Technical Field

The present disclosure relates to the fields of molecular biology, more specifically antigen-binding molecule technology. The present invention also relates to methods of medical treatment and prophylaxis.

Background

Anti-B7-H3 agents have been developed, including blocking antibodies, antibody-drug conjugates, CD16 affinity engineered antibodies and CD3-engaging bispecific antibodies. Some candidates have advanced into clinical trials (1 ,4). Early data from these clinical trials have suggested a favourable safety profile with limited toxicity, although the extent of therapeutic response remains to be established (1 ,4). In parallel, several groups have embarked on the development of B7-H3-targeting chimeric antigen receptors (CAR) T cells. These efforts have demonstrated good anti-tumor activity against multiple solid tumors, leukemia and lymphoma in preclinical studies (9-15). B7-H3 targeting CAR T cells were highly active in indications in which current therapeutic options have proven ineffective (9-11).

It has been recognized in recent years that scFv aggregation can induce tonic signalling, which in turn diminishes CAR-T activity and persistence (17-19). scFv aggregation or misfolding could be caused by low folding stabilities of the VH or VL domain, or exposure of hydrophobic residues at the VH-VL interface after deleting constant domains. Furthermore, scFv linkers can sterically constrain VH-VL domain interaction and result in oligomerization (20).

Summary

The present disclosure provides an antigen-binding molecule, optionally isolated, which binds to B7 homolog 3 (B7-H3).

In some aspects and embodiments, the antigen-binding molecule comprises a single-domain antibody sequence incorporating the following CDRs:

CDR1 having the amino acid sequence of SEQ ID NO:1 CDR2 having the amino acid sequence of SEQ ID NO:2 CDR3 having the amino acid sequence of SEQ ID NO:3.

In some embodiments, the antigen-binding molecule comprises, or consists of, an amino acid sequence having at least 70% sequence identity to the amino acid sequence of SEQ ID NO:8.

In some embodiments, the antigen-binding molecule comprises a single-domain antibody sequence incorporating the following FRs:

FR1 having the amino acid sequence of SEQ ID NO:4 FR2 having the amino acid sequence of SEQ ID NO:5 FR3 having the amino acid sequence of SEQ ID NO:6 FR4 having the amino acid sequence of SEQ ID NO:7.

In some embodiments, the antigen-binding molecule is a multispecific antigen-binding molecule, and wherein the antigen-binding molecule further comprises an antigen-binding domain that binds to an antigen other than B7-H3.

The present disclosure also provides a chimeric antigen receptor (CAR) comprising an antigenbinding molecule according to the present disclosure.

In some embodiments, the CAR comprises, or consists of, an amino acid sequence having at least 70% sequence identity to the amino acid sequence of SEQ ID NO:9.

The present disclosure also provides a nucleic acid, or a plurality of nucleic acids, optionally isolated, encoding an antigen-binding molecule or CAR according to the present disclosure.

The present disclosure also provides an expression vector, or a plurality of expression vectors, comprising a nucleic acid or a plurality of nucleic acids according to the present disclosure.

The present disclosure also provides a cell comprising an antigen-binding molecule, CAR, nucleic acid or plurality of nucleic acids, or expression vector or plurality of expression vectors according to the present disclosure.

In some embodiments, the cell is an immune cell. In some embodiments, the immune cell is a T cell.

In some embodiments, the cell is a virus-specific T cell. In some embodiments, the cell is an Epstein Barr Virus (EBV)-specific T cell.

The present disclosure also provides a method comprising culturing a cell according to the present disclosure under conditions suitable for expression of an antigen-binding molecule or CAR by the cell.

The present disclosure also provides a composition comprising an antigen-binding molecule, CAR, nucleic acid or plurality of nucleic acids, expression vector or plurality of expression vectors or cell according to the present disclosure, and a pharmaceutically acceptable carrier, diluent, excipient or adjuvant. The present disclosure also provides an antigen-binding molecule, CAR, nucleic acid or plurality of nucleic acids, expression vector or plurality of expression vectors, cell or composition according to the present disclosure, for use in a method of medical treatment or prophylaxis.

The present disclosure also provides an antigen-binding molecule, CAR, nucleic acid or plurality of nucleic acids, expression vector or plurality of expression vectors, cell or composition according to the present disclosure, for use in the treatment or prevention of a cancer. In some embodiments, the use is for the treatment or prevention of a chronic infection.

In some embodiments, the cancer is selected from the group consisting of: a B7-H3-positive cancer, lung cancer, non-small-cell lung cancer, small cell lung cancer, lung adenocarcinoma, squamous lung cell carcinoma, skin cancer, cutaneous squamous cell carcinoma, melanoma, pancreatic cancer, liver cancer, hepatocellular carcinoma, cholangiocarcinoma, intrahepatic cholangiocarcinoma, colorectal cancer, colorectal carcinoma, colon cancer, colon carcinoma, kidney cancer, clear cell renal carcinoma, Wilms' tumor, prostate cancer, ovarian cancer, ovarian carcinoma, cervical cancer, endometrial cancer, germ cell tumor, gastric cancer, gastric carcinoma, gastric adenocarcinoma, gastrointestinal adenocarcinoma, breast cancer, triple-negative breast cancer, head and neck cancer, head and neck squamous cell carcinoma, oral cavity cancer, oral squamous cell carcinoma, laryngeal cancer, oropharyngeal cancer, oropharyngeal carcinoma, nasopharyngeal carcinoma, esophageal cancer, bladder cancer, urothelial cancer, brain cancer, medulloblastoma, ependymoma medulloblastoma, glioma, diffuse intrinsic pontine glioma, diffuse midline glioma, choroid plexus carcinoma, pineoblastoma, neuroblastoma, CNS tumor, primitive neuroectodermal tumor, atypical teratoid/rhabdoid tumor, brain stem glioma, sarcoma, rhabdomyosarcoma, osteosarcoma, Ewing sarcoma, peritoneal cancer, desmoplastic small round cell tumor and mesothelioma.

In some embodiments, the antigen-binding molecule, CAR, nucleic acid or plurality of nucleic acids, expression vector or plurality of expression vectors, cell or composition according to the present disclosure are for use in targeting myeloid-derived suppressor cells (MDSCs). MSDCs are a heterogenous group of immune cells from the myeloid lineage (a family of cells that originate from bone marrow stem cells) and that expand under pathologic conditions such as a chronic infection or cancer. Tumors with high levels of infiltration by MDSCs have been associated with poor patient outcome and resistance to therapies. The MDSCs may be B7-H3 positive MDSCs. The treatment may abrogate the immunosuppressive effects of MDSCs.

In some embodiments, the treatment is not associated with cytokine release syndrome, or is minimally associated with cytokine release syndrome. In other words, compared with other CAR treatments for cancer, treatments with B7-H3 CARs as disclosed herein are associated with lower risk of CRS. For example, the treatment is associated with no changes in the level of IL-6, IL-8, IL10, IFN-y, TNFa and/or IL-2, neutrophils, monocytes and/or dendritic cells, or the levels of these cytokines of cells are lower following treatment with B7-H3 CARs as described herein than following treatment with other CARs, such as CD19-CAR.

The present disclosure also provides the use of an antigen-binding molecule, CAR, nucleic acid or plurality of nucleic acids, expression vector or plurality of expression vectors, cell or composition according to the present disclosure to deplete or increase killing of cells expressing B7-H3.

The present disclosure also provides an in vitro complex, optionally isolated, comprising an antigenbinding molecule or CAR according to the present disclosure bound to B7-H3.

The present disclosure also provides a method for detecting B7-H3 in a sample, comprising contacting a sample containing, or suspected to contain, B7-H3 with an antigen-binding molecule according to the present disclosure, and detecting the formation of a complex of the antigen-binding molecule with B7-H3.

The present disclosure also provides a method of selecting or stratifying a subject for treatment with a B7-H3-targeted agent, the method comprising contacting, in vitro, a sample from the subject with an antigen-binding molecule according to the present disclosure and detecting the formation of a complex of the antigen-binding molecule with B7-H3.

The present disclosure also provides the use of an antigen-binding molecule according to the present disclosure as an in vitro or in vivo diagnostic or prognostic agent.

Description

The present disclosure provides antigen-binding molecules that bind to B7-H3, having novel biophysical and/or functional properties as compared to antigen-binding molecules disclosed in the prior art.

The present disclosure also provides novel chimeric antigen receptor (CAR) constructs having a B7- H3-binding domain comprising the novel B7-H3-specific antigen-binding molecules of the present disclosure.

In preferred embodiments, the antigen-binding molecules are single-domain antibodies (/.e. VHH). CARs comprising VHH antigen-binding domains possess technical advantages over CARs comprising scFv antigen-binding domains. VHH are significantly smaller than scFv in size (12-15kDa versus 30kDa), and have more favourable in vivo immunogenicity, solubility, and stability qualities, while retaining the ability to bind to their target antigens with high affinity (21). VHH also avoid the potential disrupted interaction between variable domains and constant domains, and the exposure of hydrophobic patches, both which may severely affect solubility and stability (22). B7-H3

B7 homolog 3 (B7-H3, CD276) is the protein identified by UniProt Q5ZPR3. B7-H3 is an immune checkpoint member of the B7 and CD28 protein families that has a predominantly T cell inhibitory role by suppressing activation and proliferation (1).

B7-H3 isoform 1 (UniProt: Q5ZPR3-1 ; SEQ ID NO:14) comprises an N-terminal signal peptide (SEQ ID NO:15), a 438 amino acid extracellular domain (SEQ ID NO:16), a transmembrane domain (SEQ ID NO:17) and a short intracellular domain (SEQ ID NO:18). B7-H3 isoform 2 (UniProt: Q5ZPR3-2; SEQ ID NO:19) differs from B7-H3 isoform 1 in that it lacks positions 159-376 of SEQ ID NO:14. B7- H3 isoform 3 (UniProt: Q5ZPR3-3; SEQ ID NO:21) differs from B7-H3 isoform 1 in that it lacks positions 494-534 of SEQ ID NO:14, and in that positions 465-493 are replaced with ‘GPASSAVPLSPAHPPHGSMCWSHWFSRGL’. B7-H3 isoform 4 (UniProt: Q5ZPR3-4; SEQ ID NO:22) differs from B7-H3 isoform 1 in that positions 528-534 are replaced with ‘GKDTWA’. The mature sequences of human B7-H3 isoforms 1 to 4 following processing to remove the N-terminal signal peptide are shown in SEQ ID NOs:28 to 31 .

B7-H3 isoform 1 is also known as 4lg-B7-H3, with reference to the four immunoglobulin(lg)-like domains within its extracellular domain. The V-type and C2-type Ig-like domains of B7-H3 isoform 1 are shown in SEQ ID NOs:24 to 27, and are also comprised in isoforms 3 and 4. B7-H3 isoform 2 is also known as 2lg-B7-H3, and contains one single set of V-type and C2-type Ig-like domains, shown in SEQ ID NOs:24 and 27.

The detection of high levels of B7-H3 mRNA in a wide range of normal tissues contrasts with the limited expression of B7-H3 protein in these tissues, suggesting a tight regulation of B7-H3 at the transcriptional level in healthy tissues (2, 3). By contrast, B7-H3 is over-expressed at both mRNA and protein levels in multiple types of human tumors (4) with B7-H3 protein detected in cancer cells, tumor-infiltrating blood vessels and tumor stroma (5). Association studies revealed that a high expression of B7-H3 protein was correlated with bad prognosis and poor clinical outcome. These clinical observations were supported by studies that demonstrate a pro-oncogenic role for B7-H3 in various types of cancer that is independent of its immune functions. B7-H3 acts upstream from multiple signal transduction pathways, including the JAK/STAT, Ras/Raf/MEK/MAPK and PI3K/Akt/mTOR pathways. Abrogating B7-H3 expression decreases adhesion, migration, invasion, and metastasis of many types of cancer cells in vitro and in vivo (6-8). Taken together, these studies point to pleiotropic pro-oncogenic roles as well as differential expression of B7-H3 in tumors versus healthy tissues, making B7-H3 an attractive target for cancer therapy.

In this specification, ‘B7-H3’ refers to B7-H3 from any species, and includes isoforms, fragments, variants or homologues from any species. In some embodiments, B7-H3 is B7-H3 from a mammal (e.g. a therian, placental, epitherian, preptotheria, archontan, primate (rhesus, cynomolgous, nonhuman primate or human)). In some embodiments, the B7-H3 is human B7-H3 or mouse B7-H3. As used herein, isoforms, fragments, variants or homologues of a given reference protein may be characterised as having at least 70% sequence identity, preferably one of 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% amino acid sequence identity to the amino acid sequence of the reference protein.

A ‘fragment’ generally refers to a fraction of the reference protein. A ‘variant’ generally refers to a protein having an amino acid sequence comprising one or more amino acid substitutions, insertions, deletions or other modifications relative to the amino acid sequence of the reference protein, but retaining a considerable degree of sequence identity (e.g. at least 60%) to the amino acid sequence of the reference protein. A fragment of B7-H3 may have a minimum length of one of 10, 20, 30, 40, 50, 100, 200, 300, 400, 500 or 530 amino acids, and may have a maximum length of one of 10, 20, 30, 40, 50, 100, 200, 300, 400, 500 or 530 amino acids.

An ‘isoform’ generally refers to a variant of the reference protein expressed by the same species as the species of the reference protein. Isoforms of B7-H3 of course include Isoform 1 (UniProt: Q5ZPR3-1), Isoform 2 (Q5ZPR3-2), Isoform 3 (Q5ZPR3-3) and Isoform 4 (Q5ZPR3-3).

A ‘homologue’ generally refers to a variant of the reference protein produced by a different species as compared to the species of the reference protein. Homologues include orthologues.

Isoforms, fragments, variants or homologues of B7-H3 may optionally be characterised as having at least 70%, preferably one of 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% amino acid sequence identity to the amino acid sequence of an immature or mature B7-H3 isoform from a given species, e.g. human.

In some embodiments, the B7-H3 is human B7-H3. In some embodiments, the B7-H3 is mouse B7- H3 (Q8VE98-1 , SEQ ID NO:32).

Isoforms, fragments, variants or homologues may optionally be functional isoforms, fragments, variants or homologues, e.g. having a functional property/activity of the reference B7-H3, as determined by analysis by a suitable assay for the functional property/activity. For example, an isoform, fragment, variant or homologue of B7-H3 may induce B7-H3-mediated signalling.

In some embodiments, the B7-H3 comprises, or consists of, an amino acid sequence having at least 70%, preferably one of 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% amino acid sequence identity to SEQ ID NO:14, 19, 20, 21 , 22, 28, 29, 30, 31 , 32 or 39.

In some embodiments, the B7-H3 comprises an amino acid sequence having at least 70%, preferably one of 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% amino acid sequence identity to SEQ ID NO:22, 16, 31 or 28. In some embodiments, the B7-H3 comprises an amino acid sequence having at least 70%, preferably one of 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% amino acid sequence identity to SEQ ID NO:16 or 20.

In some embodiments, a fragment of B7-H3 comprises, or consists of, an amino acid sequence having at least 70%, preferably one of 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% amino acid sequence identity to SEQ ID NO:22, 16, 31 or 28. In some embodiments, a fragment of B7-H3 comprises, or consists of, an amino acid sequence having at least 70%, preferably one of 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% amino acid sequence identity to SEQ ID NO:16 or 20.

Antigen-binding molecules

The present disclosure provides antigen-binding molecules capable of binding to (/.e. which bind to) B7-H3. The present disclosure provides antigen-binding molecules which bind specifically to B7-H3. Antigen-binding molecules according to the present disclosure may be provided in purified or isolated form, i.e. from other naturally-occurring biological material.

As used herein, an ‘antigen-binding molecule’ refers to a molecule which is capable of binding to a target antigen. The term ‘antigen-binding molecule’ encompasses monoclonal antibodies, polyclonal antibodies, monospecific and multispecific antibodies (e.g., bispecific antibodies), and antibody fragments (e.g. Fv, scFv, Fab, scFab, F(ab’)2, Fab2, diabodies, triabodies, scFv-Fc, minibodies, single-domain antibodies (VHH), etc.) and aptamers.

More specifically, antigen-binding molecules according to the present disclosure comprise antigenbinding polypeptide moieties, which may be referred to as ‘antigen-binding domains’. In preferred embodiments, antigen-binding molecules according to the present disclosure comprise, or consist of, a single-domain antibody which binds specifically to B7-H3.

Single-domain antibodies (sdAbs) - also referred to variously in the art as ‘single variable domain on a heavy chain antibodies’, ‘VHHs’, ‘nanobodies’ and ‘heavy chain only antibodies (HcAbs)’ - are described e.g. in Henry and MacKenzie, Front Immunol. (2018) 9:41 and Bever ef a/., Anal Bioanal Chem. (2016) 408(22): 5985-6002, both of which are hereby incorporated by reference in their entirety.

Single-domain antibodies are formed of a single, monomeric antibody variable domain. The first single-domain antibodies were engineered from heavy-chain antibodies found in camelids, and cartilaginous fishes also have heavy-chain antibodies. Single-domain antibodies according to the present disclosure generally comprise three complementarity-determining regions CDRs: CDR1 , CDR2 and CDR3. The three CDRs together define the paratope of the molecule, which is the part through which it binds to its target antigen.

Single-domain antibodies further comprise framework regions (FRs) either side of each CDR, which provide a scaffold for the CDRs. From N-terminus to C-terminus, single-domain antibodies comprise the following structure: N term-[FR1]-[CDR1]-[FR2]-[CDR2]-[FR3]-[CDR3]-[FR4]-C term.

There are several different conventions for defining antibody CDRs and FRs, such as those described in Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD (1991), the Clothia system (Chothia et al., J. Mol. Biol. 196:901-917 (1987)), the IMGT information system (international IMGT (ImMunoGeneTics) information system (described in LeFranc et al., Nucleic Acids Res. (2015) 43 (Database issue):D413-22), which uses the IMGT V-DOMAIN numbering rules as described in Lefranc et al., Dev. Comp. Immunol. (2003) 27:55-77) and VBASE2, as described in Retter et al., Nucl. Acids Res. (2005) 33 (suppl 1): D671-D674. The CDRs and FRs of the antigen-binding molecules/single-domain antibodies described herein are defined according to VBASE2.

In some embodiments, the antigen-binding molecule comprises the CDRs of a B7-H3-binding singledomain antibody described herein, or comprises CDRs which are derived from a B7-H3-binding single-domain antibody described herein. In some embodiments, the antigen-binding molecule comprises the FRs of a B7-H3-binding single-domain antibody described herein, or comprises FRs which are derived from a B7-H3-binding single-domain antibody described herein. In some embodiments, the antigen-binding molecule comprises the CDRs and the FRs of a B7-H3-binding single-domain antibody described herein, or comprises CDRs and FRs which are derived from a B7- H3-binding single-domain antibody described herein. That is, in some embodiments the antigenbinding molecule comprises the amino acid sequence of a B7-H3-binding single-domain antibody described herein, or comprises amino acid sequence which is derived from a B7-H3-binding singledomain antibody described herein.

In some embodiments, in the amino acid sequences of SEQ ID NO:8: FR1 is formed by the amino acid sequence at positions 1 to 25; CDR1 is formed by the amino acid sequence at positions 26 to 33; FR2 is formed by the amino acid sequence at positions 34 to 50; CDR2 is formed by the amino acid sequence at positions 51 to 57; FR3 is formed by the amino acid sequence at positions 58 to 96; CDR3 is formed by the amino acid sequence at positions 97 to 1 11 ; and FR4 is formed by the amino acid sequence at positions 112 to 122.

As used herein, an amino acid sequence/domain which is ‘derived from’ a reference amino acid sequence/domain comprises an amino acid sequence having at least 60%, e.g. one of at least 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% sequence identity to the reference sequence.

In some embodiments the antigen-binding molecule comprises the CDRs, FRs and/or the complete amino acid sequence of the B7-H3-binding single-domain antibody P2A5.

In some embodiments the antigen-binding molecule comprises the CDRs, FRs and/or the complete amino acid sequence of a B7-H3-binding single-domain antibody having an amino acid sequence according to SEQ ID NO:8. In some embodiments the antigen-binding molecule comprises the CDRs (/.e. CDRs 1 , 2 and 3) of a B7-H3-binding single-domain antibody having an amino acid sequence according to SEQ ID NO:8. In some embodiments the antigen-binding molecule comprises the FRs (/.e. FRs 1 , 2, 3 and 4) of a B7-H3-binding single-domain antibody having an amino acid sequence according to SEQ ID NO:8. In some embodiments the antigen-binding molecule comprises the CDRs (/.e. CDRs 1 , 2 and 3) and FRs (/.e. FRs 1 , 2, 3 and 4) of a B7-H3- binding single-domain antibody having an amino acid sequence according to SEQ ID NO:8.

In some embodiments the antigen-binding molecule comprises, or consists of, a single-domain antibody sequence according to (1):

(1) (P2A5) a single-domain antibody sequence comprising the following CDRs:

CDR1 having the amino acid sequence of SEQ ID NO:1 CDR2 having the amino acid sequence of SEQ ID NO:2 CDR3 having the amino acid sequence of SEQ ID NO:3, or a variant thereof in which 1 or 2 or 3 amino acids in CDR1 , and/or in which 1 or 2 or 3 amino acids in CDR2, and/or in which 1 or 2 or 3 amino acids in CDR3 are substituted with another amino acid.

In some embodiments the antigen-binding molecule comprises, or consists of, a single-domain antibody sequence according to (2):

(2) (P2A5) a single-domain antibody sequence comprising the following FRs:

FR1 having the amino acid sequence of SEQ ID NO:4 FR2 having the amino acid sequence of SEQ ID NO:5 FR3 having the amino acid sequence of SEQ ID NO:6 FR4 having the amino acid sequence of SEQ ID NO:7, or a variant thereof in which 1 or 2 or 3 amino acids in FR1 , and/or in which 1 or 2 or 3 amino acids in FR2, and/or in which 1 or 2 or 3 amino acids in FR3, and/or in which 1 or 2 or 3 amino acids in FR4 are substituted with another amino acid. In some embodiments, the antigen-binding molecule comprises, or consists of, a single-domain antibody sequence comprising the CDRs according to (1), and the FRs according to (2) above.

In some embodiments, the antigen-binding molecule comprises, or consists of, a single-domain antibody sequence according to (3):

(3) (P2A5) a single-domain antibody sequence comprising an amino acid sequence having at least 70% sequence identity, more preferably one of at least 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity, to the amino acid sequence of SEQ ID NO:8.

In embodiments in accordance with the present disclosure, one or more amino acids are substituted with another amino acid. A substitution comprises substitution of an amino acid residue with a nonidentical 'replacement' amino acid residue. A replacement amino acid residue of a substitution according to the present disclosure may be a naturally-occurring amino acid residue (/.e. encoded by the genetic code) which is non-identical to the amino acid residue at the relevant position of the equivalent, unsubstituted amino acid sequence, selected from: alanine (Ala), arginine (Arg), asparagine (Asn), aspartic acid (Asp), cysteine (Cys), glutamine (Gin), glutamic acid (Glu), glycine (Gly), histidine (His), isoleucine (He): leucine (Leu), lysine (Lys), methionine (Met), phenylalanine (Phe), proline (Pro), serine (Ser), threonine (Thr), tryptophan (Trp), tyrosine (Tyr), and valine (Vai). In some embodiments, a replacement amino acid may be a non-naturally occurring amino acid residue -i.e. an amino acid residue other than those recited in the preceding sentence. Examples of non- naturally occurring amino acid residues include norleucine, ornithine, norvaline, homoserine, aib, and other amino acid residue analogues such as those described in Ellman, etal., Meth. Enzym. 202 (1991) 301-336.

In some embodiments, a substitution may be biochemically conservative. In some embodiments, where an amino acid to be substituted is provided in one of rows 1 to 5 of the table below, the replacement amino acid of the substitution is another, non-identical amino acid provided in the same row: By way of illustration, in some embodiments wherein substitution is of a Met residue, the replacement amino acid may be selected from Ala, Vai, Leu, He, Trp, Tyr, Phe and Norleucine.

In some embodiments, a replacement amino acid in a substitution may have the same side chain polarity as the amino acid residue it replaces. In some embodiments, a replacement amino acid in a substitution may have the same side chain charge (at pH 7.4) as the amino acid residue it replaces:

That is, in some embodiments, a nonpolar amino acid is substituted with another, non-identical nonpolar amino acid. In some embodiments, a polar amino acid is substituted with another, nonidentical polar amino acid. In some embodiments, an acidic polar amino acid is substituted with another, non-identical acidic polar amino acid. In some embodiments, a basic polar amino acid is substituted with another, non-identical basic polar amino acid. In some embodiments, a neutral amino acid is substituted with another, non-identical neutral amino acid. In some embodiments, a positive amino acid is substituted with another, non-identical positive amino acid. In some embodiments, a negative amino acid is substituted with another, non-identical negative amino acid. In some embodiments, substitution(s) may be functionally conservative. That is, in some embodiments, the substitution may not affect (or may not substantially affect) one or more functional properties (e.g. target binding) of the antigen-binding molecule comprising the substitution as compared to the equivalent unsubstituted molecule.

In some embodiments, the antigen-binding molecule of the present disclosure comprises one or more regions (e.g. CH1 , CH2 and/or CH3) of an immunoglobulin heavy chain constant sequence. In some embodiments, the immunoglobulin heavy chain constant sequence is, or is derived from, the heavy chain constant sequence of an IgG (e.g. lgG1 , lgG2, lgG3, lgG4), IgA (e.g. lgA1 , lgA2), IgD, IgE or IgM, e.g. a human IgG (e.g. hlgG1 , hlgG2, hlgG3, hlgG4), hlgA (e.g. hlgA1 , hlgA2), hlgD, hlgE or hlgM. In some embodiments the immunoglobulin heavy chain constant sequence is, or is derived from, the heavy chain constant sequence of a human IgG 1 allotype (e.g. G1 ml , G1 m2, G1 m3 or G1 m17).

In some embodiments, the antigen-binding molecule comprises an amino acid sequence having at least 70% sequence identity more preferably one of at least 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100%, sequence identity to the amino acid sequence of SEQ ID NO:40, 45 or 48.

In some embodiments, the antigen-binding molecule comprises an amino acid sequence having at least 70% sequence identity more preferably one of at least 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100%, sequence identity to the amino acid sequence of SEQ ID NO:49, 50 or 51 .

In some embodiments, the antigen-binding molecule comprises a CH1 region comprising an amino acid sequence having at least 70% sequence identity more preferably one of at least 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100%, sequence identity to the amino acid sequence of SEQ ID NO:41 or 46. In some embodiments, the antigen-binding molecule comprises a hinge region comprising an amino acid sequence having at least 70% sequence identity more preferably one of at least 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100%, sequence identity to the amino acid sequence of SEQ ID NO:42. In some embodiments, the antigen-binding molecule comprises a CH2 region comprising an amino acid sequence having at least 70% sequence identity more preferably one of at least 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100%, sequence identity to the amino acid sequence of SEQ ID NO:43. In some embodiments, the antigen-binding molecule comprises a CH3 region comprising an amino acid sequence having at least 70% sequence identity more preferably one of at least 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100%, sequence identity to the amino acid sequence of SEQ ID NO:44 or 47. It will be appreciated that CH2 and/or CH3 regions may be provided with further substitutions in accordance with modification to an Fc region of the antigen-binding molecule as described herein.

In some embodiments, the antigen-binding molecule of the present disclosure comprises one or more regions of an immunoglobulin light chain constant sequence. In some embodiments, the immunoglobulin light chain constant sequence is human immunoglobulin kappa constant (IGKC; CK). In some embodiments, the immunoglobulin light chain constant sequence is a human immunoglobulin lambda constant (IGLC; CA), e.g. IGLC1 , IGLC2, IGLC3, IGLC6 or IGLC7.

In some embodiments, the antigen-binding molecule comprises an amino acid sequence having at least 70% sequence identity more preferably one of at least 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100%, sequence identity to the amino acid sequence of SEQ ID NO:52, 53, 54, 55, 56 or 57.

In some embodiments, the antigen-binding molecule of the present disclosure comprises an Fc region.

An Fc region is composed of CH2 and CH3 regions from one polypeptide, and CH2 and CH3 regions from another polypeptide. The CH2 and CH3 regions from the two polypeptides together form the Fc region.

Fc-mediated functions include Fc receptor binding, antibody-dependent cellular cytotoxicity (ADCC), antibody-dependent cell-mediated phagocytosis (ADCP), complement-dependent cytotoxicity (CDC), formation of the membrane attack complex (MAC), cell degranulation, cytokine and/or chemokine production, and antigen processing and presentation. Modifications to antibody Fc regions that influence Fc-mediated functions are known in the art, such as those described e.g. in Wang etal., Protein Cell (2018) 9(1):63-73, which is hereby incorporated by reference in its entirety. Exemplary Fc region modifications known to influence antibody effector function are summarised in Table 1 of Wang et al., Protein Cell (2018) 9(1):63-73. In some embodiments, the antigen-binding molecule of the present disclosure comprises an Fc region comprising modification to increase or reduce an Fc- mediated function as compared to an antigen-binding molecule comprising the corresponding unmodified Fc region.

Where an Fc region/CH2/CH3 is described as comprising modification(s) ‘corresponding to’ reference substitution(s), equivalent substitution(s) in the homologous Fc/CH2/CH3 are contemplated. By way of illustration, L234A/L235A substitutions in human lgG1 (numbered according to the EU numbering system as described in Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD, 1991) correspond to L to A substitutions at positions 117 and 118 of the mouse Ig gamma-2A chain C region (UniProtKB: P01863-1 , v1). Where an Fc region is described as comprising a modification, the modification may be present in one or both of the polypeptide chains which together form the Fc region.

In some embodiments, the antigen-binding molecule of the present disclosure comprises an Fc region comprising modification. In some embodiments, the antigen-binding molecule of the present disclosure comprises an Fc region comprising modification in one or more of the CH2 and/or CH3 regions.

In some embodiments, the Fc region comprises modification to increase an Fc-mediated function. In some embodiments, the Fc region comprises modification to increase ADCC. In some embodiments, the Fc region comprises modification to increase ADCP. In some embodiments, the Fc region comprises modification to increase CDC. An antigen-binding molecule comprising an Fc region comprising modification to increase an Fc-mediated function (e.g. ADCC, ADCP, CDC) induces an increased level of the relevant effector function as compared to an antigen-binding molecule comprising the corresponding unmodified Fc region.

In some embodiments, the Fc region comprises modification to increase binding to an Fc receptor.

In some embodiments, the Fc region comprises modification to increase binding to an Fey receptor.

In some embodiments, the Fc region comprises modification to increase binding to one or more of

FcyRI, FcyRlla, FcyRllb, FcyRllc, FcyRllla and FcyRlllb. In some embodiments, the Fc region comprises modification to increase binding to FcyRllla. In some embodiments, the Fc region comprises modification to increase binding to FcyRlla. In some embodiments, the Fc region comprises modification to increase binding to FcyRllb. In some embodiments, the Fc region comprises modification to increase binding to FcRn. In some embodiments, the Fc region comprises modification to increase binding to a complement protein. In some embodiments, the Fc region comprises modification to increase binding to C1q. In some embodiments, the Fc region comprises modification to promote hexamerisation of the antigen-binding molecule. In some embodiments, the Fc region comprises modification to increase antigen-binding molecule half-life. In some embodiments, the Fc region comprises modification to increase co-engagement.

In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions F243L/R292P/Y300L/V305I/P396L as described in Stavenhagen et al. Cancer Res. (2007) 67:8882-8890. In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions S239D/I332E or S239D/I332E/A330L as described in Lazar et al., Proc Natl Acad Sci USA. (2006)103:4005-4010. In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions S298A/E333A/K334A as described in Shields et al., J Biol Chem. (2001) 276:6591-6604. In some embodiments, the Fc region comprises modification to one of heavy chain polypeptides corresponding to the combination of substitutions L234Y/L235Q/G236W/S239M/H268D/D270E/S298A, and modification to the other heavy chain polypeptide corresponding to the combination of substitutions D270E/K326D/A330M/K334E, as described in Mimoto et al., MAbs. (2013): 5:229-236. In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions G236A/S239D/I332E as described in Richards et al., Mol Cancer Ther. (2008) 7:2517-2527.

In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions K326W/E333S as described in Idusogie et al. J Immunol. (2001) 166(4):2571-5. In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions S267E/H268F/S324T as described in Moore et al. MAbs. (2010) 2(2):181-9. In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions described in Natsume et al., Cancer Res. (2008) 68(10):3863-72. In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions E345R/E430G/S440Y as described in Diebolder et al. Science (2014) 343(6176): 1260- 3.

In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions M252Y/S254T/T256E as described in Dall’Acqua et al. J Immunol. (2002) 169:5171- 5180. In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions M428L/N434S as described in Zalevsky et al. Nat Biotechnol. (2010) 28:157-159.

In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions S267E/L328F as described in Chu et al., Mol Immunol. (2008) 45:3926-3933. In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions N325S/L328F as described in Shang et al. Biol Chem. (2014) 289:15309-15318.

In some embodiments, the Fc region comprises modification to reduce/prevent an Fc-mediated function. In some embodiments, the Fc region comprises modification to reduce/prevent ADCC. In some embodiments, the Fc region comprises modification to reduce/prevent ADCP. In some embodiments, the Fc region comprises modification to reduce/prevent CDC. An antigen-binding molecule comprising an Fc region comprising modification to reduce/prevent an Fc-mediated function (e.g. ADCC, ADCP, CDC) induces an reduced level of the relevant effector function as compared to an antigen-binding molecule comprising the corresponding unmodified Fc region.

In some embodiments, the Fc region comprises modification to reduce/prevent binding to an Fc receptor. In some embodiments, the Fc region comprises modification to reduce/prevent binding to an Fey receptor. In some embodiments, the Fc region comprises modification to reduce/prevent binding to one or more of FcyRI, FcyRlla, FcyRllb, FcyRllc, FcyRllla and FcyRlllb. In some embodiments, the Fc region comprises modification to reduce/prevent binding to FcyRllla. In some embodiments, the Fc region comprises modification to reduce/prevent binding to FcyRlla. In some embodiments, the Fc region comprises modification to reduce/prevent binding to FcyRllb. In some embodiments, the Fc region comprises modification to reduce/prevent binding to a complement protein. In some embodiments, the Fc region comprises modification to reduce/prevent binding to C1q. In some embodiments, the Fc region comprises modification to reduce/prevent glycosylation of the amino acid residue corresponding to N297.

In some embodiments, the Fc region is not able to induce one or more Fc-mediated functions (/.e. lacks the ability to elicit the relevant Fc-mediated function(s)). Accordingly, antigen-binding molecules comprising such Fc regions also lack the ability to induce the relevant function(s). Such antigen-binding molecules may be described as being devoid of the relevant function(s).

In some embodiments, the Fc region is not able to induce ADCC. In some embodiments, the Fc region is not able to induce ADCP. In some embodiments, the Fc region is not able to induce CDC. In some embodiments, the Fc region is not able to induce ADCC and/or is not able to induce ADCP and/or is not able to induce CDC.

In some embodiments, the Fc region is not able to bind to an Fc receptor. In some embodiments, the Fc region is not able to bind to an Fey receptor. In some embodiments, the Fc region is not able to bind to one or more of FcyRI, FcyRlla, FcyRllb, FcyRllc, FcyRllla and FcyRlllb. In some embodiments, the Fc region is not able to bind to FcyRllla. In some embodiments, the Fc region is not able to bind to FcyRlla. In some embodiments, the Fc region is not able to bind to FcyRllb. In some embodiments, the Fc region is not able to bind to FcRn. In some embodiments, the Fc region is not able to bind to a complement protein. In some embodiments, the Fc region is not able to bind to C1q. In some embodiments, the Fc region is not glycosylated at the amino acid residue corresponding to N297.

In some embodiments, the Fc region comprises modification corresponding to N297A or N297Q or N297G as described in Leabman et al., MAbs. (2013) 5:896-903. In some embodiments, the Fc region comprises modification corresponding to L235E as described in Alegre et al., J Immunol. (1992) 148:3461-3468. In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions L234A/L235A or F234A/L235A as described in Xu et al., Cell Immunol. (2000) 200:16-26. In some embodiments, the Fc region comprises modification corresponding to P329A or P329G as described in Schlothauer et al., Protein Engineering, Design and Selection (2016), 29(10):457-466. In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions L234A/L235A/P329G as described in Lo et al. J. Biol. Chem (2017) 292(9):3900-3908. In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions described in Rother et al., Nat Biotechnol. (2007) 25:1256-1264. In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions S228P/L235E as described in Newman et al., Clin. Immunol. (2001) 98:164-174. In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions H268Q/V309L/A330S/P331S as described in An et al., MAbs. (2009) 1 :572-579. In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions V234A/G237A/P238S/H268A/V309L/A330S/P331S as described in Vafa et al., Methods. (2014) 65:114-126. In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions L234A/L235E/G237A/A330S/P331S as described in US 2015/0044231 A1.

The combination of substitutions ‘L234A/L235A’ and corresponding substitutions (such as e.g. F234A/L235A in human lgG4) are known to disrupt binding of Fc to Fey receptors and inhibit ADCC, ADCP, and also to reduce C1q binding and thus CDC (Schlothauer et al., Protein Engineering, Design and Selection (2016), 29(10):457-466, hereby incorporated by reference in entirety). The substitutions ‘P329G’ and ‘P329A’ reduce C1q binding (and thereby CDC). Substitution of ‘N297’ with ‘A’, ‘G’ or ‘Q’ is known to eliminate glycosylation, and thereby reduce Fc binding to C1q and Fey receptors, and thus CDC and ADCC. Lo et al. J. Biol. Chem (2017) 292(9):3900-3908 (hereby incorporated by reference in its entirety) reports that the combination of substitutions L234A/L235A/P329G eliminated complement binding and fixation as well as Fc y receptor dependent, antibody-dependent, cell-mediated cytotoxicity in both murine lgG2a and human lgG1.

The combination of substitutions L234A/L235E/G237A/A330S/P331 S in lgG1 Fc is disclosed in US 2015/0044231 A1 to abolish induction of phagocytosis, ADCC and CDC.

In some embodiments, the Fc region comprises modification corresponding to the substitution S228P as described in Silva et al., J Biol Chem. (2015) 290(9):5462-5469. The substitution S228P in lgG4 Fc reduces Fab-arm exchange (Fab arm exchange can be undesirable).

In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions L234A/L235A. In some embodiments, the Fc region comprises modification corresponding to the substitution P329G. In some embodiments, the Fc region comprises modification corresponding to the substitution N297Q.

In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions L234A/L235A/P329G.

In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions L234A/L235A/P329G/N297Q.

In some embodiments, the Fc region comprises modification corresponding to the combination of substitutions L234A/L235E/G237A/A330S/P331 S.

In some embodiments, the Fc region comprises modification corresponding to the substitution S228P, e.g. in lgG4. In some embodiments - particularly embodiments in which the antigen-binding molecule is a multispecific (e.g. bispecific) antigen-binding molecule - the antigen-binding molecule comprises an Fc region comprising modification in one or more of the CH2 and CH3 regions promoting association of the Fc region. Recombinant co-expression of constituent polypeptides of an antigen-binding molecule and subsequent association leads to several possible combinations. To improve the yield of the desired combinations of polypeptides in antigen-binding molecules in recombinant production, it is advantageous to introduce in the Fc regions modification(s) promoting association of the desired combination of heavy chain polypeptides. Modifications may promote e.g. hydrophobic and/or electrostatic interaction between CH2 and/or CH3 regions of different polypeptide chains. Suitable modifications are described e.g. in Ha et al., Front. Immnol (2016) 7:394, which is hereby incorporated by reference in its entirety.

In some embodiments, the antigen-binding molecule of the present disclosure comprises an Fc region comprising paired substitutions in the CH3 regions of the Fc region according to one of the following formats, as shown in Table 1 of Ha etal., Front. Immnol (2016) 7:394: KiH, KiH _{s s} , HA-TF, ZW1 , 7.8.60, DD-KK, EW-RVT, EW-RVTs-s, SEED or A107.

Multispecific antigen-binding molecules are also contemplated. By ‘multispecific’ it is meant that the antigen-binding molecule displays specific binding to more than one target. In some embodiments, the antigen-binding molecule is a bispecific antigen-binding molecule. In some embodiments, the antigen-binding molecule comprises at least two different antigen-binding domains.

In some embodiments, the antigen-binding molecule binds to B7-H3 and another target (e.g. an antigen other than B7-H3), and so is at least bispecific. The term ‘bispecific’ means that the antigenbinding molecule is able to bind specifically to at least two distinct antigenic determinants.

It will be appreciated that an antigen-binding molecule according to the present disclosure (e.g. a multispecific antigen-binding molecule) may comprise antigen-binding molecules capable of binding to the targets for which the antigen-binding molecule is specific. For example, an antigen-binding molecule which binds to B7-H3 and an antigen other than B7-H3 may comprise: (i) an antigenbinding molecule which binds to B7-H3, and (ii) an antigen-binding molecule which binds to an antigen other than B7-H3. In some embodiments, a component antigen-binding molecule of a larger antigen-binding molecule (e.g. a multispecific antigen-binding molecule) may be referred to e.g. as an ‘antigen-binding domain’ or ‘antigen-binding region’ of the larger antigen-binding molecule.

It will also be appreciated that an antigen-binding molecule according to the present disclosure (e.g. a multispecific antigen-binding molecule) may comprise antigen-binding polypeptides or antigenbinding polypeptide complexes capable of binding to the targets for which the antigen-binding molecule is specific. In some embodiments, the antigen other than B7-H3 in a multispecific antigen-binding molecule is an immune cell surface molecule. In some embodiments, the antigen is a cancer cell antigen. In some embodiments the antigen is a receptor molecule, e.g. a cell surface receptor. In some embodiments the antigen is a cell signalling molecule, e.g. a cytokine, chemokine, interferon, interleukin or lymphokine. In some embodiments the antigen is a growth factor or a hormone.

A cancer cell antigen is an antigen which is expressed or over-expressed by a cancer cell. A cancer cell antigen may be any peptide/polypeptide, glycoprotein, lipoprotein, glycan, glycolipid, lipid, or fragment thereof. A cancer cell antigen’s expression may be associated with a cancer. A cancer cell antigen may be abnormally expressed by a cancer cell (e.g. the cancer cell antigen may be expressed with abnormal localisation), or may be expressed with an abnormal structure by a cancer cell. A cancer cell antigen may be capable of eliciting an immune response. In some embodiments, the antigen is expressed at the cell surface of the cancer cell (/.e. the cancer cell antigen is a cancer cell surface antigen). In some embodiments, the part of the antigen which is bound by the antigenbinding molecule described herein is displayed on the external surface of the cancer cell (/.e. is extracellular). The cancer cell antigen may be a cancer-associated antigen. In some embodiments the cancer cell antigen is an antigen whose expression is associated with the development, progression or severity of symptoms of a cancer. The cancer-associated antigen may be associated with the cause or pathology of the cancer, or may be expressed abnormally as a consequence of the cancer. In some embodiments, the cancer cell antigen is an antigen whose expression is upregulated (e.g. at the RNA and/or protein level) by cells of a cancer, e.g. as compared to the level of expression by comparable non-cancerous cells (e.g. non-cancerous cells derived from the same tissue/cell type). In some embodiments, the cancer-associated antigen may be preferentially expressed by cancerous cells, and not expressed by comparable non-cancerous cells (e.g. non- cancerous cells derived from the same tissue/cell type). In some embodiments, the cancer- associated antigen may be the product of a mutated oncogene or mutated tumor suppressor gene. In some embodiments, the cancer-associated antigen may be the product of an overexpressed cellular protein, a cancer antigen produced by an oncogenic virus, an oncofetal antigen, or a cell surface glycolipid or glycoprotein.

Cancer-associated antigens are reviewed by Zarour HM, DeLeo A, Finn OJ, et al. Categories of Tumor Antigens. In: Kufe DW, Pollock RE, Weichselbaum RR, et al., editors. Holland-Frei Cancer Medicine. 6th edition. Hamilton (ON): BC Decker; 2003. Cancer-associated antigens include oncofetal antigens: CEA, Immature laminin receptor, TAG-72; oncoviral antigens such as HPV E6 and E7; overexpressed proteins: BING-4, calcium-activated chloride channel 2, cyclin-B1 , 9D7, Ep- CAM, EphA3, HER2/neu, telomerase, mesothelin, SAP-1 , survivin; cancer-testis antigens: BAGE, CAGE, GAGE, MAGE, SAGE, XAGE, CT9, CT10, NY-ESO-1 , PRAME, SSX-2; lineage restricted antigens: MARTI , Gp100, tyrosinase, TRP-1/2, MC1 R, prostate specific antigen; mutated antigens: P-catenin, BRCA1/2, CDK4, CML66, Fibronectin, MART-2, p53, Ras, TGF-0RII; post-translationally altered antigens: MUC1 , idiotypic antigens: Ig, TCR. Other cancer-associated antigens include heatshock protein 70 (HSP70), heat-shock protein 90 (HSP90), glucose-regulated protein 78 (GRP78), vimentin, nucleolin, feto-acinar pancreatic protein (FAPP), alkaline phosphatase placental-like 2 (ALPPL-2), siglec-5, stress-induced phosphoprotein 1 (STIP1), protein tyrosine kinase 7 (PTK7), and cyclophilin B. In some embodiments the cancer-associated antigen is a cancer-associated antigen described in Zhao and Cao, Front Immunol. 2019; 10: 2250, which is hereby incorporated by reference in its entirety. In some embodiments, a cancer-associated antigen is selected from CD30, CD19, CD20, CD22, B7H3, c-Met, ROR1 R, CD4, CD7, CD38, BCMA, Mesothelin, EGFR, GPC3, MUC1 , HER2, GD2, CEA, EpCAM, LeY and PSCA. In some embodiments, a cancer-associated antigen is an antigen expressed by cells of a hematological malignancy. In some embodiments, a cancer-associated antigen is selected from CD30, CD19, CD20, CD22, B7H3, c-Met, ROR1 R, CD4, CD7, CD38 and BCMA. In some embodiments, a cancer-associated antigen is an antigen expressed by cells of a solid tumor. In some embodiments, a cancer-associated antigen is selected from Mesothelin, EGFR, GPC3, MUC1 , HER2, GD2, CEA, EpCAM, LeY and PSCA.

An immune cell surface molecule may be any peptide/polypeptide, glycoprotein, lipoprotein, glycan, glycolipid, lipid, or fragment thereof expressed at or on the cell surface of an immune cell. In some embodiments, the part of the immune cell surface molecule which is bound by the antigen-binding molecule of the present disclosure is on the external surface of the immune cell (/.e. is extracellular). The immune cell surface molecule may be expressed at the cell surface of any immune cell. In some embodiments, the immune cell may be a cell of hematopoietic origin, e.g. a neutrophil, eosinophil, basophil, dendritic cell, lymphocyte, or monocyte. The lymphocyte may be e.g. a T cell, B cell, natural killer (NK) cell, NKT cell or innate lymphoid cell (ILC), or a precursor thereof (e.g. a thymocyte or pre-B cell). In some embodiments the antigen is a CD3 polypeptide (e.g. CD3e, CD36, CD3y or CD3Q.

In some embodiments, multispecific antigen-binding molecules described herein display at least monovalent binding with respect to B7-H3, and also display at least monovalent binding with respect to an antigen other than B7-H3. Binding valency refers to the number of binding sites in an antigenbinding molecule for a given antigenic determinant.

In some embodiments the antigen-binding molecule comprises a single-domain antibody capable of binding to B7-H3 (e.g. as described herein), and an antigen-binding region (e.g. a polypeptide (e.g. a single-domain antibody), Fv, Fab or antibody) capable of binding to an antigen other than B7-H3.

In some embodiments, an antigen-binding molecule comprises an immune cell-engaging moiety. In some embodiments, the antigen-binding molecule is an immune cell engager. Immune cell engagers are reviewed e.g. in Goebeler and Bargou, Nat. Rev. Clin. Oncol. (2020) 17: 418-434 and Ellerman, Methods (2019) 154:102-117, both of which are hereby incorporated by reference in their entirety. Immune cell engager molecules comprise an antigen-binding region for a target antigen of interest, and an antigen-binding region for recruiting/engaging an immune cell of interest. Immune cell engagers recruit/engage immune cells through an antigen-binding region specific for an immune cell surface molecule.

In some embodiments, the antigen-binding molecule comprises a CD3 polypeptide-binding moiety (e.g. an antigen-binding domain capable of binding to a CD3 polypeptide). The best studied immune cells engagers are bispecific T cell engagers (BiTEs), which comprise a target antigen binding domain, and a CD3 polypeptide (typically CD3e)-binding domain, through which the BiTE recruits T cells. Binding of the BiTE to its target antigen and to the CD3 polypeptide expressed by the T cell results in activation of the T cell, and ultimately directs T cell effector activity against cells expressing the target antigen. Other kinds of immune cell engagers are well known in the art, and include natural killer cell engagers such as bispecific killer engagers (BiKEs), which recruit and activate NK cells.

In some embodiments, the immune cell engaged by the immune cell engager is a T cell or an NK cell. In some embodiments, the immune cell engager is a T cell-engager. Multispecific antigenbinding molecules according to the present disclosure may be provided in any suitable format, such as those formats described in described in Brinkmann and Kontermann, MAbs (2017) 9(2): 182-212, which is hereby incorporated by reference in its entirety.

Functional properties of the antigen-binding molecules

The antigen-binding molecules described herein may be characterised by reference to certain functional properties. In some embodiments, the antigen-binding molecule described herein may possess one or more of the following properties: binds to B7-H3 (e.g. human B7-H3 (e.g. human B7-H3 isoform 1 and/or human B7-H3 isoform 2) and/or mouse B7-H3); binds to B7-H3-expressing cells; does not bind to cells that do not express B7-H3; increases killing of cells expressing B7-H3; does not increase killing of cells that do not express B7-H3; increases ADCC of cells expressing B7-H3; does not increase ADCC of cells that do not express B7-H3; inhibits tumor growth, e.g. of B7-H3-expressing cancer; and/or increases survival of subjects having a cancer, e.g. a B7-H3-expressing cancer.

It will be appreciated that a given antigen-binding molecule may display more than one of the properties recited in the preceding paragraph. A given antigen-binding molecule may be evaluated for the properties recited in the preceding paragraph using suitable assays. The assays may be e.g. in vitro assays, which may be cell-free or cell-based assays. Alternatively, the assays may be e.g. in vivo assays, i.e. performed in non-human animals. Assays may employ species labelled with detectable entities in order to facilitate their detection.

Where assays are cell-based assays, they may comprise treating cells with a given antigen-binding molecule in order to determine whether the antigen-binding molecule displays one or more of the recited properties. Assays may employ species labelled with detectable entities in order to facilitate their detection. Assays may comprise evaluating the recited properties following treatment of cells separately with a range of quantities/concentrations of a given antigen-binding molecule (e.g. a dilution series). It will be appreciated that the cells preferably express the target antigen for the antigen-binding molecule (i.e. B7-H3).

Analysis of the results of such assays may comprise determining the concentration at which 50% of the maximal level of the relevant activity is attained. The concentration of antigen-binding molecule at which 50% of the maximal level of the relevant activity is attained may be referred to as the ‘half- maximal effective concentration’ of the antigen-binding molecule in relation to the relevant activity, which may also be referred to as the ‘EC50’. By way of illustration, the EC50 of a given antigenbinding molecule for binding to B7-H3 may be the concentration at which 50% of the maximal level of binding is achieved.

Depending on the property, the EC50 may also be referred to as the ‘half-maximal inhibitory concentration’ or ‘IC50’, this being the concentration of antigen-binding molecule at which 50% of the maximal level of inhibition of a given property is observed.

The antigen-binding molecules and antigen-binding domains described herein preferably display specific binding to B7-H3. As used herein, ‘specific binding’ refers to binding which is selective for the antigen, and which can be discriminated from non-specific binding to non-target antigen. An antigen-binding molecule/domain that specifically binds to a target molecule preferably binds the target with greater affinity, and/or with greater duration than it binds to other, non-target molecules.

The ability of a given polypeptide to bind specifically to a given molecule can be determined by analysis according to methods known in the art, such as by ELISA, Surface Plasmon Resonance (SPR; see e.g. Hearty et al., Methods Mol Biol (2012) 907:411-442), Bio-Layer Interferometry (see e.g. Lad et al., (2015) J Biomol Screen 20(4): 498-507), flow cytometry, or by a radiolabelled antigen-binding assay (RIA) enzyme-linked immunosorbent assay. Through such analysis binding to a given molecule can be measured and quantified. In some embodiments, the binding may be the response detected in a given assay.

In some embodiments, the extent of binding of the antigen-binding molecule to a non-target molecule is less than about 10% of the binding of the antibody to the target molecule as measured, e.g. by ELISA, SPR, Bio-Layer Interferometry or by RIA. Alternatively, binding specificity may be reflected in terms of binding affinity where the antigen-binding molecule binds with an equilibrium dissociation constant (KD) that is at least 0.1 order of magnitude (/.e. 0.1 x 10 ⁿ , where n is an integer representing the order of magnitude) greater than the KD of the antigen-binding molecule towards a non-target molecule. This may optionally be one of at least 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.5, or 2.0.

Binding to B7-H3 may be determined by Surface Plasmon Resonance, e.g. as described in Example 1 .5 of the present disclosure.

In some embodiments, the antigen-binding molecule described herein binds to B7-H3 with submicromolar affinity, i.e. KD < 1 x 10 ⁶ M. In some embodiments, the antigen-binding molecule described herein binds to B7-H3 with an affinity in the nanomolar range, i.e. KD = 9.9 x 10 ⁷ to 1 x 10’ ⁹ M. In some embodiments, the antigen-binding molecule described herein binds to B7-H3 with sub- nanomolar affinity, i.e. KD < 1 x 10 ⁹ M. In some embodiments, the antigen-binding molecule described herein binds to B7-H3 with an affinity in the picomolar range, i.e. KD = 9.9 x 10 ^w to 1 x 10- ¹² M. In some embodiments, the antigen-binding molecule described herein binds to B7-H3 with sub- picomolar affinity, i.e. KD < 1 x 10 ¹² M.

In some embodiments, the antigen-binding molecule described herein binds to B7-H3 (e.g. human B7-H3) with a KD of 5 pM or less, preferably one of <5 pM, <2 pM, <1 pM, <500 nM, <450 nM, <400 nM, <350 nM, <300 nM, <250 nM, <200 nM, <150 nM, <100 nM, <75 nM, <50 nM, <40 nM, <30 nM, <20 nM or <10 nM.

In some embodiments, the antigen-binding molecule described herein binds to human B7-H3 isoform 1 with a KD of 5 pM or less, preferably one of <5 pM, <2 pM, <1 pM, <500 nM, <450 nM, <400 nM, <350 nM, <300 nM, <250 nM, <200 nM, <150 nM, <100 nM, <75 nM, <50 nM, <40 nM or <30 nM. In some embodiments, the antigen-binding molecule binds to human B7-H3 isoform 1 with one of <500 nM and > 5 nM, <200 nM and > 10 nM, <150 nM and > 15 r <50 nM and > 25 nM.

In some embodiments, the antigen-binding molecule described herein binds to human B7-H3 isoform 2 with a KD of 5 pM or less, preferably one of <5 pM, <2 pM, <1 pM, <500 nM, <450 nM, <400 nM, <350 nM. In some embodiments, the antigen-binding molecule binds to human B7-H3 isoform 2 with a KD ^10 pM and > 10 nM, e.g. one of <5 pM and > 50 nM, <2 pM and > 100 nM, <1 pM and > 150 nM, <750 nM and > 200 nM or <500 nM and > 250 nM.

In some embodiments, the antigen-binding molecule described herein binds to mouse B7-H3 with a KD of 5 pM or less, preferably one of <5 pM, <2 pM, <1 pM, <500 nM. In some embodiments, the antigen-binding molecule binds to mouse B7-H3 with a KD ^10 pM and > 10 nM, e.g. one of <5 pM and > 50 nM, <2 pM and > 100 nM, <1 pM and > 150 nM, <750 nM and > 200 nM or <500 nM and > 250 nM.

The antigen-binding molecules of the present disclosure may bind to a particular region of interest of B7-H3. Antigen-binding molecules according to the present disclosure may bind to linear epitope of B7-H3, consisting of a contiguous sequence of amino acids (/.e. an amino acid primary sequence).

In some embodiments, an antigen-binding molecules may bind to a conformational epitope of B7- H3, consisting of a discontinuous sequence of amino acids of the amino acid sequence.

The region of a given target molecule to which an antigen-binding molecule binds can be determined by the skilled person using various methods well known in the art, including X-ray co-crystallography analysis of antibody-antigen complexes, peptide scanning, mutagenesis mapping, hydrogendeuterium exchange analysis by mass spectrometry, phage display, competition ELISA and proteolysis-based ‘protection’ methods. Such methods are described, for example, in Gershoni et al., BioDrugs, 2007, 21 (3):145-156, which is hereby incorporated by reference in its entirety. The ability of an antigen-binding molecule to bind to a given peptide/polypeptide can be analysed by methods well known to the skilled person, including analysis by ELISA, immunoblot (e.g. western blot), immunoprecipitation, surface plasmon resonance and biolayer interferometry.

In some embodiments, the antigen-binding molecules bind (independently) to human B7-H3 (e.g. human B7-H3 isoform 1) and mouse B7-H3. In some embodiments, the antigen-binding molecules bind (independently) to human B7-H3 isoform 1 and human B7-H3 isoform 2. In some embodiments, the antigen-binding molecules bind (independently) to human B7-H3 isoform 1 , human B7-H3 isoform 2, human B7-H3 isoform 3 and human B7-H3 isoform 4. In some embodiments, the antigenbinding molecules bind (independently) to human B7-H3 isoform 1 , human B7-H3 isoform 2 and mouse B7-H3. In some embodiments, the antigen-binding molecules bind (independently) to human B7-H3 isoform 1 , human B7-H3 isoform 2, human B7-H3 isoform 3, human B7-H3 isoform 4, and mouse B7-H3.

In some embodiments, the antigen-binding molecules are cross-reactive for one or more isoforms or homologues of B7-H3 (e.g. human B7-H3 isoform 1). In some embodiments, the antigen-binding molecules are cross-reactive for human B7-H3 (e.g. human B7-H3 isoform 1) and mouse B7-H3. In some embodiments, the antigen-binding molecules are cross-reactive for human B7-H3 isoform 1 and human B7-H3 isoform 2. In some embodiments, the antigen-binding molecules are cross- reactive for human B7-H3 isoform 1 , human B7-H3 isoform 2, human B7-H3 isoform 3 and human B7-H3 isoform 4. In some embodiments, the antigen-binding molecules are cross-reactive for human B7-H3 isoform 1 , human B7-H3 isoform 2 and mouse B7-H3. In some embodiments, the antigenbinding molecules are cross-reactive for human B7-H3 isoform 1 , human B7-H3 isoform 2, human B7-H3 isoform 3, human B7-H3 isoform 4 and mouse B7-H3. As used herein, a ‘cross-reactive’ antigen-binding molecule/domain/polypeptide binds to the target antigens for which the antigen-binding molecule/domain is cross-reactive. For example, an antigenbinding molecule/domain/polypeptide which is cross-reactive for human B7-H3 and mouse B7-H3 binds to human B7-H3, and is also capable of binding to mouse B7-H3. Cross-reactive antigenbinding molecules/domains/polypeptides may display specific binding to each of the target antigens.

In some embodiments, the antigen-binding molecule of the present disclosure binds to the extracellular domain of B7-H3. In some embodiments, the antigen-binding molecule binds to the region of B7-H3 shown in SEQ ID NO:16. In some embodiments, the antigen-binding molecule binds to the region of B7-H3 shown in SEQ ID NO:20. In some embodiments, the antigen-binding molecule binds to the region of B7-H3 shown in SEQ ID NQ:30. In some embodiments, the antigenbinding molecule binds to the region of B7-H3 shown in SEQ ID NO:34.

In some embodiments, the antigen-binding molecule binds to a polypeptide comprising or consisting of the amino acid sequence shown in SEQ ID NO:16. In some embodiments, the antigen-binding molecule binds to a polypeptide comprising or consisting of the amino acid sequence shown in SEQ ID NQ:20. In some embodiments, the antigen-binding molecule binds to a polypeptide comprising or consisting of the amino acid sequence shown in SEQ ID NQ:30. In some embodiments, the antigenbinding molecule binds to a polypeptide comprising or consisting of the amino acid sequence shown in SEQ ID NO:34.

In some embodiments, the antigen-binding molecule of the present disclosure binds to an Ig-like C2- type domain of B7-H3. In some embodiments, the antigen-binding molecule contacts an Ig-like C2- type domain of B7-H3. In some embodiments, the antigen-binding molecule binds to B7-H3 via contact with one or more amino acids of an Ig-like C2-type domain of B7-H3.

In some embodiments, the antigen-binding molecule binds to the region of B7-H3 shown in SEQ ID NO:25. In some embodiments, the antigen-binding molecule contacts the region of B7-H3 shown in SEQ ID NO:25. In some embodiments, the antigen-binding molecule binds to B7-H3 via contact with one or more amino acids of the region shown in SEQ ID NO:25. In some embodiments, the epitope of the antigen-binding molecule comprises or consists of the amino acid sequence shown in SEQ ID NO:25. In some embodiments, the antigen-binding molecule binds to a polypeptide comprising or consisting of the amino acid sequence shown in SEQ ID NO:25.

In some embodiments, the antigen-binding molecule binds to the region of B7-H3 shown in SEQ ID NO:27. In some embodiments, the antigen-binding molecule contacts the region of B7-H3 shown in SEQ ID NO:27. In some embodiments, the antigen-binding molecule binds to B7-H3 via contact with one or more amino acids of the region shown in SEQ ID NO:27. In some embodiments, the epitope of the antigen-binding molecule comprises or consists of the amino acid sequence shown in SEQ ID NO:27. In some embodiments, the antigen-binding molecule binds to a polypeptide comprising or consisting of the amino acid sequence shown in SEQ ID NO:27.

In some embodiments, the antigen-binding molecule binds to the region of B7-H3 shown in SEQ ID NO:38. In some embodiments, the antigen-binding molecule contacts the region of B7-H3 shown in SEQ ID NO:38. In some embodiments, the antigen-binding molecule binds to B7-H3 via contact with one or more amino acids of the region shown in SEQ ID NO:38. In some embodiments, the epitope of the antigen-binding molecule comprises or consists of the amino acid sequence shown in SEQ ID NO:38. In some embodiments, the antigen-binding molecule binds to a polypeptide comprising or consisting of the amino acid sequence shown in SEQ ID NO:38.

In some embodiments, an antigen-binding molecule according to the present disclosure binds to the same region of B7-H3, or an overlapping region of B7-H3, to the region of B7-H3 which is bound by an antigen-binding molecule comprising the CDRs, FRs and/or the complete amino acid sequence of a B7-H3-binding single-domain antibody described herein, e.g. P2A5.

Whether a test antigen-binding molecule binds to the same or an overlapping region of a given target as a reference antigen-binding molecule can be evaluated, for example, by analysis of (i) interaction between the test antigen-binding molecule and the target in the absence of the reference binding molecule, and (ii) interaction between the test antigen-binding molecule in the presence of the reference antigen-binding molecule, or following incubation of the target with the reference antigen-binding molecule. Determination of a reduced level of interaction between the test antigenbinding molecule and the target following analysis according to (ii) as compared to (i) might support an inference that the test and reference antigen-binding molecule bind to the same or an overlapping region of the target. Suitable assays for such analysis include e.g. competition ELISA assays and epitope binning assays.

In some embodiments, an antigen-binding molecule according to the present disclosure binds to the same region of B7-H3, or an overlapping region of B7-H3, to the region of B7-H3 which is bound by a polypeptide consisting of the amino acid sequence of SEQ ID NO:8.

In some embodiments, an antigen-binding molecule according to the present disclosure may potentiate (/.e. upregulate, enhance) cell killing of cells comprising/expressing B7-H3. In some embodiments, the antigen-binding molecule does not potentiate (/.e. does not substantially potentiate) cell killing of cells lacking surface expression of B7-H3.

In some embodiments, an antigen-binding molecule according to the present disclosure may inhibit growth or reduce metastasis of a cancer comprising cells comprising/expressing B7-H3. In some embodiments, the antigen-binding molecule may potentiate (/.e. upregulate, enhance) cell killing of cells of a cancer comprising/expressing B7-H3. In some embodiments, the antigen-binding molecule may inhibit growth or reduce metastasis of a cancer comprising cells comprising/expressing B7-H3.

Cell killing can be investigated, for example, using any of the methods reviewed in Zaritskaya et al., Expert Rev Vaccines (2011), 9(6):601-616, hereby incorporated by reference in its entirety.

Examples of in vitro assays of cytotoxicity/cell killing assays include release assays such as the ⁵¹ Cr release assay, the lactate dehydrogenase (LDH) release assay, the 3-(4,5-dimethylthiazol-2-yl)-2,5- diphenyl tetrazolium bromide (MTT) release assay, and the calcein-acetoxymethyl (calcein-AM) release assay. These assays measure cell killing based on the detection of factors released from lysed cells. Cell killing of a given test cell type by a given effector immune cell type can be analysed e.g. by co-culturing the test cells with the effector immune cells, and measuring the number/proportion of viable/dead (e.g. lysed) test cells after a suitable period of time. Other suitable assays include the xCELLigence real-time cytolytic in vitro potency assay described in Cerignoli et al., PLoS One. (2018) 13(3): e0193498 (hereby incorporated by reference in its entirety). An increase in resistance to cell killing by granzyme B-expressing cells (e.g. effector immune cells), and/or a reduction in susceptibility to cell killing by such cells, relative to a reference level of cell killing (e.g. forthat cell type) can be determined by detection of a reduction in the number/proportion of dead (e.g. lysed) test cells, and/or an increase in the number/proportion of live (e.g. viable, nonlysed) test cells, after a given period of time.

In some embodiments an antigen-binding molecule according to the present disclosure is capable of reducing the number/proportion of cells expressing B7-H3. In some embodiments an antigen-binding molecule according to the present disclosure is capable of reducing the number/proportion of cells expressing B7-H3. In some embodiments, an antigen-binding molecule according to the present disclosure is capable of depleting/enhancing depletion of such cells.

Antigen-binding molecules according to the present disclosure may comprise one or more moieties for potentiating a reduction in the number/proportion of cells expressing B7-H3. For example, an antigen-binding molecule according to the present disclosure may e.g. comprise an Fc region and/or a drug moiety.

Fc regions provide for interaction with Fc receptors and other molecules of the immune system to bring about functional effects. IgG Fc-mediated effector functions are reviewed e.g. in Jefferis et al., Immunol Rev 1998 163:59-76 (hereby incorporated by reference in its entirety), and are brought about through Fc-mediated recruitment and activation of immune cells (e.g. macrophages, dendritic cells, neutrophils, basophils, eosinophils, platelets, mast cells, NK cells and T cells) through interaction between the Fc region and Fc receptors expressed by the immune cells, recruitment of complement pathway components through binding of the Fc region to complement protein C1q, and consequent activation of the complement cascade. Fc-mediated functions include Fc receptor binding, antibody-dependent cellular cytotoxicity (ADCC), antibody-dependent cell-mediated phagocytosis (ADCP), complement-dependent cytotoxicity (CDC), formation of the membrane attack complex (MAC), cell degranulation, cytokine and/or chemokine production, and antigen processing and presentation.

In some embodiments, an antigen-binding molecule according to the present disclosure comprises an Fc region capable of potentiating/directing one or more of ADCC, ADCP, CDC against, and/or potentiating formation of a MAC on or cell degranulation of, a cell expressing B7-H3 (e.g. a cell expressing B7-H3 at the cell surface).

In some embodiments, an antigen-binding molecule according to the present disclosure is capable of potentiating/directing ADCC against a cell expressing B7-H3.

The ability of, and extent to which, a given antigen-binding molecule is able to induce ADCC of a given target cell type can be analysed e.g. according to the method described in Yamashita et al., Scientific Reports (2016) 6:19772 (hereby incorporated by reference in its entirety), or by ⁵¹ Cr release assay as described e.g. in Jedema et al., Blood (2004) 103: 2677-82 (hereby incorporated by reference in its entirety). The ability of, and extent to which, a given antigen-binding molecule is able to induce ADCP can be analysed e.g. according to the method described in Kamen et al., J Immunol (2017) 198 (1 Supplement) 157.17 (hereby incorporated by reference in its entirety). The ability of, and extent to which, a given antigen-binding molecule is able to induce CDC can be analysed e.g. using a C1q binding assay, e.g. as described in Schlothauer ef al., Protein Engineering, Design and Selection (2016), 29(10):457-466 (hereby incorporated by reference in its entirety).

In some embodiments, an antigen-binding molecule according to the present disclosure comprises a drug moiety. The antigen-binding molecule may be conjugated to the drug moiety. Antibody-drug conjugates are reviewed e.g. in Parslow et al., Biomedicines. 2016 Sep; 4(3):14 (hereby incorporated by reference in its entirety). In some embodiments, the drug moiety is or comprises a cytotoxic agent, such that the antigen-binding molecule displays cytotoxicity to a cell expressing B7- H3 (e.g. a cell expressing B7-H3 at the cell surface). In some embodiments, the drug moiety is or comprises a chemotherapeutic agent.

In some embodiments, an antigen-binding molecule according to the present disclosure comprises an immune cell-engaging moiety. In some embodiments, the antigen-binding molecule comprises a CD3 polypeptide-binding moiety (e.g. an antigen-binding domain capable of binding to a CD3 polypeptide).

In some embodiments, an antigen-binding molecule according to the present disclosure is capable of potentiating/directing T cell-mediated cytolytic activity against a cell expressing B7-H3. In some embodiments, the antigen-binding molecule of the present disclosure displays anticancer activity. In some embodiments, the antigen-binding molecule of the present disclosure increases killing of cancer cells. In some embodiments, the antigen-binding molecule of the present disclosure causes a reduction in the number of cancer cells in vivo, e.g. as compared to an appropriate control condition. The cancer may be a cancer expressing B7-H3.

In some embodiments, an antigen-binding molecule according to the present disclosure reduces/inhibits growth of a cancer and/or of a tumor of a cancer. In some embodiments, an antigenbinding molecule reduces tissue invasion by cells of a cancer. In some embodiments, an antigenbinding molecule reduces metastasis of a cancer. In some embodiments, the antigen-binding molecule displays anticancer activity. In some embodiments, the antigen-binding molecule reduces the growth/proliferation of cancer cells. In some embodiments, the antigen-binding molecule reduces the survival of cancer cells. In some embodiments, the antigen-binding molecule increases the killing of cancer cells. In some embodiments, the antigen-binding molecule of the present disclosure causes a reduction in the number of cancer cells e.g. in vivo. The cancer may be a cancer comprising cells expressing B7-H3.

The antigen-binding molecule of the present disclosure may be analysed for the properties described in the preceding paragraph in appropriate assays. Such assays include e.g. in vivo models.

In some embodiments, administration of an antigen-binding molecule according to the present disclosure may cause one or more of: inhibition of the development/progression of the cancer, a delay to/prevention of onset of the cancer, a reduction in/delay to/prevention of tumor growth, a reduction in/delay to/prevention of tissue invasion, a reduction in/delay to/prevention of metastasis, a reduction in the severity of the symptoms of the cancer, a reduction in the number of cancer cells, a reduction in tumour size/volume, and/or an increase in survival (e.g. progression free survival or overall survival), e.g. as determined in an appropriate model.

In some embodiments, the antigen-binding molecule of the present disclosure is capable of reducing/inhibiting tumor growth (e.g. in an in vivo model, e.g. of a B7-H3-expressing cancer) to less than 1 times, e.g. <0.99 times, <0.95 times, <0.9 times, <0.85 times, <0.8 times, <0.75 times, <0.7 times, <0.65 times, <0.6 times, <0.55 times, <0.5 times, <0.45 times, <0.4 times, <0.35 times, <0.3 times, <0.25 times, <0.2 times, <0.15 times, <0.1 times, <0.05 times, or <0.01 times the tumor growth observed in the absence of treatment with the antigen-binding molecule (or following treatment with an appropriate control antigen-binding molecule known not to influence tumor growth), in a given assay.

In some embodiments, the antigen-binding molecule of the present disclosure is capable of reducing/inhibiting metastasis (e.g. in an in vivo model, e.g. of a B7-H3-expressing cancer) to less than 1 times, e.g. <0.99 times, <0.95 times, <0.9 times, <0.85 times, <0.8 times, <0.75 times, <0.7 times, <0.65 times, <0.6 times, <0.55 times, <0.5 times, <0.45 times, <0.4 times, <0.35 times, <0.3 times, <0.25 times, <0.2 times, <0.15 times, <0.1 times, <0.05 times, or <0.01 times the level of metastasis observed in the absence of treatment with the antigen-binding molecule (or following treatment with an appropriate control antigen-binding molecule known not to influence metastasis), in a given assay.

In some embodiments, the antigen-binding molecule of the present disclosure is capable of increasing survival of subjects having a cancer (e.g. in an in vivo model, e.g. of a B7-H3-expressing cancer) to more than 1 times, e.g. one of >1 .01 times, >1 .02 times, >1.03 times, >1 .04 times, >1 .05 times, >1.1 times, >1 .2 times, >1.3 times, >1 .4 times, >1 .5 times, >1 .6 times, >1 .7 times, >1 .8 times, >1 .9 times, >2 times, >3 times, >4 times, >5 times, >6 times, >7 times, >8 times, >9 times or >10 times the level of survival observed in the absence of treatment with the antigen-binding molecule (or following treatment with an appropriate control antigen-binding molecule known not to influence survival), in a given assay.

Chimeric antigen receptors (CARs)

In some aspects and embodiments in accordance with the present disclosure, the antigen-binding molecule is a chimeric antigen receptor (CAR). In some aspects and embodiments, the present disclosure provides a chimeric antigen receptor comprising an antigen-binding molecule or polypeptide according to the present disclosure.

CARs are recombinant receptors that provide both antigen-binding and T cell activating functions. CAR structure and engineering is reviewed, for example, in Dotti et al., Immunol Rev (2014) 257(1), hereby incorporated by reference in its entirety. CARs comprise an antigen-binding domain linked via a transmembrane domain to a signalling domain. An optional hinge or spacer domain may provide separation between the antigen-binding domain and transmembrane domain and may act as a flexible linker. When expressed by a cell, the antigen-binding domain is provided in the extracellular space, and the signalling domain is intracellular.

The antigen-binding domain mediates binding to the target antigen for which the CAR is specific. The antigen-binding domain of a CAR may be based on the antigen-binding region of an antigenbinding molecule which is specific for the antigen to which the CAR is targeted. For example, the antigen-binding domain of a CAR may comprise amino acid sequences for the complementaritydetermining regions (CDRs) of an antibody which binds specifically to the target antigen. The antigen-binding domain of a CAR may comprise or consist of the light chain and heavy chain variable region amino acid sequences of an antibody which binds specifically to the target antigen. The antigen-binding domain may be provided as a single chain variable fragment (scFv) comprising the sequences of the light chain and heavy chain variable region amino acid sequences of an antibody. Antigen-binding domains of CARs may target antigens based on other protein:protein interactions, such as ligand :receptor binding; for example an IL-13Ra2-targeted CAR has been developed using an antigen-binding domain based on IL-13 (see e.g. Kahlon et al. 2004 Cancer Res 64(24): 9160-9166).

The CAR of the present disclosure comprises an antigen-binding domain which comprises or consists of the antigen-binding molecule of the present disclosure, or which comprises or consists of a polypeptide according to the present disclosure.

An optional spacer domain may provide separation between the antigen-binding domain and the transmembrane domain, and may act as a flexible linker. Such domains may be or comprise flexible regions allowing the binding moiety to orient in different directions. Spacer domains may be derived from IgG.

The transmembrane domain is provided between the antigen-binding domain and the signalling domain of the CAR. The transmembrane domain provides for anchoring the CAR to the cell membrane of a cell expressing a CAR, with the antigen-binding domain in the extracellular space and signalling domain inside the cell. Transmembrane domains of CARs may be derived from transmembrane region sequences for cell membrane-bound proteins (e.g. CD28, CD8, CD4, CD3- etc.).

The signalling domain comprises amino acid sequences required for activation of immune cell function. The CAR signalling domains may comprise the amino acid sequence of the intracellular domain of CD3- , which provides immunoreceptor tyrosine-based activation motifs (ITAMs) for phosphorylation and activation of the CAR-expressing cell. Signalling domains comprising sequences of other ITAM-containing proteins have also been employed in CARs, such as domains comprising the ITAM containing region of FcyRI (Haynes et al., 2001 J Immunol 166(1):182-187). CARs comprising a signalling domain derived from the intracellular domain of CD3- are often referred to as first generation CARs.

The signalling domains of CARs typically also comprise the signalling domain of a costimulatory protein (e.g. CD28, 4-1 BB etc.), for providing the costimulation signal necessary for enhancing immune cell activation and effector function. CARs having a signalling domain including additional costimulatory sequences are often referred to as second generation CARs. In some cases, CARs are engineered to provide for costimulation of different intracellular signalling pathways. For example, CD28 costimulation preferentially activates the phosphatidylinositol 3-kinase (P13K) pathway, whereas 4-1 BB costimulation triggers signalling is through TNF receptor associated factor (TRAF) adaptor proteins. Signalling domains of CARs therefore sometimes contain costimulatory sequences derived from signalling domains of more than one costimulatory molecule. CARs comprising a signalling domain with multiple costimulatory sequences are often referred to as third generation CARs. Throughout this specification, polypeptides, domains and amino acid sequences which are ‘derived from’ a reference polypeptide/domain/amino acid sequence have at least 60%, preferably one of 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% amino acid sequence identity to the amino acid sequence of the reference polypeptide/domain/amino acid sequence. Polypeptides, domains and amino acid sequences which are ‘derived from’ a reference polypeptide/domain/amino acid sequence preferably retain the functional and/or structural properties of the reference polypeptide/domain/amino acid sequence.

By way of illustration, an amino acid sequence derived from the intracellular domain of CD28 may comprise an amino acid sequence having 60%, preferably one of 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% amino acid sequence identity to the intracellular domain of CD28, e.g. as shown in SEQ ID NO:12. Furthermore, an amino acid sequence derived from the intracellular domain of CD28 preferably retains the functional properties of the amino acid sequence of SEQ ID NO:12, i.e. the ability to activate CD28-mediated signalling.

The amino acid sequence of a given polypeptide or domain thereof can be retrieved from, or determined from a nucleic acid sequence retrieved from, databases known to the person skilled in the art. Such databases include GenBank, EMBL and UniProt.

Through engineering to express a CAR specific for a particular target antigen, immune cells (typically T cells, but also other immune cells such as NK cells) can be directed to kill cells expressing the target antigen. Binding of a CAR-expressing T cell (CAR-T cell) to the target antigen for which it is specific triggers intracellular signalling, and consequently activation of the T cell. The activated CAR-T cell is stimulated to divide and produce factors resulting in killing of the cell expressing the target antigen.

Antigen-binding domain

The antigen-binding domain of a CAR according to the present disclosure comprises or consists of an antigen-binding molecule that binds to B7-H3 as described herein. Accordingly, a CAR according to the present disclosure comprises an antigen-binding molecule as described herein.

It will be appreciated that an antigen-binding molecule according to the present disclosure forms, or is comprised in, the antigen-binding domain of the CAR. Accordingly, in some embodiments, the antigen-binding molecule of the present disclosure is comprised in a CAR.

It will also be appreciated that an antigen-binding molecule according to the present disclosure may be a CAR. A CAR having an antigen-binding domain comprising or consisting of an antigen-binding molecule of the present disclosure (e.g. a B7-H3-binding single-domain antibody) is an antigenbinding molecule. The antigen-binding domain of the CAR of the present disclosure may be provided with any suitable format, e.g. scFv, scFab, etc. In some embodiments, the antigen-binding domain comprises, or consists of, a B7-H3-binding single-domain antibody as described herein.

Spacer domain

In some embodiments, the CAR comprises a spacer domain. The spacer domain may be provided between the antigen-binding domain and the transmembrane domain. The spacer domain may also be referred to as a hinge domain. A spacer domain is an amino acid sequence which provides for flexible linkage of the antigen-binding and transmembrane domains of the CAR.

The presence, absence and length of spacer domains has been shown to influence CAR function (reviewed e.g. in Dotti et al., Immunol Rev (2014) 257(1) and Jayaraman et al., EBioMedicine (2020) 58:102931 , supra). Spacer length can be varied to control synaptic cleft distances, which might in turn regulate signalling. Flexible spacers can enable access to sterically hindered epitopes on the target antigen. Multimerisation of spacer domains (e.g. through homotypic associations) results in increased signal strength and activation stimulus.

In some embodiments, a spacer domain according to the present disclosure comprises, or consists of, an amino acid sequence which is, or which is derived from: the CH2-CH3 region of human lgG1 (e.g. as shown in SEQ ID NO:59), the CH2-CH3 region of human lgG2 (e.g. as shown in SEQ ID NO:61), the CH1-CH2 hinge region of human lgG1 , a spacer domain derived from CD8a, e.g. as described in WO 2012/031744 A1 , or a spacer domain derived from CD28, e.g. as described in WO 2011/041093 A1 . Hornbach et al., Gene Therapy (2010) 17:1206-1213 describes a variant CH2-CH3 region for reduced activation of FcyR-expressing cells such as monocytes and NK cells. The amino acid sequence of the variant CH2-CH3 region is shown in SEQ ID NO:60.

In some embodiments, the spacer domain comprises, or consists of, an amino acid sequence having at least 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:59. In some embodiments, the spacer domain comprises, or consists of, an amino acid sequence having at least 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NQ:60. In some embodiments, the spacer domain comprises, or consists of, an amino acid sequence having at least 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:61 . In some embodiments, the spacer domain comprises, or consists of, an amino acid sequence having at least 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:62. In some embodiments, the spacer domain comprises, or consists of, an amino acid sequence having at least 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:63. In some embodiments, the spacer domain comprises, or consists of, an amino acid sequence having at least 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:64. In some embodiments, the spacer domain comprises, or consists of, an amino acid sequence having at least 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:10.

Transmembrane domain

The CAR of the present disclosure comprises a transmembrane domain. A transmembrane domain refers to any three-dimensional structure formed by a sequence of amino acids which is thermodynamically stable in a biological membrane, e.g. a cell membrane. In connection with the present disclosure, the transmembrane domain may be an amino acid sequence which spans the cell membrane of a cell expressing the CAR.

The transmembrane domain may comprise or consist of a sequence of amino acids which forms a hydrophobic alpha helix or beta-barrel. The amino acid sequence of the transmembrane domain of the CAR of the present disclosure may be, or may be derived from, the amino acid sequence of a transmembrane domain of a protein comprising a transmembrane domain. Transmembrane domains are recorded in databases such as GenBank, UniProt, Swiss-Prot, TrEMBL, Protein Information Resource, Protein Data Bank, Ensembl, and InterPro, and/or can be identified/predicted e.g. using amino acid sequence analysis tools such as TMHMM (Krogh et al., 2001 J Mol Biol 305: 567-580).

In some embodiments, the amino acid sequence of the transmembrane domain of the CAR of the present disclosure may be, or may be derived from, the amino acid sequence of the transmembrane domain of a protein expressed at the cell surface. In some embodiments the protein expressed at the cell surface is a receptor or ligand, e.g. an immune receptor or ligand. In some embodiments the amino acid sequence of the transmembrane domain may be, or may be derived from, the amino acid sequence of the transmembrane domain of one of ICOS, ICOSL, CD86, CTLA-4, CD28, CD80, MHC class I a, MHC class II a, MHC class II p, CD3e, CD36, CD3y, CD3- , TCRa TCRp, CD4, CD8a, CD8p, CD40, CD40L, PD-1 , PD-L1 , PD-L2, 4-1 BB, 4-1 BBL, 0X40, OX40L, GITR, GITRL, TIM-3, Galectin 9, LAG3, CD27, CD70, LIGHT, HVEM, TIM-4, TIM-1 , ICAM1 , LFA-1 , LFA-3, CD2, BTLA, CD160, LILRB4, LILRB2, VTCN1 , CD2, CD48, 2B4, SLAM, CD30, CD30L, DR3, TL1A, CD226, CD155, CD112 and CD276. In some embodiments, the transmembrane is, or is derived from, the amino acid sequence of the transmembrane domain of CD28, CD3- , CD8a, CD80 or CD4. In some embodiments, the transmembrane is, or is derived from, the amino acid sequence of the transmembrane domain of CD28.

In some embodiments, the transmembrane domain comprises, or consists of, an amino acid sequence having at least 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:11 . In some embodiments, the transmembrane domain comprises, or consists of, an amino acid sequence having at least 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:65.

In some embodiments, the transmembrane domain comprises, or consists of, an amino acid sequence having at least 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:66.

Signalling domain

The chimeric antigen receptor of the present disclosure comprises a signalling domain. The signalling domain provides sequences for initiating intracellular signalling in cells expressing the CAR.

ITAM-containing sequence:

The signalling domain comprises ITAM-containing sequence. An ITAM-containing sequence comprises one or more immunoreceptor tyrosine-based activation motifs (ITAMs). ITAMs comprise the amino acid sequence YXXL/I (SEQ ID NO:69), wherein ‘X’ denotes any amino acid. In ITAM- containing proteins, sequences according to SEQ ID NO:69 are often separated by 6 to 8 amino acids; YXXL/I (X)e-s YXXL/I (SEQ ID NQ:70). When phosphate groups are added to the tyrosine residue of an ITAM by tyrosine kinases, a signalling cascade is initiated within the cell.

In some embodiments, the signalling domain comprises one or more copies of an amino acid sequence according to SEQ ID NO:69 or SEQ ID NQ:70. In some embodiments, the signalling domain comprises at least 1 , 2, 3, 4, 5 or 6 copies of an amino acid sequence according to SEQ ID NO:69. In some embodiments, the signalling domain comprises at least 1 , 2, or 3 copies of an amino acid sequence according to SEQ ID NOTO.

In some embodiments, the signalling domain comprises an ITAM-containing sequence which is, or which is derived from, the amino acid sequence of an ITAM-containing sequence of a protein having an ITAM-containing amino acid sequence. In some embodiments the signalling domain comprises an ITAM-containing sequence which is, or which is derived from, the amino acid sequence of the intracellular domain of one of CD3 FcyRI, CD3e, CD36, CD3y, CD79a, CD790, FcyRIIA, FcyRIIC, FcyRIIIA, FcyRIV or DAP12. In some embodiments the signalling domain comprises an ITAM- containing sequence which is, or which is derived from, the amino acid sequence of the intracellular domain of CD3

In some embodiments, the signalling domain comprises an ITAM-containing sequence which comprises, or consists of, an amino acid sequence having at least 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:13. Costimulatory sequence:

The signalling domain may additionally comprise one or more costimulatory sequences. A costimulatory sequence is an amino acid sequence which provides for costimulation of the cell expressing the CAR of the present disclosure. Costimulation promotes proliferation and survival of a CAR-expressing cell upon binding to the target antigen, and may also promote cytokine production, differentiation, cytotoxic function and memory formation by the CAR-expressing cell. Molecular mechanisms of T cell costimulation are reviewed in Chen and Flies, (2013) Nat Rev Immunol 13(4):227-242.

A costimulatory sequence may be, or may be derived from, the amino acid sequence of a costimulatory protein. In some embodiments the costimulatory sequence is an amino acid sequence which is, or which is derived from, the amino acid sequence of the intracellular domain of a costimulatory protein.

Upon binding of the CAR to the target antigen, the costimulatory sequence provides costimulation to the cell expressing the CAR of the kind which would be provided by the costimulatory protein from which the costimulatory sequence is derived upon ligation by its cognate ligand. By way of example in the case of a CAR comprising a signalling domain comprising a costimulatory sequence derived from CD28, binding to the target antigen triggers signalling in the cell expressing the CAR of the kind that would be triggered by binding of CD80 and/or CD86 to CD28. Thus, a costimulatory sequence is capable of delivering the costimulation signal of the costimulatory protein from which the costimulatory sequence is derived.

In some embodiments, the costimulatory protein may be a member of the B7-CD28 superfamily (e.g. CD28, ICOS), or a member of the TNF receptor superfamily (e.g. 4-1 BB, 0X40, CD27, DR3, GITR, CD30, HVEM). In some embodiments, the costimulatory sequence is, or is derived from, the intracellular domain of one of CD28, 4-1 BB, ICOS, CD27, 0X40, HVEM, CD2, SLAM, TIM-1 , CD30, GITR, DR3, CD226 and LIGHT. In some embodiments, the costimulatory sequence is, or is derived from, the intracellular domain of CD28.

In some embodiments the signalling domain comprises more than one costimulatory sequence. In some embodiments the signalling domain comprises 1 , 2, 3, 4, 5 or 6 costimulatory sequences. Plural costimulatory sequences may be provided in tandem.

Whether a given amino acid sequence is capable of initiating signalling mediated by a given costimulatory protein can be investigated e.g. by analysing a correlate of signalling mediated by the costimulatory protein (e.g. expression/activity of a factor whose expression/activity is upregulated or downregulated as a consequence of signalling mediated by the costimulatory protein). Costimulatory proteins upregulate expression of genes promoting cell growth, effector function and survival through several transduction pathways. For example, CD28 and ICOS signal through phosphatidylinositol 3 kinase (PI3K) and AKT to upregulate expression of genes promoting cell growth, effector function and survival through NF-KB, mTOR, NFAT and AP1/2. CD28 also activates AP1/2 via CDC42/RAC1 and ERK1/2 via RAS, and ICOS activates C-MAF. 4-1 BB, 0X40, and CD27 recruit TNF receptor associated factor (TRAF) and signal through MAPK pathways, as well as through PI3K.

In some embodiments the signalling domain comprises a costimulatory sequence which is, or which is derived from CD28.

Kofler et al. Mol. Ther. (2011) 19: 760-767 describes a variant CD28 intracellular domain in which the lek kinase binding site is mutated in order to reduce induction of IL-2 production on CAR ligation, in order to minimise regulatory T cell-mediated suppression of CAR-T cell activity. The amino acid sequence of the variant CD28 intracellular domain is shown in SEQ ID NO:67.

In some embodiments, the signalling domain comprises, or consists of, an amino acid sequence having at least 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:12. In some embodiments, the signalling domain comprises, or consists of, an amino acid sequence having at least 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:67. In some embodiments, the signalling domain comprises, or consists of, an amino acid sequence having at least 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:68.

In some embodiments, the signalling domain comprises, or consists of, an amino acid sequence having at least 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:74.

In some embodiments, the CAR comprises an amino acid sequence having at least 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:75.

Linkers and additional sequences

The antigen-binding molecules, polypeptides and CARs of the present disclosure may additionally comprise further amino acids or sequences of amino acids.

Antigen-binding molecules may comprise additional amino acids/sequences of amino acids in addition to the amino acid sequence required for binding to the target antigen. In some embodiments, such additional amino acids/sequences of amino acids are provided at the N-terminus of a single-domain antibody sequence according to the present disclosure. In some embodiments, such additional amino acids/sequences of amino acids are provided at the C-terminus of a singledomain antibody sequence according to the present disclosure. In some embodiments, such additional amino acids/sequences of amino acids are provided at the N-terminus and C-terminus of a single-domain antibody sequence according to the present disclosure.

The antigen-binding molecules, polypeptides and CARs of the present disclosure may comprise one or more linker sequences between sequences of amino acids. For example, a linker sequence may be provided between domains of a CAR (e.g. between the antigen-binding domain and spacer domain, and/or between the spacer domain and the transmembrane domain, and/or between the transmembrane domain and the signalling domain). By way of further example, a linker sequence may be provided between subsequences of the domains of a CAR (e.g. between the costimulatory and sequence and ITAM-containing sequence of a signalling domain).

Linker sequences are known to the skilled person, and are described, for example in Chen et al., Adv Drug Deliv Rev (2013) 65(10): 1357-1369, which is hereby incorporated by reference in its entirety. In some embodiments, a linker sequence may be a flexible linker sequence. Flexible linker sequences allow for relative movement of the amino acid sequences which are linked by the linker sequence. Flexible linkers are known to the skilled person, and several are identified in Chen et al., Adv Drug Deliv Rev (2013) 65(10): 1357-1369. Flexible linker sequences often comprise high proportions of glycine and/or serine residues.

In some embodiments, the linker sequence comprises at least one glycine residue and/or at least one serine residue. In some embodiments, the linker sequence comprises or consists of glycine and serine residues. In some embodiments, the linker sequence has the structure: (GxS)n or (GxS)nGm; wherein G = glycine, S = serine, x = 3 or 4, n = 2, 3, 4, 5 or 6, and m = 0, 1 , 2 or 3. In some embodiments, the linker sequence comprises one or more (e.g. 1 , 2, 3, 4, 5 or 6) copies (e.g. in tandem) of the sequence motif G4S. In some embodiments, the linker sequence comprises or consists of (G4S)4 or (G4S)e. In some embodiments, the linker sequence has a length of 1-2, 1-3, 1- 4, 1-5, 1-10, 1-15, 1-20, 1-25, or 1-30 amino acids.

The antigen-binding molecules, polypeptides and CARs of the present disclosure may comprise amino acid sequence(s) to facilitate expression, folding, trafficking, processing, purification or detection of the antigen-binding molecule/polypeptide. For example, antigen-binding molecules and polypeptides of the present disclosure may additionally comprise a sequence of amino acids forming a detectable moiety, e.g. as described hereinbelow.

The antigen-binding molecules, polypeptides and CARs of the present disclosure may additionally comprise a signal peptide (also known as a leader sequence or signal sequence). Signal peptides normally consist of a sequence of 5-30 hydrophobic amino acids, which form a single alpha helix. Secreted proteins and proteins expressed at the cell surface often comprise signal peptides. Signal peptides are known for many proteins, and are recorded in databases such as GenBank, UniProt and Ensembl, and/or can be identified/predicted e.g. using amino acid sequence analysis tools such as SignalP (Petersen et al., 201 1 Nature Methods 8: 785-786) or Signal-BLAST (Frank and Sippl, 2008 Bioinformatics 24: 2172-2176).

The signal peptide may be present at the N-terminus of the antigen-binding molecule/polypeptide/CAR, and may be present in the newly synthesised antigen-binding molecule/polypeptide/CAR. The signal peptide provides for efficient trafficking of the antigen-binding molecule/polypeptide/CAR. Signal peptides are often removed by cleavage, and thus are not comprised in the mature antigen-binding molecule/polypeptide/CAR.

Signal peptides are known for many proteins, and are recorded in databases such as GenBank, UniProt, Swiss-Prot, TrEMBL, Protein Information Resource, Protein Data Bank, Ensembl, and InterPro, and/or can be identified/predicted e.g. using amino acid sequence analysis tools such as SignalP (Petersen et al., 2011 Nature Methods 8: 785-786) or Signal-BLAST (Frank and Sippl, 2008 Bioinformatics 24: 2172-2176).

Labels and conjugates

In some embodiments, the antigen-binding molecule, polypeptide or CAR of the present disclosure additionally comprise a detectable moiety.

In some embodiments, a detectable moiety is a fluorescent label, phosphorescent label, luminescent label, immuno-detectable label (e.g. an epitope tag), radiolabel, chemical, nucleic acid or enzymatic label. The antigen-binding molecule, polypeptide or CAR may be covalently or non-covalently labelled with the detectable moiety.

Fluorescent labels include e.g. fluorescein, rhodamine, allophycocyanin, eosine and NDB, green fluorescent protein (GFP), chelates of rare earths such as europium (Eu), terbium (Tb) and samarium (Sm), tetramethyl rhodamine, Texas Red, 4-methyl umbelliferone, 7-amino-4-methyl coumarin, Cy3, and Cy5. Radiolabels include radioisotopes such as Hydrogen ³ , Sulfur ³⁵ , Carbon ¹⁴ , Phosphorus ³² , Iodine ¹²³ , Iodine ¹²⁵ , Iodine ¹²⁶ , Iodine ¹³¹ , Iodine ¹³³ , Bromine ⁷⁷ , Technetium ^99m , Indium ¹¹¹ , lndium ^113m , Gallium ⁶⁷ , Gallium ⁶⁸ , Ruthenium ⁹⁵ , Ruthenium ⁹⁷ , Ruthenium ¹⁰³ , Ruthenium ¹⁰⁵ , Mercury ²⁰⁷ , Mercury ²⁰³ , Rhenium ^99m , Rhenium ¹⁰¹ , Rhenium ¹⁰⁵ , Scandium ⁴⁷ , Tellurium ^121m , Tellurium ^122m , Tellurium ^125m , Thulium ¹⁶⁵ , Thuliuml ¹⁶⁷ , Thulium ¹⁶⁸ , Copper ⁶⁷ , Fluorine ¹⁸ , Yttrium ⁹⁰ , Palladium ¹⁰⁰ , Bismuth ²¹⁷ and Antimony ²¹¹ . Luminescent labels include as radioluminescent, chemiluminescent (e.g. acridinium ester, luminol, isoluminol) and bioluminescent labels. Immuno- detectable labels include haptens, peptides/polypeptides, antibodies, receptors and ligands such as biotin, avidin, streptavidin or digoxigenin. Nucleic acid labels include aptamers. In some embodiments, the antigen-binding molecule/polypeptide/CAR comprises an epitope tag, e.g. a His, (e.g. 6XHis), FLAG, c-Myc, StrepTag, haemagglutinin, E, calmodulin-binding protein (CBP), glutathione-s-transferase (GST), maltose-binding protein (MBP), thioredoxin, S-peptide, T7 peptide, SH2 domain, avidin, streptavidin, and haptens (e.g. biotin, digoxigenin, dinitrophenol), optionally at the N- or C- terminus of the antigen-binding molecule/polypeptide/CAR.

In some embodiments, the antigen-binding molecule/polypeptide/CAR comprises a moiety having a detectable activity, e.g. an enzymatic moiety. Enzymatic moieties include e.g. luciferases, glucose oxidases, galactosidases (e.g. beta-galactosidase), glucorinidases, phosphatases (e.g. alkaline phosphatase), peroxidases (e.g. horseradish peroxidase) and cholinesterases.

In some embodiments, the antigen-binding molecule/polypeptide/CAR of the present disclosure is conjugated to a chemical moiety. The chemical moiety may be a moiety for providing a therapeutic effect, i.e. a drug moiety. A drug moiety may be a small molecule (e.g. a low molecular weight (< 1000 daltons, typically between ~300-700 daltons) organic compound). Drug moieties are described e.g. in Parslow et al., Biomedicines. 2016 Sep; 4(3):14 (hereby incorporated by reference in its entirety). In some embodiments, a drug moiety may be or comprise a cytotoxic agent. In some embodiments, a drug moiety may be or comprise a chemotherapeutic agent. Drug moieties include e.g. calicheamicin, DM1 , DM4, monomethylauristatin E (MMAE), monomethylauristatin F (MMAF), SN-38, doxorubicin, duocarmycin, D6.5 and PBD.

Particular exemplary polypeptides, antigen-binding molecules and CARs

In some embodiments, the antigen-binding molecule/polypeptide of the present disclosure comprises, or consists of, an amino acid sequence comprising the CDRs of P2A5.

In some embodiments, the antigen-binding molecule/polypeptide of the present disclosure comprises, or consists of, an amino acid sequence comprising the FRs of P2A5.

In some embodiments, the antigen-binding molecule/polypeptide of the present disclosure comprises, or consists of, an amino acid sequence having at least 70%, preferably one of 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% or 100% amino acid sequence identity to SEQ ID NO:8.

In some embodiments of the present disclosure, the CAR comprises, or consists of:

An antigen-binding domain comprising or consisting of an amino acid sequence having at least 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:8; A spacer domain comprising or consisting of an amino acid sequence having at least 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:10;

A transmembrane domain comprising or consisting of an amino acid sequence having at least 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:11 ;

A co-stimulatory domain comprising or consisting of an amino acid sequence having at least 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:12; and

A signalling domain comprising or consisting of an amino acid sequence having at least 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:13.

In some embodiments of the present disclosure, the CAR comprises, or consists of an amino acid sequence having at least 60%, 65%, 70%, 75%, 80%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% sequence identity to the amino acid sequence of SEQ ID NO:9.

Nucleic acids and vectors

The present disclosure provides a nucleic acid, or a plurality of nucleic acids, encoding an antigenbinding molecule, polypeptide or CAR according to the present disclosure. In some embodiments, the nucleic acid(s) comprise or consist of DNA and/or RNA.

In some embodiments, the nucleic acid(s) may be, or may be comprised in, a vector, or a plurality of vectors. That is, the nucleotide sequence(s) of the nucleic acid(s) may be contained in vector(s). The antigen-binding molecule, polypeptide or CAR according to the present disclosure may be produced within a cell by transcription from a vector encoding the antigen-binding molecule, polypeptide or CAR, and subsequent translation of the transcribed RNA.

Accordingly, the present disclosure also provides a vector, or plurality of vectors, comprising the nucleic acid or plurality of nucleic acids according to the present disclosure. The vector may facilitate delivery of the nucleic acid(s) encoding an antigen-binding molecule, polypeptide or CAR according to the present disclosure. The vector may be an expression vector comprising elements required for expressing nucleic acid(s) comprising/encoding an antigen-binding molecule, polypeptide or CAR according to the present disclosure.

Nucleic acids and vectors according to the present disclosure may be provided in purified or isolated form, i.e. from other nucleic acid, or naturally-occurring biological material. The nucleotide sequence may be contained in a vector, e.g. an expression vector. A ‘vector’ as used herein is a nucleic acid molecule used as a vehicle to transfer exogenous nucleic acid into a cell.

The vector may be a vector for expression of the nucleic acid in the cell. Such vectors may include a promoter sequence operably linked to the nucleotide sequence encoding the sequence to be expressed. A vector may also include a termination codon and expression enhancers. Any suitable vectors, promoters, enhancers and termination codons known in the art may be used to express a peptide or polypeptide from a vector according to the present disclosure.

The term ‘operably linked’ may include the situation where a selected nucleic acid sequence and regulatory nucleic acid sequence (e.g. promoter and/or enhancer) are covalently linked in such a way as to place the expression of nucleic acid sequence under the influence or control of the regulatory sequence (thereby forming an expression cassette). Thus a regulatory sequence is operably linked to the selected nucleic acid sequence if the regulatory sequence is capable of effecting transcription of the nucleic acid sequence. The resulting transcript(s) may then be translated into a desired peptide(s)/polypeptide(s).

Suitable vectors include plasmids, binary vectors, DNA vectors, mRNA vectors, viral vectors (e.g. retroviral vectors, e.g. gammaretroviral vectors (e.g. murine Leukemia virus (MLV)-derived vectors, e.g. SFG vector), lentiviral vectors, adenovirus vectors, adeno-associated virus vectors, vaccinia virus vectors and herpesvirus vectors), transposon-based vectors, and artificial chromosomes (e.g. yeast artificial chromosomes), e.g. as described in Maus et al., Annu Rev Immunol (2014) 32:189- 225 or Morgan and Boyerinas, Biomedicines (2016) 4:9, which are both hereby incorporated by reference in their entirety.

In some embodiments, the vector may be a eukaryotic vector, e.g. a vector comprising the elements necessary for expression of protein from the vector in a eukaryotic cell. In some embodiments, the vector may be a mammalian vector, e.g. comprising a cytomegalovirus (CMV) or SV40 promoter to drive protein expression.

Constituent polypeptides of an antigen-binding molecule/CAR according to the present disclosure may be encoded by different nucleic acids of the plurality of nucleic acids, or by different vectors of the plurality of vectors.

In some embodiments, the nucleic acid encodes an antigen-binding molecule or a CAR as described herein. In some embodiments, the vector is multicistronic (e.g. bicistronic, tricistronic, etc.); that is, in some embodiments, the vector encodes mRNA with multiple protein-coding regions. In some embodiments, the vector is bicistronic. Constituent polypeptides of an antigen-binding molecule or a CAR according to the present disclosure may be encoded by different nucleic acids, or by different vectors. The present disclosure provides a retroviral vector comprising nucleic acid encoding an antigenbinding molecule or a CAR according to the present disclosure. In some embodiments, the retroviral vector comprises nucleic acid that encodes an antigen-binding molecule that binds to B7-H3. In some embodiments, the retroviral vector comprises nucleic acid that encodes a B7-H3-specifc CAR.

In some embodiments, a CAR or antigen-binding molecule according to the present disclosure may be encoded by a plasmid. The plasmid may be based on plasmid pSFG (described for example in Hakre et al Mol Cell. 2006 Oct 20. 24(2):301-8 which is incorporated by reference herein in its entirety).

Producing the antigen-binding molecules and polypeptides

Antigen-binding molecules, polypeptides and CARs according to the present disclosure may be prepared according to methods for the production of polypeptides known to the skilled person.

Antigen-binding molecules, polypeptides and CARs may be prepared by chemical synthesis, e.g. liquid or solid phase synthesis. For example, peptides/polypeptides can be synthesised using the methods described in, for example, Chandrudu et al., Molecules (2013), 18: 4373-4388, which is hereby incorporated by reference in its entirety.

Alternatively, antigen-binding molecules, polypeptides and CARs may be produced by recombinant expression. Molecular biology techniques suitable for recombinant production of polypeptides are well known in the art, such as those set out in Green and Sambrook, Molecular Cloning: A Laboratory Manual (4th Edition), Cold Spring Harbor Press, 2012, and in Nat Methods. (2008); 5(2): 135-146 both of which are hereby incorporated by reference in their entirety. Methods for the recombinant production of antigen-binding molecules are also described in Frenzel etal., Front Immunol. (2013); 4: 217 and Kunert and Reinhart, Appl Microbiol Biotechnol. (2016) 100: 3451- 3461 , both of which are hereby incorporated by reference in their entirety.

In some cases, the antigen-binding molecules and CARs of the present disclosure are comprised of more than one polypeptide chain. In such cases, production of the antigen-binding molecule/CAR may comprise transcription and translation of more than one polypeptide, and subsequent association of the polypeptide chains to form the antigen-binding molecule/CAR.

For recombinant production according to the present disclosure, any cell suitable for the expression of polypeptides may be used. The cell may be a prokaryote or eukaryote. In some embodiments, the cell is a prokaryotic cell, such as a cell of archaea or bacteria. In some embodiments, the bacteria may be Gram-negative bacteria such as bacteria of the family Enterobacteriaceae, for example Escherichia coli. In some embodiments, the cell is a eukaryotic cell such as a yeast cell, a plant cell, insect cell or a mammalian cell, e.g. a cell described hereinabove. In some cases, the cell is not a prokaryotic cell because some prokaryotic cells do not allow for the same folding or post-translational modifications as eukaryotic cells. In addition, very high expression levels are possible in eukaryotes and proteins can be easier to purify from eukaryotes using appropriate tags. Specific plasmids may also be utilised which enhance secretion of the protein into the media.

In some embodiments polypeptides may be prepared by cell-free-protein synthesis (CFPS), e.g. according to a system described in Zemella et al. Chembiochem (2015) 16(17): 2420-2431 , which is hereby incorporated by reference in its entirety.

Production may involve culture or fermentation of a eukaryotic cell modified to express the polypeptide(s) of interest. The culture or fermentation may be performed in a bioreactor provided with an appropriate supply of nutrients, air/oxygen and/or growth factors. Secreted proteins can be collected by partitioning culture media/fermentation broth from the cells, extracting the protein content, and separating individual proteins to isolate secreted polypeptide(s). Culture, fermentation and separation techniques are well known to those of skill in the art, and are described, for example, in Green and Sambrook, Molecular Cloning: A Laboratory Manual (4th Edition; incorporated by reference herein above).

Bioreactors include one or more vessels in which cells may be cultured. Culture in the bioreactor may occur continuously, with a continuous flow of reactants into, and a continuous flow of cultured cells from, the reactor. Alternatively, the culture may occur in batches. The bioreactor monitors and controls environmental conditions such as pH, oxygen, flow rates into and out of, and agitation within the vessel such that optimum conditions are provided for the cells being cultured.

Following culturing the cells that express the polypeptide(s), the polypeptide(s) of interest may be isolated. Any suitable method for separating proteins from cells known in the art may be used. In order to isolate the polypeptide, it may be necessary to separate the cells from nutrient medium. If the polypeptide(s) are secreted from the cells, the cells may be separated by centrifugation from the culture media that contains the secreted polypeptide(s) of interest. If the polypeptide(s) of interest collect within the cell, protein isolation may comprise centrifugation to separate cells from cell culture medium, treatment of the cell pellet with a lysis buffer, and cell disruption e.g. by sonification, rapid freeze-thaw or osmotic lysis.

It may then be desirable to isolate the polypeptide(s) of interest from the supernatant or culture medium, which may contain other protein and non-protein components. A common approach to separating protein components from a supernatant or culture medium is by precipitation. Proteins of different solubilities are precipitated at different concentrations of precipitating agent such as ammonium sulfate. For example, at low concentrations of precipitating agent, water soluble proteins are extracted. Thus, by adding different increasing concentrations of precipitating agent, proteins of different solubilities may be distinguished. Dialysis may be subsequently used to remove ammonium sulfate from the separated proteins.

Other methods for distinguishing different proteins are known in the art, for example ion exchange chromatography and size chromatography. These may be used as an alternative to precipitation or may be performed subsequently to precipitation.

Once the polypeptide(s) of interest have been isolated from culture it may be desired or necessary to concentrate the polype ptide(s). A number of methods for concentrating proteins are known in the art, such as ultrafiltration or lyophilisation.

Cells comprisinq/expressinq the antigen-binding molecules and polypeptides

The present disclosure also provides a cell comprising or expressing an antigen-binding molecule, polypeptide or CAR according to the present disclosure. Also provided is a cell comprising or expressing a nucleic acid, a plurality of nucleic acids, a vector or a plurality of vectors according to the present disclosure.

It will be appreciated that where cells are referred to herein in the singular (/.e. ‘a/the cell’), pluralities/populations of such cells are also contemplated.

The cell may be a eukaryotic cell, e.g. a mammalian cell. The mammal may be a primate (rhesus, cynomolgous, non-human primate or human) or a non-human mammal (e.g. rabbit, guinea pig, rat, mouse or other rodent (including any animal in the order Rodentia), cat, dog, pig, sheep, goat, cattle (including cows, e.g. dairy cows, or any animal in the order Bos), horse (including any animal in the order Equidae), donkey, and non-human primate).

In some embodiments, the cell is, or is derived from, a cell type commonly used for the expression of polypeptides for use in therapy in humans. Exemplary cells are described e.g. in Kunert and Reinhart, Appl Microbiol Biotechnol. (2016) 100:3451-3461 (hereby incorporated by reference in its entirety), and include e.g. CHO, HEK 293, PER.C6, NSO and BHK cells. In preferred embodiments, the cell is, or is derived from, a CHO cell.

The present disclosure also provides a method for producing a cell comprising a nucleic acid(s) or vector(s) according to the present disclosure, comprising introducing a nucleic acid, a plurality of nucleic acids, a vector or a plurality of vectors according to the present disclosure into a cell. In some embodiments, introducing an isolated nucleic acid(s) or vector(s) according to the present disclosure into a cell comprises transformation, transfection, electroporation or transduction (e.g. retroviral transduction). The present disclosure also provides a method for producing a cell expressing/comprising an antigen-binding molecule, polypeptide or CAR according to the present disclosure, comprising introducing a nucleic acid, a plurality of nucleic acids, a vector or a plurality of vectors according to the present disclosure in a cell. In some embodiments, the methods additionally comprise culturing the cell under conditions suitable for expression of the nucleic acid(s) or vector(s) by the cell. In some embodiments, the methods are performed in vitro.

The present disclosure also provides cells obtained or obtainable by the methods according to the present disclosure.

Cells expressing the CARs of the disclosure

In some aspects and embodiments, the present disclosure provides a cell comprising a CAR according to the present disclosure. The CAR according to the present disclosure may be used to generate CAR-expressing cells, e.g. CAR-expressing immune cells (e.g. CAR-T or CAR-NK cells).

CAR-expressing cells may comprise or express nucleic acid encoding a CAR according to the present disclosure. It will be appreciated that a CAR-expressing cell comprises the CAR it expresses. It will also be appreciated that a cell expressing nucleic acid encoding a CAR also expresses and comprises the CAR encoded by the nucleic acid.

CAR-expressing cell is preferably an immune cell. An immune cell may be a cell of hematopoietic origin, e.g. a neutrophil, eosinophil, basophil, dendritic cell, lymphocyte, or monocyte. A lymphocyte may be e.g. a T cell, B cell, NK cell, NKT cell or innate lymphoid cell (ILC), or a precursor thereof. The immune cell may express e.g. CD3 polypeptides (e.g. CD3y CD3e CD3 or CD36), TCR polypeptides (TCRa or TCRP), CD27, CD28, CD4 or CD8. In some embodiments, the immune cell is a T cell, e.g. a CD3+ T cell. In some embodiments, the T cell is a CD3+, CD4+ T cell. In some embodiments, the T cell is a CD3+, CD8+ T cell. In some embodiments, the T cell is a T helper cell (TH cell). In some embodiments, the T cell is a cytotoxic T cell (e.g. a cytotoxic T lymphocyte (CTL)).

Aspects and embodiments of the present disclosure relate particularly to T cells comprising/expressing B7-H3-specific CARs according to the present disclosure.

Immune cells useful in the methods described herein may be obtained from any suitable source. The source may be animal or human. The source may be a non-human mammal, but is more preferably human. The source may be any gender. The source may be the patient that is to be treated with adoptive cell therapy (autologous cells). As such, the source may have been diagnosed with a disease/condition requiring treatment, may be suspected of having such a disease/condition, or may be at risk of developing/contracting such a disease/condition. In some cases, the source is a different individual to the patient that is to be treated (allogeneic cells). In such cases, the source would normally be a healthy individual, or an individual that is not known to be suffering from a disease/condition or at risk of developing/contracting such a disease/condition.

In some aspects and embodiments, the immune cell may be a virus-specific immune cell. A ‘virusspecific immune cell’ as used herein refers to an immune cell which is specific for a virus. A virusspecific immune cell expresses/comprises a receptor (preferably a T cell receptor) capable of recognising a peptide of an antigen of a virus (e.g. when presented by an MHC molecule). The virusspecific immune cell may express/comprise such a receptor as a result of expression of endogenous nucleic acid encoding such antigen receptor, or as a result of having been engineered to express such a receptor. The virus-specific immune cell preferably expresses/comprises a TCR specific for a peptide of an antigen of a virus. A virus-specific T cell may display certain functional properties of a T cell in response to the viral antigen for which the T cell is specific, or in response a cell comprising/expressing the virus/antigen. In some embodiments, the properties are functional properties associated with effector T cells, e.g. cytotoxic T cells.

In some embodiments, a virus-specific T cell may display one or more of the following properties: cytotoxicity to a cell comprising/expressing the virus /the viral antigen for which the T cell is specific; proliferation, IFNy expression, CD107a expression, IL-2 expression, TNFa expression, perforin expression, granzyme expression, granulysin expression, and/or FAS ligand (FASL) expression in response to stimulation with the virus/the viral antigen for which the T cell is specific, or in response to exposure to a cell comprising/expressing the virus /the viral antigen for which the T cell is specific.

Virus-specific T cells express/comprise a TCR capable of recognising a peptide of the viral antigen for which the T cell is specific when presented by the appropriate MHC molecule. Virus-specific T cells may be CD4+ T cells and/or CD8+ T cells.

The virus for which the virus-specific immune cell is specific may be any virus. For example, the virus may be a dsDNA virus (e.g. adenovirus, herpesvirus, poxvirus), ssRNA virus (e.g. parvovirus), dsRNA virus (e.g. reovirus), (+)ssRNA virus (e.g. picornavirus, togavirus), (-)ssRNA virus (e.g. orthomyxovirus, rhabdovirus), ssRNA-RT virus (e.g. retrovirus) or dsDNA-RT virus (e.g. hepadnavirus). In particular, the present disclosure contemplates viruses of the families adenoviridae, herpesviridae, papillomaviridae, polyomaviridae, poxviridae, hepadnaviridae, parvoviridae, astroviridae, caliciviridae, picornaviridae, coronaviridae, flaviviridae, togaviridae, hepeviridae, retroviridae, orthomyxoviridae, arenaviridae, bunyaviridae, filoviridae, paramyxoviridae, rhabdoviridae and reoviridae. In some embodiments the virus is selected from Epstein-Barr virus, adenovirus, Herpes simplex type 1 virus, Herpes simplex type 2 virus, Varicella-zoster virus, Human cytomegalovirus, Human herpesvirus type 8, Human papillomavirus, BK virus, JC virus, Smallpox, Hepatitis B virus, Parvovirus B19, Human Astrovirus, Norwalk virus, coxsackievirus, hepatitis A virus, poliovirus, rhinovirus, severe acute respiratory syndrome virus, Hepatitis C virus, yellow fever virus, dengue virus, West Nile virus, TBE virus, Rubella virus, Hepatitis E virus, Human immunodeficiency virus, influenza virus, lassa virus, Crimean-Congo hemorrhagic fever virus, Hantaan virus, ebola virus, Marburg virus, measles virus, mumps virus, parainfluenza virus, picornavirus, respiratory syncytial virus, rabies virus, hepatitis D virus, rotavirus, orbivirus, coltivirus, and banna virus.

In some embodiments, the virus is selected from Epstein-Barr virus (EBV), adenovirus, cytomegalovius (CMV), human papilloma virus (HPV), influenza virus, measles virus, hepatitis B virus (HBV), hepatitis C virus (HCV), human immunodeficiency virus (HIV), lymphocytic choriomeningitis virus (LCMV), or herpes simplex virus (HSV).

In some embodiments, the virus-specific immune cell may be specific for a peptide/polypeptide of a virus e.g. selected from Epstein-Barr virus (EBV), adenovirus, cytomegalovius (CMV), human papilloma virus (HPV), influenza virus, measles virus, hepatitis B virus (HBV), hepatitis C virus (HCV), human immunodeficiency virus (HIV), lymphocytic choriomeningitis virus (LCMV), or herpes simplex virus (HSV).

A T cell which is specific for an antigen of a virus may be referred to herein as a virus-specific T cell (VST). A T cell which is specific for an antigen of a particular virus may be described as being specific for the relevant virus; for example, a T cell which is specific for an antigen of EBV may be referred to as an EBV-specific T cell, or ‘EBVST’.

Accordingly, in some embodiments the virus-specific immune cell is an Epstein-Barr virus-specific T cell (EBVST), adenovirus-specific T cell (AdVST), cytomegalovius-specific T cell (CMVST), human papilloma virus (HPVST), influenza virus-specific T cell, measles virus-specific T cell, hepatitis B virus-specific T cell (HBVST), hepatitis C virus-specific T cell (HCVST), human immunodeficiency virus-specific T cell (HIVST), lymphocytic choriomeningitis virus-specific T cell (LCMVST), or herpes simplex virus-specific T cell (HSVST).

In some preferred embodiments, the virus-specific immune cell is specific for a peptide/polypeptide of an EBV antigen. In preferred embodiments the virus-specific immune cell is an Epstein-Barr virusspecific T cell (EBVST).

EBV virology is described e.g. in Stanfield and Luftiq, F OORes. (2017) 6:386 and Odumade et al., Clin Microbiol Rev (201 1) 24(1 ):193-209, both of which are hereby incorporated by reference in their entirety.

EBV infects epithelial cells via binding of viral protein BMFR2 to pi integrins, and binding of viral protein gH/gL with integrins av06 and av08. EBV infects B cells through interaction of viral glycoprotein gp350 with CD21 and/or CD35, followed by interaction of viral gp42 with MHC class II. These interactions trigger fusion of the viral envelope with the cell membrane, allowing the virus to enter the cell. Once inside, the viral capsid dissolves and the viral genome is transported to the nucleus.

EBV has two modes of replication; latent and lytic. The latent cycle does not result in production of virions, and can take place in place B cells and epithelial cells. The EBV genomic circular DNA resides in the cell nucleus as an episome and is copied by the host cell’s DNA polymerase. In latency, only a fraction of EBV's genes are expressed, in one of three different patterns known as latency programs, which produce distinct sets of viral proteins and RNAs. The latent cycle is described e.g. in Amon and Farrell, Reviews in Medical Virology (2004) 15(3): 149-56, which is hereby incorporated by reference in its entirety.

EBNA1 protein and non-coding RNA EBER are expressed in each of latency programs l-lll. Latency programs II and III further involve expression of EBNALP, LMP1 , LMP2A and LMP2B proteins, and latency program III further involves expression of EBNA2, EBNA3A, EBNA3B and EBNA3C.

EBNA1 is multifunctional, and has roles in gene regulation, extrachromosomal replication, and maintenance of the EBV episomal genome through positive and negative regulation of viral promoters (Duellman et al., J Gen Virol. (2009); 90(Pt 9): 2251-2259). EBNA2 is involved in the regulation of latent viral transcription and contributes to the immortalisation of cells infected with EBV (Kempkes and Ling, Curr Top Microbiol Immunol. (2015) 391 :35-59). EBNA-LP is required for transformation of native B cells, and recruits transcription factors for viral replication (Szymula et al., PLoS Pathog. (2018);14(2):e1006890). EBNA3A, 3B and 3C interact with RBPJ to influence gene expression, contributing to survival and growth of infected cells (Wang et al., J Virol. (2016) 90(6):2906-2919). LMP1 regulates expression of genes involved in B cell activation (Chang et al., J. Biomed. Sci. (2003) 10(5): 490-504). LMP2A and LMP2B inhibit normal B cell signal transduction by mimicking the activated B cell receptor (Portis and Longnecker, Oncogene (2004) 23(53): 8619- 8628). EBERs form ribonucleoprotein complexes with host cell proteins, and are proposed to have roles in cell transformation.

The latent cycle can progress according to any of latency programs I to III in B cells, and usually progresses from III to II to I. Upon infection of a resting naive B cell, EBV enters latency program III. Expression of latency III genes activates the B cell, which becomes a proliferating blast. EBV then typically progresses to latency II by restricting expression to a subset of genes, which cause differentiation of the blast to a memory B cell. Further restriction of gene expression causes EBV to enter latency I. EBNA1 expression allows EBV to replicate when the memory B cell divides. In epithelial cells, only latency II occurs.

In primary infection, EBV replicates in oropharyngeal epithelial cells and establishes Latency III, II, and I infections in B-lymphocytes. EBV latent infection of B- lymphocytes is necessary for virus persistence, subsequent replication in epithelial cells, and release of infectious virus into saliva. EBV Latency III and II infections of B- lymphocytes, Latency II infection of oral epithelial cells, and Latency II infection of NK- or T cell can result in malignancies, marked by uniform EBV genome presence and gene expression.

Latent EBV in B cells can be reactivated to switch to lytic replication. The lytic cycle results in the production of infectious virions and can take place in place B cells and epithelial cells, and is reviewed e.g. by Kenney in Chapter 25 of Arvin et al., Human Herpesviruses: Biology, Therapy and Immunoprophylaxis; Cambridge University Press (2007), which is hereby incorporated by reference in its entirety.

Lytic replication requires the EBV genome to be linear. The latent EBV genome is episomal, and so it must be linearised for lytic reactivation. In B cells, lytic replication normally only takes place after reactivation from latency.

Immediate-early lytic gene products such as BZFL1 and BRLF1 act as transactivators, enhancing their own expression, and the expression of later lytic cycle genes.

Early lytic gene products have roles in viral replication (e.g. EBV DNA polymerase catalytic component BALF5; DNA polymerase processivity factor BMRF1 , DNA binding protein BALF2, helicase BBLF4, primase BSLF1 , and primase-associated protein BBLF2/3) and deoxynucleotide metabolism (e.g. thymidine kinase BXLF1 , dUTPase BORF2). Other early lytic gene products act transcription factors (e.g. BMRF1 , BRRF1), have roles in RNA stability and processing (e.g. BMLF1), or are involved in immune evasion (e.g. BHRF1 , which inhibits apoptosis).

Late lytic gene products are traditionally classed as those expressed after the onset of viral replication. They generally encode structural components of the virion such as nucleocapsid proteins, as well as glycoproteins which mediate EBV binding and fusion (e.g. gp350/220, gp85, gp42, gp25). Other late lytic gene products have roles in immune evasion; BCLF1 encodes a viral homologue of IL-10, and BALF1 encodes a protein with homology to the anti-apoptotic protein Bcl2.

An ‘EBV-specific immune cell’ as used herein refers to an immune cell which is specific for Epstein- Barr virus (EBV). An EBV-specific immune cell expresses/comprises a receptor (preferably a T cell receptor) capable of recognising a peptide of an antigen of EBV (e.g. when presented by an MHC molecule). The EBV-specific immune cell preferably expresses/comprises a TCR specific for a peptide of an EBV antigen presented by MHC class I.

In some embodiments, the EBV-specific immune cell is a T cell, e.g. a CD3+ T cell. In some embodiments, the T cell is a CD3+, CD4+ T cell. In some embodiments, the T cell is a CD3+, CD8+ T cell. In some embodiments, the T cell is a T helper cell (TH cell)). In some embodiments, the T cell is a cytotoxic T cell (e.g. a cytotoxic T lymphocyte (CTL)). EBV-specific T cells preferably express/comprise a TCR capable of recognising a peptide of the EBV antigen for which the T cell is specific when presented by the appropriate MHC molecule. EBV- specific T cells may be CD4+ T cells and/or CD8+ T cells.

An immune cell specific for EBV may be specific for any EBV antigen, e.g. an EBV antigen described herein. A population of immune cell specific for EBV, or a composition comprising a plurality of immune cells specific for EBV, may comprise immune cells specific for one or more EBV antigens.

In some embodiments, an EBV antigen is an EBV latent antigen, e.g. a type III latency antigen (e.g. EBNA1 , EBNA-LP, LMP1 , LMP2A, LMP2B, BARF1 , EBNA2, EBNA3A, EBNA3B or EBNA3C), a type II latency antigen (e.g. EBNA1 , EBNA-LP, LMP1 , LMP2A, LMP2B or BARF1), or a type I latency antigen, (e.g. EBNA1 or BARF1). In some embodiments, an EBV antigen is an EBV lytic antigen, e.g. an immediate-early lytic antigen (e.g. BZLF1 , BRLF1 or BMRF1), an early lytic antigen (e.g. BMLF1 , BMRF1 , BXLF1 , BALF1 , BALF2, BARF1 , BGLF5, BHRF1 , BNLF2A, BNLF2B, BHLF1 , BLLF2, BKRF4, BMRF2, FU or EBNA1-FUK), or a late lytic antigen (e.g. BALF4, BILF1 , BILF2, BNFR1 , BVRF2, BALF3, BALF5, BDLF3 or gp350).

In some embodiments in accordance with the various aspects of the present disclosure, cells may comprise/express more than one (e.g. 2, 3, 4, etc.) CAR.

In some embodiments, the cells may comprise/express more than one, non-identical CAR. Cells comprising/expressing more than one non-identical CAR may comprise/express CARs specific for non-identical target antigens. In some embodiments, each non-identical target antigen is independently a cancer cell antigen as described herein.

Functional properties of cells expressing the CARs of the disclosure

Cells (e.g. immune cells, e.g. T cells) expressing a CAR according to the present disclosure may display certain functional properties in response to B7-H3, or in response a cell comprising/expressing B7-H3. In some embodiments, the properties are functional properties associated with effector T cells, e.g. cytotoxic T cells.

Cells comprising a CAR/nucleic acid encoding a CAR according to the present disclosure may display one or more of the following properties: expression of one or more cytotoxic/effector factors (e.g. IFNy, TNFa, GM-CSF), proliferation/population expansion, and/or growth factor (e.g. IL-2) expression in response to cells expressing B7-H3; cytotoxicity to cells expressing B7-H3; no cytotoxicity (i.e. above baseline) to cells which do not express B7-H3; and/or anti-cancer activity (e.g. cytotoxicity to cancer cells, tumor growth inhibition, reduction of tumor burden, reduction of metastasis, etc.) against cancer comprising cells expressing B7- H3.

In some embodiments, a B7-H3-specific CAR-expressing T cell may display one or more of the following properties: cytotoxicity to a cell comprising/expressing B7-H3; proliferation, IFNy expression, CD107a expression, IL-2 expression, TNFa expression, perforin expression, granzyme expression, granulysin expression, and/or FAS ligand (FASL) expression in response to stimulation with B7-H3, or in response to exposure to a cell comprising/expressing B7-H3; proliferation/population expansion in response to stimulation with B7-H3, or in response to exposure to a cell comprising/expressing B7-H3; cytotoxicity to cells of a cancer expressing B7-H3; inhibition of tumor growth of a cancer comprising cells expressing B7-H3; reduction of tumor burden in a subject of a cancer comprising cells expressing B7-H3; reduction of metastasis of a cancer comprising cells expressing B7-H3.

In some embodiments, cells comprising a CAR/nucleic acid encoding a CAR according to the present disclosure display cytotoxicity to A549 cells, MDA-MB-231 cells and/or THP-1 cells. In some embodiments, cells comprising a CAR/nucleic acid encoding a CAR according to the present disclosure display cytotoxicity to colorectal cancer cells (e.g. DLD-1 cells, HT29 cells and/or SW480 cells). In some embodiments, cells comprising a CAR/nucleic acid encoding a CAR according to the present disclosure display cytotoxicity to gastric cancer cells (e.g. NCI-N87 cells, MKN7 cells and/or MKN45 cells). In some embodiments, cells comprising a CAR/nucleic acid encoding a CAR according to the present disclosure display cytotoxicity to breast cancer cells (e.g. MDA-MB-231 cells and/or MDA-MB-468 cells). In some embodiments, cells comprising a CAR/nucleic acid encoding a CAR according to the present disclosure display cytotoxicity to lung cancer cells (e.g. A549 cells, H1299 cells, H23 cells and/or H5967 cells).

In some embodiments, cells comprising a CAR/nucleic acid encoding a CAR according to the present disclosure display serial killing of cells expressing B7-H3, e.g. cancer cells expressing B7- H3. In some embodiments, cells comprising a CAR/nucleic acid encoding a CAR according to the present disclosure display serial killing of THP-1 cells, DLD-1 cells, HT29 cells and/or NCI-N87 cells.

Cell proliferation/population expansion can be investigated by analysing cell division or the number of cells over a period of time. Cell division can be analysed, for example, by in vitro analysis of incorporation of ³ H-thymidine or by CFSE dilution assay, e.g. as described in Fulcher and Wong, Immunol Cell Biol (1999) 77(6): 559-564, hereby incorporated by reference in entirety. Proliferating cells can also be identified by analysis of incorporation of 5-ethynyl-2'-deoxyuridine (EdU) by an appropriate assay, as described e.g. in Buck et al., Biotechniques. 2008 Jun; 44(7):927-9, and Sali and Mitchison, PNAS USA 2008 Feb 19; 105(7): 2415-2420, both hereby incorporated by reference in their entirety. As used herein, ‘expression’ may be gene or protein expression. Gene expression encompasses transcription of DNA to RNA, and can be measured by various means known to those skilled in the art, for example by measuring levels of mRNA by quantitative real-time PCR (qRT-PCR), or by reporter-based methods. Similarly, protein expression can be measured by various methods well known in the art, e.g. by antibody-based methods, for example by western blot, immunohistochemistry, immunocytochemistry, flow cytometry, ELISA, ELISPOT, or re porter- based methods.

Cytotoxicity and cell killing can be investigated, for example, using any of the methods reviewed in Zaritskaya et al., Expert Rev Vaccines (2011), 9(6):601-616, hereby incorporated by reference in its entirety. Examples of in vitro assays of cytotoxicity/cell killing assays include release assays such as the ⁵¹ Cr release assay, the lactate dehydrogenase (LDH) release assay, the 3-(4,5-dimethylthiazol- 2-yl)-2,5-diphenyl tetrazolium bromide (MTT) release assay, and the calcein-acetoxymethyl (calcein- AM) release assay. These assays measure cell killing based on the detection of factors released from lysed cells. Cell killing by a given cell type can be analysed e.g. by co-culturing the test cells with the given cell type, and measuring the number/proportion of viable/dead test cells after a suitable period of time.

In some embodiments, cell killing of cells expressing B7-H3 by CAR-expressing cells may be evaluated by flow cytometry as described in Example 1 .13 herein, or by xCELLigence assay as described in Example 1.14 herein. Cell killing by CAR-expressing cells can also be evaluated in vivo, e.g. by evaluating the number/proportion of cells expressing the target antigen for the CAR, and inferring their killing/depletion by CAR-expressing cells.

Cells may be evaluated for anti-cancer activity (e.g. cytotoxicity to cancer cells, tumor growth inhibition, reduction of tumor burden, reduction of metastasis, etc.) by analysis in an appropriate in vitro assays or in vivo models of the relevant cancer. For example, the cells may be evaluated maybe evaluated in cancer cell line-derived xenograft models, and such analysis may be performed e.g. as described in Example 1 .16 herein. By way of illustration, the Examples of the present disclosure demonstrate inhibition of tumor growth and reduction of tumor burden in HT29, SW-480 and N-87 cell line-derived xenograft models of colorectal cancer and gastric cancer, following administration of immune cells expressing B7-H3-specific CARs according to the present disclosure.

Producing cells expressing the CARs of the disclosure

Methods for producing CAR-expressing cells are well known to the skilled person. They generally involve modifying cells (e.g. immune cells, e.g. T cells or NK cells) to express/comprise a CAR, e.g. introducing nucleic acid encoding a CAR into the immune cells. Immune cells may be modified to comprise/express a CAR or nucleic acid encoding a CAR described herein according to methods that are well known to the skilled person. The methods generally comprise nucleic acid transfer for permanent (stable) or transient expression of the transferred nucleic acid.

Any suitable genetic engineering platform may be used to modify a cell according to the present disclosure. Suitable methods for modifying a cell include the use of genetic engineering platforms such as gammaretroviral vectors, lentiviral vectors, adenovirus vectors, DNA transfection, transposon-based gene delivery and RNA transfection, for example as described in Maus et al., Annu Rev Immunol (2014) 32:189-225, hereby incorporated by reference in its entirety.

Methods also include those described e.g. in Wang and Riviere Mol Ther Oncolytics. (2016) 3:16015, which is hereby incorporated by reference in its entirety. Suitable methods for introducing nucleic acid(s)/vector(s) into cells include transduction, transfection and electroporation.

Methods for generating/expanding populations of CAR-expressing immune cells in vitro/ex vivo are well known to the skilled person. Suitable culture conditions (/.e. cell culture media, additives, stimulations, temperature, gaseous atmosphere), cell numbers, culture periods and methods for introducing nucleic acid encoding a CAR into cells, etc. can be determined by reference e.g. to Hornbach et al. J Immunol (2001) 167:6123-6131 , Ramos et al. J. Clin. Invest. (2017) 127(9):3462- 3471 and WO 2015/028444 A1 , all of which are hereby incorporated by reference in their entirety.

Conveniently, cultures of cells according to the present disclosure may be maintained at 37°C in a humidified atmosphere containing 5% CO2. The cells of cell cultures can be established and/or maintained at any suitable density, as can readily be determined by the skilled person.

Cultures can be performed in any vessel suitable for the volume of the culture, e.g. in wells of a cell culture plate, cell culture flasks, a bioreactor, etc. In some embodiments cells are cultured in a bioreactor, e.g. a bioreactor described in Somerville and Dudley, Oncoimmunology (2012)

1 (8):1435-1437, which is hereby incorporated by reference in its entirety. In some embodiments cells are cultured in a GRex cell culture vessel, e.g. a GRex flask or a GRex 100 bioreactor.

Immune cells (e.g. T cells) may be activated prior to introduction of nucleic acid encoding the CAR. For example, T cells within a population of PBMCs may be non-specifically activated by stimulation in vitro with agonist anti-CD3 and agonist anti-CD28 antibodies, in the presence of IL-2.

Introducing nucleic acid(s)/vector(s) into a cell may comprise transduction, e.g. retroviral transduction. Accordingly, in some embodiments the nucleic acid(s) is/are comprised in a viral vector(s), or the vector(s) is/are a viral vector(s). Transduction of immune cells with viral vectors is described e.g. in Simmons and Alberola-lla, Methods Mol Biol. (2016) 1323:99-108, which is hereby incorporated by reference in its entirety.

Agents may be employed to enhance the efficiency of transduction. Hexadimethrine bromide (polybrene) is a cationic polymer which is commonly used to improve transduction, through neutralising charge repulsion between virions and sialic acid residues expressed on the cell surface. Other agents commonly used to enhance transduction include e.g. the poloxamer-based agents such as LentiBOOST (Sirion Biotech), Retronectin (Takara), Vectofusin (Miltenyi Biotech) and also SureENTRY (Qiagen) and ViraDuctin (Cell Biolabs).

In some embodiments the methods comprise centrifuging the cells into which it is desired to introduce nucleic acid encoding the CAR in the presence of cell culture medium comprising viral vector comprising the nucleic acid (referred to in the art as ‘spinfection’).

In some embodiments, the methods comprise introducing a nucleic acid or vector according to the present disclosure into an immune cell by electroporation, e.g. as described in Koh et al., Molecular Therapy - Nucleic Acids (2013) 2, e114, which is hereby incorporated by reference in its entirety.

The methods generally comprise introducing a nucleic acid encoding a CAR into a cell, and culturing the cell under conditions suitable for expression of the nucleic acid/CAR by the cell. In some embodiments, the methods comprise culturing immune cells into which nucleic acid encoding a CAR has been introduced in order to expand their number. In some embodiments, the methods comprise culturing immune cells into which nucleic acid encoding a CAR has been introduced in the presence of IL-7 and/or IL-15 (e.g. recombinant IL-7 and/or IL-15).

In some embodiments, the methods further comprise purifying/isolating CAR-expressing cells, e.g. from other cells (e.g. cells which do not express the CAR). Methods for purifying/isolating immune cells from heterogeneous populations of cells are well known in the art, and may employ e.g. FACS- or MACS-based methods for sorting populations of cells based on the expression of markers of the immune cells. In some embodiments the methods purifying/isolating cells of a particular type, e.g. CAR-expressing CD8+ T cells, CAR-expressing CTLs.

In preferred embodiments, B7-H3-specific CAR-expressing T cells may be generated from T cells within populations of PBMCs by a process comprising: stimulating PBMCs with antagonist anti-CD3 and andti-CD28 antibodies, transducing the cells with a viral vector (e.g. a gamma-retroviral vector) encoding the B7-H3-specific CAR, and subsequently culturing the cells in the presence of IL-7 and IL-15.

Aspects and embodiments of the present disclosure relate particularly to EBV-specific immune cells. Methods for generating/expanding populations of EBV-specific immune cells are described e.g. in WO 2013/088114 A1 , Lapteva and Vera, Stem Cells Int. (2011): 434392, Straathof et al., Blood (2005) 105(5): 1898-1904, WO 2017/202478 A1 , WO 2018/052947 A1 and WO 2020/214479 A1 , all of which are hereby incorporated by reference in their entirety. The methods typically comprise stimulating immune cells specific for a virus/viral antigen by contacting populations of immune cells with peptide(s) corresponding to EBV antigen(s) or APCs presenting peptide(s) corresponding to viral antigen(s).

The present disclosure also provides cells obtained or obtainable by the methods described herein, and populations thereof.

Compositions

The present disclosure also provides compositions comprising the antigen-binding molecules, polypeptides, CARs, nucleic acids, expression vectors and cells described herein.

The antigen-binding molecules, polypeptides, CARs, nucleic acids, expression vectors and cells described herein may be formulated as pharmaceutical compositions or medicaments for clinical use and may comprise a pharmaceutically acceptable carrier, diluent, excipient or adjuvant.

The compositions of the present disclosure may comprise one or more pharmaceutically-acceptable carriers (e.g. liposomes, micelles, microspheres, nanoparticles), diluents/excipients (e.g. starch, cellulose, a cellulose derivative, a polyol, dextrose, maltodextrin, magnesium stearate), adjuvants, fillers, buffers, preservatives (e.g. vitamin A, vitamin E, vitamin C, retinyl palmitate, selenium, cysteine, methionine, citric acid, sodium citrate, methyl paraben, propyl paraben), anti-oxidants (e.g. vitamin A, vitamin E, vitamin C, retinyl palmitate, selenium), lubricants (e.g. magnesium stearate, talc, silica, stearic acid, vegetable stearin), binders (e.g. sucrose, lactose, starch, cellulose, gelatin, polyethylene glycol (PEG), polyvinylpyrrolidone (PVP), xylitol, sorbitol, mannitol), stabilisers, solubilisers, surfactants (e.g., wetting agents), masking agents or colouring agents (e.g. titanium oxide).

The term ‘pharmaceutically-acceptable’ as used herein pertains to compounds, ingredients, materials, compositions, dosage forms, etc., which are, within the scope of sound medical judgment, suitable for use in contact with the tissues of the subject in question (e.g. a human subject) without excessive toxicity, irritation, allergic response, or other problem or complication, commensurate with a reasonable benefit/risk ratio. Each carrier, diluent, excipient, adjuvant, filler, buffer, preservative, anti-oxidant, lubricant, binder, stabiliser, solubiliser, surfactant, masking agent, colouring agent, flavouring agent or sweetening agent of a composition according to the present disclosure must also be ‘acceptable’ in the sense of being compatible with the other ingredients of the formulation. Suitable carriers, diluents, excipients, adjuvants, fillers, buffers, preservatives, anti-oxidants, lubricants, binders, stabilisers, solubilisers, surfactants, masking agents, colouring agents, flavouring agents or sweetening agents can be found in standard pharmaceutical texts, for example, Remington’s ‘The Science and Practice of Pharmacy’ (Ed. A. Adejare), 23rd Edition (2020), Academic Press.

Compositions may be formulated for topical, parenteral, systemic, intracavitary, intravenous, intraarterial, intramuscular, intrathecal, intraocular, intraconjunctival, intratumoral, subcutaneous, intradermal, intrathecal, oral ortransdermal routes of administration. In some embodiments, a pharmaceutical composition/medicament may be formulated for administration by injection or infusion, or administration by ingestion.

Suitable formulations may comprise the relevant article in a sterile or isotonic medium. Medicaments and pharmaceutical compositions may be formulated in fluid, including gel, form. Fluid formulations may be formulated for administration by injection or infusion (e.g. via catheter) to a selected region of the human or animal body.

In some embodiments, the composition is formulated for injection or infusion, e.g. into a blood vessel, tissue/organ of interest, or tumor.

The present disclosure also provides methods for the production of pharmaceutically useful compositions, such methods of production may comprise one or more steps selected from: producing an antigen-binding molecule, polypeptide, CAR, nucleic acid (or plurality thereof), expression vector (or plurality thereof) or cell described herein; isolating an antigen-binding molecule, polypeptide, CAR, nucleic acid (or plurality thereof), expression vector (or plurality thereof) or cell described herein; and/or mixing an antigen-binding molecule, polypeptide, CAR, nucleic acid (or plurality thereof), expression vector (or plurality thereof) or cell described herein with a pharmaceutically acceptable carrier, adjuvant, excipient or diluent.

For example, a further aspect the present disclosure relates to a method of formulating or producing a medicament or pharmaceutical composition for use in the treatment of a disease/condition (e.g. a cancer), the method comprising formulating a pharmaceutical composition or medicament by mixing an antigen-binding molecule, polypeptide, CAR, nucleic acid (or plurality thereof), expression vector (or plurality thereof) or cell described herein with a pharmaceutically acceptable carrier, adjuvant, excipient or diluent.

Therapeutic and prophylactic applications

The antigen-binding molecules, polypeptides, CARs, nucleic acids, expression vectors, cells and compositions described herein find use in therapeutic and prophylactic methods.

The present disclosure provides an antigen-binding molecule, polypeptide, CAR, nucleic acid (or plurality thereof), expression vector (or plurality thereof), cell or composition described herein for use in a method of medical treatment or prophylaxis. Also provided is an antigen-binding molecule, polypeptide, CAR, nucleic acid (or plurality thereof), expression vector (or plurality thereof), cell or composition described herein for use in a method of treating or preventing a disease or condition described herein. Also provided is the use of an antigen-binding molecule, polypeptide, CAR, nucleic acid (or plurality thereof), expression vector (or plurality thereof), cell or composition described herein in the manufacture of a medicament for treating or preventing a disease or condition described herein. Also provided is a method of treating or preventing a disease or condition described herein, comprising administering to a subject a therapeutically or prophylactically effective amount of an antigen-binding molecule, polypeptide, CAR, nucleic acid (or plurality thereof), expression vector (or plurality thereof), cell or composition described herein.

The methods may be effective to reduce the development or progression of a disease/condition, alleviation of the symptoms of a disease/condition or reduction in the pathology of a disease/condition. The methods may be effective to prevent progression of the disease/condition, e.g. to prevent worsening of, or to slow the rate of development of, the disease/condition. In some embodiments, the methods may lead to an improvement in the disease/condition, e.g. a reduction in the symptoms of the disease/condition or reduction in some other correlate of the severity/activity of the disease/condition. In some embodiments, the methods may prevent development of the disease/condition a later stage (e.g. a chronic stage or metastasis).

It will be appreciated that the articles of the present disclosure may be used for the treatment/prevention of any disease/condition that would derive therapeutic or prophylactic benefit from a reduction in the level/activity of B7-H3, or a reduction in the number or activity of cells comprising/expressing B7-H3.

For example, the disease/condition may be a disease/condition in which B7-H3, or cells comprising/expressing B7-H3 are pathologically-implicated, e.g. a disease/condition in which an increased level/activity of B7-H3, or an increase in the number/pro portion of cells comprising/expressing B7-H3 is positively associated with the onset, development or progression of the disease/condition, and/or severity of one or more symptoms of the disease/condition. In some embodiments, an increased level/activity of B7-H3, or an increase in the number/proportion of cells comprising/expressing B7-H3 may be a risk factor for the onset, development or progression of the disease/condition.

In some embodiments, the disease/condition to be treated/prevented in accordance with the present disclosure is a disease/condition characterised by an increase in the level of expression or activity of B7-H3, e.g. as compared to the level of expression/activity in the absence of the disease/condition.

In some embodiments, the disease/condition to be treated/prevented is a disease/condition characterised by an increase in the number/proportion/activity of cells expressing B7-H3, e.g. as compared to the level/number/proportion/activity in the absence of the disease/condition (e.g. in a healthy subject, or in equivalent non-diseased tissue). Where the disease/condition is a cancer, the level of expression or activity of B7-H3 may be greater than the level of expression or activity of B7- H3 in equivalent non-cancerous cells/non-tumor tissue. A cancer/cell thereof may comprise one or more mutations (e.g. relative to equivalent non-cancerous cells/non-tumor tissue) causing upregulation of expression or activity of B7-H3.

Treatment in accordance with the methods of the present disclosure may achieve one or more of the following in a subject (compared to an equivalent untreated subject, or subject treated with an appropriate control): a reduction in the level of B7-H3; a reduction in the activity of B7-H3; and/or a reduction in the number/proportion of cells comprising/expressing B7-H3.

In aspects and embodiments according to the present disclosure, cells (particularly immune cells, more particularly T cells) comprising/expressing a CAR according to the present disclosure are provided for therapeutic and prophylactic use. It will be appreciated that the methods generally comprise administering a population of immune cells expressing a CAR according to the present disclosure to a subject. In some embodiments, immune cells expressing a CAR according to the present disclosure may be administered in the form of a pharmaceutical composition comprising such cells.

In particular, use of immune cells expressing a CAR according to the present disclosure in methods to treat/prevent diseases/conditions by adoptive cell transfer (ACT) is contemplated.

Adoptive cell transfer generally refers to a process by which cells (e.g. immune cells) are obtained from a subject, typically by drawing a blood sample from which the cells are isolated. The cells are then typically modified and/or expanded, and then administered either to the same subject (in the case of adoptive transfer of autologous/autogeneic cells) or to a different subject (in the case of adoptive transfer of allogeneic cells). The treatment is typically aimed at providing a population of cells with certain desired characteristics to a subject, or increasing the frequency of such cells with such characteristics in that subject. Adoptive transfer may be performed with the aim of introducing a cell or population of cells into a subject, and/or increasing the frequency of a cell or population of cells in a subject.

Adoptive transfer of immune cells is described, for example, in Kalos and June (2013), Immunity 39(1): 49-60, and Davis et al. (2015), Cancer J. 21 (6): 486-491 , both of which are hereby incorporated by reference in their entirety. The skilled person is able to determine appropriate reagents and procedures for adoptive transfer of cells according to the present disclosure, for example by reference to Dai et al., 2016 J Nat Cancer Inst 108(7): djv439, which is incorporated by reference in its entirety.

The immune cells expressing a CAR according to the present disclosure may be employed in the treatment/prevention of diseases/conditions by allotransplantation or autotransplantation. As used herein, ‘allotransplantation’ refers to the transplantation to a recipient subject of cells, tissues or organs which are genetically non-identical to the recipient subject. The cells, tissues or organs may be from, or may be derived from, cells, tissues or organs of a donor subject that is genetically non-identical to the recipient subject. Allotransplantation is distinct from autotransplantation, which refers to the transplantation of cells, tissues or organs which are from/derived from a donor subject genetically identical to the recipient subject (/.e. autologous material). It will be appreciated that adoptive transfer of allogeneic immune cells is a form of allotransplantation, and that adoptive transfer of autologous immune cells is a form of autotransplantation.

The present disclosure provides methods comprising administering immune cells comprising/expressing a CAR according to the present disclosure, or immune cells comprising/expressing nucleic acid encoding a CAR according to the present disclosure, to a subject.

In some embodiments, the methods comprise modifying an immune cell to comprise/express a CAR according to the present disclosure. In some embodiments, the methods comprise modifying an immune cell specific for a virus to comprise/express nucleic acid encoding a CAR according to the present disclosure.

In some embodiments, the methods comprise:

(a) modifying an immune cell to express or comprise a CAR according to the present disclosure, or to express or comprise nucleic acid encoding a CAR according to the present disclosure, and

(b) administering the immune cell specific for a virus modified to express or comprise a CAR according to the present disclosure, or modified to express or comprise a nucleic acid encoding a CAR according to the present disclosure, to a subject.

In some embodiments, the methods comprise:

(a) isolating or obtaining immune cells;

(b) modifying an immune cell to express or comprise a CAR according to the present disclosure, or to express or comprise nucleic acid encoding a CAR according to the present disclosure, and

(c) administering the immune cell modified to express or comprise a CAR according to the present disclosure, or modified to express or comprise a nucleic acid encoding a CAR according to the present disclosure, to a subject.

In some embodiments, the methods comprise:

(a) isolating immune cells (e.g. PBMCs) from a subject; (b) generating/expanding a population of immune cells specific for a virus;

(c) modifying an immune cell specific for a virus to express or comprise a CAR according to the present disclosure, or to express or comprise nucleic acid encoding a CAR according to the present disclosure, and

(d) administering the immune cell specific for a virus modified to express or comprise a CAR according to the present disclosure, or modified to express or comprise a nucleic acid encoding a CAR according to the present disclosure, to a subject.

In some embodiments, the methods comprise administering to a subject an EBV-specific immune cell modified to express or comprise a B7-H3-specific CAR according to the present disclosure, or modified to express or comprise a nucleic acid encoding a B7-H3-specific CAR according to the present disclosure.

In some embodiments, the subject from which the immune cells (e.g. PBMCs) are isolated is the same subject to which cells are administered (/.e., adoptive transfer may be of autologous/autogeneic cells). In some embodiments, the subject from which the immune cells (e.g. PBMCs) are isolated is a different subject to the subject to which cells are administered (/.e., adoptive transfer may be of allogeneic cells).

In some embodiments the methods may comprise one or more of: obtaining a blood sample from a subject; isolating immune cells (e.g. PBMCs) from a blood sample which has been obtained from a subject; generating/expanding a population of immune cells; culturing the immune cells in in vitro or ex vivo cell culture; modifying an immune cell to express or comprise a CAR according to the present disclosure, or to express or comprise a nucleic acid encoding a CAR according to the present disclosure (e.g. by transduction with a viral vector encoding such CAR, or a viral vector comprising such nucleic acid); culturing immune cells expressing/comprising a CAR according to the present disclosure, or expressing/comprising a nucleic acid encoding a CAR according to the present disclosure in in vitro or ex vivo cell culture; collecting/isolating immune cells expressing/comprising a CAR according to the present disclosure, or expressing/comprising a nucleic acid encoding a CAR according to the present disclosure; formulating immune cells expressing/comprising a CAR according to the present disclosure, or a nucleic acid encoding a CAR according to the present disclosure to a pharmaceutical composition, e.g. by mixing the cells with a pharmaceutically acceptable adjuvant, diluent, or carrier; administering immune cells expressing/comprising a CAR according to the present disclosure, or expressing/comprising a nucleic acid encoding a CAR according to the present disclosure, or a pharmaceutical composition comprising such cells, to a subject.

In some embodiments, the methods may additionally comprise treating the cells or subject to induce/enhance expression of CAR and/or to induce/enhance proliferation or survival of virusspecific immune cells comprising/expressing the CAR.

Cancer

In some embodiments, the disease to be treated/prevented in accordance with the present disclosure is a cancer.

B7-H3 expression and B7-H3-mediated signalling is implicated in the pathogenesis of a wide variety of cancers. B7-H3 promotes cancer cell migration and invasion, and therefore promotes metastasis. B7-H3 expression and signalling, and its role in disease is reviewed e.g. in Wu-Tong Zhou and Wei- Lin Jin, Front. Immunol. (2021) 12:701006, Dong et al., Front Oncol. (2018) 8:264 and Yang et al., Int J Biol Sci. (2020) 16(11 ) :1767-1773, all of which are hereby incorporated by reference in their entirety.

Cancer may refer to any unwanted cell proliferation (or any disease manifesting itself by unwanted cell proliferation), neoplasm or tumor. The cancer may be benign or malignant and may be primary or secondary (metastatic). A neoplasm or tumor may be any abnormal growth or proliferation of cells and may be located in any tissue. A cancer may be or comprises a solid cancer (e.g. a tumor), or may be a hematological cancer. The cancer may be of tissues/cells derived from e.g. the adrenal gland, adrenal medulla, anus, appendix, bladder, blood, bone, bone marrow, brain, breast, cecum, central nervous system (including or excluding the brain) cerebellum, cervix, colon, duodenum, endometrium, epithelial cells (e.g. renal epithelia), gallbladder, oesophagus, glial cells, heart, ileum, jejunum, kidney, lacrimal glad, larynx, liver, lung, lymph, lymph node, lymphoblast, maxilla, mediastinum, mesentery, myometrium, nasopharynx, omentum, oral cavity, ovary, pancreas, parotid gland, peripheral nervous system, peritoneum, pleura, prostate, salivary gland, sigmoid colon, skin, small intestine, soft tissues, spleen, stomach, testis, thymus, thyroid gland, tongue, tonsil, trachea, uterus, vulva, and/or white blood cells.

Tumors may be nervous or non-nervous system tumors. Nervous system tumors may originate either in the central or peripheral nervous system, e.g. glioma, medulloblastoma, meningioma, neurofibroma, ependymoma, Schwannoma, neurofibrosarcoma, astrocytoma and oligodendroglioma. Non-nervous system cancers/tumors may originate in any other non-nervous tissue, examples include melanoma, mesothelioma, lymphoma, myeloma, leukemia, Non-Hodgkin’s lymphoma (NHL), Hodgkin’s lymphoma, chronic myelogenous leukemia (CML), acute myeloid leukemia (AML), myelodysplastic syndrome (MDS), cutaneous T cell lymphoma (CTCL), chronic lymphocytic leukemia (CLL), hepatoma, epidermoid carcinoma, prostate carcinoma, breast cancer, lung cancer , colon cancer, ovarian cancer, pancreatic cancer, thymic carcinoma, NSCLC, hematologic cancer and sarcoma.

In some embodiments the cancer is a cancer in which B7-H3 is pathologically implicated. That is, in some embodiments the cancer is a cancer which is caused or exacerbated by the expression of B7- H3, a cancer for which expression of B7-H3 is a risk factor and/or a cancer for which expression of B7-H3 is positively associated with onset, development, progression, severity or metastasis of the cancer. The cancer may be characterised by expression of B7-H3, e.g. the cancer may comprise cells expressing B7-H3. Such cancers may be referred to as being positive for B7-H3. A cancer which is ‘positive’ for the B7-H3 may be a cancer comprising cells expressing B7-H3 (e.g. at the cell surface). A cancer which is ‘positive’ for B7-H3 may overexpress B7-H3.

Cancers in which B7-H3 is pathologically-implicated are described e.g. in Yang et al., Int J Biol Sci. 2020; 16(11): 1767-1773 and Dong et al., Front Oncol. (2018) 8:264, and include lung cancer (e.g. non-small-cell lung cancer), skin cancer (e.g. cutaneous squamous cell carcinoma, melanoma), pancreatic cancer, liver cancer (e.g. hepatocellular carcinoma, intrahepatic cholangiocarcinoma), colorectal cancer (e.g. colorectal carcinoma), kidney cancer (e.g. clear cell renal carcinoma, Wilms' tumor), prostate cancer, ovarian cancer, cervical cancer, endometrial cancer, germ cell tumor, gastric cancer, breast cancer (e.g. triple-negative breast cancer), head and neck cancer (e.g. head and neck squamous cell carcinoma), oral cavity cancer (e.g. oral squamous cell carcinoma), esophageal cancer, bladder cancer, urothelial cancer, brain cancer (e.g. medulloblastoma (e.g. ependymoma medulloblastoma), glioma (e.g. diffuse intrinsic pontine glioma, diffuse midline glioma), choroid plexus carcinoma, pineoblastoma), neuroblastoma, CNS tumor (e.g. primitive neuroectodermal tumor, atypical teratoid/rhabdoid tumor), brain stem glioma, sarcoma (e.g. rhabdomyosarcoma, osteosarcoma, Ewing sarcoma), peritoneal cancer, desmoplastic small round cell tumor and mesothelioma.

In some embodiments, the cancer to be treated/prevented in accordance with the present disclosure is selected from the group consisting of: a B7-H3-positive cancer, lung cancer, non-small-cell lung cancer, small cell lung cancer, lung adenocarcinoma, squamous lung cell carcinoma, skin cancer, cutaneous squamous cell carcinoma, melanoma, pancreatic cancer, liver cancer, hepatocellular carcinoma, cholangiocarcinoma, intrahepatic cholangiocarcinoma, colorectal cancer, colorectal carcinoma, colon cancer, colon carcinoma, kidney cancer, clear cell renal carcinoma, Wilms' tumor, prostate cancer, ovarian cancer, ovarian carcinoma, cervical cancer, endometrial cancer, germ cell tumor, gastric cancer, gastric carcinoma, gastric adenocarcinoma, gastrointestinal adenocarcinoma, breast cancer, triple-negative breast cancer, head and neck cancer, head and neck squamous cell carcinoma, oral cavity cancer, oral squamous cell carcinoma, laryngeal cancer, oropharyngeal cancer, oropharyngeal carcinoma, nasopharyngeal carcinoma, esophageal cancer, bladder cancer, urothelial cancer, brain cancer, medulloblastoma, ependymoma medulloblastoma, glioma, diffuse intrinsic pontine glioma, diffuse midline glioma, choroid plexus carcinoma, pineoblastoma, neuroblastoma, CNS tumor, primitive neuroectodermal tumor, atypical teratoid/rhabdoid tumor, brain stem glioma, sarcoma, rhabdomyosarcoma, osteosarcoma, Ewing sarcoma, peritoneal cancer, desmoplastic small round cell tumor and mesothelioma.

In some embodiments, the cancer to be treated/prevented in accordance with the present disclosure is selected from the group consisting of: a B7-H3-positive cancer, lung cancer, non-small-cell lung cancer, skin cancer, cutaneous squamous cell carcinoma, melanoma, pancreatic cancer, liver cancer, hepatocellular carcinoma, intrahepatic cholangiocarcinoma, colorectal cancer, colorectal carcinoma, kidney cancer, clear cell renal carcinoma, Wilms' tumor, prostate cancer, ovarian cancer, cervical cancer, endometrial cancer, germ cell tumor, gastric cancer, breast cancer, triple-negative breast cancer, head and neck cancer, head and neck squamous cell carcinoma, oral squamous cell carcinoma, esophageal cancer, bladder cancer, urothelial cancer, brain cancer, medulloblastoma, ependymoma medulloblastoma, glioma, diffuse intrinsic pontine glioma, diffuse midline glioma, choroid plexus carcinoma, pineoblastoma, neuroblastoma, CNS tumor, primitive neuroectodermal tumor, atypical teratoid/rhabdoid tumor, brain stem glioma, sarcoma, rhabdomyosarcoma, osteosarcoma, Ewing sarcoma, peritoneal cancer, desmoplastic small round cell tumor and mesothelioma.

In some embodiments, the cancer may be a relapsed cancer. As used herein, a ‘relapsed’ cancer refers to a cancer which responded to a treatment (e.g. a first line therapy for the cancer), but which has subsequently re-emerged/progressed, e.g. after a period of remission. For example, a relapsed cancer may be a cancer whose growth/progression was inhibited by a treatment (e.g. a first line therapy for the cancer), and which has subsequently grown/progressed.

In some embodiments, the cancer may be a refractory cancer. As used herein, a ‘refractory’ cancer refers to a cancer which has not responded to a treatment (e.g. a first line therapy for the cancer). For example, a refractory cancer may be a cancer whose growth/progression was not inhibited by a treatment (e.g. a first line therapy for the cancer). In some embodiments a refractory cancer may be a cancer for which a subject receiving treatment for the cancer did not display a partial or complete response to the treatment.

Treatment of a cancer in accordance with the methods of the present disclosure achieves one or more of the following treatment effects: reduces the number of cancer cells in the subject, reduces the size of a cancerous tumor/lesion in the subject, inhibits (e.g. prevents or slows) growth of cancer cells in the subject, inhibits (e.g. prevents or slows) growth of a cancerous tumor/lesion in the subject, inhibits (e.g. prevents or slows) the development/progression of a cancer (e.g. to a later stage, or metastasis), reduces the severity of symptoms of a cancer in the subject, increases survival of the subject (e.g. progression free survival or overall survival), reduces a correlate of the number or activity of cancer cells in the subject, and/or reduces cancer burden in the subject. Subjects may be evaluated in accordance with the Revised Criteria for Response Assessment: The Lugano Classification (described e.g. in Cheson et al., J Clin Oncol (2014) 32: 3059-3068, incorporated by reference hereinabove) in order to determine their response to treatment. In some embodiments, treatment of a subject in accordance with the methods of the present disclosure achieves one of the following: complete response, partial response, or stable disease.

Administration

Administration of the articles of the present disclosure is preferably in a ‘therapeutically-effective’ or ‘prophylactically-effective’ amount, this being sufficient to show therapeutic or prophylactic benefit to the subject. The actual amount administered, and rate and time-course of administration, will depend on the nature and severity of the disease/condition and the particular article administered. Prescription of treatment, e.g. decisions on dosage etc., is within the responsibility of general practitioners and other medical doctors, and typically takes account of the disease/disorderto be treated, the condition of the individual subject, the site of delivery, the method of administration and other factors known to practitioners. Examples of the techniques and protocols mentioned above can be found in Remington’s ‘The Science and Practice of Pharmacy’ (ed. A. Adejare), 23rd Edition (2020), Academic Press.

Administration of the articles of the present disclosure may be topical, parenteral, systemic, intracavitary, intravenous, intra-arterial, intramuscular, intrathecal, intraocular, intravitreal, intraconjunctival, subretinal, suprachoroidal, subcutaneous, intradermal, intrathecal, oral, nasal or transdermal. Administration may be by injection or infusion. Administration of the articles of the present disclosure may be intratumoral.

In some aspects and embodiments in accordance with the present disclosure there may be targeted delivery of articles of the present disclosure, i.e. wherein the concentration of the relevant agent in the subject is increased in some parts of the body relative to other parts of the body. In some embodiments, the methods comprise intravenous, intra-arterial, intramuscular or subcutaneous administration and wherein the relevant article is formulated in a targeted agent delivery system. Suitable targeted delivery systems include, for example, nanoparticles, liposomes, micelles, beads, polymers, metal particles, dendrimers, antibodies, aptamers, nanotubes or micro-sized silica rods. Such systems may comprise a magnetic element to direct the agent to the desired organ or tissue. Suitable nanocarriers and delivery systems will be apparent to one skilled in the art.

In some cases, the articles of the present disclosure are formulated for targeted delivery to specific cells, a tissue, an organ and/or a tumor. Further intervention

Administration may be alone or in combination with other treatments, either simultaneously or sequentially dependent upon the disease/condition to be treated. The antigen-binding molecule, CAR, cell or composition described herein and another prophylactic/therapeutic agent may be administered simultaneously or sequentially.

In some embodiments, the methods comprise additional therapeutic or prophylactic intervention, e.g. for the treatment/prevention of a cancer. In some embodiments, the therapeutic or prophylactic intervention is selected from chemotherapy, immunotherapy, radiotherapy, surgery, vaccination and/or hormone therapy. In some embodiments, the therapeutic or prophylactic intervention comprises leukapheresis. In some embodiments, the therapeutic or prophylactic intervention comprises a stem cell transplant.

Simultaneous administration refers to administration of the antigen-binding molecule, polypeptide, CAR, nucleic acid (or plurality thereof), expression vector (or plurality thereof), cell or composition and therapeutic agent together, for example as a pharmaceutical composition containing both agents (combined preparation), or immediately after each other and optionally via the same route of administration, e.g. to the same artery, vein or other blood vessel. Sequential administration refers to administration of one of the antigen-binding molecule/composition or therapeutic agent followed after a given time interval by separate administration of the other agent. It is not required that the two agents are administered by the same route, although this is the case in some embodiments. The time interval may be any time interval.

In some embodiments, treatment of cancer further comprises chemotherapy and/or radiotherapy. Chemotherapy and radiotherapy respectively refer to treatment of a cancer with a drug or with ionising radiation (e.g. radiotherapy using X-rays or y-rays). The drug may be a chemical entity, e.g. small molecule pharmaceutical, antibiotic, DNA intercalator, protein inhibitor (e.g. kinase inhibitor), or a biological agent, e.g. antibody, antibody fragment, aptamer, nucleic acid (e.g. DNA, RNA), peptide, polypeptide, or protein. The drug may be formulated as a pharmaceutical composition or medicament. The formulation may comprise one or more drugs (e.g. one or more active agents) together with one or more pharmaceutically acceptable diluents, excipients or carriers.

Chemotherapy may involve administration of more than one drug. A drug may be administered alone or in combination with other treatments, either simultaneously or sequentially dependent upon the condition to be treated.

The chemotherapy may be administered by one or more routes of administration, e.g. parenteral, intravenous injection, oral, subcutaneous, intradermal or intratumoral. The chemotherapy may be administered according to a treatment regime. The treatment regime may be a pre-determined timetable, plan, scheme or schedule of chemotherapy administration which may be prepared by a physician or medical practitioner and may be tailored to suit the patient requiring treatment. The treatment regime may indicate one or more of: the type of chemotherapy to administer to the patient; the dose of each drug or radiation; the time interval between administrations; the length of each treatment; the number and nature of any treatment holidays, if any etc. For a co-therapy a single treatment regime may be provided which indicates how each drug is to be administered.

Chemotherapeutic drugs may be selected from: Abemaciclib, Abiraterone Acetate, Abitrexate (Methotrexate), Abraxane (Paclitaxel Albumin-stabilized Nanoparticle Formulation), ABVD, ABVE, ABVE-PC, AC, Acalabrutinib, AC-T, Adcetris (Brentuximab Vedotin), ADE, Ado-Trastuzumab Emtansine, Adriamycin (Doxorubicin Hydrochloride), Afatinib Dimaleate, Afinitor (Everolimus), Akynzeo (Netupitant and Palonosetron Hydrochloride), Aldara (Imiquimod), Aldesleukin, Alecensa (Alectinib), Alectinib, Alemtuzumab, Alimta (Pemetrexed Disodium), Aliqopa (Copanlisib Hydrochloride), Alkeran for Injection (Melphalan Hydrochloride), Alkeran Tablets (Melphalan), Aloxi (Palonosetron Hydrochloride), Alunbrig (Brigatinib), Ambochlorin (Chlorambucil), Amboclorin (Chlorambucil), Amifostine, Aminolevulinic Acid, Anastrozole, Aprepitant, Aredia (Pamidronate Disodium), Arimidex (Anastrozole), Aromasin (Exemestane), Arranon (Nelarabine), Arsenic Trioxide, Arzerra (Ofatumumab), Asparaginase Erwinia chrysanthemi, Atezolizumab, Avastin (Bevacizumab), Avelumab, Axicabtagene Ciloleucel, Axitinib, Azacitidine, Bavencio (Avelumab), BEACOPP, Becenum (Carmustine), Beleodaq (Belinostat), Belinostat, Bendamustine Hydrochloride, BEP, Besponsa (Inotuzumab Ozogamicin) , Bevacizumab, Bexarotene, Bexxar (Tositumomab and Iodine I 131 Tositumomab), Bicalutamide, BiCNU (Carmustine), Bleomycin, Blinatumomab, Blincyto (Blinatumomab), Bortezomib, Bosulif (Bosutinib), Bosutinib, Brentuximab Vedotin, Brigatinib, BuMel, Busulfan, Busulfex (Busulfan), Cabazitaxel, Cabometyx (Cabozantinib-S-Malate), Cabozantinib-S- Malate, CAF, Calquence (Acalabrutinib), Campath (Alemtuzumab), Camptosar (Irinotecan Hydrochloride), Capecitabine, CAPOX, Carac (Fluorouracil-Topical), Carboplatin, CARBOPLATIN- TAXOL, Carfilzomib, Carmubris (Carmustine), Carmustine, Carmustine Implant, Casodex (Bicalutamide), CEM, Ceritinib, Cerubidine (Daunorubicin Hydrochloride), Cervarix (Recombinant HPV Bivalent Vaccine), Cetuximab, CEV, Chlorambucil, CHLORAMBUCIL-PREDNISONE, CHOP, Cisplatin, Cladribine, Clafen (Cyclophosphamide), Clofarabine, Clofarex (Clofarabine), Clolar (Clofarabine), CMF, Cobimetinib, Cometriq (Cabozantinib-S-Malate), Copanlisib Hydrochloride, COPDAC, COPP, COPP-ABV, Cosmegen (Dactinomycin), Cotellic (Cobimetinib), Crizotinib, CVP, Cyclophosphamide, Cyfos (Ifosfamide), Cyramza (Ramucirumab), Cytarabine, Cytarabine Liposome, Cytosar-U (Cytarabine), Cytoxan (Cyclophosphamide), Dabrafenib, Dacarbazine, Dacogen (Decitabine), Dactinomycin, Daratumumab, Darzalex (Daratumumab), Dasatinib, Daunorubicin Hydrochloride, Daunorubicin Hydrochloride and Cytarabine Liposome, Decitabine, Defibrotide Sodium, Defitelio (Defibrotide Sodium), Degarelix, Denileukin Diftitox, Denosumab, DepoCyt (Cytarabine Liposome), Dexamethasone, Dexrazoxane Hydrochloride, Dinutuximab, Docetaxel, Doxil (Doxorubicin Hydrochloride Liposome), Doxorubicin Hydrochloride, Doxorubicin Hydrochloride Liposome, Dox-SL (Doxorubicin Hydrochloride Liposome), DTIC-Dome (Dacarbazine), Durvalumab, Efudex (Fluorouracil-Topical), Elitek (Rasburicase), Ellence (Epirubicin Hydrochloride), Elotuzumab, Eloxatin (Oxaliplatin), Eltrombopag Olamine, Emend (Aprepitant), Empliciti (Elotuzumab), Enasidenib Mesylate, Enzalutamide, Epirubicin Hydrochloride, EPOCH, Erbitux (Cetuximab), Eribulin Mesylate, Erivedge (Vismodegib), Erlotinib Hydrochloride, Erwinaze (Asparaginase Erwinia chrysanthemi), Ethyol (Amifostine), Etopophos (Etoposide Phosphate), Etoposide, Etoposide Phosphate, Evacet (Doxorubicin Hydrochloride Liposome), Everolimus, Evista (Raloxifene Hydrochloride), Evomela (Melphalan Hydrochloride), Exemestane, 5-FU (Fluorouracil Injection), 5-FU (Fluorouracil-Topical), Fareston (Toremifene), Farydak (Panobinostat), Faslodex (Fulvestrant), FEC, Femara (Letrozole), Filgrastim, Fludara (Fludarabine Phosphate), Fludarabine Phosphate, Fluoroplex (Fluorouracil-Topical), Fluorouracil Injection, Fluorouracil-Topical, Flutamide, Folex (Methotrexate), Folex PFS (Methotrexate), FOLFIRI, FOLFIRI-BEVACIZUMAB, FOLFIRI-CETUXIMAB, FOLFIRINOX, FOLFOX, Folotyn (Pralatrexate), FU-LV, Fulvestrant, Gardasil (Recombinant HPV Quadrivalent Vaccine), Gardasil 9 (Recombinant HPV Nonavalent Vaccine), Gazyva (Obinutuzumab), Gefitinib, Gemcitabine Hydrochloride, GEMCITABINE-CISPLATIN, GEMCITABINE-OXALIPLATIN, Gemtuzumab Ozogamicin, Gemzar (Gemcitabine Hydrochloride), Gilotrif (Afatinib Dimaleate), Gleevec (Imatinib Mesylate), Gliadel (Carmustine Implant), Gliadel wafer (Carmustine Implant), Glucarpidase, Goserelin Acetate, Halaven (Eribulin Mesylate), Hemangeol (Propranolol Hydrochloride), Herceptin (Trastuzumab), HPV Bivalent Vaccine, Recombinant, HPV Nonavalent Vaccine, Recombinant, HPV Quadrivalent Vaccine, Recombinant, Hycamtin (Topotecan Hydrochloride), Hydrea (Hydroxyurea), Hydroxyurea, Hyper-CVAD, Ibrance (Palbociclib), Ibritumomab Tiuxetan, Ibrutinib, ICE, Iclusig (Ponatinib Hydrochloride), Idamycin (Idarubicin Hydrochloride), Idarubicin Hydrochloride, Idelalisib, Idhifa (Enasidenib Mesylate), Ifex (Ifosfamide), Ifosfamide, Ifosfamidum (Ifosfamide), IL-2 (Aldesleukin), Imatinib Mesylate, Imbruvica (Ibrutinib), Imfinzi (Durvalumab), Imiquimod, Imlygic (Talimogene Laherparepvec), Inlyta (Axitinib), Inotuzumab Ozogamicin, Interferon Alfa-2b, Recombinant, lnterleukin-2 (Aldesleukin), Intron A (Recombinant Interferon Alfa-2b), Iodine I 131 Tositumomab and Tositumomab, Ipilimumab, Iressa (Gefitinib), Irinotecan Hydrochloride, Irinotecan Hydrochloride Liposome, Istodax (Romidepsin), Ixabepilone, Ixazomib Citrate, Ixempra (Ixabepilone), Jakafi (Ruxolitinib Phosphate), JEB, Jevtana (Cabazitaxel), Kadcyla (Ado-Trastuzumab Emtansine), Keoxifene (Raloxifene Hydrochloride), Kepivance (Palifermin), Keytruda (Pembrolizumab), Kisqali (Ribociclib), Kymriah (Tisagenlecleucel), Kyprolis (Carfilzomib), Lanreotide Acetate, Lapatinib Ditosylate, Lartruvo (Olaratumab), Lenalidomide, Lenvatinib Mesylate, Lenvima (Lenvatinib Mesylate), Letrozole, Leucovorin Calcium, Leukeran (Chlorambucil), Leuprolide Acetate, Leustatin (Cladribine), Levulan (Aminolevulinic Acid), Linfolizin (Chlorambucil), LipoDox (Doxorubicin Hydrochloride Liposome), Lomustine, Lonsurf (Trifluridine and Tipiracil Hydrochloride), Lupron (Leuprolide Acetate), Lupron Depot (Leuprolide Acetate), Lupron Depot-Ped (Leuprolide Acetate), Lynparza (Olaparib), Marqibo (Vincristine Sulfate Liposome), Matulane (Procarbazine Hydrochloride), Mechlorethamine Hydrochloride, Megestrol Acetate, Mekinist (Trametinib), Melphalan, Melphalan Hydrochloride, Mercaptopurine, Mesna, Mesnex (Mesna), Methazolastone (Temozolomide), Methotrexate, Methotrexate LPF (Methotrexate), Methylnaltrexone Bromide, Mexate (Methotrexate), Mexate-AQ (Methotrexate), Midostaurin, Mitomycin C, Mitoxantrone Hydrochloride, Mitozytrex (Mitomycin C), MOPP, Mozobil (Plerixafor), Mustargen (Mechlorethamine Hydrochloride), Mutamycin (Mitomycin C), Myleran (Busulfan), Mylosar (Azacitidine), Mylotarg (Gemtuzumab Ozogamicin), Nanoparticle Paclitaxel (Paclitaxel Albumin-stabilized Nanoparticle Formulation), Navelbine (Vinorelbine Tartrate), Necitumumab, Nelarabine, Neosar (Cyclophosphamide), Neratinib Maleate, Nerlynx (Neratinib Maleate), Netupitant and Palonosetron Hydrochloride, Neulasta (Pegfilgrastim), Neupogen (Filgrastim), Nexavar (Sorafenib Tosylate), Nilandron (Nilutamide), Nilotinib, Nilutamide, Ninlaro (Ixazomib Citrate), Niraparib Tosylate Monohydrate, Nivolumab, Nolvadex (Tamoxifen Citrate), Nplate (Romiplostim), Obinutuzumab, Odomzo (Sonidegib), OEPA, Ofatumumab, OFF, Olaparib, Olaratumab, Omacetaxine Mepesuccinate, Oncaspar (Pegaspargase), Ondansetron Hydrochloride, Onivyde (Irinotecan Hydrochloride Liposome), Ontak (Denileukin Diftitox), Opdivo (Nivolumab), OPPA, Osimertinib, Oxaliplatin, Paclitaxel, Paclitaxel Albumin-stabilized Nanoparticle Formulation, PAD, Palbociclib, Palifermin, Palonosetron Hydrochloride, Palonosetron Hydrochloride and Netupitant, Pamidronate Disodium, Panitumumab, Panobinostat, Paraplat (Carboplatin), Paraplatin (Carboplatin), Pazopanib Hydrochloride, PCV, PEB, Pegaspargase, Peg filgrastim, Peginterferon Alfa-2b, PEG-lntron (Peginterferon Alfa-2b), Pembrolizumab, Pemetrexed Disodium, Perjeta (Pertuzumab), Pertuzumab, Platinol (Cisplatin), Platinol-AQ (Cisplatin), Plerixafor, Pomalidomide, Pomalyst (Pomalidomide), Ponatinib Hydrochloride, Portrazza (Necitumumab), Pralatrexate, Prednisone, Procarbazine Hydrochloride, Proleukin (Aldesleukin), Prolia (Denosumab), Promacta (Eltrombopag Olamine), Propranolol Hydrochloride, Provenge (Sipuleucel-T), Purinethol (Mercaptopurine), Purixan (Mercaptopurine), Radium 223 Dichloride, Raloxifene Hydrochloride, Ramucirumab, Rasburicase, R-CHOP, R-CVP, Recombinant Human Papillomavirus (HPV) Bivalent Vaccine, Recombinant Human Papillomavirus (HPV) Nonavalent Vaccine, Recombinant Human Papillomavirus (HPV) Quadrivalent Vaccine, Recombinant Interferon Alfa-2b, Regorafenib, Relistor (Methylnaltrexone Bromide), R-EPOCH, Revlimid (Lenalidomide), Rheumatrex (Methotrexate), Ribociclib, R-ICE, Rituxan (Rituximab), Rituxan Hycela (Rituximab and Hyaluronidase Human), Rituximab, Rituximab and Hyaluronidase Human, Rolapitant Hydrochloride, Romidepsin, Romiplostim, Rubidomycin (Daunorubicin Hydrochloride), Rubraca (Rucaparib Camsylate), Rucaparib Camsylate, Ruxolitinib Phosphate, Rydapt (Midostaurin), Sclerosol Intrapleural Aerosol (Talc), Siltuximab, Sipuleucel-T, Somatuline Depot (Lanreotide Acetate), Sonidegib, Sorafenib Tosylate, Sprycel (Dasatinib), STANFORD V, Sterile Talc Powder (Talc), Steritalc (Talc), Stivarga (Regorafenib), Sunitinib Malate, Sutent (Sunitinib Malate), Sylatron (Peginterferon Alfa-2b), Sylvant (Siltuximab), Synribo (Omacetaxine Mepesuccinate), Tabloid (Thioguanine), TAC, Tafinlar (Dabrafenib), Tagrisso (Osimertinib), Talc, Talimogene Laherparepvec, Tamoxifen Citrate, Tarabine PFS (Cytarabine), Tarceva (Erlotinib Hydrochloride), Targretin (Bexarotene), Tasigna (Nilotinib), Taxol (Paclitaxel), Taxotere (Docetaxel), Tecentriq (Atezolizumab), Temodar (Temozolomide), Temozolomide, Temsirolimus, Thalidomide, Thalomid (Thalidomide), Thioguanine, Thiotepa, Tisagenlecleucel, Tolak (Fluorouracil-Topical), Topotecan Hydrochloride, Toremifene, Torisel (Temsirolimus), Tositumomab and Iodine I 131 Tositumomab, Totect (Dexrazoxane Hydrochloride), TPF, Trabectedin, Trametinib, Trastuzumab, Treanda (Bendamustine Hydrochloride), Trifluridine and Tipiracil Hydrochloride, Trisenox (Arsenic Trioxide), Tykerb (Lapatinib Ditosylate), Unituxin (Dinutuximab), Uridine Triacetate, VAC, Valrubicin, Valstar (Valrubicin), Vandetanib, VAMP, Varubi (Rolapitant Hydrochloride), Vectibix (Panitumumab), VelP, Velban (Vinblastine Sulfate), Velcade (Bortezomib), Velsar (Vinblastine Sulfate), Vemurafenib, Venclexta (Venetoclax), Venetoclax, Verzenio (Abemaciclib), Viadur (Leuprolide Acetate), Vidaza (Azacitidine), Vinblastine Sulfate, Vincasar PFS (Vincristine Sulfate), Vincristine Sulfate, Vincristine Sulfate Liposome, Vinorelbine Tartrate, VIP, Vismodegib, Vistogard (Uridine Triacetate), Voraxaze (Glucarpidase), Vorinostat, Votrient (Pazopanib Hydrochloride), Vyxeos (Daunorubicin Hydrochloride and Cytarabine Liposome), Wellcovorin (Leucovorin Calcium), Xalkori (Crizotinib), Xeloda (Capecitabine), XELIRI, XELOX, Xgeva (Denosumab), Xofigo (Radium 223 Dichloride), Xtandi (Enzalutamide), Yervoy (Ipilimumab), Yescarta (Axicabtagene Ciloleucel), Yondelis (Trabectedin), Zaltrap (Ziv-Aflibercept), Zarxio (Filgrastim), Zejula (Niraparib Tosylate Monohydrate), Zelboraf (Vemurafenib), Zevalin (Ibritumomab Tiuxetan), Zinecard (Dexrazoxane Hydrochloride), Ziv-Aflibercept, Zofran (Ondansetron Hydrochloride), Zoladex (Goserelin Acetate), Zoledronic Acid, Zolinza (Vorinostat), Zometa (Zoledronic Acid), Zydelig (Idelalisib), Zykadia (Ceritinib) and Zytiga (Abiraterone Acetate).

In some embodiments, the treatment may comprise administration of a corticosteroid, e.g. dexamethasone and/or prednisone.

In some embodiments, a subject is administered lymphodepleting chemotherapy prior to administration of immune cells expressing/comprising a CAR described herein (or expressing/comprising nucleic acid encoding such a CAR).

That is, in some embodiments, methods of treating/preventing a disease/condition in accordance with the present disclosure comprise: (i) administering a lymphodepleting chemotherapy to a subject, and (ii) subsequently administering an immune cell expressing/comprising a CAR according to the present disclosure, or expressing/comprising a nucleic acid encoding a CAR according to the present disclosure.

As used herein, “lymphodepleting chemotherapy” refers to treatment with a chemotherapeutic agent which results in depletion of lymphocytes (e.g. T cells, B cells, NK cells, NKT cells or innate lymphoid cell (ILCs), or precursors thereof) within the subject to which the treatment is administered. A “lymphodepleting chemotherapeutic agent” refers to a chemotherapeutic agent which results in depletion of lymphocytes.

Lymphodepleting chemotherapy and its use in methods of treatment by adoptive cell transfer are described e.g. in Klebanoff et al., Trends Immunol. (2005) 26(2) :111 -7 and Muranski et al., Nat Clin Pract Oncol. (2006) (12) :668-81 , both of which are hereby incorporated by reference in their entirety. The aim of lymphodepleting chemotherapy is to deplete the recipient subject’s endogenous lymphocyte population.

In the context of treatment of disease by adoptive transfer of immune cells, lymphodepleting chemotherapy is typically administered prior to adoptive cell transfer, to condition the recipient subject to receive the adoptively transferred cells. Lymphodepleting chemotherapy is thought to promote the persistence and activity of adoptively transferred cells by creating a permissive environment, e.g. through elimination of cells expressing immunosuppressive cytokines, and creating the ‘lymphoid space’ required for expansion and activity of adoptively transferred lymphoid cells.

Chemotherapeutic agents commonly used in lymphodepleting chemotherapy include e.g. fludarabine, cyclophosphamide, bedamustine and pentostatin.

Multiple doses of the antigen-binding molecule, polypeptide, CAR, nucleic acid (or plurality thereof), expression vector (or plurality thereof), cell or composition may be provided. One or more, or each, of the doses may be accompanied by simultaneous or sequential administration of another therapeutic agent.

Multiple doses may be separated by a predetermined time interval, which may be selected to be one of 1 , 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 , 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 , 22, 23, 24, 25, 26, 27, 28, 29, 30, or 31 days, or 1 , 2, 3, 4, 5, or 6 months. By way of example, doses may be given once every 7, 14, 21 or 28 days (plus or minus 3, 2, or 1 days).

In accordance with various aspects of the present disclosure, a method of treating and/or preventing a disease/condition may comprise one or more of the following: reducing the number/proportion of B7-H3-expressing cells; inhibiting tumor growth (e.g. of a B7-H3+ tumor); reducing metastasis of a cancer (e.g. a B7-H3+ cancer); increasing survival of a subject having a cancer (e.g. a B7-H3+ cancer).

Methods of detection

The present disclosure also provides the articles of the present disclosure for use in methods for detecting, localising or imaging B7-H3, or cells expressing B7-H3.

The antigen-binding molecules described herein may be used in methods that involve detecting binding of the antigen-binding molecule to B7-H3. Such methods may involve detection of the bound complex of the antigen-binding molecule and B7-H3. It will be appreciated that the B7-H3 may be B7-H3 expressed by a cell, e.g. in or at the cell surface of a cell expressing B7-H3. As such, a method is provided, comprising contacting a sample containing, or suspected to contain, B7-H3, and detecting the formation of a complex of the antigen-binding molecule and B7-H3. Also provided is a method comprising contacting a sample containing, or suspected to contain, a cell expressing B7-H3, and detecting the formation of a complex of the antigen-binding molecule and a cell expressing B7-H3.

Suitable method formats are well known in the art, including immunoassays such as sandwich assays, e.g. ELISA. The methods may involve labelling the antigen-binding molecule, ortarget(s), or both, with a detectable moiety, e.g. a fluorescent label, phosphorescent label, luminescent label, immuno-detectable label, radiolabel, chemical, nucleic acid or enzymatic label as described herein. Detection techniques are well known to those of skill in the art and can be selected to correspond with the labelling agent.

Methods comprising detecting B7-H3, or cells expressing B7-H3, include methods for diagnosing/prognosing a disease/condition described herein.

Methods of this kind may be performed in vitro on a patient sample, or following processing of a patient sample. Once the sample is collected, the patient is not required to be present for the in vitro method to be performed, and therefore the method may be one which is not practised on the human or animal body. In some embodiments, the method is performed in vivo.

Such methods may involve detecting or quantifying B7-H3 and/or cells expressing B7-H3, e.g. in a patient sample. Where the method comprises quantifying the relevant factor, the method may further comprise comparing the determined amount against a standard or reference value as part of the diagnostic or prognostic evaluation. Other diagnostic/prognostic tests may be used in conjunction with those described herein to enhance the accuracy of the diagnosis or prognosis or to confirm a result obtained by using the tests described herein.

Detection in a sample may be used for the purpose of diagnosis of a disease/condition (e.g. a cancer), predisposition to a disease/condition, or for providing a prognosis (prognosticating) for a disease/condition, e.g. a disease/condition described herein. The diagnosis or prognosis may relate to an existing (previously diagnosed) disease/condition.

A sample may be taken from any tissue or bodily fluid. The sample may comprise or may be derived from: a quantity of blood; a quantity of serum derived from the individual’s blood which may comprise the fluid portion of the blood obtained after removal of the fibrin clot and blood cells; a tissue sample or biopsy; pleural fluid; cerebrospinal fluid (CSF); or cells isolated from said individual. In some embodiments, the sample may be obtained or derived from a tissue or tissues which are affected by the disease/condition (e.g. tissue or tissues in which symptoms of the disease manifest, or which are involved in the pathogenesis of the disease/condition). A subject may be selected for diagnostic/prognostic evaluation based on the presence of symptoms indicative of a disease/condition described herein, or based on the subject being considered to be at risk of developing a disease/condition described herein.

The present disclosure also provides methods for selecting/stratifying a subject for treatment with a B7-H3-targeted agent. In some embodiments a subject is selected for treatment/prevention in accordance with the methods of the present disclosure, or is identified as a subject which would benefit from such treatment/prevention, based on detection/quantification of B7-H3, or cells expressing B7-H3, e.g. in a sample obtained from the individual.

Subjects

A subject in accordance with the various aspects of the present disclosure may be any animal or human. Therapeutic and prophylactic applications may be in human or animals (veterinary use).

The subject to be administered with an article of the present disclosure (e.g. in accordance with therapeutic or prophylactic intervention) may be a subject in need of such intervention. The subject is preferably mammalian, more preferably human. The subject may be a non-human mammal, but is more preferably human. The subject may be male or female. The subject may be a patient.

A subject may have (e.g. may have been diagnosed with) a disease or condition described herein, may be suspected of having such a disease/condition, or may be at risk of developing/contracting such a disease/condition. In embodiments according to the present disclosure, a subject may be selected for treatment according to the methods based on characterisation for one or more markers of such a disease/condition.

In some embodiments, a subject may be selected for therapeutic or prophylactic intervention as described herein based on the detection of cells/tissue expressing B7-H3, or of cells/tissue overexpressing B7-H3, e.g. in a sample obtained from the subject.

A subject may be an allogeneic subject with respect to an intervention in accordance with the present disclosure. A subject to be treated/prevented in accordance with the present disclosure may be genetically non-identical to the subject from which the CAR-expressing immune cells are derived. A subject to be treated/prevented in accordance with the present disclosure may be HLA mismatched with respect to the subject from which the CAR-expressing immune cells are derived. A subject to be treated/prevented in accordance with the present disclosure may be HLA matched with respect to the subject from which the CAR-expressing immune cells are derived.

The subject to which cells are administered in accordance with the present disclosure may be allogeneic/non-autologous with respect to the source from which the cells are/were derived. The subject to which cells are administered may be a different subject to the subject from which cells are/were obtained for the production of the cells to be administered. The subject to which the cells are administered may be genetically non-identical to the subject from which cells are/were obtained for the production of the cells to be administered.

The subject to which cells are administered may comprise MHC/HLA genes encoding MHC/HLA molecules which are non-identical to the MHC/HLA molecules encoded by the MHC/HLA genes of the subject from which cells are/were obtained for the production of the cells to be administered. The subject to which cells are administered may comprise MHC/HLA genes encoding MHC/HLA molecules which are identical to the MHC/HLA molecules encoded by the MHC/HLA genes of the subject from which cells are/were obtained for the production of the cells to be administered.

In some embodiments, the subject to which cells are administered is HLA matched with respect to the subject from which cells are/were obtained for the production of the cells to be administered. In some embodiments, the subject to which cells are administered is a near or complete HLA match with respect to the subject from which cells are/were obtained for the production of the cells to be administered.

In some embodiments, the subject is a >4/8 (/.e. 4/8, 5/8, 6/8, 7/8 or 8/8) match across HLA-A, -B, - C, and -DRB1. In some embodiments, the subject is a >5/10 (/.e. 5/10, 6/10, 7/10, 8/10, 9/10 or 10/10) match across HLA-A, -B, -C, -DRB1 and -DQB1. In some embodiments, the subject is a >6/12 (/.e. 6/12, 7/12, 8/12, 9/12, 10/12, 11/12 or 12/12) match across HLA-A, -B, -C, -DRB1 , -DQB1 and -DPB1 . In some embodiments, the subject is an 8/8 match across HLA-A, -B, -C, and -DRB1 . In some embodiments, the subject is a 10/10 match across HLA-A, -B, -C, -DRB1 and -DQB1. In some embodiments, the subject is a 12/12 match across HLA-A, -B, -C, -DRB1 , -DQB1 and -DPB1 .

Kits

In some aspects of the present disclosure a kit of parts is provided. In some embodiments, the kit may have at least one container having a predetermined quantity of an antigen-binding molecule, polypeptide, CAR, nucleic acid (or plurality thereof), expression vector (or plurality thereof), cell or composition described herein.

In some embodiments, the kit may comprise materials for producing an antigen-binding molecule, polypeptide, CAR, nucleic acid (or plurality thereof), expression vector (or plurality thereof), cell or composition described herein.

The kit may provide the antigen-binding molecule, polypeptide, CAR, nucleic acid (or plurality thereof), expression vector (or plurality thereof), cell or composition together with instructions for administration to a patient in order to treat a specified disease/condition. In some embodiments the kit may further comprise at least one container having a predetermined quantity of another therapeutic agent (e.g. as described herein). In such embodiments, the kit may also comprise a second medicament or pharmaceutical composition such that the two medicaments or pharmaceutical compositions may be administered simultaneously or separately such that they provide a combined treatment for the specific disease or condition.

Kits according to the present disclosure may include instructions for use, e.g. in the form of an instruction booklet or leaflet. The instructions may include a protocol for performing any one or more of the methods described herein.

Sequence identity

As used herein, ‘sequence identity’ refers to the percent of nucleotides/amino acid residues in a subject sequence that are identical to nucleotides/amino acid residues in a reference sequence, after aligning the sequences and, if necessary, introducing gaps, to achieve the maximum percent sequence identity between the sequences. Pairwise and multiple sequence alignment for the purposes of determining percent sequence identity between two or more amino acid or nucleic acid sequences can be achieved in various ways known to a person of skill in the art, for instance, using publicly available computer software such as ClustalOmega (Soding, J. 2005, Bioinformatics 21 , 951-960), T-coffee (Notredame et al. 2000, J. Mol. Biol. (2000) 302, 205-217), Kalign (Lassmann and Sonnhammer 2005, BMC Bioinformatics, 6(298)) and MAFFT (Katoh and Standley 2013, Molecular Biology and Evolution, 30(4) 772-780) software. When using such software, the default parameters, e.g. for gap penalty and extension penalty, are preferably used.

Sequences ***

The features disclosed in the foregoing description, or in the following claims, or in the accompanying drawings, expressed in their specific forms or in terms of a means for performing the disclosed function, or a method or process for obtaining the disclosed results, as appropriate, may, separately, or in any combination of such features, be utilised for realising the invention in diverse forms thereof.

While the invention has been described in conjunction with the exemplary embodiments described above, many equivalent modifications and variations will be apparent to those skilled in the art when given this disclosure. Accordingly, the exemplary embodiments of the invention set forth above are considered to be illustrative and not limiting. Various changes to the described embodiments may be made without departing from the spirit and scope of the invention.

For the avoidance of any doubt, any theoretical explanations provided herein are provided for the purposes of improving the understanding of a reader. The inventors do not wish to be bound by any of these theoretical explanations.

Any section headings used herein are for organizational purposes only and are not to be construed as limiting the subject matter described.

Throughout this specification, including the claims which follow, unless the context requires otherwise, the word ‘comprise’ and ‘include’, and variations such as ‘comprises’, ‘comprising’, and ‘including’ will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

It must be noted that, as used in the specification and the appended claims, the singular forms ‘a,’ ‘an,’ and ‘the’ include plural referents unless the context clearly dictates otherwise. Ranges may be expressed herein as from ‘about’ one particular value, and/or to ‘about’ another particular value.

When such a range is expressed, another embodiment includes from the one particular value and/or to the other particular value. Similarly, when values are expressed as approximations, by the use of the antecedent ‘about,’ it will be understood that the particular value forms another embodiment. The term ‘about’ in relation to a numerical value is optional and means for example +/- 10%.

The present disclosure includes the combination of the aspects and preferred features described except where such a combination is clearly impermissible or expressly avoided.

Where a nucleic acid sequence is disclosed herein, the reverse complement thereof is also expressly contemplated.

Methods described herein may preferably be performed in vitro. The term ‘in vitro’ is intended to encompass procedures performed with cells in culture whereas the term ‘in vivo’ is intended to encompass procedures with/on intact multi-cellular organisms.

Aspects and embodiments of the present disclosure will now be illustrated, by way of example, with reference to the accompanying figures. Further aspects and embodiments will be apparent to those skilled in the art. All documents mentioned in this text are incorporated herein by reference.

Brief Description of the Figures

Figures 1A to 1D (Figure A1A to A1D). Binding of VHH P2A5 to B7-H3. (A) Results of flow cytometric analysis of binding of purified P2A5 VHH to B7-H3-expressing HepG2 cells. (B) to (E) Sensorgrams and binding kinetic parameters determined by multi cycle kinetic analysis of binding of P2A5 VHH to (B) human B7-H3 isoform 1 (also known as 4lg-B7-H3), (C) human B7-H3 isoform 2 (also known as 2lg-B7-H3), and (D) mouse B7-H3, as determined by Surface Plasmon Resonance analysis.

Figures 2A to 2D (Figure A2A to A2D). Cross-reactivity of P2A5 to murine and human B7-H3 molecules expressed by cells of murine and human origin. Mouse CT26 cells and human MKN7 cells engineered to knock-out endogenous B7-H3 were transfected to express mouse B7-H3 (A and B) and human 4lg-B7-H3 (C and D) and stained with positive control anti-mouse or anti-human B7H3 antibodies (top rows), P2A5 as a VHH-Fc fusion molecule (middle rows) or isotype control (bottom rows).

Figure 3 (Figure A3). Generation and functionality of B7-H3 CAR T cells. (A) Schematic representation of the B7-H3 CAR consisting of a B7-H3-targeting VHH, 4-1 BB derived spacer, CD28 transmembrane domain and CD28 and CD3 signalling domains (B) Detection of B7-H3 CAR on activated T cells after transduction. (C) Transduction efficiency and fold expansion of untransduced and B7-H3 CAR T cells tracked over 11 days after transduction. Data presented are mean ± SD of 2 independent donors. (D) Cytolysis of A549 and MDA-MB-231 cells were assessed using the xCELLigence Real-Time Cell Analysis system at 1 :1 and 5:1 E:T ratio for 2 donors. (E) Serial cytolysis of THP-1 and CMK cells during multi-round co-cultures for 2 donors.

Figure 4 (Figure A4). Generation and characterisation of allogeneic B7-H3 CAR EBVSTs. (A) Schematic diagram of B7-H3 CAR EBVST manufacturing. (B) Fold expansion of EBVSTs following mock (UT) or P2A5 CAR transduction. (C) CAR+ percentages and (D) CD8/CD4 ratios of cells at 4- and 11 -days post-transduction. (E) B7-H3 expression on total cells. (F) Percentages of CD4 or CD8 cells expressing either 0, 1 , 2 or all 3 of the exhaustion markers PD-1 , Tim-3 and LAG-3. (G) Percentages of cells expressing TNF-a and IFN-y after overnight stimulation of cells with media, HIV or EBV pepmixes. (B-D) Data presented are from 6 independent donors. Each line representing data from each donor. (E-G) Marker expression data of UT or CAR EBVST final product. Data presented are from 3 independent donors.

Figure 5 (Figure A5). Cytotoxicity against B7-H3+ and B7-H3KO Tumor Cell Lines. Kinetics of cytolysis of B7-H3 CAR EBVSTs against (A) colorectal cancer cell lines DLD-1 , HT29, SW480; (B) gastric cancer cell lines NCI-N87, MKN7, MKN45; (C) breast cancer cell lines MDA-MB-231 , MDA- MB-468; (D) lung cancer cell line A549, H1299, H23 and H596, and their B7-H3 knock-out counterparts at 1 :1 EffectorTarget ratio.

Figure 6 (Figure A6). Serial Killing of B7-H3 CAR EBVSTs. (A) Serial cytotoxicity of EBVSTs following each serial encounter of target AML cell lines THP-1 and CMK. (B) Expansion of EBVSTs following each serial encounter of THP-1 . (C) Kinetics of cytolysis of EBVSTs and (D) expansion of CAR-T cells following each serial encounter of target cell lines DLD-1 , HT29 and NCI-N87 at EffectorTarget ratio of 1 :2.

Figure 7 (Figure A7). In vivo persistence and efficacy of B7-H3 CAR EBVSTs in B7-H3 expressing colorectal cancer. (A) Experiment scheme for HT-29 model. (B) Body weight monitoring twice a week. (C) HT-29 tumor volume and growth fold change following treatment. (D) Endpoint HT-29 tumor counts and B7-H3 expression quantified by flow cytometry (E) Endpoint T cell counts in blood, liver, lung, spleen, and tumor quantified by flow cytometry. (F) Experiment scheme for SW-480 model. (G) Body weight monitoring twice a week. (H) SW-480 tumor volume and growth fold change following treatment. (I) Endpoint SW-480 tumor counts and B7-H3 expression quantified by flow cytometry (J) Endpoint T cell counts in blood, liver, lung, spleen, and tumor quantified by flow cytometry.

Figure 8 (Figure A8). In vivo persistence and efficacy of B7-H3 CAR EBVSTs in B7-H3 expressing gastric cancer. (A) Experiment scheme for N-87 model. (B) Body weight monitoring twice a week. (C) HT-29 tumor volume and growth fold change following treatment. (D) Endpoint N-87 tumor counts and B7-H3 expression quantified by flow cytometry. (E) Endpoint T cell counts in blood, liver, lung, spleen, and tumor quantified by flow cytometry. Figure 9 (Figure B1). Identification and characterization of lead B7H3 VHH from a phage displayed immunized llama library. (A) B7H3 binding screening ELISA of eluted phages from round 2 (top) and round 3 (bottom) of biopanning. (B) Flow cytometric analysis of binding of purified lead VHH candidates to B7H3 expressing HepG2 cells. (C) Surface plasmon resonance screening of lead VHH candidates. Multi cycle kinetic analysis of lead candidate P2A5 binding to (D) human B7H3 4lg isoform, (E) human B7H3 2 Ig isoform and (F) murine B7H3. (G) Measured binding kinetic parameters of P2A5 binding to 4lg, 2lg isoforms of human B7H3 and murine B7H3.

Figure 10 (Figure B2). Cross-reactivity of P2A5 to murine and human B7H3 molecules expressed by cells of murine and human origins. Murine CT26 and human MKN7 cells knocked out for endogenous B7H3 were transfected to express murine B7H3 molecules (A and C) and human 4lg B7H3 molecules (B and D) and stained with positive control anti-mouse or anti-human B7H3 antibodies (top), P2A5 as a VHH-Fc fusion molecule (middle) or isotype control (bottom).

Figure 11 (Figure B3). Generation and functionality of B7H3.CAR T cells. (A) Schematic representation of the B7H3.CAR consisting of a B7-H3-targeting VHH, 4-1 BB derived spacer, CD28 transmembrane domain and CD28 and CD3 signalling domains (B) Detection of B7H3.CAR on activated T cells after transduction. (C) Transduction efficiency and fold expansion of untransduced and B7H3.CAR T cells tracked over 11 days after transduction. Data presented are mean ± SD of 2 independent donors. (D)

Cytolysis of A549 and MDA-MB-231 cells were assessed using the xCELLigence Real-Time Cell Analysis system at 1 :1 and 5:1 E:T ratio for 2 donors. (E) Serial cytolysis of THP-1 and CMK cells during multi-round co-cultures for 2 donors.

Figure 12 (Figure B4). Generation and characterisation of allogeneic B7H3.CAR EBVSTs. (A) Schematic diagram of B7H3.CAR EBVST manufacturing. (B) Fold expansion of EBVSTs following mock (UT) or P2A5 CAR transduction. (C) CAR+ percentages and (D) CD8/CD4 ratios of cells at 4- and 11 -days post-transduction. (E) B7-H3 expression on total cells. (F) Percentages of CD4 or CD8 cells expressing either 0, 1 , 2 or all 3 of the exhaustion markers PD-1 , Tim-3 and LAG-3. (G) Percentages of cells expressing TNF-aand IFN-yafter overnight stimulation of cells with media, HIV or EBV pepmixes. (B-D) Data presented are from 6 independent donors. Each line representing data from each donor. (E-G) Data presented are from 3 independent donors.

Figure 13 (Figure B5). Cytotoxicity against B7-H3+ and B7-H3KO Tumor Cell Lines. Kinetics of cytolysis of B7H3.CAR EBVSTs against (A) colorectal cancer cell lines DLD-1 , HT29, SW480; (B) gastric cancer cell lines NCI-N87, MKN7, MKN45; (C) breast cancer cell lines MDA-MB-231 , MDA- MB-468; (D) lung cancer cell lines A549, H1299, H23 and H596, and their B7-H3 knock-out counterparts at 1 :1 EffectorTarget ratio. (E) Quantification of IFNylevels in cell supernatants of untransduced or B7H3.CAR EBVST after co-incubation with tumor cell lines HT29, HT29-B7H3KO, or N87, N87-B7H3KO. (F) Cytotoxicity of EBVSTs expressing full length or untruncated B7H3. CAR or no CAR against B7-H3+ MKN-45 tumor cells.

Figure 14 (Figure B6). Serial Killing of B7H3.CAR EBVSTs. (A) Serial cytotoxicity of EBVSTs following each serial encounter of target AML cell lines THP-1 and CMK. (B) Expansion of EBVSTs following each serial encounter of THP-1 . (C) Kinetics of cytolysis of EBVSTs and (D) expansion of CAR-T cells following each serial encounter of target cell lines DLD-1 , HT29 and NCI-N87 at EffectorTarget ratio of 1 :2.

Figure 15 (Figure B7). In vitro targeting of allogeneic MDSCs by B7H3.CAR EBVSTs. (A) Flow cytometric histograms showing expression of IL-10, TGF-0, iNOS and B7-H3 in MDSCs generated from 2 independent healthy donors. (B) Cytotoxicity of CD30.CAR EBVST against KM-H2 cell line in the presence or absence of MDSCs at the MDSC:CAR:KM-H2 ratio of 4:1 :1 or 10:1 :1 . (C) Cytotoxicity of Untransduced (UT) or B7H3.CAR EBVSTs (CAR) against allogeneic MDSCs. Data compiled from co-culture of effector EBVSTs generated from 2 donors and target MDSCs generated from a different 2 donors. (D) Proliferation index of Untransduced and B7H3.CAR EBVSTs after anti- CD3/CD28 stimulation in the presence of allogeneic MDSCs. Each data point represents proliferation from each effectortarget donor pair. Proliferation index was calculated by normalising percentages of proliferated cells against that of the no MDSC control condition.

Figure 16 (Figure B8). In vivo persistence and efficacy of B7H3.CAR EBVSTs against B7-H3 expressing colorectal cancer. (A) Experiment scheme for HT-29 model. (B) Body weight monitoring twice a week. (C) HT-29 tumor volume and growth fold change following treatment. (D) Endpoint HT- 29 tumor counts and B7-H3 expression quantified by flow cytometry (E) Endpoint T cell counts in blood, liver, lung, spleen, and tumor quantified by flow cytometry. (F) Experiment scheme for SW- 480 model. (G) Body weight monitoring twice a week. (H) SW-480 tumor volume and growth fold change following treatment. (I) Endpoint SW-480 tumor counts and B7-H3 expression quantified by flow cytometry (J) Endpoint T cell counts in blood, liver, lung, spleen, and tumor quantified by flow cytometry.

Figure 17 (Figure B9). In vivo persistence and efficacy of B7H3.CAR EBVSTs with against B7-H3 expressing gastric cancer. (A) Experiment scheme for NCI-N87 model. (B) Body weight monitoring twice a week. (C) NCI-N87 tumor volume and growth fold change following treatment. (D) Endpoint NCI-N87 tumor counts and B7-H3 expression quantified by flow cytometry (E) Endpoint T cell counts in blood, liver, lung, spleen, and tumor quantified by flow cytometry.

Figure 18 (Figure B10). In vivo persistence and efficacy of B7H3.CAR EBVSTs with against B7-H3 expressing lung cancer. (A) Experiment scheme for NCI-H1299 model. (B) Body weight monitoring twice a week. (C) NCI-H1299 tumor volume and growth fold change following treatment. (D) Endpoint NCI-H1299 tumor counts and B7-H3 expression quantified by flow cytometry. Figure 19 (Figure B11). In vivo persistence and efficacy of B7H3.CAR EBVSTs with against B7-H3 expressing triple-negative breast cancer. (A) Experiment scheme for MDA-MB-468 model. (B) Body weight monitoring twice a week. (C) MDA-MB-468 tumor volume measurements and growth fold change following treatment; and tumor counts at study endpoint quantified by flow cytometry.

Figure 20 (Figure B12). (A and B) Staining of B7-H3 on breast cancer PDX by IHC. Brown colour represents positive B7-H3 staining. Blue colour represents cell nucleus.

Figure 21 (Figure B13). In vivo persistence and efficacy of B7H3.CAR EBVSTs with against B7-H3 expressing breast cancer PDX. (A) Experiment scheme for PDX model. (B) Body weight monitoring twice a week. (C) PDX tumor volume measurements and growth fold change following treatment; and (D) tumor counts at study endpoint quantified by flow cytometry.

Figure 22 (Figure B14). Stimulation and competition of soluble B7-H3 on B7H3.CAR EBVST. (A) Cell staining of IFNy, TNFaand CD25 on UT and B7H3.CAR EBVSTs after treatment with varying concentrations of coated or soluble B7-H3 in 2-lg or 4-lg forms. (B) Cytolysis of B7H3.CAR EBVST against NCI-N87 or NCI-H1299 in the presence of varying concentrations of the 2-lg or 4-lg forms of soluble B7-H3.

Figure 23 (Figure B15). Safety of B7H3.CAR EBVST against Hematopoietic Stem and Progenitor Cells (HSPCs). (A) Flow plots showing HSPC populations expressing B7-H3 after the indicated days of stimulation with Flt3L, TPO and SCF. (B) Comparison of cell surface expression of B7-H3 on HSPCs and cancer cell line NCI-N87 by flow cytometry. (C) Cytotoxicity of unstimulated and stimulated HSPCs after co-culture with B7H3.CAR EBVST. (D) Proportions of HSPC populations after co-culture with untransduced or B7H3.CAR EBVST. (E) Erythroid and myeloid development potential of HSPCs after co-culture with untransduced or B7H3.CAR EBVST.

Figure 24 (Figure B16). Safety of B7H3.CAR EBVST against peripheral blood mononuclear cells (PBMC). (A) Cell surface expressions of B7-H3 on PBMC subsets after stimulation with inflammatory cytokines and cancer cell line NCI-N87. (B) Fold changes in PBMC subset cell counts following a 2- day culture of cytokine-stimulated PBMC with B7H3.CAR EBVST. Fold change is obtained by normalising cell counts of PBMC subsets co-cultured with control untransduced EBVST. (C) Cell surface expressions of B7-H3 on monocytes after co-culturing with allogeneic EBVSTs. (D) Cytotoxicity of monocytes after co-culturing with untransduced or B7H3.CAR EBVST. Data also expressed in fold change in cytotoxicity of B7H3.CAR EBVST over untransduced control.

Figure 25 (Figure B17). Safety of B7H3.CAR EBVST against antigen-experienced T cells. (A) Proliferation and expression of TNFa and IFNy in EBVST and (B) cell surface expression of B7-H3 after stimulation with HIV or EBV pepmix-pulsed antigen presenting cells. Figure 26 (Figure B18). Safety of P2A5-mCAR in Immunocompetent Murine Model. (A) Cytotoxicity of P2A5-mCAR-T against B16F10-WT, B16F10 engineered to express human B7-H3 (B16F10- hB7H3) and B16F10 transfected with mouse B7-H3 (B16F10-mB7H3). (B) Experiment scheme for immunocompetent murine model. (C) Tumor volume measurements and growth fold change following treatment; and tumor weights at study endpoint. (D) Body weight monitoring twice a week. (E) Total counts of host hematological cell subsets in blood, bone marrow, spleen and liver of treated mice quantified by flow cytometry. (F) Serum levels of mouse cytokines and chemokines following treatment quantified by Luminex.

Figure 27 (Figure B19). In vivo CRS model in humanised mice. (A) Experiment scheme for humanised mouse model. (B) Tumor tracking of mice. Left Panel: IVIS monitoring of tumor burden in NALM-6-engrafted mice. Right panel: Fold change in tumor volumes of HT29-eng rafted mice. (C) Body weight monitoring of mice. (D) Serum levels of human cytokines and chemokines at 3-days post-treatment as quantified by multiplex bead immunoassay. (E) Total counts of myeloid cell subsets in blood, bone marrow, spleen and liver of humanized mice quantified by flow cytometry. (F) B7H3 expression on myeloid cell subsets in blood, bone marrow, spleen and liver of humanized mice quantified by flow cytometry.

EXAMPLE

In contrast to its limited protein expression in normal human tissues and lymphoid organs, B7-H3 is aberrantly expressed in a high proportion of human malignancies. The following Example describes a B7-H3 targeting nanobody, and associated CAR constructs/EBVSTs in accordance with the present invention.

The following example describes the identification and utilization of a novel B7-H3 targeting nanobody with superior biophysical qualities compared to scFv, as the antigen specific domain in the CAR construct. T cells and Epstein-Barr Virus Specific T cells (EBVSTs) expressing this nanobodybased B7-H3 CAR demonstrate good in vitro and in vivo activity in B7-H3 positive malignancies. Specifically, the excellent anti-tumor efficacy shown by the B7-H3 CAR expressing EBVSTs will provide an avenue to deliver allogeneic, off-the-shelf CAR T cells therapy to patients with B7-H3 positive malignancies.

Example 1 : Materials and Methods

1.1 Generation of llama immunized library

A naive male llama was used for the generation of an immunized library. His-tagged recombinant human B7-H3 (Sino Biological) was immunised at 1OOpg per injection for a total of 6 injections. 4 days after the final injection, 100ml of anti-coagulated blood was collected and prepared for peripheral blood lymphocytes (PBLs). Total RNA was isolated from the PBLs and used as templates for first strand cDNA synthesis with an oligodT primer. The VHH-encoding sequences were then amplified from the cDNA library and cloned into the pMECS phagemid vector between the Pstl and Notl restriction enzyme sites, upstream of a linker, haemagglutinin (HA) and Hise tag. Electro- competent E. coli TG1 cells were then transformed with the pMHCs vector to generate a VHH library of 10 ⁹ independent transformants. PCR analysis of 95 randomly selected independent transformants was performed to assess the percentage of transformants with the right insert size. Sequence analysis was then done on at 102 randomly selected colonies to assess the diversity and correctness of VHH sequences. TG1 transformants were then infected with M13K07 helper phages to amplify and repackage the VHH library as phage particles to be used for subsequent biopanning.

1 .2 Biopanninq of phage library for B7H3 specific VHHs.

Sequential rounds of biopanning of the phage displayed VHH library was performed for the identification of B7-H3-specific binders. Briefly, the phage library was first incubated with streptavidin coated beads to remove streptavidin binders. Subtracted phages were then incubated streptavidin beads pre-coated with biotinylated B7-H3. After 1 hr incubation with rotation at room temperature, the beads were separated from the non-binding phages and washed multiple times with 0.1 % PBST before elution with 0.1 M triethylamine. Eluted phage were neutralized in 1 M Tris pH8.0 and used to infect TG1 cells at 37°C for 1 hr. 2xYT with 100pg/mL ampicillin and 2% (VA/) glucose were added and the infected TG1 cells were then subsequently infected with M13K07 helper phages to help amplify and produce phage particles. TG1 cells were then spun down at 2000rpm for 10mins and the cell pellet was resuspended in 300mL of 2xTY with 100pg/mL ampicillin and 50pg/mL kanamycin to select for TG1 co-infected with binder and helper phages. Amplification was performed overnight at 37°C. Phage particles were precipitated from the TG1 culture supernatant using PEG/NaCI.

Precipitated phage was collected and quantified by OD260 and used for the subsequent round of biopanning. Stringency of subsequent rounds of biopanning was increased by reducing the concentration of biotinylated B7-H3 proteins coated on the streptavidin beads and increasing the number of PBST washes.

1.3 Phage screening ELISA

1.3.1 Periplasmic preparation

Eluted phages were diluted in 10-fold serial dilutions (dilutions from 10 ¹ -10 ⁷ fold) in PBS and used to infect TG1 cells. Infected TG1 cells were then plated on 90mm LB agar plates containing 100pg/mL ampicillin and 2% (v/v) glucose, in order to obtain single colonies. Plates were the incubated overnight at 37°C. The next day, individual colonies were then picked and grown in 1 ml Terrific Broth (TB) media containing 100pg/mL ampicillin in 96 deep well plates at 37°C for 5-7 hours until ODeoo was about 0.6-0.9. IPTG was then added to each well (to final concentration of 1 mM) and induction was allowed to happen overnight at 37°C. After overnight culture, TG1 cultures were spun down to pellet cells and the supernatants were removed. Tris/EDTA/sucrose (TES) buffer was added to the cell pellet and incubated for 1 hr at 4°C. 4-fold diluted TES was further added, and the cell suspension was incubated at 4°C for 1 hr. The cell suspension was then centrifuged at 2,500 x g for 15 min at 4°C, and the supernatant (periplasmic extraction) was then used for the binding ELISA.

1.3.2 Binding ELISA

ELISA plates were coated with 5pg/mL neutravidin at room temperature for at least 1 hr or at 4°C overnight. Wells were then washed with 0.05% PBST and blocked with casein at room temperature for 2hrs or at 4°C overnight. After blocking, 0.2pg/mL biotinylated B7H3 or control protein were added to each well and incubated for 1 hr at room temperature. After washing with 0.05% PBST, periplasmic extraction was added into individual wells and allowed to bind for 1 hr at room temperature. The wells were then washed with 0.05% PBST and anti-HA antibody (1 :2000 diluted in casein) was added and incubated for 1 hr at room temperature. Wells were washed again and goat anti-mouse HRP secondary antibody (1 :3000) was added to each well. After incubating for 1 hr at room temperature, TMB was added to detect binders and the reaction was stopped using 1 M HCI. Absorbance readings at OD450 and OD570nm were measured by spectrophotometer.

1 .4 Expression and purification of recombinant VHH

For large-scale production of purified VHH, expression was performed from the pHEN6c vector, where intact His tag can be used for subsequent purification and detection. VHH were cloned into the pHEN6c vector between the Pstel and BstEII restriction enzyme sites. Expression of the VHH was then performed using WK6 competent E. coli cells transformed with the pHEN6c vectors. Briefly, a colony of freshly-transformed WK6 cells was grown overnight as a starter culture in TB medium supplemented with 100pg/mL ampicillin and 0.1% glucose at 37°C. 1 mL of the starter culture was then further inoculated in 150mL of TB + Amp/Glu and grown at 37°C until an OD600 of 0.6-0.9 was obtained. Induction using 1 mM IPTG (final concentration) was performed at 28°C overnight (approximately 16-18 hrs). The next day, the cell pellet was harvested after centrifugation of culture, and resuspended in TES buffer. After incubating for 1 hr on ice, dilute TES was added and the suspension was further incubated on ice. The cell suspension was then centrifuged for 30min at 8000rpm at 4°C, and supernatant (periplasmic extract containing the VHH) was harvested for His tag purification using nickel-beads. Purified VHH were buffer exchanged and stored in PBS.

To generate the VHH-Fc construct, VHH sequences were cloned into the pTT5 expression vector upstream of an lgG1-CH2CH3. Subsequently, HEK2936E cells were transiently transfected and recombinant VHH-Fc was purified from the culture supernatant using Protein G beads. Pooled fractions were then concentrated and buffer exchanged into PBS pH 7.4 or MES pH 6.0.

1 .5 Analysis of binding kinetics by Surface Plasmon Resonance

Binding kinetics of VHH molecules to recombinant B7-H3 protein was assessed by surface plasmon resonance. Human 4lg-B7-H3 was obtained from Sino Biological (Cat. No. 11188-H08H; C- terminally His-tagged ECD (/.e. positions 1 to 461 of UniProt:Q5ZPR3-1)), and human 2lg-B7-H3 was obtained from Acrobiosystems (Cat. No. B73-H5253; positions 29 to 245 of UniProt:Q5ZPR3-2 comprising a C-terminal Fc tag formed by positions 100 to 330 of UniProt:P01857). All kinetic experiments were performed on a Biacore T200 (Cytiva, Uppsala, Sweden). HBS-P+ supplemented with 1 %BSA w/v (Cytiva, Marlborough, USA) was used as the running buffer and dilution buffer for the analyte and ligand. Briefly, a rabbit anti-VHH antibody (Genscript, Shanghai) was immobilized on CM5 chips. Purified VHH was then injected at a concentration of 100mM for 60s to be captured by the anti-VHH antibodies. For initial screening of VHHs, 100nM of recombinant B7- H3 was injected after stabilization. For cross-reactivity assessment, 100nM murine B7-H3 was injected after stabilization. A reference channel with no VHH captured was used to correct for bulk effect and non-specific binding while a blank run (no antigen flowed) was used to correct for surface stability. The double referenced sensorgrams were fitted with the Langmuir (1 :1) binding model to obtain the association k _a , dissociation kd and equilibrium dissociation constant KD, and the closeness of fit was evaluated with the Chi squared value.

For multi-cycle kinetic analysis, recombinant B7-H3 was injected at concentrations between 6.25nM to 200nM in a six point, two-fold dilution series. For each concentration, recombinant B7-H3 was allowed to associate for 180 seconds and dissociate for 180 seconds. Regeneration was performed with 10mM glycine (pH 1 .5). A reference channel with no scFv-mFc captured was used to correct for bulk effect and non-specific binding while a blank run (no antigen flowed) was used to correct for surface stability. The double referenced sensorgrams were fitted with the Langmuir (1 :1) binding model to obtain the association k _a , dissociation kd and equilibrium dissociation constant KD, and the closeness of fit was evaluated with the Chi square value.

1 .6 Cross-reactivity studies with transfected cell lines.

Human MKN7 cells were engineered using CRISPR-Cas9 to knock-out expression of B7-H3, as described in Example 1.12. Murine CT26 cells and the MKN7 CD276 k/o cells were then transfected to express murine B7-H3-GFPSpark and human 4lg-B7-H3-GFPSpark fusion proteins with their respective pCMV3 vectors (Sino Biological). Briefly, cells were seeded in a 6 well plate to 70% confluency and allowed to adhere overnight. Plasmids were then premixed with FuGene® 6 transfection reagent (Promega) and added to the cells. Transfection was then assessed by flow cytometry and fluorescence microscopy.

1 .7 B7-H3-tarqetinq CAR constructs and retrovirus production

B7-H3-targeting VHHs were cloned into a pSFG retrovirus vector upstream of a 4-1 BB derived spacer, followed by CD28 transmembrane domain, a CD28 and CD3 signalling domains. A truncated form of the B7-H3-specific CAR that consisted only of the extracellular domain was also cloned into pSFG retrovirus vector. Retroviruses carrying the full or truncated B7-H3 CARs were produced in RD114 packaging cell line (BioVec Pharma, Quebec, Canada) by transient transfection with the pSFG vector using PEIpro transfection reagent (Polyplus, lllkirch, FRANCE). Medium containing retroviruses were harvested at 48h and 72h post transfection and concentrated 10 to 50- fold using RetroX Concentrator (Takara Bio, Kusatsu, Shiga, Japan). The retroviruses were either used immediately or snap frozen and stored at -80°C.

The stable RD114 retrovirus packaging cell line that produces high titers of GFP-Firefly Luciferase (GFP-FFIuc) virus particles was used.

1 .8 Donors

Enriched leukapheresis products, collected from consented healthy donors by Spectra Optia® Apheresis System CMNC collection protocol and frozen in ACD-A anticoagulant, was purchased from HemaCare (Northridge, California, U.S.A.). The frozen leukopaks were thawed and PBMCs were extracted by gradient centrifugation using Ficoll-Paque PLUS (Cytiva, MA, U.S.A.). The PBMCs were either used immediately for experiments or frozen in smaller aliquots of 30-50 x 10 ⁶ cells per cryovial in CryoStor® CS10 Cell Freezing Medium (STEMCELL Technologies, Cambridge, Massachusetts, U.S.A.).

1 .9 Generation of B7-H3 targeting CAR T cells

PBMC were thawed and plated onto cell culture plates pre-coated with anti-CD3/CD28, and cultured in 10% FBS, 45% Advanced RPMI and 45% Click’s media with 5% CO2 at 37°C to generate activated T cells (ATCs). Two days after culture, IL-7 and IL-15 were added into cell culture. On the third day of culture, retrovirus containing the B7-H3-specific CAR was transduced into ATCs by spinfection. Retrovirus was washed off 24 hours later and ATCs were cultured with occasional media change to replenish the IL-7 and IL-15. On day 11 post-transduction, ATCs were frozen down using CryoStor® and stored in liquid nitrogen until needed for further studies.

All effector T cells were thawed and rested overnight in IL-7 and IL-15 prior to cell count and evaluation in in vitro or in vivo studies.

1.10 Generation of B7-H3 targeting CAR EBVSTs

PBMCs from healthy donors were thawed before CD45RA depletion was performed by negative selection using CD45RA MACS Beads (Miltenyi Biotec, Germany). CD45RA-depleted PBMCs were cultured with viral peptides consisting of overlapping Epstein-Barr Virus peptide libraries (15-mers overlapping by 11 amino acids) from JPT Technologies (Berlin, Germany) for EBV antigens BMRF2, BALF2, BNLF2b, BNLF2a, LMP1 , LMP2, EBNA1 , BZLF1 , BRLF1 , BMRF1 , BMLF1 and BARF1 and cultured in VST medium [47.5% advanced RPMI 1640 (Gibco), 47.5% Clicks’ medium (FUJIFILM Irvine Scientific), 5% human platelet lysate (Sexton Biotechnologies) and 2 mM GlutaMAX (Gibco)] supplemented with 10 ng/mL IL-7 and IL-15 (R&D Systems). Five days later, cells were transduced with B7-H3-specific CAR constructs using RetroNectin (Takara Bio, Japan). T cells were then stimulated with irradiated co-stimulatory cells expressing markers such as CD80, CD86, 4-1 BB four days post transduction. Seven to eight days later, VSTs were harvested, frozen using CryoStor® or used for cell assays. All effector T cells were thawed and rested overnight in IL-7 and IL-15 prior to cell count and evaluation in in vitro or in vivo studies.

1 .11 Transduction efficiency and phenotyping by flow cytometry and antibody staining Flow cytometry was performed with the Aurora cytometer (Cytek Biosciences) or with the FACSymphony A3 cell analyzer (BD Biosciences). Up to 200,000 T-cells were stained with Live/Dead™ NIR viability dye (Thermo Fisher) and assessed for surface presentation of epitopes using fluorescently labeled monoclonal antibodies to CD3 (clone SK7, Biolegend), CD4 (clone SK3, BD Biosciences), CD8 (clone SK1 , BioLegend), CD56 (clone B159, BD Biosciences), CD19 (clone SJ25C1 , BD Biosciences), B7-H3 (clone 7-517, BD Biosciences), PD-1 (clone EH12.1 , BD Biosciences), Tim3 (clone 7D3, BD Biosciences), and LAG3 (clone 11C3C65, BioLegend). B7-H3 CAR expression was measured with anti-Camelid VHH antibody (clone 96A3F5, Genscript). Data were analysed and gated in FlowJo v10.8.1 for Windows.

1 .12 Gene knockout of B7-H3 in cell lines

Gene knockout of B7-H3 in cell lines was performed using CRISPR-Cas9 system. Single-guide RNAs targeting seguences 5’-CTGGTGCACAGCTTTGCTGA-3’, 5’- GTGCCCACCAGTGCCACCAC-3’ and 5’-TGCCCACCAGTGCCACCACT-3’ (Integrated DNA Technologies, Inc and Synthego) were incubated with Cas9 (Integrated DNA Technologies, Inc) for 10 min to form RNP complexes. Cell lines were washed with PBS and briefly incubated with RNP complexes, and subseguently electroporated using the 4D-Nucleofector (Lonza). Cells were then allowed to recover in complete media before plating onto tissue culture plates. Knockout efficiencies were assessed using flow cytometry by staining cells with antibody against B7-H3 (clone 7-517, BD Biosciences). B7-H3- cells were selected by FACS using the BD InfluxTM Cell Sorter (BD Biosciences).

1 .13 Real-time Cytotoxicity Assays

Prior to assessment of CAR potency via cytotoxicity assay, the cell media was changed to 2% assay media containing RPMI and 2% FBS. Cytotoxicity assay was carried out using the xCELLigence® Real-Time Cell Analysis System with 5% CO2 at 37°C (Agilent). Target tumor cells were added onto PET plates following manufacturer’s manual (Agilent). 24 hours later, CAR T cells were plated into the PET plates at CAR T: target cell ratio of 0.5:1 , 1 :1 or 5:1 . The PET plates were then returned into the xCELLigence® system, and cytotoxicity was monitored for 48 hours.

1 .14 Serial Killing Potency Assay using the xCelligence Real-Time Cell Analysis System

Serial killing was assessed using the xCELLigence® Real-Time Cell Analysis System with 5% CO2 at 37°C (Agilent). Target DLD-1 , HT-29 and NCI-N87 cells were added onto PET plates following manufacturer’s instructions (Agilent). 24 hours after target cell adherence, CAR T cells were plated onto the PET plates at a defined CAR T: target cell ratio of 1 :2. The PET plates were returned to the xCELLigence® system and cytotoxicity will be monitored for 48 hours. After 48 hours, the suspensions of CAR T cells were harvested, counted and added to plates of target cells seeded 24 hours earlier at a defined CAR T: target cell ratio of 1 :2 to set up Encounter 2. After a further 48h, cells were harvested and added to a new set of target cells per well for Encounter 3. 48h after each encounter, 1 set of cells was stained with Live/Dead™ NIR viability dye (Thermo Fisher) and fluorescently labelled monoclonal antibodies to anti-camelid VHH antibody (clone 96A3F5, Genscript), CD3 (clone UCHT1 , BD Biosciences), CD4 (clone SK3, BD Biosciences), CD8 (clone RPA-T8, BD Biosciences), B7-H3 (clone 7-517, BD Biosciences), PD-1 (clone EH12.1 , BD Biosciences), Tim3 (clone 7D3, BD Biosciences) and LAG3 (clone 1 1C3C65, BioLegend) with CountBright™ Absolute Counting Beads (Invitrogen™) to enumerate the effector cell populations present after each round of target encounter. Data were analysed and gated in FlowJo v10.8.1 for Windows. Cytolysis was calculated as (cell count in target-only control well - cell count in assay well) I cell count in target-only control well.

1 .15 Serial Killing Potency Assay using flow cytometry analysis

THP-1 and CMK cells were used as the target cell line with an initial E:T ratio of 1 :2, with 50,000 effector and 100,000 target cells per well of a 96-well plate for Encounter 1 . Three sets of this set-up were prepared. After 48h, cells from 2 sets were harvested and added to 2 new sets of 100,000 target cells per well to set up Encounter 2. After a further 48h, cells from 1 set were harvested and added to a new set of 100,000 target cells per well for Encounter 3. To distinguish target cells from the different encounters, THP-1 and CMK cells for Encounters 2 and 3 were labelled with the lipophilic membrane dyes PKH67 and PKH26 (Sigma-Aldrich) respectively. 48h after each encounter, 1 set of cells was stained with Live/Dead™ NIR viability dye (Thermo Fisher) and fluorescently labelled monoclonal antibodies to anti-camelid VHH antibody (clone 96A3F5, Genscript), CD3 (clone UCHT1 , BD Biosciences), CD4 (clone SK3, BD Biosciences), CD8 (clone RPA-T8, BD Biosciences), B7-H3 (clone 7-517, BD Biosciences), PD-1 (clone EH12.1 , BD Biosciences), Tim3 (clone 7D3, BD Biosciences), LAG3 (clone 11 C3C65, BioLegend) with CountBright™ Absolute Counting Beads (Invitrogen™) to enumerate the target and effector cell populations remaining at each round of 48h culture. Data were analysed and gated in FlowJo v10.8.1 for Windows. Cytolysis was calculated as (Target count in target-only control well - Target count in assay well) / Target count in target-only control well.

1 .16 Colorectal carcinoma and gastric cancer mouse models

In the colorectal carcinoma model, 5x10 ⁶ of HT-29 or 2x10 ⁶ SW-480 cells were subcutaneously injected into right flank of NOD-sc/d IL2Rgamma ^nul1 Kb Db ^nul1 l-A ^nul1 (NSG-MHC l/ll DKO) mice. In the gastric carcinoma model, 2x10 ⁶ of NCI-N87 cells were subcutaneously injected into right flank of NOD-sc/d IL2Rgamma ^nul1 Kb Db ^nul1 l-A ^nul1 (NSG-MHC l/ll DKO) mice. On day 7, mice were randomised into treatment groups, stratified by tumor volume. 5x10 ⁶ of un-transduced or P2A5-B7- H3 CAR EBVSTs were injected into mice intravenously. Untreated mice served as controls. Physical examination, tumor and body weight measurement was carried out twice a week until endpoint. When mice were sacrificed on day 21 , blood, spleen, liver, lung and tumor were collected for endpoint flow cytometric analysis.

Endpoint antibody staining was performed with Live/Dead™ NIR viability dye (Thermo Fisher) and fluorescently labelled monoclonal antibodies to anti-camelid VHH antibody (clone 96A3F5, Genscript), mouse CD45 (clone 30-F11 , BD Biosciences), human CD45 (clone H130, BD Biosciences), CD3 (clone UCHT1 , BD Biosciences), CD4 (clone SK3, BD Biosciences), CD8 (clone RPA-T8, BD Biosciences), B7-H3 (clone 7-517, BD Biosciences), PD-1 (clone EH12.1 , BD Biosciences), Tim3 (clone 7D3, BD Biosciences), LAG3 (clone 11C3C65, BioLegend) with CountBright™ Absolute Counting Beads (Invitrogen™) to enumerate cell populations. Data were analysed and gated in FlowJo v10.8.1 for Windows.

1 .17 EBV reactivity assay:

To measure the reactivity of EBVSTs to EBV antigens, cells were stimulated with HIV or EBV pepmixes (JPTPeptide Technologies), in the presence of anti-CD49d (clone L25, BD Biosciences) and anti-CD28 (clone L293, BD Biosciences). 1 hour post-stimulation, cells were treated with GolgiSTOP and GolgiPlug solution (BD Biosciences) to accumulate cytokines in the Golgi complex. After an overnight incubation, cells were surface stained with fluorescently-tagged antibodies against CD3 (clone SK7, BioLegend), CD4 (clone SK3, BD Biosciences), CD8 (clone SK1 , BioLegend) and Camelid VHH (clone 96A3F5, Genscript). Cells were fixed and permeabilised, then stained with antibodies against TNF (clone Mab11 , BD Biosciences) and IFN-y (clone B27, BD Bioscienes). Flow cytometry was performed with the FACSymphony A3 cell analyzer (BD Biosciences) and data were analysed and gated in FlowJo v10.8.1 for Windows.

1.18 In vivo evaluation of B7H3.CAR EBVSTs in colorectal cancer (CRC), non-small cell lung cancer (NSCLC), triple negative breast cancer (TNBC) and gastric cancer (GO mouse models

In the CRC model, 5x10 ⁶ of HT-29 or 2x10 ⁶ SW480 cells were subcutaneously injected into right flank of NOD-sc/d IL2Rgammanull Kb Dbnull l-Anull (NSG-MHC l/ll DKO) mice. In the NSCLC, TNBC and GC cancer models, 2x10 ^e NCI-H1299, 5x10 ^e MDA-MB-468 or 2x10 ^e of NCI-N87 cells were subcutaneously injected into right flank of NSG-MHC l/ll DKO mice, respectively. When tumors become palpable (100-200 mm ³ ) between days 7 to 16, mice were randomized into treatment groups, stratified by tumor volume. 5x10® of un-transduced or P2A5-B7H3.CAR EBVSTs were injected into mice intravenously while untreated mice served as controls. Physical examination, tumor and body weight measurements were carried out twice a week until endpoint when tumors in control mice reaches 1000 mm ³ in size. When mice were sacrificed between days 21 to 42, blood, spleen, liver, lungs and tumor were collected for endpoint flow cytometric analysis. Endpoint antibody staining was performed with Live/Dead™ Agua viability dye (Thermo Fisher) and fluorescently labelled monoclonal antibodies to camelid VHH (clone 96A3F5, Genscript), mouse CD45 (clone 30-F11 , BD Biosciences), human CD45 (clone H130, BD Biosciences), CD3 (clone UCHT1 , BD Biosciences), CD4 (clone SK3, BD Biosciences), CD8 (clone RPA-T8, BD Biosciences), B7-H3 (clone 7-517, BD Biosciences), PD-1 (clone EH12.1 , BD Biosciences), Tim3 (clone 7D3, BD Biosciences), LAG3 (clone 11C3C65, BioLegend), with addition of CountBright™ Absolute Counting Beads (Invitrogen™) to enumerate cell populations by flow cytometry.

1 .19 Breast Cancer Patient-Derived Xenograft Model

Breast cancer patient-derived xenograft (PDX) were subcutaneously injected into the right flank of NOD-sc/d IL2Rgammanull (NSG) mice. When PDX became palpable at Day 32, mice were randomised into treatment groups, stratified by tumor volumes. 5x10 ⁶ of un-transduced or P2A5- B7H3.CAR EBVSTs expressing GFP-luciferase were injected into mice intravenously. Untreated mice served as controls. Physical examination, tumor and body weight measurements and I VIS® imaging were carried out twice a week until study endpoint at 22 days post-treatment.

1 .20 B7H3 CAR Activation and Competition Assay with Soluble B7-H3

Untransduced and B7H3.CAR EBVSTs were stimulated with either the 4-lg or 2-lg form of recombinant soluble B7-H3 (Aero Biosystems). For controls, untransduced and B7H3.CAR EBVSTs were seeded onto microwell plates that were coated with either the 4-lg or 2-lg form of recombinant B7-H3. Cells were then assayed for cytokine expression and activation by intra-cellular Interferon gamma (IFN-y) and tumor necrosis factor alpha (TNF-a) staining (ICS) and cell surface staining of CD25. For CD25 staining, cells were harvested 2 days post-stimulation and stained with Live/Dead™ NIR viability dye (Thermo Fisher) and fluorescently labelled monoclonal antibodies to anti-camelid VHH antibody (clone 96A3F5, Genscript), CD3 (clone UCHT1 , BD Biosciences), CD4 (clone SK3, BD Biosciences), CD8 (clone RPA-T8, BD Biosciences) and CD25 (clone 2A3, BD Biosciences). For ICS, GolgiSTOP and GolgiPlug (BD Biosciences) were added to cells 1 h poststimulation, and after an overnight incubation, cells were stained with Live/Dead™ NIR viability dye (Thermo Fisher) and fluorescently labelled monoclonal antibodies to camelid VHH (clone 96A3F5, Genscript), CD3 (clone UCHT1 , BD Biosciences), CD4 (clone SK3, BD Biosciences), CD8 (clone RPA-T8, BD Biosciences), before fixation and permeabilization with Cytofix/Cytoperm kit (BD Biosciences) and staining with fluorescently labelled monoclonal antibodies against IFN-y(clone B27, BioLegend) and TNF-a(clone Mab11 , BD Biosciences).

For competition assay, cytotoxicity of B7H3.CAR EBVSTs against NCI-N87 or NCI-H1299 was measured using the xCELLigence® Real-Time Cell Analysis System (Agilent) in the presence of varying concentrations of soluble B7-H3 over a period of 48 h.

1 .21 In vitro safety assays with hematopoietic stem and progenitor cells (HSPCs)

CD34+ HSPCs were stimulated with 10 ng/mL each of Flt3-ligand (FLT3L), stem cell factor (SCF) and thrombopoietin (TPO) (all from Miltenyi Biotec) at 5,000 -20,000 cells per 200pL per well for the specific durations. In cytotoxicity co-cultures of CD34+ HSPCs and T cells, stimulated HSPCs were seeded together with untransduced or B7H3.CAR EBVSTs at E:T ratio of 1 :1 for 24h. HSPC subsets were analyzed by flow cytometry after staining with Live/Dead™ NIR viability dye (Thermo Fisher) and fluorescently labelled monoclonal antibodies to CD34 (clone 561 , BD Biosciences), CD133 (clone 7, BioLegend), CD45RA (clone HI100, BD Biosciences), CD38 (clone HIT2, BD Biosciences), CD10 (clone HI10a, BD Biosciences). CountBright™ Absolute Counting Beads (Invitrogen™) were used to enumerate cell populations by flow cytometry.

Erythroid and myeloid developmental potential were assessed using the StemMACS™ HSC-CFU Assay kit (Miltenyi Biotec). Cells were labelled with fluorescently conjugated antibodies to CD14, CD15 and CD235a as part of the StemMACS kit antibody cocktail, and colony types were identified as per the manufacturer’s protocol. Colony types include colony forming units for granulocytes (CFU- G) and macrophages (CFU-M), burst forming units for erythrocytes (BFU-E). For more primitive progenitors, CFU-GM give rise to both granulocytes and macrophages, while CFU-GEMM differentiate into all three cell populations.

1 .22 In vitro safety assays with peripheral blood mononuclear cells (PBMC)

PBMC were stimulated with granulocyte-macrophage colony-stimulating factor (GM-CSF), IFNy, TNFa, Lipopolysaccharide (LPS) or media alone at the indicated concentrations before staining with Live/Dead™ NIR viability dye (Thermo Fisher) and fluorescently labelled monoclonal antibodies to B7-H3 (clone 7-517, BD Biosciences).

In cytotoxicity co-culture experiments, PBMC were stimulated with GM-CSF, IFNy, TNFa, LPS or media alone at the indicated concentrations for 1 day before cytokines were washed off and PBMCs were further cultured for 2 days with CellTrace™ Violet-labelled untransduced or B7H3.CAR EBVST. Cells were stained with Live/Dead™ NIR viability dye (Thermo Fisher) and fluorescently labelled monoclonal antibodies to human CD14 (clone M5E2, BD Biosciences), human CD56 (clone HCD56, BD Biosciences), human CD16 (clone 3G8, BD Biosciences), CD3 (clone UCHT1 , BD Biosciences), CD4 (clone SK3, BD Biosciences), CD8 (clone RPA-T8, BD Biosciences), B7-H3 (clone 7-517, BD Biosciences) and camelid VHH (clone 96A3F5, Genscript), with addition of CountBright™ Absolute Counting Beads (Invitrogen™) to enumerate cell populations by flow cytometry.

Similarly, in monocyte co-culture experiments, CD14 microbeads (Miltenyi Biotec) purified monocytes were incubated with CellTrace™ Violet-labelled allogeneic EBVST or B7H3.CAR EBVST for 2 days. Cells were stained with Live/Dead™ NIR viability dye (Thermo Fisher) and fluorescently labelled monoclonal antibodies to human CD14 (clone M5E2, BD Biosciences), CD3 (clone UCHT1 , BD Biosciences), CD4 (clone SK3, BD Biosciences), CD8 (clone RPA-T8, BD Biosciences), B7-H3 (clone 7-517, BD Biosciences) and camelid VHH (clone 96A3F5, Genscript), with addition of CountBright™ Absolute Counting Beads (Invitrogen™) to enumerate cell populations by flow cytometry. 1 .23 In vitro safety assays with antigen-experienced T cells

PBMCs were activated with plate-coated anti-CD3/CD28 antibodies for 7 days to generate activated T cells (ATCs), before pulsing with HIV or EBV pepmixes. Pulsed ATCs were then irradiated and cocultured with CellTraceTM Violet-labelled untransduced or B7H3.CAR EBVST. GolgiSTOP and GolgiPlug (BD Biosciences) were added to cells 1 h post-stimulation, and after overnight incubation, cells were stained with Live/Dead™NIR viability dye (Thermo Fisher) and fluorescently labelled monoclonal antibodies to camelid VHH (clone 96A3F5, Genscript), CD3 (clone UCHT1 , BD Biosciences), CD4 (clone SK3, BD Biosciences), CD8 (clone RPA-T8, BD Biosciences), before fixation and permeabilization with Cytofix/Cytoperm kit (BD Biosciences) and staining with fluorescently labelled monoclonal antibodies to IFNy(clone B27, BioLegend) and TNFa(clone Mab11 , BD Biosciences). For analyses of cell proliferation and B7-H3 expression, cells were collected on Day 2 and Day 5 of culture and stained with Live/Dead™NIR viability dye (Thermo Fisher) and fluorescently labelled monoclonal antibodies to camelid VHH (clone 96A3F5, Genscript), CD3 (clone UCHT1 , BD Biosciences), CD4 (clone SK3, BD Biosciences), CD8 (clone RPA-T8, BD Biosciences) and B7-H3 (clone 7-517, BD Biosciences).

1 .24 In vitro MDSC generation and co-culture assays

To generate MDSCs, monocytes were isolated from PBMCs using CD14 microbeads (Miltenyi Biotec) and cultured in IL-6 and GM-CSF for 7 days. MDSCs were then harvested on day 7 using Accutase cell detachment medium (Thermo Fisher). Cells were stained with Live/Dead™ NIR viability dye (Thermo Fisher) and fluorescently labelled monoclonal antibodies to CD14 (clone M5E2, BD Biosciences), CD11 b (clone ICRF44, BD Biosciences), CD33 (clone HIM3-4, BD Biosciences), CD15 (clone W6D3, BioLegend), CD66b (clone 6/40c, BioLegend), HLA-DR (clone G46-6, BD Biosciences) and B7-H3 (clone 7-517, BD Biosciences) to verify their surface marker expression. Cells were also fixed and permeabilised before staining with fluorescently labelled monoclonal antibodies to IL-10 (clone JES3-9D7, BD Biosciences), TGF-01 (clone S200006A, BioLegend) and iNOS (clone CXNFT, Thermo Fisher) to determine their expression of the inhibitory molecules.

For MDSC co-culture experiments, untransduced and B7H3.CAR EBVSTs were first labelled with CellTrace™ Violet (Thermo Fisher) before culturing with allogeneic MDSCs in the stated EffectorTarget ratios overnight. For assessment of proliferation, CellTrace™ Violet-labelled untransduced and B7H3.CAR EBVSTs were cultured with allogeneic MDSCs on anti-CD3/CD28 coated plate for 6 days. Cells were harvested and stained with fluorescently labelled monoclonal antibodies as detailed above to differentiate the effector and target cells. CountBright™ Absolute Counting Beads (Invitrogen) were added to enumerate cell populations by flow cytometry. Proliferation was assessed based on percentage of cells with CellTrace™ Violet dilution. Proliferation index was calculated by normalising proliferated cell percentages against proliferated cell percentages in no MDSC control condition.

1 .25 In vivo murine safety model in immunocompetent mice

B16F10-WT cells were retrovirally transduced with human B7-H3 construct to create the B16F10- hB7H3 murine tumor cell line that expresses human B7-H3. 5x10 ⁵ B16F10-hB7H3 cells were subcutaneously injected into the right flank of wild-type C57BL6/J mice. When tumors become palpable, mice were irradiated at 5 Gy to create a lymphodepleted environment. 3 days after irradiation, mice were randomised into treatment groups and intravenously injected with 10x10 ⁶ untransduced or P2A5.mCAR-T cells that were generated from activated T cells isolated from splenocytes of B6.SJL-Ptprc ^a Pepc ^b /BoyJ mice (The Jackson Laboratory). Untreated mice served as controls. Physical examination, tumor and body weight measurements were carried out twice a week until endpoint when tumors in control mice reaches 1000 mm ³ in size. Blood serum was collected from mice by cheek bleeding 3-days post-treatment. Serum cytokine levels were quantified using a Bead-Based Multiplex Assay using Luminex technology (Merck-Millipore). When mice were sacrificed 10 days post-treatment, blood, spleen, liver, lungs, tumor, brain and bone marrow were harvested for endpoint flow cytometric and pathology evaluation. Endpoint antibody staining was performed with Live/Dead™ NIR viability dye (Thermo Fisher) and fluorescently labelled monoclonal antibodies to camelid VHH (clone 96A3F5, Genscript), mouse CD45.1 (clone A20, BD Biosciences), CD45.2 (clone 104, BD Biosciences), CD11 b (clone M1/70, BD Biosciences), CD11 c (clone N418, BioLegend), Gr1 (clone RB6-8C5, BD Biosciences), CD19 (clone ID3, BD Biosciences), NK1.1 (clone PK136, BD Biosciences), CD3 (clone 145-2C11 , BD Biosciences), CD4 (clone RM4-5, BD Biosciences), CD8 (clone 53-6.7, BD Biosciences), B7-H3 (clone EPNCIR122, Abeam), with addition of CountBright™ Absolute Counting Beads (Invitrogen™) to enumerate cell populations by flow cytometry.

1.26 In vivo Cytokine Release Syndrome (CRS) model in humanised mice

To generate humanised mice, 1x10 ⁵ CD34+ cord blood cells were intravenously injected into sub- lethally irradiated triple transgenic human IL3, GM-CSF and stem cell factor (SCF) expressing NSG (NSG-SGM3) mice (The Jackson Laboratory). After 4 weeks of humanisation, mice were verified to be successfully humanised by positive staining of human CD45 on peripheral blood cells. For the B7H3 CAR test arm, 5x10® of HT-29 cells were subcutaneously injected into the right flank of the humanised mice. When tumors become palpable, mice were randomized into treatment groups, stratified by tumor volume. 5x10® of un-transduced or B7H3.CAR EBVSTs were injected into mice intravenously while untreated mice served as controls. For CRS positive control arm, 2x10® of CD19- expressing NALM-6 cells were intravenously injected into humanised mice. Tumor burden of NALM- 6 engrafted mice were tracked using I VIS imaging. 18 days post NALM-6 tumor implantation, mice were randomised and treated with CD19.CAR-T or left untreated. Physical examination, tumor and body weight measurement was carried out twice a week until endpoint at 7 days post-treatment. Blood serum was collected from mice by cheek bleeding 3-days post-treatment. Serum cytokine levels were quantified using LEGENDplex multiplex bead-based immunoassay (BioLegend). When mice were sacrificed, blood, spleen, liver and bone marrow were harvested for endpoint flow cytometric analysis.

Endpoint antibody staining was performed with Live/Dead™ Aqua viability dye (Thermo Fisher) and fluorescently labelled monoclonal antibodies to camelid VHH (clone 96A3F5, Genscript), mouse CD45 (clone 30-F11 , BD Biosciences), human CD45 (clone H130, BD Biosciences), CD3 (clone UCHT1 , BD Biosciences), CD14 (clone M5E2, BioLegend), CD19 (clone SJ25C1 , BioLegend), CD11 b (clone ICRF44, BD Biosciences), CD66b (clone G10F5, BioLegend), CD11c (clone B-ly6, BD Biosciences), CD56 (clone HCD56, BioLegend), CD4 (clone SK3, BD Biosciences), CD8 (clone RPA-T8, BD Biosciences), HLA-A3 (clone GAP.A3, BD Biosciences), B7-H3 (clone 7-517, BD Biosciences), PD-1 (clone EH12.1 , BD Biosciences), Tim3 (clone 7D3, BD Biosciences), LAG3 (clone 11C3C65, BioLegend), with addition of CountBright™ Absolute Counting Beads (Invitrogen™) to enumerate cell populations by flow cytometry.

1 .27 Flow cytometry analysis

Flow cytometry of cells were performed on the FACSymphony A3 cell analyzer (BD Biosciences) and data were analysed in FlowJo v10.8.1 for Windows.

1 .28 Statistical analysis

Statistical analysis and visualization were performed using Prism 9 software for Windows (Graphpad Software Inc.). For comparisons between two groups, a two-tailed unpaired t test was used, where appropriate. For comparisons in time courses or among three or more groups, one-way or two-way analysis of variance (ANOVA) with Tukey or Dunnett or Sidak’s post-test was applied, where appropriate.

Example 2: Results

2.1 B7-H3-specific VHH

VHHs specific for B7H3 were isolated from the immunized llama library after repeated rounds of biopanning with recombinant human 4lg-B7-H3. Increasing stringency was applied by reducing the concentrations of target protein from 100nM in the first round, to 20nM in the second round and finally 5nM in the third round. ELISA screening for B7-H3 binding was performed with phage eluted from rounds 2 and 3. After sequencing to identify unique VHH sequences, 20 shortlisted leads were then expressed and purified for further screening for binding to B7-H3 expressed on the surface of B7-H3 expressing HepG2 cells, and for binding kinetics to recombinant 41g B7H3 by surface plasmon resonance. Only 9 leads were able to bind strongly to B7-H3 expressing HepG2 cells although most leads bound to recombinant B7-H3 very strongly.

Clone P2A5 showed strong binding to human B7-H3-expressing HepG2 cells (Figure 1 A), while having a moderate off-rate (Figure 1 B).

The affinity with which a CAR binds its target antigen affects not only its efficacy, but also the safety of CAR T cell therapy. The risk of on target, off tumour toxicity can increase with higher affinity binding to targets such as B7-H3, which are expressed at low levels on non-malignant cells and tissues. Hence the balance of efficacy and specificity for higher expression of B7-H3 on tumour cells can be tuned by the selection of intermediate affinity and avoiding high affinity VHHs as the binding domain of the CAR. As such, the inventors decided to focus on the intermediate-affinity clone P2A5.

The binding affinity of P2A5 to human 4lg-B7-H3 was evaluated by multi-cycle kinetic analysis by SPR (Figure 1 B).

In humans, B7-H3 exists in two major isoforms having a similar structure, consisting of either a single pair of the extracellular immunoglobulin domains IgV-lgC for the shorter 21 g form (2lg-B7-H3) or tandem pairs IgV-lgC-lgV-lgC for the larger 4lg form (4lg-B7-H3). Murine B7-H3 comprises a single pair of IgV-lgC domains and shares 87% sequence homology with human 2lg-B7-H3. The inventors further investigated binding of P2A5 to both the 2lg-B7-H3 and mouse B7-H3 (Figure 1C and 1 D). While P2A5 binds 4lg-B7-H3 relatively strongly with an equilibrium dissociation constant KD of -30.9 nM (Figure 1 B), the affinity of binding to 2lg-B7-H3 and mouse B7-H3 is -10-fold weaker, with Ko of 340nM and 473nM respectively (Figure 1C and 1 D).

2.2 Cross-reactivity of B7H3 specific P2A5 VHH to cell expressed murine and human B7H3.

B7-H3 is a heavily glycosylated protein, and the glycosylation patterns can vary between human and murine cells. Hence to ascertain that the binding epitope of P2A5 on B7-H3 is not influenced by the glycosylation in different cells, we transfected a B7-H3-negative murine colon cancer cell line, CT26 and a gastric adenocarcinoma cell line MKN7 that has been knocked out for B7-H3 expression by CRISPR/Cas9, to express both human (4lg) and murine B7-H3 molecules. We then stained the transfected cells with either a commercial anti-mouse B7-H3 antibody (clone ab134161 , Abeam, Cat. No. EPNCIR122) or anti-human B7-H3 antibody (clone DCN.70, Biolegend, Cat. No. 331606) as positive control, P2A5 expressed in VHH-Fc format, and an isotype control (VHH to CD19). The staining of the transfected cells was then visualized by fluorescence microscopy (Figure 2). We observed that not only can P2A5 bind murine B7-H3 expressed on murine CT26 cells (Figure 2A) and human B7-H3 expressed on human MKN7 cells (Figure 2D), it could also bind murine B7H3 expressed on MKN7 cells (Figure 2B) and human B7-H3 expressed on CT26 cells (Figure 2C). Hence the cross-reactivity of P2A5 to both murine and human B7-H3 is not affected by expression on cells of different origin, and hence the glycosylation patterns of these cells. 2.3 Expression and functional cytotoxicity of B7-H3 CAR T cells

The lead anti-B7-H3 VHH candidate P2A5 was cloned into a retroviral vector pSFG as a CAR bearing a 4-1 BB spacer domain, CD28 derived transmembrane and intracellular domain, and a CD3 intracellular signalling domain (Figure 3A). Primary PBMCs from 2 healthy donors were activated on CD3- and CD28-coated plates and transduced with retrovirus particles bearing the B7- H3 CAR transgene. Efficiency of transduction was assessed by flow cytometry 6 days post transduction using labelled anti-camelid VHH antibody (Figure 3B). B7-H3 CAR T cells from both donors demonstrated good transduction efficiencies (>60%) and fold expansion (in excess of 20- fold) (Figure 3C). To determine the cytotoxic efficiency of the B7-H3 CAR T cells against B7-H3 expressing A549 non-small cell lung cancer and MDA-MB-231 triple negative breast cancer cell lines, we set up real-time cytotoxicity assays consisting of effector to target ratio of 1 :1 and 5:1 effector T cells to target cells. B7-H3 CAR T cells demonstrated good and specific killing efficacies against A549 and MDA-MB-231 cells at E:T ratios of 1 :1 and 5:1 (Figure 3D). We further assessed cytotoxic activity of B7-H3 CAR T cells in multi-rounds co-culture with B7-H3 expressing THP-1 acute myeloid leukemia cells at Effector: Target ratio of 1 :2. B7-H3 low CMK leukemia cells were used as a negative control. B7-H3 CAR T cells completely eliminated THP-1 cells through 3 sequential rounds of target encounter. In contrast, CMK cells were not lysed in the presence of B7- H3 CAR T cells (Figure 3 E).

2.4 Generation and characterisation of B7-H3-specific CAR-expressing EBVSTs

With the goal of creating off-the shelf allogeneic T cell therapy for cancer, we adopted the Virus- Specific T Cells (VSTs) platform for the manufacturing of our allogeneic B7-H3 CAR-T therapy. EBVSTs were generated by pulsing CD45RA-depleted donor PBMCs with pooled EBV pepmixes and maintained in culture media containing human platelet lysate (hPL), human IL-7 and IL-15. On day 5, EBVSTs were transduced with retroviral particles containing the above-mentioned B7-H3 CAR. Cells were re-stimulated on day 9 with irradiated MHCl/ll-knocked out universal LCL (uLCLs) in a 1 :4 EBVST:uLCL ratio. CAR EBVSTs was harvested, analysed and cryopreserved on day 16 of culture (Figure 4A). B7-H3 CAR EBVST generated from 6 donors expanded an approximate 100 to 270 -fold post-transduction (Figure 4B). B7-H3 CAR EBVSTs from all donors had good enrichment of CAR+ cells in the final product, with more than 80% of EBVSTs expressing CAR (Figure 4C). Majority of donor cells displayed enrichment in CD4+ cells in the final B7-H3 CAR EBVST product (Figure 4D).

Since B7-H3 is known to be expressed on T cells and is postulated to be an immune checkpoint molecule, we checked the B7-H3 expression of the final products and found that the CAR EBVSTs expressed lower B7-H3 than the untransduced EBVSTs (Figure 4F). Staining of known checkpoint molecules PD-1 , Tim-3 and LAG-3 revealed that the B7-H3 CAR EBVSTs expressed higher numbers of these markers over untransduced EBVSTs and the difference was greater in the CD8+ cells than the CD4+ cells (Figure 4F). In order to assess if the B7-H3 CAR EBVSTs are reactive to EBV antigens, EBVSTs were stimulated with pooled EBV-pepmixes. B7-H3 CAR EBVSTs upregulated expression of TNFa and/or IFNy in response to EBV pepmix stimulation, albeit at levels lower than the untransduced EBVSTs (Figure 4G).

2.5 In vitro cytotoxicity of B7-H3 CAR EBVSTs against B7-H3+ Cancer Cell Lines

Specific cytolysis by B7-H3 CAR EBVSTs were assessed by co-incubating EBVSTs with B7-H3+ cell lines and their B7-H3KO counterparts at an EffectorTarget ratio of 1 :1 using the xCELLigence. B7- H3 CAR EBVSTs exhibited rapid killing against DLD-1 , SW480, NCI-N87 and MDA-MB-468, with cells achieving total target lysis within 24 h of co-incubation. While cytolysis against HT29, MKN7, MKN45, MDA-MB-231 and A549 were less rapid, cells still achieved good cytolysis at end-point. Cytolysis against the B7-H3KO cell lines were delayed and attenuated, and in some donors similar to background killing levels, showing that killing of these cell lines was mediated by B7-H3 CAR activation system (Figure 5A to 5D).

To further assess the serial killing potencies of B7-H3 CAR EBVSTs, cells were subjected to multirounds of co-culture with target cells at Effector: Target ratio of 1 :2. Flow cytometry analysis revealed that B7-H3 CAR EBVSTs maintained 100% cytolysis up to 3 target encounters against B7- H3 high AML cell line THP-1, whereas cytotoxicity against B7-H3 low CMK was negligible (Figure 6A). Concomitantly, there was also an expansion in the numbers of B7-H3 CAR but not nontransduced EBVST effector cells (Figure 6B).

Serial killing potencies against solid tumor cell lines were also evaluated using the xCELLigence system. After each encounter, effector cells were counted and transferred to freshly seeded tumor cells at an EffectorTarget ratio of 1 :2. B7-H3 CAR EBVSTs displayed similar or equal killing activities for 2 serial target encounters. Subsequently, CAR EBVSTs showed a reduction in killing potencies (Figure 6C). Effector cell numbers also mostly expanded following the first 2 serial target encounters (Figure 6D).

2.6 In vivo activity of B7-H3 CAR EBVSTs against B7-H3+ tumors

2.6.1 1n vivo efficacy and safety ofB7-H3 CAR T cells in B7-H3 positive colorectal cancer

To assess the anti-tumor activity of B7-H3 CAR EBVSTs against B7-H3-expressing colorectal cancer, we implanted immunodeficient NSG-MHC l/ll DKO mice with HT-29 cells. Following tumor engraftment, mice were randomized to receive no treatment or treatment with un-transduced or B7- H3 CAR EBVSTs (Figure 7A). Body weight of mice were similar and stable across treatment groups through the 14 days post-treatment (Figure 7B).

While tumor growth in mice that received no treatment or un-transduced EBVSTs continued unbated, treatment with B7-H3 CAR EBVSTs induced significant tumor regression (Figure 7C). Likewise, endpoint flow cytometry analysis collaborated with a significantly lower population of viable HT-29 cells in tumors of mice treated with B7-H3 CAR EBVSTs, compared to untreated and untransduced mice (Figure 7D). The small population of HT-29 cells remaining in tumors of B7-H3 CAR EBVST treated mice retained high B7-H3 expression, indicating the absence of tumor antigen down-regulation (Figure 7D).

Further evaluation revealed that B7-H3 CAR EBVSTs were detected in significantly greater numbers in blood, liver, lung, and spleen of mice compared to un-transduced EBVST. More importantly, B7- H3 CAR EBVSTs in tumors were observed to be present in numbers that far exceeded untransduced EBVSTs (Figure 7E). Altogether, this demonstrates that B7-H3 CAR EBVSTs were able to migrate efficiently into the tumor site to exert their anti-tumor activity.

In vivo studies were performed to assess the anti-tumor activity of B7-H3 CAR EBVSTs against a second B7-H3-expressing colorectal cancer cell line. Immunodeficient NSG-MHC l/ll DKO mice were implanted with SW-480 cells. Following tumor engraftment, mice were randomized to receive no treatment or treatment with un-transduced or B7-H3 CAR EBVSTs (Figure 7F). Body weight of mice remained stable across treatment groups through the 15 days post-treatment (Figure 7G).

Mice treated with B7-H3 CAR EBVSTs experienced significantly improved tumor control compared to no treatment or treatment with untransduced EBVSTs (Figure 7H), consistent with endpoint analysis where significantly lower population of viable SW-480 tumor cells was observed in mice treated with B7-H3 CAR EBVSTs (Figure 7I). SW-480 tumor cells in B7-H3 CAR EBVST treated mice retained high B7-H3 expression, indicating the absence of tumor antigen down-regulation (Figure 71).

Examination revealed that greater numbers of B7-H3 CAR EBVSTs could be detected in liver, lung, and spleen of mice compared to un-transduced EBVST. Moreover, B7-H3 CAR EBVSTs in tumors were more abundant compared to untransduced EBVSTs, and they also exceeded number of B7-H3 CAR EBVSTs in other organs. Overall, this indicates that B7-H3 CAR EBVSTs were able to migrate and accumulate efficiently in tumor sites to exert their anti-tumor activity (Figure 7J).

2.6.2 In vivo efficacy and safety ofB7-H3 CAR T cells in B7-H3 positive gastric cancer

To assess the anti-tumor activity of B7-H3 CAR EBVSTs against B7-H3-expressing gastric cancer, we implanted immunodeficient NSG-MHC l/ll DKO mice with NCI-N87 cells. Following tumor engraftment, mice were randomized to receive no treatment or treatment with un- transduced or B7-H3 CAR EBVSTs (Figure 8A). Body weight of mice were similar and stable across treatment groups through the 14 days post-treatment (Figure 8B).

While tumor growth continued to increase in mice that received no treatment or un-transduced EBVSTs, treatment with B7-H3 CAR EBVSTs induced dramatic tumor regression with a hardly detectable tumor mass by endpoint (Figure 8C). Endpoint flow cytometry analysis revealed a significantly lower population of viable N-87 cells in mice treated with B7-H3 CAR EBVSTs, compared to untreated and un-transduced mice (Figure 7D). The small population of N-87 cells remaining in tumors of B7-H3 CAR EBVST treated mice retained high B7-H3 expression, indicating the absence of tumor antigen down-regulation (Figure 8D).

Further evaluation revealed that B7-H3 CAR EBVSTs were detected in significantly greater numbers in blood, liver, lung, and spleen of mice compared to un-transduced EBVST. More importantly, B7-H3 CAR EBVSTs in tumors were observed to be present in numbers that far exceeded un-transduced EBVSTs (Figure 8E). Altogether, this demonstrates that B7-H3 CAR EBVSTs were able to migrate efficiently into the tumor site to exert their anti-tumor activity.

Example 3

3.1 Biopanninq for B7H3 specific VHH

VHHs specific for B7H3 were isolated from the immunized llama library after repeated rounds of biopanning with recombinant 4lg B7H3 proteins. Increasing stringency was applied by reducing the concentrations of target proteins from 100nM in the first round, to 20nM in the second round and finally 5nM in the third round. ELISA screening for B7H3 binding was done with phage eluted from rounds 2 and 3 (Figure B1A (Figure 9A)). After sequencing to identify unique VHH sequences, 20 shortlisted leads were then expressed and purified for further screening for binding to B7H3 expressed on the surface of B7H3 expressing HepG2 cells (Figure B1 B (Figure 9B)) and for binding kinetics to recombinant 4lg B7H3 by surface plasmon resonance (Figure B1C (Figure 9C)). Only 9 leads were able to bind strongly to B7H3- expressing HepG2 cells although most leads bind to recombinant B7H3 very strongly. Notably, clone P2A5 showed strong binding to HepG2 cells while having a moderate off-rate (Figure B1 B & C (Figure 9B & C)). The binding affinity of the CAR affects not only its efficacy but also the safety of the CAR T cell (23). The risk of on-target, off-tumour toxicity can increase with higher affinity as targets such as B7H3 can be expressed at lower levels on non-malignant tissues. Hence the balance of efficacy and specificity for higher expression of B7H3 on tumour cells can be tuned by the selection of intermediate affinity and avoiding high affinity VHHs as the binding domain of the CAR. As such, we decided to focus on clone P2A5. The binding affinity of P2A5 to the 4lg form of B7H3 was further measured by multi-cycle kinetic analysis by SPR (Figure B1 D (Figure 9D)). B7H3 exists in two forms that have similar structures consisting of either a single pair of the extracellular immunoglobulin domains IgV-lgC for the shorter 21 g form or tandem pairs IgV-lgC-lgV-lgC for the larger 4 Ig form. Murine B7H3 consists of a pair of IgV-lgC domains and shares 87% sequence homology with the human 2lg form. Hence, we looked at the binding of P2A5 to both the 2lg isoform of the human B7H3 and the murine B7H3 (Figure B1 E and F (Figure 9E and F)). We saw that whilst P2A5 can bind 4lg B7H3 relatively strongly with a dissociation constant KD of around 30.9 nM, it binds both the 21g isoform and murine B7H3 about 10-fold weaker with KD of 340 nM and 473 nM respectively (Figure B1G (Figure 9G)).

3.2 Cross-reactivity of B7H3 specific P2A5 VHH to cell expressed murine and human B7H3.

B7H3 is a heavily glycosylated protein, and the glycosylation patterns can vary between human and murine cells. Hence to ascertain that the binding epitope of P2A5 on B7H3 is not influenced by the glycosylation in different cells, we transfected a B7H3 negative murine colon cancer cell line, CT26 and a gastric adenocarcinoma cell line MKN7 that has been knocked out for B7H3 expression by CRISPR/Cas9, to express both human (4lg) and murine B7H3 molecules. We then stained the transfected cells with either a commercial anti-mouse or anti-human B7H3 antibody as positive control, P2A5 expressed in VHH-Fc format, and an isotype control. The staining of the transfected cells was then visualized by fluorescence microscopy (Figure B2 (Figure 10)). We observed that not only can P2A5 bind murine B7H3 expressed on murine CT26 cells (Figure B2A (Figure 10A)) and human B7H3 expressed on human MKN7 cells (Figure B2D (Figure 10D)), it could also bind murine B7H3 expressed on MKN7 cells (Figure B2B (Figure 10B)) and human B7H3 expressed on CT26 cells (Figure B2C (Figure 10C)). Hence the cross-reactivity of P2A5 to both murine and human B7H3 is not affected by expression on cells of different origin northe glycosylation patterns of these cells.

3.3 Expression and functional cytotoxicity of B7H3.CAR T cells

The lead anti-B7-H3 VHH candidate P2A5 was cloned into a retroviral vector pSFG as a CAR bearing a 4-1 BB spacer domain, CD28-derived transmembrane and intracellular domain, and a CD3 intracellular signalling domain (Figure B3A (Figure 11 A)). Primary PBMCs from 2 healthy donors were activated and transduced with retrovirus particles bearing the B7H3.CAR transgene. Transduction efficiency was assessed by flow cytometry 6 days post transduction using labelled anti- camelid VHH antibody (Figure B3B (Figure 11 B)). B7H3.CAR T cells from both donors demonstrated good transduction efficiencies (>60%) and fold expansion (in excess of 20 fold) (Figure B3C (Figure 11C)).To determine the cytotoxic efficiency of the B7H3.CAR T cells against B7-H3 expressing A549 non-small cell lung cancer and MDA-MB-231 triple negative breast cancer cell lines, we set up realtime cytotoxicity assays consisting of effector to target ratio of 1 :1 and 5:1 effector T cells to target cells. B7H3.CAR T cells demonstrated good and specific killing efficacies against A549 and MDA- MB-231 cells at E:T ratios of 1 :1 and 5:1 (Figure B3D (Figure 11D)). We further assessed cytotoxic activity of B7H3.CAR T cells in multi-round co-cultures with B7-H3-expressing THP-1 acute myeloid leukemia cells at EffectorTarget ratio of 1 :2. B7-H3-low CMK leukemia cells were used as a negative control. B7H3.CAR T cells completely eliminated THP-1 cells through 3 sequential rounds of target encounter. In contrast, CMK cells were not lysed in the presence of B7H3.CAR T cells (Figure B3E (Figure 11 E)). 3.4 Generation and Characterisation of B7H3.CAR EBVST

With the goal of developing off-the-shelf third party T cell therapy for cancer, we adopted the Virus- Specific T Cells (VSTs) platform for the manufacturing of our allogeneic B7H3.CAR-T therapy. EBVSTs were generated by pulsing CD45RA-depleted donor PBMCs with pooled EBV pepmixes and maintained in culture media containing human platelet lysate (hPL), human IL-7 and IL-15. On day 5, EBVSTs were transduced with retrovirus bearing the B7H3.CAR transgene. Cells were restimulated on day 9 with irradiated feeder cells in a 1 :4 EBVST:feeder cell ratio. CAR EBVSTs were harvested, analysed and cryopreserved on day 16 of culture (Figure B4A (Figure 12A)). B7H3.CAR EBVSTs generated from 6 donors expanded an approximate 100 to 270-fold post-transduction (Figure B4B (Figure 12B)). B7H3.CAR EBVSTs from all donors had good enrichment of CARpositive cells in the final product, with more than 80% of EBVSTs expressing the B7H3.CAR (Figure B4C (Figure 12C)). Majority of donor cells displayed enrichment in CD4+ cells in the final B7H3.CAR EBVST product (Figure B4D (Figure 12D)).

Since B7-H3 is known to be expressed on T cells and is postulated to be an immune checkpoint molecule, we checked the B7-H3 expression of the final products and found that the B7H3.CAR EBVSTs expressed lower B7-H3 than the untransduced EBVSTs (Figure B4F (Figure 12F)). Staining of known checkpoint molecules PD-1 , Tim-3 and LAG-3 revealed that the B7H3.CAR EBVSTs expressed higher numbers of these markers over untransduced EBVSTs and the difference was greater in the CD8+ cells than the CD4+ cells (Figure B4F (Figure 12F)).

In order to assess if the B7H3.CAR EBVSTs are reactive to EBV antigens, we stimulated the EBVSTs with pooled EBV-pepmixes. B7H3.CAR EBVSTs express good TNFa and/or IFNy in response to EBV pepmix stimulation, albeit at levels lower than the untransduced EBVSTs (Figure B4G (Figure 12G)). In contrast, background or non-specific cytokine production in response to media or irrelevant HIV peptides was observed to be low.

3.5 In Vitro Cytotoxicity of B7H3.CAR EBVSTs against B7-H3+ Cancer Cell Lines

Specific cytolysis by B7H3.CAR EBVSTs was assessed by co-incubating EBVSTs with B7-H3+ cell lines and their B7-H3KO counterparts at an EffectorTarget ratio of 1 :1 , with real-time cell analysis performed using the xCELLigence system. B7H3.CAR EBVSTs exhibited rapid and efficient killing of multiple colorectal (DLD-1 , HT-29, SW480), gastric (NCI-N87, MKN-7, MKN-45), triple-negative breast (MDA-MB-231 , MDA-MB-468) and non-small cell lung (A549, NCI-H1299, NCI-H23, NCI- H596) cancer cell lines, within 48h of co-incubation. Cytolysis against the B7H3KO cell lines was delayed and attenuated, and in some donors similar to background killing levels (Figure B5A to B5D (Figure 13 A to D)). Quantification of IFNylevels in the cell supernatants after effectortarget coculture also showed B7H3.CAR EBVST responses against the wild-type tumor cell lines, but not the B7H3KO cell lines, showing that the B7H3.CAR responses are specific to B7H3 expression (Figure B5E (Figure 13E)). In addition, when we co-cultured EBVSTs expressing the full length and truncated B7H3.CAR with wild-type MKN-45 tumor cells, we noted that killing of MKN-7 cells was only observed with EBVSTs bearing B7H3 CAR with intact intracellular signalling domain (Figure B5F (Figure 13F)). Data from these studies indicates that tumor cell killing by B7H3.CAR EBVSTs must be mediated by B7H3 recognition coupled with CAR activation.

To assess the serial killing potencies of B7H3.CAR EBVSTs, cells were subjected to multi-rounds of co-culture with target cells at Effector: Target ratio of 1 :2. Flow cytometry analysis revealed that B7H3.CAR EBVSTs maintained 100% cytolysis up to 3 target encounters against B7-H3 high AML cell line THP-1 , whereas cytotoxicity against B7-H3-low CMK was negligible (Figure B6A (Figure 14A)). Concomitantly, there was also an expansion in the numbers of B7H3.CAR but not untransduced EBVST effector cells (Figure B6B (Figure 14B)).

Serial killing potencies against solid tumor cell lines were also evaluated using the xCELLigence system. After each encounter, effector cells were counted and transferred to freshly seeded tumor cells at an EffectorTarget ratio of 1 :2. B7H3.CAR EBVSTs displayed similar or equal killing activities for 2 serial target encounters. Subsequently, CAR EBVSTs showed a reduction in killing potencies (Figure B6C (Figure 14C)). Effector cell numbers also mostly expanded following the first 2 serial target encounters (Figure B6D (Figure 14C)).

3.6 B7H3.CAR EBVSTs targets Myeloid-Derived Suppressor Cells (MDSCs)

To determine if MDSCs express B7-H3 and are targeted by B7H3.CAR EBVSTs, we first differentiated MDSCs in vitro by stimulating primary CD14+ monocytes with GM-CSF and IL-6 for 7 days and stained for surface expression of B7-H3 using the anti-B7-H3 antibody. MDSCs were found to express high levels of B7-H3 , along with suppressor molecules IL-10, TGFpand iNOS (Figure B7A (Figure 15A)). Notably, B7-H3 expression on MDSCs were in the same level of magnitude as the cancer cell line, NCI-H1299. MDSCs were also verified to be immune-suppressive based on the observation that they were able to reduce the cytotoxicity of a CD30.CAR EBVST (Figure B7B (Figure 15B)), Co-culture of B7H3.CAR EBVSTs with MDSCs led to a dose-dependent cytolysis of MDSCs, as opposed to the low levels of MDSCs cytolysis observed in co-cultures with UT EBVSTs (Figure B7C (Figure 15C)). To further investigate the effect of MDSC targeting on T cell function, we stimulated untransduced or B7H3.CAR EBVSTs with anti-CD3/CD28 in the presence of MDSCs. While proliferation of untransduced EBVSTs were inhibited by MDSCs, MDSC inhibition of B7H3.CAR EBVSTs was significantly less as observed in the higher proliferation index (Figure B7D (Figure 15D)). This suggests that targeting of B7-H3-expressing MDSCs by B7H3.CAR EBVSTs led to an alleviation of MDSC-induced inhibition of T cell proliferation, leading to improved T cell function. 3.7 In Vivo Activity of B7H3.CAR EBVSTs against B7-H3+ Tumors

3.7.1 In-vivo efficacy and safety of B7H3.CAR T cells in B7-H3 positive colorectal cancer

To assess the anti-tumor activity of B7H3.CAR EBVSTs against B7-H3-expressing colorectal cancer, we implanted immunodeficient NSG-MHC l/ll DKO mice with HT-29 cells. Following tumor engraftment, mice were randomized to receive no treatment or treatment with un-transduced or B7H3.CAR EBVSTs (Figure B8A (Figure 16A)). Body weights of mice were similar and stable across treatment groups throughout the 14 days post-treatment (Figure B8B (Figure 16B)).

While tumor growth in mice that received no treatment or un-transduced EBVSTs continued unbated, treatment with B7H3.CAR EBVSTs induced significant tumor regression (Figure B8C (Figure 16C)). Likewise, endpoint flow cytometry analysis corroborated these findings with a significantly lower population of viable HT-29 cells in tumors of mice treated with B7H3.CAR EBVSTs, compared to untreated and un-transduced controls (Figure B8D (Figure 16D)). The small population of HT-29 cells remaining in tumors of B7H3.CAR EBVST treated mice retained high B7- H3 expression, indicating the absence of tumor antigen down-regulation (Figure B8D (Figure 16D)).

Further evaluation revealed that B7H3.CAR EBVSTs were detected in significantly greater numbers in blood, liver, lung, and spleen of mice compared to un-transduced EBVST. More importantly, B7H3.CAR EBVSTs in tumors were observed to be present in numbers that far exceeded un- transduced EBVSTs (Figure B8E (Figure 16E)). Altogether, this demonstrates that B7H3.CAR EBVSTs were able to migrate efficiently into the tumor site to exert their anti-tumor activity.

We next performed in vivo studies to assess the anti-tumor activity of B7H3.CAR EBVSTs against a second B7-H3-expressing colorectal cancer cell line. To this end, we implanted immunodeficient NSG-MHC l/ll DKO mice with SW-480 cells. Following tumor engraftment, mice were randomized to receive no treatment or treatment with un-transduced or B7H3.CAR EBVSTs (Figure B8F (Figure 16F)). Body weight of mice remained stable across treatment groups through the 15 days posttreatment (Figure B8G (Figure 16G)).

Mice treated with B7H3.CAR EBVSTs experienced significantly improved tumor control compared to no treatment or treatment with untransduced EBVSTs (Figure B8H (Figure 16H)), consistent with endpoint analysis where significantly lower population of viable SW-480 tumor cells was observed in mice treated with B7H3.CAR EBVSTs (Figure B8I (Figure 161)). SW-480 tumor cells in B7H3.CAR EBVST treated mice retained high B7-H3 expression, indicating the absence of tumor antigen downregulation (Figure B8I (Figure 161)).

Examination revealed that greater numbers of B7H3.CAR EBVSTs could be detected in liver, lung, and spleen of mice compared to un-transduced EBVST. More importantly, B7H3.CAR EBVSTs in tumors were more abundant compared to untransduced EBVSTs and also exceeded number of B7H3.CAR EBVSTs in other organs. Overall, this indicates that B7H3.CAR EBVSTs were able to migrate and accumulate efficiently in tumor sites to exert their anti-tumor activity (Figure B8J (Figure 16J)). 3.7.2 In-vivo efficacy and safety of B7H3.CAR T cells in B7-H3 positive gastric cancer

To assess the anti-tumor activity of B7H3.CAR EBVSTs against B7-H3-expressing gastric cancer, we implanted immunodeficient NSG-MHC l/ll DKO mice with NCI-N87 cells. Following tumor engraftment, mice were randomized to receive no treatment or treatment with un-transduced or B7H3.CAR EBVSTs (Figure B9A (Figure 17A)). Body weights of mice were similar and stable across treatment groups throughout the 14 days post-treatment (Figure B9B (Figure 17B)).

While tumor growth continued to increase in mice that received no treatment or un-transduced EBVSTs, treatment with B7H3.CAR EBVSTs induced dramatic tumor regression with a hardly detectable tumor mass by endpoint (Figure B9C (Figure 17C)). Endpoint flow cytometry analysis revealed a significantly lower population of viable N-87 cells in mice treated with B7H3.CAR EBVSTs, compared to untreated and un-transduced mice (Figure B9D (Figure 17D)). The small population of N-87 cells remaining in tumors of B7H3.CAR EBVST treated mice retained high B7-H3 expression, indicating the absence of tumor antigen down-regulation (Figure B9D (Figure 17D)).

Further evaluation revealed that B7H3.CAR EBVSTs were detected in significantly greater numbers in blood, liver, lung, and spleen of mice compared to un-transduced EBVST. More importantly, B7H3.CAR EBVSTs in tumors were observed to be present in numbers that far exceeded un- transduced EBVSTs (Figure B9E (Figure 17E)). Altogether, this demonstrates that B7H3.CAR EBVSTs were able to migrate efficiently into the tumor site to exert their anti-tumor activity.

3.7.3. In-vivo efficacy and safety of B7H3.CAR T cells in B7-H3 positive non-small cell lung cancer

To assess the anti-tumor activity of B7H3.CAR EBVSTs against B7-H3-expressing non-small cell cancer, we implanted immunodeficient NSG-MHC l/ll DKO mice with NCI-H1299 cells. Following tumor engraftment, mice were randomized to receive no treatment or treatment with un-transduced or B7H3.CAR EBVSTs (Figure B10A (Figure 18A)). Body weights of mice were similar and stable across treatment groups throughout the 14 days post-treatment (Figure B10B (Figure 18B)).

Treatment of mice with B7H3.CAR EBVSTs induced slower tumor growth as compared to mice untreated or treated with control untransduced EBVSTs (Figure B10C (Figure 18C)). Endpoint flow cytometry analysis revealed significantly lower numbers of viable H1299 tumor cells in mice treated with B7H3.CAR EBVSTs, compared to untreated and un-transduced mice (Figure B10D (Figure 18D)). There were also no significant differences in the expression levels of B7-H3 on viable H1299 tumor cells in all 3 treatment groups, indicating the absence of tumor antigen down-regulation (Figure B10D (Figure 18D)). 3.7.4. In-vivo efficacy and safety of B7H3.CAR T cells in B7-H3 positive triple-negative breast cancer

To assess the anti-tumor activity of B7H3.CAR EBVSTs against B7-H3-expressing triple-negative breast cancer, we implanted immunodeficient NSG-MHC l/ll DKO mice with MDA-MB-468 cells. Following tumor engraftment, mice were randomized to receive no treatment or treatment with untransduced or B7H3.CAR EBVSTs (Figure B11A (Figure 19A)). While there were slight weight decreases in mice treated with B7H3.CAR EBVSTs, it was not severe and remained within less than 20% loss. (Figure B11 B (Figure 19B)).

While tumor growth continued to increase in mice that received no treatment or un-transduced EBVSTs, treatment with B7H3.CAR EBVSTs induced dramatic tumor regression with a hardly detectable tumor mass by endpoint (Figure B11 C (Figure 19C)). Endpoint flow cytometry analysis revealed a significantly lower population of viable MDA-MB-468 cells in mice treated with B7H3.CAR EBVSTs, compared to untreated and un-transduced mice (Figure B11 D (Figure 19D)).

3.7.5 In-vivo efficacy and safety of B7H3.CAR T cells in B7-H3 positive breast cancer patient-derived xenograft (PDX)

In an effort to assess the efficacy of B7H3.CAR EBVST in a model that more accurately represent the biology and heterogeneity of human cancer, we identified a breast cancer patient-derived xenograft (PDX) that is positive for B7-H3 (Figure B12A and B (Figure 20 A and B)). The PDX is heterogenous in B7-H3 expression, with different areas expressing varying levels of B7-H3.

Once NSG mice were successfully engrafted with the breast cancer PDX, mice were randomized to receive no treatment or treatment with un-transduced or B7H3.CAR EBVSTs (Figure B13A (Figure 21 A)). Body weight decreases in mice treated with B7H3.CAR EBVSTs were mild and recovered at later time-points. (Figure B13B (Figure 21 B)).

While tumor growth continued to increase in mice that received no treatment or un-transduced EBVSTs, treatment with B7H3.CAR EBVSTs induced dramatic tumor regression with hardly detectable tumors by endpoint (Figure B13C (Figure 21 C)). Endpoint flow cytometry analysis revealed a significantly lower population of viable tumor cells in mice treated with B7H3.CAR EBVSTs, compared to untreated and un-transduced mice (Figure B13C (Figure 21 C)).

3.8 Stimulation and Competition of B7H3.CAR EBVST by Soluble B7-H3

Circulating soluble B7-H3 has been detected in cancer patients. To investigate whether B7H3.CAR EBVSTs can be non-specifically activated by soluble B7-H3, we co-incubated B7H3.CAR EBVSTs with plate-bound or soluble B7-H3 in both the 4-lg and 2-lg forms. While B7H3.CAR EBVSTs secreted IFNyand TNFaand upregulated CD25 in response to stimulation by plate-bound B7-H3, B7H3.CAR EBVSTs were not activated by soluble formats of B7-H34-lg and 2-lg even at high concentrations (Figure B14A (Figure 22A)). To further investigate if soluble B7-H3 diminishes B7H3.CAR EBVST cytotoxicity against B7H3- positive tumor cells, we co-incubated B7H3.CAR EBVSTs with tumor cell in the presence of soluble B7-H3. Both 2-lg and 4-lg forms of soluble B7-H3 did not block cytolysis of NCI-N87 or NCI-H1299 tumor cells by B7H3.CAR EBVST and maximal killing of tumor cells were achieved at assay endpoint (Figure B14B (Figure 22B)).

3.9 Safety of B7H3.CAR EBVST against Hematopoietic Stem and Progenitor Cells (HSPCs)

To investigate if B7H3.CAR EBVSTs target HSPCs, we first assessed B7H3 expression on cord blood CD34+ cells from two donors following the same FLT3L, SCF and TPO cytokine stimulation protocol employed by Hornbach et al. In both donors, B7H3 expression was undetectable on resting CD34+ HSPCs but progressively increased from day 3 to 14 of stimulation, albeit not at the high levels observed in NCI-N87 tumor cells (Figure B15A and B (Figure 23A and B)). B7H3 upregulation was detected across all lineage subsets of multipotent progenitors (MPPs), lymphoid-primed multipotents (LMPPs), granulocyte and macrophage-primed progenitors (GMPs) and multi-lymphoid primed progenitors (MLPs), with the largest increase observed in erythro-myeloid primed progenitors (EMPs) (Figure B15B (Figure 23B)). Co-culture of resting or cytokines-stimulated CD34+ HSPCs with B7H3.CAR EBVSTs led to the cytolysis of the more matured cytokine-stimulated HSPCs, but not the unstimulated or 3 day-stimulated HSPCs (Figure B15C (Figure 23C)). Further analysis of the HSPC population revealed that the reduction of MPP and LMPP populations was more significant (Figure B15D (Figure 23D)). To investigate if B7H3.CAR EBVSTs affect the erythroid and myeloid development of early differentiated HSPCs, we co-cultured HSPCs that have stimulated with cytokines for 3 days with B7H3.CAR EBVSTs. As shown in Figure B15D (Figure 23D), B7H3.CAR EBVSTs did not affect the differentiation capacity of HSPCs, evident by the consistent lack of major reductions of any mature progenitor subsets in colony-forming unit (CFU) assays (Figure B15D (Figure 23D)).

3.10 Safety of B7H3.CAR EBVST against Peripheral Blood Mononuclear Cells (PBMCs)

In order to evaluate B7H3.CAR EBVST on-target off-tumor targeting of human immune cells, we examined B7-H3 expression on PBMC subsets after stimulation with various inflammatory cytokines. Up to approximately 80% of monocytes were found to express B7-H3 on their cell surface, albeit at levels lower than the cancer cell line NCI-N87 (Figure B16A (Figure 24A)). In contrast, T, B and NK cells do not express significant levels of B7-H3 on the cell surface. To verify that the T, B and NK cell subsets are not targeted by B7H3.CAR EBVST, we co-cultured PBMCs with allogeneic untransduced or B7H3.CAR EBVST, in the presence of various cytokines and found that there was no loss in T, B and NK cell populations after co-culture with B7H3.CAR EBVST (Figure B16B (Figure 24B)). When we co-cultured purified monocytes with allogeneic untransduced EBVST, we found that monocytes expressed exceptionally high levels of B7-H3 (Figure B16C (Figure 24C)), likely due to an allogeneic response. Co-culturing of monocytes with allogeneic B7H3.CAR EBVST led to significantly higher cytotoxicities of the target monocytes over untransduced EBVST cells (Figure B16D (Figure 24D)).

3.11 Safety of B7H3.CAR EBVST against Antigen Experienced T Cells

In order to evaluate the safety of B7H3.CAR EBVST against antigen-experienced T cells, we stimulated EBVST with HIV or EBV-pepmix pulsed antigen-presenting cells and checked for B7-H3 expression. While there was antigen specific activation in the stimulated EBVST, as seen in EBVST proliferation and cytokine secretion (Figure B17A (Figure 25A)), B7-H3 expression was low on these stimulated cells (Figure B17B (Figure 25B)), suggesting that B7H3.CAR EBVST would minimally target memory T cells even after antigen-specific activation.

3.12 In Vivo Safety Murine Model

Based on the observation that P2A5 is cross-reactive to mouse B7-H3, we sought to test the potential toxicities of B7H3 CAR in an immunocompetent and syngeneic murine model. To generate the murine B7H3 CAR (P2A5 mCAR), we replaced the human CD28 transmembrane and signaling domains and CD3 domain with the murine homologs, while keeping the antigen recognition anti-B7- H3 VHH domain and 4-1 BB spacer. T cells from congenic CD45.1 C57BL/6J mouse splenocytes were activated and retrovirally transduced with the P2A5 mCAR. To verify activity of P2A5 mCAR-T cells, we assessed the cytotoxicity of P2A5 mCAR-T against the target cell lines B16F10-WT, B16F10WT that was retrovirally transduced with human B7-H3 (B16F10-hB7H3) and B16F10WT that was transiently transfected with murine B7-H3 (B16F10-mB7H3). A dose-dependent cytolysis of B16F10-hB7H3 and B16F10-mB7H3 cells but not B16F10-WT cells was observed, (Figure B18A (Figure 26A)), demonstrating that P2A5 mCAR-T displayed specific cytotoxic activity against both human and mouse B7-H3-expressing tumor cells.

To investigate the safety of P2A5 mCAR-T cells, wild-type C57BL/6J mice xenografted with B16F10WT-hB7H3 tumors were first lympho-depleted by total body irradiation at 5 Gy before treating them with untransduced or P2A5 mCAR murine T cells. For treatment controls, irradiated tumor bearing mice were either left untreated or administered untransduced T cells (Figure B18B (Figure 26B)). While tumors continued to grow in mice that received no treatment or untransduced T cells, treatment with P2A5 mCAR-T significantly controlled tumor growth, leading to significantly smaller tumors at study endpoint (Figure B18C (Figure 26C)). Although weight loss was observed in mice treated with P2A5 mCAR-T cells compared to mice that received no treatment or untransduced T cells, this was mild and quickly resolved (Figure B18D (Figure 26D)). Endpoint analysis revealed that mice receiving P2A5-mCAR-T only experienced significant reduction of T cells in the bone marrow and NK cells in the bone marrow and spleen. No significant losses of other hematological subsets were detected (Figure D18E (Figure 26E)). Further analyses of the mouse serum showed that mouse cytokines IFN-y, IL-10 and chemokines CXCL9, CXCL10, Eotaxin, CCL2 and CCL4 was significantly elevated in mice treated with P2A5- mCAR-T at day 3 post-treatment. However, the levels dropped to almost baseline levels by day 6 suggesting that this was an acute and transient response associated with CAR treatment, (Figure B18F (Figure 26F))).

Taken together, there were no evidence of severe toxicities experienced by mice treated with P2A5- mCAR.

3.13 In vivo Cytokine Release Syndrome (CRS) model in humanised mice

To understand if B7H3.CAR EBVST treatment was associated with potential CRS toxicities, we generated humanised mice by injecting NSG-SGM3 mice with CD34+ cord blood cells. Following successful humanisation, HT-29 tumor or luciferase-expressing NALM-6 leukemia cell lines were engrafted subcutaneously onto flanks or intravenously into mice, respectively. When HT-29 tumors became palpable, mice were stratified according to level of humanization and randomized to receive allogeneic untransduced or B7H3.CAR EBVSTs or left untreated. In parallel, untransduced or CD19.CAR ATCs were administered to mice engrafted with a high burden of NALM-6 leukemia (Figure B19A (Figure 27 A)). Both CD19.CAR ATCs and B7H3.CAR EBVSTs elicited good anti-tumor responses evident in the lower tumor burden compared to the untreated or untransduced T cells treated groups (Figure B19B (Figure 27B)). CD19.CAR ATC-treated mice showed a rapid and unremitting weight loss of more than 20% by and beyond 4 days post-treatment, whereas B7H3.CAR EBVST-treated mice experienced a weight loss at 4 days post-treatment from which they quickly recovered (Figure B19C (Figure 27C)). Serum levels of human cytokines IL-6, IL-8, IL-10, IFN-y, TNF-aand IL-2 in B7H3.CAR EBVST-treated mice were significantly lower than CD19.CAR ATC-treated mice at 3-days post-treatment, suggesting that cytokine release following B7H3.CAR EBVST treatment was contained (Figure B19D (Figure 27D)). Analyses of hematological subsets in various organs of humanised mice at study endpoint revealed that myeloid cells such as neutrophils, monocytes and dendritic cells were greatly diminished in the B7H3.CAR EBVST-treated group (Figure B19E (Figure 27E)), with reduction largely occurring in the B7-H3 expressing population (Figure B19F (Figure 27F)). Altogether, this supports the notion that B7H3.CAR EBVSTs target myeloid cells in a B7H3-dependent manner and by diminishing these primary cellular mediators of CRS (24), attenuate the risk of CAR T cell therapy associated CRS.

3.14. Impact of Invention

In contrast to its limited protein expression in normal human tissues and lymphoid organs, B7-H3 is aberrantly expressed in a high proportion of human malignancies. Here we describe the identification and utilization of a novel B7-H3 targeting nanobody, with superior biophysical qualities over scFv, as the antigen specific domain in our CAR construct. T cells and Epstein-Barr Virus Specific T cells (EBVSTs) expressing this nanobody-based B7H3.CAR demonstrated good and specific in-vitro and in-vivo activity and an excellent safety profile in B7-H3 positive malignancies. Additional data supports that this B7H3.CAR EBVST candidate targets myeloid-derived suppressor cells (MDSCs) which have high B7-H3 expression and in doing so, abrogates the immunosuppressive effects of these MDSCs. Lastly, the data reinforces that by targeting B7H3 expressing myeloid cells, B7H3.CAR EBVST cell therapy can avert the attendant risk of cytokine release syndrome in CAR T cell therapy.

In summary, the excellent anti-tumor efficacy, good safety profile and ability to ablate immunosuppressive myeloid cells shown by this B7H3. CAR. EBVSTs will provide an attractive and alternate avenue to deliver allogeneic, off-the-shelf CAR T cell therapy to patients with B7-H3 positive malignancies.

References

A number of publications are cited above in order to more fully describe and disclose the invention and the state of the art to which the invention pertains. Full citations for these references are provided below. The entirety of each of these references is incorporated herein.

1) Kontos F, Michelakos T, Kurokawa T, Sadagopan A, Schwab JH, Ferrone CR et al. B7-H3: An Attractive Target for Antibody-based Immunotherapy. Clin Cancer Res. 2021 Mar 1 ;27(5):1227- 1235.

2) Flem-Karlsen K, Fodstad O, Tan M, Nunes-Xavier CE. B7-H3 in Cancer -Beyond Immune Regulation. Trends Cancer 2018; 4:401-4.

3) Human Protein Atlas. https://www.proteinatlas.org/ENSG00000103855-CD276

4) Zhou WT and Jin WL. B7-H3/CD276: An Emerging Cancer Immunotherapy. Front Immunol. 2021 Jul 19;12:701006.

5) Seaman S, Zhu Z, Saha S, Zhang X, Yang MY, Hilton MB et al. Eradication of Tumors through Simultaneous Ablation of CD276/B7-H3-Positive Tumor Cells and Tumor Vasculature. Cancer Cell. 2017 Apr 10;31 (4):501 -515

6) Xie, C. et al. Soluble B7-H3 promotes the invasion and metastasis of pancreatic carcinoma cells through the TLR4/NF-kappaB pathway. Sci. Rep 2016. 6, 27528

7) Li Y, Zhang J, Han S, Qian Q, Chen Q, Liu L et al. B7-H3 promotes the proliferation, migration and invasiveness of cervical cancer cells and is an indicator of poor prognosis. Oncol Rep. 2017 Aug;38(2):1043-1050

8) Zhou L and Zhao Y. B7-H3 Induces Ovarian Cancer Drugs Resistance Through An PI3K/AKT/BCL-2 Signaling Pathway. Cancer Manag Res. 2019 Dec 3;11 :10205-10214

9) Du H, Hirabayashi K, Ahn S, Kren NP, Montgomery SA, Wang X et al. Antitumor Responses in the Absence of Toxicity in Solid Tumors by Targeting B7-H3 via Chimeric Antigen Receptor T Cells. Cancer Cell. 2019 Feb 11 ;35(2):221-237. 10) Theruvath JL, Sotillo E, Mount CW, Graef CM, Delaidelli A, Heitzeneder S et al.

Locoregionally administered B7-H3-targeted CAR T cells for treatment of atypical teratoid/rhabdoid tumors. Nat Med. 2020 May;26(5):712-719.

11) Majzner R, Theruvath JL, Nellan A, Heitzeneder S, Cui Y, Mount CW et al. CAR T Cells Targeting B7-H3, a Pan-Cancer Antigen, Demonstrate Potent Preclinical Activity Against Pediatric Solid Tumors and Brain Tumors. Clin Cancer Res. 2019 Apr 15;25(8):2560-2574

12) Litchman El, Du H, Shou P, Song F, Suzuki K, Ahn S et al. Preclinical Evaluation of B7-H3- specific Chimeric Antigen Receptor T Cells for the Treatment of Acute Myeloid Leukemia. Clin Cancer Res. 2021 Jun 1 ;27(11):3141 -3153

13) Zhang Y, He L, Sadogopan, Ma T, Dotti G, Wang F et al. Targeting Radiation-Resistant Prostate Cancer Stem Cells by B7-H3 CAR T Cells. Mol Cancer Ther. 2021 Mar;20(3):577-588

14) Huang B, Luo L, Wang J, He B, Feng R, Xian N et al. B7-H3 specific T cells with chimeric antigen receptor and decoy PD-1 receptors eradicate established solid human tumors in mouse models. Oncoimmunology. 2019 Nov 4;9(1):1684127

15) Zhang Z, Jiang C, Liu Z, Yang M, Tang X, Wang Y et al. B7-H3-Targeted CAR-T Cells Exhibit Potent Antitumor Effects on Hematologic and Solid Tumors. Mol Ther Oncolytics. 2020 Apr 7;17:180-189

16) Yang S, Cao B, Zhou G, Zhu L, Wang L, Zhang L et al. Targeting B7-H3 Immune Checkpoint

With Chimeric Antigen Receptor-Engineered Natural Killer Cells Exhibits Potent Cytotoxicity Against Non-Small Cell Lung Cancer. Front Pharmacol. 2020 Jul 30;11 :1089

17) Long AH, Haso WM, Shern JF, Wanhainen KM, Murgai M, Ingaramo M et al. 4-1 BB costimulation ameliorates T cell exhaustion induced by tonic signaling of chimeric antigen receptors. Nat Med. 2015 Jun;21 (6):581-90.

18) Fujiwara K, Masutani M, Tachibana M, Okada N. Impact of scFv structure in chimeric antigen receptor on receptor expression efficiency and antigen recognition properties. Biochem Biophys Res Commun. 2020 Jun 25;527(2):350-357

19) M Frigault M, Lee J, Basil MC, Carpenito C, Motohashi S, Scholler J. Identification of chimeric antigen receptors that mediate constitutive or inducible proliferation of T cells. Cancer Immunol Res 2015;3:356-67.

20) Jayapriya J. Mellody MP, Hou AJ, Desai RP, Fung AW, Pham AHT. CAR-T design: Elements and their synergistic function. EBioMedicine. 2020 Aug;58:102931.

21) Ingram J, Schmidt F, Ploegh H. Exploiting nanobodies’ singular traits. Annu Rev Immunol. 2018;36:695-715.

22) Bao C, Gao Q, Li L, Han L, Zhang B, Ding Y. The Application of Nanobody in CAR-T Therapy. Biomolecules. 2021 Feb 8;11 (2):238.

23) Sterner, R.C., Sterner, R.M. CAR-T cell therapy: current limitations and potential strategies. Blood Cancer J. 2021 ; 11 , 69 24) Norelli M, Camisa B, Giulia B, Falcone L, A Purevdorj A, Genua M, Monocyte-derived IL-1 and IL-6 are differentially required for cytokine-release syndrome and neurotoxicity due to CAR T cells. Nat Med 2018 Jun;24(6):739-748. For standard molecular biology techniques, see Sambrook, J., Russel, D.W. Molecular Cloning, A Laboratory Manual. 3 ed. 2001 , Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press