Increasing NKT cell and INKT cell-mediated cancer cell killing

Abstract

The invention relates to a pharmaceutical composition for use as a medicament, for example in the treatment of cancer and infectious diseases, in a subject in need thereof. The composition comprises isolated immune cells and modified aPD-L1 antibodies having an enhanced affinity for FcgammaRIIIa, and can further include dendritic cells. The invention further provides a method for expanding and activating NKT cells and iNKT cells or their CAR- derivatives.

Description

[0001] P3685PC00

[0002] TITLE: Increasing NKT cell and iNKT cell-mediated cancer cell killing FIELD OF THE INVENTION

[0003] The invention relates to a pharmaceutical composition for use as a medicament, for example in the treatment of cancer and infectious diseases, in a subject in need thereof. The composition comprises isolated immune cells and modified aPD-Ll antibodies having an enhanced CD16 affinity, where the composition can further include dendritic cells. The invention further provides a method for expanding and activating NKT cells and iNKT cells or their CAR- derivatives.

[0004] BACKGROUND OF THE INVENTION NKT cells can play a vital role in the upregulation of the immune system and suppression of the tumor microenvironment (TME) in the course of disease. In addition, NKT cells have increased cytolytic activity with good infiltration ability into TME, which causes the tumor cell targeting to be powerful. NKT cells have brilliant features that make them very useful cells against tumors, playing a role in the rapid detection and eradication of tumors, and the elimination of cancer cells in early cell growth. The CAR structure, when added to the NKT or iNKT cell, creates a powerful cell with additional cytotoxic features against cancers. However, there exists a barrier to exploiting the full potential of NKT cells and iNKT cells, for example that their frequency in blood is very low, where they are largely present as immature NKTs.

[0005] The present invention addresses this problem, by providing the tools needed to better exploit the potential of NKT cells and iNKT cells and their CAR- derivatives.

[0006] SUMMARY OF THE INVENTION

[0007] In a first embodiment, the invention provides a pharmaceutical composition for use as a medicament, comprising immune cells selected from the group consisting of NKT cells, iNKT cells, CAR-NKT cells and CAR-iNKT cells, and modified aPD-Ll antibodies and optionally PD-Ll+dendritic cells,

[0008] wherein said modified aPD-Ll antibodies are either:

[0009] at least partially non-fucosylated antibodies comprising 0 - 25% core-fucosylated N-glycans, or

[0010] mutant aPD-Ll antibodies, wherein two or more amino acid residues in a heavy chain of said mutant aPD-Ll antibodies are substituted, and wherein said substitutions are selected from a group consisting of:

[0011] a. S298A, E333A and K334A;

[0012] b. S239D and I332E;

[0013] c. S239D, A330L and I332E;P3685PC00

[0014] d. G236A, S239D and I332E;

[0015] e. G236A, A330L and I332E;

[0016] f. G236A, S239D, A330L, and I332E;

[0017] g. F243L, R292P, Y300L, V305I, P396L;

[0018] h. L235V, F243L, R292P, Y300L, P396L; and

[0019] i. P247I and A339Q,

[0020] wherein the heavy chain of said mutant aPD-Ll antibodies comprises an amino acid sequence derivable from a parent human IgGl heavy chain by said substitutions, wherein the corresponding amino acid sequence in said parent human IgGl heavy chain has at least 70% sequence identity to SEQ ID NO.: 10 or 12, and

[0021] wherein the substitutions are numbered relative to the first amino acid of the CHI region assigned position 118 of said heavy chain; and wherein the constant domains of said modified aPD-Ll antibodies preferably originate from a human isotype IgGl antibody.

[0022] In a second embodiment, the invention provides a method of activating and expanding immune cells selected from the group consisting of NKT cells, iNKT cells, CAR-NKT cells and CAR-iNKT cells, comprising the steps of:

[0023] i. contacting said immune cells with modified aPD-Ll antibodies and PD-Ll+dendritic cells in vitro or ex vivo; and

[0024] II. optionally isolating said immune cells obtained in step (i),

[0025] wherein said modified aPD-Ll antibodies are either:

[0026] at least partially non-fucosylated antibodies comprising 0 - 25% core-fucosylated N-glycans,

[0027] or

[0028] mutant aPD-Ll antibodies, wherein two or more amino acid residues in a heavy chain of said mutant aPD-Ll antibodies are substituted, and wherein said substitutions are selected from a group consisting of:

[0029] a. S298A, E333A and K334A;

[0030] b. S239D and I332E;

[0031] c. S239D, A330L and I332E;

[0032] d. G236A, S239D and I332E;

[0033] e. G236A, A330L and I332E;

[0034] f. G236A, S239D, A330L, and I332E;

[0035] g. F243L, R292P, Y300L, V305I, P396L;

[0036] h. L235V, F243L, R292P, Y300L, P396L; and

[0037] i. P247I and A339Q,P3685PC00

[0038] wherein the heavy chain of said mutant aPD-Ll antibodies comprises an amino acid sequence derivable from a parent human IgGl heavy chain by said substitutions, wherein the corresponding amino acid sequence in said parent human IgGl heavy chain has at least 70% sequence identity to SEQ ID NO.: 10 or 12, and

[0039] wherein said substitutions are numbered relative to the first amino acid of the CHI region assigned position 118 of said heavy chain.

[0040] In a third embodiment, the invention provides a method of treating a subject in need thereof with a pharmaceutical composition according to the first embodiment.

[0041] In a fourth embodiment, the invention provides a method of treating a subject for cancer or an infectious disease comprising the steps of:

[0042] (a) providing an isolated preparation of pre-treated immune cells selected from the group consisting of NKT cells, iNKT cells, CAR-NKT cells and CAR-iNKT cells,

[0043] (b) administering said pre-treated immune cells to said subject,

[0044] wherein said pre-treated immune cells are obtained by culturing immune cells selected from the group consisting of NKT cells, iNKT cells, CAR-NKT cells and CAR-iNKT cells with modified aPD-Ll antibodies and PD-Ll+dendritic cells in vitro or ex vivo,

[0045] wherein said modified aPD-Ll antibodies are either:

[0046] at least partially non-fucosylated antibodies comprising 0 - 25% core- fucosylated N-glycans, or

[0047] mutant aPD-Ll antibodies, wherein two or more amino acid residues in a heavy chain of said mutant aPD-Ll antibodies are substituted, and wherein said substitutions are selected from among:

[0048] a. S298A, E333A and K334A;

[0049] b. S239D and I332E;

[0050] c. S239D, A330L and I332E;

[0051] d. G236A, S239D and I332E;

[0052] e. G236A, A330L and I332E;

[0053] f. G236A, S239D, A330L, and I332E;

[0054] g. F243L, R292P, Y300L, V305I, P396L;

[0055] h. L235V, F243L, R292P, Y300L, P396L; and

[0056] i. P247I and A339Q,

[0057] wherein the heavy chain of said mutant aPD-Ll antibodies comprises an amino acid sequence derivable from a parent human IgGl heavy chain by said substitutions, wherein the corresponding amino acid sequence in said parent human IgGl heavy chain has at least 70%P3685PC00

[0058] sequence identity to SEQ ID NO.: 10 or 12, and

[0059] wherein the substitutions are numbered relative to the first amino acid of the CHI region assigned position 118 of said heavy chain; and wherein the constant domains of said modified aPD-Ll antibodies preferably originate from a human isotype IgGl antibody.

[0060] In a fifth embodiment, the invention provides a method of treating a subject for cancer or an infectious disease comprising the steps of:

[0061] (a) culturing immune cells selected from the group consisting of NKT cells, iNKT cells, CAR-NKT cells and CAR-iNKT cells with modified aPD-Ll antibodies and dendritic cells in vitro or ex vivo,

[0062] (b) isolating pre-treated immune cells obtained in step (a),

[0063] (c) administering said pre-treated immune cells to said subject

[0064] wherein said modified aPD-Ll antibodies are either:

[0065] at least partially non-fucosylated antibodies comprising 0 - 25% core- fucosylated N-glycans, or

[0066] mutant aPD-Ll antibodies, wherein two or more amino acid residues in a heavy chain of said mutant aPD-Ll antibodies are substituted, and wherein said substitutions are selected from among:

[0067] a. S298A, E333A and K334A;

[0068] b. S239D and I332E;

[0069] c. S239D, A330L and I332E;

[0070] d. G236A, S239D and I332E;

[0071] e. G236A, A330L and I332E;

[0072] f. G236A, S239D, A330L, and I332E;

[0073] g. F243L, R292P, Y300L, V305I, P396L;

[0074] h. L235V, F243L, R292P, Y300L, P396L; and

[0075] i. P247I and A339Q,

[0076] wherein the heavy chain of said mutant aPD-Ll antibodies comprises an amino acid sequence derivable from a parent human IgGl heavy chain by said substitutions, wherein the corresponding amino acid sequence in said parent human IgGl heavy chain has at least 70% sequence identity to SEQ ID NO.: 10 or 12, and

[0077] wherein the substitutions are numbered relative to the first amino acid of the CHI region assigned position 118 of said heavy chain; and wherein the constant domains of said modified aPD-Ll antibodies preferably originate from a human isotype IgGl antibody.P3685PC00

[0078] DESCRIPTION OF THE INVENTION

[0079] Brief description of the figures:

[0080] Figure 1. An image of raw data presented as dotplots of the abundance of CD3+CD56+ NKT cells as well as CD3-CD56+ (NK cells) and CD3+ CD56- (T cells) after 7 days of coculture of PMBCs with MUTZ-3-derived mDCs at a PBMC:DC ratio of 10: 1 and indicated aPD-Ll treatments, assesssed by flow cytometry. The data is from one representative experiment.

[0081] Figure 2: Graph showing the cell expansion of NKT cells after 7 days of coculture between PBMCs, MUTZ-3-derived mDCs at a PBMC:DC ratio of 10: 1 and indicated treatments. Error bars indicate + / - SEM and statistical analysis was performed by one-way-ANOVA with Tukey's multiple comparisons test (* p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0001). A. aPD-Ll treatments (n=10 independent cocultures). B. aHER2 treatments (n=2 independent cocultures.)

[0082] Figure 3: Graphs showing the percentage of NKT cells having the activation markers CD25, CD69 and CD137 after 5 and 7 days of coculture with MUTZ-3-derived mDCs in the presence of aPD-LlGE (n=2). A basal level of activation markers of 0% was set on day 0.

[0083] Figure 4: Graph showing the fold change (Y axis) in NKT cells after 7 days of coculture of CD3+CD16-cells (black) or CD3+cells (white) and MUTZ-3-derived mDCs in the presence of no antibody (-), aPD-LlNG, aPD-LlwT and aPD-LlGE. Fold change is normalized to untreated (-) cocultures. Error bars indicate + SD and statistical analysis was performed by two-way ANOVA with Sidaks's multiple comparisons test (n=2) (** p < 0.01).

[0084] Figure 5: Graph showing percent of sorted PD-L1- (black) and sorted PD-L1+ (white) MUTZ-3 DCs exhibiting PD-L1 surface expression after 3 days maturation with standard maturation protocol in the presence of no antibody (-), aPD-LlNG, aPD-Llwror aPD-LlGE, determined by flow cytometry. Error bars indicate + SD, and asterisks indicate level of statistically significant differences between samples evaluated by one-way ANOVA using Tukey's correction for multiple comparisons (n=2).

[0085] Figure 6: Graph showing the fold change in NKT cells after 7 days of coculture of PBMCs and PD-L1+ (white) or PD-L1- (black) MUTZ-3 mDCs in the presence of no antibody (-), aPD-LlNG, aPD-LlwT or aPD-LlGE. Fold change is normalized to untreated (-)P3685PC00

[0086] cocultures. Error bars indicate + SD and statistical analysis was performed by two-way ANOVA with Sidaks's multiple comparisons test (n=3) (** p < 0.01).

[0087] Figure 7: Graph showing Area Under the Curve (AUC) analysis (based on impedance-derived Cell Index (CI) measurements from an xCELLigence instrument as a measure of the killing of adherent DU-145 cancer cells after treatment with PBMCs preactivated by co-culture with MUTZ-3 mDCs in the presence of either no antibody (untreated), aPD-L1 NG, aPD-Ll WT, or aPD-Ll GE. Increased cell attachment results in higher impedance, reflected as an increased CI, while dead or detached cells cause a decrease in CI and lower AUC. Preactivated PBMCs were added 25 hours after DU-145 cell seeding and the killing was measured for 25h. Error bars indicate + SD, and asterisks indicate level of statistically significant differences between samples evaluated by one-way ANOVA using Tukey's correction for multiple comparisons (n=4, * p < 0.05, ** p < 0.01).

[0088] Figure 8: A. Representative images of 3D cancer spheroids cultured in RMPI medium for lOh before treatment and up to ~20h after treatment with PBMCs preactivated by coculture with MUTZ-3 mDCs in the presence of either no antibody (PMBC), aPD-Ll NG, aPD-Ll WT, or aPD-Ll GE, or no treatment (RMPI). B. % cytolysis of spheroids based on red integrated intensity (RI x pm) and C. % cytolysis of spheroids based on total red area / image (pm2 / well) within defined mask. Each treatment is the average value of 6-10 spheroids and graphs shows the first 25h after addition of the preactivated PBMCs as effector cells, which were added after 50h.

[0089] Figure 9: A cartoon illustrating alternative bispecific aPD-LlxaHER2 antibody formats. The bsAbs targeting PD-L1 and HER2 were generated by fusing aPD-Ll sdAb and aHER2 sdAb onto the aHER2 IgGl and aPD-Ll IgGl scaffolds, respectively. The sdAbs were fused N-terminally (NHC / NLC)) or C-terminally (CHC / CLC) on both HC and LC through a flexible (GGGGS)x2 linker.

[0090] Figure 10: A cartoon illustrating alternative domain structure of CAR proteins encoded by respective constructs, for use in CAR derivatives of NKT cells and iNKT cells.

[0091] Figure 11: A bar diagram showing the CD16A KD (molar) + / -SD binding affinities of aPD-LlGE, aPD-LlwT and aHER2GE antibodies (n=3), as determined using the Octet RED96e system (ForteBio).

[0092] Figure 12: Graph showing the glycoprofile of the isoated antibodies: aPD-Ll heavy chain-non-glycosylated HC rituximab fusion (aPD-LlNG); the aPD-Ll heavy chain-WTP3685PC00

[0093] HC rituximab fusion (aPD-LlwT)and the aPD-Ll heavy chain- non fucosylated HC rituximab fusion (aPD-LlGE). Blue square - N-acetylglucosamine (GIcNAc); Dark green circle - Mannose (Man); Light yellow circle - Galactose (Gal); A Red triangle - Fucose (Fuc).

[0094] Figure 13: Bar diagram showing the percent PD-L1+ dendritic cells present in populations of MUTZ-3 iDCs, MUTZ-3 mDCs and moDCs.

[0095] Figure 14: Graph showing fold increase in CD25 expression on CD4+and CD8+cells following 5 days co-culture with either 1 pg / mL of aPD-Llwr, aPD-LlNG, or aPD-LlGE, or relative to untreated co-culture. CD4+cells were defined as CD3+CD8“ cells.

[0096] Abbreviations, terms and definitions:

[0097] aPD-Ll is an antibody that targets PD-L1 (also called anti-PD-Ll or anti-PDLl or aPDLl)

[0098] APC is an antigen-presenting cell

[0099] CAR is a chimeric antigen receptor

[0100] CD16 is a type I transmembrane receptor, also known as FcgammaRIIIa or FcyRllla DPBS is Dulbecco's Phosphate-Buffered Saline

[0101] FACS is Fluorescence-Activated Cell Sorting

[0102] FcR Blocking Reagent is a reagent for blocking unwanted binding of antibodies to human Fc receptor-expressing cells such as B cells, monocytes, and macrophages. FBS is Fetal Bovine Serum

[0103] GM-CSF is Granulocyte-macrophage colony-stimulating factor

[0104] Human isotype IgGl (human IgGl) is a member of the five immunoglobulin isotypes, immunoglobulin G (IgG) most abundant in human serum.

[0105] iDC is an immature dendritic cell such as a MUTZ-3 iDC

[0106] iNKT cells are Invariant Natural Killer T cells

[0107] mDC is a mature dendritic cell such as a MUTZ-3 mDC

[0108] MEMa is Minimum Essential Medium Eagle

[0109] moDC is a monocyte-derived dendritic cellP3685PC00

[0110] MUTZ-3 cells are cells derived from a human cytokine-dependent myeloid cell line.

[0111] NKT cells are CD3+CD56+, herein defined as Natural Killer T cells (in accordance with Sweiboda et al., 2024).

[0112] PBMCs are Peripheral Blood Mononuclear Cells

[0113] PBS is Phosphate-Buffered Saline

[0114] PD-L1 is Programmed Death Ligand 1

[0115] Detailed description of the invention:

[0116] NKT and iNKT cells, as defined herein, are known to be cells able to kill cancer cells and thus potential candidates for development as cancer cell therapies. The advantageous properties of these cells include their ability to combine direct tumor cell killing, reprogramming of the tumor microenvironment, and promoting systemic immune responses to create a more immunogenic tumor microenvironment. However, the low frequency of NKT and iNKT cells in the mammalian T lymphocyte population, among others, presents a barrier to exploiting their therapeutic potential. The present invention addresses this problem by delivering a technology and molecules needed to achieve the therapeutic potential of both NKT cells, iNKT cells and their CAR- derivatives with respect to efficient tumor cell killing and improved tumor therapy.

[0117] The present invention is based on the surprising observation that modified IgGl-based anti-PD-Ll antibodies having an enhanced CD16 affinity, can significantly increase the activation and expansion of both NKT and iNKT cells, that are a subset of cytotoxic cells. Furthermore, it was surprisingly observed that said activation and expansion of said cells is both mediated through dendritic cells, and through binding to PD-L1 on dendritic cells and to CD16 on NKT and iNKT cells. Once activated, the cells provide an efficient and largely increased tumor cell killing, as evidenced by in vitro 2D and 3D fully human models and detailed cellular immunological studies, disclosed herein.

[0118] I: A pharmaceutical composition for use as a medicament

[0119] In a first embodiment, the invention provides a pharmaceutical composition for use as a medicament. The composition comprises immune cells selected from the group consisting of NKT cells, iNKT cells, CAR-NKT cells and CAR-iNKT cells, as well as modified aPD-Ll antibodies having an enhanced CD16 affinity. For reasons clarified in section V, the inclusion of PD-Ll+dendritic cells in the pharmaceutical composition is optional.P3685PC00

[0120] The immune cells of the composition are natural killer T cells selected from the group consisting of NKT cells, iNKT cells, CAR-NKT cells and CAR-iNKT cells. NKT cells, iNKT cells and their CAR- derivatives may be selected from the group consisting of autologous-, allogenic-, stem cell derived-, and cell line derived-cells. Preferably the composition comprises isolated forms of said natural killer T cells.

[0121] I.i NKT and iNKT cells

[0122] NKT cells-, (as defined herein) are phenotypically identifiable by their expression of the surface markers CD3+CD56+. NKT cells are present in samples of PBMCs and can be isolated from PBMCs by methods well known in the art, such as using a CD3+CD56+ NKT Cell Isolation Kit for the sequential separation of CD3+CD56+ natural killer (NK) T cells from human PBMCs (Miltenyi Biotec #130-093-064). The PBMCs can be isolated from a subject by methods well known in the art. Suitable methods include isolating PBMCs from buffy coats by a two-step density gradient centrifugation (200xg and 460xg) using Histopaque-1077 (Sigma Aldrich, #10771) as described in example 3.

[0123] INKT cells-, are Invariant Natural Killer T cells, also known as type I or classical NKT cells. These cells belong to a distinct population of T cells that express an invariant ap T-cell receptor (TCR) which is a heavily biased, invariant T cell receptor (TCR Va24-Jal8, preferentially coupled with a vpil TCR chain in humans) conferring specificity for glycolipid antigens, presented by CDld molecule. iNKT cells can be distinguished from NKT cells by the recognition of the prototypical glycolipid, alpha-galactosylceramide (alpha-GalCer). iNKT cells can be isolated from PBMCs by methods well known in the art, such as staining with 6B11 mAb and Anti-Human CD3 FITC and selection by flow cytometry (in accordance with the manufacturer: Thermofisher using products: #12-5806-42 and # 11-0038-42 respectively) or as described by Lenart et al., (2017).

[0124] I.ii CAR-NKT cells and CAR-INKT'. are NKT or iNKT cells that have been equipped with a chimeric antigen receptor (CAR) via genetic engineering, as illustrated in example 6. The CAR can be engineered to target specific antigens on tumor cells, for use in cancer therapy (Peng et al, 2024). NKT cells are cytotoxic immune cells that bridge the innate and adaptive immune systems by producing different biological factors (such as IL-2 and IL-21) and are considered potent cytotoxic cells against tumors, while having no notable side effects on normal cells. For this reason, NKT cells are potent candidates for CAR-immune cell therapy against tumors. CAR can enable the recognition of a broad range of antigens like glycosylated proteins, glycolipids, carbohydrates, protein or peptide epitopes, discontinuous or continuous epitopes, or conformational epitopes relevant for tumor targeting. The CAR structure consists of three parts from the outside to the inside: the ectodomain, the transmembrane domain, and the endodomain.P3685PC00

[0125] The ectodomain part is responsible for antigen recognition mediated by an antigen binding domain, preferably an antibody fragment such as a single-chain antibody comprising the specificity mediating variable domains derived from an antibody (ScFv) or a single domain antibody (sdAb), such as a nanobody or a humanized nanobody or a human sdAb generated from synthetic libraries with human or humanized frameworks. Many types of CARs, and methods for their production are known to those skilled in the art.

[0126] For example, ScFv can be connected to the transmembrane part by a spacer (hinge) region and is known as the most common structure for the ectodomain part. CARs are classified into first (ScFv CD3z only) lacking costimulatory domains, second (one costimulatory domain + CD3z), or third generation CARs (more than one costimulatory domain + CD3z) such as the co-stimulators: 4-1BB, CD27, or 0X40. A fourth generation of CAR, known as "T-cells redirected towards universal cytokine killing" (TRUCKS), are engineered by transducing transgenes for cytokine secretion (e.g., IL-12) or other biological factors, and include a nuclear factor of activated T cells (NFAT) that enables the cell to secrete cytokines. The ScFv part that defines the CAR antigen binding specificity can be derived from antibody clones that recognize a target specific antigen on tumor cells or generated by other technologies such as phage display, technologies well known by those skilled in the art.

[0127] The spacer may be designed to have an immunoglobulin (Ig)-like domain hinge, such as a CH3 domain, and the spacer length may be optimized to access the location of the targeted antigen.

[0128] The transmembrane domain preferably comprises a hydrophobic a helix that spans the cell membrane, for example transmembrane domains derived from CD4, CD8a, or CD28 receptors.

[0129] CAR endodomains, or intracellular domains, can be derived from costimulatory molecules such as from the CD28 family (including CD28 and ICOS) or the tumor necrosis factor receptor (TNFR) family of genes (including 4-1BB, 0X40, or CD27). Viral- (e.g. gamma retrovirus or lentivirus vector) and non-viral-based (e.g. a transposon mediated system) genetic engineering tools may be used to transfer and permanently integrate CAR encoding constructs into NKT or iNKT cells (Peng et al, 2024).

[0130] I.iii The modified aPD-Ll antibodies of the composition

[0131] The modified aPD-Ll antibodies according to the invention are human IgGl constant domain-based antibodies comprising a variable domain (or variable region) capable of binding to PD-L1, and an IgGl-based Fc domain having enhanced affinity to CD16. Said modified aPD-Ll antibody can comprise a domain (or variable region) capable of binding to PD-L1 that is linked to either a glycoengineered Fc (having a reduced content of core-fucose, also called alpha-1, 6-core fucose) or to a mutated Fc domain, wherein each saidP3685PC00

[0132] Fc domain is characterized by an increased affinity to CD16 as compared to the Fc of a parent unmodified human IgGl-based antibody.

[0133] The domain (or variable region) capable of binding to PD-L1 can be selected from among:

[0134] (i) a "normal" binding region where the variable region of the light chain is fused to the constant region of the light chain and the variable region of the heavy chain is fused to the CHI constant region of the heavy chain domain, where the fusion corresponds to a Fab region;

[0135] (ii) a single domain antibody (sdAb), e.g. a nanobody or a humanized nanobody or a human or humanized single domain antibody which is linked, preferably to the hinge region or another linker, and thereby to the IgGl-based Fc region.

[0136] (iii) A single-chain antibody (scFv), whereby the variable light chain and the variable heavy chain are linked to each other through a linker and then linked, preferably through the hinge region or another linker, to the IgGl-based Fc region.

[0137] In one embodiment, said modified aPD-Ll antibodies having an enhanced CD16 affinity are at least partially non-fucosylated aPD-Ll antibodies preferably comprising 0 - 80% core-fucosylated N-glycans. In this embodiment, the aPD-Ll antibodies of the composition are glycosylated with complex biantennary N-glycans that are N-linked to their Fc fragment, and comprise a mannosyl-chitobiose core (Man3GlcNAc2-Asn and / or GlcNAcMan3GlcNAc2-Asn). Each aPD-Ll antibody comprises two N-linked oligosaccharides. The term 0 - 80% core-fucosylated aPD-Ll antibodies refers to the percentage of aPD-Ll antibodies having an L- fucose (Fuc) residue a-1.6-linked to the N-acetylglucosamine in the innermost core of the complex biantennary N-glycans linked to their Fc fragment.

[0138] The term "partially fucose-reduced" refers to a reduction in aPD-Ll antibodies having a-1,6-core fucose attached to the first N-acetylglucosamine (GIcNac) within of any type of complex biantennary N-glycan: with or without bisecting GIcNac, with or without galactose, and with or without sialic acid, which is bound to each conserved amino acid asparagine N297 in the two CH2 domains of the Fc region of the IgGl-based antibody. This term may also be used interchangeably with the term "fucose-reduced" or "essentially lacking core-fucosylation". A fucose-reduced antibody may also be seen in view of the invention as a glyco-optimized antibody; characterized by enhanced binding to CD16 as compared to a glycosylated aPD-Ll antibody having more than 80 % core-fucosylation in the CH2 domain of the Fc region of the antibody.

[0139] The term "core fucosylated N-glycans" refers to N-glycans of a plurality of antibodies, which are core fucosylated. The plurality of partially non-fucosylated aPD-Ll antibodies of the composition may contain 0%, 1 %, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 11%, 12%, 13%, 14%, 15%, 16%, 17%, 18%, 19%, 20%, 21 %, 22%, 23%, 24%,P3685PC00

[0140] 25%, 26%, 27%, 28%, 29%, 30%, 40%, 41 %, 42%, 43%, 44%, 45.0%, 46%, 47%, 48%, 49%, 50%, 51 %, 52%, 53%, 54%, 55%, 56%, 57%, 58%, 59%, 60%, 61.0%, 62%, 63%, 64%, 65%, 66%, 67%, 68%, 69%, 70%, 71 %, 72%, 73%, 74%, 75%, 76%, 77%, 78%, 79%, or even 80% core fucosylated N-glycans.

[0141] Preferably, the plurality of partially non-fucosylated aPD-Ll antibodies contain from 0% to 25% core fucosylated N-glycans, more preferably from 0% to 20% core fucosylated N-glycans, for example any one of 0%, 1 %, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 11%, 12%, 13%, 14%, 15%, 16%, 17%, 18%, 19%, 20% core fucosylated N-glycans. The binding affinity of the Fc region of a population of aPD-Ll antibodies towards CD16 increases progressively as the level of core fucosylated N-glycans is reduced from 100% down to about 25%.

[0142] The molar amount of core fucosylated N-glycans relative to the molecular amount of total N-glycans of a plurality of aPD-Ll antibodies may be from 0% to 80 %, preferably from 0% to 25%, more preferably from 0% to 20%.

[0143] The term "essentially lacking core-fucosylation" may be used for aPD-Ll antibodies having N- linked sugar chains bound to the Fc region, wherein among the total complex N-linked sugar chains bound to their Fc region, the content of a-1 ,6-core-fucose may be from 0% to 80%, preferably from 0% to 25%, more preferably from 0% to 20%, most preferably from 0% to 10%. Such aPD-Ll antibodies having from 0% to 25% core-fucosylation are characterized by a CD16A affinity for both low- and high-affinity FcyRIIIA allotypes that is equivalent to that of aPD-LlGE antibodies (Examples 7 and 11).

[0144] In one embodiment, the non-fucosylated aPD-Ll antibody is a monospecific aPD-Ll antibody capable of binding to PD-L1, that comprises a heavy- and a light-polypeptide chain, wherein at least one of said polypeptide chains has at least 50 % sequence identity to SEQ ID NO.: 3 and / or 5 respectively. Said antibody capable of binding to PD-L1 may comprise a heavy chain to PD-L1, that has at least 50 %, 55 %, 60 %, 65 %, 70 %, 75 %, 80 %, 85 %, 90 %, 95 %, 96 %, 97 %, 98 %, or at least 99 % sequence identity to SEQ ID NO.: 3 and a light chain having at least 50 %, 55 %, 60 %, 65 %, 70 %, 75 %, 80 %, 85 %, 90 %, 95 %, 96 %, 97 %, 98 %, or at least 99 % sequence identity to SEQ ID NO.: 5. Furthermore, said antibody capable of binding to PD-L1 may be composed of a heavy chain and a light chain, wherein the amino acid sequence of heavy chain is SEQ ID NO.: 3 and the amino acid sequence of the light chain is SEQ ID NO.: 5. Alternatively, the non-fucosylated aPD-Ll antibody is a monospecific aPD-Ll antibody capable of binding to PD-L1, that comprises one or two single domain polypeptide chains, wherein each said polypeptide chain has at least 50 % sequence identity to SEQ ID NO.: 3 and / or 5 respectively.P3685PC00

[0145] One embodiment of a non-fucosylated aPD-Ll antibody (herein named aPD-LlGE) composed of heavy and light chains whose amino acid sequence is SEQ ID NO.: 3 and 5 respectively, is preferably obtainable from a mammalian cell line which

[0146] has a reduced core-fucosylation, preferably not expressing FUT8, GMD or GFUS enzymes. Preferably the cell line is selected from among CHO-K1 FUT8 KO cell line (SC-RT2707, Creative biogene); HEK293 FUT8 KO cell line (YKO-XN88623, UBIGENE); HEK293 GMDS KO (EDJ-KQ4727, EDITGENE) and HEK293 GFUS KO (EDJ-KQ5972, EDITGENE), NM-H9D8-E6 (DSM ACC 2807), NM-H9D8-E6Q12 (DSM ACC 2856), or a cell or cell line derived therefrom. Gene sequences encoding the non-fucosylated aPD-Ll antibody can be recombinantly expressed in a host cell as described in Example 1.

[0147] In an alternative embodiment, said modified IgGl-based aPD-Ll antibodies having an enhanced CD16 affinity, have a heavy chain comprising a variable domain VH and a constant domain CHI of the heavy chain, and a light chain composed of a variable domain VL and a constant domain CL1, that together form a Fab domain capable of binding to PD-L1. The Fab domain is linked to a mutant Fc domain comprising the CH2 and CH3 domains, preferably via the hinge region.

[0148] Alternatively, said modified IgGl-based aPD-Ll antibodies comprise a sdAb or scFv capable of binding to PDL1, that is linked to a mutant Fc domain, preferably via the hinge region. The scFv comprises a VL and VH linked via a linker.

[0149] The amino acid sequence of said mutant Fc domain differs from that of the Fc domain of the parent IgGl from which it is derivable. Preferably the mutations are amino acid substitutions located in the CH2 region of the Fc domain.

[0150] Said modified IgGl-based aPD-Ll antibodies having an enhanced CD16 affinity, comprising a variable region capable of binding to PD-L1, may be defined by the amino acid sequence of the heavy chain regions: constant (CHI, CH2 and CH3) regions and hinge region to which they are fused.

[0151] The amino acid substitutions in the heavy chain of IgGl-based aPD-Ll antibodies that can confer enhanced CD16 affinity, and their position in a constant region of the heavy chain are defined with respect to the amino acid sequence of the corresponding regions of the parent IgGl from which they are derived. The position of each said amino acid substitution is defined herein using the EU numbering system, whereby the first amino acid residue of the CHI region is assigned position 118.

[0152] When the amino acid sequence of the parent heavy chain regions (i.e. CHI region, hinge region and CH2 and CH3 regions), starting from the first amino acid residue of the CHI region, has at least 50 %, 55 %, 60 %, 65 %, 70 %, 75 %, 80 %, 85 %, 90 %, 95 %, 96 %, 97 %, 98 %, 99 % or 100% sequence identity to SEQ ID NO.: 10, then the amino acid substitutions may be selected from any of the nine alternative Fc modificationsP3685PC00

[0153] listed in Table 1. When the Fc domain of the parent IgGl is incapable of N-glycosylation by virtue of having residue N298A substitution, then the amino acid sequence of the corresponding parent heavy chain regions has at least 50 %, 55 %, 60 %, 65 %, 70 %, 75 %, 80 %, 85 %, 90 %, 95 %, 96 %, 97 %, 98 %, 99 % pr 100% sequence identity to SEQ ID NO.: 12.

[0154] By way of example, the amino acid sequence of the mutated heavy chain regions (i.e. CHI region, hinge region and CH2 and CH3 regions) and starting from the first amino acid residue of the CHI region, may be selected from the SEQ ID's given for the nine Fc domain mutants listed in Table 1.

[0155] Table 1

[0156]

[0157] • In accordance with the EU numbering system [http: / / www.bitgene.net / antibody / eu_number], the CHI domain of the IgG 1 Heavy chain begins with the sequence ASTK, where A is assigned position 118, and the position of the listed mutations is defined with respect to the start of this Fc sequence.

[0158] Said modified IgGl-based aPD-Ll antibodies can be engineered by introducing mutations into a polynucleotide encoding the constant heavy chain (HC) sequence of said antibodies and expressing the cloned polynucleotide as described in example 10. Said modified IgGl-based aPD-Ll antibodies comprising any one of the mutant Fc domain sequences listed in Table 1 are characterized by an Fc domain having a CD16A affinity for both low- and high-affinity FcyRIIIA allotypes that is at least 20-fold greater than antibodies having a WT Fc domain (Example 10). This increase in CD16A affinity of the modified aPD-Ll antibodies of the invention, both those partially non-fucosylated antibodies comprising 0 - 25% core-fucosylated N-glycans (e.g. aPD-LlGE in Example 7), and those comprising said mutant Fc domain, are able to enhance expansion and activation of NKT cells, iNKT cells and their CAR- derivatives over and above that obtained by aPD-Ll WT antibodies.P3685PC00

[0159] In a further embodiment, said modified aPD-Ll antibodies (i.e. at least partially non-fucosylated or mutated aPD-Ll) may be bispecific antibodies, comprising at least one binding site capable of binding to PD-L1, as well as at least one binding site capable of binding to a tumor antigen such as HER.2, EGFR, VEGF, TA-MUC1, PSA, TFa (core-1), LeY, CEA, EpCAM, CD19, and BCMA, where the sequence of the respective binding sites of antibodies binding these targets are known in the art. The binding sites of the bispecific antibodies may have a variety of stoichiometries, including where each antibody molecule is asymmetric e.g. having only one binding site against both PD-L1 and tumor targets or 1 against one and 2 against the other; or other formats having other binding site valencies such as 2+4 etc. The binding moieties can be constructed and take the form of Fab, scFv or sdAb, as described by Brinkmann et al., (2017) and Madsen et al., (2023) and (2024a, b). Preferably, each bispecific antibody has 2 binding sites towards PDL1 and 2 sites against the selected tumor antigen.

[0160] TA-MUC1 is the "tumor-associated mucin 1 epitope TA-MUC1", which is selectively accessible to antibodies when overexpressed in association with colon, breast, ovarian, lung and pancreatic cancers; and hence provides an epitope for targeting said bispecific antibody to tumor cells in a patient.

[0161] HER.2 is the "human epidermal growth factor receptor 2", which is overexpressed in associated with breast, ovarian, bladder, pancreatic, stomach, and esophageal cancers. HER2 proteins have been shown to form clusters in cell membranes that may play a role in tumorigenesis; providing epitope(s) for targeting said bispecific antibody to tumor cells in a patient.

[0162] EGFR is an "epidermal growth factor receptor", mutations in which are found in some types of cancer, including non-small cell lung cancer, and may cause cancer cells to grow and spread in the body. EGFR is a readily accessible cell surface receptor providing epitope(s) for selective antibody-based targeting of tumor cells.

[0163] VEGF is "vascular endothelial growth factor" is one of the key angiogenic factors in tumors and participates in the initial stage of tumor development, progression and metastasis. VEGF is expressed in the bone marrow and released from platelets and can thus be immunologically detected and bound in a patient's plasma.

[0164] PSA is 'Prostate specific antigen' whose expression is strongly correlated with an onset of prostate cancer. PSA is expressed by prostate epithelial cells and prostate cancer (PCa) and is the most commonly used serum marker for cancer and can thus be immunologically detected and bound in a patient's serum.

[0165] TFa (core-1) is Thomsen-Friedenreich antigen (TFA) agglutinin which is a tumor-associated, blood-group-related glycosidic precursor structure [Gal(beta l-3)GalNAc], LeY, a difucosylated or trifucosylated carbohydrate type 2 blood group related antigen that is overexpressed by malignant myeloid cells and malignant colonic tissue.P3685PC00

[0166] CEA is a Carcinoembryonic antigen (CEA) encompassing a class of glycosyl phosphatidyl inositol (GPI) cell-surface-anchored glycoproteins whose specialized sialofucosylated glycoforms serve as functional colon carcinoma L-selectin and E-selectin ligands, which may be critical to the metastatic dissemination of colon carcinoma. CEA in blood samples 5 is used as a marker for cancer.

[0167] EpCAM is an epithelial cell adhesion molecule that provides a prognostic marker, a therapeutic target, and an anchor molecule on circulating and disseminated tumor cells (CTCs / DTCs), which are considered the major source for metastatic cancer cells.

[0168] CD19 is B-lymphocyte antigen CD19 that is a marker of B cells, and has been used to 10 diagnose cancers that arise from this type of cell - notably B cell lymphomas, acute lymphoblastic leukemia (ALL), and chronic lymphocytic leukemia.

[0169] BCMA is B cell maturation antigen whose overexpression and activation are associated with multiple myeloma (MM) in preclinical models and humans. BCMA serves as a biomarker for MM having prognostic value.

[0170] 15 The amino acid sequences of the binding sites of antibodies essential for binding to each of the above tumor antigens are known in the art.

[0171] The bispecific antibodies can be obtained by expressing a fusion protein having a single chain Fv region for a given tumor antigen antibody (e.g. a Her2 antibody) fused to the 20 constant domain or the variable domain of the light chain or heavy chain of one arm of a modified aPDL-1 antibody, such that the engineered bispecific antibody retains the Fc region of the modified aPDL-laPD-Ll antibody.

[0172] Alternatively, the bispecific antibodies can be engineered by fusing a tumor antigen sdAb or a PDL-1 sdAb, preferably by means of a linker (e.g. 10 residue GS linker), onto an 25 aPD-Ll IgGl or tumor antigen IgGl scaffold, as illustrated for HER.2 / PD-L1 bispecific antibodies in figure 9; table 2 and by Madsen et aL, (2023) and (2024a, b) for bispecific antibodies against Her2 and PD-L1. Note that the engineered bispecific antibody of the invention has an Fc region that is either at least partially non-fucosylated or is mutated, thereby conferring an enhanced affinity for CD16.

[0173] 30

[0174] Table 2

[0175]

[0176] P3685PC00

[0177]

[0178] *The heavy and light chains amino acid sequences are those in the respective antibodies; ** The bispecific antibodies correspond to those shown in Figure 9.

[0179] The antigen binding specificity of the modified monospecific and bispecific fucose- 5 reduced antibodies of the present invention can be tested using methods well known in the art, such as by an ELISA binding or blocking assay, or in a cell-based assay as illustrated in Example 4, figure 5. Alternatively, antigen binding specificity can be determined using flow induced dispersion analysis (FIDA) which is especially suitable in showing simultaneous binding to both antigens as described by Madsen et al. (2022) 10 and (2023).

[0180] The at least partially non-fucosylated monospecific and bispecific antibodies of the invention are preferably produced in cells of a cell line selected from among CHO-K1 FUT8 KO cell line (SC-RT2707, Creative biogene); HEK293 FUT8 KO cell line (YKO- 15 XN88623, UBIGENE); HEK293 GMDS KO (EDJ-KQ4727, EDITGENE) and HEK293 GFUS KO (EDJ-KQ5972, EDITGENE), NM-H9D8-E6 (DSM ACC 2807), NM-H9D8-E6Q12 (DSM ACC 2856), or a cell or cell line derived therefrom, preferably under serum-free conditions. The nucleic acid encoding said non-fucosylated monospecific and non- fucosylated bispecific antibodies may be introduced in these cells; expressed and 20 subsequently isolated.

[0181] In one embodiment, said monospecific or bispecific modified aPD-Ll antibodies further comprise a cytokine domain such as a IL15, IL2, IL21, GM-CSF or type I interferon. For example, the heavy chains of each modified aPD-Ll antibody can be expressed as fusion 25 proteins comprising a cytokine domain fused to the C-terminus of the Fc region, preferably with a linker, such as a 10 amino acid GS linker.

[0182] II A pharmaceutical composition further comprising dendritic cells for use as a medicament.

[0183] 30 The pharmaceutical composition for use as a medicament, according to the first embodiment may additionally comprise dendritic cells, preferably isolated dendritic cells. Accordingly, this composition comprises dendritic cells in addition to immune cells selected from the group consisting of NKT cells, iNKT cells, CAR-NKT cells and CAR-iNKT cells, combined with modified aPD-Ll antibodies (i.e. at least partially non-fucosylated 35 or mutated aPD-Ll, as defined in section I).P3685PC00

[0184] Dendritic cells (DCs) are antigen-presenting cells (APCs) of the mammalian immune system, that can be found in mammalians in both immature and mature states. DCs are derived from hematopoietic bone marrow progenitor cells (HSC) that initially transform into immature dendritic cells. DCs that circulate in the blood are generally immature DCs (iDCs), characterized by high endocytic activity and low T-cell activation potential and have no dendrites, in contrast to their mature counterparts mainly found in other tissues.

[0185] DCs are typically identified by their surface markers, listed in table 3 below, and can be isolated from various sources, such as blood (e.g. cord blood) using flow cytometry or a 'Blood Dendritic Cell Isolation Kit II, human' supplied by Miltenyibiotec.com (product 130-091-379). Alternatively, DCs in larger quantities may be derived ex vivo from isolated monocytes (CD14+) that are cultured, as described in example 8. Preferably, the DCs belong to a PD-L1+ subpopulation, that can be identified in a blood sample using an APC conjugated anti-PD-Ll antibody and selected using the FACS method described in example 4.

[0186] Table 3

[0187]

[0188] The DCs in said pharmaceutical composition may be iDCs or mature DCs (mDCs) (see Example 12) that can for example be derived from iDCs by cultivation in standard medium supplemented with TNFa, as described in example 1. Alternatively, the iDCs may be ex vivo derived from monocytes as described in example 8, for example using monocytes derived from a blood sample of a patient to be treated.

[0189] MUTZ-3 iDC and MUTZ-3 mDC (DSMZ, #ACC 295) can be used in said pharmaceutical composition, preferably for the purpose of in vitro or pre-clinical evaluation of a composition according to the first embodiment, as illustrated in the examples disclosed herein. Since MUTZ DCs behave as the immortalized equivalent of CD34+ DC precursors, there is a need to prevent their further multiplication prior to any clinical administration, such as by irradiation.P3685PC00

[0190] in Pretreated pharmaceutical composition for use as a medicament

[0191] The pharmaceutical composition for use as a medicament, as defined above in section I and II; may be pre-treated by contacting the components of the composition, namely the isolated immune cells, modified aPD-Ll antibodies (i.e. at least partially non-fucosylated or mutated aPD-Ll as defined in section I) and dendritic cells, prior to administration of the composition. Pre-treatment, by contacting the components of the composition, is preferably performed in a culture medium that can support the activation and optionally the expansion of the isolated immune cells (NKT or iNKT cells or their CAR- derivatives). Suitable culture media are known in the art, for example using the co-culture medium and conditions as described in example 1. Pretreatment comprising co-culture in a suitable medium may be performed for a period sufficient for activation and / or expansion of the NKT cells or iNKT cells or their CAR- derivatives.

[0192] Expansion or activation of NKT cells can be measured by the expression of CD3+ / CD56+, CD25, CD69 or CD137 cell markers. Expansion of iNKT cells can be measured by the expression of the same markers as for NKT cells but in combination with staining with 6B11 mAb, as illustrated in example 1.

[0193] In a preferred embodiment, the pretreated pharmaceutical composition for administration to the patient comprises the pretreated activated and expanded immune cells selected from among NKT cells, iNKT cells and their CAR- derivatives and modified aPD-Ll antibodies, from which dendritic cells present during pretreatment have been removed. Removal of the dendritic cells following pretreatment, but prior to administration may be preferred, for example when then dendritic cell are MUTZ-3 iDC and MUTZ-3 mDC (DSMZ, #ACC 295). Dendritic cells may be partially or fully removed from the pretreated pharmaceutical composition for example by fluorescence-activated cell sorting (FACS), as illustrated in example 9, or magnetic cell sorting known to those skilled in the art. Alternatively the pretreated NKT cells, iNKT cells and their CAR-derivatives can be isolated by FACS or magnetic cell sorting positively for these cells for administration to a patient in need as described in Section VI.

[0194] IV A proposed mode of action of the pharmaceutical composition

[0195] The relative frequency of NKT or iNKT cells in the mammalian T lymphocyte population in blood samples is normally low. When PBMCs, derived from a blood sample, are cocultured with either immature or mature mDCs (such as MUTZ-3 mDCs) and modified aPD-Ll antibodies, the abundance of detectable NKT and iNKT cells amongst the PBMCs is seen to increase significantly (as illustrated for aPD-LlGE antibodies in example 1 and 12). Both NKT and iNKT cell expansion is dependent on the PDL-1 epitopeP3685PC00

[0196] recognition / binding properties of the modified aPD-Ll antibody (example 1), since treatment with an alternative antibody, aHER-2GE, failed to induce cell expansion.

[0197] Treatment of NKT and iNKT cell with modified aPD-Ll antibodies during coculture with mDCs also induces earlier NKT and iNKT cell activation, as detected by expression of the cell surface receptors CD25, CD69, and CD137 (as illustrated for aPD-LlGE antibodies in example 2).

[0198] The IgGl-based Fc region of said modified aPD-Ll antibodies have an enhanced binding affinity for CD16 receptor, as compared to the Fc region of IgG-based wild-type IgGl-based antibodies from which they are derivable (Example 7, 10 and 11). Expansion of NKT and iNKT cells is also positively enhanced by the presence of CD3+CD16+ cells, indicating the Fc region of modified aPD-Ll antibodies having enhanced CD16 binding properties, plays an important role (as illustrated for aPD-LlGE antibodies in example 3).

[0199] Surprisingly, modified aPD-Ll antibodies depend on mDC cells, more specifically on PD-L1+ sub-population of mDCs, in order to induce the expansion and activation of NKT and iNKT cell (as illustrated for aPD-LlGE antibodies in example 4). While not wishing to be bound by theory, the ability of said modified aPD-Ll antibodies to induce the expansion and activation of NKT and iNKT cell is believed to depend on both binding interactions between their Fc domain with CD16 on CD3+CD16+ cells, such as CD16+ NKT and iNKT cells, in combination with binding interactions between their variable domains and on PD-L1 receptors on mDCs.

[0200] The interaction and combined effect of said modified aPD-Ll antibodies, NKT or iNKT cells (or their CAR-derivatives) and mDCs, provides a composition that exhibits enhanced cytotoxicity against both cancer cells and cancer tumor (3D spheroids composed of fibroblasts and cancer cells) as demonstrated in example 5.

[0201] V. Administration of the pharmaceutical composition

[0202] The pharmaceutical composition for use as a medicament may comprise a combination of isolated immune cells and modified aPD-Ll antibodies as defined in section I; or may further include mDCs as defined in section II above.

[0203] mDCs are optionally included in the pharmaceutical composition according to section I, because it will be appreciated that following administration of the composition, the native PD-L+DCs present in a subject that is treated, will interact with the components of the composition so as to both expand and activate the administered NKT or iNKT cells (or their CAR-derivatives).

[0204] The pharmaceutical compositions according to I or II, may be provided in liquid form.P3685PC00

[0205] Administration of the composition may be tailored to the cancer or infections disease to be treated, for example by selecting a suitable systemic route, such as any one of: intravenous, intra-tumoral, subcutaneous, intra-nodal and intraperitoneal.

[0206] The pharmaceutical composition may be administered to a subject who is preferably in need of the administration of the pharmaceutical composition, such as a subject suffering from cancer disease and / or infectious disease. The cancer disease may be selected from any type of tumors including solid and non-solid tumors, such as but not limited to:

[0207] • Carcinoma: Cancers derived from epithelial cells. This group includes many of the most common cancers that occur in older adults. Nearly all cancers developing in the breast, prostate, lung, pancreas, and colon are carcinomas.

[0208] • Sarcoma: Cancers arising from connective tissue (i.e. bone, cartilage, fat, nerve), each of which develop from cells originating in mesenchymal cells outside of the bone marrow.

[0209] • Lymphoma and leukemia: These two classes of cancer arise from immature cells that originate in the bone marrow, and are intended to fully differentiate and mature into normal components of the immune system and the blood, respectively. Acute lymphoblastic leukemia is the most common type of cancer in children, accounting for ~30% of cases. However, far more adults than children develop lymphoma and leukemia.

[0210] • Germ cell tumor: Cancers derived from pluripotent cells, most often presenting in the testicle or the ovary (seminoma and dysgerminoma, respectively).

[0211] • Blastoma: Cancers derived from immature "precursor" cells or embryonic tissue. Blastomas are generally more common in children,

[0212] (e.g. neuroblastoma, retinoblastoma, nephroblastoma, hepatoblastoma, medulloblasto ma) than in adults.

[0213] The subject is preferably a human in need of the administration of the pharmaceutical composition. In one embodiment, the subject is an adult.

[0214] In one embodiment the subject in need thereof, has an F / F or F / V or V / V allotype of CD16A. In the case of subjects having the F / F or V / F allotype low affinity receptor, the pharmaceutical composition of the invention is particularly advantageous due to the high affinity of the aPD-Ll antibodies of the invention (as defined in section liii) towards CD16 of the F-allotype.

[0215] VI A method of activating and expanding isolated immune cells

[0216] In a second embodiment, the invention provides a method of activating and expanding isolated immune cells selected from the group consisting of NKT cells, iNKT cells, CAR-NKT cells and CAR-iNKT cells, comprising the steps of:P3685PC00

[0217] a. contacting said immune cells with modified aPD-Ll antibodies (i.e. at least partially non-fucosylated or mutated aPD-Ll as defined in section I) and dendritic cells in vitro or ex vivo;

[0218] b. optionally isolating said immune cells and

[0219] c. optionally administering the isolated immune cells obtained in step (b) to a subject in need thereof.

[0220] A suitable method for activating and expanding said isolated immune cells is in a culture medium, for example using the co-culture medium and conditions as described in example 1. Pretreatment comprising co-culture in a suitable medium may be performed for a period sufficient for activation and / or expansion of the NKT cells or iNKT cells or their CAR- derivative.

[0221] EXAMPLES

[0222] Example 1: Example 1: Enhanced expansion of NKT cells through glycoengineering of anti-PDLl

[0223] Example 1 shows how coculture of MUTZ-3 dendritic cells (mDCs), human peripheral blood mononuclear cells (PBMCs), and aPD-LlGE affects the frequency of natural killer T (NKT) cells in PBMCs, compared to other aPD-Ll glycoforms.

[0224] Materials / methods:

[0225] Production of aPD-Ll antibodies: Antibodies directed against PD-L1 were generated as follows. A gene [SEQ ID NO.: 1] encoding a fusion protein (aPD-Ll heavy chain) [SEQ ID NO.: 2] and mature fusion protein [SEQ ID NO.: 3] comprising the variable domain (VH) from an atezolizumab aPD-Ll heavy chain (Genentech, MPDL3280A) fused to the constant domain of the Rituximab heavy chain [SEQ ID NO.: 10], was cloned into a pcDNA3.1-based expression vector (supplied by Themo Fischer V79020). A gene [SEQ ID NO.: 4] encoding a fusion protein (aPD-Ll light chain) [SEQ ID NO.: 44] comprising the variable domain (VL) from a atezolizumab aPD-Ll light chain (Genentech, MPDL3280A) fused to the constant domain of the Rituximab light chain [SEQ ID NO.: 14] encoded by sequence [SEQ ID NO.: 13], was similarly cloned into a pcDNA3.1-based expression vector (supplied by Themo Fischer V79020).

[0226] The coding sequences used for expressing the aPD-Ll heavy chain and light chains were codon-optimized for expression in CHO cells and synthesized by GeneArt (Thermo Fisher Scientific). The nucleotide sequence of the gene encoding the heavy chain of the aPD-LING antibody variant was modified using Q5 site-directed mutagenesis kit (New England Biolabs) to produce a nucleotide sequence [SEQ ID NO.: 6] encoding a mutant fusion protein (aPD-LlNG light chain) chain [SEQ ID NO.: 7] and mature fusion protein [SEQ ID NO.: 8] in which asparagine in the CH2 domain was substituted with alanine at positionP3685PC00

[0227] 297 of the Rituximab heavy chain (as defined in table 1). The pcDNA3.1-expression plasmids were amplified in E. coli DH5a (New England Biolabs, #C2987H) and purified using NucleoBond Xtra Midi Plus EF (Macherey-Nagel, # 740422.50) to ensure endotoxin removal. LC and HC plasmids were co-transfected into CHO cells to establish stable independent CHO-S WT pools for expression of aPDLlWT and aPDLING antibody variants, respectively. For production of aPD-LlGE antibody variant LC and HC plasmids were co-transfected in CHO-K1 FUT8 KO cell line (SC-RT2707, Creative biogene) for development of stable pools for expression of aPD-Ll antibodies with eliminated core fucosylation (aPD-LlGE).

[0228] After co-expression of the respective LC and HC encoding plasmids in CHO cells, the supernatants were clarified (centrifugation at 1500xg for 10 min and 0.45 pm filtering) before purification using Protein A HiTrap MabSelect SuRe (Cytiva) affinity chromatography connected to an AKTA Pure system (Cytiva). Antibodies were concentrated and buffer exchanged to DPBS (Sigma Aldrich, #D8537), using Amicon Ultra-15 centrifugal filter devices (30-kDa cutoff, EMD) and the protein concentrations were measured using NanoDrop One UV-vis spectrophotometer (ThermoFisher) and theoretical extinction coefficient predicted by ProtParam (Expasy) based on the amino acid sequence of the aPD-Ll antibody.

[0229] Isolation of primary cells: PBMCs were isolated from buffy coats by a two-step density gradient centrifugation (200xg and 460xg) using Histopaque-1077 (Sigma Aldrich, #10771). Buffy coats were obtained from anonymized healthy donors, with informed written consent secured in compliance with the legislation and guidelines of the local ethics committee (Region Hovedstaden, Denmark).

[0230] MUTZ-3 differentiation and maturation-. MUTZ-3 cells (DSMZ, #ACC 295) were cultured under standard culture conditions according to manufactures protocol. For differentiation, MUTZ-3 cells were seeded at a density of lxlO5cells / mL in MEMa (Gibco, # 22571038) containing 20% FBS (Gibco, #A5256801), 1% L-glutamine (Sigma-Aldrich, #G7513) and 1% penicillin, streptomycin (Gibco, #15140163) (complete MEMa). Differentiation was induced by adding 100 ng / mL GM-CSF (PeproTech, #300-03), 10 ng / mL IL-4 (PeproTech, #200-03), and 2.5 ng / mL TNFa (PeproTech, #300-01A) to the culture, followed by incubation for 7 days. On day 4, fresh differentiation medium was added (cell dilution 1:2). For MUTZ-3 iDC maturation (mDC), 2xl05iDC / mL were incubated for three days in complete MEMa supplemented with 75 ng / mL TNFa (PeproTech, #300-01A).

[0231] Cocuiture: PBMCs were cocultured with MUTZ-3-derived mDCs at a PBMC:DC ratio of 10:1 in RPMI 1640 (Sigma-Aldrich, #42401018) supplemented with 10% FBS (Gibco,P3685PC00

[0232] #A5256801), 1% L-glutamine (Sigma-Aldrich, #G7513), and 1% penicillinstreptomycin (Gibco, #15140163) (referred to as complete RPMI). Cells were seeded at a density of lxlO6PBMCs / mL in a round-bottom 96-well plate (Thermo, #168136) and incubated for 5 days. The aPD-Ll-wr, aPD-Ll-NG and aPD-Ll-GE antibodies were added individually to separate wells comprising the PBMC cocultures at a concentration of 1 pg / mL.

[0233] Flow cytometry: The frequency of NKT cells was assessed by flow cytometry. Cells were first incubated with LIVE / DEAD™ Fixable Yellow Dead Cell Stain (ThermoFisher Scientific, #L34967) and FcR blocking reagent (Miltenyi, #130-059-901), followed by staining with the antibody panel listed in Table 4. After staining, cells were fixed in IC fixation buffer (eBioscience: #00-8222-49) and stored at 5°C until analysis. Data acquisition was performed on a MACSQuant Analyzerl6 (Miltenyi), and the results were analyzed using FlowJo vlO.7.1. Fluorescence minus one (FMO) controls were included for each antibody to establish appropriate positive gating for the specific markers, following exclusion of doublets and dead cells.

[0234] Table 4

[0235]

[0236] Results: Coculturing PBMCs and MUTZ-3 DCs (mDCs) with aPD-LlGE led to a significant fold increase in the percentage of NKT cells after 7 days that was greater than coculture with aPD-Llwt, while coculture with aPD-LlNG failed to increase the percentage of NKT cells (Figure 1 + Figure 2A). In contrast, cocultures treated with aHER2GE antibodies did not exhibit NKT cell expansion after 7 days (Figure 2B). This suggests that the observed increase in NKT cells with aPD-LlGE is specifically dependent on PD-L1 binding, rather than solely on CD16 engagement, as both aHER2GE and aPD-LlGE exhibit equivalent CD16 binding affinity due to defucosylation (see example 7).

[0237] Conclusion: aPD-LlGE induces a significant expansion of NKT cells after 7 days of coculture between PBMCs and mDCs.

[0238] Example 2: Enhanced expansion of NKT cells through glycoengineering of anti-PDLl is accompanied by earlier NKT cell activationP3685PC00

[0239] Example 2 evaluates whether the expansion of NKT cells in cocultures of mDCs, and PBMCs treated with aPD-LlGE, was associated with increased activation of the NKT cells, as indicated by the expression of the cell surface receptors CD25, CD69, and CD137.

[0240] Methods: Coculture and flow cytometry were conducted as described in example 1. The procedure was modified by staining cells on both day 5 and day 7 using the antibody panel outlined in Table 5.

[0241] Table 5

[0242]

[0243] Results: As seen in figure 3, the detected levels of activation markers CD25, CD69 and CD137 were highly elevated on day 5 in cocultures treated with aPD-LlGE as compared to cocultures treated with either aPD-Llwt, or aPD-LlNG.

[0244] Conclusions: These results demonstrate that increased expansion of NKT cells in cocultures of mDCs and PBMCs treated with aPD-LlGE is correlated with earlier activation of NKT cells.

[0245] Example 3: Enhanced expansion and activation of NKT cells is dependent on CD16 expression on lymphocytes

[0246] Example 3 investigates whether the increased expansion of NKT cells induced in cocultures treated with aPD-LlGE was dependent on CD16 expression on CD3+ cells.

[0247] Methods:

[0248] Primary cell isolation: CD3+ cells were isolated from the PBMCs (PBMC purification was performed as in Example 1) using Dynabeads untouched human T cell kit following manufactures protocol (Invitrogen, # 11344D).

[0249] Fluorescence Activated Cell Sorting (FACS). Prior to FACS, CD3+ cells were washed in DPBS (Sigma Aldrich, #D8537), and incubated for 10 min with APC conjugated anti-CD16 antibody (Invitrogen, #MHCD1605). Following this, the cells were washed in DPBSP3685PC00

[0250] to eliminate excess antibody before sorting on a SONY MA900 cell sorter utilizing a 100 pm sorting chip (#LE-C3210). Doublets were excluded prior to sorting based on forward scatter (FSC)-H versus FSC-A, and dead cells were identified and excluded based on FSC versus back scatter (BSC). The negative and positive CD16 populations were gated using unstained controls as references.

[0251] Following sorting, cocultures were established and analyzed using flow cytometry, as described in example 1, but where the step of coculture with MUTZ-3 DCs was modified by replacing the PBMCs with sorted CD3+CD16- cells and unsorted CD3+ cells. The cocultures were treated with either aPD-LlNG, aPD-Llwt, or aPD-LlGE, or untreated.

[0252] Results: As seen in figure 4, cocultures depleted of CD3+cells expressing CD16 (CD3+CD16-) resulted in a significantly lower expansion of NKT cells upon treatment with aPD-LlGE compared to coculture with CD3+cells expressing CD16 ((CD3+CD16+). No significant differences in NKT cell expansion associated with CD16 expression were observed in the other treatment groups.

[0253] Conclusion: These data show that CD16 expression on CD3+ cells has a positive effect on the observed expansion of NKT cells upon coculture with MUTZ-3 mDCs and aPD-LIGE.

[0254] Example 4: Enhanced expansion and activation of NKT cells is dependent on PD-L1 on MUTZ-3 mDCs

[0255] Examples 4 investigate whether PD-L1 expression on DCs was important for NKT cell expansion.

[0256] Methods: MUTZ-3 cells were differentiated for 7 days and then sorted into PD-L1-and PD-L1+ DC subpopulations using the FACS method described in example 3 but modified by replacing the CD16 antibody with an APC conjugated anti-PD-Ll antibody (Biolegend, #324408). After sorting, MUTZ-3 iDC were matured following the standard 3-day TNFa protocol described in example 1, but in the presence of lpg / mL aPD-LlNG, aPD-Llwr or aPD-LlGE. The MUTZ-3 mDCs were then cocultured with PBMCs following coculture procedure outlined in example 1, and NKT cell expansion was analyzed on day 7 of coculture using the flow cytometry procedure described in example 1. The mDCs obtained following the maturation step were phenotyped to assess PD-L1 expression using the staining procedure of example 2, but modified by replacing the antibody panel in Table 4 with an anti-PD-Ll antibody (Biolegend, #324408).P3685PC00

[0257] Results: Maturation of DCs in the presence of aPD-Ll antibodies completely blocked detection of PD-L1, with no differences between the different aPD-Ll glycovariants as expected, showing that the antibodies bind to PD-L1 expressed by the DCs irrespective of their glycoprofile (Figure 5). When PBMCs were cocultured with each one of the matured PD-L1+ and PD-L1- MUTZ-3 mDC populations from the 4 aPD-Ll-treatments, a significantly higher fold expansion of NKT was observed when aPD-LlGE was present during coculture with PD-L1+ mDCs as compared to PD-L1- DCs (Figure 6).

[0258] Conclusion: These results demonstrate the dependency of aPD-LlGE binding to PD-L1 on DCs on its ability to modulate the expansion of NKT cells.

[0259] Example 5 Increased tumor cell killing mediated by increased expansion and activation of NKT cells through glycoengineering of anti-PDLl

[0260] This example investigates whether the cell composition obtained by stimulating PBMCs with MUTZ-3 DCs and aPD-LlGE, characterized by a significantly higher proportion of NKT cells, results in enhanced killing of cancer cells in both 2D cultures and complex 3D cancer spheroids comprising fibroblasts and cancer cells.

[0261] Material / methods: Purification of PBMCs and maturation of MUTZ-3 DCs was performed as described in example 1.

[0262] 2D experiment: The xCELLigence RTCA (Agilent) system was utilized to monitor cell proliferation and death in real time using specialized 96-well E-plates with integrated electrodes at the bottom of each well (Agilent, #300601020). These electrodes detect cell adherence, where increased attachment raises impedance, recorded as the cell index (CI). DU-145, a human prostate cancer cell line (ACC 261) as target cells [T] (25,000 per well) were seeded in 100 pL of complete RPMI in each well, and allowed to grow for ~24 hours under standard incubation conditions until confluency was reached. Subsequently, each well received 100 pL of media containing the appropriate number of effector cells [E]. The effector cells were PBMCs that had been cocultured with MUTZ-3 mDCs and 1 pg / mL aPD-Ll glycovariants for 5 days prior to their use (as outlined in Example 1), that on day 5, were then added at an E:T ratio of 10: 1 to the confluent DU-145 cells. The cell index (CI) was assessed every 15 min for 30 h post PBMC addition. Each treatment condition was conducted in technical triplicates and the average and standard deviation were determined using the instrument's provided software. Raw CI curves were normalized to a CI value of 1.0 at the last measurement before PBMC treatment. For data analysis, CI measurements from the last pre-treatment reading up to 25 hours post-treatment were included. The area under the normalized CI curve (AUC) was computed in GraphPad Prism 10.2.1 using its built-in "Area Under the Curve" analysis feature.P3685PC00

[0263] 3D experiment: Spheroids were generated by coculturing 1BR.3.G fibroblast cells (from Bioneer) with cells of the human prostate cancer cell line DU-145 (ACC 261) as target cells. Both 1BR.3G and DU-145 cell lines were prestained with CellTrace™ Far Red (#C34564, Life Technologies) following the manufacturer's protocol, enabling subsequent spheroid quantification through red light detection. A total of 8000 target cells, at a 1: 1 ratio of fibroblasts to cancer cells, were seeded in complete RPMI medium in black ULA 96-well U-bottom plates (Corning, #CLS4515). The cells were centrifuged at 500xg for 10 min to promote spheroid formation, prior to incubation in xCELLigence RTCA at 37°C with 5% CO2 for 2 days. Images were captured every 5 h to monitor spheroid development. On the second day of spheroid formation (50h), PBMCs, which had been cocultured for 5 days with MUTZ-3 mDCs and 1 pg / mL of an aPD-Ll glycovariant, were added to the spheroids. 80000 viable PBMCs were added to each well. Since addition of PBMCs to spheroids exceed the single field of view, 2 fields of view / well were captured and eSight's 1 x 2 stitching feature was used to capture a larger area and ensure that all cells were captured. Images were acquired once per 2 h over a 30h time period using lOxobjective. Images were analyzed using xCELLigence software. Red fluorescence (CellTrace Far Red) was employed to create organoid segmentation masks to measure changes in spheroid area (pm2), as well as the total integrated red intensity (RRI x pm2 / Image) in each cell. Both exported variables were normalized to the last image before treatment to account for differences in spheroid size and plotted over a duration of 25h post treatment. Each treatment was applied to 10 different spheroids, and the results were averaged. The percentage of cytolysis for both parameters was calculated using equation:

[0264] % cytolysis = ((Parameter™ effector - Parametereffector) I Parameter™ effector) x 100.

[0265] Results: In the 2D cancer killing assay (Figure 7), no significant difference in target cell (DU-145) AUC was observed between PBMCs from cocultures prepared without antibody and those treated with aPD-LlNG. PBMCs from aPD-Llwr cocultures exhibited enhanced DU-145 cell killing compared to aPD-LlNG, while PBMCs from aPD-LlGE cocultures demonstrated a further increased killing effect relative to aPD-LlNG. To assess whether the superior cytotoxic capacity observed with PBMCs preactivated by aPD-LlGE in 2D would be sustained in a more complex model, their effect on 3D tumor spheroids was analyzed. The addition of preactivated PBMCs revealed that cocultures with untreated PBMCs and those treated with aPD-LlNG showed similar DU-145 cell killing kinetics based on the investigated parameters: Integrated Red Intensity (RRI x pm2 / Image) and total red area (pm2 / well) (Figure 8 B+C). aPD-LlwT induced stronger cytolysis than aPD-LlNG, while aPD-LlGE induced the highest percentage of cytolysis, reaching maximum killing after ~65 hours from spheroid seeding (Figure 8A-C).P3685PC00

[0266] Conclusion: This study demonstrates that PBMCs enriched with activated NKT cells through coculture with mDCs and aPD-LlGE exhibit enhanced cytotoxicity against both adherent cancer cells and 3D spheroids composed of fibroblasts and cancer cells. These results underscore the potential of aPD-LlGE-activated PBMCs to improve therapeutic efficacy.

[0267] Example 6: Design and Construction of a CAR-NKT cell and CAR-iNKT cell Methods for the design and construction of a chimeric antigen receptor (CAR) NKT cells are as follows.

[0268] Materials / Methods:

[0269] CAR Constructs and Retroviral Vector Production: In this example, primary human NKT cells and iNKT cells were engineered to express a CAR targeting the GD2 ganglioside (CAR.GD2) (figure 10; Table 6). The CAR construct includes an antigen-recognition domain comprising the single-chain variable fragment (scFv) of the GD2-specific antibody 14G2a, linked to a short hinge region derived from human IgGl, followed by a CD28-derived transmembrane domain, and intracellular signaling domains including the TCR -chain for optimal T cell activation. Other costimulatory molecules such as including 4-1BB and 0X40 could also be included.

[0270] Table 6

[0271]

[0272] A suitable vector to introduce the nucleic acid sequence of the CAR construct into host cells include plasmid or a viral vectors like a SFG retroviral backbone (Addgene #22493). The 14G2a scFv sequence was cloned into the SFG retroviral backbone in-frame with the IgGl hinge and the human TCR / CD3 -chain. Retroviral supernatants were produced by transfecting HEK293T cells (supplied by Merck # CB_12022001) with a combinationP3685PC00

[0273] of plasmids encoding the CAR construct, RDF plasmid for RD114 envelope (supplied by addgene #17576), and a retroviral expression plasmid encoding MoMLV gag-pol (supplied by addgene #14887) to produce viral particles. Retroviral supernatants were produced by transfecting HEK293T cells with a combination of plasmids encoding the CAR construct, RDF plasmid for RD114 envelope, and retroviral plasmid encoding MoMLV gag-pol to produce viral particles.

[0274] NKT and iNKT cell transduction: NKT and iNKT cells were respectively transduced with the GD2-CAR construct by inoculating them with the retroviral supernatants. CAR.GD2 expression on transduced NKT and iNKT cells was detected using the anti-idiotype 1A7 mAb followed by staining with a secondary rat anti-mouse-IgGl-PE mAb (BD Biosciences) and analyzed by flow cytometry

[0275] Results: CAR.GD2 expression can be successfully detected on transduced NKT and iNKT cells, confirming the successful engineering of the cells.

[0276] Conclusion: The protocol provides an effective means for construction and expression of functional CAR NKT and CAR iNKT cells, for inclusion in the pharmaceutical compositions of the present invention.

[0277] Example 7: CD16 affinity of aPD-LlGE, aPD-LlwT and 3HER2GE

[0278] This example examines the binding affinity of aPD-LlGE, aPD-Llwi and aHER2GEto the CD16A receptor (176V allotype), to determine whether they exhibit comparable binding.

[0279] Material / methods: Binding affinities of aPD-LlGE, aPD-Llwi and aHER2GE antibodies to monomeric FcyRIIIa (CD16A)-176V (ACRO Biosystem, #CD8-H52H4) were evaluated using the Octet RED96e system (ForteBio). Antibodies were prepared at a concentration of 5 pg / mL in an assay buffer composed of PBS (pH 7.4), 0.02% Tween20, and 0.1% BSA. Antibodies were immobilized on FAB2G biosensors (Sartorius) for 120 seconds, followed by a 180-second equilibration period in the assay buffer. The interaction with FcyRIIIa was measured by exposing the immobilized antibodies to varying concentrations of FcyRIIIa (0-500 nM) for 60 seconds, followed by a 60-second dissociation period in the assay buffer. The raw data was processed using ForteBio data analysis software 12.0. Data alignment was performed at the association step, and the Savitzky-Golay filter was applied to globally fit the sensorgrams to a 1: 1 binding model, utilizing the entire association step and the first 5 seconds of the dissociation step for fitting.

[0280] Results: As shown in figure 11, aHER2GE exhibited a CD16A binding affinity of 1.10x10"8M while aPD-LlGE had an affinity of 1.17xl0-8M, based on 3 independent experiments, equivalent to a 25-fold increase in affinity.P3685PC00

[0281] Conclusion: aPD-LlGE and aHER2GE exhibit similar binding affinities to CD16A consistent with their defucosylated Fc domains. The higher KD and corresponding lower affinity observed for aPD-LlwT confirm that Fc core fucosylation negatively impacts CD16A binding and that removal of core fucose enhances CD16A receptor affinity.

[0282] Example 8: Method for producing imoDCs and moDCs ex vivo derived from isolated monocytes (CD14+).

[0283] Dendritic cells (moDCs) can be produced from monocytes by means of the following method.

[0284] Materials / Methods: PBMCs were isolated from buffy coats by a two-step density gradient centrifugation (200xg and 460xg) using Histopaque-1077 (Sigma Aldrich, #10771). Buffy coats were obtained from anonymized healthy donors, with informed written consent secured in compliance with the legislation and guidelines of the local ethics committee (Region Hovedstaden, Denmark). Monocytes were isolated from the PBMCs using Dynabeads Untouched Human Monocytes Kit following manufactures protocol (Invitrogen, # 11350D). Isolated monocytes were differentiated into immature monocyte derived dendritic cells (imoDCs) according to a 7-day protocol in MEMa medium (Gibco) supplemented with 20% FBS, 10% 5637 conditioned medium, 250 U / mL GM-CSF (Peprotech) and 500 U / mL IL-4 (Peprotech). Medium supplemented with cytokines was refreshed at day 3 or 4. Maturation of immature imoDCs was initiated by culturing harvested immature moDCs for 1-3 days in complete MEMa supplemented with 50-75 ng / mL TNFa (PeproTech, #300-01A). The imoDCs and the resulting mature moDCs were phenotyped with antibody panel presented in table 3 to confirm successful differentiation and maturation.

[0285] Results: Phenotyping confirmed the successful differentiation of monocytes into immature imoDCs by upregulation of DC related markers including CDllc and CD80 / CD86. Maturation of moDCs was confirmed by further upregulated CD83 expression. The observed phenotypic changes were consistent with expected dendritic cell maturation profiles, validating the protocol's effectiveness.

[0286] Conclusion: This method provides a reproducible approach for differentiation and maturation of moDCs.

[0287] Example 9: Method for depletion of DCs from the pretreated pharmaceutical composition.P3685PC00

[0288] The pretreated pharmaceutical composition of the invention can be depleted for the DC component to obtain a pretreated DC free pharmaceutical composition.

[0289] Material / Methods: Prior to coculture, DCs were labelled with CellTrace™ CFSE (Invitrogen, # C34554) according to the manufacturer's instructions. After labeling, cells were washed twice with complete culture medium to remove excess dye before being introduced into the coculture as described in Coculture section of Example 1. Following coculture, CFSE labeled DCs were selectively depleted using fluorescence-activated cell sorting (FACS), with the remaining unlabeled cells now constituting the pharmaceutical composition.

[0290] Results: Fluorescence analysis confirmed efficient labeling of DC with CellTrace™ CFSE, ensuring their clear identification. After coculture, FACS successfully depleted CFSE-labeled DCs, yielding a purified population of unlabeled cells.

[0291] Conclusion: The combination of CFSE labeling and FACS-based depletion enabled precise and efficient removal of DCs while preserving the integrity of the remaining cell population. This method provides a reliable strategy for obtaining a pretreated DC free pharmaceutical composition.

[0292] Example 10 Enhanced CD16A Binding via HC Mutagenesis

[0293] Variants of the aPD-Ll WT antibody having amino acid substitutions introduced into the CH2 domain of the heavy chain (as listed in Table 1) have enhanced FcyRIIIA affinity, for example where the substitutions are S239D / A330L / I332E or G236A, S239D, A330L, and I332E.

[0294] Methods

[0295] The aPD-Ll variants with enhanced CD16 binding were produced by Fc engineering. A plasmid comprising a nucleic acid sequence in a plasmid encoding the aPD-Ll heavy chain-WT HC rituximab fusion [SEQ ID NO.: 2] having a nucleotide sequence [SEQ ID NO.: 9] encoding the wild-type rituximab HC constant region [SEQ ID NO.: 10], was mutated using the Q5 Site-Directed Mutagenesis Kit (New England Biolabs). The parent WT HC rituximab fusion as well as the mutated coding sequences encoding the 9 variant aPD-Ll heavy chains each having one of the sets of amino acid substitutions set out in Table 1, in the Fc region, was confirmed by nucleic acid sequencing (Macrogen, Amsterdam, Netherlands), where the encoded mutant rituximab HC constant region was shown to have the respective amino acid sequence (as set out in Table 1). A plasmid encoding the aPD-Ll heavy chain-non-glycosylated [NG] HC rituximab fusion [SEQ ID NO.: 6] having a nucleotide sequence [SEQ ID NO.: 7] encoding the non-glycosylated NG rituximab HC constant region [SEQ ID NO.: 12], Expression plasmids comprising theP3685PC00

[0296] respective coding sequences were amplified in E. coli DH5a (New England Biolabs) and purified using the NucleoBond Xtra Midi Plus EF kit (Macherey-Nagel) to ensure endotoxin removal. Antibodies were produced by transient expression as described in Example 1. The glycoprofiles of the aPD-Ll heavy chain-non-glycosylated [NG] HC rituximab fusion; the aPD-Ll heavy chain-WT HC rituximab fusion and the non-fucosylated aPD-LlGE (produced in example 1) are shown in Figure 12.

[0297] Binding of the Fc-mutated aPD-Ll antibodies to monomeric FcyRIIIa (CD16A)-176F (ACRO Biosystem, #CDA-H5220) and FcyRIIIa (CD16A)-176V (ACRO Biosystem, #CD8-H52H4) was measured using the Octet RED96e (ForteBio) as previously described in Example 7.

[0298] Results

[0299] Sequencing results confirmed that the desired mutations were successfully introduced while ensuring that the rest of the aPD-Ll heavy chain sequence remained identical to the wild-type. The glycoprofiles of the antibodies shown in Figure 12, confirm that aPD-LING antibodies lack glycosylation; while the aPD-Llwr antibodies comprise core-fucosylated complex N-glycans, and the aPD-LlGE antibodies comprise non-fucosylated complex N-glycans. The glycoprofile of each of the aPD-Ll variants (Table 1) corresponded to that of aPD-LlwT antibodies (not shown). The introduction of each of these combinations of mutations (Table 1) significantly increases the binding affinity to both FcyRIIIA allelic variants relative to that of aPD-LlwT antibodies, as measurable by their lower KD observed by the Octet measurements. Furthermore, the combination of mutations: S239D, A330L and I332E in a human IgGl backbone is reported to have a FcyRIIIa (CD16A)-176F affinity of 201 nM, being a 22-fold increase compared to a WT human IgGl backbone (Ahmed et al., 2016). The combination of mutations: G236A, S239D, A330L, and I332E in a human IgGl backbone is reported to have an FcyRIIIa (CD16A)-176F affinity of 110 nM, being a 20-fold increase compared to a WT human IgGl backbone (Bournazos et al., 2016).

[0300] Conclusion

[0301] By introducing the combinations of amino acid substitutions into the CH chain of the aPD-Ll antibody listed in Table 12, all of the resulting antibodies have significantly increased CD16A affinity for both low- and high-affinity FcyRIIIA allotypes. A corresponding increase in CD16A affinity for both low- and high-affinity FcyRIIIA allotypes is also observed for aPD-Ll antibodies (comprising the N297A substitution [SEQ ID NO.: 12]) when the amino acid substitutions of variant 1 - 9, are introduced in the CH region of the rituximab HC constant region (NG).

[0302] Example 11: Production of aPD-Ll antibodies comprising <25% Fc core fucosylation and determination of CD16A affinityP3685PC00

[0303] This example describes the controlled production of aPD-Ll antibodies comprising defined low levels of Fc core fucosylation (<25%) and the determination of their CD16A binding affinity.

[0304] Material / methods: aPD-Ll antibodies were produced in a HEK293 GMDS knockout (GMDS KO) cell line (EDJ-KQ4727, Editgene), in which the de novo GDP-fucose biosynthesis pathway is disrupted. In this system, intracellular GDP-fucose levels are markedly reduced, resulting in antibodies with minimal Fc core fucosylation. To generate antibodies with predefined residual levels of Fc core fucosylation, L-fucose was added to the culture medium during antibody expression at controlled concentrations. Exogenous L-fucose is converted intracellularly via the salvage pathway into GDP-fucose, thereby enabling controlled reconstitution of Fc core fucosylation. By titrating the concentration of L-fucose in the culture medium, aPD-Ll antibodies comprising defined percentages of core-fucosylated Fc N-glycans, including variants containing <25% core fucosylation, can be generated. Antibodies were purified by Protein A chromatography as described in Example 1. Binding affinities of the purified antibodies to CD16A were determined using the identical Octet RED96e protocol described in Example 7. KD values were calculated by global fitting to a 1:1 binding model.

[0305] Results: Using the described GMDS KO production system, aPD-Ll antibodies comprising defined levels of residual Fc core fucosylation, including <25% core-fucosylated glycans, can be generated in a controlled and reproducible manner. CD16A binding affinity of each variant can be directly determined under identical assay conditions as used for aPD-LlGE and aPD-Llwr, thereby enabling direct comparison of CD16A affinity of antibodies comprising <25% core fucosylation with the high-affinity binding observed for aPD-LlGE. aPD-Ll antibodies comprising <25% core-fucosylated glycans were found to have a CD16A affinity statistically equivalent to that determined for aPD-LlGE antibodies.

[0306] Conclusion: This example provides a self-contained method for producing aPD-Ll antibodies with low residual Fc core fucosylation (<25%) and for determining their CD16A binding affinity under identical conditions to those used for aPD-LlGE and aPD-LIWT. The described approach enables direct experimental comparison of CD16A affinity as a function of Fc core fucosylation within the claimed range, on the basis of which it has been observed that aPD-Ll antibodies comprising <25% core-fucosylated glycans have the same CD16A binding affinity as aPD-LlGE antibodies.

[0307] Example 12: Enhanced CD8+T Cell Activation upon Co-culture of CD3+Cells with MUTZ-3 iDCs or moDCs in the Presence of aPD-LlGEP3685PC00

[0308] This example investigates whether glycoengineered aPD-LlGE enhances CD8+cell activation when CD3+cells are co-cultured with dendritic cell populations other than MUTZ-3 mDCs, namely MUTZ-3 iDCs and moDCs.

[0309] Material / methods: moDCs were generated as described in Example 8. MUTZ-3-derived iDCs were obtained as described in Example 1. Primary CD3+cells were isolated from PBMCs as described in Example 3. PD-L1 expression on MUTZ-3 iDCs and moDCs was determined as described for MUTZ-3 mDCs in Example 4.

[0310] CD3+cells were co-cultured with either MUTZ-3 iDCs or moDCs according to the coculture procedure outlined in Example 1. The co-cultures were treated with 1 pg / mL of aPD-LlwT, aPD-LlNG, or aPD-LlGE, or a buffer as untreated control.

[0311] On day 5 of co-culture, CD25 expression on CD4+and CD8+cells was assessed by flow cytometry using the staining and analysis procedure described in Example 1, excluding CD56 from the antibody panel. CD4+cells were defined as CD3+CD8“ cells. CD25 fold change relative to untreated co-cultures was calculated.

[0312] Results: Flow cytometric analysis confirmed PD-L1 expression on both MUTZ-3 iDCs, MUTZ-3 mDCs and moDCs (Figure 13), demonstrating that each of these dendritic cell populations express the PD-L1 target molecule for aPD-Ll binding.

[0313] In both CD3+:moDC and CD3+: MUTZ-3 iDC co-cultures, treatment with aPD-LlGE resulted in a statistically significant increase in CD25 expression on CD8+cells compared to untreated controls and aPD-LlNG-treated cocultures. CD8+activation levels in aPD-LlGE-treated cocultures were consistently higher than those observed with aPD-LlwT. CD4+cells showed only minor changes across treatment groups (Figure 14).

[0314] Conclusion: These results demonstrate that glycoengineered aPD-LlGE enhances CD8+cell activation in cocultures with both moDCs and MUTZ-3 iDCs. As both DC populations express PD-L1, and the activation effect is observed across these distinct DC types, the mechanism is not restricted to MUTZ-3 mDCs but extends to additional DC types. This supports the broader applicability of the CD16A-mediated mechanism described in the invention.

[0315] Embodiments of the invention

[0316] In a first embodiment the invention provides a pharmaceutical composition for use as a medicament, comprising immune cells selected from the group consisting of NKT cells, iNKT cells, CAR-NKT cells and CAR-iNKT cells, and modified aPD-Ll antibodies, wherein said modified aPD-Ll antibodies are either:

[0317] at least partially non-fucosylated antibodies comprising 0 - 80% core-fucosylated N-P3685PC00

[0318] glycans, or

[0319] mutant aPD-Ll antibodies, wherein two or more amino acid residues in a heavy chain of said mutant aPD-Ll antibodies are substituted, and wherein said substitutions are selected from the group consisting of:

[0320] a. S298A, E333A and K334A;

[0321] b. S239D and I332E;

[0322] c. S239D, A330L and I332E;

[0323] d. G236A, S239D and I332E;

[0324] e. G236A, A330L and I332E;

[0325] f. G236A, S239D, A330L, and I332E;

[0326] g. F243L, R292P, Y300L, V305I, P396L;

[0327] h. L235V, F243L, R292P, Y300L, P396L; and

[0328] i. P247I and A339Q,

[0329] wherein the substitutions are numbered relative to the first amino acid of the CHI region assigned position 118 of said heavy chain.

[0330] In a second embodiment the invention provides the pharmaceutical composition for use as a medicament of the first embodiment, wherein

[0331] the heavy chain of said mutant aPD-Ll antibodies comprises an amino acid sequence derivable from a parent human IgGl heavy chain by said substitutions, wherein the corresponding amino acid sequence in said parent human IgGl heavy chain has at least 70% sequence identity to SEQ ID NO.: 10 or 12.

[0332] In a third embodiment the invention provides the pharmaceutical composition for use as a medicament according to the first or second embodiment, wherein the composition further comprises dendritic cells.

[0333] In a fourth embodiment the invention provides the pharmaceutical composition for use as a medicament according to any of the first to third embodiments, wherein said immune cells are pretreated; said pretreatment comprising culturing said immune cells with said modified aPD-Ll antibodies and dendritic cells.

[0334] In a fifth embodiment the invention provides the pharmaceutical composition for use as a medicament according to the third or fourth embodiment, wherein the dendritic cells comprise PD-L1 positive dendritic cells.

[0335] In a sixth embodiment the invention provides the pharmaceutical composition for use as a medicament according to any one of first to fifth embodiments, wherein the NKT cells, iNKT cells, CAR-NKT cells and CAR-iNKT cells are selected from the groupP3685PC00

[0336] consisting of autologous-, allogenic-, stem cell derived-, and cell line derived-cells.

[0337] In a seventh embodiment the invention provides the pharmaceutical composition for use as a medicament according to any one of the first to sixth embodiments, for use in the treatment of a cancer or infectious disease in a subject in need thereof.

[0338] In an eighth embodiment the invention provides the pharmaceutical composition for use as a medicament according to the seventh embodiment, wherein the subject has an F / F or F / V allotype of CD16A.

[0339] In a nineth embodiment the invention provides the pharmaceutical composition for use as a medicament according to any one of the first to eighth embodiments, wherein said at least partially non-fucosylated aPD-Ll antibodies are monospecific antibodies having an Fc chain N-glycosylated at an asparagine residue at position 297 of a heavy chain of said monospecific antibodies; wherein position 297 is numbered relative to the first amino acid of the CHI region assigned position 118 of said heavy chain.

[0340] In a tenth embodiment the invention provides the pharmaceutical composition for use as a medicament according to any one of the first to ninth embodiments, wherein said modified aPD-Ll antibodies are bispecific antibodies targeting aPD-Ll and a tumor antigen.

[0341] In an eleventh embodiment the invention provides the pharmaceutical composition for use as a medicament according to the tenth embodiment, wherein the tumor antigen is selected from the group consisting of HER.2, EGFR, VEGF, TA-MUC1, PSA, TFa (core-1), LeY, CEA, EpCAM, CD19, and BCMA.

[0342] In a twelfth embodiment the invention provides the pharmaceutical composition for use as a medicament according to the tenth or eleventh embodiment, wherein said modified aPD-Ll antibodies are fusion polypeptides comprising at least one domain having cytokine activity.

[0343] In a thirteenth embodiment the invention provides the pharmaceutical composition for use as a medicament according to the twelfth embodiment, wherein said at least one domain is selected from the group consisting of IL15, IL2, IL21, GM-CSF and a type I interferon.P3685PC00

[0344] In a fourteenth embodiment the invention provides a method of activating and expanding isolated immune cells selected from the group consisting of NKT cells, iNKT cells, CAR-NKT cells and CAR-iNKT cells, comprising the steps of:

[0345] i. contacting said immune cells with modified aPD-Ll antibodies and dendritic cells in vitro or ex vivo; and

[0346] II. optionally isolating said immune cells obtained in step (i) wherein said modified aPD-Ll antibodies are either:

[0347] at least partially non-fucosylated antibodies comprising 0 - 80% core- fucosylated N-glycans, or

[0348] mutant aPD-Ll antibodies, wherein two or more amino acid residues in a heavy chain of said mutant aPD-Ll antibodies are substituted, and wherein said substitutions are selected from among:

[0349] a. S298A, E333A and K334A;

[0350] b. S239D and I332E;

[0351] c. S239D, A330L and I332E;

[0352] d. G236A, S239D and I332E;

[0353] e. G236A, A330L and I332E;

[0354] f. G236A, S239D, A330L, and I332E;

[0355] g. F243L, R292P, Y300L, V305I, P396L;

[0356] h. L235V, F243L, R292P, Y300L, P396L; and

[0357] i. P247I and A339Q,

[0358] wherein said substitutions are numbered relative to the first amino acid of the CHI region assigned position 118 of said heavy chain.

[0359] In a fifteenth embodiment the invention provides the method of activating and expanding isolated immune cells according to the fourteenth embodiment, wherein the step of contacting is performed in a culture medium that supports the multiplication of the immune cells.P3685PC00

[0360] References:

[0361] Ahmed AA, et al., (2016). Structural characterization of GASDALIE Fc bound to the activating Fc receptor FcyRIIIa. J Struct Biol. 2016 Apr;194(l):78-89. doi:

[0362] 10.1016 / j.jsb.2016.02.001.

[0363] Bournazos et al., (2014) Broadly neutralizing anti-HIV-1 antibodies require Fc effector functions for in vivo activity. Cell. 2014;158:1243-1253. doi:

[0364] 10.1016 / j. cell.2014.08.023

[0365] Brinkmann et al., (2017) "The making of bispecific antibodies" in MAbs

[0366] Volume 9(2): 182-212. doi: 10.1080 / 19420862.2016.1268307.

[0367] Lenart et al., (2017) "Comparison of 6B11 mAb and a-GalCer-loaded CDld dextramers for detection of iNKT cells by flow cytometry" in J Immunol Methods., Volume 446:1-6. doi: 10.1016 / j.jim.2017.03.016.

[0368] Madsen et al., (2022) "Immobilization-free binding and affinity characterization of higher order bispecific antibody complexes using size-based microfluidics" in Anal Chem. 2022 Oct ll;94(40): 13652-13658

[0369] Madsen et al., (2023) Generation of robust bispecific antibodies through fusion of single-domain antibodies on IgG scaffolds: a comprehensive comparison of formats. MAbs. 2023 Jan-Dec;15(l):2189432.

[0370] doi: 10.1080 / 19420862.2023.2189432. PMID: 36939220

[0371] Madsen et al., (2024a) "IgG-VHH bispecific fusion antibodies: challenges and opportunities as therapeutic agents" in Expert Opinion on Biological Therapy. DOI: 10.1080 / 14712598.2024.2336068 PMID: 38544310

[0372] Madsen et al., (2024a) "Design and engineering of bispecific antibodies: insights and practical considerations" in Frontiers in Bioengineering and Biotechnology - Bioprocess Engineering, doi: 10.3389 / fbioe.2024.1352014. PMID: 38333084

[0373] Peng et al., (2024) "CAR-T and CAR-NK as cellular cancer immunotherapy for solid tumors" in Nature: Cellular & Molecular Immunology 21, pagesl089-1108, https: / / doi.org / 10.1038 / s41423-024-01207-0P3685PC00

[0374] Swieboda et al., (2024) "Natural killer cells and innate lymphoid cells but not NKT cells are mature in their cytokine production at birth" in Clinical and Experimental Immunology, Volume 215(1): 1-14, https: / / doi.org / 10.1093 / cei / uxad094

Claims

P3685PC00Claims1. A pharmaceutical composition for use as a medicament, comprising immune cells selected from the group consisting of NKT cells, iNKT cells, CAR-NKT cells and CAR-iNKT cells; and PD-L1 positive dendritic cells; and modified aPD-Ll antibodies,wherein said modified aPD-Ll antibodies are either:at least partially non-fucosylated antibodies comprising 0 - 25% core- fucosylated N-glycans, ormutant aPD-Ll antibodies, wherein two or more amino acid residues in a heavy chain of said mutant aPD-Ll antibodies are substituted, and wherein said substitutions are selected from the group consisting of:a. S298A, E333A and K334A;b. S239D and I332E;c. S239D, A330L and I332E;d. G236A, S239D and I332E;e. G236A, A330L and I332E;f. G236A, S239D, A330L, and I332E;g. F243L, R292P, Y300L, V305I, P396L;h. L235V, F243L, R292P, Y300L, P396L; andi. P247I and A339Q,wherein the heavy chain of said mutant aPD-Ll antibodies comprises an amino acid sequence derivable from a parent human IgGl heavy chain by said substitutions, wherein the corresponding amino acid sequence in said parent human IgGl heavy chain has at least 70% sequence identity to SEQ ID NO.: 10 or 12, andwherein the substitutions are numbered relative to the first amino acid of the CHI region assigned position 118 of said heavy chain.

2. The pharmaceutical composition for use as a medicament of claim 1, wherein the heavy chain of said mutant aPD-Ll antibodies comprises an amino acid sequence derivable from a parent human IgGl heavy chain by said substitutions, wherein the corresponding amino acid sequence in said parent human IgGl heavy chain has at least 80% sequence identity to SEQ ID NO.: 10 or 12.

3. The pharmaceutical composition for use as a medicament of claim 1 or 2, wherein the PD-L1 positive dendritic cells are immature dendritic cells or mature dendritic cells.P3685PC004. The pharmaceutical composition for use as a medicament according to any one of claims 1 - 3, wherein said immune cells are pretreated; said pretreatment comprising culturing said immune cells with said modified aPD-Ll antibodies and said PD-L1 positive dendritic cells.

5. The pharmaceutical composition for use as a medicament according to any one of claims 1 to 4, wherein the PD-L1 positive dendritic cells are MUTZ-3 dendritic cells.

6. The pharmaceutical composition for use as a medicament according to any one of claim 1 - 5, wherein the NKT cells, iNKT cells, CAR-NKT cells and CAR-iNKT cells are selected from the group consisting of autologous-, allogenic-, stem cell derived-, and cell line derived-cells.

7. The pharmaceutical composition for use as a medicament according to any one of claims 1 - 6, for use in the treatment of a cancer or an infectious disease in a subject in need thereof.

8. The pharmaceutical composition for use as a medicament according to claim 7, wherein the subject has an F / F or F / V allotype of CD16A.

9. The pharmaceutical composition for use as a medicament according to any one of claims 1 to 8, wherein said at least partially non-fucosylated aPD-Ll antibodies are monospecific antibodies having an Fc chain N-glycosylated at an asparagine residue at position 297 of a heavy chain of said monospecific antibodies; wherein position 297 is numbered relative to the first amino acid of the CHI region assigned position 118 of said heavy chain.

10. The pharmaceutical composition for use as a medicament according to any one of claims 1 to 8, wherein said mutant aPD-Ll antibodies are monospecific antibodies having an non-glycosylated Fc chain due at an asparagine to alanine residue substitution at position 297 of a heavy chain of said monospecific antibodies; wherein position 297 is numbered relative to the first amino acid of the CHI region assigned position 118 of said heavy chain.

11. The pharmaceutical composition for use as a medicament according to any one of claims 1 - 8, wherein said modified aPD-Ll antibodies are bispecific antibodies targeting aPD-Ll and a tumor antigen.P3685PC0012. The pharmaceutical composition for use as a medicament according to claim 11, wherein the tumor antigen is selected from the group consisting of HER.2, EGFR, VEGF, TA-MUC1, PSA, TFa (core-1), LeY, CEA, EpCAM, CD19, and BCMA.

13. The pharmaceutical composition for use as a medicament according to claim 11 or 12, wherein said modified aPD-Ll antibodies are fusion polypeptides comprising at least one domain having cytokine activity.

14. The pharmaceutical composition for use as a medicament according to claim 13, wherein said at least one domain is selected from the group consisting of IL15, IL2, IL21, GM-CSF and a type I interferon.

15. A method of activating and expanding isolated immune cells selected from the group consisting of NKT cells, iNKT cells, CAR-NKT cells and CAR-iNKT cells, comprising the steps of:i. contacting said immune cells with modified aPD-Ll antibodies and PD- L1 positive dendritic cells in vitro or ex vivo; andII. optionally isolating said immune cells obtained in step (i) wherein said modified aPD-Ll antibodies are either:at least partially non-fucosylated antibodies comprising 0 - 25% core- fucosylated N-glycans, ormutant aPD-Ll antibodies, wherein two or more amino acid residues in a heavy chain of said mutant aPD-Ll antibodies are substituted, and wherein said substitutions are selected from among:a. S298A, E333A and K334A;b. S239D and I332E;c. S239D, A330L and I332E;d. G236A, S239D and I332E;e. G236A, A330L and I332E;f. G236A, S239D, A330L, and I332E;g. F243L, R292P, Y300L, V305I, P396L;h. L235V, F243L, R292P, Y300L, P396L; andi. P247I and A339Q,wherein the heavy chain of said mutant aPD-Ll antibodies comprises an amino acid sequence derivable from a parent human IgGl heavy chain by said substitutions, wherein the corresponding amino acid sequence in said parent human IgGl heavy chain has at least 70% sequence identity to SEQ ID NO.: 10 or 12, andP3685PC00wherein said substitutions are numbered relative to the first amino acid of the CHI region assigned position 118 of said heavy chain.

16. The method of activating and expanding isolated immune cells of claim 15, wherein the step of contacting is performed in a culture medium that supports the multiplication of the immune cells.

17. The method of activating and expanding isolated immune cells of claim 15 or 16, wherein the PD-L1 positive dendritic cells are immature dendritic cells or mature dendritic cells.

18. The method of activating and expanding isolated immune cells of any one of claims 15 to 17, wherein the PD-L1 positive dendritic cells are MUTZ-3 dendritic cells.

Images