WO2017118613A1

WO2017118613A1 - Bispecific antibodies targeting human cd73

Info

Publication number: WO2017118613A1
Application number: PCT/EP2017/050003
Authority: WO
Inventors: Henrik Ditzel; Odd Lilleng GAMMELGAARD
Original assignee: Syddansk Universitet
Current assignee: Syddansk Universitet
Priority date: 2016-01-08
Filing date: 2017-01-02
Publication date: 2017-07-13
Anticipated expiration: 2018-07-08

Abstract

The present invention relates to a bispecific antibody targeting CD73. In particular, the present invention relates to a bispecific antibody targeting different epitopes on CD73 or a bispecific antibody targeting an epitope on CD73 and an epitope on a different antigen.

Description

Bispecific antibodies targeting human CD73

Technical field of the invention

The present invention relates to a bispecific antibody targeting CD73. In

particular, the present invention relates to a bispecific antibody targeting different epitopes on CD73 or a bispecific antibody targeting an epitope on CD73 and an epitope on a different antigen.

Background of the invention

CD73 is a 70 kDa GPI-anchored 5'-ecto-nucleotidase, which is overexpressed in several types of cancers, including triple-negative breast cancer (TNBC). The enzymatic activity of CD73 is known to dephosphorylate adenosine

monophosphate (AMP) into adenosine and inorganic phosphate. Overexpression of CD73 results in accumulation of adenosine in the tumour microenvironment, which suppress immune surveillance and induce angiogenesis and resistance towards chemotherapy. Increased levels of adenosine do also raise the migratory potential of cancer cells thereby enhancing metastasis. Additionally, non- enzymatic functions of CD73 affecting adhesion and migration have also been reported, which - when blocked - protect against spontaneous metastasis formation.

The mouse anti-human CD73 antibody called mlE9 is known to inhibit the enzymatic activity of CD73, but not to protect against experimental metastasis formation. The mouse anti-human CD73 antibody called mAD2 does only weakly inhibit the enzymatic activity of CD73, but does protect against both spontaneous and experimental metastasis formation by a mechanism, which includes clustering and internalization of CD73 (Terp et al. Anti-Human CD73 Monoclonal Antibody Inhibits Metastasis Formation in Human Breast Cancer by Inducing Clustering and Internalization of CD73 Expressed on the Surface of Cancer Cells. The Journal of Immunology, 2013, 191 : 4165-4173).

Cancer targets such as CD73 may be overexpressed in cancerous tissue, but are not exclusively expressed in the diseased tissue. Antibodies targeting these cancer antigens will conseq uently also affect healthy tissue to some degree, potentially causing adverse side effects.

Hence, in order to efficiently target and specifically contest cancerous tissue, an improved antibody would be advantageous.

Summary of the invention

In the present invention, the variable region of m l E9 and mAD2 were cloned onto a wild type or a HexaBody-mutated human IgG l backbone, respectively. In a set of experiments, the feasibility and efficiency of dual-epitope targeting of both anticancer mechanisms as well as antibody dependent cell mediated cytotoxicity (ADCC) and the ability of the antibodies to serve as vehicles for drug delivery were evaluated . Surprisingly, it was found that a bispecific antibody, that only targets CD73 with one arm, inhibited enzymatic activity of CD73 equally effective as the most potent of the two parental antibodies, which binds the CD73 epitope with two arms.

Importantly, the enzymatic inhibitory arm needs a secondary active arm to function in a non-static environment (like in the human body) . Thus, the secondary arm can be designed so that the bispecific antibody only affects cells, which express both antigens. Equally important, dual-epitope targeting enhanced the potency of both internalization and ADCC as well as efficiency of drug- delivery. Bispecific antibodies have several advantageous compared to e.g. a combination of two independent antibodies :

- Bispecific antibodies comprise both epitopes on one antibody, which may reduce production cost.

- It may be easier to get approval for at treatment, which only involves one active antibody, compared to a combinatorial treatment.

- Bispecific antibodies ensure that the bio-distribution of different Fab-arms are equal .

- Bispecific antibodies can be designed with high cell-type specificity, as both targeting epitopes has to be present on the cell type for hig h avidity binding to occur, which is critical for delivery of cytotoxic agents

predominantly to cancer cells. Thus, an object of the present invention relates to the provision of an alternative antibody for binding to CD73 with one or both arms. In particular, it is an object of the present invention to provide a bispecific antibody that solves the above mentioned problems of the prior art in regard of antibodies that simultaneous inhibit enzymatic activity of CD73 and prevent metastasis formation. Furthermore, it is an object of the present invention to provide a bispecific antibody targeting an epitope on CD73 and an epitope on a different antigen, thereby alleviating potential side effects exerted on non-cancerous tissue.

Thus, one aspect of the invention relates to a bispecific antibody, which binds to CD73, the antibody comprising

- a first antigen binding site (1), which is

o an antigen binding site (1A), which binds specifically to CD73 and inhibits the catalytic activity of CD73; OR

o an antigen binding site (IB), which binds specifically to CD73 and promotes internalization of CD73 (when CD73 is bound on a cell surface);

and

- a second antigen binding site (2), which is different from the first antigen binding site (1A or IB).

Another aspect of the present invention relates to a pharmaceutical composition comprising the bispecific antibody according to the invention and a

pharmaceutically acceptable carrier.

Yet another aspect of the present invention is to provide a composition according to the invention, for use as a medicament. Still another aspect of the present invention is to provide a composition according to the invention for use as a medicament in the treatment of cancer.

Brief description of the figures

Figure 1 shows the effect of bispecific antibodies, combination of antibodies and individual monospecific antibodies, respectively, on the catalytic activity of CD73. Top: MDA-MB-231 and bottom : MDA-MB-468.

Figure 2 shows the effect of bispecific antibodies, combination of antibodies and individual monospecific antibodies, respectively, on internalization of CD73. Top: MDA-MB-231 and bottom : MDA-MB-468.

Figure 3 shows the ADCC activity of bispecific antibodies, combination of antibodies and individual monospecific antibodies. Figure 4 shows the survival of cancer cells (MDA-MB-468) upon treatment with bispecific antibodies, combination of antibodies and individual monospecific antibodies used as vehicles for Duostatin3-delivery.

Figure 5 shows the survival of cancer cells (M4A4) upon treatment with bispecific antibodies, combination of antibodies and individual monospecific antibodies used as vehicles for Duostatin3-delivery.

Figure 6 shows the survival of cancer cells (M4A4) upon short exposure treatment with bispecific antibodies, combination of antibodies and individual monospecific antibodies used as vehicles for Duostatin3-delivery.

Figures 7-8 show antibodies inhibiting the enzymatic activity of CD73 under various circumstances. Only antibody formats which include at least one lE9-fab arm significantly inhibits the enzymatic activity of CD73. Importantly, the functional monovalent (lE9xbl2) do inhibits the enzymatic activity, but is easily washed away, indicating that it will not stick to cells in vivo (exert inhibition). Both functional bivalent antibodies comprising at least one 1E9 arm (1E9 and AD2xlE9) significantly and potently inhibits the enzymatic activity on all cells lines and circumstances investigated. Bonferroni one-way ANOVA was used to calculate stastistical significant differences relative to bl2 treated cells, * P<0.0005, ** P<0.0001.

Figure 9 shows the saturated antibody deposition on two cancer cell lines (A375 and MDA-MB-231-Luc2). Dual-epitope targeting strategies (AD2xlE9 and AD2 + 1E9) deposits more antibody on the surface of cancer cells compared to the parental antibodies AD2 and 1E9, which again deposits more antibody than their functionally monovalent counterparts (AD2xbl2 and lE9xbl2). However, the bispecific AD2xlE9 antibody consistently outperforms the combination of AD2 and 1E9 suggesting that the bispecific AD2xlE9 may exert higher antibody effector function activity.

The present invention will now be described in more detail in the following. Detailed description of the invention

Definitions

Prior to discussing the present invention in further details, the following terms and conventions will first be defined : Antibody

In the present context, an antibody is a protein that specifically binds a

corresponding antigen. Antibodies may particularly stem from the immune system of e.g. mammals, and may be directed towards antigens related to foreign bodies. An antibody is an intact immunoglobulin having two light and two heavy chains. Thus, a single isolated antibody or fragment may be originating from the non- limiting list of a polyclonal antibody, a monoclonal antibody, a synthetic antibody, a recombinant antibody, a chimeric antibody, a heterochimeric antibody, or a humanized antibody. The term antibody is used both to refer to a homogeneous molecular mixture, or a mixture such as a serum product made up of a plurality of different molecular entities.

Monoclonal antibody and chimeric antibody

In the present context, a monoclonal antibody is an antibody produced by a single clone of cells, such as a single clone of hybridoma cells. A monoclonal antibody is therefore a single pure homogeneous type of antibody. All monoclonal antibodies produced from the same clone are identical and have the same antigen specificity.

The monoclonal antibodies herein specifically include "chimeric" antibodies

(immunoglobulins) in which a portion of the heavy and/or light chain is identical with or homologous to corresponding sequences in antibodies derived from a particular species, while the remainder of the chain(s) is identical with or homologous to corresponding sequences in antibodies derived from another species, as well as fragments of such antibodies, so long as they exhibit the desired biological activity.

Typically, chimeric antibodies are antibodies whose light and heavy chain genes have been constructed, typically by genetic engineering, from antibody variable and constant region genes belonging to different species.

Antigen binding site/paratope

In the present context, antigen binding site, or paratope, refers to the portion of the antibody, which binds the antigen. The antigen-binding site is found in the variable region, Fv, within the Fab region of an antibody. Said variable region comprises a hypervariable region/complementarity determining region (CDR) which determines the specificity of the antibody against a specific antigen.

In the present context, "first antigen binding site" and "second antigen binding site" refers to two individual antigen-binding sites of a single antibody.

Bispecific antibody

In the present context, a bispecific antibody refers to an antibody, which comprises specificity against two different antigens or epitopes. The bispecific functionality enables the bispecific antibody to interfere with multiple surface receptors or ligands associated with processes or diseases, such as cancer, proliferation or inflammation. The two epitopes may be present in the same molecule (such as a cell surface protein, such as CD73), be present on two different molecules, being present on the same cell, or the epitopes being present in two different molecules not present on the same cell. The last option may be relevant for the case where one of the binding arms of the bispecific antibody is recruiting (has affinity for) e.g. an immune cell, a carbohydrate or a drug. In an embodiment, the immune cell is a T-cell, such as a CD8-positve T-cell. Vector

In the present context, a vector refers to an artificially constructed plasmid or a virus that may be used for genetic engineering, such as introduction of specific genes into a target cell and expression of proteins encoded by these genes. Epitope

In the present context, epitope refers to the antigenic determinant recognized by the CDRs of the antibody. In other words, epitope refers to that portion of any molecule capable of being recognized by, and bound by, an antibody. In general, epitopes consist of chemically active surface groupings of molecules, for example, amino acids and/or sugar side chains, and have specific three-dimensional structural characteristics as well as specific charge characteristics.

Sequence identity

In the present context, "sequence identity" is a measure of identity between proteins at the amino acid level and a measure of identity between nucleic acids at nucleotide level.

The protein sequence identity may be determined by comparing the amino acid sequence in a given position in each sequence when the sequences are aligned.

Similarly, the nucleic acid sequence identity may be determined by comparing the nucleotide sequence in a given position in each sequence when the sequences are aligned. To determine the percentage identity of two nucleic acid sequences or of two amino acids, the sequences are aligned for optimal comparison purposes (e.g ., gaps may be introduced in the sequence of a first amino acid or nucleic acid sequence for optimal alignment with a second amino or nucleic acid sequence). The amino acid residues or nucleotides at corresponding amino acid positions or nucleotide positions are then compared .

When a position in the first sequence is occupied by the same amino acid resid ue or nucleotide as the corresponding position in the second sequence, then the molecules are identical at that position . The percentage identity between the two sequences is a function of the number of identical positions shared by the sequences (i .e., % identity = # of identical positions/total # of positions (e.g ., overlapping positions) x 100) . In one embodiment, the two seq uences are the same length .

One may manually align the sequences and count the number of identical nucleic acids or amino acids. Alternatively, alignment of two sequences for the

determination of percentage identity may be accomplished using a mathematical algorithm . Such an algorithm is incorporated into the NBLAST and XBLAST prog rams. BLAST nucleotide searches may be performed with the NBLAST program, score = 100, wordlength = 12, to obtain nucleotide sequences homologous to a nucleic acid molecules of the invention . BLAST protein searches may be performed with the XBLAST prog ram, score = 50, wordlength = 3, to obtain amino acid sequences homologous to a protein molecule of the invention .

To obtain gapped alignments for comparison purposes, Gapped BLAST may be utilised . Alternatively, PSI-Blast may be used to perform an iterated search, which detects distant relationships between molecules. When utilising the NBLAST, XBLAST, and Gapped BLAST programs, the default parameters of the respective programs may be used . See http ://www. ncbi . nlm . nih .gov. Alternatively, sequence identity may be calculated after the sequences have been aligned e.g . by the BLAST program in the EMBL database (www.ncbi . nlm .gov/cgi-bin/BLAST) .

Generally, the default settings with respect to e.g . "scoring matrix" and "gap penalty" may be used for alig nment. In the context of the present invention, the BLASTN and PSI BLAST default settings may be advantageous.

The percentage identity between two sequences may be determined using techniques similar to those described above, with or without allowing gaps. In calculating percentage identity, only exact matches are counted . Active arm

In the present context, an "active arm" refers to a variable region of a single arm of a bispecific antibody, which binds an epitope (e.g. on the same target cell as the other arm of the bispecific antibody). Thus, for a bispecific antibody targeting CD73 with one arm, the second arm would be said to be active if it targets a different epitope on CD73 (or e.g. a different antigen on the same target cell). In the present context, bl2 constitutes an inactive arm, because its antigen is not found on the target cancer cell.

Inhibition of catalytic activity

In the present context, "inhibition of catalytic activity" refers to a decrease in the catalytic activity of CD73. The catalytic activity of CD73 is determined by a colorimetric measurement of the turnover of AMP to adenosine and inorganic phosphate as described in example 2.

Antigen internalization

In the present context, "antigen internalization" refers to the action in which CD73 localized on the cell surface is being absorbed into the cell by an internalization process, such as receptor-mediated endocytosis. Antigen internalization is measured by incubating target cells of the present invention at either 4 or 37 °C and subsequently stain remaining surface-bound CD73 with an antibody with a fluorophore attached directed against the antibodies of the present invention, as described in example 3.

CD73

In the present context, CD73, also known as 5'-nucleotidase (5'-NT) or ecto-5'- nucleotidase, is an enzyme that in humans are encoded by the NT5E gene. CD73 converts AMP to adenosine and inorganic phosphate.

PD-L1

In the present context, Programmed Death Ligand 1 (PD-L1), also known as CD274 or B7-H1, is a protein encoded by the CD274 gene. The multiple mechanisms of action and high activity of the bispecific antibody make it feasible to investigate the efficiency of dual-epitope targeting of CD73 in combination with other therapeutic compounds.

Bispecific antibodies

A described above, the present invention relates to bispecific antibodies binding to at least CD73. Thus, a first aspect of the invention relates to an (isolated) bispecific antibody, which binds to (human) CD73, the antibody comprising

- a first antigen binding site (1), which is

and

a second antigen binding site (2), which is different from the first antigen binding site (1A or IB).

In an even more general aspect, the invention relates to a bispecific antibody, which binds to (human) CD73, the antibody comprising a first antigen binding site (1), which is an antigen binding site (1), which binds specifically to CD73; and

a second antigen binding site (2), which is different from the first antigen binding site (1).

As explained in examples 2-4 and 6, such antibodies have several advantages compared to a mixture of two individual antibodies comprising the same antige binding sites, such as higher specificity. Preferably, the antibody is isolated . Preferably the CD73 is human CD73, since it is the most relevant seen from a medical point of view.

It may be important to inhibit the catalytic activity of CD73 the antibody should have a binding site promoting enzymatic inhibition of CD73, Thus, in an

embodiment said first antigen binding site ( 1), specifically binds to (human) CD73 and inhibits the catalytic activity of CD73. Thus, the first binding site corresponds to ( 1A) . The mouse anti-human CD73 antibody called m lE9 is known to inhibit the enzymatic activity of CD73, but not to protect against experimental metastasis formation (Terp et al . Anti-Human CD73 Monoclonal Antibody Inhibits Metastasis Formation in Human Breast Cancer by Inducing Clustering and Internalization of CD73 Expressed on the Surface of Cancer Cells. The Journal of Immunology, 2013, 191 : 4165-4173) .

It may also be important to promote internalization of CD73, the antibody should have a binding site promoting just that. Thus, in another embodiment, said first antigen binding site ( 1), specifically binds to (human) CD73 and promotes antigen internalization of CD73 (when CD73 is bound on a cell surface) . Thus, the first binding site corresponds to ( I B) . The mouse anti-human CD73 antibody called mAD2 does only weakly inhibit the enzymatic activity of CD73, but does protect against both spontaneous and experimental metastasis formation by a

mechanism, which includes clustering and internalization of CD73 (Terp et al . Anti-Human CD73 Monoclonal Antibody Inhibits Metastasis Formation in Human Breast Cancer by Inducing Clustering and Internalization of CD73 Expressed on the Surface of Cancer Cells. The Journal of Immunology, 2013, 191 : 4165-4173) . It may also be important to inhibit the catalytic activity of CD73 AND promote internalization of CD73. Thus, in a preferred embodiment said first antigen binding site ( 1), specifically binds to (human) CD73 and promotes antigen internalization of CD73 (when CD73 is bound on a cell surface) and said second antigen binding site (2), specifically binds to (human) CD73 and inhibits the catalytic activity of CD73. Thus both antigen binding sites target epitopes on CD73, albeit two different epitopes. In examples 2-6 such antibodies are described in different settings.

The antibodies of the invention may be particular relevant when the two antigen binding sites, target two epitopes (such as cell surface proteins) present on the same cell (such as cancer cells). These epitopes may be on the same protein or different proteins. Thus, in yet an embodiment, the epitope for said second antigen binding site (2) is present on a cell also expressing CD73, preferably on the cell surface. In yet an embodiment, the antigen for the second antigen binding site is on a cell surface protein .

In yet an embodiment, the epitope for said second antigen binding site (2) is present on a cell, preferably present on a cell also expressing CD73. In yet an embodiment, the antigen for the second antigen-binding site is a protein, carbohydrate or chemical substance.

In a more specific embodiment, said second antigen binding site (2), specifically binds to PD-L1, PD-L2, DR4, DR5, Tissue-factor, GITR, CTLA-4, CD137, TIM3, LAG3, CD40, EGFR, HER2, CD25, CD4, or CD39. Preferably, PD-L1.

In a further embodiment, the epitope for the first antigen binding site (1) and the epitope for the second antigen-binding site (2) are on a cancer cell, preferably the same cancer cell .

In another embodiment the epitope for the first antigen binding site (1) and the epitope for the second antigen-binding site (2) are on a single cell, such as a cancer cell .

The antigen binding sites may also be defined by specific sequences derived from the antibodies m lE9 and mAD2, Thus, in an embodiment the first antigen binding site (1) is selected from any of SEQ ID NO : 1-4 OR a sequence having at least 90% sequence identity to any of SEQ ID NO : 1-4, such as at least 95% sequence identity, such as at least 97% sequence identity, or such as at least 99% sequence identity.

In yet an embodiment, the second antigen binding site (2) is selected from SEQ ID NO : 1-4 OR a sequence having at least 90% sequence identity to SEQ ID NO : 1-4 such as at least 95% sequence identity, such as at least 97% sequence identity, or such as at least 99% sequence identity,

with the proviso that said second antigen binding site (2) is different from the first antigen binding site (1), and optionally, with the proviso that when the first antigen binding site is SEQ ID NO: 1, the second antigen binding site is not SEQ ID NO : 2 (and vice versa), and when the first antigen binding site is SEQ ID NO: 3, the second antigen binding site is not SEQ ID NO : 4 (and vice versa).

In a further embodiment, the first antigen-binding site (1) is the variable region/light chain from the mouse anti-human CD73 antibody mAD2.

In another embodiment, the second antigen-binding site (2) is the variable region/light chain from the mouse anti-human CD73 antibody mlE9.

In yet an embodiment, bispecific antibody comprises a human backbone, such as an IgGl backbone or a hexabody backbone. In yet an embodiment, the antibodies of the invention comprise a human Fc region from any isotype, either IgG, IgM, IgA, IgE, or IgD. Preferably, the Fc region is from human IgG, such as IgGl, IgG2, IgG3, or IgG4.

In yet an embodiment, the bispecific antibody is selected from the group consisting of monoclonal antibodies, conjugates, non-immunoglubulins, fusion proteins, polyclonal antibodies, antibodies wherein the heavy chain and the light chain are connected by a flexible linker, Fv molecules, Fab fragments, Fab' fragments, F(ab')₂ molecules, fully human antibodies, humanized antibodies, and chimeric antibodies.

It may be advantageous, if the antigen binding sites are positioned on a single polypeptide chain. Thus, in an embodiment, said bispecific antibody comprises at least two antigen binding sites positioned on different polypeptide chain, coupled through a linker, such as one or more disulfide bridges. It may however also be possible to have the antigen binding sites are positioned on a single polypeptide chain.

It may also be advantageous that bispecific antibody comprises more than one binding site for each epitope. Thus, in an embodiment the bispecific antibody comprises one or more first antigen binding sites (1), such as 2-6, such as 2-4, or such as 2-3 or such as 2 first antigen binding sites (1). In yet an embodiment, the bispecific antibody comprises one or more second antigen binding sites (2), such as 2-6, such as 2-4, or such as 2-3 or such as 2 second antigen binding sites (2). As described in example 8, it may though also be preferred that the antibody only comprises one binding site for each epitope. In yet an embodiment, the bispecific antibody is a monoclonal bispecific antibody, or a biclonal bispecific antibody.

If the antibody of the invention is to be used to target cancer cells, the antibody may be further coupled to a drug. Thus, in an embodiment, the antibody is further coupled to a drug. In example 5, such coupling is shown.

The drug may be coupled to the antibody in different ways. Thus, in an

embodiment, the coupling between the antibody and the drug is selected from the group consisting of a linker, a covalent coupling, a non-covalent coupling, and a ionic coupling, preferably, a covalent coupling. In yet an embodiment, the drug is selected from the group consisting of anticancer drugs, radionuclides, tubulin inhibitors, DNA alkylating agents, small-molecule drugs, Maytansines, Auristatins, Duostatins, Calicheamicins, anthracyclines, taxanes, and derivatives thereof. In example 5, a duostatin is tested.

Nucleic acid molecules

The antibodies of the invention may be encoded by one or more nucleic acid molecules. Thus, an aspect of the invention relates to one or more isolated nucleic acid molecules encoding a bispecific antibody according to the invention, preferably one nucleic acid.

Vectors

The nucleic acids according to the invention may be positioned in a vector. Thus, a further aspect of the invention relates to one or more vectors comprising the one or more nucleic acid molecules according to the invention, preferably one vector.

Cell lines

The nucleic acids and vectors according to the invention may be positioned in a cell line. Thus, yet a further aspect relates to a cell line comprising the one or more vectors according and/or the one or more isolated nucleic acid molecules according to the invention. Pharmaceutical compositions

The invention also relates to (pharmaceutical) compositions. Thus, in an aspect the invention relates to a (pharmaceutical) composition comprising the bispecific antibody according to the invention and a pharmaceutically acceptable carrier.

Preferably, the pharmaceutical composition is for use in the treatment of mammals, such as humans.

In yet an aspect, the invention relates to a composition according to the invention for use as a medicament.

In yet a further aspect, the invention relates to a composition according to the invention for use as a medicament in the treatment of cancer. In an embodiment, the cancer is selected from the group consisting of bladder cancer, leukemia, glioma, glioblastoma, melanoma, ovarian cancer, thyroid cancer, esophageal cancer, gastric cancer, colon cancer, prostate cancer, breast cancer, head and neck cancer, and cancer exosomes. In a preferred embodiment, the cancer is breast cancer. In yet an embodiment, the breast cancer is triple-negative breast cancer (TNBC) .

In an additional aspect, the invention relates to a method of inhibiting growth of a tumor in a subject, comprising administering to the subject a therapeutically effective amount of the bispecific antibody according to the invention and/or a composition according to the invention .

It should be noted that embodiments and features described in the context of one of the aspects of the present invention also apply to the other aspects of the invention .

All patent and non-patent references cited in the present application, are hereby incorporated by reference in their entirety. The invention will now be described in further details in the following non-limiting examples.

Examples

Example 1: Preparation of bispecific antibodies

Bispecific antibodies was produced as described in Labrijn et al. ("Controlled Fab- arm exchange for the generation of stable bispecific IgGl. ", Nature protocols. 2014;9(10) : 2450-63). Briefly, CH3-domain matched antibodies were mixed 1 : 1 in a 15 ml tube, and the volume adjusted by addition of PBS. 2-mercaptoethylamine was added to the antibody mixture (75 mM final concentration) and the mixture was incubated for 5 minutes at room temperature while shaking . The mixture was subsequently incubated at 31 °C for 5 hours. Upon incubation, 2- mercaptoethylamine was removed by repeated viviaspin (until the concentration was below 50 μΜ) and the samples was stored for 24 hours at 4 °C to allow re- oxidation to occur. The samples was analysed for aggregates and exchange efficiency by HP-SEC and CEX, respectively.

Example 2: Enzymatic inhibition of CD73 mediated by bispecific

antibodies

Aim

It is well known that the construction of bispecific antibodies may cause

conformational alterations, thereby abolishing the intended binding effect. Similar, the removal of one functional arm of an antibody may also abolish or minimize the binding effect of the antibody. The present example was intended to elucidate the effect of altering one arm of the binding arms of the AD2 and 1E9 antibodies, especially in respect of the enzymatic inhibition of CD73.

Materials and methods

3xl0⁵ cells (MDA-MB-231 or MDA-MB-468) were resuspended in media (DMEM AQmedia (SIGMA, D0819) supplemented with 10 % FBS (Sigma-Aldrich, F7524) and 1 % Penicillin-Streptomycin (Sigma, P0781-100 ml)) supplemented with 10 Mg/ml of antibody as indicated in Figure 1. Cells were incubated for 1 hour at 37 °C and were subsequently washed x3 in TBS and resuspended in 500 μΙ_ glycine buffer (75 mM glycine, 5 mM MgSC , pH 7.4) with and without 0.2 mM AMP. 4x100 μΙ_ was aliquoted into a flat-bottomed 96 well and incubated for 1 hour at 37 °C. Upon incubation, cells were lysed by first adding 50 μΙ_ of 0.5 M H2SO4, and the colour reaction started by adding another 50 μΙ_ phosphate reaction solution (0.4% NhU-molybdate, 10 % ascorbic acid). The plate was incubated 45 minutes at room temperature in the dark while shaking. The colour reaction product (molybdenum blue) was colorimetrically measured at 560 nm using a Victor3 Multilabel Plate Reader (PerkinElmer Life Sciences). The catalytic activity was calculated by subtracting the values in the absence of AMP relative to the values obtained from an irrelevant control antibody (anti-gpl20, bl2). All experiments were performed 3 or 4 times.

Results

The results presented in Figure 1 show that the 1E9 antibody inhibits the catalytic activity more efficient than the AD2 antibody (Figure 1, compare lanes AD2 and lanes 1E9). When mixing AD2 and 1E9 this effect is further increased (see Figure 1, lanes AD2 + 1E9 (1 : 1)).

To elucidate whether the antibodies AD2 and 1E9 would be functional when removing one of the binding arms, the antibodies AD2xB12 and 1Ε9χΒ12 were constructed. The results showed that their capability to inhibit the catalytic activity of CD73 was significantly weakened when one of the binding arms were substituted with a non-functional arm (bl2) (see Figure 1, lanes AD2xbl2 DB and lE9xbl2 DB compared to lanes AD2 and 1E9 respectively). To elucidate whether the ability to inhibit the catalytic activity of CD73 by creating a bispecific antibody comprising a binding site from AD2 and a binding site from 1E9, the bispecific antibody AD2xlE9 was constructed. Surprisingly it was found that the ability to inhibit the catalytic activity of CD73 was re-established by the bispecific antibody to a similar level to when the independent AD2 and 1E9 antibodies were mixed (see Figure 1, lanes AD2 + 1E9 (1 : 1) compared to lanes AD2xlE9 DB).

It is furthermore noted that this important finding is independent of the backbone, as both the wild type and HexaBody™ mutated backbone are equally effective (Figure 1). Conclusion

The bispecific antibody (AD2xlE9 DB) inhibits the catalytic activity of CD73 to a similar level as the combination (AD2 + 1E9 (1 : 1)). This is surprising taken into account the poor inhibition levels of the AD2 and 1E9 antibodies having one binding site substituted. Overall, the example demonstrates that if both arms of the antibody are active, then a single anti-CD73 arm is sufficient to inhibit the enzyme activity. This important finding is independent of the backbone, as both the wild type and HexaBody™ mutated backbone are equally effective.

Example 3: promotion of internalization of CD73 mediated by bispecific antibodies

Aim

As also mentioned for example 2, it is well known that the construction of bispecific antibodies may e.g. course conformational alterations, thereby abolishing the intended binding effect. Similar, the removal of one functional arm of an antibody may also abolish or at minimize the binding effect of the antibody. The present example was intended to elucidate the effect of altering one arm of the binding arms of the AD2 and 1E9 antibodies, especially in respect of promotion of internalization of CD73.

Materials and methods

2xl0⁵ cells (MDA-MB-231 or MDA-MB-468) were resuspended in media (DMEM AQmedia (SIGMA, D0819) supplemented with 10 % FBS (Sigma-Aldrich, F7524) and 1 % Penicillin-Streptomycin (Sigma, P0781-100 ml)) supplemented with 10 g/ml antibody as indicated in Figure 2. Cells were incubated for 90 minutes at either 4°C or 37°C, after which plates were washed x4 in ice cold PBS. Cells were subsequently stained with Alexa Fluor 488 goat anti-human IgG (Life

Technologies) for 60 minutes on ice in the dark. Finally, cells were washed x4 with ice cold PBS and measured immediately on a BD FACSCalibur Flow Cytometer. Flow logic software was used to analyse the data. Internalization was calculated as the ratio of geometric mean values (background subtracted) of 37°C to 4°C. The irrelevant control antibody (anti-gbl20, bl2) was used to measure

background staining at both 4°C and 37°C. All experiments were performed 3 or 4 times. Results

The same antibody constructs as described in example 2, was tested in respect of promotion of internalization of CD73.

5

The results presented in Figure 2 shows that the bispecific antibody (AD2xlE9 DB) is still capable of promoting internalization of CD73.

It is furthermore noted that this important finding is independent of the backbone, 10 as both the wild type and HexaBody™ mutated backbone are equally effective (Figure 2).

Conclusion

The bispecific antibody (AD2xlE9 DB) is a potent promotor of internalization of 15 CD73 (Figure 2). This important finding is independent of the backbone, as both the wild type and HexaBody™ mutated backbone are equally effective.

Example 4: ADCC mediated by bispecific antibody

Aim

20 To evaluate the ADCC potency of the bispecific antibodies compared to other

antibody constructs.

Materials and methods

Peripheral blood mononuclear cells (PBMCs) were isolated from buffy coats

25 according to the Ficoll-Paque PLUS protocol (GE Healthcare) and stored over night at 37 °C in a humidified atmosphere of 5% CO2. A cell suspension with a concentration of lxlO⁷ PBMCs/ml was prepared. Target cells (MDA-MB-231) were harvested, spun down and resuspended in media (RPMI supplemented with 10 % FBS). Cells were labelled by 1 hour incubation at 37°C in the presence of ΙΟΟμΟ 30 51Cr. Next, cells were washed with PBS x3 and resuspended in media. Cells were counted and added to a 96 wells plate. A serial dilution of antibodies was added, and the plate was incubated for 15 minutes at room temperature. After

incubation, PBMCs were added at a target effector ratio of 1 : 100, and the plate was incubated at 37 °C for 4 hours. The supernatant was transferred to tubes and 35 measured for radiation. Cytotoxicity was calculated by subtracting the values in the absence of PBMCs relative to max release (cells incubated with 10 % Triton-X 100). The data represent average values of 3 wells per condition.

Conclusion

The EC50 values of AD2 and the dual-epitope targeting strategies are very low (between 0.1-2 ng/ml, Figure 3). Additionally, when cancer cells are fully saturated, the dual-epitope targeting results in very high levels of cancer cell death (approximately 80 %), which may be critical for efficient tumour

eradication. The dual-epitope targeting is therefore a potent effector of antibody dependent cell mediated cytotoxicity (ADCC).

Example 5: Coupling of drug to bispecific antibody

Aim:

To evaluate the efficacy of anti-CD73 antibodies as vehicles for drug delivery to cancer cells

Materials and Methods

5xl0³ cells (MDA-MB-468, M4A4) was seeded in flat-bottomed 96 well plates, and incubated for 4 hours for attachment. Antibodies diluted in media supplemented with DuoStatin3 (final concentration 1 Mg/ml) was added. The plates were incubated for either 2 hours or 4 days at 37 °C. In the first case, Duostatin3 containing media was removed after two hours and fresh media added . The cells were subsequently incubated for 4 days at 37 °C. On day four cells were added 15 μΙ Cell titer GLO (Promega) and incubated for 90 minutes at 37 °C. 100 μΙ of supernatant was transferred to a white bottomed plate and luminescence was measured.

Staurosporin was used as background. After subtracting background values, survival % of MDA-MB-468 (Figure 4), M4A4 (Figure 5) and M4A4 with short exposure time (Figure 6) was calculated by setting each value relative to media treated cells.

Conclusion

Both dual-epitope targeting strategies are significantly more potent than the parental antibodies, which are significantly more potent than the functional monovalent antibodies. These important findings indicate a synergetic effect of dual-epitope targeting, both when treated as a combination and a bispecific antibody. Importantly, the bispecific format ensures that both fab-arms reaches the same target and thereby avoid complications with bio-distribution. Furthermore, the bispecific AD2xlE9 antibody was the only antibody, which killed a substantial fraction of cancer cells upon short toxin exposure. This difference in internalization kinetics may be very relevant in a physiological setting were antibodies are more prone to be washed away and thus need to exerts its function quickly.

Example 6: Bispecific antibodies with a first arm targeting CD73 and a second arm a different antigen on the same cell

Aim

Bispecific antibodies targeting CD73 with one arm and another antigen on the same cell with the other arm can be designed. Such an antibody do only attach strongly to cells which simultaneously express both antigens, and would thus be more cancer specific, while still blocking the cancer promoting functions of CD73 and the other antigen. An example could be a bispecific antibody comprising the anti-CD73 arm from 1E9 and a PDL-1 targeting arm on the other. Such an antibody would only recognize cancer cells and antigen presenting cells, but no vital organs, and thus allow for much more aggressive dosing . Furthermore, the immune-releasing mechanism would only be exerted locally in the tumour, which may limit autoimmune complications. Methods

Functional bivalent bispecific antibodies will be produced as described earlier. Shifts in binding constants will be used to identify candidates, which only attach to the target cell when both arms are binding. Conclusion

Bispecific antibodies, which target CD73 and other antigens, can be produced to potently inhibit several cancer promoting functions with enhanced specificity. This is supported by the fact that lE9xbl2 do inhibit the enzymatic activity of CD73 (bivalent binding is not necessary) but when the cells are washed lE9xbl2 detach from the cells (Figures 7-8). Both of the antibodies 1E9 and AD2xlE9 bind bivalently to the cell and exert its anticancer functions even after extensive washing . Thus, the anticancer activity can be completely rescued when the other arm are functional. Example 7: Antibody deposition

Aim

The amount of antibody being deposited on a cell, controls the highest antibody effector functions activity possible. We therefore investigated and compared the antibody deposition capacity of each anti-CD73 antibody.

Materials and Methods

Cells (A375 or MDA-MB-213-Luc2+) were harvested by 5 mM EDTA treatment for 20 minutes. Cells were subsequently washed in media and seeded 1.5E5 cells/well in a 96 well plate. The plate was cooled on ice, added serial diluted antibodies as indicated in the figures and incubated for 30 minutes. Upon incubation cells were washed and stained with an Alexa Flour 488 labelled anti-human IgG (Life

Technologies, A11013). Stained cells were incubated 30 minutes on ice in the dark, and were subsequently washed and resuspended in flow buffer (PBS + 0.05 BSA + 0.02 % Azid supplemented with TOPRO-3 (ThermoFischer, T3605)

(1 : 2000). The mean fluorescence intensity of Alexa Flour 488 on living single cells was subsequently measured by flow cytometry (Figure 9).

Conclusion

As shown in representative Figure 9, we consistently found that in a saturated state (10 ug/ml), higher levels of the bispecific AD2xlE9 antibody could be deposited than any other antibody treatment including the combination of AD2 and 1E9. This is important as it makes it plausible that the AD2xlE9 antibody exerts higher antibody effector function activity than any of the other treatments including the combination of AD2 and 1E9. Sequence listing

SEQ ID NO: 1 - 1E9 VH

QVQLQQPGAELVKPGASVKMSCKASGYTFTSYWITWVKQRPGQGLEWIGDIYPGSGNTNY

NEKFKTKATLTVDTSSSTAYMQLSSLTSEDSAVYYCAKEGGLTTEDYALDYWGQGTSVTVS

S

SEQ ID NO: 2 - 1E9 VL

DIVLTQSPASLAVSLGQRATISCRASKNVSTSGYSYMHWYQQKPGQPPKLLIYLASNLESGV PTRFSGSGSGTDFTLNIHPVEEEDAATYYCQHSRELPFTFGSGTKLEIK

SEQ ID NO: 3 - AD2 VH

EVQLQQSGAELVKPGASVKLSCTASGFNIKDTYIHWVKQRPEQGLEWIGRIDPATGNTEYD PKFQGKATITADTSSNTAYLHLSSLTSEDTAVYYCARGYYGSSYPPWFAYWGQGTLVTVS

SEQ ID NO: 4 - AD2 VL

DIVMTQSHKFMSTSVGDRVSITCKASQDVGSAVAWYQQKPGQSPKLLIYWASTRHTGVPD RFTGSGSGTDFTLTISNVQSEDLADYFCQQYSSYPLTFGAGTKLELK

Claims

1. A bispecific antibody, which binds to CD73, the antibody comprising

- a first antigen binding site (1), which binds specifically to CD73 and 5 promotes internalization of CD73;

and

- a second antigen binding site (2), which binds to specifically CD73 and inhibits the catalytic activity of CD73.

10 2. The bispecific antibody according to claim 1, wherein the first antigen binding site (1) is selected from any of SEQ ID NO : 3-4 OR a sequence having at least 90% sequence identity to any of SEQ ID NO: 3-4, such as at least 95% sequence identity, such as at least 97% sequence identity, or such as at least 99% sequence identity.

15

3. The bispecific antibody according to claim 1 or 2, wherein the second antigen binding site (1) is selected from any of SEQ ID NO : 1-2 OR a sequence having at least 90% sequence identity to any of SEQ ID NO: 1-2, such as at least 95% sequence identity, such as at least 97% sequence identity, or such as at least

20 99% sequence identity.

4. The bispecific antibody according to claim 1, wherein the first antigen binding site (1) is selected from any of SEQ ID NO : 3-4 AND the second antigen binding site (1) is selected from any of SEQ ID NO : 1-2.

25

5. The bispecific antibody according to any of the preceding claims, wherein the first antigen binding site (1) is the variable region/light chain from the mouse anti-human CD73 antibody mAD2 and the second antigen binding site (2) is the variable region/light chain from the mouse anti-human CD73 antibody mlE9.

30

6. The bispecific antibody according to any of the preceding claims, comprising a human backbone, such as an IgGl backbone or a hexabody backbone.

7. The bispecific antibody according to any of the preceding claims, comprising a human Fc region from any isotype, such as IgG, IgM, IgA, IgE, or IgD.

8. The bispecific antibody according to any of the preceding claims, wherein the Fc region is from human IgG, such as IgGl, IgG2, IgG3, or IgG4.

9. The bispecific antibody according to any of the preceding claims, wherein the bispecific antibody is selected from the group consisting of monoclonal antibodies, conjugates, non-immunoglubulins, fusion proteins, polyclonal antibodies, antibodies wherein the heavy chain and the light chain are connected by a flexible linker, Fv molecules, Fab fragments, Fab' fragments, F(ab')2 molecules, fully human antibodies, humanized antibodies, and chimeric antibodies.

10. The bispecific antibody according to any one of the preceding claims, wherein the antibody is further coupled to a drug.

11. The bispecific antibody according to claim 10, wherein the drug is selected from the group consisting of anticancer drugs, radionuclides, tubulin inhibitors, DNA alkylating agents, small-molecule drugs, Maytansines, Auristatins,

Duostatins, Calicheamicins, anthracyclines, taxanes, and derivatives thereof.

12. A pharmaceutical composition comprising the bispecific antibody according to any of the preceding claims and a pharmaceutically acceptable carrier.

13. The composition according to claim 12, for use as a medicament.

14. The composition according to claim 12 or 13, for use as a medicament in the treatment of cancer, such as a cancer selected from the group consisting of bladder cancer, leukemia, glioma, glioblastoma, melanoma, ovarian cancer, thyroid cancer, esophageal cancer, gastric cancer, colon cancer, prostate cancer, breast cancer, head and neck cancer, and cancer exosomes.