T Cell Modification

Abstract

The present invention provides a modified T cell or population of modified T cells comprising a heterologous recombinant T cell receptor (TCR) and heterologous recombinant co-receptor, additionally provided are methods of producing the modified T cell or population of modified T cells and their use in the treatment of cancer.

Claims

1. A modified T cell or population of modified T cells comprising a heterologous CD8 co-receptor and a heterologous T cell receptor (TCR).

2. The modified T cell or population of modified T cells according to claim 1, wherein the CD8 co-receptor is CD8.

3. The modified T cell or population of modified T cells according to claim 2, wherein the CD8 co-receptor comprises an amino acid sequence having at least 80% sequence identity to SEQ ID NO: 1.

4. The modified T cell or population according to modified T cells according to claim 1, wherein the TCR binds a cancer or tumour antigen or peptide thereof and/or is an affinity enhanced TCR.

5. The modified T cell or population of modified T cells according to claim 1, wherein the TCR is a MAGE A4 TCR and/or can bind MAGE-A4.

6. The modified T cell or population of modified T cells according to claim 5, wherein the TCR comprises an chain amino acid sequence having at least 80% sequence identity to SEQ ID NO: 3.

7. The modified T cell or population of modified T cells according to claim 5, wherein the TCR comprises a chain amino acid sequence having at least 80% sequence identity to SEQ ID NO: 5.

8. A nucleic acid or nucleic acid construct encoding the TCR and CD8 co-receptor according to claim 1.

9. The nucleic acid or nucleic acid construct according to claim 8, comprising; i. a first nucleotide sequence encoding the CD8 co-receptor; and ii. a second nucleotide sequence encoding the T cell receptor.

10-17. (canceled)

18. A vector comprising the nucleic acid construct according to claim 8.

19. The vector according to claim 18, wherein the vector is a lentiviral vector.

20. (canceled)

21. A method for preparing a modified T cell or population of modified T cells, comprising introducing the nucleic acid or nucleic acid construct of claim 8 into a T cell or population of T cells.

22-24. (canceled)

25. A method of treating cancer in an individual, comprising administering to the individual the modified T cell or population of modified T cells of claim 1.

26. (canceled)

27. The method of claim 25, wherein the cancer is a solid tumour.

28. The method of claim 25, further comprising administering to the individual a second therapeutic agent.

29. A method of preparing a modified T cell or population of modified T cells comprising: i. providing a T cell or population of T cells; ii. introducing the vector according to claim 18 into said T cell or population of T cells; and iii. expressing said vector in the T cell or population of T cells.

Description

BRIEF DESCRIPTION OF FIGURES

[0265] FIG. 1. Frequency of CD40L+ cells within the CD4+ T cell subset. The frequency of CD40L+ T cells was determined in the TCR negative (TCR) and TCR positive (TCR+) subsets of the CD4+ T cell population in unstimulated T cells (T cells only), stimulation with antigen negative target cells (Colo205) or antigen positive target cells (A375). Peptide-pulsed (+P) target cells served as controls (right-hand side graphs). Box and whisker plots summarize the data for all five Wave products tested, within these each individual data point represents one Wave product. Statistical significance was assessed using R with a 3-way repeated-measures ANOVA, with subset (i.e. CD4/8), TCR+/, and transduction level as within-subject factors followed by pairwise post-hoc tests for each combination of transduction within a subset/TCR combination and p values adjusted using the Holm method.

[0266] FIG. 2. Proliferation index of transduced CD4+ and CD8+ lymphocytes in response to antigen positive A375. The average number of divisions that all responding cells have undergone were plotted for the CD8+Valpha24+(A) and CD4+Valpha24+(B) cells expressing MAGE-A4 TCR (white) and CD8_MAGE-A4 TCR (white). The PI was shown for four separate donors and as a combined meanSEM across the four donors. Statistical significance was assessed using a 2-way ANOVA followed by Sidak post-hoc multiple comparisons test for variance among the groups.

[0267] FIG. 3. Production of IFN in response to antigen. 510.sup.4 T cells were co-incubated with either A) 510.sup.4 T2 cells and a titration of MAGE-A4 GVYDGREHTV peptide as indicated on x axis or B) 510.sup.4 antigen positive A375 cells. Supernatants were harvested after 24 hours and frozen for later analysis by Magpix. Data from 5 assays was collated and analysed using R (CEL_ANA18_021). Columns indicated T cell donor (Wave213-217), and rows different T cell fractions (mixed PBL product, isolated CD4+ or isolated CD8+). T cell transduction is indicated by colour, non-transduced ntd=grey, MAGE-A4 TCR=white and CD8_MAGE-A4 TCR=black. A). Data are represented as mean of 2 biological replicates SEM. The response curves were fitted using 4-parameter log-logistic function with baseline constrained to 0. B) Each point represents one biological replicate. Data collated in CEL_ANA18_021.

[0268] FIG. 4. Production of IL-2 in response to antigen. 510.sup.4 T cells were co-incubated with either A) 510.sup.4 T2 cells and a titration of MAGE-A4 GVYDGREHTV peptide as indicated on x axis or B) 510.sup.4 antigen positive A375 cells. Supernatants were harvested after 24 hours and frozen for later analysis by Magpix. Data from 5 assays was collated and analysed using R (CEL_ANA18_021). Columns indicated T cell donor (Wave213-217), and rows different T cell fractions (mixed PBL product, isolated CD4+ or isolated CD8+). T cell transduction is indicated by colour, ntd=grey, MAGE-A4 TCR=white and CD8_MAGE-A4 TCR=black. A) Data are represented as mean of 2 biological replicates SEM. The response curves were fitted using 4-parameter log-logistic function with baseline constrained to 0. B) Each point represents one biological replicate. Data collated in CEL_ANA18_021.

[0269] FIG. 5. Production of IL-12 (p40/p70) in dendritic cell co-culture assays. Left, centre and right hand panels show data from mixed PBLs, isolated CD4+ and isolated CD8+ T cells as indicated. The x axis indicates T cells and tumour alone (T cells), T cells, tumour and DCS (T cells +DCs), tumour cells alone (media) or DCs and tumour cells without T cells (DCs). T cell transduction is indicated by colour ntd=grey, MAGE-A4 TCR=hollow black, and CD8_MAGE-A4 TCR=filled black. Additional positive control for DC activation were LPS (filled triangle). Individual plots are shown for multiple donors (TEA and TSA=small scale T cells with autologous DCs; Wave216 and Wave217=large scale T cells co-cultured with DCs from donor TSA).

[0270] FIG. 6. Production of MIG (CXCL9) in DC co-culture assays. Left, centre and right hand panels show data from mixed PBLs, isolated CD4+ and isolated CD8+ T cells as indicated. The x axis indicates T cells and tumour alone (T cells), T cells, tumour and DCS (T cells +DCs), tumour cells alone (media) or DCs and tumour cells without T cells (DCs). T cell transduction is indicated by colour ntd=grey, MAGE-A4 TCR=hollow black, and CD8_MAGE-A4 TCR=filled black. Additional positive control for DC activation were LPS (filled triangle). Individual plots are shown for multiple donors (TEA and TSA=small scale T cells with autologous DCs; Wave216 and Wave217=large scale T cells co-cultured with DCs from donor TSA).

[0271] FIG. 7. Production of IL-6 in DC co-culture assays. Left, centre and right hand panels show data from mixed PBLs, isolated CD4+ and isolated CD8+ T cells as indicated. The x axis indicates T cells and tumour alone (T cells), T cells, tumour and DCS (T cells +DCs), tumour cells alone (media) or DCs and tumour cells without T cells (DCs). T cell transduction is indicated by colour ntd=grey, MAGE-A4 TCR=hollow black, and CD8_MAGE-A4 TCR=filled black. Additional positive control for DC activation were LPS (filled triangle). Individual plots are shown for multiple donors (TEA and TSA=small scale T cells with autologous DCs; Wave216 and Wave217=large scale T cells co-cultured with DCs from donor TSA).

[0272] FIG. 8. Production of IFN in DC co-culture assays. Left, centre and right hand panels show data from mixed PBLs, isolated CD4+ and isolated CD8+ T cells as indicated. The x axis indicates T cells and tumour alone (T cells), T cells, tumour and DCS (T cells+DCs), tumour cells alone (media) or DCs and tumour cells without T cells (DCs). T cell transduction is indicated by colour ntd=grey, MAGE-A4 TCR=hollow black, and CD8_MAGE-A4 TCR=filled black. Additional positive control for DC activation were LPS (filled triangle). Individual plots are shown for multiple donors (TEA and TSA=small scale T cells with autologous DCs; Wave216 and Wave217=large scale T cells co-cultured with DCs from donor TSA).

[0273] FIG. 9. Cytotoxic activity of ADP-A2M4CD8 T cells towards large, MAGE-A4 expressing A375.GFP microtissues (800 m diameter1200 cells/well seeded). The GFP fluorescence area of the central core of the microtissue in each well was measured over time for each treatment (mean of 6 replicates+/SEM shown for all conditions)). Dotted lines indicate T cell addition. Black arrows indicate difference in microtissue area with CD4+ADP-A2M4 and ADP-A2M4CD8 T cells at the assay end point. Data is expressed as microtissue core fluorescence area for each condition over time.

[0274] FIG. 10. Cytotoxic activity of ADP-A2M4CD8 T cells towards large, MAGE-A4 expressing A375.GFP microtissues. Scatter plots showing microtissue area under the curve (AUC) with ntd, ADP-A2M4 or ADP-A2M4CD8 for PBL, CD4+ isolated, and CD8+ isolated T cell fractions. Each data point shows the mean of n=6 for all conditions for each Wave product. All data is shown normalised to the timepoint of T cell addition. Repeated measure ANOVA were used to compare overall normalised AUC data within each donor fraction, for small microtissues (ns=not significant; ****=p<0.0001).

[0275] FIG. 11 Cytotoxic activity of ADP-A2M4CD8 T cells towards large, MAGE-A4 expressing A375.GFP microtissues (800 m diameter1200 cells/well seeded). The GFP fluorescence area of the central core of the microtissue in each well was measured over time for each treatment (mean of 6 replicates+/SEM shown for all conditions)). Dotted lines indicate T cell addition. Black arrows indicate difference in microtissue area with CD4+ADP-A2M4 and ADP-A2M4CD8 T cells at the assay end point. Data is expressed as microtissue core fluorescence area for each condition over time.

[0276] FIG. 12 Cytotoxic activity of ADP-A2M4CD8 T cells towards large, MAGE-A4 expressing A375.GFP microtissues. Scatter plots showing microtissue area under the curve (AUC) with ntd, ADP-A2M4 or ADP-A2M4CD8 for PBL, CD4+ isolated, and CD8+ isolated T cell fractions. Each data point shows the mean of n=6 for all conditions for each Wave product. All data is shown normalised to the timepoint of T cell addition. Repeated measure ANOVA were used to compare overall normalised AUC data within each donor fraction, for small microtissues (ns=not significant; ****=p<0.0001).

[0277] FIG. 13. Cytotoxic activity of CD8_MAGE-A4 T cells towards small (A) and large (B), MAGE-A4 expressing A375.GFP microtissues. (800 m diameter-1200 cells/well seeded). Representative images for Wave donor 217 T cells show the GFP fluorescent microtissue with CD4+ntd, MAGE-A4 TCR and CD8_MAGE-A4 TCR transduced T cells (80,000 cells/well) at the following time points during the assay: 4 hours pre-T cell addition (=143 hours post target seeding); 0 hours post-T cell addition (=147 hours post target seeding); 141 hours post-T cell addition (=288 hours post target seeding); 438 hours post-T cell addition (=585 hours post target seeding).

[0278] FIG. 14. IFN gamma and Granzyme B release by ntd, ADP-A2M4 and ADP-A2M4CD8 T cells in co-culture with A375.GFP 3D microtissues. Supernatants were collected from duplicate assay plates 50h post T cell addition. The levels of IFN gamma (A) and Granzyme B (B) in the supernatants were measured by ELISA following 4-fold sample dilution. Graphs display levels of cytokine produced by PBL, CD4+ or CD8+MAGE-A4 TCR T cells, MAGE-A4-CD8 TCR T cells or ntd T cells incubated with small (505-600 m diameter) or large (800 m diameter) A375.GFP 3D microtissues. Each data point shows the mean of n=6 for all conditions for each Wave product. Repeated measures ANOVAs were used to compare overall IFN and Granzyme B release within each donor fraction against small or large microtissues (ns=not significant; **** =p<0.0001). N.B. Granzyme B cytokine values measured that exceeded the maximum detectable range were assigned a value equal to the highest measurable value (=40,000 pg/ml).

EXAMPLES

[0279] In this second-generation TCR study, we added a CD8 homodimer to the specific peptide enhanced affinity receptor (SPEAR) MAGE-A4.sup.c1032 TCR (MAGE-A4 TCR herein), a first generation TCR currently being tested in a clinical trial (NCT03132922).

[0280] Transduction of HLA class I-restricted, specific peptide enhanced affinity receptor (SPEAR) TCRs into peripheral blood lymphocytes creates both cytotoxic (CD8+) and helper (CD4+) T-cells of the same specificity; however, the lack of CD8 co-receptors on CD4+ T-cells may affect binding avidity of the engineered TCR. The addition of CD8 co-receptor into CD4+ T-cells alongside the engineered TCR CD8_MAGE-A4.sup.c1032 (CD8_MAGE-A4 herein) was anticipated to increase TCR binding avidity and enhance the polyfunctional response of CD4+ T-cells against tumor antigens, thereby widening the immune response to the tumor through dendritic cell (DC) activation and enhanced cytotoxicity.

Example 1. Preparation of Modified T Cells

[0281] The intention of the following experiments was to investigate whether the stability of the TCR interaction with HLA class I and peptide antigen complex in transduced helper CD4+ T cells would be aided through the presence of CD8 co-receptor, commonly present on cytotoxic T cells, and thereby provide improved CD4+ T cell function in response to Class I antigens and enhance the polyfunctional response of CD4+ T cells against tumour antigens.

[0282] Human T cells were lentivirally transduced to constitutively co-express an affinity-enhanced MAGE-A4 TCR, [SEQ ID NO: 3 and 5], which recognises the HLA-A*02:01-restricted MAGE-A4 peptide GVYDGREHTV and the cluster of differentiation CD8 (homodimer) co-receptor [SEQ ID NO: 1]. Control T cells were also provided which lacked the CD8.

[0283] Lentiviral particles encoding the enhanced-affinity TCR, with or without CD8, were produced by transient transfection of HEK293T/17 cells. HEK293T/17 cells were seeded 48-72 hours prior to transfection in 5-layer cell factories, to ensure 60-80% confluency when they were co-transfected with a single lentiviral transgene plasmid (encoding the TCR, with or without CD8) and a set of complimentary lentiviral packaging plasmids. Replication-deficient lentiviral particles were produced over a 48-72 hour period, which were harvested from the supernatant, concentrated via centrifugation and cryopreserved. Lentiviral biological titre determination was performed to provide optimum transduction efficiencies and provide the volume of lentiviral vector required to give a Multiplicity of Infection (MOI) of 1 for T cell transduction.

[0284] Wave T cells were obtained for transduction through addition of the high-titre lentiviral suspension and expanded. For the Wave expansions, CD3+ T cells were isolated and stimulated using CD3/CD28 antibody-coated microbeads (Dynabeads CTS, Life Technologies) from cryopreserved healthy human donor leukopaks (enriched leukapheresis products collected from normal peripheral blood). The CD3+ cells were expanded using the Wave platform (Xuri Cell Expansion System, GE Healthcare Life Sciences) to provide a T cell product.

[0285] The large-scale Wave T cells were obtained from human PBMCs. A leukopak containing 2.0-2.510.sup.9 human PBMCs was thawed, CD3+ T cells were isolated using CD3/CD28 antibody coated microbeads and VueLife bags were seeded with CD3+ cells. For each donor, several VueLife bags were seeded: three to produce non-transduced (ntd) T cells, three to produce the MAGE-A4 TCR alone, two to produce T cells transduced with the CD8_MAGE-A4 TCR. Subsequently the CD3+ cells were transduced with the appropriate volume of lentiviral vector to give an MOI of 1. Following a growth period the cells were washed and placed in to a new VueLife bag. Cell count was monitored daily and the volume of media in increased as required up to a maximum of 500 mL. The cells were transferred to a Wave bag and Xuri Cell Expansion System once all conditions were above 15010.sup.6 cells. Thereafter the expanded T cells were harvested from the WAVE bags, T cells were washed and pooled. Beads were removed from the cells, followed by harvesting, washing and cryopreservation in multiple aliquots for later analysis.

[0286] For some experiments it was necessary to have separated CD4+ or CD8+ cells purified from the T cell product. The cell separation procedure was carried out on the day of cryopreservation, yielding assay-ready pre-separated vials. The CD8+ cell population was negatively selected through depletion of CD4+ cells. The Miltenyi Biotec CD4 microbeads were used for this purpose according to the manufacturer's instructions. The CD4+ cell population were negatively selected for with a CD83 antibody that only binds native CD8 cells with / CD8 heterodimer. In short 1.010.sup.7 T cells were incubated with 5 g of mouse anti-human CD83 antibody (clone 3TU9618, Creative Diagnostics) for 30 minutes at 4 C. (quantities scaled as required). Upon single wash cycle anti-mouse IgG microbeads (Miltenyi) were used according to manufacturer's protocol to negatively select an unbound fraction on LD columns (Miltenyi). The purity of resulting fractions was determined by flow cytometry.

[0287] Measures were made of T cell activation as follows, with a focus on CD4 cells, determining CD40L activation, cytokine release and direct killing of target cells and by investigating the interaction between T cells and dendritic cells.

Example 2. Assessment of CD40L Surface Expression Following Stimulation

[0288] CD40 ligand (CD40L, also known as CD154) is a member of the TNF family. It is primarily expressed on activated T cells, preferentially CD4+ T cells. It acts as a co-stimulatory molecule which binds CD40 on antigen presenting cells (APCs), which most importantly in this context licenses those APCs to activate antigen specific nave CD8+ T cells.

[0289] Surface expression of CD40L was analysed on Wave T cells following overnight stimulation. For this 0.110.sup.6 target cells were seeded into a 96-well flat bottom plate on day zero and left to adhere overnight at 37 C. and 5% CO.sub.2. A375 cells were used as an antigen positive target cell line, and Colo205 as an antigen negative target cell line. On day 1 T cells were added at an Effector:Target ratio of 5:1 (0.510.sup.6 cells/well) along with an anti-CD40L BV421 antibody (5 l/well). At the same time GolgiStop was added (2.64 l/ml of R10 final concentration) to retain/stabilise CD40L-antibody complexes on the cell surface. The plate was then incubated for 20 hours at 37 C. and 5% CO.sub.2 before staining the cells with CD3 FITC, CD4 BV650, CD8 APCeF780, Valpha24 PE and AQUA to allow identification of live, transduced T cells and specific subsets. Data was acquired on Fortessa X20 instruments using software FACSDiva version 8.0.1; data analysis was performed using FlowJo version 10.3 or version 10.4.1. Graphs were generated with GraphPad Prism version 7.02. Statistical significance was assessed using R with a 3-way repeated-measures ANOVA, with subset (i.e. CD4/8), TCR+/, and transduction level as within-subject factors followed by pairwise post-hoc tests for each combination of transduction within a subset/TCR combination and p values adjusted using the Holm method.

[0290] FIG. 1 summarises data for the CD4+ T cell subset of five Wave T cell products tested. Following co-culture with an antigen negative cell line (Colo205) no difference in the frequency of CD40L+ cells is observed between non-transduced (ntd), MAGE-A4 TCR and CD8_MAGE-A4 CD4+TCR Wave T cells. This is true for both transduced (TCR+) cells and those lacking expression of the MAGE-A4 TCR (TCR). These frequencies are also very similar to those observed in T cells that were cultured alone (T cells only). When the cells are presented with antigen positive target cells (A375) an increase in the frequency of CD40L+ cells was seen within the CD4+TCR+subset of MAGE-A4 TCR transduced Wave T cells. Compared to unstimulated T cells the frequency of CD40L+on CD4+TCR cells increased. A significantly higher proportion of CD4+CD40L+ T cells is detected in CD8_MAGE-A4 TCR transduced Wave products following antigen recognition and activation.

Example 3. T Cell Proliferation

[0291] The effect of CD8 homodimer on T cell antigen-specific functions was also assessed by calculating the proliferation index (PI) of the Valpha24+CD8+(TCR) and CD4+ T cell subsets within MAGE-A4 TCR and CD8_MAGE-A4 TCR T cell products, in response to the MAGE-A4 positive cell line A375 (FIG. 2). The proliferation index accounts for the average number of divisions the responding cells have undergone.

[0292] T cell proliferation in response to antigen was accessed by flow cytometry using a fluorescent dye that allows for the simple detection of the number of cell divisions. T cells were stained with the violet laser excitable dye VPD450, which labels the parental cells uniformly. Upon division, the dye is evenly distributed between daughter cells, each then retaining approximately half of the fluorescence intensity of its parent. Therefore, the reduction of dye intensity indicates cell division and thus proliferation.

[0293] Non-transduced (ntd), MAGE-A4 TCR and CD8_MAGE-A4 TCR wave T cell products from four different donors were thawed and rested for 26 hours at 2.010.sup.6 cells/ml in tryptophan depleted RPMI to promote cell synchronisation. T cells were then stained with VPD450 and incubated alone or in co-culture with-antigen presenting cells at a 5:1 T cells to target cells ratio, in the presence or absence of 10.sup.5M MAGE-A4 peptide GVYDGREHTV. The antigen presenting A375 (MAGE-A4 positive) and the Colo205 (MAGE-A4 negative) were irradiated (48 and 33Gy respectively) prior to co-culture, in order to prevent target cell proliferation. Following 3 days of co-culture, the cells were harvested and stained for T cell markers (CD3, CD4, CD8 and Valpha24 to mark the TCR) and viability.

[0294] Proliferation was assessed in both the CD4 and CD8 total T cell populations and within the TCR positive and negative fractions using flow cytometry. The proliferation peaks were manually gated using the T only cells from the transduced and ntd samples as a guide to set the G.sub.0 gate. This represented the undivided cells. Each generation of dividing cells that occurred after the G0 gate was gated (G1, G2 . . . Gx) as peak VPD450 reductions and antigen-driven proliferation was assessed by calculating percent divided and proliferation index. Samples were acquired on the BD LSRFortessa X-20 using the BD High Throughput Sampler (HTS) system in accordance to CBP079v00. For data acquisition, FACSDiva version 8.0.1 was used, and post-acquisition analysis was performed using FlowJo v10.4.1 and GraphPad PRISM v7.02.

[0295] Within the CD4+ fraction, across the 4 wave donors, cells transduced with of the CD8_MAGE-A4 TCR displayed greater expansion compared to the MAGE-A4 TCR (FIG. 2). The presence of the additional CD8 homodimer on the CD4+ cells, led to an increase in the percentage of cells that underwent division and proliferation index when compared to the MAGE-A4 TCR.

Example 4. Cytokine Production in Response to Tumour Cells Lines

[0296] The main effector function of CD4+ T cells is to orchestrate immune response by providing feedback signals to antigen presenting cells as well as other T cell subsets. This role can be mediated by expression of co-stimulatory signals, like CD40L, or secretion of cytokines and chemokines.

[0297] To investigate the production of cytokines by CD8_MAGE-A4 TCR T cells, T cells were co-cultured for 24 hours with either T2 cells with a titration of MAGE-A4 GVYDGREHTV peptide, or antigen positive A375 tumour cells. T cells were used as harvested (PBLs) as well as product purified for either CD4+ or CD8+ T cells.

[0298] T cells and target cells were added to the wells of 96-well U-bottomed culture plates in duplicate at 50,000 target and 50,000 T cells per well. Target cells were either T2 with added exogenous peptide, GVYDGREHTV, in range of 10.sup.6 to 10.sup.10M, or no peptide as a negative control, or the MAGE-A4 positive tumour cell line A375. Assay plates were incubated for 24 hours at 37 C./5% CO2. Culture media was collected (150 l) for cytokine analysis by Luminex MAGPIX.

[0299] Cytokine and chemokine analysis by MAGPIX was performed using the Invitrogen 25-plex human cytokine panel kit. Samples were acquired using a Bio-Rad Bio-Plex MAGPIX Multiplex Reader using acquisition software Luminex XPONENT for MAGPIX version 4.2 Build 1705. Post-acquisition analysis was performed using R (v3.3.2). Any value above the top of the standard curve was adjusted to top value. Two-way repeated-measure ANOVAs were run separately for each cytokine, with transduction and T cell fraction as within-subject factors, followed by pair-wise post-hoc tests for each combination of transduction within a transduction/T cell fraction combination and p values were adjusted using the Holm method. A pre-defined subset of relevant cytokines were analysed: Granulocyte-Macrophage Colony-Stimulating Factor (GM-CSF), IFN-, IL-2, Tumor Necrosis Factor (TNF)-, MIP-1 (CCL4), IL-17, IL-10, IL-4, IL-5, IL-13, IL-2 Receptor.

[0300] Data shown in FIG. 3 relates to the production of IFN. Data in FIG. 4 relates to the production of IL2. As shown in FIG. 3, when the co-culture included isolated CD4+ cells (top facets), although the absolute levels of cytokine were lower (especially for IFN where the maximum on the y axis is 750 g/ml vs 4000 g/ml for IL2 in FIG. 4) there is a much more consistent improvement in cytokine release by the CD8_MAGE-A4 TCR T cells (black) compared to the MAGE-A4 TCR (white). This pattern is consistent across all 5 donors (except Wave214 which did not respond well in this assay. A similar result is seen in the IFN response to antigen positive A375 tumor cells (FIG. 3, Bottom panels). In general there is little difference between the two TCR constructs in the PBLs or CD8+ fractions, but an improvement in cytokine release between constructs can be seen in 4 out of 5 donors when CD4+ is investigated.

[0301] In FIG. 4, the CD4+ cell data (top facets), demonstrates that there is also a consistent improvement in cytokine release by the CD8_MAGE-A4 TCR T cells (black) compared to the MAGE-A4 TCR (white). The same trend holds for the production of Granulocyte-Macrophage Colony-Stimulating Factor (GM-CSF), Tumor Necrosis Factor (TNF)-, and MIP-1 (CCL4), data not shown. The remaining six cytokines (IL-17, IL-10, IL-4, IL-5, IL-13, IL-2 Receptor) showed minimal production in four out of five wave donors although the levels produced were higher from the CD8_MAGE-A4 TCR T cells compared to MAGE-A4 TCR T cells for both CD4+ and CD8+ isolated cells.

[0302] In conclusion, from the cytokines analysed, and in particular IFN, IL-2, TNF, GM-CSF, and MIP1, there is an improvement in cytokine release by the CD8_MAGE-A4 TCR T cells compared to the MAGE-A4 TCR T cells.

Example 5. Cytokine Production in Response to Tumour Cells Lines in Dendritic Cells Co-Culture

[0303] It was hoped that by introducing CD8 into TCR-transduced CD4+ T cells might promote engagement with additional elements of the immune system to elicit a sustained anti-tumour response hence it was our objective to assess the interaction of CD8_MAGE-A4 TCR T cells with dendritic cells (DCs) during DC maturation and T cell activation. In the context of the tumor microenvironment, improved maturation and activity of DCs could help boost the overall anti-tumor immune response and activation of T cells. To investigate the interactions between T cells and dendritic cells during the DC maturation and T cell activation process, two assay types were set up involving the co-culture of immature DCs, T cells and tumour cell lines. This was designed to reflect the in vivo situation where DCs take up antigen from surrounding tumour cells and present on their MHC class II, rather than simply loading with exogenous peptide (which would only present on MHC class I). To address both the autologous situation and confirm application to clinical large-scale T cells, assays included small scale T cells made from blood donors with matched dendritic cells, as well as wave scale T cells.

[0304] (a) Cell Preparation and Co-Culture

[0305] Co-cultures were set up for 48 hours and each assay type included two small scale T cell preps (donors TEA and TSA, or NLA and OBA) with donor matched DCs, and two large scale donors (Wave213 and Wave217 or Wave216 and Wave217) with unmatched DCs (as specified in figure legends). For these assays antigen positive target cell lines (A375 or NCI-H1755) and an antigen negative target cell line (Nalm6) were included. As positive controls for dendritic cell maturation, wells containing either a cytokine cocktail or lipopolysaccharide (LPS) were included.

[0306] After 7 days of differentiation from CD14+ monocytes, the immature dendritic cells were washed three times in R10, counted and seeded into 48-well plates for co-culture with MAGE-A4-positive or MAGE-A4-negative tumour cell lines and TCR transduced T cells. Co-culture was set up for 48 hours and incubated at 37 C./5% CO.sub.2 to assess the effect of this co-culture on DC maturation and activation. All co-culture experiments used 100,000 target cells (A375, NCI-H1755 or Nalm6), 100,000 dendritic cells and 400,000 T cells. T cells and dendritic cells produced from the same donor were used in each assay. After 48 hours of co-culture supernatants were collected (150 l per well) and frozen at 80 C. for subsequent cytokine and chemokine analysis by MAGPIX or cells harvested for flow cytometry.

[0307] (b) Effect on Maturation Status of Dendritic Cells

[0308] To assess the maturation status of dendritic cells at the end of the 48 hour co-culture period, multicolour immunophenotypic analysis was performed using flow cytometry to determine the expression of maturation markers on dendritic cells. The target cell lines used for the immunophenotyping co-cultures were tagged with nuclear GFP to allow easy differentiation between dendritic cells and target cells. The monoclonal antibodies used were directed against CD1a, CD14, CD40, CD80, HLA-DR, CD3, CD4 and CD40L. In addition to CD1a, DCs were stained for CD40, CD80 and HLA-DR (MHC II). All three of these markers are expressed at low level in immature dendritic cells and are up-regulated upon DC maturation. The expected phenotype of immature dendritic cells is: CD1a+, HLA-DRlow, CD80low, CD40low, CD14-/low, and that of mature dendritic cells is: CD1a+, HLA-DRhigh, CD80high, CD14/low.

[0309] Dendritic cells co-incubated with CD8_MAGE-A4 TCR or MAGE-A4 TCR T cells demonstrated equivalent upregulation of CD80, CD40 and HLA-DR markers in response to antigen positive A375 cells but not with antigen negative Nalm6 cells or non-transduced T cells, hence the activation is TCR and antigen specific. There was no difference between DCs incubated with CD8_MAGE-A4 or MAGE-A4 PBLs. Both CD8_MAGE-A4 and MAGE-A4 transduced T cells specifically activated by antigen can promote the maturation of immature dendritic cells (data not shown).

[0310] (c) Cytokine Assay

[0311] The methodology for the cytokine assay is described in example 4.

[0312] (c. i.) IL-12

[0313] IL-12 is the principal cytokine produced by activated mature dendritic cells and is enhanced through CD40 signalling. Nave CD4+ T cells activated in the presence of IL-12 and/or IFN tend to differentiate into Th1 cells, which in turn support IFN production and increased cytotoxic activity of CD8+ T cells. 110.sup.5 antigen-positive (A375 and NCI-H1755) or antigen-negative (Nalm6) tumour cell lines were co-cultured in a 48 well plate with 110.sup.5 immature dendritic cells and 410.sup.5 T cells. Culture supernatants were harvested after 48 hours and cytokines analysed by Magpix. Data is shown in FIG. 5. Any data points that exceeded the top of the standard curve have been plotted as the maximum value. As the standard curves for IL-12 were consistently still in the linear range at the highest point of the standard curve, the standard range for IL-12 (p40/p70) was extended to 100,000 g/ml for the purpose of data analysis and graph generation. All experimental conditions were tested in biological duplicates and both data points are plotted. Three-way repeated-measure ANOVAs were run separately for each cytokine and positive-control target, with transduction, T cell fraction, and presence or absence of DCs as within-subject factors, followed by pair-wise post-hoc tests for each combination of transduction within a transduction/T cell fraction/DC combination and p values were adjusted using the Holm method. Only significant comparisons between MAGE-A4 TCR and CD8_MAGE-A4 TCR are signified on graph *p<0.05, **p<0.01, ***p<0.005.

[0314] In FIG. 5, IL-12 was only seen in those conditions containing DCs (T cells +DCs on the X axis) confirming this was a DC specific analyte. In response to either A375 or NCI-H1755 antigen positive target lines, large amounts of IL-12 were produced by DCs in co-culture with unseparated T cells (PBLs, left hand panel). When co-cultured with CD8_MAGE-A4 TCR T cells the amount of IL-12 was up to double the amount seen in positive control samples which were incubated with LPS (filled triangle). Consistently across every donor, significantly more IL-12 was produced by DCs exposed to CD8_MAGE-A4 TCR T cells (black points) than those cultured with MAGE-A4 TCR T cells (hollow points) (p=0.002779 A375 and p=0.0000964 NCI-H1755). No IL-12 was produced by DCs cultured with T cells, or media alone, confirming this is a MAGE-A4 antigen specific response.

[0315] When the co-culture included isolated CD4+ T cells (middle panel), the DCs still produced IL-12. The levels of IL-12 induced by the large scale CD4+CD8_MAGE-A4 TCR transduced T cells (Wave216 and Wave217) were broadly similar to that seen with mixed PBLS, yet were barely above background with MAGE-A4 TCR T cells. IL-12 levels produced in co-culture with small scale donors (TEA and TSA) were lower, but still above background. In conclusion there was a greater response with CD8_MAGE-A4 TCR T cells. The increase in IL-12 production between the CD4+ and mixed PBLs conditions highlights that although the licensing of DC maturation is mainly a CD4+ function, CD8+ T cells also play an important role in the positive feedback loop between DCs and T cells in response to antigen. When co-cultures included isolated CD8+ T cells, the DCs did not produce any IL-12 when combined with small scale T cells, and produced moderate levels when combined with large scale donors, and this was unaffected by the CD8 modification.

[0316] (c. ii.) Monokine Induced by Gamma Interferon (MIG or CXCL9),

[0317] MIG is a chemoattractant for T cells and tumour-infiltrating lymphocytes that is produced by dendritic cells, macrophages and other cell types in response to IFN. Its primary function is to recruit primed T lymphocytes to the site of inflammation.

[0318] 110.sup.5 antigen-positive (A375 and NCI-H1755) or antigen-negative (Nalm6) tumour cell lines were co-cultured in a 48 well plate with 110.sup.5 immature dendritic cells and 410.sup.5 T cells. Culture supernatants were harvested after 48 hours and cytokines analysed by Magpix.

[0319] Data is shown in FIG. 6. Any data points that exceeded the top of the standard curve have been plotted as the maximum value. All experimental conditions were tested in biological duplicates and both data points are plotted. Three-way repeated-measure ANOVAs were run separately for each cytokine and positive-control target, with transduction, T cell fraction, and presence or absence of DCs as within-subject factors, followed by pair-wise post-hoc tests for each combination of transduction within a transduction/T cell fraction/DC combination and p values were adjusted using the Holm method. Only significant comparisons between MAGE-A4 TCR and CD8_MAGE-A4 TCR are signified on graph *p<0.05, **p<0.01, ***p<0.005.

[0320] The secretion pattern of MIG (Error! Reference source not found. FIG. 6) is similar to IL-12. MIG produced by DCs co-cultured with mixed PBL T cells is above the quantitation limit of the assay and no differences can be measured. When incubated with isolated CD4+ cells however, there is a clear difference in the MIG produced by the DCs in response to CD8_MAGE-A4 TCR compared to MAGE-A4 TCR T cells with both large and small scale donors. High levels of MIG were also produced by DCs in co-culture with the large scale CD8+ T cells.

[0321] (c. iii.) IL-6

[0322] IL-6 is made by several cell types, including DCs. Clinically it is associated with poor-prognosis and is one of the major cytokines implicated in cytokine release syndrome (CRS). The Magpix data for IL-6 is shown in FIG. 7. The NCI-H1755 cell line secretes approximately 3-4000 pg/ml of IL-6, although the response when co-cultured by transduced T cells is generally above that of targets alone (media on x axis, or ntd (grey point)) so there is some production by DCs. In response to co-culture with T cells and A375 cells there is some also IL-6 production by DCs (up to 1000pg/ml).

[0323] 110.sup.5 Antigen-positive (A375 and NCI-H1755) or antigen-negative (Nalm6) tumour cell lines were co-cultured in a 48 well plate with 110.sup.5 immature dendritic cells and 410.sup.5 T cells. Culture supernatants were harvested after 48 hours and cytokines analysed by Magpix. Any data points that exceeded the top of the standard curve have been plotted as the maximum value. All experimental conditions were tested in biological duplicates and both data points are plotted. Three-way repeated-measure ANOVAs were run separately for each cytokine and positive-control target, with transduction, T cell fraction, and presence or absence of DCs as within-subject factors, followed by pair-wise post-hoc tests for each combination of transduction within a transduction/T cell fraction/DC combination and p values were adjusted using the Holm method. Only significant comparisons between MAGE-A4 TCR and CD8_MAGE-A4 TCR are signified on graph *p<0.05, **p<0.01, ***p<0.005.

[0324] Whilst more IL-6 is produced by DCs in co-culture with CD8_MAGE-A4 TCR T cells than with MAGE-A4 TCR (1.8-2.9 fold, p=0.027115 with A375) the amounts are much less than seen when DCs are stimulated with the positive control (LPS, filled triangle) or the IL-6 made by tumour cells themselves (NCI-H1755). In contrast the levels of IL-12 (a pro-T cell cytokine) produced in the same assays, was higher than either positive control (FIG. 5).

[0325] (c. iv.) IFN

[0326] IFN is a key cytokine secreted by activated T cells in response to antigen and has multiple roles in the anti-tumour response.

[0327] 110.sup.5 Antigen-positive (A375 and NCI-H1755) or antigen-negative (Nalm6) tumour cell lines were co-cultured in a 48 well plate with 110.sup.5 immature dendritic cells and 410.sup.5 T cells. Culture supernatants were harvested after 48 hours and cytokines analysed by Magpix. Data is shown in FIG. 8. IFN secreted by T cells at 48h in the absence of DCs (T cells on x axis, FIG. 8) broadly reflects what was seen in previous 24h co-culture assays (low thousands; Error! Reference source not found. FIG. 3) and there are limited (not statistically significant) differences seen between the MAGE-A4 TCR and the CD8_MAGE-A4 TCR T cell products. Any data points that exceeded the top of the standard curve have been plotted as the maximum value. The standard range for IFN in the 25-plex assay kit is up to 4,450 g/ml. As the standard curves were consistently still in the linear range at the highest point of the standard curve, the standard range was extended to 20,000 g/ml for the purpose of data analysis and graph generation and the y-axis of each graph has been set to this value. All experimental conditions were tested in biological duplicates and both data points are plotted. Three-way repeated-measure ANOVAs were run separately for each cytokine and positive-control target, with transduction, T cell fraction, and presence or absence of DCs as within-subject factors, followed by pair-wise post-hoc tests for each combination of transduction within a transduction/T cell fraction/DC combination and p values were adjusted using the Holm method. Only significant comparisons between MAGE-A4 TCR and CD8_MAGE-A4 TCR are signified on graph *p<0.05, **p<0.01, ***p<0.005.

[0328] When DCs are added to a co-culture of CD8_MAGE-A4 TCR T cells and antigen positive target cells (A375 or NCI-H1755), IFN release increases dramatically (up to 29 fold, donor TEA, NCI-H1755) whereas the effect of adding DCs on the MAGE-A4 TCR T cells is smaller (up to 7 fold, donor TEA, NCI-H1755). There is a significant difference between of IFN produced in the presence of DCs by CD8_MAGE-A4 TCR T cells as compared to MAGE-A4 TCR T cells (p=0.000111 A375, p=0.000197 NCI-H1755). Small scale CD8_MAGE-A4TCR T cells donors (TEA and TSA) showed an even more pronounced increase in IFN levels when co-cultured with antigen positive target cells and DCs. This data illustrates the other arm of the DC:T cell interaction, that the improved activation of DCs by the CD8_MAGE-A4 TCR T cells is in turn allowing the DCs to improve T cell activation. No IFN production was observed when MAGE-A4 TCR or CD8_MAGE-A4 TCR transduced T cells were co-cultured with MAGE-A4 negative Nalm6 cells, with or without addition of DCs, thus demonstrating that the observed responses are antigen specific.

Example 6. Killing of Antigen Positive Microtissues by CD8_MAGE-A4 TCR T Cells

[0329] The killing of antigen positive microtissues by CD8_MAGE-A4 TCR T cells was investigated. Cytotoxic activity against tumour cells is generally characterised as a function of CD8+ cells, but can also be a minor function of CD4+ T cells. The cytotoxic activity of MAGE-A4 TCR and CD8_MAGE-A4 TCR T cells towards GFP labelled 3D cancer cell line microtissues was determined by IncuCyte assay of 3D cancer cell microtissues.

[0330] MAGE-A4 and HLA-A2 positive A375.GFP melanoma cells transduced with cytoplasmic GFP lentivirus were seeded in ultra-low attachment (ULA) 384-well microplates at 150 cells/well and 1200 cells/well starting cell densities and briefly spun down before being incubated at 37 C./5% CO.sub.2 to allow 3D microtissues of the differing sizes to form naturally. Imaging started from the point of cell seeding and continued after addition of T cells until assay completion using the IncuCyte ZOOM 40768 (Essen Bioscience) with images acquired at 3 hour intervals at 10 magnification. Uniform microtissue formation was confirmed in each well prior to the addition of T cells. After 6 days microtissues formed from 150 cells/well seeded were 550-600 m in diameter (hereafter termed small microtissues), while those formed from 1200 cells/well seeded were 800 m in diameter (hereafter termed large microtissues). A375.GFP cells seeded into ULA plates formed stable microtissues of two uniform sizes (small and large) over 6 days prior to addition of T cells, FIGS. 13 A and B top panels shows this for small and large MAGE-A4 expressing A375.GFP microtissues respectively in the context of the Wave 217 cells.

[0331] T cell populations were added as unseparated PBLs (20,000 cells/well), as well as pure separated CD4+(80,000 cells/well) and CD8+(20,000 cells/well) fractions. 10 M GVYDGREHTV MAGE-A4 peptide was also added to designated wells as an additional control for all conditions (data not shown). Following assay completion, raw images of the green fluorescence from each well for all timepoints were analysed whereby the core microtissue fluorescence area was masked, allowing the area of the microtissue to be calculated for each replicate and treatment condition for all timepoints studied. An increase or decrease in fluorescence metrics over time was indicative of 3D microtissue growth or death respectively. 3D microtissue killing metric plots were produced.

[0332] All data were normalised to the timepoint of T cell addition to compensate for any small variances in microtissue size between replicates before T cell killing. Microtissue area over time and area under the curve (AUC) data were determined up to the endpoint of the assays, data is shown in FIGS. 9 to 13.

[0333] The data of FIGS. 9 and 11 show that following stable microtissue formation at 147 h (A) or 145 h (B), ADP-A2M4 and ADP-A2M4CD8 TCR transduced or ntd PBL (20,000 cells/well), CD4+ isolated (80,000 cells/well), or CD8+ isolated (20,000 cells/well) T cells were added from donors Wave214, Wave215 and Wave217 (A) or Wave213 and Wave216 (B). A & B). FIGS. 9 and 11panels A & B demonstrate that following addition of ntd T cells, the A375.GFP microtissues continued to grow steadily during the course of the assay before slowing and reaching a limiting size for the assay plate well 6-8 days after T cell addition. Addition of MAGE-A4 TCR or CD8_MAGE-A4 TCR T cells from unseparated PBL or pure CD8+ populations to A375.GFP microtissues of both sizes resulted in a short period of microtissue expansion as T cells infiltrated the tissue before rapid and complete destruction of the microtissues. This was measured by a rapid loss of normalised microtissue area (FIGS. 9 and 11panels A & B).

[0334] FIGS. 10 and 12 demonstrate that there was no significant difference in the rate of overall killing between MAGE-A4 TCR and CD8_MAGE-A4 TCR T cells from either unseparated PBLs or pure CD8+ T cell populations (FIGS. 10 and 12middle and right panels). However transduction of isolated CD4+ T cells with CD8_MAGE-A4 TCR elicited significantly enhanced cytotoxicity against A375.GFP microtissues of both sizes compared to CD4+ T cells transduced with the MAGE-A4 TCR alone (FIGS. 10 and 12left panels). All five Wave CD8_MAGE-A4 TCR T cells products tested consistently showed improved killing of 3D microtissues. Furthermore, complete destruction of both microtissue sizes by CD4+CD8_MAGE-A4 TCR T cells was achieved by the end of the assay with three of five donors (FIGS. 9 and 11, A-Bleft hand side panels; FIG. 13 see iii). These data demonstrate that expression of the CD8 co-receptor significantly enhances the effector function of CD4+ T cells transduced with the MAGE-A4 TCR.

Example 7. CD8_MAGE-A4 TCR T Cell Production of IFN and Granzyme B in Response to A375.GFP 3D Microtissues

[0335] The ability of MAGE-A4 TCR and CD8_MAGE-A4 TCR unseparated PBL, purified CD4.sup.+ and CD8.sup.+ T cells to produce IFN and Granzyme B in response to A375.GFP 3D microtissues was assessed. Supernatants were collected from duplicate plates set up in parallel with IncuCyte assays plates after 50h post T cell addition. Supernatants were analysed for IFN and Granzyme B by ELISA in 384 well plates. Sample supernatants were diluted 4-fold in R10 assay medium prior to addition to ELISA plates.

[0336] The plates were developed using Glo substrate luminescence HRP substrate and each plate was incubated for five minutes prior to being read on the BMG LABTECH FLUOstar Omega plate reader. Data analysis was conducted in the Omega-data analysis software (version 3.10R6) and 4-fold dilution factor applied to cytokine values obtained to account for sample dilution. Analysed data was exported to Excel and graphed using a custom R script in R version 3.2.2. Within the R script, sample wells that had a value exceeding the highest standard concentration were assigned the value of the top standard. Wells that had a value less than the standard curve range were assigned a value of 0.100 g/ml above the highest value for the corresponding ntd T cells in the presence of targets without exogenous MAGE-A4 peptide was used to distinguish between a background signal and a positive IFN response. 200 g/ml above the highest value for the corresponding ntd T cells in the presence of targets without exogenous MAGE-A4 peptide was used to distinguish between a background signal and a positive Granzyme B response. Repeated measures ANOVAs were used to compare levels of IFN and Granzyme B release within each fraction for the combined data across all waves, for each target microtissue size.

[0337] Supernatants were collected from parallel assay plates at 50 hours after T cell addition and assayed by ELISA. Robust cytokine responses were observed with MAGE-A4 TCR and CD8_MAGE-A4 TCR T cells from unseparated PBL and CD8+ subsets from all five Wave T cell products tested (FIG. 14). No significant difference was observed in the levels of IFN or Granzyme B released by MAGE-A4 TCR and CD8_MAGE-A4 TCR T cells from unseparated PBL or purified CD8+ subsets, respectively. This indicated that additional CD8 co-receptor expression in CD8+ T cells does not result in enhanced cytotoxicity towards antigen positive targets.

[0338] The levels of IFN and Granzyme B production were significantly increased by CD8_MAGE-A4 TCR in comparison to MAGE-A4 TCR CD4+ T cells in purified CD4+ T cells across all Wave products tested in response to A375.GFP 3D microtissues of both sizes (FIG. 14left panels). These data match the enhanced killing capability of CD8_MAGE-A4 TCR CD4+ T cells compared to MAGE-A4 TCR CD4+ T cells (FIGS. 9-13).

[0339] Overall, these data suggest that engineered co-expression of the CD8 homodimer with the MAGE-A4 TCR in CD4+ T cells elicits a substantial improvement in the cytotoxic response towards antigen-positive 3D microtissues compare to CD4+ T cells transduced with the MAGE-A4 TCR alone. This provides rationale for the use of CD8 to enhance the CD4+ T cells transduced with recombinant TCR to enhance potency in the cytotoxic response against antigen positive targets.

TABLE-US-00001 Sequences MALPVTALLLPLALLLHAARPSQFRVSPLDRTWNLGETVELKCQVLLSNPTSGCSWLFQPRGAAASPTFLLYLSQNK PKAAEGLDTQRFSGKRLGDTFVLTLSDFRRENEGYYFCSALSNSIMYFSHFVPVFLPAKPTTTPAPRPPTPAPTIAS QPLSLRPEACRPAAGGAVHTRGLDFACDIYIWAPLAGTCGVLLLSLVITLYCNHRNRRRVCKCPRPVVKSGDKPSLS ARYV SEQIDNO:1(CD8)CDRsboldunderlined,signalsequenceitalicunderlined ATGGCCTTACCAGTGACCGCCTTGCTCCTGCCGCTGGCCTTGCTGCTCCACGCCGCCAGGCCGAGCCAGTTCCGGGT GTCGCCGCTGGATCGGACCTGGAACCTGGGCGAGACAGTGGAGCTGAAGTGCCAGGTGCTGCTGTCCAACCCGACGT CGGGCTGCTCGTGGCTCTTCCAGCCGCGCGGCGCCGCCGCCAGTCCCACCTTCCTCCTATACCTCTCCCAAAACAAG CCCAAGGCGGCCGAGGGGCTGGACACCCAGCGGTTCTCGGGCAAGAGGTTGGGGGACACCTTCGTCCTCACCCTGAG CGACTTCCGCCGAGAGAACGAGGGCTACTATTTCTGCTCGGCCCTGAGCAACTCCATCATGTACTTCAGCCACTTCG TGCCGGTCTTCCTGCCAGCGAAGCCCACCACGACGCCAGCGCCGCGACCACCAACACCGGCGCCCACCATCGCGTCG CAGCCCCTGTCCCTGCGCCCAGAGGCGTGCCGGCCAGCGGCGGGGGGCGCAGTGCACACGAGGGGGCTGGACTTCGC CTGTGATATCTACATCTGGGCGCCCTTGGCCGGGACTTGTGGGGTCCTTCTCCTGTCACTGGTTATCACCCTTTACT GCAACCACAGGAACCGAAGACGTGTTTGCAAATGTCCCCGGCCTGTGGTCAAATCGGGAGACAAGCCCAGCCTTTCG GCGAGATACGTCGGTTCAAGAGCTAAAAGAAGTGGTAGTGGTGCCCCTGTGA SEQIDNO:2(CD8) MKKHLTTFLVILWLYFYRGNGKNQVEQSPQSLIILEGKNCTLQCNYTVSPFSNLRWYKQDTGRGPVSLTILTFSENT KSNGRYTATLDADTKQSSLHITASQLSDSASYICVVSGGTDSWGKLQFGAGTQVVVTPDIQNPDPAVYQLRDSKSSD KSVCLFTDFDSQTNVSQSKDSDVYITDKTVLDMRSMDFKSNSAVAWSNKSDFACANAFNNSIIPEDTFFPSPESSCD VKLVEKSFETDTNLNFQNLSVIGFRILLLKVAGENLLMTLRLWSSGSRAKR SEQIDNO:3(MAGEA4TCRchain)CDRsboldunderlined ATGAAGAAGCACCTGACCACCTTTCTCGTGATCCTGTGGCTGTACTTCTACCGGGGCAACGGCAAGAACCAGGTGGA ACAGAGCCCCCAGAGCCTGATCATCCTGGAAGGCAAGAACTGCACCCTGCAGTGCAACTACACCGTGTCCCCCTTCA GCAACCTGCGGTGGTACAAGCAGGACACCGGCAGAGGCCCTGTGTCCCTGACCATCCTGACCTTCAGCGAGAACACC AAGAGCAACGGCCGGTACACCGCCACCCTGGACGCCGATACAAAGCAGAGCAGCCTGCACATCACCGCCAGCCAGCT GAGCGATAGCGCCAGCTACATCTGCGTGGTGTCCGGCGGCACAGACAGCTGGGGCAAGCTGCAGTTTGGCGCCGGAA CACAGGTGGTCGTGACCCCCGACATCCAGAACCCTGACCCTGCCGTGTACCAGCTGCGGGACAGCAAGAGCAGCGAC AAGAGCGTGTGCCTGTTCACCGACTTCGACAGCCAGACCAACGTGTCCCAGAGCAAGGACAGCGACGTGTACATCAC CGACAAGACCGTGCTGGACATGCGGAGCATGGACTTCAAGAGCAATAGCGCCGTGGCCTGGTCCAACAAGAGCGACT TCGCCTGCGCCAACGCCTTCAACAACAGCATTATCCCCGAGGACACATTCTTCCCAAGCCCCGAGAGCAGCTGCGAC GTCAAGCTGGTGGAAAAGAGCTTCGAGACAGACACCAACCTGAACTTCCAGAACCTGAGCGTGATCGGCTTCAGAAT CCTGCTGCTGAAGGTGGCCGGCTTCAACCTGCTGATGACCCTGAGACTGTGGTCCAGCGGCAGCCGGGCCAAGAGA SEQIDNO:4(MAGEA4TCRchaincodingsequence) MASLLFFCGAFYLLGTGSMDADVTQTPRNRITKTGKRIMLECSQTKGHDRMYWYRQDPGLGLRLIYYSFDVKDINKG EISDGYSVSRQAQAKFSLSLESAIPNQTALYFCATSGQGAYEEQFFGPGTRLTVLEDLKNVEPPEVAVFEPSEAEIS HTQKATLVCLATGFYPDHVELSWWVNGKEVHSGVSTDPQPLKEQPALNDSRYCLSSRLRVSATFWQNPRNHFRCQVQ FYGLSENDEWTQDRAKPVTQIVSAEAWGRADCGFTSESYQQGVLSATILYEILLGKATLYAVLVSALVLMAMVKRKD SRG SEQIDNO:5(MAGEA4TCRchain)CDRsboldunderlined ATGGCCAGCCTGCTGTTCTTCTGCGGCGCCTTCTACCTGCTGGGCACCGGCTCTATGGATGCCGACGTGACCCAGAC CCCCCGGAACAGAATCACCAAGACCGGCAAGCGGATCATGCTGGAATGCTCCCAGACCAAGGGCCACGACCGGATGT ACTGGTACAGACAGGACCCTGGCCTGGGCCTGCGGCTGATCTACTACAGCTTCGACGTGAAGGACATCAACAAGGGC GAGATCAGCGACGGCTACAGCGTGTCCAGACAGGCTCAGGCCAAGTTCAGCCTGTCCCTGGAAAGCGCCATCCCCAA CCAGACCGCCCTGTACTTTTGTGCCACAAGCGGCCAGGGCGCCTACGAGGAGCAGTTCTTTGGCCCTGGCACCCGGC TGACAGTGCTGGAAGATCTGAAGAACGTGTTCCCCCCAGAGGTGGCCGTGTTCGAGCCTTCTGAGGCCGAAATCAGC CACACCCAGAAAGCCACACTCGTGTGTCTGGCCACCGGCTTCTACCCCGACCACGTGGAACTGTCTTGGTGGGTCAA CGGCAAAGAGGTGCACAGCGGCGTGTCCACCGATCCCCAGCCTCTGAAAGAACAGCCCGCCCTGAACGACAGCCGGT ACTGCCTGAGCAGCAGACTGAGAGTGTCCGCCACCTTCTGGCAGAACCCCAGAAACCACTTCAGATGCCAGGTGCAG TTTTACGGCCTGAGCGAGAACGACGAGTGGACCCAGGACAGAGCCAAGCCCGTGACACAGATCGTGTCTGCCGAAGC TTGGGGGCGCGCCGATTGTGGCTTTACCAGCGAGAGCTACCAGCAGGGCGTGCTGAGCGCCACCATCCTGTACGAGA TCCTGCTGGGAAAGGCCACACTGTACGCCGTGCTGGTGTCTGCCCTGGTGCTGATGGCCATGGTCAAGCGGAAGGAC AGCCGGGGC SEQIDNO:6(MAGEA4TCRchaincodingsequence) MKKHLTTFLVILWLYFYRGNGKNQVEQSPQSLIILEGKNCTLQCNYTVSPFSNLRWYKQDTGRGPVSLTILTFSENT KSNGRYTATLDADTKQSSLHITASQLSDSASYICVVSGGTDSWGKLQFGAGTQVVVTPD SEQIDNO:7(MAGEA4TCRchainvariableregion)136AA-CDRsbold underlined MASLLFFCGAFYLLGTGSMDADVTQTPRNRITKTGKRIMLECSQTKGHDRMYWYRQDPGLGLRLIYYSFDVKDINKG EISDGYSVSRQAQAKFSLSLESAIPNQTALYFCATSGQGAYEEQFFGPGTRLTVLE SEQIDNO:8(MAGEA4TCRchainvariableregion)133AA-CDRsbold underlined VSPFSNSEQIDNO:9;CDR1MAGEA4TCRchain,(residues48-53) LTFSENSEQIDNO:10;CDR2MAGEA4TCRchain,(residues71-76) CVVSGGTDSWGKLQFSEQIDNO:11;CDR3MAGEA4TCRchain,(residues111-125) KGHDRSEQIDNO:12;CDR1MAGEA4TCRchain,(residues46-50) SFDVKDSEQIDNO:13;CDR2MAGEA4TCRchain,(residues68-73) CATSGQGAYEEQFFSEQIDNO:14;CDR3MAGEA4TCRchain,(residues110-123) VLLSNPTSGSEQIDNO:15;CDR1CD8(residues45-53) YLSQNKPKSEQIDNO:16;CDR2CD8(residues72-79) LSNSIMSEQIDNO:17;CDR3CD8(residues118-123) GVYDGREHTV-SEQIDNO:18