98) 1630/1631 demonstrated statistically significant tumour volume inhibition.

(99) Material and Methods

(100) Leukocyte concentrates were obtained from Lund University Hospital.

(101) Female SCID-Beige mice (7-8w) from Taconic's Denmark were used in the experiments. All experiments were done by approval of Malmö/Lund ethical committee.

(102) HT-29 colon cancer were obtained from ATCC and cultivated according to ATCC recommendations. The HT-29 cell line growing in log phase was injected subcutaneously (4×10.sup.6 cells in 200 μL at day 0 (DO)). Human PBMC (7×10.sup.6 in 100 μL) isolated from leukocyte concentrates was injected intraperitoneally at the same day. Intraperitoneal treatments (100 μg) were done at days 6, 13, and 20.

(103) Tumour was measured with a calliper in width, length and height of which the tumour volume calculated (w/2×l/2×h/2×pi×(4/3)). The animals were terminated before the tumour volume reached 2 cm.sup.3, at wounding, or affected health of the mice.

(104) The data were analyzed by Mann-Whitney test using the GraphPad Prism program.

(105) Results

(106) Pooled data from mice engrafted with 4 different donors demonstrated statistically significant anti-tumour efficacy at days 12-16 in the form of inhibition of tumour growth when treated with the 1630/1631 antibody (p=0.0675 to p=0.0132, Mann-Whitney non parametric, 2-tail) in comparison to the vehicle group. The percentage of tumour volume inhibition ranged from 29-42% with 1630/1631 between days 10 and 21 (see FIG. 6 and Table 9).

(107) In conclusion, the anti-tumour effect of 1630/1631 was investigated using hPBMC humanized immunodeficient mice and subcutaneous tumour models of HT-29 colon carcinoma. 1630/1631 demonstrated statistically significant tumour volume inhibition.

(108) TABLE-US-00028 TABLE 9 Statistical analysis and percent tumour inhibition Tumour growth inhibition Day after tumour (tumour volume) compared p-value inoculation to vehicle (%) (Mann-Whitney 2-tail) D 12 42.1 0.0132 D 14 32.6 0.0675 D 16 38.7 0.0304 D 19 32.7 0.1918 D 21 29.5 0.0911

Example 12—Optimization of CD137 Parental Antibody Clone 1630/1631

(109) The aim of the optimization was to generate improved variants of the 1630/1631 antibody with regard to affinity and biophysical properties. Phage selections towards recombinant CD137 coated onto the surface of beads were performed and prior to each selection round, the phage stocks were pre-selected towards non-target proteins as well as beads. Prior to the fourth round of selection a thermal incubation step at 65° C. were performed. Overall the selection strategy was designed to promote the isolation of clones with a slow off-rate as well as a fast on-rate by prolonging the washing steps and decreasing the incubation time between the phage pool and CD137.

(110) After phage selections, screening was performed in a soluble scFv format to identify target binding clones as well as to evaluate the diversity. An extended primary screening was performed to identify clones with improved temperature stability, cynomolgus reactivity as well as affinity or off-rate. A total of 50 clones were re-cloned into the final IgG4 format having the S228P stabilizing mutation. Further evaluation of optimized variants was performed in the final format and was focused on binding in an ELISA set-up, cell-binding as determined by FACS, affinity, temperature stability as determined by both an ELISA set-up as well as DSF, SE-HPLC, Schrödinger modelling and specificity.

Example 13—Improved Stability of Clone 2674/2675

(111) The aim with the DSF analysis was to determine the Tm of the clone 2674/2675 compared to the parental clone 1630/1631 to evaluate the improvement in temperature stability after optimization.

(112) Material and Method

(113) All antibodies were analyzed with differential scanning fluorometry (DSF) at SARomics Biostructure. Samples were diluted to 0.1 mg/ml in sterile filtered PBS and a volume of 150 μl were delivered to SARomics.

(114) The samples for the DSF measurements (0.1 mg/ml in PBS buffer) were made up of 63 μl sample+7 μl PBS buffer, 1:100 fold diluted SYPRO Orange). In total, the SYPRO Orange was diluted 1:1000-fold. Duplicate measurements were made for each construct using a Stratagene MX3000P, qPCR machine. Measurements were performed in the temperature range 25° C. −95° C. The average melting temperature Tm was calculated for all samples.

(115) Results and Conclusions

(116) Melting curves for all samples were obtained and the determined Tm1 as well as Tm2 for 2674/2675 and parental clone 1630/1631 can be seen in Table 10 below. 2674/2675 had an improved Tm2 by 1-2° C. as compared to the parental clone 1630/1631.

(117) TABLE-US-00029 TABLE 10 Determined Tm1 as well as Tm2 values and the average Tm2 difference of 2674/2675 compared to 1630/1631 parental clone as measured by DSF Tm1 (° C.) Tm2 (° C.) Improved Tm2 (° C.) 2674 66.3 73.4 2 1630 66.3 71.4 —

Example 14—Reduced Aggregation Propensity of Clone 2674/2675 Analyzed by Antibody Aggregation Prediction at SchröDinger

(118) The aim with the Schrödinger analysis of the optimized variants was to evaluate the size of the hydrophobic patch as well as the aggregation propensity of 2674/2675 in comparison to the parental clone 1630.

(119) Material and Methods

(120) Sequences for the different variants were sent to Schrödinger and 3D structures were generated. The 3D models were analyzed with Protein Surface Analyzer and ranked with Aggscore. REF9-24 are clones obtained during optimization based on binding capacity for human CD137.

(121) Results and Conclusions

(122) The defined patch size and aggscore for the clones 2674/2675 and REF9-24 (clones obtained during selection) as well as parental clone 1630/1631 can be seen in FIG. 7. It can be concluded that the introduced mutations clearly disrupted the hydrophobic patch and reduced the aggregation propensity according to the modelling analysis.

Example 15—Binding of 2674/2675 to Human and Cynomolgus CD137 Measured by ELISA

(123) Aim and background The aim of the evaluation was to determine binding of 2674/2675 compared with parental clone 1630/1631 in ELISA to both human and cynomolgus CD137.

(124) Material and Methods

(125) Binding of CD137 antibodies to recombinant human CD137 was determined by sandwich ELISA. Briefly, ELISA plates (Greiner #655074) coated with recombinant human CD137-Fc (R&D #838-4B) were incubated with serial dilutions of the various CD137 antibodies to be investigated. CD137 antibodies were detected using HRP-conjugated goat-anti-human kappa light chain (AbD Serotec #STAR127P) and developed with SuperSignal ELISA Pico Chemiluminescent substrate (Pierce #37069). EC50 values of the various antibodies were determined in 2-6 separate experiments.

(126) Results and Conclusions

(127) 2674/2675 exhibits EC50 values in a similar range as parental clone 1630/1631, i.e. sub nM. Data is summarized in Table 11 below.

(128) TABLE-US-00030 TABLE 11 Determined EC50 values as measured by ELISA for 2674/2675 and parental clone 1630/1631. EC50 (nM) human EC50 (nM) cyno CD137 CD137 2674/2675 0.34 0.57 1630/1631 0.41 0.85

Example 16—Binding of 2674/2675 to Human and Cynomolgus CD137, Octet

(129) Aim

(130) The aim was to compare relative binding affinities for 2674/2675 and 1630/1631 to human and cynomolgus CD137 using the Octet platform.

(131) Materials and method CD137 affinity was determined using the Octet Red 96 platform (ForteBio). 2674/2675, 1630/1631, REF1, REF 2 and 1188 isotype control were coupled at 10 μg/ml to ARG2 biosensors (ForteBio #18-5092) by amine coupling with EDC and NHS. 7 2-fold serial dilutions of CD137 (Acro Biosystems #41B-H5227 and #41B-C52H4) from 100 nM was prepared in 1× kinetic buffer (ForteBio #18-1092). Association was measured for 180 s followed by dissociation for 180 s in 1× kinetic buffer. 10 mM Glycine pH 2.2 was used for regeneration.

(132) Data generated was referenced by reference well subtraction (1188), the baseline was aligned with the y-axis, inter-step correlation by alignment against association was performed and the data was smoothed by a Savitzky-Golay filtering in the data analysis software (v.9.0.0.14). The processed data was fitted using a 1:1 Langmuir binding model with X.sup.2 as a measurement of fitting accuracy.

(133) Results and Conclusion

(134) The binding affinities of 2674/2675, 1630/1631 and REF antibodies to human and cynomolgus CD137 are presented in Table 12. The affinity for 2674/2675 to human CD137 was improved by a factor 2 compared to 1630/1631. The affinity for 2674/2675 to cynomolgus CD137 was in the same range as 1630/1631.

(135) TABLE-US-00031 TABLE 12 Affinity for human and cynomolgus CD137 to immobilized 2674/2675, 1630/1631 and REF antibodies Human CD137 Cynomolgus CD137 Immobilized KD kon koff KD kon koff antibodies (M) (1/Ms) (1/s) (M) (1/Ms) (1/s) 2674/2675 6.9E−09 3.5E+05 2.4E−03 2.5E−08 1.6E+05 4.0E−03 1630/1631 1.4E−08 1.6E+05 2.3E−03 1.8E−08 1.4E+05 2.5E−03 REF1 3.0E−09 4.6E+05 1.4E−03 no binding REF 2 6.1E−09 6.8E+05 4.1E−03 1.0E−08 5.2E+05 5.3E−03

Example 17—Binding Affinity of 2674/2675 to Human FcγRs

(136) Aim

(137) The aim was to determine relative binding affinities for 2674/2675 to human FcγRs using the Octet platform.

(138) Materials and Methods

(139) FcγR affinity was determined using the Octet RED96 platform equipped with Anti-Human Fab-CH1 (FAB2G) sensor tips (ForteBio). Antibodies were diluted to 200 nM in 1× Kinetics Buffer (ForteBio) and loaded to a set of 8 parallel sensors for 300 seconds to reach an immobilization response of >1.5 nm. The immobilized antibodies were then assayed against 7 2-fold dilutions of FcγRs, starting at 100 nM. One immobilized sensor was assayed against 1× Kinetics Buffer for referencing and the entire assay was repeated without immobilization of antibodies to allow for double referencing. FcγRs included were obtained from R&D Systems (human FcγRI, #1257-FC-050; human FcγRIIa, #1330-CD-050; human FcγRIIb, #1460-CD-050; human FcγRIIIa (V158), #4325-FC-050; human FcγRIIIa (F158), #8894-FC-050). Binding to FcγRs was carried out for 60 seconds, followed by dissociation for 60 seconds in 1× Kinetics Buffer and regeneration of sensor tips using 10 mM glycine, pH 1.7. Data generated was referenced by standard double referencing, the baseline was aligned with the y-axis, inter-step correlation by alignment against dissociation was performed and the data was smoothed by a Savitzky-Golay filtering in the data analysis software (v.9.0.0.14). The processed data was fitted using a 1:1 Langmuir binding model with X.sup.2 as a measurement of fitting accuracy. To improve curve fitting quality of dissociation curves generated against FcγRs with very fast dissociation rates, only the initial 10 seconds of the dissociation curves were included in the curve fitting.

(140) Results and Conclusions

(141) The binding affinities human FcγRs of 2674/2675 and REF antibodies are presented in Table 13. 2674/2675 has a stronger binding to human FcγRI than all other assayed Fc receptors, as expected of an IgG4 antibody binding to the high affinity receptor FcγRI. 2674/2675 has a comparable binding to human FcγRs as REF1 antibody.

(142) TABLE-US-00032 TABLE 13 Determined binding affinities, KD (M), for 2674/2675 and REF antibodies to human FcγRs FcγRIIIa FcγRIIIa KD (M) FcγRI FcγRIIa FcγRIIb 176V 176F 2674/2675 2.11E−09 1.00E−06 7.79E−07 <det. limit <det. limit REF1 1.88E−09 6.74E−07 6.23E−07 <det. limit <det. limit REF2 <det. limit 5.87E−07 2.70E−06 <det. limit <det. limit

Example 18—Binding of 2674/2675 to Human and Cynomolgus CD137 Measured by FACS

(143) The aim of this study was to determine the binding to human and cynomolgus CD137.

(144) Material and Methods

(145) Binding and EC50 was determined using flow cytometric of CHO cells transfected with human CD137, cynomolgus CD137 or empty vector. The extracellular part of human or cynomolgus CD137 was fused to the transmembrane and intracellular part of human CD40 and cloned into pcDNA3.1. The vector was subsequently stably transfected into CHO cells. Expression of CD137 was confirmed by flow cytometry using CD137 antibody (human CD137-PE, BD Biosciences #555956) for 30 min at 4° C. CD137-transfected and empty vector-transfected cells were incubated with CD137 antibodies for at least 1 h at 4° C. to saturate the binding. In order to minimize antibody internalization, 0.05% sodium azide was used in the incubation buffer and all work was performed on ice. The CD137 antibodies were detected using PE-conjugated anti-hIgG antibody (109-115-098, Jackson Immunoresearch laboratories), incubated for 30 min at 4° C. Directly after staining the cells were fixed with a paraformaldehyde solution (10× concentrate BD CellFIX, BD biosciences #340181). Cells were analyzed by flow cytometry using FACSVerse (BD Biosciences). The median fluorescence intensity (MFI) for each sample was determined and the dose response data was analysed using Graph Pad Prism.

(146) MFI data was normalized for each antibody, where 0% is defined as the lowest value and 100% is the highest value in the dose titration for each antibody. EC50 and 95% confidence interval were calculated with Graph Pad Prism based on data from the two experiments (non-linear regression (curve fit), constraints set to 0 and 100).

(147) Results and Conclusion

(148) 2674/2675 have a comparable binding to human CD137 as the parental clone 1630 and the REF1 and REF2 CD137 mAbs FIG. 8 and Table 14. 2674/2675 and 1630/1631 have comparable binding to cynomolgus CD137 while REF1 does not bind to cynomolgus CD137 at all.

(149) TABLE-US-00033 TABLE 14 95% confidence intervals for the EC50 of each CD137 antibody determined as an average from 3 experiments of normalized data. Hu CD137 Cy CD137 EC50 (nM) EC50 (nM) 95% conf interval 95% conf interval 2674/2675 0.26-0.37 0.46-0.77 1630/1631 0.23-0.34 0.55-0.87 REF1 0.33-0.73 n.d. REF2 0.16-0.27 0.41-0.56 Nd: Not detectable

Example 19—CD137 Reporter Assay with FcγR Expressing Cells for Crosslink of CD137 mAb

(150) Aim and Background

(151) Functional evaluation of 2674/2675 with the parental clone 1630/1631 in the CD137 reporter assay when crosslinking CD137 mAbs with FcγR transfected CHO cells.

(152) Materials and Methods

(153) CHO-cells transfected with human FcγRI, FcγRIIa R131, FcγRIIb or empty vector (pcDNA3.1) were used for crosslinking. FcγR genes were cloned into pcDNA3.1. The vector was subsequently stably transfected into CHO cells. The expression of FcγRs was confirmed by staining with commercial antibody targeting CD32 or CD64.

(154) Agonistic function of the CD137 mAbs was evaluated using a CD137 reporter assay (Promega, CD137 Bioassay Kit CS196005). The assay was performed according to the manufacturer's protocol. In brief, FcγR transfected CHO cells and titrating concentrations of CD137 mAbs were diluted in RPMI containing 10% FCS and added to the assay plates before the addition of CD137 (Jurkat/CD137 cells) reporter cells. The assay plate was incubated for 6 h at 37° C. until addition of Bio-Glo™ Luciferase Assay Detection and plate read in the BMG reader.

(155) Results and Conclusions

(156) Crosslinking of the CD137 mAbs in the CD137 reporter assay with FcγRI transfected CHO cells demonstrates that 2674/2675 as well as the parental clone 1630/1631 induces a CD137 dependent activation of NF-κB in the reporter cell line (FIG. 9). It can be concluded that if the CD137 antibody is crosslinking dependent, binding affinities to FcγRI, FcγRIIa R131 and FcγRIIb (shown in previous example) correlate well with the agonistic activity induced in the CD137 reporter assay after FcγR crosslinking (FIG. 10). FcγR cross-linking independent activation of REF1 but not of 2674/2675, 1630/1631 or REF2 was determined using CHO cells transfected with empty vector (FIG. 11).

Example 20—CD8+ T Cell Agonist Assay with FcγRI Expressing CHO Cells for Crosslinking of CD137 Antibodies

(157) Aim and Background

(158) Functional evaluation of 2674/2675 with the parental clone 1630/1631 in a CD8.sup.+ T cell agonist assay when crosslinking the CD137 mAbs with FcγRI expressing cells.

(159) Materials and Methods

(160) CHO-cells transfected with human FcγRI were used for crosslinking. The FcγRI gene were cloned into pcDNA3.1. The vector was subsequently stably transfected into CHO cells. The expression of FcγRI was confirmed by staining with commercial antibody targeting CD64.

(161) Agonistic activity of CD137 antibodies was evaluated in a T cell assay based on primary human CD8.sup.+ T cells. Briefly, CD8.sup.+ T cells were separated from human peripheral blood mononuclear cells by MACS separation (Miltenyi #130-096-495) according to the manufacturer's protocol. Cells were incubated in 96-well microtiter plates (NuncThermo Scientific #268200) pre-incubated with CHO cells transfected with FcγRI and incubated with tosyl beads coated with anti-CD3 antibody (clone OKT3, Affymetrix eBioscience #16-0037) and titrated concentrations of the CD137 antibody to be tested. Following 72-hour incubation, culture medium was harvested and IFN-γ levels were determined by ELISA (BD #555142).

(162) Each clone was analyzed in at least 5 donors and compared to the reference CD137 antibody REF2. Due to large intra-donor variations, IFN-γ levels were normalized with 2674/2675 within each donor for comparison.

(163) Results and Conclusions

(164) Crosslinking with FcγRI expressing CHO cells in the CD8 T cell agonist assay of the CD137 mAbs 2674/2675 and the parental clone 1630/1631, but not REF2, induces T cell activation, measured as an increase in IFN-γ production after 72 h (FIG. 12). IgG binding affinities to FcγRs of 2674/2675, 1630/1631 and REF2 have been determined and was shown in a previous example. It can be concluded that IgG binding affinity to FcγR correlate with the agonistic activity induced in the CD8+ T cells.

Example 21—In Vivo Anti-Tumour Effect in HT-29 Colon Cancer Model

(165) Aim

(166) The anti-tumour effect of 2674/2675 was investigated using hPBMC humanized immunodeficient mice and subcutaneous tumour models of HT-29 colon carcinoma.

(167) Material and Methods

(168) Leukocyte concentrates were obtained from Lund University Hospital. Female SCID-Beige mice (7-8w) from Taconic's Denmark were used in the experiments. All experiments were done by approval of Malmö/Lund ethical committee.

(169) HT-29 colon cancer was obtained from ATCC and cultivated according to ATCC recommendations. The HT-29 cell line growing in log phase was injected subcutaneously (4×10.sup.6 cells in 200 μL at day 0 (D0)). Human PBMC (10×10.sup.6 in 100 μL) isolated from leukocyte concentrates was injected intraperitoneally at the same day. Intraperitoneal treatments (100 μg) were done twice weekly for three weeks starting at day 7.

(170) Tumour was measured with a calliper in width, length and height of which the tumour volume calculated (w/2×l/2×h/2×pi×(4/3)). The animals were terminated before the tumour volume reached 2 cm.sup.3, at wounding, or affected health of the mice.

(171) Results and Conclusion

(172) 2674/2675 demonstrated anti-tumor efficacy in humanized mouse models in comparison to the vehicle group. The percentage of tumour volume inhibition ranged from 0-35% with 2674/2675 between days 19 and 28 (Table 15).

(173) In conclusion, the anti-tumour effect of 2674/2675 was investigated using hPBMC humanized immunodeficient mice and subcutaneous tumour models of HT-29 colon carcinoma. 2674/2675 demonstrated tumour volume inhibition.

(174) TABLE-US-00034 TABLE 15 Percent tumour inhibition Tumour growth inhibition (tumour volume) compared Day after tumour inoculation to vehicle (%) D 19 9.4 D 21 24.0 D 24 27.4 D 26 24.7 D 28 35.1

Example 22—Gene Expression Analyses of FcγR and CD137 Co Expression in Human Tumor Tissue

(175) Aim

(176) Assessing the gene expression of various Fcγ receptors, as well as CD137, in a wide range of human cancers using a curated and quality-controlled database of microarray and RNA-seq datasets.

(177) Methods

(178) Mean expression values for FcγRI, FcγRIIa, FcγRIIb, FcγRIIc, FcγRIIIa, FcγRIIIb and CD137 were obtained by performing gene expression profiling using Genevestigator, a curated and quality-controlled database of RNA microarray studies for human tissues (Hruz et al. 2008, Adv Bioinformatics 2008: 420747, the disclosures of which are incorporated herein by reference). Correlation plots were obtained by plotting the mean expression values of the various Fcγ receptor genes versus CD137 (FIG. 13). Cancers displaying an above average expression (mean expression level ≥0) of both Fcγ receptor and CD137 were identified and the top 10 solid tumor types and hematological malignancies are presented in Table 16 and Table 17, respectively.

(179) Results and Conclusions

(180) Several human tumors with a high expression of Fcγ receptors and an above average expression of CD137 were identified using this method. The tables below provide an example of indications that could be highly sensitive to the antibody defined in this invention. This approach could be used to identify patient cohorts or individual patients that may benefit from treatment with agonistic CD137 antibodies. In fact, this type of approach could be used to identify patients on an individual level that may benefit from the treatment. One could also envisage other methods being used to molecularly characterize the tumor, such as next generation sequencing or methods based on protein analysis such as immunohistochemistry, flow cytometry or proteomic approaches.

(181) TABLE-US-00035 TABLE 16 Mean expression values of solid human tumors with an above average expression (mean expression level ≥10) of both Fcγ receptor and CD137 (TNFRSF9), as identified in FIG. 13. The ten tumors with the highest expression of the six Fcγ receptors are shown. Cancers with cells expressing FcγRIA and FcγRIA (Mean TNFRSF9 (Mean TNFRSF9 expression level) expression level) fibrous histiocytoma, malignant, metastatic 13.43 10.22 adenosquamous carcinoma 12.78 10.80 undifferentiated sarcoma 12.67 10.16 clear cell adenocarcinoma, NOS, 12.66 10.62 metastatic acinar cell carcinoma 12.47 10.82 dedifferentiated liposarcoma 12.45 10.67 renal cell carcinoma, unstated behavior 12.44 10.38 intraductal micropapillary carcinoma 12.36 10.68 fibrous histiocytoma, malignant 12.35 11.05 large cell neuroendocrine carcinoma 12.35 10.29 Cancers with cells expressing FcγRIIA and FcγRIIA (Mean TNFRSF9 (Mean TNFRSF9 expression level) expression level) Langerhans-cell histiocytosis, unifocal 14.58 13.85 acinar cell carcinoma 13.55 10.82 fibrous histiocytoma, malignant, metastatic 13.54 10.22 undifferentiated sarcoma 13.48 10.16 adenocarcinoma with mixed subtypes 13.43 10.77 dedifferentiated liposarcoma 13.32 10.67 undifferentiated sarcoma 13.22 10.52 fibrous histiocytoma, malignant 13.10 11.05 adenosquamous carcinoma 13.01 10.80 dedifferentiated liposarcoma 12.95 10.36 Cancers with cells expressing FcγRIIB and FcγRIIB (Mean TNFRSF9 (Mean TNFRSF9 expression level) expression level) Langerhans-cell histiocytosis, unifocal 13.23 13.85 fibrous histiocytoma, malignant, metastatic 12.64 10.22 adenocarcinoma with mixed subtypes 12.42 10.77 adenosquamous carcinoma 12.22 10.80 acinar cell carcinoma 12.19 10.82 Hodgkin's disease, NOS 12.01 10.88 renal cell carcinoma, unstated behavior 12.00 10.38 dedifferentiated liposarcoma 11.98 10.67 papillary adenocarcinoma, NOS 11.81 10.02 undifferentiated sarcoma 11.76 10.16 Cancers with cells expressing FcγRIIC and FcγRIIC (Mean TNFRSF9 (Mean TNFRSF9 expression level) expression level) Langerhans-cell histiocytosis, unifocal 12.72 13.85 Cancers with cells expressing FcγRIIIA FcγRIIIA (Mean TNFRSF9 (Mean and TNFRSF9 expression level) expression level) fibrous histiocytoma, malignant, metastatic 14.14 10.22 acinar cell carcinoma 13.14 10.82 large cell neuroendocrine carcinoma 12.97 10.29 adenocarcinoma with mixed subtypes 12.91 10.77 renal cell carcinoma, unstated behavior 12.85 10.38 intraductal micropapillary carcinoma 12.83 10.68 carcinoma, NOS 12.70 10.19 carcinoma, NOS 12.68 10.06 undifferentiated sarcoma 12.52 10.16 carcinoma, NOS 12.42 10.10 Cancers with cells expressing FcγRIIIB and FcγRIIIB (Mean TNFRSF9 (Mean TNFRSF9 expression level) expression level) carcinoma, NOS 13.09 10.19 carcinoma, NOS 12.66 10.06 carcinoma, NOS 12.61 10.10 tubular adenocarcinoma 12.06 10.18 carcinoma, NOS, micro-dissected 12.00 10.92 adenocarcinoma, intestinal type 11.97 10.26 fibrous histiocytoma, malignant, metastatic 11.89 10.22 neoplasm, malignant 11.81 10.05 adenocarcinoma, NOS 11.74 10.06 renal cell carcinoma, unstated behavior 11.61 10.38

(182) TABLE-US-00036 TABLE 17 Mean expression values of hematological malignancies with an above average expression (mean expression level ≥10) of both Fcγ receptor and CD137, as identified in FIG. 13. The ten malignancies with the highest expression of the six Fcγ receptors are shown. Cancers with cells expressing FcγRIA and FcγRIA (Mean TNFRSF9 (Mean TNFRSF9 expression level) expression level) (extranodal) NK/T-cell lymphoma, nasal and nasal-type 13.97 11.56 Hodgkin's disease, NOS 13.65 10.88 malignant lymphoma, large B-cell, diffuse 12.15 12.23 primary mediastinal B-cell lymphoma 12.13 12.43 anaplastic large cell lymphoma, T-cell and Null cell type 12.11 11.38 (ALCL), unstated behavior angioimmunoblastic T-cell lymphoma 11.93 13.85 mature T-cell lymphoma, NOS 11.86 12.53 mature T-cell lymphoma, NOS, unstated behavior 11.69 11.72 anaplastic large cell lymphoma, T-cell and Null cell type 11.67 11.43 (ALCL) angioimmunoblastic T-cell lymphoma, unstated behavior 11.63 12.98 Cancers with cells expressing FcγRIIA and FcγRIIA (Mean TNFRSF9 (Mean TNFRSF9 expression level) expression level) mycosis fungoides 13.27 11.34 Hodgkin's disease, NOS 12.33 10.88 (extranodal) NK/T-cell lymphoma, nasal and nasal-type 12.26 11.56 mature T-cell lymphoma, NOS 11.76 12.53 angioimmunoblastic T-cell lymphoma 11.69 13.85 malignant lymphoma, large B-cell, diffuse 11.37 12.23 angioimmunoblastic T-cell lymphoma, unstated behavior 11.34 12.98 mature T-cell lymphoma, NOS, unstated behavior 11.33 11.72 adult T-cell leukemia/lymphoma (HTLV-1 positive), 11.28 11.13 unstated behavior anaplastic large cell lymphoma, T-cell and Null cell type 11.23 11.38 (ALCL), unstated behavior Cancers with cells expressing FcγRIIB and FcγRIIB (Mean TNFRSF9 (Mean TNFRSF9 expression level) expression level) chronic lymphocytic B-cell leukemia, unstated behavior, 14.69 10.46 micro-dissected mantle cell lymphoma 14.06 10.11 (extranodal) marginal zone B-cell lymphoma, NOS 12.73 11.85 malignant lymphoma, nodular, NOS, unstated behavior 12.69 12.97 malignant lymphoma, nodular, NOS 12.16 12.02 Hodgkin's disease, NOS 12.01 10.88 malignant lymphoma, large B-cell, diffuse 11.97 12.23 angioimmunoblastic T-cell lymphoma 11.89 13.85 mycosis fungoides 11.76 11.34 Hodgkin's disease, NOS, micro-dissected 11.38 11.32 Cancers with cells expressing FcγRIIC and FcγRIIC (Mean TNFRSF9 (Mean TNFRSF9 expression level) expression level) juvenile myelomonocytic leukemia 10.00 10.00 Cancers with cells expressing FcγRIIIA and FcγRIIIA (Mean TNFRSF9 (Mean TNFRSF9 expression level) expression level) juvenile myelomonocytic leukemia 13.19 10.00 (extranodal) NK/T-cell lymphoma, nasal and nasal-type 12.09 11.56 mycosis fungoides 11.71 11.34 mature T-cell lymphoma, NOS 11.16 12.53 anaplastic large cell lymphoma, T-cell and Null cell type 10.80 11.43 (ALCL) angioimmunoblastic T-cell lymphoma 10.65 13.85 Hodgkin's disease, NOS 10.51 10.88 anaplastic large cell lymphoma, T-cell and Null cell type 10.38 11.38 (ALCL), unstated behavior malignant lymphoma, large B-cell, diffuse 10.31 12.23 primary mediastinal B-cell lymphoma 10.26 12.43 Cancers with cells expressing FcγRIIIB and FcγRIIIB (Mean TNFRSF9 (Mean TNFRSF9 expression level) expression level) juvenile myelomonocytic leukemia 15.03 10.00 (extranodal) NK/T-cell lymphoma, nasal and nasal-type 11.95 11.56 anaplastic large cell lymphoma, T-cell and Null cell type 11.85 11.38 (ALCL), unstated behavior mature T-cell lymphoma, NOS 11.57 12.53 mature T-cell lymphoma, NOS, unstated behavior 11.56 11.72 angioimmunoblastic T-cell lymphoma, unstated behavior 11.54 12.98 primary mediastinal B-cell lymphoma 11.44 12.43 angioimmunoblastic T-cell lymphoma 11.30 13.85 adult T-cell leukemia/lymphoma (HTLV-1 positive), 11.08 11.13 unstated behavior (extranodal) marginal zone B-cell lymphoma, NOS 10.95 11.85

REFERENCES

(183) Almeida J, Bueno C, AlgueróM C et al. Comparative analysis of the morphological, cytochemical, immunophenotypical, and functional characteristics of normal human peripheral blood lineage(−)/CD16(+)/HLA−DR(+)/CD14(−/lo) cells, CD14(+) monocytes, and CD16(−) dendritic cells. Clin Immunol. 2001 September; 100(3):325-38. Akhmetzyanova, I., Zelinskyy, G., Littwitz-Salomon, E., Malyshkina, A., Dietze, K. K., Streeck, H., Brandau, S., and Dittmer, U. (2016) CD137 Agonist Therapy Can Reprogram Regulatory T Cells into Cytotoxic CD4+ T Cells with Antitumour Activity. J. Immunol. 196, 484-492. Ascierto, P. A., Simeone, E., Sznol, M., Fu, Y. X., and Melero, I. (2010) Clinical experiences with anti-CD137 and anti-PD1 therapeutic antibodies. Semin. Oncol. 37, 508-516. Baessler T, Charton J E, Schmiedel B J, Grünebach F, Krusch M, Wacker A, Rammensee H G, Salih H R. CD137 ligand mediates opposite effects in human and mouse NK cells and impairs NK-cell reactivity against human acute myeloid leukemia cells. Blood. 2010 Apr. 15; 115(15):3058-69. doi: 10.1182/blood-2009-06-227934. Erratum in: Blood. 2010 Dec. 23; 116(26):6152. PubMed PMID: 20008791. Bartkowiak, T. and Curran, M. A. (2015) 4-1BB Agonists: Multi-Potent Potentiators of Tumour Immunity. Front Oncol. 5, 117. Bronte V, Brandau S, Chen S H et al. Recommendations for myeloid-derived suppressor cell nomenclature and characterization standards. Nat Commun. 2016 Jul. 6; 7:12150. Bruhns P, lannascoli B, England P et al. Specificity and affinity of human Fcgamma receptors and their polymorphic variants for human IgG subclasses. Blood. 2009 Apr. 16; 113(16):3716-25. Bulliard Y, Jolicoeur R, Zhang J, Dranoff G, Wilson N S, Brogdon J L. OX40 engagement depletes intratumoural Tregs via activating FcγRs, leading to antitumour efficacy. Immunol Cell Biol. 2014 July; 92(6):475-80. doi: 10.1038/icb.2014.26. PubMed PMID: 24732076. Cavnar M J, Zeng S, Kim T S et al. KIT oncogene inhibition drives intratumoral macrophage M2 polarization. J Exp Med. 2013 Dec. 16; 210(13):2873-86. Cheeseman H M, Carias A M, Evans A B et al. Expression Profile of Human Fc Receptors in Mucosal Tissue: Implications for Antibody-Dependent Cellular Effector Functions Targeting HIV-1 Transmission. PLoS One. 2016 May 10; 11(5):e0154656. Curran, M. A., Kim, M., Montalvo, W., Al-Shamkhani, A., and Allison, J. P. (2011) Combination CTLA-4 blockade and 4-1BB activation enhances tumour rejection by increasing T-cell infiltration, proliferation, and cytokine production. PLoS. ONE. 6, e19499. Dubrot, J., Milheiro, F., Alfaro, C., Palazon, A., Martinez-Forero, I., Perez-Gracia, J. L., Morales-Kastresana, A., Romero-Trevejo, J. L., Ochoa, M. C., Hervas-Stubbs, S., Prieto, J., Jure-Kunkel, M., Chen, L., and Melero, I. (2010) Treatment with anti-CD137 mAbs causes intense accumulations of liver T cells without selective antitumour immunotherapeutic effects in this organ. Cancer Immunol. Immunother. 59, 1223-1233. Gauttier, V., Judor, J. P., Le, G., V, Cany, J., Ferry, N., and Conchon, S. (2014) Agonistic anti-CD137 antibody treatment leads to antitumour response in mice with liver cancer. Int. J. Cancer 135, 2857-2867. Gray, J. C., French, R. R., James, S., Al-Shamkhani, A., Johnson, P. W., and Glennie, M. J. (2008) Optimising anti-tumour CD8 T-cell responses using combinations of immunomodulatory antibodies. Eur. J. Immunol. 38, 2499-2511. Guilliams M, Bruhns P, Saeys Y et al. The function of Fcγ receptors in dendritic cells and macrophages. Nat Rev Immunol. 2014 February; 14(2):94-108. Guo, Z., Cheng, D., Xia, Z., Luan, M., Wu, L., Wang, G., and Zhang, S. (2013) Combined TIM-3 blockade and CD137 activation affords the long-term protection in a murine model of ovarian cancer. J. Transl. Med. 11, 215. Elliott L A, Doherty G A, Sheahan K et al. Human Tumor-Infiltrating Myeloid Cells: Phenotypic and Functional Diversity. Front Immunol. 2017 Feb. 6; 8:86 Eruslanov E B, Bhojnagarwala P S, Quatromoni J G et al. Tumor-associated neutrophils stimulate T cell responses in early-stage human lung cancer. J Clin Invest. 2014 December; 124(12):5466-80. Eruslanov E, Neuberger M, Daurkin I et al. Circulating and tumor-infiltrating myeloid cell subsets in patients with bladder cancer. Int J Cancer. 2012 Mar. 1; 130(5):1109-19. Griesinger A M, Birks D K, Donson A M et al. Characterization of distinct immunophenotypes across pediatric brain tumor types. J Immunol. 2013 Nov. 1; 191(9):4880-8. Grugan K D, McCabe F L, Kinder M et al. Tumor-associated macrophages promote invasion while retaining Fc-dependent anti-tumor function. J Immunol. 2012 Dec. 1; 189(11):5457-66. Hansen B D, Schmidt H, von der Maase H et al. Tumour-associated macrophages are related to progression in patients with metastatic melanoma following interleukin-2 based immunotherapy. Acta Oncol. 2006; 45(4):400-5. Hogarth P M, Pietersz G A. Fc receptor-targeted therapies for the treatment of inflammation, cancer and beyond. Nat Rev Drug Discov. 2012 Mar. 30; 11(4):311-31. Holbrook E. Kohrt, A. Dimitrios Colevas, Roch Houot, 1,2,3 Kipp Weiskopf, Matthew J. Goldstein, Peder Lund, Antonia Mueller, Idit Sagiv-Barfi, Aurelien Marabelle, Ruth Lira, Emily Troutner, Lori Richards, 1 Amanda Rajapaska, Jonathan Hebb, Cariad Chester, Erin Waller, Anton Ostashko, Wen-Kai Weng, Lieping Chen, Debra Czerwinski, Yang-Xin Fu, John Sunwoo, and Ronald Levy. Targeting CD137 enhances the efficacy of cetuximab. The Journal of Clinical Investigation Volume 124 Number 6 Jun. 2014 Horton H M, Bernett M J, Peipp M, Pong E, Karki S, Chu S Y, Richards J O, Chen H, Repp R, Desjarlais J R, Zhukovsky E A. Fc-engineered anti-CD40 antibody enhances multiple effector functions and exhibits potent in vitro and in vivo antitumour activity against hematologic malignancies. Blood. 2010 Oct. 21; 116(16):3004-12. doi: 10.1182/blood-2010-01-265280. PubMed PMID: 20616215. Hruz T, Laule O, Szabo G, Wessendorp F, Bleuler S, Oertle L, Widmayer P, Gruissem W, Zimmermann P: Genevestigator v3: a reference expression database for the meta-analysis of transcriptomes. Adv Bioinformatics 2008, 2008:420747. Hu W, Li X, Zhang C et al. Tumor-associated macrophages in cancers. Clin Transl Oncol. 2016 March; 18(3):251-8. Kim, J. A., Averbook, B. J., Chambers, K., Rothchild, K., Kjaergaard, J., Papay, R., and Shu, S. (2001) Divergent effects of 4-1BB antibodies on antitumour immunity and on tumour-reactive T-cell generation. Cancer Res 61, 2031-2037. Kwong, B., Gai, S. A., Elkhader, J., Wittrup, K. D., and Irvine, D. J. (2013) Localized immunotherapy via liposome-anchored Anti-CD137+IL-2 prevents lethal toxicity and elicits local and systemic antitumour immunity. Cancer Res. 73, 1547-1558. Lee, H. W., Park, S. J., Choi, B. K., Kim, H. H., Nam, K. O., and Kwon, B. S. (2002) 4-1BB promotes the survival of CD8+T lymphocytes by increasing expression of Bcl-xL and Bfl-1. J Immunol 169, 4882-4888. Lee, S. J., Myers, L., Muralimohan, G., Dai, J., Qiao, Y., Li, Z., Mittler, R. S., and Vella, A. T. (2004) 4-1BB and OX40 dual costimulation synergistically stimulate primary specific CD8 T cells for robust effector function. J. Immunol. 173, 3002-3012. Li Y, Lee P Y, Kellner E S et al. Monocyte surface expression of Fcgamma receptor RI (CD64), a biomarker reflecting type-I interferon levels in systemic lupus erythematosus. Arthritis Res Ther. 2010; 12(3):R90 Li C, Luo X, Lin Y et al. A Higher Frequency of CD14.sup.+ CD169+ Monocytes/Macrophages in Patients with Colorectal Cancer. PLoS One. 2015 Oct. 28; 10(10):e0141817. Li, F. and Ravetch, J. V. (2011) Inhibitory Fcgamma receptor engagement drives adjuvant and anti-tumour activities of agonistic CD40 antibodies. Science 333, 1030-1034. Lu J, Chu J, Zou Z et al. Structure of FcγRI in complex with Fc reveals the importance of glycan recognition for high-affinity IgG binding. Proc Natl Acad Sci USA. 2015 Jan. 20; 112(3):833-8 McMillin, D. W., Hewes, B., Gangadharan, B., Archer, D. R., Mittler, R. S., and Spencer, H. T. (2006) Complete regression of large solid tumours using engineered drug-resistant hematopoietic cells and anti-CD137 immunotherapy. Hum. Gene Ther 17, 798-806. Melero, I., Shuford, W. W., Newby, S. A., Aruffo, A., Ledbetter, J. A., Hellstrom, K. E., Mittler, R. S., and Chen, L. (1997) Monoclonal antibodies against the 4-1BB T-cell activation molecule eradicate established tumours. Nat Med 3, 682-685. Melero I, Daniel Hirschhorn-Cymerman, Aizea Morales-Kastresana, et al. Agonist Antibodies to TNFR Molecules That Costimulate T and NK cells Clin Cancer Res 2013; 19:1044-1053. Published online Mar. 3, 2013. Melero I, Antonio M. Grimaldi, Jose L. Perez-Gracia, et al. Clinical Development of Immunostimulatory Monoclonal Antibodies and Opportunities for Combination. Clin Cancer Res 2013; 19:997-1008. Miller, R. E., Jones, J., Le, T., Whitmore, J., Boiani, N., Gliniak, B., and Lynch, D. H. (2002) 4-1BB-specific monoclonal antibody promotes the generation of tumour-specific immune responses by direct activation of CD8 T cells in a CD40-dependent manner. J Immunol 169, 1792-1800. Morales-Kastresana, A., Sanmamed, M. F., Rodriguez, I., Palazon, A., Martinez-Forero, I., Labiano, S., Hervas-Stubbs, S., Sangro, B., Ochoa, C., Rouzaut, A., Azpilikueta, A., Bolanos, E., Jure-Kunkel, M., Gutgemann, I., and Melero, I. (2013) Combined immunostimulatory monoclonal antibodies extend survival in an aggressive transgenic hepatocellular carcinoma mouse model. Clin. Cancer Res. 19, 6151-6162. Morimura T, Neuchrist C, Kitz K et al. Monocyte subpopulations in human gliomas: expression of Fc and complement receptors and correlation with tumor proliferation. Acta Neuropathol. 1990; 80(3):287-94. Norton S E, Dunn E T, McCall J L et al. Gut macrophage phenotype is dependent on the tumor microenvironment in colorectal cancer. Clin Transl Immunology. 2016 Apr. 29; 5(4):e76. Niu, L., Strahotin, S., Hewes, B., Zhang, B., Zhang, Y., Archer, D., Spencer, T., Dillehay, D., Kwon, B., Chen, L., Vella, A. T., and Mittler, R. S. (2007) Cytokine-mediated disruption of lymphocyte trafficking, hemopoiesis, and induction of lymphopenia, anemia, and thrombocytopenia in anti-CD137-treated mice. J. Immunol. 178, 4194-4213. Overdijk M B, Verploegen S, Ortiz Buijsse A, Vink T, Leusen J H, Bleeker W K, Parren P W. Crosstalk between human IgG isotypes and murine effector cells. J Immunol. 2012 Oct. 1; 189(7):3430-8. PubMed PMID: 22956577. Palazón A, Iván Martïnez-Forero, Alvaro Teijeira, et al. The HIF-1a Hypoxia Response in Tumour-Infiltrating T Lymphocytes Induces Functional CD137 (4-1BB) for Immunotherapy Cancer Discovery 2012; 2:608-623. Published OnlineFirst Jun. 19, 2012. Palazón A, Teijeira A, Martïnez-Forero I, Hervás-Stubbs S, Roncal C, Peñuelas I, Dubrot J, Morales-Kastresana A, Perez-Gracia J L, Ochoa M C, Ochoa-Callejero L, Martinez A, Luque A, Dinchuk J, Rouzaut A, Jure-Kunkel M, Melero I. Agonist anti-CD137 mAb act on tumour endothelial cells to enhance recruitment of activated T lymphocytes. Cancer Res. 2011 Feb. 1; 71(3):801-11. doi:10.1158/0008-5472.CAN-10-1733. PubMed PMID: 21266358. Pan, P. Y., Zang, Y., Weber, K., Meseck, M. L., and Chen, S. H. (2002) OX40 ligation enhances primary and memory cytotoxic T lymphocyte responses in an immunotherapy for hepatic colon metastases. Mol Ther 6, 528-536. Porembka M R, Mitchem J B, Belt B A et al. Pancreatic adenocarcinoma induces bone marrow mobilization of myeloid-derived suppressor cells which promote primary tumor growth. Cancer Immunol Immunother. 2012 September; 61(9):1373-85. Pulle, G., Vidric, M., and Watts, T. H. (2006) IL-15-dependent induction of 4-1BB promotes antigen-independent CD8 memory T cell survival. J Immunol 176, 2739-2748. Rabu, C., Quemener, A., Jacques, Y., Echasserieau, K., Vusio, P., and Lang, F. (2005) Production of recombinant human trimeric CD137L (4-1BBL). Cross-linking is essential to its T cell co-stimulation activity. J Biol Chem 280, 41472-41481. Roussel M, Ferrell P B Jr, Greenplate A R et al. Mass cytometry deep phenotyping of human mononuclear phagocytes and myeloid-derived suppressor cells from human blood and bone marrow. J Leukoc Biol. 2017 August; 102(2):437-447 Sallin, M. A., Zhang, X., So, E. C., Burch, E., Cai, L., Lin, W., Chapoval, A. I., and Strome, S. E. (2014) The anti-lymphoma activities of anti-CD137 monoclonal antibodies are enhanced in FcgammaRIII(−/−) mice. Cancer Immunol. Immunother. 63, 947-958. Sanmamed, M. F., Pastor, F., Rodriguez, A., Perez-Gracia, J. L., Rodriguez-Ruiz, M. E., Jure-Kunkel, M., and Melero, I. (2015) Agonists of Co-stimulation in Cancer Immunotherapy Directed Against CD137, OX40, GITR, CD27, CD28, and ICOS. Semin. Oncol. 42, 640-655. Shuford, W. W., Klussman, K., Tritchler, D. D., Loo, D. T., Chalupny, J., Siadak, A. W., Brown, T. J., Emswiler, J., Raecho, H., Larsen, C. P., Pearson, T. C., Ledbetter, J. A., Aruffo, A., and Mittler, R. S. (1997) 4-1BB costimulatory signals preferentially induce CD8+ T cell proliferation and lead to the amplification in vivo of cytotoxic T cell responses. J Exp. Med 186, 47-55. So, T., Lee, S. W., and Croft, M. (2008) Immune regulation and control of regulatory T cells by OX40 and 4-1BB. Cytokine Growth Factor Rev. 19, 253-262. Solito S, Marigo I, Pinton L et al. Myeloid-derived suppressor cell heterogeneity in human cancers. Ann N Y Acad Sci. 2014 June; 1319:47-65. Stewart R, Hammond S, Oberst M, Wilkinson R. (2014) The role of Fc gamma receptors in the activity of immunomodulatory antibodies for cancer. J Immunother. 2:29 St Rose, M. C., Taylor, R. A., Bandyopadhyay, S., Qui, H. Z., Hagymasi, A. T., Vella, A. T., and Adler, A. J. (2013) CD134/CD137 dual costimulation-elicited IFN-gamma maximizes effector T-cell function but limits Treg expansion. Immunol. Cell Biol. 91, 173-183. Sun Y, Subudhi S K, Fu Y X. Co-stimulation agonists as a new immunotherapy for autoimmune diseases. Trends Mol Med. 2003 November; 9(11):483-9. Review. PubMed PMID: 14604826. Taraban, V. Y., Rowley, T. F., O'Brien, L., Chan, H. T., Haswell, L. E., Green, M. H., Tutt, A. L., Glennie, M. J., and Al-Shamkhani, A. (2002) Expression and costimulatory effects of the TNF receptor superfamily members CD134 (OX40) and CD137 (4-1BB), and their role in the generation of anti-tumour immune responses. Eur J Immunol 32, 3617-3627. Uno, T., Takeda, K., Kojima, Y., Yoshizawa, H., Akiba, H., Mittler, R. S., Gejyo, F., Okumura, K., Yagita, H., and Smyth, M. J. (2006) Eradication of established tumours in mice by a combination antibody-based therapy. Nat. Med. 12, 693-698. Vidarsson G, Dekkers G, Rispens T. IgG subclasses and allotypes: from structure to effector functions. Front Immunol. 2014 Oct. 20; 5:520. doi:10.3389/fimmu.2014.00520. Review. PubMed PMID: 25368619; PubMed Central PMCID: PMC4202688. Vinay, D. S. and Kwon, B. S. (2012) Immunotherapy of cancer with 4-1BB. Mol. Cancer Ther. 11, 1062-1070. Wang W, Erbe A K, Hank J A, Morris Z S, Sondel P M. NK Cell-Mediated Antibody-Dependent Cellular Cytotoxicity in Cancer Immunotherapy. Front Immunol. 2015 Jul. 27; 6:368. doi: 10.3389/fimmu.2015.00368. Review. PubMed PMID: 26284063; PubMed Central PMCID: PMC4515552. Wei, H., Zhao, L., Li, W., Fan, K., Qian, W., Hou, S., Wang, H., Dai, M., Hellstrom, I., Hellstrom, K. E., and Guo, Y. (2013) Combinatorial PD-1 blockade and CD137 activation has therapeutic efficacy in murine cancer models and synergizes with cisplatin. PLoS. ONE. 8, e84927. Westwood, J. A., Darcy, P. K., Guru, P. M., Sharkey, J., Pegram, H. J., Amos, S. M., Smyth, M. J., and Kershaw, M. H. (2010) Three agonist antibodies in combination with high-dose IL-2 eradicate orthotopic kidney cancer in mice. J. Transl. Med. 8, 42. Westwood, J. A., Matthews, G. M., Shortt, J., Faulkner, D., Pegram, H. J., Duong, C. P., Chesi, M., Bergsagel, P. L., Sharp, L. L., Huhn, R. D., Darcy, P. K., Johnstone, R. W., and Kershaw, M. H. (2014a) Combination anti-CD137 and anti-CD40 antibody therapy in murine myc-driven hematological cancers. Leuk. Res. 38, 948-954. White A L, Chan H T, French R R, Willoughby J, Mockridge C I, Roghanian A, Penfold C A, Booth S G, Dodhy A, Polak M E, Potter E A, Ardern-Jones M R, Verbeek J S, Johnson P W, Al-Shamkhani A, Cragg M S, Beers S A, Glennie M J. Conformation of the human immunoglobulin G2 hinge imparts superagonistic properties to immunostimulatory anticancer antibodies. Cancer Cell. 2015 Jan. 12; 27(1):138-48. doi:10.1016/j.ccell.2014.11.001. PubMed PMID: 25500122; PubMed Central PMCID:PMC4297290. Westwood, J. A., Potdevin Hunnam, T. C., Pegram, H. J., Hicks, R. J., Darcy, P. K., and Kershaw, M. H. (2014b) Routes of delivery for CpG and anti-CD137 for the treatment of orthotopic kidney tumours in mice. PLoS. ONE. 9, e95847. Wilcox, R. A., Flies, D. B., Zhu, G., Johnson, A. J., Tamada, K., Chapoval, A. I., Strome, S. E., Pease, L. R., and Chen, L. (2002) Provision of antigen and CD137 signaling breaks immunological ignorance, promoting regression of poorly immunogenic tumours. J Clin Invest 109, 651-659. Wilson, N. S., Yang, B., Yang, A., Loeser, S., Marsters, S., Lawrence, D., Li, Y., Pitti, R., Totpal, K., Yee, S., Ross, S., Vernes, J. M., Lu, Y., Adams, C., Offringa, R., Kelley, B., Hymowitz, S., Daniel, D., Meng, G., and Ashkenazi, A. (2011b) An Fcgamma receptor-dependent mechanism drives antibody-mediated target-receptor signaling in cancer cells. Cancer Cell 19, 101-113. Wyzgol, A., Muller, N., Fick, A., Munkel, S., Grigoleit, G. U., Pfizenmaier, K., and Wajant, H. (2009) Trimer stabilization, oligomerization, and antibody-mediated cell surface immobilization improve the activity of soluble trimers of CD27L, CD40L, 41BBL, and glucocorticoid-induced TNF receptor ligand. J Immunol 183, 1851-1861. Zhang, N., Sadun, R. E., Arias, R. S., Flanagan, M. L., Sachsman, S. M., Nien, Y. C., Khawli, L. A., Hu, P., and Epstein, A. L. (2007) Targeted and untargeted CD137L fusion proteins for the immunotherapy of experimental solid tumours. Clin. Cancer Res. 13, 2758-2767. Zhang B, Wang Z, Wu L et al. Circulating and tumor-infiltrating myeloid-derived suppressor cells in patients with colorectal carcinoma. PLoS One. 2013; 8(2):e57114