Mutated humanized 12G4 antibodies and the fragments thereof against the human anti-Mullerian hormone receptor type II

Abstract

Novel mutated humanized 12G4 antibodies, and fragments thereof, directed against the anti-Mllerian hormone type II receptor.

Claims

1. A nucleic acid comprising or consisting of a sequence encoding for the light chain of a mutated humanized 12G4 monoclonal antibody and/or a sequence encoding for the heavy chain of said mutated humanized 12G4 monoclonal antibody, said mutated humanized 12G4 monoclonal antibody comprising or consisting of: a) a light chain comprising or consisting of: a variable region (VL) according to amino acid sequence SEQ ID NO: 2 or SEQ ID NO: 4, said VL consisting of three regions determining antigen recognition (CDR) surrounded by four framework regions (FR), and a constant region according to amino acid sequence SEQ ID NO: 6 or a sequence having at least 80% homology with SEQ ID NO: 6 , and b) a heavy chain comprising or consisting of: a variable region (VH) according to amino acid sequence SEQ ID NO: 8 or SEQ ID NO: 10, and a constant region according to amino acid sequence SEQ ID NO: 12 or a sequence having at least 80% homology with SEQ ID NO: 12, wherein said mutated humanized 12G4 monoclonal antibody comprises at least one mutation selected from the group consisting of: a mutation located in one of the CDR regions of the VL that is a substitution of at least one amino acid selected from the group consisting of S179P, one of E184K, E184G, or E184D, and S182F, a mutation located in one of the FR regions of VL that is a substitution of at least one amino acid selected from the group consisting of: I177T, I132T, A143T, T150A, S158P, L175Q, Y178H, V187A, S192T, G197D, and F212S, and a mutation located in the heavy chain that is a substitution of at least one amino acid selected from the group consisting of: Q1E, Q3E or Q3R, Q6E, A9T, V11A, K12R, K13R, K19E, V20A, one of A24G, A24V or A24T, Q39E, A40V, S31G, L45P, D56N, A76T, A79T, R87G, T58A, Q62R, V67M, I70N, T74A, S77P, A79T, S88P, E89D, F102S, A1031, L110P, and S114T, and said mutated humanized 12G4 monoclonal antibody has a KD for human anti-Mllerian hormone type II receptor (AMHRII) at least equal to that of the chimeric 12G4 monoclonal antibody, comprising or consisting of: a) a light chain consisting of: a variable region according to amino acid sequence SEQ ID NO: 14, and a constant region according to amino acid sequence SEQ ID NO: 6, b) a heavy chain consisting of: a variable region according to amino acid sequence SEQ ID NO: 18, and a constant region according to amino acid sequence SEQ ID NO: 12, for said receptor below 10.sup.7 M.

2. The nucleic acid according to claim 1, wherein the sequence encoding for the light chain comprises or consists of the following sequences: a) a sequence encoding for the variable region of the light chain of SEQ ID NO: 53 in which a substitution of at least one codon permitting the substitution, in one of the CDRs, of one or more of the following amino acids: S179P, E184K, E184G, E184D, S182F has been carried out, or, b) a sequence encoding for the variable region of the light chain of SEQ ID NO: 53 in which: at least one substitution of a codon permitting the substitution, in one of the CDRs, of one or more of the following amino acids: S179P, E184K, E184G, E184D, S182F has been carried out, and at least one substitution of at least one codon permitting the substitution, in one of the FRs, of one or more of the following amino acids: I177T, I132T, A143T, T150A, S158P, L175Q, Y178H, V187A, S192T, G197D, F212S, has been carried out, and a sequence encoding for the constant region represented by SEQ ID NO: 5.

3. The nucleic acid according to claim 1, in which the sequence encoding for the heavy chain comprises or consists of the following sequences: SEQ ID NO: 57 in which a substitution of at least one codon permitting the substitution of one or more of the following amino acids: Q1E, Q3E, Q3R, Q6E, A9T, V11A, K12R, K13R, K19E, V20A, A24G, A24V, A24T, Q39E, A40V, S31G, L45P, D56N, A76T, A79T, R87G, T58A, Q62R, V67M, I70N, T74A, S77P, A79T, S88P, E89D, F102S, A103T, L110P, S114T has been carried out.

4. The nucleic acid according to claim 1 comprising: a light chain encoded by a sequence comprising or consisting of the following sequences: a) a sequence for the variable region of the light chain represented by SEQ ID NO: 53 in which a substitution of at least one codon permitting the substitution, in one of the CDRs, of one or more of the following amino acids: S179P, E184K, E184G, E184D, S182F has been carried out, or b) a sequence encoding for the variable region of the light chain represented by SEQ ID NO: 53 in which: at least one substitution of a codon permitting the substitution, in one of the CDRs, of one or more of the following amino acids: S179P, E184K, E184G, E184D, S182F has been carried out, and at least one substitution of at least one codon permitting the substitution, in one of the FRs, of one or more of the following amino acids: I177T, I132T, A143T, T150A, S158P, L175Q, Y178H, V187A, S192T, G197D, F212S, has been carried out, and a heavy chain encoded by the sequence comprising or consisting of the following sequences: SEQ ID NO: 57 in which a substitution of at least one codon permitting the substitution of one or more of the following amino acids: Q1E, Q3E, Q3R, Q6E, A9T, V11A, K12R, K13R, K19E, V20A, A24G, A24V, A24T, Q39E, A40V, S31G, L45P, D56N, A76T, A79T, R87G, T58A, Q62R, V67M, I70N, T74A, S77P, A79T, S88P, E89D, F102S, A103T, L110P, S114T has been carried out.

5. The nucleic acid according to claim 1, wherein the sequence encoding for the light chain comprises or consists of a sequence encoding for a variable region and a sequence encoding for a constant region selected from the following: a) variable region: SEQ ID NO: 21 or SEQ ID NO: 23, or SEQ ID NO: 29 or SEQ ID NO: 31, or SEQ ID NO: 33 or SEQ ID NO: 35, or SEQ ID NO: 45 or SEQ ID NO: 47, or b) constant region SEQ ID NO: 5.

6. The nucleic acid according to claim 1, wherein the sequence encoding for the heavy chain comprises or consists of a sequence encoding for a variable region and a sequence encoding for a constant region selected from the following: a) variable region: SEQ ID NO: 25 or SEQ ID NO: 27, or SEQ ID NO: 7 or SEQ ID NO: 9, or SEQ ID NO: 37 or SEQ ID NO: 39, or SEQ ID NO: 41 or SEQ ID NO: 43, or SEQ ID NO: 49 or SEQ ID NO: 51, b) constant region SEQ ID NO: 11.

7. The nucleic acid according to claim 1, wherein the sequence encoding for the light chain is selected from the following sequences: SEQ ID NO: 69 or SEQ ID NO: 71, or SEQ ID NO: 77 or SEQ ID NO: 79, or SEQ ID NO: 81 or SEQ ID NO: 83, or SEQ ID NO: 93 or SEQ ID NO: 95, and the sequence encoding for the heavy chain is selected from the following sequences: SEQ ID NO: 73 or SEQ ID NO: 75, or SEQ ID NO: 57 or SEQ ID NO: 59, or SEQ ID NO: 85 or SEQ ID NO: 87, or SEQ ID NO: 89 or SEQ ID NO: 91, or SEQ ID NO: 97 or SEQ ID NO: 99.

8. The nucleic acid according to claim 1, in which the sequence encoding for the light chain and the sequence encoding for the heavy chain are as follows: a) sequence encoding for the light chain SEQ ID NO: 69 or SEQ ID NO: 71, and b) sequence encoding for the heavy chain SEQ ID NO: 73 or SEQ ID NO: 75, or, a) sequence encoding for the light chain SEQ ID NO: 77 or SEQ ID NO: 79, and b) sequence encoding for the heavy chain SEQ ID NO: 57 or SEQ ID NO: 59, or, a) sequence encoding for the light chain SEQ ID NO: 81 or SEQ ID NO: 83, and b) sequence encoding for the heavy chain SEQ ID NO: 85 or SEQ ID NO: 87, or, a) sequence encoding for the light chain SEQ ID NO: 77 or SEQ ID NO: 79, and b) sequence encoding for the heavy chain SEQ ID NO: 89 or SEQ ID NO: 91, or, a) sequence encoding for the light chain SEQ ID NO: 93 or SEQ ID NO: 95, and b) sequence encoding for the heavy chain SEQ ID NO: 97 or SEQ ID NO: 99.

9. An expression vector comprising at least one nucleic acid according to claim 1, said nucleic acid being under the control of the elements permitting its expression.

10. An expression vector according to claim 9, comprising a first nucleic acid selected from the nucleic acids with the following sequences: SEQ ID NO 71, 79, 83, or 95, said first nucleic acid being under the control of the elements permitting its expression, and a second nucleic acid selected from the nucleic acids with the following sequences: SEQ ID NO 59, 75, 87, 91, or 99, said second nucleic acid being under the control of the elements permitting its expression.

11. A host cell or cell line transformed by a nucleic acid according to claim 1 and/or an expression vector comprising at least one of said nucleic acid, said nucleic acid being under the control of the elements permitting its expression.

12. A pharmaceutical composition, and in particular vaccine composition, comprising at least a nucleic acid according to claim 1, or a vector comprising at least one of said nucleic acid, said nucleic acid being under the control of the elements permitting its expression, together with a pharmaceutically acceptable vehicle.

13. A method of treating ovarian cancer associated with the human anti-Mllerian hormone type II receptor, comprising administering to a subject in need there-of an effective amount of: 1) a nucleic acid comprising or consisting of a sequence encoding for the light chain of a mutated humanized 12G4 monoclonal antibody and comprising or consisting of a sequence encoding for the heavy chain of said mutated humanized 12G4 monoclonal antibody, said mutated humanized 12G4 monoclonal antibody comprising or consisting of: a) a light chain comprising or consisting of: a variable region (VL) according to amino acid sequence SEQ ID NO: 2 or SEQ ID NO: 4, said VL consisting of three regions determining antigen recognition (CDR) surrounded by four framework regions (FR), and a constant region according to amino acid sequence SEQ ID NO: 6 or a sequence having at least 80% homology with SEQ ID NO: 6, and b) a heavy chain comprising or consisting of: a variable region (VH) according to amino acid sequence SEQ ID NO: 8 or SEQ ID NO: 10, and a constant region according to amino acid sequence SEQ ID NO: 12 or a sequence having at least 80% homology with SEQ ID NO: 12, wherein said mutated humanized 12G4 monoclonal antibody comprises at least one mutation selected from the group consisting of: a mutation located in one of the CDR regions of the VL that is a substitution of at least one amino acid selected from the group consisting of S179P, one of E184K, E184G, or E184D, and S182F, a mutation located in one of the FR regions of VL that is a substitution of at least one amino acid selected from the group consisting of: I177T, I132T, A143T, T150A, S158P, L175Q, Y178H, V187A, S192T, G197D, and F212S, and a mutation located in the heavy chain that is a substitution of at least one amino acid selected from the group consisting of: Q1E, Q3E or Q3R, Q6E, A9T, V11A, K12R, K13R, K19E, V20A, one of A24G, A24V or A24T, Q39E, A40V, S31G, L45P, D56N, A76T, A79T, R87G, T58A, Q62R, V67M, I70N, T74A, S77P, A79T, S88P, E89D, F102S, A103T, L110P, and S114T, and said mutated humanized 12G4 monoclonal antibody has a KD for human anti-Mllerian hormone type II receptor (AMHRII) at least equal to that of the chimeric 12G4 monoclonal antibody, comprising or consisting of: a) a light chain consisting of: a variable region according to amino acid sequence SEQ ID NO: 14, and a constant region according to amino acid sequence SEQ ID NO: 6, b) a heavy chain consisting of: a variable region according to amino acid sequence SEQ ID NO: 18, and a constant region according to amino acid sequence SEQ ID NO: 12, for said receptor below 10.sup.7M, 2) a vector comprising at least one of said nucleic acid, said nucleic acid being under the control of the elements permitting its expression, or 3) a cell transformed by said nucleic acid and/or an expression vector comprising at least one of said nucleic acid, said nucleic acid being under the control of the elements permitting its expression.

14. A kit comprising at least: a nucleic acid according to claim 1, or a vector comprising at least one of said nucleic acid, said nucleic acid being under the control of the elements permitting its expression, or a cell transformed by said nucleic acid and/or an expression vector comprising at least one of said nucleic acid, said nucleic acid being under the control of the elements permitting its expression, for use in diagnosing a pathology associated with the human anti-Mllerian hormone type II receptor, in particular ovarian cancer.

15. The nucleic acid according to claim 1, wherein: a) the light chain of said mutated humanized 12G4 monoclonal antibody comprises or consists of a variable region according to amino acid sequence SEQ ID NO:2 in which said at least one substitution in said CDR region is selected from the group consisting of S179P, one of E184K, E184G, or E184D, and S182F, or the light chain comprises or consists of a variable region according to amino acid sequence SEQ ID NO:2 in which said at least one substitution in said CDR region is selected from the group consisting of: S179P, one of E184K, E184G, or E184D, and S182F, and said at least one substitution in the FR regions of VL is a substitution of at least one amino acid selected from the group consisting: of I177T, I132T, A143T, T150A, S158P, L175Q, Y178H, V187A, S192T, G197D, and F212S, and a constant region according to amino acid sequence SEQ ID NO : 6, and b) the heavy chain of said mutated humanized 12G4 monoclonal antibody comprises or consists of a variable region according to amino acid sequence SEQ ID NO: 8 in which said at least one substitution is selected from the group consisting of: Q1E, Q3E or Q3R, Q6E, A9T, V11A, K12R, K13R, K19E, V20A, one of A24G, A24V or A24T, Q39E, A40V, S31G, L45P, D56N, A76T, A79T, R87G, T58A, Q62R, V67M, I70N, T74A, S77P, A79T, S88P, E89D, F102S, A1031, L110P, and S114T.

16. The nucleic acid according to claim 1, wherein: a) the light chain of said mutated humanized 12G4 monoclonal antibody comprises or consists of a variable region according to an amino acid sequence selected from the group consisting of: SEQ ID NO: 22 or SEQ ID NO: 24, SEQ ID NO: 30 or SEQ ID NO: 32, SEQ ID NO: 34 or SEQ ID NO: 36, and SEQ ID NO: 46 or SEQ ID NO: 48, and of a constant region according to amino acid sequence SEQ ID NO: 6 , and b) the heavy chain of said mutated humanized 12G4 monoclonal antibody comprises or consists of a variable region according to an amino acid sequence selected from the group consisting of: SEQ ID NO: 38 or SEQ ID NO: 40, SEQ ID NO: 26 or SEQ ID NO: 28, SEQ ID NO: 8 or SEQ ID NO: 10, SEQ ID NO: 42 or SEQ ID NO: 44, and SEQ ID NO: 50 or SEQ ID NO: 52, and of a constant region the according to amino acid sequence SEQ ID NO : 12.

17. The nucleic acid according to claim 1, wherein the combination of the light chain and the heavy chain of said mutated humanized 12G4 monoclonal antibody is selected from the group consisting of: the light chain consisting of SEQ ID NO: 70 or SEQ ID NO: 72, and the heavy chain consisting of SEQ ID NO: 74 or SEQID NO: 76, the light chain consisting of SEQ ID NO: 78 or SEQ ID NO: 80, and the heavy chain consisting of SEQ ID NO: 58 or SEQ ID NO: 60, the light chain consisting of SEQ ID NO: 82 or SEQ ID NO: 84, and the heavy chain consisting of SEQ ID NO: 86 or SEQ ID NO: 88, the light chain consisting of SEQ ID NO: 78 or SEQ ID NO: 80, and the heavy chain consisting of SEQ ID NO: 90 or SEQ ID NO: 92, and the light chain consisting of SEQ ID NO: 94 or SEQ ID NO: 96, and the heavy chain consisting of SEQ ID NO: 98 or SEQ ID NO: 100.

Description

DESCRIPTION OF THE FIGURES

(1) FIG. 1 corresponds to a diagrammatic representation of an antibody. The black parts correspond to the constant parts of the heavy chains, the parts in dark grey correspond to the constant part of the light chain, the parts in light grey correspond to the variable part of the heavy chain, and the white parts correspond to the variable part of the light chain. SS represents the disulphide bridges established between two cysteines. The CDR and framework regions are indicated by arrows. The Fab and Fc fragments are also shown.

(2) FIG. 2 corresponds to a diagrammatic string-of-pearls representation of the amino acid sequence of a variable part of a light chain or of a heavy chain of immunoglobulin. The black beads correspond to the amino acids forming the framework regions, and the grey beads correspond to the amino acids representing the CDRs.

(3) FIG. 3 corresponds to a diagrammatic string-of-pearls representation of the amino acid sequence of the variable part of the heavy chain (FIG. 3A: amino acids 1-115, SEQ ID NO: 8) and of the light chain (FIG. 3B: amino acids 131-236, SEQ ID NO: 2) of the humanized 12G4 antibody with the numbering adopted for defining the position of the mutations. The grey-shaded beads correspond to amino acids that are not present in the sequence and which therefore are not counted in the numbering.

(4) FIG. 4 corresponds to a diagrammatic string-of-pearls representation of the amino acid sequence of the variable part of the heavy chain of the chimeric 12G4 antibody (SEQ ID NO: 66).

(5) FIG. 5 corresponds to a diagrammatic string-of-pearls representation of the amino acid sequence of the variable part of the light chain of the chimeric 12G4 antibody (SEQ ID NO: 62).

(6) FIG. 6 corresponds to a diagrammatic string-of-pearls representation of the amino acid sequence of the variable part of the heavy chain of the unmutated humanized 12G4 antibody (SEQ ID NO: 58) and of the mutated humanized 12G4 antibody (6B_78; SEQ ID NO: 58).

(7) FIG. 7 corresponds to a diagrammatic string-of-pearls representation of the amino acid sequence of the variable part of the light chain of the unmutated humanized 12G4 antibody (SEQ ID NO: 54).

(8) FIG. 8 corresponds to a diagrammatic string-of-pearls representation of the amino acid sequence of the variable part of the heavy chain of the mutated humanized 12G4 antibody (3C_23; SEQ ID NO: 74).

(9) FIG. 9 corresponds to a diagrammatic string-of-pearls representation of the amino acid sequence of the variable part of the light chain of the mutated humanized 12G4 antibody (3C_23; SEQ ID NO: 70).

(10) FIG. 10 corresponds to a diagrammatic string-of-pearls representation of the amino acid sequence of the variable part of the light chain of the mutated humanized 12G4 antibody (6B_78; SEQ ID NO: 78).

(11) FIG. 11 corresponds to a diagrammatic string-of-pearls representation of the amino acid sequence of the variable part of the heavy chain of the mutated humanized 12G4 antibody (3C_23K; SEQ ID NO: 86).

(12) FIG. 12 corresponds to a diagrammatic string-of-pearls representation of the amino acid sequence of the variable part of the light chain of the mutated humanized 12G4 antibody (3C_23K; SEQ ID NO: 82).

(13) FIG. 13 corresponds to a diagrammatic string-of-pearls representation of the amino acid sequence of the variable part of the heavy chain of the mutated humanized 12G4 antibody (4C_35; SEQ ID NO: 90).

(14) FIG. 14 corresponds to a diagrammatic string-of-pearls representation of the amino acid sequence of the variable part of the light chain of the mutated humanized 12G4 antibody (4C_35; SEQ ID NO: 78).

(15) FIG. 15 corresponds to a diagrammatic string-of-pearls representation of the amino acid sequence of the variable part of the heavy chain of the mutated humanized 12G4 antibody (5B_42; SEQ ID NO: 98).

(16) FIG. 16 corresponds to a diagrammatic string-of-pearls representation of the amino acid sequence of the variable part of the light chain of the mutated humanized 12G4 antibody (5B_42; SEQ ID NO: 94).

(17) FIG. 17 presents determination of the binding affinity of the antibody to the AMHR-II receptor in a conventional ELISA assay obtained with mutated antibodies (Fab) according to the invention.

(18) The X-axis shows the concentration of (Fab) in g/ml and the Y-axis shows the OD at 450 nm.

(19) The dotted curve with empty white circles represents the binding of the unmutated humanized 12G4 antibody.

(20) The curve with filled black triangles represents the binding of the mutated humanized 12G4 antibody, having a mutation in the CDR (E184K) of the variable region of the light chain (6B_78 antibody).

(21) The curve with empty white triangles represents the binding of the mutated humanized 12G4 antibody, having a mutation in the CDR (S179P) of the variable region of the light chain, a mutation in the FR region (I177T) of the variable region of the light chain and a mutation in the variable region of the heavy chain (Q3R) (3C_23 antibody).

(22) The curve with empty white circles represents the binding of the mutated humanized 12G4 antibody, having a mutation in the CDR (E184K) of the variable region of the light chain, a mutation in the CDR (S179P) of the variable region of the light chain, a mutation in the FR region (I177T) of the variable region of the light chain and a mutation in the variable region of the heavy chain (Q3R) (3C_23K antibody).

(23) The curve with filled black circles represents the binding of the unmutated chimeric 12G4 antibody.

(24) FIG. 18 corresponds to the diagrammatic representation of the H622-14 cloning vector for the chimeric 12G4 antibody containing the heavy chain where the leader VH AMHR-II is fused to the variable region of the heavy chain (VH AMHR-II), itself fused to the constant region of human immunoglobulin (CH T125), and the light chain where the leader VK AMHR-II is fused to the variable region of the light chain (VK AMHR-II), itself fused to the constant region of human immunoglobulin (CK T125).

(25) The various regulatory elements (promoters, chimeric introns, polyadenylation sites, etc.) as well as the antibiotic resistance genes and the origins of replication are also shown.

(26) FIG. 19 corresponds to the diagrammatic representation of the H622-18 cloning vector for the unmutated humanized 12G4 antibody containing the heavy chain where the humanized leader VH AMHR-II is fused to the variable region of the heavy chain (humanized VH AMHR-II), itself fused to the constant region of human immunoglobulin (CH T125), and the light chain where the leader VK AMHR-II is fused to the variable region of the light chain (humanized VK AMHR-II), itself fused to the constant region of human immunoglobulin (CK T125).

(27) The various regulatory elements (promoters, chimeric introns, polyadenylation sites, etc.) as well as the antibiotic resistance genes and the origins of replication are also shown.

(28) FIG. 20 corresponds to the diagrammatic representation of the H622-18 cloning vector MAO 3C23 for the mutated humanized 12G4 antibody 3C_23 containing the heavy chain where the leader VH 3C_23 is fused to the variable region of the heavy chain (VH 3C_23), itself fused to the constant region of human immunoglobulin (CH T125), and the light chain where the leader VK 3C_23 is fused to the variable region of the light chain (VK 3C_23), itself fused to the constant region of human immunoglobulin (CK T125).

(29) The various regulatory elements (promoters, chimeric introns, polyadenylation sites, etc.) as well as the antibiotic resistance genes and the origins of replication are also shown.

(30) FIG. 21 corresponds to the diagrammatic representation of the H622-18 cloning vector MAO 6B_78 for the mutated humanized 12G4 antibody 6B_78 containing the heavy chain where the humanized leader VH AMHR-II is fused to the variable region of the heavy chain (humanized VH AMHR-II), itself fused to the constant region of human immunoglobulin (CH T125), and the light chain where the leader VK 6B_78 is fused to the variable region of the light chain (VK 6B_78), itself fused to the constant region of human immunoglobulin (CK T125).

(31) The various regulatory elements (promoters, chimeric introns, polyadenylation sites, etc.) as well as the antibiotic resistance genes and the origins of replication are also shown.

(32) FIG. 22 corresponds to the diagrammatic representation of the H622-18 cloning vector MAO 3C_23K for the mutated humanized 12G4 antibody 3C_23K containing the heavy chain where the leader VH 3C_23K is fused to the variable region of the heavy chain (VH 3C_23K), itself fused to the constant region of human immunoglobulin (CH T125), and the light chain where the leader VK 3C_23K is fused to the variable region of the light chain (VK 3C_23K), itself fused to the constant region of human immunoglobulin (CK T125).

(33) The various regulatory elements (promoters, chimeric introns, polyadenylation sites, etc.) as well as the antibiotic resistance genes and the origins of replication are also shown.

(34) FIG. 23 presents the diagrammatic protocols for construction of the scFv fragments.

(35) The black arrow under region 2/3 of VH indicates the sequence encoding for the N-terminal 2/3 of the peptide bond.

(36) The black arrow under region 2/3 of VL indicates the sequence encoding for the C-terminal 2/3 of the peptide bond.

(37) FIGS. 24A and 24B show the subcloning of the nucleotide sequences of the light chains VL-CL and heavy chains VH-CH1 of the mLFB112 and huLFB112 antibodies into the pMG62-Fab expression vectors.

(38) FIG. 24A: mLFB112

(39) FIG. 24B: huLFB112

(40) FIG. 25 presents the ADCC activity of the humanized anti-AMHRII antibodies of the invention compared to that of the unmutated humanized 12G4 antibody. The results are expressed as percentage lysis of the ASC1 cell (Y-axis) as a function of the amount of antibody added in ng/ml (X-axis). MeanSEM.

(41) The curve with diamonds represents the anti-AMHRII antibody 3C_23 (R901 3C_23), the curve with triangles with the point upwards represents the anti-AMHRII antibody 6B_78 (R901 6B_78), the curve with triangles with the point downwards represents the anti-AMHRII antibody 3C_23K (R901 3C_23K), the curve with circles represents the unmutated humanized anti-AMHRII antibody 12G4.

(42) FIG. 26 shows the map of the pIRES-neo plasmid expression vector used for generating the cov434-AMHRII line.

(43) FIGS. 27A and 27B present the ADCC activity of chimeric and humanized anti-AMHRII antibodies produced in the YB2/0 cells (FIG. 27A) and the CHO cells (FIG. 27B) on the COV434-AMHRII line.

(44) The results are expressed as percentage lysis of the COV434-AMHRII cells (Y-axis) as a function of the amount of antibody added in ng/ml (X-axis). Mean of 3 assaysSEM.

(45) FIG. 27A: the curve with diamonds represents the unmutated chimeric 12G4 anti-AMHRII antibody, the curve with filled squares represents the anti-AMHRII antibody YB2/0 3C_23 (R901 3C_23), the curve with triangles with the point downwards represents the anti-AMHRII antibody YB2/0 6B_78 (R901 6B_78), the curve with triangles with the point upwards represents the anti-AMHRII antibody YB2/0 3C_23K (R901 3C_23K) and the curve with empty rectangles represents the anti-CD20 antibody used as negative control.

(46) FIG. 27B: the curve with diamonds represents the unmutated chimeric 12G4 anti-AMHRII antibody, the curve with triangles with the point upwards represents the anti-AMHRII antibody CHO 3C_23 (R901 3C_23), the curve with triangles with the point downwards represents the anti-AMHRII antibody CHO 3C_23K (R901 3C_23K), the curve with circles represents the anti-AMHRII antibody CHO 6B_78 (R901 6B_78) and the curve with empty rectangles represents the anti-CD20 antibody used as negative control.

(47) FIG. 28 shows the ADCC activity of the humanized anti-AMHRII antibodies produced in YB2/0 and CHO cells on the Asc 1 line. The results are expressed as percentage lysis of the Asc 1 cells (Y-axis) as a function of the amount of antibody added in ng/ml (X-axis). Mean of 3 testsSEM.

(48) The curve with diamonds represents the anti-AMHRII antibody YB2/0 3C_23K, the curve with triangles with the point upwards represents the anti-AMHRII antibody CHO 3C_23K.

(49) FIG. 29 shows the ADCC activity of the humanized anti-AMHRII antibodies produced in YB2/0 and CHO on the META 2815 line. The results are expressed as percentage lysis of the META 2815 cells (Y-axis) as a function of the amount of antibody added in ng/ml (X-axis). Mean of 3 testsSEM.

(50) The curve with diamonds represents the anti-AMHRII antibody YB2/0 3C_23K, the curve with triangles with the point upwards represents the anti-AMHRII antibody CHO 3C_23K, the curve with circles represents the anti-CD20 antibody used as negative control (anti-CD20 A/R603 09/045).

(51) FIG. 30 shows the effect of the anti-AMHRII antibody 3C_23K on proliferation of the COV434-AMHRII cells. The value 100% corresponds to the proliferation of the COV434-AMHRII cells observed without antibody (mean of 3 testsSD).

(52) From left to right, the histograms show:

(53) The control without antibody, an antiP24 antibody, the anti-AMHRII antibody YB2/0 3C_23K, the anti-AMHRII antibody CHO 3C_23K, an antiP24 antibody in the presence of a cross-linking agent (CK), the anti-AMHRII antibody YB2/0 3C_23K in the presence of CK, the anti-AMHRII antibody CHO 3C_23K in the presence of CK, colchicine at 1 g/ml.

(54) FIG. 31 shows the effect of the anti-AMHRII antibody 3C_23K on the proliferation of META 2815 cells. The value 100% corresponds to the proliferation of the META 2815 cells observed without antibody (mean of 3 testsSD).

(55) From left to right, the histograms show:

(56) The control without antibody, an antiP24 antibody, the anti-AMHRII antibody YB2/0 3C_23K, the anti-AMHRII antibody CHO 3C_23K, a cross-linking agent alone, an antiP24 antibody in the presence of a cross-linking agent (CK), the anti-AMHRII antibody YB2/0 3C_23K in the presence of CK, the anti-AMHRII antibody CHO 3C_23K in the presence of CK, colchicine at 1 g/ml.

(57) FIGS. 32A and 32B show the variation of tumour volumes (FIG. 32A) and the survival curves (FIG. 32B) under the effect of the treatment with 3C23K-YB2/0 with intraperitoneal injections of antibody performed at intervals of 2-3 days at a dose of 10 mg/kg/inj for a total of 18 injections (black arrows) in the cov434-AMHRII model.

(58) FIG. 32A:

(59) Y-axis: tumour volumes in mm.sup.3,

(60) X-axis: days after injection of the tumour cells.

(61) Curve with diamonds: vehicle

(62) Curve with rectangles: anti-AMHRII antibody YB2/0 3C_23K.

(63) FIG. 32B:

(64) Y-axis: percentage survival

(65) X-axis: days after injection of the tumour cells.

(66) Curve with diamonds: vehicle

(67) Curve with rectangles: anti-AMHRII antibody YB2/0 3C_23K.

(68) FIGS. 33A and 33B show the variation of the tumour volumes (FIG. 33A) and the survival curves (FIG. 33B) under the effect of the treatment with 3C_23K-YB2/0, intraperitoneal injections of antibody performed at intervals of 2-3 days at a dose of 10 mg/kg/inj for a total of 18 injections (black arrows) in an AsclaS model.

(69) FIG. 33A:

(70) Y-axis: tumour volumes in mm.sup.3,

(71) X-axis: days after injection of the tumour cells.

(72) Curve with diamonds: vehicle

(73) Curve with rectangles: anti-AMHRII antibody YB2/0 3C_23K.

(74) FIG. 33B:

(75) Y-axis: percentage survival

(76) X-axis: days after injection of the tumour cells.

(77) Curve with diamonds: vehicle

(78) Curve with rectangles: anti-AMHRII antibody YB2/0 3C_23K.

(79) FIGS. 34A and 34B show the variation of the tumour volumes (FIG. 34A) and survival curves (FIG. 34B) under the effect of the treatment with 3C_23K-YB2/0, intraperitoneal injections of antibody performed at intervals of 2-3 days at a dose of 10 mg/kg/inj for a total of 18 injections (black arrows) in the META 2815 model.

(80) FIG. 34A:

(81) Y-axis: tumour volumes in mm.sup.3,

(82) X-axis: days after injection of the tumour cells.

(83) Curve with diamonds: vehicle

(84) Curve with rectangles: anti-AMHRII antibody YB2/0 3C_23K.

(85) FIG. 34B:

(86) Y-axis: percentage survival

(87) X-axis: days after injection of the tumour cells.

(88) Curve with diamonds: vehicle

(89) Curve with rectangles: anti-AMHRII antibody YB2/0 3C_23K.

EXAMPLES

Example 1

Determination of the Affinity of the Anti-AMHR-II Antibodies

(90) The affinity of the antibodies for their antigen, AMHR-II, is determined by the SPR (Surface Plasmon Resonance) technique on BIACore X100 (BIACore, GE Healthcare). The AMHR-II recombinant receptor, expressed in the form of fusion protein with a region Fc, is immobilized by covalent coupling between its amine functions and the carboxyl groups of dextran activated in succinimide esters, present on the surface of the type CM5 sensor chip. The COOH groups of the sensor chip are activated for 7 minutes with EDC/NHS mixture (0.1 M of N-hydroxysuccinimide and 0.1 M of 3-(N,N-dimethylamino)propyl-N-ethylcarbodiimide) at a flow of 10 l/min then the AMHR-II/Fc fusion protein, diluted to 5 g/ml in 10 mM sodium acetate buffer, pH 4.0, is injected at 5 l/min on track 2 of the sensor chip so as to reach 300 RU. The ester groups that have not reacted with the amines of the fusion protein are deactivated by injection of a solution of ethanolamine-HCl 1M, pH 8.5 for 7 min at a flow of 10 l:min. Track 1, serving as negative control, was activated and deactivated like track 2.

(91) All the measurements are carried out at 25 C. The antibodies to be analysed are diluted in HBS-EP running buffer (BIACore, GE Healthcare) at concentrations from 6.25 to 3333 nM and injected on the sensor chip for 2 min at a flow of 30 l/min. The dissociation step is monitored for 10 min and then the surface is regenerated by injection of 10 mM glycine buffer, pH 1.5 for 30 s at 10 l/min.

(92) The sensorgrams obtained are analysed using the 1:2 kinetic model of the BlAevaluation 3.1 software.

(93) Results

(94) The antibodies were produced in CHO or YB 2/0 cells (Table I)

(95) TABLE-US-00001 TABLE I k.sub.a k.sub.d K.sub.D K.sub.D.sup.chimeric/ Antibody Mutations (M.sup.1s.sup.1) (s.sup.1) (nM) K.sub.D.sup.mutated 12G4 - NA 3.5 10.sup.3 7.4 10.sup.4 212 Chimeric 6B_78 VL-E184K 1.6 10.sup.4 1.3 10.sup.3 82 2.6 YB2/0 3C_23 VH-Q3R 3.6 10.sup.4 3.3 10.sup.3 92 2.3 YB2/0 VL-I177T/ S179P 3C_23K VH-Q3R .sup.4 10.sup.4 8.6 10.sup.4 21 10 YB2/0 VL-I177T/ S179P/E184K 6B_78 VL-E184K 1.5 10.sup.4 1.2 10.sup.3 81 2.6 CHO 3C_23 VH-Q3R .sup.5 10.sup.4 4.3 10.sup.3 86 2.5 CHO VL-I177T/ S179P 3C_23K VH-Q3R 4.1 10.sup.4 .sup.1 10.sup.3 25 8.5 CHO VL-I177T/ S179P/E184K

(96) The mutations introduced in the 6B_78 and 3C_23 antibodies induce an increase in affinity for the AMHR-II antigen by a factor of 2.3 to 2.6 relative to the chimeric antibody (12G4chimeric).

(97) The mutations of the two antibodies 6B_78 and 3C_23 have a synergistic effect; introduction of the mutation of the 6B_78 antibody into the 3C_23 antibody causes an increase in affinity by a factor of 10.

Example 2

Determination of the Affinity of Chimeric or Humanized Murine 12G4 Antibody on cov434-AMHR-II Cells (Epitope Peptide of Sequence: GGGGNLTQDRAQVEMQGSR (SEQ ID NO: 101) and GGGGNLTQARGQVEMQGSR (SEQ ID NO: 102) for the Negative Control Peptide)

(98) TABLE-US-00002 Association Dissociation Affinity constant constant constant at equilibrium humanized 1.83 10.sup.3 M.sup.1 .Math. s.sup.1 9.62 10.sup.3 s.sup.1 5.26 10.sup.6 M 12G4 chimeric 6.49 10.sup.3 M.sup.1 .Math. s.sup.1 1.53 10.sup.3 s.sup.1 2.35 10.sup.7 M 12G4 murine 1.47 10.sup.3 s.sup.1 of the order 12G4 of 10.sup.7 M

(99) The affinity of the chimeric antibody determined on the AMHR-II human receptor is about 10.sup.8M.

Example 3

Preparation of Mutated Humanized 12G4 Antibodies

(100) The murine antibody is substantially equivalent to the chimeric antibody and displays strong affinity.

(101) The humanized 12G4 antibody (huLFB112) was obtained by grafting hypervariable CDR loops of the murine 12G4 antibody (mLFB112) on a protein framework of a human nature (CDR grafting).

(102) The humanized antibody has an appreciable loss of affinity compared with the murine antibody.

(103) The final objective is therefore to increase the affinity of the humanized antibody so as to restore the initial binding characteristics of the murine antibody. This optimization will be performed by means of a cycle of molecular evolution by the MutaGen technology owned by the Millegen company.

(104) 3.1 Construction and Validation of the Molecular Tools

(105) 3.1.1. Construction of the scFv Fragments

(106) The nucleotide sequences coding for the variable regions of the light chains (VL) and of the heavy chains (VH) of the murine and humanized antibodies were amplified by PCR using suitable primers. The amplified sequences were then combined together so as to generate a recombinant antibody fragment of the scFv type. Several constructions were performed in this way: VH-VL or VL-VH orientation and use of two different peptide bonds (peptide bond of 15 or of 18 amino acids). A total of 8 constructions were performed, 4 for the murine antibody and 4 for the humanized antibody. The principle for construction of the scFv fragments is illustrated below (scheme I). The sequences coding for these scFv were then subcloned into the MilleGen phagemid expression vector (pMG58). This vector makes it possible to express antibody fragments of the scFv type and display them on the surface of a type M13 bacteriophage (phage-scFv).

(107) The nucleotide sequences of the VH and VL domains of the murine and humanized antibodies were verified by DNA sequencing.

(108) The protocol is summarized in FIG. 23.

(109) 3.1.2. Expression of the scFv on the Surface of the Phages and Characterization by ELISA

(110) The amount of target supplied (80 g) did not allow us to test all 8 constructions made. The murine antibody mLFB112 expressed in the form of a scFv is called mVH-VL hereinafter whereas the humanized antibody huLFB112 is called huVH-VL.

(111) 3.1.2.1. Production of the Phages-scFv

(112) The XL1-Blue bacteria transformed by the pMG58 vectors containing the DNA coding for the scFv mVH-VL on the one hand and the scFv huVH-VL on the other hand are cultured at 30 C. to an OD600 nm of 0.5-0.6. After adding IPTG and infecting the bacteria with auxiliary phages (M13KO7, New England Biolabs), the cultures are cultured at 26 C. overnight. The next day, the phage particles (phages-scFv) are recovered from the culture supernatant, precipitated by means of a PEG/NaCl solution, concentrated (100) and quantified.

(113) In this case, a concentration of the order of 810.sup.11 phages/ml is obtained for the two scFv.

(114) 3.1.2.2. ELISA-Phages Assay

(115) The functionality of the scFv mVH-VL and huVH-VL produced on the surface of the phages was verified by direct ELISA assays.

(116) Protocol: 1) Immobilization of the target: 100 g/well of the recombinant target diluted to 5 g/ml in PBS1 i.e. 500 ng/well, overnight at 4. Use of Nunc-Immuno Plate Maxisorp microtitre plates, 2) Saturation: 200 l/well of PBS1-Skimmed milk 4%, incubation 2 h at 37 C., 3) Binding: 100 l/well of the solutions of murine and humanized phages-scFv diluted in PBS1-Milk 2%-Tween20 0.05% (twofold dilution series), incubation 2 h at 37 C., 4) Detection: 100 l/well of the anti-phages M13 antibody coupled to peroxidase (dilution 1/10000, GE Healthcare), incubation 2 h at 37 C. 5) Detection: 100 l/well of TMB 6) Neutralization: 100 l/well of H.sub.2SO.sub.4 7) Measurement of OD at 450 nm
Results (Table II):

(117) TABLE-US-00003 TABLE II Number mVH-VL huVH-VL Ratio of phages Without Without mVH-VL/ per well +covering covering Diff +covering covering Diff huVH-VL 3.00E+10 2.805 0.158 2.647 1.247 0.150 1.097 2.4 3.00E+10 2.193 0.085 2.107 0.689 0.096 0.593 3.6 3.00E+09 1.570 0.064 1.506 0.395 0.072 0.323 4.7 3.00E+09 0.946 0.059 0.887 0.226 0.056 0.170 5.2 3.00E+09 0.495 0.049 0.446 0.135 0.049 0.086 5.2 Diff: specific binding (difference between the wells with covering and without covering
The ratio mVH-VL/huVH-VL was calculated with the value of specific binding (Diff )
3.1.3: Construction of the Fab Fragments

(118) The nucleotide sequences of the light chains VL-CL and heavy chains VH-CH1 of the antibodies mLFB112 and huLFB112 were subcloned into the pMG62-Fab expression vectors (FIGS. 24A and 24B).

(119) pMG62-Fab expression vectors

(120) A) The two chains VL-CL and VH-CH1 are expressed starting from the pLac promoter upstream of the light chain, the heavy chain VH-CH1 is fused to a tag for detection (peptide V5) and a tag for purification with IMAC (6 His).

(121) B) Each of the light and heavy chains is dependent on a promoter. RBS: Ribosome Binding Site.

(122) 3.2 Construction and Validation of the Molecular Tools

(123) Construction of the Library by MutaGen

(124) The objective defined for this step was to obtain a database of 510.sup.6 variants with 1 to 2 amino acid mutations per scFv.

(125) Mutations were introduced within the domains VL and VH of the humanized antibody huLFB112 by means of MutaGen technology. Finally, a large database composed of about 510.sup.7 mutated clones with 1 to 5 mutations of amino acids per scFv, i.e. 10 times the diversity initially envisaged was obtained.

(126) For this, several sub-databases were constructed according to different experimental conditions: conditions U, M, US and UE defined by different nucleotide primers, mutase enzymes used and number of replications. These sub-databases are 4 in number and are designated R20U, 45M, R20US and R20UE. For all of these sub-databases, a total of 295 sequences was carried out for accurately defining the different characteristics of mutagenesis. Table III below gives an idea of the principal data obtained from analysis of the sequencing operations.

(127) TABLE-US-00004 TABLE III Analysis of the mutations of the different sub-databases Name of database R20U 45M R20US R20UEta TOTAL Size of database 2.0E+07 6.0E+06 9.9E+07 5.0E+05 1.3E+08 Condition U M US UE Number of sequences 87.sup. 67.sup. 85.sup. 56.sup. 295 Analysis of the nucleotide sequences Frequency of 2.76 3.66 3.31 4.9 mutations per kb % of deletions 10% 20% 6% 5% % of additions 2% 0.8% % of substitutions 72% 80% 93% 95% Frequency of 2.0 2.45 3.10 3.68 mutations per kb (without deletions) Analysis of the amino acid sequences % of sequences (by 63% 46% 44% 41% weight) (+silent mutation) % of sequences with 13% 28% 13% 27% reading frame shift (+stop codon) % of sequences with 24% 25% 43% 32% mutated amino acids Number of scFv clones 4.8E+06 1.5E+06 4.3E+07 1.6E+05 4.9E+07 with mutated amino acids Number of mutations 1.3 1.2 1.8 2.11 of amino acids per scFv scFv clones with 1 65% 49% 50% 33% 2.5E+07 mutated amino acid scFv clones with 2 25% 34% 33% 39% 1.6E+07 mutated amino acids scFv clones with 3 10% 13% 8% 6% 4.2E+06 mutated amino acids scFv clones with 4 4% 6% 22% 2.5E+06 mutated amino acids scFv clones with 5 3% 1.2E+06 mutated amino acids

(128) 3.2. Elaboration of the Selection Conditions

(129) In order to evaluate different selection strategies, an artificial mixture was prepared between the murine and humanized phages-scFv ( 1/200 mixture, mLFB112/huLFB112), the objective being to simulate screening of the database. This simulation of screening must make it possible to validate different selection conditions which have the aim of rapidly amplifying the most affine clone within this artificial mixture (i.e. the murine clone mLFB112 in this case).

(130) The various strategies evaluated:

(131) i) Use of a constant amount of immobilized target in the course of the selection cycles (Cond 1) ii) Decrease of the amount of immobilized target in the course of the selection cycles (Cond 2) ii) Test for a k.sub.off condition: long incubation time of the phages-scFv with the target (Cond 3)

(132) These various conditions were performed in 3 selection cycles. Sequencing was performed on the clones retained after each selection cycle. The results are shown in Table IV below.

(133) TABLE-US-00005 TABLE IV Evaluation of the three screening strategies Bp1 Bp2 Bp3 Conditions 1 Covering 500 ng 500 ng 500 ng Number of scFv phages used 1.6E+11 1.2E+11 2.0E+11 for the selection cycle Number of scFv phages 3.3E+5 2.4E+5 2.7E+6 recovered at the end of the selection cycle % of mLFB112/huLFB112 0/100 1/99 90/10 Conditions 2 Covering 500 ng 100 ng 100 ng Number of scFv phages used 1.6E+11 1.2E+11 1.8E+11 for the selection cycle Number of scFv phages 3.3E+5 2.4E+5 2.7E+6 recovered at the end of the selection cycle % of mLFB112/huLFB112 0/100 1/99 30/70 Conditions 3 Covering 500 ng 100 ng k.sub.off 100 ng k.sub.off Number of scFv phages used 1.6E+11 1.8E+11 1.6E+11 for the selection cycle Number of scFv phages 3.3E+5 6.7E+5 2.0E+6 recovered at the end of the selection cycle % of mLFB112/huLFB112 0/100 1/99 30/70

(134) It appears from these results that condition 1 (fixed amount of target) gives best performance for amplifying the clone with better affinity, mLFB112 (9 clones out of 10). Using a smaller amount of target (100 ng/well) seems less suitable; only 3 clones out of 10 after 3 selection cycles correspond to the clone with better affinity. The same applies to condition 3 based on a long incubation time (k.sub.off selection) which does not allow sufficient amplification of clone mLFB112. Moreover, the number of phages recovered for this last-mentioned condition after 3 cycles is not high (210.sup.4 phages). It therefore seemed to us to be inadvisable to use conditions 2 and 3 for a more diversified mixture of clones, as is the case for the database constructed in the context of this project.

(135) 3.3. Primary Screening (Selection Cycles)

(136) After elaborating the screening conditions, it was therefore decided to use 2 screening conditions:

(137) Cond A: 1 g/well of target for 4 selection cycles then 2 cycles with 0.5 g/well Cond B: 0.5 g/well for 6 selection cycles.
The results obtained are presented in Table V below.

(138) TABLE-US-00006 TABLE V Results of the selection cycles Selection Selection Selection Selection Selection Selection cycle 1 cycle 2 cycle 3 cycle 4 cycle 5 cycle 6 Conditions A Covering .sup.1 g .sup.1 g .sup.1 g .sup.1 g 0.5 g 0.5 g Number of scFv phages used 4.8E+11 8.0E+11 2.8E+11 8.0E+11 4.5E+11 6.0E+11 for the selection cycle Number of scFv phages 2.0E+5 1.3E+5 2.4E+5 7.5E+5 7.3E+5 1.4E+5 recovered at the end of the selection cycle Conditions B Covering 0.5 g 0.5 g 0.5 g 0.5 g 0.5 g 0.5 g Number of scFv phages used 4.8E+11 8.0E+11 3.2E+11 6.4E+11 7.5E+10 4.2E+11 for the selection cycle Number of scFv phages 2.0E+5 8.6E+5 5.4E+5 1.3E+5 2.6E+5 6.5E+5 recovered at the end of the selection cycle

(139) After these selections, the clones obtained were sequenced starting from the 3rd selection cycle. The results obtained were compared with those obtained for the starting database (Table VI).

(140) TABLE-US-00007 TABLE VI Results of the sequencing operations Selec- Selec- Selec- Selec- Data- tion tion tion tion base cycle 3 cycle 4 cycle 5 cycle 6 Number of sequences 295 87 168 114 98 % of sequences with 39% 26% 26% 35% 20% mutated amino acids % of sequences with 14% 13% 27% 32% 49% reading frame shift % of sequences (by 47% 61% 48% 34% 32% weight) (+silent mutation) Database: initial database resulting from mixing the 4 databases R20U, 45M, R20US, R20Ueta.

(141) Analysis of these sequencing operations revealed the presence of redundant clones. In total, from all of the clones sequenced, 113 unique mutated clones were obtained.

(142) 3.4. Secondary Screening (ELISA-Phages)

(143) The secondary screening consists of analysing the clones selected at the end of the primary screening individually. For this, the 113 unique mutated clones were transferred to a culture plate (96 well-1.2 ml).

(144) After production of the phage particles, the culture supernatants containing the phages-scFv were used for carrying out an ELISA binding assay. The binding of the mutated clones was evaluated at two dilutions ( and of the supernatants containing the scFv-phages). The murine clones (mLFB112) and humanized clones (huLFB112) constructed under the scFv-phages format were used as references on each of the assay plates. Each of the mutated clones was tested at least twice.

(145) The results are expressed as a ratio, i.e. the differences in binding (OD405 nm) between the mutated clones and the references huLFB112 and mLFB112. Ratio relative to the humanized scFv, huLFB112 (Ratio/huLFB112) Ratio relative to the murine scFv, mLFB 112 (Ratio/mLFB 112)
Out of the 113 clones tested, only the best clones are presented below.

(146) Table VII below presents the various clones obtained and the mutations present (position and substitution of amino acids) in the light and/or heavy chain, as well as the binding affinity of clones determined by ELISA.

(147) The values shown after the substitutions correspond to the changes in the values of the hydropathic index as a function of the various substitutions.

(148) The values of binding affinity correspond to the ratio of the binding affinity of the antibody of the invention for the AMHRR-II receptor to the binding affinity of the unmutated humanized 12G4 antibody or the binding affinity of the unmutated chimeric 12G4 antibody.

(149) The values of binding affinity given are the mean values of at least four values and the figures in parentheses correspond to the standard deviation.

(150) Table VII shows that with the substitutions that were carried out, although the latter lead to a large change in the hydropathic index, the binding affinity of the antibody for the receptor is much greater than that of the unmutated humanized 12G4 antibody and at least equal to that of the unmutated chimeric 12G4 antibody: ratio AB invention/chimeric 12G4 greater than or equal to 1.

(151) The mutated humanized antibody displays an affinity that is restored or even greater than that of the chimeric or murine antibody.

(152) TABLE-US-00008 TABLE VII FIXATION TO THE TARGET AMHRII-Fc DETERMINED BY ELISA Numbering Numbering Ratio Ratio VH VL AB invention/ AB invention/ clone 1-115 131-236 humanized 12G4 chimeric 12G4 4C_35 L45P +3.8_1.6 E184K 3.5_3.9 4.3 (0.5) 1.9 0.4) 5B_81 L45P +3.8_1.6 3.5 (0.9) 1.6 (0.3) 6B_78 E184K 3.5_3.9 NT NT 3C_23 Q3R 3.5_4.5 I177T +4.5_0.7 2.6 (1.1) 1.2 (0.3) S179P 0.8_1.6 3C_23K Q3R 3.5_4.5 I177T +4.5_0.7 NT NT S179P 0.8_1.6 E184K 3.5_3.9 5B_42 T74A 0.7_1.8 S179P 0.8_1.6 NT NT 4F_196 Q3E 3.5_3.5 3.0 (1.0) 1.5 (0.3) Q62R 3.5_4.5 E89D 3.5_3.5 6B_87 Q1E 3.5_3.5 2.1 (0.2) 1.0 (0.1) A24V +1.8_+4.2 4F_169 Q6E 3.5_3.5 2.0 (0.7) 1.0 (0.2) T58A 0.7_+1.8 3D_74 S158P 0.8_1.6 .sup.2 (0.7) 0.8 (0.2) 4C_44 R87G 3.9_0.4 NT NT 5A_66 V67M +4.2_+1.9 F212S +2.8_0.8 2.2 (0.4) 1.1 (0.1) 6B_14 S31G 0.8_0.4 2.6 0.3) 1.1 (0.2) Q39E 3.5_3.5 4C_47 Q3E 3.5_3.5 NT NT S88P 0.8_1.6 4E_153 D56N 3.5_3.5 2.0 (0.4) 1.0 (0.1) I70N +4.5_3.5 F102S +2.8_0.8 3C_24 E184G 3.5_0.4 NT NT 5B_18 Q3E 3.5_3.5 1.8 (0.2) 0.9 (0.2) A9T +1.8_0.7 A103T +1.8_0.7 5B_84 Q1E 3.5_3.5 NT NT A24G +1.8_0.4 6B_86 Q3E 3.5_3.5 G179D 0.4_3.5 NT NT 4D_91 Q1E 3.5_3.5 1.7 (0.1) 0.8 (0.2) V11A +4.2_+1.8 6B_76 A40V +1.8_+4.2 S179P 0.8_1.6 1.7 (0.1) 0.8 (0.2) 5A_28 Y178H 1.3_3.2 1.6 (0.6) 0.8 (0.2) S179P 0.8_1.6 3D_57 A76T +1.8_0.7 NT NT A79T +1.8_0.7 6A_80 A24V +1.8_+4.2 1.5 (0.2) 1.3 (0.1) Q62E 3.5_3.5 5B_67 K12R 3.9_4.5 NT NT 5B_86 S31G 0.8_0.4 I132T +4.5_0.7 NT NT Q39E 3.5_3.5 A143T +1.8_0.7 5A_73 A24V +1.8_+4.2 NT NT 5B_33 A76T +1.8_0.7 NT NT 3B_71 S114T 0.8_0.7 S179P 0.8_1.6 NT NT 3B_87 L175Q +3.8_3.5 NT NT 3D_68 T150A 0.7_1.8 NT NT 4E_112 L110P +3.8_1.6 V187A +4.2_+1.8 NT NT S192T 0.8_0.7 5B_54 A24T +1.8_0.7 NT NT 6A_18 K13R 3.9_4.5 NT NT 3C_40 P224A 1.6_+1.8 NT NT 5A_83 Q62E 3.5_3.5 NT NT S179P 0.8_1.6 A79T +1.8_0.7 5A_19 Q3E 3.5_3.5 S182F 0.8_+2.8 NT NT 3A_29 V20A +4.2_+1.8 NT NT (Q1E, Q3E, Q6E, K19E, Q39E and Q62E: TAG codon suppressed, translated to E in the E. coli bacteria XL1-blue used) NT: not tested

Example 4

Comparison of Clones Having an Improvement of Affinity

(153) The positive clones 3C_23K, 3C_23 and 6B_78 were compared with one another by determining the binding affinity of the antibody to the AMHR-II receptor in a conventional ELISA assay obtained with soluble mutated antibodies (Fab) according to the invention.

(154) The results obtained are presented in FIG. 17.

Example 5

Establishment of the Transfected Line AMHRII cov434-AMHRII

(155) The cov434-AMHRII line was generated by transfection of a plasmid expressing the cDNA coding for AMHRII in the granulosa tumour line cov434 (van den Berg-Bakker, C., et al., 1993. Establishment and characterization of 7 ovarian carcinoma cell lines and one granulosa tumour cell line: Growth features and cytogenetics. International Journal of Cancer 53: 613; Zhang, H. et al., 2000. Characterization of an immortalized human granulosa cell line (COV434). Molecular Human Reproduction 6: 146) not expressing AMHRII.
Briefly, the cDNA of AMHRII was cloned into the commercial plasmid pIRES-neo (ClontechTakara Bio Europe, France; references 6060-1). Owing to the IRES sequence, AMHRII and neo are expressed under the control of one and the same promoter CMV (FIG. 26).

(156) This construction was stably transfected in the granulosa cancer line cov434 (transfection agent Fugene, Roche). The transfectants obtained were then screened by cytometry and by Western blotting for expression of the AMHRII receptor. After subcloning, the cellular clone cov434-AMHRII-1F3, containing a vector of the pIRES-neo type, was retained for the in vitro and in vivo studies. This line is designated cov434-AMHRII hereinafter.

Example 6

Establishment of Primary Lines from Ascitic Fluid or Biopsy from Patients with Ovarian Epithelial Carcinoma

(157) The main steps in establishing the lines derived from samples of ascitic fluid (Asc 1 line) are as follows: D0: Receiving the ascitic fluid and starting culture of the sample immediately. The sample is centrifuged for 5 min at 1000 rev/min and the pellet is taken up in 2 mL of medium for seeding a T25 flask (RPMI 10% FCS medium). D2-D46: Culture with regular microscopic observation. Fresh medium is added regularly during this period. Washing with PBS is also carried out every other day in order to remove the contaminating erythrocytes and fibroblasts. Around D13, all the contaminating cells have disappeared and a still heterogeneous confluent lawn of cells is observed (two different types). D46: Transfer: on D46 a homogeneous lawn of tumour cells is observed and the cells are then washed and transferred by scraping to be re-seeded in two T75 flasks. D61: Evaluation of AMHRII expression by FACS: after washing with PBS, the cells are detached from the scraper and analysed by flow cytometry (labelling AcM 12G4 10 g/mL, AcII anti-mouse FITC). The databases are constructed at this stage. The positive AMHRII line is cultured for 10 more days in order to confirm expression of the AMHRII receptor. D71: Confirmation of AMHRII expression by FACS.

(158) The lines thus established are kept in RPMI 10% FCS medium, with one passage per week (1/15 dilution).

(159) In the case of biopsies (META 2815 line), the primary tumour is first maintained on nude mice (grafts of the sample in the interscapular space, 3 successive passages on mice) then the tumour is removed and dilacerated before being taken up in culture medium. A protocol identical to that of the samples of ascitic fluid is then applied.

Example 7

Evaluation of the Affinity of the Various Candidate Humanized Antibodies

(160) This study was conducted on the original murine 12G4 antibody as well as on the candidate humanized antibodies 3C23, 6B78 and 3C23K (of sequence 3C_23, 6B_78 and 3C_23K respectively) produced in YB2/0. The affinity of these antibodies was evaluated on the cov434-AMHRII cells.

(161) Briefly, K.sub.D was determined by the saturation method, by adding increasing doses of radiolabelled antibody to a constant number of cov434-AMHRII cells. The cells (110.sup.6 in 50 l PBS/BSA 0.5%) were incubated (final volume 150 l) for 1 h at 4 C. in the presence of increasing doses of antibody labelled beforehand with iodine 125 (.sup.125I). For each antibody, fourfold dilutions were carried out in PBS/BSA 0.5% from the solutions of labelled and unlabelled antibody (84.4 g/ml). The nonspecific fixation was evaluated by incubating the cells in the presence of a 100 times molar excess of unlabelled antibody.

(162) After incubation, the samples were frozen in liquid nitrogen and then analysed in a gamma counter. The specific fixation was determined by subtracting the fixation obtained in the presence of an excess of unlabelled antibody.

(163) Scatchard analysis, performed on PRISM software, made it possible to determine the affinity constants presented in Table VIII.

(164) TABLE-US-00009 TABLE VIII Dissociation constant (K.sub.D) of the anti-AMHRII antibodies murine 12G4 3C23 6B78 3C23K K.sub.D 15.41 +/ 0.97 7.33 +/ 0.44 6.68 +/ 0.21 5.30 +/ 0.38 (nM)

(165) According to this study, it appears that the humanized anti-AMHRII antibody 3C23K has the best affinity (K.sub.D=5.3 nM) compared with the two other candidate antibodies 6B78 and 3C23. The 3C23K antibody also has an affinity about three times greater than that of the original murine 12G4 antibody (K.sub.D=15.4 nM).

Example 8

Comparison of the ADCC Activity of the Antibodies of the Invention Versus the ADCC Activity of the Unmutated Humanized 12G4 Antibody

(166) 1 Material and Methods

(167) 1.1 Principle of the Methods

(168) ADCC

(169) The ASC1 target cells obtained from patients are adherent and are prepared on the day before the assay. They are detached with trypsin and incubated in EMS+5% FCS in a flat-bottomed plate at a rate of 50 l per well at a concentration of 610.sup.5 cells/ml. The plates are incubated overnight at 37 C., 7% CO.sub.2.

(170) The next day, the cells have adhered to the bottom of the well. The supernatant is aspirated and the necessary volume of buffer per well is added for carrying out the assay in the presence of NK and antibody.

(171) The killer cells (NK cells) are purified beforehand by the negative depletion technique developed by the company Miltenyi (Miltenyi BiotecNK cell isolation kit human ref 130-092-657), from peripheral blood of healthy donors. The ADCC technique consists of incubating the NK cells with ASC1 target cells in the presence of different concentrations of the humanized anti-AMHRII antibody (0.005 to 5000 ng/ml) with E/T ratio of 10/1. After 4 hours of incubation, the cytotoxic activity induced by the anti-AMHRII antibodies is measured by colorimetry by determining, in the supernatants, an intracellular enzyme called lactate dehydrogenase (LDH) released by the lysed target cells (Roche DiagnosticsCytotoxicity Detection Kit LDH ref 11644793001).

(172) 1.2 Elements Studied Anti-AMHRII antibodies: 829 10 054, humanized YB2/0, R901 3C23K 829 10 050, humanized YB2/0, R901 3C23 829 10 051, humanized YB2/0, R901 6B78 632 07 107, unmutated humanized anti-AMHRII 12G4 ASC1 cells culture dossier 871 10 063
The results are presented in FIG. 25.
Table IX presents the raw data corresponding to FIG. 25.

(173) TABLE-US-00010 TABLE IX % of lysis (ADC 1193 11 081) 829 829 829 632 07 107 10 050 10 051 10 054 Anti Ac ng/ml 3C23 6B78 3C23K AMHRII Ac ng/ml (Log) YB20 YB20 YB20 hum 1st 0.001 3.000 0 0 0 0 0.005 2.301 8 14 8 18 0.05 1.301 9 5 10 2 0.5 0.301 13 3 26 0 5 0.699 35 28 51 7 50 1.699 46 42 67 24 500 2.699 52 48 79 50 5000 3.699 63 53 75 50
Table X presents the Emax and EC50 obtained with the various antibodies.

(174) TABLE-US-00011 TABLE X 829 829 829 632 07 107 10 050 10 051 10 054 Anti 3C23 6B78 3C23K AMHRII YB20 YB20 YB20 hum 1st Emax (% of lysis) 65.55 51.89 79.02 52.40 EC50 (ng/ml) 6.298 5.383 1.704 52.35

Example 9

Comparison of the ADCC Activity of the Antibodies of the Invention Versus the Chimeric 12G4 Antibody

(175) The ADCC activity of the humanized candidate antibody 3C_23K (ha12G4 of sequence 3C_23K: mutations VHQ3R (SEQ ID NO: 82 (without leader) or SEQ ID NO: 84 (with leader)), and VLI177T/S179P/E184K (SEQ ID NO: 86 (without leader), or SEQ ID NO: 88 (with leader)) was evaluated.

(176) Briefly, the effector cells (NK killer cells; NK: Natural Killer) are purified beforehand by the negative depletion technique developed by the company Miltenyi (Miltenyi BiotecNK cell isolation kit human ref 130-092-657), from peripheral blood of healthy donors, after a first step of purification of the mononucleated cells on Ficoll.

(177) The in vitro assay of ADCC activity consists of incubating the NK cells with target cells (cov434-AMHRII, Asc 1 and META 2815 lines), in the presence of different concentrations of anti-AMHRII antibodies (chimeric antibody chl2G4, humanized 3C_23K-YB2/0 antibody, produced in YB2/0, and 3C_23K-CHO, produced in CHO). The effector/target ratio applied is 15/1. The antibodies are diluted in culture medium at concentrations ranging from 0.005 to 5000 ng/ml.

(178) After 4 hours of incubation, the cytotoxic activity induced by the anti-AMHRII antibodies is measured by colorimetry by determining, in the supernatants, an intracellular enzyme called lactate dehydrogenase (LDH) released by the lysed target cells (Roche DiagnosticsCytotoxicity Detection Kit LDH ref 11644793001).

(179) The percentage lysis is calculated from the following formula:
% of lysis=[(ERSR)/(100SR)][(NCSR)/(100SR)
with: ER=release of LDH in the presence of antibodies and of NK cells SR=spontaneous release of LDH from the target cell alone NC=release of LDH in the presence of NK cells and absence of antibodies.

(180) The results are expressed in percentage lysis as a function of the amount of antibody. The Emax and EC50 values are calculated with the PRISM software.

(181) The results obtained on the cov434-AMHRII line are presented in FIG. 27. The low activity of the 3C_23K-CHO antibody does not allow a plateau to be obtained under the assay conditions. In order to compare the efficacy of the antibodies, calculation of 50% relative is carried out in this case, which represents the amount of 3C_23K-CHO antibody required to reach 50% of the plateau of the chimeric antibody (50% relative=1). According to this evaluation, the humanized antibody having the best ADCC activity on the COV434-AMRHII line is the 3C_23K antibody (50% relative: 0.84) produced in YB2/0. The humanized antibodies of sequence 3C_23 and 6B_78 have a value of 50% relative, equal to 6.41 and 36.92, respectively.

(182) The results obtained on the Asc 1 line are presented in FIG. 28. The 3C_23K-YB2/0 antibody has a dose-dependent cytotoxic activity on the Asc 1 cells with an EC50 estimated at 2.24 ng/ml. The low activity of the 3C_23K-CHO antibody does not allow a plateau to be obtained under the assay conditions. In order to compare the efficacy of the two antibodies, calculation of 50% relative is carried out in this case, which represents the amount of 3C_23K-CHO antibody required to reach 50% of the plateau of the 3C_23K-YB2/0 antibody (50% relative=1). According to this evaluation, the cytotoxic activity of the 3C_23K-YB2/0 antibody is about 40 times greater than that of the antibody produced in CHO (50% relative=39.4).

(183) Similarly, the results presented in FIG. 29 show that the antibodies 3C_23K-YB2/0 and 3C_23K-CHO induce dose-dependent lysis on the META 2815 cells. The cytotoxic activity of the 3C_23K-YB2/0 antibody (EC50=30.5 ng/ml) is about 146 times greater than that of the 3C_23K-CHO antibody (EC50=466.9 ng/ml).

(184) Taken together, these results indicate that the 3C_23K anti-AMHRII antibodies produced in YB2/0 have the capacity to induce lysis of the cells expressing the AMHRII antigen. The difference in EC50 between the anti-AMHRII-YB2/0 and anti-AMHRII-CHO antibodies suggests a particular advantage for the anti-AMHRII-YB2/0 antibody under conditions of low antigenic expression, or of low penetrance of the antibody to the tumour.

Example 10

Studies of Cellular Proliferation

(185) Inhibition of cellular proliferation was demonstrated by measuring cell growth over time in the presence or absence of the anti-AMHRII antibodies tested.

(186) Briefly, the target cells (cov434-AMHRII, Asc 1, META2815) are cultured in P6 plates (1 10.sup.5 cells/well) for 72 h at 37 C., in the presence of the anti-AMHRII antibodies (10 g/ml) expressed in CHO or YB2/0, with or without cross-linking agent (AffiniPure F(ab)2 Fragment Goat Anti-Human IgG, Fc Fragment Specific ref: 109-006-008, Jackson Immunoresearch, France). The cells are treated with trypsin for 5 minutes and then counted in the CEDEX, an automatic cell counter based on cellular viability (trypan blue). A positive control of inhibition of proliferation is established in the presence of 1 g/ml of colchicine (Ref: C3915, Sigma-Aldrich, France). A negative control is established in the presence of a non-relevant antibody (anti-P24). All the dilutions are prepared in culture medium (RPMI, 10% FCS). The results are expressed as percentage proliferation, the value 100% corresponding to the proliferation of the cells observed in the absence of antibody.

(187) The results obtained with the cov434-AMHRII line are presented in FIG. 30.

(188) According to these observations the antibodies 3C_23K-YB2/0 and 3C_23K-CHO induce about 40% inhibition of cellular proliferation of the cov434-AMHRII cells, in the presence of a cross-linking agent (CK). This cytostatic effect is not observed in the presence of a non-relevant antibody (antibody p24) whereas the colchicine positive control (10 g/ml) induces 88% inhibition.

(189) Similarly, the results presented in FIG. 31 show that the antibodies 3C_23K-YB2/0 and 3C_23K-CHO induce about 40% inhibition of cellular proliferation on the META 2815 line in the presence of a cross-linking agent (CK).

(190) This inhibition of cellular proliferation might be the consequence of cellular signalling induced by the anti-AMHRII antibodies on the cov434-AMHRII and META 2815 lines.

Example 11

Effect In Vivo of the 3C 23K-YB2/0 Antibody on COV434-AMHRII Tumours

(191) The antitumour efficacy of the 3C_23K-YB2/0 antibody was evaluated in late treatment on female Swiss nude mice after subcutaneous injection (s.c.) of COV434-AMHRII tumour cells. The intraperitoneal (i.p.) injections (inj) of antibody were performed at intervals of 2-3 days at a dose of 10 mg/kg/inj for a total of 18 injections. The group treated with the 3C_23K-YB2/0 antibody was compared with the group treated with the vehicle (PBS). Material and Methods

(192) Female Swiss nude mice were used (Harlan). On day 0 of the experiment, the mice were given a subcutaneous injection of 7.10.sup.6 COV434-AMHRII tumour cells mixed with Matrigel (ratio 1:1). The animals were then treated by i.p. injection of PBS or 3C_23K-YB2/0 with 10 mg/kg/inj starting from day 16 (tumour volume between 84 and 270 mm.sup.3, 3 injections per week for 6 weeks (total 18 injections).

(193) Tumour volume was measured 2 to 3 times per week. Tumour volume (TV) was calculated using the following formula:
TV (mm.sup.3)=(lengthwidthheight)/2, in which the length corresponds to the largest diameter of the tumour and the width corresponds to the smallest diameter of the tumour.

(194) The curves of tumour growth were plotted using the mean of the tumour volumes (MTV). The animals were euthanased when the individual tumour volume had reached 2000 mm.sup.3. In each of the groups, the curves were stopped when 30% of the animals in the group had been euthanased.

(195) The inhibition of tumour growth (T/C), defined as the ratio of the median tumour volume of the treated groups relative to the control group treated with the vehicle, was calculated as follows: T/C=(median TV of the treated group/median TV of the vehicle group)100 T/C above 42%, the product is considered to be ineffective. T/C between 42% and 10%, the product has an anti-tumour effect. T/C below 10%, the product is truly effective.

(196) The statistical differences between the different groups were obtained with the Kruskal-Wallis test, using the ANOVA comparison (Statgraphics centurion XV software). The differences were regarded as significant if P<0.05. A logrank test, for comparing the survival parameters of the study, was also performed via ANOVA (Statgraphics centurion XV software). The differences were regarded as significant if P<0.05. Results

(197) The 3C_23K-YB2/0 antibody shows anti-tumour activity, since a delay is observed in COV434-AMHRII (FIGS. 32A and 32B).

(198) Statistical comparison of the tumour volumes at each measurement point, once the treatment has been started, shows that the 3C_23K-YB2/0 antibody delays tumour growth (Kruskal-Wallis, via ANOVA). The T/C ratio calculated between the groups treated with 3C_23K-YB2/0 and vehicle shows a significant difference at all the measurement points, also once the treatment has been started. The logrank test also shows that in terms of survival, the group treated with 3C_23K-YB2/0 is statistically different from the group treated with the vehicle.

(199) Table XI below shows the evolution of the tumour volumes (treated/control ratio, T/C in %) under the effect of the treatment with 3C_23K-YB2/0 in the cov434-AMHRII model.

(200) TABLE-US-00012 TABLE XI Measurement day 15 21 25 30 32 T/C % 108 60 42 22 13

(201) Table XII below presents the statistical analyses obtained in the cov434-AMHRII model.

(202) TABLE-US-00013 TABLE XII Measurement ANOVA Kruskall-Wallis day F-ratio P-value Sig. Test P-value Sig. 15 0.03 0.8628 0.00577 0.93945 21 8.34 0.0098 * 5.67427 0.01721 * 5 12.95 0.0021 * 10.56570 0.00115 * 30 34.94 0.0000 * 13.73030 0.00021 * 32 39.04 0.0000 * 12.90670 0.00033 *

Example 12

Effect In Vivo of the 3C_23K-YB2/0 Antibody on Asc1A5 Tumours

(203) The antitumour efficacy of the 3C_23K-YB2/0 antibody was evaluated in late treatment on female Swiss nude mice after subcutaneous injection (s.c.) of Asc1A5 tumour cells (clone of the original Asc 1 line). The intraperitoneal (i.p.) injections (inj) of the antibody were performed at intervals of 2-3 days at a dose of 10 mg/kg/inj for a total of 18 injections. The group treated with the 3C_23K-YB2/0 antibody was compared with the group treated with the vehicle (PBS). Material and Methods

(204) Female Swiss nude mice were used (Harlan). On day 0 of the experiment, the mice were given a subcutaneous injection of 7.10.sup.6 Asc1A5 tumour cells mixed with Matrigel (1:1 ratio). The animals were then treated by i.p. injection of PBS or 3C_23K-YB2/0 with 10 mg/kg/inj starting from day 12 (tumour volume between 40 and 160 mm.sup.3, 3 injections per week for 6 weeks (total 18 injections)).

(205) Tumour volume was measured 2 to 3 times per week. Tumour volume (TV) was calculated using the following formula:
TV (mm.sup.3)=(lengthwidthheight)/2, in which length corresponds to the largest diameter of the tumour and width corresponds to the smallest diameter of the tumour.

(206) The curves of tumour growth were plotted using the mean of the tumour volumes (MTV). The animals were euthanased when the individual tumour volume had reached 2000 mm.sup.3. In each of the groups, the curves were stopped when 30% of the animals in the group had been euthanased.

(207) The inhibition of tumour growth (T/C), defined as the ratio of the median tumour volume of the treated groups to the control group treated with the vehicle, was calculated as follows: T/C =(median TV of treated group/median TV of vehicle group)100 T/C above 42%, the product is considered to be ineffective. T/C between 42% and 10%, the product has an anti-tumour effect. T/C below 10%, the product is truly effective.

(208) The statistical differences between the different groups were obtained with the Kruskal-Wallis test, using the ANOVA comparison (Statgraphics centurion XV software). The differences were regarded as significant if P<0.05. A logrank test, for comparing the survival parameters of the study, was also performed via ANOVA (Statgraphics centurion XV software). The differences were regarded as significant if P<0.05. Results

(209) The 3C_23K-YB2/0 antibody shows anti-tumour activity, since a delay is observed in tumour growth compared to the group treated with the vehicle in the Asc1A5 model (FIGS. 33A and 33B).

(210) Statistical comparison of the tumour volumes at each measurement point, once the treatment has been started, shows that the 3C_23K-YB2/0 antibody delays tumour growth (Kruskal-Wallis, via ANOVA). The T/C ratio calculated between the groups treated with 3C_23K-YB2/0 and vehicle shows a significant difference at all the measurement points, also once the treatment has been started. The logrank test also shows that in terms of survival, the group treated with 3C_23K-YB2/0 is statistically different from the group treated with the vehicle.

(211) Table XIII below presents the evolution of the tumour volumes (treated/control ratio, T/C in %) under the effect of the treatment with 3C_23K-YB2/0 obtained in the Asc1a5 model.

(212) TABLE-US-00014 TABLE XIII Measurement day 12 17 24 27 31 35 T/C % 101 33 8 8 7 6
Table XIV below presents the statistical analyses obtained in the Asc1a5 model.

(213) TABLE-US-00015 TABLE XIV Measurement ANOVA Kruskall-Wallis day F-ratio P-value Sig. Test P-value Sig. 12 0.02 0.8817 0.0995 0.7523 17 29.19 0.0001 * 11.3108 0.0007 * 24 103.56 0 * 11.2941 0.0007 * 27 32.08 0.0001 * 11.3108 0.0007 * 31 33.37 0 * 11.3274 0.0007 * 35 57.97 0 * 10.5788 0.0011 *

Example 13

Effect In Vivo of the 3C_23K-YB2/0 Antibody on Meta2815 Tumours

(214) The antitumour efficacy of the 3C_23K-YB2/0 antibody was evaluated in late treatment on female Swiss nude mice after subcutaneous injection (s.c.) of Meta 2815 tumour cells. The intraperitoneal (i.p.) injections (inj) of antibody were performed at intervals of 2-3 days at a dose of 10 mg/kg/inj for a total of 18 injections. The group treated with the 3C_23K-YB2/0 antibody was compared with the group treated with the vehicle (PBS). Material and Methods

(215) Female Swiss nude mice were used (Harlan). On day 0 of the experiment, the mice were given a subcutaneous injection of 8.10.sup.6 Meta2815 tumour cells. The animals were then treated by i.p. injection of PBS or 3C_23K-YB2/0 with 10 mg/kg/inj starting from day 33 (tumour volume between 45 and 240 mm.sup.3, 3 injections per week for 6 weeks (total 18 injections).

(216) Tumour volume was measured 2 to 3 times per week. Tumour volume (TV) was calculated using the following formula:
TV (mm.sup.3)=(lengthwidthheight)/2, in which length corresponds to the largest diameter of the tumour and width corresponds to the smallest diameter of the tumour.
The curves of tumour growth were plotted using the mean of the tumour volumes (MTV). The animals were euthanased when the individual tumour volume had reached 2000 mm.sup.3. In each of the groups, the curves were stopped when 30% of the animals in the group had been euthanased.

(217) The inhibition of tumour growth (T/C), defined as the ratio of the median tumour volume of the treated groups to the control group treated with the vehicle, was calculated as follows: T/C=(median TV of treated group/median TV of vehicle group)100 T/C above 42%, the product is considered to be ineffective. T/C between 42% and 10%, the product has an anti-tumour effect. T/C below 10%, the product is truly effective.

(218) The statistical differences between the different groups were obtained with the Kruskal-Wallis test, using the ANOVA comparison (Statgraphics centurion XV software). The differences were regarded as significant if P<0.05. A logrank test, for comparing the survival parameters of the study, was also performed via ANOVA (Statgraphics centurion XV software). The differences were regarded as significant if P<0.05. Results

(219) The 3C_23K-YB2/0 antibody shows anti-tumour activity, since a delay is observed in tumour growth compared to the group treated with the vehicle in the Meta2815 model (FIGS. 34A and 34B).

(220) Statistical comparison of the tumour volumes at each measurement point, once the treatment has been started, shows that the 3C_23K-YB2/0 antibody delays tumour growth (Kruskal-Wallis, via ANOVA). The T/C ratio calculated between the groups treated with 3C_23K-YB2/0 and vehicle shows a significant difference at all the measurement points, also once the treatment has been started. The logrank test also shows that in terms of survival, the group treated with 3C_23K-YB2/0 is statistically different from the group treated with the vehicle.

(221) Table XV below presents the evolution of the tumour volumes (treated/control ratio, T/C in %) under the effect of the treatment with 3C_23K-YB2/0 in the META 2815 model.

(222) TABLE-US-00016 TABLE XV Measurement day 33 38 42 47 52 T/C % 92 38 26 22 21
Table XVI presents the statistical analyses obtained in the META 2815 model.

(223) TABLE-US-00017 TABLE XVI Measurement ANOVA Kruskall-Wallis day F-ratio P-value Sig. Test P-value Sig. 33 0 0.9899 0 1 38 11.21 0.0007 * 8.30769 0.0039 * 42 9.51 0.0116 * 7.41026 0.0064 * 47 12.7 0.0052 * 8.33684 0.0038 * 52 16.14 0.0024 * 8.30769 0.0039 *