Pooled NK cells from umbilical cord blood associated with antibodies and their uses for the treatment of disease

Abstract

The invention relates to the field of cell therapy, particularly NK cell mediated therapy associated with antibodies. The present invention is directed to methods and compositions for increasing the efficiency of therapeutic natural killer cells (NK cells) and/or antibodies, wherein said methods or compositions comprise the use of pooled NK cells from umbilical cord blood units (UCBs), preferably alloreactive NK cells, in combination with a therapeutic antibody in order to enhance the efficiency of the treatment in human subjects, in particularly through an increase in antibody-dependent cell-mediated cytotoxicity (ADCC) mechanism. The present invention relates to said composition as a pharmaceutical composition, preferably for its use for the treatment of a disease in a human subject in need thereof, preferably wherein said disease is a cancer, infectious or immune disease. Finally, the present invention is also directed to a method of treatment of a disease in a human subject in need thereof, comprising the administering to said subject said pooled NK cells from UCBs, preferably alloreactive, in combination with a therapeutic antibody which can be bound to said NK cells.

Claims

1. A method for treating cancer comprising the administration to a subject in need of treatment for cancer of a therapeutically effective amount of a first composition and a second composition, wherein said first composition comprising a population of alloreactive natural killer cells (NK cells) which are derived from a mixture of at least n umbilical cord blood units (UCB units) pooled from two or more different donors, or a mixture of fraction thereof containing said NK cells, with n2; and said second composition comprising a monoclonal antibody or a specific binding fragment thereof directed to a cell receptor antigen of cells of said subject; wherein the population of alloreactive NK cells comprised in said first composition are obtained by a method comprising: (A) providing at least n umbilical cord blood units (UCB units), or fraction thereof, each UCB unit or fraction thereof comprising NK cells and T cells, with n2; (B) pooling said at least n UCB units from two or more different donors, or pooled fraction thereof, and depleting said T cells to produce the population of cells comprising pooled NK cells; (C) expanding and activating said pooled NK cells in a medium comprising accessory cells to produce a population of expanded activated NK cells; and (D) recovering said expanded activated NK cells; and wherein the pattern for major HLA class I groups genotype is the same in each of said n UCB units pooled from two or more different donors, or pooled fraction thereof, and is selected from the group consisting of HLA A3/A11, which is recognized by KIR3DL2; HLA Bw4, which recognized by KIR3DL1; HLA C group 1, which is recognized by KIR2DL2/3; and HLA C group 2, which is recognized by KIR2DL1; and wherein in step (C) said pooled NK cells and said accessory cells are HLA-KIR mismatched.

2. A method for treating cancer according to claim 1, wherein said first and second composition are administered simultaneously, separately or sequentially.

3. A method for treating cancer according to claim 1, wherein said cancer is selected from cancers overexpressing a cell receptor at the surface of the tumoral cells of the subject in need of the treatment, and wherein the monoclonal antibody of said second composition is directed against said receptor.

4. A method for treating cancer according to claim 1, wherein said second composition comprises a monoclonal antibody which is a low or non-fucosylated antibody.

5. A method for treating cancer according to claim 1, wherein said second composition comprises a monoclonal antibody selected the group consisting of anti-CD20, anti-HER2/Neu and anti-EGFR antibodies.

6. A method for treating cancer according to claim 1, wherein said second composition comprises a monoclonal antibody selected the group consisting of Ublituximab, Trastuzumab and Trastuzumab-like antibodies.

7. A method for treating cancer according to claim 1, wherein cancer to be treated are selected from the group of hematologic malignancy tumor cells, solid tumor cells or carcinoma cells, acute T cell leukemia cells, chronic myeloid lymphoma (CML) cells, acute myelogenous leukemia cells, chronic myelogenous leukemia (CML) cells, multiple myeloma cells, or lung, colon, prostate, glyoblastoma cancer.

8. The method of claim 1, wherein n is: 2<n100.

9. The method of claim 1, wherein n is: 3n50.

10. The method of claim 1, wherein said accessory cells are irradiated.

11. The method of claim 1, wherein said accessory cells are immortalized.

12. The method of claim 1, wherein said accessory cells are immortalized by Epstein Barr Virus (EBV) transformation.

13. The method of claim 1, wherein said accessory cells are cells from HLA-typed mammals.

14. The method of claim 1, wherein said provided UCB units are thawed UCB units from frozen stored UCB units.

15. The method of claim 1, wherein the pooled NK cells are expanded for from 9 to 28 days and the amplification factor for NK cells after the expanding is at least 100.

16. The method of claim 1, wherein the amplification factor for NK cells after the expanding is at least 300.

17. The method of claim 1, wherein in step A), each of the n UCB units are T cells depleted before the step B) of pooling.

Description

DESCRIPTION OF FIGURES

(1) FIGS. 1-1 to 1-3 (sub-figures1, 2 and 3 of FIG. 1) is a schema illustrating an example of a manufacture process of the present invention.

(2) FIGS. 2 and 3 illustrate the NK proliferation obtained after or without CD3 depletion.

(3) FIG. 4 illustrates the NK proliferation obtained from pooled CD3-depleted UCB units.

(4) FIG. 5 illustrates the NK proliferation obtained from 5 pooled CD3-depleted UCB units.

(5) FIG. 6 illustrates the NK proliferation obtained from pooled UCB units without prior CD3-depletion.

(6) FIG. 7 illustrates the NK proliferation from pooled UCB units after 9 days of culture with CD3-non depleted UCBs.

(7) FIG. 8 illustrates the NK proliferation amplification factor obtained with 2 KIR-HLA matched UCBs and amplified with PLH accessory cells

(8) FIG. 9 illustrates the NK proliferation amplification factor obtained with 2 KIR-HLA mismatched UCBs amplified with PLH accessory cells.

(9) FIG. 10A-10B illustrates Example 7, first experiment.

(10) FIG. 11A-11B illustrates Example 7, second experiment.

(11) FIGS. 12A-12C: Percentage of patient cells lysis induced by the combination of NK cells+mAb for p45 B lymphoma patient cells (FIG. 12A) and for B-CLL p53 and p46 patient cells (FIGS. 12B and 12C).

(12) FIGS. 13A-13D: Percentage of cell line cells lysis induced by the combination of NK cells+mAb for Daudi cell line (FIG. 13A), Raji cell line (FIG. 13B), Ri-1 cell line (FIG. 13C) and SUDHL4 cell line (FIG. 13D).

(13) FIGS. 14A-14D: Study of anti-CD20 dose effect (E:T 1/1) on cell lines.

(14) Percentage of cell line cells lysis induced by the Anti-CD20 mAb or the combination of NK cells+mAb for Daudi cell line (FIG. 14A), Raji cell line (FIG. 14B), Ri-1 cell line (FIG. 14C) and SUDHL4 cell line (FIG. 14D).

(15) FIGS. 15A and 15B illustrates Rituximab cytotoxicity dose effect and ADCC with Rituximab+NK001 dose effect on Daudi cell line

(16) FIG. 16 illustrates the NK001 and NK001+Rituximab dose effect on Daudi ADCC

(17) FIG. 17 illustrates mAb anti CD20 cytotoxicity on clinical sample

(18) FIG. 18 illustrates ADCC induced by NK-001 in synergy with anti CD20 on clinical sample

(19) FIGS. 19A-19C illustrates the reproducibility of the results obtained for ADCC with 3 lots of NK001 on p15-30 (A), p44 (B) and P16-01 (C) clinical samples, between R603 and Rituximab

(20) FIG. 20 illustrates Anti-CD20 mAb binding on NK001

(21) FIGS. 21A and 21B illustrate the direct binding analyse on Daudi (high CD20) (A) or on Raji (low CD20) (B) between TG20 and Rituximab

(22) FIGS. 22A and 22B illustrate the binding analyses by inhibition of CD20 labelling on Daudi (A) or Raji (B) cell line between TG20 and Rituximab

(23) FIGS. 23A and 23B illustrate the quantification of apoptotic cells in proportion to the dose of anti CD20 mAb after 24 h of incubation on Daudi (A) or Raji (B) cell line between TG20 and Rituximab

(24) FIGS. 24A, 24B and 24C illustrate the complement Dependent Cytotoxicity (CDC) analyse on Daudi (A) or Ri-1 (B) and SUDHL4 (C) cell line between TG20 and Rituximab

(25) FIG. 25 illustrates the cytotoxicity of anti CD20 mAb: dose effect between TG20 and Rituximab

(26) FIG. 26 illustrates ADCC on Daudi induced by NK-001+anti CD20 mAb [10 g/ml]: NK001 dose effect

(27) FIG. 27 illustrates ADCC on Daudi induced by NK-001+anti CD20 mAb: mAb dose effect

(28) FIG. 28 illustrates the cytotoxicity of anti CD20 mAb on Daudi, Raji, Ri-1, SUDHL4 and HL60 cell line between TG20 and Rituximab

(29) FIG. 29 illustrates ADCC induced by NK-001 in synergy with anti CD20 mAb on Daudi, Raji, Ri-1, SUDHL4 and HL60 cell line between TG20 and Rituximab

(30) FIG. 30 illustrates the anti CD20 mAbs cytotoxicity on 8 different clinical samples between TG20 and Rituximab

(31) FIG. 31 illustrates ADCC induced by NK-001+anti CD20 mAbs on 8 different clinical samples between TG20 and Rituximab

(32) FIGS. 32A, 32B and 32C illustrate illustrates the reproducibility of the results obtained for ADCC with 3 lots of NK001 on p15-30 (A), p44 (B) and P16-01 (C) clinical samples between TG20 and Rituximab

(33) FIG. 33 illustrates Anti Her2 mAbs binding on NK001 between Herceptine-like and Trastuzumab

(34) FIG. 34 illustrates Anti-Her2 mAbs cytotoxicity induced by 10 g/ml of anti Her2 mAbs Herceptine-like and Trastuzumab on SKBR3 and SKOV3 cell line.

(35) FIGS. 35A and 35B illustrate ADCC induced by NK-001+anti Her2 mAb [10 g/ml] on SKBR3 (A) and SKOV3 (B) cell line with E:T ratio 5:1 for 3 h

(36) FIG. 36 represents images of bright field microscopy (10) showing the synergic effect observed between NK001 and anti-Her2 mAbs cytotoxicity on SKOV3 cell line between Herceptine-like and Trastuzumab

(37) FIG. 37 represents images of fluorescence microscopy (Cyto tell green 10) showing the synergic effect observed between NK001 and anti-Her2 mAbs cytotoxicity on SKOV3 cell line between Herceptine-like and Trastuzumab.

EXAMPLE 1: MATERIALS AND METHODS

(38) Cells:

(39) PLH (Example 4): no HLA-C1, ECACC bank no 88052047, IHW number 9047

(40) This cell line was obtained by EBV immortalization of B lymphocytes coming from a scandinavian woman. This cell is completely HLA genotyped and have the particularity to express HLA Class I alleles from C group 2, A3/A11 and Bw4 types but not from C group 1 (complete informations on IMGT/HLA database).

(41) This cell line is used as accessory cell for NK amplification/activation protocol because it allows to choose a specific HLA mismatch between accessory cell and UCBs (expressing HLA C group 1, and potentially the associated inhibitory receptor KIR2DL2/3). Being transformed by EBV infection increases its NK activation ability because of membranary expression of some viral induced ligands for NK activating receptors.

(42) HOM-2 (Example 4): no HLA-C2, ID n HC107505, IHW number 9005

(43) This cell line was obtained by EBV immortalization of B lymphocytes coming from a Canadian/North American woman. This cell is completely HLA genotyped and have the particularity to express HLA Class I alleles from C group 1, A3/A11 and Bw4 types but not from C group 2 (complete informations on IMGT/HLA database).

(44) This cell line is used as accessory cell for NK amplification/activation protocol because it allows to choose a specific HLA mismatch between accessory cell and UCBs (expressing HLA C group 2, and potentially the associated inhibitory receptor KIR2DL1). Being transformed by EBV infection increases its NK activation ability because of membrane expression of some viral induced ligands for NK activating receptors.

(45) Media, Buffers and Cytokines:

(46) 1/ Density gradient cell separation medium of Ficoll and sodium diatrizoate used for the separation of lymphocytes: Histopaque-1077 from Sigma Aldrich, Saint Louis, MO, USA 2/ Kit for counting cells and looking at their viability with the Muse machine, labelling the cells with 7AAD and a fluorescent DNA probe: count and viability kit from Millipore, Darmstadt, Germany 3/ Cellular culture medium: RPMI 1640 Glutamax from Invitrogen, Carlsbad, CA, USA, purchased from France distributor Thermo Fisher Scientific 4/ Nutrient source in cellular culture medium: Ftal Bovine Serum from Invitrogen, Carlsbad, CA, USA, purchased from France distributor Thermo Fisher Scientific 5/ Organic solvent for cells freezing: dimethyl sulfoxide, DMSO from B. Braun, Melsungen, Germany 6/ Buffer for flow cytometry labelling: PBS from Invitrogen, Carlsbad, CA, USA, purchased from France distributor Thermo Fisher Scientific 7/ Cytokine for NK amplification/activation: recombinant human rhIL-2 from ebioscience, San Diego, CA, USA 8/ Cytokine for NK amplification/activation: recombinant human rh-IL15 from Miltenyi, Bergisch Gladbach, Germany

EXAMPLE 2: EXAMPLE OF MANUFACTURING PROCESS

(47) Process details: see FIGS. 1-1 to 1-3 UCBs were processed by ficoll UCB mononuclear cells isolation before first freezing. CD3 depletions were done with a manual magnetic depletion kit. Pooled UCBs present the same pattern for major HLA class I groups genotype (each HLA group is recognized by a major inhibitory KIR by NK cells): HLA A3/A11, recognized by KIR3DL2; HLA Bw4, recognized by KIR3DL1; HLA C group 1, recognized by KIR2DL2/3; HLA C group 2 recognized by KIR2DL1. Pooled UCBs are activated with an accessory cell missing one of the HLA recognized by the expressed pooled UCBs iKIRs. NK cells were amplified for 20-24 days. Cytokines used are IL-2 (100 IU/ml) and IL-15 (5 ng/ml). These concentrations can be modified to obtain similar results. Accessory cells are EBV-immortalized cell lines (cells expressing virus induced activating ligands) with specific HLA genotypes (one major HLA class I group missing). Accessory cells can be irradiated by different ways with different irradiation doses (here we mainly used 20 seconds UV irradiation, but also 105 Gy gamma irradiation for the last experiment, that showed better amplification results). Irradiated accessory cells can be used with or without prior cryopreservation: freshly irradiated cells or as irradiated cryopreserved cells (irradiation just before freezing). For the 3 last experiments, irradiated accessory cells were added to the UCB cells at NK:accessory cell ratio 1:4, each 3-4 days (days 0; 4; 8; 12; +/15; +/18). Some results (previous and other not shown results) were obtained using ratio 1:2, or ratio total cells:accessory cells from 1:1 to 1:3, with addition frequencies from 3 days to 7 days. This parameter can be changed, still obtaining similar amplification/activation results. In the experiments shown we didn't perform the CD56+ selection at the end of the process because NK cells derived from pooled CD3-depleted UCBs represented already more than 90% of alive cells at the end of the process. The CD56 selection step is not essential, but will probably improve NK purity and be preferable (and potentially totally required) for a pharmaceutical product.

(48) Some steps of the process can be changed: UCBs will be processed differently before first freezing, using a GMP-compliant method such as Hetastarch or PrepaCyte CB device (or other existing and clinically accepted method). Even if current preclinical and clinical knowledge show that a iKIR-HLA mismatch gives better results than iKIR-HLA match, it is still possible that in our case iKIR-HLA has different influence in clinical outcome. So for the moment, the literature knowledge-based development should be with a process using NK/accessory cell mismatch and NK/patient same mismatch. Future preclinical and clinical data could change this parameter if unnecessary. NK amplification culture duration can be optimized: from 14 to 28 days. IL-2 and IL-15 concentrations can be optimized. The CD3-depletion will be done with an automatic clinically accepted device such as cliniMACS. The CD3-depletion can also be done just after erythrocyte elimination and volume reduction (maybe better results in term of NK recovery). One of the results demonstrates that in some undefined cases, the CD3-depletion is not necessary for UCB pooled NK cells good amplification/activation. To obtain an important quantity of activated multi-donors-derived NK cells characterized in a unique pharmaceutically defined lot, the preferentially CD3-depleted UCB units can be pooled at various moments of the process: before amplification culture, during amplification culture, or at the end of the amplification culture.

EXAMPLE 3: OBJECTIVES

1. First Experiment

(49) Because it is known that T lymphocytes from different donors will kill each other by HLA differences recognition, and because NK cells need activator signal to be cytotoxic, we asked whether it is possible to pool CD3-depleted UCBs expressing the same major HLA groups (depending their recognition by inhibitory KIR's) but not the same HLA alleles. Total mononuclear cells and CD3-depleted mononuclear cells from 3 UCBs were pooled to verify if CD3-depletion was essential.

2. Second Experiment

(50) Because we want to produce 4 class of NK cells presenting an iKIR-HLA mismatch for each major iKIR/HLA pair, we needed to investigate if success of pooling UCBs was only due to the first particular HLA genotyping used previously or could be reproduced with another HLA genotyping of UCBs: We asked whether another accessory cell line using another iKIR-HLA mismatch will allow NK amplification/activation from a pool of 3 CD3-depleted UCBs expressing the same HLA groups.

3. Third Experiment

(51) Because to treat around 100 patients we will need to pool 10 UCBs, we asked whether a pool of 5 UCBs (half) expressing the same HLA groups allow the same NK amplification/activation.

EXAMPLE 4: EXPERIMENTS CARRIED OUT

1. First Experiment

(52) UCB mononuclear cells obtained by Ficoll separation were cryopreserved, then thawed and CD3-depleted using a stem cell kit for a part. Three CD3-depleted or total UCBs with same the major HLA class 1 groups A3/A11+, Bw4+, C1+, C2+ genotype were pooled and cultured for 21-25 days with IL-2, IL-15 and irradiated accessory cells PLH (A3/A11+,Bw4+,C1,C2+ genotype) added each 4 days.

2. Second Experiment

(53) UCB mononuclear cells obtained by Ficoll separation were cryopreserved, then thawed and CD3-depleted using a stem cell kit. Three CD3-depleted UCBs with same the major HLA class 1 groups A3/A11,Bw4+,C1,C2+ genotype were pooled and cultured for 21-25 days with IL-2, IL-15 and irradiated accessory cells HOM-2 (A3/A11+,Bw4+,C1+,C2-genotype) added each 4 days.

3. Third Experiment

(54) UCB mononuclear cells obtained by Ficoll separation were cryopreserved, then thawed and CD3-depleted using a stem cell kit. Five CD3-depleted UCBs with same the major HLA class 1 groups A3/A11,Bw4+,C1+,C2 genotype were pooled and cultured for 21 days with IL-2, IL-15 and irradiated accessory cells PLH (A3/A11+,Bw4+,C1,C2+ genotype) added each 4 days.

4. Evaluated Parameters

(55) Alive NK cells were regularly counted using the MUSE Millipore system and flow cytometry characterization of cellular composition in the culture.

(56) Expression of activating markers of NK cells was regularly evaluated by flow cytometry (CD16 for potent synergistic effect with monoclonal antibody therapies; CD69 as common activating receptor).

(57) At day 20 of culture, cytotoxicity was evaluated against well-known K562 target cells, and tumoral cells for experiment 2 and 3 (2 h incubation with NK:K562 ratio 3:1, NK:purified B lymphoma cells ratio 3:1, NK:AML cells (in total PBMC sample of the patient) ratio 10:1).

EXAMPLE 5: RESULTS

1. First Experiment (See FIGS. 2 and 3)

(58) UCB 1: HLA A11:01/A29:02, B35:01/B44:02, C04:01/C16/01>HLA A3/A11+, Bw4+, C1+, C2+ UCB2: HLA A11:01/A23:01, B35:02/B49:01, C04:01/07:01>HLA A3/A11+, Bw4+, C1+, C2+ UCB3: HLA A2/A3, B18/B51, C5/C14>HLA A3/A11+, Bw4+, C1+, C2+

(59) NK proliferation from isolated UCBs show better results after CD3-depletion because T lymphocytes are in competition with NK cells for proliferation with the cytokines used (and CD8-T lymphocytes directed against EBV antigen are also stimulated by accessory cells).

(60) NK from pooled CD3-depleted UCBs proliferate similarly than from isolated UCBs, but if UCBs are not CD3-depleted, T lymphocytes from the different donors are cytotoxic for the other one and NK cells cannot proliferate.

(61) TABLE-US-00001 TABLE 1 Pooled CD3- CD3- CD3- CD3- Pooled depleted depleted depleted depleted UCB1 UCB2 UCB3 UCBs UCB1 UCB2 UCB3 UCBs NK Amplification Factor 2.6 17.8 15.7 1.6 20 14.9 76.7 23.9 % NK CD 16+ 72.7 80.2 72.9 46 54.4 63.6 63.6 68.3 (ADCC-related) % NK CD69+ 86.7 88.6 94.7 94.3 92.9 95.1 96 86.6 Common target lysis % ND ND ND ND 64.1 58 50.9 52.7

(62) NK amplification factor is relatively low in this experiment due to technical issue.

(63) Activating receptors are well expressed, and cytotoxicity against common target K562 of cultured NK cells is highly better than with un-activated NK cells.

(64) This experiment showed that pooling UCB with same major HLA groups genotyping for NK amplification is feasible but require prior CD3-depletion. Amplified NK cells are well-activated.

2. Second Experiment (See FIG. 4)

(65) UCB1: HLA A01/02, B27:05/B40:02/C02:02/C15:02>HLA A3/A11, Bw4+, C1, C2+ UCB2: HLA A2/A31, B50/B51, C06:02/15:02>HLA A3/A11, Bw4+, C1, C2+ UCB3: HLA A23/A24, B44/B44, C4/C5>HLA A3/A11, Bw4+, C1, C2+

(66) NK proliferation from pooled CD3-depleted UCBs with this new genotype is similar to NK proliferation with isolated CD3-depleted UCBs.

(67) TABLE-US-00002 TABLE 2 Pooled CD3- CD3-depleted CD3-depleted CD3-depleted depleted UCB1 UCB2 UCB3 UCBs NK Amplification Factor 86.3 184.4 47.1 124.7 % NK CD 16+ 86.3 81.6 99.8 90 (ADCC-related) % NK CD69+ 99.6 94.9 99.2 98.5 Common target lysis % 93 97.6 90.1 87.7 B Lymphoma tumoral 37 48.2 78.4 31.6 cells lysis

(68) NK amplification factor is higher in this experiment (no technical issue), but can still be improved by protocol optimization specifically for the new accessory cell line.

(69) Activating receptors are very well expressed. Cytotoxicity against common target K562 of cultured NK cells is highly better than with unactivated NK cells, and we observe a significant cytotoxicity against B lymphoma tumoral cells with a 2 hours incubation.

(70) Pooling CD3-depleted UCBs with another major HLA groups genotype, and amplifying NK cells with another iKIR-HLA mismatch and another accessory cell line is feasible. Amplified NK cells are well-activated.

3. Third Experiment (See FIG. 5)

(71) UCBs: HLA A3/A11, Bw4+, C1+, C2

(72) TABLE-US-00003 TABLE 3 Pooled CD3-depleted UCBs NK Amplification Factor 583.2 % NK CD 16+ 81.7 (ADCC-related) % NK CD69+ 99.8 Common target lysis % 97.9 AML umoral cells lysis 10.4

(73) NK proliferation from 5 pooled CD3-depleted UCBs is good.

(74) NK amplification factor is higher in this experiment.

(75) Activating receptors are very well expressed. Cytotoxicity against common target K562 of cultured NK cells is highly better than with unactivated NK cells, and we observe a small specific cytotoxicity against AML tumoral cells with a 2 hours incubation (but we couldn't observe cytotoxicity after 20 h because at this time patient cells died because of thawing).

(76) Pooling 5 CD3-depleted UCBs and amplifying NK cells with an important amplification factor is feasible with our manufacturing process. Amplified NK cells are well-activated.

4. Complementary Results

(77) Experiment showing good amplification of NK cells from pooled UCBs without prior CD3-depletion (no reproducibility assay): (see FIG. 6)

(78) 3 iKIR-HLA mismatch UCBs amplified with PLH: UCB 1: HLA A2:01/A68:01; B38:01/B57:01; C6:02/C12:03>C1+, C2+, A3/A11, Bw4+ UCB 2: HLA A1:01/A2:01; B52:01/B57:01; C6:02/C12:02>C1+, C2+, A3/A11Bw4+ UCB 3: HLA A02/02; B15:09/B50:02; C06/C07>C1+, C2+, A3/A11, Bw4

(79) NK amplification can be similar in isolated or pooled UCBs without prior CD3-depletion. Experiments showing possibility of pooling after 9 days culture (with CD3-non depleted UCBs): 1/ same previous experiment (see FIG. 7)

(80) TABLE-US-00004 TABLE 4 UCB1 UCB2 Pool D0 Pool D9 % B Lymphoma lysis 74 91 90 91

(81) It is possible to pool 9 days activated NK cells (here without prior CD3-depletion) keeping a significant but lower NK amplification. 2/ Experiment with 2 iKIR-HLA matched UCBs amplified with PLH: (see FIG. 8) UCB 1: HLA A11:01/A29:02, B35:01/B44:02, C04:01/C16/01>HLA A3/A11+, Bw4+, C1+, C2+ UCB2: HLA A11:01/A23:01, B35:02/B49:01, C04:01/07:01>HLA A3/A11+, Bw4+, C1+, C2+

(82) When NK didn't amplify properly in CD3-non depleted, pooling UCBs after 9 days amplification (increasing NK % and NK activation status, but still with high T lymphocytes %) seemed to overcome the problem. They showed an in vitro similar good cytotoxicity against B lymphoma tumoral cells (overnight, ratio E:T 1:1). 3/ Experiment with 2 iKIR-HLA mismatched UCBs amplified with PLH: (see FIG. 9) UCB1: HLA A02:02/30:01, B42:01/B53:01, C04:01/17:01>HLA A3/A11, Bw4+, C1, C2+ UCB2: HLA A11:01/A23:01, B35:02/B49:01, C04:01/07:01>HLA A3/A11+, Bw4+, C1+, C2

(83) TABLE-US-00005 TABLE 5 UCB1 UCB2 Pool D0 Pool D9 % B Lymphoma lysis 74 92 96 97

(84) NK cells from CD3-non depleted iKIR-HLA mismatched pooled UCBs showed a lower amplification factor, and pooling these UCBs after 9 days amplification gave better NK amplification. They showed an in vitro similar good cytotoxicity against B lymphoma tumoral cells (overnight, ratio E:T 1:1).

EXAMPLE 6: PERSPECTIVES

(85) 1. Process Optimization

(86) Preferably, the manufacturing process of pooled activated/expanded NK cells according to the present invention will be adapted to the pharmaceutical regulatory obligations, and every step of the process adapted for the best quality guarantee. First, and for example, acceptance criteria of UCB units must be set, such as more than 1.4 or 1.6 10.sup.6 total nucleated cells (currently 1.85 10.sup.6 total nucleated cells for our local UCB bank), with potentially a minimal threshold for the NK percentage such as 7% (3-15% NK generally observed in UCB total nucleated cells). The Ficoll method used in the above examples for UCB mononuclear cells (UCBMC) isolation can be easily replaced by well-adapted standard and well-known method for clinical application, and pharmaceutical conditions, for example using a closed sterile single use system with bags, using adapted procedures such as HES 6% and centrifugations erythrocytes elimination and volume reduction, or Prepacyte CB isolation system. These systems certainly improve the total nucleted cell recovery in the first step. Preferably, CD3-depletion of UCBMCs can be better adapted to regulatory compliances and/or GMP process for pharmaceutical uses, for example with an adapted clinically upgradable material such as CliniMACS, and by determining the best step time for CD3-depletion whether it is needed, before or after first cryopreservation step for the best cell recovery and the best CD3-depletion quality. Preferably, the freezing, cryopreservation and thawing procedures for UCBMC can be improved using authorized procedures for clinical applications after validation of the manufacturing process. Adapted material for bag closed system can be used and cryopreservation conditions (media, cell concentration) can be easily optimized by the skilled person for the method of the present invention. These optimization steps only should certainly improve the total cell recovery after thawing. In the same time, the acceptance criteria for each thawed UCBMCs to go further into the manufacturing process according to pharmaceutical guidelines should be set. Preferably, HLA-genotyping and inhibitory KIR expression evaluation procedures should be validated to select the different UCB units allowed to be pooled for the amplification/activation step: selection criteria should be set for each lot. Preferably, GMP compliant upgradable accessory cells, whether they will be included in the method of the invention, with a final screening on NK amplification:activation for clones selection. Final accessory cells must be well-characterized for use in a therapeutic agent production procedure. This optimization step could also improve NK amplification/activation results. Preferably, irradiation procedure will be optimized and validated for the best amplification/activation results with clinically adapted quality parameters, and acceptance criteria of cryopreserved irradiated accessory cells lots will be set, including unproliferation evaluation, cells viability, EBV inactivation . . . etc. Irradiated accessory cells exact addition procedure will be optimized for the final clones used in the process including accessory cells. Preferably, a dynamic culture closed system in bioreactors will be used for amplification/activation step with at least 5, preferably 10 pooled UCB units, such as the Wave System (GE Healthcare) already tested for NK culture. Preferably, culture medium used for the amplification/activation step, using animal serum-free media such as X-VIVO media from Lonza, CellGro SCGM from Cellgenix or AIM V from Invitrogen (already tested for NK cultures) can be used. Preferably, CD56 positive selection of amplified/activated NK cells using an adapted clinically upgradable material such as CliniMACS, will be used.
2. Pharmaceutical Development: Final Product Characterization and Acceptance Criteria

(87) Preferably, a step of acceptance criteria of final amplified/activated products must will be included in the process, including product identification steps (genetic stability, chimerism, phenotype) and a standard potency evaluation procedure. Preferably, the genetic stability of NK cells before and after the process of the present invention will be checked, looking at their karyotype (for example by G-banded karyotyping or cytoscanHD microarray methods well-known by the filled person), and the chimerism of the final pooled NK cells from the different donors must be defined (for example by standard multiplex PCR STR methods). Preferably and to better identify and characterize the final product and to define acceptance criteria, the expression of more NK phenotypical markers (NKG2D, NKG2C, CD94, NKp44, NKp30, NKp46, CD158 . . . ) will be evaluated (for example by flow cytometry). Preferably, each product lot will be tested with a validated cytotoxicity assay against commonly used well-known target cells Preferably, the absence of contaminations such as bacteria, fungi, mycoplasma and viruses (particularly EBV) must be verified during or after the final step of the process, as the absence of endotoxins and cytokines used during the manufacturing process.

EXAMPLE 7: POOLED UCBS NK CELLS CYTOTOXICITY ASSOCIATED WITH ANTI-CD20 ANTIBODIES

Material and Methods

(88) UCBs:

(89) HLA A3/A11+, HLA Bw4+, HLA Cg1+et HLA Cg2; Activated in presence of PLH (HLA Cg1) and thus, no inhibition via their KIR2DL2 et KTR2DL3 Alloreactivity of the NK HLA Cg1+KIR2DL2/KIR2DL3+ against the tumoral cells (HLA Cg1)
UCBs Treatment:

(90) Production of a population of expanded and activated NK cells from 5 UCB units: From the 5 UCB units exhibiting preferably the same pattern for major HLA class I groups genotype Erythrocytes-depleting each UCB unit by density gradient separation, by Ficoll-Paque density gradient separation; Optionally (see second experiment), the population of cells obtained is frozen, kept in liquid nitrogen and thawed before their use The 5 Ficoll/depleted nUCB units cells obtained in the preceding step are pooled.

(91) The pooled NK cells obtained are thus expanding and activating by contacting the NK cells contained in the pool, in a suitable medium to produce said population of pooled expanded and activated NK cells as described before, for an expanding/activation step(s) total duration comprised between 9 and 28 days Irradiated PLH at D0, D5, D8, D12; Culture medium: RPMI FBS 10% IL-2 100 IU/ml I-15 5 ng/ml; Amplification factor at D21: 256.6; 99.5% of NK cells, 100% CD69+, 84.6% CD16+(directly used for cytotoxicity test).

(92) After freezing/Thawing, we obtained 44.9% of CD16+. Patient p45: lymphoma B tumor sample (97% tumoral cells); Patient p46: B-cell chronic lymphocytic leukemia (B-CLL) (91% tumoral cells)

(93) For these two cases, the patient cells sample are thawing and, 1-2 h after thawing, are used in cytotoxicity assay

(94) 50,000 target cells/microwell, 200 l in final (96-well microplate conical bottom), RPMI medium FBS (Foetal bovine serum) 10%, IL-2 100 IU/ml, +/rituximab 10 g/ml, incubation at 37 C., 5% CO.sub.2.

First Experiment (See FIGS. 10A-10B)

(95) Fresh (not freezed-thawed) pooled UCBs NK cells are used (21 days after their production); Overnight (18 h) incubation Ratio NK cells/tumoral cells as indicated in the FIGS. 6A-6B.

Results

(96) No cytotoxicity is observed on non NK (CD56) and non B (CD19) cells even when the ratio used is the strongest one;

(97) In presence of rituximab, the pooled UCBs NK cells mediate Antibody-Dependent Cell-mediated Cytotoxicity on cellules CD20+tumoral cells.

2.SUP.nd .Experiment (see FIGS. 11A-11B)

(98) Same Lot of pooled UCB units NK cells as the first experiment but freezed at D21 et thawed at the day of cytotoxicity test;

(99) Incubation 24 h with the ratio indicated in the FIGS. 11A-11B (we can note that the ratios at the end of the cytotoxicity test are lower than at the beginning (death of part of the NK cells after 24 h following the thawing (6 in place of 10 and 3.5-4 in place of 5).

Results

(100) In presence of rituximab, the pooled UCBs NK cells mediate Antibody-Dependent Cell-mediated Cytotoxicity on cellules CD20+tumoral cells.

EXAMPLE 8: ANTI CD20 TG20 ACT AS A SYNERGIC MANNER TO INCREASE ADCC INVOLVED BY NK CELLS

(101) A) Chimeric Anti-CD20 Antibody TG20

(102) In order to increase NK cells infusion treatment efficiency, the therapy will be combined with the use of anti CD20 monoclonal antibody: TG20 (LFB S.A., Les Ulis, France, TG Therapeutics Inc., New York, N.Y.) to decrease progression or cure hematological cancers such as non-Hodgkin B-cell lymphoma or B-Cell chronic lymphocytic leukemia.

(103) TG20 is also called TG-20, Ublituximab, LFB-R603, TG-1101 or TGTX-1101.

(104) For more details of the structure of the chimeric anti-CD20 antibody TG20/R603 (low fucose/afucosylated mAb), see the patent application WO2012/175874 (published on Dec. 27, 2012

(105) According to the National Cancer Institute drug dictionary, TG20 (Ublituximab biosimilar obtained from milk of transgenic mice according to the LFB US rPRO technology) is a chimeric recombinant IgG1 monoclonal antibody directed against human CD20 with potential antineoplastic activity. Ublituximab specifically binds to the B cell-specific cell surface antigen CD20, thereby potentially inducing a B cell-directed complement dependent cytotoxicity (CDC) and antibody-dependent cell-mediated cytotoxicity (ADCC) against CD20-expressing B cells, leading to B cell apoptosis. CD20 is a non-glycosylated cell surface phosphoprotein that is exclusively expressed on B cells during most stages of B cell development and is often overexpressed in B-cell malignancies. Ublituximab has a specific glycosylation profile, with a low fucose content, that may enhance its ADCC response against malignant B cells (obtained from Rac cell lines Y2B/0 (ATCC #CRL-1662, see WO2012/175874, page 30, Example 2).

(106) NK cells are able to interact with cytotoxic monoclonal antibodies and lyse cells which are recognized by these monoclonal antibodies. This property is well known as antibody dependent cellular cytotoxicity (ADCC).

(107) This example demonstrates in vitro on clinical samples from patients that anti CD20 TG20 act as a synergic manner to increase ADCC involved by expansed and activated NK cells from pooled umbilical cord blood.

(108) NK cells are produced regarding good manufacturing process (GMP) and characterized in order to have 99% of purity at day 21.

(109) ADCC assays were performed on patient samples and also on 5 cell lines.

(110) B) Antibody Dependent Cellular Cytotoxicity Assay

(111) a) On Patient Samples

(112) Patient samples are coming from CHU, Montpellier, France.

(113) Ratio E:T Effector (NK cells): Target (=patient cells) are 5:1, 15:1

(114) As 50 000 target cells for 250 000 or 750 000 NK cells

(115) Monoclonal Antibody (mAb) concentration: 10 g/ml

(116) TG20 is compared to gold standard Rituximab (Genentech, Roche)

(117) TG20 and Rituximab are pre-incubated 15 min at room temperature.

(118) After NK addition, incubate 24 h at 37 C. before flow cytometry analysis

(119) Results: (see FIGS. 12A, 12B and 12C)

(120) At this time, we have tested one B-lymphoma (p45) and two B-CLL (B-cell Chronic Lymphocytic Leukemia) (p46, p53) patients.

(121) Variation of E:T ratio observed between assays is technically experience-dependent.

(122) Each sample of patient respond as an independent manner to the NK treatment alone.

(123) We observed for low E:T ratio 49%, 23% or 37% of lysis induced by NK cells alone for respectively p45, p46 and p53 patients. With a higher E:T ratio lysis induced by NK cells alone increased as 58% for p45 and 39% for p46.

(124) MAb (TG20 or Rituximab) treatment alone is poorly efficient.

(125) Association of NK cells and mAb increase dramatically percentage of target lysis in all cases. Compare to treatment of NK cells alone, we observe an increase of about 40% with the combination of NK cells+mAb in patient p45, an increase of 145% for low E:T ratio and 50% for high E:T ratio in patient p46 and an increase of 29% for NK+TG20 10 g/ml and 18% for NK+Rituximab 10 g/ml in patient p53.

(126) For instance we do not observe any significant difference between efficiency of TG20 and Rituximab. Moreover each response to combination of NK+mAb is patient dependent.

(127) b) On Cell Lines

(128) Cell LINES:

(129) Daudi (Burkitt lymphoma, high CD20), Raji (Burkitt lymphoma, low CD20), SUDHL4 (DL-BCL GC (Diffuse Large B Cell Lymphoma Germinal Center)), R1-1 (DL-BCL ABC (Activated B-Cell)) and HL60 (AML (Acute Myeloid Leukaemia)).

(130) Ratio E:T Effector (NK cells): Target (=cell line cells) are 1:1, 2:1 and 3:1

(131) As 50 000 target cells for 50 000, 100 000 or 150 000 NK cells

(132) Monoclonal Antibody (mAb) concentration: 0.1; 1; 5 et 10 g/ml

(133) TG20 is compared to gold standard Rituximab (Genentech, Roche)

(134) TG20 and Rituximab are pre-incubated 15 min at room temperature.

(135) After NK addition, incubate 4 h at 37 C. before flow cytometry analysis Results: ADCC and synergic action of mAb and NK cells on cell lines (see FIGS. 13A, 13B and 13C) and anti-CD20 dose effect (ratio E:T: 1/1) ((see FIGS. 14A, 14B, 13C and 14C)

(136) 1:1 E:T ratio is more appropriate to study ADCC and synergic action of mAb and NK cells. To evaluate dose effect of mAb we should start at 5 g/ml of mAb as a minimal dose.

(137) On AML HL60 cell line we do not observe a real synergic action of mAb and NK which is a normal result as HL60 cells do not exhibit CD20 on surface (data not shown).

(138) At this time, TG20 seems to be as efficient as Rituximab. Furthermore, we observe on SUDHL4 cell line better results of synergic action between TG20-NK cells than Rituximab-NK cells.

(139) Binding of TG20 and Rituximab were also tested on Daudi and Raji cell lines and no significate results were observed (data not shown). On Daudi cells: TG20 EC50=75.6 ng/ml; Rituximab EC50=105 ng/ml. On Raji cells: TG20 EC50=229 ng/ml; Rituximab EC50=230 ng/ml. (EC50 is concentration of mAb to obtain 50% of binding).

(140) Inhibition of CD20 after mAb binding was also analyzed on Daudi and Raji cells (data not shown). On Daudi cells we observed: TG20 EC50=4 g/ml; Rituximab EC50=3.4 g/ml. On Raji cells we observed: TG20 EC50=5 g/ml; Rituximab EC50=3.7 g/ml. The differences observed for both mAb on each cell line are not significant. EC50 is concentration of mAb to obtain 50% of CD20 inhibition.

(141) Finally, we have performed complement dependent cytotoxicity on the 5 cell lines and we did not observe any difference between lysis induced by complement with TG20 or Rituximab (data not shown).

EXAMPLE 9: ANTI CD20 RITUXIMAB AND UBLITUXIMAB R603 ACT AS A SYNERGIC MANNER TO INCREASE ADCC INVOLVED BY NK CELLS

A) Experimental Study

(142) The aim of this example is to demonstrate on clinical samples of blood of patients with non-Hodgkin B lymphoma that anti CD20 mAb as Ublituximab R603 potentiate the ADCC induced by NK001 demonstrating a synergetic action between NK001 and mAb.

(143) Samples of patients come from Hemodiag collection, CHU Montpellier.

(144) Cell Line:

(145) Daudi: the Daudi line was derived from a 16-year-old male with Burkitt's lymphoma, Daudi expresses high level of CD20.

(146) Clinical Samples:

(147) Non-Hodgkin B lymphoma: Mantle cell lymphoma (p44 et p15-30) Marginal zone B cell lymphoma (p16-01 et p16-48) Diffuse large B-cells lymphoma (p16-108)
Anti CD20 mAbs UblituxiMAb R603 (afucosylated, TGTX) RituxiMAb (Roche Genentech, as a standard control)
NK001

(148) 5 different batches of NK001 (from NK001-2 to NK001-6) were produced, amplified and activated for this study.

(149) Experimental Design:

(150) 1On Daudi cell line (incubation time is 4 h): Rituximab cytotoxicity: dose effect to choose the best concentration; ADCC with Rituximab+NK001 dose effect (1a) to determine the best E:T ratio and ADCC with NK001+Rituximab dose effect (1b) to confirm Rituximab concentration and gain of lysis as a synergy between NK001+Rituximab and not as an additive effect.

(151) 2On clinical samples: mAb anti CD20 cytotoxicity (24 h at 10 g/ml)

(152) 3ADCC with an E:T (effector: target) ratio of 5:1 using 10 g/ml of anti CD20 mAb, pre-incubated 15 min at room temperature and placed at 37 C. for 24 h.

(153) 4ADCC with 5 different batches of NK001 production, example of 3 clinical samples, to demonstrate reproducibility of results.

Results

(154) On Daudi Cell Line

(155) Characterization of anti-CD20 mAb in our in vitro condition

(156) 1a) Rituximab Cytotoxicity: Dose Effect and ADCC with Rituximab+NK001 Dose Effect

(157) See FIGS. 15A and 15B and Table 6

(158) TABLE-US-00006 TABLE 6 % gain of lysis obtained between the natural cytotoxicity of NK001 and the lysis induced by combination NK001 + anti CD20 mAb. Daudi % gain of lysis Ratio E:T NK001 + Rituximab 0.5:1.sup. 30.1 1:1 21.3 2:1 11.9 3:1 1.4

(159) These experiments allow to determine the best concentration of anti CD20 mAb: 10 g/ml (see FIG. 15 B)

(160) Concerning ADCC assays on Daudi cells we conclude that ratio Effector (NK): Target (Daudi) 0.5:1 and 1:1 are the most appropriate to see a gain of target lysis using combination of NK001+Rituximab. For these 2 ratio, % of gain of targets' lysis are superior to 2

(161) 1b) ADCC with NK001+Rituximab Dose Effect

(162) See FIG. 16 and Table 7
% gain=(lysis NK001+mAblysis NK001)100/lysis NK001

(163) TABLE-US-00007 TABLE 7 ADCC with NK001 + Rituximab dose effect on Daudi cell line Daudi cell line % gain [mAb] g/ml NK001 + Rituximab 10 29.4 5 13.1 1 7.7 0.1 12.2

(164) The statistical test used is a student test (t-test) with p-value compared to NK001 alone.

(165) p-value<0.05: *; p-value<0.01: **; p-value<0.001: ***

(166) This result comforts us to use an anti CD20 mAb concentration of 10 g/ml to observe the more relevant results.

(167) 2mAb Anti CD20 Cytotoxicity on Clinical Sample

(168) See FIG. 17

(169) We have assayed 5 clinical samples. We do not have a high level of cytotoxicity of anti CD20 mAbs in our in vitro condition assay. The maximum level of lysis observed is 4% (R603). Moreover no significant difference is observed between R603 and Rituximab.

(170) 3ADCC Induced by NK-001 in Synergy with Anti CD20 mAb

(171) See FIG. 18 and Table 8

(172) TABLE-US-00008 TABLE 8 % gain of lysis obtained between the natural cytotoxicity of NK001 and the lysis induced by combination NK001 + anti CD20 mAb on 5 clinical sample % Gain Samples NK001 + R603 NK001 + Rituximab P16-48 94.7 54.4 P16-108 98.5 100.8 P15-30 94.5 75.2 P44 86.0 60.2 P16-01 45.2 45.0

(173) To demonstrate synergy between NK001 and anti CD20 mAbs we have performed ADCC on 5 clinical samples. For all presented results E:T ratio is 5:1.

(174) In our in vitro conditions assays, although we do not observe cytotoxicity of anti CD20 mAbs alone, we demonstrate an increase of specific lysis of target cells when NK001 are combined with anti CD20 mAbs, as mentioned in the table. The smaller % of gain of lysis observed is 45% and the maximum is 100%.

(175) Regarding these results we can suggest that for all the clinical samples we have a clear synergy of action between NK001 and anti CD20 mAbs.

(176) 4ADCC with 5 Lots of NK001: Reproducibility of Results

(177) See FIGS. 19A-19 C and (voir FIG. 5A-5C (Doc 6/01 de 8 pages sur Rituxi/R603)

(178) We performed ADCC with 5 batches of NK001 production on clinical samples. We present results obtained for 3 clinical samples.

(179) We confirm having a reproducibility of the results using different lots of NK001 and a similar efficacy between R603 and Rituximab.

(180) The higher difference observed is for the clinical sample p44 between batches NK001-2 and NK001-4, with 25.5% of difference.

(181) For p15-30, the mean difference between NK001-n+Rituximab is 10.2% and results obtained have a dispersion (SD) of 8.54. For p15-30, the mean difference between NK001-n+R603 is 7.9% and results obtained have a dispersion (SD) of 6.4.

(182) For p44, the mean difference between NK001-n+Rituximab is 8.9% and results obtained have a dispersion (SD) of 7. For p15-30, the mean difference between NK001-n+R603 is 13.4% and results obtained have a dispersion (SD) of 10.5.

(183) For p16-01, the mean difference between NK001-n+Rituximab is 6.1% and results obtained have a dispersion (SD) of 5. For p16-01, the mean difference between NK001-n+R603 is 8.1% and results obtained have a dispersion (SD) of 6.8

Conclusion

(184) On Daudi cell line (high CD20), 0.5:1 and 1:1 E:T ratio appear to be the most appropriate for conducting ADCC assays and for getting a difference of target lysis between treatment with NK001 alone vs NK001 combined with monoclonal antibodies.

(185) In our in vitro conditions, to observe a positive dose effect of anti CD20 mAbs in ADCC, the minimal effective concentration is 10 g/ml.

(186) On clinical samples, association of NK001 with anti CD20 mAbs increases lysis of tumor cells compare to results obtain with NK001 cytotoxicity. Indeed, we calculated the % of gain of targets' lysis and obtained a minimum of 45% and a maximum of 100% of gain of lysis.

(187) We do not observe any significant difference of efficacy between Ublituximab R603 and Rituximab.

(188) So we can conclude that we have demonstrated a synergy of action between NK001 and anti CD20 mAbs.

(189) Then we performed ADCC with 5 different batches of NK001 on clinical samples and observed a reproducibility of the results using different lots of NK001 and similar efficacy between UblituxiMAb R603 and Rituximab. The dispersion (Standard Deviation) of the % of lysis obtained with 5 lots of NK001 (+mAbs) is comprised between 5 and 10 which easily acceptable.

(190) In our in vitro assay conditions we do not observe an important lysis of tumor cells induced by treatment of anti CD20 mAb alone.

(191) Response to treatment depend on patients but are independent of the type of NH B lymphoma.

(192) In the context of NK001 project, we set up a methodology to study the efficacy of NK001+mAb anti CD20 combination therapy on both cell cultures and on biological samples of patients (hematological cancers).

(193) This methodology and its technology can be transposed with other types of monoclonal antibodies, either with new targets for hematological or solid cancer.

EXAMPLE 10: ANTI CD20 RITUXIMAB AND TG20: BINDING, CYTOTOXICITY AND INDUCED ADCC

(194) Experimental Study

(195) Samples of patients come from Hemodiag collection, CHU Montpellier.

(196) Cell Lines

(197) Daudi: the Daudi line was derived from a 16-year-old male with Burkitt's lymphoma, Daudi expresses high level of CD20. Raji: the Raji line of lymphoblast-like cells was established from a Burkitt's lymphoma of an 11-year-old male. Raji expresses a low level of CD20. Ri-1: The cell line Ri-1 has been established from the peripheral blood of a 48-year-old female patient with B-cell lymphoma. SUDHL4: The cell line SUDHL4 has been established from the peritoneal effusion of a 38-year-old man with B-NHL (diffuse large cell, cleaved cell type). HL60: HL-60 is a promyelocytic cell line. Peripheral blood leukocytes were obtained by leukopheresis from a 36-year-old Caucasian female with acute promyelocytic leukemia. HL60 does not express CD20 (negative control)
Clinical Samples: Non-Hodgkin B lymphoma: Diffuse large B-cells lymphoma (p16-108 et p45) Mantle cell lymphoma (p44 et p15-30) Marginal zone B cell lymphoma (p16-01 et p16-48) Chronic lymphocytic leukemia: BCLL (p46 et p53)
Anti CD20 mAbs TG20 (afucosylated, LFB) RituxiMAb (Roche Genentech, as a standard control)
NK001 6 different batches of NK001 (from NK001-1 to NK001-6) were produced, amplified and activated for this study.
Experimental Design:

(198) 1Anti CD20 mAb binding on NK-001 (1a) and Daudi, Raji cell lines: direct binding (1b) with a human Ig antibody and analyze of CD20 inhibition (1c).

(199) 2Quantification of apoptotic cells (Daudi, Raji) after 24 h incubation of anti CD20 mAb.

(200) 3Study of Complement Dependent Cytotoxicity (CDC) after 24 h incubation of anti CD20 mAb+20% of human serum on Daudi, Ri-1 and SUDHL4 cell lines.

(201) 4On Daudi cell line: mAb anti CD20 cytotoxicity: dose effect (4a) to choose the best concentration and ADCC with anti CD20 mAb+NK001 dose effect (4b) to determine the best E:T ratio.

(202) 5On Daudi cell line, ADCC with NK001+mAb anti CD20 dose effect to prove synergy of action or additive effect.

(203) 6On cell lines (Daudi, Raji, Ri-1, SUDHL4 and HL60 as a negative control): mAb anti CD20 cytotoxicity (6a) using 10 g/ml and ADCC (6b) with an E:T ratio of 1:1 using 10 g/ml of anti CD20 mAb, pre-incubated 15 min at room temperature and placed at 37 C. for 4 h. This experiment is done to determine the background and the acceptable % of gain of lysis for the rest of the study.

(204) 7mAb anti CD20 cytotoxicity (24 h at 10 g/ml) on clinical samples.

(205) 8ADCC on clinical samples with an E:T ratio of 5:1 using 10 g/ml of anti CD20 mAb, pre-incubated 15 min at room temperature and placed at 37 C. for 24 h.

(206) Results

(207) NK001

(208) 1aAnti CD20 mAb Binding on NK001 (See FIG. 20))

(209) As TG20 is an afucosylated mAb, we observed a better binding to the CD16 of NK001 than Rituximab. This result was expected.

(210) Cell Lines: Characterization of Anti CD20 mAb

(211) 1bDirect Binding Analyses (hIg) (See FIGS. 21A and 21B)

(212) We do not observe any difference of binding on Daudi (high CD20) or on Raji (low CD20) between TG20 and Rituximab.

(213) For Daudi, EC50 [TG20]=75.6 ng/ml and EC50 [Rituximab]=105 ng/ml.

(214) And for Raji, EC50 [TG20]=229 ng/ml and EC50 [Rituximab]=230 ng/ml.

(215) 1cBinding Analyses by Inhibition of CD20 Labelling (See FIGS. 22A and 22B)

(216) We do not observe any difference of CD20 inhibition on Daudi (high CD20) or on Raji (low CD20) cells between TG20 and Rituximab.

(217) For Daudi, EC50 [TG20]=4 g/ml and EC50 [Rituximab]=3.4 g/ml.

(218) And for Raji, EC50 [TG20]=5 g/ml and EC50 [Rituximab]=3.7 g/ml.

(219) These results are consistent with the literature.

(220) 2Quantification of Apoptotic Cells (See FIGS. 23A and 23B)

(221) Apoptosis of cell lines slightly increases in proportion to the dose of anti CD20 mAb after 24 h of incubation.

(222) 3Complement Dependent Cytotoxicity (CDC) (See FIGS. 24A, 24B and 24C)

(223) Complement dependent cytotoxicity is more important when concentration of anti CD20 mAb increases.

(224) We still do not observe any significant difference between TG20 and Rituximab.

(225) 4aCytotoxicity of Anti CD20 mAb: Dose Effect (See FIG. 25 and Table 9)

(226) TABLE-US-00009 TABLE 9 Daudi/[mAb] % Lysis g/ml NK001 + TG20 NK0021 = Rituximab 0.1 9.6 7.1 1 10.1 6.7 5 12.6 12.2 10 18 17

(227) This experiment permits to determine the best concentration of anti CD20 mAb: 10 g/ml

(228) 4bADCC Induced by NK-001+Anti CD20 mAb [10 g/Ml]: NK001 Dose Effect (See FIG. 26 and Table 10)

(229) TABLE-US-00010 TABLE 10 Daudi/[mAb] % Gain g/ml NK001 + TG20 NK001 + Rituximab 10 30.0 29.4 5 13.1 13.1 1 5.4 7.7 0.1 10.4 12.2

(230) The statistical test used is a student test (t-test) with p-value compared to NK001 alone.

(231) p-value<0.05: *; p-value<0.01: **; p-value<0.001: ***

(232) Concerning ADCC assays on cell lines we conclude that ratio Effector (NK): Target (cell line) 0.5:1 and 1:1 are the most appropriate to see a gain of target lysis using combination of NK001+mAbs. For these 2 ratio, % of gain of targets' lysis are superior to 20%.

(233) In conclusion, for the rest of the study, we are using an E:T ratio of 1:1 and a concentration of anti CD20 mAb=10 g/ml.

(234) 5ADCC Induced by NK-001+Anti CD20 mAb: mAb Dose Effect (See FIG. 27 and Table 11

(235) TABLE-US-00011 TABLE 11 Daudi/ % Gain Ratio E:T NK001 + TG20 NK0021 = Rituximab 0.5:1.sup. 30.6 30.1 1:1 23.5 21.3 2:1 12.2 11.9 3:1 5.0 1.4

(236) This experiment allows to determine if the % of gain of lysis is due to a cumulative effect or a synergetic action between NK001+anti CD20 mAbs.

(237) The statistical test used is a student test (t-test) with p-value compared to NK001 alone.

(238) p-value<0.05: *; p-value<0.01: **; p-value<0.001: ***

(239) This result comforts us to use an anti CD20 mAb concentration of 10 g/ml to observe the more relevant results.

(240) 6aCytotoxicity of Anti CD20 mAb: Cell Lines (See FIG. 28 and Table 12)

(241) TABLE-US-00012 TABLE 12 % Lysis Cell lines TG20 Rituximab Daudi 16 14.9 Raji 12.2 9.7 Ri-1 12 8.8 SUDHL4 12.3 12.9 HL60 2.5 2

(242) In CD20 positive cell lines we only observe a weak cytotoxicity of CD20 mAb for 24 h.

(243) The maximum of lyse obtained was 16% (TG20) and the minimum was 8.8% (Rituximab).

(244) HL60 is a negative control and correspond to the background.

(245) We do not observe significant difference between TG20 and Rituximab.

(246) 6bADCC Induced by NK-001 in Synergy with Anti CD20 mAb (See FIG. 29 and Table 13)

(247) TABLE-US-00013 TABLE 13 % Lysis Cell lines NK001 + TG20 NK001 + Rituximab Daudi 26.0 25.4 Raji 45.3 41.1 Ri-1 64.8 51.7 SUDHL4 30.2 26.5 HL60 20.1 10.9

(248) The statistical test used is a student test (t-test) with p-value compared to NK001 alone.

(249) p-value<0.05: *; p-value<0.01: **; p-value<0.001: ***

(250) Table of % of gain of targets' lysis of ADCC results (NK001+mAb) compare to natural NK001 cytotoxicity:

(251) NB: formula of % gain explain end of the Example.

(252) Regarding statistical analysis and negative control HL60, we can suggest for the following study that if the % of gain when NK001 is combined with mAb is less than 20%, there is no synergetic action of NK001 and mAb.

(253) Clinical Samples

(254) 7Anti CD20 mAbs Cytotoxicity (See FIG. 30)

(255) We have assayed 8 clinical samples (detailed in III-Report content). As observed on cell lines, we do not have a high level of cytotoxicity of anti CD20 mAbs in our in vitro condition assay. The maximum level of lysis observed is 11.2% (Rituximab) and the minimum is 1.7% (TG20). Moreover no significant difference is observed between TG20 and Rituximab.

(256) 8ADCC Induced by NK-001+Anti CD20 mAbs (See FIG. 31 and Table 14)

(257) To demonstrate synergy between NK001 and anti CD20 mAbs we have performed ADCC on 8 clinical samples. For all presented results E:T ratio is 5:1.

(258) TABLE-US-00014 TABLE 14 % Lysis Patients NK001 + TG20 NK001 + Rituximab p46 145.9 149.4 p16-18 94.9 99.2 p53 60.7 45.7 p15-30 124.5 136.1 p44 118.1 141.2 p45 40.7 55.9 p16-108 24.5 38.1 p16-108 19.1 25.8

(259) For clinical samples we need to use a higher E:T ratio (5:1) than which one is used for cell lines (0.5:1 or 1:1) and an incubation of 24 h to observe a relative high cytotoxicity of NK001 alone. Other E:T ratio were tested, data not shown.

(260) In our in vitro conditions assays, although we do not observe cytotoxicity of anti CD20 mAbs alone, we demonstrate an increase of specific lysis of target cells when NK001 are combined with anti CD20 mAbs, as mentioned in the table. The inferior limit is 20% and the maximum of gain of targets' lysis observed is 150%.

(261) Regarding the characterization done on cell lines we can suggest that for all the clinical samples we have a clear synergy of action between NK001 and anti CD20 mAbs.

(262) We performed ADCC with 6 batches of NK001 production on clinical samples. We present results obtained for 3 clinical samples.

(263) As we fixed a limit of 20% of acceptable difference with the in vitro study made on cell lines, we can suggest that we have a reproducibility of the results using different lots of NK001 and similar efficacy between TG20 and Rituximab.

Conclusion

(264) On NK001, TG20 is higher than binding of Rituximab. TG20 is an afucosylated mAb so its fixation to CD16 receptor is more effective.

(265) The binding of TG20 and Rituximab was tested on Daudi (high CD20) and Raji (low CD20) lines. On Daudi, the EC50 for TG20 is 75.6 ng/ml and the EC50 for Rituximab is 105 ng/ml. On Raji, the EC50 of the TG20 is 229 ng/ml and the EC50 of the Rituximab is 230 ng/ml. The small differences observed are not significant. The EC50 is to reach 50% of mAb fixation on cell surface.

(266) The inhibition of CD20 was also tested on Daudi (high CD20) and Raji (low CD20) lines.

(267) On Daudi, the EC50 for TG20 is 4 g/ml and the EC50 for Rituximab is 3.4 g/ml.

(268) On Raji, the EC50 of TG20 is 5 g/ml and the EC50 of Rituximab is 3.7 g/ml.

(269) The EC50 is to achieve 50% inhibition of CD20 labelling on cell surface.

(270) The results are consistent with the literature and the small differences observed are not significant.

(271) Apoptosis of cell lines slightly increases in proportion to the dose of anti CD20 mAb after 24 h of incubation.

(272) Also, complement dependent cytotoxicity is more important when concentration of anti CD20 mAb increases.

(273) We do not observe any significant difference between TG20 and Rituximab.

(274) On Daudi cell line, we first analyzed the lysis induced by dose effect of anti-CD20 mAb and chose 10 g/ml as a work concentration. Then we performed a dose escalation of E:T ratio for ADCC assay and 0.5:1 and 1:1 E:T ratio appear to be the most appropriate for conducting ADCC assays and for getting a difference of target lysis between treatment with NK001 alone vs NK001 combined with monoclonal antibodies. Based on these results and statistical analysis we suggest that we have a synergetic action between NK001+ and anti CD20 mAb that we also assayed on other CD20 positive cell lines.

(275) In our in vitro conditions, to observe a positive dose effect of anti CD20 mAbs in ADCC, the minimal effective concentration is 10 g/ml.

(276) On HL60 cell line (CD20 neg) no significant synergetic action of anti-CD20 combined with NK001 is observed, which is normal as a negative control of ADCC. This result permits to fix a limit of 20% of difference between % of gain of target lysis obtained with NK001 alone and NK001+anti CD20 mAbs.

(277) Moreover in all assays performed, TG20 seems to be as effective as the standard Rituximab control.

(278) On clinical samples, association of NK001 with anti CD20 mAbs increases lysis of tumor cells compare to results obtain with NK001 alone, as observed on cell lines. But, we need to use a higher E:T ratio (5:1) than which one is used for cell lines (0.5:1 or 1:1) and an incubation of 24 h to observe a relative high cytotoxicity of NK001 alone.

(279) We calculated the % of gain of targets' lysis and obtained an inferior limit of 20% (determine as acceptable by study on cell lines) and a maximum of 150% of gain of lysis.

(280) So we can conclude that we have demonstrated a synergy of action between NK001 and anti CD20 mAbs.

(281) Then we performed ADCC with 6 different batches of NK001 on clinical samples and observed a reproducibility of the results using different lots of NK001 and similar efficacy between TG20 and Rituximab.

(282) Response to treatment depend on patients but are independent of the type of NH B lymphoma.

(283) With our in vitro assay conditions we do not observe an important lyse of tumor cells induced by treatment of anti CD20 mAb alone.

(284) These results show the efficacy of NK001+mAb anti CD20 combination therapy on both cell cultures and on biological samples of patients (hematological cancers).

Abbreviation

(285) ADCC: antibody dependent cellular cytotoxicity CDC: complement dependent cytotoxicity EC50: Half maximal effective concentration E:T: Effector (=NK001): Target (cell lines or tumor cells) IV: intra venous mAb: monoclonal antibody NSG: NOD scid gamma NH: Non Hodgkin NK: natural killer cell Statistical analyses are basic Student t-test.
% gain=(lysis NK001+mAblysis NK001)100 lysis NK001

EXAMPLE 11: ANTI HER 2 ANTIBODIES ACT AS A SYNERGIC MANNER TO INCREASE ADCC INVOLVED BY NK CELLS

(286) It is demonstrated here that the efficacy of cell therapeutic treatment for solid cancer expressing Her2/neu receptor as some breast or ovarian cancer can be increased by association of NK001 cells with anti Her2/neu monoclonal antibodies (mAb) which act in synergic action.

(287) Indeed, NK cells are able to interact with cytotoxic mAb and kill cells recognized by these mAbs. This own property of NK cells is known as Antibody Dependent Cellular Cytotoxicity (ADCC).

Material and Method

(288) Experimental StudyAnti Her2 mAb Approach

(289) One of the main characteristic of NK001 cells production process is getting a high percentage of CD16 receptor (FcRIII) on NK001 which is in favor of a high level of cytotoxicity and could be potentiated with mAb association (ADCC).

(290) Demonstration, in vitro on tumor cell lines expressing Her2/neu receptor that anti Her2/neu mAb as Trastuzumab (Roche, Genentech) or Herceptin-like (LFB US) potentiate the ADCC induced by NK001.

(291) Cell Lines

(292) SKBR3: The SKBR3 line was derived from metastatic pleural effusion of a 43 years old female with a breast adenocarcinoma. SKBR3 are epithelial and adherent cells. SKOV3: The SKOV3 line was derived from the ovary tissue (ascites of an adenocarcinoma) of a 64 years old female. SKOV3 are epithelial and adherent cells.
Anti Her2 mAbs
Herceptin Like (LFB US)

(293) Herceptin-like, is a biosimilar antibody of the Trastuzumab (Herceptin) which is produced in the milk of transgenic goats, It has been obtained using the rPRO platform, a disruptive innovation, which uses genetic recombination to express a protein in mammal milk, particularly goat milk, that can be used in human medicine. The protein is then isolated, purified and made safe using the highest pharmaceutical standards. According to LFB US, rPRO is a robust, productive technology, with an excellent economic profile, largely because it increases production greatly while keeping costs down.

(294) For the process of preparation of monoclonal antibody (Mab) from milk of transgenic goats se for example U.S. Pat. No. 8,173,860, Meade et al. May 8, 2012,

(295) GTC Biotherapeutics) and for the humanized trastuzumab MAb sequence, the U.S. Pat. No. 5,821,337 (Carter et al., Oct. 13, 1998, Roche/Genentech).

(296) Trastuzumab (Roche Genentech)

(297) NK001 (Emercell)

(298) 2 different batches of NK001 (NK001-1 and NK001-2) were produced, amplified and activated for this study.

(299) Experimental Design:

(300) 1Anti Her2 mAbs binding on NK-001: direct binding with a human IgG antibody.

(301) 2Anti Her2 mAbs cytotoxicity on SKOV3 and SKBR3, 3 h, 10 g/ml.

(302) 3ADCC with 10 g/ml of anti Her2 mAbs+NK001 with E:T ratio as 5:1, 3 h.

(303) Analysis by flow cytometry and microscopy.

(304) Results

(305) NK001

(306) 1Anti Her2 mAbs Binding on NK001 (See FIG. 33)

(307) Herceptin-like and Trastuzumab binding to the CD16 of NK001 are observed, with a better binding for Herceptin-like than Trastuzumab.

(308) 2Anti-Her2 mAbs Cytotoxicity (See FIG. 34)

(309) This experiment permits to determine the lysis induced by 10 g/ml of anti Her2 mAbs for 3 h in order to compare to ADCC. In these in vitro conditions we observed a weak cytotoxicity of anti Her2 mAbs on SKBR3 and SKOV3. SKOV3 seems to be more sensitive to anti Her2 mAb treatment.

(310) 3ADCC Induced by NK-001+Anti Her2 mAb [10 g/Ml]; E:T Ratio is 5:1, 3 h

(311) (See FIGS. 35A, 35B)

(312) TABLE-US-00015 TABLE 15 % gain of lysis % Gain of SKBR3 SKOV3 Lysis Mean SD Mean SD NK001 + Trastuzumab 134.5 25.5 19.2 12.5 NK001 + Herceptin-like 121.1 28.3 88.4 10.3

(313) On SKBR3 cell line, an important synergetic action between NK001 and anti-Her2 mAbs was observed, as a result of a gain of lysis of 134% for NK001+Trastuzumab and 121% for NK001+Herceptin-like. No difference was observed between both anti Her2 mAbs.

(314) On SKOV3 cell line, an important synergetic action between NK001 and anti-Her2 mAbs was observed (See FIGS. 36 and 37)

(315) Herceptin-like appears to be more efficient than Trastuzumab on SKOV3 cell lines.

(316) Effector (NK001): Target (cells) ratio (E:T ratio) of 5:1 is acceptable for a cytotoxicity of 3 h at 37 C.

Conclusion

(317) It can be concluded that a synergy of action has been demonstrated between NK001 and anti Her2 mAbs in 2 models of adherent tumor cell lines for anti-Her2 mAbs.

(318) On NK001, binding of Herceptin like is higher than binding of Trastuzumab.

(319) Then, ADCC assay was performed on SKBR3 and SKOV3 cell line, using 10 g/ml of anti-Her2 mAbs for 3 h at 37 C. The anti-Her2 mAbs were pre-incubated at room temperature for 15 minutes.

(320) On SKBR3 no difference was observed between Trastuzumab and Herceptin like. Gain of lysis of NK001+anti Her2 mAbs compare to natural cytotoxicity of NK001 was higher for SKBR3 than for SKOV3. However it seems that association of NK001+Herceptin-like is more efficient for SKOV3 cell line.