product

Abstract

The invention relates to binding molecules which bind to a MHC-presented peptide of MAGE- A3 with reduced cross-reactivity relative to known anti-MAGE-A3 TCRs.

Description

[0001] PRODUCT

[0002] Field of the Invention

[0003] The invention relates to binding molecules for major histocompatibility complex (MHC)- restricted peptide antigen MAGE -A3, and methods and uses thereof.

[0004] Background

[0005] High-affinity, specific binding interactions between a membrane -bound or soluble binding molecule and a pathological antigen can be exploited to engineer novel therapeutic avenues. For example, cell therapies typically leverage high-affinity binding interactions between cell-presented receptors such as T cell receptors (TCRs) or chimeric antigen receptors (CARs) targeted against a cognate pathological peptide. The receptors may be heterologous receptors (e.g. for adoptive cell therapy) or modified autologous receptors (e.g. for autologous cell therapy). Meanwhile, antibodybased therapies typically leverage high-affinity binding interactions between an antibody, antibody fragment or antibody variant and a pathological antigen. Chimeric or solubilized forms of TCRs or T cells expressing the TCRs can also be applied against pathological antigens.

[0006] TCRs bind short peptide antigens presented via the major histocompatibility complex (as a ‘pMHC’). Endogenous T cells undergo screening for off-target binding against self-antigens during maturation in a process known as thymic selection; T cells bearing TCRs which display affinity for canonical self-antigens are eliminated such that mature, circulating T cells respond only to foreign antigens. However, exogenously-introduced T cells bypass thymic selection and functional central tolerance processes, increasing the risk of off-target interactions with self antigens. Although these off-target interactions are usually of far lower affinity than the binding interaction with the cognate target pMHC, they can nevertheless prove functionally significant in vivo and prevent meaningful clinical application of the TCR.

[0007] For example, the endogenous A3 TCR (also referred to as for example wildtype A3 TCR or WT A3 TCR or wt a3 herein) targets the cancer-testis antigen MAGE-A3. The A3 TCR was manipulated via engineering of its six complementary determining regions (CDRla, CDR2 a, CDR3 a, CDRip, CDR2P and CDR3P) to create the engineered a3a TCR (which can also be referred to as for example A3A or a3a), which displayed higher binding affinity and higher sensitivity for MAGE- A3. The a3a TCR was described for example in WO 2012013913 Al, which is incorporated herein by reference in its entirety. Despite initial promise in vitro and in pre-clinical trials, a3a resulted in fatalities in clinical trials, which have been attributed subsequently to off-target binding of the a3a TCR to the self-antigen Titin (Linette, Gerald P et al. “Cardiovascular toxicity and titin cross- reactivity of affinity -enhanced T cells in myeloma and melanoma.” Blood vol. 122,6 (2013): 863-71.; Cameron, Brian J et al. “Identification of a Titin-derived HLA-A1 -presented peptide as a cross-reactive target for engineered MAGE A3-directed T cells.” Science translational medicine vol. 5,197 (2013): 197ral03). This off-target, self-antigen sensitivity was not picked up in pre-clinical safety screening, likely due to its lower binding affinity. Herein, the inventors also demonstrate that a3a displays off-target binding to other self-antigens, which is not seen in the wild-type (WT) A3 TCR and which likely contributed to the clinical fatalities.

[0008] Despite that a3a binds to Titin (and other self-antigens) with only low affinity, said off- target binding interactions are sufficiently sensitive to prohibit use of a3a in the clinic. To date, no anti-MAGE-A3 TCR with an acceptable safety profile is available.

[0009] Summary

[0010] The invention provides anti-MAGE-A3 TCRs and anti-MAGE-A3 TCR-based binding molecules with reduced cross-reactivity for self-antigens expressed in healthy tissues (i.e. reduced self-reactivity), and which therefore possess an improved clinical safety profile.

[0011] Advantageously and unexpectedly, the inventors show that binding to self-antigen Titin can be reduced or essentially completely abolished by mutation of the a3a TCR alpha chain complementary determining region 2 (CDR2a) as described herein. Said mutations lead to reduced or abolished CD8+ T cell mediated cytotoxicity against Titin.

[0012] The inventors also show that binding to other self-antigens including CD 166 can be reduced or essentially completely abolished by mutation of the a3a TCR alpha chain complementary determining region 2 (CDR2a) as described herein. Said mutations likewise reduce or abolish CD8+ T cell mediated cytotoxicity against these other self-antigens e.g. CD166.

[0013] Even more surprisingly, the inventors show that the cross-reactivity and self-reactivity displayed by the a3a TCR (against various self-antigens) can be reduced by targeted mutation of the CDR2a as described herein, and can even be reduced to a level equivalent to that displayed by the endogenous WT A3 TCR, representing a best-in-class safety profile. Said reduced cross- and self-reactivity is achieved while still maintaining sensitivity against target antigen MAGE-A3 and cytotoxic activity against cells presenting the MAGE -A3 peptide by the HLA-A*01:01 complex.

[0014] The binding molecules described herein are far more sensitive for MAGE -A3 than the WT A3 TCR, but far less cross-reactive (and far less self-reactive) than known anti-MAGE-A3 TCRs. Some binding molecule embodiments are essentially no more cross-reactive than the endogenous WT A3 TCR (z. e. some binding molecule embodiments only demonstrate in vitro sensitivity to a variety of the same off-target peptide self-antigens as WT A3 TCR).

[0015] Accordingly, the invention provides a binding molecule comprising:

[0016] (A) a T cell receptor (TCR) alpha chain variable domain comprising:

[0017] (i) a complementary determining region (CDR) 1 comprising an amino acid sequence DSAIYN (SEQ ID NO: 1) or comprising an amino acid sequence differing from SEQ ID NO: 1 by one amino acid ,

[0018] (ii) a CDR2 comprising the amino acid sequence of Formula I (SEQ ID NO: 2), and

[0019] (iii) a CDR3 comprising the amino acid sequence CAVRPGGAGSYQLTF (SEQ ID NO: 3) or comprising an amino acid sequence differing from SEQ ID NO: 3 by one amino acid; and

[0020] (B) a TCR beta chain variable domain comprising:

[0021] (i) a CDR1 comprising the amino acid sequence SGHRS (SEQ ID NO: 4) or comprising an amino acid sequence differing from SEQ ID NO: 4 by one amino acid,

[0022] (ii) a CDR2 comprising the amino acid sequence YFSETQ (SEQ ID NO: 5) or comprising an amino acid sequence differing from SEQ ID NO: 5 by one amino acid, and

[0023] (iii) a CDR3 comprising the amino acid sequence CASSPNMADEQYF (SEQ ID NO: 6) or comprising an amino acid sequence differing from SEQ ID NO: 6 by one amino acid; wherein Formula I is:

[0024] Leu-Nl-N2-N3-N4 wherein

[0025] N1 is He or Vai ,

[0026] N2 is Gin or Arg,

[0027] N3 is Pro or Ser,

[0028] N4 is Tyr or Ser, and wherein: when N2 is Gin, N4 is Tyr; and when N2 is Arg, N3-N4 is Pro-Ser, Ser-Ser or Ser-Tyr.

[0029] The invention also provides a polynucleotide comprising a sequence encoding said alpha chain variable domain, optionally also comprising a sequence encoding a beta chain variable domain. The invention also provides a polynucleotide comprising a sequence encoding a binding molecule of the invention. The invention also provides a pair of polynucleotides, the first of which comprises a sequence encoding said alpha chain variable domain, and the second of which comprises a sequence encoding a beta chain variable domain.

[0030] The invention also provides a vector comprising a polynucleotide of the invention.

[0031] The invention also provides a cell (for example an isolated cell or a host cell) comprising a binding molecule of the invention, a polynucleotide or pair of polynucleotides of the invention and / or a vector of the invention. The invention also provides a pharmaceutical composition comprising a binding molecule of the invention, a polynucleotide or pair of polynucleotides of the invention, a vector of the invention, a cell of the invention and / or a plurality of cells of the invention.

[0032] The invention also provides a binding molecule of the invention, a polynucleotide or pair of polynucleotides of the invention, a vector of the invention, a cell of the invention, a plurality of cells of the invention and / or a pharmaceutical composition of the invention for use in a method of treatment, optionally wherein the method of treatment is a method of treating a MAGE-A3 -associated disease or disorder, such as a MAGE -A3 -associated cancer.

[0033] Equivalently, the invention also provides a method of treatment comprising administering a binding molecule of the invention, a polynucleotide or pair of polynucleotides of the invention, a vector of the invention, a cell of the invention, a plurality of cells of the invention and / or a pharmaceutical composition of the invention to a patient. The method of treatment may be a method of treating a MAGE-A3 -associated disease or disorder, such as a MAGE-A3 -associated cancer.

[0034] Equivalently, the invention also provides use of a binding molecule of the invention, a polynucleotide or pair of polynucleotides of the invention, a vector of the invention, a cell of the invention, a plurality of cells of the invention and / or a pharmaceutical composition of the invention for the manufacture of a medicament. The medicament may be for a MAGE -A3 -associated disease or disorder, such as a MAGE -A3 -associated cancer.

[0035] Brief Description of the Drawings

[0036] Figure 1: Validation of a3a TCR-selected peptides with co-culture assays

[0037] A) Co-culture assay methodology.

[0038] B) Flow cytometry gating strategy for assessing the activation level of a3a-TCR Jurkat cells cultured with peptide-pulsed HLA-A*01:01 K562 cells. Unless otherwise stated, the early activation marker CD69 was used as a proxy for TCR activation; due to the rapid upregulation of CD69 surface levels upon productive TCR engagement of pMHCs it is a frequently used indicator of (early) T cell and Jurkat cell activation.

[0039] Figure 2: 18 peptides activate a3a TCR-Jurkat cells

[0040] %CD69 expression on the surface of a3a TCR-Jurkat cells after 18h of culture with HLA- A*01:01 K562 cells pulsed with 18 peptides as shown at concentrations of lOOpM, lOpM and IpM. The data shown are representative of three independent experiments, displaying means of duplicates ± SEM. “Control” refers to a3a TCR-Jurkat cells cultured with HLA-A*01:01 K562 cells with no pulsed peptide.

[0041] Figure 3: Expression of CD8 on a3a TCR-Jurkat cells enhances antigen sensitivity CD8+ a3a TCR-Jurkat CD69 expression after 18h of culture with HLA-A*01:01 K562 cells pulsed with 8 peptides as shown at 100 pM. Statistical analysis by ordinary one-way ANOVA, ****p<0.0001, ***p<0.001, **p<0.01, *p<0.05, ns p>0.05. Error bars represent means of quadruplets ± SEM, pooled from two experimental repeats. “Control” refers to a3a TCR-Jurkat cells cultured with HLA-A*01:01 K562 cells with no pulsed peptide.

[0042] Figure 4: 10 peptides activate a3a TCR-Jurkat cells

[0043] %CD69 expression on the surface of a3a TCR-Jurkat cells after 18h of culture with HLA- A*01:01 K562 cells pulsed with 10 peptides as shown at 100 pM. The data shown are representative of two independent experiments, displaying means of duplicates ± SEM. “Control” refers to a3a TCR-Jurkat cells cultured with HLA-A*01:01 K562 cells with no pulsed peptide. MAGE -A3 included as positive control.

[0044] Figure 5: The a3a TCR-Jurkat cells respond to all known a3a agonists

[0045] CD8+ a3a TCR-Jurkat CD69 expression after 18h of culture with HLA-A*01:01 K562 cells pulsed with 11 human agonists as shown at 100 pM. Statistical analysis by ordinary one-way ANOVA, ****p<0.0001, ***p<0.001, **p<0.01, *p<0.05, ns p>0.05. Error bars represent means of triplets ± SEM. “Control” refers to a3a TCR-Jurkat cells cultured with HLA-A*01:01 K562 cells with no pulsed peptide.

[0046] Figure 6: Identified human agonists of the a3a TCR have distinct amino acid sequences

[0047] Amino acid sequences of the a3a TCR agonists in order of potency. Shaded amino acids are in common with the MAGE -A3 sequence, dark grey shaded amino acids at p3 and p9 as indicated show the fixed anchor residues at p3 and p9.

[0048] Figure 7: Expression patterns of the a3a TCR off-target agonists

[0049] The Human Protein Atlas depictions of the protein expression patterns of the a3a off-target antigens. hFat2, PLD 5, CD166, Titin are primarily expressed on the cell membrane. S / T MRCKa, ANKRD16, COG 4 and FXYD6 are primarily expressed intracellularly. The depictions illustrate that PLD 5, hFat2, and Titin have restricted tissue expression (PLD 5 is expressed primarily in the brain, hFat2 is expressed in the brain, oesophagus, and skin, and Titin is expressed in the heart and skeletal muscle), while the antigens CD 166, S / T MRCKa, ANKRD16, and FXYD6 show widespread tissue expression.

[0050] Figure 8: Amino acid sequences of the eight a3a TCR intermediate variants

[0051] In descending order, the CDR2a amino acid sequences of the a3a TCR (grey), the eight intermediate TCRs, and the a3 TCR. The key on the right hand side indicates the known specificities of the a3 and a3a TCRs for MAGE-A3 and Titin; ‘+’ indicating stimulatory, indicating nonstimulatory. Shared amino acids with the a3a TCR in grey; shared amino acids with the WT a3 TCR in white; all Leucines at pl in dark grey.

[0052] Figure 9: The structure of the a3a TCR PyMOL depiction of the a3a TCR (MAG-IC3 ImmTAC structure; PDB 5BRZ): alpha chain with CDR2a motif VRPY as indicated by an arrow; beta chain; bound to HLA-A*01:01, with MAGE-A3 as indicated by an arrow.

[0053] Figure 10: Generating CD8+ a3a-variant Jurkat cell lines

[0054] Left panel showing expression levels of CD8 in TCR- / - Jurkats transduced with the lentiviral pHR vector (SFFV promoter) or pHRi inducible vector (ecdysone -inducible promoter). In the absence of ecdysone the pHRi vector is transcribed at low levels, achieving physiological CD8 expression levels. Right panel showing the TCR expression levels of the pHRi CD8+ TCR -I- Jurkats transduced with the various a3a TCR variants.

[0055] Figure 11: a3a TCR variants lose sensitivity to Titin

[0056] CD69 expression levels of CD8+ a3a-variant- Jurkat cell lines after 18h of culture with HLA- A*01:01 K562 cells pulsed with peptide antigens of MAGE-A3, Titin, or a control non-agonist peptide (a PARP1 peptide with amino acid sequence SEQ ID NO: 75), all at lOOpM. Per TCR grouping, columns left to right are: MAGE -A3 (SEQ ID NO: 45), Titin (SEQ ID NO: 46), Control peptide (SEQ ID NO: 75). Error bars represent means ± SEM. Data representative of three technical repeats.

[0057] Figure 12: Correlating TCR binding profiles with TCR sequence

[0058] The activation profiles of the 10 TCRs a3, a3a, v3, v4, v5, v6, v7, v8, v9 and vlO as indicated to MAGE-A3 (SEQ ID NO: 45) and Titin (SEQ ID NO: 46), alongside their respective CDR2a sequences. The left-hand box corresponds to the TCR variants with significantly reduced sensitivity to Titin. The dotted line corresponds to the negative control peptide (SEQ ID NO: 75) gate (1.19%).

[0059] Figure 13: Images of CDR2a loop interactions with MAGE- A3

[0060] PyMOL images zooming in on the CDR2a loop interacting with the A1-MAGE-A3 complex a2 domain (PDB 5BRZ). TCR in upper half of each image: alpha chain, beta chain. CDR2a amino acids showing nitrogen atoms and oxygen atoms, with annotated distances between the CDR2a arginine and al domain glutamic acid annotated (3.5 A, 2.3 , 2.8 A). MAGE -A3 peptide. HLA- A*01:01 lower half of each image.

[0061] Figure 14: a3a TCR variants v8 and v9 lose sensitivity to off-target antigens

[0062] CD69 expression levels of CD8+ a3a-variant- Jurkat cell lines after 18h of culture with HLA- A*01:01 K562 cells pulsed with the a3a agonists (Fig 3) at 100 pM, or a control non-agonist peptide. Per TCR grouping, columns left to right are: MAGE -A3 (SEQ ID NO: 45), MAGE-A6 (SEQ ID NO: 47), MAGE-B18 (SEQ ID NO: 48), CD166 (SEQ ID NO: 50), Titin (SEQ ID NO: 46), S / MRCK alpha (SEQ ID NO: 49), ANKRD16 (SEQ ID NO: 51), FXYD6 (SEQ ID NO: 52). Error bars represent means ± SEM. Data representative of two experimental repeats. The dotted line represents the non-agonist peptide control (SEQ ID NO: 75).

[0063] Figure 15: a3a TCR variants v8 and v9 show greater specificity for MAGE family antigens CD69 expression levels of CD8+ a3a-variant-Jurkat cell lines as indicated after 18h of culture with HLA-A*01:01 K562 cells pulsed with MAGE-A3 (SEQ ID NO: 45) / -A6 (SEQ ID NO: 47) / -B18 (SEQ ID NO: 48) and Titin (SEQ ID NO: 46) peptides at a range of concentrations: (A) a3 WT and a3a; (B) v8 and v9. Error bars represent means ± SEM. Data representative of two experimental repeats. The dotted line represents the non-agonist peptide control (SEQ ID NO: 75).

[0064] Figure 16: a3a variant v8 and v9 bind to MAGE-A3-HLA-A*01:01

[0065] SPR binding curves of 8 concentrations of TCRs a3 (a) a3a (b) v8 (c) and v9 (d) flown over MAGE-A3-A*01:01 complexes docked onto a Biacore CAP chip. Data representative of three experimental repeats.

[0066] Figure 17: a3a variant v9 does not bind to Titin-HLA-A*01:01

[0067] SPR binding curves of 8 concentrations of TCRs a3 (a) a3a (b) v8 (c) and v9 (d) flown over Titin-HLA-A*01:01 complexes docked onto a Biacore CAP chip. Data representative of three experimental repeats.

[0068] Figure 18: Transduction and validation of primary CD8+ T cell lines

[0069] (Top) TCR Jurkat expression test for validation of lenti -virus stocks used for transducing CD8+ T cells. Top to bottom: control, a3 WT, a3a, v8, v9, 20a-I8 and 94a-14. (Bottom) Confirmation of successful TCR transduction by MAGE-A3-A*01:01 tetramer stain at 2.7 nM SA-PE. Left to right: control, untransduced, a3 WT, a3a, v8, v9, 20a-I8 and 94a-14. Control represents binding of cells to SA-PE without MAGE-A3 -A* 01 : 01.

[0070] Figure 19: v8 and v9 primary CD8+ T cells potently kill melanoma cancer cells

[0071] Confluence of A375 cells after 2 days cultured with or without primary CD8+ T cells transduced with various a3a TCRs. Cytotoxicity data representative of four technical repeats at a 1:5 E / T ratio. At 48h, in order of highest-to-lowest A375 confluence %: Untransduced and A375 are overlaid and indistinguishable; a3 (WT); a3av8 (also referred to as ‘v8’ herein); a3a; a3av9 (also referred to as ‘v9’ herein).

[0072] Figure 20: v8 and v9 primary CD8+ T cells from four different donors kill melanoma cancer cells reproducibly at three different E / T ratios

[0073] Confluence of A375 cells after 2 days cultured with or without primary CD8+ T cells transduced with various a3a TCRs. (a) 1:2 E / T ratio, data pooled from two technical repeats, (b) 1:5 E / T ratio, data pooled from four technical repeats, (c) 1: 10 E / T ratio, data pooled from three technical repeats. Statistical analysis by ordinary one-way ANOVA, ****p<0.000I, ***p<0.00I, **p<0.0I, *p<0.05, ns p>0.05. Error bars represent SEM.

[0074] Figure 21: v8 and v9 primary CD8+ T cells show comparable cytotoxicity to published a3a TCR-T cells

[0075] Confluence of HLA-A*01:01 Lenti -X cells pulsed with 50 pM of MAGE-A3 after 3 days cultured with or without primary CD8+ T cells transduced with various a3a TCRs. E / T ratio 1:2. Al - Lenti-X control pulsed with 50 pM of MAGE-A3 (SEQ ID NO: 45). Figure 22: v9 primary CD8+ T cells are more specific than published a3a TCR-T cells

[0076] Confluence of HLA-A*01:01 Lenti-X cells pulsed with 50 pM of (a) MAGE-A3 peptide (SEQ ID NO: 45) (b) Titin peptide (SEQ ID NO: 46) (c) CD166 peptide (SEQ ID NO: 50) after 3 days cultured with or without primary CD8+ T cells transduced with various a3a TCRs. Left: E / T ratio 1:2, pooled data from two technical repeats. Right: E / T ratio 1:5, pooled data from two technical repeats. Al-Lenti-X control pulsed with 50 pM of MAGE-A3. Statistical analysis by ordinary one-way ANOVA, ****p<0.0001, ***p<0.001, **p<0.01, *p<0.05, ns p>0.05. Error bars represent SEM.

[0077] Figure 23: v8 and v9 primary CD8+ T cells do not show alloreactivity against the most common HLA class I alleles

[0078] CD69 expression levels of CD8+ a3a-variant-primary CD8+ T cells after 18h of culture 8 common HLA class I alleles expressed on K562 cells. The top dotted line corresponds to the CD69 expression levels of primary CD8+ T cells transduced with the TCR variants a3, a3a, v8 and v9 as indicated after 18h of culture with HLA-A*01:01 K562 cells pulsed with 100 pM MAGE-A3. Data is representative of three technical repeats. Statistical analysis by ordinary one-way ANOVA, ***p<0.001, ns p>0.05. Error bars represent means ± SEM.

[0079] Figure 24: TCR v9 is more sensitive, but essentially no more cross-reactive, than the a3 WT TCR

[0080] CD69 expression levels of (a) a3 WT (b) a3a (c) v8 and (d) v9 CD8+ Jurkat cell lines after 18h of culture with HLA-A*01:01 K562 cells pulsed with the a3a agonists (peptide antigens of MAGE- A3 (SEQ ID NO: 45), MAGE-A6 (SEQ ID NO: 47), hFat (SEQ ID NO: 54), PLD 5 (SEQ ID NO: 55), MAGE-B18 (SEQ ID NO: 48), CD166 (SEQ ID NO: 50), S / MRCKa (SEQ ID NO: 49), Titin (SEQ ID NO: 46), ANKRD16 (SEQ ID NO: 51), Dyn4 (SEQ ID NO: 77), COG4 (SEQ ID NO: 53), FXYD6 (SEQ ID NO: 52)) at 100 pM, or a control non-agonist peptide (SEQ ID NO: 75) as indicated. Error bars represent means ± SEM. Data representative of one experiment. The dotted line represents a non-agonist peptide control. Statistical analysis by ordinary one-way ANOVA, ****p<0.0001, ***p<0.001, **p<0.01, *p<0.05, ns p>0.05.

[0081] Figure 25: Structural comparison of a3a, MAG-IC3, and v9 TCRs bound to MAGE-A3- HLA-A*01:01

[0082] Comparing the crystal structures of a3a, MAG-IC3 and v9 TCR crystal structures provides a mechanistic basis for differences observed in the cross-reactivity of these TCRs. (A) Superposition of the A3 A, MAG-IC3, and v9 TCRs in complex with MAGE-A3 peptide-HLA-A*01:01 (as indicated), illustrating the shared overall docking mode across all three receptors. (B) Close-up view of the CDR2a-MHC interface highlighting the distinct contact networks formed by the engineered CDR2a loops in A3 A, MAG-IC3, and v9. (C) Alignment of CDR2a sequences for A3 wild-type, A3 A, v9, and MAG-IC3 TCRs, showing the engineered residues that modulate pMHC engagement. (D) Individual structures of A3 A, v9, and MAG-IC3 TCRs with residues forming contacts to HLA- A*01:01 mapped onto their respective CDR2a loops. (E) Surface representation of HLA-A*01:01 showing the footprint of each TCR, with residues coloured by number of contacts to illustrate how each engineered CDR2a loop redistributes the interaction surface. (F) Fraction of total TCR contacts made with HLA versus peptide for each receptor (A3wt, A3 A, MAG-IC3, v9). The A3 wild-type (A3wt) distribution was determined from an AlphaFold3 prediction. (G) Distribution of contacts contributed by each of the six CDR loops (al, a2, a3, pi, 2, 3) for A3 A, MAG-IC3, and v9, illustrating how structural rewiring of CDR2a alters overall TCR-pMHC engagement.

[0083] Detailed Description

[0084] All publications, patents and patent applications cited herein, whether supra or infra, are hereby incorporated by reference in their entirety.

[0085] General definitions

[0086] Unless defined otherwise, technical and scientific terms used herein have the same meaning as commonly understood by a person skilled in the art to which this invention belongs.

[0087] In general, the term "comprising" is intended to mean including but not limited to. For example, the phrase “A binding molecule comprising a TCR alpha chain variable domain" should be interpreted to mean that the binding molecule comprises an alpha variable domain, but may also comprise further amino acid domains, sequences and / or components (such as for example a TCR beta chain varaible domain).

[0088] In some embodiments of the invention, the word “comprising” is replaced with the phrase “consisting of’. The term “consisting of' is intended to be limiting. For example, the phrase “A binding molecule consisting of a TCR alpha chain variable domain" should be understood to mean that the binding molecule consists of a step of an alpha variable domain and no further components.

[0089] In some embodiments of the invention, the word “comprising ” is replaced with the phrase “consisting essentially of” . The term “consisting essentially of’ means that specific further components can be present, namely those not materially affecting the essential characteristics of the subject matter.

[0090] The term "about" or “around” when referring to a value refers to that value but within a reasonable degree of scientific error. Optionally, a value is “about x” or “around x” if it is within 10%, within 5%, or within 1% of x.

[0091] The singular forms “a ”, “an ”, and “the ” include plural referents unless the content clearly dictates otherwise.

[0092] For the purpose of this invention, in order to determine the percent identity of two sequences (such as two polynucleotide or two polypeptide sequences), the sequences are aligned for optimal comparison purposes (e.g., gaps can be introduced in a first sequence for optimal alignment with a second sequence). The nucleotides or amino acid residues at each position are then compared. When a position in the first sequence is occupied by the same nucleotide or amino acid at the corresponding position in the second sequence, then the nucleotides or amino acids are identical at that position. The percent identity between the two sequences is a function of the number of identical positions shared by the sequences (z.e., % identity = number of identical positions / total number of positions in the reference sequence x 100).

[0093] Typically, the sequence comparison is carried out over the length of the reference sequence. For example, if the user wished to determine whether a given (“test”) sequence is 95% identical to SEQ ID NO: 1, SEQ ID NO: 1 would be the reference sequence. To assess whether a sequence is at least 95% identical to SEQ ID NO: 1 (an example of a reference sequence), the skilled person would carry out an alignment over the length of SEQ ID NO: 1, and identify how many positions in the test sequence were identical to those of SEQ ID NO: 1. If at least 95% of the positions are identical, the test sequence is at least 95% identical to SEQ ID NO: 1. If the test sequence is shorter than SEQ ID NO: 1, the gaps or missing positions should be considered to be non -identical positions.

[0094] The skilled person is aware of different computer programs that are available to align two sequences. For instance, an alignment between two sequences can be accomplished using a mathematical algorithm. In an embodiment, the two amino acid or nucleic acid sequences are aligned using the Needleman and Wunsch (1970) algorithm which has been incorporated into the GAP program in the Accelrys GCG software package using either a Blosum 62 matrix or a PAM250 matrix, and a gap weight of 16, 14, 12, 10, 8, 6, or 4 and a length weight of 1, 2, 3, 4, 5, or 6.

[0095] “T cell receptor" (TCR), as used herein, refers to an immunoglobulin superfamily member having a variable binding domain, a constant domain, a transmembrane region, and a short cytoplasmic tail (see, e.g., Janeway et al., Immunobiology: The Immune System in Health and Disease, 3rd Ed., Current Biology Publications, p. 4:33, 1997)) and which is capable of specifically binding to an antigen peptide bound to a MHC receptor. A TCR can be found on the surface of a cell and generally is comprised of a heterodimer having a and P chains (also known as alpha chain and beta chain, or TCRa and TCR , respectively), or y and 5 chains (also known as TCRy and TCR5, respectively). Like immunoglobulins, the extracellular portion of each TCR chain (e.g. a-chain, -chain) contains two immunoglobulin regions: a variable region (e.g. the alpha chain contains a ‘variable a region’ or ‘alpha chain variable domain’ or ‘alpha chain variable region’ or Va, and ‘variable P region’ or ‘beta chain variable domain’ or ‘beta chain variable region’ or Vp, typically amino acids 1 to 116 based on Kabat numbering at the N-terminus), and one constant region (e.g. the alpha chain contains a ‘constant domain a’ or ‘alpha chain constant domain’ or ‘alpha chain constant region’ or Ca which is typically amino acids 117 to 259 based on Kabat, and the beta chain contains a ‘constant domain P’ or ‘beta chain constant domain’ or ‘beta chain constant region’ or C which is typically amino acids 117 to 295 based on Kabat) adjacent to the cell membrane. Also, like immunoglobulins, the variable domains contain complementary determining regions (CDRs) separated by framework regions (FRs).

[0096] Canonically, a TCR is found on the surface of T cells (or T lymphocytes) and associates with the CD3 complex. The source of a TCR of the present disclosure may be from various animal species, such as a human, mouse, rat, rabbit or other mammal. For example, the source of a TCR may be a mouse genetically engineered to produce TCRs comprising human alpha and beta chains (see, e.g., PCT Publication No. WO 2016 / 164492).

[0097] The variable domains are involved directly in binding the TCR to the antigen (via their six total CDRs). The constant regions of the alpha chain and of the beta chain are not involved directly in binding of a TCR to an antigen, but can facilitate various effector functions.

[0098] For example, the structure of the reference TCR a3a is as follows. Numbering the residues increasing numerically N-terminal to C-terminal, residues 1 to 115 of a reference alpha chain of TCR a3a having amino acid sequence SEQ ID NO: 36 correspond to the alpha chain variable domain (SEQ ID NO: 22); residues 116 to 208 of SEQ ID NO: 36 correspond to the alpha constant domain (SEQ ID NO: 34); residues 27 to 32 of SEQ ID NO: 36 correspond to the CDRla (SEQ ID NO: 1); residues 49 to 53 of SEQ ID NO: 36 correspond to CDR2a (SEQ ID NO: 8); residues 89 to 103 of SEQ ID NO: 36 correspond to CDR3a (SEQ ID NO: 3). In the same way, as an illustrative example, residues 1 to 110 of a reference beta chain of TCR a3a having amino acid sequence SEQ ID NO: 40 correspond to the beta chain variable domain (SEQ ID NO: 33); residues 111 to 240 of SEQ ID NO: 40 correspond to the beta constant domain (SEQ ID NO 35); residues 25 to 29 of SEQ ID NO: 40 correspond to the CDRip (SEQ ID NO: 4); residues 47 to 52 of SEQ ID NO: 40 correspond to CDR2 (SEQ ID NO: 5); residues 89 to 101 of SEQ ID NO: 40 correspond to CDR3P (SEQ ID NO: 6).

[0099] “Antigen ” as used herein is meant any substance that causes the immune system to produce antibodies or specific cell-mediated immune responses against it. A disease-associated antigen is any substance that is associated with any disease that causes the immune system to produce antibodies or a specific-cell mediated response against it. Unless stated otherwise, “antigen ” herein refers to a peptide which, when presented on a MHC molecule, mediates a cellular immune response. The term also encompasses the peptide coupled to, for example, further amino acid residues which may aid in production or purification, e.g. a tag, histidine tag, mouse or human Fc, or a signal sequence such as ROR1.

[0100] “HLA” refers to the human leukocyte antigen (HLA) system or complex, which is a gene complex encoding the major histocompatibility complex (MHC) proteins in humans. These cellsurface proteins are responsible for the regulation of the immune system in humans. HLAs corresponding to MHC class I (A, B, and C) present peptides from inside the cell. “HLA-A” refers to the group of human leukocyte antigens (HLA) that are coded for by the HLA -A locus. HLA-A is one of three major types of human MHC class I cell surface receptors. The receptor is a heterodimer, and is composed of a heavy a chain and smaller p chain. The a chain is encoded by a variant HLA-A gene, and the chain (P2-microglobulin) is an invariant P2 microglobulin molecule. The term "H A-A 1" (also referred to as “HLA-A*01”) is one particular class I major histocompatibility complex (MHC) allele group at the HLA-A locus; the a chain is encoded by the HLA-A*01 gene and the P chain is encoded by the P2 -microglobulin or B2M locus. The term “HLA-A*01:07” refers to isoform 1 of the HLA-A 1 allele.

[0101] "A PC" refers to an antigen-presenting cell, which is a cell presenting a peptide antigen on a MHC molecule.

[0102] Binding molecules

[0103] Binding molecules of the invention comprise a TCR alpha chain variable domain and a TCR beta chain variable domain. A binding molecule may comprise a first polypeptide comprising the alpha chain variable domain and a separate second polypeptide comprising the beta chain variable domain, or a binding molecule may comprise the alpha chain variable domain and the beta chain variable domain as a single polypeptide.

[0104] A binding molecule of the invention comprises:

[0105] (A) a T cell receptor (TCR) alpha chain variable domain comprising:

[0106] (i) a complementary determining region (CDR) 1 comprising an amino acid sequence of SEQ ID NO: 1, or comprising an amino acid sequence differing from the sequence of SEQ ID NO: 1 by one amino acid,

[0107] (ii) a CDR2 comprising the amino acid sequence of Formula I (SEQ ID NO: 2), and

[0108] (iii) a CDR3 comprising the amino acid sequence of SEQ ID NO: 3 or comprising an amino acid sequence differing from the sequence of SEQ ID NO: 3 by one amino acid; and

[0109] (B) a TCR beta chain variable domain comprising:

[0110] (i) a CDR1 comprising the amino acid sequence of SEQ ID NO: 4 or comprising an amino acid sequence differing from the sequence of SEQ ID NO: 4 by one amino acid,

[0111] (ii) a CDR2 comprising the amino acid sequence of SEQ ID NO: 5 or comprising an amino acid sequence differing from the sequence of SEQ ID NO: 5 by one amino acid, and

[0112] (iii) a CDR3 comprising the amino acid sequence of SEQ ID NO: 6 or comprising an amino acid sequence differing from the sequence of SEQ ID NO: 6 by one amino acid.

[0113] Formula I is:

[0114] Leu-Nl-N2-N3-N4 wherein

[0115] N1 is He or Vai,

[0116] N2 is Gin or Arg,

[0117] N3 is Pro or Ser,

[0118] N4 is Tyr or Ser, and wherein: when N2 is Gin, N4 is Tyr; and when N2 is Arg, N3-N4 is Pro-Ser, Ser-Ser or Ser-Tyr.

[0119] N1 is also referred to as p2 herein, N2 as p3, N3 as p4, and N4 as p5. The Leu preceding N1 is also referred to as p 1.

[0120] In other words, the CDR2 of the alpha chain variable domain (CDR2a) comprises the amino acid sequence of any one of one of SEQ ID NOs: 9, 11, 13, 14, 15, 17, 18, 19, 20 or 21. The CDR2a may comprise the amino acid sequence of any one of the sequences of Table 1. The CDR2a may comprise the amino acid sequence of any one of one of SEQ ID NOs: 9, 11, 13, 14 or 15. The CDR2a may comprise the amino acid sequence of any one of one of SEQ ID NOs: 9, 11, 13, 14, 15, 20 or 21. In some embodiments the CDR2a may consist of the amino acid sequence of any one of one of SEQ ID NOs: 9, 11, 13, 14, 15, 17, 18, 19, 20 or 21. The CDR2a may consist of the amino acid sequence of any one of one of SEQ ID NOs: 9, 11, 13, 14 or 15. The CDR2a may consist of the amino acid sequence of any one of one of SEQ ID NOs: 9, 11, 13, 14, 15, 20 or 21.

[0121] Table 1. Embodiments of the invention. T V’ or T>V’ indicates that He at N1 of Formula I has been substituted for Vai.

[0122] In some embodiments, a binding molecule of the invention comprises a CDRla comprising an amino acid sequence of SEQ ID NO: 1, a CDR2a comprising the amino acid sequence of Formula I (SEQ ID NO: 2), a CDR3a comprising an amino acid sequence SEQ ID NO: 3, a CDRip comprising an amino acid sequence SEQ ID NO: 4, a CDR2P comprising an amino acid sequence SEQ ID NO: 5, and a CDR3P comprising an amino acid sequence SEQ ID NO: 6. In some embodiments, a binding molecule of the invention comprises a CDRla consisting of an amino acid sequence SEQ ID NO: 1 or an amino acid sequence differing therefrom by one amino acid, a CDR2a consisting of the amino acid sequence of Formula I (SEQ ID NO: 2), a CDR3a consisting of an amino acid sequence SEQ ID NO: 3 or an amino acid sequence differing therefrom by one amino acid, a CDRip consisting of an amino acid sequence SEQ ID NO: 4 or an amino acid sequence differing therefrom by one amino acid, a CDR2P consisting of an amino acid sequence SEQ ID NO: 5 or an amino acid sequence differing therefrom by one amino acid, and a CDR3P consisting of an amino acid sequence SEQ ID NO: 6 or an amino acid sequence differing therefrom by one amino acid.

[0123] In some embodiments, a binding molecule of the invention comprises a CDRla consisting of an amino acid sequence SEQ ID NO: 1, a CDR2a consisting of the amino acid sequence of Formula I (SEQ ID NO: 2), a CDR3a consisting of an amino acid sequence SEQ ID NO: 3, a CDRip consisting of an amino acid sequence SEQ ID NO: 4, a CDR2P consisting of an amino acid sequence SEQ ID NO: 5, and a CDR3P consisting of an amino acid sequence SEQ ID NO: 6.

[0124] In some embodiments, a binding molecule of the invention comprises the CDR sequences of the a3a TCR, but wherein the a3a TCR’s CDR2a (SEQ ID NO: 8, corresponding to residues 49 to 53 of SEQ ID NO: 36) has been replaced with Formula I (SEQ ID NO: 2). In some embodiments, the CDR2a comprises the amino acid sequence of any one of one of SEQ ID NOs: 9, 11, 13, 14, 15, 17, 18, 19, 20 or 21.

[0125] In other words, in some embodiments a binding molecule of the invention comprises the CDR sequences of the a3a TCR, but wherein the CDR2a of the a3a TCR (SEQ ID NO: 8, corresponding to residues 49 to 53 of SEQ ID NO: 36) has been replaced with the amino acid sequence of one of SEQ ID NOs: 9, 11, 13, 14, 15, 17, 18, 19, 20 or 21, i.e. has been replaced with the amino acid sequence of the CDR2a of one of v3, v5, v7, v8, v9, v3 I V, v5 I V, v7 I V, v8 I V or v9 I V as listed in Table 1. In some embodiments, a binding molecule of the invention comprises the CDR sequences of the a3a TCR, but wherein the CDR2a of the a3a TCR (SEQ ID NO: 8, corresponding to residues 49 to 53 of SEQ ID NO: 36) has been replaced with the amino acid sequence of one of SEQ ID NOs: 9, 11, 13, 14 or 15, i.e. has been replaced with the amino acid sequence of the CDR2a of one of v3, v5, v7, v8 or v9 as listed in Table 1.

[0126] In some embodiments, N2 is Arg. In some embodiments, the CDR2a comprises the amino acid sequence of any one of one of SEQ ID NOs: 9, 13, 14, 17, 19 or 20. In some embodiments, the CDR2a consists of the amino acid sequence of any one of one of SEQ ID NOs: 9, 13, 14, 17, 19 or 20.

[0127] In some embodiments, when N2 is Arg, N4 is Ser. In some embodiments, the CDR2a comprises the amino acid sequence of any one of one of SEQ ID NOs: 9, 11, 14, 17 or 20. In some embodiments, the CDR2a consists of the amino acid sequence of any one of one of SEQ ID NOs: 9, 11, 14, 17 or 20.

[0128] In some embodiments, when N2 is Arg, N3-N4 is Ser-Ser. In some embodiments, the CDR2a comprises the amino acid sequence of any one of one of SEQ ID NOs: 9 or 17. In some embodiments, the CDR2a consists of the amino acid sequence of any one of one of SEQ ID NOs: 9 or 17.

[0129] In some embodiments, N2 is Gin. In some embodiments, the CDR2a comprises the amino acid sequence of any one of one of SEQ ID NOs: 11, 15, 18 or 21. In some embodiments, the CDR2a consists of the amino acid sequence of any one of one of SEQ ID NOs: 11, 15, 18 or 21.

[0130] In some embodiments, N3 is Pro. In some embodiments, the CDR2a comprises the amino acid sequence of any one of one of SEQ ID NOs: 14, 15, 20 or 21. In some embodiments, the CDR2a consists of the amino acid sequence of any one of one of SEQ ID NOs: 14, 15, 20 or 21.

[0131] In some embodiments, N 1 is He. In some embodiments, the CDR2a comprises the amino acid sequence of any one of one of SEQ ID NOs: 9, 11, 13, 14 or 15. In some embodiments, the CDR2a consists of the amino acid sequence of any one of one of SEQ ID NOs: 9, 11, 13, 14 or 15.

[0132] In some embodiments, N 1 is He and N3 is Pro. In some embodiments, the CDR2a comprises the amino acid sequence of one of one of SEQ ID NO: 14 or 15. In some embodiments, the CDR2a consists of the amino acid sequence of one of one of SEQ ID NO: 14 or 15.

[0133] In some preferred embodiments, N2-N3-N4 is Gln-Pro-Tyr. In some embodiments, the CDR2a comprises the amino acid sequence of one of one of SEQ ID NO: 15 or 21. In some more preferred embodiments, the CDR2a consists of the amino acid sequence of one of one of SEQ ID NO: 15 or 21.

[0134] In some embodiments, the CDR2a comprises the amino acid sequence SEQ ID NO: 9. In some embodiments, the CDR2a consists of the amino acid sequence SEQ ID NO: 9.

[0135] In some embodiments, the CDR2a comprises the amino acid sequence SEQ ID NO: 11. In some embodiments, the CDR2a consists of the amino acid sequence SEQ ID NO: 11.

[0136] In some embodiments, the CDR2a comprises the amino acid sequence SEQ ID NO: 13. In some embodiments, the CDR2a consists of the amino acid sequence SEQ ID NO: 13.

[0137] In some embodiments, the CDR2a comprises the amino acid sequence SEQ ID NO: 20. In some embodiments, the CDR2a consists of the amino acid sequence SEQ ID NO: 20.

[0138] In some embodiments, the CDR2a comprises the amino acid sequence SEQ ID NO: 14. In some embodiments, the CDR2a consists of the amino acid sequence SEQ ID NO: 14.

[0139] In some embodiments, the CDR2a comprises the amino acid sequence SEQ ID NO: 21. In some embodiments, the CDR2a consists of the amino acid sequence SEQ ID NO: 21.

[0140] In some preferred embodiments, the CDR2a comprises the amino acid sequence SEQ ID NO: 15. In some most preferred embodiments, the CDR2a consists of the amino acid sequence SEQ ID NO: 15. Framework regions

[0141] Binding molecules of the invention can accommodate variation in the framework regions of the alpha chain variable domain whilst still retaining their function, so long as the CDR sequences are as described herein. For example a binding molecule may comprise an alpha chain variable domain having sequence SEQ ID NO: 22 or a sequence at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96% or at least 97% identical to the amino acid sequence SEQ ID NO: 22, wherein said sequence comprises a CDRla, a CDR2a and a CDR3a as set out above. In some embodiments, the alpha chain variable domain has a sequence at least 96% identical to the amino acid sequence SEQ ID NO: 22, wherein said sequence comprises a CDRla, a CDR2a and a CDR3a as set out above. In some embodiments, the alpha chain variable domain has a sequence at least 97% identical to the amino acid sequence SEQ ID NO: 22, wherein said sequence comprises a CDRla, a CDR2a and a CDR3a as set out above. In some embodiments, the alpha chain variable domain has a sequence at least 98% identical to the amino acid sequence SEQ ID NO: 22, wherein said sequence comprises a CDRla, a CDR2a and a CDR3a as set out above.

[0142] Likewise, binding molecules of the invention can accommodate variation in the framework regions of the beta chain variable domain whilst still retaining their function, so long as the CDR sequences are as described herein. For example a binding molecule may comprise a beta chain variable domain having sequence SEQ ID NO: 33 or a sequence at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96% or at least 97% identical to the amino acid sequence SEQ ID NO: 33, wherein said sequence comprises a CDRip, a CDR2P and a CDR3P as set out above. In some embodiments, the beta chain variable domain has a sequence at least 98% identical to the amino acid sequence SEQ ID NO: 33, wherein said sequence comprises a CDRip, a CDR2P and a CDR3P as set out above. In some embodiments, the beta chain variable domain has a sequence at least 99% identical to the amino acid sequence SEQ ID NO: 33, wherein said sequence comprises a CDRip, a CDR2P and a CDR3P as set out above.

[0143] In some embodiments, a binding molecule of the invention comprises: an alpha chain variable domain corresponding to the alpha chain variable domain of a3a TCR (having an amino acid sequence of SEQ ID NO: 22) but wherein the CDR2a (corresponding to residues 49 to 53 of SEQ ID NO: 22 or, equally mutatis muiandem. corresponding to residues 49 to 53 of SEQ ID NO: 36) has been replaced with Formula I (SEQ ID NO: 2); and a beta chain variable domain corresponding to the beta chain variable domain of a3a TCR (having an amino acid sequence of SEQ ID NO: 33).

[0144] In other words, in some embodiments, a binding molecule of the invention comprises: an alpha chain variable domain (Va) corresponding to the alpha chain variable domain of a3a TCR (having an amino acid sequence of SEQ ID NO: 22) but wherein the CDR2a (corresponding to residues 49 to 53 of SEQ ID NO: 22) has been replaced with the amino acid sequence of one of SEQ ID NOs: 9, 11, 13, 14, 15, 17, 18, 19, 20 or 21 ; and a beta chain variable domain (VP) corresponding to the beta chain variable domain of a3a TCR (having an amino acid sequence of SEQ ID NO: 33). In some embodiments, a binding molecule of the invention comprises: an alpha chain variable domain (Va) corresponding to the alpha chain variable domain of a3a TCR (SEQ ID NO: 22) but wherein the CDR2a (corresponding to residues 49 to 53 of SEQ ID NO: 22) has been replaced with the amino acid sequence of one of SEQ ID NOs: 9, 11, 13, 14 or 15; and a beta chain variable domain (V ) corresponding to the beta chain variable domain of a3a TCR (having an amino acid sequence of SEQ ID NO: 33).

[0145] In one embodiment, a binding molecule comprises a Va comprising the amino acid sequence SEQ ID NO: 23 and a VP comprising the amino acid sequence SEQ ID NO: 33.

[0146] In one embodiment, a binding molecule comprises a Va comprising the amino acid sequence SEQ ID NO: 24 and a VP comprising the amino acid sequence SEQ ID NO: 33.

[0147] In one embodiment, a binding molecule comprises a Va comprising the amino acid sequence SEQ ID NO: 25 and a VP comprising the amino acid sequence SEQ ID NO: 33.

[0148] In one embodiment, a binding molecule comprises a Va comprising the amino acid sequence SEQ ID NO: 26 and a VP comprising the amino acid sequence SEQ ID NO: 33.

[0149] In one preferred embodiment, a binding molecule comprises a Va comprising the amino acid sequence SEQ ID NO: 27 and a VP comprising the amino acid sequence SEQ ID NO: 33.

[0150] In one embodiment, a binding molecule comprises a Va comprising the amino acid sequence SEQ ID NO: 28 and a VP comprising the amino acid sequence SEQ ID NO: 33.

[0151] In one embodiment, a binding molecule comprises a Va comprising the amino acid sequence SEQ ID NO: 29 and a VP comprising the amino acid sequence SEQ ID NO: 33.

[0152] In one embodiment, a binding molecule comprises a Va comprising the amino acid sequence SEQ ID NO: 30 and a VP comprising the amino acid sequence SEQ ID NO: 33.

[0153] In one embodiment, a binding molecule comprises a Va comprising the amino acid sequence SEQ ID NO: 31 and a VP comprising the amino acid sequence SEQ ID NO: 33.

[0154] In one embodiment, a binding molecule comprises a Va comprising the amino acid sequence SEQ ID NO: 32 and a VP comprising the amino acid sequence SEQ ID NO: 33.

[0155] Accordingly, in some embodiments binding molecules of the invention comprise a TCR ectodomain.

[0156] Constant domains and formats

[0157] The binding molecules may comprise an alpha chain constant domain and / or a beta chain constant domain. The alpha chain constant domain may comprise the amino acid sequence SEQ ID NO: 34 or an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93% or at least 94% identity to the amino acid sequence SEQ ID NO: 34. The alpha chain constant domain may comprise an amino acid sequence having at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to the amino acid sequence SEQ ID NO: 34.

[0158] The beta chain constant domain may comprise the amino acid sequence SEQ ID NO: 35 or an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 91%, at least 92%, at least 93% or at least 94% identity to the amino acid sequence SEQ ID NO: 35. The beta chain constant domain may comprise an amino acid sequence having at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to the amino acid sequence SEQ ID NO: 35.

[0159] In some embodiments the binding molecule is a receptor, for example a TCR.

[0160] For example, the TCR may comprise:

[0161] (A) an alpha chain comprising the amino acid sequence SEQ ID NO: 36, or an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85% identity, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to the amino acid sequence SEQ ID NO: 36, wherein the CDRla, CDR2a and CDR3a of said alpha chain are as described herein; and / or

[0162] (B) a beta chain comprising the amino acid sequence SEQ ID NO: 40, or an amino acid sequence having at least 70%, at least 75%, at least 80%, at least 85% identity, at least 90%, at least 91%, at least 92%, at least 93%, at least 94%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% sequence identity to the amino acid sequence SEQ ID NO: 40, wherein the CDRip, CDR2P and CDR3P of said beta chain are as described herein.

[0163] Binding molecules of the invention comprising an alpha chain variable domain and a beta chain variable domain as described herein can be engineered into various formats known to the skilled person, for example, but not limited to, recombinant TCRs, single-chain TCRs and solubilized antibody-TCR fusions (for example wherein the alpha and beta chain variable domains are fused to antibody constant regions).

[0164] MAGE-A3 binding characteristics

[0165] “MAGE-A3 ” or '"Melanoma-Associated Antigen A3" refers to the well-known cancertestis antigen having UniProtKB Accession Code: P43357 and, for example, having amino acid sequence SEQ ID NO: 43. In certain embodiments and as will be apparent from the context, the term refers to a peptide of MAGE-A3, for example a peptide having the amino acid sequence of SEQ ID NO: 45. A MAGE-A3 peptide may be in the context of MHC, HLA, HLA-A1, linked to HLA-A1 or as displayed by HLA-A1.

[0166] Binding molecules of the invention "bind to MAGE-A3 ” , in the sense that the binding molecules bind to an HLA-A1 presented MAGE-A3 peptide (e.g. a fragment of SEQ ID NO: 43), such as a peptide comprising amino acid residues 168 to 176 of MAGE-A3 (i.e. a peptide having the amino acid sequence of SEQ ID NO: 45).

[0167] Unless otherwise specified, binding of a binding molecule to an antigen (e.g. “MAGE-A3” or “MAGE-A3 antigen” or other antigens described herein) refers to binding of the binding molecule to a HLA-presented peptide of said antigen (e.g. SEQ ID NO: 45).

[0168] Binding molecules of the invention may “ specifically bind” to MAGE-A3, in the sense that the binding molecules bind to an HLA-A1 presented MAGE-A3 peptide (e.g. a peptide having amino acid sequence SEQ ID NO: 45 as above) with higher binding affinity than to any other pMHC antigen. Nevertheless, binding molecules of the invention may also exhibit meaningful (albeit lower) binding affinity for HLA-A1 presented peptides of MAGE-A6 (e.g. a peptide having amino acid sequence SEQ ID NO: 47) and optionally PLD-5 (e.g. a peptide having amino acid sequence SEQ ID NO: 55), MAGE-B18 (e.g. a peptide having amino acid sequence SEQ ID NO: 48) and hFat2 (e.g. a peptide having amino acid sequence SEQ ID NO: 54). Thus, cells bearing binding molecules of the invention may exhibit meaningful activation by peptides from peptides such as MAGE-A6 (SEQ ID NO: 47) and optionally PLD-5 (SEQ ID NO: 55), MAGE-B18 (SEQ ID NO: 48) and hFat2 (SEQ ID NO: 54). Said binding profile is consistent with the binding profile of the endogenous anti-MAGE-A3 WT A3 TCR (see Figure 24a).

[0169] Accordingly, a binding molecule of the invention binds to a MHC -presented peptide of cancer- associated antigen MAGE-A3 (e.g. SEQ ID NO: 45), optionally a HLA-presented MAGE-A3 peptide, further optionally wherein the HLA is HLA-A, preferably HLA-A1 (e.g. HLA-A*01:01). Likewise, cells bearing a binding molecule of the invention are activated upon binding to a MHC-presented peptide of cancer-associated antigen MAGE -A3 (e.g. SEQ ID NO: 45), optionally a HLA-presented MAGE-A3, further optionally wherein the HLA is HLA-A, preferably HLA-A 1 (e.g. HLA-A*01:01). In some embodiments, binding molecules of the invention are not alloreactive (Figure 23), for example do not display meaningful sensitivity for HLA-A, HLA-B or HLA-C when measured by CD69 co-culture assay (wherein for example K562 cells expressing HLA alleles are co-cultured with CD8+ primary T cells expressing a TCR of interest or binding molecule of the invention and pulsed with 100 uM of MAGE-A3 peptide, and CD69 expression levels of the T cells are measured after 18h of co-culture).

[0170] Binding affinity is a measure of the tightness of binding of an antigen (such as a pMHC) to its binding partner (such as a binding molecule of the invention or a TCR). Binding affinity can be readily assayed via standard techniques in the art, for example but not limited to surface plasmon resonance (SPR), bio-layer interferometry (BLI), and enzyme-linked immunosorbent assay (ELISA). Binding affinity may be expressed in terms of equilibrium dissociation constant (KD). A KD of less than IO-4M indicates tight binding, and TCRs may generally bind to their pMHC with affinity KD in the order 1 to 100 pM. Unless otherwise specified, the KD values reported herein are obtained via SPR. Unless otherwise specified, the KD values reported herein are obtained for antigens presented via HLA-A* 01:01.

[0171] In some embodiments a binding molecule of the invention binds to MAGE -A3 with a binding affinity of less than 100 pM KD. In some embodiments a binding molecule of the invention binds to MAGE-A3 with a binding affinity of less than 80 pM KD. In some embodiments a binding molecule of the invention binds to MAGE -A3 with a binding affinity of less than 70 pM KD, less than 60 pM KD, less than 50 pM KD, less than 40 pM KD or less than 35 pM KD. In some embodiments a binding molecule of the invention binds to MAGE-A3 with a binding affinity of less than 30 pM KD. In some embodiments a binding molecule of the invention binds to MAGE-A3 with a binding affinity of less than 25 pM KD. In some embodiments a binding molecule of the invention binds to MAGE -A3 with a binding affinity of less than 24 pM KD. In some embodiments a binding molecule of the invention binds to MAGE-A3 with a binding affinity of less than 23 pM KD, less than 20 pM KD, less than 18 pM KD, less than 17 pM KD or less than 16 pM KD. In some embodiments a binding molecule of the invention binds to MAGE-A3 with a binding affinity of less than 15 pM KD. In some embodiments a binding molecule of the invention binds to MAGE -A3 with a binding affinity of less than 14 pM KD.

[0172] In some embodiments a binding molecule of the invention further binds to MAGE-A6. In some embodiments a binding molecule of the invention further binds to MAGE-A6 with a binding affinity of 50 pM KD or lower. In some embodiments a binding molecule of the invention binds to MAGE-A3 and further binds to one or more of the cancer-associated antigens MAGE-A6, hFat2, PLD5 and MAGE-B18. In some embodiments a binding molecule of the invention binds to MAGE -A3 and further binds to one or more of the cancer-associated antigens MAGE-A6, hFat2 and PLD5.

[0173] In some embodiments, binding molecules of the invention bind to MAGE -A3 with a KD equal to or less than half (0.5 times) that of the A3 WT TCR (wherein the A3 WT TCR has an alpha chain sequence of SEQ ID NO: 37 and a beta chain sequence of SEQ ID NO: 40). In some embodiments, binding molecules of the invention bind to MAGE-A3 with a KD equal to or less than 0.4 times that of the A3 WT TCR, optionally equal to or less than 0.3 times that of the A3 WT TCR.

[0174] As a result of said binding, immune effector cells bearing binding molecules of the invention may be sensitive to MAGE-A3 (and may optionally be sensitive to MAGE-A6 and further optionally to PLD-5, MAGE-B18 and / or hFat2 as described above).

[0175] Sensitivity is a measure of T cell response indicative of T cell activation. Sensitivity can be reported by, for example, intracellular cytokine staining, cytokine secretion (for example but not limited to IL-2 or IFNy secretion, measurable for example by antibody capture at cell surface or ELISA), cytotoxicity, T cell proliferation and / or up-regulation of cell surface markers (for example but not limited to CD69, CD25, 4-1BB or PD1) via standard techniques in the art. The skilled person will understand that co-culture assay can be performed by co-culturing antigen- presenting cells (APCs) with, for example, relevant cell lines such as a T cell line such as Jurkat T cells, patient tumour samples, healthy tissue samples or primary cells, as appropriate. Unless otherwise specified, sensitivity as described herein is reported by up-regulation of T cell surface marker CD69, which can be readily measured for example by CD69 co-culture assay. Sensitivity may be expressed in terms of % cell-marker upregulation, wherein marker upregulation relative to negative control indicates sensitivity. Unless otherwise specified, sensitivity as reported herein is obtained by CD69 co-culture assay, wherein CD8+ Jurkat T cells expressing a TCR or binding molecule of interest are co-cultured for 18 hours with HLA-A*01:01 K562 cells pulsed with 100 uM of a peptide of interest or control peptide, and CD69 expression levels of the T cells are measured, wherein sensitivity of the TCR or binding molecule for the peptide of interest is indicated by upregulation of CD69 (expressed as % CD69 expression on the surface of Jurkat T cells).

[0176] In some embodiments, cells bearing binding molecules of the invention are at least 100% more sensitive to MAGE-A3 (SEQ ID NO: 45) than cells bearing the A3 WT TCR (wherein the A3 WT TCR has an alpha chain sequence of SEQ ID NO: 37 and a beta chain sequence of SEQ ID NO: 40) when measured by CD69 co-culture assay. A molecule which is 100% more sensitive than a reference TCR by CD69 co-culture exhibits double the CD69-APC levels of the reference TCR. Solely as an illustrative example, if a reference TCR exhibits 2% CD69 surface expression and a test molecule exhibits 4% CD69 expression, the test molecule is 100% more sensitive than the reference TCR. In some embodiments, binding molecules of the invention are at least 200% more sensitive to MAGE-A3 than the A3 WT TCR when measured by CD69 co-culture assay, optionally at least 300%, at least 400% or at least 500% more sensitive to MAGE-A3 than the A3 WT TCR. In some embodiments, binding molecules of the invention are at least 600% more sensitive to MAGE -A3 than the A3 WT TCR when measured by CD69 co-culture assay. In some embodiments, binding molecules of the invention are at least 650% more sensitive to MAGE -A3 than the A3 WT TCR when measured by CD69 co-culture assay. In some embodiments, binding molecules of the invention are at least 700%, at least 800%, at least 900% or at least 1000% more sensitive to MAGE-A3 than the A3 WT TCR when measured by CD69 co-culture assay. In some embodiments, binding molecules of the invention are at least 2000%, at least 2500% or at least 3000% more sensitive to MAGE -A3 than the A3 WT TCR when measured by CD69 co-culture assay.

[0177] As a result of anti-MAGE-A3 binding, cytotoxic T cells bearing binding molecules of the invention exhibit cytotoxicity against MAGE-A3 -bearing APCs (and also optionally against APCs bearing MAGE-A6 and further optionally PLD-5, MAGE-B18 and / or hFat2 as described above).

[0178] Cytotoxicity is a measure of cell killing by CD8+ cytotoxic T cells, thereby providing a measure of TCR potency and a direct readout for CD 8+ T cell effector functions. Cytotoxicity can be readily assayed via standard techniques in the art, for example but not limited to killing assay, chromium-51 release assay, Chromium -Release Alternative Europium (Eu3+) Release Assay, Lactate Dehydrogenase (LDH) Release Assay, Flow Cytometry -Based Cytotoxicity Assays (e.g. Annexin V / Propidium Iodide (PI) Staining), CFSE labelling, Real-Time Cell Analysis (RTCA) via quantification of cell viability dyes, caspase activation assays, ELISPOT assay, Granzyme B and Perforin Release Assays, ATP Release Assay. Cytotoxicity can be expressed for example as reduction in confluence of target APCs. Unless otherwise specified, cytotoxicity herein is expressed as % confluence of APC, wherein a reduced % confluence relative to control at 96 hours using a 1:2 effector: target (E / T) ratio indicates increased cytotoxicity.

[0179] In some embodiments, T cells bearing binding molecules of the invention exhibit equivalent cytotoxicity against MAGE-A3 APCs than T cells bearing a3a, meaning that no significant difference in cytotoxicity of APCs presenting a MAGE -A3 peptide antigen (SEQ ID NO: 4%) is observed relative to a3a TCR (Figure 21 and 22A).

[0180] Titin and CD 166

[0181] The a3a TCR has been described for example in WO 2012013913 Al (as ‘TCR No. 3’ in Example 6 of WO 2012013913 Al), which is incorporated by reference in its entirety herein. The a3a TCR has an alpha chain sequence SEQ ID NO: 36 and a beta chain sequence SEQ ID NO 40. The a3a TCR is known to bind to Titin and also to CD166. The inventors have confirmed that a3a binds to Titin (for example, to the Titin peptide antigen having the amino acid sequence of SEQ ID NO: 46) (see Figures 3, 5, 14, 15A, 17B, 22B, 24B) and also to CD 166 (for example, to the CD 166 peptide antigen having the amino acid sequence of SEQ ID NO: 50) (see Figures 3, 5, 14, 22C, 24C).

[0182] The inventors have also shown that, despite suggestions to the contrary in the literature (for example in Zhao, Xiang et al. “Tuning T cell receptor sensitivity through catch bond engineering.” Science vol. 376,6589 (2022): eabl5282. doi: 10.1126 / science.abl5282) CD8+ T cells bearing known anti-MAGE-A3 TCRs 20a- 18 and 94a- 14 demonstrate significant cytotoxic activity against Titin and CD 166 (Figure 22B and 22C). Therefore, said 20a- 18 and 94a- 14 TCRs have a functionally meaningful binding affinity for these self-antigens. The TCRs 20a-18 and 94a-14 are described for example in WO 2022192087 Al, incorporated herein by reference in its entirety. WO 2022192087 Al sets out how 20a-18 and 94a-14 can be produced. As described in WO 2022192087 Al, 20a-18 has an alpha chain sequence SEQ ID NO: 38 and a beta chain sequence SEQ ID NO: 41; 94a-14 has an alpha chain sequence SEQ ID NO: 39 and a beta chain sequence SEQ ID NO: 42. Herein, effects of binding molecules of the invention may be demonstrated through comparison to one or more of TCRs a3a, 20a- 18 and / or 94a-14, which may correspondingly be referred to as comparator molecules. For example, a3a may constitute a first comparator molecule, optionally 20a- 18 a second comparator molecule and further optionally 94a- 14 a third comparator molecule. Equally, 20a- 18 may constitute a first comparator molecule, or 94a-14 may constitute a first comparator molecule, and so on.

[0183] Binding molecules of the invention, in contrast, have a lower binding affinity for Titin than a3a. In some embodiments binding molecules of the invention have a lower binding affinity for Titin than a3a and have a lower binding affinity for Titin than 20a-18 and / or 94a-14. In some embodiments binding molecules of the invention have a lower binding affinity for Titin than do all of a3a, 20a- 18 and 94a- 14.

[0184] In some embodiments binding molecules of the invention have a lower binding affinity for CD166 than a3a, and optionally also than 20a-18 and / or 94a-14. In some embodiments binding molecules of the invention have a lower binding affinity for CD 166 than do all of a3a, 20a- 18 and 94a- 14.

[0185] In some embodiments binding molecules of the invention have a lower binding affinity for both Titin and CD 166 than a3a, and optionally also than 20a- 18 and / or 94a- 14. In some embodiments binding molecules of the invention have a lower binding affinity for both Titin and CD 166 than do all of a3a, 20a-18 and 94a-14.

[0186] In some embodiments, lower binding affinity means an at least 10% higher KD, optionally an at least 20% higher KD, further optionally at least 30%, 40%, 50%, 60%, 70%, 80%, 90% or preferably at least 100% higher KD. Likewise, an X% "reduction in binding affinity’ as used herein means an X% higher KD. For example a 20% reduction in binding affinity means a 20% increase in KD which can be measured by standard techniques in the art, for example but not limited to surface plasmon resonance (SPR), bio-layer interferometry (BLI), and enzyme-linked immunosorbent assay (ELISA). Unless otherwise specified, the KD values reported herein are obtained via SPR. Unless otherwise specified, the KD values reported herein are obtained for antigens presented via HLA-A* 01:01.

[0187] Advantageously, unlike the a3a TCR, binding molecules of the invention bind to self-antigen Titin (i.e. to the Titin peptide antigen having the amino acid sequence of SEQ ID NO: 46) with only very low binding affinity, and optionally essentially do not bind to self-antigen Titin (optionally presented on HLA-A 1, optionally HLA-A*01:01). For example, in some embodiments a binding molecule of the invention displays more than 35 pM KD binding affinity for Titin, optionally more than 40 pM KD, optionally more than 50 pM KD binding affinity, optionally more than 70 pM KD binding affinity for Titin. In some embodiments, a binding molecule of the invention displays more than 100 pM KD binding affinity for Titin. In some embodiments, a binding molecule of the invention displays more than 200 pM KD binding affinity for Titin. In some embodiments, a binding molecule of the invention displays more than 300 pM KD binding affinity for Titin. In some embodiments, a binding molecule of the invention displays more than 400 pM KD binding affinity for Titin. In some embodiments, a binding molecule of the invention displays more than 500 pM KD binding affinity for Titin. In some embodiments, a binding molecule of the invention displays essentially no binding for Titin, for example displays so little binding that no read-out can be detected.

[0188] In some embodiments, immune effector cells bearing binding molecules of the invention are less sensitive to Titin than immune effector cells bearing a3a, and optionally also than immune effector cells bearing 20a- 18 and / or 94a- 14.

[0189] In some embodiments, immune effector cells bearing binding molecules of the invention are at least 50% less sensitive to Titin (SEQ ID NO: 46) than are immune effector cells bearing the a3a TCR (wherein the a3a TCR has an alpha chain sequence of SEQ ID NO: 36 and a beta chain sequence of SEQ ID NO: 40) when measured by CD69 co-culture assay. Solely as an illustrative example, if a reference TCR exhibits 4% CD69 surface expression and a test molecule exhibits 2% CD69 expression, the test molecule is 50% less sensitive than the reference TCR

[0190] In some embodiments, immune effector cells bearing binding molecules of the invention are at least 60% less sensitive to Titin than are immune effector cells bearing the a3a TCR when measured by CD69 co-culture assay, optionally at least 65% less sensitive, optionally at least 70% less sensitive, optionally at least 75% less sensitive. In some embodiments, immune effector cells bearing binding molecules of the invention are at least 80% less sensitive to Titin than are immune effector cells bearing the a3a TCR when measured by CD69 co-culture assay. In some embodiments, immune effector cells bearing binding molecules of the invention are at least 85% less sensitive to Titin than are immune effector cells bearing the a3a TCR when measured by CD69 co-culture assay. In some embodiments, immune effector cells bearing binding molecules of the invention are at least 90% less sensitive to Titin than are immune effector cells bearing the a3a TCR when measured by CD69 co-culture assay.

[0191] In some embodiments, immune effector cells bearing binding molecules of the invention are at least 50% less sensitive to Titin (SEQ ID NO: 46) than are immune effector cells bearing 20a-18 and / or 94a-14 when measured by CD69 co-culture assay. In some embodiments, immune effector cells bearing binding molecules of the invention are at least 60% less sensitive to Titin than are immune effector cells bearing 20a- 18 and / or 94a- 14 when measured by CD69 co-culture assay, optionally at least 65% less sensitive, optionally at least 70% less sensitive, optionally at least 75% less sensitive. In some embodiments, immune effector cells bearing binding molecules of the invention are at least 80% less sensitive to Titin than are immune effector cells bearing 20a-18 and / or 94a-14 when measured by CD69 co-culture assay. In some embodiments, immune effector cells bearing binding molecules of the invention are at least 85% less sensitive to Titin than immune effector cells bearing 20a-18 and / or 94a-14 when measured by CD69 co-culture assay.

[0192] In some embodiments, immune effector cells bearing binding molecules of the invention are at least 50% less sensitive to Titin (SEQ ID NO: 46) than are immune effector cells bearing any of the a3a TCR (wherein the a3a TCR has an alpha chain sequence of SEQ ID NO: 36 and a beta chain sequence of SEQ ID NO: 40), 20a-18 (the a3a TCR has an alpha chain sequence of SEQ ID NO: 38 and a beta chain sequence of SEQ ID NO: 41) and 94a-14 (the a3a TCR has an alpha chain sequence of SEQ ID NO: 39 and a beta chain sequence of SEQ ID NO: 42) when measured by CD69 co-culture assay. In some embodiments, immune effector cells bearing binding molecules of the invention are at least 60% less sensitive to Titin than are immune effector cells bearing any of a3a, 20a- 18 and 94a- 14 when measured by CD69 co-culture assay, optionally at least 65% less sensitive, optionally at least 70% less sensitive, optionally at least 75% less sensitive. In some embodiments, immune effector cells bearing binding molecules of the invention are at least 80% less sensitive to Titin than are immune effector cells bearing any of a3a, 20a-18 and 94a-14 when measured by CD69 co-culture assay. In some embodiments, immune effector cells bearing binding molecules of the invention are at least 85% less sensitive to Titin than are immune effector cells bearing any of a3a, 20a-18 and 94a-14 when measured by CD69 co-culture assay.

[0193] In some embodiments, immune effector cells bearing binding molecules of the invention are less sensitive to CD166 (SEQ ID NO: 50) than immune effector cells bearing a3a, and optionally also than immune effector cells bearing 20a- 18 and / or 94a-14.

[0194] In some embodiments, immune effector cells bearing binding molecules of the invention are at least 50%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, or at least 85% less sensitive to CD 166 than are immune effector cells bearing the a3a TCR when measured by CD69 coculture assay.

[0195] In some embodiments, immune effector cells bearing binding molecules of the invention are at least 50%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, or at least 85% less sensitive to CD 166 than are cells bearing 20a- 18 and / or 94a- 14 when measured by CD69 co-culture assay.

[0196] In some embodiments, immune effector cells bearing binding molecules of the invention are at least 50%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, or at least 85% less sensitive to CD166 than are cells bearing any of a3a TCR, 20a-18 and 94a-14 when measured by CD69 co-culture assay.

[0197] In some embodiments, immune effector cells bearing binding molecules of the invention are less sensitive to both Titin and CD 166 than immune effector cells bearing a3a, and optionally also than immune effector cells bearing 20a- 18 and / or 94a-14.

[0198] In some embodiments, immune effector cells bearing binding molecules of the invention are at least 50%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, or at least 85% less sensitive to both Titin and CD 166 than are immune effector cells bearing the a3a TCR when measured by CD69 co-culture assay.

[0199] In some embodiments, immune effector cells bearing binding molecules of the invention are at least 50%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, or at least 85% less sensitive to both Titin and CD166 than are cells bearing 20a-18 and / or 94a-14 when measured by CD69 co-culture assay.

[0200] In some embodiments, immune effector cells bearing binding molecules of the invention are at least 50%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, or at least 85% less sensitive to both Titin and CD166 than are cells bearing any of a3a TCR, 20a-18 and 94a-14 when measured by CD69 co-culture assay.

[0201] In some embodiments, immune effector cells binding molecules of the invention do not display any sensitivity to Titin (SEQ ID NO: 46) when measured by CD69 co-culture assay (meaning that no significant increase in CD69 expression is observed relative to control, for example when measured by CD69 co-culture assay as described herein and wherein the HLA- A*01:01-bearing APCs are pulsed with Titin). In some embodiments, immune effector cells binding molecules of the invention do not display any sensitivity to CD166 (SEQ ID NO: 50) when measured by CD69 co-culture assay (meaning that no significant increase in CD69 expression is observed relative to control, for example when measured by co-culture assay when HLA-A*01:01-bearing cells are pulsed with CD 166). In some embodiments, immune effector cells binding molecules of the invention do not display any sensitivity to either Titin or CD 166 when measured by CD69 co-culture assay (meaning that no significant increase in CD69 expression is observed relative to control, for example when measured by co-culture assay when HLA-A*01:01-bearing cells are pulsed with Titin or CD166 respectively).

[0202] Cytotoxic T cells bearing binding molecules of the invention exhibit less cytotoxicity for cells presenting a Titin peptide (e.g. SEQ ID NO: 46) than do cytotoxic T cells bearing a3a, and optionally also than do cytotoxic T cells bearing 20a- 18 and / or do cytotoxic T cells bearing 94a-14. In some embodiments T cells bearing binding molecules of the invention exhibit less cytotoxicity for cells presenting a Titin peptide (e.g. SEQ ID NO: 46) than do T cells bearing any of a3a, 20a-18 and 94a-14. In some embodiments cytotoxic T cells bearing binding molecules of the invention exhibit less cytotoxicity for a CD166 peptide (e.g. SEQ ID NO: 50) than do cytotoxic T cells bearing a3a, and optionally also than do cytotoxic T cells bearing 20a- 18 and / or do cytotoxic T cells bearing 94a- 14. In some embodiments T cells bearing binding molecules of the invention exhibit less cytotoxicity for a CD166 peptide (e.g. SEQ ID NO: 50) than do T cells bearing any of a3a, 20a- 18 and 94a- 14. In some embodiments, T cells bearing binding molecules of the invention exhibit less cytotoxicity for both a Titin peptide (SEQ ID NO: 46) and a CD166 peptide (SEQ ID NO: 50) than do T cells bearing a3a, and optionally also than do T cells bearing 20a-18 and / or do T cells bearing 94a-14. In some embodiments T cells bearing binding molecules of the invention exhibit less cytotoxicity for a Titin peptide (SEQ ID NO: 46) and a CD166 peptide (SEQ ID NO: 50) than do T cells bearing any of a3a, 20a-18 and 94a-14.

[0203] In some embodiments, cytotoxic T cells binding molecules of the invention do not display any cytotoxicity for Titin (meaning that no significant reduction in Titin -APC confluence is observed). In some embodiments, cytotoxic T cells binding molecules of the invention do not display any cytotoxicity for CD 166 (meaning that no significant reduction in CD166-APC confluence is observed). In some embodiments, cytotoxic T cells binding molecules of the invention do not display any cytotoxicity for either Titin or CD 166 (meaning that no significant reduction in Titin -APC or CD166-APC confluence is observed respectively).

[0204] Cross-reactivity

[0205] The a3a TCR is understood, as described herein, to bind to all of the antigens provided in Table 2. The inventors have validated in vitro (via CD69 co-culture assay) that the a3a TCR binds to all of the antigens in Table 2, and particularly all the antigens in Table 3. In other words, the a3a TCR is cross-reactive.

[0206] Cross-reactivity means exhibiting meaningful sensitivity to non-target antigens. Therefore, herein, cross-reactivity means immune effector cells bearing binding molecules exhibiting meaningful sensitivity to pMHCs other than MAGE-A3, optionally also other than MAGE-A6, PLD-5, MAGE- B 18 and / or hFATl.

[0207] Table 2. Peptide antigens to which a3a binds

[0208] Meaningful sensitivity means observing by CD69 co-culture assay that CD69 expression levels are significantly higher than those observed using a non-agonist peptide-MHC control. Unless otherwise specified the non-agonist peptide control may be SEQ ID NO: 75 (PARP-1 peptide). For example, a preferred means of testing for meaningful sensitivity is by coculturing for 18 hours CD8+ Jurkat T cells expressing a binding molecule of the invention with HLA-A*01:01 K562 cells pulsed with 100 uM of test peptide or a non-agonist peptide control SEQ ID NO: 75, and measuring CD69 expression levels of the T cells. In this example, meaningful sensitivity for the test peptide is indicated by CD69 expression levels significantly (i.e. by more than 1 standard deviation or p<0.05) higher than those observed when pulsing with the control.

[0209] Binding molecules of the invention exhibit low cross-reactivity. Binding molecules of the invention are less cross-reactive than TCRs of the prior art, meaning immune effector cells bearing the binding molecules exhibit meaningful sensitivity to fewer non-target antigens than immune effector cells bearing TCRs of the prior art. Binding molecules of the invention are less cross-reactive than a3a. Binding molecules of the invention are less cross-reactive than a3a, 20a-18 and 94a-14. Many of the peptide antigens to which a3a has been shown to bind are self-antigens. In particular, a3a has been shown to bind to at least the self-antigens shown in Table 3. In other words, the a3a TCR is self-reactive.

[0210] Self-reactivity means exhibiting meaningful sensitivity to one or more self-antigens expressed in healthy tissue. In some embodiments, binding molecules of the invention do not exhibit any meaningful sensitivity to self-antigens expressed exclusively in healthy tissue.

[0211] Table 3. Exemplary self peptide antigens expressed in healthy tissue to which the a3a TCR displays off-target binding.

[0212] Binding molecules of the invention exhibit low self-reactivity. Binding molecules of the invention are less self-reactive than TCRs of the prior art, meaning they exhibit meaningful sensitivity to fewer self-antigens expressed in healthy tissue than TCRs of the prior art. Binding molecules of the invention are less self-reactive than a3a. Binding molecules of the invention are less self-reactive than a3a, 20a-18 and 94a-14.

[0213]

[0214] Table 4. Non-target peptide antigens to which the a3a TCR displays off-target binding.

[0215] In some embodiments, binding molecules of the invention do not exhibit meaningful sensitivity to Titin. In some embodiments, binding molecules of the invention do not exhibit meaningful sensitivity to one or more, two or more, three or more or at least four of the antigens in Table 4. In some embodiments, binding molecules of the invention do not exhibit meaningful sensitivity to one or more, two or more, three or more or at least four of the selfantigens in Table 3.

[0216] In some embodiments, binding molecules of the invention do not exhibit meaningful sensitivity to one or more of Titin, CD 166, STK MRCKa, FXYD6, ANKRD16, Dyn4, COG4, protocadherin Fat2 and inactive PLD5. In some embodiments, binding molecules of the invention do not exhibit meaningful sensitivity to one or more of CD 166, S / MRCKa, Titin, ANKRD16, Dyn4, COG4 and FXYD6, optionally one or more of Titin, CD 166, S / MRCKa or FXYD6. In some embodiments, binding molecules of the invention do not exhibit meaningful sensitivity to two or more or optionally three or more of Titin, CD 166, S / MRCKa or FXYD6.

[0217] In some embodiments binding molecules exhibit less than 4%, less than 3%, less than 2.5% or less than 2.0% CD69 expression in co-culture assay with one or more of Titin, CD166, STK MRCKa, FXYD6, ANKRD16, Dyn4, COG4, protocadherin Fat2 and inactive PLD5, optionally with one or more of CD 166, S / MRCKa, Titin, ANKRD16, Dyn4, COG4 and FXYD6, optionally with one or more of Titin, CD 166, S / MRCKa or FXYD6 relative to control. In some embodiments, binding molecules exhibit less than 4%, less than 3%, less than 2.5% or less than 2.0% CD69 expression in co-culture assay with two or more or optionally three or more of Titin, CD 166, S / MRCKa or FXYD6.

[0218] In preferred embodiments, binding molecules of the invention do not exhibit meaningful sensitivity to any of Titin, CD 166, STK MRCKa and FXYD6, i.e. do not display significant CD69 expression when stimulated with any of Titin, CD 166, STK MRCKa and FXYD6 relative to a negative control. In even more preferred embodiments, binding molecules of the invention do not exhibit meaningful sensitivity to any of CD 166, S / MRCKa, Titin, ANKRD16, Dyn4, COG4 and FXYD6.

[0219] In some embodiments, binding molecules of the invention do not display any sensitivity to one or more of Titin, CD 166, S / MRCKa or COG4 (optionally presented via HLA-A*01:01) when measured by CD69 co-culture assay (meaning that no significant increase in CD69 expression is observed relative to control, for example when measured by co-culture assay when HLA-A*01 :01 - bearing cells are pulsed with Titin, CD 166, S / MRCKa or COG4 respectively). In some embodiments, binding molecules of the invention do not display any sensitivity to any of Titin, CD 166, S / MRCKa or COG4 (optionally presented via HLA-A*01:01) when measured by CD69 co-culture assay (meaning that no significant increase in CD69 expression is observed relative to control, for example when measured by co-culture assay when HLA-A* 01:01 -bearing cells are pulsed with Titin, CD 166, S / MRCKa or COG4 respectively).

[0220] Advantageously, unlike the a3a TCR, binding molecules of the invention bind to self-antigens Titin, CD 166, S / MRCKa and / or COG4 with only very low binding affinity, and optionally do not bind to any of the self-antigens Titin, CD 166, S / MRCKa or COG4. In some embodiments binding molecules of the invention do not bind to one or more, two or more, three or more or at least four of the self-antigens in Table 3. Said self-antigens may be presented on HLA-A 1, optionally HLA- A*01:01.

[0221] For example, in some embodiments a binding molecule of the invention displays more than 35 pM KD binding affinity for one or more of Titin, CD 166, S / MRCKa and COG4 (optionally when presented via HLA-A*01:01), optionally more than 40 pM KD, optionally more than 50 pM KD binding affinity, or optionally more than 70 pM KD binding affinity for one or more of Titin, CD 166, S / MRCKa and COG4 (optionally when presented via HLA-A* 01:01). In some embodiments, a binding molecule of the invention displays more than 100 pM KD binding affinity for one or more of Titin, CD 166, S / MRCKa and COG4 (optionally when presented via HLA-A* 01:01). In some embodiments, a binding molecule of the invention displays more than 200 pM KD binding affinity for one or more of Titin, CD166, S / MRCKa and COG4 (optionally when presented via HLA-A*01:01). In some embodiments, a binding molecule of the invention displays more than 300 pM KD binding affinity for one or more of Titin, CD 166, S / MRCKa and COG4 (optionally when presented via HLA-A*01:01). In some embodiments, a binding molecule of the invention displays more than 400 pM KD binding affinity for one or more of Titin, CD 166, S / MRCKa and COG4 (optionally when presented via HLA-A*01:01). In some embodiments, a binding molecule of the invention displays more than 500 pM KD binding affinity for one or more of Titin, CD 166, S / MRCKa and COG4 (optionally when presented via HLA-A* 01:01). In some embodiments, a binding molecule of the invention displays essentially no binding for one or more of Titin, CD166, S / MRCKa and COG4 (optionally when presented via HLA-A*01:01), for example displays so little binding that no read-out can be detected.

[0222] In some embodiments a binding molecule of the invention displays more than 35 pM KD binding affinity for all of Titin, CD 166, S / MRCKa and COG4 (optionally when presented via HLA-A*01:01), optionally more than 40 pM KD, optionally more than 50 pM KD binding affinity, or optionally more than 70 pM KD binding affinity for all of Titin, CD 166, S / MRCKa and COG4 (optionally when presented via HLA-A*01:01). In some embodiments, a binding molecule of the invention displays more than 100 pM KD binding affinity for all of Titin, CD 166, S / MRCKa and COG4 (optionally when presented via HLA-A*01:01). In some embodiments, a binding molecule of the invention displays more than 200 pM KD binding affinity for all of Titin, CD 166, S / MRCKa and COG4 (optionally when presented via HLA- A*01:01). In some embodiments, a binding molecule of the invention displays more than 300 pM KD binding affinity for all of Titin, CD 166, S / MRCKa and COG4 (optionally when presented via HLA-A*01:01). In some embodiments, a binding molecule of the invention displays more than 400 pM KD binding affinity for all of Titin, CD 166, S / MRCKa and COG4 (optionally when presented via HLA-A*01:01). In some embodiments, a binding molecule of the invention displays more than 500 pM KD binding affinity for all of Titin, CD 166, S / MRCKa and COG4 (optionally when presented via HLA-A* 01:01). In some embodiments, a binding molecule of the invention displays essentially no binding for all of Titin, CD 166, S / MRCKa and COG4 (optionally when presented via HLA-A* 01:01), for example displays so little binding that no read-out can be detected.

[0223] Polynucleotides, vectors and cells

[0224] Also provided is one or more isolated polynucleotide(s) encoding a binding molecule of the invention. In some cases, the nucleic acid sequence encoding the binding molecule is collectively present on more than one polynucleotide but collectively together the polynucleotides are able to encode a binding molecule of the invention. For example, provided herein is a first polynucleotide comprising a nucleic acid sequence encoding the alpha chain variable domain, optionally also comprising a nucleic acid sequence encoding the alpha chain constant domain, and a second polynucleotide comprising a nucleic acid sequence encoding the beta chain variable domain, optionally also comprising a nucleic acid sequence encoding the beta chain constant domain. In some embodiments are provided one or more polynucleotides encoding an alpha chain variable domain and / or a beta variable domain of a binding molecule of the invention.

[0225] However, in some embodiments the binding molecule is encoded by a single polynucleotide. The single polynucleotide may comprise a first open reading frame comprising the alpha chain variable domain nucleic acid sequence (optionally also the alpha chain constant domain nucleic acid sequence) and a second open reading frame comprising the beta chain variable domain nucleic acid sequence (optionally also the beta chain constant domain nucleic acid sequence) such that the alpha chain and beta chain sequences are expressed on separate transcripts. In other embodiments, the single polynucleotide may comprise a single open reading frame comprising both the alpha chain variable domain nucleic acid sequence (optionally also the alpha chain constant domain nucleic acid sequence) and the beta chain variable domain nucleic acid sequence (optionally also the beta chain constant domain nucleic acid sequence). The resulting transcript may be translated into a single polypeptide which may then be cleaved post-translationally into a first polypeptide chain comprising the alpha chain variable domain amino acid sequence (optionally also the alpha chain constant domain amino acid sequence) and a second polypeptide chain comprising the beta chain variable domain amino acid sequence (optionally also the beta chain constant domain amino acid sequence) of the binding molecules of the invention. Enzymes such as endopeptidases suitable for cleavage of the fusion protein into its first and second polypeptide chains are well known in the art. In other embodiments the resulting transcript may be translated into a first polypeptide chain comprising the alpha chain variable domain amino acid sequence (optionally also the alpha chain constant domain amino acid sequence) and a second polypeptide chain comprising the beta chain variable domain amino acid sequence (optionally also the beta chain constant domain amino acid sequence) of the binding molecules of the invention according to methods familiar to the skilled person.

[0226] Polynucleotides which encode binding molecules of the invention or their component parts can be obtained by methods well known to those skilled in the art. The polynucleotide may be a DNA sequence. The polynucleotide may be an RNA sequence, such as mRNA.

[0227] A vector may comprise the one or more polynucleotides described above.

[0228] The vector may be a viral vector. Conventional viral based expression systems could include retroviral, alpha-retroviral, lentivirus, adenoviral, adeno-associated virus (AAV) and herpes simplex virus (HSV) vectors for gene transfer. Non-viral transduction vectors include transposon based systems including PiggyBac and Sleeping Beauty systems. Methods for producing and purifying such vectors are known in the art.

[0229] The vector may be a cloning vector or an expression vector. A suitable vector may be any vector which is capable of carrying a sufficient amount of genetic information, and allowing expression of a polypeptide of the invention. The vector may be an RNA vector. Suitable RNA vectors include the RNA vectors as described in Schutsky, Keith, et al., Oncotarget 6.30 (2015): 28911 and Beatty, Gregory L., et al., Gastroenterology 155.1 (2018): 29-32.

[0230] General methods by which the vectors may be constructed, transfection methods and culture methods are well known to those skilled in the art. In this respect, reference is made to “Current Protocols in Molecular Biology”, 1999, F. M. Ausubel (ed), Wiley Interscience, New York and the Maniatis Manual produced by Cold Spring Harbor Publishing.

[0231] A polynucleotide may be provided in the form of an expression construct (or ‘expression cassette’), which includes control sequences operably linked to the inserted sequence, thus allowing for expression of a binding molecule of the invention in vivo. Hence, also provided is one or more expression cassettes encoding the one or more polynucleotides that encode a binding molecule of the invention. These expression cassettes, in turn, are typically provided within vectors (e.g. plasmids or recombinant viral vectors). Hence, also provided is a vector comprising a polynucleotide encoding a binding molecule of the invention. Also provided is a vector comprising a polynucleotide encoding an alpha chain variable domain of a binding molecule of the invention. Further provided are vectors which collectively encode a binding molecule of the invention.

[0232] The vector may be a human artificial chromosome. Human artificial chromosomes are described in e.g. Kazuki et al., Mol. Ther. 19(9): 1591-1601 (2011), and Kouprina et al., Expert Opinion on Drug Delivery 11(4): 517-535 (2014).

[0233] The vector may be a non-viral delivery system, such as DNA plasmids, naked nucleic acid (e.g. naked RNA), and nucleic acid complexed with a delivery vehicle, such as a liposome.

[0234] The binding molecules, polynucleotides, expression cassettes or vectors described herein may be introduced into a cell (for example a host cell), e.g. by transfection. Hence, also provided is a cell comprising a binding molecule of the invention. Also provided is a cell comprising one or more nucleic acids, expression cassettes or vectors of the invention. The nucleic acids, expression cassettes or vectors described herein may be introduced transiently or permanently into the cell, allowing expression of binding molecule from the one or more nucleic acids, expression cassettes or vectors. Such cells include transient, or preferably stable higher eukaryotic cell lines, such as mammalian cells or insect cells, lower eukaryotic cells, such as yeast, or prokaryotic cells, such as bacteria cells. Particular examples of cells include mammalian HEK293, such as HEK293F, HEK293T, HEK293S or HEK Expi293F, CHO, HeLa, NS0 and COS cells, or any other cell line used herein.

[0235] The cell may be an isolated cell.

[0236] Typically, the cell is an immune cell, optionally an immune effector cell. In some embodiments the cell is a T lymphocyte, optionally an inflammatory T lymphocyte, a cytotoxic T lymphocyte, a regulatory T lymphocyte, or a helper T lymphocyte. A cell of the invention may be a CD8+ cytotoxic T lymphocyte. A cell of the invention may be a CD4+ helper T lymphocyte.

[0237] The polynucleotide or vector of the invention may be an mRNA for administering to patients, e.g. mRNA vaccination. The patient’s T cells may then express the receptor of the invention and / or the accessory receptor of the invention in vivo. Such mRNA molecules and associated methods are described in Rurik et al., Science 375(6576):91-96 (2022).

[0238] Also provided is a population of cells of the invention.

[0239] Also provided is a kit suitable for transforming and / or transfecting an immune effector cell or a population of immune effector cells to generate an immune effector cell or population of immune effector cells of the invention. The kit comprises a polynucleotide or vector as described herein. The kit may comprise further agents such as those discussed herein that improve transfection or transformation efficacy.

[0240] Pharmaceutical composition

[0241] Also provided are pharmaceutical compositions comprising a binding molecule, polynucleotide, pair of polynucleotides, vector and / or cell or cells of the invention and further comprising a pharmaceutically acceptable carrier and / or excipient. Remington ’s Pharmaceutical Sciences, by E. W. Martin, Mack Publishing Co., Easton, PA, 19th Edition, 1995, describes compositions and formulations suitable for pharmaceutical delivery of the binding molecules herein disclosed.

[0242] Suitable pharmaceutically acceptable carriers comprise aqueous carriers, diluents or excipients. Examples of suitable carriers include all aqueous and non-aqueous isotonic sterile injection solutions which may contain anti-oxidants, buffers and solutes, which render the composition isotonic with the blood of the intended recipient; aqueous and non-aqueous sterile suspensions, which may include suspending agents and thickening agents, dispersion media, antifungal and antibacterial agents, isotonic and absorption agents and the like. It will be understood that compositions of the invention may also include other supplementary physiologically active agents.

[0243] The carrier is typically pharmaceutically “acceptable” in the sense of being compatible with the other ingredients in the composition and not injurious to the subject. Compositions include those suitable for parenteral administration, including subcutaneous, intramuscular, intravenous, ophthalmic, optic and intradermal administration. The compositions may conveniently be presented in unit dosage form and may be prepared by any method well known in the art of pharmacy. Such methods include preparing the carrier for association with isolated T cells. In general, the compositions are prepared by uniformly and intimately bringing into association any active ingredients with liquid carriers.

[0244] In general, the nature of the carrier or excipient will depend on the particular mode of administration being employed. For instance, parenteral formulations usually comprise injectable fluids that include pharmaceutically and physiologically acceptable fluids, which include, but are not limited to, water, physiological saline, balanced salt solutions, aqueous dextrose, glycerol or the like as a vehicle. For solid compositions (e.g., powder, pill, tablet, or capsule forms), conventional non-toxic solid carriers can include, for example, pharmaceutical grades of mannitol, lactose, starch, or magnesium stearate. In addition to biologically neutral carriers, pharmaceutical compositions to be administered can contain minor amounts of non-toxic auxiliary substances, such as wetting or emulsifying agents, preservatives, and pH buffering agents and the like, for example sodium acetate or sorbitan monolaurate.

[0245] The composition may be suitable for parenteral administration. Compositions suitable for parenteral administration include aqueous and non- aqueous isotonic sterile injection solutions which may contain anti-oxidants, buffers, bactericides and solutes, which render the composition isotonic with the blood of the intended recipient; and aqueous and non-aqueous sterile suspensions which may include suspending agents and thickening agents. In another embodiment, the composition is suitable for intravenous administration.

[0246] The composition described herein may be prepared in a manner known in the art and are those suitable for parenteral administration to mammals, particularly humans, comprising a therapeutically effective amount of the composition with one or more pharmaceutically acceptable carriers or diluents.

[0247] Provided in some embodiments is a composition comprising an immune effector cell or population of immune effector cells of the invention. The immune effector cell or population of immune effector cells may be at least 1% of the total cells in the composition, such as at least 5%, at least 10%, at least 15 at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, at least 99% or at least 99.9% of the total cells in the composition. The total cells in the composition may consist or consist essentially of the immune effector cell or population of immune effector cells of the invention, i.e. no other cells are detectable in the composition. The composition may comprise at least about IxlO6to about IxlO12of the immune effector cells of the invention.

[0248] The present disclosure also contemplates the combination of the composition described herein with other active agents and / or in addition to other treatment regimens or modalities such as radiation therapy or surgery. When the composition described herein is used in combination with known active agents, the combination may be administered either in sequence (either continuously or broken up by periods of no treatment) or concurrently or as an admixture.

[0249] Suitable anti-cancer agents will be known to persons skilled in the art.

[0250] Treatment in combination is also contemplated to encompass the treatment with either the composition of the invention followed by a known treatment, or treatment with a known agent followed by treatment with the composition of the invention, for example, as maintenance therapy.

[0251] For example, in the treatment of cancer it is contemplated that the composition of the present invention may be administered in combination with an alkylating agent (such as mechlorethamine, cyclophosphamide, chlorambucil, ifosfamidecysplatin, or platinum -containing alkylating agents such as cisplatin, carboplatin and oxaliplain), and anti -metabolite (such as a purine or pyrimidine analogue or an anti-folate agent, such as azathioprine and mercaptopurine), an anthracycline (such as daunorubicin, doxorubicin, epirubicin idarubicin, valrubicin, mitoxantrone or anthracycline analog), a plant alkaloid (such as a vinca alkaloid or a taxane, such as vincristine, vinblastine, vinorelbine, vindesine, paclitaxel or doestaxel), a topoisomerase inhibitor (such as a type I or type II topoisomerase inhibitor), a podophyllotoxin (such as etoposide or teniposide), a tyrosine kinase inhibitor (such as imatinib mesylate, nilotinib or dasatinib), an adenosine receptor inhibitor (such as A2aR inhibitors, SCH58261, CPI-444, SYN115, ZM241385, FSPTP or A2BR inhibitors such as PSB-1115), adenosine receptor agonists (such as CCPA, IB-MECA and CI-IB-MECA), a checkpoint inhibitor, including those of the PDL-1:PD-1 axis, nivolumab, pembrolizumab, atezolizumab, BMS- 936559, MEDI4736, MPDL33280A or MSB0010718C), an inhibitor of the CTLA-4 pathway (such as ipilimumab and tremelimumab), an inhibitor of the TIM- 3 pathway or an agonist monoclonal antibody that is known to promote T cell function (including anti-OX40, such as MEDI6469; and anti-4-BB, such as PF-05082566).

[0252] The invention also provides a kit or article of manufacture including a pharmaceutical composition as described above.

[0253] The invention also provides a kit for use in a therapeutic application mentioned above, the kit comprising: (a) a container holding a polypeptide, nucleic acid, vector or pharmaceutical composition of the invention; and (b) a label or package insert with instructions for use.

[0254] Suitable containers include, for example, bottles, vials, syringes, blister pack, etc. The containers may be formed from a variety of materials such as glass or plastic. The container holds a therapeutic composition which is effective for treating the condition and may have a sterile access port (e.g, the container may be an intravenous solution bag or a vial having a stopper pierceable by a hypodermic injection needle). The label or package insert indicates that the therapeutic composition is used for treating the condition of choice. In an embodiment, the label or package insert includes instructions for use and indicates that the therapeutic or prophylactic composition can be used to treat a cancer or other condition described herein.

[0255] The kit may further comprise a container comprising a pharmaceutically -acceptable buffer, such as bacteriostatic water for injection (BWFI), phosphate -buffered saline, Ringer's solution and dextrose solution. It may further comprise other materials desirable from a commercial and user standpoint, which would be known to persons skilled in the art, suitable examples of which include other buffers, diluents, filters, needles, and syringes. Therapeutic uses

[0256] Also described herein is use of a binding molecule, polynucleotide, pair of polynucleotides, vector, cell or cells, or a pharmaceutical composition of the invention, in a method of treatment of the human or animal body by therapy, e.g. for use as a medicament.

[0257] For instance, also provided is a method of treating cancer in a subject, the method comprising administering to the subject an effective amount of an immune effector cell or a population of immune effector cells of the invention. Hence, the invention also provides a binding molecule, polynucleotide, pair of polynucleotides, vector, cell or cells, or a pharmaceutical composition of the invention for use in a method of treating cancer. The invention also provides the use of a binding molecule, polynucleotide, pair of polynucleotides, vector, cell or cells, or a pharmaceutical composition of the invention for the manufacture of a medicament for the treatment of cancer. The invention also provides the use of a binding molecule, polynucleotide, pair of polynucleotides, vector, cell or cells, or a pharmaceutical composition of the invention to treat cancer. The cancer may be any cancer, such as a solid tumour cancer. The cancer may be a melanoma, non-small cell lung carcinoma, head and neck squamous cell carcinoma, bladder cancer, oesophageal cancer, gastric cancer, hepatocellular carcinoma, breast cancer, ovarian cancer, colorectal cancer, multiple myeloma, sarcoma, prostate cancer, pancreatic cancer, cervical cancer, testicular cancer and / or glioblastoma.

[0258] Also provided is a method of treating any MAGE-A3 associated disease or disorder in a subject, the method comprising administering to the subject an effective amount of an immune effector cell or a population of immune effector cells of the invention. Hence, the invention also provides a binding molecule, polynucleotide, pair of polynucleotides, vector, cell or cells, or a pharmaceutical composition of the invention for use in a method of treating a MAGE -A3 associated disease or disorder. The invention also provides the use of a binding molecule, polynucleotide, pair of polynucleotides, vector, cell or cells, or a pharmaceutical composition of the invention for the manufacture of a medicament for the treatment of a MAGE -A3 associated disease or disorder. The invention also provides the use of a binding molecule, polynucleotide, pair of polynucleotides, vector, cell or cells, or a pharmaceutical composition of the invention to treat a MAGE -A3 associated disease or disorder.

[0259] Also provided is a method of performing adoptive cell therapy in a subject, the method comprising administering to the subject an effective amount of an immune effector cell or a population of immune effector cells of the invention.

[0260] Hence, the invention also provides a binding molecule, polynucleotide, pair of polynucleotides, vector, cell or cells, or a pharmaceutical composition of the invention for use in adoptive cell therapy. The invention also provides the use of a binding molecule, polynucleotide, pair of polynucleotides, vector, cell or cells, or a pharmaceutical composition of the invention for the manufacture of a medicament for adoptive cell therapy.

[0261] The invention also provides the use of a binding molecule, polynucleotide, pair of polynucleotides, vector, cell or cells, or a pharmaceutical composition of the invention for adoptive cell therapy.

[0262] The therapeutic uses and methods may comprise administering a therapeutically effective amount of the immune effector cell or population of immune effector cells.

[0263] Also provided is a method of formulating a composition for treating cancer, wherein said method comprises mixing an immune effector cell or population of immune effector cells of the invention with an acceptable carrier to prepare said composition.

[0264] The subject may have been previously treated for the cancer, such as using adoptive cell therapy.

[0265] The therapeutic methods and uses may comprise, prior to treatment with an immune effector cell or population of immune effector cells of the invention, determining whether the cancer expresses a target antigen specifically targeted by immune effector cell or population of immune effector cells of the invention, for example MAGE-A3.

[0266] The method may comprise selecting an immune effector cell or population of immune effector cells based on the expression of the target antigen by the cancer, so that the immune effector cell or population of immune effector cell is specific for the cancer. The method may comprise transfecting or transforming an immune effector cell with a nucleic acid of the invention in response to information on the expression of the target antigen by the cancer.

[0267] The therapeutic methods and uses described herein may comprise inhibiting the disease state (i.e. the cancer), for example by arresting its development and / or causing regression of the disease state until a desired end point is reached. The therapeutic methods and uses of the invention may comprise achieving a partial response, or a full response by the cancer. The therapeutic methods and uses of the invention may achieve remission of the cancer.

[0268] The therapeutic methods and uses described herein may delay the growth of the cancer, arrest the growth of the cancer and / or reverse the growth of the cancer. The therapeutic methods and uses of the invention may reduce the size of the cancer by at least 10%, such as at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90% or by 100%.

[0269] Typically, the therapeutic methods and uses are for a human subject in need thereof. However, non-human animals such as non-human mammals are also contemplated. The non-human mammals may be for example (but not limited to) rats, rabbits, sheep, pigs, cows, cats or dogs.

[0270] The dose of the immune effector cell or population of immune effector cells may vary depending on the age and size of a subject, as well as on the disease, conditions and route of administration. The immune effector cell or population of immune effector cells may be administered at a dose of about IxlO6to about IxlO12cells. The immune effector cell or population of immune effector cells may be administered at a dose of about IxlO5cells / kg to about IxlO11cells / kg body weight.

[0271] The immune effector cell or population of immune effector cells may be administered as a single dose. The immune effector cell or population of immune effector cells may be administered in a multiple dose regimen. For example, the initial dose may be followed by administration of a second or plurality of subsequent doses. The second and subsequent doses may be separated by an appropriate time. For example, the doses between doses may be administered once about every week, once about every 2 weeks, once about every 3 weeks, once about every four weeks, or once about every month.

[0272] The immune effector cell or population of immune effector cells may be administered intravenously.

[0273] The immune effector cell or population of immune effector cells may be administered with one or more additional therapy, such as one or more additional therapeutic agents. The additional therapeutic agent may be an anti-tumour agent. The additional therapeutic may be an additional immune effector cell.

[0274] Combined administration of the immune effector cell or population with the additional therapeutic agent may be achieved in a number of different ways. All the components may be administered together in a single composition. Each component may be administered separately as part of a combined therapy.

[0275] For example, the immune effector cell or the population of immune effector cells of the invention may be administered before, after or concurrently with the additional therapeutic agent. The additional therapy may be chemotherapy, radiotherapy and / or surgery.

[0276] Prior to administration of the immune effector cell or population of immune effector cells of the invention, the subject may undergo lymphodepletion. Lymphodepletion may be achieved via administration to the subject with fluradabine, cyclophosphamide and / or bendamustine. Lymphodepletion may be carried out for at least about one day, such as about 2 days or about 3 days.

[0277] The biological activity and / or therapeutic efficacy of the administered immune effector cell or population of immune effector cells may be measured by known methods. For example, the method may comprise imaging, such as magnetic resonance imaging. Reduced bonds to the HLA molecule and increased bonds to MAGE A-3

[0278] The present invention also provides a binding molecule comprising:

[0279] (A) a T cell receptor (TCR) alpha chain variable domain comprising:

[0280] (i) a complementary determining region (CDR) 1 comprising an amino acid sequence DSAIYN (SEQ ID NO: 1) or comprising an amino acid sequence differing from SEQ ID NO: 1 by one amino acid, (ii) a CDR2 comprising the amino acid sequence of Formula I (SEQ ID NO: 2), and

[0281] (iii) a CDR3 comprising the amino acid sequence CAVRPGGAGSYQLTF (SEQ ID NO: 3) or comprising an amino acid sequence differing from SEQ ID NO: 3 by one amino acid; and

[0282] (B) a TCR beta chain variable domain comprising:

[0283] (i) a CDR1 comprising the amino acid sequence SGHRS (SEQ ID NO: 4) or comprising an amino acid sequence differing from SEQ ID NO: 4 by one amino acid,

[0284] (ii) a CDR2 comprising the amino acid sequence YFSETQ (SEQ ID NO: 5) or comprising an amino acid sequence differing from SEQ ID NO: 5 by one amino acid, and

[0285] (iii) a CDR3 comprising the amino acid sequence CASSPNMADEQYF (SEQ ID NO: 6) or comprising an amino acid sequence differing from SEQ ID NO: 6 by one amino acid; wherein Formula I (SEQ ID NO: 2) is Leu-Nl-N2-N3-N4, the binding molecule binds to a complex of an epitope of SEQ ID NO: 45 (MAGE -A3) and an HLA molecule, and wherein the binding molecule forms fewer bonds to the HLA molecule and more bonds to the epitope of SEQ ID NO: 45 compared to a TCR comprising an alpha chain comprising an amino acid sequence of SEQ ID NO: 36 and a beta chain comprising an amino acid sequence of SEQ ID NO: 40.

[0286] Optionally, the binding molecule forms fewer hydrogen bonds and / or salt bridges to the HLA molecule compared to a TCR comprising an alpha chain comprising an amino acid sequence of SEQ ID NO: 36 and a beta chain comprising an amino acid sequence of SEQ ID NO: 40. Optionally, the binding molecule forms more hydrogen bonds and / or salt bridges to the epitope of SEQ ID NO: 45 compared to a TCR comprising an alpha chain comprising an amino acid sequence of SEQ ID NO: 36 and a beta chain comprising an amino acid sequence of SEQ ID NO: 40. Whether a binding molecule fewer hydrogen bonds and / or salt bridges to the HLA molecule compared to a TCR comprising an alpha chain comprising an amino acid sequence of SEQ ID NO: 36 and a beta chain comprising an amino acid sequence of SEQ ID NO: 40 and / or forms more hydrogen bonds and / or salt bridges to the epitope of SEQ ID NO: 45 compared to a TCR comprising an alpha chain comprising an amino acid sequence of SEQ ID NO: 36 and a beta chain comprising an amino acid sequence of SEQ ID NO: 40 can be determined by solving the crystal structure of the TCR in the presence of MAGE-A3-HLA- A*01:01 (for example as described in Example 9) and determining the number of bonds, for examples hydrogen bonds and / or salt bridges between the binding molecule and the MAGE-A3 peptide.

[0287] Optionally the structure is solved at 2.4 A and / or 2.8 A. Optionally, a hydrogen bond is defined as two atoms capable of forming a hydrogen bond that are less than 3.6 A apart. Optionally, a salt bridge is defined as two atoms capable of forming a salt bridge that are less than 4 A apart.

[0288] Optionally, N 1 is He or Vai,

[0289] N2 is Gin or Arg,

[0290] N3 is Pro or Ser, N4 is Tyr or Ser, and wherein: when N2 is Gin, N4 is Tyr; and when N2 is Arg, N3-N4 is Pro-Ser, Ser-Ser or Ser-Tyr.

[0291] Optionally, the binding molecule has one of more of the features described above in relation to binding molecules of the invention.

[0292] Examples

[0293] Example 1 - Materials and methods

[0294] Cloning

[0295] The pHR vector with Mini and Notl restriction sites was used for the generation of TCR- Jurkat and A1-K562 cell lines. For CD8 expression on Jurkat cells, the pHRi vector was used.

[0296] The a3a TCR Va (SEQ ID NO: 102) and VP (SEQ ID NO: 108) fragments were synthesised with the respective wild-type a3a signal peptides (IDT) with suitable bp overhangs for Gibson Assembly into separate pHR vectors alongside the full length Ca (SEQ ID NO: 109) and C (SEQ ID NO: 110) fragments, respectively, with V and C fragments in equimolar quantities. The HLA- A*01:01 heavy chain sequence was obtained from IPD-IMGT / HLA and synthesised along with its endogenous signal peptide (IDT) for cloning into the pHR vector. The products of the a3aa, a3ap, and HLA-A*01:01 ITA reactions were used to transform DH5a E. Colt (NEB C2987H) that were subsequently grown on LB agar plates with 1 pg / mL carbenicillin overnight. Single bacterial colonies were grown in liquid LB bacterial culture medium with 1 pg / mL carbenicillin overnight, and the cloned DNA plasmids were extracted and purified the following morning. All cloned DNA plasmids were sent for sequencing (GENEWIZ) to confirm sequence integrity prior to downstream application.

[0297] Lentivirus transfections

[0298] Lentivirus infections were used to generate stably transfected cell lines expressing the desired genes (a3a TCRa, a3a TCRp, and HLA-A*01:01 heavy chain): HEK-293T cells, grown in DMEM supplemented with 10% FBS, 1% GlutaMax, and 1% penicillin / streptomycin (v / v), were plated in six well plates with 3xl05cells / plate (2 mb total) (all reagents from Gibco, Thermo Scientific, UK). After 24h, for each well of cells, 750 ng plasmid of interest, 500 ng psPAX, and 260 ng pMD2.G were mixed with 4.5 pL Fugene transfection reagent (Promega) in 100 pL Opti-MEM and rested for 20 minutes. In the meantime, cDMEM was removed from each well, and fresh cRPMI was added to each well. After 20 minutes, the DNA / Fugene mixture was added to each well, the plates were gently rocked up-down, side-side, and then incubated for 3 days. All cell lines grown at 37°C, 5% CO2

[0299] Cell line transductions

[0300] 72 hours later, the lentivirus supernatants were removed and spun down at 3300 rpm for 10 minutes to remove any unwanted cells. For a3a TCR Jurkat cells, 500 pL of a3a TCRa lentivirus and 500 pL of a3a TCRp lentivirus were added to IxlO6TCR -I- Jurkat cells or CD8+ TCR -I- Jurkat cells (TCR Jurkat cells transduced with a CD8-pHRi vector). For HLA-A*01:01 K562 cells, 500 pL of the HLA-A*01:01 heavy chain was added to 0.5x106WT K562 cells, assembling with endogenous P2M in the ER. Transduced Jurkat and K562 cells were grown in cRPMI. All cell lines grown at 37°C, 5% CO2 Cells were analysed using FACS after 5 days of lentiviral infection to confirm expression of the desired proteins (anti-TCR PE, BioLegend; anti-CD8 PB, BioLegend).

[0301] Co-culture of a3a antigens

[0302] K562 cells were stably transduced with the HLA-A*01:01 heavy chain. K562 cells are a lymphoblast cell line that (at rest) do not express surface MHC molecules, but do possess free cytosolic P2M, enabling assembly and cell surface expression of the HLA-A*01:01-p2M complex for the presentation of peptides. Additionally, Jurkat T-cells were stably transduced with the a3a TCR. TCR and MHC expression were determined by flow cytometry with appropriate antibodies. Validation of HLA-A*01:01 expression on K562 cells was conducted by comparing the activation levels of a3a-Jurkat cells cultured with HLA-A*01:01 K562 cells pulsed with or without the MAGE- A3 peptide, understanding that successful HLA-A*01:01 expression would lead to a significant increase in activation levels of the a3a-Jurkat cells exposed to MAGE -A3 -A*01:01. “Pulsing” APCs with HLA-restricted peptides is an effective way of presenting desired peptides to T cells: the presence of a high concentration of free peptide in the medium leads to a shift in equilibrium that forces the exchange of the endogenously presented peptide for the desired synthetic peptide. Peptide- pulsing is a widely used and straightforward method for testing and titrating the effect of a large number of pre-defined peptides on T cell responses.

[0303] Peptides (GenScript, purity >70%) were dissolved in DMSO, aliquoted, and stored at -80°C (maximum x5 freeze-thaw cycles). A1-K562 cells (split the day before) were resuspended at 1 x 106cells / mL, and 100 pL were aliquoted to each well of a 96-well U-bottom plate, supplemented with 100 pL fresh cRPML A1-K562 cells were pulsed with single concentrations of thawed a3a antigen peptides (<1% DMSO), or if conducting a peptide titration, the starting peptide concentration was prepared at the desired volume in a separate 96-well U-bottom plate, and titrated in thirds before transfer to the cells. The A1-K562 cells were peptide pulsed for 2 hours at 37°C, and negative controls (A1-K562 cells not pulsed with peptide) were prepared in the same way without peptide. Pulsed A1-K562 cells were then washed once to remove excess peptides (1300rpm, 3mins). a3a Jurkat cells (split the day before) were resuspended at 5x 105cells / mL and 200 pL added directly to each well of washed peptide-pulsed A1-K562 cells (i.e. a 1: 1 E / T ratio). The stimulation was performed at 37°C for 14-16 hours.

[0304] Following stimulation, the cells were stained with Zombie Violet viability dye, anti-CD69- APC (BioLegend; cat no. 310910) and anti-a TCR-PE (clone IP26, BioLegend; cat no. 306708) or anti-CD45-PE (BioLegend; cat no. 368510) on ice for 45 minutes, washed twice, and immediately analysed by an Attune flow cytometer (Thermo Fisher).

[0305] Engineering of a3a TCR variants

[0306] Eight CDR2a variants were designed (v3, v4, v5, v6, v7, v8, v9 and vlO, see Example 2). Each variant is based on the a3a TCR Va sequence (SEQ ID NO: 102), wherein the CDR2a portion of the a3a TCR Va (SEQ ID NO: 93) was replaced with one of the sequences SEQ ID NOs: 94 to 101. In this way a Va of each of v3, v4, v5, v6, v7, v8, v9 and vlO respectively was provided. The resulting Va gene fragments were synthesised with the a3a TCRa wild-type signal peptide by IDT. Cloning into pHR vectors was conducted as previously above: the a3a TCR Va fragments (8 variants as above and the a3 WT having a CDR2a portion SEQ ID NO: 92) were cloned into pHR vectors with the full-length, unmodified Ca (SEQ ID NO: 109), using Gibson Assembly.

[0307] Mapping of protein-protein interfaces with PyMOL and PDBePISA

[0308] The PDB reference of the MAG-IC3 ImmTAC (5BRZ) was inputted into PDBePISA for a description of the structural and chemical properties of the TCR / pMHC interface. In PyMOL, the distances between the identified bonds were calculated and annotated.

[0309] Preparation of TCR-Jurkat cell lines for co-culture assays

[0310] Identical protocol to that described above, but with 1.0 x 106CD8+ / - TCR- / - Jurkat cells cotransduced with 500 pL of the a3a TCRp lentivirus and 500 pL of the various a3a TCRa variant lentiviruses. TCR expression levels were confirmed by flow cytometry.

[0311] Expression of recombinant TCRs and pMHC single chain trimers a3a TCR variants: TCR Va fragments for all a3a variants were amplified from the pHR vectors by PCR using appropriate oligonucleotide primers; forward primer providing sufficient bp overlap with the destination pD649 plasmid, and the reverse primer providing sufficient bp overlap with the soluble TCR Ca fragment. All amplified TCR a3a Va fragments with appropriate overhangs were cloned into the pD649 vector (between Notl and Xbal restriction sites) alongside the Ca fragment in equimolar quantities using the Gibson Assembly as per the manufacturer’s (Thermo Fisher’s) instructions. Gibson Assembly is a DNA cloning method that allows multiple DNA fragments to be seamlessly joined in a single, isothermal reaction without the need for restriction sites. The process relies on the activity of three key enzymatic activities: 3ZExonuclease, DNA polymerase and a DNA Ligase in a single, one-pot reaction at a single temperature, making Gibson Assembly a versatile and efficient tool for cloning complex constructs. The pD649 a3a TCRa plasmids were co -transfected into Expi293T cells along with the shared pD649 a3a TCRp plasmid. pMHC single chain trimers (SCT): The DNA sequence encoding the human MHC-I was obtained from IPD-IMGT / HLA (Acc no. 142800), and the sequence encoding the P2M was obtained from UniProt (Acc no. P61769). The SCT construct consists of the same HA signal peptide sequence as the soluble TCRs, immediately followed by the peptide presented by the MHC-I (MAGE-A3 or Titin). The P2M domain is directly linked to the C-terminus of the peptide through a flexible (648)3 linker (LI), and to the MHC-I heavy chain via a comparatively longer (648)4 linker (L2). To accommodate the LI linker, the SCT features a Y84A mutation. As with the TCR constructs, the SCTs contain a C-terminal AviTag™ and 6xHis-tag for biotinylation and purification of the recombinant proteins, respectively. The recombinant SCT constructs have also been designed for cloning into the pD649 plasmid using Gibson Assembly, facilitating transfection into Expi293F cells, as per the manufacturer’s (Thermo Fisher’s) instructions.

[0312] Surface Plasmon Resonance

[0313] SPR analysis was conducted using a BIAcore S200 system (Cytiva), using HBS-P+(0.1 M HEPES, 1.5 M NaCl and 0.5% v / v Surfactant P20, pH 7.4 at lOx dilution) as a running buffer. Biotinylated MAGE -A3 -A*01:01 and Titin-A*01:01 SCT complexes were immobilised onto two flow cells of Series S Sensor Chip CAP (according to the instructions in the Biotin CAPture kit, Cytiva). Between 150 and 250 RUs were immobilised in separate experiments. An irrelevant protein was also immobilised at similar levels in one flow cell to act as a control surface. Equilibrium binding analyses of TCR / p-A*01:01 interactions were carried out at 25°C by measuring binding responses at equilibrium of serial injections (twofold dilutions) of soluble non-biotinylated, 3C-treated a3a TCR variants (a3 WT, a3a, v8 and v9), from high to low concentration, at a flow rate of 20 pl min1. Biacore software was used to conduct curve fittings and derive the equilibrium dissociation constant, AD. Experiments were repeated three times, varying the order of injected TCRs onto the chip.

[0314] Generation of primary CD 8+ T cell lines

[0315] Cloning: For maximal CD8+ T cell transduction efficiency, TCRp-P2A-TCRa lentiviral vector constructs were cloned and used to transfect HEK293T cells for lentivirus expression, as described above. Large lentivirus batches were made on separate occasions, harvested, and frozen at - 20°C for <1 month. On the day of transduction, lentivirus was thawed at 37°C and added directly to the cells as described in the next section (Day 2).

[0316] CD8+ T cell isolation, transduction, and expansion: All cell culture of human T cells was done using cRPMI (10% FBS, 1% glutamax, 1% sodium pyruvate, 1% HEPES, 1% penicillin / streptomycin) at 37°C, 5% CO2.

[0317] Day 0: T cells were isolated from whole blood from healthy donor leukocyte cones purchased from the NHS Blood and Transplantation service at the John Radcliffe Hospital. This project has been approved by the Medical Sciences Interdivisional Research Ethics Committee, with ethics approval reference: R93485 / RE001. CD8+ T cells were isolated from fresh leukocyte cones using RosetteSep Human CD8+ enrichment cocktail (STEMCELL) at 150 pL / ml. After a 20 min incubation at room temperature, blood cone samples were diluted 3-fold with PBS and layered onto Ficoll (STEMCELL Technologies) at a 0.8: 1.0 Ficolksample ratio, and spun at 1200g for 20 min at room temperature, acceleration 2, deceleration 0. Buffy coats were harvested with Pasteur pipettes, washed twice in cRPMI, counted, and the cells were rested for l-4h at 1.0 x 106 / mL in 24 well plates (1 mb per well) with 30 U / mL IL-2 (BioLegend). The cells were then treated with CD3 / CD28 Human T-activator dynabeads (Thermo Fisher) at a 1: 1 bead:cell ratio for 48h for proliferation and expansion (according to the manufacturer’s instructions). Day 2: after 48h, all cells were harvested, counted and resuspended at 0.5 x 106 / mL in cRMPI. 2 mL cell suspension was then added to each well of a 12 well plate (i.e. 1 x 106cells per well) and treated with 1.5 mL TCR lenti -virus. Each well was supplemented with 400 pL cRMPI with IL-2 at a concentration of 300 U / mL, for a final concentration of 30 U / mL (in 4 mL total volume). Two wells were also reserved for un-transduced controls.

[0318] Day 4: after 48h, the cells were expanded at a 1: 1 ratio with cRPMI and fresh IL-2 to a final concentration of 30 U / mL.

[0319] Day 6: after 48h, the dynabeads were magnetically removed, and the CD8+ T cells were suspended at 1.0 x 106 / mL in cRPMI with fresh IL-2 to a final concentration of 30 U / mL. The cells were left overnight and used for Incucyte cytotoxicity experiments the following morning. TCR expression was confirmed by flow cytometry staining with a MAGE-A3-A*01:01-PE-conjugated tetramer at 0.2 pM SA-PE.

[0320] Preparation of target cell lines for Incucyte cytotoxicity assays

[0321] A375 cells were grown in adherent T75 flasks in cDMEM (+10% PBS, 1% GlutaMax, 1% sodium pyruvate, and 1% penicillin / streptomycin (v / v)), and passaged every 3 days. A1-HEK293T cells were transduced with HLA-A*01:01 lentivirus (produced as previously described; example 1) by culturing 2.0 x 106HEK293T cells (at 1.0 x 106 / mL) with 2 mL HLA-A*01:01 lentivirus and 6 mL fresh cDMEM in a T25 flask for 48h. Once confluent, the cells were expanded into a T75 flask and passaged every 3 days with fresh cDMEM. On the afternoon of the day before the Incucyte experiment (see ‘CD8+ T cell isolation, transduction, and expansion’ day 6), the target cells were seeded in fresh cDMEM into flat bottom 96 well plates (in 100 pL / well) and rested overnight: (5000 A375 cells per well, or 3000 A1-HEK293T cells per well).

[0322] Incucyte cytotoxicity assays

[0323] To prepare the cytotoxicity assays, 50 pL of media was carefully removed per well, leaving 50 pL of target cells per well. To 5000 A375 cells, 50 pL of primary CD8+ T cell lines were subsequently added at the desired concentration to achieve a 1: 10 (1.0xl04 / mL), 1:5 (2.0xl04 / mL) and / or 1:2 (5.0xl04 / mL) E / T ratio. To 3000 A1-HEK293T cells, the cells were first pulsed with peptide at 50 pM for Ih prior to directly adding 50 pL of the primary CD8+ T cell lines at the desired concentration to achieve a 1: 10 (0.6xl04 / mL), 1:5 (1.2xl04 / mL) and / or 1:2 (3.0xl04 / mL) E / T ratio. All conditions were prepared in triplicate with three control wells included per experiment (target cells + / - peptide, without T cells). All outer wells were free of cells, instead containing 200 pL PBS per well.

[0324] Following addition of the primary CD8+ T cells to the target cells, the cells were placed in an incubator fitted with an Incucyte live cell imaging microscope (Sartorius). Starting 1 hour after incubator placement, four images were taken per well every hour for 48h-96h depending on the experiment. Target cell killing was subsequently assessed by comparing the confluency of the target cells in the control wells to the confluency of the target cells in the wells with primary CD8+ T cells. Analysis was conducted using the Sartorius Live Cell Analysis Software, computing the target cell confluency per condition by averaging all wells within a triplicate group.

[0325] Alloreactivity screens

[0326] The DNA sequences encoding the human MHC-I heavy chain alleles (A*01:01, A*02:01, A*24:02, A*03:01, B*07:02, B*08:01, C*04:01, C*07:01) and their respective endogenous signal peptides were obtained from IPD-IMGT / HLA, and the respective gene fragments were synthesised by IDT with appropriate bp overhangs for cloning into the pHR plasmid, as described above. Lentiviruses for all heavy chain alleles were produced and used to transduce K562 cells as described above. Co-culture assays with the primary CD8+ T cell lines were conducted identically to the coculture assays with TCR-Jurkat effector cells.

[0327] Developing primary CD8+ T cell lines

[0328] Primary CD8+ T cells were isolated from leukapheresis chambers (blood cones) from healthy donors and transduced with the TCRs a3 WT, a3a, v8, and v9. Additionally, two other engineered MAGE-A3 -reactive TCRs previously described in the literature with reduced Titinreactivity were included for comparison: TCRs 20a-18 and 94a-14 (Zhao, X. et al. Tuning T cell receptor sensitivity through catch bond engineering. Science 376, (2022). PMID: 35389803 PMCID: PMC9513562 DOI: 10.1126 / science.abl5282).

[0329] CD8+ T cells were isolated from blood cones and their purity confirmed by flow cytometry (data not shown) prior to resting for two days. In parallel, pMHC tetramers were used to confirm expression of the transduced TCRs. Antigen-specific tetramer staining was required because the endogenous TCRs were not knocked out of the primary CD8+ T cells; thus, staining with an anti-TCR antibody would not demonstrate successful transduction. Staining a3a TCR Jurkat cells with both MAGE-A3-A*01:01 and Titin -A*01:01 tetramers confirmed the utility and sensitivity of this technique (data not shown), as the observed ~ 10-fold difference in EC50s was reflective of the difference in affinities observed by SPR (Figure 16 and 17). Next, as a standard lentiviral quality control step prior to primary cell transduction, TCR- / - Jurkat cells were transduced with the identical TCR-encoding lentivirus batches prior to each primary CD8+ T cell transduction experiment, with representative flow cytometry staining shown in Figure 18.

[0330] Primary CD8+ T cells were subsequently transduced with the same lentivirus batches, and TCR expression was assessed by staining with a MAGE -A3 -A*01:01 tetramer (Figure 18 Bottom Panel). All transduced primary CD8+ T cell lines (a3 WT, a3a, v8, v9, 20a-18 and 94a-14 TCRs) demonstrated tetramer binding, whereas the SA-PE control tetramer (i.e. lacking MAGE-A3- A*01:01) did not. While there was a small degree of binding by the untransduced CD8+ T cells, tetramer staining of the transduced CD8+ cell lines correlated with the respective affinities of the TCRs for MAGE-A3. This confirmed a MAGE -A3 -A*01:01 specific effect, enabling progression to the cytotoxicity assays.

[0331] Example 2 - Identifying a3a agonists

[0332] A series of 18 9-mer HLA-A*01:01 peptides (Table 5) were tested for ability to activate a3a- bearing T cells via co-culture with a3a Jurkat T-cells. The 18 peptides were pulsed onto HLA- A*01:01 K562 cells at three different concentrations and cultured with the a3a-Jurkat cells. The majority of the 18 tested peptides were found to upregulate CD69 levels on the a3a-Jurkat cells (Figure 2), demonstrating that the majority of the selected peptides were a3a agonists.

[0333] Table 5.

[0334] For further validation, since the CD8 co-receptor increases the sensitivity of TCRs to pMHC and Jurkat cells do not naturally express CD8 (only CD4), a3a-Jurkat cells were transduced with the CD 8 co-receptor after cloning the CD8aP heterodimer into the pHRi lenti viral plasmid. The activation levels of the resulting CD8+ a3a-Jurkat cells were subsequently investigated in the presence of eight known and suspected a3a agonists (Table 6): the cancer-testis antigens MAGE-A3, MAGE-A6, MAGE-B18, Titin, the serine / threonine -protein kinase MRCKa, the activated leukocyte adhesion molecule CD 166, Ankyrin Repeat Domain 16 (ANKRD16) and FXYD domain -containing ion transport regulator 6 (FXYD6).

[0335] Table 6.

[0336] The CD69 levels were elevated by all antigens (Figure 3, although elevation with FXYD6 was not statistically significant). Activation to all tested MAGE -family antigens was expected given their high degree of sequence similarity. Fig 3 shows that the a3a TCR can target three different cancertestis antigens (MAGE-A3 / A6 / B 18), which can be for targeting a wide range of different cancers. However, a crucial disadvantage of the a3a TCR is the extent of cross-reactivity exhibited towards self-antigens expressed by healthy tissues, particularly: Titin, CD 166, STK MRCKa and, it was suspected, also FXYD6.

[0337] CD166-induced CD69 levels were particularly striking as they were similar to Titin -induced levels; a concerning observation because it is expressed in most healthy tissues. The serine / threonine - protein kinase MRCKa also induced comparable CD69 levels as Titin; another alarming observation as it is abundantly expressed in the heart, brain, skeletal muscle, kidney, and pancreas. Further, while ANKRD16 is found at high levels in male testes, it also has low tissue specificity. In other words, the broad expression profiles of CD 166, MRCKa and, ANKRD16 in healthy tissue are likely to have contributed to a3a-driven toxicity.

[0338] The agonistic effect of FXYD6 was found to not be statistically significant (at least by one-way ANOVA, significance was indeed observed with an unpaired t-Test). Nonetheless, given that the FXYD6 protein is an important regulator of ion homeostasis found in almost all human tissues, such widespread expression levels could still lead to significant toxicity. This is because the imperfect discriminatory powers of the TCR means that T cells can become activated by ultra-low affinity antigens if the expression of these antigens is high enough. A side-by-side comparison of these sequences illustrates the extensive and unpredictable crossreactivity of the a3a TCR. Excluding the anchor residues, CD 166 and Titin share 3 / 7 amino acids with MAGE-A3, and STK MRCKa, ANKRD16 and FXYD6 share only 2 / 7 amino acids with MAGE-A3. There is no predictable pattern or justification that can rationalise how all of these peptide sequences can bind to the same TCR.

[0339] It is known that additional human agonists exist: COG complex subunit 4 (COG4) required for normal Golgi morphology, Protocadherin Fat2 expressed in brain tissue and some tumours, and Inactive PLD 5 expressed in the retina and many other healthy tissues. Therefore, a further co-culture assay was performed with the 11 peptides listed in Table 7 (see Figures 4 and 5).

[0340] Table 7.

[0341] All 11 peptides of Table 7 resulted in CD8+ T cell stimulation (elevated CD69 expression in CD8+ a3a TCR-Jurkat) (Figure 4). The amino acid sequence diversity within the newly identified agonists protocadherin Fat2, inactive PLD 5, and COG complex subunit 4 further illuminates the challenge of predicting TCR agonists, and the remarkable degeneracy of TCR / pMHC interactions. Ignoring the obvious differences in the p9 anchor residue and the conserved DP motif, protocadherin Fat2 and inactive PLD 5 share only three total amino acids with MAGE-A3, and yet these peptides were the third and fourth most potent agonists, respectively, despite MAGE -A3 sharing four additional amino acids with MAGE-B18.

[0342] Taken together, all eight of these off-target antigens (Titin, CD 166, STK MRCKa, FXYD6, COG4, protocadherin Fat2 and inactive PLD5) have expression profiles in healthy tissues of a large variety of organs that could result in significant and destructive off-target toxicity if presented by MHC molecules and targeted by the a3a TCR (Figure 7). Therefore, while recognition of Titin resulted in rapid and extreme toxicity, it is likely that recognition of any of these other antigens would also cause serious and potentially fatal toxicity, particularly because the already highly potent response observed in Jurkat cells (Figure 5) would be significantly amplified in vivo.

[0343] Example 3 - Engineering TCRs with high anti-MAGE-A3 affinity and reduced cross-reactivity for self-antigens

[0344] Decoupling MAGE -A3 recognition from Titin recognition

[0345] It is well established that, unlike antibodies and cytokines, TCRs can mediate robust functional responses at relatively low affinities for peptide antigens ( 1 pM < K\ ) < 100 pM) (Davis, M. M. et al. Dynamics of cell surface molecules during T cell recognition. Annu Rev Biochem 72, 717-742 (2003). PMID: 14527326 DOI: 10.1146 / annurev.biochem.72.121801. 161625), a balance between functional responses to non-self and efficient peptide scanning. This favouring of TCRs with intermediate dissociation rates primarily occurs in the thymus where a progressive narrowing of the TCR repertoire occurs throughout thymocyte maturation, biasing the naive peripheral T cell repertoire towards TCR / pMHC interactions that fall within a moderate half-life “window” (Savage, P. A. & Davis, M. M. A kinetic window constricts the T cell receptor repertoire in the thymus. Immunity 14, 243-252 (2001). PMID: 11290334 DOI: 10. 1016 / sl074-7613(01)00106-6.

[0346] The present invention arises from the novel idea that the cross-reactivity of the a3a TCR arises from excessive affinity-maturation, such that the sequence of a3a was so far removed from that of its endogenous predecessor, the a3 WT TCR, that the benefits of thymic selection were lost. The inventors posited for the first time that, given that the dramatic affinity enhancement observed in a3a had been attained through mutations to the MHC -centric CDR2a loop, the a3a TCR had been overly stabilised for the MHC independently of the peptide.

[0347] Adopting a rational protein design approach for the identification of “intermediate” TCRs, the a3a CDR2a sequence “LVRPY” was reverted to the wild-type a3 CDR2a sequence “LIQSS” one or two or three amino acids at a time, generating eight novel TCRs (Figure 8). This was not a completely exhaustive list, primarily because the existing crystal structure of the a3a ImmTAC (Figure 9) indicated that the CDR2a valine interacted the least with the pMHC molecule (highlighted with an arrow), supporting a heavier focus on the latter three amino acids.

[0348] The TCRs tested herein therefore have the following sequences shown in Table 8, wherein the TCRs differ only by their alpha chain variable domain CDR2 sequence, as shown in Figures 8 or 12 or in Table 9.

[0349] Table 8.

[0350] Table 9.

[0351] Engineered TCRs maintain potency to MAGE -A3 but show reduced cross-reactivity

[0352] To initially explore whether these minor changes in the CDR2a amino acid sequence had altered the sensitivity of the TCRs to MAGE -A3 and / or Titin, a co-culture assay was set up per Example 1. Jurkat cell lines were generated expressing all 10 TCRs (the a3, a3a, and all eight intermediate variants; Figure 10) alongside the co-stimulatory molecule CD8 to maximise the sensitivity of the Jurkat cells to antigen.

[0353] When MAGE-A3 and Titin were pulsed onto HLA-A*01:01 K562 cells and co-cultured with the various TCR- Jurkat cell lines, the activation profdes of the intermediate TCR variants changed dramatically (Figure 11). Interestingly, most of the TCR variants maintained enhanced sensitivity to MAGE-A3 compared to the a3 WT, but lost sensitivity to Titin (v3, v5, v7, v8, v9). Two TCRs lost sensitivity to both antigens (v4 and v6), and one TCR exhibited enhanced sensitivity to Titin (vlO).

[0354] To understand how these marginally different CDR2a sequences altered the TCR / pMHC binding profdes, the sequences were matched to their respective activation profiles (from Figure 11), resulting in four broad groups of TCRs (Figure 12).

[0355] While it is difficult to form absolute conclusions, a few patterns can be inferred. It appears that an L RPY motif is required for sensitivity to both MAGE -A3 and Titin, independent of a V / I at p2, because sensitivity to Titin is effectively abolished with an RPS or RSY motif. Accordingly, binding molecules of the invention may comprise either He (I) or Vai (V) at p2 (‘NT of Formula I). A positively charged arginine (R) is required at p3 for potent sensitivity to MAGE -A3. This is best illustrated by the single amino acid substitution between the a3 WT’s LIQSS and v3’s LIRSS, where an alteration of only 6 atoms transformed a weak response into a highly potent response. Indeed, the crystal structure of the a3a ImmTAC MAG-IC3 shows that this CDR2a arginine contributes three salt bridges at distances of 2.31A, 3.51 and 1.11k, and one hydrogen bond at a distance of 2.31A, by interacting with the MHC al domain glutamic acid (E) (Figure 13); thus, this one amino acid substitution stabilises the interaction with the MHC considerably. Furthermore, the proline (P) and tyrosine (Y) of the RPY motif both have very high buried surface areas. The crystal structure suggests that the tyrosine is at a perfect angle to reach down and interact with a hydrophobic patch of the MHC al domain (Figure 13), likely stabilising the interaction even further. Moreover, given that the RPY motif facilitates robust reactivity to Titin, whereas an RSY motif does not, perhaps the small, kinked proline achieves a change in loop direction that places the tyrosine in a better position to form a stronger hydrophobic interaction with the MHC, facilitating a peptide -independent interaction. This would also explain how, in the absence of the arginine at p3, the PY motif is required for maintained sensitivity to MAGE -A3, such that all other CDR2a sequences with a Q at p3 that lack the PY motif show comparable sensitivity with the a3 WT. Finally, as predicted, the introduction of a V alone (LVQSS) had no effect on the sensitivity of the TCR to MAGE -A3 (v6), validating the decision to focus the amino acid substitutions on the latter three amino acids of the loop. Based on these observations, it could be hypothesised that the LIQPY motif of TCR v9 will be the most potent, but least cross-reactive TCR compared to the other TCRs that contain an arginine at p3, given the four strong bonds that it forms with the MHC.

[0356] Given these different peptide reactivity profiles explored through the lens of only two peptide antigens, the next step was to investigate how sensitivity had changed to the other agonists of the a3a TCR originally identified in Example 2 (Figure 3). The Jurkat cell lines expressing the TCRs that showed specificity to MAGE-A3 from Figure 13 were thus co-cultured with HLA-A*01:01 cells pulsed with these a3a agonists (Figure 14). Notably, the sensitivity to most of the a3a agonists had significantly dropped, especially for TCRs v8 and v9 which had importantly maintained sensitivity to MAGE -A3 and MAGE-A6 and additionally exhibited some weak sensitivity to MAGE-B18.

[0357] This was confirmed to be a TCR-dependent effect, as shown by the titration curves in Figure 15. It was incredible to see that these two TCRs (v8 and v9) that differ by only two amino acids from the a3 WT and the a3a, discovered through a non-exhaustive rational protein design strategy, had markedly improved specificity to three different cancer-testis antigens, potentially representing suitable therapeutic candidates.

[0358] Example 4 - TCRs v8 and v9 show negligible binding to Titin

[0359] Keen to pursue these TCRs further to understand how such minor sequence differences could translate to such dramatic alterations in specificity, the TCRs, A1-MAGE-A3 and Al -Titin complexes were all expressed recombinantly and tested for binding by SPR (Figures 16 and 17).

[0360] Using the a3 WT TCR and a3a TCR as controls, both TCRs bound to Titin and / or MAGE -A3 with affinities within the range of what has previously been reported in the literature (Figures 16a, b and 17a, b) (Vazquez-Lombardi, R. et al. High-throughput T cell receptor engineering by functional screening identifies candidates with enhanced potency and specificity. Immunity 55, 1953-1965. el 1 (2022). PMID: 36174557 DOI: 10.1016 / j.immuni.2022.09.004, validating the use of single-chain trimer formats of A1-MAGE-A3 and Al -Titin as SPR ligands. v8 and v9 affinities to MAGE -A3 remained much higher than that of the original a3 WT TCR (originally reported XD = 500 pM) (Cameron, B. J. et al. Identification of a titin-derived HLA-A1- presented peptide as a cross-reactive target for engineered MAGE A3 -directed T cells. Sci Transl Med 5, (2013). PMID: 23926201 PMCID: PMC6002776 DOI: 10.1126 / scitranslmed.3006034), and likely higher in general than thymically selected TCRs with affinities for over-expressed self-antigens such as MAGE-A3 (v8 / MAGE-A3 KD= 13.58 pM, v9 / MAGE-A3 KD= 23.87 pM). Furthermore, whilst the a3a TCR bound with relatively high affinity to Titin (AD = 32.36 pM) TCRs v8 and v9 (and the a3 WT) did not bind to Titin.

[0361] Example 5 - TCRs v8 and v9 efficiently recognise and eliminate melanoma cancer cells ex vivo The functional responses of primary CD8+ T cells transduced with the a3a, v8, v9, and additional control TCRs were determined using cytotoxic functional assays. Cytotoxicity is a sensitive measure of TCR potency and a direct readout for CD8+ T cell effector functions.

[0362] Using an Incucyte live microscope, cytotoxicity assays were conducted, initially targeting tumour cell lines endogenously expressing peptide antigens. The A375 cell line is a human melanoma cell line with endogenous expression of MAGE -A3 that was used in the pre- and post-trial investigations of the a3a clinical trial (Cameron, B. J. et al. Identification of a titin -derived HLA-A1- presented peptide as a cross-reactive target for engineered MAGE A3 -directed T cells. Sci Transl Med 5, (2013). PMID: 23926201 PMCID: PMC6002776 DOI: 10.1126 / scitranslmed.3006034). Therefore, an initial assessment of the potency of TCRs v8 and v9 compared to the a3a and a3 WT was conducted by measuring the confluency of A375 cells every hour over 48h, such that a reduction in the confluence of A375 cells cultured with the primary CD8+ T cell lines compared to the A375 cells alone indicated CD8+ T cell-mediated cytotoxicity. A representative experiment is shown in Figure 19. At a 1:5 E / T ratio, no cytotoxicity was observed by the untransduced CD8+ T cells, unlike for TCRs v8 and v9, where cytotoxic activity was comparable to that elicited by the a3a, in contrast to the lower activation levels seen in Figures 10 to 13.

[0363] To corroborate these initial observations, primary CD8+ T cells from four donors were isolated and transduced on separate occasions and tested for cytotoxic responses across different E / T ratios over 48h (Figure 20). Across all donors, significant cytotoxicity was elicited by v8 and v9 comparable to that elicited by the a3a, and this held true even when there were lOx as many tumour cells as T cells (Figure 10c). Interestingly, despite the lower affinity of v9 for MAGE -A3 compared to v8 (Figure 16c, d), v9 elicited more potent cytotoxic activity than v8 at the lowest E / T ratio (1: 10). The comparable cytotoxicity to the a3a was also a surprise for this same reason, perhaps suggesting an imperfectly linear relationship between affinity and functional capabilities, particularly at higher affinities where T cell dysfunction has been observed. A small but insignificant degree of cytotoxic activity by the a3 WT was visible at the highest E / T ratio, which is in agreement with the very low affinity of this TCR for MAGE-A3.

[0364] Example 6 - TCR v9 displays reduced cross-reactivity with no loss of cytotoxic activity against MAGE-A3

[0365] As a proxy for cell lines expressing Titin and CD 166 peptide antigens, Lenti-X cells were transduced with HLA-A*01:01 (Al -Lenti-X) and the resulting Al -Lenti-X cells pulsed with a physiological concentration of Titin and CD 166 peptides. To validate this setup, the MAGE-A3 peptide was also pulsed onto Al -Lenti-X cells, allowing cytotoxicity under these more artificial circumstances to be compared to the A375 melanoma cell line. All transduced CD8+ T cell lines displayed cytotoxic activity, whereas the untransduced CD8+ T cells did not, validating both the assay and transduction protocol (Figure 21). Interestingly, all engineered TCRs elicited similar levels of cytotoxicity up to 48h, but v9 showed superior cytotoxic control up to 96h. To next assess the cytotoxic potential of these cell lines to Titin and CD 166, two different E / T ratios were trialled (Figure 22) (top panel 1:2, bottom panel 1:5), and these results are reflective of two technical repeats with CD8+ T cells isolated and transduced from different donors on both occasions.

[0366] First, for Al -Lenti-X cells pulsed with MAGE -A3 (Figure 22a) all transduced primary CD8+ T cell lines displayed significant cytotoxic capabilities at both E / T ratios, including the a3 WT, with v8 and v9 showing comparable cytotoxicity to the a3a as previously observed when cultured with the A375 melanoma cell line (Figures 19 and 20). TCR 94a-14 appeared to be the most potent; an effect most obvious at a lower E / T ratio of 1:5. Moreover, given that no cytotoxic response was observed by the a3 WT cell line for the A375 cells at any ratio, this implied that the density of MAGE -A3 antigen expressed by the A375 cells was lower than when the peptide was pulsed onto the Al-Lenti-X cells (at 50 pM).

[0367] Second, for Al-Lenti-X cells pulsed with Titin (Figure 22b), as expected, no cytotoxicity was observed by the a3 WT, whereas significant cytotoxicity was observed by the a3a at both E / T ratios. 20a- 18 and 94a- 14 displayed significant cytotoxicity at both E / T ratios, despite the published data pointing to the contrary (Zhao, X. et al. Tuning T cell receptor sensitivity through catch bond engineering. Science 376, (2022). PMID: 35389803 PMCID: PMC9513562 DOE 10.1126 / science.abl5282), albeit at a lower level than that induced by the a3a TCR. Notably, no cytotoxic activity was displayed by v9 at either E / T ratio.

[0368] Finally, for Al-Lenti-X cells pulsed with CD 166 (Figure 22c) no cytotoxic activity was elicited by the a3 WT, whereas significant cytotoxicity was elicited by the a3a at both E / T ratios. No cytotoxic response was elicited by v9 at either E / T ratio.

[0369] Based on these results, it appeared that v9 was as potent as the a3a in targeting the MAGE -A3 peptide, but most importantly, it did not target Titin or CD166. Therefore, v9 represents a potential therapeutic candidate for MAGE-A3- (and -A6-) expressing tumours.

[0370] Example 7 - TCRs v8 and v9 do not show alloreactivity ex vivo

[0371] Humans are often MHC heterozygous, carrying two different alleles at the HLA-A / B / C loci such that each person can theoretically express six different HLA class-I molecules. Though many TCRs are HLA-restricted, many have also been reported to be alloreactive, responding to MHC molecules from a foreign haplotype; a phenomenon at the core of transplant rejection. Therefore, it was important to assess whether v9 displayed alloreactivity to other HLA class-I molecules.

[0372] The alloreactivity of the a3a TCR was assessed in pre-clinical testing by culturing a3a TCR primary CD8+ T cells with a panel of cell lines covering more than 95% of the HLA serotype population, concluding no signs of alloreactivity. Due to lack of access to such a large panel of cell lines, K562 cells were instead transduced with eight of the most common HLA class-I alleles (Pearlman, A. H. et al. Targeting public neoantigens for cancer immunotherapy. Nature Cancer 2:5 2, 487-497 (2021). PMID: 34676374 PMCID: PMC8525885 DOI: 10.1038 / s43018-021-00210-y), covering a significant proportion of all ethnicities across the globe.

[0373] The transduced primary CD8+ T cells were subsequently incubated with the HLA-A / B / C transduced K562 cell lines, and alloreactivity was assessed by monitoring CD8+ T cell CD69 upregulation (Figure 23). Only the a3a CD8+ T cells showed a (minimal) response to HLA-A*01:01, likely due to the high affinity of the a3a TCR for the MHC molecule and / or due to reactivity to an agonist peptide processed or presented by the K562 cells. Importantly, no obvious alloreactivity was evident by v8 or v9-CD8+ T cells, suggesting that the risk of these TCRs responding to peptides presented by other HLA haplotypes was minimal.

[0374] Example 8 - The cross-reactivity of engineered TCRs is dependent on the cross-reactivity of the wild-type TCR

[0375] The sensitivity of the a3 WT TCR for a panel of a3a agonists was investigated, to compare the binding profde of TCRs of the invention against the WT TCR. The a3 WT, v8, v9 and the a3a TCRs were tested for antigen sensitivity in co-culture assays as previously described (Figure 24). All a3a agonists were pulsed onto HLA -A*01:01 K562 cells, including one additional peptide, Dynactin subunit 4 (Dyn4), which is a further suspected agonist of a3a TCR that showed insignificant activation of v9.

[0376] With a3a, a significant difference (p<0.0001) in CD69 surface-expression (as a % of APCs) was observed between control (non-agonist peptide) each of MAGE -A3, MAGE-A6, hFat2, PLD 5, MAGE-B18, CD 166, S / MRCKa, Titin and ANKRD16, i.e. a3a displays meaningful sensitivity to each of MAGE-A3, MAGE-A6, hFat2, PLD 5, MAGE-B18, CD166, S / MRCKa, Titin and ANKRD16. With v8, no significant difference in CD69 surface-expression was observed between control and any of CD 166, S / MRCKa, Titin and FXYD6, i.e. v8 displays no meaningful sensitivity to any of CD166, S / MRCKa, Titin and FXYD6. With v9, no significant difference in CD69 surfaceexpression was observed between control and any of CD 166, S / MRCKa, Titin, ANKRD16, Dyn4, COG4 and FXYD6, i.e. v9 displays no meaningful sensitivity to any of CD 166, S / MRCKa, Titin, ANKRD16, Dyn4, COG4 and FXYD6.

[0377] No v9 agonists outside of the a3a TCR’s binding landscape have been identified. This observation suggests that v9 has a similar pMHC docking profile as the a3a, despite its modified CDR2a loop. Therefore, it appears that the two CDR2a mutations within v9 (a3a LVRPY -> v9 LIQPY) were sufficient to abolish sensitivity to almost all off-target a3a.

[0378] The v8 and v9 TCRs exhibit far greater discrimination for cancer testis antigens (particularly MAGE-A3 and MAGE-A6) than the a3a TCR. v8 and v9 also display higher affinity for said cancertestis antigens MAGE-A3 and also MAGE-A6 than the a3 WT TCR.

[0379] While some cross-reactivity still remains in the v8 and the v9 TCRs, the level of crossreactivity observed is comparable to the cross-reactivity seen in the a3 WT TCR. In particular, the v9 TCR exhibited the same (or even less) cross-reactivity than the a3 WT TCR (as well as advantageously exhibiting higher affinity for MAGE -A3 than the a3 WT TCR).

[0380] In other words, v9 was much more specific than the a3a TCR (i.e. far less cross-reactive than the a3a TCR) and was also just as cross-reactive as the a3 WT TCR (Figure 24d), implying that it would be difficult to re-engineer a more specific TCR using the a3 WT TCR as a scaffold than the TCRs of the present invention. Therefore, v9 represents the most optimal affinity -enhanced version of the a3 WT TCR generated to date.

[0381] Example 9 - Comparison of TCR v9 and a3a TCR structures

[0382] To define how reversing mutations in the CDR2a loop alter recognition, the inventors solved the crystal structures of the v9 and a3a TCRs bound to MAGE-A3-HLA-A*01:01 at 2.4 A and 2.8 A resolution, respectively. The two complexes were highly similar overall (r.m.s.d. 0.26 A) and exhibited nearly identical docking (56.9° for v9; 55.9° for a3a) and incident angles (13.5° and 13.6°), with comparable interface chemistries (Fig. 25A, Table 10). Thus, functional differences between a3a and v9 arise not from altered global orientation but from local interactions within the CDR2a-MHC interface.

[0383] In the a3a structure, the engineered VRPY motif makes extensive contacts with the MHC a2- helix. Arg57 projects deeply into the MHC surface, forming two hydrogen bonds, four salt bridges, and three polar contacts with Hisl51 and Glul54 (Fig. 25A-D, Table 10). Pro58 and Tyr59 contribute additional van der Waals and polar interactions along the a2 -helix. These contacts, likely absent in the wild-type a3 TCR containing the IQSS sequence, explain the dramatic affinity enhancement of a3a relative to a3 WT (KD 2.3 pM versus 500 pM). Comparison with the affinity -enhanced MAG-IC3 TCR, which also incorporates a VRPY loop, confirms that this motif robustly amplifies MHC -centric stabilization across a3-derived TCRs. The ‘hotspot’ of the interactions formed between the TCR with the HLA-A*01:01 is denoted by an arrow in Fig. 25E. The number of interactions between the MAGE-A3 targeting TCRs and Hisl51, E154, and Argl57 of HLA-A*01:01 is significantly lower for v9 compared to those formed by A3A and MAG-IC3 TCRs (Fig. 25D, E).

[0384] Replacing the VRPY motif in the a3a TCR with the IQPY motif in the v9 TCR removes Arg57- dependent contacts, resulting in the loss of three hydrogen bonds and four salt bridges while retaining the stabilizing contributions of Pro58 and Tyr59 (Fig. 25D).

[0385] Collectively, these structural data show that the VRPY motif shifts the energetic balance of recognition toward MHC engagement and away from peptide -focussed binding, providing a mechanistic basis for the heightened cross-reactivity observed in a3a and MAG-IC3 (Fig. 25E, F). In contrast, the v9 IQPY sequence partially restores peptide -focused binding, recovering a peptide contribution of -27% versus 20% in the a3a, while preserving moderate binding affinity through the Pro58 and Tyr59 residues (Fig. 25F, G). This redistribution of binding energy underlies the regained selectivity of v9.

[0386]

[0387]

[0388] Table 10.

[0389] Aspects

[0390] 1. A binding molecule comprising:

[0391] (A) a T cell receptor (TCR) alpha chain variable domain comprising:

[0392] (i) a complementary determining region (CDR) 1 comprising an amino acid sequence DSAIYN (SEQ ID NO: 1) or comprising an amino acid sequence differing from SEQ ID NO: 1 by one amino acid,

[0393] (ii) a CDR2 comprising the amino acid sequence of Formula I (SEQ ID NO: 2), and

[0394] (iii) a CDR3 comprising the amino acid sequence CAVRPGGAGSYQLTF (SEQ ID NO: 3) or comprising an amino acid sequence differing from SEQ ID NO: 3 by one amino acid; and

[0395] (B) a TCR beta chain variable domain comprising:

[0396] (i) a CDR1 comprising the amino acid sequence SGHRS (SEQ ID NO: 4) or comprising an amino acid sequence differing from SEQ ID NO: 4 by one amino acid,

[0397] (ii) a CDR2 comprising the amino acid sequence YFSETQ (SEQ ID NO: 5) or comprising an amino acid sequence differing from SEQ ID NO: 5 by one amino acid, and

[0398] (iii) a CDR3 comprising the amino acid sequence CASSPNMADEQYF (SEQ ID NO: 6) or comprising an amino acid sequence differing from SEQ ID NO: 6 by one amino acid; wherein Formula I (SEQ ID NO: 2) is:

[0399] Leu-Ni-N2-N3-N4wherein

[0400] N1 is He or Vai,

[0401] N2 is Gin or Arg,

[0402] N3 is Pro or Ser,

[0403] N4 is Tyr or Ser, and wherein: when N2 is Gin, N4 is Tyr; and when N2 is Arg, N3-N4 is Pro-Ser, Ser-Ser or Ser-Tyr.

[0404] 2. The binding molecule of aspect 1, wherein when N2 is Arg, N4 is Ser.

[0405] 3. The binding molecule of aspect 1 or 2, wherein when N2 is Arg, N3-N4 is Ser-Ser.

[0406] 4. The binding molecule of aspect 1, wherein N2 is Arg.

[0407] 5. The binding molecule of aspect 1, 2 or 4, wherein N3 is Pro. 6. The binding molecule of any one of the preceding aspects, wherein N1 is He.

[0408] 7. The binding molecule of any one of the preceding aspects, wherein N2-N3-N4 is Gln-Pro- Tyr.

[0409] 8. The binding molecule of any one of the preceding aspects, wherein N1-N2-N3-N4 is Ile-Gln- Pro-Tyr (SEQ ID NO: 76).

[0410] 9. The binding molecule of any one of the preceding aspects, wherein:

[0411] (A) (i) the T cell receptor (TCR) alpha chain variable domain CDR1 comprises the amino acid sequence of sequence SEQ ID NO: 1, and

[0412] (ii) the T cell receptor (TCR) alpha chain variable domain CDR3 comprises the amino acid sequence of sequence SEQ ID NO: 3; and

[0413] (B) (i) the TCR beta chain variable domain CDR1 comprises the amino acid sequence of sequence SEQ ID NO: 4,

[0414] (ii) the TCR beta chain variable domain CDR2 comprises the amino acid sequence of sequence SEQ ID NO: 5, and

[0415] (iii) the TCR beta chain variable domain CDR3 comprises the amino acid sequence of sequence SEQ ID NO: 6.

[0416] 10. The binding molecule of any one of the preceding aspects, wherein the alpha chain variable domain comprises an amino acid sequence of any one of SEQ ID NOs: 23-32, or a sequence at least 80% identical to the amino acid sequence any one of SEQ ID NOs: 23-32; and / or wherein the beta chain variable domain comprises an amino acid sequence SEQ ID NO: 33, or a sequence at least 80% identical to the amino acid sequence SEQ ID NO: 33.

[0417] 11. The binding molecule of any one of the preceding aspects, wherein the alpha chain variable domain comprises an amino acid sequence of SEQ ID NO: 27, or a sequence at least 80% identical to the amino acid sequence of SEQ ID NO: 27.

[0418] 12. The binding molecule of any one of the preceding aspects, wherein the alpha chain variable domain comprises an amino acid sequence of any one of SEQ ID NOs: 23-32, or a sequence at least 95% identical to the amino acid sequence of any one of SEQ ID NOs: 23-32.

[0419] 13. The binding molecule of any one of the preceding aspects, wherein the alpha chain variable domain comprises an amino acid sequence of any one of SEQ ID NOs: 23-32, or a sequence at least 98% identical to the amino acid sequence of any one of SEQ ID NOs: 23-32. 14. The binding molecule of any one of the preceding aspects, wherein the alpha chain variable domain comprises an amino acid sequence of any one of SEQ ID NOs: 23-32, or a sequence at least 99% identical to the amino acid sequence of any one of SEQ ID NOs: 23-32.

[0420] 15. The binding molecule of any one of the preceding aspects, wherein the alpha chain variable domain comprises an amino acid sequence of SEQ ID NO: 27, or a sequence at least 95% identical to the amino acid sequence of SEQ ID NO: 27.

[0421] 16. The binding molecule of any one of the preceding aspects, wherein the alpha chain variable domain comprises an amino acid sequence of SEQ ID NO: 27, or a sequence at least 98% identical to the amino acid sequence of SEQ ID NO: 27.

[0422] 17. The binding molecule of any one of the preceding aspects, wherein the alpha chain variable domain comprises an amino acid sequence of SEQ ID NO: 27, or a sequence at least 99% identical to the amino acid sequence of SEQ ID NO: 27.

[0423] 18. The binding molecule of any one of the preceding aspects, wherein the beta chain variable domain comprises an amino acid sequence of SEQ ID NO: 33, or a sequence at least 95% identical to the amino acid sequence SEQ ID NO: 33.

[0424] 19. The binding molecule of any one of the preceding aspects, wherein the beta chain variable domain comprises an amino acid sequence of SEQ ID NO: 33, or a sequence at least 98% identical to the amino acid sequence SEQ ID NO: 33.

[0425] 20. The binding molecule of any one of the preceding aspects, wherein the beta chain variable domain comprises an amino acid sequence of SEQ ID NO: 33, or a sequence at least 99% identical to the amino acid sequence SEQ ID NO: 33.

[0426] 21. The binding molecule of any one of the preceding aspects, wherein the binding molecule comprises the ectodomain of a T cell receptor (TCR).

[0427] 22. The binding molecule of any one of the preceding aspects, wherein the binding molecule is a receptor.

[0428] 23. The binding molecule of any one of the preceding aspects, wherein the binding molecule is a TCR. 24. The binding molecule of any one of the preceding aspects, wherein the binding molecule comprises an alpha chain comprising an amino acid sequence of any one of SEQ ID NOs: 78, 80 or 82 to 84, or a sequence at least 80% identical to the amino acid sequence any one of SEQ ID NOs: 78, 80 or 82 to 84; and / or wherein the beta chain variable domain comprises an amino acid sequence SEQ ID NO: 40, or a sequence at least 80% identical to the amino acid sequence SEQ ID NO: 40.

[0429] 25. The binding molecule of aspect 24, wherein the alpha chain comprises an amino acid sequence of SEQ ID NO: 84, or a sequence at least 80% identical to the amino acid sequence of SEQ ID NO: 84.

[0430] 26. The binding molecule of aspect 24, wherein the alpha chain comprises an amino acid sequence of any one of SEQ ID NOs: 78, 80 or 82 to 84, or a sequence at least 95% identical to the amino acid sequence of any one of SEQ ID NOs: 78, 80 or 82 to 84.

[0431] 27. The binding molecule of aspect 24 or 26, wherein the alpha chain comprises an amino acid sequence of any one of SEQ ID NOs: 78, 80 or 82 to 84, or a sequence at least 98% identical to the amino acid sequence of any one of SEQ ID NOs: 78, 80 or 82 to 84.

[0432] 28. The binding molecule of any one of aspects 24, 26 or 27, wherein the alpha chain comprises an amino acid sequence of any one of SEQ ID NOs: 78, 80 or 82 to 84, or a sequence at least 99% identical to the amino acid sequence of any one of SEQ ID NOs: 78, 80 or 82 to 84.

[0433] 29. The binding molecule of any one of aspects 24 to 28, wherein the alpha chain comprises an amino acid sequence of SEQ ID NO: 84, or a sequence at least 95% identical to the amino acid sequence of SEQ ID NO: 84.

[0434] 30. The binding molecule of any one of aspects 24 to 29, wherein the alpha chain comprises an amino acid sequence of SEQ ID NO: 84, or a sequence at least 98% identical to the amino acid sequence of SEQ ID NO: 84.

[0435] 31. The binding molecule of any one of aspects 24 to 30, wherein the alpha chain comprises an amino acid sequence of SEQ ID NO: 84, or a sequence at least 99% identical to the amino acid sequence of SEQ ID NO: 84. 32. The binding molecule of any one of aspects 24 to 31, wherein the beta chain comprises an amino acid sequence SEQ ID NO: 40, or a sequence at least 95% identical to the amino acid sequence SEQ ID NO: 40.

[0436] 33. The binding molecule of any one of aspects 24 to 32, wherein the beta chain comprises an amino acid sequence SEQ ID NO: 40, or a sequence at least 98% identical to the amino acid sequence SEQ ID NO: 40.

[0437] 34. The binding molecule of any one of aspects 24 to 33, wherein the beta chain comprises an amino acid sequence SEQ ID NO: 40, or a sequence at least 99% identical to the amino acid sequence SEQ ID NO: 40.

[0438] 35. The binding molecule of any one of the preceding aspects, wherein the binding molecule binds specifically to a melanoma-associated antigen 3 (MAGE-3) peptide antigen.

[0439] 36. The binding molecule of aspect 35, wherein the MAGE-3 peptide antigen is a HLA-A1 presented MAGE-3 peptide antigen.

[0440] 37. The binding molecule of aspect 35 or 36, wherein the MAGE-3 peptide antigen comprises or consists of the amino acid sequence of SEQ ID NO: 45.

[0441] 38. The binding molecule of any one of aspects 35 to 37, wherein the binding molecule binds to to a MAGE-A3 peptide antigen with a KD of less than 50 pM.

[0442] 39. The binding molecule of any one of aspects 25 to 38, wherein the binding molecule binds to a MAGE-A3 peptide antigen with a KD of less than 25 pM.

[0443] 40. The binding molecule of any one of the preceding aspects, wherein the binding molecule exhibits reduced binding to a Titin peptide relative to:

[0444] (a) a first comparator molecule, wherein the first comparator molecule is a TCR comprising an alpha chain comprising an amino acid sequence of SEQ ID NO: 36 and a beta chain comprising an amino acid sequence of SEQ ID NO: 40; and / or

[0445] (b) a second comparator molecule, wherein the second comparator molecular is a TCR comprising an alpha chain comprising an amino acid sequence of SEQ ID NO: 38 and a beta chain comprising an amino acid sequence of SEQ ID NO: 41; and / or (c) a third comparator molecule, wherein the third comparator molecule is a TCR comprising an alpha chain comprising an amino acid sequence of SEQ ID NO: 39 and a beta chain comprising an amino acid sequence of SEQ ID NO: 42.

[0446] 41. The binding molecule of aspect 40, wherein the Titin peptide is a HLA-A1 presented Titin peptide.

[0447] 42. The binding molecule of aspect 40 or 41, wherein the reduced binding is an at least 20% reduction in binding affinity when measured by surface plasmon resonance (SPR) and / or when measured by bio-layer interferometry (BLI), preferably when measured by SPR.

[0448] 43. The binding molecule of any one of aspects 40 to 42, wherein the Titin peptide comprises or consists of an amino acid sequence of SEQ ID NO: 46.

[0449] 44. The binding molecule of any one of the preceding aspects, wherein the binding molecule does not exhibit meaningful sensitivity to one or more of the peptide antigens listed in Table 3, optionally wherein the binding molecule does not exhibit meaningful sensitivity to any of a peptide antigen having an amino acid sequence of SEQ ID NOs: 46, 49, 50 or 52.

[0450] 45. A binding molecule comprising:

[0451] (A) a T cell receptor (TCR) alpha chain variable domain comprising:

[0452] (i) a complementary determining region (CDR) 1 comprising an amino acid sequence DSAIYN (SEQ ID NO: 1) or comprising an amino acid sequence differing from SEQ ID NO: 1 by one amino acid,

[0453] (ii) a CDR2 comprising the amino acid sequence of Formula I (SEQ ID NO: 2), and

[0454] (iii) a CDR3 comprising the amino acid sequence CAVRPGGAGSYQLTF (SEQ ID NO: 3) or comprising an amino acid sequence differing from SEQ ID NO: 3 by one amino acid; and

[0455] (B) a TCR beta chain variable domain comprising:

[0456] (i) a CDR1 comprising the amino acid sequence SGHRS (SEQ ID NO: 4) or comprising an amino acid sequence differing from SEQ ID NO: 4 by one amino acid,

[0457] (ii) a CDR2 comprising the amino acid sequence YFSETQ (SEQ ID NO: 5) or comprising an amino acid sequence differing from SEQ ID NO: 5 by one amino acid, and

[0458] (iii) a CDR3 comprising the amino acid sequence CASSPNMADEQYF (SEQ ID NO: 6) or comprising an amino acid sequence differing from SEQ ID NO: 6 by one amino acid; wherein Formula I (SEQ ID NO: 2) is Leu-Nl-N2-N3-N4, the binding molecule binds to a complex of an epitope of SEQ ID NO: 45 (MAGE -A3) and an HLA molecule, and wherein the binding molecule forms fewer bonds to the HLA molecule and more bonds to the epitope of SEQ ID NO: 45 compared to a TCR comprising an alpha chain comprising an amino acid sequence of SEQ ID NO: 36 and a beta chain comprising an amino acid sequence of SEQ ID NO: 40.

[0459] 46. The binding molecule of aspect 45, wherein the binding molecule forms fewer hydrogen bonds and / or salt bridges compared to a TCR comprising an alpha chain comprising an amino acid sequence of SEQ ID NO: 36 and a beta chain comprising an amino acid sequence of SEQ ID NO: 40.

[0460] 47. The binding molecule of aspect 45 or 46, wherein:

[0461] N1 is He or Vai,

[0462] N2 is Gin or Arg,

[0463] N3 is Pro or Ser,

[0464] N4 is Tyr or Ser, and wherein: when N2 is Gin, N4 is Tyr; and when N2 is Arg, N3-N4 is Pro-Ser, Ser-Ser or Ser-Tyr.

[0465] 48. The binding molecule of any one of aspects 45 to 47, wherein the binding molecule is further characterized by one or more features defined in any of aspects 2 to 44.

[0466] 49. A polynucleotide comprising a sequence encoding an alpha chain variable domain as set forth in any one of aspects 1-48.

[0467] 50. A polynucleotide comprising a sequence encoding the binding molecule of any one of aspects 1-48.

[0468] 51. A pair of polynucleotides comprising a first polynucleotide comprising a sequence encoding an alpha chain variable domain as set forth in any one of aspects 1-41 and a second polynucleotide comprising a sequence encoding a beta chain variable domain as set forth in any one of aspects 1-48.

[0469] 52. A vector comprising the polynucleotide of aspect 49 or 50.

[0470] 53. A cell comprising the binding molecule of any one of aspects 1-48, the polynucleotide of and aspect 49 or 50, the pair of polynucleotides of aspect 51 and / or the vector of aspect 52, optionally wherein the cell is an isolated cell.

[0471] 54. The cell of aspect 53, wherein the cell is a T lymphocyte. 55. A pharmaceutical composition comprising the binding molecule of any one of aspects 1-48, the polynucleotide of aspect 49 or 50, the pair of polynucleotides of aspect 51, the vector of aspect 52, the cell of aspect 53 or 54 and / or a plurality of the cells of aspect 53 or 54, and a pharmaceutically acceptable carrier.

[0472] 56. A binding molecule of any one of aspects 1-48, the polynucleotide of aspect 49 or 50, the pair of polynucleotides of aspect 51, the vector of aspect 52, the cell of aspect 53 or 54, a plurality of the cells of aspect 53 or 54, or the pharmaceutical composition according to aspect 55, for use in a method of treatment.

[0473] 57. A binding molecule of any one of aspects 1-48, the polynucleotide of aspect 49 or 50, the pair of polynucleotides of aspect 51, the vector of aspect 52, the cell of aspect 53 or 54, a plurality of the cells of aspect 53 or 54, or the pharmaceutical composition according to aspect 55, for use in a method of treating a subject having a MAGE-A3 -associated disease or disorder, wherein the method comprises administering the binding molecule, polynucleotide, pharmaceutical composition, vector, cell or cells to the subject.

[0474] 58. The binding molecule, polynucleotide, vector, pharmaceutical composition, cell or cells for use according to aspect 57, wherein the MAGE-A3 -associated disease or disorder is a MAGE -A3 - associated cancer, optionally wherein the MAGE -A3 -associated cancer is a solid tumour cancer.

[0475] 59. A method of treating a subject comprising administering to the subject a binding molecule of any one of aspects 1-48, the polynucleotide of aspect 49 or 50, the pair of polynucleotides of aspect 51, the vector of aspect 52, the cell of aspect 53 or 54, a plurality of the cells of aspect 53 or 54, or the pharmaceutical composition according to aspect 55.

[0476] 60. A method of treating a MAGE-A3 -associated disease or disorder in a subject, wherein the method comprises administering to the subject a binding molecule any one of aspects 1-48, the polynucleotide of aspect 49 or 50, the pair of polynucleotides of aspect 51, the vector of aspect 52, the cell of aspect 53 or 54, a plurality of the cells of aspect 53 or 54, or the pharmaceutical composition according to aspect 55.

[0477] 61. The method of aspect 60, wherein the MAGE-A3 -associated disease or disorder is a MAGE-A3 -associated cancer, optionally wherein the MAGE-A3 -associated cancer is a solid tumour cancer. 62. Use of a binding molecule any one of aspects 1-48, the polynucleotide of aspect 49 or 50, the pair of polynucleotides of aspect 51, the vector of aspect 52, the cell of aspect 53 or 54, a plurality of the cells of aspect 53 or 54, or the pharmaceutical composition according to aspect 55 for the manufacture of a medicament.

[0478] 63. Use of a binding molecule of any one of aspects 1-48, the polynucleotide of aspect 49 or 50, the pair of polynucleotides of aspect 51, the vector of aspect 52, the cell of aspect 53 or 54, a plurality of the cells of aspect 53 or 54, or the pharmaceutical composition according to aspect 55 for the manufacture of a medicament for a MAGE -A3 -associated disease or disorder.

[0479] 64. The use of aspect 63, wherein the MAGE-A3 -associated disease or disorder is a MAGE -A3 - associated cancer, optionally wherein the MAGE -A3 -associated cancer is a solid tumour cancer.

[0480] Sequence Listing

[0481]

[0482]

[0483]

[0484]

Claims

CLAIMS1. A binding molecule comprising:(A) a T cell receptor (TCR) alpha chain variable domain comprising:(i) a complementary determining region (CDR) 1 comprising an amino acid sequence DSAIYN (SEQ ID NO: 1) or comprising an amino acid sequence differing from SEQ ID NO: 1 by one amino acid,(ii) a CDR2 comprising the amino acid sequence of Formula I (SEQ ID NO: 2), and(iii) a CDR3 comprising the amino acid sequence CAVRPGGAGSYQLTF (SEQ ID NO:3) or comprising an amino acid sequence differing from SEQ ID NO: 3 by one amino acid; and(B) a TCR beta chain variable domain comprising:(i) a CDR1 comprising the amino acid sequence SGHRS (SEQ ID NO: 4) or comprising an amino acid sequence differing from SEQ ID NO: 4 by one amino acid,(ii) a CDR2 comprising the amino acid sequence YFSETQ (SEQ ID NO: 5) or comprising an amino acid sequence differing from SEQ ID NO: 5 by one amino acid, and(iii) a CDR3 comprising the amino acid sequence CASSPNMADEQYF (SEQ ID NO: 6) or comprising an amino acid sequence differing from SEQ ID NO: 6 by one amino acid; wherein Formula I (SEQ ID NO: 2) is:Leu-Ni-N2-N3-N4whereinN1 is lie or Vai,N2 is Gin or Arg,N3 is Pro or Ser,N4 is Tyr or Ser, and wherein: when N2 is Gin, N4 is Tyr; and when N2 is Arg, N3-N4 is Pro-Ser, Ser-Ser or Ser-Tyr.

2. The binding molecule of claim 1, wherein when N2 is Arg, N4 is Ser.

3. The binding molecule of claim 1 or 2, wherein when N2 is Arg, N3-N4 is Ser-Ser.

4. The binding molecule of claim 1 or 2, wherein N3 is Pro.

5. The binding molecule of any one of the preceding claims, wherein N1 is He.

6. The binding molecule of any one of the preceding claims, wherein N2-N3-N4 is Gin- Pro -Tyr.

7. The binding molecule of any one of the preceding claims, wherein N1-N2-N3-N4 is Ile- Gln-Pro-Tyr (SEQ ID NO: 76).

8. The binding molecule of any one of the preceding claims, wherein:(A) (i) the T cell receptor (TCR) alpha chain variable domain CDR1 comprises the amino acid sequence of sequence SEQ ID NO: 1, and(ii) the TCR alpha chain variable domain CDR3 comprises the amino acid sequence of sequence SEQ ID NO: 3; and(B) (i) the TCR beta chain variable domain CDR1 comprises the amino acid sequence of sequence SEQ ID NO: 4,(ii) the TCR beta chain variable domain CDR2 comprises the amino acid sequence of sequence SEQ ID NO: 5, and(iii) the TCR beta chain variable domain CDR3 comprises the amino acid sequence of sequence SEQ ID NO: 6.

9. The binding molecule of any one of the preceding claims, wherein the alpha chain variable domain comprises an amino acid sequence of any one of SEQ ID NO: 23 to 32, or a sequence at least 80% identical to the amino acid sequence of any one of SEQ ID NO: 23 to 32; and / or wherein the beta chain variable domain comprises an amino acid sequence SEQ ID NO: 33, or a sequence at least 80% identical to the amino acid sequence SEQ ID NO: 33.

10. The binding molecule of any one of the preceding claims, wherein the binding molecule comprises the ectodomain of a T cell receptor (TCR).

11. The binding molecule of any one of the preceding claims, wherein the binding molecule is a receptor.

12. The binding molecule of any one of the preceding claims, wherein the binding molecule is a TCR.

13. The binding molecule of any one of the preceding claims, wherein the binding molecule binds specifically to a melanoma-associated antigen 3 (MAGE-3) peptide.

14. The binding molecule of any one of the preceding claims, wherein the MAGE-3 peptide is a HLA-A1 presented MAGE-3 peptide.

15. The binding molecule of any one of the preceding claims, wherein the MAGE-3 peptide comprises or consists of the amino acid sequence of SEQ ID NO: 45.

16. The binding molecule of any one of the preceding claims, wherein the binding molecule exhibits reduced binding to a Titin peptide relative to:(a) a first comparator molecule, wherein the first comparator molecule a TCR comprising an alpha chain comprising an amino acid sequence of SEQ ID NO: 36 and a beta chain comprising an amino acid sequence of SEQ ID NO: 40; and / or(b) a second comparator molecule, wherein the second comparator molecular is a TCR comprising an alpha chain comprising an amino acid sequence of SEQ ID NO: 38 and a beta chain comprising an amino acid sequence of SEQ ID NO: 41; and / or(c) a third comparator molecule, wherein the third comparator molecule is a TCR comprising an alpha chain comprising an amino acid sequence of SEQ ID NO: 39 and a beta chain comprising an amino acid sequence of SEQ ID NO: 42; optionally wherein the Titin peptide is a HLA-A1 presented Titin peptide, further optionally wherein the reduced binding is an at least 20% reduction in binding affinity when measured by surface plasmon resonance (SPR) and / or when measured by biolayer interferometry (BLI), preferably when measured by SPR.

17. The binding molecule of claim 16, wherein the Titin peptide comprises or consists of an amino acid sequence of SEQ ID NO: 46.

18. The binding molecule of any one of the preceding claims, wherein the binding molecule does not exhibit meaningful sensitivity to one or more of the peptides listed in Table 3, optionally wherein the binding molecule does not exhibit meaningful sensitivity to any peptide having an amino acid sequence of SEQ ID NOs: 46, 49, 50 or 52.

19. A polynucleotide comprising a sequence encoding an alpha chain variable domain as set forth in any one of claims 1-9 or a sequence encoding the binding molecule of any one of claims 1-18.

20. A pair of polynucleotides comprising a first polynucleotide comprising a sequence encoding an alpha chain variable domain as set forth in any one of claims 1 -9 and a second polynucleotide comprising a sequence encoding a beta chain variable domain as set forth in any one of claims 1-9.

21. A vector comprising the polynucleotide of claim 19.

22. A cell comprising the binding molecule of any one of claims 1-18, the polynucleotide of claim 19, the pair of polynucleotides of claim 20 and / or the vector of claim 21, optionally wherein the cell is an isolated cell.

23. A pharmaceutical composition comprising the binding molecule of any one of claims 1-18, the polynucleotide of claim 19, the pair of polynucleotides of claim 20, the vector of claim 21 and / or the cell of claim 22 or a plurality of the cells of claim 22, and a pharmaceutically acceptable carrier.

24. A binding molecule according to any one of claims 1-18, a polynucleotide according to claim 19, a pair of polynucleotides according to claim 20, a vector according to claim 21, a cell according to claim 22 or a plurality of the cells of claim 22, or a pharmaceutical composition according to claim 23, for use in a method of treatment.

25. A binding molecule according to any one of claims 1-18, a polynucleotide according to claim 19, a pair of polynucleotides according to claim 20, a vector according to claim 21, a cell according to claim 22 or a plurality of the cells of claim 22, or a pharmaceutical composition according to claim 23, for use in a method of treating a subject having a MAGE- A3 -associated disease or disorder, wherein the method comprising administering the binding molecule, polynucleotide, pharmaceutical composition, vector, cell or cells to the subject, optionally wherein the MAGE-A3 -associated disease or disorder is a MAGE-A3 -associated cancer, further optionally wherein the MAGE-A3 -associated cancer is a solid tumour cancer.

26. A binding molecule comprising:(A) a T cell receptor (TCR) alpha chain variable domain comprising:(i) a complementary determining region (CDR) 1 comprising an amino acid sequence DSAIYN (SEQ ID NO: 1) or comprising an amino acid sequence differing from SEQ ID NO: 1 by one amino acid,(ii) a CDR2 comprising the amino acid sequence of Formula I (SEQ ID NO: 2), and(iii) a CDR3 comprising the amino acid sequence CAVRPGGAGSY QLTF (SEQ ID NO: 3) or comprising an amino acid sequence differing from SEQ ID NO: 3 by one amino acid; and(B) a TCR beta chain variable domain comprising:(i) a CDR1 comprising the amino acid sequence SGHRS (SEQ ID NO: 4) or comprising an amino acid sequence differing from SEQ ID NO: 4 by one amino acid,(ii) a CDR2 comprising the amino acid sequence YFSETQ (SEQ ID NO: 5) or comprising an amino acid sequence differing from SEQ ID NO: 5 by one amino acid, and(iii) a CDR3 comprising the amino acid sequence CASSPNMADEQYF (SEQ ID NO: 6) or comprising an amino acid sequence differing from SEQ ID NO: 6 by one amino acid; wherein Formula I (SEQ ID NO: 2) is Leu-Nl-N2-N3-N4, the binding molecule binds to a complex of an epitope of SEQ ID NO: 45 (MAGE -A3) and an HLA molecule, and wherein the binding molecule forms fewer bonds to the HLA molecule and more bonds to the epitope of SEQ ID NO: 45 compared to a TCR comprising an alpha chain comprising an amino acid sequence of SEQ ID NO: 36 and a beta chain comprising an amino acid sequence of SEQ ID NO: 40.

Images