immunotherapy

Abstract

The invention relates to immunotherapy, and to invariant natural killer T (iNKT) cells, and iNKT cell immunotherapy. The invention concerns CAR-iNKT cell immunotherapy, in particular, mono- and bi-specific CAR-iNKT cell immunotherapy, as well as pharmaceutical compositions comprising these iNKT cells, and to their use in therapy, as well as therapies and methods for treating, preventing, or ameliorating cancer or infection. The invention also concerns the treatment of various leukaemias, such as acute lymphoblastic leukaemia (ALL), including infant ALL.

Description

[0001]Immunotherapy The present invention relates to immunotherapy, and particularly, although not exclusively, to invariant natural killer T (iNKT) cells, and iNKT cell immunotherapy. The invention is especially concerned with CAR-iNKT cell immunotherapy, in particular, mono- and bi-specific CAR-iNKT cell immunotherapy, as well as pharmaceutical compositions comprising these iNKT cells, and to their use in therapy, as well as therapies and methods for treating, preventing, or ameliorating cancer or infection. The invention is particularly concerned with the treatment of various leukaemias, such as acute lymphoblastic leukaemia (ALL), including infant ALL. Acute leukaemias initiated by rearrangement and fusion of the KMT2A (MLL) gene to a variety of partners are some of the worst prognosis haematological malignancies1. KMT2A-rearranged (KMT2Ar) B cell acute lymphoblastic leukaemia (B-ALL) is the commonest form of infant leukaemia (80%)2and comprises a small fraction of childhood (3%) and adult (10%) leukaemias as well. Compared to the >90% survival of childhood B-ALL, event free survival of infants and children with KMT2Ar B-ALL, partly due to chemo-resistance and relapse, is 38% and 60% respectively. Recent use of the bispecific CD3-CD19 T cell engager blinatumomab appears to improve survival while CD19 CAR-T cell immunotherapy offers promise of rescuing a subset of patients with relapsed disease who would otherwise have dismal prognosis3. Nevertheless, after CD19 CAR-T therapy, one third of patients relapse with either CD19+ or CD19- disease, with a fraction of relapses being due to lineage-switched disease4. The latter describes a state of transcriptional plasticity whereby by means of trans-differentiation or expansion of pre-existing KMT2Ar myeloid / multipotent progenitors, leukemic cells lose their B cell identity and assume a myeloid phenotype associated with partial or complete loss of CD19 expression5; hence escape from CD19 CAR-T. Finally, treatment failure in KMT2Ar ALL is also linked to inability of current therapeutic approaches to effectively tackle or protect from central nervous system disease, usually leptomeningeal infiltration6. There is, therefore, a pressing need to provide improved cancer therapies, particularly for treating leukaemia, such as acute leukaemia and acute lymphoblastic leukaemia (ALL). A characteristic phenotypic feature of KMT2Ar ALL is its expression of PROM1 / CD1337,8, a marker also present on haematopoietic stem and progenitor cells (HSPC)9. Therefore, the inventors considered dual targeting of CD19 and CD133 as potentially being a rational approach to enhance anti-leukemic activity and limit immune escape. Indeed, a pre-clinical approach involving a tandem CD19-CD133 CAR has been described10. However, direct comparison of the bi-specific approach with CD133 CAR-T and an in vivo study of potential haematological toxicity were not performed. It will be appreciated, however, that the dual targeting of CD19 and CD133 represents a proof of concept, and any target antigen may be utilised, including CD33. Moreover, previous pre-clinical data suggest that tandem CARs may even be less effective than bispecific designs11,12, possibly due to steric hindrance mechanisms that interfere with high affinity binding by each CAR to their target antigens. As well as optimal CAR design and target selection, effective immunotherapy would also benefit from availability of powerful and multi-functional effector cells. In this regard, the inventors previously developed the CD1d-restricted, glycolipid-reactive iNKT cells13-15as a versatile ‘off-the-shelf’ platform16-19that lacks risk of aGVHD20,21, has inherent anti-tumour activity that can complement that of the CAR modules and ability, when equipped with CAR, to eradicate brain lymphoma more effectively than CAR-T counterparts18,22. In view of the above problems with current therapies, the inventors have now developed immune cells equipped with exemplar monospecific and bispecific CARs against target antigens to investigate their anticancer activity. Specifically, the inventors have developed iNKT cells equipped with a monospecific anti-CD33 CAR, as discussed in Example 8, and a bispecific anti-CD19 / CD133 CAR, as discussed in Examples 1-4 and shown in Figure 2B, to investigate their effect on NKG2D expression upon co-culture with AML cell lines, and their antileukemic activity against medullary and extramedullary KMT2Ar-ALL. As described in the Examples, a powerful bi-specific CAR strategy, exemplified using a bispecific CD19-CD133 CAR as a proof of concept, harnesses the unique properties of iNKT cells, including their lack of risk of aGVHD when allogeneically sourced and their CD1d-independent inherent anti-cancer activity, to demonstrate the enhanced anti-leukaemia activity of bi-specific CAR-iNKT over both mono-specific CD19 or CD133 CAR-iNKT and also bi-specific CAR-T in faithful and clinically relevant models of KMT2Ar-ALL. In all cases, enhanced avidity of the bi- specific CAR-iNKT cells, correlated with in vivo efficacy. This observed increased avidity of bi-specific over mono-specific CAR-iNKT is particularly important for the effective targeting of CAR antigen-low disease, as the inventors demonstrate in one of their in vivo models, and it can potentially mitigate subsequent CAR target-low relapse often seen in KMT2Ar-ALL as well as other forms of ALL treated with CD19 CAR-T. Furthermore, as described in Example 5 and with reference to Figures 10 and 11, when compared to CAR-T, the inventors also show that higher expression of NKG2D by CAR-iNKT contributes to their higher avidity. Moreover, the inventors have surprisingly demonstrated a novel process of dynamic NKG2D expression by both CD4- and CD4+ iNKT subsets. As shown in Figure 17, this process, that requires and is triggered by CAR engagement, allows subsequent targeting of leukaemia cells with variable expression of CAR targets, including leukaemia cells that show reduced expression or lack expression of both CAR targets in an NKG2D-dependent manner. In other words, the bi-specific CAR-iNKT was surprisingly able to also kill bystander leukaemia cells which did not express (or showed reduced expression levels of) CD19 and CD133, i.e., examples of the target antigens of the bi-specific CAR-iNKT. Importantly, as shown in Figure 18, dynamic expression of NKG2D can also be demonstrated by CAR-iNKT targeting CD33 on acute myeloid leukaemia cells. This mechanism is surprisingly more powerful in CAR-iNKT than CAR-T and would be important, not only for eradication of leukaemia with partial expression of CD133 or CD19 or low expression of both CAR targets, but also for reducing the risk of lineage switch which is associated with partial or complete loss of CD19 and CD133, and emergence of CD33, expression by directly targeting lineage switched blasts. Importantly, expression of NKG2D ligands in lineage-switched blasts remains high or even higher than in the original lymphoblasts, thereby highlighting the importance of NKG2D-mediated anti-leukaemia activity of CAR-iNKT cells. While CD1d expression by leukemic blasts has been reported in KMT2Ar-ALL, here the superior anti-leukemic activity of bispecific CAR-iNKT was shown in CD1d- ALL cells suggesting that CAR-iNKT could exert additional anti-leukaemia activity against CD1d-expressing targets which, as the inventors show, can further be enhanced by the high affinity aGalCer glycolipid ligand. The ability of the bi-specific CAR-iNKT cells to kill bystander cancer cells that lack (or show reduced) expression of the CAR target antigens was completely unexpected. This characteristic is highly advantageous because it combats immune escape, i.e., the ability to kill tumour cells which do not express (or have reduced expression of) the antigens targeted by the CARs as well as those tumour cells which do express the CAR-targeted antigens. Thus, according to a first aspect of the invention, there is provided an immune cell, for use in treating, preventing or ameliorating cancer, wherein the immune cell comprises one or more antigen-binding domain, which targets at least one target antigen which is expressed above a threshold level by a first cancer cell, and the at least one target antigen is expressed below the threshold level by a second cancer cell, wherein the second cancer cell expresses one or more NKG2D stress ligand, and the immune cell expresses one or more NKG2D stress ligand receptor, and wherein the immune cell kills the second cancer cell upon engagement between the one or more NKG2D stress ligand and the one or more NKG2D stress ligand receptor. In a second aspect of the invention, there is provided a method of treating, preventing or ameliorating cancer, the method comprising, administering, or having administrated, to a subject in need of such treatment, a therapeutically active amount of an immune cell, wherein the immune cell comprises one or more antigen-binding domain, which targets at least one target antigen which is expressed above a threshold level by a first cancer cell, and the at least one target antigen is expressed below the threshold level by a second cancer cell, wherein the second cancer cell expresses one or more NKG2D stress ligand, and the immune cell expresses one or more NKG2D stress ligand receptor, and wherein the immune cell kills the second cancer cell upon engagement between the one or more NKG2D stress ligand and the one or more NKG2D stress ligand receptor. Firstly, with reference to Step 1 of Figure 17, the inventors were surprised that the expression levels of the NKG2D stress ligand receptor increase in the immune cell upon engagement between the one or more antigen-binding domain (e.g., a CAR) on the immune cell with the at least one target antigen on the first cancer cell. This was unexpected. Secondly, as shown in Step 2 of Figure 17, they were also surprised that the second (i.e., bystander) cancer cells, which are not expressing the at least one target antigen, are being killed by the immune cell. Therefore, the inventors believe that the expression level of a NKG2D stress ligand receptor may be increased above a threshold level on the immune cell by engaging the one or more antigen-binding domain (e.g., a CAR) with the at least one target antigen on the first cancer cell, and thereby increasing the expression level of the NKG2D stress ligand receptor on the immune cell above the threshold level, and, in so doing, target and kill bystander cancer cells with lower (or undetectable) expression levels of the target antigen. The inventors believe that this is a novel and unexpected finding. Thus, in a third aspect, there is provided a method of increasing the expression level of a NKG2D stress ligand receptor on an immune cell above a threshold level, the method comprising transducing the immune cell with one or more antigen-binding domain, which targets at least one target antigen which is expressed above a threshold level by a first cancer cell, and the at least one target antigen is expressed below the threshold level by a second cancer cell, and engaging the one or more antigen-binding domain with the at least one target antigen on the first cancer cell, and thereby increasing the expression level of a NKG2D stress ligand receptor on the immune cell above the threshold level. The methods described herein may be carried out in vivo, in vitro or ex vivo. In an embodiment, the method comprises targeting the first and / or second cancer cell with the immune cell with the increased expression level of the NKG2D stress ligand receptor. Typically, the method comprises targeting the first and second cancer cell with the immune cell with the increased expression level of the NKG2D stress ligand receptor. In a further embodiment, the method comprises killing the first and / or second cancer cell upon engagement between the one or more NKG2D stress ligand on the or each cancer cell, and the one or more NKG2D stress ligand receptor on the immune cell. The immune cell may be a T-cell. The immune cell may be a Natural Killer T (NKT) cell. Natural killer T (NKT) cells are a heterogeneous group of T cells that share properties of both T cells and natural killer cells. NKT cells are classified into two groups based on differences in T cell receptor (TCR) usage. Type I NKT cells or invariant NKT cells have an invariant TCRα-chain and are readily detectable by α- galactosylceramide-loaded CD1d tetramer, whereas Type II NKT cells have a more diverse TCR repertoire, and their direct identification is more difficult. Thus, typically, the immune cell is an invariant NKT (iNKT) cell. iNKT cells are a rare subset of T cells characterised by an invariant TCRVα24Jα18 nearly always pairing with a TCRVβ11 chain, and are restricted by the non- polymorphic, glycolipid-presenting, MHC-like molecule CD1d. Advantageously, iNKT cells have features of both innate and adaptive immunity and possess effector, as well as immunoregulatory activity. Their notable direct and indirect anti-tumour activity is mediated by CD1d-dependent and CD1d-independent mechanisms, including direct killing of tumour cells, antigen presenting cell (APC) maturation and activation of NK cells and tumour-specific T. iNKT cells comprise CD4+ and CD4- subsets, with the former displaying a Th0 / 2 phenotype while the latter display Th1 polarisation, with higher expression of IFNγ and cytolytic granules and enhanced anti-tumour activity, with CD4-CD8+ iNKT cells having the highest cytotoxic potential. Importantly, in pre-clinical models, iNKT cells deplete immunosuppressive, CD1d-expressing tumour-associated myeloid cells while extensive data, including from the inventors’ lab, show that allogeneic iNKT cells do not cause acute graft-versus-host disease (aGVHD), and hence their development as an ‘off-the-shelf’ immunotherapy platform without need for deletion of the endogenous TCR. Accordingly, the iNKT cell may be a CD4-negative iNKT cell or CD4-positive iNKT cell. The iNKT cell, optionally the CD4-negative invariant natural killer T (iNKT) cell or the CD4-positive iNKT cell, may be enriched, which may be achieved by contacting it with a glycolipid. The glycolipid may be alpha-Galactosylceramide (αGalCer). Advantageously, the therapies described herein harness the unique properties of iNKT cells, including their lack of risk of acute graft vs host disease when allogeneically sourced and their CD1d-independent, inherent anti-cancer activity. In some embodiments, the immune cell is autologously-sourced. However, in a typical embodiment, the immune cell is allogeneically-sourced. In an embodiment, the immune cell may be generated using iNKT cells from a prospective stem cell transplant donor. The at least one target antigen on the first cancer cell may comprise a tumour antigen. The at least one target antigen may comprise polypeptide tumour antigen or a glycoprotein tumour antigen displayed on the surface of the first cancer cell. The tumour antigen may be (a) a full length molecule associated with cancer cells, (b) a homologue or modified form of the same, including a molecule with deleted, added and / or substituted portions, or (c) a fragment of the same. The at last one target antigen may be selected from a list of surface antigens expressed in tumours such as: a testis cancer, melanoma, lung cancer, head and neck cancer, lung cancer, breast cancer, gastrointestinal cancer, bladder cancer, colorectal cancer, pancreatic cancer, lymphoma, leukaemia, multiple myeloma, renal cancer, hepatic cancer, ovarian cancer, gastric cancer, brain cancer and prostate cancer. Such tumour antigens include: BCMA, CS1, CD38, FCRL5, CD19, CD20, CD22, CD30, CD123, CLL1, CD33, FTL3, CD133, CLAUDIN 18.2, NKG2D ligands, TCRVbeta / alpha variable and constant chains, CD5, CD7, B7-H3, EGFR, HER2, EGFR806, Mesothelin, PSCA, MUC1, EpCAM, GD2, VEGFR2, AFP, Nectin4 / FA, PCEA, Lewis Y, Glypican-3, EGFRIII, IL-13Rα2, CD171, MUC16, PSMA, AFP, AXL, c-MET, DLL-3, DR5, EpHA2, FRα gp100, MAGE-A1 / 3 / 4, LMP1 and others. Suitable tumour antigens may also include intracellular proteins which are presented as class I-restricted antigens recognised by CD 8+ lymphocytes or class II-restricted antigens recognised by CD4+ lymphocytes. In this case, the tumour-targeting domain or arm of the iNKT engager would be derived from an antibody that would specifically bind to the MHC class I / II-peptide complex. Such MHC-resented peptides may be derived from: (a) cancer-testis antigens, such as NY-ESO-I, SSX2, SCP-1, as well as RAGE, BAGE, GAGE and MAGE family polypeptides, for example, GAGE-I, GAGE-2, MAGE-I, MAGE-2, MAGE-3, MAGE-4, MAGE-5, MAGE-6, and MAGE-12 (which can be used, for example, to address melanoma, lung, head and neck, NSCLC, breast, gastrointestinal, and bladder tumours); (b) mutated antigens, for example, p53 (associated with various solid tumours, e.g., colorectal, lung, head and neck cancer), p21 / Ras (associated with, e.g., melanoma, pancreatic cancer and colorectal cancer), CDK4 (associated with, e.g., melanoma), MUM- 1 (associated with, e.g., melanoma), caspase-8 (associated with, e.g., head and neck cancer), CIA 0205 (associated with, e.g., bladder cancer), HLA-A2-R1701, beta catenin (associated with, e.g., melanoma), TCR (associated with, e.g., T- cell non-Hodgkins lymphoma), BCR-abl (associated with, e.g., chronic myelogenous leukaemia), triosephosphate isomerase, KIA 0205, CDC-27, and LDLR-FUT; (c) over-expressed antigens, for example, Galectin 4 (associated with, e.g., colorectal cancer), Galectin 9 (associated with, e.g., Hodgkin's disease), proteinase 3 (associated with, e.g., chronic myelogenous leukaemia), WT 1 (associated with, e.g., various leukaemias), carbonic anhydrase (associated with, e.g., renal cancer), aldolase A (associated with, e.g., lung cancer), PRAME (associated with, e.g., melanoma), HER- 2 / neu (associated with, e.g., breast, colon, lung and ovarian cancer), alpha-fetoprotein (associated with, e.g., hepatoma), KSA (associated with, e.g., colorectal cancer), gastrin (associated with, e.g., pancreatic and gastric cancer), telomerase catalytic protein, MUC- I (associated with, e.g., breast and ovarian cancer), G-250 (associated with, e.g., renal cell carcinoma), p53 (associated with, e.g., breast, colon cancer), and carcinoembryonic antigen (associated with, e.g., breast cancer, lung cancer, and cancers of the gastrointestinal tract such as colorectal cancer); (d) shared antigens, for example, melanoma-melanocyte differentiation antigens, such as MART-1 / Melan A, gplOO, MClR, melanocyte-stimulating hormone receptor, tyrosinase, tyrosinase related protein- 1 / TRPl and tyrosinase related protein-2 / TRP2 (associated with, e.g., melanoma); (e) prostate-associated antigens, such as PAP, PSA, PSMA, PSH-Pl, PSM-Pl, PSM-P2, associated with e.g., prostate cancer; and / or (f) immunoglobulin idiotypes (associated with myeloma and B cell lymphomas. In one embodiment, the at least one target antigen expressed by the first cancer cell is CD19. In another embodiment, the at least one target antigen expressed by the first cancer cell is CD133. In another embodiment, the at least one target antigen expressed by the first cancer cell is CD33. Typically, the first cancer cell expresses both CD19 and CD133. Typically, the first cancer cell expresses both CD19 and CD33. Typically, the first cancer cell expresses both CD133 and CD33. Typically, the first cancer cell expresses CD19, CD133, and CD33. As described in the Examples (see Figure 2Aa), a bi-specific CD19 / CD133 CAR and monospecific CD19 and CD133 CARs were used for proof-of-concept. Typically, therefore, the second cancer cell exhibits lower expression of the at least one target antigen, which may be CD19, CD133, and / or CD33, and typically lower than both CD19 and CD133, or both CD19 and CD33, or both CD133 and CD33, or each of CD19, CD33, and CD133. Ideally, the second cancer cell does not express the at least one target antigen. In an embodiment, the threshold level of expression of the at least one antigen on the first cancer cell is a detectable amount using known techniques, and the second cancer cell either expresses the at least one antigen at an amount which is below the threshold level (of the first cancer cell) or at undetectable levels, or no expression occurs in the second cancer cell. The threshold levels of expression of the at least one target antigen (e.g., CD19, CD133, and / or CD33) on the first and second cancer cell may be detected by transcriptome analysis or flow cytometry. For example, the threshold levels of antigen expression may be measured by flow cytometry using unstained and an Ig isotype control as background. Accordingly, cells that have the same intensity of staining as the unstained and Ig isotype control-stained cells are considered to not have detectable expression (e.g., the second cancer cell). However, those that show a staining intensity above the controls are said to express the antigen (e.g., the first cancer cell). Thus, the intensity of staining may be used to classify them as low, medium and high expression of the at least one target antigen. In one embodiment, the amount of the at least one target antigen expressed by the second cancer cell may be at medium, low or undetectable levels compared to the expression of the at least one antigen expressed by the first cancer cell. In an embodiment, the amount of the at least one target antigen expressed by the second cancer cell may be at least 1%, 2%, 3%, 4% or 5% lower than the expression level of the at least one target antigen expressed by the first cancer cell. In an embodiment, the amount of the at least one target antigen expressed by the second cancer cell may be at least 6%, 7%, 8%, 9% or 10% lower than the expression level of the at least one target antigen expressed by the first cancer cell. In an embodiment, the amount of the at least one target antigen expressed by the second cancer cell may be at least 15%, 20%, 25%, 30% or 35% lower than the expression level of the at least one target antigen expressed by the first cancer cell. In an embodiment, the amount of the at least one target antigen expressed by the second cancer cell may be at least 40%, 45%, 50%, 55% or 60% lower than the expression level of the at least one target antigen expressed by the first cancer cell. In an embodiment, the amount of the at least one target antigen expressed by the second cancer cell may be at least 65%, 70%, 75%, 80% or 85% lower than the expression level of the at least one target antigen expressed by the first cancer cell. In an embodiment, the amount of the at least one target antigen expressed by the second cancer cell may be at least 90%, 95%, or 100% lower than the expression level of the at least one target antigen expressed by the first cancer cell. The inventors have observed that their process, which is triggered by engagement between the antigen-binding domain on the immune cell and the antigen on the first cancer cell, allows targeting and subsequent killing of the second cancer cell with variable or low expression of the antigens compared to the first cancer cell, including leukaemia cells that lack expression of both antigens in an NKG2D-dependent manner. This mechanism is powerful in CAR-iNKT cell and is important not only for eradication of leukaemia with partial or low expression of one antigen (e.g., CD133, CD19, or CD33), two different antigens (e.g., CD19 and CD133, CD19 and CD33, or CD133 and CD33), or three different antigens (e.g., CD19, CD133, and CD33) in the second cancer cell compared to their corresponding level(s) in the first cancer cell. Moreover, it would also be expected to reduce the risk of lineage switch which is associated with partial or complete loss of antigen (e.g., CD19, CD133, and CD33) expression by directly targeting lineage switched blasts. The first cancer cell may be comprised in or part of a tumour. The second cancer cell may be comprised in or part of a tumour. The first and second cancer cells may be comprised in or part of the same tumour. The first and second cancer cells may be substantially adjacent to each other. The first and second cancer cells may contact each other or apart from one another. The first cancer cell and / or the second cancer cell may express CD1d. The inventors have shown that the superior anti-cancer activity of CAR-iNKT cells was shown in CD1d-expressing cells suggesting that CAR-iNKT could exert additional anti- leukaemia activity against CD1d-expressing targets which, as shown here, can further be enhanced by the high affinity αGalCer glycolipid ligand. In one embodiment, the one or more antigen-binding domain is configured to specifically target the at least one target antigen on the first cancer cell, optionally CD19. In another embodiment, the one or more antigen-binding domain is configured to specifically target CD133. In another embodiment, the one or more antigen-binding domain is configured to specifically target CD33. The one or more antigen-binding domain is configured to specifically target two or more target antigens on the first cancer cell. Typically, the one or more antigen- binding domain is configured to specifically target both CD19 and CD133. Typically, the one or more antigen-binding domain is configured to specifically target both CD19 and CD33. Typically, the one or more antigen-binding domain is configured to specifically target both CD133 and CD33. Typically, the one or more antigen-binding domain is configured to specifically target CD19, CD133, and CD33. In an embodiment, the antigen-binding domain may comprise a hypervariable Complementarity-Determining Region (CDR), or a functional portion thereof. The term “functional portion” of a CDR can mean a sequence within the CDR which shows specific affinity for the target antigen, which is on the first cancer cell. The term “CDR” can mean a hypervariable region in the heavy and light variable chains of an antibody. There may be one, two, three, or more CDRs in each of the heavy and light chains of the antibody. Normally, there are at least three CDRs on each chain which, when configured together, form the antigen-binding domain, i.e., the three-dimensional combining site with which the antigen binds or specifically reacts. The definition of CDR also includes overlapping or subsets of amino acid residues when compared against each other. The exact residue numbers which encompass a particular CDR, or a functional portion thereof, will vary depending on the sequence and size of the CDR. Those skilled in the art can routinely determine which residues comprise a particular CDR given the variable region amino acid sequence of the antibody. The amino acid sequence boundaries of a CDR can be determined by using any of a number of known numbering schemes, including those described by Kabat et al., supra (“Kabat” numbering scheme); Al-Lazikani et al., 1997, J. Mol. Biol., 273:927-948 (“Chothia” numbering scheme); MacCallum et al., 1996, J. Mol. Biol. 262:732-745 (“Contact” numbering scheme); Lefranc et al., Dev. Comp. Immunol., 2003, 27:55-77 (“IMGT” numbering scheme); and Honegge and Plückthun, J. Mol. Biol., 2001, 309:657-70 (“AHo” numbering scheme). The antigen-binding domain may comprise any antibody fragments produced by protease digestion or reduction of a human monoclonal antibody and by recombinant DNA methods known to those skilled in the art. Human monoclonal antibody functional fragments include, for example, individual heavy or light chains and fragments thereof, such as VL, VH and Fd; monovalent fragments, such as Fv, Fab, and Fab', single domain Ab, bivalent fragments such as F(ab')2, single chain Fv (scFv), bivalent or trivalent scFvs, scFv-Fc-scFv, an IgG-scFv, an IgG-dAb, KIH-IgG, ^^-BODY, KIH-Fc Fab / scFv, tri / tetraspecific, KIH tri / tetra-specific, or a tandem scFv, bispecific antibody (BsAb), and Fc fragments. The Fc fragment of the antibody may be disabled by introducing amino acid substitutions into the Fc region, which silence or reduce the effector function of the antibody. Thus, the antigen-binding domain may comprise a domain selected from the group consisting of: an antibody or antigen-binding fragment thereof, CDR, VL, VH and Fd; Fv, Fab, and Fab'; bivalent fragments such as F(ab')2; single chain Fv (scFv); bivalent or trivalent scFvs, scFv-Fc-scFv, an IgG-scFv, an IgG-dAb, KIH-IgG, ^^-BODY, KIH-Fc Fab / scFv, tri / tetraspecific, KIH tri / tetra-specific, or a tandem scFv, bispecific antibody (BsAb), and Fc fragment. The term “VL fragment” can mean a fragment of the light chain of a human monoclonal antibody which includes all or part of the light chain variable region, including the CDRs. A VL fragment can further include light chain constant region sequences. The term “VH fragment” can mean a fragment of the heavy chain of a human monoclonal antibody which includes all or part of the heavy chain variable region, including the CDRs. The term “Fd fragment” can mean the heavy chain variable region coupled to the first heavy chain constant region, i.e., VH and CH-1. The “Fd fragment” does not include the light chain, or the second and third constant regions of the heavy chain. The term “Fv fragment” can mean a monovalent antigen-binding fragment of a human monoclonal antibody, including all or part of the variable regions of the heavy and light chains, and absent of the constant regions of the heavy and light chains. The variable regions of the heavy and light chains include, for example, the CDRs. For example, an Fv fragment includes all or part of the amino terminal variable region of about 110 amino acids of both the heavy and light chains. The term “Fab fragment” can mean a monovalent antigen-binding fragment of a human monoclonal antibody that is larger than an Fv fragment. For example, a Fab fragment includes the variable regions, and all or part of the first constant domain of the heavy and light chains. Thus, a Fab fragment additionally includes, for example, amino acid residues from about 110 to about 220 of the heavy and light chains. The term “Fab' fragment” can mean a monovalent antigen-binding fragment of a human monoclonal antibody that is larger than a Fab fragment. For example, a Fab' fragment includes all of the light chain, all of the variable region of the heavy chain, and all or part of the first and second constant domains of the heavy chain. For example, a Fab' fragment can additionally include some or all of amino acid residues 220 to 330 of the heavy chain. The term “F(ab')2 fragment” can mean a bivalent antigen-binding fragment of a human monoclonal antibody. An F(ab')2 fragment includes, for example, all or part of the variable regions of two heavy chains-and two light chains, and can further include all or part of the first constant domains of two heavy chains and two light chains. The term “single chain Fv (scFv)” can mean a fusion of the variable regions of the heavy (VH) and light chains (VL) connected with a short linker peptide. The term “bispecific antibody (BsAb)” can mean a bispecific antibody comprising two scFv linked to each other by a shorter linked peptide. In one embodiment, the antigen-binding domain may be a single domain antibody (sdAb) (also referred to as a nanobody). The skilled person would understand that an sdAb is an antibody fragment consisting of a single monomeric variable antibody domain (referred to as a VHH). Alternatively, in another embodiment, the antigen-binding domain is a single-chain antibody, an intrabody, a peptide (e.g., a bicyclic peptide), or any other type of fragment or protein scaffold. The one or more antigen-binding domain may be monospecific and target one target antigen. Alternatively, the one or more antigen-binding domain may be bi-specific and target two target antigens. In some embodiments, the one or more antigen-binding domain may be tri-specific and target three target antigens. In one embodiment, the one or more antigen-binding domain comprises one or more chimeric antigen receptor (CAR). The one or more CAR may be first, second, third or fourth generation CAR. The CAR may be monospecific, bi-specific or tri-specific. The skilled person will appreciate that a CAR comprises an antigen-binding domain, and a hinge region (collectively called the ectodomain), a transmembrane domain (which spans the membrane of the T cell) and an intracellular T cell signalling domain (called the endodomain). The CAR’s antigen-binding domain may comprise a “single chain Fv (scFv)”, i.e., a fusion of the variable regions of the heavy (VH) and light chains (VL) connected with a short linker peptide. The CAR may therefore comprise a hinge, a transmembrane domain, a costimulatory domain 1, a costimulatory domain 2, and / or an activation domain. Embodiments of the CAR are shown in Figure 2A. As described in the Examples, a bi- specific CD19 / CD133 CAR was used for proof-of-concept. The inventors have demonstrated that it is a highly effective immunotherapeutic strategy. The dual antigen-binding domain approach engages two surface antigens that may often be, although not always, highly co-expressed on cancer cells. Typically, therefore, the immune cell, typically an iNKT cell, comprises a bi-specific CD19-CD133 CAR. Typically, the immune cell comprises a bi-specific CD19-CD33 CAR. Typically, the immune cell comprises a bi-specific CD133-CD33 CAR. Thus, the CAR may comprise a CD19 CAR, a CD133 CAR, and / or a CD33 CAR, i.e., monospecific CARs. In an embodiment, the CAR may comprise a CD19 CAR and CD133 CAR, a CD19 CAR and CD33 CAR, or a CD133 CAR and CD33 CAR, i.e., a bispecific CAR. The CAR may comprise a CD8α hinge and transmembrane domain. The intracellular domain may comprise a signalling domain of CD28, a signalling domain of 4-1BB and / or a CD3ζ chain. Typically, the intracellular domain comprises a signalling domain of CD28 and / or a signalling domain of 4-1BB, and a CD3ζ chain. In one embodiment, the immune cell may be transduced with an anti-CD19 single- target CAR. As described in the Examples, the inventors employed the FMC63 antibody, fused with the hinge, transmembrane, and partial intracellular domain of CD8α (amino acids 128-210 of CD8α), followed by the CD28 co-stimulatory domain and CD3ζ activation domain. Thus, the CAR may comprise an antigen-binding domain (e.g., CD19 specific), a hinge, a transmembrane domain, a domain of CD8α, CD28 co- stimulatory domain and CD3ζ activation domain. In another embodiment, the immune cell may be transduced with an anti-CD133 single-target CAR. For this CAR, the inventors employed the AC133 antibody clone, incorporating the same CD8α components as the anti-CD19 CAR, but coupled with the 4-1BB co-stimulatory domain and CD3ζ activation domain. Thus, the CAR may comprise an antigen-binding domain (e.g., CD133 specific), a hinge, a transmembrane domain, a domain of CD8α, 4-1BB co-stimulatory domain and CD3ζ activation domain. In another embodiment, the immune cell may be transduced with an anti-CD33 single-target CAR. For the anti-CD33 target CAR, the inventors employed the CD33 antibody clone hP67.6 (US20230330139A1, the entire contents of which are incorporated herein by reference), fused with the hinge, transmembrane, and partial intracellular domain of CD8α (amino acids 128-210 of CD8α), followed by the CD28 co-stimulatory domain and CD3ζ activation domain. Thus, the CAR may comprise an antigen-binding domain (e.g., CD33 specific), a hinge, a transmembrane domain, a domain of CD8α, a CD28 co-stimulatory domain, and a CD3ζ activation domain. In yet another embodiment, the immune cell may be transduced with a bispecific anti- CD19 / anti-CD133 dual CAR, an anti-CD19 / anti-CD33 dual CAR, or an anti- CD133 / CD33 dual CAR. A bispecific CAR may comprise a suitable linker, which links the single-target CARs. A linker may comprise a T2A peptide sequence. The schematic structures of some of these embodiments of the CAR constructs are illustrated in Figure 2A. Following synthesis of the complete coding sequences, these constructs may be cloned into a suitable transfer plasmid, such as a lentiviral transfer plasmid, typically the pSIEW lentiviral transfer plasmid, using appropriate restriction enzyme digestion and ligation approaches. In a first embodiment, the amino acid sequence of the monospecific CD19 CAR may be represented herein as SEQ ID No: 1, as follows: It will be appreciated that the order of the various domains of the CD19 CAR may be set out as follows: N-terminus-CD8α signal peptide; Accordingly, typically, the one or more CAR comprises an amino acid sequence substantially set out as SEQ ID No: 1, or a fragment or variant thereof. In one embodiment, the nucleic acid sequence of the monospecific CD19 CAR may be represented herein as SEQ ID No: 2, as follows: ATGGCTCTGCCTGTGACAGCTCTGCTGCTGCCTCTGGCCCTGCTGCTGCATGCCGCAAGACCTATGGAGACCG ACACCCTGCTGCTGTGGGTGCTGCTGCTGTGGGTGCCAGGCAGCACCGGCGACATCCAGATGACCCAGACCAC CAGCAGCCTGAGCGCCAGCCTGGGCGACCGGGTGACCATCAGCTGCAGAGCCAGCCAGGACATCAGCAAGTAC CTGAACTGGTACCAGCAGAAGCCCGACGGCACCGTGAAGCTGCTGATCTACCACACCAGCCGGCTGCACAGCG GCGTGCCCAGCCGGTTCAGCGGCAGCGGCAGCGGCACCGACTACAGCCTGACCATCAGCAACCTGGAGCAGGA GGACATCGCCACCTACTTCTGCCAGCAGGGCAACACCCTGCCCTACACCTTCGGAGGCGGCACCAAGCTGGAG ATCACCAAGGCCGGAGGCGGAGGCTCTGGCGGAGGCGGCTCTGAGGTGAAGCTGCAGGAGTCTGGCCCAGGCC TGGTGGCCCCAAGCCAGAGCCTGAGCGTGACCTGCACCGTGAGCGGCGTGAGCCTGCCCGACTACGGCGTGAG CTGGATCAGGCAGCCCCCACGGAAGGGCCTGGAGTGGCTGGGCGTGATCTGGGGCAGCGAGACCACCTACTAC AACAGCGCCCTGAAGAGCCGGCTGACCATCATCAAGGACAACAGCAAGAGCCAGGTGTTCCTGAAGATGAACA GCCTGCAGACCGACGACACCGCCATCTACTACTGCGCCAAGCACTACTACTATGGCGGCAGCTACGCTATGGA CTACTGGGGCCAGGGCACCAGCGTGACCGTGAGCTCGTTCGTGCCTGTGTTTCTGCCTGCCAAGCCCACCACA ACCCCTGCCCCTAGACCTCCTACACCCGCCCCTACAATCGCCAGCCAGCCTCTGTCTCTGAGGCCCGAGGCTT GTAGACCTGCTGCTGGCGGAGCCGTGCACACCAGAGGACTGGATTTCGCCTGCGACATCTACATCTGGGCCCC TCTGGCCGGCACATGTGGCGTGCTGCTGCTGAGCCTCGTGATCACCCTGTACTGCAACCACCGGAACAGAAGC AAGCGGAGCCGGCTGCTGCACAGCGACTACATGAACATGACCCCCAGACGGCCTGGCCCCACCAGAAAGCACT ACCAGCCTTACGCCCCTCCCAGAGACTTCGCCGCCTACCGGTCCAGAGTGAAGTTCAGCAGAAGCGCCGACGC CCCTGCCTATCAGCAGGGCCAGAACCAGCTGTACAACGAGCTGAACCTGGGCAGACGGGAAGAGTACGATGTG CTGGACAAAAGACGTGGCCGGGACCCTGAGATGGGGGGAAAGCCGAGAAGGAAGAACCCTCAGGAAGGCCTGT ACAATGAACTGCAGAAAGATAAGATGGCGGAGGCCTACAGTGAGATTGGGATGAAAGGCGAGCGCCGGAGGGG CAAGGGGCACGATGGCCTTTACCAGGGTCTCAGTACAGCCACCAAGGACACCTACGACGCCCTTCACATGCAG GCCCTGCCCCCTCGCTAA [SEQ ID No: 2] Accordingly, typically, the one or more CAR is encoded by a nucleic acid sequence substantially set out as SEQ ID No: 2, or a fragment or variant thereof. In a second embodiment, the amino acid sequence of the monospecific CD133 CAR may be represented herein as SEQ ID No: 3, as follows: It will be appreciated that the order of the various domains of the CD133 CAR may be set out as follows: N-terminus-CD8α signal peptide; AC133 light chain; GGGGS*2 linker; costimulatory domain; CD3ζ stimulation domain-C-terminus. Accordingly, typically, the one or more CAR comprises an amino acid sequence substantially set out as SEQ ID No: 3, or a fragment or variant thereof. In one embodiment, the nucleic acid sequence of the monospecific CD133 CAR may be represented herein as SEQ ID No: 4, as follows: ATGGCTCTGCCTGTGACAGCTCTGCTGCTGCCTCTGGCCCTGCTGCTGCATGCCGCCAGACCTGACGTGGTGG TGACCCAGACCCCCCTGAGCCTGCCCGTGAGCTTCGGCGACCAGGTGAGCATCAGCTGCAGAAGCAGCCAGAG CCTGGCCAACAGCTACGGCAACACCTACCTGAGCTGGTACCTGCACAAGCCCGGCCAGAGCCCCCAGCTGCTG ATCTACGGCATCAGCAACAGATTCAGCGGCGTGCCCGACAGATTCAGCGGCAGCGGCAGCGGCACCGACTTCA CCCTGAAGATCAGCACCATCAAGCCCGAGGACCTGGGCATGTACTACTGCCTGCAGGGCACCCACCAGCCCTA CACCTTCGGCGGCGGCACCAAGCTGGAGATCAAGGGAGGCGGAGGCTCTGGCGGAGGCGGCTCTCAGGTGCAG CTGCAGCAGAGCGGCGCCGAGCTGGTGAGACCCGGCGCCAGCGTGAAGCTGAGCTGCAAGGCCAGCGGCTACA CCTTCAGCGACTTCGAGATGCACTGGGTGAAGCAGACCCCCGTGCACGGCCTGGAGTGGATCGGCGACATCGA CCCCGGCACCGGCGACACCGCCTACAACCTGAAGTTCAAGGGCAAGGCCACCCTGACCACCGACAAGAGCAGC AGCACCGCCTACATGGAGCTGAGAAGCCTGACCAGCGAGGACAGCGCCGTGTACTACTGCACCCTGGGCGCCT TCGTGTACTGGGGCCAGGGCACCCTGGTGACCGTGAGCGCCTTCGTGCCTGTGTTTCTGCCTGCCAAGCCCAC CACAACCCCTGCCCCTAGACCTCCTACACCCGCCCCTACAATCGCCAGCCAGCCTCTGTCTCTGAGGCCCGAG GCTTGTAGACCTGCTGCTGGCGGAGCCGTGCACACCAGAGGACTGGATTTCGCCTGCGACATCTACATCTGGG CCCCTCTGGCCGGCACATGTGGCGTGCTGCTGCTGAGCCTCGTGATCACCCTGTACTGCAACCACCGGAACAA GAGAGGCAGAAAGAAGCTGCTGTACATCTTCAAGCAGCCCTTCATGAGACCCGTGCAGACCACCCAGGAGGAG GACGGCTGCAGCTGCAGATTCCCCGAGGAGGAGGAGGGCGGCTGCGAGCTGAGAGTGAAGTTCAGCAGAAGCG CCGACGCCCCTGCCTATCAGCAGGGCCAGAACCAGCTGTACAACGAGCTGAACCTGGGCAGACGGGAAGAGTA CGATGTGCTGGACAAAAGACGTGGCCGGGACCCTGAGATGGGGGGAAAGCCGAGAAGGAAGAACCCTCAGGAA GGCCTGTACAATGAACTGCAGAAAGATAAGATGGCGGAGGCCTACAGTGAGATTGGGATGAAAGGCGAGCGCC GGAGGGGCAAGGGGCACGATGGCCTTTACCAGGGTCTCAGTACAGCCACCAAGGACACCTACGACGCCCTTCA CATGCAGGCCCTGCCCCCTCGCTAA [SEQ ID No: 4] Accordingly, typically, the one or more CAR is encoded by a nucleic acid sequence substantially set out as SEQ ID No: 4, or a fragment or variant thereof. In a third embodiment, the amino acid sequence of the bi-specific CD19CD133 CAR may be represented herein as SEQ ID No: 5, as follows: MALPVTALLLPLALLLHAARPDIQMTQTTSSLSASLGDRVTISCRASQDISKYLNWYQQKPDGTVKLLIYHTS RLHSGVPSRFSGSGSGTDYSLTISNLEQEDIATYFCQQGNTLPYTFGGGTKLEITKAGGGGSGGGGSEVKLQE SGPGLVAPSQSLSVTCTVSGVSLPDYGVSWIRQPPRKGLEWLGVIWGSETTYYNSALKSRLTIIKDNSKSQVF LKMNSLQTDDTAIYYCAKHYYYGGSYAMDYWGQGTSVTVSSFVPVFLPAKPTTTPAPRPPTPAPTIASQPLSL RPEACRPAAGGAVHTRGLDFACDIYIWAPLAGTCGVLLLSLVITLYCNHRNRSKRSRLLHSDYMNMTPRRPGP TRKHYQPYAPPRDFAAYRSRVKFSRSADAPAYQQGQNQLYNELNLGRREEYDVLDKRRGRDPEMGGKPRRKNP QEGLYNELQKDKMAEAYSEIGMKGERRRGKGHDGLYQGLSTATKDTYDALHMQALPPRRAEGRGSLLTCGDVE ENPGPMALPVTALLLPLALLLHAARPDVVVTQTPLSLPVSFGDQVSISCRSSQSLANSYGNTYLSWYLHKPGQ SPQLLIYGISNRFSGVPDRFSGSGSGTDFTLKISTIKPEDLGMYYCLQGTHQPYTFGGGTKLEIKGGGGSGGG GSQVQLQQSGAELVRPGASVKLSCKASGYTFSDFEMHWVKQTPVHGLEWIGDIDPGTGDTAYNLKFKGKATLT TDKSSSTAYMELRSLTSEDSAVYYCTLGAFVYWGQGTLVTVSAFVPVFLPAKPTTTPAPRPPTPAPTIASQPL SLRPEACRPAAGGAVHTRGLDFACDIYIWAPLAGTCGVLLLSLVITLYCNHRNKRGRKKLLYIFKQPFMRPVQ TTQEEDGCSCRFPEEEEGGCELRVKFSRSADAPAYQQGQNQLYNELNLGRREEYDVLDKRRGRDPEMGGKPRR KNPQEGLYNELQKDKMAEAYSEIGMKGERRRGKGHDGLYQGLSTATKDTYDALHMQALPPR [SEQ ID No: 5] It will be appreciated that the order of the various domains of the bi-specific CD19CD33 CAR may be set out as follows: N-terminus-CD8α signal peptide; FMC63 light chain and heavy chain; GGGGS*2 terminus. Accordingly, typically, the one or more CAR comprises an amino acid sequence substantially set out as SEQ ID No: 5, or a fragment or variant thereof. In one embodiment, the nucleic acid sequence of the bi-specific CD19CD133 CAR may be represented herein as SEQ ID No: 6, as follows: ATGGCTCTGCCTGTGACAGCTCTGCTGCTGCCTCTGGCCCTGCTGCTGCATGCCGCAAGACCTATGGAGACCG ACACCCTGCTGCTGTGGGTGCTGCTGCTGTGGGTGCCAGGCAGCACCGGCGACATCCAGATGACCCAGACCAC CAGCAGCCTGAGCGCCAGCCTGGGCGACCGGGTGACCATCAGCTGCAGAGCCAGCCAGGACATCAGCAAGTAC CTGAACTGGTACCAGCAGAAGCCCGACGGCACCGTGAAGCTGCTGATCTACCACACCAGCCGGCTGCACAGCG GCGTGCCCAGCCGGTTCAGCGGCAGCGGCAGCGGCACCGACTACAGCCTGACCATCAGCAACCTGGAGCAGGA GGACATCGCCACCTACTTCTGCCAGCAGGGCAACACCCTGCCCTACACCTTCGGAGGCGGCACCAAGCTGGAG ATCACCAAGGCCGGAGGCGGAGGCTCTGGCGGAGGCGGCTCTGAGGTGAAGCTGCAGGAGTCTGGCCCAGGCC TGGTGGCCCCAAGCCAGAGCCTGAGCGTGACCTGCACCGTGAGCGGCGTGAGCCTGCCCGACTACGGCGTGAG CTGGATCAGGCAGCCCCCACGGAAGGGCCTGGAGTGGCTGGGCGTGATCTGGGGCAGCGAGACCACCTACTAC AACAGCGCCCTGAAGAGCCGGCTGACCATCATCAAGGACAACAGCAAGAGCCAGGTGTTCCTGAAGATGAACA GCCTGCAGACCGACGACACCGCCATCTACTACTGCGCCAAGCACTACTACTATGGCGGCAGCTACGCTATGGA CTACTGGGGCCAGGGCACCAGCGTGACCGTGAGCTCGTTCGTGCCTGTGTTTCTGCCTGCCAAGCCCACCACA ACCCCTGCCCCTAGACCTCCTACACCCGCCCCTACAATCGCCAGCCAGCCTCTGTCTCTGAGGCCCGAGGCTT GTAGACCTGCTGCTGGCGGAGCCGTGCACACCAGAGGACTGGATTTCGCCTGCGACATCTACATCTGGGCCCC TCTGGCCGGCACATGTGGCGTGCTGCTGCTGAGCCTCGTGATCACCCTGTACTGCAACCACCGGAACAGAAGC AAGCGGAGCCGGCTGCTGCACAGCGACTACATGAACATGACCCCCAGACGGCCTGGCCCCACCAGAAAGCACT ACCAGCCTTACGCCCCTCCCAGAGACTTCGCCGCCTACCGGTCCAGAGTGAAGTTCAGCAGAAGCGCCGACGC CCCTGCCTATCAGCAGGGCCAGAACCAGCTGTACAACGAGCTGAACCTGGGCAGACGGGAAGAGTACGATGTG CTGGACAAAAGACGTGGCCGGGACCCTGAGATGGGGGGAAAGCCGAGAAGGAAGAACCCTCAGGAAGGCCTGT ACAATGAACTGCAGAAAGATAAGATGGCGGAGGCCTACAGTGAGATTGGGATGAAAGGCGAGCGCCGGAGGGG CAAGGGGCACGATGGCCTTTACCAGGGTCTCAGTACAGCCACCAAGGACACCTACGACGCCCTTCACATGCAG GCCCTGCCCCCTCGCAGAGCCGAGGGCAGAGGCAGCCTGCTGACCTGCGGCGACGTGGAGGAGAACCCAGGCC CCATGGCTCTGCCTGTGACAGCTCTGCTGCTGCCTCTGGCCCTGCTGCTGCATGCCGCCAGACCTGACGTGGT GGTGACCCAGACCCCCCTGAGCCTGCCCGTGAGCTTCGGCGACCAGGTGAGCATCAGCTGCAGAAGCAGCCAG AGCCTGGCCAACAGCTACGGCAACACCTACCTGAGCTGGTACCTGCACAAGCCCGGCCAGAGCCCCCAGCTGC TGATCTACGGCATCAGCAACAGATTCAGCGGCGTGCCCGACAGATTCAGCGGCAGCGGCAGCGGCACCGACTT CACCCTGAAGATCAGCACCATCAAGCCCGAGGACCTGGGCATGTACTACTGCCTCCAGGGCACCCACCAGCCC TACACCTTCGGCGGCGGCACCAAGCTGGAGATCAAGGGAGGCGGAGGCTCTGGCGGAGGCGGCTCTCAGGTGC AGCTGCAGCAGAGCGGCGCCGAGCTGGTGAGACCCGGCGCCAGCGTGAAGCTGAGCTGCAAGGCCAGCGGCTA CACCTTCAGCGACTTCGAGATGCACTGGGTGAAGCAGACCCCCGTGCACGGCCTGGAGTGGATCGGCGACATC GACCCCGGCACCGGCGACACCGCCTACAACCTGAAGTTCAAGGGCAAGGCCACCCTGACCACCGACAAGAGCA GCAGCACCGCCTACATGGAGCTGAGAAGCCTGACCAGCGAGGACAGCGCCGTGTACTACTGCACCCTGGGCGC CTTCGTGTACTGGGGCCAGGGCACCCTGGTGACCGTGAGCGCCTTCGTGCCTGTGTTTCTGCCTGCCAAGCCC ACCACAACCCCTGCCCCTAGACCTCCTACACCCGCCCCTACAATCGCCAGCCAGCCTCTGTCTCTGAGGCCCG AGGCTTGTAGACCTGCTGCTGGCGGAGCCGTGCACACCAGAGGACTGGATTTCGCCTGCGACATCTACATCTG GGCCCCTCTGGCCGGCACATGTGGCGTGCTGCTGCTGAGCCTCGTGATCACCCTGTACTGCAACCACCGGAAC AAGAGAGGCAGAAAGAAGCTGCTGTACATCTTCAAGCAGCCCTTCATGAGACCCGTGCAGACCACCCAGGAGG AGGACGGCTGCAGCTGCAGATTCCCCGAGGAGGAGGAGGGCGGCTGCGAGCTGAGAGTGAAGTTCAGCAGAAG CGCCGACGCCCCTGCCTATCAGCAGGGCCAGAACCAGCTGTACAACGAGCTGAACCTGGGCAGACGGGAAGAG TACGATGTGCTGGACAAAAGACGTGGCCGGGACCCTGAGATGGGGGGAAAGCCGAGAAGGAAGAACCCTCAGG AAGGCCTGTACAATGAACTGCAGAAAGATAAGATGGCGGAGGCCTACAGTGAGATTGGGATGAAAGGCGAGCG CCGGAGGGGCAAGGGGCACGATGGCCTTTACCAGGGTCTCAGTACAGCCACCAAGGACACCTACGACGCCCTT CACATGCAGGCCCTGCCCCCTCGCTAA [SEQ ID No: 6] Accordingly, typically, the one or more CAR is encoded by a nucleic acid sequence substantially set out as SEQ ID No: 6, or a fragment or variant thereof. In a fourth embodiment, the amino acid sequence of the monospecific CD33 CAR may be represented herein as SEQ ID No: 12, as follows: It will be appreciated that the order of the various domains of the CD33 CAR may be set out as follows: N-terminus-CD8α signal peptide; anti-CD33 scFv, hP67.6 clone; Accordingly, typically, the one or more CAR comprises an amino acid sequence substantially set out as SEQ ID No: 12, or a fragment or variant thereof. In one embodiment, the nucleic acid sequence of the monospecific CD33 CAR may be represented herein as SEQ ID No: 13, as follows: ATGGCTCTGCCTGTGACCGCCCTGCTGCTGCCTCTGGCTCTGCTGCTGCACGCCGCTCGGCCTCAAGTGCAGC TGCAGGAGAGCGGAGGAGGTTTGGTGCAGGCTGGCGGTTCCCTACGCCTGTCATGCGCTGCAAGTGGGCGGAC ATTTTCTAGCTACGCCATGGGCTGGTTCCGCCAGGCCCCGGGCAAGGAGCGCGAGTTCGTCGCGGCCATCACC TGGTCTGGCGGCTCCACTTATTACGCGGACAGCGTGAAGGGCCGTTTCACCATTTCCCGCGACAACGCGAAAA ACACGCTCTACCTGCAGATGAATTCTCTGAAGCCCGAAGACACCGCCGTGTACTACTGTGCTGCCATGCTGCT GAGGGGTGGGCTTTACGATTACACCGACTACATCCTGTACAACTATTGGGGCCAGGGCACCCAGGTCACTGTT TCGTCCTTTGTGCCGGTGTTTCTGCCGGCGAAACCGACCACCACCCCGGCGCCGCGCCCGCCGACCCCGGCGC CGACCATTGCGAGCCAGCCGCTGAGCCTGCGCCCGGAAGCGTGCCGCCCGGCGGCGGGCGGCGCGGTGCATAC CCGCGGCCTGGATTTTGCGTGCGATATTTATATTTGGGCTCCTCTTGCTGGTACTTGTGGTGTTCTTCTTCTT TCTCTTGTTATTACTCTGTATTGCAACCATCGCAACCGCAGCAAACGCAGCCGCCTGCTGCATAGCGATTATA TGAACATGACCCCGCGCCGCCCGGGCCCGACCCGCAAACATTATCAGCCGTATGCGCCGCCGCGCGATTTTGC GGCGTATCGCAGCCGCGTGAAGTTTAGCCGGTCCGCCGATGCCCCTGCCTACCAGCAGGGCCAGAATCAGCTG TATAACGAGCTGAATCTGGGCCGGAGAGAGGAGTACGACGTGCTGGATAAGAGGAGGGGAAGAGATCCCGAGA TGGGAGGCAAGCCACGGAGAAAGAACCCCCAGGAGGGCCTGTACAATGAGCTGCAGAAGGACAAGATGGCCGA GGCCTATTCTGAGATCGGCATGAAGGGAGAGAGGCGCCGGGGCAAGGGACACGATGGCCTGTACCAGGGCCTG AGCACCGCCACAAAGGACACCTATGATGCCCTGCACATGCAGGCCCTGCCACCAAGG [SEQ ID No: 13] Accordingly, typically, the one or more CAR is encoded by a nucleic acid sequence substantially set out as SEQ ID No: 13, or a fragment or variant thereof. It will be appreciated that NKG2D is an activating receptor (i.e., a transmembrane protein) belonging to the NKG2 family of C-type lectin-like receptors. NKG2D is encoded by KLRK1 (killer cell lectin like receptor K1) gene which is located in the NK- gene complex (NKC) situated on chromosome 12 in humans. In humans, it is expressed by NK cells, γδ T cells and CD8+αβ T cells. NKG2D recognises as its ligand the induced-self proteins from MIC (e.g., MICA-B) and RAET1 / ULBP (e.g., ULBP1-6) families which appear on the surface of stressed, malignant transformed, and infected cells. The threshold level of the NKG2D stress ligand receptor on the immune cell may be detectable by transcriptome analysis and / or flow cytometry. For example, the threshold levels of NKG2D receptor expression may be measured by flow cytometry using unstained and an Ig isotype control as background. Surprisingly, in an embodiment, engagement between the one or more antigen- binding domain (typically a CAR) on the immune cell, and the at least one antigen (e.g., CD19 / CD133, CD19 / CD33, or CD133 / CD33) on the first cancer cell, causes an increase in the expression of the NKG2D stress ligand receptor on the immune cell. Following engagement between the antigen-binding domain on the immune cell and the antigen on the first cancer cell, the expression level of the NKG2D stress ligand receptor on the immune cell may increase by at least 1%, 2%, 3%, 4% or 5%. In another embodiment, following engagement between the antigen-binding domain on the immune cell and the antigen on the first cancer cell, the expression level of the NKG2D stress ligand receptor on the immune cell may increase by at least 6%, 7%, 8%, 9% or 10%. In another embodiment, following engagement between the antigen-binding domain on the immune cell and the antigen on the first cancer cell, the expression level of the NKG2D stress ligand receptor on the immune cell may increase by at least 15%, 20%, 25%, 30% or 35%. In another embodiment, following engagement between the antigen-binding domain on the immune cell and the antigen on the first cancer cell, the expression level of the NKG2D stress ligand receptor on the immune cell may increase by at least 40%, 45%, 50%, 55% or 60%. In another embodiment, following engagement between the antigen-binding domain on the immune cell and the antigen on the first cancer cell, the expression level of the NKG2D stress ligand receptor on the immune cell may increase by at least 65%, 70%, 75%, 80% or 85%. In another embodiment, following engagement between the antigen-binding domain on the immune cell and the antigen on the first cancer cell, the expression level of the NKG2D stress ligand receptor on the immune cell may increase by at least 90%, 95%, or 100%. In an embodiment, the cancer being treated is a malignancy, which may be a solid tumour or a liquid tumour. In some embodiments, the cancer is myeloid or lymphoid leukaemia, B cell and T cell lymphoma, plasma cell dyscrasia and solid tumours cancer of the brain, respiratory tract, head and neck, or skin etc. In one embodiment, leptomeningeal cancer may be treated. In another embodiment, brain, blood and solid tumour malignancies, be they primary or secondary (i.e., metastasis), may be treated. Typically, the cancer is leukaemia, such as acute lymphoblastic leukaemia (ALL), which may be B cell acute lymphoblastic leukaemia (B-ALL) or T cell acute lymphoblastic leukaemia (T-ALL). Typically, the leukaemia is a KMT2A-rearranged leukaemia. More typically, infant ALL may be treated. Medullary and leptomeningeal leukaemia may also be treated. In another embodiment, the leukaemia may be acute myeloid leukaemia (AML). Advantageously, the therapies described herein address chemoresistance, the propensity to involve the brain and in particular the leptomeninges, and also transcriptional and phenotypic plasticity associated with switching from a lymphoid to a myeloid identity. The inventors have appreciated that infected cells (e.g., with a virus or bacterium) up- regulate NKG2D ligands, and so may also be treated by the invention which results in elevated expression levels of the NKG2D stress ligand receptor on immune cells transduced with the antigen-binding domain. Thus, in other embodiments, infectious disease may be treated, including bacterial or viral diseases, such as HIV infection. It will be appreciated that the immune cell according to the invention (collectively referred to herein as “agents”) may be used in a monotherapy (e.g., the use of the immune cell comprising the one or more antigen-binding domain alone), for treating, preventing or ameliorating cancer or any infectious disease, including viral disease, such as HIV infection. Alternatively, the immune cell comprising the one or more antigen-binding domain according to the invention may be used along with adoptively transferred iNKT cells as an adjunct to, or in combination with, known immunotherapies or for treating cancer or infectious disease. The immune cells according to the invention may be combined in compositions having a number of different forms depending, in particular, on the manner in which the composition is to be used. Thus, for example, the composition may be in the form of a liquid, suitably delivered intravenously to a person in need of treatment. It will be appreciated that the vehicle of medicaments according to the invention should be one which is well-tolerated by the subject to whom it is given. In a preferred embodiment, immune cells according to the invention may be administered to a subject by injection into the blood stream or directly into a site requiring treatment. Injections may be intravenous (bolus or infusion) or subcutaneous (bolus or infusion), or intradermal (bolus or infusion), or directly into tumours. It will be appreciated that the amount of the immune cell that is required is determined by its biological activity and bioavailability, which in turn depends on the mode of administration, the physiochemical properties of the agent, and whether it is being used as a monotherapy or in a combined therapy. The frequency of administration will also be influenced by the in vivo persistence of the agent within the subject being treated. Optimal dosages to be administered may be determined by those skilled in the art, and will vary with the particular agent in use, the strength of the pharmaceutical composition, the mode of administration, and the advancement of the disease being treated, for example cancer, or autoimmune disease. Additional factors depending on the particular subject being treated will result in a need to adjust dosages, including subject age, weight, gender, diet, and time of administration. Thus, according to a fourth aspect of the invention, there is provided a pharmaceutical composition comprising an immune cell and pharmaceutically acceptable vehicle, wherein the immune cell comprises one or more antigen-binding domain, which targets at least one target antigen which is expressed above a threshold level by a first cancer cell, and the at least one target antigen is expressed below the threshold level by a second cancer cell, wherein the second cancer cell expresses one or more NKG2D stress ligand, and the immune cell expresses one or more NKG2D stress ligand receptor, and wherein the immune cell kills the second cancer cell upon engagement between the one or more NKG2D stress ligand and the one or more NKG2D stress ligand receptor. Typically, the pharmaceutical composition is for use in treating, preventing or ameliorating cancer. In a fifth aspect, there is provided a method for preparing the pharmaceutical composition according to the fourth aspect, the method comprising contacting an immune cell with a pharmaceutically acceptable vehicle, wherein the immune cell comprises one or more antigen-binding domain, which targets at least one target antigen which is expressed above a threshold level by a first cancer cell, and the at least one target antigen is expressed below the threshold level by a second cancer cell, wherein the second cancer cell expresses one or more NKG2D stress ligand, and the immune cell expresses one or more NKG2D stress ligand receptor, and wherein the immune cell kills the second cancer cell upon engagement between the one or more NKG2D stress ligand and the one or more NKG2D stress ligand receptor. A “subject” may be a vertebrate, mammal, or domestic animal. Most suitably, the subject is a human being. A “therapeutically effective amount” of the immune cell is any amount which, when administered to a subject, is the amount of agent that is needed to treat the disease being treated, for example cancer, or produce the desired effect. For example, the therapeutically effective amount of the immune cell used may be at least 50, 100, 1000, or 10,000 immune cells. Suitably, at least 100,000, or at least 1,000,000 or at least 10,000,000 engagers is used. A “pharmaceutically acceptable vehicle” as referred to herein, is any known compound or combination of known compounds that are known to those skilled in the art to be useful in formulating pharmaceutical compositions. Suitably, for a successful therapy, the immune cell is prepared and then delivered as a cell suspension, most suitably intravenously. The pharmaceutical vehicle may be a liquid, and the pharmaceutical composition is in the form of a solution. Liquid vehicles are used in preparing solutions, suspensions, emulsions, syrups, elixirs and pressurised compositions. The active agent according to the invention may be dissolved or suspended in a pharmaceutically acceptable liquid vehicle such as water, an organic solvent, a mixture of both or pharmaceutically acceptable oils or fats. The liquid vehicle can contain other suitable pharmaceutical additives such as solubilisers, emulsifiers, buffers, preservatives, sweeteners, flavouring agents, suspending agents, thickening agents, colours, viscosity regulators, stabilisers or osmo-regulators. Suitable examples of liquid vehicles for oral and parenteral administration include water (partially containing additives as above, e.g., cellulose derivatives, suitably sodium carboxymethyl cellulose solution), alcohols (including monohydric alcohols and polyhydric alcohols, e.g., glycols) and their derivatives, and oils (e.g., fractionated coconut oil and arachis oil). For parenteral administration, the vehicle can also be an oily ester such as ethyl oleate and isopropyl myristate. Sterile liquid vehicles are useful in sterile liquid form compositions for parenteral administration. The liquid vehicle for pressurised compositions can be a halogenated hydrocarbon or other pharmaceutically acceptable propellant. Liquid pharmaceutical compositions, which are sterile solutions or suspensions, can be utilised by, for example, intramuscular, intrathecal, epidural, intraperitoneal, intravenous and particularly subcutaneous injection. The agent may be prepared as a sterile solid composition that may be dissolved or suspended at the time of administration using sterile water, saline, or other appropriate sterile injectable medium. It will be appreciated that the invention extends to any nucleic acid or peptide or variant, derivative or analogue thereof, which comprises substantially the amino acid or nucleic acid sequences of any of the sequences referred to herein, including variants or fragments thereof. The terms “substantially the amino acid / nucleotide / peptide sequence”, “variant” and “fragment”, can be a sequence that has at least 40% sequence identity with the amino acid / nucleotide / peptide sequences of any one of the sequences referred to herein, for example 40% identity with the sequences identified herein, and so on. Amino acid / polynucleotide / polypeptide sequences with a sequence identity which is greater than 65%, more suitably greater than 70%, even more suitably greater than 75%, and still more suitably greater than 80% sequence identity to any of the sequences referred to are also envisaged. Suitably, the amino acid / polynucleotide / polypeptide sequence has at least 85% identity with any of the sequences referred to, more suitably at least 90% identity, even more suitably at least 92% identity, even more suitably at least 95% identity, even more suitably at least 97% identity, even more suitably at least 98% identity and, most suitably at least 99% identity with any of the sequences referred to herein. The skilled technician will appreciate how to calculate the percentage identity between two amino acid / polynucleotide / polypeptide sequences. In order to calculate the percentage identity between two amino acid / polynucleotide / polypeptide sequences, an alignment of the two sequences must first be prepared, followed by calculation of the sequence identity value. The percentage identity for two sequences may take different values depending on:- (i) the method used to align the sequences, for example, ClustalW, BLAST, FASTA, Smith-Waterman (implemented in different programs), or structural alignment from 3D comparison; and (ii) the parameters used by the alignment method, for example, local vs global alignment, the pair-score matrix used (e.g., BLOSUM62, PAM250, Gonnet etc.), and gap-penalty, e.g., functional form and constants. Having made the alignment, there are many different ways of calculating percentage identity between the two sequences. For example, one may divide the number of identities by: (i) the length of shortest sequence; (ii) the length of alignment; (iii) the mean length of sequence; (iv) the number of non-gap positions; or (v) the number of equivalenced positions excluding overhangs. Furthermore, it will be appreciated that percentage identity is also strongly length dependent. Therefore, the shorter a pair of sequences is, the higher the sequence identity one may expect to occur by chance. Hence, it will be appreciated that the accurate alignment of protein or DNA sequences is a complex process. The popular multiple alignment program ClustalW (Thompson et al., 1994, Nucleic Acids Research, 22, 4673-4680; Thompson et al., 1997, Nucleic Acids Research, 24, 4876-4882) is a preferred way for generating multiple alignments of proteins or DNA in accordance with the invention. Suitable parameters for ClustalW may be as follows: For DNA alignments: Gap Open Penalty = 15.0, Gap Extension Penalty = 6.66, and Matrix = Identity. For protein alignments: Gap Open Penalty = 10.0, Gap Extension Penalty = 0.2, and Matrix = Gonnet. For DNA and Protein alignments: ENDGAP = -1, and GAPDIST = 4. Those skilled in the art will be aware that it may be necessary to vary these and other parameters for optimal sequence alignment. Suitably, calculation of percentage identities between two amino acid / polynucleotide / polypeptide sequences may then be calculated from such an alignment as (N / T)*100, where N is the number of positions at which the sequences share an identical residue, and T is the total number of positions compared including gaps and either including or excluding overhangs. Suitably, overhangs are included in the calculation. Hence, a most suitable method for calculating percentage identity between two sequences comprises (i) preparing a sequence alignment using the ClustalW program using a suitable set of parameters, for example, as set out above; and (ii) inserting the values of N and T into the following formula:- Sequence Identity = (N / T)*100. Alternative methods for identifying similar sequences will be known to those skilled in the art. For example, a substantially similar nucleotide sequence will be encoded by a sequence which hybridizes to DNA sequences or their complements under stringent conditions. By stringent conditions, the inventors mean the nucleotide hybridises to filter-bound DNA or RNA in 3x sodium chloride / sodium citrate (SSC) at approximately 45ºC followed by at least one wash in 0.2x SSC / 0.1% SDS at approximately 20-65ºC. Alternatively, a substantially similar polypeptide may differ by at least 1, but less than 5, 10, 20, 50 or 100 amino acids from the sequences shown herein. Due to the degeneracy of the genetic code, it is clear that any nucleic acid sequence described herein could be varied or changed without substantially affecting the sequence of the protein encoded thereby, to provide a functional variant thereof. Suitable nucleotide variants are those having a sequence altered by the substitution of different codons that encode the same amino acid within the sequence, thus producing a silent (synonymous) change. Other suitable variants are those having homologous nucleotide sequences but comprising all, or portions of, sequence, which are altered by the substitution of different codons that encode an amino acid with a side chain of similar biophysical properties to the amino acid it substitutes, to produce a conservative change. For example, small non-polar, hydrophobic amino acids include glycine, alanine, leucine, isoleucine, valine, proline, and methionine. Large non-polar, hydrophobic amino acids include phenylalanine, tryptophan and tyrosine. The polar neutral amino acids include serine, threonine, cysteine, asparagine and glutamine. The positively charged (basic) amino acids include lysine, arginine and histidine. The negatively charged (acidic) amino acids include aspartic acid and glutamic acid. It will therefore be appreciated which amino acids may be replaced with an amino acid having similar biophysical properties, and the skilled technician will know the nucleotide sequences encoding these amino acids. All of the features described herein (including any accompanying claims, abstracts and drawings), and / or all of the steps of any method or process so disclosed, may be combined with any of the above aspects in any combination, except combinations where at least some features and / or steps are mutually exclusive. For a better understanding of the invention, and to show how embodiments of the same may be carried into effect, reference will now be made, by way of example, to the accompanying Figures, in which:- Figure 1 shows a. Representative flow-cytometry plots showing variable CD133 and CD1d co-expression in CD19+ blasts from 3 primary KMT2Ar ALL patient samples (chALL; childhood ALL, iALL: infant ALL) and a human fetal liver (FL) HSPC derivedCRISPRKMT2A-AF4 ALL model. b. Left: Percentage of CD19+ blasts that express CD133 in KMT2Ar iALL (n=10), KMT2Ar chALL (n=5), andCRISPRKMT2A-AF4 ALL (n=10). Right: Percentage of CD19+ blasts that express CD1d in KMT2Ar iALL (n=4), KMT2Ar chALL (n=5), andCRISPRKMT2A-AF4 ALL (n=4). c. Data shown as percentage of primitive CD34+38- progenitors and CD34+CD19+ B-progenitors that express CD133 from fetla liver (FL, n=8) and fetal bone marrow (FBM, n=8). Figure 2A shows the design and in vitro validation of bispecific CD19-CD133 CAR- iNKT cells. a. Design of CD19 and CD133 mono-specific and CD19-CD133 bispecific CARs. b. Flow-cytometric analysis of highly pure, expanded iNKT cells, gated on CD3 and co-expressing invariant TCRVa24 (identified with the 6B11 mAb) and TCRVb11, expressing the indicated CARs, as assessed by L-protein staining. c. Cytotoxicity assay of indicated mono- and bi-specific CAR-iNKT against SEM cells at 4 and 24 hrs. d. Flow-cytometric analysis of CD19 and CD133 expression in parental and gene-edited SEM leukaemia cells. e. Cytotoxicity assay of indicated mono- and bi-specific CAR- iNKT against the target cells shown in d. f. Flow chamber avidity assay of indicated CAR-iNKT over a range of pressures (left) and representative images of CAR-iNKT bound to SEM cells at the beginning and end of the assay (right). Figure 2B illustrates one embodiment of a CAR-immune cell according to the invention. In this embodiment, the CAR-immune cell is a bispecific CAR-iNKT cell. Anti-CD19 CARs and anti-CD133 CARs are expressed by the iNKT cell. These antigens (i.e., CD19 and CD133), as well as CD1d, are expressed by MLLr ALL leukaemia cells. Figure 3 shows eradication of medullary, extramedullary and meningeal KMT2Ar ALL by bispecific CAR-iNKT. a & b. Leukaemia burden as assessed by bioluminescence imaging (BLI) and overall survival of SEM leukaemia-bearing mice treated with 106mono- or bi-specific CAR-iNKT. c-e. Representative BLI images, leukaemia burden and overall survival in mice treated with 5x106bispecific CAR-iNKT 6 days after iv leukaemia cell transfer. f-h. Representative BLI images, leukaemia burden and overall survival in mice treated with 107bispecific CAR-iNKT 12 days after iv leukaemia cell transfer. i. Left: Representative Immunohistochemistry against CD19 of brains of mice treated or not on days 6 with bispecific CAR-iNKT cells. Right. CNS infiltration scores in untreated and treated mice sacrificed on day 61. j. Meningeal infiltration scores in untreated mice sacrificed on days 16 or 25 and bi-specific CAR-iNKT treated mice sacrificed on day 35 Figure 4 shows a. Growth curve of CD19, CD133 and CD19-CD133 CAR-INKT. b. Right: Cytotoxicity of mono- and bi-specific CAR-iNKT against the CD19+CD133+ KMT2Ar RS4;11 leukaemia cell line (left). c. In vivo activity of the mono-specific CD133 CAR-iNKT in NSG mice engrafted with Luc-dsRED SEM cells.5x106CAR-iNKT were iv transferred or not on day 6 post-leukaemia cell injection. Leukaemia burden as assessed by BLI is shown. d. An SEM subline propagated in vivo in treatment-free mice. Variable co-expression of CD19 and CD133. e & f. Leukaemia burden and survival in mice first injected with the SEM subline shown in c followed by treatment with indicated numbers of bi-specificCD19-CD133 CAR-iNKT. g. Representative flow- cytometric analysis of BM and spleen in sacrificed mice treated with bi-specific CAR- iNKT. Figure 5 shows a. In vitro growth curve of parental and gene-edited SEM cells. b. Frequency of SEM cells with the indicated phenotype in the BM, spleen and liver of mice engrafted with parental SEM cells co-expressing CD19 and CD133 admixed with CD19-CD133+ and CD19+CD133- SEM cells at an 8:1:1 ratio respectively and treated with 106mono- or bi-specific CAR-iNKT cells (n=5-7 mice per group). With reference to untreated controls, frequency of CD19-CD133+ and CD19+CD133- SEM cells is not significantly different amongst the mono-specific CAR-iNKT-treated animals; with the only exception being a significant higher fraction of CD19-CD133+ cells in the spleens of animals treated with CD19 CAR-iNKT (p=0.03 vs untreated control; one-way ANOVA) c. Overall survival of mice treated as described in b. Figure 6 shows a. Flow-cytometric identification of SEM cells after staining with anti- HLA-ABC mAb. Right: cumulative data of leukaemia burden in BM on the indicated timepoints. b. Flow-cytometric examples for SEM (top) and iNKT (bottom) from both d6 and d12 high dose CAR-iNKT treated animals culled at day 60 and 55 respectively. d. Leukaemia burden assessed by BLI (left) and overall survival (right) of mice treated with bispecific CAR-iNKT on day 16 after leukaemia transfer. e. Representative immunophenotypic analysis of leukaemia in BM and spleen in untreated CAR-iNKT- treated mice from each group. Figure 7 shows CAR-iNKT outperform CAR-T against MLLr ALL. a. Cytotoxic activity at 24hrs of untransduced T and iNKT and of their bispecific CAR-transduced counterparts against parental SEM cells and their single or double CD19 and CD133 gene edited sublines. b. Expression level of indicated cytokines by bi-specific CAR-T and CAR-iNKT after 4 and 2hr co-colture with SEM leukaemia cells. c. Avidity assay comparing bi- specific CAR-T vs CAR-iNKT against SEM cells. d&e. Leukaemia burden by BLI and overall survival of SEM leukaemia-bearing mice treated with indicated dose of bi- specific CAR-T and CAR-iNKT cells. f&g. Absolute numbers of CAR-T and CAR-iNKT cells in bone marrow and spleen at sacrifice of mice treated as shown in d&e. Figure 8 shows Bi-specific CD19-CD133 CAR-iNKT cells eradicate primary KMT2Ar ALL including of ALL cells lacking CAR target expression a. CD19 and CD133 co-expression pattern inCRISPRKMT2A-AFF1 cells. b. Avidity assay of mono- and bi-specific CAR-iNKT againstCRISPRKMT2A-AFF1 cells. c. Overall survival ofCRISPRKMT2A-AFF1 leukaemia- bearing mice treated with 106bi-specific CAR-T and CAR-iNKT and 5x106CAR-iNKT. d. iNKT numbers as identified by CD3 expression in BM and spleen of untreated and 106treated CAR-T and CAR-iNKT. e. Representative flow-cytometric analysis ofCRISPRKMT2A-AFF1 leukaemia cells as identified with HLA-ABC staining in bone marrow and spleen of leukaemia-bearing mice treated with 106or 5x106bi-specific CAR-iNKT. f. cumulative data of e. g. CD19 and CD133 co-expression pattern in PDX KMT2A-AFF1 cells as compared to SEM cells. h. Avidity assays of indicated effectors against PDX KMT2A-AFF1 leukaemia cells i. Leukaemia burden in bone marrow and spleen of untreated and 5x106bispecific CAR-iNKT-treated PDX KMT2A-AFF1-leukaemia bearing mice 16 days post treatment (n= 5 mice). Representative flow-cytometric analysis and cumulative data. Figure 9 shows a. Bispecific CAR transduction of T and iNKT cells from the same donor. b. Cytotoxic activity at 4 and 24hrs of untransduced T and iNKT and of their bispecific CAR-transduced counterparts against parental RS4;11 leukaemia cells. c. Representative flow-cytometric analysis of cytokine production by CAR-iNKT and CAR- T after their 4 and 24hr co-culture with SEM cells. d. CD19 and CD133 mono-specific CAR-transduced and untransduced T and iNKT against the parental SEM and RS4;11 cells and the CD19+CD133- KOPN8 KMT2Ar cells. e. CD1d surface expression as assessed by flow-cytometry in SEM cells. f. Cytotoxicity assay of indicated effectors against SEM and CRISPR-generated MLL-AF4 leukaemia cells. g. Representative flow- cytometric example of day 6 bloodCRISPRKMT2A-AFF1 cells identified as HLA- ABC+CD19+CD133+. h. Cytotoxicity assay of indicated effectors against the PDX KMT2A-AFF1 cells. i. Frequency of HLA-ABC cells in the peripheral blood of mice injected with PDX ALL cells as assessed by flow-cytometry on day 14. Figure 10 shows NKG2D-dependent anti-leukemic activity of CAR-iNKT. a. Flow- cytometric analysis of NKG2D expression on bispecific CD19-CD133 CAR-T and -iNKT cells before and after overnight co-culture with SEM cells. b. NKG2D expression as measured by per cent of cells and mean fluorescence intensity in SEM and RS4;11 MLLr cells. c. NKG2DL expression in the indicated MLLr leukaemia cell lines as assessed by staining with biotinylated NKG2D-Fc protein followed by fluorescent streptavidin. d & e. Cytotoxicity of bispecific CAR-T and-iNKT that had been pre- cultured with SEM cells against SEM cells in the presence of different concentrations (d) or 5mg of NKG2D mAb or Ig isotype control (e). f. Avidity measurement of bispecific CAR-iNKT vs CAR-T against SEM cells in the presence of NKG2D mAb or Ig isotype control. g. Left: Schematic of experiment. Right: NKG2D expression (% and MFI) in T / iNKT and CAR-T / iNKT before co-culture with leukaemia cells. h. Left: Schematic of experiment. Right: NKG2D expression after co-culture with the different SEM cell lines as shown. i. Left: Schematic of experiment. Middle: Cytotoxicity of bispecific CAR-T and-iNKT that had been pre-cultured with CD19+CD133+ SEM cells against CD19-CD133- SEM cells in the presence of NKG2D mAb or Ig isotype control. Right: Cytotoxicity of bispecific CAR-T and -iNKT that had been pre-cultured with CD19-CD133- SEM cells against CD19-CD133- SEM cells in the presence of NKG2D mAb or Ig isotype control. Figure 11 shows a. mRNA expression of NKG2D ligands in SEM and RS4;11 cells as assessed by RNA-seq. b.24hr cytotoxicity of bispecific CAR-T and-iNKT that had been pre-cultured with SEM cells against SEM cells in the presence of 10mg of NKG2D-Fc protein or PBS control. c. Representative examples of the FACS analysis of data shown in Figure 4h. e. Representative flow-cytometric examples of NKG2D expression on CAR-iNKT / T pre-cultured with parental and gene-edited SEM cells as shown in Figure 4f&h. d. Upregulation of NKG2D in CAR-T vs CAR-iNKT after 24hr co-culture withCRISPRKMT2A-AFF1 leukaemia cell. e. Left: Bi-specific CAR-iNKT pre-cultured withCRISPRKMT2A-AFF1 leukaemia cells are subsequently more cytotoxic than CAR-T against CAR target-negative CD19-CD133-SEM cells in an NKG2D-dependent manner. Right: Cytotoxicity of CAR-iNKT / T that had not been pre-cultured withCRISPRKMT2A-AFF1. f. mRNA expression of indicated genes as assessed by RNA-seq of paired presentation KMT2Ar lymphoblasts and relapse myeloid blasts. Figure 12 shows Ex vivo transcriptome analysis of bi-specific CAR-iNKT cells. a. Immunophenotypic profiling of pre-infusion CAR-iNKT cells and design of the experiment. b. UMAP embedding of gene expression data after batch correction coloured by cluster ID (top) and by sample (bottom). c. Volcano plot showing differential gene expression between clusters 0 & 1 (log2FC>0.5 and padj< 0.05) d. Bubble plot showing comparative expression levels of genes of immunological relevance in clusters 0,1&2. e. Pathway enrichment analysis in genes over-expressed in C0 over C1 with reference to MSigDB Hallmarks genesets. f. Volcano plot showing differential gene expression between cells isolated from the bone marrow of leukaemia-bearing- vs leukaemia-free mice on day 15. g. Pathway enrichment analysis in genes over-expressed in iNKT isolated from the bone marrow of day 15 leukaemia-bearing mice compared to those from leukaemia-free mice.. h. Frequency of cells in different clusters over the time course of the experiment. Figure 13 shows a. Numbers of iNKT and SEM cells in the bone marrow of mice receiving bispecific CAR-iNKT only or SEM leukaemia cells and CAR-iNKT. b. Table showing the number of cells in which TCR transcripts were identified. c. Overlay of TRA expressing cells on the UMAP map. d. Clone size of the invariant TCRVa24Ja18 clonotype. e. Volcanoplots showing differential gene expression in iNKT cells recovered from the different experimental subgroups. f. Relative expression of indicated gene in the different experimental subgroups. g. Cell cycle signatures projected on UMAP. Figure 14 shows a. Left: Flow-cytometric identification of iNKT cells in the BM at sacrifice of animals described in Figure 2a&b. Right: cumulative data for a. b. Schematic of expansion and functional analysis of iNKT cells form a. c. Purity and CAR expression by iNKT of day 9 post ex vivo selection and expansion. d. Cytotoxic activity at 4 and 24hr of day 16 ex vivo expanded CAR-iNKT against parental SEM, SEM with variable co-expression of CD19 and CD133 (BM; Figure 2c) and gene-edited SEM cells lacking expression of CD19 and CD133. e.4 and 24hr cytotoxicity assay with day 23 CAR-iNKT against the CD19+CD133- C1R and C1R-CD1d cells in the presence or not of aGalCer. Figure 15 shows impact of bispecific CAR-iNKT on haematopoiesis in humanised mice a. Schematic of xenograft experiment to test haematological toxicity of CD19 / CD133 CAR-iNK-T in vivo. b. Peripheral blood human CD45 engraftment levels in mice treated with PBS (n=5) or 107CD19 / CD133 CAR-iNKT (n=7). c) Peripheral blood human CD19+ cells in mice treated with PBS (n=5) or 107CD19 / CD133 CAR-iNKT (n=7), in two separate experiments, showing a transient drop in B cell proportions 1-3 days post CAR injection. d. Long term engraftment in the bone marrow of mice treated with PBS or 107CD19 / CD133 CAR-iNKT: from left to right: proportion of hCD45 cells in BM (PBS, n=4, CAR, n=6); B cells, T cells and myeloid cells in the BM expressed as proportion of hCD45+ cells (PBS, n=4, CAR, n=6); immature CD34+ cells in the BM expressed as proportion of hCD45+ cells (PBS, n=4, CAR, n=5). Figure 16 shows a. Representative flow plots showing gating strategy used to determine peripheral blood (PB) and bone marrow (BM) engraftment and lineage output. The data shown is from BM at cull for PBS treated (left) and CAR-iNKT treated (right) mice. b. representative flow plots showing frequency of CD19+ B cells and CD3+ T cells in the PB of control mice treated with PBS (top row) and mice treated with CAR-iNKT at 9 weeks (bottom row). Data shown as % of hCD45+ cells. c. representative flow plots showing gating strategy used to determine immature CD34+ cells in the BM at cull. d. proportion of hCD45 cells in BM of secondary engrafted NSG mice. Each mouse was injected with 1M whole BM from primary engrafted mice treated with PBS (n=4) or 10M CAR (n=4). Figure 17 shows a schematic diagram illustrating the hypothesised mechanism of action for killing bystander tumour cells by the CAR-immune cell according to the invention. As shown, an immune cell (IC), such as an iNKT cell, expresses a CAR, for example for targeting the CD19 antigen. The immune cell can express multiple monospecific CARs, or bispecific CARs by virtue of appropriate linkers, but just CD19 is shown in the figure for simplicity. The immune cell also expresses the NKG2D receptor. A tumour cell (TC) expresses the same CD19 antigen for the CAR, and also the NKG2D ligand specific for the NKG2D receptor. Upon engagement between the CAR (anti-CD19) on the immune cell and the antigen on the tumour cell (CD19), the expression level of NKG2D receptor on the immune cell surprisingly increases. The immune cell therefore now expresses an elevated concentration of NKG2D receptors on its surface, which enable it to bind to the NKG2D ligands expressed on different tumour cells which do not express the CAR antigen (i.e., CD19), and can thereby kill those bystander tumour cells, and so address immune evasion. Figure 18 shows dynamic expression of NKG2D on CD33 CAR-iNKT. Expression of NKG2D as assessed by flow cytometry on CD33 CAR-iNKT cells with and without overnight exposure to the THP1 and KMTA2r MV4-11 AML cells. Two different iNKT donors were used. Examples Current therapies, including autologous CAR-T immunotherapy, fail to cure half of infants with KMT2A-rearranged acute lymphoblastic leukaemia (KMT2Ar-ALL). Here, the inventors deploy allogeneic iNKT cells, ‘innately’ more powerful effectors than T cells, and equip them with CD19- and / or CD133-targeting CARs as a proof of concept. The inventors show that compared to mono-specific counterparts and bispecific CAR-T, CD19-CD133 bispecific CAR-iNKT not only have enhanced anti-leukaemia activity, but are also able to eradicate medullary and leptomeningeal leukaemia and induce sustained remissions without discernible haematologic toxicity. Critically, CAR-iNKT outperform CAR-T and target CAR antigen-negative leukaemia cells through a mechanism that requires dynamic, CAR- and CAR target-dependent upregulation of the activating innate receptor NKG2D and its engagement by its ligands in KMT2Ar- ALL cells. Indeed, the inventors have also shown that co-culturing CD33 CAR-iNKT with two AML cell lines results in upregulation of NKG2D. Thus, by engaging with two different types of leukaemia-associated targets, CAR-iNKT provide a powerful platform for the treatment of KMT2Ar-ALL. This approach can be readily adapted for other high- risk malignancies, including those with otherwise difficult to target leptomeningeal and brain involvement. Materials and Methods Primary human samples iNKT and T cells. Peripheral blood mononuclear cells (PBMCs), obtained from healthy donors were isolated by density gradient centrifugation and used as a source of CD3+ lymphoid cells for CAR engineering. Human fetal haematopoietic stem and progenitor cells: These were provided for the purposes of this research by the Human Developmental Biology Resource (HDBR, www.hdbr.org), regulated by the UK Human Tissue Authority (HTA, www.hta.gov.uk) and covered under ethics granted by NHS Health Research Authorities: North East – Newcastle & North Tyneside Research Ethics Committee (REC: 23 / NE / 0135) and London - Fulham Research Ethics Committee (23 / LO / 0312). Informed consent was obtained from all participants, who donated human fetal tissue for research without receiving any monetary compensation. FL samples used for CRISPR / Cas9 KMT2A-AF4 translocation experiments underwent CD34 magnetic bead selection at the time of sample processing and were cryopreserved for future use as described(1). Cord blood samples were obtained from NHSBT under ethical approval (REC: 21 / LO / 0195) and CD34 cells were selected as above. Leukaemia samples: ALL patient samples were obtained from VIVO Biobank, UK after appropriate review of our research project to ensure that it was covered under their ethics approval granted by NHS HRA South West - Central Bristol Research Ethics Committee (REC: 23 / EM / 0130). Two infant KMT2Ar ALL samples were obtained from Our Lady’s Children’s Hospital, Crumlin, Dublin, Ireland and one from Oxford University Hospital Trust under ethical approval (REC: 21 / LO / 0195). Informed consent was obtained from all participants or those with parental responsibility, and participants did not receive any monetary compensation. Infant and paediatric KMT2Ar ALL samples from patients being treated at Great Ormond Street Hospital for Children, London were analysed by flow cytometry as part of their diagnostic workup after informed consent. All patient samples / data were anonymised at source, assigned a unique study number and linked. Patient derived xenografts: KMT2Ar ALL PDX cells were provided by the Halsey lab (Glasgow). Primary HSPC. Donated Cord blood sample were obtained from NHSBT and covered under ethical approval (REC: 21 / LO / 0195). Donated fetal tissue was provided for purposes of this research by the Human Developmental Biology Resource (HDBR, www.hdbr.org), regulated by the UK Human Tissue Authority (HTA, www.hta.gov.uk) and covered under ethics (REC: 18 / NE / 0290 and 18 / LO / 0822). PDX ALL BM cells were obtained from the Halsey Lab, University of Glasgow. KMT2A- AFF1 model. Model cells of infant B- acute lymphoblastic leukaemia were derived by introducing a KMT2A -AFF1 translocation into CD34 selected human fetal liver cells using CRISPR and transplanted into NSG mice for of in vivo expansion as described in Rice et al 2021. Cell lines SEM, RS4;11 and KOPN8 cell lines were available in the Milne lab. SEM cell line was further modified in the following ways: SEM cells were transduced with non-replicative MIGR1 retrovirus to co-express Luciferase and dsRed as an expression marker. All cell lines were purchased and maintained under recommended conditions. SEM cells were cultured in IMDM (Gibco) supplemented with 10% FCS. RS4;11cells were cultured in RPMI (Gibco) supplemented with 10% FCS. KOPN-8 cells were cultured in RPMI (Gibco) supplemented with 10% FCS. CRISPR / Cas9 gene editing of SEM cells CD19 and PROM1 knockouts were performed in cell lines using pools of targeted CRISPR RNP guides (Table 1) purchased from Synthego. Following the recommended procedures and concentrations for the SEM cells using the Neon Transfection System (Invitrogen: MPK1096) and Cas9 Nuclease V3 (IDT: 108105 (2)9). Knockout cells recovered in standard culture conditions for 3-7 days post knockout before FACS was performed to select for complete knockout before continuing on to additional experiments. Table 1: Guides for CRISPR knockout CD19-28932346 CACAGCGUUAUCUCCCUCUG [SEQ ID No: 7] CD19+28932373 CGCUGUGCUGCAGUGCCUCA [SEQ ID No: 8] PROM1-16075877 UCGGCUCCCUGUUGCUGCUG [SEQ ID No: 9] PROM1+16075896 GCAACAGGGAGCCGAGUACG [SEQ ID No: 10] PROM1-16075896 CUAGCUAUGGCCCUCGUACU [SEQ ID No: 11] CAR constructs For the anti-CD19 single-target CAR, the inventors employed the FMC63 antibody clone, fused with the hinge, transmembrane, and partial intracellular domain of CD8α (amino acids 128-210 of CD8α), followed by the CD28 co-stimulatory domain and CD3ζ activation domain. The anti-CD133 single-target CAR was constructed using the AC133 antibody clone, incorporating the same CD8α components as the anti-CD19 CAR, but coupled with the 4-1BB co-stimulatory domain and CD3ζ activation domain. For the anti-CD33 target CAR, the inventors employed the CD33 antibody clone hP67.6 (US20230330139A1), fused with the hinge, transmembrane, and partial intracellular domain of CD8α (amino acids 128-210 of CD8α), followed by the CD28 co-stimulatory domain and CD3ζ activation domain. The bispecific CD19-CD133 dual CAR was engineered by linking the aforementioned single-target CARs via a T2A peptide sequence. The schematic structures of the anti-CD19 and anti-CD133 CAR constructs are illustrated in Figure 2a. Following synthesis of the complete coding sequences, these constructs were cloned into the pSIEW lentiviral transfer plasmid using appropriate restriction enzyme digestion and ligation approaches. CAR-iNKT and CAR-T generation TCRVα24Jα18+ lymphocytes were immunomagnetically sorted from PBMC using anti- human iNKT cell microbeads (Miltenyi Biotec). Purified iNKT cells were seeded in 24- or 48-well plates at a 1:1 ratio with irradiated (3500 rad) autologous mononuclear cells (iAPC) and activated with Dynabeads Human T-Activator CD3 / CD28 (Gibco™) at a 1:1 beads-to-cell ratio in T cell medium at a density of 1-5 x 104cells per ml. IL-15 (Miltenyi Biotec) at 30 IU / ml and 150 IU / ml was added at the time of seeding and 12 hours later, respectively. Within 48 hours, activated iNKT cells were transduced with concentrated CAR lentivirus using an MOI of 2-5 in the presence of 8 μg / ml pre-coated retronectin, with spinoculation for 90 minutes at 1000G. After 8-12 hours, cells were resuspended in fresh medium supplemented with 150 IU / ml of IL-15 and allowed to rest for 4 days before assessment of viability and CAR expression. CAR transduction efficiency was determined by flow cytometry as the percentage of L-protein+ cells relative to untransduced controls. CAR+ cells were re-stimulated with a 1:1 ratio of irradiated C1R-CD1d cells loaded with αGalCer (100 ng / ml), IL-15 (30 IU / ml), and with an additional 150 IU / ml of IL-15 added 12-24 hours later. Subsequently, cells were expanded for 14 to 35 days, assessed for purity, and used for in vitro assays. Alternatively, CAR iNKT cells were harvested during the exponential growth phase, cryopreserved in 10% DMSO, and stored in liquid nitrogen until use. Untransduced iNKT were generated in the same way, with the omission of lentiviral transduction step. CAR-T were generated as previously described (3, 4). Flow cytometry. Cells were stained with fluorophore-conjugated monoclonal antibodies in PBS with 2% FBS and 1mM EDTA for 30 minutes and analysed using BD LSR II / Fortessa X50 or FACS sorted using BD Aria instruments using BD FACSDiva software (v8.0.2). Antibodies used are detailed in Table 2. Flow cytometry antibodies were validated by titration in-house using primary human fetal mononuclear cells (MNC) or NSG mouse BM. Analysis was performed using FlowJo software (v10.7.1) where gates were set using unstained and fluorescence minus one (FMO) controls. Table 2 - Antibodies used in this work Antibody Colour Company Cat. No. clone TCR Va24-Ja18 BV421 Biolegend 342916 6B11 FMC63 CD19-CAR Idiotype Miltenyi 130-127-343 FMC63 CD56 BV605 Biolegend 318334 HCD56 CD56 FITCLife Tech 11-0566-42HCD56L-Protein PE Stratech(SinoBiological) 11044-H07E-P RecProtein Viability7AAD Cayman 11397 mCD45.1 APCcy7 Biolegend 103116 30-F11 CD45 af700 Life Tech 56-9459-42 2D1 CD20 ef450 Life Tech 48-0209-42 2H7 CD16 PerCPcy5.5 Biolegend 302028 3G8 CD34 PEcy7 Life Tech 25-0349-42 4H11 CD133 / 1 APC Miltenyi Biotec 130-113-106 AC133 CD133 PE Miltenyi 130-113-108 AC133 CD10 FITC Life Tech 11-0106-42 eBioCB-CALLA (CB-CALLA) CD56 PerCPcy5.5 Biolegend 318322 HCD56 CD19 APC Biolegend 302212 HIB19 CD19 PEcy7 Life Tech 25-0199-42 HIB19 CD19 FITC eBioscience 11-0199-42 REA675 CD235a PerCPcy5.5 Biolegend 306614 HIR2 CD38 BV605 Biolegend 303532 HIT2 CD14 PerCPcy5.5 Biolegend 301824 M5E2 CD3 af700 Life Tech 56-0037-42 okt3 CD3 BV711 Biolegend 317328 OKT3 CD3 PerCPcy5.5 Biolegend 317336 OKT3 CD2 beads Miltenyi 130-091-114 CD2 PerCPcy5.5 Biolegend 300216 RPA-2.10 Viability Hoescht58 Life Tech H3569 Viability ef506 Life Tech 65-0866-18 CD1d PE BD Pharmingen™ 550255 CD1d42 CD1d PE Biolegend 350306 51.1 CD1d BB790 BD Bioscience custom CD45 FITC BD Bioscience 3458082D1mCD45APCcy7 Biolegend 103116 30-F11 TCR Va24-Ja18 (iNKT cells) BV421 Biolegend 342916 6B11 Viability 7AAD Cayman 11397 Labeling Check Reagent APC Miltenyi 130-122-219 CD11bFITCBiolegend 301330 ICRF44 CD33FITCBiolegend 366620 P67.6 Brilliant stain buffer plus BD Bioscience 566385 CD90 BV421 Biolegend 328122 5E10 CD16 PerCPcy5.5 Biolegend 302028 3G8 CD34 PEcy7 Life Tech 25-0349-42 4H11 CD123 BV650 Biolegend 306020 6H6 CD133 PE Miltenyi 130-113-108 AC133 CD10 FITC Life Tech 11-0106-42 eBioCB-CALLA CD45 APC / Cyanine7 BioLegend 103116 30-F11 Brilliant Violet CD45 510™ BioLegend 368526 2D1 CD45RA APCef780 Life Tech 47-0458-42 HI100 CD45 BUV395 BD Bioscience 563792 HI30 CD235 PerCP cy5.5 Biolegend 306614 HIR2 CD38 af700 Life Tech 56-0389-42 HIT2 Brilliant stain buffer plus BD Bioscience 566385 Brilliant Ultra CD4 Violet 395 eBioscience 363-0042-80 RM4-5 CD4 PE BioLegend 317410 OKT4 CD4 PE BioLegend 300539 RPA-T4 CD8a FITC BioLegend 301005 RPA-T8 CD8 FITC BioLegend 344704 SK1 Viability Staining 7-AAD Solution BioLegend 420404 TCR Vb11 APC Miltenyi Biotec 130-125-508 REA559 Protein L PE Stratech Scientific 11044-H07E-P HLA-A,B,C Pacific Blue™BioLegend311418 W6 / 32APC / Fire™ CD3 750 BioLegend 300470 UCHT1 NKG2D Fc Bio-Techne 1299-NK-050 CD314 (NKG2D) eBioscience 16-5878-82 1D11 iNKT MicroBeads Miltenyi Biotec 130-094-842 iNKT APC Miltenyi Biotec 130-094-839 EasySep™ Release Human PE Positive Selection Kit STEMCELL Technologies Granzyme B PE BioLegend 372207 QA16A02 Granzyme B PE-eFluor 610 eBioscience 61-8898-82 gb11 Brilliant Violet IFN-gamma 510™ BioLegend 502544 4S.B3 Alexa Fluor IFN gamma 700 eBioscience 56-7319-42 4S.B3 Brilliant Violet TNF-alpha 711™ BioLegend 502940 MAb11 Perforin FITC eBioscience 11-9994-42 dG9 IL-4 PE-Cyanine7 eBioscience 25-7049-41 8D4-8 IL-2 FITC BioLegend 500304 MQ1-17H12 CD3 MicroBeads 130-097-043 OKT3 CD28 MicroBeads 130-093-375 Cytotoxicity assays. These were performed a previously described (Rotolo 2018). Briefly, CellTrace™ Violet (Invitrogen)-labelled targets were incubated at the indicated ratios with effector cells for 4 or 16-24 hours. As controls, targets and effectors alone were simultaneously incubated to determine spontaneous cell death. Cells were then harvested and 7-AAD was added prior to flow cytometric analysis on BD Fortessa Flow Cytometer, using BD FACSDiva software version 6.0. Specific cytotoxic activity was determined as ((% sample (7-AAD+, Violet+) − % spontaneous (7-AAD+, Violet+)) / (100 - %spontaneous (7-AAD+, Violet+)) x 100. All assays were run in duplicates or triplicates and analysed using FlowJo 10.9.0. Intracellular cytokine expression were performed as previously described (3, 4). Flow-chamber avidity A µ-VI 0.4-A 6 channel slide (Ibidi) and a pump power system were used for the cell- binding avidity assay. Cell lines or primary leukaemia cells, were attached to poly-L- lysine-coated chips as a monolayer for at least 3 hours prior to testing. CellTrace far- red-labelled (Thermo Fisher Scientific) target cells were allowed to bind for 5 minutes before the flow rate and pressure were ramped up. The flow rate was increased from 0 to 25.6 ml / min of flow through the chamber, with real-time imaging using a EVOS M5000 Imaging System (ThermoFisher) conducted under both static and flow conditions. Counting of target cells remaining attached at each flow rate was analysed using ImageJ software. In vivo experiments Animals. All experiments were performed under two separate project licenses approved by the UK Home Office under the Animal (Scientific Procedures) Act 1986 after approval by the Oxford and Imperial College Animal Welfare and Ethical Review Bodies; and in accordance with the principles of 3Rs (replacement, reduction and refinement) in animal research. Experimental animals were 6-8 week old female NOD.Cg-PrkdcscidIl2rgtm1Wjl / SzJ (NSG) mice or 6 week old female NSGS mice. Male mice were used for some secondary and tertiary xenotransplantation experiments. Mice were housed in IVC cages, and kept at a 12-hour light / dark cycle, 21-22°C temperature and 45-65% relative humidity. They had red tunnels or houses and balconies in the cages as enrichment. Bioluminescence Imaging (BLI) BLI were collected on an IVIS Lumina XR III Imaging System using Living Image software (PerkinElmer). mice were anesthetised and maintained under inhalational anaesthesia via a nose cone with 2% isoflurane (Zoetis UK) / medical oxygen. A single intraperitoneal (IP) injection of 150 mg / kg D-luciferin (Goldbio) in PBS was administered to all mice 10 minutes before scanning. Up to three mice were imaged simultaneously in a 12.5 cm field of view (FOV) with a minimum target count of 30,000 and exposure times ranging from 0.5 to 3 minutes at medium binning, with additional images acquired at low binning levels to maximise sensitivity and spatial resolution where required. Both ventral and dorsal scans were acquired for each mouse. The dorsal and ventral signals were quantitated separately through region of interest (ROI) analysis using Living Image software (Aura-4.0.8) and expressed in radiance (units of photons / sec) as a total signal summation normalised to the ROI area. Where required, normalised background signal from similarly sized ROIs was subtracted. Leukaemia Models SEM model. Six-week-old NOD / SCID / IL-2Rγ-null (NSG) female mice (Charles River, UK) were handled in accordance with the 1986 Animal Scientific Procedures Act and under a United Kingdom Government Home Office–approved project licence PP8553679. The animals were housed at the Hammersmith Central Biomedical Services (CBS) facility, Imperial College London. On day 1, all animals were injected with 5 × 10^6 or 1x 10^6 luciferase-expressing SEM cells via the tail vein (iv), followed by bioluminescence imaging (BLI) monitoring on day 6 to confirm engraftment. On day 7, day 12, or day 16, the mice were randomised to either no treatment or immunotherapy with CAR-T or CAR-iNKT cells generated from the same donor. Thereafter, BLI was performed twice a week until day 21 and weekly until the end of the experiment. All mice were sacrificed according to protocol when either experimental or humane endpoints were reached.CRISPRMLL-AF4 ALL model: As with the SEM cells, on day 1, all animals were injected with 1 × 10^6CRISPRMLL-AF4 cells via the tail vein (iv), followed by tail vein blood collection to determine HLA-ABC or CD19, CD133 expression by FACS on day 6 and thus confirm engraftment. On day 7, the mice were randomised to either no treatment or immunotherapy with CAR-T or CAR-iNKT cell generated from the same donor. All mice were sacrificed according to protocol when either experimental or humane endpoints were reached. CAR-iNKT haematological toxicity assays in humanised mice 7-9 week old NSGS mice (n=) or NSG mice (n=) were sub-lethally irradiated with two doses of 1.25Gy six hours apart (2.5Gy total) and injected via the tail vein with 60,000 cord blood CD34+ cells. Engraftment was monitored by peripheral blood sampling every 3 weeks. Engrafted mice (>1% human CD45 cells in peripheral blood), were divided into control group (received PBS only) or treatment group, treated with 10 million bispecific CD19 / 133 CAR-iNKT cells at either 9 weeks or 12 weeks post CB CD34+ transplantation. Additional blood samples were taken at D+1 and D+3 post CAR-iNKT injection. Animals were monitored regularly using a standardised physical scoring system, and any mouse found to be in distress was humanely killed. All well mice were culled between 15-21 weeks post CB transplantation to assess long term bone marrow engraftment. Bone marrow was harvested from these mice for analysis. One million cells from the harvested BM from primary NSG mice were transplanted into secondary recipients. Secondary mice were monitored by peripheral blood sampling every 3 weeks, and culled at 18 weeks to assess long term bone marrow engraftment. Brain histopathology Murine heads were stripped of soft tissues, fixed in 10% neutral-buffered formalin (CellPath) and decalcified in Hilleman and Lee EDTA solution (5.5% EDTA in 10% formalin) for 2–3 weeks. Following paraffin embedding, haematoxylin and eosin staining (Sigma-Aldrich) and CD19 staining was performed on 5-mm brain sections. Anti-CD19 immunohistochemistry on paraffin-embedded sections was performed as previously described(5). Imaging used Axiostar Plus or Axio Imager M2 microscopes with AxioVision and ZEN software (Carl Zeiss, Cambridge, United Kingdom). CNS infiltration was assessed by an experienced pathologist (Halsey) blinded to treatment allocation.5-6 coronal slices and 2 blocks per head were examined and each block was assigned a score from 0= no infiltrate seen, 1= scattered occasional cells, 2= mild infiltrate, 3=moderate infiltrate, 4= heavy infiltrate. Single cell transcriptome-TCR combined analysis Experimental design. Bone marrows from either day 3 or day 15 post CAR-iNKT injection were pooled from 3 NSG mice per group of mice that either did or did not receive SEM cells before CAR-iNKT enrichment and purification. Ex vivo selection of iNKT. CAR-iNKT cells were enriched by magnetic bead selection following the manufacturers protocol against human CD2 (Miltenyi 130-091-114). CD2 positive cells were then further purified by FACS sorting staining using Labelling Check Reagent (Miltenyi 130-122-219) and 7AAD viability stain (Cayman 11397) with additional antibodies against mouse-CD45, human-CD45, and human CD19. The number of CD2+ cells recovered were 2768 to 7645 per pool (from 3 mice). In addition, 30,000 pre infusion CAR-iNKT cells from the same donor were also FACS sorted for single cell analysis. The inventors then immediately proceeded with the TCR / RNAseq protocol for the sorted cells (10x genomics CG000331 Rev C). Public datasets Publicly available RNA sequencing datasets for SEM and RS4;11 cell lines were obtained from NCBI (GSE149158)(6) and Cancer Cell Line Encyclopedia (CCLE) DepMap 2019 {Ghandi, 2019 #1}. Transcripts per million (TPM) values for the following 14 genes were extracted from both datasets and collated: PROM1 (CD133), CD19, HLA-A, HLA- B, HLA-C, MICA, MICB, ULBP1, ULBP2, ULBP3, RAET1E (ULBP4), RAET1G (ULBP5) and RAET1L (ULBP6) and CD1D. For NCBI (GSE149158), GEO2R was used to calculate the normalised TPM for each gene across triplicate samples for SEM (GSM4491229, GSM4491230 and GSM4491231) and RS4;11 (GSM4491211, GSM4491212 and GSM4491213) and the average TPM were subsequently enumerated. For CCLE DepMap, the TPM values for both cell lines were extracted from CCLE 2019 dataset “CCLE RNAseq gene expression data fpr 1019 cell lines (RSEM, gene)”. Statistics Two-tailed Mann-Whitney, Log-rank (Mantel-Cox) tests and ANOVA followed by multiple comparisons testing were used to compare experimental groups as indicated in the figure legends. Statistical analyses were performed using GraphPad Prism v10. Data are expressed as mean ± SEM unless otherwise indicated. References for materials and methods 1. Roy A, Cowan G, Mead AJ, Filippi S, Bohn G, Chaidos A, et al. Perturbation of fetal liver hematopoietic stem and progenitor cell development by trisomy 21. Proc Natl Acad Sci U S A. 2012;109(43):17579-84. 2. Rice S, Jackson T, Crump NT, Fordham N, Elliott N, O'Byrne S, et al. A human fetal liver-derived infant MLL-AF4 acute lymphoblastic leukemia model reveals a distinct fetal gene expression program. Nat Commun.2021;12(1):6905. 3. Rotolo A, Caputo VS, Holubova M, Baxan N, Dubois O, Chaudhry MS, et al. Enhanced Anti-lymphoma Activity of CAR19-iNKT Cells Underpinned by Dual CD19 and CD1d Targeting. Cancer Cell.2018;34(4):596-610 e11. 4. Rowan AG, Ponnusamy K, Ren H, Taylor GP, Cook LBM, Karadimitris A. CAR-iNKT cells targeting clonal TCRVbeta chains as a precise strategy to treat T cell lymphoma. Front Immunol. 2023;14:1118681. 5. Williams MT, Yousafzai Y, Cox C, Blair A, Carmody R, Sai S, et al. Interleukin-15 enhances cellular proliferation and upregulates CNS homing molecules in pre-B acute lymphoblastic leukemia. Blood.2014;123(20):3116-27. 6. Heczey A, Courtney AN, Montalbano A, Robinson S, Liu K, Li M, et al. Anti-GD2 CAR-NKT cells in patients with relapsed or refractory neuroblastoma: an interim analysis. Nat Med. 2020;26(11):1686-90. Background to the experiments described below Acute leukaemias initiated by rearrangement and fusion of the KMT2A (MLL) gene to a variety of partners are amongst the worst prognosis haematological malignancies. KMT2A-rearranged (KMT2Ar) B-cell acute lymphoblastic leukaemia (B-ALL) is the commonest form of infant leukaemia (80%)1and comprises a small fraction of childhood (3%) and adult (10%) leukaemias as well. Compared to the >90% survival of childhood B-ALL, event free survival of infants and children with KMT2Ar B-ALL, mainly due to chemo-resistance and relapse, is 38% and 60% respectively. Recent use of bi-specific CD19-targeting T cell engagers appears to improve survival while CD19 CAR-T cell immunotherapy offers promise of rescuing a subset of patients with relapsed disease who would otherwise have a dismal prognosis2. Nevertheless, after CD19 CAR-T therapy, one third of patients relapse with either CD19+ or CD19- disease, including lineage-switched disease3. The latter describes a state of transcriptional plasticity where trans-differentiation or expansion of pre-existing KMT2Ar myeloid / multipotent progenitors, results in B lineage lymphoblasts assuming a myeloid phenotype associated with partial or complete loss of CD19 expression3,4and escape from CD19 CAR-T control3,5. Treatment failure in KMT2Ar-ALL is also linked to inability of current therapeutic approaches to effectively eradicate central nervous system (CNS) disease, usually manifesting as leptomeningeal infiltration6. The inventors previously showed that PROM1 (CD133) is a direct target of leukemic KMT2A fusion genes and that KMT2Ar-ALL typically express PROM1 / CD1337,8. Therefore, dual targeting of antigens (e.g., CD19 and CD133, CD19 and CD33, or CD133 and CD33) provides a rational approach to enhance anti-leukemic activity and limit immune escape. In support of this, a pre-clinical approach involving a tandem CD19-CD133 CAR has been described9; however, direct comparison of the bi-specific approach with CD133 CAR-T and in vivo study of potential haematological toxicity were not performed. This is particularly important as CD133 is also expressed on normal haematopoietic stem and progenitor cells (HSPC)10. Of note, previous pre- clinical data suggest that tandem CARs may be less effective than bi-specific designs11,12, possibly due to steric hindrance mechanisms that interfere with high affinity binding of each CAR to their target antigens. As well as optimal CAR design and target selection, effective immunotherapy would also benefit from the co-operative activity of powerful and multi-functional effector cells. In this regard, the inventors and others have developed the CD1d-restricted, glycolipid-reactive iNKT cells13-15as a versatile ‘off-the-shelf’ platform16-19that not only lacks the risk of inciting aGVHD20,21, but also has inherent anti-tumour activity that may complement that of the CAR modules. Notably, when iNKT are equipped with CAR, they are able to eradicate brain lymphoma more effectively than their CAR-T counterparts18,22. Here, the inventors describe the development of iNKT cells equipped with a bi-specific CAR against CD19 and CD133, as a proof of concept, and investigate their activity against high risk medullary and meningeal KMT2Ar-ALL expressing or not the CAR targets. As described below, the inventors first developed novel iNKT cells equipped with a bi-specific CAR against CD19 and CD133 and then investigated their activity against high risk medullary and meningeal KMT2Ar-ALL expressing or not the CAR targets. The inventors also show that NKG2D expression is increased on CD33 CAR- iNKT after co-culture with CD33-epxressing AML cells. Results Example 1 - Enhanced anti-leukemic activity of bispecific CAR-iNKT With reference to Figure 1, immunophenotypic analysis of KMT2Ar-ALL primary blast cells showed co-expression of CD19 and CD133 and in agreement with a previous study23, with a 67% median level of CD133 co-expression (range 0-100). Although it was previously reported that KMT2Ar-ALL primary blasts express CD1d24, the inventors found that the majority of the patient samples tested were negative for CD1d expression (Figure 1a-c). Similarly, ALL blasts isolated from the bone marrow of the inventors’ recently described KMT2Ar-ALL model where expression of the KMT2A::AFF1 fusion gene in human fetal liver CD34+ cells was achieved by CRISPR- Cas9 mediated gene editing (CRISPRKMT2A-AFF1 ALL)25, demonstrated variable expression of CD133 on CD19+ blasts (Figure 1a&b). In contrast, CD133 but not CD19 is expressed on normal primitive fetal haematopoietic stem cell / multipotent progenitors (Lin-CD34+CD38-) while fetal B cell progenitors (CD34+CD19+) express CD19 but not CD133 confirming that CD133 / CD19 co-expression is leukaemia-specific (Figure 1a&d). Starting from two mAb clones, the inventors generated mono-specific 2ndgeneration 28z CD19 and 4-1BB CD133 CARs and the corresponding bi-specific CAR such that each CAR would be expressed stoichiometrically and independently of each other in the same cell (Figure 2a). Using their previously described protocol16,17, the inventors generated peripheral blood-derived CAR iNKT cells with all three CARs stably and similarly highly expressed and corresponding CAR-iNKT growing at similar rate (Figure 2b and Figure 4a). When tested against the KMT2A::AFF1 CD19+CD133+B-ALL cell line SEM, bi-specific CAR-iNKT were more cytotoxic than their mono-specific counterparts (Figure 2c). To test the specificity of each CAR, using CRISPR-Cas9 gene editing, the inventors generated SEM cell lines that lack expression of either CD19, CD133 or both (Figure 2d). They then tested cytotoxic activity of mono-specific CD19 or CD133 and bi-specific CD19-CD133 CAR-iNKT cells against these and the parental CD133+CD19+SEM cells (Figure 2e). The inventors observed that mono- and bi- specific CAR-iNKT displayed the lowest cytotoxicity against CD19-CD133- SEM cells while mono-specific CAR-iNKT only displayed background cytotoxicity against SEM cells lacking expression of their respective target. By contrast, bi-specific CAR-iNKT cells displayed the highest cytotoxic potential against the parental CD19+CD133+SEM cells, but were also equally effective as the corresponding mono-specific CAR-iNKT cells against CD19+CD133- and CD19-CD133+SEM cells (Figure 2e). Bi-specific CAR- iNKT were also more active than the mono-specific CAR counterparts against a second CD19+CD133+KMT2Ar-ALL cell line, RS4;11 (Figure 4b). To confirm the anti-leukemic efficacy of CD133 CAR-iNKT cells, mice were transplanted with luciferase- dsRed labelled KMT2Ar SEM cells. In this model, mice develop an aggressive form of universally fatal ALL with medullary and extramedullary (spleen, liver and leptomeningeal) disease by day 23-2523. The inventors found that CD133 CAR iNKT at a dose of 5x106 / mouse very effectively controlled leukaemia burden by reducing bioluminescence activity to background levels (Figure 4c). These findings confirm that both CD19 and CD133 CARs are active as bi- as well as mono-specific CARs when transduced into iNKT cells, with bi-specific CAR-iNKT exerting a more robust anti-leukaemia activity than their mono-specific counterparts in vitro. This likely reflects increased avidity mediated by their ability to engage both CD19 and CD133 on leukaemia cells. To investigate this, the inventors applied a modified flow chamber assay in which avidity was assessed by measuring the fraction of immune cells attached to leukaemia cells under an increasing pressure gradient. They found that while mono-specific CD133 CAR-iNKT demonstrated slightly higher avidity than the CD19 counterparts, bi-specific CAR-iNKT displayed the highest avidity (Figure 2f). Example 2 - Bispecific CAR-iNKT eradicate medullary and extramedullary leukaemia and outperform mono-specific CAR-iNKT Referred to Figure 3, the inventors next tested the anti-leukemic efficacy of bi-specific CAR-iNKT cells in vivo. First, they defined the lowest dose that could improve survival in mice engrafted with 106luciferase-labelled SEM leukaemia cells that displayed a spectrum of CD19 and CD133 co-expression, from high, low to no expression of both molecules (Figure 4d). Of the three doses tested, (105, 3x105and 106cells), only the highest dose significantly delayed leukaemia growth and prolonged survival (Figure 4e & f). Importantly, in line with the higher avidity of bi-specific CAR-iNKT cells, all the residual leukaemia cells were CD19-CD133- (Figure 4g) suggesting complete elimination of leukaemia cells co-expressing CD19 and CD133 even when expressed at low levels. Next, to test the ability of the bi-specific CAR-iNKT to mitigate immune evasion, the inventors injected NSG mice with 106parental SEM cells co-expressing CD19 and CD133 admixed with CD19-CD133+ and CD19+CD133- SEM cells at an 8:1:1 ratio respectively, followed by treatment with mono- or bi-specific CAR-iNKT cells at the limiting dose of 106cells per mouse. Of note, comparable growth curves of all three SEM cell lines in vitro suggested similar proliferative potential (Figure 5a). Treatment with mono-specific as well as bi-specific CAR-iNKT cells resulted in a significant delay in leukaemia progression and improved survival, with bi-specific CAR- iNKT cells conferring the longest survival (Figure 5b). The inventors performed immunophenotypic analysis of the bone marrow, spleen and liver at termination expecting preferential expansion of the CD19-CD133+and CD19+CD133- fraction of leukaemia cells in animals treated with mono-specific CAR19 and CAR133 CAR-iNKT respectively but not in those treated with bi-specific CAR-iNKT. Against expectation, with the exception of a significant higher fraction of CD19-CD133+cells in the spleens of animals treated with CD19 CAR-iNKT, the inventors found no significant difference in the frequency of the three different leukaemic populations in mice treated with any of the three effectors (Figure 5c). This suggests a CAR-independent anti-leukaemia effect of iNKT cells, shared by all three CAR-iNKT effectors. Repeating the same experiment using a different iNKT donor and only parental CD19+CD133+SEM cells as targets confirmed the enhanced ability of bi-specific CAR- iNKT cells to control leukaemia and prolong overall survival when compared to each of the mono-specific CAR-iNKT cells (Figure 3a & b). To investigate whether bi-specific CAR-iNKT can truly eradicate established leukaemia the inventors treated SEM leukaemia bearing mice with 5x106(Figure 3c-e) and 107CAR-iNKT (Figure 3f-h) on days 6 and 12 respectively when leukaemia burden in the BM is at ~2% and ~12% (Figure 6a). Remarkably, leukaemia burden as assessed by BLI rapidly declined to baseline with 100% of animals surviving long term (Figure 3c-e & f-h). Furthermore, detailed analysis of the BM and spleen at day 60 showed lack of leukaemia as well as effector cells (Figure 6b). Critically, in both experiments, while in untreated animals the meninges were heavily infiltrated by leukaemia cells as identified by CD19 staining, none were found in CD19-CD133 CAR-iNKT treated animals (Figure 3i). Therefore, CD19-CD133 bi-specific CAR-iNKT cells can eradicate leukaemia from the BM and spleen and protect from meningeal disease thus resulting in deep, lasting remissions. To investigate whether bi-specific CAR-iNKT cells can eradicate established meningeal leukaemia, the inventors used three leukaemia-bearing groups (Figure 3j): one that was untreated and when assessed for meningeal disease on day 16 was found to have a mean leukaemia meningeal infiltration grade of 1.25, while a second group that was also untreated at sacrifice had an infiltration grade of 3.8. By contrast, a third group of animals treated with bispecific CAR-iNKT on day 16 showed no evidence of leukaemia by both BLI and upon immunophenotypic analysis of the BM and spleen respectively (Figurea 6c-e). Importantly, consistent with meningeal leukaemia eradication, detailed histological and immunohistochemical analysis of the brain of these animals showed lack of any CD19-positive cells (infiltration grade 0; Figure 3j). The inventors conclude that bi-specific CAR-iNKT cells outperform their mono-specific counterparts and at high doses display curative potential against medullary and extramedullary disease, underpinned by their enhanced, CAR-dependent avidity. Critically, they also display ability to eradicate as well as protect from meningeal leukaemia. Example 3 - CAR-iNKT cells outperform their CAR-T counterparts Referring now to Figure 7, the inventors also compared the anti-leukemic activity of bi-specific CD19-CD133 CAR-iNKT cells with their CAR-T cell counterparts manufactured in parallel for 28 and 14 days respectively, using the same donor (Figure 9a). CAR-iNKT cells were more cytotoxic than CAR-T cells against the parental and gene-edited KMT2Ar SEM cells and the RS4;11 cells (Figure 7a, Figure 9b) and showed higher avidity (Figure 7B). Consistent with their innate functional profile, CAR- iNKT cells, upon co-culture with SEM cells for 4 and 24hrs produced similar peak levels of IFNg and TNFa as CAR-T cells, but at an earlier time point, and a higher cytotoxic potential as indicated by production of higher amounts of PRF, GRZB and IL-2 compared to CAR-T cells (Figure 7c & Figure 9c). Of note, CD19 and CD133 mono- specific CAR-iNKT are also more cytotoxic against KMT2Ar leukaemia cell lines than their CAR-T counterparts ( Figure 9d). As SEM and RS4;11 cell lines do not express CD1d (Figure 9e and Figure 11a) the higher anti-leukemic activity of CAR-iNKT is not due to CD1d targeting. Next, the inventors compared the in vivo efficacy of bi-specific CAR-T vs CAR-iNKT cells. They found that at the limiting dose of 106CAR+ cells, CAR-iNKT-treated SEM leukaemia-bearing mice had a significantly lower disease burden and longer survival than CAR-T-treated mice (Figure 7d & e). While CAR-T cells were not detected in the BM and spleen of treated animals at sacrifice on days 39 and 46 respectively, CAR- iNKT cells were readily detectable at sacrifice one week later (Figure 7f & g). In summary, these findings show that compared to CAR-T, CAR-iNKT cells have higher capacity for in vivo persistence and display superior anti-leukaemia activity involving a mechanism that is independent of CD1d expression by the leukaemia cells. Example 4 - Bi-specific CD19-CD133 CAR-iNKT cells eradicate primary KMT2Ar ALL and induce remissions, even where ALL cells lack CAR target expression Referring to Figure 8, to better emulate primary KMT2Ar ALL and the pattern of CD19- CD133 co-expression in patient primary blast cells, the inventors took advantage of the CD19+CRISPRKMT2A-AFF1 cells, in which there is negligible CD1d expression, CD133 is expressed by 66% of the CD19+ blasts and ~5% of the cells lack expression of both CD19 and CD133 (Figure 8a). As for the KMT2Ar ALL cell lines, bi-specific CAR-iNKT were more cytotoxic againstCRISPRKMT2A-AFF1 ALL blasts than CAR-T in vitro (Figure 9f) and displayed higher avidity than mono-specific CAR-iNKT (Figure 8b). The inventors treatedCRISPRKMT2A-AFF1 leukaemia-engrafted mice (assessed by presence of leukaemia cells in peripheral blood; Figure 9g) with 106bi-specific CAR- iNKT vs CAR-T generated from the same donor and manufactured over 28 and 14 days respectively. As with SEM cells, they found thatCRISPRKMT2A-AFF1 leukaemia- bearing mice treated with CAR-iNKT survived significantly longer than CAR-T treated animals (Figure 8c) and while CAR-T were not detectable at sacrifice, CAR-iNKT cells were readily detectable in the BM and spleen of mice sacrificed nearly one week later (Figure 8d). Increasing the therapeutic dose to 5x106bi-specific CAR-iNKT resulted in 100% ofCRISPRKMT2A-AFF1 leukaemia-bearing mice surviving beyond 60 days (Figure 8c). While untreated and 106bi-specific CAR-iNKT-treated mice succumbed to disease that comprised HLA-ABC-expressing CD19+CD133+ and notably CD19-CD133- populations in both bone marrow and spleen of all mice, in 5x106bi-specific CAR-iNKT-treated animals there were not detectable HLA-ABC+ cells suggesting that both CD19+CD133+ and CD19-CD133- cells were eliminated (Figure 8e). Finally, the inventors tested the activity of bi-specific CAR-iNKT against patient- derived xenograft (PDX) leukaemia cells (from BM of NSG mice transplanted with blasts from an 11-year old patient with KMT2A-AFF1 ALL). The PDX cells comprised ~70% CD19+CD133+ population within the human HLA class I-expressing fraction (Figure 8f). In vitro, bispecific CAR-iNKT were more cytotoxic than their CAR-T counterparts (Figure 9h) and demonstrated higher avidity than mono-specific CAR- iNKT (Figure 8g). In vivo, animals with detectable PDX cells in blood on day 14 (Figure 9i) were treated or not with 5x106bi-specific CAR-iNKT on day 15. On day 33, two animals from each arm were sacrificed. While ~8% and ~18% hCD45+ CD19-CD133- as well as CD19+CD133+ cells were detected in the BM and spleen of untreated animals, hCD45+ cells were not detectable in treated animals (Figure 8g). Similarly, hCD45+ cells were not detectable in the BM and spleen of treated animals sacrificed of day 45. In conclusion, bispecific CAR-iNKT persist longer and are more effective than CAR-T against primary ALL, and when delivered at high doses effectively eliminate primary ALL with variable expression of the CAR targets including of cells that lack expression of both CAR targets. Example 5 - Dynamic NKG2D upregulation in CAR-iNKT cells accounts for their ability to outperform CAR-T and for their enhanced activity against CAR target-negative leukaemia Referring to Figure 10, the inventors hypothesised that the mechanism that dictates enhanced anti-leukaemia activity of CAR-iNKT over CAR-T and ability to eliminate leukaemia cells that lack expression of both CAR targets could involve activating NK receptors, such as NKG2D, which is known to be expressed on CD4-iNKT cells24,25. Indeed, NKG2D CAR-T approaches are currently tested in clinical trials in leukaemia and other blood cancers while recent evidence suggests that CD8+ T cells can effectively control tumours that lack MHC expression via NKG2D-mediated activation26. To address these hypotheses, the inventors first compared kinetics of NKG2D expression on CAR effectors. They surprisingly found that in three different donors, expression of NKG2D at baseline was significantly higher (as assessed by both % and expression intensity) in CD19-CD133 CAR-iNKT than CD19-CD133 CAR-T cells (Figure 10a & b). Following a 16hr exposure to SEM or RS4;11 cells, while in CAR-T cells expression of NKG2D increased modestly in CAR-iNKT cells it increased to nearly 100% (Figure 10a & b). Of note, while as previously reported, only CD4- and not CD4+ iNKT expressed NKG2D at baseline, upon exposure to leukaemia cells, both fractions displayed expression of NKG2D to nearly 100%. Next, by analysing previous transcriptomes of KMT2Ar cell lines (SEM and RS4;11) and primary leukaemia cells, the inventors found evidence of expression of at least one of the NKG2D ligands (including MICA-B and ULBP1-6) in both cell lines, and primary leukaemias analysed. (Figure 11a & f). The inventors further investigated expression of NKG2D ligands by staining KMT2Ar cell lines with a NKG2D-Fc chimaeric protein and they indeed confirmed expression in all cases tested (Figure 10c). To investigate the functional significance of these observations, the inventors tested the cytotoxic activity of bi-specific CAR-iNKT vs CAR-T that had been pre-exposed to leukaemia cells, in the presence of an NKG2D blocking mAb or Ig control. They found a concentration-dependent decrease in the cytotoxic activity of CAR-iNKT and -T cells against the KMT2Ar SEM cell line in the presence of NKG2D mAb (Figure 10d & e) suggesting that the higher anti-leukemic activity of CAR-iNKT is largely NKG2D- mediated. Consistent with these results, the modestly higher avidity displayed by bi- specific CAR-iNKT when compared to CAR-T counterparts was also mitigated upon NKG2D blockade (Figure 10f). Using NKG2D-Fc soluble protein to block the NKG2D ligands on target cells also abrogated the enhanced cytotoxicity of CAR-iNKT vs CAR-T thus further confirming the role of NKG2D in the differential reactivity of CAR-iNKT vs CAR-T against leukaemia (Figure 11b). To investigate further the mechanism of upregulation of NKG2D expression, the inventors then co-cultured untransduced or CAR-transduced T and iNKT with SEM cells expressing CD19, CD133, both or none (Figure 10g & h). They found that expression of NKG2D in untransduced T and iNKT did not increase after exposure to CAR target- expressing or non-expressing SEM cells. By contrast, the higher baseline expression of NKG2D on CAR-iNKT than CAR-T commensurately increased in CAR-iNKT and CAR-T after their co-culture with CAR target-expressing cells i.e., CD19+CD133+, CD19+CD133- or CD19-CD133+ SEM cells; however, upon co-culture with CD19- CD133- SEM cells, both CAR-iNKT and CAR-T failed to upregulate NKG2D (Figure 10h and Figure 11c). Together these results show that upregulation of NKG2D requires both expression of CARs and their engagement by their corresponding targets on leukaemia cells. Since bispecific CAR-iNKT can eliminate CAR target-expressing and non-expressing leukaemia cells in vivo, the inventors next tested whether ability of CAR-iNKT to kill leukaemia cells lacking expression of the CAR targets is NKG2D-mediated and requires their prior engagement with CAR target-expressing leukaemia cells (Figure 10i). They found that CAR-iNKT and to a lesser extent CAR-T cells that had been pre-exposed to CAR target-expressing leukaemia cells, displayed higher cytotoxicity against CAR target-negative cells than CAR-iNKT and CAR-T that had been pre-exposed to leukaemia cells lacking expression of CAR target (Figure 10i) and this effect was attenuated by NKG2D blockade. Similarly, CAR-iNKT pre-cultured withCRISPRKMT2A- AFF1 ALL blasts upregulated NKG2D more efficiently than CAR-T (Figure 11d) and subsequently were able to kill CAR target-negative SEM cells more efficiently than CAR-T in an NKG2D-dependent manner (Figure 11e). These findings highlight the ability of CAR-iNKT cells to enhance their innate, NKG2D- dependent effector functions in a CAR- and CAR target-dependent manner. In this way they can kill both leukaemia cells that express the CAR target in vivo and those that do not. Such a mechanism would contribute to higher efficacy and limit immune escape of CAR target-negative leukaemia cells, including those that might have undergone myeloid lineage switch, provided they express NKG2DL. In line with this, transcriptome analysis of paired diagnostic and relapse KMT2Ar ALL patient samples27shows that while expression of CD19 and PROM1 / CD133 is variably reduced upon lineage switch of B ALL to myeloid blasts, in most cases there is residual low-level expression of either CD19 or CD133 while expression of NKG2D ligands and in particular of MICA-B remains the same or is higher in lineage switched myeloid blasts than in the paired presentation lymphoblasts (Figure 11f). Figure 17 illustrates the hypothesised mechanism of action for killing bystander tumour cells by the CAR-immune cell. A tumour cell (TC) expresses a target antigen, such as CD19 antigen, CD133, and / or CD33, and also the NKG2D ligand specific for the NKG2D receptor. An immune cell (IC), such as an iNKT cell, expresses a CAR, for targeting the TC’s CD19 / CD133 / CD33 antigen, and also the NKG2D receptor. Upon engagement between the CAR (anti-CD19) on the immune cell and the antigen on the tumour cell (CD19), the expression level of NKG2D receptor on the immune cell surprisingly increases. The immune cell therefore now expresses an elevated concentration of NKG2D receptors on its surface, as shown in the Figure, which enable the iNKT cell to bind to the NKG2D ligands expressed on different tumour cells which do not express the CAR antigen (i.e., CD19), and can thereby kill those bystander tumour cells, and so address immune evasion. Example 6 – Ex vivo functional characterisation of CAR-iNKT cells at the single cell level Referring to Figure 12, to gain insights into how the functional state of CAR-iNKT cells is altered in vivo upon their engagement with leukaemia, the inventors subjected pre- infusion and CAR-iNKT isolated from the bone marrow of CAR-iNKT-treated, leukaemia-bearing or leukaemia-free mice (i.e., five groups) to single cell TCRab and transcriptome analysis (Figure 12a). Specifically, cells were first isolated using CD2 immunomagnetic beads and pooled (n=3 mice) on day 3 when no leukaemia cells were detected in the BM and spleen of CAR-iNKT-treated animals and on day 15 when CAR-iNKT cells were still readily detectable (Figure 13a). Single cell gene expression and TCR-sequencing was performed and dimensionality reduction of all cells identified three clusters of cells (C0, C1 & C2) with all five groups represented in each cluster (Figure 13b). As expected, although not captured in all cells, the invariant TCRAV10-TRAJ18 sequence was by far the most dominant TRA clonotype (Figure 13b-d). Differential gene expression analysis showed only a handful of variable genes between the day 3 groups but a larger number of variable genes between pre-infusion and day 15 cells (Figure 13e). Focusing on genes over-expressed in cluster 0 vs 1 (n=830, Log2FC > 0.5, Padj< 0.05) revealed enrichment for oxidative phosphorylation and enhanced protein synthetic capacity (Myc Targets V1; MTORC1 signalling; Figure 12c&d) suggesting that in a sizeable fraction of CAR-iNKT favourable energy production and protein anabolic pathways support their function. At a single gene expression level, cluster 0, compared to clusters 1&2 emerged with a robust profile of cytotoxic potential with perforin (PRF), several granzyme genes (GZMB, GZMH, GZMK) and activating NK cell receptor genes (KLRK1 / NKG2D, NCR3, LTB) preferentially and highly expressed in cluster 0 cells (Figure 12e). Of note, with the exception of GZMH and GZMK the expression of which was reduced, cells retained their expression of GZMA, GZMB and PRF1 after 3 and 15 days (Figure 13f). Day 15 iNKT cells from leukaemia-bearing mice over-expressed 72 genes compared to iNKT isolated from leukaemia-free mice, with the most enriched pathways corresponding to cell cycle (G2M checkpoint, E2F targets) and protein synthesis (Myc Targets V1; Figure 12f & g) thus highlighting the fact that >10 days after clearing leukaemia CAR-iNKT retain robust proliferative potential and metabolic fitness. In line with this, while frequency of C0 declined, frequency of cells in cluster 2, which is the most proliferative of the three clusters (Figure 13g) increased on day 15 only in leukaemia-bearing but not leukaemia free mice (Figure 12h). Together these findings suggest that CAR-iNKT retain their proliferative and effector functions after their in vivo tackling of leukaemia. To corroborate this prediction the inventors isolated iNKT cells from the BM of bispecific CAR-iNKT-treated, leukaemia- bearing animals shown in Figure 3a & b involving a different donor. The inventors found presence of iNKT cells in the BM of all treated animals treated with 106cells on day 6 (Figure 14a), isolated them and investigated their functional properties in vitro. Explanted CAR-iNKT cells proliferated robustly in the presence of alphaGalCer (>100- fold over two weeks; (Figure 14b), retained the same level of CAR expression as the originally injected cells (Figure 14c), and following expansion they retained their ability to exert a powerful cytotoxic effect when cultured with CD19hi / loCD133hi / lo SEM leukaemia cells (Figure 14d). When tested against the CD19-expressing C1R and C1R- CD1d B cells with or without alphaGalCer, explanted bi-specific CAR-iNKT cells exerted the highest cytotoxicity against alphaGalCer-pulsed C1R-CD1d cells (Figure 14e), thus confirming that their iTCR-CD1d axis is functionally preserved and can contribute to the anti-tumour function of CAR-iNKT cells. Therefore, as predicted by transcriptome analysis, even after 4 weeks of in vitro expansion, ~5 weeks of in vivo presence and another 3 weeks of ex vivo expansion, bi-specific CAR-iNKT cell retain robust proliferative and cytolytic anti-leukaemia activity. Example 7 - Impact of bi-specific CAR-iNKT cells on normal human haematopoiesis Referring to Figure 15, given the expression of CD133 by a proportion of normal primitive HSPC (Figure 1c), the inventors used a humanised xenograft model to assess potential off-target effects of CD19 / CD133 bispecific CAR-iNKT by transplanting cord blood (CB) CD34+ cells into sub-lethally irradiated NSG-SGM3 mice (Figure 15 a & b and Figure 16a). In two independent experiments, engrafted NSG-SGM3 mice (confirmed as hCD45 expression by >1% of PB cells) were treated with either PBS or 107CD19-CD133 bi-specific CAR iNKT cells. On-target, off-leukaemia activity of CAR iNKT cells was confirmed by a significant reduction in the frequency of CD19+ B cells in the PB at day 1 and day 3 post CAR-iNKT injection, compared to controls (Figure 15c). Long-term engraftment in PB (18 weeks) or BM (at cull 18-21 weeks post- transplantation) showed no significant difference between CAR iNKT- and PBS-treated mice (Figure 15a & d), either in total hCD45+ cells or, in BM, in the frequencies of B, T, myeloid or CD34+ cell within the hCD45 compartment (median values for B cells: 25.7% vs 26.1%; T cells: 22.4% vs.25.8%, myeloid cells: 43.5% vs.46.4% and CD34: 3.2% vs.2.3% of hCD45 respectively; Figure 15d). This confirms comparable multi-lineage long term human haematopoietic reconstitution in PBS- and CAR iNKT- treated mice, suggesting there is no significant, CAR iNKT-mediated toxicity against primitive HSPC. Example 8 - NKG2D expression is increased on CD33 CAR-iNKT after co-culture with CD33-epxressing AML cells The schematic structures of the anti-CD19 and anti-CD33 CAR constructs are equivalent to those illustrated in Figure 2a except that CD133 is replaced by CD33. Referring to Figure 18, the inventors co-cultured CD33 CAR-iNKT with two AML cell lines, which resulted in upregulation of NKG2D in both cell lines, thus providing proof- of-principle that this property would be useful for CAR-iNKT treatment of acute myeloid leukaemia to. Discussion Effective immunotherapy of cancer requires that the specific biological features of the malignancy of interest inform the therapeutic strategy. In the case of KMT2A- rearranged leukaemias, including infant ALL, the adverse prognosis is due mainly to chemoresistance, the propensity to involve the brain and in particular the leptomeninges; and transcriptional and phenotypic plasticity associated with switching from a lymphoid to a myeloid identity. Here, the inventors developed a highly effective immunotherapeutic strategy that can address these biological adverse features. Firstly, the dual CAR target approach engages two surface antigens that are often, although not always, highly co- expressed. Informed from previous work that demonstrates higher pre-clinical anti- tumour potential of bi-specific over tandem CAR designs, the inventors developed and validated a powerful bi-specific CD19-CD133 CAR strategy. They harness the unique properties of iNKT cells, including their lack of risk of aGVHD when allogeneically sourced and their CD1d-independent, inherent anti-cancer activity, to demonstrate the enhanced anti-leukaemia activity of bi-specific CAR-iNKT over mono-specific CAR-iNKT and bi-specific CAR-T in faithful and clinically relevant models of KMT2Ar-ALL. In all cases, enhanced avidity of bi-specific CAR-iNKT, in line with recent work (11,32,33), correlated with in vivo efficacy. Increased avidity of bi-specific over mono- specific CAR-iNKT would be particularly important for the effective targeting of CAR antigen-low disease as demonstrated in in vivo models and it can potentially mitigate subsequent CAR target-low relapse often seen in KMT2Ar-ALL as well as other forms of ALL treated with CD19 CAR-T (3,5). When compared to CAR-T, the inventors show that higher expression of NKG2D by CAR-iNKT contributes to their modestly higher avidity. Previous work demonstrated the importance of NKG2D expression by CD4- iNKT for enhancing iNKT activation independent of iTCR-CD1d interaction when engaged by NKG2D stress ligands (28). Here, the inventors demonstrate a novel process of dynamic NKG2D expression by both CD4- and CD4+ iNKT subsets. This process, that requires and is triggered by CAR engagement, allows subsequent targeting of leukaemia cells with variable expression of CAR targets, including leukaemia cells that lack expression of both CAR targets in an NKG2D-dependent manner. These findings are analogous to those demonstrating that TCR-MHC class I-dependent upregulation of NKG2D in conventional CD8+ T cells allows them to target tumour cells with loss of MHC molecules in a NKG2D-dependent manner30. This mechanism is more powerful in CAR-iNKT than CAR-T and would be important not only for eradication of leukaemia with partial expression of CD133 or low expression of both CAR targets; it would also be expected to reduce the risk of lineage switch which is associated with partial or complete loss of CD19 and CD133 expression by directly targeting lineage switched blasts. Pertinently, expression of NKG2D ligands in lineage-switched blasts remains high or even higher than in the original lymphoblasts highlighting the importance of NKG2D-mediated anti-leukaemia activity of CAR-iNKT cells. Consistent with this, in the CRISPRKMT2A-AFF1 model, CD19-CD133- cells, comprising a small fraction of the leukemic clone, were effectively eliminated by bi-specific CD19-CD133 CAR iNKT in vivo. While CD1d expression by leukemic blasts has been reported in KMT2Ar-ALL24, here the superior anti-leukemic activity of CAR-iNKT was shown in CD1d- ALL cells suggesting that CAR-iNKT could exert additional anti-leukaemia activity against CD1d- expressing targets which, as shown here, can further be enhanced by the high affinity αGalCer glycolipid ligand. Although the superior anti-leukemic activity of CAR-iNKT over CAR-T was evident at limiting dose levels it did not demonstrate curative potential despite longer persistence and retention of functionality of ex vivo isolated CAR-iNKT. Instead, CAR-iNKT demonstrate curative potential against different levels of leukaemia burden only when used at higher dose levels in vivo while the in vitro NKG2D-mediated anti-leukaemia activity against CAR target-negative leukaemia was also more evident at higher E:T ratios. These findings should inform clinical development and call for use of high doses of CAR-iNKT in future clinical trials. This notion is in line with recent pre-clinical data showing that failure of immunotherapy is often not due to tumour-intrinsic mechanisms of immune escape but simply due to not enough numbers of effector cells reaching all tumour sites when relatively low therapeutic doses are used (34). A further major advantage of higher bi-specific CAR-iNKT cell doses is their ability to confer protection from, and eradication of established meningeal leukaemia, in agreement with higher ability of CAR-iNKT than CAR-T to clear brain lymphoma. Delineating the mechanism(s) of this critical property of iNKT cells may have therapeutic implications for a variety of leptomeningeal metastases seen in breast cancer, melanoma and lung cancers (35), while in the case of ALL CAR-iNKT could be an alternative to toxic intrathecal methotrexate for CNS prophylaxis. The inventors’ transcriptome analysis suggested a higher expression in CAR-iNKT than CAR-T of VLA- 4 (18), an integrin required for endothelial adhesion and crossing of the blood brain barrier and in the case of leptomeningeal leukaemia, the choroid plexus-brain barrier (18). The transcriptional analysis of CAR-iNKT cells shows that gene expression programmes associated with cytotoxic and proliferative potential as well as metabolic fitness that are present in the pre-infusion CAR-iNKT cells, are preserved at two weeks following their in vivo interaction with leukaemia cells. These findings are in line with the fact that explanted CAR-iNKT retain their ability to proliferate and exert robust cytotoxicity against leukaemia cells. Finally, the inventors demonstrate that at dose levels that induce B cell depletion, bi- specific CAR-iNKT do not significantly impact haematopoiesis, which remains without discernible defects, at least immunophenotypically, thus providing initial insights into the safety profile of this approach in patients. Of note, in a clinical trial of autologous CD133 CAR-T in patients with solid tumour cancers haematologic toxicity comprised grade 2 immediately after lymphodepletion and CAR-T infusion with haematologic parameters recovering to normal thereafter (36). Projecting into a future clinical application, the inventors propose that allogeneically- sourced bi-specific CAR-iNKT could be deployed as a swift bridging therapy to allogeneic stem cell transplantation to achieve MRD-negative / low disease in relapsed / refractory KMT2Ar-ALL, as a pre-requisite for a curative allo-HSCT (38). A variation of this approach could entail generating CAR-iNKT cells from the prospective stem cell transplant donors. While not as developed as other forms of cellular immunotherapy such as CAR-T and CAR-NK cells, allogeneic CAR-iNKT and iNKT-based immunotherapeutics are at the verge of clinical development with early clinical results suggesting high efficacy with, as predicted, lack of aGVHD and importantly, mild or no cytokine release syndrome and neurotoxicity (39,40). Conclusions Bi-specific CAR-iNKT cell immunotherapy is a very effective treatment for pre-clinical aggressive KMT2Ar-ALL, as it outperforms CAR-T cell immunotherapy in an NKG2D- dependent manner and has the potential to protect from immune escape, leptomeningeal disease and potentially lineage switch without discernible haematological toxicity. These findings provide the basis for clinical development of bi- specific CAR-iNKT cells, such as CD19-CD133 CAR-iNKT cells, CD19-CD33 CAR-iNKT cells, or CD133-CD33 CAR-iNKT cells, as an ‘off-the-shelf’ treatment for MLLr-ALL. More broadly, the ‘innate’ ability of CAR-iNKT cells to effectively target CAR target- negative as well as CAR target-expressing leukaemia cells calls for their development for a wide range of high-risk malignancies with leptomeningeal involvement. References 1. Hunger SP, Mullighan CG. Acute Lymphoblastic Leukemia in Children. N Engl J Med. 2015;373(16):1541-1552. 2. van der Sluis IM, de Lorenzo P, Kotecha RS, et al. Blinatumomab Added to Chemotherapy in Infant Lymphoblastic Leukemia. N Engl J Med.2023;388(17):1572-1581. 3. Gardner R, Wu D, Cherian S, et al. Acquisition of a CD19-negative myeloid phenotype allows immune escape of MLL-rearranged B-ALL from CD19 CAR-T-cell therapy. Blood. 2016;127(20):2406-2410. 4. Tirtakusuma R, Szoltysek K, Milne P, et al. Epigenetic regulator genes direct lineage switching in MLL / AF4 leukemia. Blood.2022;140(17):1875-1890. 5. Liao W, Kohler ME, Fry T, Ernst P. Does lineage plasticity enable escape from CAR-T cell therapy? Lessons from MLL-r leukemia. Exp Hematol.2021;100:1-11. 6. Frishman-Levy L, Izraeli S. Advances in understanding the pathogenesis of CNS acute lymphoblastic leukaemia and potential for therapy. Br J Haematol.2017;176(2):157-167. 7. Mak AB, Nixon AM, Moffat J. The mixed lineage leukemia (MLL) fusion-associated gene AF4 promotes CD133 transcription. Cancer Res.2012;72(8):1929-1934. 8. Godfrey L, Crump NT, O'Byrne S, et al. H3K79me2 / 3 controls enhancer-promoter interactions and activation of the pan-cancer stem cell marker PROM1 / CD133 in MLL-AF4 leukemia cells. Leukemia.2021;35(1):90-106. 9. Li D, Hu Y, Jin Z, et al. TanCAR T cells targeting CD19 and CD133 efficiently eliminate MLL leukemic cells. Leukemia.2018;32(9):2012-2016. 10. Yin AH, Miraglia S, Zanjani ED, et al. AC133, a novel marker for human hematopoietic stem and progenitor cells. Blood.1997;90(12):5002-5012. 11. Fernandez de Larrea C, Staehr M, Lopez AV, et al. Defining an Optimal Dual-Targeted CAR T-cell Therapy Approach Simultaneously Targeting BCMA and GPRC5D to Prevent BCMA Escape-Driven Relapse in Multiple Myeloma. Blood Cancer Discov.2020;1(2):146-154. 12. Leung I, Templeton ML, Lo Y, et al. Compromised antigen binding and signaling interfere with bispecific CD19 and CD79a chimeric antigen receptor function. Blood Adv. 2023;7(12):2718-2730. 13. Kronenberg M, Gapin L. The unconventional lifestyle of NKT cells. NatRevImmunol. 2002;2(8):557-568. 14. Cerundolo V, Kronenberg M. The role of invariant NKT cells at the interface of innate and adaptive immunity. Seminars in Immunology.2010;22(2):59-60. 15. Salio M, Silk JD, Jones EY, Cerundolo V. Biology of CD1- and MR1-restricted T cells. Annu Rev Immunol.2014;32:323-366. 16. Heczey A, Courtney AN, Montalbano A, et al. Anti-GD2 CAR-NKT cells in patients with relapsed or refractory neuroblastoma: an interim analysis. Nat Med.2020;26(11):1686-1690. 17. Heczey A, Liu D, Tian G, et al. Invariant NKT cells with chimeric antigen receptor provide a novel platform for safe and effective cancer immunotherapy. Blood.2014. 18. Rotolo A, Caputo VS, Holubova M, et al. Enhanced Anti-lymphoma Activity of CAR19- iNKT Cells Underpinned by Dual CD19 and CD1d Targeting. Cancer Cell.2018;34(4):596-610 e511. 19. Rowan AG, Ponnusamy K, Ren H, Taylor GP, Cook LBM, Karadimitris A. CAR-iNKT cells targeting clonal TCRVbeta chains as a precise strategy to treat T cell lymphoma. Front Immunol. 2023;14:1118681. 20. Chaidos A, Patterson S, Szydlo R, et al. Graft invariant natural killer T-cell dose predicts risk of acute graft-versus-host disease in allogeneic hematopoietic stem cell transplantation. Blood.2012. 21. Mavers M, Maas-Bauer K, Negrin RS. Invariant Natural Killer T Cells As Suppressors of Graft-versus-Host Disease in Allogeneic Hematopoietic Stem Cell Transplantation. Front Immunol.2017;8:900. 22. Delfanti G, Dellabona P, Casorati G, Fedeli M. Adoptive Immunotherapy With Engineered iNKT Cells to Target Cancer Cells and the Suppressive Microenvironment. Front Med (Lausanne). 2022;9:897750. 23. . !!! INVALID CITATION !!! 21. 24. Fais F, Tenca C, Cimino G, et al. CD1d expression on B-precursor acute lymphoblastic leukemia subsets with poor prognosis.2005;19(4):551-556. 25. Rice S, Jackson T, Crump NT, et al. A human fetal liver-derived infant MLL-AF4 acute lymphoblastic leukemia model reveals a distinct fetal gene expression program. Nat Commun. 2021;12(1):6905. 26. Savino AM, Fernandes SI, Olivares O, et al. Metabolic adaptation of acute lymphoblastic leukemia to the central nervous system microenvironment is dependent on Stearoyl CoA desaturase. Nat Cancer.2020;1(10):998-1009. 27. Ransohoff RM, Kivisakk P, Kidd G. Three or more routes for leukocyte migration into the central nervous system. Nat Rev Immunol.2003;3(7):569-581. 28. Kuylenstierna C, Bjorkstrom NK, Andersson SK, et al. NKG2D performs two functions in invariant NKT cells: direct TCR-independent activation of NK-like cytolysis and co-stimulation of activation by CD1d. Eur J Immunol.2011;41(7):1913-1923. 29. Gumperz JE, Miyake S, Yamamura T, Brenner MB. Functionally distinct subsets of CD1d- restricted natural killer T cells revealed by CD1d tetramer staining. JExpMed.2002;195(5):625- 636. 30. Lerner EC, Woroniecka KI, D'Anniballe VM, et al. CD8(+) T cells maintain killing of MHC- I-negative tumor cells through the NKG2D-NKG2DL axis. Nat Cancer.2023;4(9):1258-1272. 31. Chen C, Yu W, Alikarami F, et al. Single-cell multiomics reveals increased plasticity, resistant populations, and stem-cell-like blasts in KMT2A-rearranged leukemia. Blood. 2022;139(14):2198-2211. 32. Larson RC, Kann MC, Bailey SR, et al. CAR T cell killing requires the IFNgammaR pathway in solid but not liquid tumours. Nature.2022;604(7906):563-570. 33. Leick MB, Silva H, Scarfo I, et al. Non-cleavable hinge enhances avidity and expansion of CAR-T cells for acute myeloid leukemia. Cancer Cell.2022;40(5):494-508 e495. 34. Goddard ET, Linde MH, Srivastava S, et al. Immune evasion of dormant disseminated tumor cells is due to their scarcity and can be overcome by T cell immunotherapies. Cancer Cell. 2024;42(1):119-134 e112. 35. Nayar G, Ejikeme T, Chongsathidkiet P, et al. Leptomeningeal disease: current diagnostic and therapeutic strategies. Oncotarget.2017;8(42):73312-73328. 36. Wang Y, Chen M, Wu Z, et al. CD133-directed CAR T cells for advanced metastasis malignancies: A phase I trial. Oncoimmunology.2018;7(7):e1440169. 37. Vora P, Venugopal C, Salim SK, et al. The Rational Development of CD133-Targeting Immunotherapies for Glioblastoma. Cell Stem Cell.2020;26(6):832-844 e836. 38. Takachi T, Watanabe T, Miyamura T, et al. Hematopoietic stem cell transplantation for infants with high-risk KMT2A gene-rearranged acute lymphoblastic leukemia. Blood Adv. 2021;5(19):3891-3899. 39. Ramos C, Courtney A, Robinson S, et al. Allogeneic NKT Cells Expressing a CD19- Specific CAR in Patients with Relapsed or Refractory B-Cell Malignancies: An Interim Analysis. Blood.2021;138. 40. Hammond TC, Purbhoo MA, Kadel S, et al. A phase 1 / 2 clinical trial of invariant natural killer T cell therapy in moderate-severe acute respiratory distress syndrome. Nat Commun. 2024;15(1):974.

Claims

1. Claims 1. An immune cell, for use in treating, preventing or ameliorating cancer, wherein the immune cell comprises one or more antigen-binding domain, which targets at least one target antigen which is expressed above a threshold level by a first cancer cell, and the at least one target antigen is expressed below the threshold level by a second cancer cell, wherein the second cancer cell expresses one or more NKG2D stress ligand, and the immune cell expresses one or more NKG2D stress ligand receptor, and wherein the immune cell kills the second cancer cell upon engagement between the one or more NKG2D stress ligand and the one or more NKG2D stress ligand receptor.

2. The immune cell, for use according to claim 1, wherein the immune cell is a T- cell, optionally a Natural Killer T (NKT) cell.

3. The immune cell, for use according to either claim 1 or claim 2, wherein the immune cell is an invariant NKT (iNKT) cell.

4. The immune cell, for use according to claim 3, wherein the iNKT cell is a CD4- negative iNKT cell or CD4-positive iNKT cell.

5. The immune cell, for use according to either claim 3 or claim 4, wherein the iNKT cell is enriched by contacting it with a glycolipid, optionally alpha- Galactosylceramide (αGalCer).

6. The immune cell, for use according to any preceding claim, wherein the immune cell is allogeneically-sourced or autologously-sourced.

7. The immune cell, for use according to any preceding claim, wherein the immune cell is generated using iNKT cells from a prospective stem cell transplant donor.

8. The immune cell, for use according to any preceding claim, wherein the at least one target antigen on the first cancer cell comprises a tumour antigen, which is either a polypeptide tumour antigen or a glycoprotein tumour antigen displayed on the surface of the first cancer cell.

9. The immune cell, for use according to any preceding claim, wherein the tumour antigen is selected from a group of antigens including: BCMA, CS1, CD38, FCRL5, CD19, CD20, CD22, CD30, CD123, CLL1, CD33, FTL3, CD133, CLAUDIN 18.2, NKG2Dligands, TCRVbeta / alpha variable and constant chains, CD5, CD7, B7-H3, EGFR, HER2, EGFR806, Mesothelin, PSCA, MUC1, EpCAM, GD2, VEGFR2, AFP, Nectin4 / FA, PCEA, Lewis Y, Glypican-3, EGFRIII, IL-13Rα2, CD171, MUC16, PSMA, AFP, AXL, c-MET, DLL- 3, DR5, EpHA2, FRα gp100, MAGE-A1 / 3 / 4, and LMP1.

10. The immune cell, for use according to any preceding claim, the at least one target antigen expressed by the first cancer cell is: (i) CD19; (ii) CD133; (iii) CD33; (iv) CD19 and CD133; (v) CD19 and CD33; and / or (vi) CD133 and CD33.

11. The immune cell, for use according to any preceding claim, wherein the second cancer cell exhibits lower expression of the at least one target antigen, optionally CD19, CD133, and / or CD33, and typically lower than CD19 and CD133, or CD19 and CD33, or CD133 and CD33.

12. The immune cell, for use according to any preceding claim, wherein the threshold levels of expression of the at least one target antigen on the first and second cancer cell are detected by transcriptome analysis or flow cytometry.

13. The immune cell, for use according to any preceding claim, wherein the threshold levels of antigen expression are measured by flow cytometry using unstained and an Ig isotype control as background, wherein cells that have the same intensity of staining as the unstained and Ig isotype control-stained cells are considered to not have detectable expression and / or wherein cells that show a staining intensity above the controls are said to express the antigen.

14. The immune cell, for use according to any preceding claim, wherein the amount of the at least one target antigen expressed by the second cancer cell may be: (i) at least 1%, 2%, 3%, 4% or 5% lower than the expression level of the at least one target antigen expressed by the first cancer cell; (ii) at least 6%, 7%, 8%, 9% or 10% lower than the expression level of the at least one target antigen expressed by the first cancer cell; (iii) at least 15%, 20%, 25%, 30% or 35% lower than the expression level of the at least one target antigen expressed by the first cancer cell;(iv) at least 40%, 45%, 50%, 55% or 60% lower than the expression level of the at least one target antigen expressed by the first cancer cell; (v) at least 65%, 70%, 75%, 80% or 85% lower than the expression level of the at least one target antigen expressed by the first cancer cell; or (vi) at least 90%, 95%, or 100% lower than the expression level of the at least one target antigen expressed by the first cancer cell.

15. The immune cell, for use according to any preceding claim, wherein the first cancer cell is comprised in or part of a tumour, and / or the second cancer cell is comprised in or part of a tumour.

16. The immune cell, for use according to any preceding claim, wherein the first and second cancer cells are comprised in or part of the same tumour.

17. The immune cell, for use according to any preceding claim, wherein the first cancer cell and / or the second cancer cell may express CD1d.

18. The immune cell, for use according to any preceding claim, wherein (i) the one or more antigen-binding domain is configured to specifically target the at least one target antigen on the first cancer cell, optionally CD19, CD133, and / or CD33; and / or (ii) the one or more antigen-binding domain is configured to specifically target two or more target antigens on the first cancer cell, optionally CD19 and CD133, CD19 and CD33, or CD133 and CD33.

19. The immune cell, for use according to any preceding claim, wherein the one or more antigen-binding domain comprises a domain selected from the group consisting of: an antibody or antigen-binding fragment thereof, CDR, VL, VH and Fd; Fv, Fab, and Fab'; bivalent fragments such as F(ab')2; single chain Fv (scFv); bivalent or trivalent scFvs, scFv-Fc-scFv, an IgG-scFv, an IgG-dAb, KIH-IgG, ^^-BODY, KIH-Fc Fab / scFv, tri / tetraspecific, KIH tri / tetra-specific, or a tandem scFv, bispecific antibody (BsAb), and Fc fragment.

20. The immune cell, for use according to any preceding claim, wherein the one or more antigen-binding domain is monospecific and targets one target antigen.

21. The immune cell, for use according to any preceding claim, wherein the one or more antigen-binding domain is bi-specific and targets two target antigens.

22. The immune cell, for use according to any preceding claim, wherein the one or more antigen-binding domain comprises one or more chimeric antigen receptor (CAR).

23. The immune cell, for use according to claim 22, wherein the CAR comprises: (i) a CD19 CAR, a CD133 CAR, and / or a CD33 CAR; or (ii) a CD19 CAR, CD133 CAR, and CD33 CAR.

24. The immune cell, for use according to either claim 22 or claim 23, wherein the immune cell is an iNKT cell comprising a bi-specific CD19-CD133 CAR, a bi-specific CD19-CD33 CAR, or a bi-specific CD133-CD33 CAR.

25. The immune cell, for use according to any one of claims 22-24, wherein the CAR comprises an antigen-binding domain, optionally a CD19 specific binding domain, a hinge, a transmembrane domain, a domain of CD8α, a CD28 co-stimulatory domain, and a CD3ζ activation domain.

26. The immune cell, for use according to any one of claims 22-24, wherein the CAR comprises an antigen-binding domain, optionally a CD133 specific bind domain, a hinge, a transmembrane domain, a domain of CD8α, a 4-1BB co-stimulatory domain, and a CD3ζ activation domain.

27. The immune cell, for use according to any one of claims 22-24, wherein the CAR comprises an antigen-binding domain, optionally a CD33 specific bind domain, a hinge, a transmembrane domain, a domain of CD8α, a CD28 co-stimulatory domain, and a CD3ζ activation domain.

28. The immune cell, for use according to any preceding claim, wherein the immune cell is transduced with a bispecific anti-CD19 / anti-CD133 dual CAR, a bispecific anti-CD19 / anti-CD33 dual CAR, or a bispecific anti-CD133 / anti-CD33 dual CAR, optionally comprising a linker, which links the single-target CARs.

29. The immune cell, for use according to any one of claims 22-28, wherein the one or more CAR comprises an amino acid sequence substantially set out as SEQ ID No: 1, or a fragment or variant thereof, and / or is encoded by a nucleic acid sequence substantially set out as SEQ ID No: 2, or a fragment or variant thereof.

30. The immune cell, for use according to any one of claims 22-28, wherein the one or more CAR comprises an amino acid sequence substantially set out as SEQ ID No: 3, or a fragment or variant thereof and / or is encoded by a nucleic acid sequence substantially set out as SEQ ID No: 4, or a fragment or variant thereof.

31. The immune cell, for use according to any one of claims 22-28, wherein the one or more CAR comprises an amino acid sequence substantially set out as SEQ ID No: 12, or a fragment or variant thereof and / or is encoded by a nucleic acid sequence substantially set out as SEQ ID No: 13, or a fragment or variant thereof.

32. The immune cell, for use according to any one of claims 22-28, wherein the one or more CAR comprises an amino acid sequence substantially set out as SEQ ID No: 5, or a fragment or variant thereof and / or is encoded by a nucleic acid sequence substantially set out as SEQ ID No: 6, or a fragment or variant thereof.

33. The immune cell, for use according to any preceding claim, wherein the threshold level of the NKG2D stress ligand receptor on the immune cell is detectable by transcriptome analysis and / or flow cytometry.

34. The immune cell, for use according to any preceding claim, wherein following engagement between the antigen-binding domain on the immune cell and the antigen on the first cancer cell, the expression level of the NKG2D stress ligand receptor on the immune cell may increase by: (i) at least 1%, 2%, 3%, 4% or 5%; (ii) at least 6%, 7%, 8%, 9% or 10%; (iii) at least 15%, 20%, 25%, 30% or 35%; (iv) at least 40%, 45%, 50%, 55% or 60%; (v) at least 65%, 70%, 75%, 80% or 85%; or (vi) at least 90%, 95%, or 100%.

35. The immune cell, for use according to any preceding claim, wherein the cancer being treated is a malignancy, which is a solid tumour or a liquid tumour.

36. The immune cell, for use according to any preceding claim, wherein the cancer is myeloid or lymphoid leukaemia, B cell and T cell lymphoma, plasma cell dyscrasia and solid tumours cancer of the brain, respiratory tract, head and neck, or skin.

37. The immune cell, for use according to any preceding claim, wherein the cancer leptomeningeal, brain, or blood cancer, optionally primary or secondary cancer.

38. The immune cell, for use according to any preceding claim, wherein the cancer is leukaemia.

39. The immune cell, for use according to any preceding claim, wherein the leukaemia is acute lymphoblastic leukaemia (ALL), B cell acute lymphoblastic leukaemia (B-ALL), T cell acute lymphoblastic leukaemia (T-ALL), or acute myeloid leukaemia (AML).

40. The immune cell, for use according to any preceding claim, wherein the leukaemia is a KMT2A-rearranged leukaemia, infant ALL, or medullary and leptomeningeal leukaemia.

41. The immune cell, for use according to any preceding claim, wherein infectious disease is treated, including bacterial or viral disease.

42. A pharmaceutical composition comprising an immune cell and pharmaceutically acceptable vehicle, wherein the immune cell comprises one or more antigen-binding domain, which targets at least one target antigen which is expressed above a threshold level by a first cancer cell, and the at least one target antigen is expressed below the threshold level by a second cancer cell, wherein the second cancer cell expresses one or more NKG2D stress ligand, and the immune cell expresses one or more NKG2D stress ligand receptor, and wherein the immune cell kills the second cancer cell upon engagement between the one or more NKG2D stress ligand and the one or more NKG2D stress ligand receptor.

43. The pharmaceutical composition according to claim 42, wherein the pharmaceutical composition is for use in treating, preventing or ameliorating cancer.

44. A method for preparing the pharmaceutical composition according to either claim 42 or claim 43, the method comprising contacting an immune cell with a pharmaceutically acceptable vehicle, wherein the immune cell comprises one or more antigen-binding domain, which targets at least one target antigen which is expressed above a threshold level by a first cancer cell, and the at least one target antigen is expressed below the threshold level by a second cancer cell, wherein the second cancer cell expresses one or more NKG2D stress ligand, and the immune cellexpresses one or more NKG2D stress ligand receptor, and wherein the immune cell kills the second cancer cell upon engagement between the one or more NKG2D stress ligand and the one or more NKG2D stress ligand receptor.

45. The pharmaceutical composition according to either claim 42 or claim 43, or the method according to claim 44, wherein the immune cell is as defined in any one of claims 1-41.

Images