NOVEL ANTIBODIES AND NUCLEOTIDE SEQUENCES, AND USES THEREOF

Abstract

Described are novel anti-CTLA-4 antibody molecules and nucleotide sequences and expression vectors, such as viruses, encoding such antibody molecules. The novel antibody molecules are Treg depleting antibody molecules, and they have an improved depleting effect on CTLA-4 positive cells, such as Tregs, compared to ipilimumab. Described is also the use of such antibody molecules or nucleotide sequences or viruses in medicine, such as in the treatment of cancer, such as solid tumours.

Claims

1. An antibody molecule that specifically binds to CTLA-4, and has improved depleting effect on CTLA-4 positive cells compared to ipilimumab.

2. An antibody molecule according to claim 1, which has improved depleting effect on CD4 positive cells compared to ipilimumab and/or which has improved depleting effect on Tregs compared to ipilimumab.

3. (canceled)

4. An antibody molecule according to claim 1, which is considered as having improved depleting effect on CTLA-4 positive cells, CD4 positive cells and/or Tregs compared to ipilimumab if it gives an improved depletion in: (i) an in vitro ADCC test which is performed using an NK-92 cell line stably transfected to express the CD16-158V allele together with GFP, wherein the ADCC test comprises the following consecutive steps: 1) CTLA-4 positive cells, CD4 positive cells or Tregs as target cells are isolated from peripheral blood of healthy donors; 2) the target cells are then stimulated with CD3/CD28 and rhIL-2; 3) the target cells are then pre-incubated with the antibody molecule and are then mixed with NK cells; 4) the target cells are then incubated in RPMI 1640+GlutaMAX medium containing 1 HEPES buffer, sodium pyruvate and FBS low IgG; 5) lysis is determined by flow cytometry; 6) steps 1-5 are repeated, or performed in parallel, with ipilimumab used instead of the antibody molecule in step 3; and 7) the results of the lysis for the antibody molecule are compared to the results of the lysis for ipilimumab, and an improved lysis for the antibody molecule compared to ipilimumab demonstrates that the antibody molecule has improved depleting effect on CTLA-4 positive cells, CD4 positive cells and/or Tregs; and/or (ii) an in vivo test in a PBMC-NOG/SCID model, wherein the in vivo test comprises the following consecutive steps: 1) human PBMCs are isolated, washed and resuspended in sterile PBS; 2) NOG mice are injected i.v. with cell suspension from step 1); 3) the spleens from the NOG mice are isolated and rendered into a single cell suspension; 4) the cell suspension from step 3) is resuspended in sterile PBS; 5) SCID mice are injected i.p. with suspension from step 4; 6) the SCID mice are then treated with either the antibody molecule, ipilimumab or an isotype control monoclonal antibody; 7) the intraperitoneal fluid of the treated SCID mice is collected; 8) human T cell subsets are identified and quantified by FACS using following markers: CD45, CD4, CD8, CD25, CD127; 9) the results from identification and quantification of the T cell subsets from the mice treated with the antibody molecule is compared to the results from identification and quantification of the T cell subsets from the mice treated with ipilimumab and to the results from identification and quantification of the T cell subsets from the mice treated with isotype control monoclonal antibody, and a lower number of CTLA-4 positive cells, CD4 positive cells and/or Tregs in the intraperitoneal fluid from mice treated with the antibody molecule to be tested compared to the number of CTLA-4 positive cells, CD4 positive cells and/or Tregs in the intraperitoneal fluid from mice treated with ipilimumab demonstrates that the antibody molecule has improved depleting effect on CTLA-4 positive cells, CD4 positive cells and/or Tregs compared to ipilimumab.

5. An antibody molecule according to claim 1, wherein the antibody molecule is selected from the group consisting of antibody molecules comprising 1-6 of the CDRs selected from the group consisting of SEQ ID. Nos: 3, 6, 8, 10, 12 and 14.

6. An antibody molecule according to claim 1, wherein the antibody molecule is selected from the group consisting of antibody molecules comprising 1-6 of the CDRs VH-CDR1, VH-CDR2, VH-CDR3, VL-CDR1, and VL-CDR3, wherein VH-CDR1, if present, is selected from the group consisting of SEQ ID. Nos: 15, 22, 29 and 35; wherein VH-CDR2, if present, is selected from the group consisting of SEQ ID. Nos: 16, 23, 30, and 36; wherein VH-CDR3, if present, is selected from the group consisting of SEQ ID. Nos: 17, 24, 31 and 37; wherein VL-CDR1, if present, is selected from the group consisting of SEQ ID. Nos: 10 and 38; wherein VL-CDR2, if present, is selected from the group consisting of SEQ ID. Nos: 18, 25, 32 and 39; wherein VL-CDR3, if present, is selected from the group consisting of SEQ ID. Nos: 19, 26 and 40.

7. An antibody molecule according to claim 1, wherein the antibody molecule comprises the 6 CDRs having SEQ ID. NOs: 15, 16, 17, 10, 18 and 19 or the 6 CDRs having SEQ ID. NOs: 22, 23, 24, 10, 25 and 26.

8. (canceled)

9. An antibody molecule according to claim 1, wherein the antibody molecule comprises a variable heavy chain selected from the group consisting of SEQ. ID. NOs: 20 and 27 and/or a variable light chain selected from the group consisting of SEQ. ID. NOs: 21 and 28, or wherein the antibody molecule comprises the heavy chain constant region SEQ ID NO: 43 and/or the light chain constant region SEQ ID NO: 44.

10. (canceled)

11. An antibody molecule according to claim 1, wherein the antibody molecule is an antibody molecule that is capable of competing for binding to CTLA-4 with an antibody molecule that comprises the 6 CDRs having SEQ ID. NOs: 15, 16, 17, 10, 18 and 19 or the 6 CDRs having SEQ ID. NOs: 22, 23, 24, 10, 25 and 26.

12. An antibody molecule according to claim 1, wherein the antibody molecule is selected from the group consisting of a monoclonal antibody, a full-size antibody, a chimeric antibody, a single chain antibody, a Fab, a Fv, an scFv, a Fab′, and a (Fab′).sub.2.

13. An antibody molecule according to claim 1, which binds to human CTLA-4 (hCTLA-4) and/or to cynomolgus monkey CTLA-4 (cmCTLA-4) and/or to murine CTLA-4 (mCTLA-4), or which does not bind to human CD28.

14. (canceled)

15. An antibody molecule according to claim 1, wherein the antibody molecule is selected from the group consisting of a human IgG antibody, such as a human IgG1 antibody, a humanized IgG antibody and an IgG antibody of human origin.

16-17. (canceled)

18. An isolated nucleotide sequence encoding an antibody molecule as defined in claim 1.

19. An isolated nucleotide sequence according to claim 18, comprising or consisting of a sequence selected from the group consisting of SEQ ID 45-52.

20. A plasmid, virus or cell comprising a nucleotide sequence as defined in claim 18.

21. (canceled)

22. A plasmid, virus or cell according to claim 20, which is an oncolytic virus and preferably an oncolytic poxvirus.

23. A plasmid, virus or cell to claim 22, wherein said poxvirus belongs to the Chordopoxviridae subfamily, more preferably to the Orthopoxvirus genus preferably selected from the group consisting of Vaccinia virus, cowpox virus, canarypox virus, ectromelia virus and myxoma virus.

24. A plasmid, virus or cell to claim 23, wherein said oncolytic virus is a vaccinia virus defective for both thymidine kinase (TK) and/or ribonucleotide reductase (RR) activities and comprising nucleotide sequences encoding SEQ. ID. NO: 20 and ID. NO: 21 or SEQ. ID. NO: 53 and ID. NO: 54.

25. A plasmid, virus or cell according to claim 24, wherein said oncolytic vaccinia virus further comprises a nucleotide sequence encoding a GM-CSF, with a specific preference for a human GM-CSF (e.g. having SEQ ID NO: 55 or SEQ ID NO: 56) or a murine GM-CSF (e.g. having SEQ ID NO: 57 or SEQ ID NO: 58).

26. A plasmid, virus or cell according to claim 20, wherein the cassette encoding the heavy chain is inserted at the J2R locus and the cassette encoding the light chain is inserted at the I4L locus.

27-30. (canceled)

31. A pharmaceutical composition comprising or consisting of antibody molecule as defined in in claim 1, and optionally a pharmaceutically acceptable diluent, carrier, vehicle and/or excipient.

32. A pharmaceutical composition comprising the nucleotide sequence according to claim 18, and optionally a pharmaceutically acceptable diluent, carrier, vehicle and/or excipient.

33. A method for treatment of cancer in a subject comprising administering to the subject a therapeutically effective amount of an antibody molecule as defined in claim 1.

34. The method according to claim 33, wherein the cancer is a solid cancer.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0195] In the examples below, reference is made to the following figures:

[0196] FIG. 1: Antibodies of the invention specifically bind CTLA-4

[0197] The antibodies were shown by ELISA to bind to human and cynomolgous CTLA-4 but not human CD28 protein. Binding for 2-006 (FIG. 1A), 4-E-03 (FIG. 1B), 5-B07 (FIG. 1C) was compared to Yervoy (FIG. 1D).

[0198] FIG. 2: Dose-dependent binding of anti-CTLA-4 mAb to hCTLA-4-transfected cells

[0199] Anti-CTLA-4 mAb (FIG. 2A-2D) show strong binding to CTLA-4-expressing 293T cells similar to Yervoy (FIG. 2E).

[0200] FIG. 3: Anti-CTLA-4 mAb bind to in vitro-activated human CD4+ T cells

[0201] CD4+ T cells obtained from peripheral human blood were stimulated in vitro. Binding of anti-CTLA-4 mAb (solid lines, top row) was analysed by FACS and compared to Yervoy (dotted line, top row) and a commercial FACS antibody (bottom row).

[0202] FIG. 4: Binding block on in vitro-activated CD4+ T cells

[0203] Human in vitro-activated CD4+ T cells were stained with Alexa 647-labelled anti-CTLA-4 mAb (black line). Antibody binding was blocked by rhCTLA-4-Fc protein (grey line).

[0204] FIG. 5: Binding to in vitro-activated cynomologous CD4+ T cells

[0205] CD4+ T cells obtained from peripheral cynomologous blood were stimulated in vitro with CD3/CD28 dynabeads. Binding of anti-CTLA-4 mAb (solid lines, top row) was analysed by FACS and compared to Yervoy (dotted line, top row) and a commercial FACS antibody (bottom row).

[0206] FIG. 6: Block of cell binding by human and cynomologous CTLA-4 protein

[0207] 293T-CTLA-4 cells were stained with Alexa 647-labelled anti-CTLA-4 mAb (black line). Antibody binding was blocked by rhCTLA-4-Fc protein (light grey line) and rcmCTLA-4-Fc protein (dark grey line).

[0208] FIG. 7: Binding to 293T cells expressing cynomologous CTLA-4

[0209] 293T cells were transiently transfected with cynomologous CTLA-4 and binding of anti-CTLA-4 mAb at different concentrations was analysed by FACS.

[0210] FIG. 8: Expected lack of binding to resting human/cynomologous PBMCs

[0211] 4-E03—as well as 2-006, 5-B07 and 2-F09 (data not shown)—does not show any unspecific binding to different cell subsets in human (top row) and cynomologous (bottom row) PBMCs as analysed by FACS.

[0212] FIG. 9: Expected lack of direct agonistic activity

[0213] CFSE-labelled CD4+ T cells from healthy donors were stimulated with coated anti-CD3 plus soluble anti-CTLA-4 mAb or anti-CD28. % dividing cells (CFSElow CD25+ cells) were determined after 3 days by FACS. (A) FACS plots of one representative experiment (B) summarizing graph of 6 donors.

[0214] FIG. 10: CD80/CD86 blocking activity

[0215] Anti-CTLA-4 mAb block the binding of CD80 (FIG. 8A) and CD86 (FIG. 8b) to its ligand CTLA-4 as shown by ELISA.

[0216] FIG. 11: Functional ligand block in vitro

[0217] PBMCs were stimulated with SEB plus titrating doses of anti-CTLA-4 antibodies. Amount of secreted IL-2 in the supernatant was determined by MSD. In this figure, 1 representative donor out of 6 is shown.

[0218] FIG. 12: ADCC assay on in vitro-activated CD4+ T cells

[0219] In vitro-activated CD4+ T cells from heathy donors pre-opsonized with anti-CTLA-4 mAbs at 10 μg/ml were co-cultured with NK cells (NK-92 cell line) at 2:1 ratio. ADCC activity was assessed by FACS as described below. The figure shows the mean+SD of 4-8 donors.

[0220] FIG. 13: CTLA-4 is highest expressed on tumour resident Treg cells.

[0221] Samples of freshly excised ovarian tumours and blood were obtained from patients at surgery. Ascites was collected from patients with different cancer indications.

[0222] CTLA-expression on this patient material was compared to healthy PBMCs. Tumour samples were minced and digested. Peripheral blood mononuclear cells were separated by centrifugation. CTLA-4 expression was assessed on CD4+CD25+CD127− Treg cells, CD4+ non-Treg cells and CD8+ effector T cells by flow cytometry. Data represent individual patients/donors with n=12 for healthy PBMCs, n=20 for ascites, n=9 for tumour and n=5 for patient blood.

[0223] CTLA-4 expression was also analysed on human T cells which were activated in NOG mice in vivo and then isolated from the spleen of these mice (see FIG. 14).

[0224] FIG. 14: Anti-CTLA-4 mAb mediate Treg depletion in vivo.

[0225] Human PBMCs were injected i.v. into NOG mice. After approximately 2 weeks, spleens were taken and expression of CTLA-4 on human Treg cells and CD8+ T cells was analysed by FACS. Splenic cells isolated from NOG mice were transferred i.p. into SCID mice. 1 h later, mice were treated i.p. with CTLA-4 hIgG1 or control mAb. Intraperitoneal fluid was collected after 24 h and frequency of human T cell subsets (14A: Tregs and 14B: CD8+ T cells) was determined by flow cytometry.

[0226] FIG. 15: Table summarizing characteristics for anti-CTLA-4 antibodies

[0227] FIG. 16: FIG. 16: Characterization of mouse surrogate anti-CTLA-4 mAb

[0228] FIG. 16 A-B: Blocking ELISAs were performed with m5-B07 to evaluate ligand blocking characteristics. The antibody blocks the binding of FIG. 16A) CD80 and FIG. 16B) CD86 to its ligand CTLA-4 in a dose-dependent manner.

[0229] FIG. 16 C-D: 5-B07 in mouse IgG2a format mediated Treg deletion in CT26 tumor model. Balb/c mice were injected subcutaneously with 1×10.sup.6 CT26 cells and treatment started at tumor size of ca 7×7 mm. After 3 injections of 10 mg/kg antibody, tumor single cell suspensions were analyzed for immune cell content by FACS. FIG. 16C: Ligand blocking surrogate antibody 5-B07 causes T reg depletion. This causes a shift in CD8.sup.+/Treg T cell ratio as depicted in FIG. 16D.

[0230] FIG. 17: Generation of COPTG19384 and COPTG19385

[0231] Schematic representation of COPTG19384 and COPTG19385 used in this study. COPTG19385 contains a deletion of J2R gene in TK locus replaced by the heavy chain of anti-CTLA-4 driven by p7.5K, and a deletion of I4L gene in RR locus replaced by the light chain of anti-CTLA-4 driven by p7.5K. COPTG19384 contains a deletion of J2R gene in TK locus replaced by the heavy chain of anti-CTLA-4 driven by p7.5K, and a deletion of I4L gene in RR locus replaced by the light chain of anti-CTLA-4 driven by p7.5K, and by GM-CSF driven by pSE/L.

[0232] FIG. 18: Expression analysis of 4-E03 monoclonal antibody in supernatant of CEF cells infected with COPTG19384

[0233] A) by Western Blot: CEF cells were infected at MOI 0.05 with COPTG19384 in triplicate. Cell supernatants were harvested after 48 h and were analysed by WB after an electrophoresis in non-reducing condition and using either an anti-Ig (left blot) or an anti-light chain (right blot) HRP conjugated antibody.

[0234] B) by ELISA: CEF cells were infected at MOI 0.05 with COPTG19384 in triplicate or VVTG17137. Cell supernatants were harvested after 48 h and were analysed by ELISA for detection of either 4-E03 monoclonal antibody.

[0235] FIG. 19: Expression analysis of GM-CSF in supernatant of CEF cells infected with COPTG19384 by ELISA

[0236] CEF cells were infected at MOI 0.05 with COPTG19384 primary research stock in triplicate or VVTG17137. Cell supernatants were harvested after 48 h and were analysed by ELISA for detection of GM-CSF.

[0237] FIG. 20: Replication studies of COP WT, COPTG19384 (two batches) and VVTG17137 (two batches) on normal and tumoral hepatocytes

[0238] A) Replication rate on normal human hepatocytes.

[0239] B) Replication rate on malignant HepG2.

[0240] C) Therapeutic indexes calculated from the replication rates measured on HepG2 and hepatocytes.

[0241] FIG. 21: Replication of COPTG19384 and VVTG17137 in reconstructed human skin

[0242] Replication of COPTG19384 and VVTG17137 were evaluated after 7 days and with inoculums varying from 10 to 10.sup.5 pfu. Results are the means and SEM of three measurements.

[0243] FIG. 22: Oncolytic activities of COPTG19384 and VVTG17137 in three human tumoral cell lines: MIA PaCa-2 (A), LoVo (B) and HepG2 (C)

[0244] FIG. 23: Expression level of both 4-E03 monoclonal antibody and GM-CSF in (A) supernatants of infected HepG2 and LoVo and (B) supernatants of 5 different infected human tumoral cell lines

[0245] A) 4-E03 and GM-CSF expression levels were evaluated after 5 days of incubation, at MOI from 10.sup.−5 to 10.sup.−2 for COPTG19384 and at MOI of 10.sup.−2 for VVTG17137 used as negative control.

[0246] B) 4-E03 and GM-CSF expression levels were evaluated 48 hours after infection by COPTG19384 at MOI 0.05.

[0247] FIG. 24: Binding of different batches of 4-E03 to CTLA-4 protein

[0248] Binding of recombinantly produced 4-E03 by CHO (4-E03 research batch) or HEK (4-E03 tox batch) cells to (A) human and (B) cynomolgus recombinant protein was compared to the binding of 4-E03 purified from the supernatant of infected MIA PaCa-2 tumor cells (4-E03 TG) by ELISA.

[0249] FIG. 25: Binding of different batches of 4-E03 to CTLA-4 expressing cells

[0250] Binding of recombinantly produced 4-E03 by CHO (4-E03 research batch) or HEK (4-E03 tox batch) cells to (A) human and (B) cynomolgus CTLA-4 expressing cells was compared to the binding of 4-E03 purified from the supernatant of infected MIA PaCa-2 tumor cells (4-E03 TG) by flow cytometry.

[0251] FIG. 26: Kinetic of virus accumulation in LoVo xenografted tumor

[0252] Kinetic of virus accumulation was evaluated in LoVo xenografted tumor after a single i.t. injection of either COPTG19384 or VVTG17137 at two different doses (10.sup.4 or 10.sup.5 pfu). The solid or dashed lines represent the median of the three values determined at each time points.

[0253] FIG. 27: Kinetic of 4-E03 mAb and GM-CSF accumulation in LoVo xenografted tumor

[0254] A) Kinetic of 4-E03 mAb accumulation in LoVo xenografted tumor was evaluated after a single i.t. injection of either COPTG19384 or VVTG17137 at two different doses (10.sup.4 or 10.sup.5 pfu) or after a single i.p. injection of 3 mg/kg of 4-E03 monoclonal antibody. The solid or dashed lines represent the median of the three values.

[0255] B) Kinetic of GM-CSF accumulation in LoVo xenografted tumor was evaluated after a single i.t. injection of either COPTG19384 at two different doses (10.sup.4 or 10.sup.5 pfu) or VVTG17137 (10.sup.5 pfu). The solid lines represent the median of the three values determined at each time points.

[0256] FIG. 28: Kinetic of 4-E03 mAb and GM-CSF concentrations in sera of LoVo xenografted mice

[0257] A) Kinetic of 4-E03 mAb concentrations in sera was evaluated after a single i.t. injection in LoVo xenografted tumor of either COPTG19384 or VVTG17137 at two different doses (10.sup.4 or 10.sup.5 pfu) or after a single i.p. injection of 3 mg/kg of 4-E03 monoclonal antibody. The solid lines represent the median of the three values.

[0258] B) Kinetic of GM-CSF concentrations in sera after a single i.t. injection in LoVo xenografted tumor of either COPTG19384 at two different doses (10.sup.4 or 10.sup.5 pfu) or VVTG17137 (10.sup.5 pfu). The dashed lines represent the median of the three values determined at each time points.

[0259] FIG. 29: Kinetic of virus accumulation in CT26 tumors

[0260] Kinetic of virus accumulation in CT26 tumors after three i.t. injections (at D0, D2 and D4) of either VVTG18058, COPTG19421 or COPTG19407 at 10.sup.7 pfu/injection.

[0261] FIG. 30: Kinetic of m5-B07 mAb and mGM-CSF accumulation in CT26 tumor

[0262] A) Kinetic of m5-B07 mAb concentrations in CT26 tumor during and after three i.t. injections of either VVTG18058, COPTG19421 or COPTG19407 (10.sup.7 pfu/injection) or after a single i.p. injection of 3 mg/kg of m5-B07 monoclonal antibody. The solid lines represent the median of the three values.

[0263] B) Kinetic of mGM-CSF concentrations in CT26 tumor during and after three i.t. injections of either VVTG18058, COPTG19421 or COPTG19407 (10.sup.7 pfu/injection) or after a single i.p. injection of 3 mg/kg of m5-B07 monoclonal antibody. The solid lines represent the median of the three values determined at each time points.

[0264] FIG. 31: Kinetic of m5-B07 mAb concentrations in sera of CT26 model

[0265] Kinetic of m5-B07 mAb concentrations in sera after three i.t. injections in CT26 tumor of either VVTG18058, COPTG19421 or COPTG19407 (10.sup.7 pfu/injection) or after a single i.p. injection of 3 mg/kg of m5-B07 monoclonal antibody. The solid lines represent the median of the three values determined at each time points.

[0266] FIG. 32: Antitumoral activity in CT26 model: effect of COPTG19347+/−anti-PD1 on CT26 tumor growth (A) and mice survival (B)

[0267] CT26 cells were injected SC in BalB/c mice at D-7. COPTG19347 (10.sup.7 pfu), VVTG18058 (10.sup.7 pfu), VVTG18058 or buffer were injected IT, at D0, D2 and D4, possibly followed by the injection i.p. of 250 μg/mouse of anti-PD1 RMPI-14 at D7, D10, D14, D17 and D21.

[0268] FIG. 33: Dose-effect evaluation in CT26 model (compilation of survival data observed after treatment of COPTG19407, COPTG19421 and VVTG18058) FIG. 34: Anti-tumoral activity of COPTG19407 compared to VVTG18058 plus m5-B07 in CT26 tumor model

[0269] CT26 cells were injected s.c. into Balb/C mice. Treatment of the mice was started when tumors reached approx. 100 mm.sup.3. Mice were injected at D0, D2 and D5 with COPTG19407 (8.5×10.sup.6 pfu i.t.), VVTG18058 (8.5×10.sup.6 pfu i.t.), m5-B07 (10 mg/kg i.p.) or the combination of VVTG18058 (8.5×10.sup.6 pfu i.t.) plus m5-B07 (10 mg/kg i.p.). FIG. 30 A-D: Tumor growth and FIG. 16 E: survival were followed over time.

[0270] FIG. 35: Individual tumor volume curves of BALB/c mice bearing subcutaneous A20 tumors.

[0271] A20 cells were injected s.c. into Balb/C mice. Treatment of the mice was started when tumors reached 80-100 mm.sup.3. Mice were injected at D0, D2 and D4 with vehicle (i.t), COPTG19407 (4.75×10.sup.6 pfu i.t.), VVTG18058 (4.75×10.sup.6 pfu i.t.), RMP1-14 (anti-mPD-1) (250 μg/mouse i.p.) or the combination of COPTG19407 (4.75×10.sup.6 pfu i.t.) plus RMP1-14 (250 μg/mouse i.p.).

[0272] A) Group 1 animals treated with vehicle

[0273] B) Group 2 animals treated with VVTG18058

[0274] C) Group 3 animals treated with COPTG19407

[0275] D) Group 4 animals treated with murine anti-PD1

[0276] E) Group 5 animals treated with COPTG19407 and murine anti-PD1

[0277] FIG. 36: Mean tumor volume curves of BALB/cN mice bearing subcutaneous A20 tumors.

[0278] Each point represents the mean of the recorded tumor volume per group. The tumor volumes of all animals were monitored over 64 days. Mice were treated with vehicle (group 1), VVTG18058 (group 2), COPTG19407 (group 3), the murine anti-PD1 antibody RMP1-14 (BioXCell) (group 4) and COPTG19407 and RMP1-14 (group 5). Animals were randomized on D7 and treated during the period D7 to D31. Last mice were sacrificed on D61.

[0279] FIG. 37: Antitumoral activity in A20 model: effect of COPTG19407+/−anti-PD1 on A20 tumor growth (A) and mice survival (B)

[0280] A20 cells were injected s.c. into Balb/C mice. Treatment of the mice was started when tumors reached 80-100 mm.sup.3. Mice were treated with vehicle (i.t.) (group 1), anti-PD-1 (group 2), isotype (group 3), VVTG18058 (10.sup.5 pfu i.t.) (group 4), VVTG18058 (10.sup.5 pfu i.t.)+isotype (group 5), VVTG18058 (10.sup.5 pfu i.t.)+anti-PD-1 (group 6), VVTG19407 (10.sup.5 pfu i.t.) (group 7), VVTG19407 (10.sup.5 pfu i.t.)+isotype (group 8) and VVTG19407 (10.sup.5 pfu i.t.)+anti-PD-1 (group 9).

[0281] FIG. 38: Individual tumor volume curves of C57BL/6 mice bearing subcutaneous C38 tumors.

[0282] C38 cells were injected s.c. into C57bl/6 mice. Treatment of the mice was started when tumors reached 80-100 mm.sup.3. Mice were injected at D0, D2 and D4 with vehicle (i.t), COPTG19407 (4.75×10.sup.6 pfu i.t.), VVTG18058 (4.75×10.sup.6 pfu i.t.), RMP1-14 (anti-mPD-1) (250 μg/mouse i.p.) or the combination of COPTG19407 (4.75×10.sup.6 pfu i.t.) plus RMP1-14 (250 μg/mouse i.p.).

[0283] A) Group 1 animals treated with vehicle

[0284] B) Group 2 animals treated with VVTG18058

[0285] C) Group 3 animals treated with COPTG19407

[0286] D) Group 4 animals treated with murine anti-PD1

[0287] E) Group 5 animals treated with COPTG19407 and murine anti-PD1

[0288] FIG. 39: Mean tumor volume curves of C57BL/6 mice bearing subcutaneous C38 tumors.

[0289] Each point represents the mean of the recorded tumor volume per group. The tumor volumes of all animals were monitored over 61 days. Mice were treated with vehicle (group 1), VVTG18058 (group 2), COPTG19407 (group 3), the murine anti-PD1 antibody RMP1-14 (BioXCell) (group 4) and COPTG19407 and RMP1-14 (group 5). Animals were randomized on D7 and treated during the period D7 to D31. Last mice were sacrificed on D61.

[0290] FIG. 40: Individual tumor volume curves of BALB/c mice bearing subcutaneous EMT6 tumors.

[0291] EMT6 cells were injected s.c. into Balb/C mice. Treatment of the mice was started when tumors reached 80-100 mm.sup.3. Mice were injected at D0, D2 and D4 with vehicle (i.t), COPTG19407 (4.75×10.sup.6 pfu i.t.), VVTG18058 (4.75×10.sup.6 pfu i.t.), RMP1-14 (anti-mPD-1) (250 μg/mouse i.p.) or the combination of COPTG19407 (4.75×10.sup.6 pfu i.t.) plus RMP1-14 (250 μg/mouse i.p.).

[0292] A) Group 1 animals treated with vehicle

[0293] B) Group 2 animals treated with VVTG18058

[0294] C) Group 3 animals treated with COPTG19407

[0295] D) Group 4 animals treated with murine anti-PD1

[0296] E) Group 5 animals treated with COPTG19407 and murine anti-PD1

[0297] FIG. 41: Mean tumor volume curves of BALB/c mice bearing subcutaneous EMT6 tumors.

[0298] Each point represents the mean of the recorded tumor volume per group. The tumor volumes of all animals were monitored over 61 days. Mice were treated with vehicle (group 1), VVTG18058 (group 2), COPTG19407 (group 3), the murine anti-PD1 antibody RMP1-14 (BioXCell) (group 4) and COPTG19407 and RMP1-14 (group 5). Animals were randomized on D7 and treated during the period D7 to D31. Last mice were sacrificed on D56. Curves were stopped after the death of more than 20% of mice.

EXAMPLES

[0299] Specific, non-limiting examples which embody certain aspects of the invention will now be described. In the examples, rh protein denotes a human recombinant protein (e.g. rhIL-2 denotes human recombinant IL-2 protein) and rcm protein denotes a cynomologous recombinant protein (e.g. rcmCTLA4 denotes cynomologous recombinant CTLA-4 protein).

[0300] In addition to sequences mentioned above, some additional sequences are used in the examples, and these are set ut in Table 5 below.

TABLE-US-00005 TABLE 5 Additional sequences used in the examples 'dfj Sequence SEQ. ID. NO Human GM-CSF APARSPSPSTQPWEHVNAIQEARRLLNLSRDTAAEMNETVEVIEMF 55 without signal DLQEPTCLQTRLELYKQGLRGSLTKLKGPLTMMASHYKQHCPPTPE peptide TSCATQTITFESFKENLKDFLLVIPFDCWEPVQE Human GM-CSF MWLQSLLLLGTVACSISAPARSPSPSTQPWEHVNAIQEARRLLNLS 56 with signal RDTAAEMNETVEVISEMFDLQEPTCLQTRLELYKQGLRGSLTKLKG peptide PLTMMASHYKQHCPPTPETSCATQTITFESFKENLKDFLLVIPFDC WEPVQE Murine GM-CSF APTRSPITVTRPWKHVEAIKEALNLLDDMPVTLNEEVEVVSNEFSF 57 without signal KKLTCVQTRLKIFEQGLRGNFTKLKGALNMTASYYQTYCPPTPETD peptide CETQVTTYADFIDSLKTFLTDIPFECKKPVQK Murine GM-CSF MWLQNLLFLGIVVYSLSAPTRSPITVTRPWKHVEAIKEALNLLDDM 58 with signal PVTLNEEVEVVSNEFSFKKLTCVQTRLKIFEQGLRGNFTKLKGALN peptide MTASYYQTYCPPTPETDCETQVTTYADFIDSLKTFLTDIPFECKKP VQK Promoter CCACCCACTTTTTATAGTAAGTTTTTCACCCATAAATAATAAATAC 59 p7.5K AATAATTAATTTCTCGTAAAAGTAGAAAATATATTCTAATTTATTG CACGGTAAGGAAGTAGAATCATAAAGAACAGT Promoter TTTATTCTATACTTAAAAAATGAAAATAAATACAAAGGTTCTTGAG 60 pH5.4 GGTTGTGTTAAATTGAAAGCGAGAAATAATCATAAATTATTTCATT ATCGCGATATCCGTTAAGTTTG Promoter AAAAATTGAAATTTTATTTTTTTTTTTTGGAATATAAATA 61 pSE/L

Example 1—Generation of CTLA-4 Specific Antibodies

[0301] Isolation of scFv Antibody Fragments

[0302] The n-CoDeR® scFv library (BioInvent; Söderlind E, et al Nat Biotechnol. 2000; 18(8):852-6) was used to isolate scFv antibody fragments recognizing human CTLA-4.

[0303] The phage library was used in three consecutive pannings against recombinant human protein. After phage incubation, the cells were washed to remove unbound phages. Binding phages were eluted with trypsin and amplified in E. coli. The resulting phage stock was converted to scFv format. E. coli was transformed with scFv bearing plasmids and individual scFv clones were expressed.

Identification of Unique CTLA-4 Binding scFv

[0304] Converted scFv from the third panning were assayed using a homogeneous FMAT analysis (Applied Biosystems, Carlsbad, Calif., USA) for binding to transfected 293 FT cells expressing human CTLA-4 or a non-target protein.

[0305] Briefly, transfected cells were added to clear-bottom plates, together with the scFv-containing supernatant from expression plates (diluted 1:7), mouse anti-His Tag antibody (0.4 μg/ml; R&D Systems) and an APC-conjugated goat anti-mouse antibody (0.2 μg/ml; cat.no. 115-136-146, Jackson Immunoresearch). FMAT plates were incubated at room temperature (approximately 20-25° C.) for 9 h prior to reading. Target-specific bacterial clones were classified as actives and cherry picked into 96-well plate.

IgG Binding to CTLA-4 in ELISA

[0306] 96-well plates (Lumitrac 600 LIA plate, Greiner) were coated overnight at 4° C. with recombinant human CTLA-4-Fc protein (R&D Systems) at 1 pmol/well, recombinant cynomologous (cm) CTLA-4-Fc protein (R&D Systems) at 1 pmol/well, recombinant mouse CTLA-4-Fc protein (R&D Systems) at 0.3 pmol/well, recombinant human CD28-Fc protein (R&D Systems) at 1 pmol/well or recombinant mouse CD28-Fc protein (R&D Systems) at 0.3 pmol/well. After washing, titrated doses of anti-CTLA-4 mAbs from 0 μg/ml to 0.06 ng/ml (66 nM to 0.3 pM) were allowed to bind for 1 hour. Plates were then washed again and bound antibodies were detected with an anti-human-F(ab)-HRP secondary antibody (Jackson ImmunoResearch) diluted in 50 ng/ml. Super Signal ELISA Pico (Thermo Scientific) was used as substrate and the plates were analysed using Tecan Ultra Microplate reader.

[0307] All antibodies were shown to bind to human and cynomologous CTLA-4 protein but not to human CD28 protein as assayed by ELISA. In addition, 5-B07 was shown to bind to mouse CTLA-4 but not mouse CD28 (see FIG. 1).

IgG Binding to CTLA-4-Expressing 293T Cells in Flow Cytometry

[0308] Converted IgG clones were analysed for binding to CTLA-4-expressing 293T cells (purchased from Crownbio). Cells were incubated with different concentrations (as indicated in FIG. 2) of anti-CTLA-4 mAb at 4° C. for 20 min prior to washing and staining with an APC-labelled goat anti-human secondary antibody (cat.no. 109-136-088, Jackson ImmunoResearch). Dead cells were excluded from analyses using Fixable Viability Dye eFluor780 (eBiosciences). Data acquisition was performed on a FACSVerse (BD Biosciences, Franklin Lakes, N.J.) and analysed with FlowJo software (Tree Star, Ashland, Oreg.).

[0309] The anti-CTLA-4 mAbs were shown to bind human CTLA-4-expressing 293T cells in a dose-dependent manner with a similar EC50 value as Yervoy (FIG. 2).

[0310] 293T cells stably transfected with human CTLA-4, 293T cells transiently transfected with cynomologous CTLA-4, naive human or cynomologous PBMCs, in vitro-activated human or cyno CD4+ T cells were incubated with the concentrations of anti-CTLA-4 mAb indicated at 4° C. for 20 min prior to washing and staining with a APC-labelled anti-human secondary antibody (Jackson ImmunoResearch).

Example 2—Anti-CTLA-4 mAb Specifically Bind Human and Cynomologous CTLA-4-Expressing (Primary) Cells

[0311] CTLA-4 Specific mAb Bind Primary Human and Cynomologous In Vitro-Activated CD4+ T Cells but not Naïve PBMCs Isolated from Healthy Donors

[0312] PBMCs were isolated from buffy coats. Briefly, buffy were diluted 1:3 in PBS and were loaded onto Ficoll-Paque Plus (Amersham) cushions. Samples were centrifuged at 800×g for 20 min at 20° C. The upper, plasma-containing phase was removed and mononuclear cells were isolated from the distinct white band at the plasma/Ficoll interphase.

[0313] Human peripheral CD4.sup.+ T-cells were purified by negative selection using MACS CD4 T-cell isolation kit (Miltenyi Biotec). CD4+ T cells were activated in vitro with CD3/CD28 dynabeads (Life Technologies) plus 50 ng/ml rhIL-2 (R&D Systems) in R10 medium (RPMI containing 2 mM glutamine, 1 mM pyruvate, 100 IU/ml penicillin and streptomycin and 10% FBS (GIBCO by Life Technologies) for 3 days to upregulate CTLA-4 expression. Cynomologous CD4+ T cells were isolated using non-human CD4 microbeads (Miltenyi Biotec) and incubated with 50 ng/ml PMA (Sigma-Aldrich) and 100 ng/ml lonomycin (Sigma-Aldrich) for 3 days.

[0314] Naive human or cynomologous PBMCs, in vitro-activated human or cyno CD4+ T cells were incubated with the indicated concentrations of anti-CTLA-4 mAb at 4° C. for 20 min prior to washing and staining with a APC-labelled anti-human secondary antibody (Jackson ImmunoResearch). Binding of anti-CTLA-4 mAb was analysed by FACS using a BD FACS Verse.

[0315] The antibodies were shown to bind to in vitro-activated human (FIG. 3) and cynomologous (FIG. 5) CD4+ T cells but not to resting PBMCs (FIG. 8). The binding to endogenously CTLA-4 expressing T cells is similar to the staining with Yervoy (FIG. 3, upper row, dotted line) and a as a positive control a commercial anti-CTLA-4 FACS-antibody from BD Biosciences (clone BNI3; FIG. 3, bottom row).

[0316] As shown in FIG. 4, the staining of human in vitro-activated CD4+ T cells (black line) can be completely blocked by rhCTLA-4-Fc (grey line) demonstrating the specificity of the antibodies. In this competitive binding assay, 2 μg/ml Alexa 647-labelled anti-CTLA-4 mAb was mixed with recombinant human CTLA-4-Fc protein (50 μg/ml) prior to incubation with CTLA-4 expressing cells. IgG binding was detected by FACS.

Transfected 293T Cells Expressing Human and Cynomologous CTLA-4 Confirm Cynocrossreactivity of the Tested Antibodies

[0317] The cyno-crossreactivity of the antibodies was further confirmed on transfected CTLA-4 expressing 293T cells.

[0318] As demonstrated in FIG. 6, the binding of CTLA-4 specific antibodies to human CTLA-4 expressing transfected cells can be inhibited by human and cynomologous recombinant protein (both R&D Systems). The antibodies were also shown to bind to transfected cells expressing cynomologous CTLA-4 (FIG. 7, upper row). This binding can be blocked again by cynomologous recombinant protein (bottom row, grey line). The experiments were performed similar to the competitive assay described above in Example 2 in connection with FIG. 4.

Expected Lack of Direct Agonistic Activity

[0319] In vitro-proliferation assays were performed to exclude unanticipated direct agonistic activity (e.g. due to unspecific binding).

[0320] Human peripheral CD4+ T-cells were purified from healthy PBMCs by negative selection using MACS CD4 T-cell isolation kit (Miltenyi Biotec) and were thereafter labelled with CFSE (2 μM, Molecular Probes). Antibodies were cross-linked with F(ab′).sub.2 goat anti-human IgG, Fcγ fragment specific or F(ab′).sub.2 goat anti-mouse IgG, Fcγ fragment specific in a molar ratio IgG:F(ab′).sub.2=1.5:1 for 1 h at RT. 1×10.sup.5 purified human CD4+ T cells were stimulated with plate-bound anti-CD3 (0.5 μg/ml; clone UCHT1, R&D Systems) and 4 μg/ml of soluble, cross-linked anti-CTLA-4 or crosslinked anti-CD28 (clone CD28.2, BioLegend) for 72 hours at 37° C. Cells were washed and stained with a BV421-conjuagated anti-CD25 antibody (clone M-A251, BD Biosciences). The percentage of CD25+/CFSElow dividing cells was analysed by FACS.

[0321] FIG. 9 demonstrates that none of the tested anti-CTLA-4 mAb induces T cell proliferation in contrast to anti-CD28 stimulation.

Example 3—Anti-CTLA-4 mAb Block Ligand Binding of CD80/CD86

Ligand Blocking ELISA

[0322] The ligand blocking activity of anti-CTLA-4 IgGs was assessed by ELISA. To this end, recombinant human CTLA-4-Fc protein (R&D Systems) was coated to 96-well plates (Lumitrac 600 LIA plate, Greiner) at 1 pmol/well. After washing, titrated doses of anti-CTLA-4 mAbs were allowed to bind for 1 hour. His-tagged ligands were added at 200 nM and 100 nM, respectively (rhCD80 and rhCD86; R&D Systems) and the plates were further incubated for 15 minutes. After washing, bound ligand was detected with an HRP-labelled anti-His antibody (R&D Systems). Super Signal ELISA Pico (Thermo Scientific) was used as substrate and the plates were analysed using Tecan Ultra Microplate reader.

[0323] As shown in FIG. 10, anti-CTLA-4 antibodies tested show similar ligand blocking activity as Yervoy.

Functional Ligand Block In Vitro

[0324] For the SEB PBMC assay, total PBMCs from healthy donors were seeded on 96-well plates (1×10.sup.5 cells/well) and stimulated with 1 μg/ml Staphylococcus enterotoxin B (SEB, Sigma Aldrich) in the presence of titrated doses of anti-CTLA-4 IgGs. IL-2 secretion was measured by MSD (Mesoscale) on day 3 according to manufactures' instructions.

[0325] The antibodies 4-E03 and 2-006 were shown to enhance IL-2 production and their potency was shown to be similar to that of Yervoy. In FIG. 11, one representative donor out of 6 is shown.

Example 4—Anti-CTLA-4 mAb Deplete CTLA-4 Expressing Cells In Vitro and In Vivo

Antibody Dependent Cellular Cytotoxicity (ADCC)

[0326] ADCC assays were performed using an NK-92 cell line stably transfected to express the CD16-158V allele together with GFP (purchased from Conkwest, San Diego, Calif.; Binyamin, L., et al., 2008, Blocking NK cell inhibitory self-recognition promotes antibody-dependent cellular cytotoxicity in a model of anti-lymphoma therapy. Journal of immunology 180, 6392-6401). CD4+ target T cells were isolated from peripheral blood of healthy donors using CD4+ T cell isolation kit (Miltenyi Biotec). Cells were stimulated for 48 hours with CD3/CD28 dynabeads (Life Technologies, Thermo Fisher) and 50 ng/ml rhIL-2 (R&D Systems) at 37° C. Target cells were pre-incubated with mAb at 10 μg/ml for 30 min at 4° C. prior to mixing with NK cells. The cells were incubated for 4 h in RPMI 1640+GlutaMAX medium (Invitrogen) containing 10 mM HEPES buffer, 1 mM sodium Pyruvate and 10% FBS low IgG at a 2:1 effector:target cell ratio. Lysis was determined by flow cytometry. Briefly, at the end of the incubation, the cell suspension was stained with BV510-conjugated anti-CD4 (clone RPA-T4, BD Biosciences) together with 10 nM SYTOX Red dead cell stain (Invitrogen) or Fixable Viability Dye eFluor780 (eBioscience) for 20 min in the dark at 4° C. and the cells were then analysed using a FACSVerse (BD Biosciences).

[0327] 4-E03 showed a significantly improved deletion of CTLA-4+ T cells in vitro compared to Yervoy (FIG. 12).

CTLA-4 Expression on Primary Patient Material

[0328] In order to validate the translational potential of the finding above on the depleting activity of anti-CTLA-4 mAb, the CTLA-4 expression was examined on primary patient material.

[0329] Ethical approval for the use of clinical samples was obtained by the Ethics Committee of Sickle University Hospital. Informed consent was provided in accordance with the Declaration of Helsinki. Samples were obtained through the Department of Gynocology and Department of Oncology at, Skånes University Hospital, Lund. Ascitic fluid was assessed as single cell suspensions that had been isolated. Tumour material was cut into small pieces and incubated in R10 with DNase I (Sigma Aldrich) and Liberase™ (Roche Diagnostics) for 20 min at 37° C. Remaining tissue was mechanically crashed and, together with the cell suspension, passed through a 70 μm cell strainer. Cells isolated from ascitic fluid and tumours were stained. To identify different T cell subsets following antibodies were used: CD4-BV510 (RPA-T4), CD25-BV421 (M-A251), anti-CD127-FITC (HIL-7R-M21), CTLA-4-PE (BNI3), CD8-PeCy7 (RPA-T8), CD3-APC (UCHT1), CD45-PercP-Cy5.5 (HI30), mouse IgG2a isotype,κ control-PE (G155-178; all from BD Biosciences). Data acquisition was performed using FACSVerse and data analysed using FlowJo.

[0330] As shown in FIG. 13, CTLA-4 is highest expressed on intratumoral Treg cells which makes them a good target for depleting CTLA-4-specific antibodies.

PBMC-NOG/SCID Model

[0331] To confirm the in vitro findings on the depleting activity of the CTLA-4 specific antibodies, we analysed the depleting capacity of anti-CTLA-4 mAb in a PBMC-NOG/SCID model in vivo. The model is based on the well-established hu-PBMC-NOG model (Søndergaard H. et al., Clin Exp Immunol. 2013 May; 172(2):300-10. doi: 10.1111/cei.12051; Cox J H et al., PLoS One. 2013 Dec. 23; 8(12):e82944. doi: 10.1371/journal.pone.0082944. eCollection 2013) and was modified in-house as described below.

[0332] Mice were bred and maintained in local facilities in accordance with home office guidelines. Eight weeks-old female C.B. 17 scid (Bosma G C et al., Nature. 1983 Feb. 10; 301(5900):527-30) and NOG (NOD/Shi-scid/IL-2Rγ.sup.null; Ito M et al, 2002, NOD/SCID/γ.sub.C.sup.null mouse: an excellent recipient mouse model for engraftment of human cells. Blood 100(9):3175-3182) mice were supplied by Taconic (Bomholt, Denmark) and maintained in local animal facilities. For the PBMC-NOG/SCID (primary human xenograft) model, human PBMCs were isolated using Ficoll Paque PLUS and after washing the cells were resuspended in sterile PBS at 75×10.sup.6 cells/ml. NOG mice were i.v. injected with 200 μl cell suspension corresponding to 15×10.sup.6 cells/mouse. 2 weeks after injection, the spleens were isolated and rendered into a single cell suspension. Thereafter, a small sample was taken to determine the expression of CTLA-4 on human T cells by FACS. As indicated in FIG. 13, CTLA-4 is higher expressed on Treg cells compared to other T cells reflecting the situation in human patients. The majority of the cells was resuspended in sterile PBS at 50×10.sup.6 cells/ml. SCID mice were injected i.p. with 200 μl of the suspension corresponding to 10×10.sup.6 cells/mouse. 1 h later, mice were treated with 10 mg/kg of either anti-CTLA-4 hIgG1, Yervoy or isotype control mAb. The intraperitoneal fluid of the mice was collected after 24 h. Human T cell subsets were identified and quantified by FACS using following markers: CD45, CD4, CD8, CD25, CD127 (all from BD Biosciences).

[0333] All antibodies tested showed a similar or better Treg depleting activity than Yervoy. Other T cell populations, such as CD8+ effector T cells, were not affected (FIG. 14).

Example 5—the Selected Surrogate Antibody m5-B07 Shows the Same Functional Characteristics as 4-E03

[0334] In some of the examples, in particular the in vivo examples, the antibody clone 5-B07 in mIgG2a format has been used (also denoted m5-B07). This is a mouse antibody, which is a surrogate antibody to the human antibodies disclosed herein. It has been selected as a surrogate antibody since it binds murine CTLA-4 and thereby blocks ligand binding (FIG. 16 A-B). Furthermore, it also shows Treg depleting activity (FIG. 16 C-D).

Ligand Blocking ELISA

[0335] The ligand blocking activity of 5-B07 was assessed by ELISA. To this end, recombinant mouse CTLA-4-Fc protein (Sino Biological Inc.) was coated to 96-well plates (Lumitrac 600 LIA plate, Greiner) at 1 pmol/well. After washing, titrated doses of anti-CTLA-4 mAbs were allowed to bind for 1 hour. His-tagged ligands were added at 200 nM and 100 nM, respectively (rmCD80 and rmCD86; Sino Biological Inc.) and the plates were further incubated for 15 minutes. After washing, bound ligand was detected with an HRP-labelled anti-His antibody (R&D Systems). Super Signal ELISA Pico (Thermo Scientific) was used as substrate and the plates were analysed using Tecan Ultra Microplate reader.

[0336] As shown in FIG. 16, the antibody blocks the binding of (A) CD80 and (B) CD86 to its ligand CTLA-4.

Treg Depleting Activity In Vivo

[0337] The effects of CTLA-4 specific antibodies on the T cell subsets in the tumor in vivo was investigated in the CT26 tumor model as described below.

[0338] Mice were bred and maintained in local facilities in accordance with home office guidelines. Six to eight weeks-old female Balb/C were supplied by Taconic (Bomholt, Denmark) and maintained in local animal facilities. CT26 cells (ATCC) were grown in glutamax buffered RPMI, supplemented with 10% FCS. When cells were semi confluent they were detached with trypsin and resuspended in sterile PBS at 10×10.sup.6 cells/ml. Mice were s.c. injected with 100 μl cell suspension corresponding to 1×10.sup.6 cells/mouse. When the tumors reached approximately 7×7 mm, the mice were treated twice weekly i.p. with 10 mg/kg of the indicated antibodies as indicated in figures. After the third administration, tumors were dissected out, mechanically divided into small pieces and digested using a mixture of 100 μg/ml liberase (Roche) and 100 μg/ml Dnase (Sigma) in 37° C. for 2×5 min with Vortex in between. The cell suspension was then washed (400 g for 10 min) with PBS containing 10% FBS. Thereafter, the cells were resuspended in MACS buffer and stained with an antibody panel staining CD45, CD3, CD8, CD4 and CD25 (all from BD Biosciences). Before staining, the cells were blocked for unspecific binding using 100 μg/ml IVIG. Cells were analyzed using a FACS Verse (BD Biosciences). Mouse Treg cells were identified as CD45.sup.+CD3.sup.+CD4.sup.+CD25.sup.+ cells.

[0339] As shown in FIG. 16 C, 5-B07 in mouse IgG2a format mediates Treg deletion in the tumor associated with D) increased CD8/Treg ratio compared to other CTLA-4 specific n-CoDeR antibodies and the well-described commercially available clone 9H10.

Example 6—Generation of a Virus Expressing Anti-CTLA4 mAb (COPTG19385) or Anti-CTLA4 mAb and GM-CSF (COPTG19384), Expression of Transgenes and Characterization of Genetic Stabilities

[0340] COPTG19384 and COPTG19385 are vaccinia viruses (Copenhagen strain) encoding the monoclonal antibody anti-CTLA4 (4-E03). COPTG19384 further encodes the human GM-CSF. More particularly, COPTG19384 and COPTG19385 are both defective for thymidine kinase (TK, J2R locus) and ribonucleotide reductase (RR, I4L locus) activities. As illustrated in FIG. 17, the expression cassette encoding the 4-E03 heavy chain (HC; SEQ ID NO: 54) under the control of p7.5K promoter (SEQ ID NO: 59) was inserted at the J2R locus, and the expression cassette encoding the light chain (LC, SEQ ID NO: 53) of the 4-E03 IgG under the control of p7.5K promoter SEQ ID NO: 59) was placed at the I4L locus. For COPTG19384, the expression cassette encoding the human GM-CSF (SEQ ID NO: 56) under the control of pSE/L promoter (SEQ ID NO: 61) was also placed at the I4L locus.

[0341] The same promoter (p7.5K) was used to control the expression of HC and LC to obtain to same level of expression for both chains and therefore an optimal assembly of the antibody as a hetero-tetrameric protein (i.e. to avoid excess of non-associated chain). However, the same promoter for both chains of the antibody precludes from inserting them at the same locus (identical DNA sequences increase the risk of recombination and then elimination of transgenes). Therefore, the cassette encoding the 4-E03 HC was inserted at the J2R locus and the cassette encoding the 4-E03 LC at the I4L locus. The cassette encoding the GM-CSF transgene but under a different promoter (pSE/L) was also inserted into the I4L locus, like the antibody light chain.

Generation of COPTG19384

[0342] The vaccinia virus transfer plasmids, pTG19339 and pTG19341, were designed to allow insertion of nucleotide sequences by homologous recombination in J2R and in I4L loci of the vaccinia virus genome, respectively. They originate from the plasmid pUC18 into which were cloned the flanking sequences (BRG and BRD) surrounding the J2R (pTG19339) or I4L (pTG19341) locus. Each plasmid contains also the p7.5K promoter.

[0343] A synthetic fragment named “Fragment HC” of 1436 bp containing the HC gene of 4-E03 antibody was produced. A fragment “LC fragment” containing the LC gene of 4-E03 antibody and hGM-CSF gene under the control of the pSE/L was generated by a synthetic way and inserted in a plasmidic vector. The coding sequences were optimized for human codon usage, a Kozak sequence (ACC) was added before the ATG start codon and a transcriptional terminator (TTTTTNT) was added after the stop codon. Moreover, some patterns were excluded: TTTTTNT, GGGGG, CCCCC which are deleterious for expression in poxvirus.

[0344] The HC fragment was inserted in pTG19339 restricted with PvuII by homologous recombination, giving rise to pTG19367. The LC-carrying plasmid was restricted by SnaB1 and the resulting fragment “LC-GMCSF” was inserted by homologous recombination in pTG19341 restricted with PvuII, giving rise to pTG19384. In this plasmid, the expressions cassettes were inserted head to tail between recombination arms allowing homologous recombination in the I4L locus of vaccinia virus genome.

[0345] COPTG19384 was generated on chicken embryo fibroblast (CEF) by two successive homologous recombination for successive insertion in I4L and J2R loci and by using COPTG19156 as parental virus and the two transfer plasmids pTG19367 and pTG19384. CEF were isolated from 12 day-old embryonated SPF eggs (Charles River). The embryos were mechanically dilacerated, solubilized in a Tryple Select solution (Invitrogen) and cultured in a MBE (Eagle Based Medium; Gibco) supplemented with 5% FCS (Gibco) and 2 mM L-glutamine.

[0346] The homologous recombination between the transfer plasmids and parental vaccinia virus enables the generation of recombinant vaccinia viruses which have lost the GFP and the mCherry expression cassettes and gained the antibody and GM-CSF expression cassettes. COPTG19156 contains the expression cassette of the mCherry gene in its I4L locus and the expression cassette of the GFP gene in its J2R locus. The homologous recombination between the transfer plasmid pTG19367 and the parental COPTG19156 enables the generation of recombinant vaccinia viruses which have lost their GFP expression cassette and gained the 4-E03 heavy chain expression cassette and the selection was performed by isolation of red fluorescent plaques. This intermediary recombinant virus (COPTG19367) was used as parental virus for a second round of homologous recombination with pTG19384 as transfer plasmid for the generation of recombinant vaccinia viruses which have lost their mCherry expression cassette and gained the 4-E03 light chain and GM-CSF expression cassettes. The selection of COPTG19384 (FIG. 17) was performed by isolation of white non-fluorescent plaques.

[0347] The viral stock of COPTG19384 was amplified on CEFs in two F175 flasks to generate appropriate stocks of viruses which can be aliquoted and stored at −80° C. until use. Viral stock was titrated on CEF cells and infectious titers were expressed in pfu/mL and calculated with the following formula: number of lytic areas×dilution factor×4. For illustrative purposes, the produced viral stock titrated 6.8×10.sup.6 pfu/mL. This stock was analyzed by PCR to verify the integrity of the expression cassettes and recombination arms using appropriate primer pairs. The stock was also analyzed by sequencing of both expression cassettes. Alignment of sequencing results showed 100% homology with the theoretical expected sequence. If needed, viral preparations were purified using conventional techniques (e.g. as described in WO2007/147528).

Generation of COPTG19385

[0348] The vaccinia virus transfer plasmids, pTG19339 and pTG19341, were designed to allow insertion of nucleotide sequences by homologous recombination in J2R and in I4L loci of the vaccinia virus genome, respectively. They originate from the plasmid pUC18 into which were cloned the flanking sequences (BRG and BRD) surrounding the J2R (pTG19339) or I4L (pTG19341) locus. Each plasmid contains also the p7.5K promoter.

[0349] A synthetic fragment named “Fragment HC” of 1436 bp containing the HC gene of 4-E03 antibody was produced. The coding sequences were optimized for human codon usage, a Kozak sequence (ACC) was added before the ATG start codon and a transcriptional terminator (TTTTTNT) was added after the stop codon. Moreover, some patterns were excluded: TTTTTNT, GGGGG, CCCCC which are deleterious for expression in poxvirus.

[0350] The HC fragment was inserted in pTG19339 restricted with PvuII by homologous recombination, giving rise to pTG19367.

[0351] The plasmid containing the expression cassette encoding only the 4-E03 light chain was obtained by elimination of the cassette encoding the hGM-CSF gene under the control of the pSE/L in the plasmid pTG19384 (described above). pTG19384 was restricted with NheI and XbaI (compatible cohesive ends) and was religated, giving rise to pTG19385.

[0352] COPTG19385 was generated on chicken embryo fibroblast (CEF) by two successive homologous recombination for successive insertion in J2R and I4L loci and by using COPTG19156 as parental virus and the two transfer plasmids pTG19367 and pTG19385. CEF were isolated from 12 day-old embryonated SPF eggs (Charles River). The embryos were mechanically dilacerated, solubilized in a Tryple Select solution (Invitrogen) and cultured in an MBE (Eagle Based Medium; Gibco) supplemented with 5% FCS (Gibco) and 2 mM L-glutamine.

[0353] The homologous recombination between the transfer plasmids and parental vaccinia virus enables the generation of recombinant vaccinia viruses which have lost the GFP and the mCherry expression cassettes and gained the antibody expression cassettes. COPTG19156 contains the expression cassette of the mCherry gene in its I4L locus and the expression cassette of the GFP gene in its J2R locus. The homologous recombination between the transfer plasmid pTG19367 and the parental COPTG19156 enables the generation of recombinant vaccinia viruses which have lost their GFP expression cassette and gained the 4-E03 heavy chain expression cassette and the selection was performed by isolation of red fluorescent plaques. This intermediary recombinant virus (COPTG19367) was used as parental virus for a second round of homologous recombination with pTG19385 as transfer plasmid for the generation of recombinant vaccinia viruses which have lost their mCherry expression cassette and gained the 4-E03 light chain expression cassette. The selection of COPTG19385 was performed by isolation of white non-fluorescent plaques.

[0354] The viral stock of COPTG19385 was amplified on CEFs in two F175 flasks to generate appropriate stocks of viruses which can be aliquoted and stored at −80° C. until use. Viral stock was titrated on CEF cells and infectious titers were expressed in pfu/mL and calculated with the following formula: number of lytic areas×dilution factor×4. For illustrative purposes, the produced viral stock titrated 1.04×10.sup.7 pfu/mL. This stock was analyzed by PCR to verify the integrity of the expression cassettes and recombination arms using appropriate primer pairs. The stock was also analyzed by sequencing of both expression cassettes. Alignment of sequencing results showed 100% homology with the theoretical expected sequence. If needed, viral preparations were purified using conventional techniques (e.g. as described in WO2007/147528).

Expression of Transgenes

[0355] Virus-mediated expression of 4-E03 monoclonal antibody was evaluated in supernatants of CEF cells infected with COPTG19384 by Western Blot (WB) and compared to the recombinantly produced antibody (40 ng of 4-E03). WB allows to visualize the presence of non-functional molecules that do not bind the CTLA4 (e.g. molecules with incomplete chain assembly, aggregates). CEF cells were infected at MOI 0.05 with COPTG19384 viral stock in triplicate. Cell supernatants were harvested after 48 h and were analyzed by WB after an electrophoresis in non-reducing condition and using either an anti-Ig (left blot) or an anti-light chain (right blot) HRP conjugated antibody. The results illustrated in FIG. 18A indicate that the WB profile in non-reducing condition of the mAb produced by infected CEF is close to that of the purified 4-E03 with a similar apparent size between 100 to 150 kDa, indicating a correct chains-folding and assembly.

[0356] Quantification in supernatants of the functional secreted 4-E03 antibodies and GM-CSF was performed with ELISA. ELISA allowed to measure quantitively the amount of functional polypeptide produced in cell supernatants. VVTG17137 was used as a negative control. It is a vaccinia virus (Copenhagen strain) deleted in J2R and I4L loci encoding the suicide gene FCU1 (described in WO2009/065546).

[0357] For estimation of 4-E03 antibody, microplates were coated by an overnight incubation at 4° C. with 100 μL per well of CTLA4-Fc at 0.25 μg/mL. After incubation, the coating solution was discarded, blocking solution was added and plates were incubated for 1 to 2 hours at RT before being washed. 4-E03 calibration standards (from 0.097 to 100 ng/mL), or samples (in triplicate) diluted in blocking solution were added to the wells and the plates were incubated for 2 hours at 37° C. before being washed. HRP conjugated antibody diluted in blocking solution was added to each well and plates were incubated 1 hour at 37° C. before being washed. After incubation with TMB solution 30 min at RT in darkness, H.sub.2SO.sub.4 2 M (stop solution) was added to stop the enzymatic reaction. Absorbances were read at 450 nm on microplate reader. Absorbances were plotted versus antibody concentration of the calibration standard. As illustrated in FIG. 18B, functional 4-E03 mAb was produced in cells infected by COPTG19384, reaching concentrations close to 1 μg/mL.

[0358] Virus-mediated expression of GM-CSF was also evaluated in the same supernatants using Quantikine® ELISA (R&D Systems Ref SGM00). Briefly, this assay is using two anti-hGM-CSF antibodies. The first antibody used to capture the hGM-CSF in the samples was coated on the well surface of a 96-well plate. The second one is conjugated and added to the plate in solution to detect the captured hGM-CSF. The concentration of hGM-CSF in the sample is then calculated by interpolation from a calibration curve established with some purified hGM-CSF provided by the kit. As illustrated in FIG. 19, the level of expression of GM-CSF was around 6 μg/mL.

[0359] In conclusion, concentrations equal or above 1 μg/mL were detected for both transgenes indicating a satisfying level of expression.

Genetic Stability:

[0360] Genetic stability tests were performed after five passages of the virus on CEF in serum free medium at a multiplicity of infection (MOI) of 10.sup.−4. Passage P5 was diluted and inoculated on CEF cells in 60 mm culture dishes to obtain from 20 to 40 viral plaques per dish. One hundred viral plaques were isolated and sub-cultured. After one amplification cycle, isolated viral plaques were inoculated on CEF cells and tested by PCR and by ELISA. PCR analysis and expression of transgenes showed that more than 90% of clones have a correct profile. As the acceptance criteria for clinical development of a product is a genetic stability superior or equal to 90%, COPTG19384 was considered as genetically stable.

Example 7—In Vitro Characterization of COPTG19384

Replication Studies in Hepatocytes: Tumor-Selectivity of a Virus Expressing Anti-CTLA4 mAb and GM-CSF

[0361] COPTG19384 carries two gene deletions encoding viral enzymes (TK and RR) involved in nucleotide metabolism. When functional, these enzymes allow the virus to replicate in cytoplasm of most of cells including the ones in resting state (i.e. with low nucleotide pool available). A host range study was performed in primary and malignant cells to verify that the insertion of the different transgenes in the J2R and I4L loci does not modify the host range selectivity. The replication of COPTG19384 was assessed on normal primary human cells (hepatocytes, prepared by Biopredic) and on tumoral cells from the same organ (HepG2 from hepatocarcinoma, ATCC® HB-8065™). Replication rates and therapeutic indexes were calculated and compared to those of both the wild type Copenhagen vaccinia (COP WT, virus without any deletion) as references for non-selective vaccinia virus and the recombinant double deleted VVTG17137 virus (deleted in J2R and 14L genes with the suicide FCU1 gene inserted in place of J2R), respectively. Two batches were assayed, a research batch (batch1) and a GMP-produced batch (batch 2). The replication rate was determined as the ratio of total infectious particles at the end of incubation/initial infectious particle (inoculum). The therapeutic index of each virus was determined as the ratio: replication rate on HepG2 cells/replication rate on hepatocytes. The higher the ratio the better the selectivity of the virus toward the tumoral cells.

[0362] Primary hepatocytes were grown in Basal Hepatic cell medium supplemented with 1.6% Additives for hepatocyte culture medium. HepG2 were seeded in 12 well plate at 4 E+05 cells/well and incubated for 24 H at 37° C. with 5% CO.sub.2. Before infection the culture medium was removed and 70 pfu/well of virus in either PBS for hepatocytes or FCS-supplemented PBS for HepG2 were added to each well. The infected cells were incubated 30 min at 37° C. with 5% CO.sub.2 and then 1.5 mL/well of culture medium were added. Plates were incubated at 37° C. with 5% CO.sub.2 for 3 days and then stored at −80° C. Then, plates were thawed and wells sonicated 30 seconds with 40% of amplitude before titration on Vero cells.

[0363] Replication Rate in Normal Human Hepatocytes:

[0364] Normal hepatocytes were chosen to monitor the capacity of COPTG19384 in normal human cells as these primary cells can be obtained regularly directly from donors. In those cells, the COP WT spread well with a replication rate of more than 50,000 (FIG. 20A). In other words, each initial infection viral particle produced about 50,000 new viruses. In the case of the two recombinant double deleted viruses (i.e. VVTG17137 and COPTG19384), this replication rate was dramatically reduced to 5 to 15 according to the virus or batch of virus (FIG. 20A). This last result indicates that the attenuated replication toward normal cells brings along by the two deletions was conserved between VVTG17137 and COPTG19384.

[0365] Replication Rate in Tumoral Cells HepG2:

[0366] HepG2 cells were chosen to monitor the capacity of COPTG19384 to replicate in tumoral human cells as these cells are a malignant counterpart of normal hepatocytes. In those cells, the five viruses tested had a quite similar replication rate reaching about 100,000 new viruses whatever the initial virus tested (FIG. 20B). Therefore, the double deletion in both VVTG17137 and COPTG19384 and vectorization of transgenes did not impair their capacity to replicate in malignant cells.

[0367] Therapeutic Index:

[0368] As illustrated in FIG. 20C,the calculated index is only two for COP WT indicating a poor selectivity of the COP WT for the tumoral versus normal cells. In contrary, for both VVTG17137 and COPTG19384 (and for both lots of virus tested) this index varies from 8.2 E+03 to 1.8 E+04. This confirms that the two recombinant viruses have the same good selectivity toward tumoral versus normal cells.

[0369] These results demonstrated that COPTG19384 and VVTG17137 have a very similar replicative properties on both tumoral and healthy cells. Compared to COP WT, their replication on tumoral Hep G2 is similar whereas it is highly impaired on healthy hepatocytes. Therefore, the deletion of the two genes (J2R and 140 restricts the replication of the deleted virus to the multiplying cells (i.e. with high nucleotide pool) including the tumoral cells. Since the transgenes expression and the replication are tightly linked, COPTG19384 is an efficient vector for the selective delivery of therapeutic proteins into tumour.

Replication Assays on CEF and LoVo:

[0370] Replication of COPTG19384 was evaluated on CEF (producer cells) isolated from 11 or 12 day-old embryonated Specific Pathogens Free eggs (Charles Rivers) and on a tumoral human cell line (LoVo; ATCC® CCL-229™). CEF and LoVo cells were prepared in suspension and infected at MOI of 10.sup.−3 for CEF and 10.sup.−2 for LoVo (three wells per cells and per time point). After different times of incubation, Viral titration was done on Vero cells (CCL-81™). COPTG19384 replication was compared to the one of VVTG17137 as benchmark. The results show that the replication of COPTG19384 and VVTG17137 were similar in both CEF and LoVo (data not shown).

Replication Assays on Reconstructed Human Skin:

[0371] Replication of COPTG19384 was also evaluated on reconstructed human skin (T-Skin™/Human Full Thickness Skin Model). Thirty-six T-Skin™ samples obtained from (EPISKIN SA) were cultured in 6-well plates and maintained in fresh culture medium. VVTG17137 and COPTG19384 were distributed into each well (in triplicate) in order to obtain the attended final concentration (i.e. 10.sup.1 to 10.sup.5 pfu/well). A negative control corresponding of medium without virus was also tested (Mock). Plates were incubated at 37° C. with 5% CO.sub.2 for 7 days and T-Skin™ samples were collected and cut in two pieces. Infectious titer was determined on one of the two pieces using Vero cells for virus titration. FIG. 21 shows that COPTG19384 replicates in reconstructed skin to the same extent as the benchmark VVTG17137 supporting the fact that the vectorization of both GM-CSF and 4-E03 mAb did not modify the replication behavior of the vaccinia virus on the human reconstructed skin.

Oncolytic Assay

[0372] Oncolytic activity is representative of the lytic activity of the tested viral samples on tumor cells. It was assessed by quantification of cell viability after 5 days of incubation on different tumor cell lines:the human colorectal adenocarcinoma cell line LoVo (ATCC® CCL-229™), the human pancreatic tumor cell line MIA PaCa-2 (ATCC® CCL-1420) and the human hepatocarcinoma cell line HepG2 (ATCC® HB-8065™). COPTG19384 oncolytic activities were compared to the ones of VVTG17137 as benchmark. A negative control corresponding of uninfected cells was also plated (Mock infected cells).

[0373] Cells were prepared, distributed in Eppendorf tubes (1.2×10.sup.6 cells/tube) before being infected with the virus at a MOI of 10.sup.−6 to 10.sup.−2 and incubated 30 min at 37° C. Appropriate complete medium was added to Eppendorf tube and an aliquote of this suspension was added in each well (in triplicate) in 6-well plate containing 2 mL of appropriate complete medium. Plates were incubated at 37° C. with 5% CO.sub.2 for 5 days and cell viability was determined on Vi-Cell counter. Results were expressed as a percentage of the cell viability of mock infected cells. The cell supernatants were also recovered for the determination of concentration of 4-E03 mAb and GM-CSF. FIG. 22 shows that the oncolytic activities of COPTG19384 and VVTG17137 are similar in the three tumoral cell lines assessed.

Level of Expression of Transgenes

[0374] The levels of expression of both 4-E03 monoclonal antibody and GM-CSF were measured by ELISA (as described in Example 6) in culture supernatants of HepG2 and LoVo cells recovered after oncolysis activity determination (5 day of infection at variable MOI).

[0375] The levels of expression of 4-E03 and GM-CSF were also measured by ELISA (see example 6) in supernatants of 5 cell lines cultured in the following conditions, respectively the human gastric carcinoma cell line Hs-746 T (ATCC® HTB-135T™), the human ovarian tumor cell line SK-OV-3 (ATCC® HTB-77T™), the human pancreatic tumor cell line MIA PaCa-2 (ATCC® CCL-1420), the human colorectal adenocarcinoma cell line LoVo (ATCC® CCL-229™) and the human colorectal carcinoma cell line HCT 116 (ATCC® CCL-247T™). Each cell line was cultured (in triplicate) in 6-well plates (10.sup.6 cells/well) and incubated at 37° C. with 5% CO.sub.2 for 24 h before being infected at MOI 0.05. The cell supernatants were then recovered 48 h post infection for the determination of concentration of 4-E03 mAb and GM-CSF. As expected, the MOI, time post-infection and cell line are important parameters that impact the level of transgenes expression in supernatants of infected cells. FIG. 23A shows that a rather permissive to replication (HepG2) and a rather resistant (LoVo) tumoral cell line, when infected by COPTG19384, are able to produce in their culture supernatants approximately the same amount of 4-E03 mAb and GM-CSF. However, the maximum of expression for the HepG2 is reached at a MOI 10-fold lower than for the LoVo. Moreover, for the 5 tumoral cell lines tested, the expression of transgenes was above 0.1 and above 1 μg/mL for 4-E03 mAb and GM-CSF respectively (FIG. 23B). To be noted that the ELISA assay used to measure the concentration of 4-E03 mAb uses the antigen CTLA4 to capture the antibody. In other words, the antibodies measured by this assay are at least partially functional (i.e. recognizing their antigen, the other functions of the antibody are carried by the Fc part),

Purification of 4-E03 mAb and Glycosylation Profile Analysis

[0376] In order to produce a rather large quantity of 4-E03 mAb from infected cells, 15 F175 flasks containing ˜4.7 10.sup.7 Mia-PACA cells/flask were infected at MOI 0.01 with COPTG19384 and incubated 72 h. MIA Paca-2 cell culture supernatant (about 450 mL containing 670 μg of mAb 4-E03 as determined by ELISA) was harvested, pooled and clarified by centrifugation to remove most of the cell debris. Cleared supernatants were filtered on 0.2 μm filters and 2 mM EDTA (to inhibit putative metal proteases) and 20 mM Tris pH7.5 (to raise the pH) were added. The filtered supernatant was then passed through a protA Hitrap column (GE healthcare, ref 17-5079-01). The column was transferred and connected to Purifier FPLC (GE Healthcare) and purification program (THM/ProtA 1 mL injection loop frac bleu) was applied. The eluted fractions containing the mAb were loaded on NuPage Bis-Tris gels 4-12% (Thermo NP0323) after addition of Laemlli buffer (Biorad) containing or not beta-mercaptoethanol for reduction or not of the disulphide bonds of the mAb. The gel was stained with InstantBlue (Expedeon, ISB1L). Three fractions corresponding to the main peak of elution were pooled and, after dialysis against the formulation buffer, the antibody concentration was determined by absorbance at 280 nm. The final concentration of the purified mAb was 0.29 mg/mL.

[0377] The first characterization was an assessment of chains assembly by electrophoresis in reducing and non-reducing conditions. In non-reducing conditions the 2 light and 2 heavy chains assemble to form the native and functional antibody. It appeared that the purified 4-E03 from infected MIA PaCa-2 and the recombinantly produced 4-E03 have an undistinguishable electrophoresis profiles in both reducing and non-reducing conditions. In other words, the purified 4-E03 from infected MIA PaCa-2 has the expected ratio of light and heavy chains and is correctly assembled into 2 light and 2 heavy chains heterotetramer. The presence of this heterotetramer was also confirmed by mass spectrometry. The purified mAb was subject to mass spectrometry for glycosylation analysis. Briefly, the mAb was digested, or not, with IdeS protease that cleave specifically the IgG at the hinge (resulting in F(ab′).sub.2 and Fc parts). The masses of the whole antibody or Fc parts that carry the N-glycosylation were determined and each mass was fitted with a theoretical mass calculated from the primary sequence of Fc and a glycosylation pattern. The glycosylation profile of 4-E03 mAb purified from infected MIA PaCa-2 was compared to the one of the recombinantly produced and purified 4-E03 and MabThera as benchmark for human IgG1 used in clinic. The results show that the glycosylation profiles of 4-E03 produced from infected MIA PaCa2 had a glycosylation profile different from the two antibody references with a majority of G0F (88%) whereas both recombinant 4-E03 and MabThera have a similar glycosylation profile with the typical G0F, G1F and G2F distribution. However, MIA PaCa-2 was suspected to be the cause of the low G1F and G2F species in the purified 4-E03 mAb due to a low level of Beta-1,4-Galactosyltransferase 1 transcript which could be the cause of the lack of galactosyl moiety (and therefore a lack of G1f and G2F) of the 4-E03 expressed in MIA PaCa-2.

[0378] The same type of purification followed by mass analysis was also performed from permeate recovered during the purification of the COPTG19384 produced on CEF. The results show that the glycosylation profiles of 4-E03 from infected CEF was very similar to the ones of MabThera or recombinant 4-E03. This latest result suggests that the glycosylation profile of the antibody is more impacted by the cell line used than by the infection itself.

[0379] 4-E03 purified from the supernatant od infected MIA PaCa-2 cells (4-E03 TG) also exhibit the same binding characteristics as recombinantly produced 4-E03 by CHO (research batch) or HEK (tox batch) cells (FIGS. 24 and 25). This was demonstrated by ELISA (described in Example 1) to test the binding to recombinant (FIG. 24A) human and (FIG. 24B) cynomolgus CTLA-4 protein. A FAGS analysis where the binding to (FIG. 25A) human and (FIG. 25B) cynomolgus CTLA-4 expressing cells was tested (see Example 1 and 2) confirmed similar cross-reactivity and binding affinities for the different 4-E03 batches.

GM-CSF Glycosylation and Disulphide Bonds Pattern

[0380] In order to investigate the glycosylation pattern and the presence of some disulphide bonds, different human tumoral cells line were infected by COPTG19384 and their supernatants were analysed by the same WB method. MIA-Paca-2, LoVo, HepG2 and HCT116 cells were infected at MOI 0.01 and incubated 72 h in culture medium without serum. The culture supernatants were harvested, clarified by centrifugation and then filtered on 0.2 μm filter. The supernatants were stored at −20° C. until analysis. They were treated by addition 8 μl of Rapid PNGase F Buffer 5× followed by an incubation at 75° C. for 5 minutes. One μL of PNGase F was then added (to remove N-glycans from glycoprotein) and the mixture incubated 30 minutes at 50° C. Twenty-five μL of samples were prepared by addition of 5 μL Laemmli buffer×4 with or without beta-mercaptoethanol (reducing and non-reducing conditions) before being submitted to western blotting. Immune complexes were detected using the Amersham ECL Prime Western Blotting and chemiluminescence was recorded with a Molecular Imager ChemiDOC XRS (Biorad).

[0381] GM-CSF from infected HCT116, LoVo and MIA PaCa-2 displayed the same pattern of glycosylation whereas the GM-CSF produced by infected HepG2 migrated as a non-N-glycosylated molecule. These results indicate that the GM-CSF produced by the COPTG19384 infected human tumoral cells has the expected post-translational modifications (i.e. disulphide bonds and N-glycosylation). However, these modifications probably vary depending of the tumour cell lines used for infection and their specific metabolic status.

Example 8: Pharmacokinetics after Intratumoral Injections of COPTG19384

[0382] Kinetic of Expression in Tumor and Bloodstream of Anti-CTLA4 Antibodies, GM-CSF and Viruses after Intratumoral (i.t.) Injection of Vaccinia Viruses in a LoVo Xenografted Model.

[0383] Protocol

[0384] 5×10.sup.6 cells LoVo cells were implanted in right flank of Swiss nude mice (Charles River, France). After about two weeks when the tumours volume reached ˜120 mm.sup.3, the mice were randomized and split in 6 groups of 15 animals. [0385] Mice from group 1 received an i.t. administration of COPTG19384 at a dose of 1×10.sup.4 pfu/mouse at D0 (first day of treatment). [0386] Mice from group 2 received an i.t. administration of COPTG19384 at a dose of 1×10.sup.5 pfu/mouse at D0. [0387] Mice from group 3 received an i.t. administration of VVTG17137 at a dose of 1×10.sup.4 pfu/mouse at D0. [0388] Mice from group 4 received an i.t. administration of VVTG17137 at a dose of 1×10.sup.5 pfu/mouse at D0. [0389] Mice from group 5 received an intraperitoneal (i.p.) administration of 4-E03 at a dose of 3 mg/kg at D0. [0390] Mice from group 6 received an i.p. administration of Ipilimumab (Yervoy) at a dose of 3 mg/kg at D0.

[0391] Tumour and blood from 3 animals were collected at days 1, 3, 6, 10 and 20. Tumors were weighted and homogenized for immediate processing. One quarter of the homogenized tumours was collected for virus titration and the remaining suspension was centrifuged and supernatants were stored at −20° C. until use. Blood was split in two parts: one was added to heparin tube (25 IU/100 μL of blood) for titration assay and frozen at −80° C. until analysis. Clarified sera were produced from the other part and stored at −20° C. until use. Virus titer was determined in tumor and blood samples by titration on Vero cells.

Kinetic of Virus Replication in LoVo Model:

[0392] In the LoVo model, where COPTG19384 was injected once at two doses (1×10.sup.4 or 1×10.sup.5 pfu) the virus replication was monitored and compared to the one of VVTG17137 injected in same conditions. The results displayed in FIG. 26 show an important dispersion of the three values of virus titers measured for each time point. Anyway, the results show also that both viruses and at both doses replicate in the tumour and maintain a rather high titer/g of tumour from day 3 to up to 20 days after the injection. There is no obvious difference of virus titer, for a given time point, between the two doses of virus or between the two viruses used. Interestingly, all the blood samples were negative for virus detection excepted one sample (VVTG17137, dose: 1×10.sup.7 pfu at Day 10) for which only 13 pfu/mL were detected (data not shown).

[0393] Together these results indicate that, after one i.t. injection of either 1×10.sup.4 or 1×10.sup.5 pfu, the virus' replication was maintained in LoVo tumours for at least 20 days with a barely detectable presence in the blood stream. It has to be noted that the LoVo xenografted model is very favourable to the virus replication as it uses permissive human tumour cells and Swiss nude mice that have a severely impaired immune system with, therefore, a limited anti-viral activity.

Kinetic of Transgenes Expression in LoVo Model:

[0394] As expected, the kinetic of transgenes expression in tumour followed kinetic of virus replication with a maximum concentration (Cmax) at days 6 or 10 for both 4-E03 mAb (FIG. 27A) and GM-CSF (FIG. 27B). In the case of the single injection of 4-E03 mAb (or Ipilimumab), the Cmax in tumour and blood were observed at the first time point (day 1) and the thereafter measured concentrations of mAb were in accordance with the pharmacokinetic of a human IgG1 in mouse (FIG. 28).

[0395] Moreover, the concentrations of 4-E03 into the tumour at Cmax and thereafter (i.e. from 6-10 to 20 days after injection) was around 10-fold higher following COPTG19384 treatment (for both doses) than after a single injection of 4-E03 mAb at a therapeutic dose of 3 mg/kg (FIG. 27A). In contrast, the concentration in blood of the mAb after COPTG19384 treatment was always inferior to those measured after i.p. injection of 3 mg/kg of 4-E03 (FIG. 28A). This result indicates that vectorization of mAb allows to reach high concentration into the tumour without exceeding or even reaching the blood concentration obtained at therapeutic dosing of mAb.

[0396] The kinetic of expression of GM-CSF after COPTG19384 treatment follows the one observed with 4-E03 (FIG. 27B). Interestingly, the levels of GM-CSF measured into the tumour are below the level of 4-E03 for the same samples, although in vitro LoVo infected by COPTG19384 express about two-fold more GM-CSF than 4-E03. The blood concentrations of GM-CSF were also very low compared to the 4-E03 ones (FIG. 28B). This result is in accordance with the in vivo half-life of GM-CSF that is very short compared to the one of a human IgG1.

[0397] These results indicate that the vectorized antibodies and GM-CSF are expressed mainly in the tumour after i.t. injection of COPTG19384 with a minimal systemic exposure. These results confirm that vectorization is particularly suitable for transgenes with toxicological (e.g. anti-CTLA4) or pharmacokinetic (e.g. GM-CSF) issues.

Kinetic of Expression in Tumor and Bloodstream of Anti-CTLA4 Antibodies, GM-CSF and Viruses after Intratumoral Injection of Vaccinia Viruses in a CT26 Syngenic Model

[0398] Evaluation of the viral activities in the CT26 immunocompetent murine model requires the generation of several surrogate viruses encoding a murine anti-mCTLA4 with or without the murine GM-CSF: [0399] COPTG19407 is a Vaccinia virus (Copenhagen strain) containing an expression cassette encoding the heavy chain of the murine m5-B07 IgG2 (SEQ ID NO: 63) under p7.5 promoter at the J2R locus, and an expression cassette encoding the light chain of the m5-B07 (SEQ ID NO: 62) under p7.5 promoter and murine GM-CSF (SEQ ID NO: 58) under pSE/L promoter at the I4L locus. [0400] COPTG19421 is a Vaccinia virus (Copenhagen strain) containing an expression cassette encoding the m5-B07 heavy chain under p7.5 promoter at the J2R locus, and an expression cassette encoding the light chain of the m5-B07 under p7.5 promoter at the I4L locus. [0401] VVTG18058, used as benchmark, is a Vaccinia virus (Copenhagen strain) deleted in J2R and I4L genes, without any transgene (“empty” virus).

[0402] These vaccinia viruses were generated as for the human counterparts by two successive homologous recombinations at J2R (TK) and then I4L (RR) loci following the process described in example 6. The ELISA method to quantify the m5-B07 antibody and mGM-CSF was similar to that described above (example 6, “expression of transgene” for 4-E03 and GM-CSF) except that murine CTLA4-Fc antigen was used to capture the murine antibody and Quantikine ELISA kit Mouse GM-CSF (R&D Systems) was used to quantify mGM-CSF. Oncolytic activity of these virus was also evaluated in various cell lines (one sarcoma: MCA205 and two colon carcinoma CT26 and MC38) and found similar to the one of VVTG18058 showing that vectorization of the murine antibody with or without the mGM-CSF did not impact the oncolytic abilities of the vaccinia virus (data not shown).

[0403] Protocol:CT26 cells (2×10.sup.5 cells) were implanted in right flank of Balb/c mice (Charles River, France). After about one week when the tumours volume reached 25-50 mm.sup.3 the mice were randomized and split in 3 groups of 20 animals (groups 1 to 3) and one group 4 of 10 animals. Tumor and blood were collected and treated as described for the LoVo model except that they were collected at days 1, 4, 8 and 10 for the first 3 groups and day 1 for the group 4. [0404] Mice from group 1 received an i.t. administration of VVTG18058 at a dose of 1×10.sup.7 pfu/mouse at D0, D2 and D4. [0405] Mice from group 2 received an i.t. administration of COPTG19407 at a dose of 1×10.sup.7 pfu/mouse at D0, D2 and D4. [0406] Mice from group 3 received an i.t. administration of COPTG19421 at a dose of 1×10.sup.7 pfu/mouse at D0, D2 and D4. [0407] Mice from group 4 received an i.p. administration of m5-1307 at a dose of 3 mg/kg at D0.

Kinetic of Virus Replication in CT26 Model:

[0408] In the CT26 model, where two surrogate viruses were injected thrice (1×10.sup.7 pfu/injection) the virus replication was monitored and compared to the one of VVTG18058 injected in the same conditions. The results displayed in FIG. 29 show, as for the LoVo model, an important dispersion of the three values of virus titers measured for each time point. However, the titers for the three viruses were maintained over the time and up to 10 days, indicating that the two transgenes did not impact the virus clearance or replication, at least in this window of time. No virus infectious particle was detected in any of the blood samples (data not shown).

Kinetic of Transgenes Expression in CT26 Model

[0409] Like for the LoVo model, the transgenes expression into the tumour mirrored the virus replication. In other words, the m5-B07 antibody (FIG. 30A) and mGM-CSF (FIG. 30B) were detected in the tumour at a rather constant level over the 10 days of the monitoring.

[0410] In the case of the monoclonal antibody, the Cmax reached after either COPTG19421 or COPTG19407 injections were about 10-fold lower than the Cmax observed with a single i.p. injection of the m5-B07 antibody at 3 mg/mL (FIG. 30A). In the serum, the difference was even more pronounced with a circulating concentration of m5-B07 ˜100-fold lower after virus treatments versus m5-B07 injection at 3 mg/kg (FIG. 31).

[0411] For GM-CSF, only the treatment by COPTG19407 yielded to measurable concentration of mGM-CSF in CT26 tumors indicating that the measured cytokine had a recombinant, rather than endogen, origin. Like in LoVo model, the mGM-CSF concentrations measured in tumour were lower than the ones of m5-B07 (FIG. 30B). Moreover, the mGM-CSF produced by the tumour was not detectable in any of sera sample probably due to a short half-life of the molecule that preclude any systemic accumulation.

Example 9: Antitumoral Activity Studies

[0412] COPTG19347 is a Vaccinia virus (Copenhagen strain) deleted in J2R and I4L genes and encoding a whole murine antibody (namely m5-B07, heavy and light chain) recognizing the murine CTLA4 antigen. COPTG19421 versus COPTG19347 expressed both m5-B07 but under different promoters: namely p7.5K and pH5.R respectively. Quantification of m5-B07 was assessed in supernatants of infected cells reaching about 1 μg/mL in infected CT26 at MOI 10.sup.−1 to about 4 μg/mL in infected MCA205 cells at MOI 10.sup.−2. The higher expression in MCA205 versus CT26 was observed also for mGM-CSF in the culture supernatant of cells infected by COPTG19407 (data not shown).

Antitumoral Activity in Mice Bearing CT26 Model in Combination with Anti-PD1 Protocol:

[0413] CT26 cells (2×10.sup.5 cells) were implanted in right flank of Balb/c mice (Charles River, France). When tumors reached a volume of 25-50 mm.sup.3, the mice were randomized in five groups of ten animals. Briefly, the mice were treated by three i.t. administrations, 2 days apart, of virus followed of i.p. treatment of murine anti-PD1 (RMP1-14 BioXcell) twice a week for three weeks. More specifically, [0414] Mice from group 1 received vehicle; [0415] Mice from group 2 received an i.t. administration of 1×10.sup.7 pfu of COPTG19347 at D0, D2 and D4; [0416] Mice from group 3 received an i.t. administration of 1×10.sup.7 pfu of COPTG19347 at D0, D2 and D4 and i.p. intraperitoneal administration of 250 μg/mice of RMP1-14, at D7, D11, D14, D18 and D22; [0417] Mice from group 4 received an i.p. administration of 250 μg/mice of RMP1-14, at D7, D11, D14, D18 and D22; [0418] Mice from group 5 received an i.t. administration of 1×10.sup.7 pfu of VVTG18058 at D0, D2 and D4.

[0419] Tumor dimensions were measured twice a week with calipers and their volumes calculated using the formula (π/6)(length×width.sup.2). The animals were euthanized when their tumor volume reached 2000 mm.sup.3.

Antitumoral Activities of COPTG19347 in CT26 Model

[0420] As illustrated in FIG. 32, COPTG19347 treatment yielded not only to tumour growth inhibition (FIG. 32A) but also to tumour regression that finally translate into tumour free mice that survive up to 100 days (FIG. 32B). Treatment with COPTG19347 yielded to 60% tumour free mice at day 100. The co-treatment with anti-PD-1 antibody did not improve significantly the tumour growth inhibition or the percentage of long surviving mice (about 70% tumour free mice at day 100). In comparison, RPMI-14 treatment did not provide any anti-tumor effect (same behavior as non-treated mice (all dead within the first 40 days) whereas VVTG18058 had a poor activity (about 10% tumour free mice at day 100).

Dose-Effect Evaluation in CT26 Model:

[0421] The three surrogate viruses were compared (different promoter to drive m5-B07 and with or without m-GM-CSF) and a dose escalation of COPTG19407 versus VVTG18058 was performed (7.5×10.sup.4, 7.5×10.sup.5 or 7.5×10.sup.6 pfu). The experimental conditions were exactly as the ones described above except that the co-treatment with anti-PD1 was omitted.

[0422] The results of two independent experiments demonstrated clearly that the three viruses tested COPTG19407, COPTG19421 and COPTG19347 had a strong anti-tumoral activity at the dose of 7.5×10.sup.6 pfu. This confirm that neither the mGM-CSF encoded in COPTG19407 nor the use of a weakest promoter in COPTG19421 and COPTG19407 did impair the anti-tumoral activity of the armed viruses. At the highest dose tested (i.e. 7.5×10.sup.6 pfu), the number of tumor free mice at 80 days were between 5/10 and 7/10 depending of the virus and of experiment versus 0/10 for the mice treated with the empty virus as summarized in the following Table.

TABLE-US-00006 TABLE 6 Effect of the surrogate viruses COPTG19407, COPTG19421 and COPTG19347 on the tumor growth after i.t. injections Number of tumor free Virus mice at name TK locus RR locus Dose (pfu) D 100 COPTG19407 p7.5K -HC* p7.5K -LC*; 7.5 × 10.sup.6 7/10 pSE/L-GM-CSF COPTG19421 p7.5K -HC* p7.5K -LC* 7.5 × 10.sup.6 5/10 COPTG19347 pH5.R -HC* pH5.R -LC* 7.5 × 10.sup.6 7/10 VVTG18058 — — 7.5 × 10.sup.6 0/10 Mock — 0/10 *HC and LC stand for heavy and light chain of m5-B07 murine anti-mCTLA4 antibody respectively

[0423] Moreover the dose escalation performed with both the “empty” virus (VVTG18058) and the COPTG19384 surrogate (COPTG19407) demonstrated that even at a relative low dose (7.5×10.sup.4 pfu) the antibody expressing virus had still some clear antitumoral activities with 4/10 and 2/10 tumor free mice at 80-98 days versus 0/10 for the treatment with VVTG18058 at the same low dose as shown in the following Table.

TABLE-US-00007 TABLE 7 effect of the dose of COPTG19407 or VVTG18058 on the tumor growth after i.t. injections Number of tumor free Virus mice at name TK locus RR locus Dose (pfu) D 100 COPTG19407 p7.5K -HC* p7.5K -LC*; 7.5 × 10.sup.6 7/10 pSE/L-GM-CSF 7.5 × 10.sup.5 8/10 7.5 × 10.sup.4 4/10 VVTG18058 — — 7.5 × 10.sup.6 0/10 7.5 × 10.sup.5 0/10 7.5 × 10.sup.4 0/10 *HC and LC stand for heavy and light chain of m5-B07 murine anti-mCTLA4 antibody respectively

[0424] A compilation of survival data (global survival plot compiled from two independent studies) are presented in FIG. 33. A statistical analysis, using logrank test, has been performed to figure out if there were significant differences between survival of each group.

Anti-Tumoral Activity of COPTG19407 Compared to the Combination of VVTG18058 Plus m5-B07

[0425] CT26 tumor-bearing mice were set up as described previously (Example 5). Briefly, CT26 cells were injected s.c. into Balb/C mice. The treatment of the mice was started when tumors reached approx. 100 mm.sup.3. Mice were then injected at D0, D2 and D5 with COPTG19407 (8.5×10.sup.6 pfu i.t.), VVTG18058 (8.5×10.sup.6 pfu i.t.), m5-B07 (10 mg/kg i.p.) or the combination of VVTG18058 (8.5×10.sup.6 pfu i.t.) plus m5-B07 (10 mg/kg i.p.). Tumor dimensions were then measured twice a week and mice were euthanized when tumors reached 2000 mm.sup.3. As shown in FIGS. 34A to 34D, tumor growth was significantly inhibited when mice were treated with the virus COPTG19407 expressing anti-CTLA-4 and GM-CSF whereas the combination of the unarmed virus plus anti-CTLA4 m5-B07 did not result in improved therapy compared to single agent use. In the groups treated with only m5-B07, only virus VVTG18058 or the combination of both m5-B07 and VVTG18058, only 20% of the mice survived after day 70 (FIG. 34E). In contrast, 90% of the mice survived more than 100 days after administration of COPTG19407 demonstrating the potency of the vectorization strategy.

Antitumoral Activity of VVTK RR-Encoding Anti-CTLA-4 and GM-CSF in Mice Bearing A20 Subcutaneous Murine B-Cell Lymphoma

[0426] The A20 cell line is a BALB/c B cell lymphoma line derived from a spontaneous reticulum cell neoplasm found in an old BALB/cAnN mouse (ATCC TIB-208™).

[0427] Protocol (1):

[0428] Tumors were induced by subcutaneous injection of 5×10.sup.6 of A20 cells into the right flank of female Balb/cN mice (Charles River, France). When tumors reached a mean volume of 95 mm.sup.3, 50 mice were randomized into 5 groups of ten animals. [0429] Mice from group 1 received an i.t. administration of vehicle at D0, D2 and D4, [0430] Mice from group 2 received an i.t. administration of VVTG18058 at a dose of 4.75×10.sup.6 pfu at D0, D2 and D4, [0431] Mice from group 3 received an i.t. administration of COPTG19407 at a dose of 4.75×10.sup.6 pfu at D0, D2 and D4, [0432] Mice from group 4 received an i.p. administration of anti-PD-1 antibody at a dose of 250 μg at D7, D10, D14, D17, D21 and D24, [0433] Mice from group 5 received an i.t. administration of COPTG19407 at a dose of 4.75×10.sup.6 pfu at D0, D2 and D4 combined with an i.p. administration of anti-PD-1 antibody at a dose of 250 μg at D7, D10, D14, D17, D21 and D24.

[0434] Antitumoral Activity:

[0435] The tumor volumes of all animals was monitored throughout the study. The anti-tumoral activity of treatments is based on the evaluation of the criteria of tumor doubling time, tumor growth delay and tumor growth inhibition (T/C %).

[0436] Tumor doubling time was similar for Groups 1, 2 and 4 ranging from 5.14 days (Group 1) to 6.37 days (Group 2). For Group 3, the tumor doubling time could not be precisely calculated since the tumors did not grow exponentially indicating an increased efficacy of treatment, compared to Groups 1, 2 and 4. Similarly, tumor doubling time was calculated using only one animal in Group 5. 9 out of 10 mice in Group 3 had tumors which regressed on D15 and did not grow substantially in volume from D25 until the end of the study on D64. On D64, tumor volumes of these 9 mice ranged from 4 mm.sup.3 (technical limit of tumor detection) to 59.77 mm.sup.3. Similarly, in Group 5, tumors regressed in 9 mice following the start of treatment, to reach values ranging from 7.24 to 63.21 mm.sup.3 at the end of the study on D64. As it can be seen in FIG. 35 presenting the individual tumor volume curves of each group of treatment (Groups 1 to 5 corresponding to FIG. 35 A to E), tumors did not grow in animals of groups 3 and 5 receiving COPTG19407 confirming the strong anti-tumor activity of the antibody-expressing virus with or without anti-PD1.

[0437] The tumor growth delay was calculated by estimating the time taken for tumors to reach a mean target volume of 300 mm.sup.3. The results were similar to those obtained with tumor doubling time, since this parameter could only be calculated for tumors that reached the target volume of 300 mm.sup.3, which was not the case for the majority of animals in Groups 3 and 5. Groups 1, 2 and 4 had mean tumor growth delays of 16, 21 and 17 days respectively which were not significantly different to each other. In addition, Groups 3 (n=2) had a mean tumor delay of 14 days, indicated that the tumors that did grow in this group, grew at the same rate as Groups 1, 2 and 4 however these tumors did actually regress in both animals. In comparison, the single tumor in Group 5 (n=1) which grew had a significantly (p≤0.0026) longer tumor growth delay than all other groups, of 43 days.

[0438] Tumor growth inhibition (T/C %) was calculated by comparing the median tumor volume of the vehicle treated Group 1 with the other treatment groups. Group 2 had an optimum T/C % of 34% on D22, indicating a transient marginal anti-tumoral activity but this value increased up to 71% by D31. Moderate anti-tumoral activity (10-30% T/C %) was observed in Group 4. In comparison, Groups 3 and 5 both showed marked anti-tumoral activity (T/C % less than 10%) from D27 to D31 (last calculable value of T/C %).

[0439] FIG. 36 illustrates the mean tumor volume curves on BALB/cN mice bearing subcutaneous A20 tumors demonstrating the dramatic effect of the anti-CTLA4 and Gm-CSF expressing virus COPTG19407 with or without anti-PD1 on tumor growth.

[0440] Protocol (2):

[0441] Tumors were induced by subcutaneous injection of 5×10.sup.6 of A20 cells into the right flank of female BALB/cN mice. When tumors reached a mean volume of 80-100 mm.sup.3, 90 animals were randomized into 9 groups of 10 animals. [0442] Mice from group 1 received an i.t. administration of vehicle at D0, D2 and D4 [0443] Mice from group 2 received an i.p. administration of anti-PD-1 antibody at a dose of 250 μg/mouse/injection at D0, D4, D7, D10, D14 and D17 [0444] Mice from group 3 received an i.p. administration of isotype at a dose of 250 μg/mouse/injection at D0, D4, D7, D10, D14 and D17 [0445] Mice from group 4 received an i.t. administration of VVTG18058 at a dose of 1×10.sup.5 pfu at D0, D2 and D4 [0446] Mice from group 5 received an i.t. administration of VVTG18058 at a dose of 1×10.sup.5 pfu at D0, D2 and D4, and an i.p. administration of isotype at a dose of 250 μg/mouse/injection at D0, D4, D7, D10, D14 and D17 [0447] Mice from group 6 received an i.t. administration of VVTG18058 at a dose of 1×10.sup.5 pfu at D0, D2 and D4 and an i.p. administration of anti-PD-1 antibody at a dose of 250 μg/mouse/injection at D0, D4, D7, D10, D14 and D17 [0448] Mice from group 7 received an i.t. administration of COPTG19407 at a dose of 1×10.sup.5 pfu at D0, D2 and D4 [0449] Mice from group 8 received an i.t. administration of COPTG19407 at a dose of 1×10.sup.5 pfu at D0, D2 and D4, combined with an i.p. administration of isotype at a dose of 250 μg/mouse/injection at D0, D4, D7, D10, D14, D17 and D24. [0450] Mice from group 9 received an i.t. administration of COPTG19407 at a dose of 1×10.sup.5 pfu at D0, D2 and D4, combined with an i.p. administration of anti-PD-1 antibody at a dose of 250 μg/mouse/injection at D0, D4, D7, D10, D14 and D17.

[0451] Antitumoral Activity:

[0452] The dose of COPTG19407 is suboptimal and demonstrated a mild antitumoral activity similar to the one of the anti-PD-1 treatment in terms of tumor volume and mice survival.

[0453] In contrast, the combination of COPTG19407 with anti-PD-1 showed an strong anti-tumoral activity resulting in a small tumor volume compared to the other groups as showed in FIG. 37A (approximately 290 mm.sup.3 at Day 36 compared to approximatively 630 mm.sup.3 and 750 mm.sup.3 at day 24 for mice receiving anti-PD-1 alone or COPTG19407 alone respectively), and a much better survival of the animals as presented in FIG. 37B (7 animals still alive at Day 57 in group 9 versus only two or one in groups 2 or 7 respectively).

Antitumoral Activity Study of VVTK-RR-Encoding Anti-CTLA-4 and GM-CSF in Mice Bearing C38 Subcutaneous Colon Tumor Cells

[0454] C38 is a murine colon adenocarcinoma originating from the American Type Culture Collection (ATCC CRL-2779™).

[0455] Protocol:

[0456] Tumors fragments (30-50 mg) were subcutaneously implanted into the right flank of female C57BL/6J mice (Janvier, France). When tumors reached a mean volume of approximately 60 mm.sup.3, 50 animals were randomized into five groups of ten animals. [0457] Mice from group 1 received an i.t. administration of vehicle at D0, D2 and D4, [0458] Mice from group 2 received an i.t. administration of an VVTG18058 at a dose of 4.75×10.sup.6 pfu at D0, D2 and D4, [0459] Mice from group 3 received an i.t. administration of COPTG19407 at a dose of 4.75×10.sup.6 pfu at D0, D2 and D4, [0460] Mice from group 4 received an i.p. administration of the murine anti-PD-1 antibody at a dose of 250 μg at D7, D10, D14, D17, D21 and D24, [0461] Mice from group 5 received an i.t. administration of COPTG19407 at a dose of 4.75×10.sup.6 pfu at D0, D2 and D4 combined with an i.p. administration of anti-PD-1 antibody at a dose of 250 μg at D7, D10, D14, D17, D21 and D24.

[0462] Antitumoral Activity:

[0463] As before, the tumor volumes of all animals was monitored throughout the study. Tumor doubling time was similar for Groups 1 and 2 at approximately 6.7 days. Group 4 (n=5) had a longer tumor doubling time (10.4 days) but there were no significant differences between groups. For Groups 3 and 5, the tumor doubling time was calculable but with fewer animals (n=2) since the majority of tumors on mice in these groups did not grow exponentially indicating an increased efficacy of treatment, compared to Groups 1, 2 and 4. As it can be seen from FIG. 38, 8 out of 10 mice in Group 3 had tumors which regressed from D15 and did not grow substantially in volume, with five mice having no detectable tumors at the end of the study on D61. Similarly, in Group 5, tumors regressed in 8 mice following the start of treatment, to reach values ranging from 0 (n=2) to 47.82 mm.sup.3 at the end of the study on D61.

[0464] The tumor growth delay was calculated by estimating the time taken for tumors to reach a mean target volume of 300 mm.sup.3. The results were similar to those obtained with tumor doubling time, since this parameter could only be calculated for tumors that reached the target volume of 300 mm.sup.3, which was not the case for the majority of animals in Groups 3 and 5. There were no significant differences between groups. Groups 1, 2 and 4 (n=5) had mean tumor growth delays of 23 to 27 days respectively. In addition, Groups 3 (n=2) and 5 (n=3) had mean growth delays of 18 and 24 days, respectively. This indicated that the tumors that did grow in these two groups, grew at similar rates as those in Groups 1, 2 and 4.

[0465] Tumor growth inhibition (T/C %) was calculated by comparing the median tumor volume of the vehicle treated Group 1 with the other treatment groups. Group 2 did not show any tumor growth inhibition, since T/C % s remained above 100% for the duration of the study. In comparison, Groups 3, 4 and 5 all showed marked anti-tumoral activity (T/C % less than 10%) from D31 (Group 3 only) to D42 (last calculable value of T/C %). FIG. 39 illustrates the mean tumor volume curves of C57BL/6 mice bearing subcutaneous C38 tumors demonstrating the dramatic effect of the anti-CTLA4/GM-CSF expressing virus COPTG19407 with or without anti-PD1 on tumor growth.

Antitumoral Activity Study of VVTK-RR-Encoding Anti-CTLA-4 and GM-CSF in Mice Bearing EMT6 Subcutaneous Breast Tumor Cells

[0466] EMT6 is a murine breast carcinoma originating from the ATTC (ATCC CRL-2755™)

[0467] Protocol:

[0468] Tumors were induced by subcutaneous injection of 1×10.sup.6 EMT6 cells in female BALB/cByJ mice (Charles River, France). When tumors reached a mean volume of approximately 51 mm.sup.3, fifty mice were randomized by individual tumor volume into five groups of ten animals. [0469] Mice from group 1 received an i.t. administration of vehicle at D0, D2 and D4, [0470] Mice from group 2 received an i.t. administration of VVTG18058 at a dose of 4.75×10.sup.6 pfu at D0, D2 and D4, [0471] Mice from group 3 received an i.t. administration of COPTG19407 at a dose of 4.75×10.sup.6 pfu at D0, D2 and D4, [0472] Mice from group 4 received an i.p. administration of anti-PD-1 antibody at a dose of 250 μg at D7, D10, D14, D17, D21 and D24, [0473] Mice from group 5 received an i.t. administration of COPTG19407 at a dose of 4.75×10.sup.6 pfu at D0, D2 and D4 combined with an i.p. administration of anti-PD-1 antibody at a dose of 250 μg at D7, D10, D14, D17, D21 and D24.

[0474] The tumor volumes of all animals was monitored throughout the study by evaluating the criteria of tumor doubling time, tumor growth delay and tumor growth inhibition (T/C %).

[0475] Antitumoral Activity:

[0476] Tumor doubling time was similar for Groups 1, 2 and 4 at approximately 5.4 days. For Group 3, the tumor doubling time could not be calculated since the majority of tumors did not grow exponentially indicating an increased efficacy of treatment, compared to Groups 1, 2 and 4. A similar effect was observed in Group 5, where only one animal was used for the calculation of tumor doubling time. As can be seen on the graphs of individual tumor volume (FIG. 40), 8 out of 10 mice in Group 3 had tumors which regressed from D15 and did not grow substantially in volume, with seven mice having no detectable tumors at the end of the study on D61. Similarly, in Group 5, tumors regressed in 9 mice following the start of treatment, to reach values ranging from 0 (n=8) to 13.24 mm.sup.3 at the end of the study on D56.

[0477] The tumor growth delay was calculated by estimating the time taken for tumors to reach a mean target volume of 200 mm.sup.3. The results were similar to those obtained with tumor doubling time, since this parameter could only be calculated for tumors that reached the target volume of 200 mm.sup.3, which was not the case for the majority of animals in Groups 3 and 5. Groups 1, 2 and 4 had mean tumor growth delays of approximately 19 days. In addition, Groups 3 and 5 (n=2 for both groups) had mean growth delays of 24 and 12 days, respectively. This indicated that the tumors that did grow in these two groups, grew at similar rates as those in Groups 1, 2 and 4. There were no significant differences between groups.

[0478] Tumor growth inhibition (T/C %) was calculated by comparing the median tumor volume of the vehicle treated Group 1 with the other treatment groups. Group 2 showed transient marginal tumor growth inhibition on D28 but increased to 79% on D31. Group 4 showed no anti-tumoral activity, with a T/C % >60% for the duration of the study. In comparison, Groups 3 and 5 both showed marked anti-tumoral activity (T/C % less than 10%) from D24 to D31 (last calculable value of T/C %).

[0479] FIG. 41 illustrates the mean tumor volume curves on BALB/cByJ mice bearing subcutaneous EMT6 tumors demonstrating the dramatic effect of the anti-CTLA4/GM-CSF expressing virus COPTG19407 with or without anti-PD1 on tumor growth.

CT26 Rechallenge

[0480] Balb/c mice challenged with CT26 tumor cells who survived after treatment with 10.sup.4, 10.sup.5 or 10.sup.6 pfu of COPTG19421 or 10.sup.6 pfu of COPTG19407 were rechallenged with CT26 tumor cells or challenged with Renca cells (renal adenocarcinoma cells:control), in order to study if a specific antitumoral immune response was raised.

TABLE-US-00008 TABLE 8 Effect of the rechallenge with CT26 tumor cells or challenge with Renca tumor cells on the number of tumor free mice Group Tumor free/Total mice CT26 control naïve 0/5 VVTG18058 10.sup.5 + CT26 rechallenge 0/1 COPTG19421 10.sup.6 + CT26 rechallenge 3/3 COPTG19421 10.sup.6 + RenCa challenge 0/2 COPTG19407 10.sup.4 + CT26 rechallenge 1/1 COPTG19407 10.sup.4 + RenCa challenge 0/1 COPTG19407 10.sup.5 + CT26 rechallenge 1/2 COPTG19407 10.sup.5 + RenCa challenge 0/2 COPTG19407 10.sup.6 + CT26 rechallenge 2/4 COPTG19407 10.sup.6 + RenCa challenge 0/3

[0481] Results presented in table 8 show that 0/8 mice having received COPTG19421 or COPTG19407 were tumor free after RenCa challenge, while 7/10 mice having received COPTG19421 or COPTG19407 were tumor free after CT26 rechallenge. This indicates that COPTG19421 and COPTG19407 raised a specific immune memory against CT26 cells.