Use of ADCC-optimized antibodies for treating low-responder patients

Abstract

The invention concerns the use of human or humanized chimeric monoclonal antibodies which are produced in selected cell lines, said antibodies bringing about a high ADCC activity as well as a high secretion of cytokines and interleukins, for treating underpopulations of so-called weak-response patients exhibiting CD16 FCGR3A-158F homozygote or FCGR3A-158V/F heterozygote polymorphism.

Claims

1. A method for treating haemolytic disease of the newborn, Sezary Syndrome, chronic myeloid leukaemias, chronic lymphoid leukaemias (CLL-B), cancer, breast cancer, conditions related to the environment, infectious diseases, chronic fatigue syndrome (CFS), parasitic infections, or viral infections, comprising: administering a composition of antibodies specific to the condition to be treated, wherein the condition to be treated has a number of antigenic sites below 250,000 per target cell, to a patient homozygous for phenylalanine in position 158 of CD16 (FCGR3A-158F homozygotes) or heterozygous for valine/phenylalanine in position 158 of CD16 (FCGR3A-158V/F), wherein said antibodies are biantennary type, wherein said antibodies have an ADCC level via CD16 (FcgammaRIIla) above 60% compared with the same antibody produced in a CHO line.

2. The method according to claim 1, wherein said patient is homozygous for phenylalanine in position 158 of CD16 (FCGR3A-158F homozygotes).

3. The method according to claim 1, wherein the condition to be treated has a number of antigenic sites below 100,000 per target cell.

4. The method according to claim 3, wherein the condition to be treated has a number of antigenic sites below 50,000 per target cell.

5. The method according to claim 3, wherein the condition to be treated has a number of antigenic sites below 10,000 per target cell.

6. The method according to claim 1, wherein the dose of said antibody administered to the patient is between 2 and 100 times lower than a dose of an antibody of the same specificity but of different glycosylation or produced in a CHO line.

7. The method according to claim 6, wherein the dose of said antibody administered to the patient is between 5 and 25 times lower than a dose of an antibody of the same specificity but of different glycosylation or produced in a CHO line.

8. The method according to claim 1, wherein said method is for treating a cancer of positive HLA class-II cells, B-cell lymphomas, acute B-cell leukaemias, Burkitt's syndrome, Hodgkin's lymphoma, myeloid leukaemias, chronic B-cell lymphoid leukaemias (CLL-B), non-Hodgkin's T-cell leukaemias and lymphomas and chronic myeloid leukaemias.

9. The method according to claim 1, wherein the antibody is selected from the group consisting of anti-HLA-DR, anti-CD20, anti Ep-CAM, anti HER2, anti CD52, anti HER1, anti GD3, anti CA125, anti GD, anti GD2, anti CD-23 and anti Protein C, anti-KIR3DL2, anti-EGFR, anti-CD25, anti-CD38, anti-CD30, anti-CD33, anti- CD44, anti-Rhesus, and anti-viral antibodies.

10. The method according to claim 9, wherein the antibody is anti-HLA-DR.

11. The method according to claim 9, wherein the antibody is anti-CD20.

12. The method according to claim 1, wherein said antibodies have an ADCC level via CD16 (FcgammaRIIIa) above 70% compared with the same antibody produced in a CHO line.

13. The method according to claim 1, wherein said antibodies have an ADCC level via CD16 (FcgammaRIIIa) above 80% compared with the same antibody produced in a CHO line.

14. The method according to claim 1, wherein said antibodies have an ADCC level via CD16 (FcgammaRIIIa) above 90% compared with the same antibody produced in a CHO line.

Description

LEGENDS OF THE FIGURES

(1) FIG. 1: ADCC activity of the anti-Rhesus D monoclonal antibody R297 (IgG1 T125 produced in YB2/0) and polyclonal antibodies WinRho in the presence of effector cells (PBMC: Peripheral Blood Mononuclear Cells) of different blood donors and polyvalent immunoglobulins (Tegeline 500 g/ml).

(2) FIG. 2: ADCC induced by anti-Rhesus D antibodies (polyclonal antibodies WinRho, T125 produced in YB2/0 and T125 produced in CHO) on NK cells of donors of homozygous phenotype CD16 FCGR3A-158F.

(3) FIG. 3: ADCC induced by anti-Rhesus D antibodies (polyclonal antibodies WinRho, T125 produced in YB2/0 and T125 produced in CHO) on NK cells of donors of homozygous phenotype CD16 FCGR3A-158F.

(4) FIG. 4: Activation of homozygous FCGR3A-158F CD16 Jurkat by the anti-Rhesus D antibodies T125 expressed in YB2/0 and CHO.

(5) FIG. 5: Activation of homozygous FCGR3A-158V CD16 Jurkat by the anti-Rhesus D antibodies T125 expressed in YB2/0 and CHO.

(6) FIG. 6: Activation of homozygous FCGR3A-158F CD16 Jurkat induced by the anti-HLA-DR antibodies expressed in YB2/0 and CHO.

(7) FIG. 7: Activation of homozygous FCGR3A-158V CD16 Jurkat induced by the anti-HLA-DR antibodies expressed in YB2/0 and CHO.

EXAMPLES

Example 1: ADCC Activity of the Anti-Rhesus D Monoclonal Antibody R297 Compared with the Polyclonal Anti-D Antibodies on a Group of 107 Blood Donors

(8) The respective capacities of the anti-Rhesus D R 297 monoclonal antibody and polyclonal anti-D antibodies to lyse erythrocytes in the presence of effector cells of different individual donors are compared (FIG. 1).

(9) The effector cells come from a group of 107 blood donors. The mononuclear cells (PBMC) are isolated from a blood bag by centrifugation on a Ficoll gradient (Pack Plus Pharmacia). The platelets are removed by centrifugation (190 g, 15 min) and the residual red blood cells are lysed with NH4Cl. The cells are washed and resuspended at 810.sup.7 cells/ml in IMDM. The red blood cells obtained from therapeutic concentrates (group O, Rhesus+) are treated for 10 min with papain (1 mg/ml) then washed three times in a saline buffer and adjusted to a concentration of 410.sup.7/ml or 210.sup.7/ml (NK test) in IMDM.

(10) The test is performed in a plate with 96 wells (NUNC). The culture supernatants or the purified antibodies (100 l to 200 ng/ml in IMDM+0.5% FBS), the effector cells (25 l), the red blood cells (25 l) and the polyvalent immunoglobulins (Tegeline, LFB) (50 l) are incubated for 16 h at 37 C. in a CO2-enriched atmosphere. For the non-specific lysis, the effector cells are replaced with IMDM. After 16 h at 37 C., the plates are centrifuged. 60 l of supernatants are collected and mixed with 60 l of 2.7 diaminofluorene (DAF, Sigma).

(11) After 5 min, the OD is measured at 620 nm.

(12) The percentage of lysis is estimated by using a calibration curve obtained with different dilutions of red blood cells lysed with NH4Cl, corresponding to 100%, 75%, 50%, 25% and 0% lysis, respectively.

(13) On the basis of the genotypic study conducted on donors of the same geographic region, the present study estimates that the average distribution of the 107 donors with regard to the CD16 polymorphism is 27 FCGR3A-158F homozygotes, 20 FCGR3A-158V homozygotes and 60 FCGR3A-158V/F.

(14) The results show a wide variability in the capacity of the effector cells of the different subjects to induce lysis of the positive Rhesus erythrocytes, regardless of the antibodies tested. The antibodies expressed in the YB2/0 cell line have a cytolytic activity comparable to that of the polyclonal antibodies, regardless of the donor studied, and consequently regardless of the CD16 polymorphism of the effector cells of the donor (see FIG. 1).

Example 2: ADCC Efficacy of the Antibodies Produced in CHO and YB2/0 According to the CD16 Polymorphism

(15) The same sequence encoding a specific IgG1 of the Rhesus D antigen was transfected into CHO and YB2/0 cell lines. The antibodies were incubated with positive Rhesus erythrocytes (target cells) and NK cells from 6 different donors (3 FCGR3A-158V homozygotes and 3 FCGR3A-158F homozygotes) previously genotyped for their CD16 phenotype in position 158. The NK cells are isolated using the magnetic bead separation technique (MACS) of Myltenyi. The NK cells are washed and resuspended at 210.sup.7/ml and/or 610.sup.7/ml in IMDM. The red blood cells are adjusted to a concentration of 210.sup.7/ml in IMDM. The Tegeline is replaced by the IMDM. Aside from these modifications, the test is identical to the ADCC test with PBMC.

(16) The cytotoxic activity of the antibodies on the erythrocytes (ADCC) was evaluated (FIG. 2 and FIG. 3).

(17) The antibody produced in the CHO line induced a lower lysis than the antibody produced in YB2/0, regardless of the donor's phenotype. The R297 antibody expressed in the YB2/0 line induced, from the lowest concentrations, a stronger cytolytic activity. At the maximum concentration of 25 ng/ml, the two antibodies induced the same percentage of ADCC.

(18) At concentrations below 25 ng/ml, the difference in lysis between the antibody produced in CHO and that produced in YB2/0 is greater in the homozygous FCGR3A-158F donors than in the homozygous FCGR3A-15BV donors. At a concentration of 2.5 ng/ml, in the presence of homozygous FCGR3A-158V NK cells, the antibody produced by CHO induced 54% lysis while that produced by YB2/0 induced 89% lysis, that is, a 56% increase. By contrast, at the same concentration, in the presence of homozygous FCGR3A-158F NK cells, the antibody produced by CHO induced only 22% lysis while that produced by YB2/0 induced 74% lysis, that is a 236% increase.

(19) The antibody expressed in the YB2/0 line therefore proved to be a better product for treating the patients giving a low lysis with the antibodies produced in CHO. Thus, the difference in ADCC activity between the FCGR3A-158V and homozygous FCGR3A-158F patients is lower with the antibody expressed in YB2/0 (89% and 74%) by comparison with that observed with the antibody expressed in CHO (56% and 22%).

(20) The optimised antibodies have a response that therefore appears to be less dependent on the polymorphic forms of the CD16.

(21) In addition, the monoclonal antibody expressed in YB2/0 always induces a lysis greater than or equal to the polyclonal antibodies.

Example 3: Comparison of the Activation of Homozygous FCGR3A-158F CD16 Jurkat Cells and Homozygous FCGR3A-158F CD16 Jurkat Cells Induced by Anti-Rhesus Antibodies Produced in CHO and YB2/0 Respectively: Evaluation of IL2 Production

(22) This test estimates the capacity of the antibodies to bind to the CD16 receptor (Fc gamma RIII) expressed on the CD16 Jurkat cells and to induce the secretion of IL2.

(23) The same sequence encoding an IgG1 (T125) specific to the Rhesus D antigen was transfected into the CHO and YB2/0 cell lines. The antibodies are incubated with positive Rhesus erythrocytes (target cell) and CD16 Jurkat cells (effector cells). Two types of Jurkat cells were used: 1cells transfected with the gene encoding an RFc bearing the amino acid phenylalanine F in position 158 (form F), 2cells transfected with the gene encoding an RFc bearing the amino acid valine V in position 158 (form V). The amount of cytokine (IL2) secreted by the CD16 Jurkat cells was measured by ELISA.

(24) In a 96-well plate in mixture: Antibody: 50 l of a dilution of 50; 37.5; 25; 18.75;

(25) 12.5; 9.4; 6.25; 3.125 ng/ml in IMDM (Iscove's Modified Dulbecco's) Medium 5% FBS (foetal bovine serum)

(26) PMA 50 l of a dilution at 40 ng/ml in IMDM 5% FBS

(27) Red blood cells treated with papain. 50 l at 810.sup.6/ml in IMDM 5% FBS

(28) Jurkat CD16. 50 l at 210.sup.6/ml in IMDM 5% FBS

(29) Incubation 1 night at 37 C.

(30) Then centrifugation of plates, collection of 100 l of supernatants and assay of IL2 with the commercial kit (Quantikine of R/D). Reading at 450 nm.

(31) The antibody expressed in the YB2/0 line is capable of inducing a higher secretion of IL2, contrary to the antibody expressed in CHO, regardless of the CD16 phenotype (FIGS. 4 and 5).

(32) The antibody produced in CHO does not induce the secretion of IL2 from homozygous FCGR3A-158F CD16 Jurkat and only a very small production of IL2 with the homozygous FCGR3A-158V form.

(33) As one of the special features of the Rhesus system is the low expression of the antigen at the membrane surface, it appears that under these conditions, the antibody expressed in the YB2/0 line is a much better product for activating the effector cells with the homozygous form FCGR3A-158F of CD16 which do not appear to be capable of being activated with the antibody produced in CHO. As regards the homozygous form FCGR3A-158V, very small amounts of antibodies produced by YB2/0 (<1.56) make it possible to induce an activation comparable to that obtained with higher concentrations of antibodies produced in CHO (12.5 ng/ml).

(34) With the homozygous form FCGR3A-158F and at a concentration of 12.5 ng/ml, the antibody produced in CHO induces a secretion of IL2 (18 pg/ml) lower than 2% of that induced by the antibody produced in YB2/0 (1435 pg/ml). This corresponds to an increase of more than 7000%, when the antibody produced in YB2/0 is used, by comparison with the antibody produced in CHO.

(35) With the homozygous form FCGR3A-158V and at a concentration of 12.5 ng/ml, the antibody produced in CHO induces a secretion of IL2 (869 pg/ml) lower than 8% of that induced by the antibody produced in YB2/0 (12312 pg/ml). This corresponds to an increase of more than 1300%, when the antibody produced in YB2/0 is used, by comparison with the antibody produced in CHO.

Example 4: Comparison of the Activation of Homozygous FCGR3A-158F CD16 Jurkat Cells and Homozygous FCGR3A-158V Jurkat Cells Induced by Two Anti-HLA-DR Antibodies Expressed in CHO and YB2/0, Respectively: Evaluation of the Secretion of IL2

(36) The same sequence encoding an IgG1 specific to the HLA-DR antigen was transfected into the CHO and YB2/0 cell lines. The antibodies are incubated with Raji cells (positive HLA-DR target cell) and CD16 Jurkat cells (effector cells). Two types of Jurkat cells were used: 1cells transfected with the gene encoding an RFc bearing the amino acid phenylalanine F in position 158 (form F), 2cells transfected with the gene encoding an RFc having the amino acid valine V in position 158 (form V). The quantity of cytokines (IL2) secreted by the CD16 Jurkat cells was measured by ELISA.

(37) In 96-well plates in mixture:

(38) Antibody: 50 l of a dilution of 50; 37.5; 25; 18.75; 12.5; 9.4; 6.25; 3.125 ng/ml in IMDM 5% FBS

(39) PMA 50 l of a dilution at 40 ng/ml in IMDM 5% FBS

(40) Raji cells: 50 l at 610.sup.5/ml in IMDM 5% FBS

(41) Jurkat CD16. 50 l at 2010.sup.6/ml in IMDM 5% FBS

(42) Incubation 1 night at 37 C.

(43) Then centrifugation of plates, collection of 100 l of supernatants and assay of IL2 with the commercial kit (Quantikine of R/D). Reading at 450 nm.

(44) The antibody expressed in the YB2/0 line is capable of inducing a higher secretion of IL2, contrary to the antibody expressed in CHO, regardless of the CD16 phenotype (FIGS. 6 and 7).

(45) Unlike the Rhesus system on erythrocytes, the expression of the HLA-DR antigen at the membrane surface of the Raji cell is not low (200,000 to 400,000 copies). Under these conditions, it appears that the antibody expressed in the YB2/0 line is a much better product for activating the effector cells with the form F158 as well as V158 of CD16, and more specifically with low antibody concentrations. Thus, the concentration of 1.56 ng/ml, the antibody produced in CHO induces a secretion of IL2 (2410 pg/ml) lower than 15% of that induced by the antibody produced in YB2/0 (16952 pg/ml). This corresponds to an increase of more than 600% when the antibody produced in YB2/0 by comparison with the antibody produced in CHO.

(46) At a concentration of 12.5 ng/ml, the antibody produced in CHO induces a secretion of IL2 (14597 pg/ml) lower than 45% of that induced by the antibody produced in YB2/0 (34823 pg/ml). This corresponds to an increase of more than 100% when the antibody produced in YB2/0 is used, by comparison with the antibody produced in CHO.

(47) The antibody expressed in the YB2/0 line is therefore a much better product for inducing the secretion of cytokines of effector cells with the polymorphic form V158 (FCGR3A-158V homozygotes) and F158 (FCGR3A-158F homozygotes).