EA043895B1 - ADENOVIRUS ARMED WITH A BISPECIFIC T-CELL ACTIVator - Google Patents

Description

The present invention provides a modified adenovirus, in particular Enadenotucirev (EnAd), armed with a bispecific T-cell activator comprising at least two binding domains, wherein at least one of said domains is specific for a surface antigen on a T-cell of interest. The present invention also provides a composition, for example a pharmaceutical formulation, comprising the virus, the use of said virus and virus-containing formulations, in particular in treatment, especially in the treatment of cancer. The present invention also extends to methods for producing said virus and DNA encoding said virus.

State of the art

Cancer continues to impose a huge social burden on society in terms of the hardship and suffering of patients and their loved ones, and in terms of the high financial costs of treatment, care and support for patients.

A wide variety of treatment modalities have been developed to treat cancer, including chemotherapeutic agents, radiotherapy, and more recently, biologics such as antibodies. Over the past 15 years, antibody-based cancer therapy has been developed and is currently one of the most successful and important treatment strategies for patients with hematological malignancies and solid tumors. Examples of monoclonal antibody-based cancer therapeutics currently in clinical use include rituximab, which targets CD20, bevacizumab, which targets VEGF, cetuximab, which targets EGFR, and labetuzumab, which targets CEA.

Among the various antibody formats developed, bispecific T cell activators appear to be very promising. They are relatively simple bispecific molecules that are specific for the CD3ε subunit of the TCR complex of T cells and also target an antigen of interest, such as a cancer antigen. Since bispecific T cell activators are specific for the TCR complex, this allows the bispecific T cell activator to activate resident T cells to kill cells expressing a specific target antigen on their cell surface, such as cancer cells. An important property of bispecific T cell activators is their ability to target CD4+ and non-activated CD8+ T cells to cancer cells. In other words, T cells activated by bispecific T cell activators can be used to kill cells regardless of MHC expression on the surface of these cells. This is important because some tumor cells exert MHC inhibition, making them resistant to agents such as CAR-T cells and immTACs.

Unfortunately, bispecific T cell activators have poor kinetic parameters in the circulation compared to full-length antibodies. This means that when administered to a patient, a large proportion of the bispecific T cell activators do not reach their target cells. In addition, the use of high-affinity anti-CD3 ScFv as part of bispecific T cell activators can result in strong binding to T cells in the blood, which also hinders delivery to the tumor. As a result, bispecific T cell activators cannot fully realize their potential as an anti-cancer therapeutic because they cannot be effectively delivered to tumor cells.

The need to ensure efficient delivery of therapeutic agents such as bispecific T cell activators to tumor cells is becoming increasingly important as it is becoming increasingly clear that solid tumors protect themselves in vivo in several ways, such as by forming a peritumoral stroma. Progression to carcinoma is associated with epithelial cell proliferation (mitotic cells) along with the development of an activated tumor stroma. In this case, extracellular matrix (ECM) components such as collagen bundles are degraded due to increased metabolism. Inflammatory cells increase in number and fibroblasts differentiate into myofibroblasts, resulting in their expression of growth factors, matrix components and degradative proteases. Angiogenesis is maintained, resulting in a large number of permeable vessels in the tumor. Once the tumor stroma is activated with persistent angiogenesis, tumor cells invade through the disrupted basement membrane and blood vessels infiltrate the tumor tissue.

This stroma provides physical protection to the tumor in that it can act as a trap for immune cells that are targeting the tumor. In addition, the stroma shields the hypoxic tumor microenvironment, which provides permissive and optimized conditions for tumor growth. There are some theories that the cells in the stroma represent a source of energy for the tumor.

A significant component of the tumor stroma are fibroblasts, whose function has been altered by the tumor for its own purposes. Other cells infiltrating the stroma are tumor-associated macrophages (TAMs), which are type 2 macrophages (M2) that can stimulate tumor growth by secreting cytokines and chemokines such as IL-10, which suppress immune responses.

- 1 043895

Targeting the tumor stroma is particularly challenging because the cells that make up its environment are native immune cells or connective tissue cells found throughout the body. Thus, targeting these cells with therapeutic agents can lead to serious side effects.

Based on this, there is a need for an improved method of delivering bispecific T cell activators directly to tumor cells where they can provide maximum therapeutic effect, in particular delivery to tumor cells surrounded by stromal fibroblasts.

Brief description of the invention

The present inventors believe that one of the most effective ways to deliver therapeutic agents directly to a tumor is to use an oncolytic adenovirus engineered to express agents that, for example, activate T cells and target an antigen, such as a stromal antigen.

Accordingly, the present invention provides an adenovirus comprising a sequence of formula (I):

5'ITR-B ₁ -BA-B ₂ -B _x -B _b -B _y -B ₃ -3'ITR (I) where B1 is a bond or contains E1A, E1B or E1A-E1B;

BA contains -E2B-L1-L2-L3-E2A-L4;

B2 is a bond or contains E3;

BX is a DNA linkage or sequence containing a restriction site, one or more transgenes, or both;

BB contains L5;

BY is a DNA linkage or sequence containing a restriction site, one or more transgenes, or both;

B3 is a bond or contains E4;

wherein said adenovirus encodes a bispecific T cell activator comprising at least two binding domains, and wherein at least one of said domains is specific for a surface antigen on an immune cell of interest, such as a T cell of interest; and wherein said adenovirus is EnAd or Ad11.

The bispecific T cell activators of the present invention do not contain a transmembrane domain and are therefore not expressed on the surface of cancer cells, but contain a signal sequence that promotes the release of the bispecific T cell activator molecule from the cancer cell.

The following paragraphs provide a summary of the present invention.

1. An adenovirus containing a sequence of formula (I):

5'ITR-B ₁ -B _a -B ₂ -B _x -B _b -B _Y -B ₃ -3'ITR (I) where B1 is a bond or contains E1A, E1B or E1A-E1B;

B _A contains -E2B-L1-L2-L3-E2A-L4;

B ₂ is a bond or contains E3;

B _X is a DNA linkage or sequence containing a restriction site, one or more transgenes, or both;

B _B contains L5;

BY is a DNA linkage or sequence containing a restriction site, one or more transgenes, or both;

B3 is a bond or contains E4;

wherein said adenovirus encodes a bispecific T cell activator comprising at least two binding domains, and wherein at least one of said domains is specific for a surface antigen on an immune cell of interest, such as a T cell of interest; and wherein said adenovirus is EnAd or Ad11.

2. The adenovirus according to paragraph 1, characterized in that said adenovirus is EnAd.

3. An adenovirus according to paragraph 1 or 2, wherein the surface antigen is a component of a T cell receptor (TCR) complex, such as CD3, TCR-α and TCR-β.

4. The adenovirus according to paragraph 3, characterised in that the surface antigen is CD3, such as CD3ε, CD3y and CD3δ, in particular CD3ε.

5. An adenovirus according to any one of paragraphs 1-4, characterized in that one of the binding domains is specific for a tumor antigen, such as CEA, MUC-1, EpCAM, HER receptors HER1, HER2, HER3, HER4, PEM, A33, G250, carbohydrate antigens Le ^y , Lex, Leb, PSMA, TAG-72, STEAP1, CD166, CD24, CD44, E-cadherin, SPARC, ErbB2 and ErbB3.

6. An adenovirus according to paragraph 5, characterized in that one of the binding domains is

- 2 043895 specific to EpCAM, for example EpCAM comprising the amino acid sequence set forth in SEQ ID NO: 28.

7. An adenovirus according to any one of paragraphs 1-4, characterized in that one of the binding domains is specific for an antigen that is present in the tumor stroma, for example, fibroblast activation protein (FAP), TREM1, IGFBP7, FSP-1, platelet-derived growth factor receptor-α (PDGFR-α), platelet-derived growth factor receptor-β (PDGFR-β) and vimentin.

8. The adenovirus of paragraph 7, wherein one of the binding domains is specific for a FAP, for example a FAP comprising the amino acid sequence as set forth in SEQ ID NO: 30.

9. An adenovirus according to paragraph 7 or 8, wherein the antigen that is present in the stroma is an antigen selected from myeloid suppressor cell antigen, tumor-associated macrophage antigen, and combinations thereof.

10. The adenovirus of paragraph 9, wherein the antigen is selected from CD163, CD206, CD68, CD11c, CD11b, CD14, CSF1 receptor, CD15, CD33, CD66b and a combination of two or more of them.

11. An adenovirus according to any one of paragraphs 1-3 and 5-10, wherein one of the binding domains in the bispecific T cell activator is specific for a non-TCR activating protein such as CD31, CD2 and CD277.

12. An adenovirus according to any of paragraphs 1-11, wherein at least one of BX or BY is not a link.

13. An adenovirus according to any of paragraphs 1-12, characterized in that said adenovirus is chimeric.

14. An adenovirus according to any of paragraphs 1-13, characterized in that said adenovirus is oncolytic.

15. An adenovirus according to any of paragraphs 1-14, characterized in that said adenovirus is capable of replication.

16. An adenovirus according to paragraph 13, characterized in that said adenovirus is replication competent.

17. An adenovirus according to any of paragraphs 1-14, characterized in that said adenovirus is replication defective.

18. An adenovirus according to any of paragraphs 1-17, characterized in that B _X contains one or more transgenes or a transgene cassette.

19. An adenovirus according to any of paragraphs 1-16, characterized in that B _Y contains one or more transgenes or a transgene cassette.

20. An adenovirus according to any one of paragraphs 1-19, wherein one or more transgenes or transgene cassettes are under the control of an endogenous or exogenous promoter, such as an endogenous promoter.

21. An adenovirus according to paragraph 20, characterized in that the transgene or transgene cassette is under the control of an endogenous promoter selected from the group consisting of the E4 promoter and the major late promoter, in particular the major late promoter.

22. The adenovirus of paragraph 19, characterized in that the transgene or transgene cassette is under the control of an exogenous promoter, such as CMV.

23. An adenovirus according to any one of paragraphs 1-22, characterized in that the transgene cassette further comprises a regulatory element independently selected from:

a. splice acceptor sequences,

b. sequences of the internal ribosome entry site or a peptide with high self-cleavage efficiency A,

c. Kozak sequences and d. their combinations.

24. The adenovirus according to paragraph 23, characterized in that the transgene cassette contains a Kozak sequence, which is located at the beginning of the protein coding sequence.

25. An adenovirus according to any of paragraphs 1-24, characterized in that the transgene cassette encodes a peptide A with high self-cleavage efficiency.

26. An adenovirus according to any of paragraphs 1-25, characterized in that the transgene cassette additionally contains a polyadenylation sequence.

27. An adenovirus according to any of paragraphs 1-26, characterized in that the transgene cassette additionally contains a restriction site at the 3' end of the DNA sequence and/or at the 5' end of the DNA sequence.

28. An adenovirus according to any one of paragraphs 1-27, characterized in that at least one transgene cassette encodes a monocistronic mRNA.

29. An adenovirus according to any one of paragraphs 1-28, wherein the bispecific T cell activator has a short half-life, such as 48 hours or less.

30. An adenovirus according to any of paragraphs 1-29, characterized in that the bispecific activator

- 3 043895

T cells are encoded in a region selected from E1, E3, BX, BY and combinations thereof.

31. The adenovirus according to paragraph 30, characterised in that the bispecific T-cell activator is encoded at least in the BX position, for example under the control of the major late promoter.

32. An adenovirus according to any one of paragraphs 1-29, wherein the adenovirus additionally encodes a second bispecific T cell activator.

33. The adenovirus of paragraph 32, wherein the first molecule of the bispecific T cell activator is specific for an antigen, such as a tumor antigen (such as those specified in the present invention), and the second molecule of the bispecific T cell activator is specific for an antigen that is present in the tumor stroma, such as an antigen that is present in the stroma (such as those specified in the present invention).

34. An adenovirus according to any one of paragraphs 1-33, characterized in that said adenovirus further comprises a cytokine or chemokine or an immunomodulator (such as a cytokine or chemokine].

35. Adenovirus according to paragraph 34, characterized in that the cytokine or chemokine is selected from MIP1a, IL-1a, IL-1e, IL-6, IL-9, IL-12, IL-13, IL-17, IL-18, IL-22, IL-23, IL-24, IL-25, IL-26, IL-27, IL-33, IL-35, IL2, IL-4, IL-5, IL-7, IL-10, IL-15, IL-21, IL-25, IL-1RA, IFNa, IFNe, IFNy, TNFa, lymphotoxin a (LTA), Flt3L, GM- CSF, IL-8, CCL2, CCL3, CCL5, CCL17, CCL20, CCL22, CXCL9, CXCL10, CXCL11, CXCL13, CXCL12, CCL2, CCL19, CCL21, (e.g. IL-1a, ]Γ-1β, IL-6, IL-9, IL-12, IL-13, IL-17, IL-18, IL-22, IL23, IL-24, IL-25, IL-26, IL-27, IL-33, IL-35, IL-2, IL-4, IL-5, IL-7, IL-10, IL-15, IL-21, IL-25, IL-1RA, IFNa, IFNe, IFNy, TNFa, lymphotoxin a (LTA), GM-CSF, IL-8, CCL2, CCL3, CCL5, CCL17, CCL20, CCL22, CXCL9, CXCL10, CXCL11, CXCL13, CXCL12, CCL2, CCL19, CCL21), e.g. IL-12, IL-18, IL-22, IL7, IL-15, IL-21, IFNy , TNFa, lymphotoxin a (LTA), CCL3, CCL5, CXCL9, CXCL10, CXCL12, CCL2, CCL19 and CCL21.

36. The adenovirus of any one of paragraphs 1-35, wherein said adenovirus further comprises an immunomodulator such as an antibody or antibody fragment, or a protein or peptide ligand specific for a checkpoint protein such as CTLA-4, PD-1, PD-L1, PD-L2, VISTA, B7-H3, B7-H4, HVEM, ILT-2, ILT-3, ILT-4, TIM-3, LAG-3, BTLA, LIGHT or CD160, such as CTLA-4, PD-1, PD-L1 and PD-L2, or for a costimulatory molecule such as CD28, CD80, CD86, CD83, ICOS, B7H2, TL1A and 4-1BB.

37. An adenovirus according to any one of paragraphs 1-36, wherein the bispecific T cell activator comprises a VH domain comprising an amino acid sequence as set forth in any one of SEQ ID NOs: 8, 13 or 18, or an amino acid sequence at least 95% identical thereto.

38. An adenovirus according to any one of paragraphs 1-37, wherein the bispecific T cell activator comprises a VL domain comprising an amino acid sequence as set forth in any one of SEQ ID NOs: 9, 12 or 17, or an amino acid sequence at least 95% identical thereto.

39. The adenovirus of any one of paragraphs 1-36, wherein the bispecific T cell activator comprises an scFv comprising the amino acid sequence set forth in any one of SEQ ID NOs: 7, 11 or 16, or an amino acid sequence at least 95% identical thereto.

40. The adenovirus of any one of paragraphs 1-39, wherein the bispecific T cell activator comprises the amino acid sequence set forth in SEQ ID NO: 2 or 4, or an amino acid sequence that is at least 95% identical thereto, such as the amino acid sequence set forth in SEQ ID NO: 73 or 75.

41. An adenovirus according to any one of paragraphs 1-40, characterized in that the adenovirus comprises a DNA sequence indicated in any one of SEQ ID NOs: 34-37, or an amino acid sequence of DNA that is at least 95% identical thereto, for example a DNA sequence presented in any one of SEQ ID NOs: 79-82.

42. A composition comprising an adenovirus according to any one of paragraphs 1-41 and a diluent or carrier.

43. A method of treating a patient comprising administering a therapeutically effective amount of an adenovirus according to any of paragraphs 1-41 or a composition according to paragraph 42.

44. The method according to paragraph 43, for the treatment of cancer, in particular a solid tumor.

In one embodiment, the adenovirus of the present invention encodes at least one additional transgene, such as 1, 2, 3 or 4 additional transgenes.

In one embodiment, a different cleavage peptide is encoded between each of the genes.

In one embodiment, all transgenes in the virus are in the same position, such as in the BY position.

The present inventors have discovered the advantage that arming an adenovirus with a bispecific T cell activator molecule allows the bispecific antibody fragment molecule to take advantage of the adenovirus's ability to selectively infect cancer cells, thereby providing targeted delivery of the bispecific T cell activator to tumor cells.

Another advantage is that the bispecific T cell activators are small and can be produced in mammalian cells. Therefore, after infection with the adenoviruses of the present invention, the bispecific T cell activator molecules are synthesized by the tumor cells, secreted and can act locally, spreading beyond the immediate sphere of presence of the virus. This thus allows the bispecific T cell activator to spread beyond the immediate site of infection, but at the same time limits the spread of the virus too far beyond the group of infected tumor cells. This minimizes the risk of unwanted side effects.

In one embodiment, the adenovirus is EnAd. EnAd has been shown to have enhanced oncolytic activity compared to prior art adenoviruses. EnAd has also been shown to have high selectivity for human epithelial-derived carcinoma cells, such as colon, lung, bladder, and kidney cancer cells. This makes it an ideal delivery vehicle for bispecific T cell activator molecules because the bispecific T cell activator molecule can activate T cells to attack target cells, while EnAd simultaneously infects and lyses cancer cells. This results in a two-pronged attack on the tumor, providing a synergistic oncolytic effect.

In one embodiment, the surface antigen is a component of a T cell receptor (TCR) complex, such as CD3, TCR-α, and TCR-β.

In one embodiment, the surface antigen is CD3, such as CD3ε, CD3y and CD3δ, especially CD3ε.

In one embodiment, one of the binding domains is specific for a tumor antigen, such as CEA, MUC-1, EpCAM, HER receptor (such as HER1, HER2, HER3, HER4), PEM, A33, G250, carbohydrate antigens Le ^y , Lex, Le ^b , PSMA, TAG-72, STEAP1, CD166, CD24, CD44, Ecadherin, SPARC, ErbB2 and ErbB3, especially EpCAM.

In one embodiment, one of the binding domains is specific for EpCAM, such as EpCAM comprising the amino acid sequence as set forth in SEQ ID NO: 28, or a sequence at least 95% identical thereto.

In one embodiment, one of the binding domains is specific for an antigen that is present in the tumor stroma, such as fibroblast activation protein (FAP), TREM1, IGFBP7, FSP-1, platelet-derived growth factor receptor-α (PDGFR-α), platelet-derived growth factor receptor-β (PDGFR-β), and vimentin. The advantage is that stromal cells (transformed cells) expressing these antigens are not subject to the same level of selection for mutational resistance as transformed cells. Thus, these cells are easier to target for cancer treatment since they are not a moving target. In addition, the types of receptors found in stromal cells are common in different types of cancer. Therefore, targeting one of the above antigens is likely to be effective in multiple types of cancer.

In one embodiment, one of the binding domains is specific for a FAP, such as a FAP comprising the amino acid sequence set forth in SEQ ID NO: 30 or a sequence at least 95% identical thereto. An advantage is that FAP is activated on tumor-associated fibroblasts. Fibroblasts are a vital component of solid carcinomas, supporting their growth, invasion, and recovery from interventions. They typically comprise 40-60% of the cells in advanced carcinomas. An advantage is that fibroblasts are genetically stable cells that are less likely to escape therapy than cancer cells. Activated fibroblasts are also relatively similar across tumor types. Thus, by activating T cells to target and kill FAP-expressing tumor-associated fibroblasts, the adenoviruses of the present invention are able to help reduce the spectrum of immunosuppressive pathways such as IL-10, TGFe, and IDO-mediated pathways.

Other stromal targets include tumor-associated macrophages and myeloid-derived suppressor cell antigen, such as CD163, CD206, CD68, CD11c, CD11b, CD14, CSF1 receptor, CD15, CD33, CD66b, and combinations of two or more of these.

In one embodiment, one of the binding domains in the bispecific T cell activator is specific for a non-TCR activating protein, such as CD31, CD2, and CD277.

In one embodiment, one of the binding domains is specific for a surface antigen on a T cell of interest, such as selected from CD3 (such as CD3delta, CD3epsilon, or CD3gamma), a TCR-α chain, and a TCR-β chain, and one binding domain is specific for a tumor antigen.

In one embodiment, one of the binding domains is specific for CD3 and the other binding domain is specific for a tumor antigen, such as selected from the group consisting of: CEA, MUC-1, EpCAM, HER receptors HER1, HER2, HER3, HER4, PEM, A33,

G250, carbohydrate antigens Le ^y , Lex, Le ^b , PSMA, TAG-72, StEAPI, CD166, CD24, CD44, E-cadherin,

SPARC, ErbB2 and ErbB3.

In one embodiment, one of the binding domains is specific for CD3 (e.g., CD3delta, CD3epsilon, or CD3gamma) and the other binding domain is specific for EpCAM.

In one embodiment, one of the binding domains is specific for CD3ε and the other binding domain is specific for EpCAM.

In one embodiment, one of the binding domains is specific for a surface antigen on a T cell of interest, such as selected from CD3 (e.g. CD3delta, CD3epsilon or CD3gamma), TCR-α and TCR-β, and the other binding domain is specific for an antigen that is present in the tumor stroma.

In one embodiment, one of the binding domains is specific for CD3 (e.g., CD3 delta, CD3 epsilon, or CD3 gamma), and the other binding domain is specific for an antigen that is present in the tumor stroma, e.g., selected from the group consisting of: fibroblast activation protein (FAP), TREM1, IGFBP7, FSP-1, platelet-derived growth factor receptor α (PDGFR-α), platelet-derived growth factor receptor β (PDGFR-β), and vimentin.

In one embodiment, one of the binding domains is specific for CD3 (e.g., CD3delta, CD3epsilon, or CD3gamma) and the other binding domain is specific for FAP.

In one embodiment, one of the binding domains is specific for CD3ε and the other binding domain is specific for FAP.

In one embodiment, one of the binding domains is specific for a surface antigen on the T cell of interest, such as CD3, TCR-α, and TCR-β, and the other binding domain is specific for a non-TCR activating protein.

In one embodiment, one of the binding domains is specific for CD3 (e.g., CD3delta, CD3epsilon, or CD3gamma) and the other binding domain is specific for a non-TCR activating protein selected from the group consisting of CD31, CD2, and CD277.

In one embodiment, one of the binding domains is specific for CD3ε and the other binding domain is specific for a non-TCR activating protein selected from the group consisting of CD31, CD2, and CD277.

In one embodiment, at least one of BX or BY is not a relationship.

In one embodiment, BX is not a relationship.

In one implementation, BY is not a relationship.

In one implementation, both BX and BY are not relationships.

In one embodiment, the adenovirus is chimeric.

In one embodiment, the adenovirus is oncolytic.

In one embodiment, the adenovirus is chimeric and oncolytic.

In one embodiment, the adenovirus is replicatingly capable.

In one embodiment, the adenovirus is chimeric, oncolytic, and replication competent.

In one embodiment, the adenovirus is replication competent.

In one embodiment, the adenovirus is chimeric, oncolytic, and replication competent.

In one embodiment, the adenovirus is replication defective, i.e., it is a vector.

In one embodiment, BX comprises a transgene or transgene cassette, in particular a transgene cassette encoding a bispecific T cell activator according to the present invention.

In one embodiment, B _Y comprises a transgene or transgene cassette, in particular a transgene cassette encoding a bispecific T cell activator according to the present invention.

In one embodiment, B _Y comprises a transgene or transgene cassette, in particular a transgene cassette encoding a bispecific T cell activator according to the present invention, and B _X is a link.

In one embodiment, both BX and BY comprise a transgene or transgene cassette.

In one embodiment, one or more transgenes or transgene cassettes are under the control of an endogenous or exogenous promoter, such as an endogenous promoter. The advantage is that while under the control of these promoters, the virus remains replication competent and is also capable of expressing the bispecific T cell activator and/or another protein. Thus, the selected bispecific T cell activator will be expressed by the cancer cell. The use of an exogenous promoter may be advantageous in some embodiments because it promotes strong and constitutive expression of the antibody or fragment, which may be particularly useful in some situations, such as when a patient has a highly

- 6 043895 ability to spread. The use of an endogenous promoter may be advantageous because it reduces the size of the transgene cassette that must be included to express the bispecific T cell activator, i.e. the cassette can be smaller because inclusion of an exogenous promoter is not required.

Accordingly, in one embodiment, the transgene or transgene cassette is under the control of an endogenous promoter selected from the group consisting of E4 and a major late promoter, in particular a major late promoter. The use of an endogenous promoter in a virus may also be advantageous in a therapeutic context, since the transgene is only expressed when the virus replicates in a cancer cell, as opposed to a constitutive exogenous promoter, which would continuously transcribe the transgene and may result in inappropriate concentration of the antibody or fragment.

In one embodiment, the transgene or transgene cassette (e.g., encoding a bispecific T cell activator) is under the control of an exogenous promoter, such as CMV. An advantage is that the use of a constitutive exogenous promoter results in continuous transcription of the transgene, which may be desirable in some cases.

In one embodiment, one transgene or transgene cassette (e.g., encoding a bispecific T cell activator) is under the control of an endogenous promoter, and the other transgene or transgene cassette (e.g., encoding a bispecific T cell activator) is under the control of an exogenous promoter.

In one embodiment, all transgenes or transgene cassettes (e.g., encoding a bispecific T cell activator) in the virus are/are under the control of an endogenous promoter.

In another embodiment, all transgenes or transgene cassettes (e.g., encoding a bispecific T cell activator) in the virus are/are under the control of an exogenous promoter.

In one embodiment, the transgene or transgene cassette further comprises a regulatory element independently selected from:

i) splice acceptor sequences, ii) internal ribosome entry site sequences or highly self-cleaving peptide A, iii) Kozak sequences, and iv) combinations thereof.

Thus, in one embodiment, the transgene cassette comprises i) or ii) or iii) or iv).

In one embodiment, the transgene cassette comprises i) and ii), or i) and iii), or i) and iv), or ii) and iii), or ii) and iv), or iii) and iv).

In one embodiment, the transgene cassette comprises i) and ii) and iii), or i) and ii) and iv), or i) and iii) and iv), or ii) and iii) and iv).

In one embodiment, the transgene cassette comprises i) and ii) and iii) and iv).

In one embodiment, the transgene cassette comprises a Kozak sequence at the beginning of the coding sequence of a protein (e.g., a bispecific T cell activator) that promotes translation of mRNA.

In one embodiment, the transgene cassette encodes a highly self-cleaving peptide 2A.

In one embodiment, the transgene cassette further comprises a polyadenylation sequence.

In one embodiment, the transgene cassette further comprises a restriction site at the 3' end of the DNA sequence and/or at the 5' end of the DNA sequence.

In one embodiment, at least one transgene cassette encodes a monocistronic mRNA.

In one embodiment, the bispecific T cell activator molecule has a short half-life, such as 48 hours or less.

In one embodiment, the bispecific T cell activator molecule is encoded in a region selected from E1, E3, BX, BY and a combination thereof. The present inventors have found the advantage that various transgenes can be inserted into B _X and/or B _Y under the control of an exogenous or endogenous promoter, without negatively affecting the life cycle of the virus or the stability of the vector.

In one embodiment, the bispecific T cell activator molecule is encoded at least at position BX, for example under the control of the major late promoter. The advantage is that the transgene or transgene cassette allows the bispecific T cell activator or any additional molecule to be expressed together with the adenovirus itself. Importantly, the inventors have successfully demonstrated that expression of the bispecific T cell activator does not significantly affect the ability of EnAd to replicate and does not

- 7 043895 has a negative effect on its oncolytic activity.

In one embodiment, the bispecific T cell activator molecule is encoded at least at the BY position, for example under the control of the major late promoter. The advantage is that the transgene or transgene cassette allows the bispecific T cell activator or any additional molecule to be expressed together with the adenovirus itself. Importantly, the inventors have successfully demonstrated that expression of the bispecific T cell activator does not significantly affect the ability of EnAd to replicate and does not negatively affect its oncolytic activity.

In one embodiment, the adenovirus further encodes a second bispecific T cell activator.

In one embodiment, the first bispecific T cell activator molecule is specific for a tumor antigen, such as a tumor antigen as described above, and the second bispecific T cell activator molecule is specific for an antigen that is present in the tumor stroma, such as an antigen that is present in the stroma as described above.

In one embodiment, the first bispecific T cell activator molecule is specific for a tumor antigen selected from the group consisting of: CEA, MUC-1, EpCAM, HER receptors HER1, HER2, HER3, HER4, PEM, A33, G250, carbohydrate antigens Le ^y , Lex, Le ^b , PSMA, TAG-72, STEAP1, CD166, CD24, CD44, E-cadherin, SPARC, ErbB2 and ErbB3, and the second bispecific T cell activator molecule is specific for an antigen present in the tumor stroma selected from the group consisting of: fibroblast activation protein (FAP), TREM1, IGFBP7, FSP-1, platelet-derived growth factor receptor-α (PDGFR-α), platelet-derived growth factor receptor-β (PDGFR-β) and vimentin.

In one embodiment, the first bispecific T cell activator molecule is specific for EpCAM, and the second bispecific T cell activator molecule is specific for an antigen present in the tumor stroma, selected from the group consisting of: fibroblast activation protein (FAP), TREM1, IGFBP7, FSP-1, platelet-derived growth factor receptor-α (PDGFR-α), platelet-derived growth factor receptor β (PDGFR-β), and vimentin.

In one embodiment, the first bispecific T cell activator molecule is specific for a tumor antigen selected from the group consisting of: CEA, MUC-1, EpCAM, HER receptors HER1, HER2, HER3, HER4, PEM, A33, G250, carbohydrate antigens Le ^y , Lex, Le ^b , PSMA, TAG-72, STEAP1, CD166, CD24, CD44, E-cadherin, SPARC, ErbB2 and ErbB3, and the second bispecific T cell activator molecule is specific for FAP.

In one embodiment, the first bispecific T cell activator molecule is specific for EpCAM and the second bispecific T cell activator molecule is specific for FAP.

In another embodiment, the first molecule of the bispecific T cell activator is specific for a tumor antigen and the second molecule of the bispecific T cell activator is specific for a non-TCR activating protein.

In one embodiment, the first molecule of the bispecific T cell activator is specific for a tumor antigen selected from the group consisting of: CEA, MUC-1, EpCAM, HER receptors HER1, HER2, HER3, HER4, PEM, A33, G250, carbohydrate antigens Le ^y , Lex, Le ^b , PSMA, TAG-72, STEAP1, CD166, CD24, CD44, E-cadherin, SPARC, ErbB2 and ErbB3.

In one embodiment, the first bispecific T cell activator molecule is specific for EpCAM and the second bispecific T cell activator molecule is specific for a non-TCR activating protein selected from the group consisting of: CD31, CD2, and CD277.

In one embodiment, the first bispecific T cell activator molecule is specific for an antigen that is present in the tumor stroma, and the second bispecific T cell activator molecule is specific for a non-TCR activating protein.

In one embodiment, the first bispecific T cell activator molecule is specific for an antigen present in the tumor stroma, selected from the group consisting of: fibroblast activation protein (FAP), TREM1, IGFBP7, FSP-1, platelet-derived growth factor receptor-α (PDGFR-α), platelet-derived growth factor receptor-β (PDGFR-β) and vimentin, and the second bispecific T cell activator molecule is specific for a non-TCR activating protein selected from the group consisting of: CD31, CD2 and CD277.

In one embodiment, the first bispecific T cell activator molecule is specific for FAP and the second bispecific T cell activator molecule is specific for a non-TCR activating protein selected from the group consisting of: CD31, CD2, and CD277.

In one embodiment, the adenovirus contains only one bispecific T cell activator.

In another embodiment, the adenovirus contains two bispecific T cell activators.

In another embodiment, the adenovirus contains three bispecific T cell activators.

In addition to encoding one, two, or three bispecific T-cell activators, the virus

- 8 043895 may also encode 1, 2, 3 or 4 additional transgenes.

In one embodiment, the adenovirus further encodes a cytokine or chemokine.

In one embodiment, the adenovirus further encodes a cytokine.

In one embodiment, the adenovirus additionally encodes a chemokine.

In another embodiment, the adenovirus additionally encodes a cytokine and a chemokine.

In one embodiment, the adenovirus comprises one bispecific T cell activator and at least one cytokine or chemokine, such as 1, 2, or 3 cytokines, 1, 2, or 3 chemokines, or a combination of 2 or 3 genes, each independently encoding a cytokine or chemokine.

In another embodiment, the adenovirus comprises two bispecific T cell activators and at least one cytokine or chemokine, such as 1 or 2 cytokines, 1 or 2 chemokines, or a combination of a cytokine and a chemokine.

In another embodiment, the adenovirus comprises three bispecific T cell activators and at least one cytokine or chemokine.

In one embodiment of the invention, the cytokine or chemokine is selected from IL-1a, IL-1e, IL-6, IL-9, IL-12, IL-13, IL-17, IL-18, IL-22, IL-23, IL-24, IL-25, IL-26, IL-27, IL-33, IL-35, IL-2, IL-4, IL-5, IL-7, IL-10, IL-15, IL-21, IL-25, IL-1RA, IFNa, IFNp, IFNy, TNFa, lymphotoxin a (LTA) and GM-CSF, IL-8, CCL2, CCL3, CCL5, CCL17, CCL20, CCL22, CXCL9, CXCL10, CXCL11, CXCL13, CXCL12, CCL2, CCL19, CCL21, for example IL-12, IL-18, IL-22, IL-7, IL-15, IL-21, IFNy, TNFa, lymphotoxin a (LTA), CCL3, CCL5, CXCL9, CXCL12, CCL2, CCL19 and CCL21.

In one embodiment, the encoded cytokine is selected from the TNF-alpha superfamily (TNFRSF includes TNF-alpha, TNF-C, OX40L, CD154, FasL, LIGHT, TL1A, CD70, Siva, CD153, 4-1BB ligand, TRAIL, RANKL, TWEAK, APRIL, BAFF, CAMLG, NGF, BDNF, NT-3, NT-4, GITR ligand, EDA-A, EDAA2), TGF-beta superfamily, IL-1 family (i.e., IL-1 and IL-8), IL-2 family, IL-10 family, IL-17 family, interferon family.

In one embodiment, the chemokine is selected from the group consisting of MIP-1 alpha, RANTES, IL8, CCL5, CCL17, CCL20, CCL22, CXCL9, CXCL10, CXCL11, CXCL13, CXCL12, CCL2, CCL19 and CCL21.

In one embodiment, the adenovirus further comprises an immunomodulator, such as an antibody or antibody fragment, or a protein or peptide ligand specific for a checkpoint protein or costimulatory molecule, or specific binding ligands for such molecules.

In one embodiment, said immunomodulator is an antibody or antibody fragment, or a protein or peptide ligand specific for a checkpoint protein such as CTLA-4, PD-1, PD-L1, PD-L2, VISTA, B7-H3, B7-H4, HVEM, ILT-2, ILT-3, ILT-4, TIM-3, LAG-3, BTLA, LIGHT or CD160, such as CTLA-4, PD-1, PD-L1 and PD-L2.

In one embodiment, the immunomodulator is an inhibitor, such as a checkpoint inhibitor.

In one embodiment, the immunomodulator is an agonist.

In another embodiment, the immunomodulator is an antibody or antibody fragment, or a protein or peptide ligand specific for a costimulatory molecule such as CD28, CD80, CD86, CD83, ICOS, B7H2, TL1A, and 4-1BB.

In one embodiment, the adenovirus comprises a first antibody, antibody fragment, protein or peptide ligand specific for a checkpoint protein and a second antibody, antibody fragment, protein or peptide ligand specific for a costimulatory molecule.

In one embodiment, the bispecific T cell activator comprises a VH domain comprising an amino acid sequence set forth in any one of SEQ ID NOs: 8, 13 or 18, or an amino acid sequence at least 95% identical thereto.

In one embodiment, the bispecific T cell activator comprises a VL domain comprising an amino acid sequence as set forth in any one of SEQ ID NOs: 9, 12 or 17, or an amino acid sequence at least 95% identical thereto.

In one embodiment, the bispecific T cell activator comprises an scFv comprising an amino acid sequence set forth in any one of SEQ ID NOs: 7, 11 or 16, or an amino acid sequence at least 95% identical thereto.

In one embodiment, the bispecific T cell activator used in the present invention (i.e., encoded by an adenovirus) comprises a VH domain binding domain having the sequence shown in SEQ ID NO: 8, or a sequence at least 95% identical thereto, and a VL domain having the sequence shown in SEQ ID NO: 9, or a sequence at least 95% identical thereto.

In one embodiment, the bispecific T cell activator used in the present invention (i.e., encoded by an adenovirus) comprises a VH domain binding domain having the sequence shown in SEQ ID NO: 13, or a sequence at least 95% identical thereto, and a VL domain having the sequence shown in SEQ ID NO: 12, or a sequence at least 95% identical thereto.

- 9 043895

In one embodiment, the bispecific T cell activator used in the present invention (i.e., encoded by an adenovirus) comprises a VH domain binding domain having the sequence shown in SEQ ID NO: 18, or a sequence at least 95% identical thereto, and a VL domain having the sequence shown in SEQ ID NO: 17, or a sequence at least 95% identical thereto.

In one embodiment, the bispecific T cell activator used in the present invention (i.e., encoded by an adenovirus) comprises a VH domain binding domain with the sequence shown in SEQ ID NO: 8, or a sequence at least 95% identical thereto, and a VL domain with the sequence shown in SEQ ID NO: 9, or a sequence at least 95% identical thereto, and a VH domain binding domain with the sequence shown in SEQ ID NO: 13, or a sequence at least 95% identical thereto, and a VL domain with the sequence shown in SEQ ID NO: 12, or a sequence at least 95% identical thereto.

In one embodiment, the bispecific T cell activator used in the present invention (i.e., encoded by an adenovirus) comprises a VH domain binding domain with the sequence shown in SEQ ID NO: 8, or a sequence at least 95% identical thereto, and a VL domain with the sequence shown in SEQ ID NO: 9, or a sequence at least 95% identical thereto, and a VH domain binding domain with the sequence shown in SEQ ID NO: 18, or a sequence at least 95% identical thereto, and a VL domain with the sequence shown in SEQ ID NO: 17, or a sequence at least 95% identical thereto.

In one embodiment, the bispecific T cell activator used in the present invention (i.e., encoded by an adenovirus) comprises the sequence shown in SEQ ID NO: 7, 11, 16, or a sequence at least 95% identical to any of them.

In one embodiment, the bispecific T cell activator used in the present invention (i.e., encoded by an adenovirus) comprises the sequence shown in SEQ ID NO: 7, or a sequence at least 95% identical thereto.

In one embodiment, the bispecific T cell activator used in the present invention (i.e., encoded by an adenovirus) comprises the sequence shown in SEQ ID NO: 11, or a sequence at least 95% identical thereto.

In one embodiment, the bispecific T cell activator used in the present invention (i.e., encoded by an adenovirus) comprises the sequence shown in SEQ ID NO: 16, or a sequence at least 95% identical thereto.

In one embodiment, the bispecific T cell activator comprises an amino acid sequence as set forth in SEQ ID NO: 2 or 4, or an amino acid sequence that is at least 95% identical thereto, such as an amino acid as set forth in SEQ ID NO: 73 or 75.

In one embodiment, the adenovirus of the present invention comprises a DNA sequence as set forth in any one of SEQ ID NOs: 34-37, or an amino acid sequence of DNA that hybridizes thereto under stringent conditions.

In one embodiment, the adenovirus of the present invention comprises a DNA sequence set forth in any one of SEQ ID NOs: 79-82.

In one embodiment, the adenovirus of the present invention comprises a DNA sequence shown in any of SEQ ID NOs: 34, 35, 36, 37, 79, 80, 82, 96, 97, 98, 99, 100, 101, 102, 103, 120, 298, or a sequence encoding the same virus, or a sequence that hybridizes with any of them under stringent conditions.

In one embodiment, the adenovirus of the present invention comprises a DNA sequence shown in any of SEQ ID NOs: 34, 35, 36, 37, 79, 80, 82, 96, 97, 98, 99, 100, 101, 102, 103, 120, or 298.

The skilled person knows that the DNA code is degenerate, therefore the present invention extends to End or Ad11 encoding a bispecific T cell activator with an amino acid disclosed in the present invention.

The C-terminal deca-His affinity tag (HHHHHHHHH SEQ ID NO: 24) is useful for purifying the bispecific T cell activator or adenovirus. However, it is not required and can be omitted, for example, in the final product. The skilled artisan will also be aware that other affinity tags other than deca-His can be used and can also be omitted without affecting the biological function of the bispecific T cell activator or adenovirus. Accordingly, in one embodiment, the bispecific T cell activator comprises an amino acid sequence as set forth in any of SEQ ID NOs: 2 or 4, but excludes a deca-His affinity tag at the C-terminus of the sequence, such as set forth in SEQ ID NOs: 73 or 75. In another embodiment, the adenovirus comprises a DNA sequence as set forth in any of SEQ ID NOs: 34-37, but excludes a deca-His affinity tag, such as a DNA sequence as set forth in any of SEQ ID NOs: 79-82.

The exclusion of the deca-His affinity tag also extends to all other sequences disclosed in the present invention that include the deca-His affinity tag, i.e., the present invention includes the same amino acid or DNA sequences that lack the C-terminal deca-His-MeTKa (HHHHHHHHHH or CATCASSACCATCASSACCATCASSACCATCASSAT), for example, as set forth in any of SEQ ID NOs: 72-82.

In one embodiment, the bispecific T cell activator encoded by the virus of the present invention is under the control of an exogenous promoter, such as a CMV promoter. The exogenous promoter may be placed between MPL and the encoded transgene when the transgene is located between the L5 and E4 regions.

An exogenous promoter can be placed between the encoded transgene and L5 when the transgene is located between the L5 and E3 regions.

In one aspect, a composition is provided comprising an adenovirus as described in the present invention and a diluent or carrier.

In one aspect, a method of treating a patient is provided, comprising administering a therapeutically effective amount of an adenovirus or composition described in the present invention.

In one embodiment, the described method is for the treatment of cancer, such as epithelial cancer, in particular a solid tumor.

In one embodiment, a method of treatment is provided comprising administering a virus according to the present invention in combination with a checkpoint inhibitor (such as a PD-1 or PDL1 inhibitor), particularly where the checkpoint inhibitor is encoded in the virus.

In one embodiment, a method of treatment is provided comprising administering a virus according to the present invention that is NOT combined with a checkpoint inhibitor (e.g., as described elsewhere in the present invention, such as a PD-1 or PDL1 inhibitor), particularly where the checkpoint inhibitor is not encoded in the virus.

The bispecific T cell activators encoded by the virus of the present invention have the ability to enhance the cytotoxicity of the virus.

Surprisingly, it was found that the bispecific T cell activators encoded by the virus of the present invention can activate CD4+ cells and/or CD8+ cells, for example even cells in the suppressive environment of the tumor, including T cells in the fluid environment of the tumor, such as ascites.

An advantage is that the bispecific T cell activators encoded by the virus according to the present invention can activate cytotoxic T cells, for example even T cells in the suppressive environment of the tumor, including T cells in the fluid environment, such as ascites, of the tumor.

Even more unexpectedly, it was found that the bispecific T cell activators encoded by the virus according to the present invention are capable of stimulating (activating) T cell proliferation.

The viruses encoding the bispecific T cell activators of the present invention appear to be able to bypass, overcome, or reverse the immunosuppressive tumor microenvironment.

In one embodiment, T cell activation results in upregulation of a T cell marker, such as CD25.

In one embodiment, binding of the bispecific T cell activator in a virus of the present invention is specific to a neoantigen.

The invention also extends to the novel sequences disclosed in the present invention.

Detailed description

As used herein, the term immune cell refers to a cell with a functional role in the immune system, including (among others) macrophages, neutrophils, dendritic cells, NK cells, lymphocytes such as T lymphocytes (particularly T cells and NKT cells).

As used in the present invention, the term "antibody" refers to an immunoglobulin molecule capable of specifically binding to a target antigen, such as a carbohydrate, polynucleotide, lipid, polypeptide, peptide, etc., through at least one antigen recognition site (also referred to in the present invention as a binding site) located in the variable region of the immunoglobulin molecule. Unless the context otherwise indicates, the term extends to full-length antibodies and multispecific antibody molecules comprising full-length antibodies.

As used herein, the term antibody molecule includes antibodies and binding fragments thereof, as well as multispecific formats of any of them.

As used herein, the term antigen-binding site refers to a portion of a molecule that contains a pair of variable regions, particularly a cognate pair, that specifically interact with a target antigen.

In particular, in the present invention, the said term is intended to denote a binding site that recognizes only the antigen to which it is specific, or a binding site that has a significantly higher binding affinity for the antigen to which it is specific, compared to the affinity for antigens to which it is non-specific, for example, 5, 6, 7, 8, 9, 10 times higher binding affinity.

- 11 043895

As used herein, the terms binding fragments or antibody-binding fragments refer to antibody-binding fragments and multispecific antibody molecules comprising antibody-binding fragments, including, but not limited to, Fab, modified Fab, Fab', modified Fab', F(ab') ² , Fv, single-domain antibodies, scFv, di-, tri- or tetravalent antibodies, Bis-scFv, diabodies, triabodies, tetrabodies, and epitope-binding fragments of any of the foregoing (see, e.g., Holliger and Hudson, 2005, Nature Biotech. 23(9):1126-1136; Adair and Lawson, 2005, Drug Design Reviews - Online 2(3), 209-217). Methods for creating and making these antibody fragments are well known in the art (see, e.g., Verma et al., 1998, Journal of Immunological Methods, 216:165-181). Other antibody fragments for use in the present invention include the Fab and Fab' fragments described in International Patent Applications WO 05/003169, WO 05/003170, and WO 05/003171. Multivalent antibodies may have multiple specificities, such as being bispecific or monospecific (see, e.g., WO92/22853 and WO05/113605).

In one embodiment, the adenovirus comprises a multispecific antibody molecule.

As used in the present invention, the term "multispecific antibody molecule" refers to an antibody molecule that has two or more antigen-binding domains, such as two (bispecific) or three (trispecific) or four (tetraspecific) binding domains.

The multispecific antibody molecules of the present invention can be constructed from various antibody fragments, such as those described above. For example, a diabody is a bispecific antibody molecule consisting of a non-covalent dimer of ScFv fragments, while F(ab') ² is a bispecific antibody molecule consisting of 2 Fab fragments linked by a hinge region. Thus, one skilled in the art will recognize that various antibody fragments can be arranged in various combinations to produce a bi- or multispecific antibody molecule.

Examples of tri-specific or tetra-specific antibody formats include, but are not limited to, Fab3, triabody, tetrabody, tribody, DVD-Ig, IgG-scFv, ScFv2-Fc, tandAbs, and DNL-Fab3.

As used in the present invention, the term bispecific antibody molecule refers to a molecule with two antigen-binding domains that can bind the same or different antigens. A bispecific T cell activator is a subclass of bispecific antibody molecules.

Domains can bind different antigens.

Alternatively, all domains may bind the same antigen, including binding the same epitope on the antigen or binding different epitopes on the same antigen.

Examples of bispecific antibody formats include, but are not limited to, bispecific T cell activator, F(ab') ₂ , F(ab')-ScFv2, di-scFv, diabody, minibody, scFv-Fc, DART, TandAb, ScDiabody, ScDiabody-CH3, Diabody-CH3, tribody, minibody, minibody, TriBi minibody, ScFv-CH3 KIH (knobs-into-holes), Fab-ScFv, SCFv-CH-CL-scFv, scFv-KIH, Fab-scFv-Fc, tetravalent HCAb, scDiabody-Fc, Diabody-Fc, intrabody, dock and lock'' antibodies, ImmTAC, HSAbody, ScDiabody-HAS, Humabody antibodies, and Tandem ScFv-toxic (see, for example Christoph Spiess et al., Molecular Immunology 67 (2015) p. 95–106).

The adenovirus of the present invention comprises a bispecific T cell activator that is specific for at least a surface antigen on a T cell of interest. Examples of T cell surface antigens include, but are not limited to: CD3, CD2, VLA-1, CD8, CD4, CCR6, CXCR5, CD25, CD31, CD45RO, CD197, CD127, CD38, CD27, CD196, CD277 and CXCR3, especially CD2, CD3, CD31 and CD277.

As used in the present invention, the term bispecific T cell activator (bisp.T cell activator) refers to a class of artificial bispecific monoclonal antibodies comprising 2 scFv fragments of different antibodies or amino acid sequences from 4 different genes in a single peptide chain of approximately 55 kDa in length. One of these scFv is specific for an immune cell, such as a T cell antigen, for example the CD3 receptor expressed on the surface of T cells. The other scFv typically binds to a tumor cell via a tumor-specific molecule. Accordingly, bispecific T cell activators are able to form a link between T cells and tumor cells due to their specificity for an antigen on the T cell and an antigen on the tumor cell. This leads to the activation of T cells and triggers the T cells to exert their cytotoxic effect on tumor cells independently of MHC class I or costimulatory molecules. Examples of bispecific T-cell activator-based drugs currently approved or in clinical trials include blinatumomab (Blincyto®), which targets CD19 and is indicated for the treatment of non-Hodgkin lymphoma and acute lymphoblastic leukemia, and solitomab, which targets EpCAM and is indicated for the treatment of gastrointestinal and lung cancer.

In one embodiment, an immune cell activator (such as a T cell activator) is assembled in the format VL1-linker 1-VH1-linker 2-VH2-linker 3-VL2, e.g., using linkers, e.g., independently selected from the linker sequences disclosed herein.

- 12 043895 research institute.

In one embodiment, the linkers in the bispecific T cell activator of the present invention are independently selected from SEQ ID NOs: 10, 14, 23, 124-162, and 166-297.

In one embodiment, linker 1 and linker 3 have the same sequence, such as the sequence shown in any of SEQ ID NOs: 10, 14, 23, 124-162, and 166-296, in particular 10, 14, and 23.

In one embodiment, linker 1 and linker 3 have different amino acid sequences, such as independently selected from SEQ ID NOs: 10, 14, 23, 124 to 162, and 166 to 296, in particular 10, 14, and 23.

In one embodiment, linker 1 is SEQ ID NO: 10.

In one embodiment, linker 1 is SEQ ID NO: 14.

In one embodiment, linker 3 is SEQ ID NO: 10.

In one embodiment, linker 3 is SEQ ID NO: 14.

In one embodiment, linker 1 and linker 3 are SEQ ID NO: 10.

In one embodiment, linker 1 and linker 3 are SEQ ID NO: 14.

In one embodiment, linker 1 is SEQ ID NO: 10 and linker 3 is SEQ ID NO: 14.

In one embodiment, linker 1 is SEQ ID NO: 14 and linker 3 is SEQ ID NO: 10.

In one embodiment, linker 2 is different from both linker 1 and linker 3.

In one embodiment, linker 2 is selected from any of SEQ ID NOs: 10, 14, 23, 124-162, and 166297, such as SEQ ID NO: 297.

In one embodiment, the length of linker 1 is from 10 to 30 amino acids, such as 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29 or 30.

In one embodiment, the length of linker 3 is from 10 to 30 amino acids, such as 10,

11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29 or 30.

In one embodiment, linker 2 is from 2 to 10 amino acids in length, such as 2, 3, 4, 5, 6, 7, 8, 9, or 10.

In one embodiment, VH1 and VL1 are specific for a T cell antigen of the present invention, such as CD3.

In one embodiment, VH2 and VL2 are specific for an immune cell antigen, such as a T cell antigen, according to the present invention, such as CD3.

In one embodiment, VH1 and VL1 are specific for a target antigen, such as a cancer antigen or an antigen that is present in the stroma, etc.

In one embodiment, VH2 and VL2 are specific for a target antigen, such as a cancer antigen or an antigen that is present in the stroma, etc.

As used in the present invention, the term "stroma" or "stromal antigen" (an antigen that is present in the stroma) refers to a therapeutic target antigen in the stroma, including those expressed in the molecular structure of the stromal matrix, such as connective tissue molecules or molecules associated with this matrix, or antigens associated with cellular components of the stroma, such as those expressed on fibroblasts, tumor-associated macrophages, dendritic cells, NK cells and/or T cells that have infiltrated the stroma. Examples of antigens that are present in the stroma include, but are not limited to, FAP, TGFe, TREM1, IGFBP7, FSP-1, fibroblast-associated antigen, NG2, endosialin (CD248), platelet-derived growth factor receptor-α (PDGFR-α), platelet-derived growth factor receptor-β (PDGFR-β), and vimentin.

Fibroblasts can be targeted using a fibroblast activation protein (FAP) antigen, particularly a FAP-specific antibody that does not bind CD26 (see US2012/0258119, incorporated herein by reference).

FAP was initially identified as a serine protease on reactive stromal fibroblasts. Subsequent molecular cloning revealed that FAP is identical to seprase, a 170 kDa membrane-associated gelatinase that is expressed by melanoma cell lines. The full-length cDNA encoded a 760 amino acid (aa) type H transmembrane protease highly homologous to dipeptidyl peptidase IV (DPPIV) with 52% sequence identity and nearly 70% identity in the catalytic domain. U.S. Patent No. 5,587,299, incorporated herein by reference, describes nucleic acid molecules encoding FAP and uses thereof.

FAP and DPPIV have similar gene sizes and are chromosomally adjacent to each other at 2q24, suggesting a gene duplication event (Genebank accession number U09278). Both proteins are members of the prolyl peptidase family. This class of enzymes is inducible, active at the cell surface or in extracellular fluids, and is uniquely capable of cleaving N-terminal dipeptides from polypeptides with a proline or alanine at the penultimate position. DPPIV, also called CD26,

- 13 043895 is constitutively expressed by several cell types including fibroblasts, endothelial and epithelial cells, leukocyte subsets such as NK cells, T lymphocytes and macrophages. A small fraction of DPPIV circulates in the blood as a soluble protein. In contrast to DPPIV, FAP is generally not expressed in normal adult tissue and its proteolytically active soluble form is termed antiplasmin cleavage enzyme 2 (APCE). FAP is highly expressed in conditions associated with activated stroma including wound healing, epithelial cancer, osteoarthritis, rheumatoid arthritis, cirrhosis and pulmonary fibrosis.

The structure of FAP has been solved (PDB ID 1Z68) and is very similar to that of DPPIV. FAP is anchored to the plasma membrane by an uncleaved signal sequence of approximately 20 amino acids and has a short amino-terminal cytoplasmic domain of six amino acids. The major portion of this protein, including the catalytic domain, is exposed to the extracellular environment. The FAP glycoprotein is a homodimer consisting of two identical 97 kDa subunits. Each subunit of the FAP monomer consists of two domains: an αβ-hydrolase domain (aa 27-53 and 493-760) and an eight-lobed β-propeller domain (aa 54-492), which enclose a large cavity. A small pocket within this cavity at the interface of the two domains contains the catalytic triad (Ser624, Asp702, and His734). FAP acquires its enzymatic activity by homodimerization of subunits, and in addition to its dipeptidyl peptidase activity, FAP also possesses collagen type I specific gelatinase and endopeptidase activity, the β-propeller acts as a scaffold for protein-protein interactions and determines the binding of the substrate and the extracellular matrix (ECM). In addition, the β-propeller is involved in the formation of supramolecular complexes of FAP with other prolyl peptidases or with other membrane-bound molecules. It has been shown that the formation of heteromeric or tetrameric complexes of FAP and DPPIV is associated with invadopodia of migrating cells on a collagen substrate. Collagen type I induces a close association of FAP with β 1-integrins, thereby playing a major organizational role in the formation and adhesion of invadopodia. Although the mechanisms involved are not understood in detail, the formation of such proteinase-rich membrane domains at the cell invasion front promotes targeted pericellular degradation of the ECM. This suggests that FAP-ECM interactions may be closely linked to cell invasive behavior by influencing cell adhesion, migration, proliferation, and apoptosis via integrin pathways, and supports a role for FAP in disease pathogenesis and progression. Thus, FAP is recognized as a multifunctional protein that exerts its biological functions in a cell-dependent manner through a combination of its protease activity and the ability to form complexes with other cell surface molecules. Overexpression of FAP in epithelial and fibroblast cell lines promotes malignant behavior, suggesting a clinical situation where cellular FAP expression levels correlate with worse clinical outcome.

Through paracrine signaling molecules, cancer cells activate stromal fibroblasts and induce FAP expression, which in turn influences cancer cell proliferation, invasion, and migration. Recent studies have shown that the dominant factor in stimulating FAP protein expression is TGF-β (Chen, H et al. (2009) Exp and Molec Pathology, doi: 10.1016/j.yexmp. 2009.09.001). FAP is abundantly expressed on reactive stromal fibroblasts in 90% of human epithelial carcinomas, including breast, lung, colon, and ovarian cancers (Garin-Chesa, P et al. (1990) PNAS USA 87: 7236-7239). Chen et al recently showed that FAPa affects the invasion, proliferation, and migration of ovarian cancer HO-8910PM cells (Chen, H et al. (2009) Exp and Molec Pathology, doi: 10.1016/j.yexmp. 2009.09.001).

FAP can be targeted by binding to said antigen and sterically blocking its interaction with biologically relevant molecules. Alternatively or additionally, cross-linking of a FAP molecule to another FAP molecule or another molecule, such as an antigen on the surface of a cancer cell, can be achieved using a multispecific, such as a bispecific antibody molecule. This cross-linking increases the visibility of cells bearing antigens to the immune system, which can then be activated to neutralize or kill these cells.

Tumor-associated macrophages (TAMs) are thought to express TREM1, CD204, CD68 (alone or in combination with CD163 or CD206). These markers can be used to target TAM macrophages.

The adenovirus of the present invention has the ability to infect tumor cells and, in particular, is selected so as to preferentially infect tumor cells. The infection of the oncolytic virus causes the death and lysis of the cancer cell with the release of newly generated viral particles. The incorporated transgenes encoding antibodies, a bispecific T-cell activator and other useful charges are newly synthesized and actively secreted by the tumor cells before their death, and some molecules are also released during cell lysis.

Antibody molecules with short half-lives may be particularly suitable for

- 14 043895 use in the present invention, as it minimizes side effects, since the body quickly eliminates the molecules if they become systemically available.

NKT cells have the ability to target and kill tumor-associated macrophages. However, their activity appears to be inhibited by the hypoxic environment of the tumor. This activity can be restored by providing NKT cells with IL-2 and/or IL-15 cytokines, such as those encoded by the virus of the present invention.

Thus, in one embodiment, the virus of the present invention further encodes a cytokine for activating NKT cells, for example selected from IL-2, IL-15 and combinations thereof. The gene encoding the cytokine may be in the same location or in a different location than the gene encoding the antibody molecule, for example independently selected from E1, E3, E4, BX and BY.

Thus, the adenovirus of the present invention has at least two or three mechanisms for attacking the tumor, including indirect mechanisms that undermine the tumor stroma.

As used herein, the term transgene refers to a gene that has been inserted into a genomic sequence that is a gene that is unnatural to the virus (exogenous) or is not normally found at that particular location in the virus. Examples of transgenes are provided below. As used herein, the term transgene also includes a functional fragment of a gene, which is a portion of a gene that, when inserted, is suitable to perform the function or most of the function of the full-length gene.

The terms transgene and coding sequence are used interchangeably in the present invention in the context of insertion into a viral genome, unless the context otherwise indicates. As used in the present invention, the term coding sequence means, for example, a DNA sequence encoding a functional RNA, peptide, polypeptide or protein. Typically, the coding sequence is cDNA for a transgene that encodes a functional RNA, peptide, polypeptide or target protein. Functional RNAs, peptides, polypeptide and target proteins are described below.

It is obvious that the genome of the virus contains coding DNA sequences. Endogenous (naturally occurring genes) in the genomic sequence of the virus are not considered transgenes in the context of the present application, unless they have been modified by recombinant methods so that they are in a non-natural position or in a non-natural environment.

In one embodiment, the term transgene, as used in the present invention, refers to a segment of DNA containing a gene or cDNA sequence that has been isolated from one organism and introduced into another organism, i.e., a virus according to the present invention. In one embodiment, said non-native segment of DNA may retain the ability to produce functional RNA, peptide, polypeptide or protein.

Thus, in one embodiment, the inserted transgene encodes a human or humanized protein, polypeptide, or peptide.

In one embodiment of the invention, the inserted transgene encodes a non-human protein, polypeptide or peptide molecule (such as a protein, polypeptide or peptide of a non-human mammal) or an RNA molecule, for example, from a mouse, rat, rabbit, camel, llama or the like. Advantageously, the viruses of the present invention allow the transgenes to be transported into the cancer cell. Thus, responses generated by a human patient to a non-human sequence (e.g., a protein) can be minimized by this intracellular delivery.

A DNA sequence may contain more than one transgene, such as 1, 2, 3, or 4 transgenes, such as 1 or 2.

A transgene cassette may contain more than one transgene, such as 1, 2, 3, or 4 transgenes, such as 1 or 2.

In one or more embodiments, the cassette is positioned as shown in one or more figures or examples.

As used herein, the term transgene cassette refers to a DNA sequence encoding one or more transgenes in the form of one or more coding sequences and one or more regulatory elements.

The transgene cassette may encode one or more monocistronic and/or polycistronic mRNA sequences.

In one embodiment, the transgene or transgene cassette encodes a monocistronic or polycistronic mRNA, and, for example, such a cassette is suitable for insertion into the adenoviral genome at a specific position under the control of an endogenous promoter or an exogenous promoter or a combination thereof.

As used in the present invention, the term monocistronic mRNA refers to an mRNA molecule encoding a single functional RNA, peptide, polypeptide or protein.

In one embodiment, the transgene cassette encodes a monocistronic mRNA.

- 15 043895

In one embodiment, a transgene cassette in the context of a cassette encoding a monocistronic mRNA means a segment of DNA optionally comprising an exogenous promoter (which is a regulatory sequence that will determine where and when said transgene is active) or a splice site (which is a regulatory sequence that will determine when the mRNA molecule will be cleaved by the spliceosome), a coding sequence (i.e., a transgene), typically derived from cDNA for a target protein, optionally comprising a polyA signal sequence, and a terminator sequence.

In one embodiment, the transgene cassette may encode one or more polycistronic mRNA sequences.

The term polycistronic mRNA in the context of the present application refers to an mRNA molecule encoding two or more functional RNAs, peptides or proteins, or a combination thereof. In one embodiment, the transgene cassette encodes a polycistronic mRNA.

In one embodiment, a transgene cassette, in the context of a cassette encoding a polycistronic mRNA, comprises a DNA segment optionally comprising an exogenous promoter (which is a regulatory sequence that will determine where and when the transgene is active) or a splice site (which is a regulatory sequence that will determine when the mRNA molecule will be cleaved by the spliceosome), two or more coding sequences (i.e., transgenes), typically derived from cDNA for a target protein or peptide, such as where each coding sequence is separated by either an IRES or a 2A peptide. After the last coding sequence to be transcribed, the cassette may optionally comprise a polyA sequence and a terminator sequence.

In one embodiment, the transgene cassette encodes a monocistronic mRNA followed by a polycistronic mRNA. In another embodiment, the transgene cassette is a polycistronic mRNA followed by a monocistronic mRNA.

In one embodiment, the adenovirus is a human adenovirus. As used herein, the terms adenovirus, serotype, or adenoviral serotype refer to any adenovirus that can be classified into any of the more than 50 currently known adenoviral serotypes classified into subgroups A-F, and further extends to any as yet unidentified or unclassified adenoviral serotypes. See, e.g., Strauss, Adenovirus Infection in Humans, in The Adenoviruses, Ginsberg, ea., Plenum Press, New York, NY, pp. 451-596 (1984) and Shenk, Adenoviridae: The Viruses and Their Replication, in Fields Virology, Vol. 2, Fourth Edition, Knipe, 35ea., Lippincott Williams & Wilkins, pp. 2265-2267 (2001), as shown in Table 1. 1.

Table 1

Subgroup Adenovirus serotype A 12,18,31 IN 3,7,11,14,16,21,34,35,51 WITH 1,2,5,6 D 8-10,13,15,17,19,20,22-30,32,33,36-39,42-49,50 E 4 F 40,41

The adenoviruses of the present invention are viruses of subgroup B, namely Ad11, in particular Ad11p (Slobitski strain) and derivatives thereof such as EnAd.

Adenoviruses are classified into their groups/serotypes based on the capsid, such as hexon and/or fiber.

The adenovirus of the present invention is not a group A, C, D, E or F virus. The viruses of the present invention do not contain an adenoviral death protein.

In one embodiment, the adenovirus of the present invention is chimeric. When an adenovirus is chimeric, then the characteristics of the outer capsid will be used to determine the serotype. As used herein, the term chimeric refers to a virus that contains DNA from at least two different viral serotypes, including different serotypes within the same group.

In one embodiment, the oncolytic virus has hexon, fiber and penton proteins of the same serotype, such as Ad11, in particular Ad11p, such as those located at positions 3081231789, 18254-21100 and 13682-15367 of the genomic sequence of the latter, where the nucleotide positions are indicated relative to GenBank ID 217307399 (Accession Number: GC689208).

In one embodiment, the adenovirus is enadenotucirev (also known as EnAd and previously as EnAd). As used herein, the term Enadenotucirev refers to the chimeric adenovirus of SEQ ID NO: 38. It is a replication competent, oncolytic chimeric adenovirus that has improved therapeutic properties compared to wild-type adenoviruses (see WO2005/118825). EnAd has a chimeric E2B region that contains DNA from Ad11p and Ad3 and deletions in E3/E4. These structural changes in enadenotucirev result in a genome that is approximately 3.5 kb smaller than Ad11p, thereby providing additional space for transgene insertion.

Enadenotucirev (EnAd) is a chimeric oncolytic adenovirus, previously known as EnAd (WO2005/118825), with the fiber, penton and hexon of Ad11p, hence it is a subgroup B virus. It has a chimeric E2B region that includes DNA from Ad11p and Ad3. In EnAd, almost all of the E3 region and part of the E4 region have been deleted. It therefore has significant genomic space to accommodate additional genetic material while remaining viable. Furthermore, since EnAd is a subgroup B adenovirus, pre-existing immunity in humans is less common than for example Ad5. Other examples of chimeric oncolytic viruses with the fiber, penton and hexon of Ad11 include OvAd1 and OvAd2 (see WO2006/060314).

EnAd appears to preferentially infect tumor cells, rapidly replicate in these cells, and cause their lysis. This, in turn, may generate inflammatory immune responses, thereby stimulating the body to also fight cancer. Part of EnAd’s success is thought to be due to the virus’s rapid replication in vivo.

Although EnAd selectively lyses tumor cells, it may be possible to introduce additional beneficial properties, such as enhancing the therapeutic activity of the virus or reducing the side effects of the virus, by arming it with transgenes, such as a transgene that encodes a protein, signaling protein, or antibody, or a transgene that encodes an entity that stimulates cellular signaling protein(s).

What is useful is the ability to weaponize a virus by providing it with DNA encoding specific proteins, such as a bispecific T-cell activator, that can be expressed inside a cancer cell to exploit the body's own defense mechanisms to more effectively fight tumor cells, for example by making these cells more visible to the immune system or by delivering a therapeutic gene/protein that preferentially targets tumor cells.

In addition, the ability to insert transgenes into the genome that are reporters may aid in clinical or preclinical studies.

It is important that the expression of the transgenes does not adversely affect the replication or other beneficial properties of the virus. Therefore, the gene or genes must be inserted in a position that does not compromise the ability to replicate or other beneficial properties of the virus. In addition, the adenovirus genome is tightly packed, and therefore it can be difficult to find a suitable place to insert transgenes. This also limits the size of the transgenes that can be accommodated.

OvAd1 and OvAd2 are also chimeric adenoviruses similar to enadenotucirev, which also have additional space in the genome (see WO2008/080003). Thus, in one embodiment, the adenovirus is OvAd1 or OvAd2.

In one embodiment, the adenovirus is oncolytic. As used herein, the term oncolytic adenovirus means an adenovirus that preferentially kills cancer cells over non-cancerous cells.

In one embodiment, the oncolytic virus is apoptotic, meaning it accelerates programmed cell death.

In one embodiment, the oncolytic virus is cytolytic. The cytolytic activity of the oncolytic adenoviruses of the present invention can be determined in representative tumor cell lines and the data can be converted to a measure of activity, for example, using an adenovirus belonging to subgroup C, such as Ad5, as a standard (i.e., assigning it an activity = 1). A suitable method for determining cytolytic activity is the MTS assay (see Example 4, Fig. 2 of WO 2005/118825, incorporated herein by reference).

In one embodiment, the oncolytic virus is necrolytic. That is, it causes or accelerates cell death or death of immunogenic cells. In one embodiment, necrolytic cell death is advantageous because it triggers, induces immune responses in patients (the host).

Unless otherwise indicated in the context, adenovirus as used in the present invention refers to a replication-competent virus (e.g., a replication-competent virus) as well as replication-defective viral vectors.

As used herein, the term "replication competent" refers to a replication competent virus or a virus whose replication is dependent on a factor in cancer cells, such as an activated factor such as p53 or the like.

In one embodiment, the virus is replication competent. The term replication competent in the context of the present application refers to a virus that has all the necessary mechanisms to replicate in cells in vitro and in vivo, i.e. without the help of a packaging cell line. A viral vector, for example, deleted in the E1 region, capable of replicating in a complementary packaging cell line is not a replication competent virus in this context.

Viral vectors are replication defective and require a packaging cell to provide a complementary gene to allow replication.

As used herein, the term adenovirus genome means a DNA sequence encoding structural proteins and elements relevant to the function/life cycle of an adenovirus.

All human adenovirus genomes examined to date have the same general organization, i.e., genes encoding specific functions are located at the same position in the viral genome (referred to herein as structural elements). Each end of the viral genome has a short sequence known as an inverted terminal repeat (or ITR), which is essential for viral replication. The viral genome contains five early transcription units (E1A, E1B, E2, E3, and E4), three delayed early units (IX, IVa2, and E2 late), and one late unit (major late), which is processed to generate five families of late mRNAs (L1-L5). Proteins encoded by the early genes are primarily involved in replication and modulation of the host cell response to infection, whereas the late genes encode viral structural proteins. Early genes begin with the prefix E, and late genes begin with the prefix L.

The genome of adenoviruses is tightly packed, i.e. there is only a small non-coding sequence, and therefore it can be difficult to find a suitable site for insertion of transgenes. The present inventors have identified two regions of DNA where transgenes are tolerated, in particular the identified sites are suitable for placement of complex transgenes, such as encoding antibodies. In this way, the transgene is expressed without compromising viability, native properties of the virus, such as oncolytic properties or replication.

In one embodiment, an oncolytic or partially oncolytic virus of the present invention may be the result of a deletion in the E4 and/or E3 region, such as with a deletion in part of the E4 region or a complete deletion in the E3 region, or alternatively with a deletion in part of the E4 region (such as E4orf4) and a complete deletion in the E3 region, such as shown in the examples disclosed herein.

In one embodiment, the oncolytic virus of the invention is chimeric. As used herein, the term chimeric refers to a virus that contains DNA from two or more different serotypes and has the properties of an oncolytic virus.

In one embodiment, the oncolytic virus is EnAd or an active derivative thereof that retains the essential beneficial properties of the virus. EnAd is disclosed in WO 2005/118825 (incorporated herein by reference), and the complete sequence for the virus is provided herein as SEQ ID NO: 38. The chimeric E2B region is disclosed herein as SEQ ID NO: 71.

Alternative oncolytic viruses include OvAd1 and OvAd2, which are respectively disclosed as SEQ ID NO: 2 and 3 in WO2008/080003 and incorporated herein by reference.

An advantage of the adenoviruses of the present invention is that they exhibit similar profiles of viral activity, such as replication and/or infectivity, compared to EnAd after infection of various colon cancer cell lines in vitro.

Structural elements of adenoviruses

The present invention also relates to novel sequences of viruses or viral components/constructs, such as plasmids, disclosed in the present invention.

In one embodiment, the adenovirus comprises a genome comprising a sequence of formula (I)

5TTR-Bi-B _a -B ₂ -B _x -B _b -B _y -B ₃ -3'ITR (I) where B1 comprises E1A, E1B or E1A-E1B; B _A comprises E2B-L1-L2-L3-E2A-L4; B ₂ is a linkage or comprises E3; B _X is a linkage or a DNA sequence comprising a restriction site, one or more transgenes (in particular a transgene encoding at least one bispecific T cell activator according to the present invention, for example under the control of an exogenous promoter) or both; BB comprises L5; BY is a linkage or a DNA sequence comprising a restriction site, one or more transgenes (in particular a transgene encoding at least one bispecific T cell activator according to the invention, for example under the control of an endogenous promoter such as MPL or under the control of an exogenous promoter such as the CMV promoter), or both; _B3 is a linkage or comprises E4; wherein at least one of _BX and BY is not a linkage and comprises a transgene or a restriction site, or both; and encodes a multispecific antigen molecule comprising at least two binding domains and at least one of said domains is specific for a surface antigen on a T cell of interest. In one embodiment, the adenovirus comprises a genome comprising a sequence of formula (I), wherein B1 and _BX is a linkage.

In one embodiment, the adenovirus comprises a genome comprising a sequence of formula (I), where BY is a linkage.

In one embodiment, the adenovirus comprises a genome comprising a sequence of formula (Ia):

5'ITR-B _a -B ₂ -B _x -B _b -BY-B ₃ -3'ITR (Ia) wherein BA comprises E2B-L1-L2-L3-E2A-L4; B ₂ is a linkage or comprises E3; BX is a linkage or a DNA sequence comprising a restriction site, one or more transgenes (in particular a transgene encoding at least one bispecific T cell activator according to the present invention, for example under the control of an exogenous promoter), or both; BB comprises L5; BY is a linkage or a DNA sequence comprising a restriction site, one or more transgenes (in particular a transgene encoding at least one bispecific T cell activator according to the present invention, for example under the control of an endogenous promoter such as MPL or under the control of an exogenous promoter such as the CMV promoter), or both; B ₃ is a linkage or comprises E4; wherein at least one of BX and BY is not a linkage and at least one comprises a transgene or a restriction site such as a transgene.

In one embodiment, the adenovirus comprises a genome comprising a sequence of formula (Ia), wherein B _X is a bond.

In one embodiment, the adenovirus comprises a genome comprising a sequence of formula (Ia), wherein B _Y is a bond.

In one embodiment, the adenovirus comprises a genome comprising a sequence of formula (Ib):

5TTR-BA-B _x -B _b -B _y -B ₃ -3TTR (Ib) where BA comprises E2B-L1-L2-L3-E2A-L4; BX is a linkage or a DNA sequence comprising a restriction site, one or more transgenes (in particular a transgene encoding at least one bispecific T cell activator according to the present invention, for example under the control of an exogenous promoter), or both; BB comprises L5; BY is a linkage or a DNA sequence comprising a restriction site, one or more transgenes (in particular a transgene encoding at least one bispecific T cell activator according to the present invention, for example under the control of an endogenous promoter such as MPL or under the control of an exogenous promoter such as the CMV promoter), or both; B ₃ is a linkage or comprises E4; wherein at least one of B _X and B _Y is not a linkage and comprises a trans gene or a restriction site or both, such as a transgene.

In one embodiment, the adenovirus comprises a genome comprising a sequence of formula (Ib), wherein BX is a bond.

In one embodiment, the adenovirus comprises a genome comprising a sequence of formula (Ib), wherein BY is a linkage.

In one embodiment, the adenovirus comprises a genome comprising a sequence of formula (Ic):

5TTR-BA-B2-BX-BB-BY-3TTR (Ic) wherein BA comprises E2B-L1-L2-L3-E2A-L4; B ₂ is E3; BX is a linkage or a DNA sequence comprising a restriction site, one or more transgenes (in particular a transgene encoding at least one bispecific T cell activator according to the present invention, for example under the control of an exogenous promoter), or both; BB comprises L5; BY is a linkage or a DNA sequence comprising a restriction site, one or more transgenes (in particular a transgene encoding at least one bispecific T cell activator according to the present invention, for example under the control of an endogenous promoter such as MPL or under the control of an exogenous promoter such as the CMV promoter), or both; wherein at least one of BX and BY is not a linkage and comprises a transgene or a restriction site or both, such as a transgene.

In one embodiment, the adenovirus comprises a genome comprising a sequence of formula (Ic), wherein B _X is a bond.

In one embodiment, the adenovirus comprises a genome comprising a sequence of formula (Ic), wherein B _Y is a bond.

In one embodiment, the adenovirus comprises a genome comprising a sequence of the formula (Id):

5TTR-B ₁ -BA-B _x -B _b -B _y -B ₃ -3TTR (Id) where B1 comprises E1A, E1B or E1A-E1B; Bd comprises E2B-L1-L2-L3-E2A-L4; BX is a linkage or DNA sequence comprising a restriction site, one or more transgenes (in particular a transgene encoding at least one bispecific T cell activator according to the present invention, for example under the control of an exogenous promoter) or both; BB comprises L5; BY is a linkage or a DNA sequence comprising a restriction site, one or more transgenes (in particular a transgene encoding at least one bispecific T cell activator according to the present invention, for example under the control of an endogenous promoter such as MPL or under the control of an exogenous promoter such as the CMV promoter), or both; _B3 is a linkage or comprises E4; wherein at least one of BX and BY is not a linkage and comprises a transgene or a restriction site, or both.

In one embodiment, the adenovirus comprises a genome comprising a sequence of formula (Id), wherein BX is a bond.

In one embodiment, the adenovirus comprises a genome comprising a sequence of formula (Id), where B _Y is a bond.

In one embodiment, the adenovirus comprises a genome comprising a sequence of formula (Ie):

5TTR-B1-B _a -B2-B _b -BY-3TTR (Ie) wherein B1 comprises E1A, E1B or E1A-E1B; B2 comprises E2B-L1-L2-L3-E2A-L4; _B2 comprises E3; _Bb comprises L5; BY is a linkage or a DNA sequence comprising a restriction site, one or more transgenes (in particular a transgene encoding at least one bispecific T cell activator according to the present invention, for example under the control of an endogenous promoter such as MPL or under the control of an exogenous promoter such as the CMV promoter); wherein at least one of _BX and _BY is not a linkage and comprises a transgene or a restriction site or both.

In one embodiment, a compound of formula (I), (Ia), (Ib), (Ic), (Id) or (Ie) is provided, wherein at least one of B _X and B _Y is not a linkage and comprises a transgene or a restriction site or both, such that BX and BY are both transgenes.

In one embodiment of formula (I), (Ia), (Ib), (Ic) or (Id), only BX encodes one or two bispecific T cell activators, such as one bispecific T cell activator (and BY does not encode a bispecific T cell activator), in particular said bispecific T cell activator or more bispecific T cell activators are under the control of an exogenous promoter, such as the CMV promoter. In one embodiment of formula (I), (Ia), (Ib), (Ic) or (Id), only B _Y encodes one or two bispecific T cell activators, such as one bispecific T cell activator (and B _X does not encode a bispecific T cell activator), in particular said bispecific T cell activator or more bispecific T cell activators are under the control of an exogenous promoter, such as the CMV promoter. In one embodiment of formula (I), (Ia), (Ib), (Ic) or (Id), B _X encodes a bispecific T cell activator (e.g., under the control of an exogenous promoter, such as the CMV promoter) and BY encodes a bispecific T cell activator (e.g., under the control of an endogenous promoter, such as MPL, or under the control of an exogenous promoter, such as the CMV promoter).

The term "linkage" refers to a covalent bond joining one DNA sequence to another DNA sequence, such as joining one region of a viral genome to another. Thus, when a variable in formula (I) (Ia), (Ib), (Ic), (Id) or (Ie) in the present invention represents a linkage, the feature or element represented by this linkage is absent, i.e., removed.

Since the structure of adenoviruses is generally similar, the elements below are discussed in terms of the structural elements and the commonly used nomenclature relating to them, which are known to those skilled in the art. When an element is mentioned in the present invention, we mean the DNA sequence encoding the element or the DNA sequence encoding the same structural protein of the element in an adenovirus. The latter is applicable due to the redundancy of the DNA code. It may be necessary to take into account the preferences of viruses for codon usage to obtain optimal results.

Any structural element from an adenovirus used in the viruses of the present invention may comprise or consist of a natural sequence or may have at least 95%, such as 96, 97, 98, 99 or 100% similarity to it over a given length. The original sequence may be modified to exclude 10, 9, 8, 7, 6, 5, 4, 3, 2% or 1% of the genetic material. The skilled person knows that when making changes, the reading frames of the virus should not be disrupted so that the expression of the structural proteins is disrupted.

In one embodiment, the element in question is a full-length sequence, i.e., a full-length gene.

In one embodiment, the element in question is less than the full length and retains the same or corresponding function as the full length sequence.

In one embodiment, for a subject element that is optional in the constructs of the present invention, the DNA sequence may be smaller than the full length and lack functionality.

Structural genes encoding structural or functional proteins of adenovirus are typically linked by non-coding regions of DNA. Thus, there is some flexibility as to where to cut the genomic sequence of a target structural element (especially its non-coding regions) for the purpose of inserting a transgene into the viruses of the present invention. Thus, for the purposes of the present application, the element will be considered a reference structural element in that

- 20 043895 the extent to which it is relevant to the target and does not encode extraneous material. Thus, if necessary, the gene will be associated with suitable non-coding regions, such as occurs in the natural structure of the virus.

Thus, in one embodiment, the insertion, for example, when DNA encoding a restriction site and/or a transgene, is inserted into a non-coding region of the viral genomic DNA, such as an intron or intergenic sequence. However, some non-coding regions of the adenovirus may have some function, for example, in alternative splicing, transcriptional regulation or translational regulation, and this may need to be taken into account.

The sites identified in the present invention that are associated with the L5 region (e.g. between L5 and the E4 region) are suitable for accommodating a variety of DNA sequences encoding complex objects such as RNAi, cytokines, single-stranded or multimeric proteins such as antibodies, e.g. bispecific T cell activators.

As used herein, the term gene refers to coding and any non-coding sequences associated therewith, such as introns and associated exons. In one embodiment, a gene comprises or consists of only essential structural components, such as a coding region.

A discussion concerning specific structural elements of adenoviruses follows.

Inverted terminal repeat (ITR) sequences are common to all known adenoviruses, are named for their symmetry, and are the origin of viral chromosome replication. Another property of these sequences is their ability to form a hairpin.

As used herein, the designation 54TR refers to a portion or all of the ITR from the 5' end of an adenovirus that retains ITR function when incorporated into the adenovirus at the appropriate location. In one embodiment, the 5'ITR comprises or consists of a sequence of approximately 1-138 bp of SEQ ID NO: 38 or a sequence that is 90%, 95%, 96%, 97%, 98% or 99% identical thereto over its entire length, in particular a sequence consisting of approximately 1-138 bp of SEQ ID NO: 38.

As used herein, the term 3'ITR refers to a portion or all of the ITR from the 3' end of an adenovirus that retains ITR function when incorporated into the adenovirus at the appropriate location. In one embodiment, the 3'ITR comprises or consists of a sequence of approximately bp 32189-32326 of SEQ ID NO: 38 or a sequence that is 90%, 95%, 96%, 97%, 98% or 99% identical thereto over its entire length, in particular a sequence consisting of approximately bp 32189-32326 of SEQ ID NO: 38.

As used in the present invention, the designation B1 refers to a DNA sequence encoding part or all of the E1A region of an adenovirus, part or all of the E1B region of an adenovirus, and independently part or all of the E1A and E1B region of an adenovirus.

When B1 is a link, the E1A and E1B sequences will be excluded from the virus. In one embodiment, B1 is a link, and thus the virus is a vector.

In one embodiment, B1 also contains a transgene. It is known in the art that the E1 region can accommodate a transgene, which can be disruptively introduced into the E1 region (i.e., into the middle of the sequence), or part or all of the E1 region can be removed to provide more space for the genetic material.

As used herein, the designation E1A refers to the coding portion of the DNA sequence or the entire E1A region of an adenovirus. The latter in this case refers to the E1A polypeptide/protein. It may be mutated such that the protein encoded by the E1A gene has conservative or non-conservative amino acid substitutions such that it has the same function as the wild type (i.e., the corresponding non-mutated protein); an increased function compared to the wild type protein; a decreased function, such as no function compared to the wild type protein; or a new function compared to the wild type protein, or a combination thereof, as appropriate.

As used in the present invention, the designation E1B refers to a portion of the coding DNA sequence or the entire E1B region of an adenovirus (i.e., a polypeptide or a protein), it may be mutated such that the protein encoded by the E1B gene/region has conservative or non-conservative amino acid substitutions, such as that it has the same function as the wild type (i.e., the corresponding non-mutated protein); an increased function compared to the wild type protein; a decreased function, such as no function compared to the wild type protein; or has a new function compared to the wild type protein, or a combination thereof, as appropriate.

Thus, B1 may or may not be modified relative to the wild-type _E1 region, such as wild-type E1A and/or E1B. One skilled in the art can easily determine whether E1A and/or E1B are present or (partially) deleted or mutated.

As used herein, the term wild type refers to a known adenovirus. A known adenovirus is an adenovirus that has been identified and named, regardless of whether its sequence is available.

- 21 043895

In one embodiment, B1 has a sequence from 139 to 3932 bp of SEQ ID NO: 38.

As used herein, the designation BA refers to a DNA sequence encoding the E2B-L1-L2-L3-E2A-L4 regions, including any non-coding sequences as appropriate. Typically, this sequence does not contain a transgene. In one embodiment, the sequence is substantially similar to or identical to an adjacent sequence from a known adenovirus, such as a serotype shown in Table 1, in particular a group B virus, such as Ad3, Ad7, Ad11, Ad14, Ad16, Ad21, Ad34, Ad35, Ad51, or a combination thereof, such as Ad3, Ad11, or a combination thereof. In one embodiment, E2B-L1-L2-L3-E2A-L4 refers to the inclusion of these elements and other structural elements associated with the region, for example, Bd typically includes the sequence encoding the IV2a protein, for example as follows: IV2A IV2a-E2B-L1-L2-L3E2A-L4.

In one embodiment, the E2B region is chimeric. That is, it contains DNA sequences from two or more different adenovirus serotypes, such as Ad3 and Ad11, such as Ad11p. In one embodiment, the E2B region has a sequence from 5068 to 10355 bp of SEQ ID NO: 38 or a sequence that is 95, 96, 97, 98, or 99% identical thereto over its entire length.

In one embodiment, E2B in component Bd comprises the sequences shown in SEQ ID NO: 71 (which corresponds to SEQ ID NO: 3 disclosed in WO2005/118825).

In one embodiment, Bd has a sequence from 3933 to 27184 bp of SEQ ID NO: 38.

As used herein, the term E3 refers to a portion of the DNA coding sequence or the entire E3 region of an adenovirus (i.e., protein/polypeptide), which may be mutated such that the protein encoded by the E3 gene has conservative or non-conservative amino acid substitutions such that it has the same function as the wild type (the corresponding non-mutated protein); an increased function compared to the wild type protein; a decreased function, such as no function compared to the wild type protein; or a new function compared to the wild type protein, or a combination thereof, as appropriate.

In one embodiment, the E3 region is derived from an adenovirus serotype listed in Table 1, or a combination thereof, in particular a group B serotype, such as Ad3, Ad7, Ad11 (in particular Ad11p), Ad14, Ad16, Ad21, Ad34, Ad35, Ad51, or a combination thereof, such as Ad3, Ad11 (in particular Ad11p), or a combination thereof.

In one embodiment, region E3 is partially removed, such as removed by 95, 90, 85, 80, 75, 70, 65, 60, 55, 50, 45, 40, 35, 30, 25, 20, 15, 10, 5%.

In one embodiment, _B2 is a linkage where the DNA encoding the E3 region is absent.

In one embodiment, the DNA encoding the E3 region may be replaced or interrupted by a transgene. As used herein, the term E3 region replaced by a transgene includes part or all of the E3 region replaced by a transgene.

In one embodiment, region _B2 comprises the sequence from 27185 bp to 28165 bp of SEQID NO: 38.

In one embodiment, _B2 consists of the sequence from 27185 bp to 28165 bp of SEQ ID NO: 38.

As used herein, the designation BX refers to a DNA sequence in the vicinity of the 5' end of the L5 gene in B _B . As used herein, the term "near or proximal to the 5' end of the L5 gene" means adjacent to the 5' end of the L5 gene or a non-coding region inherently associated therewith, i.e., adjacent to or contiguous with the 5' end of the L5 gene or a non-coding region inherently associated therewith. Alternatively, "near or proximal" may refer to proximity to the L5 gene such that there are no coding sequences between the BX region and the 5' end of the L5 gene.

Thus, in one embodiment, BX is attached directly to the L5 base, which represents, for example, the beginning of the coding sequence of the L5 gene.

Thus, in one embodiment, B _X is attached directly to the base of L5, which is, for example, the beginning of a non-coding sequence, or is attached directly to a non-coding region naturally associated with L5. As used in the present invention, the term non-coding region naturally associated with L5 refers to the portion of all non-coding regions that is part of the L5 gene or is adjacent to it, but is not part of another gene.

In one embodiment, the BX comprises the sequence of SEQ ID NO: 39. This sequence is an artificial non-coding sequence into which a DNA sequence can be inserted, for example, containing a transgene (or transgene cassette), a restriction site, or a combination thereof. This sequence is useful because it acts as a buffer in that it provides some flexibility regarding the exact location of the transgene, while minimizing the disruptive effect on the stability and viability of the virus.

- 22 043895

The insertion(s) may occur anywhere in SEQ ID NO: 39 from the 5' end, the 3' end, or at any point between 1 and 201 bp, such as between base pairs 1/2, 2/3, 3/4, 4/5, 5/6, 6/7, 7/8, 8/9,

9/10, 10/11, 11/12, 12/13, 13/14, 14/15, 15/16, 16/17, 17/18, 18/19, 19/20, 20/21, 21/ 22, 22/23, 23/24, 24/25,

25/26, 26/27, 27/28, 28/29, 29/30, 30/31, 31/32, 32/33, 33/34, 34/35, 35/36, 36/37, 37/ 38, 38/39, 39/40, 40/41,

41/42, 42/43, 43/44, 44/45, 45/46, 46/47, 47/48, 48/49, 49/50, 50/51, 51/52, 52/53, 53/ 54, 54/55, 55/56, 56/57,

57/58, 58/59, 59/60, 60/61, 61/62, 62/63, 63/64, 64/65, 65/66, 66/67, 67/68, 68/69, 69/ 70, 70/71, 71/72, 72/73,

73/74, 74/75, 75/76, 76/77, 77/78, 78/79, 79/80, 80/81, 81/82, 82/83, 83/84, 84/85, 85/ 86, 86/87, 87/88, 88/89,

89/90, 90/91, 91/92, 92/93, 93/94, 94/95, 95/96, 96/97, 97/98, 98/99, 99/100, 100/101, 101/ 102, 102/103, 103/104, 104/105, 105/106, 106/107, 107/108, 108/109, 109/110, 110/111, 111/112, 112/113, 113/114, 114/115,

115/116, 116/117, 117/118, 118/119, 119/120, 120/121, 121/122, 122/123, 123/124, 124/125, 125/126, 126/127,

127/128,128/129, 129/130, 130/131, 131/132, 132/133, 133/134, 134/135, 135/136, 136/137, 137/138, 138/139, 139/140, 140/141, 141/142, 142/143, 143/144, 144/145, 145/146, 146/147, 147/148, 148/149, 150/151, 151/152,

152/153, 153/154, 154/155, 155/156, 156/157, 157/158, 158/159, 159/160, 160/161, 161/162, 162/163, 163/164,

164/165, 165/166, 166/167, 167/168, 168/169, 169/170, 170/171, 171/172, 172/173, 173/174, 174/175, 175/176,

176/177, 177/178, 178/179, 179/180, 180/181, 181/182, 182/183, 183/184, 184/185, 185/186, 186/187, 187/188,

189/190, 190/191, 191/192, 192/193, 193/194, 194/195, 195/196, 196/197, 197/198, 198/199, 199/200 or 200/201.

In one embodiment, BX comprises SEQ ID NO: 39 with a DNA sequence inserted between bp 27 and bp 28, or at a location corresponding to a location between positions 28192 bp and 28193 bp of SEQ ID NO: 38.

In one embodiment, the insert is a restriction site insert. In one embodiment, the restriction site insert comprises one or two restriction sites. In one embodiment, the restriction site is a 19 bp restriction site insert comprising 2 restriction sites. In one embodiment, the restriction site insert is a 9 bp restriction site insert comprising 1 restriction site. In one embodiment, the restriction site insert comprises one or two restriction sites and at least one transgene, such as one or two transgenes. In one embodiment, the restriction site is a 19 bp restriction site insert comprising 2 restriction sites and at least one transgene, such as one or two transgenes. In one embodiment, the restriction site insertion is a 9 bp restriction site insertion comprising 1 restriction site and at least one transgene, such as one, two or three transgenes, such as one or two transgenes. In one embodiment, the two restriction sites sandwich one or more, such as two transgenes (e.g., in a transgene cassette). In one embodiment, when the BX comprises two restriction sites, the restriction sites are different from each other. In one embodiment, the one or more restriction sites in the BX do not naturally occur in the particular adenovirus genome into which they have been inserted. In one embodiment, the one or more restriction sites in BX are different from other restriction sites located elsewhere in the adenovirus genome, such as different from naturally occurring restriction sites and/or restriction sites introduced into other parts of the genome, such as the restriction site introduced into BY. Thus, in one embodiment, the restriction site or sites allow DNA to be cut at a particular location.

The use of unique restriction sites provides the advantage of allowing selectivity and control over where exactly the viral genome is cut, simply by using the appropriate restriction enzyme.

Site-specific cutting, as used in the present invention, refers to the case where the use of a restriction site-specific enzyme cuts the virus only at the desired site, usually at one site, although sometimes it may be a pair of sites. As used in the present invention, the term site pair refers to two restriction sites in close proximity to each other that are intended to be cut by the same enzyme (i.e., cannot be distinguished from each other).

In one embodiment, the restriction site insert is SEQ ID NO: 50.

In one embodiment, BX has a sequence from 28166 bp to 28366 bp of SEQ ID NO: 38.

In one embodiment, B _X represents a bond.

In one embodiment, BX comprises a restriction site, such as 1, 2, 3 or 4 restriction sites, such as 1 or 2. In one embodiment, _BX comprises at least one transgene, such as 1 or 2 transgenes. In one embodiment, BX comprises at least one transgene, such as 1 or 2 transgenes, and one or more restriction sites, such as 2 or 3 restriction sites, particularly where the restriction sites sandwich a gene or DNA sequence containing genes, allowing it/them to be excised at a specific location from the genome and/or replaced. Alternatively, the restriction sites may sandwich each gene, such that when there are two transgenes, three different restriction sites are needed to ensure that the genes can be selectively excised and/or replaced. In one embodiment, one or more, e.g., all, of the transgenes are in the form of a transgene cassette. In one embodiment, _BX comprises SEQ ID NO: 39. In one embodiment, SEQ ID NO: 39 is interrupted by, e.g., a transgene. In one embodiment, SEQ ID NO: 39 is continuous. In one embodiment, BX does not comprise a restriction site. In one embodiment, BX is a linkage. In one embodiment, BX comprises or consists of one or more transgenes.

In one embodiment, BY comprises a restriction site, such as 1, 2, 3 or 4 restriction sites, such as 1 or 2. In one embodiment, BY comprises at least one transgene, such as 1 or 2 transgenes. In one embodiment, B _Y comprises at least one transgene, such as 1 or 2 transgenes, and one or more restriction sites, such as 2 or 3 restriction sites, particularly where the restriction sites sandwich a gene or DNA sequence containing genes, allowing it/them to be excised at a specific location from the genome and/or replaced. Alternatively, the restriction sites may sandwich each gene, such that when there are two transgenes, three different restriction sites are required to ensure that the genes can be selectively excised and/or replaced. In one embodiment, one or more, such as all, transgenes are in the form of a transgene cassette. In one embodiment, BY comprises SEQ ID NO: 40. In one embodiment, SEQ ID NO: 40 is interrupted by, such as, a transgene. In one embodiment, SEQ ID NO: 40 is continuous. In one embodiment, B _Y does not comprise a restriction site. In one embodiment, B _Y is a linkage. In one embodiment, B _Y comprises or consists of one or more transgenes.

In one embodiment, B _X and B _Y each comprise a restriction site, such as 1, 2, 3 or 4 restriction sites, such as 1 or 2. In one embodiment, B _X and B _Y each comprise at least one transgene, such as 1 or 2 transgenes. In one embodiment, BX and BY each comprise at least one transgene, such as 1 or 2 transgenes, and one or more restriction sites, such as 2 or 3 restriction sites, particularly where the restriction sites sandwich the genome or DNA sequence containing the genes, allowing it or her to be excised from the genome and/or replaced. Alternatively, the restriction sites may sandwich each gene, such that when there are two transgenes, three different restriction sites are required to ensure that the genes can be selectively excised and/or replaced. In one embodiment, one or more, such as all, of the transgenes are in the form of a transgene cassette. In one embodiment, BX and BY comprise SEQ ID NO: 39 and SEQ ID NO: 40, respectively. In one embodiment, B _X and B _Y do not comprise a restriction site. In one embodiment, B _X is a linkage and BY is not a linkage. In one embodiment, BY is a linkage and BX is not a linkage.

As used in the present invention, the term BB means a DNA sequence encoding the L5 region. In this context, the L5 region refers to a DNA sequence containing a gene encoding a fiber polypeptide/protein, as the case may be. The fiber gene/region encodes the fiber protein, which is the main component of the capsid of adenoviruses. Fiber functions in receptor recognition and contributes to the ability of the adenovirus to selectively bind and infect cells.

In the viruses of the present invention, the fiber may be from any adenovirus strain of serotype 11, such as Ad11p.

In one embodiment, BB has a sequence from 28367 bp to 29344 bp of SEQ ID NO: 38.

A DNA sequence relative to BY as used herein refers to a DNA sequence near the 3' end of the L5 gene of BB. Near or proximal to 3' near the L5 gene as used herein means adjacent to the 3' end of the L5 gene or a non-coding region inherently associated therewith, i.e. adjacent to the 3' end of the L5 gene or a non-coding region inherently associated therewith (i.e. all or part of the non-coding sequence endogenous to L5). Alternatively, near or proximal may refer to proximity to the L5 gene such that there are no coding sequences between the BY region and the 3' end of the L5 gene.

Thus, in one embodiment, BY is linked directly to base L5, which represents the end of the coding sequence.

Thus, in one embodiment, B _Y is linked directly to the base of L5, which represents the end of the non-coding sequence, or is linked directly to the non-coding region naturally associated with L5.

The terms are by their nature and naturally used interchangeably in the present invention. In one embodiment, BY comprises the sequence of SEQ ID NO: 40. This sequence is a non-coding sequence into which a DNA sequence can be inserted, for example, containing a transgene (or transgene cassette), a restriction site, or a combination thereof. This sequence is useful because it acts as a buffer that provides some flexibility regarding the exact location of the transgene, while minimizing the disruptive effect on the stability and viability of the virus.

- 24 043895

The insertion(s) may occur at any location of SEQ ID NO: 40 from the 5' end, the 3' end, or at any point from 1 to 35 bp, such as between base pairs 1/2, 2/3, 3/4, 4/5, 5/6, 6/7, 7/8, 8/9, 9/10, 10/11,

11/12, 12/13, 13/14, 14/15, 15/16, 16/17, 17/18, 18/19, 19/20, 20/21, 21/22, 22/23, 23/ 24, 24/25, 25/26, 26/27,

27/28, 28/29, 29/30, 30/31, 31/32, 32/33, 33/34 or 34/35.

In one embodiment, BY comprises SEQ ID NO: 40 with a DNA sequence inserted between positions 12 and 13 bp or the location corresponding to bp 29356 and bp 29357 in SEQ ID NO: 38. In one embodiment, the insert is a restriction site insert. In one embodiment, the restriction site insert comprises one or two restriction sites. In one embodiment, the restriction site is a 19 bp restriction site insert containing 2 restriction sites. In one embodiment, the restriction site insert is a 9 bp restriction site insert containing 1 restriction site. In one embodiment, the restriction site insert comprises one or two restriction sites and at least one transgene, such as one or three transgenes, such as one or two transgenes. In one embodiment, the restriction site is a 19 bp restriction site insert comprising 2 restriction sites and at least one transgene, such as one or two transgenes. In one embodiment, the restriction site insert is a 9 bp restriction site insert comprising 1 restriction site and at least one transgene, such as one or two transgenes, such as one or two transgenes. In one embodiment, the two restriction sites sandwich one or more, such as two, transgenes (e.g., in a transgene cassette). In one embodiment, when B _Y comprises two restriction sites, the restriction sites are different from each other. In one embodiment, the one or more restriction sites in BY are not naturally occurring (e.g., unique) in the particular adenovirus genome into which they have been inserted. In one embodiment, the one or more restriction sites in BX are different from other restriction sites located elsewhere in the adenovirus genome, such as different from naturally occurring restriction sites and/or restriction sites introduced into other parts of the genome, such as BX. Thus, in one embodiment, the restriction site or sites allow DNA to be cut at a specific location.

In one embodiment, the restriction site insert is SEQ ID NO: 51.

In one embodiment, BY has a sequence from 29345 bp to 29379 bp of SEQ ID NO: 38.

In one embodiment, B _Y represents a bond.

In one embodiment, the insert is located after 12 bp of SEQ ID NO: 40.

In one embodiment, the insert is located at approximately position 29,356 bp of SEQ ID NO: 38.

In one embodiment, the insert is a transgene cassette comprising one or more transgenes, such as 1, 2 or 3, such as 1 or 2 transgenes.

As used herein, the designation E4 refers to a portion of the DNA coding sequence or the entire E4 region of an adenovirus (i.e., a polypeptide/protein region) that may be mutated such that the protein encoded by the E4 gene has conservative or non-conservative amino acid substitutions such that it has the same function as the wild type (i.e., the corresponding non-mutated protein);

an increased function compared to the wild-type protein; a decreased function, such as no function compared to the wild-type protein; or has a new function compared to the wild-type protein, or a combination thereof, as appropriate. In one embodiment, the E4 region is deleted E4orf4.

In one embodiment, the E4 region is partially deleted, such as deleted by 95, 90, 85, 80, 75, 70, 65, 60, 55, 50, 45, 40, 35, 30, 25, 20, 15, 10, or 5%. In one embodiment, the E4 region has a sequence from 32188 bp to 29380 bp of SEQ ID NO: 38.

In one embodiment, _B3 is a bond, meaning that E4 is absent.

In one embodiment, _B3 has a sequence from 32188 to 29380 bp of SEQ ID NO: 38.

The ranges of numbers and numbers used in the present invention include endpoints.

The skilled person will understand that the elements in the claims of the present invention, such as the formulas (I), (Ia), (Ib), (Ic), (Id) and (Ie), are contiguous and may contain non-coding DNA sequences as well as genes and coding DNA sequences (structural features) mentioned in the present invention. In one or more embodiments, the claims of the present invention attempt to describe a naturally occurring sequence in the genome of an adenovirus. In this context, the skilled person will understand that the claim refers to the basic elements characterizing the corresponding region of the genome and is not intended to be an exhaustive description of the genomic region of DNA.

As used herein, the designations E1A, E1B, E3 and E4 each independently refer to the wild type and its equivalents, mutated or partially deleted forms of each region as described herein, in particular the wild type sequence from a known adenovirus.

As used in the present invention, the term insert refers to a DNA sequence that is incorporated either at the 5' end, at the 3' end, or in a given reference segment of a DNA sequence, such that it interrupts the reference sequence. The latter is a reference sequence used as a reference point relative to which the insert is located. In the context of the present invention, inserts typically occur in either SEQ ID NO: 10 or SEQ ID NO: 11. The insert may be either a restriction site insertion, a transgene cassette insertion, or both. When the sequence is interrupted, the virus will still contain the original sequence, but will typically consist of two fragments sandwiching the insert.

In one embodiment, the transgene or transgene cassette does not comprise an unbiased insertional transposon, such as a TN7 transposon, or a portion thereof. As used herein, the term Tn7 transposon refers to an unbiased insertional transposon as described in WO2008/080003.

In one embodiment, one or more restriction sites in BX and BY are independently selected from a restriction site specific for an enzyme described in the present invention, such as NotI, FseI, AsiSI, SgfI and SbfI, in particular the inserted restriction sites are all different, such as NotI-specific sites and FseI-specific sites located in BX, and SgfI and SbfI located in BY.

As discussed above, in one embodiment, the BX and/or BY region does not contain a restriction site. An advantage is that the viruses and constructs of the present invention can be produced without restriction sites, for example, using synthetic methods. These techniques allow greater flexibility in creating viruses and constructs. In addition, the inventors have found that the properties of the viruses and constructs are not reduced when they are produced by synthetic methods.

Promoters

As used herein, the term "promoter" refers to a region of DNA that initiates transcription of a particular gene or genes. Promoters are typically located proximal to the genes they transcribe, on the same strand and upstream (i.e., 5') of the DNA. Proximal in this context means close enough to function as a promoter. In one embodiment, the promoter is within 100 bp of the transcription start site. Thus, as used herein, the term "endogenous promoter" refers to a promoter that occurs naturally (i.e., is native) to the adenovirus (or construct) into which the transgene is inserted. In one or more embodiments, the endogenous promoter used is a naturally occurring promoter in the virus at its original position in the viral genome, in particular, it is the major or only promoter used to express the transgene or transgenes. In one embodiment, the endogenous promoter used to stimulate translation and, optionally, transcription of the transgene is resident, i.e., integrated into the adenovirus genome, and not previously introduced by recombinant methods.

As used in the present invention, the term under the control of an endogenous promoter means that the transgene/transgene cassette is inserted in an appropriate orientation to be under the control of said endogenous promoter. That is, when the promoter is normally on the antisense strand, the cassette is inserted, for example, in the antisense orientation.

With that said, genes can be expressed in one of two orientations. However, typically one orientation will provide higher levels of expression than the other orientation for a given transgene.

In one embodiment, the cassette is in the sense orientation. That is, it is transcribed in the 5' to 3' direction. In one embodiment, the cassette is in the antisense orientation. That is, it is transcribed in the 3' to 5' orientation.

Endogenous promoters in a virus can be used, for example, by using a gene encoding a transgene and a splice acceptor sequence. Thus, in one embodiment, the cassette will contain a splice acceptor sequence when under the control of an endogenous promoter. Thus, in one embodiment, the coding sequence, for example, a sequence encoding an antibody or antibody-binding fragment, further comprises a splice acceptor sequence.

In one embodiment, the transgene, transgenes, or transgene cassette are under the control of the E4 promoter or a major late promoter, such as the major late promoter (MLpromoter).

By controlled in the present invention is meant that the transgene is activated, i.e. transcribed, when a particular promoter dictates so.

The major late promoter (ML promoter or MLP) in the context of the present application refers to an adenoviral promoter that controls the expression of late expressed genes, such as the L5 gene. The MLP is a sense strand promoter. That is, the promoter affects genes that are downstream of the promoter in the 5'-3' direction. The term major late promoter in the present

- 26 043895 invention relates to the original major late promoter located in the genome of the virus.

As used herein, the term E4 promoter refers to the adenoviral promoter E4 region. The E4 region is an antisense region; therefore, the promoter is an antisense promoter. That is, the promoter is located to the left of the E4 region in the 3'-5' direction. Therefore, any transgene cassette under the control of the E4 promoter must be oriented accordingly. In one embodiment, the cassette under the control of the E4 promoter is in the antisense orientation. In one embodiment, the cassette is under the control of the E4 promoter in the sense orientation. As used herein, the term E4 promoter refers to the original E4 promoter located in the viral genome.

Thus, in one embodiment, there is provided a replication competent oncolytic adenovirus serotype 11 (such as Ad11p) or a viral derivative thereof, wherein the fiber, hexon and capsid are serotype 11 (such as Ad11p), wherein the viral genome comprises a DNA sequence encoding a therapeutic antibody or antibody-binding fragment (such as a bispecific T cell activator), wherein said DNA sequence is under the control of a promoter endogenous to the adenovirus selected from the E4 promoter and the major late promoter (i.e. the E4 promoter or the major late promoter), such that the transgene does not interfere with viral replication, such as being associated with the L5 region (i.e. before or after said region), such as being located after L5 in the viral genome, in particular located between L5 and the E4 region.

In one embodiment, the endogenous promoter is introduced into the viral genome at the desired location by recombinant methods, for example, introduced into a transgene cassette. However, in the context of the present application, this location will generally be referred to as an exogenous promoter.

In one embodiment, the transgene cassette comprises an exogenous promoter. As used herein, the term exogenous promoter refers to a promoter that does not naturally occur in the adenovirus into which the transgene is inserted. Typically, exogenous promoters are derived from other viruses or are mammalian promoters. As used herein, the term exogenous promoter refers to a DNA element, typically located upstream of a target gene, that regulates transcription of the gene.

In one embodiment, the gene expression regulator is an exogenous promoter, such as a CMV (cytomegalovirus) promoter, a CBA (chicken beta-actin) promoter, or a PGK (phosphoglycerate kinase 1) promoter, such as the CMV promoter.

In one embodiment, the exogenous CMV promoter used has the nucleotide sequence of SEQ ID NO: 52. In one embodiment, the exogenous PGK promoter used has the nucleotide sequence of SEQ ID NO: 53. In one embodiment, the exogenous CBA promoter used has the nucleotide sequence of SEQ ID NO: 54.

In one embodiment, there is provided a replication competent oncolytic adenovirus serotype 11 (such as Ad11p) or a viral derivative thereof, wherein the fiber, hexon and capsid are serotype 11 (such as Ad11p), wherein the viral genome comprises a DNA sequence encoding a therapeutic antibody or antibody-binding fragment (such as the bispecific T cell activator of the present invention) located in a portion of the viral genome that is expressed at the end of the viral replication cycle and such that the transgene does not interfere with viral replication, wherein said DNA sequence is under the control of a promoter exogenous to the adenovirus (e.g., the CMV promoter). In one embodiment, the DNA sequence encoding the antibody or fragment (such as the bispecific T cell activator of the present invention) is associated with the L5 region as described elsewhere in the present invention, in particular located between the L5 and E4 region.

In one embodiment, the exogenous promoter is an antigen-presenting cell promoter. As used herein, the term antigen-presenting cell promoter refers to a gene promoter that is selectively expressed by antigen-presenting cells, such as dendritic cells or macrophages. Such genes include, but are not limited to: FLT-3, FLT-3 ligand, TLR, CD1a, CD1c, CD11b, CD11c, CD80, CD83, CD86, CD123, CD172a, CD205, CD207, CD209, CD273, CD281, CD283, CD286, CD289, CD287, CXCR4, GITR ligand, IFN-a2, IL12, IL-23, ILT1, ILT2, ILT3, ILT4, ILT5, ILT7, TSLP receptor, CD141, CD303, CADM1, CLEC9a, XCR1 or CD304; mediators of antigen processing and presentation such as CTIIA or GILT. Thus, in one embodiment, the exogenous promoter is suitable for selectively expressing transgenes in said antigen-presenting cells.

Other regulatory sequences.

As used herein, the term gene expression regulator (or regulator/regulatory element) refers to a genetic feature, such as a promoter, enhancer, or splice acceptor sequence, that plays a role in the expression of a gene, typically by initiating or enhancing transcription or translation.

As used in the present invention, the terms splice acceptor sequence,

- 27 043895 splice acceptor or splice site refers to a regulatory sequence that determines when an mRNA molecule will be recognized by the small nuclear ribonucleoproteins of the spliceosome complex. Once assembled, the spliceosome catalyzes splicing between the splice acceptor site of the mRNA molecule and the left splice donor site, producing a mature mRNA molecule that can be translated to produce a single polypeptide or protein.

In the present invention, splice acceptor sequences of different sizes can be used, and they can be described as a short splice acceptor (small), a splice acceptor (medium), and a branched splice acceptor (large).

As used herein, the term SSA means a short splice acceptor, typically comprising only a splice site, for example 4 base pairs. As used herein, the term SA means a splice acceptor, typically comprising a short splice acceptor and a polypyrimidine tract, for example 16 bp. As used herein, the term bSA means a branched splice acceptor, typically comprising a short splice acceptor, a polypyrimidine tract and a branch point, for example 26 base pairs.

In one embodiment, the splice acceptor used in the constructs of the invention is set forth in SEQ ID NO: 55-57. In one embodiment, SSA has the nucleotide sequence of SEQ ID NO: 55. In one embodiment, SA has the nucleotide sequence of SEQ ID NO: 56. In one embodiment, bSA has the nucleotide sequence of SEQ ID NO: 57. In one embodiment, the splice acceptor sequence is independently selected from the group consisting of TGCTAATCTT CCTTTTCTCTC TTCAGG (SEQ ID NO: 57), CCTTTCTCTCT CAGG (SEQ ID NO: 56), and CAGG (SEQ ID NO: 55).

In one embodiment, the splice site is directly processed (i.e., followed in the 5' to 3' direction) by a Kozak consensus sequence comprising CCACC. In one embodiment, the splice site and the Kozak sequence are interspersed for up to 100 base pairs or less. In one embodiment, the Kozak sequence has the nucleotide sequence of SEQ ID NO: 58.

Typically, when under the control of an endogenous or exogenous promoter (such as an endogenous promoter), the coding sequence will be immediately preceded by a Kozak sequence. The start of the coding region is designated by an initiation codon (AUG), for example, in the context of the sequence (gcc)gccRccAUGg [SEQ ID NO: 59], the start of the coding sequences is designated by the bases in bold. Lower case letters indicate common bases at that position (which may, however, vary), and upper case letters indicate highly conserved bases, i.e., the sequence AUGG is constant or rarely, if ever, varies; R indicates that a purine (adenine or guanine) is commonly found at that position, and the sequence in brackets (gcc) is of uncertain significance. Thus, in one embodiment, the initiation codon AUG is included in the Kozak sequence.

As used in the present invention, the term internal ribosome entry DNA sequence refers to a DNA sequence encoding an internal ribosome entry site (IRES). An IRES is, as used in the present invention, a nucleotide sequence that allows for the initiation of translation of a messenger RNA (mRNA) sequence, including initiation starting within the mRNA sequence. This is particularly useful when the cassette encodes a polycistronic mRNA. The use of an IRES results in a polycistronic mRNA that is translated into several distinct proteins or peptides. In one embodiment, the internal ribosome entry DNA sequence has the nucleotide sequence of SEQ ID NO: 60. In one embodiment, a particular IRES is used only once in a genome. This may have advantages in terms of genome stability.

As used herein, the term 2A peptide with high self-cleavage efficiency or 2A peptide refers to a peptide that is efficiently cleaved after translation. Suitable 2A peptides include P2A, F2A, E2A and T2A. The present inventors have noted that if a particular DNA sequence encoding a given 2A peptide is used once, the same particular DNA sequence cannot be used a second time. However, redundancy in the DNA code can be used to generate a DNA sequence that is translated into the same 2A peptide. The use of 2A peptides is particularly useful when the cassette encodes a polycistronic mRNA. The use of 2A peptides results in the translation of a single polypeptide chain, which after translation is modified to generate multiple individual proteins or peptides.

In one embodiment, the encoded peptide P2A used has the amino acid sequence of SEQ ID NO: 61. In one embodiment, the encoded peptide F2A used has the amino acid sequence of SEQ ID NO: 62. In one embodiment, the encoded peptide E2A used has the amino acid sequence of SEQ ID NO: 63. In one embodiment, the encoded peptide T2A used has the amino acid sequence of SEQ ID NO: 64.

In one embodiment, the or each mRNA encoded by the transgene(s) comprises a polyadenylation signal sequence, such as typically at the end of the mRNA sequence, such as shown in SEQ ID NO: 65. Thus, in one embodiment, the transgene or transgene cassette comprises at least one sequence encoding a polyadenylation signal sequence.

PolyA, polyadenylation signal or polyadenylation sequence in the present invention means a DNA sequence, typically containing an AATAAA site, which, after transcribing, can be recognized by a multiprotein complex that cleaves and polyadenylates the nascent mRNA molecule.

In one embodiment, the polyadenylation sequence has the nucleotide sequence of SEQ ID NO: 65.

In one embodiment, the construct does not include a polyadenylation sequence. In one embodiment, the gene expression regulator is a splice acceptor sequence.

In one embodiment, the sequence encoding a protein/polypeptide/peptide, such as an antibody or antibody fragment (such as a bispecific T cell activator according to the present invention), further comprises a polyadenylation signal.

Molecules encoded by the transgene.

As described in the present invention, at least one transgene in the virus encodes a bispecific T cell activator, wherein one binding domain is specific for a T cell surface antigen. The second binding domain can target a suitable antigen, such as a pathogenic antigen, a cancer antigen, an antigen that is present in the stroma.

Cancer antigens (also called tumor antigens) are one category of particular interest and include, for example, those selected from CEA, MUC-1, EpCAM, HER receptors HeR1, HER2, HER3, HER4, PEM, A33, G250, carbohydrate antigens Le ^y , Lex, Le ^b , PSMA, TAG-72, STEAP1, CD166, CD24, CD44, E-cadherin, SPARC, ErbB2, ErbB3, WT1, MUC1, LMP2, idiotype, HPV E6&E7, EGFRvIII, HER-2/neu, MAGE A3, p53 non-mutant, p53 mutant, NY-ESO-1, GD2, PSMA, PCSA, PSA, MelanA/MART1, Ras mutant, proteinase 3 (PR1), bcr-abl, irosinase, survivin, PSA, hTERT, in particular WT1, MUC1, HER-2/neu, NY-ESO-1, survivin and hTERT.

Antigens that are present in the stroma include fibroblast antigens, such as those described herein, such as FAP, tumor-associated macrophage antigens, and myeloid-derived suppressor cell antigens, such as CD163, CD206, CD68, CD11c, CD11b, CD14, CSF1 receptor, CD15, CD33, and CD66b.

The following list of targets may, if desired, be encoded in the bispecific T cell activator of the present invention or, alternatively, may be provided as an additional therapeutic transgene, or both.

In one embodiment, the transgene or transgenes independently encode a protein, a peptide, an RNA molecule, such as an RNA molecule. The advantage is that the transgene can be delivered intracellularly and subsequently transcribed and, if necessary, translated. Examples of genetic material encoded by a transgene include, for example, antibodies or binding fragments thereof, chemokines, cytokines, immunomodulators, enzymes (e.g., capable of converting a prodrug into an active agent), and an RNAi molecule.

The term peptide as used in the present invention refers to an amino acid sequence of 2 to 50 residues, such as 5 to 20 residues. The term polypeptide as used in the present invention refers to an amino acid sequence of more than 50 residues without tertiary structure, in particular without secondary and tertiary structure. The term protein refers to an amino acid sequence of more than 50 residues with secondary and/or tertiary structure, in particular with secondary and tertiary structure.

In one embodiment, the coding sequence encodes a therapeutic RNA, a therapeutic peptide, a therapeutic polypeptide, or a therapeutic protein (i.e., is a therapeutic gene).

As used in the present invention, the term immunomodulatory gene or transgene means a gene that encodes a peptide or protein molecule that can qualitatively or quantitatively modify the activity of cells of the immune system.

As used in the present invention, the term therapeutic gene means a gene encoding an entity that may be useful for treating, alleviating or preventing a disease, for example, a gene expresses a therapeutic protein, polypeptide, peptide or RNA that at least slows down, stops or reverses the progression of a disease such as cancer.

In one embodiment, the transgene-encoded event, when transcribed or translated in a cell, such as a cancer cell, increases the cell's production of danger signals. As used herein, the term danger signals refers to a variety of molecules produced by cells undergoing injury, stress, or non-apoptotic death that act as alarm signals, such as stimulating cells of the innate immune system to respond directly, and also serve to enhance the activation of cells of the adaptive immune system.

It is known that the tumor microenvironment is often altered in such a way that the person’s natural immune responses are suppressed. Thus, the ability to re-initiate immune responses from within the tumor is potentially very interesting for cancer treatment.

In one embodiment, the encoded therapeutic peptide or protein is intended to be secreted into the extracellular environment. In one embodiment, the functional RNA, peptide, polypeptide or protein, such as an antibody, is released into the external microenvironment of the cell, such as a culture supernatant or in vivo: into tissue, stroma, circulation, blood and/or lymphatic system.

In one embodiment, the peptide, polypeptide or protein (including the bispecific T cell activator of the present invention) encoded by the transgene comprises a signal sequence. As used herein, the term signal peptide refers to a short peptide sequence of 13-36 residues located at the N-terminus of proteins that facilitates entry of the protein into the secretory pathway for secretion or membrane expression. In one embodiment, the leader sequence (signal peptide) has the amino acid sequence of SEQ ID NO: 66 or 67.

In another embodiment, the encoded therapeutic peptide or protein, such as an antibody, is intended to be expressed as a membrane-anchored form in the cell surface membrane, for example, by including encoding a transmembrane domain in the protein or a site for attachment of a lipid membrane anchor. Typically, the bispecific T cell activator of the present invention is not expressed in a cell surface anchor format.

In one embodiment, a functional RNA, peptide, polypeptide or protein, such as an antibody, is released from a cell infected with an adenovirus, such as by active secretion or by cell lysis. Thus, in one embodiment, the adenovirus lyses the cell, thereby releasing a functional RNA, peptide, polypeptide or protein, such as an antibody.

In another embodiment, the encoded additional therapeutic peptide or protein, such as an antibody, is intended to be retained within an intact cell.

The advantage is that the functional RNA, peptide, polypeptide or protein, such as antibodies, expressed by the adenoviruses of the present invention can be detected in tissue in vivo as both mRNA and antibody protein. In addition, the expressed functional RNA, peptide or protein, such as antibody, can bind its ligand in ELISA. In addition, the functional RNA, peptide, polypeptide or protein, such as antibody, is detectable at an early stage (e.g. within 3 days after infection), and this expression is maintained for several weeks.

In one embodiment, the adenoviruses of the present invention express functional RNA, peptide, polypeptide or protein, such as antibodies, for about 3 days or more after infection, such as for about 36, 48, 60 or 72 hours or, such as, 2, 3, 4, 5 or 6 days.

In one embodiment, the adenoviruses of the present invention express functional RNA, peptide, polypeptide or protein, such as antibodies, for several weeks, such as about 1, 2, 3, 4, 5 or 6 weeks. For example, for 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41 or 42 days.

The advantage is that the expression of a functional RNA, peptide or protein, such as antibody expression, is high enough that the functional RNA, peptide, polypeptide or protein, such as an antibody, can be detected in the blood.

In one embodiment, functional RNA, peptide or protein, such as antibodies, expressed by the adenovirus of the present invention are released into the bloodstream and/or lymphatic system.

In one embodiment, the adenovirus of the present invention is an oncolytic virus that has an increased therapeutic index for cancer cells.

In one embodiment, the coding sequence further encodes a functional RNA, such as a therapeutic RNA.

As used in the present invention, the term functional RNA refers to an RNA that performs a function other than encoding a protein or peptide, and includes, for example, RNA constructs suitable for inhibiting or reducing gene activity, including RNAi such as shRNA and microRNA. As used in the present invention, the term shRNA refers to a short hairpin RNA, which is an RNA sequence that makes a tight hairpin turn that can be used to silence the expression of a target gene through RNAi (RNAi). The term microRNA (miRNA) as used in the present invention,

- 30 043895 refers to a small non-coding RNA molecule (containing approximately 22 nucleotides) that functions through base pairing with complementary sequences in mRNA molecules to regulate gene expression at the transcriptional or post-transcriptional level.

mRNA strands bound by microRNAs are silenced because they can no longer be translated into proteins by ribosomes, and such complexes are often actively disassembled by the cell.

In one embodiment, the transgene encodes a protein. As used herein, the term protein includes a protein ligand, a protein receptor, or an antibody molecule.

As used herein, the term protein ligand refers to a cell surface membrane or secreted proteins thereof that bind to or otherwise interact with cellular receptors to affect cell function, such as by stimulating intracellular signaling and modulating gene transcription in a cell. In one embodiment, the expressed protein is engineered to be expressed on the cell surface and/or secreted from the cell.

In one embodiment, the encoded protein is a bispecific antibody, such as a bispecific T cell activator.

In one embodiment, the transgene further encodes an enzyme, such as an enzyme that helps degrade the extracellular matrix of a tumor, such as DNase, collagenase, matrix metalloproteinase (such as MMP2 or 14), or the like.

Suitable antibodies and antibody fragments may be agonistic or antagonistic and include antibodies with anti-cancer activity and antibodies that modify host cell responses to cancer, such as an agonistic or antagonistic antibody or antibody fragment that can reduce vascularization or normalize vascularization of a tumor. In one embodiment, agonistic antibodies or other encoded proteins can make a host cell more visible to the host's innate and adaptive immune responses, such as by expressing antigens, danger signals, cytokines or chemokines to attract and activate them, or by binding to costimulatory or checkpoint pathway molecules to enhance adaptive immune responses.

As used herein, the terms "therapeutic antibody" or "antibody-binding fragment" refer to an antibody or antibody-binding fragment that, when introduced into an oncolytic virus, exerts a beneficial effect on a pathology in a patient, such as a cancer being treated.

As used herein, the term beneficial effect refers to a desirable and/or beneficial effect of an antibody expressed in vivo.

Classes of therapeutic antibodies and antibody-binding fragments include anti-EGF antibodies, anti-VEGF antibodies, anti-PDGF antibodies, anti-CTLA antibodies, anti-PD1 antibodies, anti-PDL1 antibodies, and anti-FGF antibodies.

Registered therapeutic antibodies suitable for incorporation into the viruses of the present invention include abciximab, adalimumab, alemtuzumab, basiliximab, belimumab, bevacizumab, brentuximab vedotin, canakinumab, cetuximab, certolizumab, daclizumab, denosumab, eculizumab, efalizumab, gemtuzumab, golimumab, ibritumomab tiuxetan, infliximab, ipilimumab, muromonab-CD3, ofatumumab, palivizumab, panitumumab, ranibizumab, rituximab, tocilizumab, tositumomab and trastuzumab.

In one embodiment, the antibody variable region sequences for the antibody or antibody fragment used are 95-100% similar or identical to the variable region of bevacizumab (also known as Avastin®), such as 96, 97, 98, or 99% similar or identical.

Also suitable for inclusion in the viruses of the present invention are coding sequences for those antibodies and binding fragments thereof that are approved for cancer indications, such as trastuzumab, tositumomab, rituximab, panitumumab, ofatumumab, ipilimumab, ibritumomab tiuxetan, gemtuzumab, denosumab, cetuximab, brentuximab vedotin, avastin and adalimumab.

In one embodiment, the antibody variable region sequences for the antibody or antibody fragment used are 95-100% similar or identical to the variable regions of a known antibody or an antibody disclosed herein.

As used herein, the term antibody molecule includes antibodies and binding fragments thereof.

As used herein, the term antibody generally refers to a full-length antibody and bispecific or multispecific formats containing it.

Antibody-binding fragments include a fragment of an antibody capable of targeting an antigen with the same, similar, or better specificity than the parent antibody from which they were derived. Antibody fragments include Fab, modified Fab, Fab', modified Fab', F(ab') ₂ , Fv, single-domain antibodies (e.g., VH or VL or VHH), scFv, bi-, tri-, or tetravalent antibodies, bis-scFv, diabodies, triabodies, tetrabodies, and epitope-binding fragments of any of

- 31 043 895 of the above (see, e.g., Holliger and Hudson, 2005, Nature Biotech. 23 (9): 1126-1136; Adair and Lawson, 2005, Drug Design Reviews - Online 2 (3), 209-217). Methods for creating and making these antibody fragments are well known in the art (see, e.g., Verma et al., 1998, Journal of Immunological Methods, 216, 165-181). Other antibody fragments for use in the present invention include the Fab and Fab' fragments described in International Patent Applications WO2005/003169, WO2005/003170 and WO2005/003171. Multivalent antibodies may have multiple specificities, such as being bispecific or monospecific (see, for example, WO 92/22853, WO05/113605, WO2009/040562 and WO2010/035012).

The term specific as used in the present invention is intended to mean an antibody or fragment that recognizes only the antigen to which it is specific, or an antibody or fragment that has a significantly higher binding affinity for the antigen to which it is specific compared to its binding affinity for antigens to which it is not specific, for example, a binding affinity that is 5, 6, 7, 8, 9, 10 times higher.

Known antibodies or antibody-binding fragments can be used to create alternative antibody formats with the same CDRs or the same variable regions, for example, a full-length antibody can be easily converted into a Fab, Fab' or scFv fragment.

A wide range of different forms of antibodies can be used in the constructs of the present invention, including antibody molecules from non-human animals, human antibody molecules, humanized antibody molecules, and chimeric antibody molecules.

In one embodiment, the antibody or binding fragment is monoclonal. Monoclonal antibodies can be produced by any method known in the art, such as hybridoma technology (Kohler & Milstein, 1975, Nature, 256:495-497), trioma technology, human B-cell hybridoma technology (Kozbor et al., 1983, Immunology Today, 4:72), and EBV hybridoma technology (Cole et al., Monoclonal Antibodies and Cancer Therapy, pp77-96, Alan R Liss, Inc., 1985).

In one embodiment, the antibody or binding fragment is non-human, i.e., entirely non-human in origin. This is possible because antibodies and fragments can be delivered into a cancer cell by a virus.

In one embodiment, the antibody is chimeric, e.g., has human constant region(s) and non-human variable regions.

In one embodiment, the antibody or binding fragment is human, i.e., of entirely human origin.

In one embodiment, the antibody or binding fragment is humanized. Humanized antibodies (which include CDR-grafted antibodies) are antibody molecules having one or more complementarity determining regions (CDRs) from a non-human animal species and a framework region from a human immunoglobulin molecule (see, e.g., US 5,585,909; WO91/09967). It is understood that only the specificity-determining residues of the CDRs may need to be transferred, rather than the entire CDR (see, e.g., Kashmiri et al., 2005, Methods, 36, 25-34). Humanized antibodies may optionally contain one or more framework residues derived from a non-human animal species, such as that from which the CDRs were derived.

In one embodiment, the coding sequence encodes a heavy chain of an antibody, a light chain of an antibody, or an antibody fragment. The term heavy chain (HC) in the present invention refers to a large polypeptide subunit of an antibody. The term light chain (LC) in the present invention refers to a small polypeptide subunit of an antibody. In one embodiment, the light chain of an antibody comprises a CL domain, either kappa or lambda.

Antibodies for use in the present invention may be produced using any suitable method known in the art. To produce antibodies that specifically recognize an antigen, an antigen polypeptide/protein, including fusion proteins, may be used, including cells (recombinant or natural) expressing said polypeptide (e.g., activated T cells). Said polypeptide may be a mature polypeptide or a biologically active fragment or derivative thereof.

Antibody screening can be performed using assays to measure binding to an antigen and/or assays to measure the ability to antagonize a receptor. An example of a quantitative binding assay is ELISA, in particular using a fusion protein (optionally containing a reporter) that is immobilized on plates and using a conjugated secondary antibody to detect anti-antigen antibody bound to the fusion protein.

The constant region domains of the antibody molecule of the present invention, if present, may be selected taking into account the intended function of the antibody molecule and, in particular, the effector functions that may be required. For example, the constant region domains may be human IgA, IgD, IgE, IgG or IgM domains. In particular, human IgG constant region domains, especially the IgG1 and IgG3 isotypes, may be used when the antibody molecule is intended for therapeutic purposes and the effector functions of the antibody are required. Alternatively,

- 32 043895 IgG2 and IgG4 isotypes may be used when the antibody molecule is intended for therapeutic purposes and the effector functions of the antibody are not required, for example, for simple agonizing activity or for neutralizing the target. It is understood that sequence variants of these constant region domains may also be used. For example, IgG4 molecules may be used in which serine at position 241 is replaced by proline, as described in Angal et al., Molecular Immunology, 1993, 30(1), 105-108.

For certain antibody functions, such as delivering activation signals to cells that carry the antibody's target molecule, such as cells of the immune system, it may be advantageous to use membrane-anchored versions of the antibody so that the antibody is expressed on the surface of the expressing cell. Such cell-surface-expressed molecules provide efficient multimeric interactions between the target signaling molecule on the surface of another cell, which improves the delivery of activation signals from the target molecule to the recipient cell.

An advantage of the adenoviruses of the present invention is the ability to express full-length antibodies, antibody fragments such as scFv, multispecific antibodies, in particular bispecific antibodies such as the bispecific T cell activator as described herein.

In one embodiment, the sequence encoding the antibody or antibody fragment (such as the bispecific T cell activator of the present invention) comprises or further comprises an internal ribosome entry site. An internal ribosome entry site (IRES) in the context of the present application means a nucleotide sequence that provides for translation initiation in the middle of a messenger RNA (mRNA) sequence.

In one embodiment, the encoded therapeutic proteins or peptides are target specific proteins, polypeptides or peptides.

Target specific proteins or peptides in the context of the present application refer either to the target proteins themselves or to various proteins or peptides that directly bind (e.g., are specific to the target) or otherwise modify the levels of the target proteins or peptides. An example of the former would be a cytokine, while an example of the latter would be an antibody against that cytokine.

The targets in question typically refer to specific cells, cellular products, antigens, or signaling pathways associated with a disease, especially cancer. Depending on the context, a target also refers to an mRNA or its analogue transcribed from a gene encoding a protein or polypeptide that, for example, can be inhibited using RNAi-type technology. Thus, in the context of RNA, such as RNAi technology, the target is the mRNA that is encoded by the target gene.

Examples of targets under consideration include, among others, stimulatory T cell coreceptors and ligands therefor, inhibitory T cell coreceptor molecules and ligands therefor, receptors and ligands therefor expressed by regulatory T cells, myeloid-derived suppressor cells and immunosuppressive immune cells, dendritic cell and antigen-presenting cell receptors and ligands therefor, mediators of antigen processing and presentation, cytokines and cytokine receptors, chemokines and chemokine receptors, transcription factors and transcription regulators, intracellular transport molecules and regulators of cell function, receptors and products of the tumor cell and tumor cell microenvironment, intracellular tumor cell enzymes such as IDO, antigens for recognition by immune cells.

Thus, in one embodiment, the term target as used herein refers to a protein or polypeptide that can, for example, be inhibited, neutralized or activated, for example, by an antibody or a binding fragment thereof, as appropriate. A target in the context of cytokines refers to the cytokine itself or an antibody or a binding fragment thereof specific for the cytokine. Thus, a virus can encode and express the cytokine itself, since its release can stimulate immune responses of the host. In the context of ligands, mutated forms of the ligand can be encoded by the virus that compete with the natural ligand for binding to the receptor. The mutated ligand can have an increased binding affinity for the receptor, for example, such that it has a low dissociation rate, thereby occupying the receptor and increasing or decreasing signaling from it. Alternatively, the activity of the mutated ligand may be reduced compared to the wild-type ligand, thereby reducing binding and overall activity through the receptor from the natural ligand.

In one embodiment, the virus or construct of the present invention encodes a prodrug, an immunomodulator and/or an enzyme.

As used herein, the term prodrug means a molecule that is administered as an inactive (or less than fully active) derivative that is subsequently converted into an active pharmacological agent in the body, often through normal metabolic processes. A prodrug serves as a precursor to the intended drug. A prodrug-converting enzyme serves as an enzyme that converts a prodrug into its

- 33 043895 pharmacologically active form.

As used in the present invention, the term immunomodulator means a modulator of the immune response. Immunomodulators function in regulating the immune response to a desired level, such as immunopotentiation, immunosuppression, or induction of immunological tolerance.

T cells require two signals to become fully activated. The first signal, which is antigen-specific, is provided via the T cell receptor, which interacts with peptide-MHC molecules on the membrane of antigen-presenting cells (APCs). The second signal, the costimulatory signal, is non-antigen-specific and is provided by the interaction between costimulatory molecules expressed on the membrane of the APC and the T cell. Thus, a costimulatory molecule in the context of the present application means a molecule that provides an additional signal to the antigen-specific signal that is required by T cells for their activation, proliferation and survival. Examples of costimulatory molecules include, among others, CD28, CD80, CD86, CD83 and 4-1BB.

As used herein, the term enzyme means a substance that acts as a catalyst in living organisms, regulating the rate at which chemical reactions occur without being changed in the process.

A non-exhaustive discussion of illustrative target peptides/polypeptides and proteins is provided below.

In one embodiment, the target is a checkpoint protein, such as an immune checkpoint protein or a cell cycle checkpoint protein. Examples of checkpoint proteins include, but are not limited to, CTLA-4, PD-1, PD-L1, PD-L2, VISTA, B7-H3, B7-H4, HVEM, ILT-2, ILT-3, ILT4, TIM-3, LAG-3, BTLA, LIGHT, or CD160, such as CTLA-4, PD-1, PD-L1, and PD-L2. In one embodiment, an antibody or binding fragment thereof is provided that is specific for one of these. Thus, in one embodiment, the transgene or transgene cassette encodes an antibody or antibody fragment specific for CTLA-4, PD-1, PD-L1, or PD-L2. In one embodiment, the adenovirus expresses an antibody or antibody fragment specific for CTLA-4, PD-1, PD-L1, or PD-L2.

In one embodiment, the antibody is a checkpoint inhibitor antibody, such as αntu-PD-L1. In one embodiment, the adenovirus expresses a full-length human anti-PD-L1 antibody. In one embodiment, expression of the full-length human anti-PD-L1 antibody is under the control of an endogenous promoter, such as the major late promoter (MLP), in particular at position BY. In one embodiment, the adenovirus expresses an scFv form of the human anti-PD-L1 antibody. In one embodiment, expression of the scFv form of the human anti-PD-L1 antibody is under the control of an endogenous promoter, such as the major late promoter, in particular at position BY.

In one embodiment, a virus or construct of the present invention is provided encoding an antibody or binding fragment thereof, for a full-length antibody or scFv specific for CTLA-4, for example, as exemplified herein.

In one embodiment, the target is one or more independently selected from the group consisting of CD16, CD25, CD33, CD332, CD127, CD31, CD43, CD44, CD162, CD301a, CD301b and galectin-3. In one embodiment, an antibody or binding fragment thereof specific therefor is provided, such as a full-length antibody or scFv.

In one embodiment, the target, for example, to which the antibody or binding fragment can be directed, is one or more independently selected from the group consisting of FLT-3 ligand, FLT-3 ligand, TLR, TLR ligands, CCR7, CD1a, CD1c, CD11b, CD11c, CD80, CD83, CD86, CD123, CD172a, CD205, CD207, CD209, CD273, CD281, CD283, CD286, CD289, CD287, CXCR4, GITR ligand, IFN-a2, IL-12, IL-23, ILT1, ILT2, ILT3, ILT4, ILT5, ILT7, TSLP receptor, CD141, CD303, CADM1, CLEC9a, XCR1 and CD304.

In one embodiment, the target of the bispecific T cell activator used in the present invention is a tumor cell antigen.

In one embodiment, the target is one or more independently selected from the group consisting of CEA, MUC-1, EpCAM, HER receptors HER1, HER2, HER3, HER4, PEM, A33, G250, carbohydrate antigens Le ^y , Lex, Le ^b , PSMA, TAG-72, STEAP1, CD166, CD24, CD44, E-cadherin, SPARC, ErbB2, ErbB3.

In one embodiment, the target of the bispecific T cell activator used in the present invention is an antigen that is present in the tumor stroma.

In one embodiment, the target of the bispecific T cell activator used in the present invention is one or more independently selected from the group consisting of FAP, TREM1, IGFBP7, FSP-1, platelet-derived growth factor receptor-a (PDGFR-a), platelet-derived growth factor receptor-β (PDGFR-β), and vimentin.

- 34 043895

In one embodiment, the target, for example, to which an antibody or binding fragment (e.g., a bispecific T cell activator) can be directed, is a cancer target.

In one embodiment, the target is one or more independently selected from the group consisting of OX40, OX40 ligand, CD27, CD28, CD30, CD40, CD40 ligand, TL1A, CD70, CD137, GITR,

4-1BB, ICOS or ICOS ligand, for example CD40 and CD40 ligand.

In one embodiment, the transgene cassette encodes a ligand comprising CD40 or a CD40 ligand, or an antibody, antibody fragment, or shRNA targeting CD40 or a CD40 ligand. In one embodiment, the adenovirus expresses a ligand comprising CD40 or a CD40 ligand, or an antibody, antibody fragment, or shRNA targeting (specific for) CD40 or a CD40 ligand.

In one embodiment, the target is one or more cytokines independently selected from the group consisting of IL-1a, IL-1e, IL-6, IL-9, IL-12, IL-13, IL-17, IL-18, IL-22, IL-23, IL-24, IL-25, IL-26, IL-27, IL-33, IL-35, Interleukin -2 (IL-2), IL-4, IL-5, IL-7, IL-10, IL-15, IL-21, IL-25, IL-1RA, IFNa, IFNe, IFN©, TNFa, TGFe, lymphotoxin a (LTA), and GM-CSF.

In one embodiment, the transgene cassette encodes an antibody or antibody fragment specific for IL-12, IL-18, IL-22, IL-7, IL-15, IL-21, IFNa, IFN©, TNFa, TGFe or lymphotoxin a (LTA). In one embodiment, the adenovirus expresses IL-12, IL-18, IL-22, IL-7, IL-15, IL-21, IFNa, IFN©, TNFa, TGFe or lymphotoxin a (LTA).

In one embodiment, the amino acid sequence of IFNy is SEQ ID NO: 68. In one embodiment, the amino acid sequence of IFNa is SEQ ID NO: 69. In one embodiment, the amino acid sequence of TNFa is SEQ ID NO: 70.

In one embodiment, the target is a chemokine, such as one or more independently selected from the group consisting of IL-8, CCL3, CCL5, CCL17, CCL20, CCL22, CXCL9, CXCL10, CXCL11, CXCL13, CXCL12, CCL2, CCL19, CCL21, CXCR2, CCR2, CCR4, CCR5, CCR6, CCR7, CCR8, CXCR3, CXCR4, CXCR5, and CRTH2.

In one embodiment, the transgene cassette encodes an antibody or antibody fragment specific for CCL5, CXCL9, CXCL12, CCL2, CCL19, CCL21, CXCR2, CCR2, CCR4, or CXCR4. In the context of chemokines, the target includes situations where viruses encode and express a chemokine, such as to induce or enhance host immune responses to cancer.

In one embodiment, the adenovirus expresses an antibody or antibody fragment specific for CCL5, CXCL9, CXCL12, CCL2, CCL19, CCL21, CXCR2, CCR2, CCR4, or CXCR4.

In one embodiment, the target is one or more independently selected from the group consisting of STAT3, STAT1, STAT4, STAT6, CTIIA, MyD88 and members of the NFkB family, for example, the protein is the target of an inhibitor, such as an antibody or a binding fragment thereof, or mRNA transcribed from the corresponding gene is inhibited by a mechanism such as RNAi.

In one embodiment, the target is HSp70 or a cell survival and death regulator such as survivin, for example, the protein is targeted by an inhibitor, such as an antibody or binding fragment thereof, or mRNA transcribed from the corresponding gene is inhibited by a mechanism such as RNAi.

In one embodiment, the target is one or more independently selected from the group consisting of amphiregulin, BTC, NRG1a, NRG1b, NRG3, TGFa, LRIG1, LRIG3, EGF, EGF-L6, epigen, HB-EGF, EGFR, Her2, Her3 and Her4, for example, the protein is a target for an inhibitor, such as an antibody or a binding fragment thereof, or mRNA transcribed from the corresponding gene is inhibited by a mechanism such as RNAi.

In one embodiment, the target is a ligand or receptor for one or more independently selected from the group consisting of hedgehog, FGF, IGF, Wnt, VEGF, TNF, TGFe, PDGF, and Notch.

In one embodiment, the adenovirus expresses an antibody or antibody fragment specific for VEGF. In one embodiment, the antibody is an anti-VEGF antibody. For example, such as an antibody having the amino acid sequence of the antibody bevacizumab or an equivalent thereof. In one embodiment, the adenovirus expresses a full-length anti-human VEGF antibody. In one embodiment, expression of the full-length anti-human VEGF antibody is under the control of an endogenous promoter, such as the major late promoter (MLP), in particular at the BY position. In one embodiment, the adenovirus expresses an scFv form of the anti-human VEGF antibody. In one embodiment, expression of the scFv form of the anti-human VEGF antibody is under the control of an endogenous promoter, such as the major late promoter, in particular at the _BY position.

In one embodiment, the target is an IDO.

In one embodiment, the target is an antigen for recognition by immune cells (such as T cells activated by a bispecific T cell activator), one or more proteins or peptides independently selected from the group consisting of immunogenic proteins from infectious

- 35 043895 organisms such as cytomegalovirus antigens, influenza antigens, hepatitis B surface and core antigens, diphtheria toxoid, Crm197, tetanus toxoid; peptides derived from such antigens that are known T-cell or antibody epitopes, or genetically engineered composites or multimers of such antigens; tumor-derived proteins as antigens; peptides derived from such antigens that are known T-cell or antibody epitopes; and genetically engineered composites or multimers of such antigens, such as WT1, MUC1, LMP2, idiotype, HPV E6&E7, EGFRvIII, HER-2/neu, MAGE A3, non-mutant p53, mutant p53, NY-ESO-1, GD2, PSMA, PCSA, PSA, gp100, CEA, MelanA/MART1, mutant Ras, proteinase 3 (PR1), bcr-abl, tyrosinase, survivin, PSA, hTERT, in particular WT1, MUC1, HER-2/neu, NY-ESO-1, survivin or hTERT.

The skilled person will appreciate that there are many possibilities for nucleic acid sequences that encode a given amino acid sequence due to codon redundancy, that silent mutations of nucleic acid base pairs lack immunogenic properties, and the present invention includes all nucleic acid sequences that encode a given amino acid sequence defined in any of SEQ ID NO.

In one embodiment, the peptide, polypeptide or protein encoded by the transgene is a mimotope. As used herein, the term mimotope means a molecule, often a peptide, that mimics the structure of an epitope. The latter property elicits an antibody response similar to that elicited by the epitope. An antibody to a given epitope antigen recognizes a mimotope that mimics the epitope. Mimotopes are typically obtained from phage display libraries by biopanning. Vaccines using mimotopes are under development. Thus, antibodies of known specificity can be used to screen libraries (e.g., phage display peptide libraries - e.g., Ab sequence libraries or non-antibody peptide libraries, especially optimized to obtain peptides with more stable three-dimensional conformations). The production of mimotopes is well described in the art (see Tribbick G, Rodda S. Combinatorial methods for discovery of peptide ligands which bind to antibody-like molecules. J Mol Recognit. 2002 15(5):306-10; Masuko T, Ohno Y, Masuko K, Yagi H, Uejima S, Takechi M, Hashimoto Y. Towards therapeutic antibodies to membrane oncoproteins by a robust strategy using rats immunized with transfectants expressing target molecules fused to green fluorescent protein. Cancer Sci. 2011 102(1):25-35).

In one embodiment, mimotope or other engineered vaccine antigens are encoded by a transgene and expressed to induce an antibody response in a recipient patient, wherein the induced antibodies exert a desired therapeutic effect. In one embodiment, GFP-peptide fusion proteins with peptide sequences from a desired human ligand are used to induce autoreactive antibody responses, such as a peptide region of PD-L1 that is known to be important for binding to a target molecule. PD-1 can be genetically linked to GFP or other highly immunogenic foreign carrier proteins such that the immune antibody response to the peptide includes antibodies that cross-react with the natural PDL1 molecule and thus serve to block PD-L1:PD-1 interactions in the same way as a direct encoding anti-PDL1 antibody. Ideas for creating vaccines that induce autoreactive antibody responses are well described in the art (see Spohn G, Bachmann MF. Therapeutic vaccination to block receptorligand interactions. Expert Opin Biol Ther. 2003 3(3):469-76; Link A, Bachmann MF. Immunodrugs: breaking B- but not T-cell tolerance with therapeutic anticytokine vaccines. Immunotherapy 2010 2(4):561-74; Delavallee L, Assier E, Semerano L, Bessis N, Boissier MC. Emerging applications of anticytokine vaccines. Expert Rev Vaccines. 2008 7(10):1507-17).

In one or more embodiments, the transgene used encodes a sequence shown in any of SEQ ID NOs: 2, 4, 7, 11, or 16.

In another embodiment, the transgene used encodes a sequence that excludes the deca-His affinity tag at the C-terminus, such as shown in the virus set forth in any of SEQ ID NOs: 72-78.

An advantage of the adenoviruses of the present invention is their ability to express and release antibody forms (such as a bispecific T cell activator) and other proteins such as cytokines encoded by a transgene therein into a culture supernatant in vitro or into the stroma of tumor tissue in vivo. Leader sequences can facilitate the encoding of proteins/polypeptides or peptides released from a cancer cell. Therefore, in one embodiment, the encoded protein comprises a leader sequence. As used herein, the term leader sequence refers to a polynucleotide sequence located between a promoter sequence and a coding region that can regulate gene expression at the transcriptional or translational level.

In one embodiment, the coding sequence encodes a peptide. As used herein, the term peptide refers to an amino acid chain that is not a fully functional protein. Typically, it is a fragment that retains some or all of the function of the protein of which it is a fragment or can be recognized by the immune system, such as peptides from

- 36,043,895 or more amino acids that can be recognized by T cells.

In one embodiment, the transgene is a reporter gene encoding, for example, an imaging agent, including bioluminescent, fluorescent imaging agents (including activatable fluorescent imaging agents), such as luciferase, GFP or eGFP, or red fluorescent protein.

As used in the present invention, the term reporter gene or reporter sequence means a gene or DNA sequence that produces a product that is easily detected in eukaryotic cells and can be used as a marker to determine the activity of another gene to which its DNA has been closely linked or combined. Reporter genes provide the cells or organisms that express them with characteristics that are easy to detect and measure, or are selectable markers. Reporter genes are often used as an indicator of whether a particular gene has been taken up or expressed by a population of cells or organisms. Examples of common reporter genes include, but are not limited to, LacZ, luciferase, GFP, eGFP, neomycin phosphotransferase, chloramphenicol acetyltransferase, sodium iodide symporter (NIS), nitroreductase (e.g., NfsA, NfsB), intracellular metalloproteins, HSV1-tk, or estrogen receptor.

In one embodiment, the genetic material (particularly the transgene) does not encode or express a reporter gene, such as an imaging agent, luciferase, GFP, or eGFP.

The viruses of the present invention can be tested for their preference for a particular tumor type by examining its lytic potential in a panel of tumor cells, for example, colon tumor cell lines include HT-29, DLD-1, LS174T, LS1034, SW403, HCT116, SW48 and Colo320DM. Any available colon tumor cell lines will be equally useful for such an assessment.

Prostate cell lines include DU145 and PC-3 cells. Pancreatic cell lines include Panc-1 cells. Breast tumor cell lines include the MDA231 cell line, and ovarian cell lines include the OVCAR-3 cell line. Hematopoietic cell lines include, but are not limited to, Raji and Daudi B lymphoid cells, K562 erythroblastoid cells, U937 myeloid cells, and HSB2 T lymphoid cells. Other available tumor cell lines would be equally useful.

The present invention also extends to the novel sequences disclosed herein. In one embodiment, the virus is shown in any of the sequences disclosed herein, such as any of SEQ ID NOs: 34-37 or a sequence at least 95% identical thereto, such as as set forth in any of SEQ ID NOs: 79-82.

Compositions.

The present invention also extends to a pharmaceutical composition containing the virus described in the present invention.

In one embodiment, a liquid parenteral preparation, for example for infusion or injection of a replication-capable oncolytic, according to the present invention is provided, characterized in that the composition provides a dose in the range of from 1x10 ¹⁰ to 1x10 ¹⁴ viral particles per dose volume.

A parenteral formulation means a formulation intended for delivery via the gastrointestinal tract. Typical routes of parenteral delivery include injection, implantation, or infusion. In one embodiment, the formulation is provided in a bolus delivery form.

In one embodiment, the parenteral formulation is in the form of an injection. Injection includes intravenous, subcutaneous, intratumoral or intramuscular injection. Injection in the present invention means introducing a liquid into the body through a syringe. In one embodiment, the method of the present invention does not include intratumoral injection.

In one embodiment, the parenteral formulation is in the form of an infusion.

Infusion in the present invention means the administration of fluids at a slower rate using a drip, an infusion pump, a syringe pump or an equivalent device. In one embodiment, the infusion is administered over a period of time in the range of 1.5 minutes to 120 minutes, such as about 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 15, 16, 17, 18, 19 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 65, 80, 85, 90, 95, 100, 105, 110 or 115 minutes.

In one embodiment, a single dose of the composition is less than 100 ml, such as 30 ml, such as administered using a syringe pump. In one embodiment, a single dose of the composition is less than 10 ml, such as 9, 8, 7, 6, 5, 4, 3, 2 or 1 ml. In one embodiment, a single dose of the composition is less than 1 ml, such as 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2 or 0.1 ml.

In one embodiment, the injection is administered as a slow injection, such as over 1.5 to 30 minutes.

In one embodiment, the composition is intended for intravenous (IV) administration. This route is particularly effective for delivery of the oncolytic virus, as it provides rapid access to most organs and tissues and is particularly useful for the treatment of metastases, such as established metastases, especially those located in highly vascularized areas such as the liver.

- 37 043895 and lungs.

Therapeutic formulations will generally be sterile and stable under the conditions of manufacture and storage. The composition may be formulated as a solution, microemulsion, liposome or other parenteral formulation suitable for administration to humans, and may be prepared in a pre-filled device such as a syringe or vial, especially as a single dose.

Such a formulation typically contains a pharmaceutically acceptable diluent or carrier, such as a non-toxic, isotonic carrier, that is compatible with the virus and in which the virus is stable for the required period of time.

The carrier may be a solvent or dispersion medium containing, for example, water, ethanol, a polyol (for example, glycerol, propylene glycol and liquid polyethylene glycol and the like) and suitable mixtures thereof. Proper fluidity can be maintained, for example, by using a dispersant or a surfactant such as lecithin or a nonionic surfactant such as polysorbate 80 or 40. In dispersions, the presence of a surfactant may help maintain the desired particle size. Examples of isotonic agents include sugars, polyalcohols such as mannitol, sorbitol or sodium chloride in the composition.

In one embodiment, the parenteral preparations used may contain one or more of the following buffers, such as 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, phosphate buffer and/or Tris buffer, a sugar, such as dextrose, mannose, sucrose or the like, a salt, such as sodium chloride, magnesium chloride or potassium chloride, a detergent, such as a non-ionic surfactant, such as Brij, PS-80, PS-40 or the like. The formulation may also contain a preservative, such as EDTA or ethanol, or a combination of EDTA and ethanol, which are believed to prevent one or more possible degradation pathways.

In one embodiment, the composition will comprise a purified oncolytic virus according to the present invention, for example, from 1x10 ¹⁰ to 1x10 ¹⁴ viral particles per dose, for example from 1x10 ¹⁰ to 1x10 ¹² viral particles per dose. In one embodiment, the concentration of virus in the composition is in the range of 2x10 ⁸ to 2x10 ¹⁴ viral particles/mL, for example 2x10 ¹² viral particles/mL.

In one embodiment, the parenteral formulation comprises glycerin.

In one embodiment, the composition comprises an oncolytic adenovirus as described herein, HEPES (N-2-hydroxyethylpiperazine-N-2-ethanesulfonic acid), glycerol and a buffer.

In one embodiment, the parenteral formulation comprises the virus of the present invention, HEPES, e.g. 5 mM, glycerol, e.g. 5-20% (v/v), hydrochloric acid, e.g. to adjust the pH to the range 7-8, and water for injection.

In one embodiment, 0.7 ml of the virus of the invention at a concentration of 2x10 ¹² viral particles/ml is prepared in 5 mM HEPES, 20% glycerol with a final pH of 7.8.

A detailed discussion of pharmaceutically acceptable carriers is provided in Remington's Pharmaceutical Sciences (Mack Publishing Company, N.J. 1991).

In one embodiment, the composition is provided as a composition for topical administration, including inhalation.

Suitable inhalation preparations include inhalation powders, metered-dose aerosols containing propellant gases, or inhalation solutions not containing propellant gases. Inhalation powders according to the present invention typically contain the virus described in the present invention with a physiologically acceptable filler.

These inhalation powders may contain monosaccharides (eg glucose or arabinose), disaccharides (eg lactose, sucrose, maltose), oligo- and polysaccharides (eg dextrans), polyalcohols (eg sorbitol, mannitol, xylitol), salts (eg sodium chloride, calcium carbonate) or mixtures thereof. It is advisable to use mono- or disaccharides, using lactose or glucose, for example, but not exclusively, in the form of their hydrates.

Particles for deposition in the lungs require a particle size of less than 10 μm, such as 1-9 μm, such as 0.1 to 5 μm, in particular 1 to 5 μm. The size of the virus-bearing particles is of primary importance, and thus, in one embodiment, the virus according to the present invention can be adsorbed or absorbed onto a particle, such as a lactose particle of a given size.

Propellant gases that can be used to prepare inhalable aerosols are known in the art. Suitable propellant gases are selected from hydrocarbons such as n-propane, n-butane or isobutane, and halogenated hydrocarbons such as chlorinated and/or fluorinated derivatives of methane, ethane, propane, butane, cyclopropane or cyclobutane. The above-mentioned gaseous propellants can be used alone or in mixtures thereof.

Particularly suitable propellant gases are halogenated derivatives of alkanes selected from TG 11, TG 12, TG 134a and TG227. Of the above-mentioned halogenated hydrocarbons, TG134a (1,1,1,2-tetrafluoroethane) and TG227 (1,1,1,2,3,3,3-heptafluoropropane) and mixtures thereof are particularly suitable.

Inhalation aerosols containing propellant gas may also contain other ingredients,

- 38 043895 such as cosolvents, stabilizers, surfactants, antioxidants, lubricants and pH adjusting agents. All of these ingredients are known in the art.

Inhalation aerosols containing propellant gas according to the invention may contain up to 5% by weight of active substance. Aerosols according to the invention contain, for example, from 0.002 to 5% by weight, from 0.01 to 3% by weight, from 0.015 to 2% by weight, from 0.1 to 2% by weight, from 0.5 to 2% by weight or from 0.5 to 1% by weight of active ingredient.

Alternatively, local administration to the lungs may also be accomplished by administering the formulation as a liquid solution or suspension, for example, using a device such as a nebulizer, for example a nebulizer connected to a compressor (e.g., a Pari LC-Jet Plus® nebulizer connected to a Pari Master® compressor manufactured by Pari Respiratory Equipment, Inc., Richmond, VA).

The virus of the invention can be delivered dispersed in a solvent, for example in the form of a solution or suspension, for example as already described above for parenteral formulations. It can be suspended in a suitable physiological solution, for example, physiological saline or another pharmacologically acceptable solvent or a buffered solution. Buffered solutions known in the art can contain from 0.05 to 0.15 mg disodium edetate, from 8.0 to 9.0 mg NaCl, from 0.15 to 0.25 mg polysorbate, from 0.25 to 0.30 mg anhydrous citric acid and from 0.45 to 0.55 mg sodium citrate per 1 ml of water, to obtain a pH of about 4.0-5.0.

Therapeutic suspensions or solution formulations may also contain one or more excipients. Excipients are well known in the art and include buffers (e.g., citrate buffer, phosphate buffer, acetate buffer, and bicarbonate buffer), amino acids, urea, alcohols, ascorbic acid, phospholipids, proteins (e.g., serum albumin), EDTA, sodium chloride, liposomes, mannitol, sorbitol, and glycerol. The solutions or suspensions may be encapsulated in liposomes or biodegradable microspheres. The formulation will typically be provided in substantially sterile form using sterile manufacturing techniques.

This may include preparing and sterilizing by filtration the buffered solvent/solution used to prepare the formulation, aseptically suspending the antibody in the sterile buffered solvent/solution, and dispensing the formulation into sterile vessels by methods known to those skilled in the art.

The spray composition according to the present invention can be provided, for example, in the form of unit doses (for example, sealed plastic containers or vials) packed in foil envelopes. Each vial contains a unit dose in a volume of, for example, 2 ml of solvent/buffer solution.

Treatment.

In a further aspect, the present invention extends to a virus or composition thereof as described herein for use in therapy, in particular for the treatment of cancer.

In one embodiment, the method of treatment is intended for use in the treatment of a tumor, in particular a solid tumor.

A tumor in the context of this application means an abnormal mass of tissue resulting from uncontrolled and progressive excessive cell division, also called a neoplasm. Tumors can be benign (non-malignant) or malignant. The term tumor includes all forms of cancer and metastases.

In one embodiment, the tumor is a solid tumor. The solid tumor may be localized or metastatic.

In one embodiment, the tumor is of epithelial origin.

In one embodiment, the tumor is a malignancy such as colorectal cancer, hepatoma, prostate cancer, pancreatic cancer, breast cancer, ovarian cancer, thyroid cancer, kidney cancer, bladder cancer, head and neck cancer, or lung cancer.

In one embodiment, the tumor is a colorectal malignancy.

Malignant tumor in the context of the present application means cancer cells.

In one embodiment, the oncolytic adenovirus is used to treat or prevent metastasis.

In one embodiment, the method or composition of the present invention is used to treat drug-resistant forms of cancer.

In one embodiment, the virus is administered in combination with the administration of an additional cancer treatment or therapy.

In one embodiment, a virus or composition according to the present invention is provided for use in the preparation of a medicament for the treatment of cancer, such as the cancer described above.

In another aspect, a method of treating cancer is provided, comprising administering a therapeutically effective amount of a virus or drug according to the present invention to a patient in need thereof, such as a human patient.

In one embodiment, the oncolytic virus or composition of the present invention is administered in combination with another therapy.

The term in combination in the present invention is intended to cover cases where the oncolytic virus is administered before, simultaneously with and/or after cancer treatment or therapy.

Cancer treatments include surgery, radiation therapy, targeted therapy and/or chemotherapy.

As used herein, the term "treatment of cancer" refers to treatment with a therapeutic compound or biological agent, such as an antibody, intended for the treatment of cancer and/or supportive care.

In one embodiment, the cancer treatment is selected from any other anti-cancer therapy, including a chemotherapeutic agent, a targeted anti-cancer agent, radiation therapy, radioisotope therapy, or any combination thereof.

In one embodiment, a virus of the present invention, such as an oncolytic adenovirus, can be used as a pre-treatment prior to therapy, such as surgery (neoadjuvant therapy), to shrink a tumor, to treat metastasis and/or prevent metastasis or further metastasis. An oncolytic adenovirus can be used after therapy, such as surgery (adjuvant therapy), to treat metastasis and/or prevent metastasis or further metastasis.

As used herein, the term "concurrently" refers to the administration of an additional cancer treatment at or about the same time as the oncolytic adenovirus formulation. Such treatment may be contained in the same formulation or administered as a separate formulation.

In one embodiment, the virus is administered in combination with administration of a chemotherapeutic agent.

As used in the present invention, the term chemotherapeutic agent is intended to refer to specific antitumor chemical agents or drugs that selectively destroy malignant cells and tissues. For example, alkylating agents, antimetabolites, anthracyclines, plant alkaloids, topoisomerase inhibitors and other antitumor agents. Other examples of chemotherapy include doxorubicin, 5-fluorouracil (5-FU), paclitaxel, capecitabine, irinotecan and platins such as cisplatin and oxaliplatin. The preferred dose can be selected by the practitioner depending on the nature of the cancer to be treated.

In one embodiment, the therapeutic agent is ganciclovir, which can help control immune responses and/or tumor vascularization.

In one embodiment, one or more treatments used in the method of the present invention are metronomic, i.e., continuous or frequent treatment with low doses of anticancer drugs, often administered concurrently with other therapies.

Oncolytic adenoviruses of the B subgroup, particularly Ad11 and those derived from it such as EnAd, may be particularly synergistic with chemotherapeutic agents because they appear to have a mechanism of action that is largely independent of apoptosis, killing cancer cells by a predominantly necrolytic mechanism. In addition, the immunosuppression that occurs during chemotherapy may allow the oncolytic virus to function with greater efficiency.

Therapeutic dose in the present invention refers to the amount of virus, such as oncolytic adenovirus, which is suitable to achieve the desired therapeutic effect when used in a suitable treatment regimen, for example, alleviate the symptoms of a disease or condition. The dose can be considered a therapeutic dose in the treatment of cancer or metastasis, when the amount of virus particles can be sufficient to lead to the following: the growth of a tumor or metastasis is slowed or stopped, or it is found that the tumor or metastasis is reduced in size and / or the life expectancy of the patient is increased. Suitable therapeutic doses are generally a balance between therapeutic effect and tolerable toxicity, for example, when the side effect and toxicity are tolerable, taking into account the beneficial effect achieved by this therapy.

In one embodiment, the virus or therapeutic construct of the present invention (including a formulation containing the same) is administered weekly, such as one week 1 dose is administered on day 1, 3, 5, and then one dose every subsequent week.

In one embodiment, the virus or therapeutic construct of the present invention (including a formulation containing the same) is administered every two weeks or three weeks, such as administered in week 1 on days 1, 3, and 5, and in week 2 or 3 also administered on days 1, 3, and 5. This dosing regimen can be repeated as many times as necessary.

In one embodiment, the virus or therapeutic construct of the present invention (including a formulation containing the same) is administered monthly.

In one embodiment, the viruses and constructs of the present invention are produced by recombinant methods. The skilled person will understand that the weaponized adenovirus genome can be produced by other technical means, including complete genome synthesis or a plasmid comprising a portion of the entire genome. The skilled person will understand that in the case of genome synthesis, the insertion region may not contain restriction site nucleotides, since the latter are artifacts after gene insertion using cloning methods.

In one embodiment, the weaponized adenovirus genome is entirely synthetically produced, such as according to SEQ ID NO: 34-37.

The invention in the present invention further extends to an adenovirus of formula (I) or a subformula thereof obtained or capable of being obtained by insertion of a transgene or a transgene cassette.

In the present invention, is means containing.

In the context of this description, containing should be interpreted as including.

Embodiments of the invention comprising certain features/elements are also intended to be expandable to alternative embodiments consisting of or consisting essentially of the corresponding elements/features.

Where technically feasible, embodiments of the invention may be combined.

Technical references such as patents and applications are incorporated into this patent by reference.

Any embodiments specifically and expressly disclosed in the present invention may form the basis of a disclaimer either individually or in combination with one or more additional embodiments.

This application claims priority under GB1614607.8, GB1700663.6, GB1706219.1 and GB1713765.4, incorporated herein by reference. These documents may be used to correct errors in this specification, in particular errors in the sequence listing.

The present invention is further described, for purposes of illustration only, in the following examples which refer to the accompanying figures, in which the following is shown:

Description of figures

Fig. 1. (A) Schematic representation of the bispecific T cell activator antibody of the present invention, with or without the optional decahistidine affinity tag. Ig SP: signal peptide; 10His: decahistidine affinity tag; L: GS linker; VL: variable light domain; VH variable heavy domain. (B) Plasmid map for pSF-CMV-EpCAM Bisp.activ.T cells. (C) Plasmid map for pSF-CMV-FAP Bisp.activ.T cells. (D) Plasmid map for pSF-CMV-Control Bisp.activ.T cells.

Fig. 2. (A) Dot blot showing quantification of recombinant bispecific T cell activators. (B) Graph showing ELISA results for FAP. (C) Graph showing ELISA results for EpCAM.

Fig. 3. CD69 (A) and CD25 (B) expression levels of T cells co-cultured alone or with NHDF cells in the presence of FAP bispecific T cell activator and control bispecific T cell activator measured using flow cytometry.

Fig. 4. (A) IFNγ expression levels for T cells co-cultured alone or with NHDF cells in the presence of FAP bispecific T cell activator and control bispecific T cell activator measured by intracellular cytokine staining. Graphs (B) and (C) show CD69 and CD25 expression levels for T cells co-cultured alone or with DLD cells in the presence of EpCAM bispecific T cell activator and control bispecific T cell activator analyzed by flow cytometry.

Fig. 5. (A) IFN-γ expression levels for T cells co-cultured with DLD cells in the presence of EpCAM bispecific T cell activator and control bispecific T cell activator measured by intracellular cytokine staining. Graphs (B) and (C) showing CD69 and CD25 levels for PBMCs co-cultured with DLD cells in the presence of EpCAM bispecific T cell activator and control bispecific T cell activator analyzed by flow cytometry.

Fig. 6. (A) Results of LDH assay showing cytotoxicity of NHDF cells co-cultured with T cells and FAP bispecific T cell activator or control bispecific T cell activator. (B) Graph of results of LDH assay showing cytotoxicity of BTC100 cells co-cultured with T cells and FAP bispecific T cell activator or control bispecific T cell activator. (C) Images of NHDF cells after co-culture with T cells and FAP bispecific T cell activator compared to control bispecific T cell activator.

Fig. 7. (A) Dot plots showing FAP expression in different patient-derived cells. (B) Graph showing % of cells expressing EpCAM and FAP across different cells and cell lines.

- 41 043895

Fig. 8. (A) Dose-response relationship of NHDF for FAP bispecific T cell activator with increasing concentration of bispecific T cell activator. (B) and (C) Results of LDH assay showing cytotoxicity of DLD cells co-cultured with T cells and EpCAM bispecific T cell activator or control bispecific T cell activator.

Fig. 9. (A) Results of LDH assay showing cytotoxicity of SKOV cells co-cultured with T cells and bispecific T cell activator to EpCAM or control bispecific T cell activator. (B) Results of LDH assay showing cytotoxicity of MCF7 cells co-cultured with T cells and bispecific T cell activator to EpCAM or control bispecific T cell activator.

Fig. 10. Dose dependence of NHDF for the bispecific T cell activator to EpCAM with increasing concentration of the bispecific T cell activator.

Fig. 11. (A) FAP expression in CHO cells detected with FAP or isotype control antibody and analyzed by flow cytometry. (B) Results of LDH assay showing cytotoxicity of CHO or CHO-FAP cells cocultured with T cells and FAP bispecific T cell activator or control bispecific T cell activator.

Fig. 12. T cell activation (based on CD69 and CD25 expression levels) by CHO cells compared to CHO-FAP cells, flow cytometric analysis.

Fig. 13. (A) Graphs of EpCAM expression of parental cell lines compared to the stable transfected variant, determined by EpCAM staining or isotype control antibody and analyzed by flow cytometry. (B) Results of LDH assay showing cytotoxicity of CHO or CHO-EpCAM cells cocultured with T cells and bispecific T cell activator to EpCAM or control bispecific T cell activator.

Fig. 14. T cell activation (based on CD69 and CD25 expression levels) by CHO cells compared to CHO-EpCAM, flow cytometric analysis.

Fig. 15. (A) Ability of FAP bispecific T cell activator to activate CD4+ or CD8+ T cells (based on CD69 and CD25 expression levels) as analyzed by flow cytometry. (B) Graph of LDH assay results showing cytotoxicity of NHDF cells co-cultured with CD4+ or CD8+ T cells and FAP bispecific T cell activator or control bispecific T cell activator.

Fig. 16. (A) Activation of CD4+ and CD8+ T cells (based on CD69 and CD25 expression levels) by DLD cells in the presence of EpCAM or control bispecific T cell activator as analyzed by flow cytometry. (B) Results of LDH assay showing cytotoxicity of DLD cells co-cultured with CD4+ or CD8+ T cells and bispecific T cell activator to EpCAM or control bispecific T cell activator.

Fig. 17. (A) Number of CD3+ T cells from ascites cultured with control or FAP bispecific T cell activator. (B) CD25 expression levels of T cells from ascites cultured with control or FAP bispecific T cell activator. (C) Number of FAP+ cells from ascites cultured with control or FAP bispecific T cell activator.

Fig. 18. (A) Schematic representation of the genome of adenoviruses of the present invention. (B) Graphs comparing the kinetics of expression driven by the CMV and SA promoters.

Fig. 19. (A) Quantification of the number of detected viral genomes per cell for NG-601, NG-602, NG-603, NG-604, NG-605, NG-606, and EnAd. (B) Oncolytic activity of NG-601, NG-602, NG-603, NG-604, NG-605, NG-606, or EnAd assessed by infection of A549 cells.

Fig. 20. (A) T cell activation (based on CD69 and CD25 expression levels) by NG601, NG-602, NG-605, and NG-606 viruses co-cultured with CHO-FAP, analyzed by flow cytometry. (B) T cell activation (based on CD69 and CD25 expression levels) by NG601, NG-602, NG-605, and NG-606 viruses co-cultured with CHO-EpCAM, analyzed by flow cytometry.

Fig. 21. Results of experiments to determine the amount of bispecific T cell activator FAP obtained from NG-605 and NG-606.

Fig. 22. Results of experiments to determine the amount of bispecific T cell activator to EpCAM obtained from NG-601 and NG-602.

Fig. 23. Microscopic images of Ad293 cells infected with NG-607, NG-608, NG-609 and NG-610.

Fig. 24. (A) Cytotoxicity of EnAd-infected DLD cells assayed using XCELLigence. (B) Cytotoxicity of EnAd-infected SKOV cells assayed using XCELLigence. (C) Cytotoxicity of EnAd-infected NHDF cells assayed using XCELLigence.

Fig. 25. (A) Ability of NG-603, NG-604, NG-605, NG-606 and EnAd to kill NHDF cells as analyzed using XCELLigence. (B) Ability of NG-603, NG-604, NG-605, NG-606 and EnAd to kill NHDF cells as determined by LDH assay.

Fig. 26. T cell activation graphs (based on CD69 and CD25 expression levels) by viruses

NG-603, NG-604, NG-605, NG-606 co-cultured with NHDF, SKOV and T cells, flow cytometry analysis.

Fig. 27. (A) Graph of T cell activation (based on CD69 and CD25 expression levels) by NG-603, NG-604, NG-605, NG-606 viruses co-cultured with NHDF and SKOV cells compared to SKOV alone, flow cytometric analysis. (B) Graph of cytotoxicity of NHDF cells infected with NG-605 and NG-606, as determined by LDH assay.

Fig. 28. Still images from time-lapse videos of NHDF cell lysis with recombinant bispecific T cell activator to FAP, EnAd, NG-603 or NG-605.

Fig. 29. Still images from time-lapse videos of NHDF cell lysis by NG-607, NG-608, NG-609, or NG-610 viruses.

Fig. 30. Cytotoxicity of DLD cells infected with EnAd, NG-601, NG-602, NG-603, and NG-604 in the presence or absence of T cells as analyzed by XCELLigence.

Fig. 31. Cytotoxicity of DLD cells infected with EnAd, NG-601, NG-602, NG-603 and NG-604 in the presence or absence of T cells as determined by LDH assay.

Fig. 32. T cell activation (based on CD69 and CD25 expression levels) by EnAd, NG601, NG-602, NG-603 and NG-604 viruses, analyzed by flow cytometry.

Fig. 33. Results of experiments to determine the ability of NG-601 to kill DLD tumor cells at different multiplicities of infection (MOI) in the presence or absence of CD3+ T cells, assessed by xCELLigence.

Fig. 34. Ability of EnAd and NG-601, NG-602, NG-603 and NG-604 to kill SKOV tumor cells in the presence or absence of CD3+ T cells as assessed by xCELLigence.

Fig. 35. Ability of EnAd and NG-601, NG-602, NG-603 and NG-604 to kill SKOV tumor cells in the presence or absence of CD3+ T cells as determined by LDH assay.

Fig. 36. T cell activation (based on CD69 and CD25 expression levels) with EnAd, NG-601, NG-602, NG-603 and NG-604 viruses co-cultured with SKOV tumor cells, flow cytometric analysis.

Fig. 37. T cell activation (based on CD69 and CD25 expression levels) by EnAd, NG601, NG-602, NG-603 and NG-604 viruses co-cultured with ascites cells, analyzed by flow cytometry.

Fig. 38. Still images from a video of NHDF cell lysis by EpCAM-specific T cell activator, EnAd, NG-601, or NG-603 viruses.

Fig. 39. Microscopic images of ascites cells obtained from a patient infected with the viruses of the present invention and stained with EnAd-CMV-GFP and EnAd-SA-GFP as reporters for infection and late stage viral gene expression.

Fig. 40. (A) A graph showing the expression levels of CD25 on CD3+ T cells in ascites samples that were infected with the viruses of the present invention. (B) A graph showing the number of FAP+ cells in ascites samples that were infected with the viruses of the present invention.

Fig. 41. Microscopic images of ascites cells obtained from a cancer patient infected with the viruses of the present invention and stained with EnAd-CMV-GFP and EnAd-SA-GFP as reporters for infection and late stage viral gene expression.

Fig. 42. Number of CD3+ T cells in ascites samples obtained from a cancer patient and infected with viruses of the present invention.

Fig. 43. CD25 expression levels on CD3+ T cells in ascites samples obtained from a cancer patient and infected with viruses of the present invention.

Fig. 44. Number of FAP+ cells in ascites samples obtained from a cancer patient and infected with viruses of the present invention.

Fig. 45. Characterization of the EpCAM bispecific T cell activator and its effect on PBMC-derived T cells. (A) Schematic structure of the EpCAM-targeted bispecific T cell activator and the non-specific control bispecific T cell activator. The VL and VH domains are linked to flexible peptide linkers (L) rich in serine and glycine for flexibility and solubility. IgSP, immunoglobulin light chain signal peptide; 10His, decahistidine affinity tag. (B) Induction of the activation markers CD69 and (C) CD25 on CD3-purified PBMCs cultured alone or with DLD cells (5:1) in the presence of bispecific T cell activator-containing supernatants. CD69 and CD25 were measured by flow cytometry after 24 h of co-culture. Significance was assessed by comparison with IgG isotype. (D) Percentage of IFNγ-positive T cells after 6 h in co-culture with DLD cells (5:1) and bispecific T cell activator-containing supernatants. (E) Proliferation, represented by division index and percentage of parental T cell population participating in proliferating, of CFSE-stained T cells in co-culture with DLD cells (5:1) and bispecific T cell activator-containing supernatants. Fluorescence was measured by flow cytometry 5 days after co-culture. Division index was simulated using Flowjo proliferation software. (F) T cell degranulation measured by CD107a externalization in co-culture with DLD cells (5:1) and bispecific T cell activator-containing supernatants. Externalization was assessed by co-culture with CD107a-specific antibody for 6 h followed by flow cytometric analysis. (G) Cytokine levels were measured with the LEGENDplex human Th cytokine panel using supernatants from T cell co-cultures with DLD cells (5:1) in the presence of bispecific T cell activator-containing supernatants for 48 h. Each experimental condition was measured in triplicate and data are presented as mean ± SD. Significance was assessed relative to untreated cells unless otherwise stated using one-way ANOVA with Tukey's post hoc test, *p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 46. Characterization of recombinant bispecific T cell activator to EpCAM (A) Dot blot to assess the amount of bispecific T cell activator to EpCAM produced by transfected HEK293A cells. (B) ELISA measuring the level of EpCAM binding by controls or recombinant EpCAM or non-specific bispecific T cell activator. Significance was assessed by comparison with the empty vector control using one-way ANOVA with Tukey's post hoc test, ***p<0.001.

Fig. 47. Evaluation of antigen specificity of bispecific T cell activator EpCAM-mediated cytotoxicity for T cells. (A) Induction of the activation marker CD25 on CD3+ T cells co-cultured with CHO or CHO-EpCAM cells (5:1) and bispecific T cell activator-containing supernatants, as determined by FACS analysis after 24 h of co-culture. (B) Cytotoxicity of CHO or CHO-EpCAM cells cultured with bispecific T cell activator-containing supernatants, alone or in co-culture with T cells. Cytotoxicity was assessed by LDH release into culture supernatants after 24 h of incubation. (C) Cytotoxicity of multiple EpCAM-positive carcinoma cells after 24 h in co-culture with T cells (1:5) and bispecific T cell activator-containing supernatants. Viability was measured by MTS assay after 24 h of co-culture. (D) EpCAM expression levels (N=1) assessed by FACS analysis of EpCAM-positive cell lines in (C) compared to background fluorescence measured using an isotype control antibody. (AC) Each experimental condition was measured in triplicate and data are presented as mean ± SD. Significance was assessed compared to untreated wells or T-cell-only control wells using one-way ANOVA with Tukey's post hoc test, *p<0.05, **p<0.01, ***p<0.001.

Fig. 48. Cytotoxicity of EnAd expressing a bispecific T cell activator to EpCAM in SKOV3 cells.

SKOV3 cells were incubated with EnAd or recombinant viruses in the absence (A) or presence (B) of T cells, and cytotoxicity was measured by LDH release at the indicated time points. Significance was assessed by comparison with uninfected control wells using one-way ANOVA with Tukey's post hoc test, ***p<0.001.

Fig. 49. Identification of T cells responsible for bispecific T cell activator-mediated cytotoxicity. (A) Bispecific T cell activator-mediated activation of CD4 and CD8 T cells 24 h after co-culture of CD3 T cells with DLD cells (5:1) and bispecific T cell activator-containing supernatant. Activation was assessed by surface expression of CD69 and CD25 and measured by flow cytometry. (B) Proliferative response of CFSE-stained CD4 and CD8 T cells co-cultured with DLD cells and incubated with bispecific T cell activator-containing supernatants. Fluorescence was measured after 5 days of incubation by FACS analysis. (C) Degranulation of CD4 and CD8 cells after 6 h co-culture with DLD cells and bispecific T cell activator-containing supernatants. CD107a-specific antibody was added to the culture medium during co-culture, and degranulation was assessed by flow cytometry. (D) Cytotoxicity of the CD4 or CD8 T cell subset was assessed by LDH release into the supernatant after 24 h incubation of DLD cells with CD4- or CD8-purified T4 cells (1:5) and bispecific T cell activator-containing supernatant. Each experimental condition was measured in triplicate, and data are presented as mean ± SD. Treatment with the bispecific T cell activator EpCAM was compared with the control bispecific T cell activator unless otherwise stated and significance was assessed using one-way ANOVA with Tukey's post hoc test, *p<0.05, **p<0.01, ***p<0.001.

Fig. 50. Cytotoxicity and T cell activation by EnAd expressing the bispecific T cell activator to EpCAM in DLD cells Cytotoxicity to infected DLD cells in the absence (A) or presence (B) of T cells. DLD cells were infected and co-cultured with T

- 44 043 895 cells and measured cytotoxicity by LDH release at the indicated time points. (C-D) T cells from (B) were harvested and stained for the activation markers CD69 (C) or CD25 (D) and analyzed by flow cytometry. (E-F) Quantification of the bispecific T cell activator to EpCAM produced from DLD cells infected with recombinant viruses. Standard curve of LDH release (Abs) from DLD in co-culture with CD3+ cells and various known amounts of recombinant bispecific T cell activator to EpCAM (E). Co-cultures were incubated in parallel with diluted supernatants (10,000-fold) of 3-day-infected DLD cells (F). The standard curve allowed the approximate determination of the bispecific T-cell activator to EpCAM produced at 165 μg and 50 μg per million DLD cells for EnAd-CMV-EpCAM Bisp.active.T-cells and EnAd-SA-EpCAM Bisp.act.T-cells, respectively. Significance was assessed by comparison with uninfected control wells using one-way ANOVA with Tukey's post hoc test, ***p<0.001.

Fig. 51. Characterization of the oncolytic virus EnAd expressing the bispecific T cell activator EpCAM using cell lines and PBMC-derived T cells. (A) DLD cells were infected with parental EnAd or recombinant virus (100 vp/cell) and harvested from wells after 24 or 72 h. Replication was assessed by measuring genomes using qPCR with respect to viral hexon. (B) Cytotoxicity of DLD cells infected with EnAd or recombinant virus at increasing virus concentrations. Cytotoxicity was measured by the MTS assay 5 days after infection. (C) Supernatants from 3-day-old uninfected or virus-naive HEK293A cells were assessed for transgene expression by immunoblot analysis and probed with anti-His antibody. (D) Induction of the activation marker CD25 of CD3-positive T cells cultured with CHO or CHO-EpCAM (E:T 5:1) and diluted HEK293A supernatants from (D). Activation was measured by surface expression of CD25 by flow cytometry. (E) Cytotoxicity of CHO or CHO-EpCAM cells incubated with HEK293A supernatants from (D) alone or in co-culture with CD3-purified PBMCs (E:T 5:1). HEK293A supernatants were diluted 300-fold.

Cytotoxicity was assessed by LDH released into the supernatant after 24 h of incubation. Each experimental condition was measured in triplicate and data were presented as mean ± SD. Significance was assessed using one-way ANOVA with Tukey's post hoc test comparing each experimental condition to untreated cells, *p<0.05, **p<0.01, ***p<0.001.

Fig. 52. Cellular composition of malignancy exudate (A) Representative image (pleural effusion sample, patient 3 from Fig. 57) showing screening of ascitic and exudate fluids for their cellular composition assessed by flow cytometry. (B) The table shows the absolute number of each cell type (in a sample of 10,000 cells).

Fig. 53. Superior activity of EnAd expressing bispecific T cell activator to EpCAM in a partially EnAd-resistant cell line (A-B) SKOV3 cell viability was monitored in real time for 160 h using an xCELLigence-based cytotoxicity assay. SKOV3 cells were seeded and infected with EnAd viruses or EnAd armed with the bispecific T cell activator at 0 h, with uninfected cells serving as a negative control. In (B), CD3-purified PBMCs (5:1) were added 2 h post infection and impedance was measured at 15-min intervals. (C-D) CD3-purified PBMCs were co-cultured with SKOV3 cells (5:1) that were infected with parental EnAd or recombinant armed viruses. At each time point, T cells were harvested and analyzed for surface expression of CD69 (C) or CD25 (D) by flow cytometry. (E) Time-lapse image sequences showing co-cultures of SKOV3 carcinoma cells (unstained), NHDF fibroblasts (red), and CD3-purified PBMCs (blue) infected with EnAd, EnAdCMVEpCAM Bisp.activ.T cells, or uninfected. Apoptosis was visualized using CellEvent Caspase 3/7 reagent (green). Images were acquired on a Nikon TE 2000-E Eclipse inverted microscope at 15-min intervals spanning 96 h. Representative images were recorded at the indicated time points; original magnification ×10; scale bar 100 μm. (A–D). Each experimental condition was measured in triplicate and data were presented as mean ± SD. Significance was assessed by comparison with uninfected controls using one-way ANOVA with Tukey's post hoc test, *p<0.05, **p<0.01, ***p<0.001.

Fig. 54. PD1 expression and the effect of PD1 antibodies on bispecific T cell activator-mediated T cell activation. (A) PD1 expression by endogenous T cells after their initial isolation from pleural effusions was assessed by flow cytometry. (B–D) Unpurified total cells from pleural effusions (from three different patients) were incubated in 100% fluid from the same pleural exudate in the presence of free bispecific T cell activator, EnAd, or recombinant virus. After 5 days, the total cell population was harvested and the number of (B) CD3+ T cells and cells that were (C) CD25+ were determined. (D) The number of EpCAM+ cells was measured by flow cytometry. Significance was assessed by comparison with untreated control wells using one-way ANOVA with Tukey's post hoc test, ***p<0.001.

Fig. 55. EnAd expressing a bispecific T cell activator to EpCAM can selectively kill primary human tumor cells from chemotherapy-pretreated patients. (A) Cytotoxicity of EpCAM+ cells or (B) FAP+ fibroblasts first isolated from ascites of three patients and expanded ex vivo, then incubated with a recombinant bispecific T cell activator or infected with EnAd or recombinant virus. Cytotoxicity was measured by flow cytometry after 5 days. (C) Induction of the activation marker CD25 on CD3-positive T cells cultured with ascites EpCAM+ and FAP+ cells from (A + B). Each experimental condition was measured in triplicate, and data are presented as mean ± SD. Significance was assessed by comparison with untreated wells using one-way ANOVA with Tukey's post hoc test, *p<0.05, **p<0.01, ***p<0.001.

Fig. 56. Bispecific T cell activator to EpCAM can overcome the immunosuppressive effects of ascites fluid and activate endogenous T cells (A-B) PBMC-derived T cells were incubated with anti-CD3 antibodies in RPMI culture medium or in the presence of 100% peritoneal ascites fluid from five ovarian cancer patients. (A) After 24 h of T cell induction, activation markers CD69 and CD25 were analyzed and (B) T cell degranulation was measured by CD107a externalization using flow cytometry. (C) MCF7 cell viability was monitored in real time for 60 h using an xCELLigence-based cytotoxicity assay. MCF7 cells were seeded and incubated with control or bispecific T cell activator to EpCAM for 25 h in the presence of RPMI medium or 100% ascites fluid #1 or #2. Untreated cells served as a negative control. CD3-purified PBMCs were added simultaneously (5:1) and impedance was measured at 15-min intervals. (D) Endogenous crude total cells from peritoneal ascites were incubated in 100% ascites fluid in the presence of free EpCAM or control bispecific T cell activator. After 24 h, the total cell population was collected and the number of CD3+/CD69+ and CD3+/CD25+ cells was measured by flow cytometry. Each experimental condition was measured in triplicate and the data are presented as mean ± SD. Significance was assessed by comparison with RPMI (A + B), untreated samples (D), or control bispecific T cell activator (E) using one-way ANOVA with Tukey's post hoc test, *p<0.05, **p<0.01, ***p<0.001.

Fig. 57. EnAd expressing a bispecific T cell activator to EpCAM can activate endogenous T cells to kill endogenous tumor cells in pleural exudates of malignancies.

Crude total cells from pleural effusions (from four different patients) were incubated in 100% fluid from the same pleural exudate in the presence of free bispecific T cell activator, EnAd, or recombinant virus. After 5 days, the total cell population was harvested and the number of (A) CD3+ T cells and cells that were (B) CD25+ were determined. (C) The number of EpCAM+ cells was measured by flow cytometry. (D) Representative images (10× magnification; scale bar 100 μm) and flow cytometric analysis of pleural effusion cells from patient 3 (cancer cells and lymphocytes) after treatment with EnAd or EnAd-CMVEpCAM Bisp.activ.Tcells. (E) After 5 days, cytokine levels were measured using the LEGENDplex human cytokine panel using pleural effusion cultures after incubation with free recombinant bispecific T cell activator or infection with EnAd or recombinant virus. Each experimental condition was measured in triplicate and data are presented as mean ± SD. Significance was assessed by comparison with untreated controls using one-way ANOVA with Tukey's post hoc test, *p<0.05, **p<0.01, ***p<0.001.

Fig. 58. The amount of IL-10 measured in normal serum (NS) or patient malignancy exudate fluid (A: peritoneal ascites, P: pleural effusion) using the IL-10 Human ELISA MAX kit (Biolegend, 430604).

Fig. 59. CD3/28 bead-mediated T cell activation of PBMCs (based on CD69/CD25 levels) in patient fluids compared to normal serum, as determined by flow cytometry. A: patient exudate fluid, P: pleural fluid.

Fig. 60. CD3/28 bead-mediated degranulation of PBMC T cells (based on CD107a expression) in patient fluids. A: ascites, P: pleural fluid.

Fig. 61. Correlation between IL-10 levels in patient fluids and CD3/CD28 microsphere-mediated T cell degranulation.

- 46 043895

Fig. 62. EpCAM bispecific T cell activator-mediated microspheres in PBMC T cells (based on CD69/CD25 expression) in patient fluids. A: ascites, P: pleural fluid.

Fig. 63. EpCAM bispecific T cell activator microsphere-mediated T cell degranulation of PBMCs (based on CD107a expression) in patient fluids. A: ascites, P: pleural fluid.

Fig. 64. Cytotoxicity of SKOV3 mediated by EpCAM bispecific T cell activator microspheres in patient fluids. A: ascites, P: pleural fluid.

Fig. 65. EpCAM bispecific T cell activator-mediated T cell activation (based on CD25/CD69 expression) in RPMI medium compared to ascites fluid.

Fig. 66. Ability of EnAd-SA-EpCAM Bisp. act.T cells and EnAd-SA-Control Bisp. act.T cells to induce T cell-mediated target cell lysis in RPMI medium compared to ascites fluid. ((A) CD3+ counts. (B) CD25 expression by T cells. (C) EpCAM+ cell counts determined by flow cytometry.

Fig. 67. Ability of EnAd-SA-EpCAM Bisp.act.T cells and EnAd-SA-Control Bisp.act.T cells to induce T cell-mediated lysis of target cells in ascites fluid (samples from 7 patients). (A) CD3+ counts. (B) CD25 expression by T cells. (C) EpCAM+ cell counts determined by flow cytometry. See legend for Fig. 67A.

Fig. 68. Comparison of activation of T cell cytokine production by recombinant FAP bispecific T cell activator protein in the presence of human fibroblasts and polyclonal activation with anti-CD3/CD28 beads. (A) IFNy levels measured by ELISA. (B) Cytokine levels measured by cytokine bead array.

Fig. 69. FAP-targeted bispecific T cell activator induces T cell degranulation and specific cytotoxicity of FAP+ cells. (A) T cell degranulation in culture with NHDF cells (5:1) and (B) bispecific T cell activator-containing supernatants. Degranulation was assessed by externalization of CD107a after 6 h of culture with a specific CD107a antibody and measured by flow cytometry. CD3/CD28 Dynabeads were used as a positive control. (C) Cytotoxicity of NHDF cells after 24 h of co-culture with T cells (1:5) and 10-fold serial dilutions of bispecific T cell activator-containing supernatants. Cytotoxicity was assessed by LDH release into culture supernatants. (D) NHDF lysis as assessed by LDH release (left) and CD25 induction on T cells (right) were assessed after 24 h co-culture with PBMC-derived T cells (1:5) from six healthy donors and bispecific T cell activator-containing supernatants.

Fig. 70. EnAd expressing a bispecific FAP T cell activator selectively kills FAP+ fibroblasts and reduces TGFb in peritoneal ascites samples. (A, B) Numbers of FAP+ fibroblasts (A) and EpCAM+ tumor cells (B) after 72 h of culture with PBMC-derived T cells and EnAd or recombinant viruses. Ascites cells were first isolated from three patients with ascites and expanded ex vivo. Cell numbers were measured 72 h post-infection by flow cytometry. (C) Induction of the activation marker CD25 on PBMC-derived CD3 cells from (A) was measured 72 h post-infection. (D) TGFb levels were measured by ELISA using supernatants collected from (A).

Fig. 71. Activation of endogenous tumor-associated T cells and associated FAP+ cell death in ascites biopsy specimens from a patient's malignancy by FAP bispecific T cell activator protein and EnAd-FAP Bisp.act.T-cell viruses. (A) T cell activation measured by CD25 expression. (B) Residual FAP+ cell counts measured by flow cytometry.

Fig. 72. Effect of PD-L1 blocking antibodies on bispecific T cell activator-mediated T cell activation in a patient sample. (A) Expression of PD1 by endogenous T cells and PD-L1 on FAP+ cells after their initial isolation from peritoneal ascites was assessed by flow cytometry. (B) Unpurified total cells from peritoneal ascites were incubated in 50% fluid from the same exudate in the presence of free bispecific T cell activator, EnAd, or recombinant virus with or without blocking anti-PD-L1 antibody. After 2 days, the total cell population was harvested and the number of CD25+ T cells was quantified by flow cytometry. (C) The amount of gamma interferon in culture supernatants from (B, D) was measured by ELISA. (D) The number of residual FAP+ cells in (B) was measured by flow cytometry.

Fig. 73. EnAd expressing a bispecific T cell activator activates and redirects T cells from patient biopsy specimens to lyse NHDF fibroblasts. (A) PD-1 expression by endogenous T cells after isolation from healthy donors or cancer exudate biopsy specimens. PD-1 expression was measured by flow cytometry. (B) Proportion of CD3+ cells in the crude cell population of PBMCs and cancer biopsy specimens measured by flow cytometry. (C) Interferon-gamma levels measured by ELISA in culture supernatants collected from (B) 120 h after treatment. (D) NHDF fibroblast viability was monitored in real time for 130 h using the xCELLigence cytotoxicity assay in co-culture with PBMC or total cancer biopsy cells (1:5) and bispecific T cell activator-containing supernatant.

Fig. 74. Effect of immunosuppressive ascites fluid samples on FAP bispecific T cell activator and anti-CD3/CD28 Dynabeads-mediated activation of PBMC T cells. (A) PBMC T cells activated by anti-CD3/CD28 Dynabeads. (B) PBMC T cells activated by control or FAP bispecific T cell activator in the presence of NHDF cells. NS: normal serum, A: peritoneal ascites.

Fig. 75. EnAd expressing bispecific T cell activator to FAP polarizes CD11b+ macrophages from patients with ascites toward a more inflammatory phenotype. (A) Unpurified total cells from an ascites sample were incubated in 50% ascites fluid in the presence of free bispecific T cell activator or virus expressing bispecific T cell activator. Treatment with gamma interferon was used as a positive control. After 3 days, the total cell population was harvested and the induction of the activation marker CD25 on CD3+ cells was measured by flow cytometry. (B) Gamma interferon levels in culture supernatants from (A) were measured by ELISA. (C) After 3 days, the expression levels of CD68, CD86, CD206, and CD163 on CD11b+ cells from (A) were measured by flow cytometry. Representative flow cytometry spectra from triplicates are shown, along with the full data set.

Fig. 76. Characterization of the structure and cellular composition of solid prostate tumor. (A) EpCAM staining, (B) CD8 staining, (C) FAP staining. (D) Representative immunohistochemical images of CD25 induction in prostate tumor sections after treatment with viruses expressing a bispecific T cell activator. Tumor biopsies were cut at 300 μm thickness with a Leica vibratome, cultured and infected in inserts, and collected after 7 days of treatment. (E) IFNg levels in tissue section culture medium measured by ELISA. Supernatants were collected from malignant and benign tissue culture sections at the indicated time points. (F) IL-2 levels in malignant and benign tissue culture medium measured by ELISA.

Fig. 77A-C. Schematic representation of the transgene cassettes used in Example 33.

Fig. 77D. Number of viral genomes detected per cell in tumor cells treated with NG-611, NG-612, and NG-617.

Fig. 78. Percentage of T cells expressing CD69 (a), CD25 (b), HLA-DR (c), CD40L (d), or cell surface CD107a (e) after co-culture with EpCam expressing SKOV cells and supernatants collected from A549 cells 24, 48, or 72 h after treatment with NG-611 viral particles compared to NG-612, enadenotucirev, or untreated control supernatants.

Fig. 79. Percentage of T cells expressing CD69 (a), CD25 (b), HLA-DR (c), CD40L (d), or cell surface CD107a (e) after co-culture with FAP expressing MRC-5 cells and supernatants collected from A549 cells 24, 48, or 72 h after treatment with NG-612 viral particles compared to NG-611, enadenotucirev, or untreated control supernatants.

Fig. 80. Percentage of MRC-5 cells that express EpCAM and FAP.

Fig. 81. IFNγ expression in supernatants of T cell cocultures with SKOV (A) or MRC-5 (B) cells incubated with supernatants collected from A549 cells 24, 48, or 72 h after treatment with NG-611, NG-612, or enadenotucirev viral particles, or untreated control supernatants.

Fig. 82. Antitumor efficacy and immune activation of bispecific T cell activator expressing viruses in vivo, (a) tumor volume in mice treated with saline, enadenotucirev or NG-611. (b) CD8 to CD4 T cell ratio in NG-611 treated tumors compared to enadenotucirev virus treated or untreated controls.

Fig. 83. Schematic representation of transgene cassettes, (a) NG-615, (b) NG-640, (c) NG-641.

Fig. 84. Number of viral genomes detected per cell in tumor cells treated with NG-612 and NG-615.

Fig. 85. Expression of IFNa, MIP1a and Flt3L in NG-615 cell supernatant compared to supernatant of enadenotucirev virus-treated and untreated control tumor cells.

Fig. 86. Number of T cells expressing CD69 (a), CD25 (b), HLA-DR (c), CD40L (d), or cell surface CD107a (e) after co-culture with FAP-expressing MRC-5 cells and supernatant collected from A549 cells 24, 48, or 72 h after treatment with NG-615 viral particles compared to NG-612, enadenotucirev, or untreated control supernatants.

Fig. 87. IFNy expression in supernatants of co-culture of T cells with MRC-48 043895 cells

5, incubated with supernatants collected from A549 cells 24, 48 or 72 h after treatment with NG-612, NG-615 or enadenotucirev viral particles or untreated control supernatants.

Fig. 88. Schematic representation of the NG-618 transgene cassette.

Fig. 89. Detection of FAP surface expression on MRC-5 cells (a) or EpCam expression on SKOV cells (b) after incubation with supernatants collected from A549 cells 72 h after treatment with NG-611, NG-612, NG-615 or enadenotucirev viral particles.

Fig. 90. Percentage of T cells expressing cell surface CD24 (a), CD40L (b), or CD107a (c) after co-culture with FAP-expressing MRC-5 cells and supernatants collected from A549 cells 72 h after treatment with NG-618 viral particles compared to enadenotucirev or untreated controls.

Fig. 91. Percentage of T cells expressing cell surface CD24 (a), CD40L (b), or CD107a (c) after co-culture with EpCam-expressing SKOV cells and supernatants collected from A549 cells 72 h after treatment with NG-618 viral particles, compared to enadenotucirev or untreated controls.

Fig. 92. Percentage of MRC-5 (a) or SKOV (b) cells killed after co-culture with T cells and supernatants collected from A549 cells for 72 h after treatment with NG-618 viral particles, compared to enadenotucirev or untreated controls.

Sequences

SEQ ID NO: 1: DNA coding sequence of anti-EpCAM bispecific T cell activator with N-terminal signal sequence and C-terminal deca-His affinity tag.

SEQ ID NO: 2: anti-EpCAM bispecific T cell activator protein sequence with N-terminal signal sequence and C-terminal deca-His affinity tag.

SEQ ID NO:3: DNA coding sequence of anti-FAP bispecific T cell activator with N-terminal signal sequence and C-terminal deca-His affinity tag.

SEQ ID NO: 4: Amino acid sequence of anti-FAP bispecific T cell activator, with N-terminal signal sequence and C-terminal deca-His affinity tag.

SEQ ID NO: 5: DNA coding sequence of the control (anti-FHA) bispecific T cell activator, with N-terminal signal sequence and C-terminal affinity deca-His tag.

SEQ ID NO: 6: Amino acid sequence of control (anti-FHA) bispecific T cell activator with N-terminal signal sequence and C-terminal affinity deca-His tag.

SEQ ID NO: 7: Amino acid sequence of anti-CD3 ScFv.

SEQ ID NO: 8: VH anti-CD3.

SEQ ID NO: 9: VL anti-CD3.

SEQ ID NO: 10: anti-CD3 ScFv linker sequence.

SEQ ID NO: 11: Anti-FAP ScFv.

SEQ ID NO: 12: VL domain of anti-FAP.

SEQ ID NO: 13: Anti-FAP VH domain.

SEQ ID NO: 14: anti-FAP and anti-EpCAM linker sequence.

SEQ ID NO: 15: leader sequence of bispecific T cell activator.

SEQ ID NO: 16: anti-EpCAM ScFv.

SEQ ID NO: 17: VL anti-EpCAM.

SEQ ID NO: 18: VH anti-EpCAM.

SEQ ID NO:19: control bispecific T cell activator (anti-FHA).

SEQ ID NO:20: Control (anti-FHA) ScFv.

SEQ ID NO:21: Control (anti-FHA) VL.

SEQ ID NO:22: Control (anti-FHA) VH.

SEQ ID NO:23: Linker sequence of control (anti-FHA) ScFv.

SEQ ID NO:24: deca-His tag sequence.

SEQ ID NO:25: FAP bispecific T cell activator-P2A-RFP [ITALIC = leader, BOLD = furin cleavage site, UNDERLINE = P2A sequence, lowercase = RFP).

SEQ ID NO:26: Control (anti-FHA) bispecific T cell activator-P2A-RFP [ITALIC = leader, BOLD = furin cleavage site, UNDERLINE = P2A sequence, lowercase = RFP).

SEQ ID NO: 27: coding sequence of human EpCAM DNA.

SEQ ID NO: 28: Amino acid sequence of human EpCAM.

SEQ ID NO:29: DNA coding sequence of human FAP.

SEQ ID NO: 30: Amino acid sequence of human FAP.

SEQ ID NO:31: CMV promoter sequence.

SEQ ID NO:32: SV40 late polyadenylation sequence.

- 49 043895

SEQ ID NO:33: null sequence.

SEQ ID NO:34: NG-601 (EnAd-CMV-EpCAM bispecific T cell activator).

SEQ ID NO: 35: NG-602 (EnAd-SA-EpCAM bispecific T cell activator).

SEQ ID NO:36: NG-605 (EnAd-CMV-FAP bispecific T cell activator).

SEQ ID NO:37: NG-606 (EnAd-SA-FAP bispecific T cell activator).

SEQ ID NO: 38: EnAd genome.

SEQ ID NO:39: BX DNA sequence corresponding to and including bp 28166-28366 of the EnAd genome.

SEQ ID NO: 40: BY DNA sequence corresponding to and including bp 29345-29379 of the EnAd genome.

SEQ ID NO: 41: HIS tag.

SEQ ID NO:42: splice acceptor sequence.

SEQ ID NO:43: SV40 polyadenylation sequence.

SEQ ID NO:44: Nucleotide sequence of bispecific T cell activator to EpCam (OKT3).

SEQ ID NO: 45: Nucleotide sequence of bispecific T cell activator to FAP (OKT3).

SEQ ID NO:46: Nucleotide sequence of bispecific T cell activator to FAP (aCD3).

SEQ ID NO: 47: NG-611 transgene cassette.

SEQ ID NO: 48: NG-612 transgene cassette.

SEQ ID NO: 49: NG-613 transgene cassette.

SEQ ID NO:50: restriction site insertion (BX).

SEQ ID NO:51: restriction site insertion (BY).

SEQ ID NO:52: CMV promoter sequence.

SEQ ID NO: 53: PGK promoter sequence.

SEQ ID NO: 54: CBA promoter sequence.

SEQ ID NO:55: short splice acceptor (SSA) DNA sequence.

SEQ ID NO:56: DNA sequence of splice acceptor (SA).

SEQ ID NO:57: branched splice acceptor DNA sequence (bSA).

SEQ ID NO:58: Kozak sequence (Zero sequence).

SEQ ID NO:59: Example of a start codon.

SEQ ID NO:60: internal ribosome entry site (IRES).

SEQ ID NO:61: Peptide P2A.

SEQ ID NO: 62: F2A peptide.

SEQ ID NO: 63: E2A peptide.

SEQ ID NO: 64: T2A peptide.

SEQ ID NO:65: polyadenylation sequence (polyA).

SEQ ID NO:66: leader sequence.

SEQ ID NO:67: leader sequence.

SEQ ID NO: 68: amino acid sequence of IFNF.

SEQ ID NO: 69: amino acid sequence of IFNa.

SEQ ID NO: 70: amino acid sequence of TNFa.

SEQ ID NO: 71: DNA sequence corresponding to the E2B region of the EnAd genome (bp 103555068).

SEQ ID NO: 72: DNA coding sequence of anti-EpCAM bispecific T cell activator with N-terminal signal sequence and C-terminal deca-His affinity tag.

SEQ ID NO: 73: anti-EpCAM bispecific T cell activator protein sequence with N-terminal signal sequence without C-terminal deca-His affinity tag.

SEQ ID NO: 74: DNA coding sequence of anti-FAP bispecific T cell activator, with N-terminal signal sequence without C-terminal deca-His affinity tag.

SEQ ID NO: 75: Amino acid sequence of anti-FAP bispecific T cell activator, with N-terminal signal sequence without C-terminal deca-His affinity tag.

SEQ ID NO: 76: DNA coding sequence of control (anti-FHA) bispecific T cell activator, with N-terminal signal sequence without C-terminal decaHis affinity tag.

SEQ ID NO: 77: Amino acid sequence of control (anti-FHA) bispecific T cell activator with N-terminal signal sequence without C-terminal affinity deca-His tag.

SEQ ID NO:78: Control bispecific T cell activator (anti-FHA) without C-terminal deca-His affinity tag.

- 50 043895

SEQ ID NO: 79: NG-601 (EnAd-CMV-EpCAM bispecific T cell activator) without affinity deca-His-MeTKu.

SEQ ID NO: 80: NG-602 (EnAd-SA-EpCAM bispecific T cell activator) without deca-His affinity tag.

SEQ ID NO: 81: NG-605 (EnAd-CMV-FAP bispecific T cell activator) without decaHis affinity tag.

SEQ ID NO:82: NG-606 (EnAd-SA-FAP bispecific T cell activator) without decaHis affinity tag.

SEQ ID NO: 83: Nucleotide sequence of bispecific T cell activator to EpCam (OKT3).

SEQ ID NO:84: null sequence.

SEQ ID NO: 85: Nucleotide sequence of bispecific T cell activator to FAP (OKT3).

SEQ ID NO:86: null sequence.

SEQ ID NO: 87: Nucleotide sequence of bispecific T cell activator to FAP (aCD3).

SEQ ID NO: 88: NG-611 transgene cassette.

SEQ ID NO: 89: NG-612 transgene cassette.

SEQ ID NO: 90: NG-613 transgene cassette.

SEQ ID NO: 91: NG-614 transgene cassette.

SEQ ID NO: 92: NG-617 transgene cassette.

SEQ ID NO: 93: Amino acid sequence of bispecific T cell activator to EpCam (OKT3).

SEQ ID NO: 94: Amino acid sequence of bispecific T cell activator to FAP (OKT3).

SEQ ID NO: 95: Amino acid sequence of bispecific T cell activator to FAP (aCD3).

SEQ ID NO: 96: NG-611 genome.

SEQ ID NO: 97: NG-612 genome.

SEQ ID NO: 98: NG-613 genome.

SEQ ID NO: 99: NG-614 genome.

SEQ ID NO: 100: NG-617 genome.

SEQ ID NO: 101: NG-615 genome.

SEQ ID NO: 102: NG-640 genome.

SEQ ID NO: 103: NG-641 genome.

SEQ ID NO: 104: null sequence.

SEQ ID NO: 105: Nucleotide sequence of Flt3L.

SEQ ID NO: 106: null sequence.

SEQ ID NO: 107: nucleotide sequence of MIP1a.

SEQ ID NO: 108: Flexible linker sequence.

SEQ ID NO: 109: nucleotide sequence of IFNa.

SEQ ID NO: 110: nucleotide sequence of CXCL10.

SEQ ID NO: 111: nucleotide sequence of CXCL9.

SEQ ID NO: 112: NG-615 transgene cassette.

SEQ ID NO: 113: NG-640 transgene cassette.

SEQ ID NO: 114: NG-641 transgene cassette.

SEQ ID NO: 115: amino acid sequence of FLT3L.

SEQ ID NO: 116: amino acid sequence of MIP1a.

SEQ ID NO: 117: amino acid sequence of IFNa.

SEQ ID NO: 118: amino acid sequence of CXCL9.

SEQ ID NO: 119: amino acid sequence of CXCL10.

SEQ ID NO: 120: NG-618 genome.

SEQ ID NO: 121: Nucleotide sequence of the bispecific T cell activator EpCam NG-618.

SEQ ID NO: 122: Nucleotide sequence of the bispecific T cell activator to FAP NG-618.

SEQ ID NO: 123: NG-618 transgene cassette.

SEQ ID NO: 124-297 are linker sequences.

SEQ ID NO: 298: NG-616 genome.

- 51 043895

Examples

Example 1.

Recombinant bispecific T cell activators were constructed and proteins produced as described in this example.

Construction of a bispecific T cell activator.

The bispecific T cell activator was constructed by joining two single chain antibody fragments (ScFv) of different specificities with a flexible Gly4Ser linker. The ScFv fragments were constructed by joining the VH and VL domains of the parent monoclonal antibodies using a linker. Each bispecific T cell activator was engineered with an N-terminal signal sequence for secretion by mammalian cells and a C-terminal decahistidine affinity tag for detection and purification. The bispecific T cell activators were constructed by standard DNA cloning techniques and inserted into protein expression vectors (Fig. 1). The anti-EpCAM bispecific T cell activator is referred to in patent WO 2005040220 (SEQ ID NO: 63) with the addition of a signal sequence and affinity tag. An anti-FAP bispecific T cell activator was generated de novo using the anti-FAP ScFv from WO2010037835A2 and the anti-CD3 ScFv from WO2005040220 (SEQ ID 63 therein) with the addition of a signal sequence and an affinity tag. Anti-FHA (filamentous hemagglutinin from Bordetella pertussis) ScFv from Hussein et al., 2007 (Hussein AH et al. (2007) Construction and characterization of single-chain variable fragment antibodies directed against the Bordetella pertussis surface adhesins filamentous hemagglutinin and pertactin. Infect Immunity 75, 5476-5482) and anti-CD3 ScFv from WO 2005040220 (SEQ ID NO: 63 therein) with the addition of a signal sequence and an affinity tag were used as control bispecific T cell activators. The coding DNA and amino acid sequences for these bispecific T cell activators correspond to SEQ ID NOs: 1-6. Production of Recombinant Bispecific T Cell Activator Recombinant bispecific T cell activator proteins were produced by cloning the corresponding sequences into the pSF-CMV vector using the CMV promoter (SEQ ID NO: 31) to drive protein expression (Fig. 1). The plasmid DNA concentration for the plasmids, pSF-CMV-EpCAM bispecific T cell activator, pSF-CMV-FAP bispecific T cell activator, and pSF-CMV-Control bispecific T cell activator (Table 2), were measured using a NanoDrop instrument. Empty pSF-CMV vector was included as a negative control. 54.7 μg of each was diluted in 4 ml OptiMEM. 109.2 μg PEI (linear, MW 25000, Polysciences, USA) was diluted in 4 ml OptiMEM medium and mixed with 4 ml of diluted DNA to obtain DNA-PEI complexes (DNA:PEI ratio 1:2 (w/w)). After incubation at room temperature for 20 min, this complex mixture was made up to 18 ml with OptiMEM medium, and this transfection mixture was added to a T175 flask containing Ad293 cells at 90% confluence. After incubation of the cells with the transfection mixture for 4 h at 37°C, 5% _CO2 , 30 ml of cell medium (high glucose DMEM supplemented with glutamine, without phenol red) were added to the cells and the flasks were incubated at 37°C, 5% CO2 for 48 h. Another flask of cells was transfected in parallel with the pSF-CMV-GFP mixture to ensure efficient transfection efficiency. To collect the secreted protein, the supernatant of the transfected cells was collected and centrifuged at 350 g at 4°C for 5 min to remove cellular components (Allegra X-15R, Beckman Coulter). The supernatants were transferred to 10 k Amicon Ultra-15 MWCO centrifugal filter units (Millipore). After centrifugation at 4750 rpm and 4°C, the retentate volume was adjusted by flow filtration to obtain a 50-fold higher concentration. Aliquots of the concentrated protein were stored at -80°C.

Table 2

The prefix p in the notation indicates that the construct is a plasmid.

Plasmid ID Coding sequence SEQ ID NO: [plasmid DNA] ng/ml pSF-CMV- EpSAMB-specific T-cell activator SEQ ID NO: 1 3717 pSF-CMV-EARB-specific T-cell activator SEQ ID NO: 3 6700 pSF-CMV- ConPo1Bi-specific T-cell activator SEQ ID NO: 5 5300 pSF-Lenti-EpCAM SEQ ID NO: 27 2529.3 pSF-Lenti-FAP SEQ ID NO: 29 659.6

- 52 043895

Detection of a recombinant bispecific T cell activator.

To detect the bispecific T cell activator, the C-terminal deca-histidine affinity tag can be probed with anti-His antibody using Western blotting. Protein samples were adjusted to a final volume of 15 μl with lysis buffer including 2.5 μl of 6x Laemmli SDS sample buffer containing p-mercaptoethanol and SDS. Samples were incubated for 5 min at 95°C to denature proteins and loaded onto 15-well 10% precast polyacrylamide gels (Mini-PROTEAN TGX Precast Gels, BioRad, UK). Gels were run at 180 V for 45 min in 1x running buffer on a Mini-PROTEAN Tetra System (BioRad, UK). Proteins from SDS gels were transferred to nitrocellulose membranes by wet electroblotting at 300 mA and 4°C for 90 min in 1x transfer buffer in a Mini Trans-Blot Cell (BioRad, UK). Transfer was performed in the presence of an ice pack to limit heating. The nitrocellulose membrane was then blocked with 5% milk in PBS-T on a shaker for 1 h at room temperature and probed with anti-His antibody (C-terminal) (mouse a-6xHis, clone 3D5, Invitrogen, UK, #46-0693) diluted 1:5000 in PBS/5% milk. After overnight incubation on a shaker at 4°C, the membrane was washed and probed with a polyclonal α-mouse immunoglobulin secondary antibody labeled with HRP (1:10,000 in PBS/5% milk, Dako, #P0161) for 1 h at room temperature. Super Dignal West Dura Extended Duration Substrate (Thermo Fisher Scientific, UK) was used for visualization following the manufacturer's instructions, with exposure to X-ray film and development in an automated film processor. The results demonstrated expression and secretion of bispecific T cell activator protein from Ad293 cells transfected with bispecific T cell activator expression plasmids but not with the parental vector. Quantification of recombinant bispecific T cell activator.

To measure the amount of recombinant bispecific T cell activator protein, a dot blot assay was used to compare the bispecific T cell activator signal with a His-tagged (C-terminal 10His) protein standard (10χHis-tagged human cathepsin D, Biolegend, #556704). Two-fold serial dilutions of bispecific T cell activator and protein standard samples were prepared, and 1.5 μl of each were directly loaded onto a nitrocellulose membrane and air dried for 20 min. The blocking and staining protocol described above for Western blotting was then performed. The molar concentration of the protein standard was adjusted to provide a bispecific T cell activator concentration of 250 μg/ml. The results (Fig. 2A) demonstrated the expression and secretion of bispecific T cell activator protein from Ad293 cells transfected with bispecific T cell activator expression plasmids. FAP binding ELISA assay.

FAP-binding activity of the FAP bispecific T cell activator and control (anti-FHA) bispecific T cell activator (SEQ ID NOs: 4 and 6) secreted from cells transfected with pSF-CMV-FAP bispecific T cell activator or pSF-CMV-Control bispecific T cell activator were assessed by enzyme-linked immunosorbent assay (ELISA). Supernatants of empty pSF-CMV vector were included as a negative control. ELISA plates (Nunc Immuno MaxiSorp 96-well microplate) were prepared by coating overnight at 4°C with human FAP/sepase protein (100 ng/well, Sino Biological Inc, No. 10464-H07H-10) in PBS buffer. Between all subsequent binding steps, the plates were washed with PBS 0.05% Tween 20. The plates were blocked for 1 h at room temperature with 5% BSA in PBS 0.05% Tween 20. Aliquots of the bispecific T cell activator protein or the protein harvested from pSF-CMV empty vector-transfected wells were diluted 10-fold in PBS/5% BSA/0.05% Tween 20. All samples were added to the FAP-coated plates and incubated for 2 h at room temperature. The detection antibody, anti-His (C-terminus) antibody (mouse anti-6xHis, clone 3D5, Invitrogen, UK, #46-0693), was diluted 1:1000 and used for 1 h at room temperature. HRP-conjugated anti-mouse Fc (1:1000 in PBS/5% milk, Dako) was then applied for 1 h at room temperature, followed by HRP detection using the HRP substrate solution, 3.3.5.5'tetramethylethylenediamine (TMB, Thermo-Fisher). Stop solution was used to stop the reaction, and the color developed was measured at 450 nm in a plate reader. The absorbance at 450 nm was plotted for FAP bispecific T cell activator, control bispecific T cell activator, and empty vector supernatants, demonstrating specific binding of FAP bispecific T cell activator to FAP protein. The results (Fig. 2B) show specific binding of FAP bispecific T cell activator but not control bispecific T cell activator to recombinant FAP protein. EpCAM binding ELISA assay.

The EpCAM-binding activity of the EpCAM bispecific T cell activator and the Control bispecific T cell activator (SEQ ID NOs: 2 and 6) secreted from cells transfected with pSF-CMV-EpCAM bispecific T cell activator or pSF-CMV-Control bispecific T cell activator was assessed by enzyme-linked immunosorbent assay (ELISA). Empty pSF-CMV vector supernatants were included as negative controls. ELISA plates (96-well microplate

- 53 043895 Nunc Immuno MaxiSorp plate) was prepared by coating overnight at 4°C with human EpCAM/TROP-1 protein (50 ng/well, Sino Biological Inc, No. 10694-H02H-50) in PBS buffer. Between all subsequent binding steps, the plates were washed with PBS 0.05% Tween 20. The plates were blocked for 1 h at room temperature with 5% BSA in PBS 0.05% Tween 20. Aliquots of the bispecific T cell activator protein or the protein harvested from pSF-CMV empty vector-transfected wells were diluted 10-fold in PBS/5% BSA/0.05% Tween 20. All samples were added to EpCAM-coated plates and incubated for 2 h at room temperature. The detection antibody, anti-His (C-terminus) antibody (mouse anti-6xHis, clone 3D5, Invitrogen, UK, #46-0693), was diluted 1:5000 and applied for 1 h at room temperature. HRP-conjugated anti-mouse Fc (1:5000 in PBS/5% milk, Dako) was then applied for 1 h at room temperature, followed by HRP detection using the HRP substrate solution, 3,3,5,5'-tetramethylethylenediamine (TMB, Thermo-Fisher). Stop solution was used to stop the reaction, and the color developed was measured at 450 nm in a plate reader. The absorbance at 450 nm was plotted for the EpCAM bispecific T cell activator, control bispecific T cell activator, and empty vector supernatants, demonstrating specific binding of the EpCAM bispecific T cell activator to recombinant EpCAM. The results (Fig. 2C) show specific binding of the bispecific T cell activator to EpTEAM, but not of the control bispecific T cell activator to recombinant EpCAM protein.

Example 2.

The functional activity of recombinant bispecific T cell activator proteins was assessed in a number of different assays prior to the construction of transgene-containing bispecific T cell activator EnAd viruses.

Isolation of human peripheral blood mononuclear cells (PBMCs) Human PBMCs were isolated by density gradient centrifugation from either fresh human blood samples from healthy donors or from whole blood leukocyte concentrates obtained from NHS Blood and Transplant UK, Oxford. In each case, samples were diluted 1:2 in PBS and 25 ml of this mixture was layered onto 13 ml Ficoll (1.079 g/ml, Ficoll-Paque Plus, GE Healthcare) in a 50 ml Falcon tube. Samples were centrifuged (Allegra X-15R, Beckman Coulter) at 1600 rpm for 30 min at 22°C with the lowest deceleration setting to maintain phase separation. After centrifugation, four layers could be observed, comprising a plasma layer at the top, followed by an interface containing PBMCs, a Ficoll layer, and a layer of red blood cells and granulocytes at the bottom. PBMCs were collected using a Pasteur pipette and washed twice with PBS (1200 rpm for 10 min at room temperature) and resuspended in RPMI medium supplemented with 10% FBS.

Isolation of CD3-positive T cells.

CD3-positive (CD3+) T cells were extracted from PBMCs by depletion of non-CD3 cells using the Pan T Cell Isolation Kit (Miltenyi Biotec, #130-096-535) according to the manufacturer's protocol.

Processing of primary ascites samples.

Primary human ascites samples were obtained from the Oncology Unit of Churchill Hospital (Oxford University Hospitals) from patients with multiple indications including, but not limited to, ovarian, pancreatic, breast and gastric cancer. Upon receipt, cellular and fluid fractions were separated and fluid aliquots were frozen at -20°C for storage and subsequent analysis. The cellular fraction was treated with red blood cell lysis buffer (Roche, #11814389001) to remove red blood cells following the manufacturer's instructions. The cell types present in each sample were determined by staining for EpCAM, EGFR, FAP, CD45, CD11b, CD56, CD3, CD4, CD8, PD1 and CTLA4 and analyzed by flow cytometry. Cells were then used fresh for ex vivo T cell activation experiments and target cell lysis experiments. In some cases, cells were passaged in DMEM supplemented with 10% FBS for use in later experiments.

Maintenance of cell line.

All cell lines were maintained in DMEM (Sigma-Aldrich, UK) or RPMI (Sigma-Aldrich, UK) as indicated in Table 3 supplemented with 10% (v/v) fetal bovine serum (FBS, GibcoTM) and 1% (v/v) penicillin/streptomycin (10 mg/ml, Sigma-Aldrich, UK) in a humidified incubator (MCO-17AIC, Sanyo) at 37°C and 5% _CO2 unless otherwise stated. Cells were split every 2-3 days until confluency was achieved by enzymatic dissociation with trypsin/EDTA (0.05% trypsin, 0.02% EDTA, Sigma-Aldrich, UK). In this process, the culture medium was aspirated and the cells were washed with 15 ml PBS, and then the cells were treated with 2 ml trypsin/EDTA for 2-10 min at 37°C. Trypsin was neutralized with 10 ml DMEM medium containing 10% FBS, and a portion of the cells were transferred to new flasks containing fresh medium. For normal cell culture, 10% FBS was added to the medium, 2% FBS was added for viral plasmid infection and transfections, and FBS was not added for recombinant bispecific T cell activator plasmid transfections.

- 54 043895

Table 3

Cell line Origin of cells Cultural environment Source Ascites derived cell lines Primary human ascites DMEM NHS Blood & Transplant UK ВТС100 Human lung cancer-associated primary fibroblasts (CAF) DMEM Oxford University SNO-K1 Chinese hamster ovary cells, adhesive RPMI ATCC Stable cell lines CHO-K1 Chinese hamster ovary cells, adhesive RPMI DLD1 Human colorectal adenocarcinoma RPMI ATCC HEK293A Human embryonic kidney cells, adhesive DMEM ATCC Stable cell lines HEK293A Human embryonic kidney cells, adhesive DMEM HEK293T Human embryonic kidney cells, adhesive DMEM ATCC MCF-7 Human, mammary gland, breast, adhesive DMEM ATCC Normal human dermal fibroblasts (NHDF) Normal primary dermal fibroblasts of an adult DMEM ATCC SKOV3 Human ovarian adenocarcinoma DMEM ATCC

Statistics.

In cases where two conditions were compared, statistical analysis was performed using the t-test. In all other cases, statistical analysis was performed using one-way analysis of variance.

Characterization of human T cell activation by a recombinant bispecific T cell activator to FAP.

The ability of a bispecific T cell activator to FAP to induce T cell activation in the presence or absence of normal human dermal fibroblast (NHDF) cells was compared. Human CD3+ T cells (70,000 cells/well in 96-well U-bottom plates) were cocultured alone or with NHDF cells (10:1 T:NHDF) in the presence of medium alone or 300 ng/ml FAP or control bispecific T cell activator. Cells were cocultured for 24 h at 37°C and then harvested with enzyme-free cell dissociation buffer (Thermo, #13151014). Expression levels of CD69 (Fig. 3A) and CD25 (Fig. 3B) on CD45+ T cells were then analyzed by antibody staining and flow cytometry and data were presented as geometric mean fluorescence intensity (gMFI). Plate-coated anti-CD3 antibody (7.5 μg/ml) was used as a positive control for T cell activation. The FAP bispecific T cell activator selectively induced expression of the activation markers CD69 and CD25 on T cells, indicating that it is capable of activating T cells.

In a second similar experiment, T cells were assessed by intracellular cytokine staining 6 h after co-culture with NHDF cells (200,000 CD3+ cells plus 40,000 NHDF in a 96-well plate) and 300 ng/ml FAP or control bispecific antibody.

- 55 043895 T cell activator. CD45+ T cells were intracellularly stained for IFNy expression using brefeldin A added to the culture medium 5 h before harvest. As a positive control, T cells were stimulated with soluble PMA (10 ng/ml) and ionomycin (1 μg/ml). The results shown in Fig. 4A indicate that the FAP bispecific T cell activator in the presence of NHDF resulted in significantly more IFNy-expressing T cells compared to the control bispecific T cell activator.

Example 3.

A series of experiments similar to those in Example 2 were conducted to characterize the recombinant EpCAM bispecific T cell activator protein. Characterization of Human T Cell Activation by the Recombinant EpCAM Bispecific T Cell Activator.

The ability of a bispecific T cell activator to EpCAM to induce T cell activation in the presence or absence of the EpCAM-positive cell line DLD was compared. Human CD3+ T cells (70,000 cells/well in 96-well U-bottom plates) were cocultured alone or with DLD cells (10:1 TtDLD) in the presence of medium alone or 600 ng/ml EpCAM or control bispecific T cell activator. Cells were cocultured for 24 h at 37°C and then harvested with enzyme-free cell dissociation buffer. CD69 and CD25 expression levels on CD45+ T cells were then analyzed by antibody staining and flow cytometry, and data were presented as geometric mean fluorescence intensity (gMFI). Plate-immobilized anti-CD3 antibody (7.5 μg/ml) was used as a positive control for T cell activation. The bispecific T cell activator to EpCAM selectively induced the expression of the activation markers CD69 and CD25 on T cells, indicating that it is capable of activating T cells (Fig. 4B and C).

In a similar experiment, T cells were assessed by intracellular cytokine staining 6 h after coculture with DLD cells (200,000 CD3+ T cells plus 40,000 DLD cells per well of a 96-well plate) and 300 ng/ml EpCAM or control bispecific T cell activator. CD45+ T cells were intracellularly stained for IFNy expression using brefeldin A added to the culture medium 5 h before harvest. As a positive control, T cells were stimulated with soluble PMA (10 ng/ml) and ionomycin (1 μg/ml). The results showed that the bispecific T cell activator to EpCAM in the presence of DLD resulted in significantly more IFNγ-expressing T cells compared to the control bispecific T cell activator (Fig. 5A).

In another similar experiment, PBMCs from 8 different blood donors were used to assess donor-dependent changes in bispecific T cell activator-mediated T cell activation. DLD (7,000 cells) were cocultured with 100,000 PBMCs in a 96-well U-bottom plate in the presence of medium alone or 300 ng/ml control or bispecific T cell activator to EpCAM. Cells were cocultured for 24 h at 37°C and then harvested. CD69 and CD25 expression levels on CD45+ T cells were then analyzed by antibody staining and flow cytometry, and data were presented as geometric mean fluorescence intensity (gMFI). The results showed that the bispecific T cell activator to EpCAM induced the expression of the activation markers CD69 and CD25 in CD3+ T cells from all 8 donors (Fig. 5B and C).

Example 4.

In this example, the ability of recombinant T cells activated with a bispecific T cell activator to FAP to induce cell death of fibroblast target cells was assessed.

FAP bispecific T cell activator induces T cell-mediated lysis of FAPpositive cell lines and primary NHDF cells (7,000 cells) were cocultured with 70,000 T cells in wells of a 96-well U-bottom plate in the presence of medium alone or 300 ng/ml control or FAP bispecific T cell activator. After 24 h of coculture, supernatants were collected and cytotoxicity was determined by LDH assay following the manufacturer's instructions. The results in Fig. 6A show that FAP bispecific T cell activator significantly increased NHDF cell lysis. In a similar experiment, 7,000 primary lung fibroblast cells (BTC100) were cocultured with 70,000 CD3+ T cells with or without 300 ng/ml control or FAP bispecific T cell activator. After 24 h of coculture, supernatants were collected and cytotoxicity was determined by LDH assay. The results in Fig. 6B and C show that FAP bispecific T cell activator significantly increased cell lysis of primary human cancer-associated fibroblasts (CAF). FAP expression by these and other patient-derived cell lines is shown in Fig. 7.

The dose-response relationship for FAP bispecific T cell activator-mediated cell lysis was assessed by co-culturing 8,000 NHDF cells with 40,000 T cells and bispecific T cell activator concentrations ranging from 2x103 to 2x10-2 ng/mL. After co-cultivation for 24 h at 37°C, LDH assay was performed on supernatants to determine target cell cytotoxicity. Dose-response curves were fitted using a four-parameter

- 56 043895 nonlinear fitting model integrated in GraphPad Prism yielded an EC50 value for the FAP bispecific T cell activator of 3.2 ng/mL. The results (Fig. 8A) show a dose-response relationship between the FAP bispecific T cell activator concentration and cytotoxicity measured by the LDH assay (shown as Abs ₄₉₀ ).

Example 5.

Studies similar to those in Example 4 were used to demonstrate the ability of T cells activated with a recombinant bispecific T cell activator to EpCAM to induce cell death of target tumor cells. The bispecific T cell activator to EpCAM induces T cell-mediated lysis of EpCAM-positive cell lines.

DLD tumor cells (7,000 cells) were co-cultured with 70,000 T cells in wells of a 96-well U-bottom plate in the presence of medium alone or 300 ng/ml control or EpCAM bispecific T cell activator. After 24 h of co-culture, supernatants were collected and cytotoxicity was determined by LDH assay. The results in Fig. 8B show that EpCAM bispecific T cell activator significantly increased DLD cell lysis (EpCAM expression on DLD cells is shown in Fig. 8C).

In a similar experiment, 4,000 SKOV cells were co-cultured with 40,000 CD3+ T cells with or without 300 ng/ml control or EpCAM bispecific T cell activator. After 24 h of co-culture, the supernatants were collected and cytotoxicity was determined by LDH assay. The results in Fig. 9A show that EpCAM bispecific T cell activator significantly increases SKOV cell lysis. In another similar experiment, 5,000 MCF7 cells were co-cultured with 50,000 CD3+ T cells with or without 300 ng/ml control or EpCAM bispecific T cell activator. After 24 h of co-culture, the supernatants were collected and cytotoxicity was determined by LDH assay. The results in Fig. 9B show that the bispecific T cell activator to EpCAM also significantly increases MCF7 cell lysis.

The dose response for EpCAM bispecific T cell activator-mediated cell lysis was assessed by co-culturing 8,000 DLD with 40,000 T cells and concentrations of EpCAM or control bispecific T cell activator ranging from 2x103 to ^2x10-2 ng/mL. After co-cultivation for 24 h at 37°C, LDH assay was performed on supernatants to determine target cell cytotoxicity. Dose response curves were fitted using a four-parameter nonlinear fitting model integrated in GraphPad Prism, generating an EC50 value for FAP bispecific T cell activator of 7.4 ng/mL. The results in Fig. 10 show the dose response between EpCAM bispecific T cell activator concentration and cytotoxicity.

In conclusion, the results of this example demonstrate that a bispecific EpCAM T cell activator is able to induce T cell-mediated lysis of multiple EpCAM-positive tumor cell lines.

Example 6.

Stable FAP-expressing CHO and Ad293 cell lines were generated as a means to demonstrate the specificity of the bispecific FAP T cell activator for the FAP antigen by comparison with parental untransfected cells. Generation of FAP-expressing stably transfected cell lines The protein sequence of the FAP gene was obtained from the NCBI database (SEQ ID 30) and reverse transcribed to generate the coding DNA sequence, which was synthesized by Oxford Genetics Ltd (Oxford, UK). The FAP gene was cloned into the pSF-Lenti vector by standard cloning methods, yielding the pSFLenti-FAP vector. HEK293T cells were transfected with the FAP lentiviral expression vector along with pSFCMV-HIV-Gag-Pol, pSF-CMV-VSV-G, pSF-CMV-HIV-Rev. Lipofectamine 2000 was used as the transfection reagent and was added to the vector DNA at a DNA:lipofectamine ratio of 1:2 and incubated with the cells at 37°C. The supernatant containing lentivirus was collected after 48 h and mixed with polybrene (final concentration, 8 μg/ml). The lentivirus/polybrene mixture was added to the seeded Ad293 or CHO cells and incubated at 37°C. On the fourth day, the supernatant was replaced with medium containing puromycin (2 μg/ml for Ad293 and 7.5 μg/ml for CHO). Stable variants were then clonally selected and FAP expression of the parental cell lines or the stably transfected variant was determined by staining with FAP or isotype control antibody and analyzed by flow cytometry (Fig. 11A).

Bispecific T cell activator to FAP mediated target cell lysis is specific for FAP-expressing cells.

CHO or CHO-FAP cells (7,000 cells) were co-cultured alone or with human T cells (70,000) in the presence of medium alone or 2 μg/ml control or FAP bispecific T cell activator in wells of a 96-well U-bottom plate. After 24 h of incubation, supernatants were collected and target cell cytotoxicity was measured by LDH cytotoxicity assay as described in Example 4 (Fig. 11B). T cell activation was also determined by analyzing CD69 and CD25 expression levels by flow cytometry (Fig. 12). Cytotoxicity was observed only at

- 57 043895 in CHO-FAP cells with T cells and a bispecific FAP T cell activator. This indicates that FAP bispecific T cell activator-mediated T cell activation and target cell lysis are highly specific and restricted to cells expressing

FAP but not the FAP-negative parental cell line.

Example 7.

Stable EpCAM-expressing CHO and Ad293 cell lines were generated as a means to demonstrate the specificity of the EpCAM bispecific T cell activator for the EpCAM antigen by comparison with parental untransfected cells.

Generation of EpCAM-expressing stably transfected cell lines The protein sequence of the EpCAM gene was obtained from the NCBI database (SEQ ID 28) and reverse transcribed to generate the coding DNA sequence, which was synthesized by Oxford Genetics Ltd (Oxford, UK). The EpCAM gene was cloned into the pSF-Lenti vector by standard cloning methods, yielding the pSF-Lenti-EpCAM vector. HEK293T cells were transfected with the EpCAM lentiviral expression vector along with pSF-CMV-HIV-Gag-Pol, pSF-CMV-VSV-G, pSFCMV-HIV-Rev. Lipofectamine 2000 was used as a transfection reagent and was added to the vector DNA at a DNA:lipofectamine ratio of 1:2 and incubated with the cells at 37°C. The supernatant containing lentivirus was collected after 48 h and mixed with polybrene (final concentration, 8 μg/ml). The lentivirus/polybrene mixture was added to seeded Ad293 or CHO cells and incubated at 37°C. On the fourth day, the supernatant was replaced with medium containing puromycin (2 μg/ml for Ad293 and 7.5 μg/ml for CHO). Stable variants were then clonally selected and EpCAM expression of the parental cell lines or the stably transfected variant was determined by staining with EpCAM or isotope control antibody and analyzed by flow cytometry (Fig. 13A).

EpCAM bispecific T cell activator-mediated target cell lysis is specific for EpCAM-expressing cells CHO or EpCAM cells (7,000 cells) were cocultured alone or with human T cells (70,000) in the presence of medium alone or 2 μg/ml control or EpCAM bispecific T cell activator in wells of a 96-well U-bottom plate. After 24 h of incubation, supernatants were collected and target cell cytotoxicity was measured by LDH cytotoxicity assay (Fig. 13B). T cell activation was also determined by analyzing CD69 and CD25 expression levels by flow cytometry (Fig. 14). Cytotoxicity was observed only when CHO-EpCAM cells were co-cultured with T cells and the EpCAM bispecific T cell activator, indicating that the EpCAM bispecific T cell activator-mediated T cell activation and target cell lysis are highly specific and restricted to EpCAM-expressing cells but not to the EpCAM-negative parental cell line.

Example 8.

In the next experiment, the ability of the recombinant FAP bispecific T cell activator protein to activate CD4 or CD8 T cells and the ability of each T cell subset to lyse NHDF cells were assessed. CD3+ T cells (35,000) were cocultured with 7,000 NHDF cells in the presence of 300 ng/ml control or FAP bispecific T cell activator in wells of a 96-well U-bottom plate and incubated at 37°C for 24 h. Cells were harvested and stained with anti-CD4 or anti-CD8 and anti-CD69 and CD25 antibodies and analyzed by flow cytometry. The results (Fig. 15A) demonstrated that FAP bispecific T cell activator induced increases in the activation markers CD69 and CD25 in both CD4+ and CD8+ T cells.

In a similar experiment, the ability of each T cell subset (CD4 and CD8) to kill target cells was assessed. CD4+ T cells were extracted from CD3-purified cells by positive selection using the CD4 T Cell Isolation Kit (Miltenyi Biotec, #130-045-101) according to the manufacturer's protocol with CD8 cells in a non-isolated filtered flow-through. In 96-well U-bottom plates, 7,000 NHDFs with 35,000 CD4+ or CD8+ T cells were co-cultured with 300 ng/ml control or FAP bispecific T cell activator and incubated at 37°C. After 24 h, supernatants were collected and target cell cytotoxicity was measured using the LDH cytotoxicity assay. The results (Fig. 15B) show that the FAP bispecific T cell activator induced both CD4+ and CD8+ T cells to kill NHDF cells.

Example 9.

The ability of the EpCAM bispecific T cell activator to activate CD4+ or CD8+ T cells and the ability of each subset to lyse DLD tumor cells were assessed. CD3+ T cells (35,000) were co-cultured with 7,000 DLD cells in the presence of 300 ng/ml control or EpCAM bispecific T cell activator in a 96-well U-bottom plate and incubated at 37°C for 24 h. Cells were harvested and stained with anti-CD4 or anti-CD8 and anti-CD69 and anti-CD25 antibodies and analyzed by flow cytometry. The results (Fig. 16A) demonstrated that the EpCAM bispecific T cell activator induced an increase in the activation markers CD69 and CD25 in both CD4+ and CD8+ T cells. In a similar experiment, the method was evaluated

- 58 043895 the ability of each T cell subset (CD4 and CD8) to kill target cells. CD4+ T cells were extracted from CD3-purified cells by positive selection using a CD4 T cell isolation kit according to the manufacturer's protocol, with CD8 cells in the unselected filtered flow-through. In 96-well U-bottom plates, 7,000 DLD were cocultured with 35,000 CD4+ or CD8+ T cells with 300 ng/ml control or EpCAM bispecific T cell activator and incubated at 37°C. After 24 h, supernatants were collected and target cell cytotoxicity was measured using an LDH cytotoxicity assay (Fig. 16B). The results show that the bispecific T cell activator to EpCAM induces both CD4+ and CD8+ T cells to kill DLD cells.

Example 10.

Characterization of FAP bispecific T cell activator-mediated activation of autologous tumor-associated lymphocytes from primary ascites of malignancy.

To assess the activity of the bispecific T cell activator proteins using cells obtained from cancer patients, primary ascites fluid samples from malignancies containing both CD3+ T cells and FAP+ cells were collected for testing. Unpurified ascites cells (i.e., unaltered after collection) were seeded at a density of 250,000 cells per well of a 96-well U-bottom plate in either 100% ascites fluid or in medium supplemented with 1% human serum in the presence of 500 ng/ml control or FAP bispecific T cell activator. Untreated wells served as negative controls. After incubation at 37°C for 5 days, the total cell population was harvested and the number of CD3+ T cells (Fig. 17A) and the levels of CD25 expression on CD3+ T cells (Fig. 17B) were determined. Total cell numbers per well were determined using precision counting beads. These results demonstrate that the FAP bispecific T cell activator significantly increases T cell activation by tumor-associated T cells in cancer patients. As a continuation of the experiment described above, cells were pooled from duplicate wells and the number of FAP+ cells was determined by flow cytometry (Fig. 17C). Total cell numbers per well were determined using precision counting beads. These results demonstrate that the FAP bispecific T cell activator significantly reduces the number of autologous FAP-expressing cells in the ascites sample.

Example 11.

Recombinant EnAd bispecific T cell activator-expressing viruses were constructed, produced, and purified using the methods described below. Generation of Enadenotucirev bispecific T cell activator-expressing EnAd is a replication-competent chimeric group B adenovirus that contains frequent non-homologous nucleotide substitutions of Ad3 to Ad11p in the E2B region, a nearly complete deletion of E3, and a smaller deletion of E4 located with E4orf4 (Kuhn et al., Directed evolution generates a novel oncolytic virus for the treatment of colon cancer, PLoS One, 2008 Jun 18; 3(6): e2409). A schematic representation of the genome of the adenoviruses used in this study is shown in Fig. 18A.

Plasmid pEnAd2.4 was used to generate plasmids pEnAd2.4-CMV-EpCAM Bisp.active T-cell, pEnAd2.4-SA-EpCAM Bisp.active T-cell, pEnAd2.4-CMV-FAP Bisp.active T-cell, pEnAd2.4-SA-FAP Bisp.active T-cell, pEnAd2.4-CMV-ControlBi, pEnAd2.4-SA-Control Bisp.active T-cell (Table 4) by direct insertion of the cassette encoding the bispecific T cell activator to EpCAM (SEQ ID NO: 1), the bispecific T cell activator to FAP (SEQ ID NO: 3), or the control bispecific T cell activator (SEQ ID NO: 5). The transgene cassette contained a short 5' splice acceptor sequence (SEQ ID NO: 33) or the exogenous CMV promoter (SEQ ID NO: 31), the cDNA sequence of EpCAM, FAP, or a control bispecific T cell activator, and a 3' polyadenylation sequence (SEQ ID NO: 32). The exogenous CMV promoter is constitutively active and thus leads to early expression of the transgenes. The splice acceptor sequence drives expression under the control of the viral major late promoter and leads to late expression of the transgene after initiation of viral genome replication. The kinetics of this promoter-driven expression can be observed in Fig. 18B, where GFP was used as the transgene.

- 59 043895

Table 4

Plasmid ID [plasmid DNA] ng/ml repa02.4-SMU-ErSAMBisp.active.Tcl. 205.3 repa42.4-5A-ErSAMBisp.active.T-cl. 325.2 рЕпА02.4-СMU-EARBisp.active.T-cl. 1322.8 repA42.4-5A-RARBisp.aktiv.T-cl. 3918,3 pEnAd2.4-CMV-ControlBHcn. active Tcl. 189.1 repA42.4-5A-Sop1go1Bisp.active.T-cl. 236.2 repA42.4-SMU-RARBisp.activ.T-cl,- RFP 1599 rPpA42.4-5A-RARBisp.activ.T-cl.-PPR 1872 pEnAd2.4-CMV-ControlBHcn. active. Tcl.-RFP 1294 repaD2.4-5A-Sop1go1Bisp.active.T-cl.RFP 2082

Obtaining and characterizing viruses.

Plasmids EnAd2.4-CMV-EpCAM Bisp.active T-cell, pEnAd2.4-SA-EpCAM Bisp.active T-cell, pEnAd2.4-CMV-FAP Bisp.active T-cell, pEnAd2.4-SA-FAP Bisp.active T-cell, pEnAd2.4-CMV-Control Bisp.active T-cell, pEnAd2.4-SA-Control Bisp.active T-cell were linearized by restriction digestion with AscI enzyme to obtain a linear viral genome.

Digested DNA was purified by isopropanol extraction and precipitated for 16 h at -20°C in 300 μl >95% molecular biology grade ethanol and 10 μl 3 M sodium acetate. The precipitated DNA was pelleted by centrifugation at 14,000 rpm for 5 min and washed in 500 μl 70% ethanol before centrifugation again at 14,000 rpm for 5 min. Pure DNA pellets were air dried and resuspended in 100 μl water. 6.25 μg DNA was then mixed with 15.6 μl transfection reagent, lipofectamine in OptiMEM, and incubated for 20 min at room temperature. The transfection mixture was then added to a T-25 flask containing Ad293 cells grown to 80% confluence. After incubating the cells with the transfection mixture for 4 h at 37°C, 5% _CO2, 4 ml of cell medium (DMEM with high glucose and glutamine supplemented with 10% FBS) were added to the cells and the flasks were incubated at 37°C, 5% _CO2 . Transfected Ad293 cells were monitored every 24 h and supplemented with additional media every 48-72 h. Virus production was monitored by observing a significant cytopathic effect (CPE) in the cell monolayer. Once intense CPE was observed, the virus was harvested from the Ad293 cells by three freeze-thaw cycles. Individual viral clones were selected by serial dilution of the pooled lysate and re-infection of cells with Ad293 and harvesting from wells containing individual plaques. Serial infections of cells with Ad293 were performed after infection had reached complete CPE to amplify viral material. Production of viable virus during amplification was confirmed by observation of significant CPE in the cell monolayer.

Virus cleaning.

Following amplification of the active viral material, the viruses were purified by double cesium chloride density gradient centrifugation (banding) to obtain viral material NG-601, NG-602, NG-603, NG-604, NG-605, and NG-606. These materials were titrated using the micoBCA assay (Life Technologies) following the manufacturer's instructions (Table 5).

Table 5

ID EnAd NG ID No: SEQ ID of the viral genome vh/ml TCID50/M l EnAd-CMV-ErOAMBisp.active.Tcl. NG-601 SEQ ID NO: 34 2,2494x1012 l,26xl0ii EnAd-SA-ErSAMBisp.activ.T-cl. NG-602 SEQ ID NO: 35 4,21746x1012 l,58xl0ii

- 60 043895

EnAd-CMV-ControlBncn.active.Tcl. NG-603 l,42607xl0i ² 5.01x101° Epab-5A-SopGgo1Bisp.active.T-cl. NG-604 3,31073x1012 2,00xl0ii EnAd-CMV-EARBisp.active.T-cl. NG-605 SEQ ID NO: 36 1,64653x1012 l,58xl0n EnAd-SA-EARBis.activ.T-cl. NG-606 SEQ ID NO: 37 l,28148xl0 ¹² 3.98xl0 ¹⁰ EnAd-CMV-ControlBncn.active.Tcl.-P2A-YAEP NG-607 5,963x1012 l,26xl0 ⁹ EnAd- SA-ControlBncn. active. Tcl.-P2A-KEP NG-608 1,51848x1012 6.31x109 EnAd-CMV-RARBisp. active. T-cl.- P2A-RFP NG-609 1,57517x1012 7.94x109 EnAd-SA-EARBis.activ.T-cl.- P2A-RFP NG-610 7,74881xl0ii 5.01x101°

Example 12.

The activity of viruses NG-601, NG-602, NG-603, NG-604, NG-605 and NG-606 was characterized using the methods described below.

Characterization of the activity of EnAd encoding a bispecific T cell activator compared with EnAd in carcinoma cell lines.

The ability of NG-601, NG-602, NG-603, NG-604, NG-605, NG-606, or EnAd to replicate was analyzed by infecting A549 lung carcinoma cells and assessed by qPCR. A549 cells were seeded into wells of a 24-well plate at a cell density of 2x105 cells/well. Plates were incubated for 18 h at 37°C, 5% _CO2 , then cells were either infected with 100 viral particles per cell (ppc) or left uninfected. Wells were harvested at 24, 48, or 72 h post infection, and DNA was purified using the PureLink Genomic DNA Assay Mini Kit (Invitrogen) according to the manufacturer's protocol. Total viral genome content was quantified by qPCR for each extracted sample or standard using a set of primers specific for the EnAd gene in the reaction mixture detailed in Table 6. qPCR was performed according to the program in Table 7.

Table 6

Reagent Volume/well (µl) 2 x qPCRBIO, probe mix (PCRBiosystems) 10 EnAd, direct primer 0.08 EnAd, reverse primer 0.08 EnAd, probe — 0.8 NFW — 4.04 Sample — 5 Volume of the hole Table 7 20 Number of cycles Temperature (°C) Duration (sec) 1 95 120 40 95 60-65 5 20-30

Quantification of detected viral genomes per cell demonstrated that replication of NG-601, NG-602, NG-603, NG-604, NG-605, NG-606 and EnAd viruses was comparable in the A549 cell line (Fig. 19A).

The oncolytic activity of NG-601, NG-602, NG-603, NG-604, NG-605, NG-606, or EnAd was assessed by infection with A549 (Fig. 19B). A549 cells were seeded in a 96-well plate at a cell density of ^1.5x104 cells/well. The plates were incubated for 18 h, 37°C, 5% _CO2 , then the cells were infected with increasing numbers of ppc virus (5-fold serial dilution, from 4.1x10-7 to 5000 ppc virus).

- 61 043895 rus) or left uninfected. A549 cytotoxicity was measured on day 5 using the CellTiter 96® AQueous One Solution (MTS) Cell Proliferation Assay (Promega, #G3582). Dose response curves were fitted using a four-parameter nonlinear fitting model integrated in GraphPad Prism. IC50 values obtained for each virus demonstrated that the oncolytic activity of NG-601, NG-602, NG-603, NG-604, NG-605, NG-606 and EnAd was comparable for each virus. Confirmation of the expression of a functional bispecific T cell activator transgene from NG-601, NG-602, NG-603, NG-604, NG-605, NG-606.

To determine whether NG-601, NG-602, NG-605, NG-606 viruses produce functional bispecific T cell activators, T cell activation assays were performed using CHO, CHO-EpCAM, and CHO-FAP cell lines as target cells. 10,000 target cells were co-cultured with 50,000 CD3+ T cells in a 96-well U-bottom plate with Ad293 viral supernatants diluted 100-fold in culture medium and incubated for 24 h at 37°C, 5% CO ₂ . T cells were harvested and stained with antibodies specific for CD25 and CD69 and analyzed by flow cytometry. The results (Fig. 20A and 20B) showed that NG601 and NG-602 viruses expressed a functional bispecific T cell activator transgene that activated T cells when co-cultured with CHO-EpCAM cells, and NG-605 and NG606 expressed a functional bispecific T cell activator transgene that activated T cells when co-cultured with CHO-FAP cells but not when co-cultured with CHO cells.

Quantification of bispecific T cell activator expression in a colon carcinoma cell line.

The expression of bispecific T cell activator was assessed by infection of human colon carcinoma cell line with NG-601, NG-602, NG-605, NG-606 viruses. DLD cells were seeded in 6-well culture plates at a density of 1.2x 10 ⁶ cells per well. Eighteen hours after seeding, DLD cells were infected with EnAd, NG-601, NG-602, NG-603, NG-604, NG-605, NG-606 at a density of 100 ppc. Cells were cultured for 72 h, then supernatants were collected from wells and centrifuged for 5 min at 1200 rpm to remove cell debris. The clarified supernatants were then used for killing assays and cytotoxicity was compared to a standard curve generated with recombinant bispecific T cell activator of known concentration, allowing the amount of bispecific T cell activator in viral supernatants to be determined.

To quantify the FAP bispecific T cell activator produced from NG605 and NG-606, a cytotoxicity assay was performed in which 8,000 NHDFs were co-cultured with 40,000 CD3+ T cells and DLD viral supernatants diluted 1 in 103, 1 in ¹⁰⁴ , and 1 in 105. A standard curve was generated by incubating NHDFs and CD3+ T cells with FAP or control bispecific T cell activator at 10-fold serial dilutions from 3,333 to 3.33 ^{x 10-4} ng/μl. Supernatants were collected 24 h after treatment and cytotoxicity was measured by LDH assay. The amount of bispecific T cell activator expressed was determined by comparing the cytotoxicity of viral supernatants with that of a standard curve of recombinant bispecific T cell activator. The results (Fig. 21) showed that NG-605 and NG-606 viruses produced 9.8 and 49.2 μg of bispecific T cell activator to FAP per million DLD cells, respectively.

To quantify the EpCAM bispecific T cell activator produced from NG-601 and NG-602, a cytotoxicity assay was performed in which 8,000 DLD cells were co-cultured with 40,000 CD3+ T cells and DLD viral supernatants diluted 1 in 103, 1 in ^104, and 1 in 105. A standard curve was generated by incubating DLD and CD3+ T cells with EpCAM or control bispecific T cell activator at 10-fold serial dilutions from 3,333 to 3.33 x ^10-4 ng/μl. Supernatants were collected 24 h after treatment and cytotoxicity was measured by LDH assay (Fig. 22). The amount of bispecific T cell activator expressed was determined by comparing the cytotoxicity of viral supernatants with that of a standard curve of recombinant bispecific T cell activator. The results showed that NG-601 and NG-602 viruses produced 165 and 50.3 μg EpCAM bispecific T cell activator per million DLD cells, respectively.

Example 13.

In addition to encoding FAP or a control bispecific T cell activator, viruses NG-607, NG-608, NG-609, NG-610 also carry a red fluorescent protein (RFP) transgene for visualization of infected cells using fluorescence microscopy methods (SEQ ID NOs: 25 and 26, Table 4). The functional activities of these viruses were characterized using the methods described below.

Confirmation of Transgene Expression from NG-607, NG-608, NG-609, NG-610 The ability of NG-607, NG-608, NG-609, and NG-610 viruses to produce their bispecific T cell activator transgene was assessed by infecting Ad293 cells. Ad293 cells were seeded in a 6-well plate at a density of 1x10 ⁶ cells/well. The plates were incubated for 24 h, 37°C, 5% CO ₂ , then the cells were infected with viruses at a density of 100 ppc or left uninfected. Forty-eight h after infection, the plaques were irradiated with a mercury fluorescent lamp and photographed (Fig. 23). The results showed that viruses NG-607, NG-608, NG-609 and NG-610 express the RFP transgene.

Example 14.

In the next series of experiments, the ability of EnAd and viruses NG-603, NG-604, NG-605, NG-606, NG-607, NG-608, NG-609, NG-610 expressing the bispecific T cell activator to FAP or the control bispecific T cell activator to kill target cells, including tumor cells and fibroblasts, was assessed.

In the first study, the ability of EnAd to kill DLD cells was assessed using xCELLigence technology. DLD cells were seeded in a 48-well E-plate at a density of 1.2 x ¹⁰⁴ cells/well and incubated for 18 h, 37°C, 5% CO2, then the cells were infected with EnAd viruses at a density of 100 ppc or left uninfected. XCELLigence was used every 15 min to measure target cell cytotoxicity over an 8-day incubation period. The results (Fig. 24A) show that EnAd could effectively kill DLD cells during this time period. A similar experiment assessed the ability of EnAd to kill SKOV cells using xCELLigence technology. SKOV cells were seeded in a 48-well E-plate at a density of 1x104 cells/well and incubated for 18 h, 37°C, 5% _CO2 , then cells were infected with EnAd viruses at a density of 100 ppc or left uninfected. XCELLigence was used to measure target cell cytotoxicity every 15 min during the 8-day incubation period. The results (Fig. 24B) show that SKOV cells are resistant to EnAd-mediated cytotoxicity during this time period.

In a similar experiment, the ability of EnAd to kill NHDF cells was also assessed using xCELLigence technology. NHDF cells were seeded in a 48-well E-plate at a density of 4x103 cells/well and incubated for 18 h, 37°C, 5% _CO2 , then the cells were either infected with 100 ppc EnAd or left uninfected. XCELLigence was used every 15 min to measure cytotoxicity on target cells for the same period of time as for A549 and SKOV cells. The results (Fig. 24C) show that EnAd was unable to kill NHDF cells during the observation time.

In a similar experiment, the ability of NG-603, NG-604, NG-605, NG-606, and EnAd to kill NHDF cells co-cultured with SKOV tumor cells and CD3+ T cells was assessed using xCELLigence. NHDF cells and SKOV cells were seeded in a 48-well E-plate at a density of 4x103 and 1x103 cells/well, respectively. Plates were incubated for 18 h at 37°C, 5% _CO2 , then cells were either infected with 100 ppc of EnAd, NG-603, NG-604, NG-605, or NG-606 or left uninfected. After 2 h of incubation, 37,500 CD3+ T cells were added to each well. To measure cytotoxicity on target cells, xCELLigence was used every 15 min. The results (Fig. 25A) demonstrate that FAP bispecific T cell activator-expressing viruses NG-605 and NG606, but not EnAd or the control bispecific T cell activator-expressing viruses NG-603 and NG-604, were able to induce NHDF cell lysis, with kinetics dependent on the promoter used to express the bispecific T cell activator (faster with the CMV promoter).

In a similar experiment, the ability of NG-603, NG-604, NG-605, NG-606, and EnAd to kill NHDF cells in co-culture with SKOV and CD3+ T cells was assessed using an LDH cytotoxicity assay. NHDF cells and SKOV cells were seeded in a 96-well U-bottom plate at a density of 8x103 and 2x103 cells/well, respectively, and either infected with 100 ppm of EnAd, NG-603, NG-604, NG-605, or NG-606 or left uninfected. After a 2-hour incubation, 75,000 CD3+ T cells were added to each well and the plates were incubated at 37°C, 5% _CO2 . Supernatants were collected at 0, 24, 48, and 96 h after treatment and cytotoxicity was measured by LDH cytotoxicity assay. The results (Fig. 25B) demonstrate that FAP bispecific T cell activator-expressing viruses NG-605 and NG606, but not EnAd or the control bispecific T cell activator-expressing viruses NG-603 and NG-604, were able to induce NHDF cell lysis, with kinetics dependent on the promoter used to express the bispecific T cell activator. As a continuation of the LDH assay experiment described above, cells were also collected at 0, 24, 48, and 96 h after treatment, stained with antibodies to CD45, CD69, and CD25, and analyzed by flow cytometry. The results (Fig. 26) demonstrate that FAP bispecific T cell activator-expressing viruses NG-605 and NG-606, but not EnAd or the control bispecific T cell activator-expressing viruses NG-603 and NG-604, were able to induce T cell activation with kinetics dependent on the promoter used to express the bispecific T cell activator.

In a similar experiment, FAP dependence for the ability to induce FAP bispecific T cell activator-mediated T cell activation was assessed. In 96-well U-bottom plates, SKOV cells were seeded at 2 × 10 ³ cells/well alone or in combination with NHDF cells at 8 × 10 ³ cells/well. Viral particles were added to each well at 100 ppc, and the plates were incubated at 37°C, 5% CO ₂ . After 2 h, 75,000 CD3+ T cells were added and the plates were incubated further. At 96 h post infection, cells were harvested and stained for CD45 and CD25 and analyzed by flow cytometry (Fig. 27A). The results demonstrate that FAP-binding T cell activator-expressing viruses NG-605 and NG606 induced T cell activation only in the presence of FAP-positive NHDF cells.

In a similar experiment, the specificity of promoter-driven (CMV or MLP/SA virus) expression of the bispecific T cell activator in NG-605 and NG-606 was investigated. NHDF cells were seeded in a 96-well U-bottom plate at 4 x 10 ³ cells/well. 100 viral particles per cell were added to each well and the plates were incubated at 37°C, 5% CO ₂ . After 2 h, 40,000 CD3 cells were added and the plates were incubated further. Supernatants were collected 72 h after infection and cytotoxicity was measured using an LDH cytotoxicity assay. The results (Fig. 27B) demonstrate that CMV-driven NG-605, but not SA-driven NG-606, was able to mediate NHDF cell killing when infecting NHDF cells only.

These results indicate that both NG-605 and NG-606 are capable of inducing T cell activation and target cell lysis, although the kinetic profile differs slightly depending on the promoter used.

Time-lapse videos were acquired to observe target cell lysis mediated by viruses or T cells recombinant with bispecific T cell activator to FAP, EnAd, NG-603, or NG-605. NHDF cells were stained with CellTracker Orange CMTMR (Life Tech, #C2927) and CD3+ T cells were stained with CellTrace Violet Cell Proliferation Assay Kit (Life Tech, #C34557) according to the manufacturer's protocols. Stained NHDF were seeded in 24-well plates at 7.5 x 10 ³ cells/well in co-culture with DLD or SKOV tumor cells 1.35 x 10 ⁴ . Plates were incubated for 18 h, 37°C, 5% CO2. Cells were then treated with 300 ng/ml FAP bispecific T cell activator, infected with 100 ppb of EnAd, NG-603, and NG-605, or left untreated. After a 2-hour incubation, 100,000 stained CD3+ T cells were added to the appropriate wells in addition to 1.5 μM CellEventCaspase 3-7 reagent (Life Tech, #C10423). Videos were acquired on a Nikon TE 2000-E Eclipse inverted microscope, with images acquired every 15 min for 96 h. Still images from the video are shown in Fig. 28. The results indicate that recombinant FAP bispecific T cell activator and NG-605, but not EnAd or NG-603, are able to induce rapid lysis of NHDF cells.

In a similar experiment, NHDF cells were stained with CellTracker Green CMFDA (Life Tech, #C2925) and CD3+ T cells were stained with CellTrace Violet Cell Proliferation Kit (Life Tech, #C34557) according to the manufacturer's protocols. Stained NHDF were seeded in 24-well plates at 7.5 x 10 ³ cells/well in co-culture with 1.35 x 10 ⁴ DLD or SKOV tumor cells. The plates were incubated for 18 h, 37°C, 5% CO ₂ . Cells were then infected with 100 ppm of NG-607, NG-608, NG-609, or NG-610 or left uninfected. After a 2-hour incubation, 100,000 stained CD3+ T cells were added to the desired wells. Videos were acquired on a Nikon TE 2000-E Eclipse inverted microscope, with images acquired every 15 min for 96 h. Stills from the video are shown in Fig. 29. The results show that all viruses resulted in tumor cell infection (RFP, red fluorescence, positive), but only NG-609 and NG-610 were able to induce rapid lysis of cocultured NHDF cells.

Example 15.

In this series of experiments, the ability of EnAd and the EpCAM bispecific T cell activator or control bispecific T cell activator viruses NG-601, NG-602, NG603, and NG-604 to kill target cells, including tumor cells and fibroblasts, was assessed.

Characterization of human T cell activation and lysis of EPCAM-positive target cells by EnAd, NG-601, NG-602, NG-603 and NG-604.

The ability of EnAd and NG-601, NG-602, NG-603, and NG-604 to kill DLD tumor cells in the presence or absence of CD3+ T cells was assessed using xCELLigence technology. DLD cells were seeded in a 48-well E-plate at 1.2 x ¹⁰⁴ cells/well. Plates were incubated for 18 h, 37°C, 5% _CO2 , then cells were either infected with 100 ppc EnAd or left uninfected. Two h after infection, 75,000 CD3+ T cells were added to the appropriate wells. XCELLigence was used every 15 min to measure cytotoxicity on target cells. The results (Fig. 30) demonstrate that NG-601 and NG-602 result in significantly more rapid cytotoxicity against DLD in a T cell-dependent manner.

In a similar experiment, the ability of EnAd and NG-601, NG-602, NG-603 and NG-604 to kill DLD tumor cells in the presence or absence of CD3+ T cells using LDH was assessed.

- 64 043895 cytotoxicity assay. DLD cells were seeded in a 96-well U-bottom plate at 2 x 10 ⁴ cells/well and either infected with 100 ppc EnAd or left uninfected. Two hours after infection, 150,000 CD3+ T cells were added to the appropriate wells. Plates were incubated at 37°C, 5% CO ₂ , and the supernatant was collected and analyzed using the LDH cytotoxicity assay at 0, 24, 48, and 72 hours after infection. The results (Fig. 3I) demonstrate that NG-601 and NG-602 result in more rapid cytotoxicity against DLD in a T cell-dependent manner.

As a continuation of the LDH assay experiment described above, cells were also harvested at 0, 24, 48, and 96 h post-treatment, stained with antibodies to CD45, CD69, and CD25, and analyzed by flow cytometry to determine the activation status of CD3+ T cells. The results (Fig. 32) demonstrate that the EpCAM bispecific T cell activator-expressing viruses NG-601 and NG602, but not EnAd or the control bispecific T cell activator-expressing viruses NG-603 and NG-604, were able to induce T cell activation with kinetics dependent on the promoter used to express the bispecific T cell activator.

In another experiment, the ability of NG-601 to kill DLD tumor cells at different multiplicities of infection (MOI) in the presence or absence of CD3+ T cells was assessed using xCELLigence technology. DLD cells were seeded in a 48-well E-plate at ² x 104 cells/well. Plates were incubated for 18 h, 37°C, 5% CO2, then cells were either infected with NG-601 at MOIs ranging from 0.001 to 10 or left uninfected. Two hours after infection, 150,000 CD3+ T cells were added to the desired wells. xCELLigence was used every 15 min to measure target cell cytotoxicity. The results (Fig. 33) demonstrate that NG-601 results in more rapid cytotoxicity of DLD in a T cell-dependent manner at an MOI of only 0.001.

In a similar experiment, the ability of EnAd and NG-601, NG-602, NG-603, and NG-604 to kill SKOV tumor cells in the presence or absence of CD3+ T cells was assessed using xCELLigence technology. SKOV cells were seeded in a 48-well E-plate at 1 x 10 ⁴ cells/well. Plates were incubated for 18 h, 37°C, 5% CO ₂ , then cells were infected with EnAd (100 ppc) or left uninfected. Two hours after infection, 50,000 CD3+ T cells were added to the desired wells. xCELLigence was used every 15 min to measure target cell cytotoxicity. The results (Fig. 34) indicate that SKOV cells are resistant to EnAd-mediated cytotoxicity during the period of this study, however, NG-601 and NG-602 were able to induce rapid lysis of SKOV cells in the presence of CD3+ T cells.

In a similar experiment, the ability of EnAd and NG-601, NG-602, NG-603, and NG-604 to kill SKOV cells in the presence or absence of CD3+ T cells was assessed using an LDH cytotoxicity assay. SKOV cells were seeded in 96-well U-bottom plates at 2 x 10 ⁴ cells/well and either infected with EnAd (100 ppc) or left uninfected. Two hours after infection, 150,000 CD3+ T cells were added to the desired wells. The plates were incubated at 37°C, 5% CO ₂ , and the supernatant was collected and analyzed using an LDH cytotoxicity assay at 0, 24, 48, and 72 hours after infection. The results (Fig. 35) are consistent with previous data and suggest that SKOV cells are resistant to EnAd-mediated cytotoxicity during this time period, however, NG-601 and NG-602 are able to induce rapid lysis of SKOV cells in the presence of CD3+ T cells. As a continuation of the LDH assay experiment described above, cells were also harvested at 0, 24, 48, and 96 h post-treatment, stained with antibodies to CD45, CD69, and CD25, and analyzed by flow cytometry to determine the activation status of CD3+ T cells (Fig. 36). The results show that the EpCAM bispecific T cell activator-expressing viruses NG-601 and NG-602, but not EnAd or the control bispecific T cell activator-expressing viruses NG-603 and NG-604, were able to induce T cell activation with kinetics dependent on the promoter used to express the bispecific T cell activator.

In a similar experiment, the ability of EnAd and NG-601, NG-602, NG-603, and NG-604 to activate cancer patient-derived CD3+ T cells from a CD3+ EpCAM-positive primary ascites sample was assessed. EpCAM-positive DLD cells were seeded at ^1x104 cells/well in a 96-well U-shaped plate and co-cultured with 100,000 ascites cells (unchanged). Cells were either infected with 100 ppc of viral particles or left uninfected. After incubation at 37°C for 48 h, the total cell population was collected and the expression level of CD25 on CD3+ T cells was determined by flow cytometry. The results (Fig. 37) demonstrate that the EpCAM bispecific T cell activator-expressing viruses NG-601 and NG-602, but not EnAd or the control bispecific T cell activator-expressing viruses NG-603 and NG-604, are able to induce T cell activation in patient-derived CD3+ T cells.

The results show that both EpCAM bispecific T cell activator viruses, NG-601 and NG-602, were able to induce T cell activation and target cell lysis, although the kinetics

- 65 043895 The profile differed slightly depending on the promoter used.

Time-lapse videos were acquired to observe target cell lysis mediated by viruses or T cells recombinant with the bispecific T cell activator to EpCAM, EnAd, NG-603, or NG-605. NHDF cells were stained with CellTracker Orange CMTMR (Life Tech, #C2927) and CD3+ T cells were stained with the CellTrace Violet Cell Proliferation Assay Kit (Life Tech, #C34557) according to the manufacturer's protocols. Stained NHDF were seeded in 24-well plates at 7.5 x 103 cells/well in co-culture with 1.35 x ¹⁰⁴ DLD or SKOV tumor cells. Plates were incubated for 18 h, 37°C, 5% CO ₂ . Cells were then treated with 300 ng/ml bispecific T cell activator to EpCAM, infected with 100 ppb EnAd, NG-603, and NG-605, or left untreated. After 2 h of incubation, 100,000 stained CD3+ T cells were added to the appropriate wells in addition to 1.5 μM CellEvent Caspase 3-7 reagent (Life Tech, #C10423). Movies were acquired on a Nikon TE 2000-E Eclipse inverted microscope, with images acquired every 15 min for 96 h. Stills from the movies are shown in Fig. 28. The results show that the recombinant bispecific T cell activator EpCAM and NG-605 lead to rapid lysis of both DLD and SKOV target cells, but NHDF remained unaffected.

Example 16.

In this example, activation of autologous tumor-associated lymphocytes from FAP+ primary malignant ascites from cancer patients by EnAd, NG-603, NG-604, NG-605, and NG-606 viruses was assessed. Patient samples considered suitable for further analysis included CD3+ T cells and FAP+ cells. In the first experiment, crude (i.e., unchanged since collection) ascites cells from a patient were seeded at 250,000 cells/well of a 96-well U-bottom plate in 100% ascites fluid. Cells were infected with 100 ppc of viruses, and untreated wells served as negative controls. EnAd-CMV-GFP and EnAd-SA-GFP were also included in the experiment as reporters for infection and late viral gene expression, respectively, and the resulting photomicrographs are shown in Fig. 39. After incubation at 37°C for 5 days, the total cell population was harvested and the expression level of CD25 on CD3+ T cells was determined (Fig. 40A). The total number of cells per well was determined using precision counting beads. The results show that the FAP bispecific T cell activator viruses NG-605 and NG-606 result in a significant increase in T cell activation by tumor-associated lymphocytes. As a continuation of the experiment described above, cells were harvested from duplicate wells and the number of endogenous FAP+ cells was determined by flow cytometry. The total number of cells per well was determined using precision counting beads. The results (Fig. 40B) show that NG-605 and NG-606 resulted in a significant reduction in the number of autologous FAP-expressing cells in ascites samples, suggesting that some FAP+ cells were killed by activated T cells.

In a second experiment, crude (i.e., not altered after collection) ascites cells from a cancer patient were seeded at 250,000 cells/well of a 96-well U-bottom plate in either 100% ascites fluid or medium supplemented with 1% human serum. The cells were infected with 100 ppc of viruses, and untreated wells served as negative controls. EnAd-CMV-GFP and EnAd-SA-GFP were also included as a reporter for infection and late-stage viral gene expression, respectively, and the resulting micrographs are shown in Fig. 41. After incubation at 37°C for 5 days, the total cell population was harvested and the number of CD3+ T cells (Fig. 42) and the level of CD25 expression on CD3+ T cells (Fig. 43) were determined. Total cell counts per well were determined using precision counting beads. The results show that for this patient, recombinant bispecific T cell activator to FAP and NG-605, but not NG-606, resulted in significant increases in T cell activation by tumor-associated lymphocytes in the media. Neither virus resulted in activation in ascites fluid.

As a continuation of the experiment described above, cells were collected from duplicate wells and the number of FAP+ cells was determined by flow cytometry (Fig. 44). The total number of cells per well was determined using precision counting beads. The results show that recombinant bispecific T cell activator to FAP and NG-605, but not NG-606, significantly reduced the number of autologous FAP-expressing cells in the media. Neither virus reduced the number of FAP+ cells in the ascites fluid.

Example 17. Materials and methods.

Cell lines.

HEK293A, DLD, SKOV3, MCF7, A431, A549 and PC3 (ATCC) cells were cultured in Dulbecco's modified Eagle's medium (DMEM, Sigma-Aldrich, UK), and CHO cells (ATCC) were cultured in Roswell Park Memorial Institute medium (RPMI-1640, Sigma, Aldrich, UK). The culture medium was supplemented with 10% (v/v) fetal bovine serum (FBS, Gibco, UK) and

- 66 043895

1% (v/v) penicillin/streptomycin (10 mg/ml, Sigma-Aldrich) and the cells were maintained in a humidified atmosphere at 37°C and 5% CO2. For viral infection and transfection of viral plasmids, cells were maintained in DMEM supplemented with 2% FBS. For transfection of recombinant bispecific T cell activator plasmids, cells were maintained in DMEM without FBS. EpCAM expression in target cell lines was determined by flow cytometry. Generation of stable EpCAM-expressing cell lines The protein sequence of the EpCAM gene was obtained from the NCBI database (SEQ ID 28) and reverse transcribed to generate the coding DNA sequence, which was synthesized by Oxford Genetics Ltd (Oxford, UK). The EpCAM gene was cloned into the pSF-Lenti vector by standard cloning methods to yield the pSF-Lenti-EpCAM vector. HEK293T cells were transfected using Lipofectamine 2000 with the EpCAM lentiviral expression vector along with pSF-CMV-HIV-Gag-Pol, pSF-CMV-VSV-G, pSF-CMV-HIV-Rev (Oxford Genetics Ltd). The supernatants containing lentivirus were collected after 48 h and mixed with polybrene (8 μg/ml). The lentivirus/polybrene mixture was added to CHO cells and incubated at 37°C. On the fourth day, the supernatant was replaced with medium containing 7.5 μg/ml puromycin. Stable variants were then clonally selected and EpCAM expression of the parental cell lines or the stably transfected variant was determined by staining with EpCAM or an isotope control antibody and analyzed by flow cytometry. Positive clones were expanded and used for further experiments. Preparation of peripheral blood mononuclear cells (PBMCs) and isolation of T cells PBMCs were isolated by density gradient centrifugation (Boyum, 1968) from whole blood leukocyte concentrates obtained from NHS Blood and Transplant UK (Oxford, UK). Blood was diluted 1:2 with PBS and overlaid with Ficoll (1079 g/mL, Ficoll-Paque Plus, GE Healthcare) before centrifugation at 400 g for 30 min at 22°C with a long deceleration. After centrifugation, PBMCs were collected and washed twice with PBS (300 g for 10 min at room temperature) and resuspended in RPMI-1640 medium supplemented with 10% FBS. CD3-positive T cells were extracted from PBMCs by depletion of non-CD3 cells using the Pan T Cell Isolation Kit (Miltenyi Biotec, #130-096-55) according to the manufacturer's protocol. To further isolate CD4- and CD8-positive T cells, CD3 T cells underwent another round of purification using CD4+ beads (Miltenyi Biotec, #130-045-101).

Treatment of primary ascites and pleural effusion.

Primary human ascites and pleural effusion samples from malignancy were obtained from Churchill Hospital, Oxford University Hospitals (Oxford, UK) after informed consent from patients with multiple features of advanced carcinoma, including ovary, pancreas, breast and lung. This work was approved by the Oxford Histopathology Research Centre Research Ethics Committee. Upon receipt, cellular and fluid fractions were separated and the fluid was used immediately or aliquots were stored at -20°C for subsequent analysis. The cellular fraction was treated with red blood cell lysis buffer (Roche, UK) according to the manufacturer's instructions. Cell numbers and viability were determined by trypan blue staining. The cell types present in each sample were determined by staining with antibodies to EpCAM, EGFR, FAP, CD45, CD11b, CD56, CD3, CD4, CD8, PD1, and CTLA4 and analyzed by flow cytometry. Fresh cells and fluid were used for ex vivo T cell activation and target cell lysis experiments. In some cases, adherent cells were passaged in DMEM supplemented with 10% FBS and expanded for later use.

Construction and production of a bispecific T cell activator.

Bispecific T cell activators were generated by joining two scFvs of different specificities with a flexible GS linker. Each scFv was constructed by joining the VH and VL domains of the parental monoclonal antibodies using a linker. Each bispecific T cell activator possessed an N-terminal immunoglobulin (Ig) light chain signal sequence for secretion by mammalian cells and a C-terminal decahistidine affinity tag for detection and purification. Bispecific T cell activators were constructed by standard DNA cloning methods and inserted into a protein expression vector (pSFCMV-Amp) for cytomegalovirus (CMV) promoter-driven expression and secretion of the constitutive protein. Plasmid DNA pSF-CMVЕрCAM Bisp.activ.T-cell or pSF-CMV-Control Bisp.activ.T-cell was used to generate the bispecific T cell activators. were transfected into HEK293A cells using polyethyleneimine (PEI, linear, MW 25000, Polysciences, USA) under the following conditions, 55 μg plasmid DNA:110 μg PEI (DNA:PEI ratio 1:2 (w/w)) were added to the cells, incubated at 37°C for 4 h, then replaced with fresh serum-free DMEM medium and further incubated at 37°C, 5% _CO2 for 48 h. The cells were transfected in parallel with pSFCMV-GFP to ensure the transfection efficiency. To collect the secreted protein, the supernatant of the transfected cells was collected and centrifuged at 350 g, 4°C for 5 min to remove cellular components. The supernatants were transferred to a 10,000 MWCO Amicon Ultra-15 Centrifugal Filter (Millipore). After centrifugation at 4750 g and 4°C, the volume of the retentate was adjusted by flow filtration to obtain a 50-fold higher concentration. Aliquots were concentrated

- 67 043895 protein was stored at -80°C. Generation of EnAdenotucirev bispecific T cell activator expressing plasmids pEnAd2.4-CMV-EpCAM Bisp.act.T-cell, pEnAd2.4-SA-EpCAM Bisp.activ.T-cell, pEnAd2.4-CMV-Control Bisp.activ.T-cell, pEnAd2.4-SA-Control Bisp.activ.T-cell were generated by direct insertion of the transgene cassette encoding the bispecific T cell activator to EpCAM or the control bispecific T cell activator into the EnAd base plasmid pEnAd2.4 using the Gibson assembly technology. The transgene cassette contained a short 5' splice acceptor sequence or the exogenous CMV promoter followed on the right by the cDNA sequence of the bispecific T cell activator to EpCAM or the control bispecific T cell activator and a 3' polyadenylation sequence. The schematic of the inserted transgene cassette is shown in Fig. 18. The correct construction of the plasmid was confirmed by DNA sequencing. Plasmids EnAd2.4-CMV-EpCAM Bisp.active.T-cell, pEnAd2.4-SA-EpCAM Bisp.active.T-cell, pEnAd2.4-CMVControl Bisp.active.T-cell and pEnAd2.4-SA-Control Bisp.active.T-cell were linearized by restriction digestion with AscI enzyme before transfection in HEK293 cells. Virus production was monitored by observing a significant cytopathic effect (CPE) in the cell monolayer. Once a strong CPE was observed, virus was harvested from HEK293A cells by three freeze-thaw cycles. Individual viral clones were selected by serially diluting the pooled lysate and re-infecting HEK293A cells and harvesting from wells containing individual plaques. Serial infections of HEK293A cells were performed after infection had achieved a complete CPE to amplify viral material. Following amplification of the active viral material, the viruses were purified by double cesium chloride density gradient centrifugation (banding) to obtain the viral material EnAd-CMVEpCAM Bisp.active T-cell, EnAd-SA-EpCAM Bisp.active T-cell, EnAdCMV-Control Bisp.active T-cell, EnAd-SA-Control Bisp.active T-cell. These materials were titrated by TCID50 and the PicoGreen assay (Life Technologies) following the manufacturer's instructions.

Preparation of supernatants.

To assess bispecific T cell activator-mediated cytokine release, DLD cells (20,000) were seeded with 100,000 CD3+ T cells in a 96-well flat-bottom plate alone or with 2 ng/μl EpCAM or control bispecific T cell activator. After 48 h of incubation at 37°C and 5% _CO2, supernatants were collected, cellular components were removed by centrifugation, and aliquots were stored at -20°C. To assess the expression of the bispecific T cell activator transgene from the recombinant viruses, HEK293A cells (1x106) or DLD cells (1.2x106) were infected with EnAd-CMV-EpCAM Bisp.active T cells, EnAd-SA-EpCAM Bisp.active T cells, EnAd-CMVControl Bisp.active T cells, EnAd-SAControl Bisp.active T cells or EnAd at 100 vp/cell. The cells were cultured for 72 h, after which the cytopathic effect (CPE) was enhanced. The supernatants were collected and centrifuged for 5 min at 300 g to remove cell debris and stored at -20°C for subsequent analysis.

Immunoblotting.

Dot blotting was used to measure the concentration of recombinant bispecific T cell activator obtained from plasmid transfections. Two-fold serial dilutions of each bispecific T cell activator and protein standard (10χHis-tagged (C-terminus) human cathepsin D, Biolegend, #556704) were prepared. The molar concentration of the protein standard was adjusted to provide a bispecific T cell activator concentration of 100 μg/mL. Two μL of each sample and protein standard were applied directly to a nitrocellulose membrane. The membrane was air dried, blocked and probed with a-6xHis (C-terminal) antibody (1:5000, clone 3D5, Invitrogen, UK, #46-0693) to detect proteins tagged with His at the C-terminus, followed by washing and incubation with secondary anti-mouse antibody (1:10000, Dako, #P0161) and detection with SuperSignal West Dura Extended Duration Substrate (Thermo Fisher, #34075) according to the manufacturer's instructions. Supernatants of virus-infected HEK293A cells were analyzed by Western blotting for bispecific T cell activator expression. Supernatants were fractionated by SDS-PAGE and transferred to nitrocellulose membrane according to the manufacturer's protocols (Bio-Rad). The membranes were then processed identically to the dot blot protocol described above.

Enzyme-linked immunosorbent assay (ELISA).

To assess EpCAM binding, ELISA plates were prepared by coating overnight at 4°C with human EpCAM/TROP-1 protein (50 ng/well, Sino Biological Inc, #10694-H02H-50). Plates were blocked for 1 h at ambient temperature with 5% BSA, followed by incubation with diluted supernatants of HEK293A transfected with EpCAM bispecific T cell activator, control bispecific T cell activator, and empty pSF-CMV vector (2 h, room temperature). Plates were washed three times with PBS-T and then after each subsequent binding step. Plates were incubated with anti-His antibody (C-terminus) (1:5000, clone 3D5, #46-0693, Invitrogen, UK) for 1 h at room temperature, followed by HRP-conjugated anti-mouse-Fc (1:1000 in PBS/5% milk, Dako) for 1 h at room temperature. HRP detection was performed using 3.3.5.5'-tetramethylethylenediamine (TMB, Thermo-Fisher), and for

- 68 043895 Stop solution was used to terminate the reaction. Absorbance at 450 nm was measured on a plate reader.

Wallac 1420 (Perkin Elmer).

Flow cytometry.

Flow cytometry analysis was performed on a FACSCalibur flow cytometer (BD Biosciences), and data were processed using Flowjo v10.0.7r2 software (TreeStar Inc., USA). Antibodies specific for CD45 (HI30, Biolegend), CD11b (ICRF44, Biolegend), EpCAM (9C4, Biolegend), and FAP (427819, R&D Systems) were used to classify different cell populations. The following antibody clones linked to different fluorophores were used for the analysis of T cell populations: CD69 (FN50, Biolegend), CD25 (BC96, Biolegend), IFNy (4S.B3, Biolegend), aCD107a antibody (H4A3, Biolegend), CD3 (HIT3a, Biolegend), CD4 (OKT4, Biolegend), CD8a (HIT8a, Biolegend), PD1 (H4A3, Biolegend). In each case, the corresponding isotype control antibody was used.

Characterization of human T cell activation Expression levels of CD69 and CD25.

The ability of recombinant bispecific T cell activators to EpCAM or EpCAM bispecific T cell activator-bearing viruses to induce T cell activation was assessed by surface expression of CD69 and CD25. Human CD3 cells (75,000 cells/well in 96-well flat-bottom plates) from PBMC or ascites samples were cultured alone or with DLD, SKOV, CHO, CHOepCAM target cells or from ascites (15,000) in the presence of medium alone, EpCAM, or bispecific T cell activator control protein (2 ng/μl) or recombinant virus (100 vp/cell). In some cases, anti-PD1 antibody (Invivogen, #hpdlni-mab7) was added at a final concentration of 2.5 μg/ml. CD3 cells were incubated with CD3/CD28 Dynabeads (Thermo Fisher, #11131D) as a positive control for T cell activation. Cells were cultured for 24 h at 37°C unless otherwise noted and then harvested with Enzymatic Cell Dissociation Buffer (Gibco, #13151014). All cells were stained with antibodies for surface expression of CD69, CD25, CD3, CD4, or CD8 and analyzed by flow cytometry. The effect of ascites fluid on T cell activation (CD69, CD25) was investigated by polyclonal activation of CD3-purified PBMCs (100,000) by incubation with plate-immobilized anti-CD3 antibody (7.5 μg/ml, HIT3a, Biolegend, #300313) in RPMI-1640 or fluids isolated from ascites samples of malignant neoplasms.

Expression of IFNy.

The ability of the EpCAM bispecific T cell activator to induce T cell activity was assessed by IFNγ expression by co-culture of T cells for 6 h with DLD cells (200,000 CD3 cells/well, 40,000 DLD cells/well in a 96-well flat-bottom plate) and 2 ng/μl recombinant EpCAM or control bispecific T cell activator. As a positive control, T cells were stimulated with soluble PMA/ionomycin cell activating cocktail (Biolegend, #423301). Brefeldin A (GolgiPlug, BD Biosciences) was added to the culture medium 5 h before harvest, after which CD3+ T cells were collected and intracellularly stained for IFNγ expression and analyzed by flow cytometry.

T cell proliferation.

To study T cell proliferation, 100,000 CFSE-labeled (CellTrace CFSE kit, Invitrogen, #C34554) CD3+ T cells were incubated with 20,000 DLD cells in a 96-well plate with 2 ng/μl EpCAM or control bispecific T cell activator. Five days after co-culture, cells were stained for CD3, CD4, or CD8, and CFSE fluorescence of viable CD3+ T cells was measured by flow cytometry, normalized for total cell counts, using Precision counting beads (5000/well, Biolegend, #424902). Fluorescence data were analyzed and modeled using the proliferation fit function of Flowjo v7.6.5. Data are presented as the percentage of parent cells that entered a proliferation cycle (% divided) or the average number of cell divisions a cell in the parent population underwent (division index).

Degranulation of CD107a.

DLD cells (15,000 cells/well) were co-cultured with 75,000 CD3+ T cells in a 96-well flat-bottom plate in the presence of medium alone or 2 ng/μl control bispecific T cell activator or EpCAM bispecific T cell activator. aCD107a or isotype control antibodies were added directly to the culture medium. Monensin (GolgiStop, BD Biosciences) was added after 1 h of incubation at 37°C and 5% CO ₂ , followed by a further 5 h incubation. Cells were then harvested, stained for CD3, CD4, or CD8, and analyzed by flow cytometry.

Release of cytokines.

Cytokine levels in supernatants collected from DLD/PBMC cultures or pleural effusion cells were determined using the LEGENDplex Human T Helper Cytokine panel (Biolegend, #740001) and flow cytometry following the manufacturer's instructions. Cytokines included in the analysis were:

- 69 043895

IL-2, IL-4, IL-5, IL-6, IL-9, IL-10, IL-13, IL-17A, IL-17F, IL-21, IL-22, IFNy and TNFa.

In vitro cytotoxicity assay for target cells.

Target cell cytotoxicity mediated by recombinant bispecific T cell activator or viruses was assessed by LDH release or MTS assay. Target cells (DLD, SKOV, HT-29, A431, A549, PC3, CHO, CHO-EpCAM) were co-cultured with CD3, CD4 or CD8 T cells (E:T 5:1) in a 96-well flat-bottom plate in the presence of medium alone, diluted supernatants or virus (100 vp/cell). After 24 h of co-culture (unless otherwise stated), supernatants and cells were collected and cytotoxicity was determined using LDH assay (CytoTox 96 Non-Radioactive Cytotoxicity Assay, Promega, #G1780) or MTS viability assay (CellTiter 96, Cell Proliferation Assay, Promega, #G3580) according to the manufacturer's instructions. The amount of bispecific T cell activator produced from DLD virus-infected cells was determined by comparing the cytotoxicity induced by diluted viral supernatants with the cytotoxicity of a standard curve generated using recombinant bispecific T cell activator.

To assess the oncolytic activity of the viruses, DLD cells were seeded in a 96-well plate (25,000 cells/well) for 18 h at 37°C and 5% CO ₂ , then infected with increasing numbers of vp/cell (5-fold serial dilution, from 100 to 5.12x10'5 vp/cell) or left uninfected. DLD cytotoxicity was measured on day 5 using the MTS viability assay. Dose-response curves were constructed and IC50 was determined using a four-parameter nonlinear fitting model integrated in Prism 7 (GraphPad Software). Cell viability was monitored in real time using xCELLigence RTCA DP technology (Acea Biosciences). DLD, SKOV3, or MCF7 cells were seeded in 48-well E-plates at 12,000 cells/well. Plates were incubated for 18 h, 37°C, 5% _CO2 , then cells were treated with bispecific T cell activator (2 ng/μl), infected with virus (100 vp/cell), or left untreated. Two h after infection, 75,000 CD3+ cells were added to the desired wells. Cell impedance was measured every 15 min for up to 160 h. For ex vivo cytotoxicity assays, crude cells from ascites or pleural effusion samples were resuspended in ascites fluid and seeded (1.5 x 105/well) in 96-well flat-bottom plates. After incubation for the indicated duration at 37°C, 5% _CO2 , supernatants were analyzed by LDH assay or total cells were harvested with cell dissociation buffer, stained for CD3, CD25, and EpCAM, and analyzed by flow cytometry. For PD1 blocking experiments, anti-PD1 antibody (2.5 μg/ml, Invivogen, #hpd1ni-mab7) was included.

Viral genome replication and qPCR.

The ability of EnAd-CMV-EpCAM Bisp.active T cells, EnAd-SA-EpCAM Bisp.active T cells, EnAdCMV-Control Bisp.active T cells, EnAd-SAControl Bisp.active T cells or EnAd to replicate their genomes was analyzed by seeding DLD cells in 24-well plates (150,000 cells/well) for 18 h, 37°C, 5% CO ₂ , prior to infection with 100 vp/cell virus. Cells from wells were harvested at 24 and 72 h post-infection and DNA was purified using the PureLink Genomic DNA Isolation Mini Kit (Invitrogen, #K182001) according to the manufacturer's protocol. Total viral genome content was quantified by qPCR against EnAd hexon using specific primer-probe sets (primers: TACATGCACATCGCCGGA/CGGGCGAACTGCACCA, probe: CCGGACTCAGGTACTCCGAAGCATCCT).

Microscopy.

Brightfield and fluorescence images were acquired on a Zeiss Axiovert 25 microscope. Time-lapse movies were captured to observe virus- or T-cell-mediated target cell lysis with EnAd or EnAd-CMVEpCAM Bisp.activ.T-cells. Uninfected cells were used as negative controls. NHDF cells were stained with CellTracker Orange CMTMR (Life Technologies, #C2927) and CD3+ cells were stained with CellTrace Violet Cell Proliferation Kit (Life Technologies, #C34557) according to the manufacturer's protocols. Stained NHDF were seeded in 24-well plates at 7,500 cells/well in coculture with SKOV3 at 13,500 cells/well. Plates were incubated for 18 h, 37°C, 5% _CO2 . Cells were then treated with 300 ng/ml bispecific T cell activator to EpCAM, infected with 100 hpc/cell EnAd or EnAd2.4-CMV-EpCAM Bisp.activ.T cells, or left untreated. After 2 h of incubation, 100,000 stained CD3+ cells were added to the desired wells in addition to 1.5 μM CellEvent Caspase 3-7 reagent (Life Technologies, #C10423). Images were acquired on a Nikon TE 2000-E Eclipse inverted microscope (10x optical objective) at 15 min intervals covering a period of 96 h. Time-lapse movies (12 fps) were created using ImageJ software.

Statistics.

In all cases where more than two experimental conditions were compared, statistical analysis was performed using one-way ANOVA with Tukey's post hoc test. All data are presented as mean ± standard deviation. Significance levels used were P = 0.01–0.05 (*), 0.001–0.01 (**), 0.0001–0.001 (***). All in vitro experiments were performed in triplicate unless otherwise stated.

Example 18. Preparation and production of a bispecific T cell activator targeting EpCAM.

A bispecific T cell activator targeting EpCAM was constructed by fusing two scFvs specific for CD3ε and EpCAM with a flexible glycine-serine (GS) linker. A control bispecific T cell activator recognizing CD3ε and an irrelevant antigen (Bordetella pertussis filamentous adhesion hemagglutinin (FHA)) was also generated. Both bispecific T cell activators were designed to contain an N-terminal signal sequence for mammalian secretion and a C-terminal decahistidine affinity tag for detection and purification (Fig. 45A). To characterize the functionality of the recombinant bispecific T cell activators, they were cloned into expression vectors under the transcriptional control of the CMV immediate-early promoter (pSF-CMV-EpCAM Bisp.activ.T-cell and pSF-CMV-Control Bisp.activ.T-cell, respectively). Adherent HEK293 cells (HEK293A) were transfected with the expression vectors and the supernatants were collected and concentrated 50-fold for further analysis. To assess the amount of bispecific T cell activator produced, samples were serially diluted and assayed with anti-His in a dot blot assay using decahistidine-labeled cathepsin D as a standard. Thus, it was possible to estimate that the level of bispecific T cell activators produced in the supernatant was approximately 20 μg/ml 48 h after transfection (1.8 x 10 ⁷ HEK293A cells) (Fig. 46A). Specific binding of the bispecific T cell activator to EpCAM, but not the control bispecific T cell activator, to the recombinant EpCAM protein was demonstrated by ELISA (Fig. 46B).

Example 19. Characterization of human T-cell activation by a recombinant bispecific T-cell activator to EpCAM.

The ability of the recombinant EpCAM bispecific T cell activator protein to activate PBMC-derived T cells was assessed by adding unstimulated primary human CD3+ cells to cultured human DLD colorectal carcinoma cells, which are known to express EpCAM on their surface (Karlsson et al., 2008). Addition of 2.5 ng/ml EpCAM bispecific T cell activator (as supernatant from transduced HEK293A cells) resulted in a significant increase in the T cell activation markers CD69 and CD25 (Fig. 45B and C), whereas the control bispecific T cell activator had no effect. Exposure of CD3 cells to the bispecific EpCAM T cell activator in the absence of tumor cells resulted in very modest increases in CD69 and CD25, indicating that antibody-mediated CD3 clustering is required for full activation by this anti-CD3 binding. T cells stimulated with the bispecific EpCAM T cell activator in the presence of tumor cells also showed significant increases in gamma interferon production (Fig. 45D) and cell proliferation (Fig. 45E), whereas the control bispecific T cell activator had no effect. The purpose of T cell activation is to induce degranulation-mediated cytotoxicity, and surface CD107a/LAMP1 expression (indicative of degranulation, Aktas et al.) was strongly increased by treatment with the bispecific T cell activator to EpCAM but not control (Fig. 45F).

Cytokine release following EpCAM bispecific T cell activator-mediated activation of PBMC-derived T cells in the presence of DLD cells was characterized by flow cytometry using a cytokine bead array. As before, the control bispecific T cell activator showed little activity, although EpCAM bispecific T cell activator induced the release of several cytokines, including high levels of IL-2, IL-6, IL-10, IL-13, IFN-gamma, and TNF (Fig. 45G). IL-2, IFN-gamma, and TNF production are typically associated with a Th1 response, whereas IL-6 and IL-10 are more commonly associated with a Th2 response (Mosmann & Sad, 1996).

Example 20. Specificity of recombinant bispecific T-cell activators to EpCAM.

Most human epithelial cells express EpCAM, so to assess whether the effect of the EpCAM bispecific T cell activator is antigen specific, Chinese hamster ovary cells (CHO cells) were engineered using a lentiviral vector to express human EpCAM on their surface. In the presence of the EpCAM bispecific T cell activator and CHO-EpCAM cells, exogenously added PBMC-derived T cells showed strong activation (as assessed by CD25 expression, see Fig. 47A) and associated cytotoxicity (Fig. 47B) that was not observed with parental control CHO cells or control bispecific T cell activator cells. This indicates that the cytotoxicity of the EpCAM bispecific T cell activator is antigen specific.

We next assessed whether the bispecific EpCAM T cell activator could kill a range of tumor cells and whether the level of observed EpCAM bispecific T cell activator-mediated cytotoxicity depended on the density of EpCAM expression. T cell cytotoxicity in the presence of the bispecific EpCAM T cell activator was measured in six different carcinoma cell lines, with the highest cytotoxicity observed in DLD and A431 and the lowest in A549 and PC3 (Fig. 47C). This showed a weak association with surface EpCAM levels (as determined by flow cytometry), with A549 and PC3 cells showing the lowest levels and DLD the highest (Fig. 47D). This suggests that the presence and expression level of EpCAM do influence the degree of cytotoxicity, although other factors (possibly the intrinsic resistance of cells to granzyme-mediated apoptosis) also play a role in determining the overall level of cell killing.

Example 21. Bispecific T cell activator-mediated activation of CD4+ and CD8+ T cell subsets.

To determine which T cell types are activated by the EpCAM bispecific T cell activator, PBMC-derived T cells were incubated with DLD cells and activated using the bispecific T cell activators and then analyzed by flow cytometry. Both CD4+ and CD8+ cells displayed high levels of CD69 and CD25 expression (Fig. 49A), although the percentage of activated CD4 cells was generally slightly higher. EpCAM bispecific T cell activator-mediated T cell proliferation was assessed using CFSE staining (Fig. 49B) and degranulation by CD197a/LAMP1 expression (Fig. 49C), and as in the previous example, similar levels of activation were observed for both CD4+ and CD8+ cells. Finally, tumor cell cytotoxicity levels were compared using the bispecific T cell activator to EpCAM to activate purified CD4+ and CD8+ subsets. All T cell preparations showed similar cytotoxicity (Fig. 49D), indicating that both CD4+ and CD8+ cells can contribute to the observed bispecific T cell activator-mediated cytotoxicity.

Example 22. Expression of a bispecific T-cell activator to EpCAM from an oncolytic adenovirus, EnAdenotucirev.

EnAdenotucirev (EnAd) is an oncolytic adenovirus, a chimera of group B adenovirus type 11 and type 3 with a mosaic E2B region, a nearly complete deletion of E3 and a smaller deletion of E4 localized to E4orf4 (Kuhn 2008). This virus is currently in several early-phase clinical trials for the treatment of cancer, combining good systemic pharmacokinetics and promising clinical activity with the ability to encode and express transgenes (Calvo 2014, Boni 2014). A bispecific T-cell activator to EpCAM was encoded in EnAd immediately downstream of the fiber gene, using a shuttle vector inserted into the viral backbone by Gibson assembly (Fig. 18). The bispecific T cell activator was placed either under the transcriptional control of the CMV immediate-early promoter (EnAd-CMV-EpCAM Bisp.activ.T-cell) or was placed downstream of the splice acceptor site for the adenovirus major late promoter (MLP; EnAd-SA-EpCAM Bisp.activ.T-cell). In the first configuration, the bispecific T cell activator would be expressed whenever the virus successfully infects a cell, whereas expression from the MLP splice acceptor site would only occur when MLP is activated in cells that are permissive for viral replication. A control bispecific T cell activator (recognizing CD3 and FHA) was also introduced to generate two corresponding control viruses.

Viruses were cloned, reconstituted in HEK293A cells, and each was prepared in bulk in a HYPERFlask and purified twice by cesium chloride banding. Infection of DLD with parental EnAd and the bispecific T cell activator recombinant viruses yielded similar amounts of viral genomes (as measured by qPCR) at all time points tested, indicating that the bispecific T cell activator transgene does not affect viral replication kinetics (Fig. 5A). We next examined the replicative and oncolytic properties of the viruses in the absence of human T cells. DLD cells were infected with batches of virus at increasing numbers of viral particles (vp)/cell, and cytotoxicity was measured by the MTS assay on day 5. All recombinant viruses, including those with the bispecific T cell activator to EpCAM and control bispecific T cell activators driven by the CMV promoter or splice acceptor, showed cytotoxic activity indistinguishable from the parental virus, indicating that the genetic modification did not alter the intrinsic oncolytic activity of the virus (Fig. 51B).

To assess the expression and secretion of the bispecific T cell activator, EnAd bispecific T cell activator-expressing viruses were used to infect HEK293A cells and the supernatants were examined by Western blotting with an anti-His antibody after 72 h. As shown in Fig. 51C, all four viruses (two expressing the control bispecific T cell activator and two expressing the EpCAM bispecific T cell activator) showed similar levels of bispecific T cell activator secreted into the supernatant.

- 72 043895

Example 23. Selective killing of EpCAM-positive cells using a virus-produced bispecific EpCAM T-cell activator.

Supernatants from EnAd-EpCAM bisp.activ.T cells-infected HEK293A cells were added to CHO and CHO-EpCAM cultures either with or without PBMC-derived T cells; T cell activation and cytotoxicity toward CHO/CHO-EpCAM cells were measured after 24 h. In the case of CHO cells, neither an increase in CD25 expression on T cells was observed (Fig. 51D) nor any cytotoxicity was observed with either treatment (Fig. 51E). However, T cells incubated with CHOpCAM cells showed a significant increase in CD25 expression when using supernatants from HEK293A cells that were infected with EnAd-CMV-EpCAM bisp.activ.T cells or EnAdSA-EpCAM bisp.activ.T cells (Fig. 51D). As expected, this translated into selective cytotoxicity against CHO-EpCAM cells only when T cells were added in the presence of supernatant from 293A cells that had been infected with EnAd-CMV-EpCAM Bisp.activ.T cells or EnAd-SA-EpCAM Bisp.activ.T cells (Fig. 51E). Importantly, cytotoxicity was absent in the absence of T cells or when using supernatants from HEK293A cells when the cells were infected with EnAd expressing a control bispecific T cell activator.

Example 24. Superior cytotoxicity of EnAd expressing a bispecific T cell activator to EpCAM.

EnAd rapidly kills most carcinoma cells by direct oncolysis (Kuhn 2008), although some cells, notably SKOV3 ovarian carcinoma cells, are partially resistant and are killed more slowly. We therefore reasoned that the consequences of arming End virus to secrete a bispecific T cell activator to EpCAM, leading to cytotoxic T cell activation, might be particularly evident in SKOV3 cells. Therefore, cells were exposed to virus (100 vp/cell) 24 h after seeding and cell death was monitored by the xCELLigence system. PBMC-derived T cells were added (or not added) to the SKOV3 cell culture after 2 h. In the absence of T cells, tumor cells grew for approximately 72 h (as evidenced by the increase in the cell index signal in Fig. 53A), but then cell growth plateaued and remained stable, independent of virus infection, until at least 160 h. All viruses tested, including the parental EnAd, did not induce detectable cytotoxicity against target cells during the time course of the measurements. However, when co-cultured with PBMC-derived T cells, both CMV- and SA-EpCAM bisp.activ.T cell-armed viruses induced rapid SKOV3 lysis, with CMV inducing lysis within 16 h and SA within 44 h after T cell addition (Fig. 53B). Importantly, parental EnAd or non-specific control viruses expressing the bispecific T cell activator failed to lyse target cells during this time period, even with the addition of T cells. This result was confirmed by an LDH assay in which co-cultures identical to those above were established and cytotoxicity was measured at 24, 48, and 96 h post-infection (Fig. 48). These results are supported by similar results in DLD cells, in which viruses expressing the bispecific T cell activator to EpCAM induced cytotoxicity significantly faster than control bispecific T cell activator viruses (Fig. 50A+B).

To confirm that target cell cytotoxicity was mediated by T cell activation, CD3 cells were harvested at each time point and activation status was determined by CD69 and CD25 expression, showing expression kinetics similar to that observed for cytotoxicity (Fig. 53C and D, Fig. 50C and D). The approximate amount of EpCAM bispecific T cell activator produced from infected DLD cells was determined by comparing the cytotoxicity (Abs490) induced by infected DLD supernatants with that induced by known amounts of recombinant bispecific T cell activator (i.e., creating a standard curve (Abs490)). DLD in co-culture with CD3-purified PBMCs (1:5) were incubated with recombinant bispecific T cell activator (Fig. 50E) or infected DLD supernatant (Fig. 50F), and LDH release was measured after 24 h. This allowed us to determine that the bispecific T cell activator to EpCAM was produced at levels of 165 μg and 50 μg per million DLD for EnAd-CMVEpCAM Bisp.activ.T cells and EnAd-SAEpCAM Bisp.activ.T cells, respectively. The EC50 for the bispecific T cell activator to EpCAM is 7.4 ng/mL (Fig. 50E and F), and therefore the bispecific T cell activator to EpCAM is produced by the recombinant virus at levels that may reach therapeutic doses.

The cytotoxicity of EnAd expressing a bispecific T cell activator to EpCAM was visualized using time-lapse video microscopy.

Tumor cells.

SKOV3 (untagged) were co-incubated with normal human fibroblasts (EpCAM-negative, labeled red, serving as non-targeting controls) and PBMC-derived T cells (labeled blue) in the presence of caspase staining (CellEvent Caspase 3-7 reagent produces a green spot upon caspase activation). As in the previous example, the combination of EnAd expressing the bispecific T cell activator to EpCAM in combination with exogenous T cells resulted in exceptionally high cytotoxicity against SKOV3 tumor cells, which showed strong induction of apoptosis upon infection with EnAd-CMVEpCAM Bisp.activ.T cells but not with the parental EnAd. Importantly, EpCAM-negative NHDFs in co-culture remained viable throughout the process. Representative fluorescence images at different time points from a video of SKOV3 cells are shown in Fig. 53E. Time-lapse videos showing DLD cells (which are inherently more susceptible to the virus) co-cultured with NHDFs are also shown.

Example 25. A bispecific T-cell activator to EpCAM can overcome immunosuppression, activate endogenous T cells, and kill endogenous tumor cells in peritoneal ascites of malignancy.

Three clinical peritoneal ascites samples from malignancy containing EpCAM-positive tumor cells and primary fibroblasts (non-EpCAM-expressing cells as a control) were expanded ex vivo and mixed primary cell populations were incubated with PBMC-derived T cells and then treated with free bispecific T cell activator or 100 ppm EnAd-EpCAM Bisp.activ.T cells in culture medium. After 72 h, the level of EpCAM-positive target cells (Fig. 55A) or non-target FAP-positive fibroblasts (Fig. 55B) were measured by flow cytometry. T cell activation was analyzed by measuring CD25 expression (Fig. 55C). Free EpCAM bispecific T cell activator and EpCAM bispecific T cell activator-expressing viruses induced T cell activation resulting in depletion of EpCAM-positive tumor cells, while primary FAP-positive (EpCAM-negative) fibroblasts showed no change in numbers. This was observed in all patient samples, and none of the other drugs showed any significant effects. This demonstrates that EpCAM bispecific T cell activator (or its encoding oncolytic virus) can mediate activation and selective cytotoxicity of PBMC-derived T cells against human ovarian ascites tumor cells.

Malignancy exudates likely represent a potential immune tolerance environment with suppressed immune responses commonly observed in patients with advanced metastatic cancer. To test this hypothesis, we polyclonally stimulated PBMC-derived T cells with anti-CD3 antibodies in culture media or in the presence of 100% ascites fluid from five patients with abdominal malignancies. While in RPMI medium the anti-CD3 antibody resulted in approximately 50% of T cells being positive for both CD25 and CD69, the presence of ascites fluid appeared to attenuate T cell activation as evidenced by a reduction in antibody-mediated increase in CD69/CD25 expression, and this was particularly evident in the fluid of patient #2 (Fig. 56A). This supports our notion that ascites fluid components may exert immunosuppressive or tolerance-inducing effects. However, this attenuation in the increase in activation markers did not correlate with the suppression of T cell degranulation, with CD107a externalized in ascitic fluid similar to that in culture medium (Fig. 5B). This suggests that bispecific T cell activators may be able to bypass tumor microenvironment-associated T cell immunosuppression mechanisms (Nakamura & Smyth, 2016). We therefore examined the ability of PBMC-derived T cells and an EpCAM bispecific T cell activator to mediate target cell cytotoxicity in the presence of immunosuppressive ascitic fluid. T cells incubated with ascites fluid 1 and 2 induced lysis of the human breast adenocarcinoma MCF7 cell line similar to that seen in RPMI culture medium (measured using xCELLigence), although cytotoxicity was delayed by approximately 8 h in the presence of ascites fluid from Patient #2 (Fig. 56C). In addition to containing immunosuppressive fluid and tumor cells, ascites contains tumor-associated lymphocytes and tumor-stromal support cells, providing a unique tumor-like model system for testing bispecific T cell activator-mediated activation of endogenous patient-derived T cells. Following 24 h incubation of total endogenous cells and ascites fluid with free recombinant bispecific T cell activator, patient T cell activation was assessed (Fig. 56D). In this clinically important setting, the EpCAM bispecific T cell activator (but not the control analogue) induced expression of CD69 and CD25, although at lower levels of CD25 when the experiment was performed in 100% ascites fluid rather than in plain medium. These data indicate that the EpCAM bispecific T cell activator is able to overcome at least some of the immunosuppressive effects of peritoneal ascites fluid to activate endogenous T cells. Cytotoxicity was assessed by measuring LDH release, and the bispecific T cell activator induced significant increases in both the medium and 100% ascites experiments. This indicates that some of the ascites cells were killed by bispecific T cell activator-mediated cytotoxicity, although given the multiple cell types present in primary ascites, it is impossible to determine what proportion of tumor cells are killed.

- 74 043895

Example 26. EnAd expressing a bispecific T cell activator to EpCAM can activate endogenous T cells to kill endogenous tumor cells in pleural exudates of malignant neoplasms.

To study the impact of EpCAM bispecific T cell activator expressing viruses in a different clinical setting, we obtained several pleural effusion samples from patients with different malignancies. In an initial screening (an example is shown in Fig. 52), samples containing CD3 and EpCAM positive cells were considered suitable for further analysis. We also assessed PD1 expression by endogenous T cells after their initial isolation, and although only 10% of PBMC-derived T cells express PD1, all T cells from malignancy effusion samples were at least 40% positive for PD1, and this percentage sometimes approached 100% (Fig. 54). Crude cells in bulk (isolated by centrifugation and resuspended) were incubated at fixed concentrations in 100% pleural effusion fluid in the presence of 500 ng/ml free EpCAM bispecific T cell activator or 100 vp/cell virus encoding the bispecific T cell activator. After 5 days, the total cell population was harvested and total CD3+ cells were measured (Fig. 57A). Compared with untreated controls, T cell proliferation was only evident in samples treated with free EpCAM bispecific T cell activator or EnAd encoding the EpCAM bispecific T cell activator. This confirms that the EpCAM bispecific T cell activator bound to the EpCAM target and cross-linked CD3 to stimulate endogenous T cells. The level of CD25 expression on CD3 cells was also determined (Fig. 57B). Free EpCAM bispecific T cell activator induced significant T cell activation of tumor-associated lymphocytes (as assessed by CD25 expression) in all patient samples, even in the likely immunoallergic environment of pleural exudate fluid. The addition of anti-PD1 blocking antibody had no effect on EpCAM bispecific T cell activator-mediated T cell activation in this setting (Fig. 54B and C). There was marked variation between patients (although little between samples from the same patient), with activation ranging from 50% to 90% depending on the donor. Similarly, samples treated with EnAd expressing a bispecific T cell activator to EpCAM showed high activation in some patients (from 10-20% to 80% for both EnAd-CMV-EpCAM Bisp.activ.T cells and EnAd-SA-EpCAM Bisp.act.T cells).

Interestingly, the patient showing the lowest bispecific T cell activator-mediated activation also had the lowest background T cell activation. Parental EnAd, EnAd expressing control bispecific T cell activators, or free control bispecific T cell activators did not induce stimulation above background. We assessed the ability of bispecific T cell activator-expressing viruses to mediate EpCAM-targeted cytotoxicity by measuring residual levels of EpCAM-positive cells by flow cytometry at the end of the 5-day incubation (Fig. 57C). Free EpCAM bispecific T cell activator and the two EpCAM bispecific T cell activator-encoding viruses caused a marked depletion of autologous EpCAM-expressing cells in each case, whereas the other treatments had little or no effect on the level of EpCAM-positive cells. For sample #1, the viability of all EnAd-based viruses was slightly reduced compared with the untreated control, likely reflecting the effects of direct viral oncolysis. Because the PD1 blocking antibody had no effect on T cell activation, it had no effect on EpCAM bispecific T cell activator-mediated target cell killing, with almost complete cytotoxicity of EpCAM+ cells (patients 2, 3, and 4) in the absence of the PD1 blocker (Fig. 54D).

The differential effects of parental EnAd and EnAd-CMV-EpCAM Bisp.activ.T-cells are shown by microscopy in Fig. 57D, where expression of the bispecific T cell activator reduces the presence of tumor cells and expands the T cell population. Corresponding flow cytometry plots confirm significant expansion and activation of T cells following treatment with the EpCAM bispecific T cell activator expressing virus.

Finally, the effects of the different treatments were characterized by measuring the levels of key cytokines produced using the LEGENDplex protein array (Fig. 57E). By far the largest increase was in interferon-gamma expression, which increased nearly 1,000-fold after treatment with free EpCAM bispecific T cell activator or EnAd encoding the EpCAM bispecific T cell activator. These two treatments also induced approximately 10-fold increases in the expression of IL-5, IL-13, tumor necrosis factor (TNF), IL17A, and IL17F, all characteristic of activated T cells. EnAd alone (or expressing a control bispecific T cell activator) also induced a 10-fold increase in interferon-gamma, but otherwise neither treatment induced any detectable changes in cytokine expression.

Example 27. Discussion.

Oncolytic viruses offer an intriguing new strategy that combines several

- 75,043,895 therapeutic modalities in a single targeted, self-amplifying agent (Keller & Bell, 2016; Seymour & Fisher, 2016). Because they selectively replicate in cancer cells and spread from cell to cell, some oncolytic viruses are thought to mediate cell death through non-apoptotic death pathways (Ingemarsdotter et al., 2010; Li et al., 2013), a process that allows viral particles to escape from dying cells. In particular, EnAd kills cells through a proinflammatory process known as oncosis or ischemic cell death (Dyer, 2017). This non-apoptotic death mechanism results in the release of several pro-inflammatory cellular components such as ATP, HMGB1 and calreticulin activation (known as damage-associated molecular patterns, DAMPs) (Weerasinghe & Buja, 2012) and is likely critical to the ability of the virus to promote an effective anti-cancer immune response. However, in addition to the effects of direct lysis, viruses offer the opportunity to encode and express other anti-cancer biologics, eliminating delivery issues and ensuring that the biologic reaches its highest concentration in the tumor microenvironment. Imlygic encodes GM-CSF, however the potential for weaponizing viruses is virtually limitless and provides many exciting opportunities for the development of multimodal therapeutic strategies with additive or synergistic anti-cancer effects (de Gruijl et al., 2015; Hermiston & Kuhn, 2002).

Encoding a bispecific T cell activator in oncolytic viruses provides a potent means to activate tumor-infiltrating lymphocytes to become cytotoxic and lyse antigen-positive target cells, providing an entirely separate therapeutic modality from the effects of direct viral lysis. In this study, we show that bispecific T cell activator-directed cytotoxicity is entirely antigen-specific, can be mediated by both CD4 and CD8 T cells (Brischwein et al., 2006), and can be incorporated into an oncolytic adenovirus and expressed only in cells that permit viral replication. Furthermore, the present study shows for the first time that endogenous T cells in liquid cancer biopsies can be activated by a bispecific T cell activator and virus-encoded bispecific T cell activators and can kill endogenous tumor cells without any additional stimulation or reversal of immune suppression. Importantly, this can occur even in primary fluids that include the microenvironment of peritoneal ascites or pleural effusions as surrogates for the immunosuppressive microenvironment of solid tumors. Arming oncolytic viruses to express a bispecific T cell activator combines two very different therapeutic mechanisms: the first mediates lytic death of tumor cells that are permissive to viral infection, and the second targets T cell cytotoxicity via a specific, selected antigen. This provides considerable flexibility in the design of a therapeutic approach, perhaps using a bispecific T cell activator to deliver cytotoxicity to tumor-associated cells that are relatively resistant to direct viral killing. For example, although we have exemplified herein the use of bispecific T cell activator technology that recognizes a carcinoma-associated antigen (EpCAM), the bispecific T cell activator approach could also be used to target cytotoxicity to tumor-associated fibroblasts or other stromal cells. Indeed, even when the targets for recognition by the bispecific T cell activator are not restricted to expression in the tumor microenvironment, coupling bispecific T cell activator production to viral replication allows for spatial restriction of bispecific T cell activator expression within the tumor, minimizing systemic toxicity. This is important because intravenously administered bispecific T cell activator exhibits relatively short circulation kinetics (Klinger et al., 2012) and is often associated with significant off-target toxicity outside the tumor (Teachey et al., 2013).

The possibility of encoding a bispecific T cell activator in oncolytic viruses has been previously studied using the oncolytic vaccinia virus with an Ephrin A2-targeted bispecific T cell activator. This agent showed that the Ephrin bispecific T cell activator could mediate PBMC activation and antigen-directed tumor cell killing both in vitro and in vivo. Interestingly, although the bispecific T cell activator could activate T cells, it did not result in T cell proliferation without the addition of exogenous IL-2, whereas the bispecific T cell activator used in the present study resulted in extensive proliferation of both PBMC in vitro and tumor-associated lymphocytes using clinical biopsy specimens ex vivo.

We speculate that the observed differences may reflect different design of the bispecific T cell activator, different oncolytic viruses used, or perhaps depend on the density of antigen that allows sufficient cross-linking of CD3 on T cells.

One of the main goals of oncolytic virus therapy is to generate an antitumor T cell response that recognizes patient-specific neoantigens as well as common tumor-associated antigens. Lytic viruses can do this by stimulating improved

- 76 043895 antigen presentation by tumor cell lysis in the context of DAMPs along with virus-associated pathogen-associated molecular patterns (PAMPs). Immunohistochemical staining of resected colon tumors after intravenous delivery of EnAd shows that the virus promotes a strong influx of CD8+ T cells into tumor tissue (Garcia-Carbonero, 2017). However, although this is a potentially very powerful approach, adaptive T cell responses ultimately depend on tumor cell expression of MHC class I antigens, which allows targeted cell killing. Loss of MHC expression is a well-documented immune evasion strategy for tumors (Garrido et al., 2016).

It should be noted that both cytotoxic strategies, which are directly applied by bispecific T-cell activator-armed oncolytic viruses, act independently of MHC class I expression by tumor cells and, therefore, can be used to kill cancer cells even when tumor cells have lost MHC expression.

In summary, the present study demonstrates that encoding a bispecific T cell activator in EnAd provides a particularly promising strategy to achieve targeted expression in disseminated tumors, taking advantage of the known blood stability and systemic bioavailability of the virus, which has already been investigated in several early phase clinical trials. Notably, in a study where the virus was administered intravenously several days prior to resection of a primary colon cancer, subsequent immunohistological assessment of tumor sections revealed that the virus had reached sites across the tumors and yielded strong intranuclear hexon signals, indicating successful infection and viral replication selectively in tumor cells. This supports preclinical data (Di et al., 2014; Illingworth, 2017) indicating that this virus is stable in 100% of human blood and should be capable of tumor-directed infection of disseminated and metastatic malignancies in human patients. Bispecific T cell activators can be encoded by EnAd without loss of oncolytic virulence (Fig. 51B), reflecting the remarkable ability of the virus to package the transgene. The presence of the transgene will not affect the physicochemical properties of the viral particles, so the modified viruses should exhibit exactly the same clinical pharmacokinetics as the parent agent and should be able to express the encoded bispecific T cell activator selectively in tumors throughout the body. This provides an exciting and potentially highly effective new approach to systemic targeted cancer immunotherapy that should now be prioritized for clinical evaluation.

Example 28.

Immunosuppression of human T-cell activation and target cell cytotoxicity by patient-derived malignant neoplasm exudates.

Malignancy exudates represent a potential immune tolerance environment with suppressed immune responses, which is commonly observed in patients with advanced metastatic cancer. IL-10, considered an anti-inflammatory cytokine, was measured in normal serum or patient malignancy exudates (A, peritoneal ascites; P, pleural effusion) using the Human IL-10 ELISA MAX Kit (Biolegend, 430604). IL-10 levels in exudates (88.1-633.4 pg/mL) were significantly higher than those measured in normal serum (7.2-10 pg/mL). See Fig. 58.

The ability of CD3/CD28 beads (Gibco, 11161D) to activate PBMC T cells in the presence of normal serum, ascites, or pleural fluid was examined. Human PBMC T cells (100,000 cells/well in a 96-well plate) were treated with CD3/CD28 beads (following the manufacturer's instructions) in normal patient serum or exudate (50%). As a negative control, T cells were left untreated in each fluid. After 24 h of culture, cells were harvested, and the expression levels of CD69 and CD25 on CD3+ T cells were analyzed by antibody staining and flow cytometry and presented as a percentage of double-positive cells (CD69+ CD25+) (Fig. 59). In normal serum, anti-CD3/CD28 microspheres yielded approximately 60% of T cells double-positive for both CD25 and CD69, whereas the presence of ascites fluid attenuated T cell activation in 6/12 fluids.

In a similar experiment, 100,000 T cells were exposed to CD3/CD28 beads in the presence of normal serum, ascites, or pleural fluid (50%). Anti-CD107a or isotype control antibody was added directly to the culture medium. After 1 h, monensin (BD Golgistop, BD Biosciences) was added according to the manufacturer's instructions. After an additional 5 h, cells were harvested and analyzed by flow cytometry to determine degranulation (Fig. 60). In normal serum, anti-CD3/CD28 beads yielded approximately 22.5% degranulated T cells, whereas the presence of ascites fluid attenuated T cell activation in 10/12 fluids. The level of degranulation was significantly correlated (Pearson's r = -0.7645; p = 0.0038) with the amount of IL-10 in each fluid (Fig. 61).

In a similar experiment, 75,000 T cells were co-cultured with 15,000 SKOV3 and EpCAM in the presence of normal serum, ascites, or pleural fluid (50%). T cells were treated

- 77 043895 control bispecific T cell activator in each fluid as a negative control. After 24 h of culture, cells were harvested and the expression levels of CD69 and CD25 on CD3+ T cells were then analyzed by antibody staining and flow cytometry and presented as a percentage of double-positive cells (CD69+ CD25+) (Fig. 62). In normal serum, the EpCAM bispecific T cell activator contributed approximately 67.6% of T cells double-positive for both CD25 and CD69, whereas the presence of ascites fluid attenuated T cell activation in 0/12 fluids and slightly induced activation in 4/10 fluids.

In a similar experiment, 75,000 T cells were cocultured with 15,000 SKOV3 and EpCAM in the presence of normal serum, ascites, or pleural fluid (50%). T cells were treated with control bispecific T cell activator in each fluid as a negative control. Anti-CD107a or isotype control antibody was added directly to the culture medium. After 1 h, monensin (BD Golgistop, BD Biosciences) was added according to the manufacturer's instructions. After an additional 5 h, cells were harvested and analyzed by flow cytometry to determine degranulation (Fig. 63). In normal serum, EpCAM bispecific T cell activator-bearing beads yielded approximately 41.4% degranulated T cells, whereas the presence of ascites fluid attenuated T cell activation in 2/12 fluids.

The ability of EnAd-SA-EpCAM Bisp.activ.T-cells and EnAd-SA-Control Bisp.activ.T-cells to induce T-cell lysis of target cells in cancer exudate fluids was assessed using xCELLigence technology. SKOV cells were seeded in a 48-well E-plate at ¹⁰⁴ cells/well, respectively. Plates were incubated for 18 h, 37°C, 5% _CO2 , before cells were either infected with 100 viral particles per cell (ppc) or left uninfected. After 2 h, PBMC T cells (5:1) in normal serum or patient exudate fluid (final concentration, 50%) were added. xCELLigence was used to measure target cell cytotoxicity every 10 min (Fig. 64). The results show that bispecific T cell activator-mediated lysis of SKOV3 by T cells is independent of the fluid used. Unpurified ascites cells (therefore unaltered after preparation) were seeded at 100,000 cells/well of a 96-well flat-bottomed plate in RPMI or ascites fluid. Cells were treated with EpCAM or control bispecific T cell activator, with untreated wells serving as a negative control. After incubation at 37°C for 24 h, cells were harvested and the expression levels of CD25 and CD69 on CD3 cells were determined (Fig. 65). The results demonstrate that bispecific T cell activator to EpCAM resulted in a significant increase in T cell activation (CD69/CD25 double positive) of tumor-associated lymphocytes, with little increase by ascites fluid.

In a similar experiment, crude ascites cells (therefore unaltered after preparation) were seeded at 100,000 cells/well of a 96-well flat-bottomed plate in RPMI or ascites fluid. Cells were treated with EpCaM, control bispecific T cell activator viruses, or recombinant bispecific T cell activator viruses (100 vp/cell), with untreated wells serving as a negative control (Fig. 66B). After incubation at 37°C for 5 days, the total cell population was harvested and the number of CD3+ cells (Fig. 66A) and the level of CD25 expression on CD3 cells (Fig. 66B) were determined, as well as the number of endogenous EpCaM ⁺ cells determined by flow cytometry (Fig. 66C). Total cell counts per well were determined using Precision counting beads. The results show that EpCAM bispecific T cell activator and EnAd expressing EpCAM bispecific T cell activator result in significant increases in T cell activation (CD3, CD25) of tumor-associated lymphocytes and cytotoxicity of EpCAM ⁺ cells in both RPMI and ascites fluid.

As a continuation of the experiment described above, six more patient exudate samples (7 in total) were processed identically in ascitic fluid (Fig. 67) and the CD3+ (Fig. 67A), CD25 T cell expression (Fig. 67B) and EpCAM ⁺ cell numbers (Fig. 67C) were determined by flow cytometry. The results show that EpCAM bispecific T cell activator and EnAd expressing EpCAM bispecific T cell activator reproducibly result in significant increases in T cell activation (CD3, CD25) of tumor-associated lymphocytes and EpCAM ⁺ cell cytotoxicity in a number of exudate biopsy samples.

Example 29.

A bispecific FAP T cell activator mediates T cell activation and killing of FAP+ cells by diverse donor T cells.

In other experiments, the methods described in Example 2 were used to further evaluate the T cell activating properties of the recombinant FAP bispecific T cell activator protein tested in NHDF and T cell cocultures, compared to a control bispecific T cell activator and polyclonal T cell activation using anti-CD3/CD28 Dynabeads. Supernatants taken after 24 h of culture were tested by ELISA for IFNy (Fig. 68A) and with a cytokine bead array (Human T helper cytokine panel

- 78 043895

LEGENDplex, BioLegend #74001) for a panel of cytokines (Fig. 68B). The control bispecific T cell activator did not significantly change any cytokine, but the FAP bispecific T cell activator resulted in strong increases in IFN-gamma, IL-2, TNFa, IL-17, and IL10, consistent with the different T cell subsets stimulated, and IFNy production was much higher than that induced by anti-CD3/CD28.

Stimulation with the FAP bispecific T cell activator, but not the control bispecific T cell activator, in the presence of NHDF cells also induced rapid degranulation (within 6 h) of both CD4+ and CD8+ T cell subsets, as assessed by the externalization of CD107a/LAMP1 on the T cell surface (by flow cytometry), which strongly correlates with their ability to kill target cells (Fig. 69A and B). This induction of degranulation by the FAP bispecific T cell activator translated into potent fibroblast lysis (Fig. 69C) as measured by LDH release after 24 h of co-culture with PBMC T cells (EC50 ~ 2.5 ng/ml) when induced T cell activation and cytotoxicity was observed using 6/6 donor T cells (Fig. 69D). The control bispecific T cell activator did not induce any cytotoxicity, consistent with the T cells remaining inactivated.

Example 30.

Effect of bispecific T-cell activator to FAP and EnAd-FAP - Bis.activ.T-cell viruses on cells of primary ascites samples of malignant neoplasms from different patients.

As a continuation of the studies described in Example 16, fresh primary peritoneal ascites of malignant neoplasms were obtained from other cancer patients to study the activity of the EnAd virus with the bispecific T-cell activator kFAP.

Three patient samples containing both EpCAM ⁺ tumor cells and FAP+ fibroblasts were expanded ex vivo, and mixed (adherent) cell populations were co-cultured with PBMC-derived T cells and unmodified or bispecific T cell activator-expressing EnAd viruses. After 72 h, total cells were harvested, and the number of FAP+ (Fig. 70A) and EpCAM ⁺ (Fig. 70B) cells was determined by flow cytometry. In addition, T cell activation status (as measured by CD25 expression) was measured (Fig. 70C). Infection with both EnAd-CMV-FAP Bisp.activ.T cells and EnAd-SA-FAP Bisp.activ.T cells induced T cell activation and depletion of FAP+ cells in all patient samples, without significantly altering EpCAM ⁺ tumor cell levels. Parental EnAd or control viruses did not induce observable T cell activation, with FAP+ cell numbers remaining similar to uninfected controls.

Importantly, this depletion in FAP+ fibroblasts consistently resulted in a strong reduction in the levels of the immunosuppressive cytokine TGFe detected in the supernatants (Fig. 70D).

In a second set of experiments, total (and crude) cells from five patient biopsies were assessed to evaluate endogenous tumor-associated T cell activity in the samples. Cells were seeded in 50% ascites fluid and treated with recombinant control or FAP bispecific T cell activator proteins or EnAd or EnAd-bispecific T cell activator viruses at 100 vp/cell. After 5 days of incubation, T cell activation (as measured by CD25 expression) and residual FAP+ cell numbers were measured by flow cytometry (Fig. 7A and B). In all 3 patient samples, recombinant FAP bispecific T cell activator and EnAd-CMV-FAP Bisp.activ.T-cell induced robust T cell activation, with up to ~80% of patient-derived T cells being activated, causing marked depletion of FAP+ fibroblasts. Interestingly, EnAd-SA-FAP-Bisp.activ.T-cells induced CD25 expression in 2/3 of the samples without visible activation or depletion of FAP+ cells in patient 1. This is likely due to the lack of tumor cells present to be infected by the virus and produce the bispecific T cell activator protein (no tumor EpCAM+ cells were detected in this sample by flow cytometry), consistent with the requirement for tumor cells to express the MLP(SA)-driven transgene (this would also likely explain the lack of T cell activation and FAP+ cell depletion by the EnAd-SA-FAP-Bisp.activ.T cell virus in the patient's ascites sample shown in Figs. 42-44). Taken together, these data demonstrate that EnAd, expressing a bispecific T cell activator to FAP, can, following infection of tumor cells, reproducibly lead to the activation of tumor-associated T cells to kill endogenous fibroblasts.

In another experiment, we examined whether the activity of a bispecific FAP T cell activator could be enhanced by blocking the PD-1 checkpoint using a patient biopsy specimen in which T cells were 73.6% PD-1 positive and FAP+ cells were 62.9% PDL1 positive (Fig. 72A). Co-cultures similar to those described above were performed in the presence or absence of purified mouse anti-human PDL1 IgG2b blocking antibody (BioLegend clone 29E.2A3) at a final concentration of 2.5 μg/mL. After 2 days of culture, all cells were harvested and residual FAP+ cells and T cell activation were measured. Inclusion of a blocking anti-PDL1 antibody resulted in a modest increase in CD25 induction (Fig. 72B) and a two-fold increase in IFNy production (Fig. 72C) without altering FAP+ cell depletion (Fig. 72D), with nearly complete lysis by day 2 in either setting.

Tumor-associated lymphocytes (TALs) isolated from ascites of a patient with ovarian cancer have been reported to have increased PD-1 expression and impaired effector functions, including cytotoxicity and IFNg production. Consistent with this, PD-1 expression was 2-fold higher on CD3+ cells from six ascites biopsies of cancer patients than on peripheral blood mononuclear cells (PBMCs) from three healthy donors (Fig. 73A). To assess T cell functionality in these cancer biopsy samples, NHDF cells and crude PBMCs or ascites cells (% CD3+ cells for each sample are shown in Fig. 73B) were co-cultured with control or FAP Bisp.activ.T-cell-containing supernatants, and the supernatants were collected after 5 days and tested for IFNg by ELISA (Fig. 73C). The control bispecific T cell activator did not induce IFNy. Three ascites cell samples produced IFNy at levels similar to PBMC samples, while the other three showed an attenuated response to the bispecific T cell activator to FAP. We next examined the ability of these T cells to induce bispecific T cell activator-mediated lysis of NHDF cells. NHDF were plated, PBMC or ascites cells were added along with bispecific T cell activator-containing supernatants, and cell viability in culture was monitored in real time using the xCELLigence cytotoxicity assay system. Despite the variability in IFNγ production, all ascites samples induced complete cytotoxicity of NHDF cells when added with FAP bispecific T cell activator, with the overall rate of NHDF bispecific T cell activator-mediated lysis being similar to that observed with PBMCs (Fig. 73D). To investigate whether FAP bispecific T cell activator could mediate T cell activation in patient cancer exudate samples (all at 50%), PBMC T cells were activated with control or FEP bispecific T cell activator in the presence of NHDF cells or activated with anti-CD3/CD28 Dynabeads, either in 50% normal human serum (NS) or in other (acellular) cancer exudate samples. While 74% of T cells in normal serum were activated (double positive for both CD25 and CD69) 24 h after stimulation with anti-CD3/CD28 microspheres, 3/5 of the ascites fluid samples tested significantly attenuated T cell activation compared to the response in NS (normal serum) (Fig. 74A). However, when PBMCs were cultured with NHDF and stimulated with the FAP bispecific T cell activator, no significant suppression of T cell activation was observed in the presence of either exudate fluid (Fig. 74B), demonstrating that the FAP bispecific T cell activator can overcome immunosuppressive mechanisms to activate T cells.

Example 31.

EnAd-FAP Bisp.activ.T-cell-mediated oncolysis and T-cell stimulation polarize CD11b+ TAMs from patients with ascites toward a more activated phenotype.

To examine whether FAP bispecific T cell activator-mediated Th1 cytokine production, including IFNy, TNFa, and IL-2, and subsequent depletion of FAP+ fibroblasts (and associated decrease in TGFe1) are associated with other shifts in the tumor microenvironment away from immunosuppressive and pro-tumorigenic effects toward anti-tumor activity, the effects on tumor-associated macrophages (TAMs) were assessed in unseparated ascites cells. Whole bulk unpurified patient ascites cells were seeded in 50% ascites fluid and treated with free control or FAP bispecific T cell activator or infected with EnAd-SA-control Bisp.activated T cells or EnAd-SA-FAP Bisp.activated T cells virus (100 vp/cell). In parallel, some cells were treated with IFNy to induce the activated CD11b myeloid cell phenotype. After 3 days of incubation, T cell activation status was first determined; CD25+ cells were measured by flow cytometry and IFNy secretion by ELISA.

Treatment with FAP bispecific T cell activator and EnAd-SA-FAP Bisp.activ.T cells resulted in approximately 60% of CD3+ T cells being converted to CD25+ (Fig. 75A) and high amounts of IFNy in the culture supernatants (Fig. 75B). No increase over background with control bispecific T cell activator or control virus was observed in CD25 or IFNy expression. To assess TAM polarization, expression levels of CD64 and CD86 (M1 or activated macrophage markers) and CD206 and CD163 (M2 or TAM markers) were measured on CD11b+ cells by flow cytometry (Fig. 75C). Treatment with free FAP bispecific T cell activator or EnAd expressing FAP bispecific T cell activator induced a more activated phenotype, as evidenced by a significant increase in CD64 expression and a strong decrease in CD206 and CD163 - similar to what is observed when IFNy is added at measured levels to the cultures. While in this experiment treatment with free FAP bispecific T cell activator or control virus did not cause an obvious change in CD86 above background levels, EnAd expressing FAP bispecific T cell activator induced a significant

- 80 043895 increased CD86 expression, indicating that EnAd virus infection and FAP bispecific T cell activator activity can synergistically activate primary myeloid cells in suppressive tumor microenvironments, such as the ascetic fluid samples from malignancies tested in the present invention. In this study, IFNy treatment induced a moderate decrease in CD86, indicating that the robust increase in CD86 observed by EnAd-SA-FAP Bispecific T cell activator may occur via an IFNY-independent mechanism.

Example 32.

EnAd-FAP Bisp.activ.T-cell activates tumor-infiltrating lymphocytes and induces cytotoxicity in solid prostate tumor biopsies ex vivo.

Tissue section cultures represent one of the most realistic preclinical models of various tissues, organs, and tumors. To evaluate the activity of viruses expressing a bispecific T cell activator for FAP in this highly clinically relevant setting, several paired core biopsies of malignant and benign prostate tissue from resected human prostates were studied. Initial screening reproducibly showed that prostate tissue contained annular rings of tumor-associated EpCAM+ cells (Fig. 76A) interspersed with large areas of stroma containing scattered CD8 T cells (Fig. 76B). FAP staining was detected on fibroblasts adjacent to tumor areas (Fig. 76C).

Organ cores were vibratome-cut to 300 μm thickness and section cultures were prepared in the presence of virus (1.5 x ^{10 9} vp/section) or left uninfected. After 7 days, preparations were fixed, embedded in paraffin, sectioned, and assessed for T-cell activation by immunohistochemistry (IHC) staining for CD25 expression (Fig. 76D). Only samples treated with EnAd-CMV-FAP Bisp.activ.T cells or EnAd-SA-FAP Bisp.activ.T cells showed tumor-infiltrating T cell activation as evidenced by strong CD25 staining. Neither untreated nor control virus-treated samples had detectable CD25-positive cells. Supernatants from these slice cultures taken at 4 and 7 days post-infection were tested for IFNy and IL-2 by ELISA, with increased IFNy detected in malignant but not benign prostate slice cultures infected with FAP bispecific T cell activator virus (Fig. 76E) and increased IL-2 detected in EnAd-SA-FAP Bisp.activ.T cell cultures (Fig. 76F). EnAd-SA-FAP Bisp.activ.T cell induced higher amounts of IFNy that became detectable earlier than CMV-driven FAP bispecific T cell activator virus.

Example 33. EnAd viruses expressing a bispecific T-cell activator to EpCAM or FAP.

Five viruses were created (NG-611, NG-612, NG-613, NG-614, NG-617) that encode a single bispecific T-cell activator (Table 8).

Table 8

Virus ID Transgenic cassette NG-611 (SEQ ID NO: 96) 55A ¹ -ErSashBisp.active.T-cl. ² -N15 ³ -RA ⁴ NG-612 (SEQ ID NO: 97) BBA^EARBisp.activ.T-clAEPzZ-RA ⁴ NG-613 (SEQ ID NO: 98) 5A ⁶ -EARBisp.activ.T-cl. ⁵ -H15 ³ -RA ⁴ NG-614 (SEQ ID NO: 99) 5A ⁶ -EARBisp.activ.T-cl. ⁷ -N15 ³ -RA ⁴ NG-617 (SEQ ID NO: 100) BBA^EARBisp.activ.T-CLARA ⁴

¹ SEQ ID NO: 55; ² SEQ ID NO: 83; ³ SEQ ID NO: 84; ⁴ SEQ ID NO: 65; ⁵ SEQ ID NO: 85; ⁶ SEQ ID NO: 86; ⁷ SEQ ID NO: 87.

In each transgene cassette, the cDNA encoding the bispecific T cell activator was flanked at the 5' end by either a short splice acceptor sequence (SSA, SEQ ID NO: 55) or a longer splice acceptor sequence (SA, SEQ ID NO: 86). At the 3' end of the bispecific T cell activator, the SV40 late poly(A) sequence (PA, SEQ ID NO: 65) was encoded preceded by either a histidine tag (HIS, SEQ ID NO: 41) or no tag. In viruses NG-611, NG-612, NG-613 and NG-617, a single-chain version of the mouse monoclonal anti-human CD3ε antibody OKT3 was used in the anti-CD3 portion of the bispecific T cell activator molecule.

Getting a virus.

Plasmid pEnAd2.4 was used to generate plasmids pNG-611, pNG-612, pNG-613, pNG-614 and pNG-617 by direct insertion of synthesized transgene cassettes (SEQ ID NOs: 88-92, respectively). Transgene cassette pNG-611 encodes EpCam-targeted bispecific T cell activator (SEQ ID NO: 93), transgene cassettes pNG-612, pNG-613 and pNG-617 encode FAP-targeted bispecific T cell activator (SEQ ID NO: 94), and transgene cassette pNG-614 encodes FAP-targeted bispecific T cell activator with SEQ ID NO: 95. Schematics of transgene cassettes are shown.

- 81 043895 in Fig. 77A-C. Construction of plasmid DNA was confirmed by restriction analysis and DNA sequencing.

Plasmids pNG-611, pNG-612, pNG-613, pNG-614, and pNG-617 were linearized by restriction digestion with AscI to obtain viral genomes. Viruses were amplified and purified according to the methods described below.

Digested DNA was purified by phenol/chloroform extraction and precipitated for 16 h at 20°C in 300 μl molecular biology grade ethanol >95% and 10 μl 3 M sodium acetate. The precipitated DNA was pelleted by centrifugation at 14,000 rpm for 5 min and washed in 500 μl 70% ethanol before centrifugation again at 14,000 rpm for 5 min. The pure DNA pellet was air dried, resuspended in 500 μl OptiMEM containing 15 μl Lipofectamine transfection reagent, and incubated for 30 min at room temperature. The transfection mixture was then added dropwise into a T-25 flask containing 293 cells grown to 70% confluence. After incubating the cells with the transfection mixture for 2 h at 37°C, 5% _CO2 , 4 ml of cell medium (high glucose DMEM with glutamine supplemented with 2% FBS) were added to the cells and the flasks were incubated at 37°C, 5% _CO2 .

Transfected 293 cells were monitored every 24 h and additional media were added every 48-72 h. Virus production was monitored by observing a significant cytopathic effect (CPE) in the cell monolayer. Once intense CPE was observed, the virus was harvested from the 293 cells by three freeze-thaw cycles. The harvested viruses were used to re-infect 293 cells to amplify viral material. Production of viable virus during amplification was confirmed by observing significant CPE in the cell monolayer. After observation of CPE, the virus was harvested from the 293 cells by three freeze-thaw cycles. The amplified viral material was used for further amplification and then the viruses were purified by cesium chloride double banding to obtain purified viral material.

Viral activity assessed by qPCR.

A549 cells either infected for 72 h with 1ppc NG-611, NG-612, NG-617, enadenotucirev viruses or left uninfected were used for quantitative determination of viral DNA by qPCR. Cell supernatants were collected and clarified by centrifugation for 5 min at 1200 rpm. DNA was extracted from 45 μl of supernatant using the Qiagen DNeasy kit according to the manufacturer's protocol. Enadenotucirev virus particles (2.5x1010-2.5x105 vp), also generated and extracted using the DNeasy kit, were used for the standard curve. Each extracted sample or standard was analyzed by qPCR using a viral gene-specific primer-probe for the early gene E3.

Quantification of the number of detected viral genomes per cell demonstrated that NG-611, NG-612, and NG-617 showed significant genome replication in A549 cell lines (Fig. 77D). This result was similar for all viruses tested, including the parental enadenotucirev virus, indicating that inclusion of the bispecific T cell activator transgene does not affect viral replicative activity. No viral genomes were detected in uninfected cells (data not shown).

T cell activation and degranulation mediated by viruses expressing a bispecific T cell activator.

Infection of carcinoma cells.

A549 cells were seeded in 24-well plates at a density of 2.5 x 105 cells/well. Plates were incubated for 4 h at 37°C and 5% _CO2 , then cells were infected with 1ppc NG-611, NG-612, enadenotucirev or left uninfected. At 24, 48 or 72 h post infection, supernatants were collected from the cells, cleared by centrifugation for 5 min at 1200 rpm and snap frozen.

T cell analysis.

FAP-expressing lung fibroblast cell lines MRC-5 or EpCam-expressing ovarian carcinoma cells SKOV3 were seeded in 48-well plates at densities of 5.7x104 cells/well and 1.2x105 cells/well, respectively. Plates were incubated for 4 h, 37°C, 5% _CO2 , then the medium was replaced with 150 μl/well of thawed supernatant collected from A549 plates. Purified CD3 T cells isolated from donor human PBMCs were then also added to the plates to give a 2:1 ratio of T cells to MRC-5 or SKOV3. Co-cultures were incubated for 16 h, 37°C, 5% CO ₂ , then cell supernatants were collected for ELISA analysis and T cells were harvested for flow cytometric analysis. Culture media containing non-adherent cells were removed from the co-culture wells and centrifuged (300 x g). The supernatant was carefully removed, diluted 1 in 2 with PBS 5% BSA and stored for ELISA analysis. Adherent cell monolayers were washed once with PBS and then detached with trypsin. Trypsin was inactivated with RPMI medium with 10% FBS and the cells were added to the cell pellets that had been previously

- 82 043895 were harvested from culture supernatants. Cells were centrifuged (300xg), the supernatant was discarded and the cell pellet was washed in 200 μl PBS. Cells were centrifuged again then resuspended in 50 μl PBS-containing Live/Dead Aqua (Life tech) for 15 min at room temperature. Cells were washed once in FACs buffer before staining with panels of directly conjugated antibodies: anti-CD3 conjugated to AF700; anti-CD25 conjugated to BV421; anti-HLA-DR conjugated to PE/CY5; anti-CD40b conjugated to BV605; anti-CD69 conjugated to PE and anti-CD107a conjugated to FITC. A sample of cells from each co-culture situation was also stained with the appropriate isotype control antibodies. All staining was performed in FACs buffer in a total volume of 50 μl/well for 15 min, 4°C. Cells were then washed twice with FACs buffer (200 μl), then resuspended in 200 μl FACs buffer and analyzed by flow cytometry (Attune).

Increased expression of T cell activation markers.

Flow cytometric analysis of T cell activation was assessed by expression of the T cell activation markers CD25, CD69, HLA-DR, and CD40L or the T cell degranulation marker CD107a on live single cells. These data showed that when cocultured with EpCam+ SKOV3 cells, the number of T cells expressing cell surface CD25, CD69, HLA-DR, CD40L, or CD107a was significantly increased by the addition of NG-611 supernatants to the cells compared with NG-612, enadenotucirev, or untreated control supernatants (Fig. 78). For all of these markers, supernatants from infected A549 cells stimulated little T cell activation at 24 h, but at 48 h postinfection, supernatants stimulated significant T cell activation for all markers. This also occurred 72 h after infection.

When co-cultured with FAP+ MRC-5 cells, the number of T cells expressing cell surface CD25, CD69, HLA-DR, CD40L, or CD107a was significantly increased by the addition of NG-612 supernatants to the cells compared to NG-611, enadenotucirev, or untreated control supernatants (Fig. 79). Some T cell activation could also be observed with NG-611 virus, likely related to the low but detectable EpCam expression (~5%) on MRC-5 cell lines harboring the bispecific EpCam T cell activator expressed by NG-611 virus (Fig. 80). For all these markers, little T cell activation was stimulated by supernatants from A549 cells infected for 24 h, however, by 48 h post infection, supernatants stimulated significant T cell activation for all markers. Markers CD25 and CD69 were also activated after incubation with supernatants collected at 72 h post infection, however, the activation markers HLA-DR, CD40L and CD107a were detected at lower levels with supernatants collected at 72 h post infection than at 48 h post infection. This may be due to the high level of bispecific T cell activator present at this later stage of infection leading to rapid and potent T cell activation, meaning that effector functions need to be measured at time points earlier than 16 h after incubation with supernatants.

To detect IFNy expression, co-culture supernatants were diluted in 5% BSA/PBS assay buffer (ranging from 1:10 to 1:1000) and ELISA was performed using the Human IFN gamma Quantikine ELISA kit (R&D systems) according to the manufacturer's protocol. The concentration of secreted IFNy was determined by interpolation from the standard curve. IFNy expression could only be detected in co-culture supernatants using NG-611 on SKOV3 cells (Fig. 81A) or NG-611, NG-612 on MRC-5 cells (Fig. 81B).

Example 34: Immune activation and antitumor efficacy of viruses expressing a bispecific T cell activator in vivo.

Humanized hematopoietic stem cells from CD34+ NSG mice (Jackson Labs) were implanted with HCT116 tumor cells subcutaneously on both flanks of the animals 18 weeks after engraftment. Once tumors reached 80-400 ^mm3 , mice were grouped so that each treatment group had an equivalent distribution of tumor volumes, 7 mice per group. Mice were injected intravenously with saline, enadenotucirev, or NG-611 at 5x109 particles per injection, 2 injections per tumor. Tumors on both flanks were treated. Tumor volume measurements 34 times per week showed that NG-611 treatment resulted in a significant antitumor response up to 20 days post-dose compared with enadenotucirev or untreated controls (Fig. 82a). At 20 days post-dose, one tumor from 4 mice in each group was processed for flow cytometric analysis, and the remaining tumors were frozen on dry ice.

Flow cytometry.

Tumor samples were mechanically disaggregated immediately after resection in a small volume of RPMI medium. Disaggregated tumors were then passed through a 70 μm cell strainer and centrifuged at 300 g for 10 min. Cell pellets were resuspended in 100 μl PBS-containing Live/Dead Aqua (Life tech) for 15 min on ice. Cells were washed once in FACs buffer (5% BSA PBS) before staining with a panel of directly conjugated antibodies: anti-CD8

- 83 043895 (RPA-T8, AF700); aHTu-CD4 (RPA-T4, PE); aHTu-CD45 (2D1, APC-Fire 750); aHTu-CD3 (ОКТ3, PerCPСу5.5); anti-CD25 (М-А251, PE-Dazzle 594); anti-CD69 (FN50, АРС); anti-HLA-DR (L243, BV605); anti-CD107а (Н4А3, FITC). The pool of tumor cell suspensions was also stained with the corresponding control isotype antibodies. All staining was performed in FACs buffer in a total volume of 50 μl/well for 20 min at 4°C. Cells were washed three times with FACs buffer (200 μl) before resuspension in 200 μl FACs buffer and flow cytometric analysis (Attune). FACs analysis showed that the tumor CD8 to CD4 T cell ratio was significantly increased in NG-611-treated tumors compared to controls treated or not with enadenotucirev virus (Fig. 82b).

Example 35. EnAd viruses coexpressing a bispecific T-cell activator to FAP and immunomodulatory cytokines and chemokines.

Three viruses (NG-615, NG-640, and NG-641) were generated that encoded a bispecific FAP T-cell activator and immunomodulatory proteins (Table 9).

Table 9

Virus ID Transgenic cassette NG-615 (SEQ ID NO: 101) 55A ¹ -EARBisp.activ.T-cl. ² -E2A ³ -EISE ⁴ -P2A ⁵ -M1P1a ⁶ -T2A ⁷ - IFNa ⁸ -PA ⁹ NG-640 (SEQ ID NO: 102) 55A1-EARBisp.activ.T-cl. ² -Р2А ⁵ -СХСЭ101°-Т2А ⁷ -СХСЭ911-RA ⁶ NG-641 (SEQ ID NO: 103) 55A1-EARBisp.activ.T-cl.5-P2A5-CXCE1O1o-T2A ⁷ -CXCE9i- E2A ³ -IFNa ⁸ -PA ⁶ NG-615 (SEQ ID NO: 298) 5A ¹² -RARB. active.T-cl. ² -E2A ³ -RIZE ⁴ -P2A ⁵ -M1P1a ⁶ -T2A ⁷ - IFNa ⁸ -PA ⁹

¹ SEQ ID NO: 55; ² SEQ ID NO: 87; ³ SEQ ID NO: 63; ⁴ SEQ ID NO: 105; ⁵ SEQ ID NO: 61; ⁶ SEQ ID NO: 107; ⁷ SEQ ID NO: 64; ⁸ SEQ ID NO: 109; ⁹ SEQ ID NO: 65; ¹⁰ SEQ ID NO: 110; ¹¹ SEQ ID NO: 111; ¹² SEQ ID NO: 86.

Getting a virus.

Plasmid pEnAd2.4 was used to generate plasmids pNG-615, pNG-616, pNG-640, and pNG-641 by direct insertion of the synthesized transgene cassettes (SEQ ID NOs: 112-114, respectively). NG-615 and NG-616 contain four transgenes encoding FAP-targeted bispecific T cell activator (SEQ ID NO: 94), Flt3L (SEQ ID NO: 115), MIP1a (SEQ ID NO: 116), and IFNa (SEQ ID NO: 117). NG-640 and NG-641 encode a FAP-targeted bispecific T cell activator (SEQ ID NO: 94), CXCL9 (SEQ ID NO: 118), and CXCL10 (SEQ ID NO: 119), NG-641 also contains a fourth transgene encoding IFNa (SEQ ID NO: 117). Schematics of the transgene cassettes are shown in Fig. 83A-C. Plasmid DNA construction was confirmed by restriction analysis and DNA sequencing.

Plasmids pNG-615, pNG-616, pNG-640, and pNG-641 were linearized by restriction digestion with AscI to obtain the virus genomes. The viruses were amplified and purified according to the methods detailed in Example 33.

Viral activity assessed by qPCR and transgenic ELISA.

Infection of carcinoma cells.

A549 cells infected for 72 h with 1 ppc of NG-615, enadenotucirev, or left uninfected were used for quantitative determination of viral DNA by qPCR and analysis of transgene expression by ELISA. Cell supernatants were collected and clarified by centrifugation for 5 min at 1200 rpm. Then, 45 μl of supernatant was used for DNA analysis, and the remaining supernatant was used for ELISA.

qPCR.

DNA was extracted from the supernatant sample using the Qiagen DNeasy kit according to the manufacturer's protocol. A standard curve was prepared using enadenotucirev virus particles (2.5 x 10 ¹⁰ -2.5 x ^{10 5} pbp), also generated and extracted using the DNeasy kit. Each extracted sample or standard was analyzed by qPCR using a viral gene-specific E3 early gene primer-probe. Quantification of the number of viral genomes detected per cell demonstrated that NG-615 exhibits significant genome replication in A549 cell lines at a level similar to that of the parental enadenotucirev virus (Fig. 84). These data indicate that inclusion of the bispecific T cell activator and three immunomodulatory transgenes does not significantly affect the replicative activity of the virus. No viral genomes were detected in uninfected cells.

IFA.

ELISA for IFNa was performed using the Verikine Human IFNa kit (Pbl assay science), ELISA for MIP1a was performed using the Human CCL3 Quantikine ELISA kit (R & D systems), and ELISA

- 84 043895 for Flt3L was performed using the Flt3L human ELISA kit (Abcam).

All analyses were performed according to the manufacturer's protocol.

Concentrations of secreted IFNa, MIP1a or FLt3L were determined by interpolation from standard curves. Expression of IFNa, MIP1a and Flt3L could be detected in the cell supernatant.

NG-615, but not enadenotucirev or untreated control cells (Fig. 85).

T cell activation and degranulation mediated by viruses expressing a bispecific T cell activator.

Infection of carcinoma cells.

A549 cells were seeded in 24-well plates at a density of 2.5 x 105 cells/well. The plates were incubated for 4 h at 37°C and 5% _CO2 , then the cells were infected with 1ppc NG-612, NG-615, enadenotucirev or left uninfected. At 24, 48 or 72 h post infection, the supernatants were collected from the cells, cleared by centrifugation for 5 min at 1200 rpm and snap frozen.

T cell analysis.

FAP-expressing MRC-5 lung fibroblast cell lines were seeded in 48-well plates at densities of 5.7 x 104 cells/well. The plates were incubated for 4 h, 37°C, 5% _CO2 , then the medium was replaced with 150 μl/well of thawed supernatant collected from A549 plates. Purified CD3 T cells isolated from human PBMC donors were also added to the plates to give a 2 to 1 T cell to MRC-5 ratio. The co-cultures were incubated for 16 h, 37°C, 5% _CO2 , then the cell supernatants were collected for ELISA analysis and the T cells were collected for flow cytometric analysis according to the methods described in Example 29.

Increased expression of T cell activation markers.

Flow cytometric analysis of T cell activation was assessed by expression of the T cell activation markers CD25, CD69, HLA-DR, and CD40L or the T cell degranulation marker CD107a on live, CD3+, single cells. These data showed that when co-cultured with FAP+ MRC5 cells, the number of T cells expressing CD25, CD69, HLA-DR, CD40L, or CD107a was significantly increased by the addition of NG-615 or 612 supernatants to the cells compared with enadenotucirev or untreated control supernatants (Fig. 8b). Secretion of the stimulatory cytokine IFNy

To detect IFNy expression, co-culture supernatants were diluted in 5%BSA/PBS assay buffer (ranging from 1:10 to 1:1000) and ELISA was performed using the Human IFN gamma Quantikine kit (RandD Systems) according to the manufacturer's protocol. The concentration of secreted IFNy was determined by interpolation from the standard curve. IFNy expression could only be detected in co-culture supernatants using NG-612- or NG-615-infected A549 supernatants (Fig. 87).

Example 36. EnAd virus co-expressing a bispecific T cell activator targeting FAP and a bispecific T cell activator targeting EpCam.

The NG-618 virus was generated, which encoded two bispecific T cell activator molecules, one targeting EpCam (EpCam bispecific T cell activator) and one targeting FAP (FAP bispecific T cell activator) (Table 10).

Table 10

Virus ID

Transgene cassette

NG-618 (SEQID NO: 120)

55A ¹ -ErSAMBisp.active.T-cl. ² -P2A ³ EARBisp.active.T-clLRA ^{5 1} SEQ ID NO: 55; ² SEQ ID NO: 121; ³ SEQ ID NO: 106; ⁴ SEQ ID NO: 122; ⁵ SEQ ID NO: 65.

Getting a virus.

Plasmid pEnAd2.4 was used to generate plasmid pNG-618 by direct insertion of the synthesized transgene cassettes (SEQ ID NO: 123). The NG-618 virus contains two transgenes encoding a bispecific T cell activator targeting EpCat (SEQ ID NO: 93) and a bispecific T cell activator targeting FAP (SEQ ID NO: 95). The schematic of the transgene cassette is shown in Fig. 88. The construction of plasmid DNA was confirmed by restriction analysis and DNA sequencing.

Plasmid pNG-618 was linearized by restriction digestion with AscI to obtain viral genomes. Viruses were amplified and purified according to the methods detailed in Example 33.

T cell activation and degranulation mediated by viruses expressing a bispecific T cell activator.

Infection of carcinoma cells.

A549 cells were seeded in 24-well plates at a density of 1.2x106 cells/well. The plates were incubated for 4 h, 37°C, 5% CO ₂ , then the cells were infected with viruses NG-611, NG-612, NG- 85 043895

618, enadenotucirev or left uninfected. Supernatants were collected from the cells 72 h after infection and cleared by centrifugation for 5 min at 1200 rpm.

T cell analysis.

FAP-expressing lung fibroblast cell lines MRC-5 and EpCam-expressing A549 cells were seeded in 24-well plates at a density of 1.5 x 105 cells/well. MRC-5 and A549 cells were also mixed at a 1:1 ratio and seeded in 24-well plates at a total density of 1.5 x 105 cells/well. The plates were incubated for 4 h, 37°C, 5% _CO2 , then the medium was replaced with 300 μl/well of thawed supernatant collected from A549 plates. Purified CD3 T cells isolated from donor human PBMCs were then also added to the plates to give a 2 to 1 ratio of T cells to MRC-5 or SKOV3 cells.

Joint cultures.

incubated for 16 h, 37°C, 5% CO2, then cell supernatants were collected for ELISA analysis, and T cells, MRC-5 and A549 were collected for flow cytometric analysis.

Detection of FAP and EpSat on MRC-5 or SKOV cells.

Flow cytometric analysis of detectable FAP or EpCam on the surface of MRC-5 or SKOV cells, respectively, was assessed by washing the cells once in FACs buffer before staining with panels of directly conjugated antibodies: anti-FAP conjugated to AF647; anti-EpCam conjugated to PE. Analysis showed that FAP expression was no longer detectable on MRC-5 cells that were incubated with supernatant from cells infected with a virus expressing the FAP-targeting bispecific T cell activator, NG-618, but was detected on >80% of cells incubated with supernatants from EnAd-treated or untreated cells (Fig. 89A). These data indicate that the FAP-targeting bispecific T cell activator produced by NG-618 viruses binds to its FAP target on MRC-5 cells, interfering with anti-FAP antibody binding. Live, large, single SKOV cells were assessed for detectability of EpCam expression. EpCam expression was detectable only at low levels in SKOV cells that were incubated with supernatants from cells infected with the EpCam-targeting bispecific T cell activator virus, NG-618 (17% of cells), but was detected in >40% of cells incubated with supernatants from EnAd-treated or untreated cells (Fig. 89B). Taken together, these data demonstrate that NG-618 produces bispecific T cell activator molecules that bind to the target proteins EpCam and FAP.

Increased expression of T cell activation markers.

Flow cytometric analysis of T cell activation was assessed by expression of the T cell activation markers CD25, CD69, HLA-DR, and CD40L or the T cell degranulation marker CD107a on live, CD3+, single cells. These data showed that when co-cultured with FAP+ MRC-5 cells, the number of T cells expressing CD25, CD40L, or CD107a was significantly increased when NG-618 supernatants were added to the cells compared to enadenotucirev or untreated control supernatants (Fig. 90). The number of T cells expressing CD25, CD40L, or CD107a was also significantly increased when NG-618 supernatants were added to EpCam+ SKOV3 cells compared to enadenotucirev or untreated control supernatants (Fig. 91). These data demonstrate that both bispecific T cell activator molecules expressed by the NG-618 virus are functional in inducing T cell activation.

Analysis of T cell-mediated targeted cell killing (MRC-5 and SKOV.

For flow cytometric analysis, MRC-5 and SKOV cell viability was assessed by staining cells in 50 μl PBS containing Live/Dead Aqua (Life tech) for 15 min at room temperature. Cells were washed once in FACs buffer before staining with panels of directly conjugated antibodies: AF647-conjugated anti-FAP; PE-conjugated anti-EpCam. MRC-5 and SKOV cell viability was significantly reduced after incubation with NG-618 supernatant samples, whereas no significant cell death was detected in enadenotucirev or untreated control supernatants, Fig. 92. These data demonstrate the functional ability of co-expressed NG-618 bispecific T cell activators targeting FAP and EpCam to induce T cell-mediated killing of target cells.

- 86 043895

Sequences

SEQ ID NO:25: FAP Bisp.active.T-cl.-P2A-KEP (ITALIC = leader, BOLD = furin cleavage site, UNDERLINE = P2A sequence, lowercase = RFP)

MGWC//LFLK4TATCPH5DIVMTQSPDSLAVSLGERATINCKSSQSLLYSRNQKNYLAWYQQKPGQPPKLL IFWASTRESGVPDRFSGSGFGTDFTLTISSLQAEDVAVYYCQQYFSYPLTFGQGTKVEIKGGGGSGGGGSGGG GSQVQLVQSGAEVKKPGASVKVSCKTSRYTFTEYTIHWVRQAPGQRLEWIGGINPNNGIPNYNQKFKGRV TITVDTSASTAYMELSSLRSEDTAVYYCARRRIAYGYDEGHAMDYWGQGTLVTVSSGGGGSDVQLVQSGAE VKKPGASVKVSCKASGYTFTRYTMHWVRQAPGQGLEWIGYINPSRGYTNYADSVKGRFTITTDKSTSTAY MELSSLRSEDTATYYCARYYDDHYCLDYWGQGTTVTVSSGEGTSTGSGGSGGSGGADDIVLTQSPATLSLSP GERATLSCRASQSVSYMNWYQQKPGKAPKRWIYDTSKVASGVPARFSGSGSGTDYSLTINSLEAEDAATYY CQOWSSNPLTFGGGTKVEIKHHHHHHHHHHRRKRGSGATNFSLLKQAGDVEENPGPmselikenmhmkly megtvnnhhfkctsegegkpyegtqtmkikweggplpfafdilatsfmygskafinhtqgipdffkqsfpegftwerittyedggvltat qdtsfqngciiynvkingvnfpsngpvmqkktrgweantemlypadgglrghsqmalklvgggylhcsfkttyrskkpaknlkmpgf hfvdhrlerikeadketyveqhemavakycdlpsklghr

SEQ ID NO:26: Control (anti-FHA) Bisp.activ.T-cell-P2A-KRR (ITALIC = leader, BOLD = furin cleavage site, UNDERLINE = P2A sequence, lowercase = RFP)

MGWC//LFLK4TATCPH5ELDIVMTQAPASLAVSLGQRATISCRASKSVSSSGYNYLHWYQQKPGQPPKLLI YLASNLESGVPARFSGSGSGTDFTLNIHPVEEEDAATYYCQHSREFPLTFGAGTKLEIKSSSGGGGSGGGGGGS SRSSLEVQLQQSGPELVKPGASVKISCKTSGYTFTGYTMHWVRQSHGKSLEWIGGINPKNGGIIYNQKFQGK ATLTVDKSSSTASMELRSLTSDDSAVYYCARRVYDDYPYYYAMDYWGQGTSVTVSSAKTTPPSVTSGGGGS DVQLVQSGAEVKKPGASVKVSCKASGYTFTRYTMHWVRQAPGQGLEWIGYINPSRGYTNYADSVKGRFTI TTDKSTSTAYMELSSLRSEDTATYYCARYYDDHYCLDYWGQGTTVTVSSGEGTSTGSGGSGGSGGADDIVL TQSPATLSLSPGERATLSCRASQSVSYMNWYQQKPGKAPKRWIYDTSKVASGVPARFSGSGSGTDYSLTIN SLEAEDAATYYCOQWSSNPLTFGGGTKVEIKHHHHHHHHHRRKRGSGATNFSLLKQAGDVEENPGPm selikenmhmklyrnegtvnnhhfkctsegegkpyegtqtmkikweggplpfafdilatsfmygskafinhtqgipdffkqsfpegftw erittyedggvltatqdtsfqngciiynvkingvnfpsngpvmqkktrgweantemlypadgglrghsqmalklvgggylhcsfkttyrs kkpaknlkmpgfhfvdhrlerikeadketyveqhemavakycdlpsklghr

SEQ ID NO:33: splice acceptor sequence

CAGG

SEQ ID NO: 55 Short splice acceptor (SSA) DNA sequence (null sequence)

CAGG

Claims (21)

CLAIM 1. An oncolytic adenovirus containing the sequence of formula (I) 5'ITR-B ₁ -B _a -B ₂ -B _x -B _b -B _y -B ₃ -3'ITR (I) where B1 is a linkage or contains E1A , E1B or E1A-E1B; BA contains -E2B-L1-L2-L3-E2A-L4; B ₂ is a bond or contains E3; BX is a DNA linkage or sequence containing a restriction site, one or more transgenes, or both; BB contains L5; BY is a DNA sequence containing a restriction site, one or more transgenes, or both; B ₃ is a bond or contains E4; wherein said adenovirus at position B _Y encodes a first bispecific T cell activator that does not bind to a tumor antigen, containing at least two binding domains, wherein one of said binding domains is specific to a component of the T cell receptor (TCR) complex, and one of said binding domains of said bispecific T cell activator is specific to an antigen that is present in the tumor stroma selected from fibroblast activation protein (FAP), TREM1, IGFBP7, FSP-1, platelet-derived growth factor receptor-α (PDGFR-α), platelet-derived growth factor receptor-α (PDGFR-α), growth factor-β (PDGFR-β) and vimentin; and wherein said adenovirus is Enadenotucirev (EnAd) or adenovirus serotype 11 (Ad11). 2. Oncolytic adenovirus according to claim 1, characterized in that said component of the TCR complex is CD3, for example CD3ε, CD3y or CD3δ, in particular CD3ε. 3. Oncolytic adenovirus according to any one of claims 1, 2, characterized in that said encoded bispecific T-cell activator contains a VH domain containing the amino acid sequence SEQ ID NO: 8 or an amino acid sequence at least 95% identical to the specified sequence, and VL containing the amino acid sequence of SEQ ID NO: 9 or a sequence at least 95% identical to the specified sequence. 4. The oncolytic adenovirus of claim 3, wherein said encoded bispecific T cell activator comprises a scFv containing the amino acid sequence of SEQ ID NO: 7 or a sequence at least 95% identical to said sequence. 5. Oncolytic adenovirus according to any one of claims 1-4, characterized in that said antigen, which is present in the tumor stroma, is FAP. 6. Oncolytic adenovirus according to any one of claims 1 to 5, characterized in that said encoded bispecific T cell activator contains a VH domain containing the amino acid sequence of SEQ ID NO: 13 or an amino acid sequence at least 95% identical to the specified sequence, and VL containing the amino acid sequence of SEQ ID NO: 12 or a sequence at least 95% identical to the specified sequence. 7. The oncolytic adenovirus of claim 6, wherein said encoded bispecific T cell activator comprises a scFv containing the amino acid sequence of SEQ ID NO: 11 or 75 or a sequence at least 95% identical to any of these sequences. 8. An oncolytic adenovirus according to any one of claims 1 to 7, characterized in that said adenovirus is chimeric, capable of replication, for example, is replication competent, or is replication defective. 9. Oncolytic adenovirus according to any one of claims 1 to 8, characterized in that said bispecific T-cell activator, encoded at the BY position, is under the control of a major late promoter. 10. Oncolytic adenovirus according to any one of claims 1 to 9, characterized in that said adenovirus additionally encodes a second bispecific T-cell activator. 11. Oncolytic adenovirus according to claim 10, characterized in that both bispecific T cell activators are encoded at position B _Y . 12. Oncolytic adenovirus according to claim 10 or 11, characterized in that said second bispecific T-cell activator contains a VH domain containing the amino acid sequence of SEQ ID NO: 18 or an amino acid sequence at least 95% identical to the specified sequence, and VL containing the amino acid sequence of SEQ ID NO: 17 or a sequence at least 95% identical to the specified sequence. 13. Oncolytic adenovirus according to claim 12, characterized in that said encoded bispec- - 88 043895 specific T cell activator contains scFv containing the amino acid sequence SEQ ID NO: 16 or 73 or a sequence at least 95% identical to any of these sequences. 14. Oncolytic adenovirus according to claim 1 or 7, characterized in that said adenovirus contains the sequence SEQ ID NO: 36, 37, 81, 82, 97, 98, 99 or 100. 15. Oncolytic adenovirus according to any one of claims 1 to 14, characterized in that said adenovirus encodes 2, 3 or 4 additional transgenes, for example, characterized in that a different cleavage peptide is encoded between each of the transgenes. 16. Oncolytic adenovirus according to claim 15, characterized in that said additional transgene(s) encode a cytokine, chemokine and/or immunomodulator. 17. Oncolytic adenovirus according to any one of claims 15, 16, characterized in that all of these additional transgenes are encoded in the BY position. 18. Oncolytic adenovirus according to any one of claims 15-17, characterized in that at least one additional transgene encodes a cytokine, for example, selected from MIP1a, IL-1a, IL-1e, IL-6, IL-9, IL- 12, IL-13, IL-17, IL-18, IL-22, IL-23, IL-24, IL-25, IL-26, IL-27, IL-33, IL-35, IL-2, IL-4, IL-5, IL-7, IL-10, IL-15, IL-21, IL-25, IL-1RA, IFNa, JFNp, IFNy, TNFa, lymphotoxin α (LTA), Flt3L, GM- CSF and IL8. 19. Oncolytic adenovirus according to any one of claims 15-18, characterized in that at least one additional transgene encodes a chemokine, for example, selected from CCL2, CCL3, CCL5, CCL17, CCL20, CCL22, CXCL9, CXCL10, CXCL11, CXCL13, CXCL12, CCL2, CCL19 and CCL21. 20. A composition for the treatment of cancer containing a therapeutically effective amount of an adenovirus according to any one of claims 1 to 19 and a diluent or carrier. 21. A method of treating cancer in a patient in need thereof, comprising administering a therapeutically effective amount of an oncolytic adenovirus according to any one of claims 1 to 19 or a composition according to claim 20.