Method for producing gamma delta t cells

Abstract

The present invention relates to a method for producing gamma delta (γδ) T cells. The present invention also relates to the cells obtained by said method and said cells for their use as a medicament and in an immunotherapy, preferably in the treatment or prevention of a cancer.

Description

[0001] METHOD FOR PRODUCING GAMMA DELTA T CELLS

[0002] The present invention relates to a method for producing gamma delta (y6) T cells. The present invention also relates to the cells obtained by this method and said cells for their use as a medicament, preferably in an immunotherapy, more preferably in the treatment or prevention of a cancer.

[0003] Cellular immunotherapy is part of the therapeutic arsenal that can be used to treat cancers and hematological malignancies such as leukemia. It consists of isolating immune cells with antitumor activity of autologous origin (from the patient) or allogeneic origin (from a donor) and reinjecting them into the patient either in native form or after genetic modification. The cellular immunotherapies most widely used currently include T lymphocytes (T cells) and Natural Killer (NK) cells. For example, injections of native T cells from donors are regularly performed following allogeneic hematopoietic stem cell (HSC) transplantation when the patient's disease relapses (Abbi et al., Bone Marrow transplant., 48:357-362, 2013). However, such injections are often accompanied by the risk of triggering a graft-versus-host (GvH) reaction. This complication results in an attack of the recipient's epithelial tissues by the donor's T cells and is mainly due to a subpopulation of T cells having a T cell receptor alpha beta (TCR a ). This complication is very frequent after an allogeneic HSC transplantation and can, in the most severe forms, be lifethreatening for the patient.

[0004] In recent years, T cells have also been used as a platform for the production of chimeric antigenic receptor-T cells (or CAR-T cells). These are cells genetically modified to express a specific receptor for an antigen expressed by tumors and they are particularly appropriate for the treatment malignant blood diseases. Because of this genetic modification, CAR-T cells have specificity against the tumor and a more targeted cytotoxic effect. Today, only autologous CAR-T cells from the patient are used because allogeneic CAR-T cells, like native T cells, are likely to trigger a GvH reaction. However, CAR-T cells are very toxic because they generate a cytokine release syndrome, the most severe grades of which can induce a cardiogenic shock or even respiratory distress. In addition, CAR-T cells are very expensive and time-consuming to produce.

[0005] It is thus necessary to develop efficient cell platforms that would be safer than conventional T cells. T cells with gamma delta TCR (y6 T cells or gd T cells) represent 5 to 10% of peripheral blood lymphocytes, of which 50 to 70% are composed of the subpopulation expressing y962 TCR. y6 T cells have the capacity to respond quickly after TCR engagement and to expand at the tumor site. Unlike aP T cells, y6 T cells recognize antigens in an M HC-independent manner and detect abnormal cells through their danger signal molecular profile. Therefore, these cells can be usedtherapeutically without triggering a GvH reaction. Their mechanism of action is based on their ability to destroy a target cell (tumor or infected) by triggering cell death via the release of perforin and granzyme and to recruit other immune cells via the release of interferon gamma (IFN y) and interleukin 17 (IL-17), depending on y6 T cell subtype. y6 T cells have been identified as the most favorable cells in terms of prognosis among the immune subpopulations infiltrated in 18,000 tumors, representing 39 tumor pathologies (Gentles et al., Nat. Med., 21:938-945, 2015).

[0006] Therapeutic uses of y6 T cells (TCR y962) were tested in some clinical trials on 559 patients with malignant blood diseases or solid tumors and 396 patients showed a measurable response (Feng & Zhou, Front. Immunol., 14, 2023). The different therapeutic strategies used were in vivo stimulation of y6 T cells and adoptive transfer of autologous or allogeneic y6 T cells, alone or combined with interleukin 2 (IL-2). Trials based on adoptive transfer showed good tolerance of y6 T cells with few adverse effects and an overall response rate of 35%.

[0007] In practice, in the cases of relapse of patients with malignant hematological diseases after allogeneic HSC transplantation, clinicians usually administer injections of fresh or thawed donor lymphocytes in order to enhance the graft versus leukemia effect (GvL). However, it must be taken into account that the administered product is not composed solely of pure T cells. Indeed, it is usually obtained either from leukapheresis or from a surplus of the HSC graft. In addition, among the T cells contained in the product, only a very small proportion are y6 T cells, the majority being ap T cells likely to trigger a GvH reaction. Thus, this limits the dose of T cells infused and therefore its effectiveness in the treatment of relapses.

[0008] In this context, there thus remains a genuine need for means allowing the production of a high amount of T cells enriched with the y6 phenotype.

[0009] The present invention is believed to meet such a need by providing a new method for expanding y6 T cells from mononuclear cells.

[0010] Unexpectedly, the Inventors discovered that sequential agitation during the first days of mononuclear cells culture allows the targeted amplification of y6T cells subpopulation. Indeed, the Inventors showed that this step favors an enrichment of the circulating population present in the blood from healthy donors (from 0.5 to 5% depending on the individual) to reach a cell population composed almost exclusively of y6 T cells (> 80%) and almost devoid of a T cells (< 20%). This ability to selectively expand the y6 T cells subpopulation is highly advantageous in that it provides a high amount of y6T cells, which then makes it possible to amplify y6T cells on a larger scale withininstrumented and controlled bioreactors. This method permits the production of a high quantity and quality of therapeutic products with good manufacturing practices (GMP).

[0011] In a main aspect, the present invention thus relates to a method for producing gamma delta (y6) T cells comprising:

[0012] - a step (i) wherein mononuclear cells are incubated in presence of at least an activating compound selected from the group consisting of a phosphoantigen and a bisphosphonate, and

[0013] - a step (ii) wherein the mononuclear cells are cultured in a liquid culture medium comprising at least interleukin 2 (IL-2),

[0014] wherein step (ii) begins with a static phase and alternates static and agitated phases during at least 7 days.

[0015] As used herein, the term "T cell" (or "T lymphocyte") refers to an immune cell expressing CD3 (CD3+) and a T cell receptor (TCR+). T cells are important players in the adaptive arm (ap subtype) and innate (y6 subtype) of the immune system. They respond to specific antigens presented by a major histocompatibility complex (MHC) dependent (a subtype) or independent manner (y6 subtype).

[0016] As used herein, the term "y6 T cell" (or "gd T cell" or "gamma delta T cell") refers to a subset of T cells that express a distinct TCR, y6 TCR, on their surface, composed of one y-chain and one 6-chain. The predominant subset in the blood, expresses a V62 chain associated with a Vy9 chain and represent 50 to 70% of the circulating y6 T cells in human adults, while a minor subset (approximately 30%) expresses a V61 chain linked to a chain different from Vy9. In the present invention, the term "y6 T cells" includes, without limitation, all subsets of y6 T cells, in particular y962 T cells.

[0017] As used herein, the term "mononuclear cells" (or "peripheral blood mononuclear cells" or "PBMC") designates any blood cells having a round nucleus. These cells include lymphocytes (T cells, B cells, NK cells), monocytes and dendritic cells. While mononuclear cells are typically isolated from the peripheral circulation, in particular from buffy coat or leukapheresis samples, they can also be derived from bone marrow or cord blood.

[0018] As used herein, the term "culture medium" designates any medium appropriate for the culture of blood cells, in particularT cells. It includes, but is not limited to, RPMI 1640 medium, T vivo medium, CTS T, 4Cell® Nutri-T GMP Medium, CellGenix® GMP Advanced TCM, LymphoONE T-Cell Expansion Xeno-Free Medium cell culture media.The step (i) allows the activation of y6T cell subpopulation among the mononuclear cells by a boost with the activating compound.

[0019] The step (i) is preferably performed in static conditions to allow the adhesion of the accessory cells (other cells than y6 T cells) from the mononuclear cells at the bottom of the recipient.

[0020] As used herein, the term "phosphoantigen" refers to pyrophosphate-containing metabolites, which are recognized by y6 TCR and thus trigger the activation of y6 T cells. In particular, isopentenyl pyrophosphate (IPP) is generated from the endogenous mevalonate (MVA) pathway and accumulates intracellularly during dysregulated metabolism in many types of tumor cells. Also, (E)-4-hydroxy-3-methyl-but-2-enyl pyrophosphate (HMBPP) is a microbial metabolite from the isoprenoid pathway which represents "non-self" pathogen signals.

[0021] As used herein, the term "bisphosphonates" designates nitrogen-containing bisphosphonates, such as zoledronic acid or alkylamines, which cause intracellular IPP accumulation through inhibition of farnesyl pyrophosphate synthase.

[0022] The step (ii) allows the expansion and enrichment of the y6 T cell subpopulation.

[0023] As used herein, the term "static phase" means that no movement is applied to the culture.

[0024] The static phase allows the contact by sedimentation of y6 T cells (non-adherent cells) with the accessory cells (adherent cells) priorly (step (i)) adhered at the bottom of the recipient.

[0025] As used herein, the term "agitated phase" means that at least one movement is applied to the cell culture. This term includes, with no limitation, stirring, shaking, rocking, rotating, shocking, agitating or inverting the recipient containing the cell culture.

[0026] The agitated phase allows the suspension of y6 T cells, whereas the static phase allows the sedimentation of y6 T cells which results in their contact with accessory cells adhered to the bottom of the recipient and allows paracrine communications between cells.

[0027] In an embodiment, the invention relates to the method as defined above, wherein step (ii) begins with a static phase of 24 hours.

[0028] In an embodiment, the invention relates to the method as defined above, wherein step (ii) cyclically alternates static and agitated phases.

[0029] In an embodiment, the invention relates to the method as defined above, wherein step (ii) comprises at least one static phase and one agitated phase per day, preferably only one static phase and only one agitated phase per day during the first 7 days of culture.In an embodiment, the invention relates to the method as defined above, wherein each agitated phase lasts at least 1 hour.

[0030] In an embodiment, the invention relates to the method as defined above, wherein each static phase lasts at least 23 hours.

[0031] In an embodiment, the invention relates to the method as defined above, wherein each static phase lasts 23 hours and each agitated phase lasts 1 hour.

[0032] In an embodiment, the invention relates to the method as defined above, wherein step (ii) begins with a static phase, alternates static and agitated phases during at least 7 days, and ends with an agitated phase.

[0033] In an embodiment, the invention relates to the method as defined above, wherein step (ii) comprises:

[0034] - a static phase of 24 hours, followed by

[0035] - a cycle of an agitated phase of 1 hour and a static phase of 23 hours during the next 7 days, optionally followed by

[0036] - an agitated phase during the next 14 days.

[0037] Step (i) allows accessory cells adhesion on the culture vessel surface. Step (ii) begins with a static phase allowing a significant contact time between adhered cells and sedimented ones including cells of interest. Then, a cyclic alternation of 1 hour agitation of the culture, followed by 23 hours of static cultivation is applied and repeated for the rest of step (ii). Such an agitation allows to correctly homogenize cultivation media (improvement of mass-transfer of nutrients and gas at the vicinity of cells), leading to an appropriate aeration, but also to avoid any nutrients depletion and / or toxicity of local concentrations of metabolites (lactate, ammonium...). A 23 hour static culture phase then enables the non-adherent cells such as y6 T cells to sediment again and to enter in contact with adherent one to simulate the activation and proliferation of y6T cells, while respecting a time scale in accordance with mean mammal cells generation time and paracrine communication. The agitation can be performed by using any device which prevents sedimentation of the cells in the culture medium.

[0038] In an embodiment, the invention relates to the method as defined above, wherein the agitated phases during step (ii) are carried out with an orbital shaker, a rocking shaker or a rotator shaker.In an embodiment, the invention relates to the method as defined above, wherein the agitated phases are carried out at speed not exceeding 75 rotations per minute (rpm), preferably 70 rpm. In an embodiment, the agitated phase is carried out at a speed selected from the group comprising 75, 70, 65, 60, 55, 50, 45, 40, 35, 30, 25, 20, 15, 10 and 5 rpm.

[0039] In an embodiment, the invention relates to the method as defined above, wherein the mononuclear cells have been obtained on peripheral blood.

[0040] In an embodiment, the invention relates to the method as defined above, wherein the mononuclear cells have been obtained by leukapheresis or buffy coat.

[0041] In an embodiment, the invention relates to the method as defined above, wherein the activating compound is a phosphoantigen, preferably selected from the group comprising isopentenyl pyrophosphate (IPP), ethyl pyrophosphate (EtPP) and hydroxy-methyl-butyl-pyrophosphate (HMBPP), more preferably IPP.

[0042] In an embodiment, the invention relates to the method as defined above, wherein the activating compound is a bisphosphonate, preferably zoledronic acid (Zometa).

[0043] In an embodiment, the invention relates to the method as defined above, wherein, during step (ii), the cells are maintained at a cell density not exceeding 3 x 106cells / ml.

[0044] In an embodiment, the invention relates to the method as defined above, wherein, during step (ii), the cells are maintained at a cell density not exceeding 2 x 106cells / ml.

[0045] In an embodiment, the invention relates to the method as defined above, wherein, among the population of cells obtained after step (ii), the population of CD3+ cells represents at least 75%, preferably 80%, of the total nucleated cells, and is composed of more than 70 % of y6 T cells. Once a sufficient y6 T cells quantity and quality are reached (generally between day 8 and day 14 of culture), the cells obtained from step (ii) can be expanded in an agitated bioreactor under controlled conditions.

[0046] As used herein, the term "controlled conditions" means that the physicochemical parameters of the culture (e.g. pH, temperature, oxygen, carbon dioxide tension, agitation) are regulated.

[0047] In an embodiment, the invention relates to the method as defined above, further comprising: - a step (iii), wherein the cells obtained from step (ii) are expanded in an agitated bioreactor under controlled conditions.In an embodiment, cells obtained from step (ii) are cultivated in a continuous agitated mode, preferably with an orbital shaker, a rocking shaker or a rotator shaker, before being inoculated in a bioreactor.

[0048] In an embodiment, the invention relates to the method as defined above, wherein the agitation is carried out at rate not exceeding 75 rotations per minute (rpm), preferably 70 rpm, before reactor inoculation.

[0049] - In an embodiment, the agitation is carried out at a speed selected from the group comprising 75, 70, 65, 60, 55, 50, 45, 40, 35, 30, 25, 20, 15, 10 and 5 rpm, before reactor inoculation.

[0050] Seeding of the bioreactor is preferably performed at least at 1.5 x 106cells / ml as soon as the minimal desired inoculation volume is reached. In addition to that, it is preferred that a minimal purity of the inoculum is achieved such as the population of CD3+ cells represents at least 75%, preferably 80%, of the total nucleated cells, and is composed of more than 70 % of y6 T cells. In an embodiment, step (iii) is carried in an agitated bioreactor containing at least 1 liter of culture medium.

[0051] In an embodiment, step (iii) is carried in an agitated bioreactor containing from 1 to 10 liters of culture medium.

[0052] In an embodiment, the invention relates to the method as defined above, wherein, during step (iii), the cells are maintained at a cell density not exceeding 3 x 106cells / ml.

[0053] In an embodiment, the invention relates to the method as defined above, wherein, among the population of cells obtained after step (iii), the population of y6 T cells represents at least 80 % of CD3+ cells.

[0054] In an embodiment, the invention relates to the method as defined above, wherein the population of cells obtained after step (iii) comprises at least 80 % of y6 T cells.

[0055] In an embodiment, the invention relates to the method as defined above, wherein the population of cells obtained after step (iii) comprises at least 80 % of y6 T cells and is obtained in less than 22 days.

[0056] In an embodiment, the invention relates to the method as defined above, further comprising after step (iii):

[0057] - a step (iv), wherein the y6 T cells are recovered, and optionally sorted.In an embodiment, the invention relates to the method as defined above, further comprising after step (iii):

[0058] - a step (iv), wherein the y6 T cells are recovered, and sorted if the population obtained after step (iii) still contains more than 20% of ap T cells.

[0059] In a preferred embodiment, the invention relates to the method as defined above, which is carried out in fed-batch conditions.

[0060] In the present invention, the expression "fed-batch conditions" indicates that one or more nutrients is added continuously or intermittently into the medium after the start of cultivation.

[0061] In an embodiment, the invention relates to the method as defined above, wherein at least step (ii) or step (iii) is carried out in fed-batch conditions.

[0062] In an embodiment, the invention relates to the method as defined above, wherein both step (ii) and step (iii) are carried out in fed-batch conditions.

[0063] In another aspect, the invention relates to y6 T cells obtained by the method as defined above. In an embodiment, the invention relates to the y6 T cells obtained by the method as defined above, wherein said y6 T cells mainly belong to the y962 subpopulation.

[0064] In an embodiment, the invention relates to the y6 T cells obtained by the method as defined above, wherein at least 75% of said y6 T cells belongs to the y962 subpopulation.

[0065] In an embodiment, the invention relates to the y6 T cells obtained by the method as defined above, wherein said y6 T cells are y962 T cells.

[0066] In an embodiment, the invention relates to the y6 T cells obtained by the method as defined above which are stored frozen.

[0067] In another aspect, the invention relates to y6 T cells obtained by the method as defined above, for use as a medicament.

[0068] In an embodiment, the medicament is an advanced therapeutic medicinal product (ATMP).

[0069] As used herein, the expression "advanced therapeutic medicinal product" (or ATMP) designates, but is not limited to, gene therapeutics, somatic cell therapeutics and tissue engineered products. In an embodiment, the invention relates to y6 T cells obtained by the method as defined above, for use in an immunotherapy, preferably in the treatment or prevention of a cancer.In an embodiment, said cancer is a leukemia, preferably an acute leukemia, most preferably a post-hematopoietic stem cell transplantation acute leukemia relapse.

[0070] The following Examples are put forth so as to provide those of ordinary skill in the art with a complete disclosure and description of how to make and use the present invention, and are not intended to limit the scope of what the Inventors regard as their invention nor are they intended to represent that the experiments below are all or the only experiments performed. While the present invention has been described with reference to the specific embodiments thereof, it should be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the true spirit and scope of the invention. In addition, many modifications may be made to adapt a particular situation, material, composition of matter, process, process step or steps, to the objective, spirit and scope of the present invention. All such modifications are intended to be within the scope of the claims appended hereto.FIGURE LEGENDS

[0071] Figure 1. Principle of PBMC collection from a buffy coat bag using a density gradient separation technique. Mononuclear cells rings are isolated after centrifugation. PBS (Phosphate Buffer Saline), LSM (Lymphocyte Separation Medium), PBMC (Peripheral Blood Mononuclear Cells).

[0072] Figure 2. y6 T-cells sequential shaking program during expansion period and monitoring of cell characterization, expansion rate and biochemistry parameters.

[0073] Figure 3. Flow cytometry gating strategy for analysis. This strategy assesses the percentage of cells of interest based on surface marker expression (CD3) and T-cell Receptor (TCR) expression (y6 or ap).

[0074] Figure 4. Comparative kinetics of surface marker expression during yS T cell culture in static (black) and agitated (doted lines) conditions. Identity markers: CD3, a TCR, y6TCR, CD16. Activation / exhaustion markers: DNAM-1, LAG-3, TIG IT, PD1.

[0075] Figure 5. Growth kinetics of viable cells during the different phases of T cells culture. Cell expansion is expressed as total cells (concentration * volume), to avoid the dilution effects of regularly adding variable volumes of culture medium (fed batch, dotted arrows). The removal of 200 mL of culture indicates that the maximum reactor volume has been reached. The entire process was punctuated by daily sampling for process monitoring. Samples were taken at DO, D7, D13, D17, D21 and D27 for phenotypic analyses by flow cytometry.

[0076] Figure 6. Phenotype characterization of expanded cells (D0-D2) by flow cytometry. The initial y6 T subpopulation of interest was significantly expanded through the culture and reached 90% of the total population at D27. a T cell subpopulation was significantly depleted in shaking culture conditions, compared to static culture (control culture).

[0077] Figure 7. Growth kinetics of viable cells during the different phases of T cells culture for N=4 replicates. Cell expansion is expressed here as total viable cells quantity (concentration * volume), to avoid the dilution effects of on demands adding of optimal volumes of fresh supplemented culture media, L-Glutamine or glucose solution (fed batch mode), and daily withdrawals.

[0078] Figure 8. Expansion factor of y6 T cells at Day 21 in static and in shaking cell culture. y6 T cells are significantly more expanded using shaking culture method than in static culture.Figure 9. Subpopulation repartition for static and shaking culture conditions. (A) Evolution of the percentage of y6, ap and other subpopulation of cells between Day 0 and Day 21 (harvest day) using either static or shaking culture conditions. y6 T cell subtypes repartition using static (B) and shaking (C) culture modes. N=4.

[0079] Figure 10. Activation markers (NKG2D and DNAM-1) of y6T cells. (A) Static culture conditions. (B) shaking culture conditions (N=5). (C) Comparison of both culture methods regarding activation markers expression between DO and D21.

[0080] Figure 11. Exhaustion markers of y6 T cells (PD-1, TIM-3, TIGIT). (A) Static culture conditions. (B) Shaking culture conditions N=5. (C) Comparison of both culture methods regarding exhaustion markers expression between DO and D21.

[0081] Figure 12. Kinetic of cell expansion in conventional mode of culture (static), in alternate shaking culture (shaken) and controls (Tl: continuous shaking; T2: without feeder cells; T3: without Zometa pulse). Total cell quantity was determined by flow cytometry numeration.

[0082] Figure 13. Evolution of the percentage of CD3+ cells (A), y6 and a T cells (B) depending on culture conditions (static, alternate shaking, continuous shaking (Tl), without feeder cells (T2), without zometa pulse (T3)).

[0083] Figure 14. Effect of the type of culture medium used in CSC, and comparison with GS culture method on y6 T cell number and fold expansion. GS: RPMI static culture (control condition); Ml: 4Cell® Nutri-T GMP Medium; M2: CellGenix® GMP Advanced TCM; M3: LymphoONE T-Cell Expansion Xeno-Free Medium. * step (iii) inoculation of a controlled bioreactor in M3 generating a high amount of foam leading to cell death.

[0084] Figure 15. y6 T cells enrichment comparison between GS method (RPMI static - control) and with the three commercial serum-free medium (Ml, M2 and M3) performed with the CSC method.

[0085] Figure 16. Expansion kinetic comparison of y6 T cells cultivated in RPMI and Ml media in CSC, with y6 T cells culture performed using the GS method. GS: RPMI static culture (control condition); Ml: 4Cell® Nutri-T GMP Medium.

[0086] Figure 17. Phenotype characterization of expanded cells (D0-D14) by flow cytometry. The initial y6 T subpopulation of interest was significantly expanded through the culture, since the ap T cellsubpopulation was significantly depleted in cyclic shaking culture conditions (RPMI and Ml), compared to static culture (GS - control culture).

[0087] Figure 18. y6 T cells enrichment (A) and fold expansion (B) in cyclic shaking (RPMI and Ml) and static (RPMI) culture conditions.

[0088] Figure 19. Expression of activation markers in y6 T cells expanded in static condition (RPMI media (A), in cyclic shaking condition in RPMI medium (B) and Ml media (C).

[0089] Figure 20. Kinetics of exhaustion markers expression (PD1, TIM3, TIGIT) of y6 T cells cultivated in RPMI (static vs cyclic agitation) and in Ml media (cyclic agitation) (A) and comparison of exhaustion markers expression levels between culture conditions, at the end of the culture (D21) (B).

[0090] Figure 21. Expansion kinetic comparison of y6T cell quantities and expansion folds cultivated in (A) RPMI and (B) Ml medium with CSC method, and (C) according to US 2024 / 344026 Al ("026). (D) Comparative y6 T cell expansion kinetics in each condition (n= total number of y6 T cells). Independent biological replicates have been performed, and the mean values are shown.

[0091] Figure 22. Phenotype characterization of expanded cells (D0-D14) by flow cytometry. The initial y6 T cells subpopulation of interest was significantly expanded through the culture, since the ap T cells subpopulation was significantly depleted in cyclic shaking culture conditions (RPMI (A) and Ml (B)), and in culture condition of US 2024 / 344026 Al ("026) (C). Independent biological replicates have been performed, and the mean values are shown.

[0092] Figure 23. Kinetics of exhaustion markers expression (PD1 (A), TIGIT(B) and TIM3 (C)), and cytotoxic activity / activation markers (NKG2D, DNAM1 and CD16) (D) of y6 T cells cultivated in RPMI and in Ml media in CSC condition and based on the protocol of US 2024 / 344026 Al ("026). Independent biological replicates have been performed, and the mean values are shown.

[0093] Figure 24. Expansion kinetic of y6 T cells, and specific y6 T cells fold expansion, cultivated in Ml medium according to CSC method (N= 3). Three biological replicates have been performed, and the mean standard deviation values are shown.

[0094] Figure 25. CD3+ cells and its subpopulation evolution kinetics in Ml medium, using the CSC method. Three biological replicates have been performed, and the mean ± standard deviation values are shown.

[0095] Figure 26. Comparative analysis of y6 T enrichment at DO (A), D7 (B), D14 (C) (N=3) and D18 (D)(N=1)). Cells of interest (y6T cells) are progressively enriched through CSC process (Ml medium),whereas percentages of a T cells, and other cells impurities such as NK decreased. The results shown above correspond to the average of three independent biological replicates.

[0096] Figure 27. Kinetics of exhaustion markers expression (PD1, TIM3, TIGIT) (A) and kinetics of activation markers (NKG2D, DNAM1) expression (B) of y6 T cells cultivated in Ml medium and according to CSC culture conditions. The results shown above represented the mean ± standard deviation of three independent biological replicates.EXAMPLES

[0097] Materials & methods

[0098] Culture media

[0099] Three different commercial culture media and supplementation have been used for the example presented below.

[0100] RPMI medium 1640 (Fisher Scientific) supplemented with Human serum type AB (MALE) (SAB) at 10 % (V / V) (Sigma), Human IL-2 IS, premium gr. (200 pg) (Miltenyi Biotec) at 500 UI / mL, Antibiotic Antimycotic Solution (100x) (Sigma) at 1% (V / V) and L-Glutamine solution (200 mM) (Ref: Sigma) at 2% (V / V) is used. Media can punctually be supplemented by adding L-Glutamine solution (200 mM) (Ref: Sigma) and / or D-(+)-Glucose solution 45% (Sigma).

[0101] 4Cell® Nutri-T GMP Medium (Sartorius) supplemented with Human IL-2 IS, premium gr. (200 pg) (Miltenyi Biotec) at 500 UI / mL, and Antibiotic Antimycotic Solution (100x) (Sigma) at 1% (V / V) is used. Media can punctually be supplemented by adding L-Glutamine solution (200 mM) (Ref: Sigma) and / or D-(+)-Glucose solution 45% (Sigma).

[0102] CellGenix® GMP Advanced TCM (Sartorius) supplemented with Human IL-2 IS, premium gr. (200 pg) (Miltenyi Biotec) at 500 UI / mL, and Antibiotic Antimycotic Solution (100x) (Sigma) at 1% (V / V) is used. Media can punctually be supplemented by adding L-Glutamine solution (200 mM) (Ref: Sigma) and / or D-(+)-Glucose solution 45% (Sigma).

[0103] LymphoONE T-Cell Expansion Xeno-Free Medium (Takara) with supplemented Human IL-2 IS, premium gr. (200 pg) (Miltenyi Biotec) at 500 UI / mL, Antibiotic Antimycotic Solution (100x) (Sigma) at 1% (V / V) and L-Glutamine solution (200 mM) (Ref: Sigma) at 1% (V / V) is used. Media can punctually be supplemented by adding L-Glutamine solution (200 mM) (Ref: Sigma) and / or D-(+)-Glucose solution 45% (Sigma).

[0104] Cells

[0105] First, Peripheral Blood Mononuclear Cells (PBMC) are isolated from a buffy coat (Human volunteer adult donor, Etablissement Frangais du Sang (EFS)) using a density gradient centrifugation method. Buffy coat diluted with Phosphate Buffer Saline (PBS) (dilution factor: %) is gently poured on a Lymphocyte Separation Medium (LSM) (Eurobio Scientific, France) or Ficoll Paque Premium (Cytiva) cushion (Figure 1). The gradient density separation is performed by centrifugation (400 x g, room temperature, 20 min, Acceleration / break: 0). The PBMC ring is then transferred and washed withPBS (200 x g, room temperature, 20 min, Acceleration: 7, Break: 0). The PBMC pellet is resuspended in PBS and washed (400 x g, room temperature, 10 min, Acceleration: 9, Break: 9). The supernatant is discarded, and the cell pellet is finally resuspended in culture media. The volume is adjusted to reach a total cell concentration of 2.5xl06cells.mL-1

[0106] Activation mechanism of yd T-cells

[0107] After separation by centrifugation, PBMC are seeded at a concentration of 2.5 x 106viable cells mL-1in culture media Erlenmeyer flasks (total volume 125 mL, maximal working volume 25 mL). Zoledronic acid (ZOMETA, Novartis AG) (lOpM) is added to cultures. Cells are incubated for 14 -18h at 37°C, 80% H2O, 5% CO2 under static conditions. Cells are then harvested by centrifugation (400 x g, room temperature, 10 min, Acceleration: 9, Break: 9) and resuspended in culture media at final concentration of 2.5xl06viable cells mL1in the same Erlenmeyer flasks, to keep adherent accessory cells for y6 T-cells activation.

[0108] Culture conditions

[0109] Two expansion methods have been compared: one in static condition considered as the gold standard (GS) and one in a cyclic shaking condition (CSC).

[0110] y6 T cells expanded using GS conditions are cultivated for 21-25 days in ventilated culture flasks (Day 0 - 7) to keep initial adherent accessory cells, and then in T-Flask (75cm2) after day 7. (37°C, 80% H2O, 5% CO2). Fresh media is added every 2 days in order to maintain the cell concentration between 1 - 3xl06viable cells. mL1.

[0111] In comparison with GS cultures, cells in shaking conditions (Figure 2) are cultivated in Erlenmeyer flasks with ventilated lids in static conditions, during 24 hours. Then, a sequential shaking cycle is applied during 168 hours (Day 1 to day 8) with a cycle of 1 hour at 70 rpm an orbital shaking incubator (Kuhner, LT-X LabTherm), and then followed by 23 hours of static culture conditions (Phase 1). This alternate regimen between shaking (1 hour) and static (23 hours) steps is performed during 7 days in order to allow either the physical contact between accessory and y6 T-cells or resuspension and homogenization during the agitated phase.

[0112] After 168 hours of culture (7 days), the cells in suspension (containing activated T-cells) are transferred in larger Erlenmeyer flasks to allow the culture scale-up, respecting a working volume / total volume of 1 / 5 ratio, optimal for gases exchanges. From day 8 until days 12 / 14, cells are expanded under continuous shaking condition (37°C, 80% H2O, 5% CO2, 70 rpm - 50mm) in an incubator (Kuhner, LT-X LabTherm). This second phase (Phase 2) allows further enrichment in y6 T-cells and scale up. During both culture conditions (GS and CSC), daily sampling is carried out under sterile conditions to evaluate VCD (Viable Cell Density), TCD (Total Cell Density) and cell viability using an automated cell counter based on the method of trypan blue exclusion (Vi-Cell XR -Beckmann Counter.) Biochemical characterisation of substrates (glutamine, glucose) and metabolic products (lactate, ammonium ions, lactate dehydrogenase (LDH)) is also carried out daily, using an automated biochemical analyzer (Gallery, ThermoFisher). Addition of fresh media or supplements (Glucose, L-Glutamine) is determined according to the cell density measurements in order to maintain cell concentration between 1 - 3 x 106viable cells mL1. Occasional additions of concentrated Glucose and L-Glutamine solution are performed to maintain at least lg L1of glucose, and 2 mM glutamine, to limit nutrient deprivations. During the expansion of y6 T-cells, both in agitated Erlenmeyer flasks and later in bioreactors, regular flow cytometry analyses are carried out to determine their phenotypic characteristics (see section below, Cells characterization).

[0113] Culture in bioreactor

[0114] Between day 8 and 14, depending on the cell density and y6 T-cell enrichment, a stirred-tank bioreactor with elephant ear mobile (pumping from bottom up) is inoculated (37°C, 50% O2 relative to concentration in air, pH 7.20, 120 rpm). The quantity of y6 T-cells cells reached after phase 2 (total volume of flasks x cell concentration) is used to determine both the initial working volume and the date of the transfer of cells in the agitated bioreactor. To avoid any growth latency at the beginning of the culture in the bioreactor, a minimum concentration of y6 T-cells is required (1.5 x 106cells. mL1) for a 1 L bioreactor. The pH and pC values are regulated using the Air or O2 / Nitrogen and CC / NaOH (0.1 M) regulator pairs respectively. During the expansion of T cells in the agitated bioreactor, the GS culture is still maintained in static conditions (See section Culture conditions). Daily sterile sampling is carried out to monitor VCD, TCD, cell's viability and biochemical parameters (Glucose, Glutamine, Lactate, LDH) using both cell counter (ViCell XR, Beckman) and automated biochemical analysers (Gallery, Thermo). Cells are characterized by flow cytometry right after bioreactor's inoculation (Day 8 -14), and at day 21 (end of the expansion process).

[0115] Cells characterization

[0116] Cells from GS and CSC methods are characterized by flow cytometry (M ACSQuant analyzer, Miltenyi Biotec). Identity, but also activation and exhaustion markers are monitored at day 0, day 7, bioreactor inoculation day, day 14, day 21 and final day of the culture) using two different cytometry multiple-markers-panels of antibodies (Table 1).

[0117]

[0118] Table 1: Identity and activity / exhaustion flow cytometry panel markers for y6 T-cells phenotypic characterization during expansion processes in static and cyclic shaking conditions (Identity = to identify cells as T lymphocytes and also to differentiate between a , and y6 sub-phenotype; Activation = to ensure that y6 T-cells are functional; Exhaustion = Cell exhaustion related to replicative senescence due to culture time, or cytokine stimulation).

[0119] For y6 T cell enrichment follow-up, flow cytometry analyses are performed weekly between day 0 and day 21. The "Identity" panel was designed to follow y6 and a T cell sub-populations during expansion. Within the y6 population, cells expressing a TCR with a 62 chain are particularly followed as well as cells with a naive, central memory or effector memory phenotype based on CCR7 and CD45RA expression. The y6 T cell exhaustion is followed using PD1 (programmed cell death protein 1), TIM-3 and TIGIT. High levels of these proteins generally result in an exhaustion as well as a loss of T cell functions. Regarding activating receptor expression, DNAM-1 is an activating molecule that promotes tumor cell lysis through its interaction with the PVR / CD155 (Polioma Virus Receptor) protein expressed on the surface of tumor target cells. The DNAM-l / PVR pair is thus involved in tumor and viral immunosurveillance. NKG2D recognizes ULBP and MICA / B molecules expressed on leukemia cells. 7-Aminoactinomycin D (7-AAD) was also added to the sample a few minutes before analysis to check cell viability.The gating strategy implemented for y6 T-cell phenotyping is shown in Figure 3. Cells are visualized by labelling with anti-CD3 antibody PEVio-770A and by adding 7AAD to the sample. CD3+ / 7AAD-cells correspond to the viable T lymphocytes. Finally, y6 or ap TCR expression was determined after gating on CD3+ / 7AAD- cells. The expression of the markers belonging to the 2 panels were then determined on CD3+ / 7AAD- / TCR y6+ cells.

[0120] In addition to Identity and activity / exhaustion flow cytometry panel analysis, a Maturation, terminal differentiation and impurities / subtype analysis by flow cytometry were punctually performed to further characterize the expanded cells (Table 2).

[0121]

[0122] Table 2: Maturation, terminal differentiation and impurities flow cytometry panel markers for y6 T-cells phenotypic characterization during expansion processes performed in static and cyclic shaking culture conditions (Maturation= to quantify maturation level in expanded cells, proportional to CD85J expression; Terminal differentiation = y6 T-cells highly differentiated associated with important cytotoxic properties (KLRG1 +); Impurities = to quantify the amount of residual cells as Natural Killer ( CD56 +), Monocytes or macrophages (CD14+); Sub-type = to identify the exact sub type of y6 T-cells (V62 + and Vy9+).

[0123] A similar gating strategy applied to Identity and activity / exhaustion panel is also implemented for Maturation, terminal differentiation and impurities panel. Cells are visualized by labelling with anti-CD3 antibody PEVio-770A, TCR V62 Antibody PE, and TCR Vy9-PC5. CD3+ / V62+ / Vy9+ cells correspond to the subpopulation of interest. Within it, the expression level KLRG1, is quantified to determine cytotoxic properties, as the CD85J expression level is proportional with maturation level. On CD3- cell population, the proportion of CD56+ cells, corresponding to NK cells, and CD14+ cells, corresponding to macrophage or monocytes is determined to quantify residual populations proportion.1. v6 T cell expansion in cyclic shaking conditions

[0124] The method of the invention aims to expand clinical-grade T y6 cells intended for the treatment of post-transplant leukemia or hematological malignancies relapse. To achieve this goal, the culture method needs to specifically amplify from the PBMC the y6 phenotypic subpopulation of T lymphocytes, which is a scarce population in healthy donor's blood (between 0.5 and 5% depending on the individuals).

[0125] A selective amplification should allow to obtain a high enrichment (above 80%) of y6 T cells and, at the same time, an efficient depletion (below 15%) of the ap T cell subpopulation responsible for the GvH reaction. Also, a selective amplification should produce a high quantity of y6 T cells rendering possible a culture of y6 T cells in controlled bioreactors, at a scale up to 5-10 liters. In addition, it is necessary to limit the cell "exhaustion", which can result in a therapeutic product with little or no activity.

[0126] In the present invention, it was hypothesized that a cyclic agitation during the first step of the culture could improve gas and nutrient transfers while keeping a contact with the adherent accessory cells at the bottom of the flasks.

[0127] For this purpose, PBMC were isolated using a density gradient, treated with Zometa (zoledronic acid), and seeded in static conditions (GS) and in cyclic shaking conditions (with intermittent agitation in shaking flasks). An initial flow cytometry analysis was performed at DO to evaluate surface markers. The same analyses were performed at D7, D14 and D21. Weekly sample collections were performed for cell counting and phenotyping analyses. The kinetic expression of each target marker during the culture, in both shaking and static conditions, is shown in Figure 4.

[0128] Specific surface markers related to the identity of cells in suspension, such as CD3, TCR y6, TCRaP and CD16 were quantified. With regard to the purity of cultures, 71% CD3+ lymphocytes at DO have been measured, and reached up to 91% and 83% in static and shaking conditions respectively at D21. More specifically, 92% and 96% of cells expressed y6 TCR in static and in agitated conditions respectively, with a multiplication rate higher than 800x from DO to D21.

[0129] With regard to exhaustion markers, LAG3 increased for both culture conditions until D7 and then stabilized until the end of the culture. Maximum in PD-1 expression was reached at D14 and then declined until the end of culture. This increase in cell expression was more pronounced in cyclic shaking cultures compare to static cultures (33 vs. 20% respectively). TIGIT+ cells decreased completely during static culture whereas this marker expression was maintained in shakingconditions until D14, then decreased after a peak. At D21, expression of this marker is completely abolished in both culture conditions. Interestingly, a strong induction in DNAM-l expression was observed in almost 100% of cells after D7 and until the end of the culture, whatever the cell culture conditions, indicating gain in cytotoxic activity and therefore an interesting functionality of cells. The experiments carried out in both static and agitated conditions showed that cells had a strong proliferative capacity. Additionally, exhaustion markers such as PD-1 and TIG IT were lost during the culture whereas activation markers (i.e. DNAM-l) were strongly induced. Only the shaking condition allows to reach enough y6 T cells to seed the bioreactor used for the next step of expansion.

[0130] Example 2. v6T cell expansion in cyclic shaking conditions and passage in controlled bioreactors In order to confirm its efficiency and robustness despite biological variability, the shaking process was carried out using independent donor buffy coats. The following experiment generated a reference growth kinetic (Figure 5), that was used to orient further replicates in terms of cell sampling and phenotyping using flow cytometry, or in terms of feeding and cell harvesting timing. In addition, complete culture medium (fed batch mode) was added at regular time intervals from day 3, depending on cell growth and residual glucose and glutamine concentration in culture media, to maintain a sufficient amount of nutrients in the bioreactors.

[0131] Moreover, flow cytometry analyses of cells confirmed the progressive and substantial enrichment in y6 T cells (Figure 6). Across all checkpoints (DO, D7, D13, D17, D21 and D28), y6 T cells identity markers indicated an increase in purity up to 97.65% of CD3+ cells (T cells identity), including 90% of y6 T cells among them at D27.

[0132] Compared with the control experiment (static culture), the total expansion rate increased 76-fold (vs. 3-fold for the control), with a 1,600-fold enrichment in y6 T cells (vs. 34-fold for the control), while ap T cells were significantly reduced, representing less than 10% of T cells in the agitated bioreactor vs. more than 25% in the static control culture.

[0133] The cell identity and activation markers measured (CD3 positive at D21 = 96.96% and DNAM-l positive at D27 = 99.45%), indicate strong activation marker expression, whereas cell exhaustion markers such as PD-1+ at D27 remained low (less than 2.5%).Repetition of experiments were performed to confirm results and trends, using the method combining cyclic shaking program during the first phase. All replicates (using different sex, age and blood group donor buffy coats) led to similar trends, despite variations in initial cell quantities and purities, and consequently leading to variable final amount of y6 T cells after expansion.

[0134] In four described replicates (N=l; N=2; N=3; N=4), the cell initial quantity has been significantly expanded (Figure 7). y6 T cells were significantly more expanded in shaking culture conditions than in static culture conditions (Figure 8).

[0135] Variation in subpopulations in GS and CSC conditions displayed the same pattern (Figure 9, A), with a significant depletion in CD3 negative cells. The ap T cells, initially predominant at day 0, are efficiently depleted at day 21 in both culture conditions (bellow 15 % of the cells). y6 T cells, are significantly enriched at day 21 compared to the beginning of the culture and reached approximately 80% in static conditions, and near 90% in bioreactor cultures. Experiments showed a better purity of the final product in bioreactor cultures compared to conventional static culture. In each experiment, more than 80% of cells were of y6 T cells whereas a T cells remained below 15%. In addition, other y6 T phenotype markers were similar whatever the culture conditions (Figure 9, B and C).

[0136] In addition, a higher quantity of cells of y6 Effector memory (TEM) T cells were obtained in the bioreactor compared to static culture.

[0137] Other markers such as those related to activation and exhaustion were monitored by flow cytometry (Figure 10, A and B). Activation markers (NKG2D, and DNAM1) appeared not to be affected by the cultivation method suggesting that after expansion, the functionality, and consequently their therapeutic properties were maintained. Major cons of cells expansion in bioreactor could be an earlier cell exhaustion. However, bioreactor cultures exhibited lower level of TIM 3, PD1 and TIGIT exhaustion markers expression at the end of the process (Figure 11, B), compared to static cultures (Figure 11, A). Despite the mechanical stress potentially induced by the agitation in a bioreactor, an efficient expansion of y6 T cells (CD3+, y6 TCR+, V62+), displaying a good activation (Figure 10, C) and a low exhaustion (Figure 11, C) was obtained.

[0138] Example 3. Further experiments with various controls

[0139] In order to demonstrate the efficiency of the method of the invention, based on a cyclic shaking culture, different experiments have been carried out.Starting from a single buffy coat, five different culture conditions experiments have been compared (n=2): (i) GS culture in flasks, (ii) CSC (as described above) and (iii) three control cultures : with continuous shaking during the three phases (Tl), (iv) with sequential shaking (phase I) but without accessory cells (T2) and (v) with sequential shaking, including accessory cells but without Zometa (T3).

[0140] Comparison of cell quantity kinetics between D0-D18 confirmed the cell expansion efficiency using the sequential shaking conditions (Figure 12). Conventional static culture, and controls (Tl, T2, T3) exhibited similar poor cell expansion, indicating that sequential shaking of cells, preliminary treatment with Zometa and the presence of adherent accessory cells are essential to promote total cell expansion. Taking into account these results, the y6 T cells enrichment as well as the ap T cells depletion were also monitored by flow cytometry (Figure 13).

[0141] Starting from a single buffy coat, initial CD3+ cells in each culture conditions were equivalent (84 %), then decreased at day 18, 79% of the cell in shaken condition are CD3 + cells, compared to 59 % in static conditions and 67,83%, 64,7 % and 71,7 % for Tl, T2, T3 respectively. Within the CD3+ cells, the subpopulation percentages of cells of interest (y6 TCR+ cells) and of a TCR+ cells were compared. Both static and shaking conditions resulted in a similar enrichment in y6 TCR+ cells with 91,87 % and 90,215 % respectively and in depletion in ap TCR + cells, with 5,63% and 4,43 % respectively. As expected, controls (Tl, T2, T3) exhibited lowered enrichment in y6 T cells and depletion in ap T cells, during the cultures (Figure 13). Taking together, these results (Figures 12 and 13) confirm the poor performance of the expansion process in continuous shaking, without feeder cells and without Zometa, but also the superiority of the sequential shaking process compared to conventional static method in terms of y6 T cells quantity after expansion.

[0142] Example 4. Tests with different commercial media

[0143] To produce y6 T cells in serum free expansion conditions and according to GMP regulatory requirements, three different media, 4Cel I® Nutri-T GMP Medium (Ml), CellGenix® GMP Advanced TCM (M2) and LymphoONE T-Cell Expansion Xeno-Free Medium (M3) were compared regarding production yield, purity, and phenotypic quality.

[0144] To this end, PBMC were isolated using a density gradient, treated with a Zometa (zoledronic acid) pulse, and seeded either in static conditions (GS - control) or in cyclic shaking conditions (CSC) for each medium starting with same initial total nucleated cells quantity and y6 T cells amount (Figure14, A, Day 0). Weekly CMF analysis (DO, D7, D14 and D21) were performed to evaluate expanded cells' purity and quality.

[0145] After day 7, y6 T cells in GS or CSC (Ml, M2, and M3) started to diverge from each other. y6 T cell proportion was higher using Ml and M3 than using GS method or M2. This trend was reinforced over days. At day 21, the quantity of y6 T cells obtained in CSC and in Ml media (2.9xl09y6 T cells) was 19 times higher than the quantity of y6 T cells obtained with GS method (around 1.5xl08y6 T cells).

[0146] A higher fold expansion of y6 T cells (361 y6 T cells fold expansion in RPMI vs 675 fold expansion in M2 at day 21) was observed using M2 in CSC at day 21 (Figure 14, B) than in RPMI medium in static conditions. However, no significant improvement regarding final y6T cell number (1.31xl08in RPMI vs 9.78xl07in M2 at day 21) was observed, probably due to a lower y6 T cells percentage in M2 compared to GS with 58.86% and 78% respectively (Figure 15) at the end of the culture. Thus, M2 media in CSC allowed to quickly increase cell quantities compared to GS, but lead to a poor enrichment in y6 T cells.

[0147] The use of M3 in CSC, allowed to expand y6 T cells efficiently until day 14. Actually, 2.10xl07of y6 T cells were obtained using RPMI, and l,91xl08y6 T cells were obtained in M3 (Figure 14 A).

[0148] Unfortunately, this medium seemed not to be appropriated for step (iii) in agitated bioreactor (Figure 14, A and B), since a strong foam formation was observed, leading to a massive cells death, and finally resulting in reduced fold expansion performances. Medium M3, has therefore been removed from our test panel, due to its intrinsic characteristic and composition incompatibility with agitation and aeration in agitated bioreactor.

[0149] Cultures in Ml medium in cyclic shaking conditions exhibit high performance compared to the static method in terms of y6T cells quantity and expansion fold (Figure 14). At the end of the culture (Day 21), a total amount of 1.65xl09y6 T cells have been obtained (approximately 10 times more than the number obtained with the GS method). In addition, an efficient and progressive enrichment in cells of interest was observed after day 7, with 77 % of y6 T cells (vs 55 % in RPMI static or 51% in M3 at day 7), and rose until 94% after day 14 (vs 74 % y6 T cells in RPMI static) (Figure 15).

[0150] Experiments performed in Ml medium also allowed to obtain high y6 T cells fold expansion rates over days, with values always superior to those obtained with the other conditions and media tested after day 7, with a final expansion fold of 2331 times at day 21 (vs 361 fold expansion in RPMI, 675 fold expansion in M3).Taken together, the results obtained confirmed the efficiency of the CSC method compared to the GS method, independently of culture media. The culture of y6 T cells in serum free commercial media with a cyclic shaking is possible and efficient. The medium Ml (4Cell® Nutri-T GMP Medium) exhibits better results compared to M2 and M3 in terms of total y6 T cells quantity, purity and fold expansion, without detrimental effect on expanded cells quality, and with low expression of exhaustion markers whereas activation markers were maintained. Thereby, the4Cell® Nutri-T GMP Medium, i.e. the Ml medium, has been chosen for further testing.

[0151] To confirm the results described above, comparative experiments between GS (static culture in RPMI), CGS (in RPMI medium with Serum Ab (SAB), and in 4Cell® Nutri-T GMP Medium (Ml)), have been replicated (N=2). As previously described, PBMC were isolated using a density gradient, treated with Zometa (zoledronic acid) pulse, and seeded in static (GS - control) and in cyclic shaking conditions (CSC) for both media (RPMI and Ml), starting with same initial total nucleated cells quantity and y6T cells amount (Figure 16, Day 0). Weekly CM F analyses (DO, D7, D14 and D21) were performed to evaluate expanded cells purity and phenotypic quality (viability, cell population, activation, exhaustion markers).

[0152] Cell quantity produced in CSC using both media was higher than cells quantity obtained using static method. At day 21, 1.31xl08y6 T cells are produced in GS, whereas l,02xl09and 1.65xl09y6 T cells were respectively produced in RPMI CSC and Ml CSC.

[0153] Flow cytometry analyses were performed weekly, to estimate population proportion in y6 T cells, ap T cells and CD3- residuals cells (Figure 17) proportion within the total cell population.

[0154] CD3+ cells population represented almost 80 % of the total cells at day 0 in each tested condition, progressively increased and finally reached 87 % in RPMI static, 95.36 % in RPMI CSC, and 98 % in Ml CSC (Figure 17, A) at day 14, respectively. Within CD3+ cell population, the y6 T subpopulation has been strongly enriched through expansion process. At Day 14, 95 % CD3 + cells are y6 T cells and only 4 % of a T cells still remained, using Ml media with cyclic shaking. CSC in RPMI allowed to obtain 73 % of y6 T cells, whereas GS method only 69 % of y6 T cells with 26 % of ap T cells remaining at day 14 (Figure 17, B).

[0155] At day 14, the highest y6 T cell enrichments were obtained using the cyclic shaking condition. This trend was confirmed until day 21, with more than 90% of y6 T cells obtained in Ml CSC compared to 78% in RPMI static condition (Figure 18, A). In addition to a progressive enrichment in cells of interest, y6 T cells fold expansion rate significantly improved in cyclic shaking compared to staticcondition (Figure 18, B). At the end of the culture (day 21), an expansion fold of y6 T cells of 2331 was obtained in Ml CSC culture versus 723 in RPMI CSC, and 361 in RPMI static condition.

[0156] These results demonstrate an efficient and quicker enrichment in cells of interest, as well as a depletion in ap T cells with CSC method, and even in a more efficient manner using the serum-free Ml medium. Since the quality and functionality of the produced cells are keypoints, flow cytometry comparative analyses of both activation and exhaustion markers have been performed during the expansion kinetics.

[0157] Flow cytometry analyses confirmed that y6 T cells produced via both cultivation method (static and cyclic shaking) and in both media (Ml and RPMI) conserved the CD16 marker, (over than 93 % of CD16+ cells) (Figure 19). Additionally, more than 99% of the y6 T cells expanded are DNAM1+ / NKG2D +. The expression of these two activation markers confirmed that the expanded y6 T cells may have a cytotoxic activity and that they have undergone activation (Figure 19). As the y6 T cells are not a continuous cell line, they are limited in terms of numbers of cell divisions and are subject to exhaustion through cultivation process due to replicative senescence. In addition to that, y6 T cells proliferation stimulation by interleukin addition in the media could lead to a cytokine overstimulation and finally to cell exhaustion. Previous experiments have shown that cyclic shaking conditions did not increase exhaustion rate (in RPMI). This observation was confirmed in comparative experiments (Figure 20, A). At day 21, exhaustion markers PD1, TIG IT and TIM3 were respectively expressed by 69%, 67 % and 97% of the cells in GS method, whereas only 48% were PD1+, 37% TIG IT + and 98% TIM3+ in cyclic shaking condition in RPMI. Taken together, expression levels of exhaustion markers remained lower in cyclic shaking than in static culture in RPMI medium (Figure 20, A and B).

[0158] In Ml media (CSC), only 17% PD1 +, less than 1% TIGIT +, and 80% of TIM3+ of y6 T cells were detected by FCM at day 21. Taken together, these results suggested a lower expression of exhaustion marker levels on cells cultured in cyclic shaking condition. This apparent decrease in exhaustion markers expression is even more pronounced in Ml (CSC) than in RPMI medium. A low level of exhaustion markers, is associated to optimal y6 T cells cytotoxic activity, suggesting that both cyclic shaking condition and the absence of serum (Ml serum free), does not have any negative influence on cell activity and functionality.

[0159] In conclusion, the comparative analysis of y6 T cells culture in static (RPMI) and cyclic shaking conditions (RPMI, Ml, M2, M3) indicates the superiority of cyclic shaking over static culture regardless of the media and in absence of serum, in terms of y6 T cells fold expansion andphenotype. Thus, to comply with GMP guidelines, the switch from a medium supplemented with human serum AB (RPMI) to a serum-free media is possible, while maintaining quantity and quality standards. Selecting 4Cell® Nutri-T GMP Medium (Ml), is an appropriate manner to comply with cell therapies manufacturing regulatory authorities. Culture in CSC and in Ml, allowed to produce active (CD16+, DNAM1+ / NKG2D+), poorly exhausted (low percentage of PD1+, TIGIT+, and TIM3+), with high purity (over than 90 % of TCR y6+ cells) and with high fold expansion rate of y6 T cells at the end of the process (over 2000 specific y6 T cells fold expansion).

[0160] Example 5. Comparative results with the method of US 2024 / 344026 Al

[0161] The Cyclic Shaking Culture (CSC) parameters and especially shaking frequencies and cycles have been designed to improve expansion of y6T cells. During the first step, cells undergo a daily shaking phase followed by a static phase. As previously described, the alternation of shaken and static culture conditions is cyclically repeated to allow daily homogenization of the media to prevent gradient formation (pH, nutrient, etc.) while maintaining prolonged contact phase between y6 T cells and adhered accessory cells which is essential for the cells of interest activation.

[0162] In US 2024 / 344026 Al (hereafter "026), a method to activate y6 T cells for antibody dependent cytotoxicity has been described, with an expansion phase comprising a static phase followed by a continuous shaking phase.

[0163] To assess the originality of the CSC protocol of the invention, this protocol was compared to a culture method based on the static phase followed by a continuous shaking, replicating the protocol of "026. For this purpose, two biological replicated experiments, including 6 technical replicates of each condition have been performed: CSC in RPMI medium, CSC in Cell® Nutri-T GMP Medium (Ml) and OpTmizer CTS medium and supplements (Thermo Fisher) as described in US 2024 / 344026 Al. To allow a comparison with the method described in "026, its whole expansion protocol has been strictly replicated.

[0164] To this end, PBMC were isolated using a density gradient, treated with a Zometa (zoledronic acid, ZOL) pulse, and seeded either: in Cyclic Shaking Conditions (CSC) for each medium (RPMI or Ml), and starting with the same initial concentrations of Total Nucleated Cells (TNC) and y6 T cells, or according to US 2024 / 344026 Al expansion protocol parameters in terms of seeding concentration, ZOL pulse concentrations and frequencies.1

[0165] Weekly CMF analyses (DO, D7, Dll (bioreactor inoculation day of CSC in Ml), D14 and D18) were performed to evaluate expanded cell's purity and quality.

[0166] Thanks to the CSC method, y6 T cells are progressively expanded from DO to Dll (from 1.09 xlO6at DO to 3.22 xlO8cells at Dll and from 1.37xlOsat DO to 6.48x10scells at Dll in RPMI and Ml media, respectively) (Figure 21, A and B). Then, at Dll, cells expanded in CSC - Ml medium were inoculated in an agitated bioreactor (step ill of CSC protocol and according to the example 4 "media preference"), whereas culture in RPMI media is continued in shaken flasks as the control. Contrary to the CSC culture method, a cell quantity decline was observed after D7 under US 2024 / 344026A1 expansion protocol conditions (Figure 21, C). This cell loss triggered also a decrease in y6 T cells fold expansion after D7 (3.12-fold expansion at D7 to 0.64-fold expansion at D14). In CSC method, the y6 T cells specific fold expansion increased, to finally reached 1674-fold expansion in bioreactor (Ml), compared to a 768-fold expansion obtained in flasks (RPMI) at D14. However, comparison between culture conditions was stopped after D14, due to the impossibility to analyze the phenotype of cells cultivated using method of US 2024 / 344026 Al. In fact, the strong decline in cell quantity prevented to properly compare expansion performances. Less than 8% of cells detected by flow cytometry showed the morphologic features specific to lymphocytes, such as granularity and size, meaning that most of the detected events were cells' debris or aggregates. In addition to that, a low viability (less than 48%) in each replicate of culture based on "026 was observed at the end of the experiment. This expansion failure seemed to be only linked to "026 expansion protocol parameters, and not due to donors' variability, since y6 T cells expansion has been successfully obtained in the CSC protocol-based cultures of the present invention.

[0167] CD3+ cells population represented over 50% of the total cells at DO in each tested condition (69.19% in CSC - RPMI, 59.49% in CSC- Ml and 60.07% in "026-based protocol (Figure 22), and progressively increased and finally reached over 86% in RPMI and Ml media in CSC conditions at day 14 (Figure 22, A and B). Whereas CD3+ cells population increased until Dll (65.97%) and then dropped to 25.41% in "026-based protocol on D14. Cells impurities (CD3- cells) percentage evolved in inverse proportion to CD3 + cells proportions. Among CD3+ cells, y6 T cells enrichment has been monitored. Within CD3+ cell population, the y6 T subpopulation has been strongly enriched through the expansion process (CSC protocol, Figure 22, A and B) until D14 (end of the experiments) with more than 81% of y6T cells obtained in the CSC - Ml culture condition. In culture conditions of "026, the proportion CD3+ / y6 TCR+ subpopulation also increased, to finally represent the majority of CD3+ cells (73%, Figure 22, C). However, as the CD3+ cells percentage decreased since Dll, the finalpurity of y6 T cells cultivated on the basis of "026 at D14, was only about 35%, with consequently 65% of cells debris, in contrary to expected results described in "026 (over 96% of y6 T cells expected). Taken together, these results suggested that the CSC method (independently to media) allowed to obtain a significantly higher quantity of y6 T cells, with better purity and a reduced cell impurity, contrary to "026-based protocol.

[0168] The activity and exhaustion markers expression levels of expanded cells have been measured throughout the culture period to evaluate the impact of the agitation protocol (cyclic shaking vs static phase followed by continuous shaking) on cells activity and exhaustion.

[0169] Also, some markers were monitored, associated either with the activation (NKG2D, DNAM1) or exhaustion (TIM3, TIGIT or PD1) of cells. Concerning the exhaustion markers, most of the produced cells are TIM3+ (Figure 23). However, the TIM3 expression profile is associated with different consequences on cell functionality depending on the co-expression of additional markers. The simultaneous expression of DNAM1, NKG2D and TIM3 markers is generally associated with high cytotoxic properties, whereas TIM3 single expression is linked to cell exhaustion. These results are reinforced by activation markers' expression levels. At D14, the majority of the produced y6 T cells were NKG2D+ / DNAM+ for 99.8% in CSC- Ml, 99.6% in CSC RPMI and 99.17% in "026-based cultures, and CD16+ for 95.62% in CSC - RPMI, 93.92% in CSC- Ml and 97.78% in "026-based cultures (Figure 23, D). Triple positive y6 T cells (TIM3+ / NKG2D+ / DNAM1+) represented 99.67%, 99.69% and 100% of the total analyzed cells in Ml - CSC, RPMI-CSC and "026-based cultures, respectively. Taken together, these results suggest that both shaking protocols enable the production of highly active y6 T cells with cytotoxic properties. At the end of the culture (D14), less than 1.30% and 15% of TIM3+ y6 T cells were also PD1+ / TIGIT+ with the CSC method (Ml media and RPMI media, respectively). Based on the method of US 2024 / 344026 Al, no exhausted cells have been detected. However, due to the low number of detected events by flow cytometry, the exhaustion levels of cells cultivated according to "026 could not be properly determined. Taken together, these results suggested that the CSC shaking culture method do not trigger cells' exhaustion (Figure 23, A, B, and C).

[0170] In conclusion, the method of US 2024 / 344026 Al and the CSC expansion protocol allowed both the production and enrichment in activated y6 T cells, with no exhaustion of y6 T cells detected using the CSC culture method. Cyclic Shaking Condition also allowed to amplify poorly exhausted cells of interest. However, y6 T cells fold expansion obtained with both methods could not been compared due to a significant culture scale difference (2 L bioreactor vs 10 mL). In addition, the final purposesof the 2 methods differ. The culture method of US 2024 / 344026 Al aims to produce activated y6 T cells, whereas the CSC method has been developed and optimized to produce large batches of activated y6 T cells, allowing the production of several therapeutic doses for an immunotherapy. Despite a strict replication of the method, the results obtained (6 replicates) seemed not to be in adequation with those described in US 2024 / 344026 Al. Experiments led neither to the expected fold expansion nor purity, as claimed in US 2024 / 344026 Al (over 400-fold expansion). The transition from a static phase to a continuous shaking condition seemed to be sufficient to produce a small quantity of activated y6 T cells starting from a donor buffy coat (method of US 2024 / 344026 Al). In contrast, CSC (cyclic alternation of static and shaken phase) method enabled to expand y6 T cells with a significantly higher fold expansion, a better purity, and without cell loss over time.

[0171] 6. Robust expansion of activated v6 T cells with a low level of exhaustion markers. The method of the invention aims to expand clinical grade y6 T cells intended for the treatment of post-transplant leukaemia or haematological malignancies relapse. To achieve this goal, the culture method needs to specifically amplify from the PBMC the y6 phenotypic subpopulation of T lymphocytes, which is a scarce population in healthy donor's blood (between 0.5 and 5% depending on the individuals). Previous experiments have shown that a selective amplification was obtained with a high enrichment (above 80%) of y6 T cells and, while allowing an efficient depletion (below 15%) of the aPT cell subpopulation responsible for the GvH reaction. Also, a selective amplification should produce a high quantity of y6T cells allowing a culture of y6T cells in controlled bioreactors. To confirm the efficiency and robustness of the CSC method despite biological variability of donors, the shaking process was carried out using three independent buffy coats, but with the selected serum-free GMP medium (Ml). The following experiments allowed to establish a reference growth kinetic in medium Ml (Figure 24), which will be used to perform further replicates in terms of cell sampling and phenotyping using flow cytometry, or in terms of feeding and cell harvest timing. Cells were cultivated following the CSC method until D14, in a controlled bioreactor (2 L) as the reference culture for further step of a scale-up (5-10 L) for two replicates. One additional replicate culture was cultivated until D18 to allow the extrapolation of culture expansion and purification.

[0172] For this purpose, PBMC were isolated using a density gradient, treated with Zometa (zoledronic acid) and cultivated according to the CSC method (with intermittent agitation in shaking flasks). An initial flow cytometry analysis was performed at DO to evaluate surface markers. Weekly flow cytometry analyses at DO, D7 and Dll (bioreactor inoculation day of CSC in Ml), D14 and D18) wereperformed to monitor the expanded cell purity and quality. In addition, complete culture medium (fed batch mode) was added at regular time intervals starting from D4, depending on cell growth, and residual glucose and glutamine concentrations in the culture media, to maintain a sufficient level of nutrients in the bioreactors and an optimal Viable Cells Density (VCD).

[0173] From DO, the initial y6 T cells quantity was around 1.36x10scells and has been progressively amplified and reached an average of 2.28xl09cells at D14 (Figure 24). This significant and specific expansion of y6 T cells allowed to reach a 1761-fold expansion on day D14, demonstrating the success of CSC protocol performed in Ml medium and in a controlled 2 L bioreactor.

[0174] The flow cytometry analyses during the process allowed to monitor the enrichment in y6 T cells, and conversely the ap T cells depletion. Total CD3+ cells increase from 79.25% (DO) to over 85% on D14 (Figure 25). This enrichment was improved, as shown by results obtained with the replicated culture performed until D18 (more than 98% of CD3+ cells). Within this CD3+ population, the expected progressive enrichment in y6 T cells was achieved (0.76% at DO vs over 86% y6 T cells at D14), with a strong depletion of a T cells (from 76.8% at DO to 10.78% at D14) (Figure 25).

[0175] Flow cytometry phenotyping data, allowed to determine subpopulation percentage (Figure 26).

[0176] Thus, the ap T cells (67.3% at DO) have been strongly depleted, (10.5% ap T cells at D14), while y6 T cells have been enriched (from 2.4% to 77.2%) to reach a final percentage of 89.4% at D18 (Figure 26, D). Other cell subpopulations were also monitored through flow cytometry. NK cells (CD3- / CD56+) represented initially 8.1% of the total cells. This percentage dropped progressively to 3.6% at D18 (Figure 26). Other CD3- cells, such as monocytes, were also efficiently depleted (22.1% at DO vs 0.6% at D14). The specific subpopulation of y6 T cells (especially V62y9 phenotype) reached a value over 83% for V62y9 T cells at D18 (data not shown). Thus, the final expected purity in cells of interest (more than 80% of y6 T cells and less than 15% ap T cells) was successfully achieved, demonstrating the robustness of CSC expansion method despite biological variability of donors. The CSC expansion method has been developed not only to specifically amplify y6 T cells, but also to ensure reduced cells' exhaustion, activation of cytotoxic properties and the maintenance of the cytotoxic potential. Regarding the exhaustion markers (Figure 27, A) PD1, TIGIT and TIM3 markers were monitored. The ratio of y6 T cells PD1+ cells (23.61% on DO) decreased to 2.72% on D14. The same trend was observed with the percentage of TIGIT+ cells, that dropped to 28.66% on D14 (and 14.99% on D18 for N=l).

[0177] As previously shown, most of the expanded cells were TIM3+ (Figure 27, A). However, the TIM3 expression pattern is associated with different consequences on cell functionality, depending onthe co-expression of additional markers. The simultaneous expression of DNAM1, NKG2D and TIM3 markers is associated with high cytotoxic properties, whereas TIM3 single expression is linked to cell exhaustion. At D14, the majority of produced y6 T cells were NKG2D+ / DNAM+ (99.8%) and CD16+ (91.40%) (Figure 27, B). Thus, TIM3, DNAM-1 and NKGD2 co-expression on amplified y6 T cells indicated that the expanded cells displayed the characteristics of cytotoxic properties. This profile was also observed later on D18 (99.52% of NKG2D+ / DNAM1+). Taken together, these results suggest that the cyclic shaking protocol enabled the production of highly active y6 T cells with cytotoxic properties, even at a bioreactor scale, according to their surface markers expression. A low level of exhaustion markers, as well as the presence of activation markers, is associated with an optimal y6 T cells cytotoxic activity. These results suggested that both Cyclic Shaking Condition (CSC) and the absence of human serum (with serum free Ml medium) do not have any negative influence on cell activity and functionality, besides these conditions promote a strong y6 T cells' expansion, independently from the biological variability associated with buffy coat donors. In conclusion, the CSC culture method in a serum-free GMP medium enable a robust expansion, resulting in an important quantity of y6 T cells (V62y9 T cells subpopulation of interest mainly), sufficient to produce large batches of therapeutic doses.

Claims

CLAIMS1. A method for producing gamma delta (y6) T cells comprising:- a step (i) wherein mononuclear cells are incubated in presence of at least one activating compound selected from the group consisting of a phosphoantigen and a bisphosphonate, and- a step (ii) wherein the mononuclear cells are cultured in a liquid culture medium comprising at least interleukin 2 (IL-2),wherein step (ii) begins with a static phase and alternates static and agitated phases during at least 7 days.

2. The method according to claim 1, wherein step (ii) cyclically alternates static and agitated phases.

3. The method according to claim 1 or 2, wherein step (ii) comprises at least one static phase and one agitated phase per day, preferably only one static phase and only one agitated phase per day during the first 7 days of culture.

4. The method according to any one of claims 1 to 3, wherein each static phase lasts 23 hours and each agitated phase lasts 1 hour.

5. The method according to any one of claims 1 to 4, wherein step (ii) begins with a static phase of 24 hours.

6. The method according to any one of claims 1 to 5, wherein step (ii) comprises:- a static phase of 24 hours, followed by- a cycle of an agitated phase of 1 hour and a static phase of 23 hours during the next 7 days, optionally followed by- an agitated phase during the next 14 days.

7. The method according to any one of claims 1 to 6, wherein the agitated phases during step (ii) are carried out with an orbital shaker, a rocking shaker or a rotator shaker.

8. The method according to any one of claims 1 to 7, wherein the mononuclear cells have been obtained on peripheral blood, in particular by leukapheresis or buffy coat.

9. The method according to any one of claims 1 to 8, wherein the activating compound is a phosphoantigen selected from the group comprising isopentenyl pyrophosphate (IPP), ethyl pyrophosphate (EtPP), hydroxy-methyl-butyl-pyrophosphate (HMBPP), more preferably IPP.

10. The method according to any one of claims 1 to 9, wherein the activating compound is a bisphosphonate, preferably zoledronic acid (Zometa).

11. The method according to any one of claims 1 to 10, wherein, during step (ii), the cells are maintained at a cell density not exceeding 3 x 106cells / ml.

12. The method according to any one of claims 1 to 11, further comprising:- a step (iii), wherein the cells obtained from step (ii) are expanded in an agitated bioreactor under controlled conditions.

13. The method according to any one of claims 1 to 12, further comprising after step (iii):- a step (iv), wherein the y6 T cells are recovered, and optionally sorted.

14. y6 T cells obtained by the method as defined in any one of claims 1 to 13.

15. y6 T cells obtained by the method as defined in any one of claims 1 to 13, for use in an immunotherapy, preferably in the treatment or prevention of a cancer.

Images