WO2024235848A1

WO2024235848A1 - Novel t cell receptors

Info

Publication number: WO2024235848A1
Application number: PCT/EP2024/062926
Authority: WO
Inventors: Sandra HERVÁS STUBBS; Enric VERCHER HERRÁEZ
Original assignee: Fundacion para la Investigacion Medica Aplicada
Current assignee: Fundacion para la Investigacion Medica Aplicada
Priority date: 2023-05-12
Filing date: 2024-05-10
Publication date: 2024-11-21
Anticipated expiration: 2025-11-12

Abstract

The present disclosure provides T-cell receptors binding to a HLA-A2-GPC3(522-530) complex. Also provided are TCR constructs comprising said TCR and a fusion molecule, for example, a single chain antibody fragment. The disclosure also provides polynucleotides and expression vectors encoding for said TCRs or constructs, as well as host cells comprising those. The TCRs, constructs, polynucleotides and vectors are for use in the therapy, in particular, for use in cancer therapy.

Description

Novel T Cell Receptors

This application claims the benefit of European Patent Application EP23382444.0 filed on 12.05.2023.

Technical Field

The present invention relates to the field of cancer immunotherapy, in particular, it relates to gly pican-3- specific T-cell receptors.

Background Art

Hepatocellular carcinoma (HCC) is the primary form of liver cancer and ranks as one of the most prevalent types of cancer. Unfortunately, its prognosis is exceedingly poor, and the mortality rate is alarmingly high. The chief factor behind the dismal prognosis is the limited options for treating progressive HCC. Currently, only symptomatic treatments like local excision and multikinase inhibitor sorafenib administration are available for such patients. Sorafenib, in particular, shows a low response rate in elderly patients and is often accompanied by adverse effects. This highlights the urgent need for developing novel treatment methods that minimize side effects and improve the survival rates of patients with progressive HCC.

Immunotherapy is a growing treatment approach for HCC, owing to its potential benefits. Glypican-3 (GPC3) is a cell surface glycophosphatidylinositol-anchored protein belonging to the heparan sulfate family of proteoglycans, which plays an important role in cell growth, differentiation, and migration. GPC3 is highly expressed in hepatocellular carcinoma (HCC) and other tumors, but rarely or not expressed in non-malignant adult tissues, except in the placenta. GPC3 has been proposed as an attractive target for cancer immunotherapy. Recently, chimeric antigen receptors (CARs) specific for GPC3 have been designed and clinical trials using GPC3-targeted CAR-T cells in HCC patients are currently underway. The GPC3-targeted CAR-T have been reported to be safe, with manageable toxicity, but the response rate is far from satisfactory. Apart from the well-known drawbacks of CAR-T cells in the treatment of solid tumors, inherent properties of GPC3 impairs the efficacy of GPC3 CAR-T cell therapy. In particular, GPC3 is cleaved in vivo by sheddase and released from the cell surface, being found in soluble form locally in the tumor and in the circulation. Shed GPC3 competes with surface GPC3 for CAR binding and this, together with loss of surface GPC3 expression, blunts CAR-T cell efficacy.

There is a need in the art to develop effective immunotherapeutic strategies for the treatment of HCC.

Summary of Invention

The present inventors have obtained highly-tumor reactive murine GPC3-specific T-cell receptors (TCRs) useful in the treatment of HCC and other types of tumors.

Thus, a first aspect of the disclosure provides a T-cell receptor binding to GPC3(522-530) epitope, said epitope comprising or consisting of the sequence SEQ ID NO: 2, or to an HLA-A2-GPC3(522-530) epitope complex. Within the obtained tumor reactive GPC3-specific TCRs, some of them also bound to HLA-A2- GPC3(522-532) complex, ie, an 11-mer comprising the minimal GPC3(522-530) epitope. Said larger 11-mer epitope has SEQ ID NO: 100. Thus, the first aspect of the disclosure encompasses a T-cell receptor binding to GPC3(522-532) epitope, said epitope consisting of the sequence SEQ ID NO: 100, or to an HLA-A2- GPC3(522-532) epitope complex.

The TCRs of the first aspect can be used to redirect human T cells against GPC3+ tumors, in particular GPC3+ HLA-A2+ tumors. This can be done by genetically engineering human T cells to express the identified GPC3 TCRs, for example using retroviruses (as in the example below) or lentiviruses that express the genes for the TCRo and TCRp chains of the TCRs. These T lymphocytes can be used as an alternative to GPC3- specific CAR-T cells. Intrinsic characteristics of TCRs allow to overcome some of the deficiencies and limitations of CAR-T cells. One important advantage is that, unlike CARs, TCRs are not affected by shedding of surface tumor-associated antigens (TAA) or competition with soluble TAA. Additionally, as compared with CARs, TCR-T cells exhibit a less differentiated phenotype with lower expression of coinhibitory molecules, and a greater ability to expand under high antigenic pressure and differentiate into memory T cells, all features that enable transferred T cells to mount a more efficient tumor growth control.

As shown in the examples below, the herein disclosed TCRs are specific for the human HLA-A2 restricted immunodominant epitope GPC3(522-530), some are specific for the GPC3(522-532) epitope, present in all GPC3 isoforms. Moreover, human T cells engineered to express the disclosed GPC3-specific murine TCRs recognize GPC3(522-530)-peptide pulsed HLA-A2+ cells and recognized HLA-A2+GPC3+ tumor cells in vitro. Interestingly, human T cells expressing the herein disclosed TCRs were able to recognize all tested GPC3+ HLA-A2+ tumor cells, including those expressing very low levels of GPC3 or low levels of HLA-A2, showing strong effector functions. More importantly, the T cells engineered with TCRs of the present disclosure demonstrated great efficiency in vivo, being able to eradicate GPC3+HLA-A2+ hepatocellular carcinoma (HCC) tumor xenograft in mouse models. Interestingly, as shown in example 2, T cells engineered to express a TCR according to the disclosure outperformed a GPC3-specific CAR-T equivalent to one currently being tested in clinical trials (NCT05003895).

In a second aspect, the present disclosure refers to a functional fragment of the T-cell receptor of the first aspect. Said functional fragment maintains the ability of the original TCR from which it derives to recognise the GPC3(522-530) antigen or HLA-A2-GPC3(522-530) and/or the GPC3(522-532) antigen or HLA-A2- GPC3(522-532) complex.

Another way to redirect human T cells to GPC3+ tumors is to use soluble bispecific molecules that incorporate an anti-CD3 effector function and a TCR of the first aspect, a functional fragment thereof as defined in the second aspect or an affinity-enhanced version of any of these. Unlike genetically modified T-cell therapies, bispecific therapies are "off-the-shelf" therapies, can redirect any T cell from the patient to the tumor, and can be given in repeated doses.

In a third aspect, the disclosure thus refers to a T-cell receptor construct comprising: (1) a TCR as defined in the first aspect or functional fragment thereof as defined in the second aspect, and (2) at least one fusion component. The fusion component can be selected from the group consisting of Fc receptors and/or Fc domains, cytokines, such as IL-2 or IL-15, toxins, an antibody or a single chain antibody fragment (scFv), CD3-zeta chain(s) and/or other TCR stimulation domains, such as the intracellular CD28, CD137 or CD134 domain, and wherein the T-cell receptor (1) is bound to the at least one fusion component (2). The TCR contained in this construct is preferably devoid of its transmembrane domain to facilitate solubility of the construct.

As outlined above, T cells, in particular, human T cells, may be genetically engineering to express the disclosed GPC3-specific TCRs, e.g. using retroviruses (as in the example below) or lentiviruses that express the genes for the TCRo and TCRp chains of the TCRs.

A fourth aspect thus refers to a polynucleotide encoding for a TCR of the first aspect, or a functional fragment thereof as defined in the second aspect, or a TCR construct as defined in the third aspect. A fifth aspect refers to an expression vector comprising said polynucleotide. A sixth aspect further refers to a cell, in particular a T cell, comprising a TCR as defined in the first aspect, or a functional fragment thereof as defined in the second aspect, or a TCR construct as defined in third aspect, or a polynucleotide as defined in the fourth aspect, or an expression vector as defined in the fifth aspect.

A seventh aspect refers to a method for producing a cell according to the sixth aspect, the method comprising a step of introducing into the cell a polynucleotide as defined in the fourth aspect or an expression vector as defined in the fifth aspect.

An eight aspect discloses a method for producing a TCR as defined in the first aspect, or a functional fragment thereof as defined in the second aspect, or a TCR construct as defined in third aspect, said method comprising the steps of: (a) culturing a cell according to the sixth aspect under conditions suitable for producing the TCR, functional fragment thereof, or TCR construct, thereby obtaining a culture containing said TCR, functional fragment thereof, or TCR construct; and (b) isolating or recovering said TCR, functional fragment thereof, or TCR construct from the culture.

A ninth aspect refers to a pharmaceutical composition comprising a TCR as defined in the first aspect, or a functional fragment thereof as defined in the second aspect, or a TCR construct as defined in third aspect, or a polynucleotide as defined in the fourth aspect, or an expression vector as defined in the fifth aspect, or a cell as defined in the sixth aspect, together with pharmaceutically acceptable excipients and/or carriers.

A tenth aspect refers to a TCR as defined in the first aspect, or a functional fragment thereof as defined in the second aspect, or a TCR construct as defined in third aspect, or a polynucleotide as defined in the fourth aspect, or an expression vector as defined in the fifth aspect, a cell as defined in the sixth aspect, or a pharmaceutical composition as defined in the ninth aspect, for use as a medicament. This aspect may be reworded as use of a TCR as defined in the first aspect, or a functional fragment thereof as defined in the second aspect, or a TCR construct as defined in third aspect, or a polynucleotide as defined in the fourth aspect, or an expression vector as defined in the fifth aspect, or a cell as defined in the sixth aspect, or a pharmaceutical composition as defined in the ninth aspect, for the preparation of a medicament. Also disclosed is a method of treatment which comprises administering to a subject in need thereof a TCR as defined in the first aspect, or a functional fragment thereof as defined in the second aspect, or a TCR construct as defined in third aspect, or a polynucleotide as defined in the fourth aspect, or an expression vector as defined in the fifth aspect, a cell as defined in the sixth aspect, or a pharmaceutical composition as defined in the ninth aspect. An eleventh aspect refers to a TCR as defined in the first aspect, or a functional fragment thereof as defined in the second aspect, or a TCR construct as defined in third aspect, or a polynucleotide as defined in the fourth aspect, or an expression vector as defined in the fifth aspect, a cell as defined in the sixth aspect, or a pharmaceutical composition as defined in the ninth aspect, for use in treating cancer. This aspect may be reworded as use of a TCR as defined in the first aspect, or a functional fragment thereof as defined in the second aspect, or a TCR construct as defined in third aspect, or a polynucleotide as defined in the fourth aspect, or an expression vector as defined in the fifth aspect, a cell as defined in the sixth aspect, or a pharmaceutical composition as defined in the ninth aspect, for the preparation of a medicament for treating cancer. Also disclosed is a method of treating cancer which comprises administering to a subject in need thereof a TCR as defined in the first aspect, or a functional fragment thereof as defined in the second aspect, or a TCR construct as defined in third aspect, or a polynucleotide as defined in the fourth aspect, or an expression vector as defined in the fifth aspect, a cell as defined in the sixth aspect, or a pharmaceutical composition as defined in the ninth aspect.

A twelfth aspect refers to a method for detecting the presence of cancer in a subject, the method comprising: (I) administering a T-cell receptor fragment according to the second aspect or a construct as defined in the third aspect, and (ii) determining the presence of cells bound to the T-cell receptor construct or fragment, wherein the presence of cells bound to the T-cell receptor construct or fragment is indicative of the presence of cancer.

Brief Description of Drawings

Figure 1. Characterization of the T cell response induced in HHD-DR1 mice against hGPC3. (A) Schematic procedure to identify the relevant GPC3 T cell epitopes. (B) Graphs show compiled data from 1^st and 2^nd screening. Percentage of I FNY⁺ cells in CD8 and CD4 T lymphocytes stimulated with the peptide mixes and the individual peptides from Mix 7. The dotted line indicates the threshold to consider a positive response (3 times the % IFNy⁺ cells in the naive mouse). One experiment representative of three.

Figure 2. Single-cell sorting and expansion of GPC3-specific T cell clones for TCR sequencing. (A) Experimental design to identify murine HLA-A2-restricted TCR specific for pGPC3(522-530). On day 8 (d8) after immunization, splenocytes from ADV-GPC3-immunized HHD-DR1 mice were stimulated with pGPC3(522-530) (as in figure 1) to check for the presence of specific T cells (B), cells were labeled with the Mouse I FNy Secretion Assay kit. CD8⁺ and highly I FNy⁺ (I FNy^hi) cells were isolated as single cells in 96-well U plates and expanded to establish clones. Two experiments were performed. Finally, twenty-four clones were established, which were used to prepare the TCR library and for TCR sequencing. (B) Percentage of CD8 I FNy* cells in ADV-GPC3-immunized HHD-DR1 mice from experiment 1 and 2. Each point represents a mouse. The dotted line indicates the % I FNy* cells in a naive mouse. (C) Staining for isolation of pGPC3(522- 530)-specific T cells. The dot plots show I FNy-producing CD8⁺ cells after stimulation without (w/o) or with peptide pGPC3(522-530). The gate shows the I FNy^hi cells that were sorted.

Figure 3. Retroviral vector used to transduce human T cells. (A) Overview of TCR acceptor cassette and Vo and Vp plasmids. TCR acceptor cassette. The murine (m) TCRo and TCRp2 constant regions (mTCR-Co and mTCR-Cp2, respectively) were modified to insert an additional interchain disulfide bond (white circle) to improve the expression and biological activity of transgenic TCRs. Both constant chains were joined by a linker consisting of a furin cleavage site (RAKR), a SGSG spacer and "self-cleaving” peptide (P2A). The mTCR-Ca and mTCR-Cp DNA sequences were flanked in their 5' ends by two Bbsl sites (triangles) separated by a short unstable stuffer and four base overhangs including conserved coding sites to enable Vo and Vp region fusion with the respective constant regions. Vo and VP plasmids. For each TCR, the sequences of Vo and Vp fragments were extracted from the sequencing data analyzed by MIXCR and flanked with the Bbsl enzyme cleavage sites (triangles) in both 5' and 3' DNA segments. (B) TCR reconstruction by Golden gating assembly and retroviral production. For each of the final TCR constructs (indicated as x), a mixture containing T4 ligase buffer, MSCV-Mu(*C) Acceptor plasmid, pUC57-TCRx-Vo, pUC57-TCRx-Vp, Bbsl and T4 ligase was incubated as indicated. After digestion of unligated linear DNA, the cloning products were transformed in bacteria for plasmid isolation. After verification by Sanger, the vectors were used for retrovirus production.

Figure 4. TCR expression on human T cells and study of the dependence of the CD8 co-receptor in the recognition of the peptide/MHC complex. (A) Transduction of human T cells with retrovirus encoding GPC3-specific TCRs. Transduction efficiency was assessed by measuring murine TCRp (mTCRp) surface expression by FACS analysis using an anti-mouse TCRp mAb. Dot plots are gated in CD4 (Up) or CD8 (Down) T cells. The numbers refer to the percentage of mTCRp⁺ cells in CD4 and CD8 T cells at day 4 upon retroviral transduction. (B) Staining with pGPC3(522-530)/HLA-A2 tetramer. T cells were stained with pGPC3(522-530) /HLA-A2 tetramer, anti-human CD4 and CD8 mAbs and anti-mouse TCRp mAb, and analyzed by FACS. Dot plots are gated in mTCRp⁺ CD4 (Up) or mTCRp⁺CD8 (Down) T cells. The numbers refer to the percentage of Tetramer cells in mTCRp⁺ T cells at day 5 upon retroviral transduction. (C) Coculture with peptide-pulsed T2 cells. The bulk of genetically modified T cells, including CD4 and CD8 T cells, were co-cultured with T2 cells previously pulsed with a saturating concentration of pGPC3(522-530) or with medium. Twenty hours later, cells were recovered and stained with anti-human CD4, CD8 and CD137 mAbs and anti-mouse TCRp mAb and analyzed by FACS. The graphs show the response measured as the percentage of cells expressing surface CD137 within transduced (mTCRp⁺) CD4 or CD8 T cells. Data are show as mean + SD (2 replicates per condition). One experiment representative of two.

Figure 5. Analysis of GPC3 and HLA-A2 expression in liver tumor-derived cell lines. (A) The histograms show the surface expression of HLA-A2 and total HLA-I (HLA-A/B/C) molecules as assessed by FACS. (B) Relative expression of GPC3 vs HLA-A2 molecules for the different tumor cell lines used in this study. GPC3 and HLA-A2 expression is shown as CT and Median Fluorescence intensity (MFI), respectively. SKHEP1- GP'^nt account for SKHEP1-GPC3'^nt. Vertical and horizontal dotted lines shown the lowest positive value for GPC3 and HLA-A2 expression detected in our tumor cell line collection. The gray arrows indicate the position (open circle) on the graph of each cell type.

Figure 6. TCR-engineered human CD8 T cells recognize GPC3⁺HLA-A2⁺ tumor cells. Co-culture of TCR- T cells and GPC3^+/ HLA^^ tumor cell lines. TCR-T cells and target cells were co-cultured (37 °C) and twenty-four hours later cells were harvested and stained to assess CD 137 expression by FACS. Heat map shows the average percentage of CD137⁺ cells within mTCRp⁺ CD8 T cells stimulated with GPC3 ^/-HLA-A2’^,/- tumor cells. A compilation of seven experiments is shown. SKHEP1-GP'^nt account for SKHEP1-GPC3'^nt.

Figure 7. TCR-4 T cells recognize HLA-A2-engineered tumor cell lines naturally expressing low levels of GPC3. (A and B). Co-culture of TCR-T cells and GPC3⁺HLA-A2^+/~ COS7 and A431 tumor cell lines. COS7 and A431 tumor cells (both negative for HLA-A*02:01 allele) were genetically modified cells to express HLA- A*02:01 allele. TCR-4 and TCR-5 T cells were co-cultured with HLA-A2^+A C0S7 or A431 cells. As positive control, HLA-A2^+A C0S7 or A431 cells were previously pulsed with pGPC3(522-530) (Pept) at the concentration indicated and, after extensive washing, they were also co-cultured with TCR-4 and TCR-5 T cells. The graphs in panels A and B show the percentage of CD137⁺ cells within mTCRp⁺ CD8 T cells. One representative experiment from two. Data are shown as Mean ± SD (2 replicates per condition).

Figure 8. Effect of GPC3 silencing using shRNA on the recognition of tumor cell lines by GPC3 TCR-T cells. (A) GPC3 silencing using shRNA and quantification of GPC3 mRNA levels by qRT-PCR. GPC3 expression was silenced by transduction of tumor cells with lentivirus encoding shRNA complementary to GPC3 (shGPC3-630 and shGPC3-1344) or renilla (shREN), as negative control. Relative levels of GPC3 mRNA were compared with Histone 3 (H3F3A) using the 2^AAACt method. The graph shows GPC3 mRNA levels in shGPC3-630 and shGPC3-1344 treated cells normalized to the shREN condition for each cell line. (B) GPC3 protein levels in GPC3-silensed tumor cell lines. Total GPC3 protein levels were assessed in shGPC3-1344 and shREN treated cells on day 26 of drug selection by staining with anti-human GPC3 mAb and FACS analysis. The numbers represent the median fluorescence intensity of each condition. The percentages indicate the relative decrease in GPC3 mRNA levels in cells treated with shGPC3-1344 with respect to their counterparts treated with shREN. (C) Recognition of GPC3-silenced tumor cell lines by GPC3 TCR-T cells. TCR-3, 4 and 5-T cells were cocultured with the indicated GPC3-silenced tumor cell lines and their respective shREN-treated control cells. The response was evaluated by measuring the production of I FNy by ELISA. As a negative control, PLCPRF5-derived cells were used for TCR-3 and TCR-4, and HEP3B- A2-derived cells for TCR-5. Panel C is one experiment representative of two experiments. Data are shown as Mean ± SD (2 replicates per condition). SKHEP1-GP'^nt account for SKHEP1-GPC3'^nt.

Figure 9. HLA-I restriction of TCR-3, -4 and -5. GPC3⁺HLA-A2^+A tumor cells were incubated with anti (a)- pan-HLA-l, a-HLA-A2 or control IgG mAbs before co-culture with TCR-T cells. The response was evaluated by measuring surface CD137 expression in mTCRp⁺ cells by FACS (up panels) and the production of I FNy (low panels) by ELISA. One experiment representative of three experiments is shown. Data are depicted as Mean ± SD (2 replicates per condition). SKHEP1-GP'^nt account for SKHEP1-GPC3'^nt.

Figure 10. TCR-4-engineered T cells stood out for their high proliferative capacity in response to HEPG2 and PLCPRF5-A2 cells. (A) Tumor cell titration assay. TCR-3-, -4- and -5-T cells were co-cultured with serial dilutions of target cells. The response was evaluated by measuring surface CD137 expression in mTCRp⁺ cells by FACS. (B) Proliferating assay. CellTrace Violet (CTV)-labelled TCR-3-, -4-, and -5-T cells were cultured alone (medium) or with irradiated HEPG2, PLCPRF5-A2 or PLCPRF5 cells in the presence or not of anti (a)-pan-HLA-l mAbs. Ninety-six hours later, cells were collected and surface stained with anti-CD8 and anti-mTCRp mAb and analyzed by FACS. The graph shows the percentage of proliferating T cells (CTV^|OW) within mTCRp⁺ cells. (A and B) One experiment representative of two experiments is shown. (A and B) Data are depicted as Mean ± SD (2 replicates per condition).

Figure 11. Cytokine production by GPC3-specific TCR-T cells upon recognition of GPC3⁺ tumor cells. GPC3⁺HLA-A2^+A tumor cells were incubated with anti-pan-HLA-l or anti (a)-HLA-A2 mAbs or with medium before co-culture with TCR-T (or UTD) cells. Twenty hours later, supernatants were recovered and the response was evaluated by measuring the levels of I FNy, IL-2 and TNFa by ELISA. The graphs show the cytokine levels in the culture supernatant. Data are depicted as Mean ± SD (three replicates per condition). One experiment representative of two experiments.

Figure 12. Degranulation and cytotoxic capacity of TCR-3-, -4- and -5-T cells. (A) GzmB ELISPOT.

GPC3⁺HLA-A2^+/- tumor cells were incubated with anti-pan-HLA-l mAbs or with medium and then co-cultured with TCR-T (or UTD) cells in ELISPOT plates coated with anti-human GzmB capture mAb. As a negative control, TCR-T cells were incubated alone (medium). On the following day, cells were washed from wells and the ELISPOT plate was revealed. The graph shows the number of GzmB-producing cells (spots) per 5 x 10⁴ mTCRb⁺ cells seeded. (B and C) Real time killing cell assay using xCELLigence Real-Time Cell Analysis. GPC3⁺HLA-A2^+/- tumor cells were cultured alone (medium) or with TCR-T (or UTD) cells in the presence of a- pan-HLA-l, O-HLA-A2 mAbs or medium in a xCELLigence E-plate 16-well flat-bottom plates. Impedance measurements were recorded each 15-minute intervals for up to 50 hours. The cell index curve was normalized to the time point at which T cells were added using RTCA Software Pro. Normalized cell index data for each TCRs is represented. (A) Data are depicted as Mean ± SD (two replicates per condition). (A-C) one experiment representative of two experiments.

Figure 13. TCR-4 T cells efficiently control tumor progression in vivo. (A-D) Adoptive transfer of TCR- engineered T cells. (A) Schematic draw of the experimental approach. 6-8-week-old gender-matched NSG mice were subcutaneously (sc) implanted with PLCPRF5-A2 tumor cells. On day 9, mice were adoptively transferred with TCR-4 CD8 T cells containing 8 x 10⁶ mTCRp⁺ cells injected intravenously (i.v.) (N=7 mice/group). Control groups received a number of UTD CD8 T cells similar to the total number of CD8 T cells in the TCR-treated groups from the same donor (N=6 mice/group). Mice received human IL-2 on days 1 , 2, 3, 5, 7, 9, 11 , 13 and 15 upon ACT. (B and C) Tumor size from individual mice (B) and average tumor size (C). (D) Overall survival. Data are represented as mean as mean±SEM (0). Statistical significance was determined using nonlinear regression (curve fit) (0), and Mantel-Cox test (D). ***p < 0.0005. One experiment representative of 2 is shown.

Figure 14. TCR-4 may recognize a different epitope configuration that it is present in the 11-mer GPC3(522-532) peptide. (A and C) Peptide-induced HLA stabilization assay. Surface expression of HLA-A2 in T2 (HLA-A2⁺, TAP deficient) cells pulsed with different peptides containing the minimal HLA-A2-restricted GPC3(522-530) epitope. T2 cells were cultured (26°C, 18h) with decreasing equimolar concentrations of 11 - mer FLAELAYDLDV and 10-mer RFLAELAYDL peptides (A) or the 20-mer peptide (FLAELAYDLDVDDAPGNSQQ) (C). As control, the 9-mer GPC3(522-530) peptide (FLAELAYDL) was used. Then, cells were staining with anti-HLA-A2 mAb and the MFI was analyzed by FACS. Mean ± SD for each concentration (2 replicates per condition). (B and D) TOR- recognition of p/HLA-A2 complex. T2 cells were pulsed with decreasing equimolar concentrations of the indicated peptides and then TCR-engineered CD8 T cells were added. To properly compare the different TCRs, the percentage of CD8 TCR⁺ T cells in the different TCR-T cell lines was equalized by adding non-transduced CD8 T cells. The response was evaluated by measuring I FNy secretion. Nonlin fit curve (thick line) and EC50 are represented. Data are depicted as: Mean ± SD (two replicates per condition). One experiment representative of two experiments.

Figure 15. Recognition of pGPC3(522-530)/HLA-C4 complex by TCR-5 T cells. (A) 721-04 and 721-A2 cells were generated by retrovirally transducing 721.221 cells to express HLA-C*04:01/|32m or HLA- A*02:01/|32m, respectively. Subsequently, these cells were used to generate 721-C4-ICP47 and 721 -A2- ICP47 cells, respectively, by retroviral transduction to express and the TAP inhibitor ICP47. Cells were stained with ant-HLA-A/B/C antibody to check the surface expression of the respective HLA molecule. (B) Peptide- induced HLA stabilization assay. 721-C4-ICP47 and 721 -A2-ICP47 cells were pulsed (26°C, 18h) with decreasing concentration of pGPC3(522-530), or vehicle (DMSO). Cells were stained with anti-HLA-A/B/C mAb and the MFI of HLA was assessed by FACS. Mean ± SD for each concentration (2 replicates per condition). (C) TCR- recognition of pGPC3(522-530)/HLA-C4. 721 -C4-ICP47 and 721 -A2-ICP47 cells were pulsed with (or without) 1 piM of pGPC3(522-530) and then TCR-engineered CD8 T cells were added. To properly compare the different TCRs, the percentage of CD8 TCR⁺ T cells in the different TCR-T cell lines was equalized by adding non-transduced CD8 T cells. The response was evaluated by measuring I FNy secretion. Data are depicted as: Mean ± SD (two (B) or three (C) replicates per condition). One experiment representative of two experiments.

Figure 16. TCR-4 T cells outperformed CAR-T cells in ACT schedules. (A) GPC3-CAR-engineered CD8 T cells were stained with anti-EGFR(t) and GPC3-Fc/anti-Fc mAb and analyzed by FACS. (B) CAR- and TCR- engineered CD8 T cells were stained with anti-EGFR(t) and GPC3-Fc/anti-Fc mAb or with anti-mouse TCRp mAb, respectively, and sorted as EGFR(t)⁺GPC3-Fc⁺ or mTCRp⁺, respectively, and finally expanded. Levels of CAR (stained with GPC3-Fc/anti-Fc) and mTCRp before and after sorting are shown. (C) Surface GPC3 protein levels were assessed by staining with anti-human GPC3 mAb (clone YP7) or Isotype mAb (Isot) and FACS analysis. (D) TCR-4, GPC3-CAR or UTD T cells (5x10⁴ cells/well) were co-cultured (24h) with serial dilutions of target cells (from 1 :1 to 1 :16 ratio). Cells were then harvested and stained to evaluate surface CD137 in CD8 T cells by FACS. (E and F) Real time killing cell assay. HEPG2 were cultured into xCELLigence E-plate. After 18h, T cells were added and impedance measurements were recorded for up to 19h. As control, medium were also added to tumor cells. (E) Normalized cell index data. (F) Percentage of cytotoxicity at 4h and 16h from the addition of T cells (G and H) 12-day HEPG2 tumor bearing mice were treated with TCR-4, GPC3-CAR or UTD CD8 T cells (8x10⁶ cells) together with IL-2. (G) Average tumor size (mm²). The inset is a zoomed-in image of the grey area on the main graph. (H) Overall survival. Data are depicted as: Mean ± SEM in G (tumor growth, 7 mice); Mean ± SD in D and F(two replicates per condition). Statistical significance was determined using: nonlinear regression (curve fit) in G (tumor growth) and the Mantel-Cox test for the overall survival (H). ****p < 0.0001, ***p < 0.0005, **p < 0.005, *p < 0.05. One experiment representative of two experiments.

Figure 17. TCR-4-T cells exhibited lower level of exhaustion and enhanced response after transfer, as compared to CAR-T cells. (A) Blood analysis. On days 2, 6 and 12, blood samples were collected and the absolute number of human CD8 T cells in 50 uL of blood was assessed by volumetric flow cytometry. (B and C) On day 13, mice were sacrificed and the tumor cell suspension was stained and analyzed by FACS. (B) Graph showing the percentage of human CD8 T cells within total living cells in the tumor. The graph shows the percentage relative to UTD group. (C) Graphs showing the percentage of GzmB⁺ cells (left), Ki-67 MFI (middel) and PD-1 MFI (right) in CD8 TILs. (D) PLCPRF5-A2-tumor bearing mice were treated with TCR-4-T cells. At day 102 of ACT, cured mice were sacrificed and the presence of human T cells was analyzed in blood and bone marrow. Representative dot plots and graph showing the percentage of human CD8 T cells in the blood and in the bone marrow (BM). Data are depicted as: Mean ± SEM in A (4-5 mice) and Median and individual mice (B, C and D). Statistical significance was determined using: two-way ANOVA. ***p < 0.0005, **p < 0.005, *p < 0.05. One experiment representative of two experiments. Detailed description of the invention

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as is commonly understood by one with skill in the art to which this invention belongs at the time of filling. However, in the event of any latent ambiguity, definitions provided herein take precedent over any dictionary or extrinsic definition.

As used herein, the indefinite articles "a” and "an” are synonymous with "at least one” or "one or more.” Unless indicated otherwise, definite articles used herein, such as "the” also include the plural of the noun.

T cell receptors

The present disclosure refers to a T-cell receptor which binds to GPC3(522-530) antigen, said antigen consisting of the sequence SEQ ID NO: 2 (FLAELAYDL), or bind to HLA-A2-restricted GPC3(522-530), i.e. the complex formed by HLA-A2-and GPC3(522-530).

According to the present disclosure, "T cell receptor” or "TCR” means a receptor constituted of a dimer of two TCR chains (a chain, p chain). Each chain of the TCR is a member of the immunoglobulin superfamily and possesses one N-terminal immunoglobulin (Ig)-variable (V) domain, one Ig-constant (C) domain, a transmembrane/cell membrane-spanning region, and a short cytoplasmic tail at the C-terminal end. The variable domain of both the TCR a chain and p chain have three hypervariable or complementarity determining regions (CDRs) interspersed with four regions that are more conserved, termed framework regions (FRs). The TCR recognises a short antigen that is presented by the major histocompatibility complex (MHC)Zhuman leucocyte antigen (HLA) and activates the T cell through signal transduction, that is, a series of biochemical events mediated by associated enzymes, co-receptors, specialized adaptor molecules, and activated or released transcription factors.

The TCR of the present disclosure encompasses not only one in which the a chain and the p chain of the TCR constitute a heterodimer but also one in which they constitute a homodimer. Furthermore, the TCR of the present invention also encompasses a fragment lacking a part of or whole constant region, for example, one lacking the transmembrane domain of the a and p chains but retaining its functionality. The disclosure also encompasses TCRs which (i) contain one or more conservative or non-conservative amino acid residues substitutions with respect to the TCRs herein disclosed, or (ii) contain a substituent group or amino acid analogue in one or more amino acid residues; or (iii) conjugates formed by fusing any of the above and another compound (said conjugates will be explained to more detail below). All these variants also retain the functionality of the TCR regarding antigen recognition and binding. These TCRs are herein termed as "functional TCR fragments” or "functional TCR variants”. Throughout the disclosure, the term "TCR” encompasses "functional TCR fragment”.

Functional variant forms of the herein disclosed TCRs preferably comprise the same CDR regions and may include, but are not limited to, deletion, insertions and/or substitutions of one or several (for example, 1-30, 1- 20, or 1-10) amino acids, and addition of one or several (generally less than 20, or less than 10, or less than 5) amino acids at C-terminus and/or N-terminus. For example, in the art, the substitution of amino acids (natural or non-natural amino acids) with analogical or similar properties usually does not alter the function of the protein. In another example, addition of one or several amino acids at the C-terminus and/or N-terminus usually does not change the function of the protein. The contemplated variations also include polypeptides having a substituent group in one or more amino acid residues, as well as homologous sequences, conservative variants, allelic variants, natural mutants, and induced mutants.

In the present disclosure, "a conservative variant" refers to the polypeptides in which there are amino acids substituted by amino acids having analogical or similar properties, compared to the amino acid sequence of the TCR of the present invention. Usually, substituted amino acids are up to 10, up to 8, up to 5, or most usually up to 3. These conservative variant polypeptides may be produced according to the amino acid substitutions in Table 1.

The TCRs obtained by the present inventors are murine. Even though no side effects are expected due to this

Table 1

fact, sometimes it is convenient to "humanise” a murine TCR. Thus, in one embodiment of the first aspect, the TCR is humanized. In another embodiment, the TCR is recombinant. In another embodiment, the TCR is isolated. The term "humanized” TCR refers to a TCR whose peptide sequence has been modified to increase its similarity to variants produced naturally in humans. This can be done replacing one or more amino acid residues in the amino acid sequence of the murine TCR sequence (and in particular in the framework sequences) by one or more of the amino acid residues that occur at the corresponding position(s) in a conventional TCR from a human being. The term "recombinant” refers to such molecules created, expressed, isolated or obtained by technologies or methods known in the art as recombinant DNA technology which include, e.g., DNA splicing and transgenic expression. The term includes TCRs expressed in a non-human mammal (including transgenic non-human mammals, e.g., transgenic mice), or a cell (e.g., CHO cells) expression system, or a non-human cell expression system (e.g., yeast, bacteria, insect), or isolated from a recombinant combinatorial human antibody library. The term "isolated” refers to a substance that is at least partially free of other biological molecules from the cells or cell culture from which they are produced, for example, at least partially free of expression system components such as biological molecules from a host cell or of the growth medium thereof, or at least partially free of cell components of the organism (e.g., animal) from which they are derived. Generally, the term "isolated” is not intended to refer to a complete absence of such biological molecules or to an absence of water, buffers, or salts or to components of a pharmaceutical formulation that includes the referred substances.

"GPC3(522-530)” or "HLA-A2-restricted GPC3(522-530)" or, herein simply "GPC3”, is herein understood as a peptide fragment of gly pican-3 (GPC3) comprising or, preferably, consisting of the amino acid sequence shown in SEQ ID NO: 2. In a particular embodiment, the TCRs of the disclosure specifically recognize and are capable of binding to a complex of GPC3(522-530) and HLA-A2. "GPC3(522-532)” or "HLA-A2-restricted GPC3(522-532)" is herein understood as a peptide fragment of gly pican-3 (GPC3) consisting of the amino acid sequence shown in SEQ ID NO: 100. In a particular embodiment, the TCRs of the disclosure specifically recognize and are capable of binding to a complex of GPC3(522-532) and HLA-A2

It can be confirmed by any known method that the TCRs of the present invention specifically recognize and are capable of binding to the above-mentioned HLA-A2-GPC3(522-530), sometimes HLA-A2-GPC3(522- 532), complex. Suitable methods are immunoassays, including, for example, tetramer assay, activationdependent molecule expression (CD137, IFNy, and other cytokines), etc., using GPC3(522-530) or GPC3(522-532) antigen pulsed on HLA-A2. By performing these immunoassays, it can be confirmed that T cells expressing the TCR on the cell surface specifically recognize GPC3(522-530) and/or, in particular cases, GPC3(522-532) antigen when it was pulsed on HLA-A2.

There are three parameters to measure the strength of a TCR: two of them are physical (affinity and avidity) and the third one is functional (functional avidity).

The TCR affinity refers to the physical strength of the monomeric interaction between the TCR and a pMHC- complex. TCR affinity is evaluated by the dissociation constant (KD), that requires surface plasmon resonance . The biochemical determination of KD values is rather complicated, and it requires the availability of soluble pMHC-complexes and a soluble form of the TCR. Moreover, it needs to be considered that the binding kinetics can significantly vary depending on whether the interaction is measured with soluble or membrane-bound ligands.

TCR avidity: a more practical but less precise way to assess the strength of pMHC-TCR interaction is to stain living T cells with pMHC-multimers. Binding kinetics can then be determined by measuring fluorescent intensity of cell-surface bound multimers. These measurements are described by the term avidity, which is normally used to refer to the strength of multimeric receptor-ligand engagement. The TCR functional avidity is determined by ex vivo quantification of biological functions such as IFN-y production, cytotoxic activity (ability to lyse target cells), or proliferation. The concentration needed to induce a half-maximum response (EC50) is often used to describe the functional avidity of T cells. The functional avidity of a T-cell clone is primarily impacted by (a) the affinity of the TCR for the pMHC-complex, that is, the strength of the interaction between the TCR and pMHC, (b) expression levels of the TCR and the formation of clusters that comprise several TCRs, and (c) the distribution and composition of signaling molecules, as well as expression levels of molecules that attenuate T-cell function and TCR signaling.

In contrast to the physical parameters affinity and avidity, the functional avidity describes how well a T-cell responds to antigens. Though all of the three parameters correlate in most cases, that is, high-affinity T cells often have a high functional avidity, this does not need to be the case. There are several factors besides the antigen recognition ability of the TCR that can impact the T-cell response. In principle, T cells could express a high-affinity TCR, but due to other factors, for example, inhibitory molecules, it might show a very weak response to antigen stimulation. Thus, determining the functional avidity is not only often more practical, but is also the only one out of the three parameters that actually describes the functional outcome of the stimulation.

The TCRs of the present disclosure have a very high avidity for the GPC3(522-530) antigen and/or, in particular cases, for the GPC3(522-532) antigen. In one embodiment, the avidity of the TCR for the GPC3(522-530) or GPC3(522-532) antigen, as expressed by EC50 for pGPC3(522-530)/HLA-A2 or pGPC3(522-532)/HLA-A2 complex binding (described as the complex dilution at which a half-maximal number of complex+ cells is reached), is in the range from 0.5 x 10³to 10 x 10³, in particular from 1. x 10³ to 6 x 10- ³, more in particular, from 4.0 x 10’³ to 6 x 10³, more particularly around 5x 10³, for example 4.9x 10³, 5x 10’³, 5.1 x 10’³, or 5.2x 10’³. The avidity of the TCR for the antigen can be determined by any known method. In particular, by staining TCR-transduced T cells with decreasing concentration of HLA-A2-GPC3(522-530) or GPC3(522-532)/HLA-A2 tetramers conjugated with a fluorochrome and measuring the percentage of tetramer labelled T cells. The EC50 was calculated with the model log(agonist) vs. response (three parameters) of Graphpad 8.0.1.

The TCRs of the present disclosure also have a very high functional avidity for the GPC3(522-530) and/or GPC3(522-532) antigen. The functional avidity of the TCR for the antigen can be determined by any known method. In particular, the functional avidity may be assessed by culturing TCR-expressing T cells with T2 (HLA-A2+ TAP-) cells pulsed with serial dilutions of pGPC3(522-530) and measuring surface CD137 expression and the production of I FNy, IL-2 and TNF-o, preferably I FNy. The EC50 was calculated with the model log(agonist) vs. response (three parameters) of Graphpad 8.0.1. In one embodiment, the functional avidity of the TCR for the GPC3(522-530) antigen, as expressed by EC50 peptide concentration (representing the peptide dose at which a half-maximal I FNy response is reached), is equal or below 1 .20 nM, in particular equal or below 1 .05 nM, more in particular equal or below 0.9 nM, even more in particular equal or below 0.85. In another embodiment, the functional avidity is equal or above 0.02 mM, in particular equal or above 0.03 mM, more in particular equal or above 0.5 mM, even more in particular equal or above 0.7 mM. In some embodiments, the functional avidity of the TCR for the GPC3(522-530) is in the range from 0.020 to 1 .20 nM, in particular from 0.030 to 1 .05 nM, more in particular from 0.035 to 0.9 nM, more in particular from 0.05 to 0.9 mM, even more in particular, from 0.77 to 0.85 nM. In one embodiment, T cells engineered with the T-cell receptor have a functional avidity for the GPC3(522-530)-HLA-A2 complex, depicted as EC50, equal or below 1 .20 nM, wherein the functional avidity is defined as the degree of activation of TCR-engineered T cells, measured as IFNy production, following interaction of the TCR-engineered T cell with the exogenous GPC3(522-530) peptide presented by the surface HLA-2 molecule, and the EC50 is the GPC3(522-530) peptide concentration required to induce a half-maximal IFNy response. In another embodiment, T cells engineered with the T-cell receptor have a functional avidity for the GPC3(522-532)-HLA-A2 complex, depicted as EC50, equal or below 1.20 nM, wherein the functional avidity is defined as the degree of activation of TCR-engineered T cells, measured as IFNy production, following interaction of the TCR- engineered T cell with the exogenous GPC3(522-532) peptide presented by the surface HLA-2 molecule, and the EC50 is the GPC3(522-532) peptide concentration required to induce a half-maximal IFNy response. The functional avidity of T cells engineered with the TCR for GPC3(522-530)/HLA-A2 complex may be preferably measured in a standard IFNy production assay using as target cells HLA-A2+ T2 cells pulsed with graded amounts of exogenous GPC3(522-530) peptide, for example, the functional avidity may be measured as described by Hillerdal et al (2016).

In particular embodiments, the above-mentioned avidity and functional avidity correspond to that of the presently disclosed TCRs when expressed in the surface of a cell, in particular, a T cell, more in particular, a CDS T cell.

In an embodiment, the TCRs of the first aspect, or cells expressing the same, recognize GPC3⁺ HLA-A2⁺ tumor cells (GPC3 endogenously expressed by HLA-A2⁺ tumor cells), for example HEPG2, PLCPRF5-A2, HEP3B-A2, HUH7-A2, COS7-A2 and/or A431-A2 cells, more in particular HEPG2 and/or PLCPRF5-A2 cells, more in particular HEPG2 and PLCPRF5-A2. In another embodiment, the TCRs of the first aspect, or cells expressing the same, do not recognize tumor cells not expressing GPC3.

GPC3 expression may be quantified by assessing GPC3 RNA levels from total RNA (200 ng/sample) by qRT- PCR and calculating the Ct values. Ct refers to Cycle threshold and is defined as the number of cycles required for the fluorescent signal to cross the threshold. Ct levels are inversely proportional to the amount of target nucleic acid in the sample. Cts < 25 are strong positive reactions indicative of abundant target nucleic acid in the sample. Cts of 25-30 are positive reactions indicative of moderate amounts of target nucleic acid. Cts > 35 are weak reactions indicative of minimal amounts of target nucleic acid, which could represent environmental contamination. Surface HLA-A2 expression may be quantified by assessing median fluorescence intensity (MFI) of HLA-A2 molecules by flow cytometry using anti-HLA2 mAb (clone BB7.2) conjugated to phycoerythrin. The classification of cells according to their HLA-A2 MFI may be stablished as follows: high levels of HLA-A2 (MFI in the range of 1-2 10⁴), Intermedium levels of HLA-A2 (MFI in the range of 3-7 10³), low levels of HLA-A2 (MFI in the range of 0.2-1 10³) and null expression of HLA-A2 (MFI <100).

In a particular embodiment, the TCRs, or cells expressing the same, are able to recognize GPC3+HLA-A2+ tumor cells with CT values of GPC3 RNA as determined by qRT-PCR equal or below 35, in particular in the range from 16 to 36, more particularly, in the range of 26 to 30. In another very particular embodiment, the TCRs of the first aspect, or cells expressing the same, recognize GPC3+HLA-A2+ tumor cells expressing HLA-A2 levels with MFI in the range of 0.2-2 10⁴, more particularly, in the range of 0.2-1 10³. In a more particular embodiment, the TCRs, or cells expressing the same, are able to recognize GPC3+HLA-A2+ tumor cells with CT values of GPC3 RNA as determined by qRT-PCR equal or below 35, in particular in the range from 16 to 36, more particularly, in the range of 26 to 30 and expressing HLA-A2 levels with MFI in the range of 0.2-2 10⁴, more particularly, in the range of 0.2-1 10³. In one embodiment, the a chain complementary determining regions (CDRs) of the TCR comprises: as CDR1 the SEQ.ID.NO 3, as CDR2 the SEQ.ID.NO 4, and as CDR3 a sequence selected from the group consisting of SEQ.ID.NO 7, SEQ.ID.NO 8, and SEQ.ID.NO 9.

In one embodiment, the p chain complementary determining regions (CDRs) of the TCR comprise: as CDR1 the SEQ.ID.NO 5, as CDR2 the SEQ.ID.NO 6, and as CDR3 a sequence selected from the group consisting of SEQ.ID.NO 10, and SEQ.ID.NO 11.

In one embodiment, the TCR comprises a chain complementary determining regions (CDRs) selected from the group consisting of:

- SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 7 [CDR3],

- SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 8 [CDR3], and

- SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 9 [CDR3],

In another embodiment, the TCR comprises p chain CDRs selected from the group consisting of:

- SEQ ID NO: 5 [CDR1], SEQ ID NO: 6 [CDR2], and SEQ ID NO: 10 [CDR3], and

- SEQ ID NO: 5 [CDR1], SEQ ID NO: 6 [CDR2], and SEQ ID NO: 11 [CDR3],

In one embodiment, the TCR comprises a chain CDRs consisting of SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 8 [CDR3], and p chain CDRs consisting of SEQ ID NO: 5 [CDR1], SEQ ID NO: 6 [CDR2], and SEQ ID NO: 10 [CDR3],

In another embodiment, the TCR comprises a chain CDRs consisting of SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 9 [CDR3], and p chain CDRs consisting of SEQ ID NO: 5 [CDR1], SEQ ID NO: 6 [CDR2], and SEQ ID NO: 11 [CDR3],

In another embodiment, the T- cell receptor comprises a chain CDRs consisting of SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 7 [CDR3], and p chain CDRs consisting of SEQ ID NO: 5 [CDR1], SEQ ID NO: 6 [CDR2], and SEQ ID NO: 10 [CDR3],

In particular embodiments, the TCR comprises an a chain variable region comprising or consisting of a sequence selected from SEQ ID NO: 12, SEQ ID NO: 13, and SEQ ID NO: 14, or, alternatively, selected from SEQ ID NO: 17, SEQ ID NO: 18, and SEQ ID NO: 19.

In particular embodiments, the TCR comprises a p chain variable region comprising or consisting of a sequence selected from SEQ ID NO: 15, and SEQ ID NO: 16, or alternatively, selected from SEQ ID NO: 20, and SEQ ID NO: 21.

In particular embodiments, the TCR comprises a chain variable region comprising or consisting of SEQ ID NO: 13 or SEQ ID NO: 18, and p chain variable region comprising or consisting of SEQ ID NO: 15 or SEQ ID NO: 20. In particular embodiments, TCR comprises a chain variable region comprising or consisting of SEQ ID NO: 14 or SEQ ID NO: 19, and p chain variable region comprising or consisting of SEQ ID NO: 16 or SEQ ID NO: 21 . In particular embodiments, TCR comprises a chain variable region comprising or consisting of SEQ ID NO: 12 or SEQ ID NO: 17, and p chain variable region comprising or consisting of SEQ ID NO: 15 or SEQ ID NO: 20.

In particular embodiments, the a and p chains of the TOR are preceded at N-terminus by a signal peptide. Thus, each of SEQ ID NO: 17, SEQ ID NO: 18 and SEQ ID NO: 19 comprise the sequence of the signal peptide of the Mus musculus T-cell receptor a chain, while each of SEQ ID NO: 20 and SEQ ID NO: 21 comprise the sequence of the signal peptide of the Mus musculus T-cell receptor p chain. The skilled person will easily appreciate that, in particular embodiments, a and p chains of the TOR can be preceded at N- terminus by other alternative signal peptide sequence used for constructing the TOR of the invention. In other embodiments, the a and p chains of the TOR do not comprise a signal peptide, such as in SEQ ID NO: 12, SEQ ID NO: 13, or SEQ ID NO: 14 (for a chains) and SEQ ID NO: 15 or SEQ ID NO: 16 (for p chains).

In some embodiments, the TOR comprises an a and/or p chain variable region(s) having a sequence identity of at least 90% with respect to any one of a or p chain variable regions sequences defined above. In some embodiments said sequence identity is preferably at least 91%, or at least 92%, or at least 93%, or at least 94%, more preferably, at least 95%, or at least 96%, or at least 97%, or at least 98% or at least 99%. As will be apparent to the skilled person, the sequence variability preferably or solely affects the framework regions of the TCR, while the CDRs are preferably conserved.

In the present disclosure the term "identity” refers to the percentage of residues that are identical in the two sequences when the sequences are optimally aligned. If, in the optimal alignment, a position in a first sequence is occupied by the same amino acid residue as the corresponding position in the second sequence, the sequences exhibit identity with respect to that position. The percentage of identity determines the number of identical residues over a defined length in a given alignment. Thus, the level of identity between two sequences or ("percent sequence identity”) is measured as a ratio of the number of identical positions shared by the sequences with respect to the number of positions compared (i.e., percent sequence identity = (number of identical positions/total number of positions compared) x 100). A gap, i.e., a position in an alignment where a residue is present in one sequence but not in the other, is regarded as a position with nonidentical residues and is counted as a compared position.

A number of mathematical algorithms for rapidly obtaining the optimal alignment and calculating identity between two or more sequences are known and incorporated into a number of available software programs. For purposes of the present invention, the sequence identity between two amino acid sequences is preferably determined using algorithms based on global alignment, such as the Needleman-Wunsch algorithm (Needleman and Wunsch, J. Mol. Biol. 48: 443-453, 1970. DOI: 10.1016/0022-2836(70)90057-4), preferably implemented in the Needle program of the EMBOSS package (EMBOSS: The European Molecular Biology Open Software Suite, Rice et al., Trends Genet. 16: 276-277, 2000. DOI: 10.1016/s0168-9525(00)02024-2); or the BLAST Global Alignment tool (Altschul et al., "Basic local alignment search tool”, 1990, J. Mol. Biol, v. 215, pages 403-410, 1990. DOI: 10.1016/S0022-2836(05)80360-2), using default settings. Local alignment also can be used when the sequences being compared are substantially the same length.

In a very particular embodiment, the TCR comprises a chain variable region comprising SEQ ID NO: 13 and p chain variable region comprising SEQ ID NO: 15. In some embodiments, the constant region of the a chain comprises or consists of SEQ ID NO: 22 or a sequence having at least 90% identity, in particular at least 95% identity, more in particular at least 98% identity, with SEQ ID NO: 22. In some embodiments the constant region of the p chain comprises or consists of SEQ ID NO: 23 or a sequence having at least 90% identity, in particular at least 95% identity, more in particular at least 98% identity, with SEQ ID NO: 23. The a and p chains of the TCR are preferably linked to each other, usually through a disulfide bridge between the a and p constant regions. To this end, SEQ ID NO: 22 and SEQ ID NO: 23, which derive from the a and p murine constant region, contain a mutated cysteine, sited in position 49 of SEQ ID NO: 22 and position 57 of SEQ ID NO: 23.

In a very particular embodiment, the TCR comprises or consists of:

- a chain variable region comprising SEQ ID NO: 13,

- a chain constant region comprising SEQ ID NO: 22 or a sequence having at least 98% identity with SEQ ID NO: 22,

- p chain variable region comprising SEQ ID NO: 15, and

- p chain constant region comprising SEQ ID NO: 23 or a sequence having at least 98% identity with SEQ ID NO: 23.

In another very particular embodiment the TCR comprises or consists of:

- a chain variable region comprising SEQ ID NO: 14,

- p chain variable region comprising SEQ ID NO: 16, and

In another embodiment the, TCR comprises or consists of:

- a chain variable region comprising SEQ ID NO: 12,

- p chain variable region comprising SEQ ID NO: 15, and

In most embodiments the TCR is a murine TCR. In other embodiments, the TCR is humanized. For example, the TCR may comprise human constant domains. In some embodiments, the TCR is chimeric. In another embodiment, the TCR is a recombinant protein. Such recombinant proteins are expressed, preferably in E.coli or mammalian cells, and then purified before further use or application. Alternatively, as will be further described above, the TCR can be encoded by an expression vector and engineered to be expressed in the surface of a T cell.

Functional TCR fragments and constructs

As mentioned above, the present disclosure encompasses functional fragments of the TCRs of the first aspect. By "functional fragment” it is understood that the fragment maintains the functional features of the original TCR, in particular regarding its ability to recognise and bind GPC3(522-530) and/or GPC3(522-532) antigen, or to the HLA-A2-GPC3(522-530) and/or HLA-A2-GPC3(522-532) complex. Preferably, the fragment retains the functional avidity for the GPC3(522-530)-HLA-A2 and/or GPC3(522-532)-HLA-A2 complex as defined above. Moreover, the functional fragment comprises a and p chain CDRs as defined above for the first aspect. In some embodiments, the fragment comprises a and p chain variable regions as defined for the first aspect. All embodiments described above for the CDRs and variable regions of the TCRs of the first aspect also apply to the functional fragments of this second aspect.

The functional fragments may also comprise the a and p chain constant regions as defined for the first aspect. In a particular embodiment, the functional fragments lack the transmembrane domain of the a and p chain. In particular embodiments, the functional fragment is soluble. The transmembrane domains of the a and p chains as illustrated by SEQ ID NO: 22 and SEQ ID NO: 23 are SVMGLRILLLKVAGFNLLMTL (SEQ ID NO: 38) for a chain and ILYEILLGKATLYAVLVSGLVLMAMV (SEQ ID NO: 39) for the p chain. In some embodiments the functional fragment is bound to a label, such as radionuclides, gold (particles), fluorophores (such as fluorescein), which are preferably covalently attached/coupled to the TCR fragment. Throughout the description, the term "TOR” of the disclosure encompasses "functional TCR fragments thereof”.

As outlined above, in a third aspect the present disclosure provides TCR constructs comprising a and p chains as defined for the first aspect and a fusion component. Preferably, the construct retains the functional avidity for the GPC3(522-530)-HLA-A2 and/or GPO3(522-532)-HLA-A2 complex as defined above. The construct contains the CDRs as defined for the first aspect. In some embodiments, the construct comprises a and p chain variable regions as defined for the first aspect. All embodiments described above for the CDRs and variable regions of the TCRs of the first aspect also apply to the construct of this third aspect.

The construct may also comprise the a and p chain constant regions as defined for the first aspect. In a preferred embodiment, the constructs lack the transmembrane domain of the a and p chain. In particular embodiments, the construct is soluble. The soluble TCR constructs are suitable to be directly used for the detection and destruction of liver cancer cells, such as HCC cells, without the need of being expressed on T- cells. In one embodiment, the TCR construct is a bispecific TCR. The transmembrane domains of the a and p chains as illustrated by SEQ ID NO: 22 and SEQ ID NO: 23 are SVMGLRILLLKVAGFNLLMTL (SEQ ID NO: 38) for a chain and ILYEILLGKATLYAVLVSGLVLMAMV (SEQ ID NO: 39) for the p chain.

The a and p chains of the construct are preferably linked to each other to form heterodimers or multimers. Moreover, at least one of the a or p chain is bound to the fusion component. The at least one of the a or p chain may be bound to the fusion component through a linker. As used herein, "linkers" are peptides of 1 to 50 amino acids length and are typically chosen or designed to be unstructured and flexible. These include, but are not limited to, synthetic peptides rich in Gly, Ser, Thr, Gin, Glu or further amino acids that are frequently associated with unstructured regions in natural proteins. An appropriate linker is, for example, a flexible glycine-serine linker. Both the a and p chains can be linked to the fusion component. Moreover, the conjugate may comprise more than one fusion component. In one embodiment, the fusion component is an effector molecule. "Effector molecule” as used herein, refers to a molecule that selectively binds to a protein and regulates its biological activity. In one embodiment, the fusion molecule is a cytokine, the cytokine may be selected from the group consisting of IL-12, IL-2, IL-15, IL-18, IL-21, IL-33, IL-7, IFN-gamma, IFN-alpha, and IFN-p, in particular, IL-2 or IL-15. In another embodiment, the fusion component is a Fc receptor or Fc domain. In another embodiment, the fusion component is a toxin. In another embodiment, the fusion component is an antibody or a single chain antibody fragment (scFv). In another embodiment, the fusion component is a CD3-zeta chain or other TCR stimulation domain, such as the intracellular CD28, CD137 or CD134 domain. In a particular embodiment the fusion component is a scFv, for example, selected from anti- CD3, anti-CD28, anti-CD5, anti-CD 16 and anti-CD56.

In a particular embodiment, the construct comprises a purification tag, for example, a His-tag.

Polynucleotides, expression vectors and recombinant technology

The TCRs, functional fragments thereof, and TCR constructs disclosed herein may be administered directly, i.e. in protein form, or as polynucleotides encoding for said polypeptides, wherein the polypeptides are expressed in vitro or in vivo by recipient cells. This means that the polynucleotides are delivered into the cells, for instance, by intratumoral administration of DNA or RNA vectors encoding for the TCRs, functional fragments thereof or TCR constructs, whereby the host cells produce TCRs, functional fragments thereof or TCR constructs in situ. For example, an expression vector comprising a TCR of the invention may be delivered to T cells, whereby the T cells express the TCR on their surface.

The fourth aspect thus refers to a polynucleotide encoding for a TCR, functional fragments thereof, or TCR construct as defined above. All embodiments defined above for the TCRs, functional fragments thereof, or TCR constructs also apply to the fourth aspect. The term "polynucleotide encoding for a TCR, functional fragments thereof, or TCR construct includes a polynucleotide that encodes for said TCR, functional fragments thereof or TCR constructs, and may also contain additional coding and/or non-coding sequences. Polynucleotides of the fourth aspect may be in the form of DNA or RNA. DNA forms include cDNA, genomic DNA, or synthetic DNA. DNA can be single-stranded or double-stranded. DNA can be a coding strand or a non-coding strand.

The polynucleotides according to the present disclosure may comprise monomers other than deoxyadenosine 3'-monophosphate, deoxyguanosine 3'-monophosphate, deoxycytidine 3'-monophosphate, deoxythymidine 3'-monophosphate, adenosine 3'-monophosphate, guanosine 3'-monophosphate, cytidine 3'-monophosphate, or uridine 3'-monophosphate, but are functionally and structurally similar thereto. These are also referred to as oligonucleotide analogues and are well known to the skilled person. Such oligonucleotides may be naturally- occurring or not and are sometimes preferred over native forms because of properties such as, for example, enhanced binding ability, enhanced cellular uptake, reduced immunogenicity, and increased stability in the presence of nucleases. The polynucleotides of the present disclosure may also comprise well known modified oligonucleotides, such as 2'-O-methylation.

In one embodiment, the polynucleotide comprises a sequence encoding for the a chain variable region selected from the group consisting of: SEQ ID NO: 24, SEQ ID NO: 26, and SEQ ID NO: 27. In another embodiment, the polynucleotide comprises a sequence encoding for the p chain variable region selected from the group consisting of: SEQ ID NO: 25, and SEQ ID NO: 28.

In one embodiment, the polynucleotide comprises SEQ ID NO: 26 (encoding for the a chain variable region) and SEQ ID NO: 25 (encoding for the p chain variable region). In another embodiment, the polynucleotide comprises SEQ ID NO: 27 (encoding for the a chain variable region) and SEQ ID NO: 28 (encoding for the p chain variable region). In another embodiment, the polynucleotide comprises SEQ ID NO: 24 (encoding for the a chain variable region) and SEQ ID NO: 25 (encoding for the p chain variable region).

In one embodiment, the polynucleotide comprises SEQ ID NO: 29 (encoding for the a chain constant region). In another embodiment, the polynucleotide comprises SEQ ID NO: 30 (encoding for the p chain constant region).

In some embodiments, the polynucleotide of the fourth aspect comprises a sequence having a sequence identity of at least 90% with respect to a nucleotide sequence disclosed above. In some embodiments said sequence identity is preferably at least 95%, or at least 96%, or at least 97%, or at least 98% or at least 99%. As will be apparent to the skilled person, the sequence variability may be greater in the sequence encoding for the constant region than that encoding the variable region and, within the variable region, preferably affects the sequences encoding for the framework regions of the TCR. Silent nucleotide mutations (that result in no change of the encoded amino acid) are contemplated. Moreover, codon optimization may be performed on the basis of the above sequences to improve polypeptide production.

The full-length nucleotide sequence encoding for the TCR of the first aspect, the functional fragment of the second aspect or the construct of the third aspect can generally be obtained by PCR amplification method or/and recombination method. Once the polynucleotide sequence has been obtained, the concerned sequences can be produced in large scale using recombinant methods. Usually, sequences can be obtained by cloning into a vector, transferring it into cells, and then isolating the sequences from the proliferated host cells by conventional methods. The polynucleotide sequence encoding for the TCR, functional fragment thereof or construct of the present disclosure can also be obtained by chemical synthesis.

The polynucleotide sequence then can be introduced into various existing or synthetic DNA molecules (e.g. vectors) and cells known in the art. The invention therefore also relates, in a fifth aspect, to vectors comprising the above-mentioned polynucleotides of the fourth aspect. In particular embodiments of this fifth aspect, the vector is an expression vector also containing a suitable promoter. The vector of the present disclosure may contain a transcription and translation regulatory sequence, a ribosome binding site, an enhancer, a replication origin, a polyA addition signal, and/or a selection marker gene. Non-limiting examples of the selection marker gene include dihydrofolate reductase gene, neomycin resistance gene, puromycin resistance gene and the like.

Moreover, the vectors of the fifth aspect can comprise additional sequences encoding for other proteins or peptides of interest, such as a cell surface marker, purifications tags, or helper peptides. In one embodiment, the expression vector further comprises a sequence encoding for truncated human epidermal growth factor receptor polypeptide (EGFRt). The expression of EGFRt successfully provides a cell surface marker for tracking adoptively transferred transduced cells and also acts as a suicide gene. Recognition of EGFRt by the clinically available chimeric anti-EGFR monoclonal antibody, cetuximab (Erbitux™), mediates antibodydependent cellular cytotoxicity (ADCC) leading to in vivo elimination of TCR-T cells.

The expression vector that can be used in the present disclosure is not particularly limited, so long as it can express TCR for a sufficient period of time for preventing or treating a disease when introduced into a cell. Examples thereof include viral vector, plasmid vector and the like. Non-limitative appropriate expression vectors in the sense of the present disclosure may be alphavirus vector, adenovirus vector, adeno-associated virus vector, herpes virus vector, lentivirus vector, retrovirus vector, poxvirus vector, and Newcastle disease virus vector. In particular embodiments, the viral vector is replication-defective. A transposon expression system (PiggyBac system) may also be used. As the plasmid vector, animal cell expression plasmid (e.g., pa1-11, pXT1, pRc/CMV, pRc/RSV, pcDNAI/Neo) and the like can be mentioned. The vectors can be used to express the TCR, functional fragment or construct in vitro or in vivo. In a particular embodiment, the expression vector is a lentivirus or retrovirus vector. Although both viral systems have traditionally been used to modify human T lymphocytes (retroviruses being the most commonly used to express TCR), lentiviral vectors are less genotoxic than gammaretroviral vectors and are becoming the vector of choice in clinical trials.

In one embodiment of the present disclosure, heterodimers of an a chain and a p chain of the TCR can be constructed in a target cell, for example, on the cell surface, by introducing an expression vector containing a nucleic acid encoding the a chain and a nucleic acid encoding the p chain of the TCR into the target cell. In this case, the nucleic acid encoding the a chain of the TCR and the nucleic acid encoding p chain of the TCR may be incorporated into separate expression vectors or a single expression vector. When they are incorporated into a single expression vector, these two kinds of nucleic acids are preferably incorporated via a sequence enabling polycistronic expression. Using a sequence enabling polycistronic expression, plural genes incorporated in one kind of expression vector can be more efficiently expressed.

Host cells

The sixth aspect of the disclosure refers to a cell (host cell) comprising a TCR as defined in the first aspect, or a functional fragment as defined in the second aspect, or a TCR construct as defined in third aspect, or a polynucleotide as defined in the fourth aspect, or an expression vector as defined in the fifth aspect. All embodiments defined above for the TCRs, functional fragments thereof, TCR constructs, polynucleotides, or expression vectors, also apply to the cells of the sixth aspect.

Delivery of the polynucleotides or expression vectors into the host cell may be carried out by conventional transforming methods, e.g. by electroporation, microinjection, liposome packaging, calcium phosphate coprecipitation, etc. The present disclosure also encompasses introducing the polynucleotide of the present invention into the genome of the host cell by genome editing (for example, CRISPR system, TALEN system. An embodiment of the sixth aspect refers to a cell into which both a polynucleotide encoding for the TCR a chain and a polynucleotide encoding for the TCR |3 chain are introduced. In another embodiment, a polynucleotide encoding for the TCR construct is introduced into the cell.

In one embodiment, the host cell expresses the TCR on its surface. In these embodiments, the cell may be a lymphocyte. "Lymphocyte" is one of the subtypes of leukocytes in the immune system of vertebrata. Examples of lymphocytes include T cells, B cells, and natural killer cells (NK cells). Since a TCR plays an important role in recognizing a T cell antigen, in particular embodiments of the disclosure the cell of the sixth aspect is a T cell. "T cells" are known as one of the important types of lymphocytes and are distinguished from other lymphocytes by the presence of a TCR on their cell surface. Examples of T cells appropriate for the present invention include cytotoxic T lymphocytes which are CD8 positive cells, helper T cells which are CD4 positive cells, regulatory T cells, and effector T cells. In particular embodiments, the cell of the sixth aspect is a CD8 T cell (expressing the herein disclosed TCR). In some particular embodiments, the cell is a chimeric cell comprising a human cell, in particular a human T cell, more in particular, a human CD8 T cell, and a murine TCR. The T cell of the present disclosure (e.g., cytotoxic T cell) usually has, in addition to the TCR gene inherently present in the T cell, an exogenous TCR gene derived from the polynucleotide or the vector of the present invention. On this point, the cell of the present invention is different from the cells harvested from the living body.

T cells expressing the TCR of the present disclosure can be obtained by introducing the polynucleotide or the vector of the present invention into T cells collected from a living body. Preferably, T cells expressing the TCR of the present disclosure (namely, T cells derived from the progenitor cells) can be obtained by inducing from lymphocyte progenitor cells (e.g., pluripotent stem cells) into which the polynucleotide or the vector of the present disclosure has been introduced. The disclosure also contemplates gene therapy whereby the herein disclosed polynucleotide or vector is delivered to a living body and the encoded TCR, functional fragment thereof or TCR construct is expressed in the cells in vivo. In particular embodiments the T cells expressing a TCR of the present disclosure are transgenic T-cells.

The aforementioned lymphocytes can be collected from, for example, peripheral blood, bone marrow and cord blood of a human or a non-human mammal. When a cell transfected with the TCR gene of the present disclosure is used for the treatment of diseases such as cancer, the cell population is preferably collected from the subject to be treated or from a donor matched with the HLA type of the treatment target. Preferably, the subject or donor is a human.

In some embodiments, when the polynucleotide or the vector of the present disclosure is introduced into a cell and TCR is present on the cell surface, the cell recognizes GPC3(522-530)+HLA-A2+ and/or GPC3(522- 532)+HLA-A2+ tumor cells. In particular embodiments, the cell has HLA-A2 restricted GPC3(522-530) and/or GPC3(522-532) specific cytotoxic activity against the target cell. In some embodiments, the cell expressing TCR or functional fragment thereof of the present disclosure may also recognize GPC3(522-530)+HLA-A2- tumor cells and may additionally have non-HLA-A2 restricted GPC3(522-530) specific cytotoxic activity against the target cell. Whether the cell of the present invention has a cytotoxic activity can be confirmed by a known method, for example, measurement of a cytotoxic activity on HLA-A2 positive target cells, for example, by Real time killing cell assay using xCELLigence Real-Time Cell Analysis Instrument (ACEA Biosciences).

In one embodiment, the vectors of the fifth aspect or the host cells of the sixth aspect, comprising polynucleotides encoding for a TCR, functional fragment thereof or construct as defined above, may also be used to recombinantly produce the TCR, functional fragment thereof or construct in vitro, which may then be delivered to tumoral cells as a recombinant product. A further aspect thus provides a method for producing TCRs, functional fragments thereof or TCR constructs as defined above, said method comprising the steps of: (a) culturing a host cell comprising an expression vector as defined above under conditions suitable for producing the TCRs, functional fragments thereof or TCR constructs, thereby obtaining a culture containing said TCRs, functional fragments thereof or TCR constructs; and (b) isolating or recovering said TCRs, functional fragments thereof or TCR constructs from said culture.

In general, the host cell for expressing the TCRs, functional fragments thereof, or TCR constructs in vitro can be a prokaryotic cell, such as a bacterial cell; or a lower eukaryotic cell, such as a yeast cell; or a higher eukaryotic cell, such as a mammalian cell. Non-limiting examples are: bacterial cells such as Escherichia coli, Streptomyces or Salmonella typhimurium, fungal cells such as yeast, insect cells, such as Drosophila S2 or Sf9, animal cells, such as BHK, CHO, Vero, Vero E6, COS7, 293 cells, and the like. In one embodiment, the host cell is E. coli. In another embodiment, the host cell is selected from Saccharomyces cerevisiae and Pichia pastoris. In another particular embodiment, the host cell is a mammal cell, for example selected from BHK, CHO-K1, Vero, and Vero E6, more particularly a BHK cell, such as BHK-21.

The transformation of the host cell with the recombinant DNA can be performed using conventional techniques well known to those skilled in the art. When the host is a prokaryotic organism such as E. coli, competent cells capable of absorbing DNA can be harvested after the exponential growth phase and treated with the CaCb method. Another method is to use MgCh. If necessary, conversion can also be performed by electroporation. When the host is eukaryotic, transfection can be done by electroporation, microinjection, liposome packaging, calcium phosphate coprecipitation, and the like. The obtained transformants can be cultured in a conventional manner to express the polypeptide encoded by the gene of the present invention. As already mentioned above, depending on the host cells used, the medium used in the culture may be selected from various conventional media and the culture is performed under conditions suitable for the host cells growth. After the host cells are grown to an appropriate cell density, the selected promoter may be induced by a suitable method (such as temperature shift or chemical induction) and the cells are incubated for a further period of time.

The recombinant polypeptide in the above method may be expressed intracellularly, or on the cell membrane, or secreted extracellularly. If necessary, the recombinant protein can be isolated and purified by various separation methods by utilizing its physical, chemical and other characteristics. These methods are well- known to those skilled in the art. Examples of these methods include, but are not limited to: conventional renaturation treatment, treatment with a protein precipitation agent (salting out method), centrifugation, osmotic disruption, ultracentrifugation, molecular sieve chromatography (gel filtration), adsorption layer analysis, ion exchange chromatography, high performance liquid chromatography (HPLC), and various other liquid chromatography techniques and combinations thereof.

The present disclosure also contemplates recombinant TCRs or TCR constructs obtainable or obtained by the above methods.

Compositions

The ninth aspect of the disclosure refers to a pharmaceutical composition comprising a TCR, functional fragment thereof, TCR construct, polynucleotide, expression vector, or cell as defined above, together with pharmaceutically acceptable excipients and/or carriers. All embodiments described above for the TCR, functional fragment thereof or TCR construct, polynucleotide, expression vector, or cell, also apply to the pharmaceutical composition. The expression "pharmaceutically acceptable excipients or carriers" refers to pharmaceutically acceptable materials, compositions or vehicles. Each component must be pharmaceutically acceptable in the sense of being compatible with the other ingredients of the pharmaceutical composition. It must also be suitable for use in contact with the tissue or organ of humans and animals without excessive toxicity, irritation, allergic response, immunogenicity or other problems or complications commensurate with a reasonable benefit/risk ratio.

The election of the pharmaceutical formulation will depend upon the nature of the active compound and its route of administration. Any route of administration may be used. In some embodiments, the route of administration is parenteral, and the composition is then appropriate for parenteral administration. In a particular embodiment, the route of administration is by injection. In a more particular embodiment, the route of administration is systemic, for example, by intramuscular, intravenous, intraarterial, intraperitoneal, subcutaneous, or transdermal injection. In a particular embodiment, the route of administration is local, for example, intratumoral injection. Topical administration is also contemplated, such that the pharmaceutical composition may be a topical composition.

The pharmaceutical compositions may be in any form, including, among others, tablets, pellets, capsules, aqueous or oily solutions, suspensions, emulsions, aerosols, or dry powdered forms suitable for reconstitution with water or other suitable liquid medium before use, for immediate or retarded release.

The appropriate excipients and/or carriers, and their amounts, can readily be determined by those skilled in the art according to the type of formulation being prepared. Examples of suitable pharmaceutically acceptable excipients are solvents, dispersion media, diluents, or other liquid vehicles, dispersion or suspension aids, surface active agents, isotonic agents, thickening or emulsifying agents, preservatives, solid binders, lubricants and the like. Except insofar as any conventional excipient medium is incompatible with a substance or its derivatives, such as by producing any undesirable biological effect or otherwise interacting in a deleterious manner with any other component(s) of the pharmaceutical composition, its use is contemplated to be within the scope of this invention.

The term "carrier" is to be understood as a pharmaceutically acceptable vehicle. The carrier can be organic, inorganic, or both. Suitable carriers well known to those of skill in the art and include, without limitation, large, slowly metabolized macromolecules such as proteins, polysaccharides, polylactic acids, polyglycolic acids, polymeric amino acids, amino acid copolymers, lipid aggregates (such as oil droplets or liposomes) and inactive virus particles. Carriers may also include, saline, buffer, dextrose, water, glycerol, ethanol, and the combinations thereof. In particular embodiments, the carries may be a polycationic polymer, a vesicle, a liposome, or a nanoparticle.

Usually, the pharmaceutical composition comprises a therapeutically effective amount of the TCR, functional fragment, TCR construct, polynucleotide, expression vector, or cell. The expression "therapeutically effective amount" as used herein, refers to the amount of a compound that, when administered, is sufficient to prevent development of, or alleviate to some extent, one or more of the symptoms of the disease which is addressed. The particular dose of compound administered according to this disclosure will of course be determined by the particular circumstances surrounding the case, including the compound administered, the route of administration, the particular condition being treated, and the similar considerations. In addition, the TCR, functional fragment thereof, TCR construct, polynucleotide, expression vector, or cell of the present disclosure may also be used with other therapeutic agents.

The disclosure also contemplates diagnostic compositions comprising a TCR, functional fragment thereof or TCR construct, as defined above. Herein disclosed is also a kit of parts that comprises:

(a) the TCR, functional fragment thereof or TCR construct, as defined above, optionally together with pharmaceutically acceptable excipients or carriers;

(b) optionally a further therapeutic agent; and

(c) optionally, instructions for its use.

Herein disclosed is also a vessel or injection device which comprises the TCR, functional fragment thereof or TCR construct, as defined above, preferably together with pharmaceutically acceptable excipients or carriers.

The disclosure also contemplates a kit of parts that comprises:

(a) the TCR, functional fragment thereof or TCR construct as defined above;

(b) optionally further means for performing a diagnosis (e.g. buffers, reagents, controls); and

(c) optionally, instructions for its use.

Uses

The disclosure provides, in the tenth aspect, a TCR, functional fragment thereof, TCR construct, polynucleotide, expression vector, or cell, pharmaceutical composition, or kit, all of them as defined above, for use in therapy. All embodiments described above for the TCR, functional fragment thereof, TCR construct, polynucleotide, expression vector, cell, pharmaceutical composition, or kit, also apply to the therapeutical uses. In a particular embodiment, the therapy is immunotherapy. In another embodiment, the therapy comprises mounting effector responses against tumor cells, in particular against GPC3+HLA-A2+ and/or GPC3+HLA-C4+ tumor cells. In another embodiment, the therapy comprises a cytotoxic effect against tumor cells, in particular against GPC3+HLA-A2+ and/or GPC3+HLA-C4+ tumor cells. The invention also contemplates the TCR, functional fragment thereof, TCR construct, polynucleotide, expression vector, or cell, or pharmaceutical composition, all of them as defined above, for use as a cytotoxic agent, in particular against tumor cells, more particularly against GPC3+HLA-A2+ and/or GPC3+HLA-C4+ cancer cells. In particular embodiments, the tumor cells are GPC3+HLA-A2+.

An eleventh aspect provides a TCR, functional fragment thereof, TCR construct, polynucleotide, expression vector, or cell, pharmaceutical composition, or kit, all of them as defined above, for use in treating cancer. In other words, the eleventh aspect provides TCR, TCR construct, polynucleotide, expression vector, or cell, pharmaceutical composition, or kit, all of them as defined above, for use in treating a tumor. As used herein the terms "cancer” or "tumor” are used interchangeably and refer as generally understood in the art to a malignant abnormal cell growth with the potential to invade or spread to other parts of the body. In another embodiment, the therapy is cancer therapy. In another embodiment, the therapy is cancer immunotherapy.

"Treating cancer” in the sense of the present disclosure includes a prophylactic treatment before the clinical onset of cancer or a therapeutic treatment after the clinical onset of cancer and may be achieved by arresting the development or reversing the symptoms of cancer. In some embodiments, the disclosure contemplates treating metastatic cancer, as well as treating refractory cancer which has not responded to conventional treatments. In a particular embodiment of the eleventh aspect, the TCR, functional fragment thereof, TCR construct, polynucleotide, expression vector, or cell, pharmaceutical composition, or kit, all of them as defined above, are for use in combination therapy for the treatment of cancer, for example, in combination with surgery, radiation or a further therapeutic agent. The TCR, functional fragment thereof, TCR construct, polynucleotide, expression vector, or cell, pharmaceutical composition, and the further therapeutic agent may be administered sequentially, simultaneously or within a therapeutic interval. In a particular embodiment the combination therapy comprises surgery. In another particular embodiment the combination therapy comprises radiation. In a particular embodiment the combination therapy comprises an antitumoral agent. The antitumoral agent may be another cytotoxic agent, such as, for example, alkylating agents, antimetabolites, including folate antagonists, purine and pyrimidine analogues, antibiotics and other natural products, including anthracyclines and vinca alkaloids, and antibodies, which improve specificity. The antitumoral agent may also be a hormonal agent or a signal transduction inhibitor or a checkpoint inhibitor. In a particular embodiment the combination therapy comprises chemotherapy. The combination therapy may also comprise a different immunotherapy. For example, in a particular embodiment the combination may comprise activated natural killer cells, (CAR) T-cells, tumor-infiltrating lymphocytes or tumor antigen-loaded dendritic cells.

Preferably, the cancer to be treated is a GPC3 positive cancer. For example, the cancer may be liver cancer, ovarian cancer, melanoma, squamous cell carcinoma of the lung, hepatoblastoma, nephroblastoma [Wilms tumor], yolk sac tumor, or some paediatric cancers. In a particular embodiment, the cancer to be treated is selected from hepatocellular carcinoma and hepatoblastoma.

In one embodiment, the subject to be treated is a human. In a particular embodiment, the subject is an HLA- A2+ human subject. In another particular embodiments, the subject to be treated is an HLA-C4+ human subject. In particular embodiments, when the subject to be treated is an HLA-C4+ human subject, then the TCR to be used comprises a chain CDRs consisting of SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 9 [CDR3], and p chain CDRs consisting of SEQ ID NO: 5 [CDR1], SEQ ID NO: 6 [CDR2], and SEQ ID NO: 11 [CDR3], More particularly, the TCR comprises or consists of a chain variable region comprising or consisting of SEQ ID NO: 14, and p chain variable region comprising or consisting of SEQ ID NO: 16.

In one embodiment, the therapeutic use refers to a TCR, functional fragment thereof, TCR construct, cell, pharmaceutical composition, or kit, all of them as defined above, wherein the CDR1, CDR2 and CDR3 of the a chain consist of SEQ ID NO: 3, SEQ ID NO: 4, and SEQ ID NO: 8, respectively, and the CDR1, CDR2 and CDR3 of the p chain consists of SEQ ID NO: 5, SEQ ID NO: 6, and SEQ ID NO: 10, respectively. This embodiment also contemplates polynucleotides, expression vectors or cells comprising the same, which comprise a polynucleotide encoding for said a and p chain CDRs. In more particular embodiments, the therapeutic use refers to a TCR, functional fragment thereof, TCR construct, or cell, wherein the a chain variable region comprises SEQ ID NO: 13 and the p chain variable region comprises SEQ ID NO: 15. This embodiment also contemplates polynucleotides, expression vectors or cells comprising the same, which comprise a polynucleotide encoding for said a and p variable regions. In a particular embodiment, the therapeutic use refers to a TCR, functional fragment thereof, TCR construct, cell, pharmaceutical composition, or kit, all of them as defined above, wherein the CDR1, CDR2 and CDR3 of the a chain consist of SEQ ID NO: 3, SEQ ID NO: 4, and SEQ ID NO: 9, respectively, and the CDR1, CDR2 and CDR3 of the p chain consists of SEQ ID NO: 5, SEQ ID NO: 6, and SEQ ID NO: 11, respectively. This embodiment also contemplates polynucleotides, expression vectors or cells comprising the same, which comprise a polynucleotide encoding for said a and p chain CDRs. In more particular embodiments, the therapeutic use refers to a TCR, functional fragment thereof, TCR construct, or cell, wherein the a chain variable region comprises SEQ ID NO: 14 and the p chain variable region comprises SEQ ID NO: 16. This embodiment also contemplates polynucleotides, expression vectors or cells comprising the same, which comprise a polynucleotide encoding for said a and p variable regions. In a very particular embodiment, the therapeutic use refers to a TCR, functional fragment thereof, TCR construct, or cell, wherein the CDR1, CDR2 and CDR3 of the a chain consist of SEQ ID NO: 3, SEQ ID NO: 4, and SEQ ID NO: 7, respectively, and the CDR1, CDR2 and CDR3 of the p chain consists of SEQ ID NO: 5, SEQ ID NO: 6, and SEQ ID NO: 10, respectively. This embodiment also contemplates polynucleotides, expression vectors or cells comprising the same, which comprise a polynucleotide encoding for said a and p chain CDRs. In more particular embodiments, the therapeutic use refers to a TCR, functional fragment thereof, TCR construct, or cell, wherein the a chain variable region comprises SEQ ID NO: 12 and the p chain variable region comprises SEQ ID NO: 15. This embodiment also contemplates polynucleotides, expression vectors or cells comprising the same, which comprise a polynucleotide encoding for said a and p variable regions.

The present disclosure moreover contemplates the use of the TCR, functional fragment thereof, TCR construct, or kit, all of them as defined above, in diagnosis. In some embodiments, the TCR, functional fragment thereof, or TCR construct is a diagnosis agent. The method for detecting the presence of cancer in a subject comprises: (i) administering a TCR according to the first aspect (which can be associated to a cell), or a TCR functional fragment according to the second aspect, or a TCR construct as defined in the third aspect, and (ii) determining the presence of cells bound to the TCR, or construct, or fragment, or construct, wherein the presence of cells bound to the TCR, or fragment, or construct, is indicative of the presence of cancer. In most embodiments related to diagnosis, the TCR, functional fragment thereof, or TCR construct, is linked to a detectable molecule. Also, in most embodiments related to diagnosis, the TCR, functional fragment thereof, or TCR construct is soluble (for example, it does not contain the transmembrane domain). All the embodiments described above for the TCR, functional fragment thereof, TCR construct, also apply to the diagnostic uses described in this section.

The method of diagnosis may be in vivo or in vitro. In particular embodiments, the diagnosis of the present disclosure is carried out in vitro on a sample obtained from the subject, for example, a tumor biopsy or a blood sample.

For completeness, the present description is also disclosed in the following numbered embodiments:

1. T-cell receptor binding to GPC3(522-530) antigen, said antigen consisting of the sequence SEQ ID NO: 2, or to an HLA-A2-GPC3(522-530) complex.

2. The T-cell receptor according to embodiment 1, which:

- has a functional avidity for the GPC3(522-530)-HLA-A2 complex defined by an EC50 peptide concentration for I FNy equal or below 1.20 nM, for example from 0.020 to 1.20 nM, in particular equal or below 1.05 nM, for example from 0.03 to 1 .05 nM or from 0.035 to 0.9 nM, more in particular equal or below 0.85 mM, for example from 0.7 to 0.85 nM, or alternatively

- recognizes GPC3+HLA-A2+ tumor cells with Ct values of GPC3 RNA as determined by qRT-PCR equal or below 35 and expressing HLA-A2 levels with MFI levels as determined by flow cytometry using anti-HLA2 mAb (clone BB7.2) conjugated to phycoerythrin in the range of 0.2-2 10⁴.

3. The T-cell receptor according to embodiment 1, wherein T cells engineered with the T-cell receptor have a functional avidity for the GPC3(522-530)-HLA-A2 complex, depicted as EC50, equal or below 1.20 nM, for example from 0.020 to 1 .20 nM, wherein the functional avidity is defined as the degree of activation of TCR- engineered T cells, measured as IFNy production, following interaction of the TCR-engineered T cell with the exogenous GPC3(522-530) peptide presented by the surface HLA-2 molecule, and the EC50 is the GPC3(522-530) peptide concentration required to induce a half-maximal IFNy response.

4. The T-cell receptor according to the preceding embodiment, wherein the EC50 is equal or below 1.05 nM, for example from 0.03 to 1 .05 nM or from 0.035 to 0.9 nM, more in particular equal or below 0.85 mM, for example from 0.7 to 0.85 nM.

5. The T-cell receptor according to any one of embodiments 3-4, wherein the functional avidity of T cells engineered with the TCR for GPC3(522-530)/HLA-A2 complex is measured in a standard IFNy production assay using as target cells HLA-A2+ T2 cells pulsed with graded amounts of exogenous GPC3(522-530) peptide.

6. The T-cell receptor according to any one of embodiments 1-5, comprising: a) a chain complementary determining regions (CDRs) consisting of SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 8 [CDR3], and p chain CDRs consisting of SEQ ID NO: 5 [CDR1], SEQ ID NO: 6 [CDR2], and SEQ ID NO: 10 [CDR3], or b) a chain CDRs consisting of SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 9 [CDR3], and p chain CDRs consisting of SEQ ID NO: 5 [CDR1], SEQ ID NO: 6 [CDR2], and SEQ ID NO: 11 [CDR3], or c) a chain CDRs consisting of SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 7 [CDR3], and p chain CDRs consisting of SEQ ID NO: 5 [CDR1], SEQ ID NO: 6 [CDR2], and SEQ ID NO: 10 [CDR3],

7. The T-cell receptor according to any one of embodiments 1-6, comprising:

- a chain CDRs consisting of SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 8 [CDR3], and

- p chain CDRs consisting of SEQ ID NO: 5 [CDR1], SEQ ID NO: 6 [CDR2], and SEQ ID NO: 10 [CDR3],

8. The T-cell receptor according to any one of embodiments 1-7, comprising:

- a chain CDRs consisting of SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 9 [CDR3], and

- p chain CDRs consisting of SEQ ID NO: 5 [CDR1], SEQ ID NO: 6 [CDR2], and SEQ ID NO: 11 [CDR3],

9. The T-cell receptor according to any of the embodiments 1 to 8, which is a murine T-cell receptor.

10. The T-cell receptor according to any one of embodiments 1 to 9, comprising: a) a chain variable region comprising or consisting of SEQ ID NO: 13 or SEQ ID NO: 18, and p chain variable region comprising or consisting of SEQ ID NO: 15 or SEQ ID NO: 20, or b) a chain variable region comprising or consisting of SEQ ID NO: 14 or SEQ ID NO: 19, and p chain variable region comprising or consisting of SEQ ID NO: 16 or SEQ ID NO: 21, or c) a chain variable region comprising or consisting of SEQ ID NO: 12 or SEQ ID NO: 17, and p chain variable region comprising or consisting of SEQ ID NO: 15 or SEQ ID NO: 20.

11. The T-cell receptor according to any one of embodiments 1 to 10, comprising:

- the a chain variable region comprises or consist of SEQ ID NO: 13 or SEQ ID NO: 18, and

- the p chain variable region comprises or consist of SEQ ID NO: 15 or SEQ ID NO: 20.

12. The T-cell receptor according to any one of embodiments 1 to 10, comprising:

- a chain variable region comprising or consisting of SEQ ID NO: 14 or SEQ ID NO: 19, and

- p chain variable region comprising or consisting of SEQ ID NO: 16 or SEQ ID NO: 21.

13. A T-cell receptor binding to a HLA-A2-GPO3(522-532) antigen complex, said antigen consisting of the sequence SEQ ID NO: 100.

14. The T-cell receptor according the preceding embodiment, wherein T cells engineered with the T-cell receptor have a functional avidity for the GPO3(522-532)-HLA-A2 complex, depicted as EC50, equal or below 1 .20 nM, for example from 0.020 to 1 .20 nM, wherein the functional avidity is defined as the degree of activation of TCR-engineered T cells, measured as IFNy production, following interaction of the TCR- engineered T cell with the exogenous GPC3(522-532) peptide presented by the surface HLA-2 molecule, and the EC50 is the GPC3(522-532) peptide concentration required to induce a half-maximal IFNy response.

15. The T-cell receptor according to the preceding embodiment, wherein the EC50 is equal or below 1.05 nM, for example from 0.03 to 1 .05 nM or from 0.035 to 0.9 nM, more in particular equal or below 0.85 mM, for example from 0.7 to 0.85 nM.

16. The T-cell receptor according to any one of embodiments 14-15, wherein the functional avidity of T cells engineered with the TCR for GPC3(522-532)/HLA-A2 complex is measured in a standard IFNy production assay using as target cells HLA-A2+ T2 cells pulsed with graded amounts of exogenous GPC3(522-532) peptide

17. The T-cell receptor according to any one of claims 13-16, wherein the T-cell receptor binds to GPC3+ HLA-A2+ tumor cells.

18. The T-cell receptor according to any one of embodiments 13-17, comprising: a chain complementary determining regions (CDRs) consisting of SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 8 [CDR3], and p chain CDRs consisting of SEQ ID NO: 5 [CDR1], SEQ ID NO: 6 [CDR2], and SEQ ID NO: 10 [CDR3],

19. The T-cell receptor according to any one of embodiments 1 to 18, wherein:

- the constant region of the a chain comprises or consists of SEQ ID NO: 22 or a sequence having at least 90% identity, in particular at least 95% identity, more in particular at least 98% identity, with SEQ ID NO: 22, and - the constant region of the p chain comprises or consists of SEQ ID NO: 23 or a sequence having at least 90% identity, in particular at least 95% identity, more in particular at least 98% identity, with SEQ ID NO: 23.

20. A functional fragment of the T-cell receptor according to any one of embodiments 1-19, wherein said functional fragment:

(I) binds to a HLA-A2-GPC3(522-530) complex, or to a HLA-A2-GPC3(522-532) antigen complex, preferably with a functional avidity for the GPC3(522-532)-HLA-A2 complex, depicted as EC50, equal or below 1.20 nM, for example from 0.020 to 1 .20, in particular equal or below 1 .05 nM, for example from 0.03 to 1 .05 nM or from 0.035 to 0.9 nM, more in particular equal or below 0.85 mM, for example from 0.7 to 0.85 nM, wherein the functional avidity is defined as the degree of activation of TCR-engineered T cells, measured as I FNy production, following interaction of the TCR-engineered T cell with the exogenous GPC3(522-532) peptide presented by the surface HLA-2 molecule, and the EC50 is the GPC3(522-532) peptide concentration required to induce a half-maximal IFNy response, or alternatively, recognizes GPC3+HLA-A2+ tumor cells with Ct values of GPC3 RNA as determined by qRT-PCR equal or below 35 and expressing HLA-A2 levels with MFI levels as determined by flow cytometry using anti-HLA2 mAb (clone BB7.2) conjugated to phycoerythrin in the range of 0.2-2 10⁴; and

(II) comprises a and p chain CDRs as defined in any one of embodiments 6-9, or alternatively, a and p chain variable regions as defined in any one of embodiments 10-12.

21 . A T-cell receptor construct comprising

(1) a and p chains comprising CDRs as defined in embodiment 6 (a)-(c), or variable regions as defined in embodiment 10 (a)-(c), wherein the a and p chains are preferably linked to each other to form TCR heterodimers or multimers, and

(2) fusion component(s) selected from: Fc receptors and/or Fc domains, cytokines, such as IL-2 or IL-15, toxins, an antibody or a single chain antibody fragment (scFv),

CD3-zeta chain(s) and/or other TCR stimulation domains, such as the intracellular CD28, CD137 or CD134 domain; wherein the at least one of the T-cell receptor a or p chain (1) is bound to the fusion component(s) (2), preferably through a linker.

22. The T-cell receptor construct according to the preceding embodiment, wherein the fusion component is an antibody or a single chain antibody fragment (scFv) selected from anti-CD3, anti-CD28, anti-CD5, anti-CD 16 or anti- CD56.

23. The T-cell receptor construct according to any one of embodiments 21-22, further comprising the extracellular domain of the a and p chain constant regions defined in embodiment 19.

24. The T-cell receptor construct according to any one of embodiments 21-23, further comprising a label.

25. The functional fragment of the T-cell receptor according to embodiment 20 or the T-cell receptor construct according to any one of embodiments 21-24, said functional fragment or construct lacking the transmembrane domains of the a and p chains.

26. A polynucleotide encoding for a T-cell receptor as defined in any one of embodiments 1-19, or a functional fragment thereof as defined in embodiments 20 or 25, or a T-cell receptor construct as defined in any one of embodiments 21-25.

27. The polynucleotide according to embodiment 26, comprising: a) SEQ ID NO: 26 or a sequence having at least 90% identity with SEQ ID NO: 26 (encoding for the a chain variable region), and SEQ ID NO: 25 or a sequence having at least 90% identity with SEQ ID NO: 25 (encoding for the p chain variable region), or b) SEQ ID NO: 27, or a sequence having at least 90% identity with SEQ ID NO: 27 (encoding for the a chain variable region), and SEQ ID NO: 28 or a sequence having at least 90% identity with SEQ ID NO: 28 (encoding for the p chain variable region), or c) SEQ ID NO: 24 or a sequence having at least 90% identity with SEQ ID NO: 24 (encoding for the a chain variable region), and SEQ ID NO: 25 or a sequence having at least 90% identity with SEQ ID NO: 25 (encoding for the p chain variable region).

28. The polynucleotide according to any one of embodiments 26-27, comprising:

SEQ ID NO: 26 or a sequence having at least 90% identity with SEQ ID NO: 26 (encoding for a chain variable region), and

SEQ ID NO: 25 or a sequence having at least 90% identity with SEQ ID NO: 25 (encoding for the p chain variable region).

29. The polynucleotide according to any one of embodiments 26-28, comprising:

SEQ ID NO: 27, or a sequence having at least 90% identity with SEQ ID NO: 27 (encoding for the a chain variable region), and SEQ ID NO: 28 or a sequence having at least 90% identity with SEQ ID NO: 28 (encoding for the p chain variable region).

30. The polynucleotide according to any one of embodiments 26-29, further comprising

SEQ ID NO: 29 (encoding for the a chain constant region) and

SEQ ID NO: 30 (encoding for the p chain constant region).

31 . An expression vector comprising a polynucleotide as defined in any one of embodiments 26-30.

32. The expression vector according to the preceding embodiment, which is a retrovirus or lentivirus vector, in particular lentivirus.

33. The expression vector according to any one of embodiments 32-32, which is replication-defective.

34. A cell comprising a T-cell receptor as defined in any one of embodiments 1-19, or a functional fragment thereof as defined in embodiments 20 or 25, or a T-cell receptor construct as defined in any one of embodiments 21-25, or a polynucleotide as defined in any one of embodiments 26-30, or an expression vector as defined in any one of embodiments 31-33. 35. The cell according to the preceding embodiment, which is an HLA-A2 positive cell or an HLA-C4 positive cell, in particular HLA-A2 positive cell.

36. The cell according to the preceding embodiment, which is a T cell.

37. The cell according to the preceding embodiment, which is a CD8 T cell.

38. The cell, in particular the T cell, more in particular the CD8 T cell, according to any one of embodiments 34-37, which has a functional avidity for the GPC3(522-530)-HLA-A2 complex, measured as I FNy production, of EC50, equal or below 1.20 nM, for example from 0.020 to 1.20 nM, wherein the functional avidity is defined as the degree of activation of TCR-engineered T cells, measured as I FNy production, following interaction of the TCR-engineered T cell with the exogenous GPC3(522-530) peptide presented by the surface HLA-2 molecule, and the EC50 is the GPC3(522-530) peptide concentration required to induce a half-maximal I FNy response.

39. The cell, in particular the T cell, more in particular the CD8 T cell, according to any one of embodiments 34-37, which has a functional avidity for the GPC3(522-532)-HLA-A2 complex, depicted as EC50, equal or below 1.20 nM, for example from 0.020 to 1.20 nM, wherein the functional avidity is defined as the degree of activation of TCR-engineered T cells, measured as IFNy production, following interaction of the TCR- engineered T cell with the exogenous GPC3(522-532) peptide presented by the surface HLA-2 molecule, and the EC50 is the GPC3(522-532) peptide concentration required to induce a half-maximal IFNy response.

40. The cell according any one of embodiments 38-39, wherein the EC50 is equal or below 1 .05 nM, for example from 0.03 to 1 .05 nM or from 0.035 to 0.9 nM, more in particular equal or below 0.85 mM, for example from 0.7 to 0.85 nM.

41 . The cell according to any one of embodiments 38-40, wherein the functional avidity of T cells engineered with the TCR for GPC3(522-530)/HLA-A2 complex or with the TCR for GPC3(522-532)/HLA-A2 complex is measured in a standard IFNy production assay using as target cells HLA-A2+ T2 cells pulsed with graded amounts of exogenous GPC3(522-530) or GPC3(522-532) peptide.

42. The cell according to any one of embodiments 34-41, which is transgenic.

43. The cell according to any one of embodiments 34-42, which is a chimeric cell comprising a human cell and a murine T-cell receptor.

44. A method of producing the cell according to any of embodiments 34-43, the method comprising a step of delivering the polynucleotide according to any of embodiments 26-30, or the expression vector according to any of embodiments 31-33, into a host cell.

45. A pharmaceutical composition comprising a T-cell receptor as defined in any one of embodiments 1-19, or a functional fragment thereof as defined in embodiments 20 or 25, or a T-cell receptor construct as defined in any one of embodiments 21-25, or a polynucleotide as defined in any one of embodiments 26-30, or an expression vector as defined in any one of embodiments 31-33, or a cell as defined in any one of embodiments 34-43, together with pharmaceutically acceptable excipients and/or carriers.

46. a kit of parts comprising:

(a) a T-cell receptor as defined in any one of embodiments 1-19, or a functional fragment thereof as defined in embodiments 20 or 25, or a T-cell receptor construct as defined in any one of embodiments 21-25, or a polynucleotide as defined in any one of embodiments 26-30, or an expression vector as defined in any one of embodiments 31-33, or a cell as defined in any one of embodiments 34-43, or a pharmaceutical composition as defined in embodiment 45;

(b) optionally a further therapeutic agent; and

(c) optionally, instructions for its use.

47. A T-cell receptor as defined in any one of embodiments 1-19, or a functional fragment thereof as defined in embodiments 20 or 25, or a T-cell receptor construct as defined in any one of embodiments 21-25, or a polynucleotide as defined in any one of embodiments 26-30, or an expression vector as defined in any one of embodiments 31-33, or a cell as defined in any one of embodiments 34-43, or a pharmaceutical composition as defined in embodiment 45; or a kit as defined in embodiment 46, for use as a medicament.

48. A T-cell receptor as defined in any one of embodiments 1-19, or a functional fragment thereof as defined in embodiments 20 or 25, or a T-cell receptor construct as defined in any one of embodiments 21-25, or a polynucleotide as defined in any one of embodiments 26-30, or an expression vector as defined in any one of embodiments 31-33, or a cell as defined in any one of embodiments 34-43, or a pharmaceutical composition as defined in embodiment 45; or a kit as defined in embodiment 46, for use in the treatment of cancer, in particular, for treating a GPC3 positive cancer.

49. The T-cell receptor as defined in any one of embodiments 1-19, or the functional fragment thereof as defined in embodiments 20 or 25, or the T-cell receptor construct as defined in any one of embodiments 21- 25, or the polynucleotide as defined in any one of embodiments 26-30, or the expression vector as defined in any one of embodiments 31-33, or the cell as defined in any one of embodiments 34-43, or the pharmaceutical composition as defined in embodiment 45; or the kit as defined in embodiment 46, for use according to embodiment 48, wherein the cancer is selected from the group consisting of liver cancer and ovary cancer, melanoma, squamous cell carcinoma of the lung, hepatoblastoma, nephroblastoma [Wilms tumor], yolk sac tumor, and some paediatric cancers, in particular, the cancer is selected from hepatocellular carcinoma, and hepatoblastoma.

50. The T-cell receptor as defined in any one of embodiments 1-19, or the functional fragment thereof as defined in embodiments 20 or 25, or the T-cell receptor construct as defined in any one of embodiments 21- 25, or the polynucleotide as defined in any one of embodiments 26-30, or the expression vector as defined in any one of embodiments 31-33, or the cell as defined in any one of embodiments 34-43, or the pharmaceutical composition as defined in embodiment 45; or the kit as defined in embodiment 46, for use according to embodiment 49, wherein the cancer is selected from hepatocellular carcinoma and hepatoblastoma.

51. The T-cell receptor as defined in any one of embodiments 1-19, or the functional fragment thereof as defined in embodiments 20 or 25, or the T-cell receptor construct as defined in any one of embodiments 21- 25, or the polynucleotide as defined in any one of embodiments 26-30, or the expression vector as defined in any one of embodiments 31-33, or the cell as defined in any one of embodiments 34-43, or the pharmaceutical composition as defined in embodiment 45; or the kit as defined in embodiment 46, for use according to any one of embodiments 48-50, that is for the treatment of a human subject, in particular, an HLA-A2 positive human subject.

52. The T-cell receptor as defined in any one of embodiments 1-19, or the functional fragment thereof as defined in embodiments 20 or 25, or the T-cell receptor construct as defined in any one of embodiments 21- 25, or the polynucleotide as defined in any one of embodiments 26-30, or the expression vector as defined in any one of embodiments 31-33, or the cell as defined in any one of embodiments 34-43, or the pharmaceutical composition as defined in embodiment 45; or the kit as defined in embodiment 46, for use according to any one of embodiments 48-50, that is for the treatment of a human subject; in particular, an HLA-C4 positive human subject, wherein:

- The TCR comprises CDRs as defined in embodiment 8, or variable regions as defined in embodiment 12, or

- The polynucleotide encodes for the CDRs as defined in embodiment 8, or variable regions as defined in embodiment 12, or comprises the sequence as defined in embodiment 29.

53. The use of the T-cell receptor as defined in any one of embodiments 1-19, or the functional fragment thereof as defined in embodiments 20 or 25, or the T-cell receptor construct as defined in any one of embodiments 21-25, or the kit as defined in embodiment 46, for detecting the presence of cancer in a subject.

Throughout the description and claims the word "comprise" and variations of the word, are not intended to exclude other technical features, additives, components, or steps. Furthermore, the word "comprise” encompasses the case of "consisting of”. Additional objects, advantages and features of the invention will become apparent to those skilled in the art upon examination of the description or may be learned by practice of the invention. The following examples and drawings are provided by way of illustration, and they are not intended to be limiting of the present invention. Reference signs related to drawings and placed in parentheses in a claim, are solely for attempting to increase the intelligibility of the claim, and shall not be construed as limiting the scope of the claim. Furthermore, the present invention covers all possible combinations of particular and preferred embodiments described herein.

Examples

Example 1

T cells receptors specific for Glypican 3 (GPC3) in the context of HLA-A2 molecules to redirect human T cells to GPC3⁺ tumors.

Identification of relevant T cell epitopes of GPC3 in the HHD-DR¹ model

To identify relevant GPC3 T cell epitopes, HHD-DR1 mice were immunized with an adenovirus encoding human GPC3 (ADV-GPC3) (Fig. 1A). HHD-DR1 mice were immunized with ADV-GPC3 (10⁸ viral particles per mouse). On day 8, splenocytes were isolated and stimulated with seven mixtures of synthetic GPC3 peptides covering the entire GPC3 protein (Table 1). Table 1. Sequence of synthetic overlapping peptide covering the entire human GPC3 protein. Each peptide contains 20 amino acids and overlaps 10 amino acids with the adjacent peptides. For each peptide its position in the GPC3 protein (SEQ ID NO: 1)(^a) and the mixture it was part of in the first epitope screening (^b) are shown.

For the first screening, cells were stimulated (37 °C/6h) with seven mixtures of synthetic GPC3 peptides covering the entire GPC3 protein or with medium ( ). Each mix contained an equimolar amount of 8 or 9 consecutive GPC3 peptides. The final concentration of each peptide in the culture was 1 g/mL. One hour after starting the culture, brefeldin A (10 ug/mL) was added. Five hours later, cells were stained with antimouse CD8 (clone 53.6-7) and CD4 (clone GK1.5) fluorochrome-labeled monoclonal antibodies (mAbs). Then, they were fixed and permeabilized, and finally stained intracellularly with fluorochrome-labeled antimouse IFNy (clone XMG1.2) mAbs. Once the reactive mix was identified (Mix7), a second screening was performed, but this time splenocytes were stimulated with each one of the single peptides (1 pg/mL) from the reactive mix. In both screenings, splenocytes from a naive mouse were stimulated in parallel as negative control.

The induced T cell response was analyzed by fluorescence-activated cell sorter (FACS) analysis, detecting IFN-y (IFNyj CD4 and CD8 T lymphocytes. A specific response was only detected in the CD8⁺ population against mix 7 (Fig. 1 B). When analyzing the response against each of the single peptides of mix 7, each of them identified in Fig. 1 by position number of its first residue, reactivity was only detected against the peptide pGPC3(511-530) [i.e. p511 in Fig. 1], This peptide contains a naturally processed, HLA-A2-restricted CD8 T cell epitope [pGPC3(522-530): FLAELAYDL], For subsequent studies, the 9mer peptide pGPC3(522-530) containing this minimal epitope was used.

Single-cell sorting and expansion of GPC3-specific T cell clones

To identify GPC3-specific TOR, the process illustrated in Fig. 2A was conducted in two different experiments. Briefly, HHD-DR1 mice were immunized with ADV-GPC3, and splenocytes were obtained. The presence of GPC3-specific T lymphocytes was confirmed by stimulating a cell sample with pGPC3(522-530) and analyzing the IFNy production as before (Fig. 2B). Then, pGPC3(522-530)-specific CD8 T cells were singlecell sorted as high IFNy producing cells (Fig. 2A and C). Briefly, splenic CD8 T cells from immunized mice were magnetically sorted by negative selection using the EasySep Mouse CD8+ T cell Isolation Kit (Stemcell) and stimulated with autologous bone marrow-derived dendritic cells (BM-DC) pulsed with or without pGPC3(522-530) (1 ug/ml) for 4 hours at 37 °C. Cells were labeled with the Mouse IFNy Secretion Assay kit (Miltenyi), following the manufacturer's instructions. Finally, cells were stained with fluorochrome-labeled antimouse IFNy (from the Mouse IFNy Secretion Assay kit) and CD8 (clone 53.6-7) mAbs. CD8⁺ and highly IFNy⁺ (I FNy^hi) cells were isolated as single cells (1 cell/well) in 96-well U plates using an Aria II cytometer (BD Bioscience). Two experiments were performed, in which 380 (1^st experiment) and 300 (2^nd experiment). Given that the system chosen to make the TCR libraries (SMARTer Mouse TCRa/b Profiling kit, Takara) requires 10³-10⁴ cells or 10-500 ng of RNA, single T cells specific for pGPC3(522-530) epitope were expanded to establish clones. By the term clone, we refer here to cells expanded from a single well. For clone expansion, sorted cells were cultured in the presence of 3,000 rad irradiated 5 x10⁴ autologous BM-DC together with pGPC3(522-530) peptide (1 pg/ml). Every 15 days the stimulation with peptide-pulsed BM-DC was repeated. Three rounds of stimulation were given. BM-DCs used in this process were previously obtained from the BM of HHD-DR1 mice upon culture (6 days) in the presence of murine granulocyte-macrophage colonystimulating factor (20 ng/mL). The number of established clones was 21 and 13 clones from experiments 1 and 2, respectively (Table 2).

Table 2. Summary of the number of expanded clones passing quality control (QC) until the final TCR sequencing phase. The yields of each step are indicated. (^a) For library preparation, "extracted total RNA" (experiment 1) or "whole cells" (experiment 2) were used. N.A.: not applicable. NGS: next Generation Sequencing

Successfully expanded clones were spun down and homogenized in RLT buffer (Qiagen) (experiment 1) or in 1 x Reaction Buffer containing RNase Inhibitor (Takara) (experiment 2) and frozen for TCR library preparation.

TCR sequencing and bioinformatic analysis

TCR libraries were prepared from "extracted total RNA" (experiment 1) or "whole cells" (experiment 2) using SMARTer Mouse TCR a/b Profiling kit (Takara Bio). In experiment 1, total RNA from T-cell clones was isolated using the RNeasy Micro kit (Qiagen) according to the manufacturer's instructions. TRA and TRB libraries were prepared on 5-10 ng of extracted RNA with the SMARTer Mouse TCR a/b Profiling Kit following the manufacturer's instructions. Previous tests with the control RNA included in the kit showed that 5 ng of RNA isolated from a T-cell population was sufficient to amplify the TCR genes. From 12 out of 21 clones established in experiment 1, sufficient quality RNA was obtained to prepare libraries ( >5 ng) (Table 2).

All libraries prepared from extracted total RNA passed quality control (Table 2). In experiment 2, the RNA present in the sample (i.e. without isolation) was directly reverse-transcribed to cDNA using SMARTer Mouse TCR a/b Profiling kit. Twelves out of 13 clones from experiment 2 passed the library preparation quality controls (Table 2). Libraries were purified using "Agencourt AMPure XP Beads". High Sensitivity D1000 ScreenTape (Agilent 4200 TapeStation System"; Agilent technology) was used to determine amplified fragment size and the library concentration. TCRo and TCRp sequencing was performed using a 600-cycle MiSeq Reagent Kit V3 for 2 x 300 base-pair reads (Illumina). Library preparation was carried out in our laboratory. Library validation and sequencing were performed at the IMIBIC Genomics Unit (Hospital Reina Sofia, Cordoba). The 24 libraries, encompassing 24 clones, were successfully sequenced and analyzed using MiXCR (Bolotin et al) and VDJtools (Shugay et al).

TCR clonotypes were identified based on their nucleotide sequence. From the 24 T-cell clones sequenced, 4 clone pairs share the same TCR clonotype, therefore 20 different TCR clonotypes were finally identified, of which we show data of three clonotypes (Table 3).

Table 3. TCRa and TCR chain sequences of the expanded T cell clones. Twenty-four T-cell clones expanded in two different experiments (Exp) (Exp 1 and Exp 2) were sequenced and analyzed. Four T-cell clone pairs share the same TCR clonotype. Therefore, 20 different TCR clonotypes were identified, of which we show data of three. The Vo, Jo, Vp, Dp and Jp segment genes of each TCR clonotype are shown. For each TCRo and TCRp chain, the amino acid sequence and length (residue numbers) of the CDR3o and CDR3p regions are also shown.

The average residues number of the CDR3o and CDR3p regions was 10.9 (minimum 9 and maximum 15) and 13.45 (minimum 11 and maximum 16), respectively. Regarding the Va and Vp usage, 75% (15/20) and 87% (17/20) of the identified TCRs presented TRAV9-2 and TRBV1, respectively. Other Va and Vp gene segments used were TRAV9-4 (3/20) and TRAV9D-2 (2/20); and TRBV2 (1/20), TRBV3 (1/20) and TRBV26 (1/20), respectively. As for the J usage, the most frequent TRBJ gene segments were TRBJ2-7 (7/20) and TRBJ2-1 (5/20). A great diversity of TRAJ segments was detected among the identified TCRs. A convergent recombination event (different nucleotide sequences coding for the same CDR3) was observed in the CDR3p of clones 1 H4 and 3E9, from experiment 1 and 2, respectively.

Of the 20 identified TOR clonotypes, three were selected to be cloned and tested on human T cells: clones 1 H4 and 3E9 (TCR-3 and TCR-4, respectively), which were the clones in which the convergent recombination event was observed; and clone pair 1 F2/4B5 (TCR-5).

TCR reconstruction by Golden gating assembly and retroviral production

For TCR reconstruction, the Murine Stem Cells Virus (MSCV)-Mu(*C) Acceptor vector was used. This vector was generated by cloning a TCR acceptor cassette in pMSCV neo Vector (631461, Takara) between EcoRI and BamHI restriction site (Fig. 3A). This cassette contains the murine TCR alpha (mCa) and beta (mCp2) constant chains. The use of murine constant regions avoids mispairing of the transgenic TCR with the endogenous TCR when transducing human T lymphocytes. The mCo and mCp were modified to insert an additional interchain disulfide bond to improve the expression and biological activity of transgenic TCRs (Cohen et al, Thomas et al). Both constant chains were joined by a linker consisting of a furin cleavage site and an autocatalytic region P2A. This approach allows equimolar expression of both chains while cleaving them without adding any additional amino acid residues. The vector was designed to allow directed and unidirectional cloning of Vo and V|3 regions in front of Co and Cp2, respectively, by "Golden Gate Assembly", following a strategy similar to that described by Coren et al for the design of the MSGV1 Hu Acceptor vector (Coren et al). Thus, mCo and mCp2 DNA sequences contained two Bbsl (type IIS enzyme) sites (GAAGAC (2/6)^A) in the 5'-end separated by a short unstable stuffer (which provides spacing for efficient cleavage) and four base overhangs including conserved coding sites to enable V region fusion with the respective constant regions. The DNA sequence coding the TCR acceptor cassette was codon optimized to ensure maximal expression in human cells and synthesized by GenScript.

For each TCR, the sequences of Vo and Vp fragments were codon optimized to ensure maximal expression in human cells. They were also flanked with the Bbsl enzyme cleavage sites in both 5' and 3' DNA segments to allow cloning into the MSCV-Mu(*C) Acceptor vector. The modified Vo and Vp sequences were synthesized (Genscript) and cloned into the pUC57 plasmid. The sequences encoding Vo and Vp of the murine TCRs were cloned into the MSCV-Mu(*C) Acceptor by Golden Gate Assembly, as described by Coren et al. (Coren et al) (Fig. 3B). Briefly, for each TCR constructs (indicated as x), a mixture containing T4 ligase buffer, MSCV-Mu(*C) Acceptor plasmid (100 ng), pUC57-TCRx-Vo (100 ng), pUC57-TCRx-Vp (100 ng), Bbsl (10 U) (New England Biolab, NEB) and T4 ligase (2000 U) (NEB) in nuclease-free water was incubated in a thermal cycler for 20 cycles of 37 °C/5' and 16 °C/5', followed by a 37 °C/5' extension step and an 80 °C/5' inactivation step. Eventually, unligated linear DNA was digested with Plasmid-Safe™ ATP-dependent DNase (Lucigen) (10 U) in the presence of ATP solution (Lucigen) (1 mM) at 37 °C (1 h) and 70 °C (30'). As a control, a reaction mixture containing all reagents unless pUC57-Vo and pUC5-Vp were run in parallel to check the golden gating assembly. Once the reaction finished, 50 uL of chemically competent bacteria (XL1 Blue) were transformed with 5 ul of Golden Gate reaction mix following heat shock methodology (42 °C/45” and immediately 57ice). To continue, all the volume was added to 450 uL of SOC medium (Sigma), incubated (37 °C/1h) and seeded onto 100 pg/ml ampicillin Luria Broth agar for 37°C/18h. Typical transformations yielded an average of 60-90 total colonies/transformation. The plasmid was isolated from the bacterial cultures using the NucleoSpin Plasmid Kit (Macherey-Nagel). Colony pre-screening was done by double enzymatic digestion with EcoRI-HF and BamHI enzymes (NEB) (37°C/1h) and electrophoresis in agarose gel. The detection of two bands, with 5.6 and 1 .9 kb, was indicative of effective cloning. Colonies showing the appropriate banding pattern were confirmed by Sanger sequencing (performed at CIMA-LAB). Two primers, complementary to the MSGV1-Mu(*C) acceptor plasmid sequence upstream of Vo (Forward (F1, SEQ ID NO: 31): 5'- ACCTCGCTGGAAAGGACCTTAC-3') and within the mCp2 gene region (Reverse (R2, SEQ ID NO: 32): 5'AATTGGACTCCTTGTAGGCCTG-3') were designed to verify the re-constructed 5'-Va-Ca-linker-V[3-Cp-3' sequence. Finally, verified colonies were expanded and plasmids were isolated form bacterial cultures using the NucleoBond Xtra Maxi™ kit (Macherey-Nagel). The resulting plasmid was used to produce retrovirus to transduced human T cells (Fig. 3B). Briefly, Platinum-A (Plat-A) Retroviral Packaging cells (Cell Biolabs, Inc.) were plated 24 hours before transfection in 6-well plates (8 x 105 cells per well) onto previously poly-d-lysine- coated plates (50 ug/mL, GIBCO) in a final volume of 2mL/well of infection medium (DMEM-Glutamax (GIBCO), FBS 10%, 1% sodium pyruvate, 1% non-essential amino acids, 10 mM HEPES). Twenty-four hours later, when cells reached 70% of confluence, 500pL/well of a mix in OPTIMEM medium (GIBCO) containing 2 pg of the TCR expressing vector, 1 pg pMD2.G plasmid (Addgene plasmid # 12259), and Lipofectamine 2000® (Thermo Fisher) was added. Infection medium was replaced by PLAT-A medium (DMEM-Glutamax (GIBCO), FBS 10%, 1 % sodium pyruvate, 1 % non-essential amino acids, 10 mM HEPES, 100 U/mL P/S and 10 pig/mL gentamicin) 24 hours post-transfection. The supernatant containing the retroviruses was collected 48 hours and 72 hours post-transfection and kept at 4 °C or frozen at -80 °C until cell transduction. Plat-A cells were maintained in "PLAT-A medium” supplemented with Puromycin (1 pg/ml) and Blasticidin (10 pg/ml).

TCR expression on human T cells and study of the dependence of the CD8 coreceptor in the recognition of the peptide/MHC complex

To assess the capacity of the cloned TCR to recognize GPC3, we retrovirally engineered human T cells with the selected TCR genes to generate TCR-T cells (Fig. 4A). Briefly, peripheral blood leukocytes (PBL) from healthy donors were isolated by Ficoll density gradient from blood collected in EDTA tubes. Next, PBLs were incubated with a mix containing anti-human CD8 and CD4 coated magnetic beads (Miltenyi Biotec) in "PEF Buffer” (PBS with 5% EDTA, 5% FBS, and 100 U/mL P/S) following the manufacturer's instructions. CD4⁺ and CD8⁺ populations were jointly isolated by positive selection using an LS column (Miltenyi). Human T cells (containing both CD4⁺ and CD8⁺ cells) (10⁶ cells/mL) were activated (2 days) with TransAct reagent (anti-CD3 and -CD28 mAbs) (Miltenyi) (10 uL/mL) in "T-cell medium” [1 :1 mixture of AIMV (Invitrogen) and RPM1 1640- glutamax (GIBCO), containing 5% heat-inactivated human AB serum (SIGMA), 12.5 mM HEPES, 100 U/ml of P/S and 10 pig/mL gentamicin] supplemented with 100ng/mL (h)IL-7 (Immunotools) and 50 ng/mL of (h)IL-15 (Immunotools) (T-cell-l L-7-IL-15 medium). T cells were transduced twice on days 2 and 3 using retroviral particles encoding the TCRs and retronectin-coated plates. Briefly, non-tissue culture-treated 24-well plates (Life sciences) were coated overnight with retronectin (Takara Bio) at 4 °C (30 pg/mL, 0.3 mL/well). After 24 hours, the retronectin solution was transferred to a clean plate to prepare the microwells that will be used in the second infection. Retronectin-coated plates were subsequently blocked with PBS 2% BSA (RT/30') and then washed with PBS. After aspirating the wash buffer, the 48-hour retroviral supernatants were added to the retronectin-coated wells (1 mL/well). The plates were then centrifuged at 2000g for 2 h at 32 °C. Meanwhile, activated T cells were collected and resuspended (10⁶/ml) in retrovirus-containing supernatant supplemented with 2x IL-7 and IL-15 (200ng/mL and 100 ng/mL, respectively). Once the centrifugation of the plate with the retronectin and the virus had elapsed, 1 mL/well of the solution containing T cells, the retrovirus and the cytokines was added to the volume of virus present on the plate and the plate was again centrifuged at 2,000 g for 20 min at 32°C and finally incubated for 3-4 h at 37 °C. Next, half of the medium was removed and replaced by "T-cell-IL-7-IL-15 medium”. Transductions were repeated on day 3 post-stimulation using the same protocol with the 72-hour retroviral supernatants. Five hours after the second transduction, T cells were harvested and resuspended in "T-cell-IL-7-IL-15 medium” and allowed to expand in vitro. A sample of activated T cells was left untransduced (UTD) as a control. Transduction efficiency was assessed by measuring murine TCR|3 (mTCRp) surface expression fluorescence-activated cell sorter (FACS) analysis using an anti-mouse TCR|3 mAb (clone H57-597).

The three selected TCRs were efficiently expressed on the surface of human CD4 and CD8 T cells, as depicted by staining with an antibody that recognized the constant region (C|32) of the murine TCR|3 chain. In general, the percentage of transduced (mTCR|3⁺) cells was slightly higher in CD4 than in CD8 T cells. To know if TCRs expressed by CD4 and CD8 T cells could recognize the GPC3(522-530) epitope presented by the HLA-A2 molecule, we stained T cells with a phycoerythrin (PE)-labeled pGPC3(522-530)/HLA-A2 tetramer, at saturating concentration. Briefly, empty loadable HLA-A2 tetramer conjugated to phycoerythrin (PE) was purchased from Tetramer shop. pGPC3(522-530) peptide was loaded to the empty tetramer (4°C/30') following manufacturer's instructions. T cells were incubated (37 °C/15') with 1/20 dilution of pGPC3(522-530) /HLA-A2 tetramer (saturating concentration), and then (4 °C/30') with anti-human CD4 (clone RPAT4) and/or CD8 (clone RPAT8) mAbs and anti-mouse TCRp mAb (clone H57-597). Cells were analyzed by FACS. All TCRs, when expressed by CD8 T cells, were capable of binding to pGPC3(522- 530)/HLA-A2 tetramer (Fig. 4B). However, when the TCRs were expressed by CD4 T cells, only TCR-3 was able to bind to the tetramer (~ 1 % Tetramer cells) but to a much lower extent that when expressed in CD8 T cells. These data indicated that the recognition of the pGPC3(522-530)/HLA-A2 tetramer complex by the GPC3 TCRs was strongly dependent on the CD8 co-receptor.

To find out if the GPC3 TCRs expressed by CD4 and CD8 T cells were functional, the bulk of genetically modified T cells (5 10⁴ cells/well), including CD4 and CD8 T cells, were co-cultured with T2 cells (HLA-A2⁺, TAP deficient cells) (1 10⁵ cells/well) previously pulsed with a saturating concentration of pGPC3(522-530) (1 g/mL) or with medium. Twenty hours later, cells were recovered and stained with anti-human CD4 (clone RPAT4), CD8 (clone RPAT8) and CD137 (clone 4B4-1 ) mAbs and anti-mouse TCRp mAb (clone H57-597) and analyzed by FACS. Confirming data with tetramer staining, only the mTCRp⁺CD8⁺ cell population efficiently responded to stimulation with peptide-pulsed T2 cells, as depicted by CD137 upregulation on the surface of transduced (mTCRp⁺) cells (Fig. 4C). TCR-3 and TCR-5 showed the highest response followed by TCR-4. Only TCR-3-transduced CD4 T cells were able to recognize T2 cells pulsed with pGPC3(522-530), although the response was very weak. Taken together, our data showed that GPC3 TCRs can be expressed efficiently on the surface of human T cells, which can respond to stimulation with the pGPC3(522-530)/HLA- A2 complex, but that recognition of this complex depends largely on the CD8 co-receptor, so the identified GPC3-TCRs are only functional when expressed on CD8 T cells. Based on these results, and to facilitate comparison between the different TCRs, the rest of the study was carried out with CD8 T cells.

Analysis of avidity and functional avidity of the selected TCRs

TCR avidity measures the strength of TCR/peptide/MHC interactions and considers the effect of other molecules, such as co-receptors, in the interaction. It is usually determined by staining with decreasing amounts of MHC multimers pulsed with the peptide of interest. Instead, functional avidity is a biological measure that describes how well a T cell responds to a target antigen in terms of its activation and effector functions, namely, upregulation of activation markers, cytokine production, etc. In addition to the interactions of the TCR and co-receptor with the peptide/MHC complex, it considers the TCR signaling necessary for T- cell activation. To assess the avidity of the three TCR, TCR-T cells were stained with decreasing concentrations of the pGPC3(522-530)/HLA-A2 tetramer (from 1/20 to 1/10240) (37°C/15'), and then with anti-human CD8 and anti-mouse TCRp mAbs (4°C/30') and analyzed by FACS. For each dilution factor, the percentage of tetramer cells within the mTCRp⁺ population was calculated. To assess functional avidity, T2 (HLA-A2⁺TAP ) cells were pulsed with 10-fold serial dilutions of pGPC3(522-530) (from 1 to 10-5 piM) and then 2 x 10⁴ cells/well were cultivated with TCR-engineered CD8 T cells (5 x 10³ mTCRp⁺ cells/well) from HLA-A2^nes donors. Twenty hours later, cells and supernatant were recovered. The response was evaluated by measuring surface CD137 expression in mTCRp⁺ cells by FACS, and the production of I FNy, IL-2 and TNFo by analyzing the culture supernatants with specific ELISA (BD OptEIA Set, BD Bioscience). To properly compare the different TCRs, the percentage of CD8 TCR⁺ T cells in the different TCR-T cell lines was equalized by adding non-transduced CD8 T cells. To compare the different TCRs in terms of avidity, the EC50 for tetramer binding (described as the tetramer dilution at which a half-maximal number of tetramer cells is reached) was calculated. TCR-3 stood out by their high avidity, followed by TCR-4 and eventually TCR-5 (Table 4).

Table 4. Avidity and Functional avidity (EC50 values) of GPC3-specific TCRs. (^a) Avidity was assessed by staining TCR-T cells with decreasing concentrations of the pGPC3(522-530)/HLA-A2 tetramer (from 1/20 to 1/10240) (37°C/15'), and then with anti-human CD8 and anti-mouse TCRp mAbs (4°C/30') and analyzed by FACS. For each dilution factor, the percentage of tetramer cells within the mTCRp⁺ population was calculated. The EC50 for tetramer binding is described as the tetramer dilution at which a half-maximal number of tetramer cells is reached. (^b) Functional avidity. T2 (HLA-A2⁺, TAP deficient) cells were pulsed with 10-fold serial dilutions of pGPC3(522-530) (from 1 to 10-5 piM) and then 2 x 10⁴ cells/well were cultivated with TCR-engineered CD8 T cells (5 x 10³ mTCRp⁺ cells/well) from HLA-A2^ne9 donors. Twenty hours later, cells and supernatant were recovered. The response was evaluated by measuring surface CD137 expression in mTCRp⁺ cells by FACS (as described in figure 4C), and the production of I FNy, IL-2 and TNFo by analyzing the culture supernatants with specific ELISA (BD OptEIA Set, BD Bioscience). To properly compare the different TCRs, the percentage of CD8 TCR⁺ T cells in the different TCR-T cell lines was equalized by adding non-transduced CD8 T cells. EC50 peptide concentration represents the peptide dose at which a half-maximal response is reached. The EC50 was calculated with the model log(agonist) vs. response (three parameters) of Graphpad 8.0.1 . Data are the mean of three independent experiments with different donors. (^c) Classification of the different TCRs according to their avidity and functional avidity.

Similarly, to compare the functional avidity of the different TCRs, the EC50 peptide concentration (representing the peptide dose at which a half-maximal response is reached) was also estimated (Table 4). For each TCR, the EC50 values varied greatly depending on the parameter used to evaluate T-cell activation, with the expression of CD 137 and the production of TNFa being the determinations that presented the highest EC50 values, followed by the release of IL-2 and I FNy (Table 4). The classification of the TCRs according to their functional avidity was relatively consistent between the different determinations, with TCR-3 and -5 being the best positioned in the ranking (lowest EC50 values), while TCR-4 showed the lowest functional avidity (Table 4). Finally, the classifications of the TCRs in terms of avidity and functional avidity, were quite consistent, except for TCR-5, the TCR with the lowest avidity (according to tetramer staining), but which instead had very high functional avidity when tested with peptide-pulsed T2 cells. Characterization of different tumor cells derived from liver cancer in terms of the expression of GPC3 and HLA-A2 for functional studies of the selected TCRs

To study the capacity of the different TCR-T cells to recognize naturally processed pGPC3(520-530) in the context of the HLA-A2 molecule, we selected seven tumor cell lines, five of them (HEPG2, PLCPRF5-A2, HEP3B-A2, HUH7-A2, and SKHEP1) were liver-tumor derived cells, A431-A2 was derived from A431 cells (epidermoid carcinoma), and COS7-A2 cells was derived from COS7 cell lines (fibroblast-like african green monkey CV-1 cells transformed with SV40 encoding for wild type T antigen). Of the seven chosen cell lines, two of them (HEPG2 and SKHEP1) naturally expressed the HLA-A*02:01 allele (Table 5) and the other five tumor cell lines (PLCPRF5-A2, HEP3B-A2, HUH7-A2, COS-A2 and A431-A2) were retrovirally modified from parental cells to express this allele. Briefly, these cells were transduced with MSCV-HLA(A2)-IRES-Thy1 .1 retroviral particles (produced in PLAT-A cells) to generate PLCPRF5-A2, Hep3B-A2, HuH7-A2, COS7-A2 and A431-A2. MSCV-HLA(A2)-IRES-Thy1.1 vector was produced by inserting the human HLA-A*02:01 coding regions (synthesized by GenScript) into MSCV-IRES-Thy1.1 plasmid (Addgene plasmid # 17442). Thy1 .1 encodes for CD90.1 molecule and works as reporter gene. Four days after infection, cells were assessed for transduction efficiency by measuring the expression of CD90.1 with Alexa Fluor(AF)647-labelled anti-mouse CD90.1 mAb (clone OX-7). Transduced (CD90.1⁺) cells were further sorted by FACS and expanded.

Table 5. HLA class I genotype of the different tumor cell lines used. Only HEPG2 and SKHEP1 naturally express HLA-A*02:01 allele. (') Ambiguity on four-digit level. Chosen allele had the highest number of reads. Information was obtained from TRON Cell Line Portal (http://celllines.tron-mainz.de).

Surface expression of HLA-A2 was quantified by assessing median fluorescence intensity (MFI) of HLA-A2 molecules by flow cytometry. Briefly, cells were stained with PE-labeled anti-human HLA-A2 mAb (clone BB7.2) and analyzed by FACS. The classification of cells according to their HLA-A2 MFI was: high levels of HLA-A2 (PLCPRF5-A2, A431-A2 and C0S7, MFI in the range of 1-2 10⁴), Intermedium levels of HLA-A2 (HEPG2, SKHEP1 and SKHEP1-GPint, MFI in the range of 3-7 10³), low levels of HLA-A2 (HEP3B-A2 and HUH7-A2, MFI in the range of 0.2-1 10³) and null expression of HLA-A2 (MFI <100) (Table 6).

Table 6. Analysis of HLA-A2 and GPC3 expression in tumor-cell lines used in this study.

Parental (*) and genetic modified cells (a and b) are shown, (a) Those tumor cells negative for HLA-A*02:01 allele (PLCPRF5, HEP3B, HUH7, COS7 and A431) were genetically modified cells to express this allele, (b) SKHEP1 were genetically modified cells to express human GPC3. A clone with intermedium expression of GPC3 (SKHEP1-GP'^nt) was selected, (c) HLA-A2 surface expression assessed by FACS. Data are shown as the median fluorescence intensity (MFI) of HLA-A2 staining ± SD (3 replicates), (d) GPC3 mRNA quantification by qRT-PCR. Numbers refer to the PCR cycle threshold (CT) for each cell line ± SD (3 replicates).

Despite efficient transduction of HEP3B-A2 and HUH7-A2 cells (90.2% and 80%, for HEP3B-A2 and HUH7- A2 cells respectively, measured by detection of the CD90.1 reporter protein simultaneously expressed with HLA-A2 molecules by a bicistronic IRES element -HLA-A2-\RES-Thy1.1-) these cells expressed lower surface levels of HLA-A2 than other transduced tumor cells lines. This resembled the surface staining of total HLA- A/B/C molecules on parental cell lines (Fig. 5A). The surface expression of class I HLA molecules depends on the expression level of HLA-I genes but also key components of the antigen-presenting machinery (APM). Notably, transcriptome analysis of these cell lines in the TRON cell database showed impaired expression of some AMP-related genes in HEP3B cells and HUH7, as compared to other cell lines (Table 7), which may account for the lower expression of natural HLA-I molecules as well as transgenic HLA-A2 in HEP3B-A2 and HUH7-A2 cells.

Table 7. Gene expression of the most important antigen-presenting machinery (APM) and HLA class I genes in the different tumor cell lines used. Gene expression is expressed as reads per kilobase of transcript per million reads (RPKM) mapped. Information was obtained from TRON Cell Line Portal

(http://celllines.tron-mainz.de). Of the seven tumor cell lines chosen, we show data for six of them. The C0S7 cell line is not registered in the TRON Cell Line Portal. For each gene, the underlined digits refer to the lowest values detected in the tumor cell col ection used in this study.

Regarding GPC3 expression, six cell lines (HEPG2, PLCPRF5-A2, HEP3B-A2, HUH7-A2, COS7-A2 and A431-A2) naturally expressed GPC3 and one tumor cell line (SKHEP1) did not (Nakatsura et al). We retrovirally transduced SKHEP1 to overexpress human GPC3. Briefly, these cells were transduced with MSCV-GPC3-IRES-Thy1.1 retroviral particle (produced in PLAT-A cells). MSCV-GPC3-IRES-Thy1.1 vector was produced by inserting the human GPC3 coding region (synthesized by GenScript) into MSCV-IRES- Thy1 .1 plasmid. Transduced cells were single cell sorted as CD90.1⁺ cells by FACS upon staining with AF647-labelled anti-mouse CD90.1 mAb. Single sorted cells were expanded and a clone with intermedium expression of GPC3 (SKHEP1-GPC3'^nt) was selected. To better characterize these tumor cells, we assessed the expression of GPC3 by quantitative reverse transcription polymerase chain reaction (qRT-PCR). Briefly, Tumor cells (5 x 10⁵) were resuspended in 200 piL of Homogenization buffer containing 20 uL of 1- Thioglycerol per mL (Promega). Total cellular RNA was isolated using Maxwell RSC simply RNA tissue kit (Promega) according to the manufacturer's instructions. Extracted RNA was quantified using a UV spectrophotometer (Thermo Scientific™ NanoDrop™). RNA isolated from the tumor cell lines (200 ng/sample) was treated (37°C/20') with DNase I (1 Ul/ piL) (Invitrogen) and RNaseOUT (40 U/ piL) (Invitrogen). The reaction was stopped by adding EDTA (1 mM) and incubating for 10' at 35°C and then for T at 90°C. RT reaction was performed (37°C/1h) with Moloney murine leukemia virus (MLV) reverse transcriptase (120 UI/piL) (Invitrogen). The qPCR was performed in a CFX96™ Real-Time PCR Detection System (BIO-RAD) with 10 ng of the cDNA preparation and 300 nM of validated primers in 10 l IQ SYBR Green Supermix (BIORAD). The following program was used: 95°C/3', 45 cycles of amplification (95°C/15”, 58°C/15”, 72°C/25”), and 80°C/10”. Experiments were performed in triplicate. Primers used were: (F-GPC3, SEQ ID NO: 33) 5'- TTCTCAACAACGCCAAT -3' and (R-GPC3, SEQ ID NO: 34) 5'- GATGTAGCCAGGCAAAGC -3'. The cycle threshold (CT) value was recorded. CT is defined as the number of cycles required for the fluorescent signal to cross the threshold. Ct levels are inversely proportional to the amount of target nucleic acid in the sample. Cts < 25 are strong positive reactions indicative of abundant target nucleic acid in the sample. Cts of 25-35 are positive reactions indicative of moderate amounts of target nucleic acid. Cts > 35 are weak reactions indicative of minimal amounts of target nucleic acid which could represent environmental contamination As shown in table 6, HEPG2, HEP3B-A2, and HuH7-A2 exhibited the highest GPC3 mRNA expression (CT values between 16.5 and 16), followed by PLCPRF5-A2, COS7-A2, SKHEP1-GPC3int and A431-A2 cells (CT values between 23.7 and 29.89) and SKHEP1 cells (CT values ~ 35), which were on the limited of detection.

In summary, we have a battery of tumor cells, which differ in terms of the expression levels of HLA-A2 and GPC3 molecules (Fig. 5B), which will be very useful to study the capacity of the different GPC3 TCRs to recognize naturally processed GPC3(522-530) epitope in the context of the HLA-A2 molecule.

Reactivity of the TCRs for GPC3⁺HLA-A2⁺ Tumor cells

Then, we tested the different TCR-T cells against the GPC3⁺HLA-A2⁺ tumor cell lines. As controls, we used GPC3⁺ HLA-A2- (PLCPRF5, HEP3B and HUH7) and GPC3- HLA-A2⁺ (SKHEP1) target cells (Fig. 6). Briefly, TCR-T cells (containing 5x10⁴ mTCRp⁺ cells/well) and target cells (5x10⁴ cells/well) were co-cultured (37 °C) in a 96-well round-bottom plate in 200 pL of T-cell media. Twenty-four hours later, cells were harvested and stained to assess CD137 expression by FACS. TCR-3 and TCR-5 proficiently recognized HEPG2 and PLCPRF5-A2 cells (as depicted by CD137 expression). Interestingly, TCR-4, classified as an intermedium/low avidity/functional avidity TCR, recognized all GPC3⁺ HLA-A2⁺ tumor cells tested, including GPC3⁺HLA-A2⁺ cell lines expressing low levels of HLA-A2⁺ (HEP3B-A2 and HuH7-A2) (Fig. 6). Compared to TCR-3 and TCR- 5-engineered T cells, TCR-4 T cells also exhibited higher levels of activation against those targets commonly recognized by these three TCRs (Fig. 6). Interestingly, while TCR-3 and TCR-4 did not recognize parental HLA-A2 negative PLCPRF5 and HEP3B cells, TCR-5 did recognize PLCPRF5 cells.

In summary, from the three cloned TCRs, 3 TCRs (TCR-3, TCR-4 and TCR-5) recognized GPC3⁺ HLA-A2⁺ tumor cells, with TCR-4 being the most reactive one. The better tumor-reactivity of TCR-4 was further confirmed when it was tested against other HLA-A2-engineered cell lines naturally expressing low levels of GPC3, such as COS7 and A431 (Fig. 7). In summary, despite being classified as an intermediate avidity/functional avidity TCR, TCR4 was the most reactive TCR against GPC3⁺HLA-A2⁺ tumors, being able to recognize even tumor cells expressing low levels of GPC3 (GPC3 CT values of >29.89) or low levels of HLA- A2 (HLA-A2 MFI > 267). For the following experiments, we focused on TCR-3, TCR-4 and TCR-5.

GPC3-specific recognition of GPC3⁺ tumor cell lines by GPC3 TCRs

To verify the GPC3 specificity of TCR-3, -4, and -5, two different short hairpins (sh) RNAs were designed to silence GPC3 expression (shGPC3-630 (SEQ ID NO: 36) and shGPC3-1344 (SEQ ID NO: 37)) (Table 8).

Table 8. Selection of short hairpin RNA sequences complementary to GPC3 [sh(GPC3], The splashRNA software ( Nat. Biotechnol. 35, 350-353 (2017)) was used to select sh-RNA-GPC3 sequences. As input, we used the transcript variant 1 of human GPC3 (NM_001164617.2). The selection criteria was: (i) being among the top 10 (score > 1 .4). (ii) being present in the other three GPC3 transcript variants (NM_004484.4, NM_001164618.2 and NM_001164619.2) and (iii) being included in the CDS region. The shGPC3 sequences fulfilling these criteria were those at position 630 (shGPC3-630, SEQ ID NO: 36) and 1344 (shGPC3-1344, SEQ ID NO: 37). shGPC3-630, shGPC3-1344 were designed, synthesized (GenScript) and inserted into pLentiN plasmid, which contains a pre-built hairpin structure to clone the sense and antisense sequences directly and a blasticidin resistance gene. As a negative control, we used a pLentiN plasmid expressing sh- RNA to Renilla mRNA (shREN, SEQ ID NO: 35) (antisense 5'-TAGATAAGCATTATAATTCCTA-3') that do not suppress the expression of genes expressed in humans and provide a baseline for the experiments.

For GPC3 silencing using shRNA, lentivirus encoding shRNA complementary to GPC3 (shGPC3-630 and shGPC3-1344) (Table 8) or renilla (shREN) were generated using the Lenti-X 293T cells (Takara Bio) (6 x 106 cells in 10 mL) transfected with 6.9 g of the different sh-RNA- plasmid along with 3.41 pg of the lentiviral packaging plasmid pMDLG/pRRE (Plasmid #12251 , Addgene), 1.7 pg of pRSV-Rev plasmid (Plasmid #12253, Addgene) and 2 pg of the helper envelope plasmid pMD2.G (Plasmid #12259, Addgene), in the presence of lipofectamine. The lentivirus-containing cell culture medium was collected and immediately used to transduce target cells (HEPG2, SKHEP1 -GPC3int, PLCPRF5, PLCPRF5-A2, and HEP3B-A2) in the presence of polybrene (Thermo Fischer) (10 pig/ml). Three days later, cells started to be selected with 10 pg/ml blasticidin for 21 days. From day 22, cells were maintained with 1 pg/ml blasticidin. GPC3 silencing was assessed on day 24 by RT-PCR. The most effective shRNA silencing GPC3 was shGPC3-1344 (Fig. 8A). This was consistent with shGPC3-1344 having a higher score than shGPC3-630 in the splashRNA software (Table 8).

According to RT-PCR data, the HEPG2 cells exhibited the highest GPC3 silencing (90%), followed by PLCPRF5 and PLCPRF5-A2 (82% and 84% respectively), SKHEP1-GPC3^int (81 %) and HEP3B-A2 (50%). GPC3 silencing with shGPC3-1344 was also confirmed by staining of total GPC3 protein with anti-GPC3 mAb (clone YP-7) and FACS analysis (Fig. 8B). Briefly, for total GPC3 staining, cells were first permeabilized and fixed with cytofix/cytoperm (BD Biosciences) and then incubated (4°C/30') with the anti-human GPC3 mAb (clone YP-7, mouse lgG1) in the presence of Beriglobin P, as Fc Block. As a negative control, a control isotype (Isot) (mouse lgG1 MOPC-21) was used. Cells were subsequently stained (4°C/30') with secondary APC-labelled anti-mouse lgG1 mAb (RMG1-1). As depicted in figure 8B, HEPG2 cells showed the highest GPC3 silencing (-80%), followed by HEP3B-A2 (-60%) and SKHEP1-GPC3^int, PLCPRF5-A2 and PLCPRF5 (-35%). According to this results, shGPC3-1344 was selected to silence GPC3.

To assess the GPC3 specificity of the selected TCRs, GPC3-silenced tumor cell lines were cocultured (6 hours) with TCR-3, -4 and -5 T cells and the response was evaluated by measuring CD137 expression and I FNy release (Fig. 8C). The intensity of the T-cell response against GPC3-silenced target cells was lower than that against control cells (cells treated with shREN). The decrease was more pronounced in I FNy production. Recognition of HEPG2 and PLCPRF5-A2 by TCR-3 T cells was abrogated upon GPC3 silencing. Recognition of GPC3-silenced HEPG2 by TCR-4 and -5 was also seriously impaired. However, TCR-4 T cells still mounted a significant response (as depicted by CD 137 expression) against GPC3-silenced PLCPRF5-A2, HEP3B-A2 and SKHEP1-GPC3'^nt cells, and the same happens to TCR-5 in response to GPC3-silenced PLCPRF5-A2 and PLCPRF5 cells. The partial inhibition of GPC3 expression at the protein level in these cell types (Fig. 8B) and the ability of some TCRs to recognize cells expressing low GPC3 levels (such as TCR-4) may be the cause of the residual response. The fact that TCR-4 was not able to recognize SKHEP1 cells (Fig. 6), non-expressing GPC3 (Table 6), confirm its specificity for GPC3. In summary, our data indicate that the target antigen of TCR-3, -4 and -5 in GPC3⁺ tumor cell lines is GPC3.

Identification of HLA class I molecules presenting GPC3 to TCR-3, -4 and -5

Assays with tetramers and T2 cells, both using soluble pGPC3(522-530), indicated that the selected TCRs recognized this epitope in the context of HLA-A2 (Fig. 4B). In addition, assays with HLA-A2⁺ and HLA-A2’ cells naturally expressing GPC3 indicated that TCR-3 and -4 also recognized the endogenously processed epitope in the context of the HLA-A2 molecule (Fig. 6). However, TCR-5 T cells were also capable of recognizing GPC3⁺HLA-A2’ tumor cells, such as PLCPRF5 (Fig. 6), this recognition being specific to GPC3, as verified with the silencing of GPC3 (Fig. 8C). To further study the HLA-I restriction of TCR-3, -4 and -5 against the endogenously processed peptide, we cultured TCR-T cells with GPC3⁺HLA-A2'^,/’ tumor cells in the presence of anti-pan-HLA-l or anti-HLA-A2 blocking mAbs. Briefly, GPC3⁺HLA-A2'^,/’ tumor cells (5x10⁴ cells/well) were incubated (1 h/37°C) with anti (a)-pan-HLA-l (clone W6/32), O-HLA-A2 (clone BB7.2) or control IgG (mouse lgG1) mAbs (10 pg/mL) before co-culture (24h/37°C) with TCR-T cells (containing 5x10⁴ mTCRp⁺ cells/well). The mAbs were present during all the culture period. Cells and supernatant were recovered and the response was evaluated by measuring surface CD137 expression in mTCRp⁺ cells by FACS and the production of I FNy by ELISA. As depicted in figure 9, the ability of TCR-3 to recognize HEPG2 and PLCPRF5-A2 cells, and that of TCR-4 to recognize HEPG2, SKHEP1-GPC3^int, PLCPRF5-A2 and HEP3B-A2 cells, was abrogated in the presence of anti-pan-HLA-l and anti-HLA-A2 blocking mAbs, confirming the HLA- A2-restricted recognition of these tumor cells lines by these TCRs. Regarding TCR-5, whereas anti-pan-HLA- I mAb abrogated recognition of HEPG2, PLCPRF5-A2 and PLCPRF5, anti-HLA-A2 mAb did not affect recognition of HEPG2 and PLCPRF5 and partially inhibited recognition of PLCPRF5-A2 by TCR-5. Altogether, our data suggest that TCR-5 can recognize GPC3 in the context of HLA-A2 molecule and another HLA-I molecule shared by HEPG2 and PLCPRF5-derived cells. The only HLA-I allele shared between the two cell lines is HLA-C*04:01 (coding HLA-C4 molecule) (Table 5), for which the NetMHCPan program (Reynisson et al) predicts that pGPC3(522-530) (FLAELAYDL) could be a weak binder (Table 9). These data suggest that the HLA-I molecule that could be responsible for the recognition of PLCPRF5 by TCR-5, as well as the HLA- A2-non-restricted recognition of HEPG2 and PLCPRF5-A2 by this TCR, is HLA-C4. However, given the good prediction of pGPC3(522-530) binding to HLA-C17*01:01 (present only in PLCPRF5-derived cells), we cannot rule out that part of the recognition of PLCPRF5-derived cells by TCR-5 was also mediated by this allele. Further experiments are necessary to confirm these hypotheses.

Table 9. HLA-I molecules and peptides containing the GPC3(522-530) epitope (FLAELAYDL) that may be presented by HEPG2 and PLCPRF5 cells. Using the NetMHCpan4.1 algorithm, we predicted the ability of the peptide GPC3(520-532) (LRFLAELAYDLDV) to generate epitopes of 8 to 11 amino acids that were capable of binding to the HLA-I alleles expressed by HEPG2 and PLCPRF5 (Table 5). (^a) Cell line expressing the indicated HLA-I allele. The only HLA allele shared between HEPG2 and PLCPRF5 cell lines is HLA- C*04:01 . (^b) Only the peptides with the best binding score (categorized as strong binders [SB] or weak binders [WB]) and/or Affinity (Aff)< 50 nM are shown.

TCR-4 modified T cells stood out for their high proliferative capacity after recognition of GPC3⁺HLA- A2⁺ tumor cells

To better characterize TCR-3, -4, and -5, we studied the ability of human T cells genetically modified with these TCRs to proliferate upon recognition of tumor cells. First, we performed an assay with decreasing amounts of HEPG2, PLCPRF5-A2 or PLCPRF5 cells to find a target cell/effector cell ratio that would allow comparative studies between the different TCR-T cells. Briefly, TCR-3, -4 and -5-T cells (5x10⁴ mTCRp⁺ cells/well) were co-cultured with serial dilutions of target cells (from ratio 1 :1 to 1 :16) for 24h at 37 °C. Cells were then harvested and stained to evaluate surface detection of CD137 in TCR|3⁺ CD8 T cells by FACS. As observed in figure 10A, TCR-engineered T cells responded in a dose-dependent manner, as depicted by the expression of the CD137 marker. Confirming previous observations, TCR-4 exhibited the highest reactivity against both HEPG2 and PLCPRF5-A2, followed by TCR-5 and TCR-3. In addition, TCR-5 responded against PLCRF5 cells, but at lower levels compared to its response to PLCPRF5-A2. Based on these results, we chose the target cell/effector cell ratio 1/4 to study the proliferative capacity of TCR-T cells upon recognition of tumor cells. Thus, TCR-T cells were labeled (RT/157darkness) with CellTrace Violet (CTV, Biolegend) (5 uM). Stained TCR-T cells (1 x10⁵ mTCRp⁺ cells/well) were cultured (37 °C/96h) alone (medium) or with irradiated (20,000 rads) HEPG2, PLCPRF5-A2 or PLCPRF5 cells (2.5 x10⁴ cells/well) in the presence or not of anti (a)- pan-HLA-l mAbs (10 pg/mL). IL-2 (10 U/ml) was added 48 h post-stimulation. Ninety-six hours later, cells were collected and surface stained with anti-CD8 and anti-mTCRp mAb and then analyzed by FACS. TCR-4- engineered T cells stood out for their high proliferative capacity in response to HEPG2 and PLCPRF5-A2 cells (Fig. 10B). TCR-5 T cells were the only ones capable to proliferate upon stimulation with PLCPRF5 cells. In all cases, the proliferation was abrogated by blockade with anti-pan-HLA-l mAbs.

TCR-4 and TCR-5 assembled comparable effector functions on TCR-T cells.

To study how the expression of the selected GPC3 TCRs allowed human T lymphocytes to mount effector responses against the GPC3⁺HLA-A2^+A target cells, we analyzed the production of effector molecules (cytokines and cytotoxicity-related molecules) as well as the killing capacity of TCR-T cells. The release of I FNy, IL-2 and TNFa was first analyzed by ELISA. As shown in figure 11, the PLCPRF5-A2 cell line was the one that stimulated a higher production of cytokines, with TCR-4 T cells being the best responders to this cell line. In contrast, TCR-5 was distinguished by being the one that induced the best cytokine response against HEPG2 and the only one that responded against PLCPRF5. The cytokine response of TCR-3- and TCR-4 T cells against HEPG2 and PLCPRF5-A2 was discontinued upon blockade of pan-HLA-l and HLA-A2 molecules. Confirming previous observations, anti-pan-HLA-l mAbs blocked the cytokine release by TCR-5 T cells to all three target cells, while anti- HLA-A2 partially inhibited the response to PLCPRF5-A2 cells but not to HEPG2 and PLCPRF5 cells.

To characterize the cytotoxic capacity of the three TCR-T cell lines, we cultured them with tumor cells and analyzed Granzyme B (GzmB) release by ELISPOT. Briefly, GPC3⁺HLA-A2^+A tumor cells were incubated with anti-pan-HLA-l mAbs or with medium before co-culture with TCR-T (or UTD) cells in a nitrocellulose-backed multiscreen (PVDF) ELISPOT plates (Millipore; cat no. MSIP4w10) previously coated with anti-human GzmB capture mAb (GB10, Mabtech) (15 pg/mL in PBS). As a negative control, TCR-T cells were incubated with medium alone. On the following day, cells were washed from ELISPOT plates and wells were incubated (RT/2h) with biotinylated anti-human GzmB detection mAb (GB-11 , Mabtech), then with streptavidin, alkaline phosphatase conjugate (SAV-ALP) (Mabtech) (RT/1 h) and finally developed with BCIP/NBT substrate (Mabtech) (RT/10-15'). The enzymatic reaction was stopped with water and the plates were allowed to dry. Spots were scanned and counted using an ImmunoSpot plate reader and associated software (Cellular Technologies). As depicted in figure 12A, TCR-4 and TCR-5 T cells secreted comparable levels of GzmB against HEPG2 and PLCPRF5-A2. TCR-5 T cells were the only one producing GzmB in response to PLCPRF5 cells. Eventually, GzmB release was suppressed in the presence of pan-HLA-l blocking mAbs.

We next examined the killing activity of the three TCR-expressing T cells using an xCELLigence Real-Time Cell Analyzer-Multiple Plate system. Briefly, HEPG2 (2 x 10⁴ cells/well) and PLCPRF5-derived cells (5 x 10³ cells/well) were plated (100 l/well) overnight in tumor-cell medium [MEM-glutamax (GIBCO), 10% Fetal Bovine Serum (FBS) (SIGMA), 1% sodium pyruvate (GIBCO), 1% non-essential amino acids (GIBCO), 10 mM of HEPES(GIBCO), 100 U/mL Penicillin/Streptomycin (P/S) (GIBCO), and 10 pig/mL gentamicin (GIBCO)] into xCELLigence E-plate 96-well flat-bottom plates. Before adding target cells, a baseline impedance measurement was taken. At 24h of culture, anti (a)-pan-HLA-l or O-HLA-A2 (final concentration 10 g/mL) mAbs in T-cell medium or T-cell medium (50 pl/well) were added and incubated for 1 hour. Then, TCR- engineered (containing 5 x 10⁴ mTCRp+ T cells/well) or UTD T cells were added in T-cell medium (50 pl/well) and impedance measurements were recorded each 15-minute intervals for up to 50 hours. As control, T-cell medium (50 pl/well) without TCR-T cells were also added to tumor cells. The cell index curve was normalized to the time point at which T cells were added using RTCA Software Pro. Of the three TCR-T cell lines, those engineered with TCR-3 showed the least killing activity against GPC3⁺HLA-A2⁺ target cells (Fig. 12B and C). The most sensitive target cell line to be lysed by TCR-T cells was HEPG2, with 15 h of culture being sufficient for its complete lysis by TCR-4- and TCR-5 T cells (Fig. 12B). Otherwise, PLCPRF5-derived cells required > 35 h of the assay to be completely lysed (Fig. 12C). TCR-4- and TCR-5 T cells showed comparable killing activity to HEPG2 cells. In contrast, TCR-5 T cells were slightly more effective in killing PLCPRF5-A2 target cells, as evidenced by their enhanced cytotoxic activity at early times (15 h), as compared to TCR-4 T cells. The killing activity of all three TCR-T cell lines was inhibited by blocking with anti-pan-HLA-l mAbs (Fig. 12B and C). Regarding the role of HLA-A2⁺ in the lysis of PLCPRF5-A2 cells, its blockade completely prevented the lysis of these cells by TCR-3- and TCR-4 T cells, but only partially in the case of TCR-5 (Fig. 12C), confirming once again that the recognition of PLCPRF5-A2 cells by TCR-5 was restricted by HLA-A2⁺ and other HLA-I molecules. Finally, TCR-5-engineered T cells were the only ones capable of killing PLCPRF5 cells. The lytic activity of TCR-5 against the parental PLCPRF5 cell line was lower than against PLCPRF5-A2 cells. The possibility of recognizing the GPC3 through two different HLA-I molecules would have allowed TCR-5 T cells to establish more intercellular junctions with PLCPRF5-A2 cells, which would explain the better lysis of this cell line compared to PLCPRF5. The same explanation would account for the greater lysis of PLCPRF5-A2 cells by TCR-5 T cells, compared to that of TCR-4. Together, TCR-4 and TCR-5 showed the best effector function profiles.

TCR-4 cells efficiently control tumor progression thanks to their increased engraftment capacity and ability to mount a vigorous response after adoptive transfer.

To assess the therapeutic efficacy of GPC3-specific TCRs, NSG mice bearing PLCPRF5-A2 tumors were treated with TCR-4 T cells. We chose this TCR for the in vivo testing because it was the TCR that endowed the modified T cells with the greatest proliferative capacity after stimulation with GPC3⁺HLA-A2⁺ tumor cells and for assembling potent effector functions. Briefly, 6-8-week-old gender-matched NSG mice were subcutaneously (sc) implanted with 2.5 x 10⁶/mouse PLCPRF5-A2 human hepatocellular tumor cells into the right flank. Eight days post tumor implantation, tumors were measured with calipers, and mice were distributed among the different treatment cohorts so that the mean tumor size and standard deviation were similar among all groups. On day 9, mice were adoptively transferred with TCR-4 CD8 T cells containing 8 x 10⁶ mTCR|3⁺ cells injected intravenously (i.v.) by the retro-orbital route. Control groups received a number of UTD CD8 T cells similar to the total number of CD8 T cells in the TCR-treated groups from the same donor. Similar to how it is done in cancer patients treated with TCR-transduced T cells (Leidner et al), mice received IL-2 by systemic route to support the expansion of transferred cells (Fig. 13A). Briefly, human IL-2 (Proleukin) was administered intraperitoneally (ip) (4 xio⁴ lU/mouse) on days 1, 2, 3, 5, 7, 9, 11, 13 and 15, and sc (near to the tumor site) (2 xio⁴ lU/mouse) on days 3, 5, 7, 9, 11, 13 and 15 upon T-cell transfer. The tumor growth rate was determined by measuring the perpendicular diameters of tumors three times per week using digital calipers. The survival rate was also monitored. In line with ethics requirements (protocol 048-21), mice were sacrificed when they showed one or more of the following criteria: mean diameter [(major length + minor length)/2] of the tumor reaching 18 mm, ulcerated/necrotic tumor and/or physical impairment (impaired mobility, signs of lethargy, lack of physical activity and weight loss). As depicted in figures 13B-D, TCR-4 T cells were able to cure all treated mice.

Example 2

Alternative GPC3 peptides and peptide/HLA complexes recognized by GPC3 TCR

The fact that TCR-4 showed greater reactivity against GPC3-expressing tumor cells than against cells pulsed with pGPC3(522-530) suggested that the endogenous GPC3 epitope may be somewhat different from the minimal predicted epitope. Interestingly, an in-silico analysis using NetMHCpan4.1 algorithm showed that the 11-mer FLAELAYDLDV (SEC ID NO: 100) and the 10-mer RFLAELAYDL (SEC ID NO: 102) peptides containing the minimal epitope have also good predicted HLA-A2-binding affinity (< 50 nM), with the 11-mer peptide exhibiting also a good predicted binding score (<1.5%) (Table 10).

Table 8. HLA-I molecules and peptides containing the GPC3(522-530) epitope (FLAELAYDL) that may be presented by HEPG2, PLCPRF5-A2 and PLCPRF5 cells.

Predicted

MHC Peptide (a) Aa length Affinity rank Aff (nM) Bind Level

HLA-A*02:01 FLAELAYDL 9 0.05 % 2.98 SB

FLAELAYDLDV 11 1.45% 23.66 WB

RFLAELAYDL 10 2.54% 46.06 (-)

HLA-C*04:01 FLAELAYDL 9 0.98% 4690.87 WB

Empirically, both peptides stabilized the surface expression of HLA-A2 in T2 cells, with the 11-mer peptide showing the highest binding capacity, being even better than the 9-residue peptide (Fig. 14A). Interestingly, while TCR-3 and -5 recognized the 11-mer peptide much worse than those with 9 and 10 amino-acid length, TCR-4 recognized the 11-mer peptide as well as, or slightly better than, the 9-mer peptide. On the contrary, TCR-4 showed low reactivity towards the 10-mer peptide (Fig. 14B). These data suggest that the TCR-4 may recognize a different epitope configuration that it is conserved, or even better reflected, in the 11-mer peptide.

Although it is generally accepted that MHC-I molecules bind peptides of 8-11 amino acid lengths, longer peptide binders have also been observed. Using unbiased mass spectrometry, a 20-mer peptide derived from GPC3 containing the minimal epitope (FLAELAYDLDVDDAPGNSQQ; SEQ ID NO: 101) has been identified in the immunopeptidome of hepatocytes isolated from an HOC patient carrying the HLA-A*02:01 allele (de Beijer et al., 2022). This peptide was able to stabilize the surface expression of HLA-A2 in T2 cells, although much less efficiently than the 9-mer peptide (Fig. 14C). Interestingly, TCR-4 was the only TOR showing some reactivity against this peptide (Fig. 14D), supporting the idea that it may recognize a different epitope conformation. Regarding TCR-5, our data suggested that this TCR could also recognize GPC3 in the context of another HLA-I molecule common to HEPG2 and PLCPRF5-derived cells. The only HLA-I allele shared by these cell lines was HLA-C*04:01 (coding HLA-C4 molecule) (Table 9).

Table 9. HLA class I genotype of the different tumor cell lines used.

(') Ambiguity on four-digit level. Chosen allele had the highest number of reads. Information was obtained from TRON Cell Line Portal (http://celllines.tron-mainz.de). Of the seven tumor cell lines chosen, we show data for six of them. The primate C0S7 cell line is not registered in the TRON Cell Line Portal.

To assess anti-GPC3 reactivity of TCR-5 in the context of HLA-C4 molecule, we generated HLA-C*04:01⁺, TAP-deficient cells (721-C4-ICP47) by sequentially transducing the HLA-I negative cell line 721.221 with HLA- C*04:01/ p -2 microglobulin (|32m) and then with the TAP inhibitor ICP47 (Fig. 15A). As control, we also generated HLA-A*02:01⁺, TAP-deficient 721.221 cells (721 -A2-ICP47). pGPC3(522-530) stabilized HLA-C4 on the cell surface, although much more weakly than HLA-A2 under similar conditions (Fig. 15B). Interestingly, TCR-5 T cells, but no TCR-3 and TCR-4 T cells, recognized pGPC3(522-530) in the context of HLA-C4 molecules but less efficiently that when the peptide was presented by HLA-A2 (Fig. 15C). These data suggest that the non-HLA-A2-restricted, GPC3-specific recognition of HEPG2 and PLCPRF5-derived cells by TCR-5 may be mediated by HLA-C4 molecule.

TCR-4 T cells outperformed CAR-T cells in ACT schedules

In order to compare TCR-4 with a GPC3-specific CAR equivalent to one currently being tested in clinical trials (NCT05003895), we constructed an YP7-based second-generation 4-1 BB CARs (Zhang & Ho, 2016) (Sun et al., 2021) and packaged into the retrovirus vector used for the expression of GPC3-TCRs. The gene encoding the truncated version of EGFR (EGFRt, GenBank accession no. AH010139.2) was also inserted as reporter system. Human CD8 T cells were efficiently transduced with the retrovirus GPC3-CAR and the CAR exhibited high binding activity to GPC3-Fc chimera protein (Fig. 16A). Next, GPC3-CAR and TCR-4 transduced cells were isolated and expanded to better compare their effector and antitumor properties (Fig. 16B). First, we confronted the engineered T cells to HEPG2 and PLCPRF5-A2 cells, expressing high and intermediate levels of surface GPC3, respectively, according to the staining with YP7 anti-GPC3 mAb (Fig. 16C) and published data (Yu et al., 2018). Surface expression of the activation marker CD 137 revealed that GPC3-CAR-T cells recognize HEPG2 cells better than PLCPRF5-A2 cells. Although without prominent differences, at high tumor:T cell ratios CAR-T cells appeared to recognize HEPG2 cells better than TCR-4 T cells, and the opposite was true at low ratios (Fig. 16D). TCR-4 T cells were better than CAR-T cells at recognizing PLCPRF5-A2. From now on, the HEPG2 cell line was chosen for the rest of the comparative studies. When the cytotoxicity capacity was compared it was observed that GPC3-CAR T cells lysed HEPG2 cells very quickly, reaching a plateau (90% lysis) after 6-7 hours of culture. In contrast, TCR-4 T cells killed more slowly, reaching the maximum lytic capacity of CAR-T cells at 11-12 hours, and were able to continue lysing tumor cells up to 100% cytotoxicity at 16 hour (Fig. 16 E and F). Next, we compared TCR-4 and GPC3-CAR T cells in ACT experiments with HEPG2-tumor bearing mice. Interestingly, during the first 6 days of treatment, CAR-T cells showed slightly better control of tumor growth compared to TCR-T cells, but then they lost efficacy and TCR-4 T cells proved to be better at long-term tumor control (Fig. 16G and H). Our data regarding anti-tumor efficacy of YP7-based CAR-T cells in the HEPG2 xenograft model were comparable to published data.

To examine the causes of the different behavior of TCR-4 and GPC3-CAR T cells in ACT schedules, we analyzed transferred T cells in the blood and in the tumor. The total number of circulating CD8 T cells increased between days 2 and 6 upon transfer, with this rise being significantly greater in the group receiving GPC3-CAR (Fig. 17A). However, on day 12 blood CD8 T cells levels were maintained in the TCR-4 group, while they decreased abruptly in the GPC3-CAR group. At day 13, mice were sacrificed and transferred CD8 T cells were analyzed in the tumor infiltrate. As depicted in figure 17B, the percentage of human CD8 TILs in the group treated with TCR-4 T cells was higher than in the GPC3-CAR group. In addition, TCR-4 TILs expressed higher levels of Gzmb and ki-67 than GPC3-CAR T cells, indicating a greater lytic and proliferative capacity in tumors (Fig. 17C). They also exhibit intermedium level of expression of PD-1 molecule, suggesting that they did not reach a state of overactivation (Fig. 17C). Interestingly, TCR-4 T cells could be detected in the bone marrow, but not in the blood, of cured mice on day 102 after ACT, indicating that these cells engrafted efficiently and possibly differentiated into long-term memory T cells (Fig. 17D). A higher level of exhaustion of GPC3-CAR-T cells, which is possibly hindering their cytotoxic capacity, proliferation and, ultimately, their persistence, together with their lower sensitivity to respond to low antigen densities, could explain the worse performance of GPC3-CAR-T cells, compared to TCR-4 T cells, in ACT.

Sequence list:

Citation List

- Needleman and Wunsch, J. Mol. Biol. 48: 443-453, 1970. DOI: 10.1016/0022-2836(70)90057-4) - EMBOSS: The European Molecular Biology Open Software Suite, Rice et al., Trends Genet. 16: 276-277,

2000. DOI: 10.1016/s0168-9525(00)02024-2)

Altschul et al., "Basic local alignment search tool”, 1990, J. Mol. Biol, v. 215, pages 403-410, 1990. DOI:

10.1016/S0022-2836 (05)80360-2

Bolotin, D.A., Poslavsky, S., Mitrophanov, I., Shugay, M., Mamedov, I.Z., Putintseva, E. v, and Chudakov, D.M. (2015). MIXCR: software for comprehensive adaptive immunity profiling. Nat Methods 12, 380-381. 10.1038/nmeth.3364.

Shugay, M., Bagaev, D. v, Turchaninova, M.A., Bolotin, D.A., Britanova, O. v, Putintseva, E. v, Pogorelyy, M. v, Nazarov, V.I., Zvyagin, I. v, Kirgizova, V.I., et al. (2015). VDJtools: Unifying Post-analysis of T Cell Receptor Repertoires. PLoS Comput Biol 11, e1004503. 10.1371/journal.pcbi.1004503. - Cohen, C.J., Zhao, Y., Zheng, Z., Rosenberg, S.A., and Morgan, R.A. (2006). Enhanced antitumor activity of murine-human hybrid T-cell receptor (TOR) in human lymphocytes is associated with improved pairing and TCR/CD3 stability. Cancer Res 66, 8878-8886. 10.1158/0008-5472. CAN-06- 1450.

Thomas, S., Xue, S.-A., Cesco-Gaspere, M., San Jose, E., Hart, D.P., Wong, V., Debets, R., Alarcon, B., Morris, E., and Stauss, H.J. (2007). Targeting the Wilms Tumor Antigen 1 by TCR Gene Transfer: TCR Variants Improve Tetramer Binding but Not the Function of Gene Modified Human T Cells. The Journal of Immunology 179, 5803-5810. 10.4049/jimmunol.179.9.5803.

Coren, L. v, Jain, S., Trivett, M.T., Chien, C., and Ott, D.E. (2015). Production of retroviral constructs for effective transfer and expression of T-cell receptor genes using Golden Gate cloning. Biotechniques 58, 135-139. 10.2144/000114265.

Nakatsura, T., Yoshitake, Y., Senju, S., Monji, M., Komori, H., Motomura, Y., Hosaka, S., Beppu, T., Ishiko, T., Kamohara, H., et al. (2003). Glypican-3, overexpressed specifically in human hepatocellular carcinoma, is a novel tumor marker. Biochem Biophys Res Commun 306, 16-25. 10.1016/S0006- 291X(03)00908-2.

Reynisson, B., Alvarez, B., Paul, S., Peters, B., and Nielsen, M. (2020). NetMHCpan-4.1 and NetMHCIIpan-4.0: improved predictions of MHC antigen presentation by concurrent motif deconvolution and integration of MS MHC eluted ligand data. Nucleic Acids Res 48, W449-W454. 10.1093/nar/gkaa379.

Leidner, R., Sanjuan Silva, N., Huang, H., Sprott, D., Zheng, C., Shih, Y.-P., Leung, A., Payne, R., Sutcliffe, K., Cramer, J., et al. (2022). Neoantigen T-Cell Receptor Gene Therapy in Pancreatic Cancer. N Engl J Med 386, 2112-2119. 10.1056/NEJMoa2119662.

Pelossof R., et al. (2017). Prediction of potent shRNAs with a sequential classification algorithm. Nat. Biotechnol. 35, 350-353. http://celllines.tron-mainz.de

Hillerdal V, Boura VF, Bjbrkelund H, Andersson K, Essand M. Avidity characterization of genetically engineered T-cells with novel and established approaches. BMC Immunol. 2016 Jul 13;17(1):23. doi: 10.1186/sl 2865-016-0162-z. de Beijer, M. T. A., Bezstarosti, K., Luijten, R., Doff, W. A. S., Boor, P. P. C., Pieterman, R. F. A., Bouzid, R., Biesta, P. J., Ijzermans, J. N. M., Doukas, M., de Man, R. A., Woltman, A. M., Demmers, J. A. A., & Buschow, S. I. (2022). Immunopeptidome of hepatocytes isolated from patients with HBV infection and hepatocellular carcinoma. JHEP Reports : Innovation in Hepatology, 4(11), 100576. https://doi.Org/10.1016/i.jhepr.2022.100576

NCT05003895. GPC3 Targeted CAR-T Cell Therapy in Advanced GPC3 Expressing Hepatocellular Carcinoma (HCC). https://classic.clinicaltrials.gov/ct2/show/NCT05003895

Sun, L, Gao, F., Gao, Z., Ao, L, Li, N., Ma, S., Jia, M., Li, N., Lu, P., Sun, B., Ho, M., Jia, S., Ding, T., & Gao, W. (2021). Shed antigen-induced blocking effect on CAR-T cells targeting Glypican-3 in Hepatocellular Carcinoma. Journal for ImmunoTherapy of Cancer, 9(4), e001875. https://doi.org/10.1136/iitc-2020-0Q1875

Zhang, Y.-F., & Ho, M. (2016). Humanization of high-affinity antibodies targeting glypican-3 in hepatocellular carcinoma. Scientific Reports, 6, 33878. https://doi.org/10.1038/srep33878

Yu, M., Luo, H. , Fan, M., Wu, X., Shi, B., Di, S., Liu, Y., Pan, Z. , Jiang, H. , & Li, Z. (2018). Development of GPC3-Specific Chimeric Antigen Receptor-Engineered Natural Killer Cells for the Treatment of Hepatocellular Carcinoma. Molecular Therapy : The Journal of the American Society of Gene Therapy, 26(2), 366-378. https://doi.Org/10.1016/j.ymthe.2017.12.012

Claims

1. T-cell receptor binding to a HLA-A2-GPC3(522-530) antigen complex, said antigen consisting of the sequence SEQ ID NO: 2, wherein T cells engineered with the T-cell receptor have a functional avidity for the GPC3(522-530)-HLA-A2 complex, depicted as EC50, equal or below 1.20 nM, wherein the functional avidity is defined as the degree of activation of TCR-engineered T cells, measured as I FNy production, following interaction of the TCR-engineered T cell with the exogenous GPC3(522-530) peptide presented by the surface HLA-2 molecule, and the EC50 is the GPC3(522-530) peptide concentration required to induce a half- maximal I FNy response.

2. The T-cell receptor according to claim 1, wherein the functional avidity of T cells engineered with the TCR for GPC3(522-530)/HLA-A2 complex is measured in a standard I FNy production assay using as target cells HLA-A2+ T2 cells pulsed with graded amounts of exogenous GPC3(522-530) peptide.

3. The T-cell receptor according to any one of claims 1-2, which binds to GPC3+ HLA-A2+ tumor cells.

4. The T-cell receptor according to any one of claims 1-3, comprising: a) a chain complementary determining regions (CDRs) consisting of SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 8 [CDR3], and p chain CDRs consisting of SEQ ID NO: 5 [CDR1], SEQ ID NO: 6 [CDR2], and SEQ ID NO: 10 [CDR3], or b) a chain CDRs consisting of SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 9 [CDR3], and p chain CDRs consisting of SEQ ID NO: 5 [CDR1], SEQ ID NO: 6 [CDR2], and SEQ ID NO: 11 [CDR3], or c) a chain CDRs consisting of SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 7 [CDR3], and p chain CDRs consisting of SEQ ID NO: 5 [CDR1], SEQ ID NO: 6 [CDR2], and SEQ ID NO: 10 [CDR3],

5. The T-cell receptor according to any one of claims 1 to 4, comprising: a) a chain variable region comprising or consisting of SEQ ID NO: 13, and p chain variable region comprising or consisting of SEQ ID NO: 15, or b) a chain variable region comprising or consisting of SEQ ID NO: 14, and p chain variable region comprising or consisting of SEQ ID NO: 16, or c) a chain variable region comprising or consisting of SEQ ID NO: 12, and p chain variable region comprising or consisting of SEQ ID NO: 15.

6. T-cell receptor binding to a HLA-A2-GPC3(522-532) antigen complex, said antigen consisting of the sequence SEQ ID NO: 100.

7. The T-cell receptor according to claim 6, wherein T cells engineered with the T-cell receptor have a functional avidity for the GPC3(522-532)-HLA-A2 complex, depicted as EC50, equal or below 1.20 nM, wherein the functional avidity is defined as the degree of activation of TCR-engineered T cells, measured as I FNy production, following interaction of the TCR-engineered T cell with the exogenous GPC3(522-532) peptide presented by the surface HLA-2 molecule, and the EC50 is the GPC3(522-532) peptide concentration required to induce a half-maximal IFNy response.

8. The T-cell receptor according to claim 7, wherein the functional avidity of T cells engineered with the TCR for GPC3(522-532)/HLA-A2 complex is measured in a standard I FNy production assay using as target cells HLA-A2+ T2 cells pulsed with graded amounts of exogenous GPC3(522-532) peptide

9. The T-cell receptor according to any one of claims 6-8, wherein the T-cell receptor binds to GPC3+ HLA- A2+ tumor cells.

10. The T-cell receptor according to any one of claims 4, and 6-9, comprising: a chain complementary determining regions (CDRs) consisting of SEQ ID NO: 3 [CDR1], SEQ ID NO: 4 [CDR2], and SEQ ID NO: 8 [CDR3], and p chain CDRs consisting of SEQ ID NO: 5 [CDR1], SEQ ID NO: 6 [CDR2], and SEQ ID NO: 10 [CDR3],

11. The T-cell receptor according to any one of claims 1 to 10, wherein:

- the constant region of the a chain comprises or consists of SEQ ID NO: 22 or a sequence having at least 90% identity, in particular at least 95% identity, more in particular at least 98% identity, with SEQ ID NO: 22, and

- the constant region of the p chain comprises or consists of SEQ ID NO: 23 or a sequence having at least 90% identity, in particular at least 95% identity, more in particular at least 98% identity, with SEQ ID NO: 23.

12. A T-cell receptor construct comprising:

(1) a and p chains comprising CDRs as defined in claim 4 (a)-(c) or variable regions as defined in claim 5 (a)- (c), wherein the a and p chains are preferably covalently linked to each other to form TCR heterodimers or multimers, and

(2) a fusion component selected from the group consisting of a Fc receptor and/or Fc domain, a cytokine, such as IL-2 or IL-15, a toxin, an antibody or a single chain antibody fragment (scFv), CD3-zeta chain and/or other TCR stimulation domain, such as the intracellular CD28, CD137 or CD134 domain, and combinations thereof, wherein the at least one of the T-cell receptor a or p chains (1) is bound to the fusion component(s) (2), preferably through a linker.

13. The T-cell receptor according to any one of claims 1-11 or the T-cell receptor construct according to claim 12, said TCR or construct lacking the transmembrane domains of the a and p chains.

14. A polynucleotide encoding for a T-cell receptor as defined in any one of claims 1-11 or 13, or a T-cell receptor construct as defined in any one of claims 12-13.

15. An expression vector comprising a polynucleotide as defined in the preceding claim.

16. A cell, in particular a T cell, more in particular a CD8 T cell, comprising a T-cell receptor as defined in any one of claims 1-11 or 13, or a T-cell receptor construct as defined in any one of claims 12-13, or a polynucleotide as defined in claim 14, or an expression vector as defined in claim 15.

17. The T-cell, in particular the CD8 T cell, according to claim 16, which has a functional avidity for the GPC3(522-530)-HLA-A2 complex, measured as IFNy production, of EC50, equal or below 1.20 nM.

18. A pharmaceutical composition comprising a T-cell receptor as defined in any one of claims 1-11 or 13, or a T-cell receptor construct as defined in any one of claims 12-13, or a polynucleotide as defined in claim 14, or an expression vector as defined in claim 15, or a cell according to any one of claims 16-17, together with pharmaceutically acceptable excipients and/or carriers.

19. A T-cell receptor as defined in any one of claims 1-11 or 13, or a T-cell receptor construct as defined in any one of claims 12-13, or a polynucleotide as defined in claim 14, or an expression vector as defined in claim 15, or a cell according to any one of claims 16-17, or a pharmaceutical composition according to claim 18, for use as a medicament.

20. A T-cell receptor as defined in any one of claims 1-11 or 13, or a T-cell receptor construct as defined in any one of claims 12-13, or a polynucleotide as defined in claim 14, or an expression vector as defined in claim 15, or a cell according to any one of claims 16-17, or a pharmaceutical composition according to claim 18, for use in the treatment of a GPC3 positive cancer, more in particular, for treating hepatocellular carcinoma or hepatoblastoma.

21. The T-cell receptor as defined in any one of claims 1-11 or 13, or a T-cell receptor construct as defined in any one of claims 12-13, or a polynucleotide as defined in claim 14, or an expression vector as defined in claim 15, or a cell according to any one of claims 16-17, or the pharmaceutical composition according to claim 18, for use according to any one of claims 19-20, that is for the treatment of an HLA-A2 positive human subject.