Multispecific antibodies with conditional activity

Abstract

The present invention relates to a multispecific antibody which binds to human serum albumin (HSA) and to a second antigen, wherein the apparent affinity of the antibody to the second antigen in the presence of the average concentration of HSA in human cerebrospinal fluid is higher compared to the affinity of the antibody to the second antigen in the presence of the average concentration of HSA in human peripheral blood.

Description

Technical Field

[0001] This invention relates to antibodies that preferentially bind to target antigens in human cerebrospinal fluid compared to the same targets in human peripheral blood, and methods for using the same antibodies. Background Technology

[0002] For certain therapeutic applications, it may be desirable to generate therapeutic molecules, such as proteins with conditional activity, to avoid on-target effects at undesirable locations. For example, such molecules may be almost inactive in the patient's systemic circulation but activated in certain organs or microenvironments. Besides temperature, other environmental factors that may cause proteins to exhibit conditional activity include changes in pH, osmotic pressure, redox potential, and electrolyte concentration.

[0003] There is still a need for conditionally active proteins that exhibit higher activity and / or selectivity in specific environments and / or under specific conditions (e.g., in the brain). Summary of the Invention

[0004] This invention relates to a multispecific antibody that binds to human serum albumin (HSA) and a second antigen, wherein the apparent affinity of the antibody for the second antigen is higher in the presence of an average concentration of HSA in human cerebrospinal fluid than in the presence of an average concentration of HSA in human peripheral blood.

[0005] This invention is partly based on the discovery that HSA concentration gradients can be used for conditional antibody activation, preferably against multispecific antibodies that bind to HSA and the second antigen in a mutually exclusive manner, thereby causing competitive blocking of the second antigen binding site by high concentrations of HSA in human peripheral blood, and in the presence of much lower concentrations of HSA in human cerebrospinal fluid, the competitive blocking of the binding site is relatively much lower.

[0006] In one embodiment of the invention, the second antigen is a therapeutic target antigen. In a preferred embodiment, the second antigen is a target antigen in the human brain.

[0007] One embodiment of the present invention relates to a multispecific antibody, wherein the antibody comprises a bispecific Fab fragment. In one embodiment, the multispecific antibody is a bispecific Fab fragment.

[0008] One embodiment of the present invention relates to a multispecific antibody, wherein the antibody comprises a bispecific Fab fragment, wherein the bispecific Fab fragment comprises a first complementary site capable of binding to a first antigen and a second complementary site capable of binding to a second antigen, wherein one of the complementary sites from the first and second complementary sites comprises amino acid residues from H-CDR2, L-CDR1, and L-CDR3, and the other complementary site comprises amino acid residues from H-CDR1, H-CDR3, and L-CDR2.

[0009] One embodiment of the present invention relates to a multispecific antibody, wherein the first antigen is HSA, and wherein the second antigen is selected from human VEGF-A, IL-1β, CD32A, CD32B, CD79b, 41BB, IFNα, and Ox40.

[0010] Another aspect of the present invention is a nucleic acid according to the present invention, which encodes a multispecific antibody.

[0011] Another aspect of the present invention is a carrier according to the invention, which contains nucleic acids.

[0012] Another aspect of the invention is a host cell according to the invention, which contains nucleic acids.

[0013] Another aspect of the present invention is a method for generating multispecific antibodies according to the present invention, the method comprising culturing host cells of the present invention under conditions suitable for antibody expression.

[0014] Another aspect of the invention is a pharmaceutical composition comprising the antibody of the invention and a pharmaceutically acceptable carrier.

[0015] According to the present invention, antigen binding at a complementary site within a multispecific antibody is activated due to the concentration gradient of certain antigens between peripheral blood and cerebrospinal fluid. For example, this allows the therapeutic antibody to be conditionally active in the brain rather than in the periphery. Attached Figure Description

[0016] Figure 1: Surface plasmon resonance experimental data of bispecific Fab and control bound to albumin and human VEGF-A from humans, cynomolgus monkeys, and mice. Figure 1A: P1AJ5849, Figure 1B: P1AJ0293, Figure 1C: P1AJ0400. Figure 1D Figure 1E: P1AJ5862, Figure 1F: P1AJ5895 (HSA binding control antibody), Figure 1G: P1AL9477 (VEGF binding control antibody).

[0017] Figure 2: Studies of bispecific Fab binding to its targets simultaneously or in a mutually exclusive manner. SPR-based bridging assays were applied as described in Example 3. Figure 2A: Possible results of bridging assays (simultaneous and mutually exclusive) of albumin / VEGF-A bispecific Fab (top panel) and illustrations of bridging assay results of VEGF-A and albumin monospecific control Fab. Figure 2B: Results of the VEGF-A / albumin bispecific Fab shown.

[0018] Figure 3: Biochemical ELISA assay demonstrating that conditional VEGF-A binding activity depends on the serum albumin concentration in the solution. The competitive ELISA assays were performed in the absence of HSA or in the presence of elevated HSA concentrations of 0.3 µM, 3 µM, 30 µM, 300 µM, and 600 µM. Figure 3A: Results of the VEGF-A / albumin bispecific Fab assay shown. Figure 3B: Results of the VEGF-A and albumin monospecific control Fab assays.

[0019] Figure 4: In vitro cell-based reporter gene assay demonstrating serum albumin concentration-dependent VEGF-A binding as described in Example 4. The reporter gene assay response is shown in the absence of HSA or in the presence of elevated HSA concentrations of 0.3 µM, 3 µM, 30 µM, and 300 µM. Figures 4A and 4B: Results of the VEGF-A / albumin bispecific Fab shown. Figure 4C: Results of the VEGF-A and albumin monospecific control Fabs.

[0020] Figure 5: Target binding of the bispecific Fab fragment comprising the anti-HSA complementary site from antibody P1AJ5849, as described in Example 6. The results of binding of the bispecific Fab to serum albumin and the second target are shown in Figures 5A to 5D. An overview of the bispecific molecule is given in Table 6.

[0021] Figure 6: Mutually repulsive binding of bispecific Fab fragments comprising the anti-HSA complementary site from antibody P1AJ5849, as described in Example 6. The results of the albumin bispecific Fabs shown are displayed in Figures 6A to 6D. An overview of the bispecific molecules is given in Table 7.

[0022] Figure 7: Target binding of the bispecific Fab fragment comprising the anti-HSA complementary site from antibody P1AJ0293, as described in Example 6. The results of the albumin bispecific Fab shown are illustrated in Figures 7A through 7D.

[0023] Figure 8: Mutually repulsive binding of bispecific Fab fragments comprising the anti-HSA complementary site from antibody P1AJ0293, as described in Example 6. The results of the albumin bispecific Fab shown are illustrated in Figures 8A through 8D.

[0024] Figure 9: Biochemical ELISA assay demonstrating the conditional IL-1β binding activity of the serum albumin / IL-1β bispecific Fab P1AL6714. P1AL9447, a molecule that binds IL-1β but not serum albumin, was used as a control not expected to show serum albumin-dependent activity. The competitive ELISA assays were performed in the absence of HSA or in the presence of elevated HSA concentrations of 0.3 µM, 3 µM, 30 µM, 300 µM, and 600 µM. Detailed Implementation

[0025] 1. Definition

[0026] Unless otherwise defined herein, scientific and technical terms related to this invention shall have the meanings commonly understood by one of ordinary skill in the art. Furthermore, unless the context requires otherwise, singular terms shall include plural terms, and plural terms shall include singular terms. The methods and techniques of this disclosure are generally performed according to conventional methods known in the art. Typically, terms and techniques related to biochemistry, enzymology, molecular and cell biology, microbiology, genetics, and protein and nucleic acid chemistry and hybridization as described herein are well-known and commonly used in the art.

[0027] Unless otherwise defined herein, the term "comprising" shall include the term "consisting of".

[0028] When the term “about” is used in conjunction with specific values ​​(such as temperature, concentration, time, etc.) in this article, it should refer to a change of + / - 1% in the specific value referred to by the term “about”.

[0029] The term "antibody" is used in the broadest sense and includes a variety of antibody structures, including but not limited to monoclonal antibodies, polyclonal antibodies, multispecific antibodies (e.g., bispecific antibodies), and antibody fragments, as long as they exhibit the desired antigen-binding activity. When the KD of an antibody is 1 µM or less, the antibody is said to "specifically bind" to the target antigen.

[0030] "Isolated" antibodies are antibodies that have been separated from components of their natural environment. In some respects, antibodies are purified to a purity greater than 95% or 99%, as determined by methods such as electrophoresis (e.g., SDS-PAGE, isoelectric focusing (IEF), capillary electrophoresis) or chromatography (e.g., ion exchange or reversed-phase HPLC). For a review of methods used to assess antibody purity, see, for example, Flatman et al., J. Chromatogr. B 848:79-87 (2007).

[0031] "Antibody fragment" refers to a molecule other than a complete antibody that contains a portion of the complete antibody and binds to the antigen bound by the complete antibody. Examples of antibody fragments include, but are not limited to, Fv, Fab, Fab', Fab'-SH, F(ab')2; bisomatic antibodies; linear antibodies; single-chain antibody molecules (e.g., scFv and scFab); single-domain antibodies (dAb); and multispecific antibodies formed from antibody fragments. For a review of some antibody fragments, see Holliger and Hudson, Nature Biotechnology 23:1126-1136 (2005).

[0032] As used herein, the term "monoclonal antibody" refers to an antibody derived from a substantially homogeneous population of antibodies, meaning that the individual antibodies comprising the population are identical and / or bind to the same epitopes, except for possible variant antibodies (e.g., containing naturally occurring mutations or generated during the production of the monoclonal antibody formulation), which are typically present in small quantities. In contrast to polyclonal antibody formulations, which typically comprise different antibodies targeting different determinants (epitopes), each monoclonal antibody in a monoclonal antibody formulation targets a single determinant on the antigen. Therefore, the modifier "monoclonal" indicates that the antibody is characterized by being derived from a substantially homogeneous population of antibodies and should not be interpreted as requiring the antibody to be produced by any particular method. For example, monoclonal antibodies according to the invention can be prepared by a variety of techniques, including but not limited to hybridoma methods, recombinant DNA methods, phage display methods, and methods utilizing transgenic animals containing all or part of the human immunoglobulin loci, such methods and other exemplary methods for preparing monoclonal antibodies are described herein.

[0033] A "multispecific antibody" is a monoclonal antibody that specifically binds to at least two different antigens. When used herein, this term should exclude antibodies that bind to at least two different epitopes of the same antigen. In some respects, a multispecific antibody has three or more binding specificities. In some respects, one of the specificities is against HSA, while the others are against any other antigen.

[0034] A “bispecific Fab fragment” or “bicomplementary Fab fragment” is a Fab fragment capable of binding to two different antigens. Bispecific Fab fragments containing Fv fragments that bind to two target antigens have been described, for example, in Fagète, S. et al. mAbs 1, 288-296 (2009); Fagète, S. et al. J. Biol. Chem. 287, 1458-1467 (2012); Bostrom, J. et al. Science 323, 1610-1614 (2009); WO2008082651; WO03002609 and WO2012163520. Specifically, WO2012163520 and Beckmann, R. et al., Nature Communications 12, 708 (2021) describe a so-called “DutaFab”, which is a bispecific Fab fragment containing so-called H-side complementary sites covering HCDR1, HCDR3, and LCDR2, and so-called L-side complementary sites covering LCDR1, LCDR3, and HCDR2. The two complementary sites can be independently selected and combined with each other to form the desired bispecific DutaFab.

[0035] In some embodiments, DutaFab comprises a double-complementary-site Fab fragment, wherein a first complementary site is capable of binding to one antigen and a second complementary site is capable of binding to another antigen different from the first antigen, wherein one of the complementary sites from the first and second complementary sites comprises amino acid residues from H-CDR2, L-CDR1, and L-CDR3, and the other complementary site comprises amino acid residues from H-CDR1, H-CDR3, and L-CDR-2. In some embodiments, the complementary site containing amino acid residues from H-CDR2, L-CDR1, and L-CDR3 does not contain amino acids from any of the other CDRs H-CDR1, H-CDR3, and L-CDR-2. In some embodiments, the complementary site containing amino acid residues from H-CDR2, L-CDR1, and L-CDR3 comprises amino acids from one or both of the other CDRs H-CDR1, H-CDR3, and L-CDR-2. In some embodiments, the complementary sites of amino acid residues from H-CDR2, L-CDR1, and L-CDR3 contain amino acids from one of the other CDRs H-CDR1, H-CDR3, and L-CDR-2. In some embodiments, the complementary sites of amino acid residues from H-CDR2, L-CDR1, and L-CDR3 do not contain amino acids from H-CDR1, H-CDR3, and L-CDR-2. In some embodiments, the complementary sites of amino acid residues from H-CDR1, H-CDR3, and L-CDR-2 do not contain amino acids from any of the other CDRs H-CDR2, L-CDR1, and L-CDR3. In some embodiments, the complementary sites of amino acid residues from H-CDR1, H-CDR3, and L-CDR-2 contain amino acids from one or both of the other CDRs H-CDR2, L-CDR1, and L-CDR3. In some embodiments, the complementary sites of amino acid residues comprising H-CDR1, H-CDR3, and L-CDR-2 contain amino acids from one of the other CDRs H-CDR2, L-CDR1, and L-CDR3. In some embodiments, the complementary sites of amino acid residues comprising H-CDR1, H-CDR3, and L-CDR-2 do not contain amino acids from H-CDR2, L-CDR1, and L-CDR3.

[0036] The term "variable region" or "variable domain" refers to a domain of the antibody heavy or light chain involved in antibody-antigen binding. The variable domains (VH and VL, respectively) of the heavy and light chains of natural antibodies typically have similar structures, with each domain containing four conserved frame regions (FRs) and three complementarity-determining regions (CDRs). (See, for example, Kindt et al., Kuby Immunology, p. 6) (WH Freeman and Co., p. 91 (2007)) A single VH or VL domain is sufficient to confer antigen binding specificity. Furthermore, antibodies binding to a specific antigen can be separated using either the VH or VL domain from the antibody binding that antigen, to screen libraries of complementary VL or VH domains. See, for example, Portolano et al., J. Immunol. 150:880-887 (1993); Clarkson et al., Nature 352:624-628 (1991). The “complementary site” is the portion of the antibody that recognizes and binds to the antigen.

[0037] "Affinity" refers to the strength of the sum of non-covalent interactions between a single binding site of a molecule (e.g., an antibody) and its binding partner (e.g., an antigen). Unless otherwise stated, as used herein, "binding affinity" refers to intrinsic binding affinity, which reflects a 1:1 interaction between members of a binding pair (e.g., antibody and antigen). The affinity of molecule X for its partner Y can generally be determined by the dissociation constant (K). D Affinity can be measured by conventional methods known in the art, including those described herein. Specific illustrative and exemplary methods for measuring binding affinity are described below.

[0038] As used herein, "apparent affinity" refers to the affinity of an antibody as measured under defined conditions (e.g., HSA concentration). Therefore, the affinity of the antibody of the present invention for its second target varies depending on the HSA concentration in the bodily fluid containing the antibody. Thus, the apparent affinity of the antibody of the present invention can be determined in the presence of an average concentration of HSA in human cerebrospinal fluid (preferably about 3 µM) or in the presence of an average concentration of HSA in human peripheral blood (preferably about 650 µM).

[0039] The terms “host cell,” “host cell line,” and “host cell culture” are used interchangeably and refer to cells in which exogenous nucleic acids have been introduced, including progeny cells. Host cells include “transformations” and “transformed cells,” which include primary transformed cells and progeny derived from those primary transformed cells, regardless of passage number. Progeny cells may not have completely identical nucleic acid contents to the parent cells and may contain mutations. This article includes mutant progeny with the same function or biological activity as those screened or selected from the original transformed cells.

[0040] As used herein, the term "vector" refers to a nucleic acid molecule capable of carrying another nucleic acid linked to it. This term includes vectors that function as self-replicating nucleic acid structures, as well as vectors incorporated into the genome of a host cell into which they have been introduced. Some vectors are capable of directing the expression of the nucleic acid to which they are operatively linked. Such vectors are referred to herein as "expression vectors."

[0041] The term "nucleic acid molecule" or "polynucleotide" includes any compound and / or substance comprising a nucleotide polymer. Each nucleotide consists of a base, specifically a purine or pyrimidine base (i.e., cytosine (C), guanine (G), adenine (A), thymine (T), or uracil (U)), a sugar (i.e., deoxyribose or ribose), and a phosphate ester group. Typically, nucleic acid molecules are described by a base sequence, where the bases represent the primary structure (linear structure) of the nucleic acid molecule. Base sequences are typically represented from 5' to 3'. In this document, the term nucleic acid molecule encompasses deoxyribonucleic acid (DNA) (including, for example, complementary DNA (cDNA) and genomic DNA), ribonucleic acid (RNA) (particularly messenger RNA (mRNA)), synthetic forms of DNA or RNA, and mixed polymers comprising two or more of these molecules. Nucleic acid molecules can be linear or circular. Furthermore, the term nucleic acid molecule includes both sense and antisense strands, as well as single-stranded and double-stranded forms. Furthermore, the nucleic acid molecules described herein may contain naturally occurring or non-naturally occurring nucleotides. Examples of non-naturally occurring nucleotides include modified nucleotide bases having derivatized sugar or phosphate backbone bonds or chemically modified residues. Nucleic acid molecules also encompass DNA and RNA molecules suitable as vectors for direct in vitro and / or in vivo (e.g., in a host or patient) expression of antibodies used in this invention. Such DNA (e.g., cDNA) or RNA (e.g., mRNA) vectors may be unmodified or modified. For example, mRNA may be chemically modified to enhance the stability of the RNA vector and / or the expression of the encoding molecule, enabling the mRNA to be injected into a subject to generate in vivo antibodies (see, for example, Stadler et al., Nature Medicine 2017, published online June 12, 2017, doi:10.1038 / nm.4356 or EP 2101 823 B1).

[0042] "Isolated" nucleic acids refer to nucleic acid molecules that have been separated from components of their natural environment. Isolated nucleic acids include nucleic acid molecules that are contained in cells that normally contain nucleic acid molecules, but which are located outside the chromosome or at a chromosomal location different from their natural chromosomal location.

[0043] The terms “pharmaceutical composition” or “pharmaceutical formulation” refer to a formulation in which the active ingredient contained therein is in a biologically effective form and does not contain any additional components that would have unacceptable toxicity to a subject to whom the pharmaceutical composition will be administered.

[0044] "Pharmaceutically acceptable carriers" refer to components in a pharmaceutical composition or formulation other than the active ingredient that are non-toxic to the subject. Pharmaceutically acceptable carriers include, but are not limited to, buffer solutions, excipients, stabilizers, or preservatives.

[0045] As used herein, "neurological disorder" refers to a disease or condition affecting the CNS and / or having a cause in the CNS. Exemplary CNS diseases or conditions include, but are not limited to, neuropathy, amyloidosis, cancer, eye diseases or conditions, viral or microbial infections, inflammation, ischemia, neurodegenerative diseases, seizures, behavioral disorders, and lysosomal storage diseases. For the purposes of this application, the CNS will be understood to include the eye, which is generally isolated from the rest of the body by the blood-retinal barrier. Specific examples of neurological disorders include, but are not limited to, neurodegenerative diseases (including but not limited to Lewy body disease, post-poliomyelitis syndrome, Shy-Draeger syndrome, oligopontocerebellar atrophy, Parkinson's disease, multiple system atrophy, striatal substantia nigra degeneration, tau protein diseases (including but not limited to Alzheimer's disease and supranuclear palsy), prion diseases (including but not limited to bovine spongiform encephalopathy, scrapie, Creutzfeldt-Jakob disease, kuru, Gerstmann-Straussler-Scheinker disease, chronic wasting diseases, and fatal familial insomnia), bulbar palsy, motor neuron diseases, and neurodegenerative disorders (including but not limited to Canavan disease, Huntington's disease, neuronal ceroid lipofuscinosis, Alexander disease, Tourette syndrome, Menkes curly hair syndrome, Cockayne syndrome, Halervorden-Spatz syndrome, Lafra disease, Rett syndrome, Wilson's disease, Lesch-Nyhan syndrome, and Unverricht-Lundborg syndrome). Syndromes), dementia (including but not limited to Pick's disease and spinocerebellar ataxia), and cancer (such as CNS and / or brain cancer, including brain metastases caused by cancer in other parts of the body).

[0046] "Neurological disease drugs" are drugs or therapeutic agents for treating one or more neurological disorders. The neurological disease drugs of the present invention include, but are not limited to, small molecule compounds, antibodies, peptides, proteins, one or more natural ligands of CNS targets, modified forms of one or more natural ligands of CNS targets, aptamers, inhibitory nucleic acids (i.e., small inhibitory RNA (siRNA) and short hairpin RNA (shRNA)), ribozymes, and small molecules, or active fragments of any of the foregoing substances. Exemplary neurological disease drugs of the present invention are described herein, and include, but are not limited to: antibodies, aptamers, proteins, peptides, inhibitory nucleic acids, and small molecules, and active fragments of any of the foregoing substances, which themselves or specifically recognize and / or act on (i.e., inhibit, activate, or detect) CNS antigens or target molecules, such as, but not limited to, amyloid precursor protein or portions thereof, amyloid β, β-secretase, γ-secretase, tau, α-synuclein, parkin, huntingtin, DR6, presenilin, ApoE, glioma or other CNS cancer markers, and neurotrophic factors. Non-limiting examples of neurological drugs and the corresponding disorders they may treat: brain-derived neurotrophic factor (BDNF), chronic brain injury (neurogenesis), fibroblast growth factor 2 (FGF-2), anti-epidermal growth factor receptor brain cancer, (EGFR)-antibody, glial cell-derived neurotrophic factor (GDNF) for Parkinson's disease, amyotrophic lateral sclerosis (ALS) with BDNF, depression, lysosomal storage disease of the brain, ciliary neurotrophic factor (CNTF) for ALS, neuromodulation protein-1 for schizophrenia, and anti-HER2 antibodies (e.g., trastuzumab) for brain metastases from HER2-positive cancers.

[0047] "CNS antigens" or "brain targets" are antigens and / or molecules expressed in the CNS (including the brain) that can be targeted by antibodies or small molecules. Examples of such antigens and / or molecules include, but are not limited to: β-secretase 1 (BACE1), amyloid β (Aβ), epidermal growth factor receptor (EGFR), human epidermal growth factor receptor 2 (HER2), Tau protein, apolipoprotein E4 (ApoE4), α-synuclein, CD20, huntingtin protein, prion protein (PrP), leucine-rich repeat kinase 2 (LRRK2), parkin protein, presenilin 1, presenilin 2, γ-secretase, death receptor 6 (DR6), amyloid precursor protein (APP), p75 neurotrophic factor receptor (p75NTR), and caspase 6. In one embodiment, the antigen is BACE1. Detailed Implementation

[0048] This invention relates to a multispecific antibody that binds to human serum albumin (HSA) and a second antigen, wherein the apparent affinity of the antibody for the second antigen is higher in the presence of an average concentration of HSA in human cerebrospinal fluid than in the presence of an average concentration of HSA in human peripheral blood.

[0049] This invention is based on the observation that the concentration gradient of human serum albumin between peripheral blood and cerebrospinal fluid (CSF) can be used to conditionally activate therapeutic antibodies in the brain rather than in the periphery. While the average albumin concentration in human CSF is approximately 3 µM, the albumin concentration in human peripheral blood is approximately 650 µM (see, for example, S. Seyfert et al., J Neurol Sci. 2004;219(1-2):31-3). Multispecific antibodies are provided that bind to HSA and a second antigen in a mutually exclusive manner based on the HSA concentration in the surrounding body fluids, such as human CSF, allowing conditional activation of antigen-binding specificity against a second target antigen in these fluids where the HSA concentration is lower than in peripheral blood.

[0050] This invention relates to multispecific antibodies, wherein the antibody binds to at least a first target antigen (i.e., HSA) and a second target antigen. In one embodiment, the second target antigen is a therapeutic antigen. In another embodiment, the second target antigen is a therapeutic target antigen in the human brain. In some embodiments, the antibodies of the present invention are bispecific.

[0051] Ideally, the antibodies of the present invention can bind to their target antigens in a mutually exclusive manner, meaning that the antibody can only bind to one of its target antigens at a time.

[0052] Mutually exclusive antigen binding

[0053] When present in body fluids containing HSA at the average concentration of HSA in human cerebrospinal fluid, the antibodies of the present invention preferably bind to the second antigen in the presence of HSA at the cerebrospinal fluid concentration, rather than in the presence of its higher concentration in peripheral blood. This allows for conditional activity of the antibody in the human brain, thereby reducing undesirable activity in the periphery. In some embodiments, antigen binding to the second antigen in peripheral blood is at least an order of magnitude lower than antigen binding to the second antigen in CSF. Thus, in one embodiment, the apparent affinity of the antibody of the present invention for its second target antigen is at least an order of magnitude lower in body fluid containing 3 µM HSA than the affinity for its second target antigen in body fluid containing 600 µM HSA.

[0054] In one embodiment, the antibody of the present invention specifically binds to the second antigen in the absence of HSA. In another embodiment, the antibody of the present invention does not specifically bind to the second antigen in the presence of HSA.

[0055] Structural arrangement of complementary positions

[0056] The mutual repulsion binding of its targets can be achieved, for example, by providing multispecific antibodies with two complementary sites and / or overlapping complementary sites within a pair of VH / VL.

[0057] In some embodiments, the antibodies of the present invention comprise a pair of VH and VL domains (i.e., Fv fragments) that bind to a first antigen and a second antigen. Therefore, according to these embodiments, the antibodies of the present invention may comprise a bicomplementary Fv fragment, i.e., an Fv fragment comprising two complementary site regions, as opposed to natural antibodies. Bicomplementary Fv fragments have been described in the art, for example, as exemplified by the DutaFab antibody as outlined above.

[0058] In some embodiments, the antibody of the present invention comprises a bicomplementary Fv fragment having two overlapping complementary sites. In this context, the term "overlapping" means that the complementary site regions are spatially arranged within the Fv region in such a way that antigen binding of one antigen (e.g., HSA) spatially hinders the binding of the antibody to another antigen (e.g., a second antigen). In some embodiments, at least one amino acid residue from the first complementary site binding to the first antigen is also contained within the second complementary site binding to the second antigen. The complementary site regions can be defined by methods known in the art, such as X-ray crystallography. The functional effect of the overlapping complementary sites as defined herein can be determined experimentally, for example in so-called "bridging" ELISA or SPR measurements, demonstrating that the bispecific antibody cannot bind to two targets simultaneously.

[0059] In one embodiment of the invention, the multispecific antibody comprises the VH and VL domains of DutaFab. Therefore, in one embodiment, the bispecific antibody comprises a bispecific Fv fragment, the bispecific Fv fragment comprising a first complementary site capable of binding to a first antigen (HSA in one embodiment) and a second complementary site capable of binding to a different antigen (a second target antigen in one embodiment), wherein one of the complementary sites from the first and second complementary sites comprises amino acid residues from H-CDR2, L-CDR1, and L-CDR3, and the other complementary site comprises amino acid residues from H-CDR1, H-CDR3, and L-CDR2.

[0060] In addition to amino acid residues from the CDR region shown, the first and second complementary sites may also contain amino acid residues contained within the framework regions of the VH and VL domains of the Fv fragment. In one embodiment, at least one amino acid residue contained in the first complementary site is also contained in the second complementary site.

[0061] HSA was selected as the first target antigen.

[0062] The first antigen bound by the antibody of the present invention is a target antigen of the human body, which exhibits a concentration gradient between peripheral blood and cerebrospinal fluid. Specifically, the concentration of the first antigen in peripheral blood is higher than that in cerebrospinal fluid. In some embodiments, the ratio between the average concentration of the first antigen in peripheral blood and the average concentration of the first antigen in cerebrospinal fluid is greater than about 10, preferably greater than about 100, and even more preferably greater than 200.

[0063] The average concentration of HSA in human peripheral blood is approximately 42.9 g / L (650 µM) (S. Seyfert et al., J Neurol Sci. 2004;219(1-2):31-3). The average concentration of HSA in human cerebrospinal fluid is approximately 0.21 g / L (3.2 µM) (S. Seyfert et al., J Neurol Sci. 2004;219(1-2):31-3). Therefore, the ratio between the average concentration of HSA in peripheral blood and the average concentration of HSA in cerebrospinal fluid is approximately 200.

[0064] Other antigens exhibiting concentration gradients between human cerebrospinal fluid (CSF) and peripheral blood include IgG (with an average concentration of approximately 0.024 g / L in CSF and 10.9 g / L in peripheral blood, see, for example, S. Seyfert et al., J Neurol Sci. 2004;219(1-2):31-3) and transferrin (with an average concentration of 0.02 g / L in CSF and 2-3.6 g / L in peripheral blood, see Levine SM et al., Brain Res. 1999;821(2):511-5). Therefore, in addition to HSA, conditionally active multispecific antibodies in the human brain can also target IgG or transferrin and secondary antigens.

[0065] Second antigen

[0066] The second antigen bound to the antibody of the present invention can be any antigen other than HSA. Because the antibody of the present invention is suitable for therapeutic applications, in some embodiments the second antigen is a therapeutic target antigen, and in one embodiment it is a brain target.

[0067] In one embodiment, the second antigen is a cell surface antigen, cytokine, or growth factor.

[0068] In one embodiment, the second antigen is selected from VEGF-A, IL-1β, CD32A, CD32B, CD79b, 41BB, IFNα, and Ox40.

[0069] Exemplary antibody

[0070] In some embodiments, the antibodies of the present invention comprise the VH and VL domains as disclosed in the Embodiments section herein.

[0071] In one embodiment, the anti-HSA / anti-VEGF-A antibody of the present invention comprises a pair of VH and VL domains selected from the following:

[0072] a) The VH domain of SEQ ID NO:01 and the VL domain of SEQ ID NO:02,

[0073] b) The VH domain of SEQ ID NO:03 and the VL domain of SEQ ID NO:04,

[0074] c) The VH domain of SEQ ID NO:05 and the VL domain of SEQ ID NO:06,

[0075] d) The VH domain of SEQ ID NO:07 and the VL domain of SEQ ID NO:08, and

[0076] e) The VH domain of SEQ ID NO:09 and the VL domain of SEQ ID NO:10.

[0077] In one embodiment, the anti-HSA / anti-IL1β antibody of the present invention comprises the VH domain of SEQ ID NO:19 and the VL domain of SEQ ID NO:20.

[0078] In one embodiment, the anti-HSA / anti-CD32a antibody of the present invention comprises the VH domain of SEQ ID NO:21 and the VL domain of SEQ ID NO:22.

[0079] In one embodiment, the anti-HSA / anti-CD32b antibody of the present invention comprises the VH domain of SEQ ID NO:23 and the VL domain of SEQ ID NO:24.

[0080] In one embodiment, the anti-HSA / anti-CD79b antibody of the present invention comprises the VH domain of SEQ ID NO:25 and the VL domain of SEQ ID NO:26.

[0081] In one embodiment, the anti-HSA / anti-41BB antibody of the present invention comprises the VH domain of SEQ ID NO:27 and the VL domain of SEQ ID NO:28, or the VH domain of SEQ ID NO:33 and the VL domain of SEQ ID NO:34.

[0082] In one embodiment, the anti-HSA / anti-IFNα antibody of the present invention comprises the VH domain of SEQ ID NO:29 and the VL domain of SEQ ID NO:30.

[0083] In one embodiment, the anti-HSA / anti-OX40 antibody of the present invention comprises the VH domain of SEQ ID NO:31 and the VL domain of SEQ ID NO:32; or the VH domain of SEQ ID NO:35 and the VL domain of SEQ ID NO:36.

[0084] antibody fragments

[0085] In some respects, the antibodies presented in this article are antibody fragments.

[0086] In one respect, antibody fragments are Fab, Fab', Fab'-SH, or F(ab')2 fragments, particularly Fab fragments. Papain digestion of an intact antibody produces two identical antigen-binding fragments called "Fab" fragments, each containing a heavy chain variable domain and a light chain variable domain (VH and VL, respectively), as well as a constant domain (CL) of the light chain and a first constant domain (CH1) of the heavy chain. Thus, the term "Fab fragment" refers to an antibody fragment comprising a light chain containing the VL and CL domains and a heavy chain containing the VH and CH1 domains. A "Fab' fragment" differs from a Fab fragment in that a residue is added to the carboxyl terminus of the CH1 domain, including one or more cysteine ​​residues from the antibody hinge region. Fab'-SH is a Fab' fragment in which the cysteine ​​residues of the constant domain have a free thiol group. Pepsin treatment produces an F(ab')2 fragment, which has two antigen-binding sites (two Fab fragments) and a portion of the Fc region. For a discussion of Fab and F(ab')2 fragments containing salvage receptor-binding epitope residues and having an increased in vivo half-life, see U.S. Patent No. 5,869,046.

[0087] In another respect, antibody fragments are bisomatic, trisomatic, or tetrasomatic antibodies. A “bisomatic antibody” is an antibody fragment having two antigen-binding sites (which can be bivalent or bispecific). See, for example, EP 404,097; WO1993 / 01161; Hudson et al., Nat. Med. 9:129-134 (2003); and Hollinger et al., Proc. Natl. Acad. Sci. USA 90: 6444-6448 (1993). Trisomatic and tetrasomatic antibodies are also described by Hudson et al. in Nat. Med. 9:129-134 (2003).

[0088] In another aspect, the antibody fragment is a single-chain Fab fragment. A “single-chain Fab fragment” or “scFab” is a polypeptide composed of an antibody heavy chain variable domain (VH), an antibody heavy chain constant domain 1 (CH1), an antibody light chain variable domain (VL), an antibody light chain constant domain (CL), and a linker, wherein the antibody domains and linker have one of the following sequences in the N-terminal to C-terminal direction: a) VH-CH1-linker-VL-CL, b) VL-CL-linker-VH-CH1, c) VH-CL-linker-VL-CH1, or d) VL-CH1-linker-VH-CL. Specifically, the linker is a polypeptide of at least 30 amino acids, preferably between 32 and 50 amino acids. The single-chain Fab fragment is stabilized via a native disulfide bond between the CL domain and the CH1 domain. Furthermore, these single-chain Fab fragments can be further stabilized by generating interchain disulfide bonds via the insertion of cysteine ​​residues (e.g., position 44 in the variable heavy chain and position 100 in the variable light chain according to Kabat numbering).

[0089] On the other hand, antibody fragments are single-chain variable fragments (scFv). A "single-chain variable fragment" or "scFv" is a fusion protein of the antibody's heavy chain variable domain (VH) and light chain variable domain (VL), linked by a linker. Specifically, the linker is a short polypeptide of approximately 10 to 25 amino acids, typically rich in glycine for flexibility and serine or threonine for solubility, and can link the N-terminus of the VH to the C-terminus of the VL, or vice versa. Despite the removal of the constant region and the introduction of the linker, the protein retains the specificity of the original antibody. For reviews of scFv fragments, see, for example, Plückthun, in The Pharmacology of Monoclonal Antibodies, Vol. 113, edited by Rosenburg and Moore, (Springer-Verlag, New York), pp. 269–315 (1994); see also WO 93 / 16185; and U.S. Patent Nos. 5,571,894 and 5,587,458.

[0090] On the other hand, antibody fragments are single-domain antibodies. A "single-domain antibody" is an antibody fragment containing all or part of the variable heavy chain domain or all or part of the variable light chain domain of an antibody. In some respects, single-domain antibodies are human single-domain antibodies (Domantis, Inc., Waltham, MA; see, for example, U.S. Patent No. 6,248,516 B1).

[0091] Antibody fragments can be prepared using various techniques, including but not limited to the proteolytic digestion of intact antibodies and recombinant production from recombinant host cells (e.g., E. coli), as described herein.

[0092] Antibody variants

[0093] In some respects, amino acid sequence variants of the antibodies presented herein are envisioned. For example, it may be desirable to alter the binding affinity and / or other biological properties of the antibody. Amino acid sequence variants of antibodies can be prepared by introducing appropriate modifications into the nucleotide sequence encoding the antibody or by peptide synthesis. Such modifications include, for example, deletion, and / or insertion and / or substitution of residues within the antibody amino acid sequence. Any combination of deletions, insertions, and substitutions can be performed to achieve the final construct, provided that the final construct possesses the desired characteristics, such as antigen binding.

[0094] a) Substitution, insertion, and deletion variants

[0095] In some respects, antibody variants with one or more amino acid substitutions are provided. Sites of interest for substitution mutagenesis include CDR and FR. Conserved substitutions are shown under the heading “Preferred Substitutions” in Table D1. Further substantial variations are provided under the heading “Exemplary Substitutions” in Table 1, and are described further below with reference to the amino acid side chain categories. Amino acid substitutions can be introduced into the target antibody, and the product can be screened for desired activities (e.g., preserved / improved antigen binding, reduced immunogenicity, or improved ADCC or CDC).

[0096] Table D1: Amino Acid Substitutions

[0097]

[0098] Amino acids can be grouped based on common side-chain characteristics:

[0099] (1) Hydrophobicity: Leucine, Met, Ala, Val, Leu, Ile;

[0100] (2) Neutral hydrophilicity: Cys, Ser, Thr, Asn, Gln;

[0101] (3) Acidic: Asp, Glu;

[0102] (4) Alkaline: His, Lys, Arg;

[0103] (5) Residues affecting chain orientation: Gly, Pro;

[0104] (6) Fang ethnic group: Trp, Tyr, Phe.

[0105] Non-conservative substitution would require swapping members of one of these categories for members of another category.

[0106] One type of substitution variant involves replacing one or more highly variable region residues of a parent antibody (e.g., a humanized antibody or a human antibody). Typically, one or more resulting variants selected for further research will alter (e.g., improve) certain biological properties (e.g., increased affinity, decreased immunogenicity) and / or will substantially retain certain biological properties of the parent antibody, relative to the parent antibody. An exemplary substitution variant is an affinity-matured antibody, which can be conveniently generated, for example, using phage display-based affinity maturation techniques such as those described herein. Briefly, one or more CDR residues are mutated and the variant antibody is displayed on a phage and screened for specific biological activities (e.g., binding affinity).

[0107] For example, alterations (e.g., substitutions) can be made in the CDR to improve antibody affinity. Such alterations can be made in CDR “hotspots,” which are codon-encoded residues that are frequently mutated during somatic maturation (see, for example, Chowdhury, Methods Mol. Biol. 207:179-196 (2008)), and / or residues that come into contact with the antigen, and the binding affinity of the resulting variant VH or VL is tested. Affinity maturation achieved by constructing and reselecting from a secondary library has been described, for example, by Hoogenboom et al. in Methods in Molecular Biology 178:1-37 (edited by O'Brien et al., Human Press, Totowa, NJ, (2001)). In some aspects of affinity maturation, diversity is introduced into the variable gene selected for maturation by any of a variety of methods (e.g., error-prone PCR, strand shuffling, or oligonucleotide directed mutagenesis). A secondary library is then created. This library is subsequently screened to identify any antibody variant with the desired affinity. Another approach to introducing diversity involves CDR targeting, where several CDR residues (e.g., 4 to 6 residues at a time) are randomized. CDR residues involved in antigen binding can be specifically identified, for example, using alanine scanning mutations or modeling. Specifically, CDR-H3 and CDR-L3 are often targeted.

[0108] In some respects, substitution, insertion, or deletion can occur within one or more CDRs, as long as such changes do not substantially reduce the antibody's ability to bind to the antigen. For example, conserved changes that do not substantially reduce binding affinity (e.g., conserved substitutions as described herein) can be made within the CDR. Such changes can, for example, be external to the antigen-contacting residues in the CDR. In some variant VH and VL sequences provided above, each CDR either remains unchanged or contains more than one, two, or three amino acid substitutions.

[0109] A method for identifying antibody residues or regions that can be targeted for mutation is called "alanine scan mutation," as described in Cunningham and Wells, (1989) Science, 244:1081-1085. In this method, a residue or a group of target residues (e.g., charged residues such as arg, asp, his, lys, and glu) is identified and replaced with a neutral or negatively charged amino acid (e.g., alanine or polyalanine) to determine if the antibody-antigen interaction is affected. Additional substitutions can be introduced at amino acid positions that exhibit functional sensitivity to the initial substitution. Alternatively or additionally, the contact points between the antibody and antigen can be identified using the crystal structure of the antigen-antibody complex. Such contact residues and adjacent residues can be targeted or eliminated as candidates for substitution. Variants can be screened to determine if they possess the desired properties.

[0110] Amino acid sequence insertions include the fusion of amino and / or carboxyl terms of peptides ranging in length from one residue to one hundred or more residues, as well as intra-sequence insertions of one or more amino acid residues. Examples of terminal insertions include antibodies having an N-terminal methionine residue. Other insertion variants of antibody molecules include the fusion of the N-terminus or C-terminus of the antibody with an enzyme (e.g., for ADEPT (antibody-directed enzyme prodrug therapy)) or peptide that increases the antibody's serum half-life.

[0111] b) Glycosylated variants

[0112] In some respects, the antibodies presented herein can be modified to increase or decrease the degree of antibody glycosylation. The addition or deletion of glycosylation sites to antibodies can be conveniently achieved by altering the amino acid sequence to create or remove one or more glycosylation sites.

[0113] When an antibody contains an Fc region, the oligosaccharide associated with it can be modified. Natural antibodies produced by mammalian cells typically contain branched biantennary oligosaccharides, which are usually linked to Asn297 of the CH2 domain of the Fc region via N-bonding. See, for example, Wright et al., TIBTECH 15:26-32 (1997). Oligosaccharides can include various carbohydrates, such as mannose, N-acetylglucosamine (GlcNAc), galactose, and sialic acid, as well as fucose of GlcNAc attached to the “backbone” of the biantennary oligosaccharide structure. In some aspects, the oligosaccharides in the antibodies of the present invention can be modified to produce antibody variants with certain improved properties.

[0114] On the one hand, antibody variants with unfucosylated oligosaccharides are provided, i.e., oligosaccharide structures lacking (directly or indirectly) fucose linked to the Fc region. Such unfucosylated oligosaccharides (also known as "defucosylated" oligosaccharides) are particularly N-linked oligosaccharides that lack the fucose residues linking the first GlcNAc in the stem of the biantennary oligosaccharide structure. On the other hand, antibody variants with an increased proportion of unfucosylated oligosaccharides in the Fc region compared to natural or parental antibodies are provided. For example, the proportion of unfucosylated oligosaccharides can be at least about 20%, at least about 40%, at least about 60%, at least about 80%, or even about 100% (i.e., no fucose oligosaccharides present). The percentage of defucosylated oligosaccharides, as described, for example, in WO 2006 / 082515, and measured by MALDI-TOF mass spectrometry, is the (average) amount of oligosaccharides lacking fucosylate residues relative to the sum of all oligosaccharides linked to Asn 297 (e.g., complex, hybrid, and high-mannose structures). Asn 297 refers to the asparagine residue (EU number of Fc region residues) located at approximately position 297 in the Fc region; however, due to minor sequence variations in antibodies, Asn 297 can also be located approximately ±3 amino acids upstream or downstream of position 297, i.e., between positions 294 and 300. Such antibodies with an increased proportion of non-fucosylated oligosaccharides in the Fc region may exhibit improved FcγRIIIa receptor binding and / or improved effector function, particularly improved ADCC function. See, for example, US 2003 / 0157108 and US 2004 / 0093621.

[0115] Examples of cell lines capable of producing antibodies with reduced fucosylation include Lec13 CHO cells lacking protein fucosylation (Ripka et al., Arch. Biochem. Biophys. 249:533-545 (1986); US 2003 / 0157108; and WO 2004 / 056312, especially in Example 11); and knockout cell lines, such as CHO cells with the α-1,6-fucosylation gene FUT8 knocked out (see, for example, Yamane-Ohnuki et al., Biotech. Bioeng. 87:614-622 (2004); Kanda, Y. et al., Biotechnol. Bioeng., 94(4):680-688 (2006); and WO 2003 / 085107); or cells with reduced or absent GDP-fucose synthesis or transporter activity (see, for example, US2004259150, US2005031613, US2004132140, US2004110282).

[0116] On the other hand, antibody variants provide bipartite oligosaccharides, for example, in which biantennary oligosaccharides linked to the Fc region of the antibody are bipartitely divided by GlcNAc. As mentioned above, such antibody variants can have reduced fucosylation and / or improved ADCC function. Examples of such antibody variants are described, for example, in Umana et al., Nat Biotechnol 17, 176-180 (1999); Ferrara et al., Biotechn Bioeng 93, 851-861 (2006); WO 99 / 54342; WO 2004 / 065540, WO 2003 / 011878.

[0117] Antibody variants having at least one galactose residue in the oligosaccharide linked to the Fc region are also provided. Such antibody variants may have improved CDC function. Such antibody variants are described, for example, in WO 1997 / 30087, WO 1998 / 58964 and WO 1999 / 22764.

[0118] c) Fc region variant

[0119] In some aspects of this invention, the antibody is a full-length IgG antibody.

[0120] In some respects, one or more amino acid modifications may be introduced into the Fc region of the antibody provided herein, thereby generating an Fc region variant. The Fc region variant may contain a human Fc region sequence (e.g., a human IgG1, IgG2, IgG3, or IgG4 Fc region) containing an amino acid modification (e.g., substitution) at one or more amino acid positions.

[0121] In some respects, the present invention considers antibody variants possessing some, but not all, of the effector functions, making them ideal candidates for applications where the in vivo antibody half-life is important but certain effector functions (such as complement-dependent cytotoxicity (CDC) and antibody-dependent cell-mediated cytotoxicity (ADCC)) are unnecessary or detrimental. In vitro and / or in vivo cytotoxicity assays can be performed to confirm a reduction / depletion of CDC and / or ADCC activity. For example, Fc receptor (FcR) binding assays can be performed to ensure that the antibody lacks FcγR binding (and therefore may lack ADCC activity), but retains FcRn binding capacity. The primary cells mediating ADCC, NK cells, express only FcγRIII, while monocytes express FcγRI, FcγRII, and FcγRIII. FcR expression on hematopoietic cells is summarized in Table 3 on page 464 of Ravetch and Kinet, Annu. Rev. Immunol. 9:457-492 (1991). Non-limiting examples of in vitro assays for evaluating ADCC activity of target molecules are described in U.S. Patent No. 5,500,362 (see, for example, Hellstrom, I. et al., Proc. Nat'l. Acad. Sci. USA 83:7059-7063 (1986)) and Hellstrom, I. et al., Proc. Nat'l. Acad. Sci. USA 82:1499-1502 (1985); 5,821,337 (see Bruggemann, M. et al., J. Exp. Med. 166:1351-1361 (1987)). Alternatively, non-radioactive assays may be used (see, for example, the ACTI™ non-radioactive cytotoxicity assay for flow cytometry (Cell Technology, Inc. Mountain View, CA); and the CytoTox 96® non-radioactive cytotoxicity assay (Promega, Madison, WI)). Useful effector cells for such assays include peripheral blood mononuclear cells (PBMCs) and natural killer (NK) cells. Alternatively or additionally, the ADCC activity of the target molecule can be assessed in vivo in animal models, such as those disclosed in Clynes et al., Proc. Nat'l Acad. Sci. USA 95:652-656 (1998). A C1q binding assay can also be performed to confirm that the antibody cannot bind to C1q and therefore lacks CDC activity.See, for example, C1q and C3c binding ELISAs in WO 2006 / 029879 and WO 2005 / 100402. To assess complement activation, CDC assays can be performed (see, for example, Gazzano-Santoro et al., J. Immunol. Methods 202:163 (1996); Cragg, MS et al., Blood 101:1045-1052 (2003); and Cragg, MS and MJ Glennie, Blood 103:2738-2743 (2004)). FcRn binding and in vivo clearance / half-life assays can also be performed using methods known in the art (see, for example, Petkova, SB et al., Int'l. Immunol. 18(12):1759-1769 (2006); WO2013 / 120929 A1).

[0122] Antibodies with reduced effector function include those with substitutions of one or more of the Fc region residues 238, 265, 269, 270, 297, 327, and 329 (US Patent No. 6,737,056). Such Fc mutants include Fc mutants with substitutions at two or more of the amino acids at positions 265, 269, 270, 297, and 327, including the so-called “DANA” Fc mutant, in which residues 265 and 297 are substituted with alanine (US Patent No. 7,332,581).

[0123] Certain antibody variants with improved or reduced binding to FcR are described. (See, for example, U.S. Patent No. 6,737,056; WO 2004 / 056312, and Shields et al., J. Biol. Chem. 9(2):6591-6604(2001).)

[0124] In some respects, antibody variants contain an Fc region with one or more amino acid substitutions that improve ADCC, such as substitutions at positions 298, 333, and / or 334 of the Fc region (EU numbers of the residues).

[0125] In some aspects, the antibody variant includes an Fc region with one or more amino acid substitutions that reduce FcγR binding, such as substitutions at positions 234 and 235 of the Fc region (EU numbers of the residues). In one aspect, the substitutions are L234A and L235A (LALA). In some aspects, the antibody variant further includes D265A and / or P329G in an Fc region derived from the human IgG1 Fc region. In one aspect, in an Fc region derived from the human IgG1 Fc region, the substitutions are L234A, L235A, and P329G (LALA-PG). (See, for example, WO 2012 / 130831). In another aspect, in an Fc region derived from the human IgG1 Fc region, the substitutions are L234A, L235A, and D265A (LALA-DA).

[0126] In some aspects, such as those described in U.S. Patent Nos. 6,194,551, WO 99 / 51642, and Idusogie et al. J. Immunol. 164: 4178-4184 (2000), alterations are made in the Fc region resulting in altered (i.e., improved or reduced) C1q binding and / or complement-dependent cytotoxicity (CDC).

[0127] Antibodies with prolonged half-life and improved neonatal Fc receptor (FcRn) binding (responsible for transferring maternal IgG to the fetus) (Guyer, RL et al., J. Immunol. 117:587 (1976), and Kim, JK et al., J. Immunol. 24:249 (1994)) are described in US2005 / 0014934 (Hinton et al.). These antibodies contain an Fc region with one or more substitutions that improve the binding of the Fc region to FcRn. Such Fc variants include Fc variants with substitutions at one or more of the following Fc region residues: 238, 252, 254, 256, 265, 272, 286, 303, 305, 307, 311, 312, 317, 340, 356, 360, 362, 376, 378, 380, 382, ​​413, 424, or 434, for example, a substitution of Fc region residue 434 (see, for example, U.S. Patent No. 7,371,826; Dall'Acqua, WF et al. J. Biol. Chem. 281 (2006) 23514-23524).

[0128] Studies of the human Fc-human FcRn complex have shown that residues I253, S254, H435, and Y436 are crucial for the interaction (Firan, M. et al., Int. Immunol. 13 (2001) 993; Shields, RL et al., J. Biol. Chem. 276 (2001) 6591-6604). Various mutants of residues 248 to 259, 301 to 317, 376 to 382, ​​and 424 to 437 have been reported and examined in Yeung, YA et al. (J. Immunol. 182 (2009) 7667-7671).

[0129] In some respects, antibody variants contain Fc regions with one or more amino acid substitutions that reduce FcRn binding, for example, substitutions at positions 253, and / or 310, and / or 435 of the Fc region (EU numbers of the residues). In some respects, antibody variants contain Fc regions with amino acid substitutions at positions 253, 310, and 435. In one instance, in the Fc region derived from the human IgG1 Fc region, the substitutions are I253A, H310A, and H435A. See, for example, Grevys, A. et al., J. Immunol. 194 (2015) 5497-5508.

[0130] In some respects, antibody variants comprise Fc regions with one or more amino acid substitutions that reduce FcRn binding, for example, substitutions at positions 310, and / or 433, and / or 436 of the Fc region (EU numbers of the residues). In some respects, antibody variants comprise Fc regions with amino acid substitutions at positions 310, 433, and 436. In one instance, in the Fc region derived from human IgG1, the substitutions are H310A, H433A, and Y436A. (See, for example, WO 2014 / 177460A1).

[0131] In some respects, antibody variants comprise an Fc region having one or more amino acid substitutions that increase FcRn binding, such as substitutions at positions 252, and / or 254, and / or 256 of the Fc region (EU numbers of the residues). In some respects, antibody variants comprise an Fc region having amino acid substitutions at positions 252, 254, and 256. In one respect, in the Fc region derived from the human IgG1 Fc region, the substitutions are M252Y, S254T, and T256E. See also Duncan & Winter, Nature 322:738-40 (1988); U.S. Patent No. 5,648,260; U.S. Patent No. 5,624,821; and WO 94 / 29351 for other examples of Fc region variants.

[0132] The C-terminus of the heavy chain of the antibody reported herein may be a full C-terminus ending with the amino acid residue PGK. The C-terminus of the heavy chain may be a shortened C-terminus in which one or two C-terminal amino acid residues have been removed. In a preferred aspect, the C-terminus of the heavy chain is a shortened C-terminus ending with PG. In one aspect of all aspects reported herein, as specified herein, an antibody comprising a heavy chain including a C-terminal CH3 domain comprises a C-terminal glycine-lysine dipeptide (G446 and K447, EU index numbers of amino acid positions). In one aspect of all aspects reported herein, as specified herein, an antibody comprising a heavy chain including a C-terminal CH3 domain comprises a C-terminal glycine residue (G446, EU index number of amino acid position).

[0133] d) Cysteine-engineered antibody variants

[0134] In some respects, it may be desirable to generate cysteine-engineered antibodies, such as THIOMAB™ antibodies, in which one or more residues of the antibody are replaced by cysteine ​​residues. In certain respects, the substituted residues are present at accessible sites on the antibody. By replacing those residues with cysteine, reactive thiol groups are thereby located at accessible sites on the antibody and can be used to conjugate the antibody to other parts (e.g., pharmaceutical parts or linker-pharmaceutical parts) to form immunoconjugates, as further described herein. Cysteine-engineered antibodies can be generated, for example, as described in U.S. Patent Nos. 7,521,541, 8,30,930, 7,855,275, 9,000,130, or WO 2016040856.

[0135] e) Antibody derivatives

[0136] In some respects, the antibodies provided herein can be further modified to include additional non-protein moieties known in the art and readily available. Suitable moieties for antibody derivatization include, but are not limited to, water-soluble polymers. Non-limiting examples of water-soluble polymers include, but are not limited to, polyethylene glycol (PEG), copolymers of ethylene glycol / propylene glycol, carboxymethyl cellulose, dextran, polyvinyl alcohol, polyvinylpyrrolidone, poly-1,3-dioxolane, poly-1,3,6-trioxane, ethylene / maleic anhydride copolymers, polyamino acids (homogeneous or random copolymers) and dextran or poly(n-vinylpyrrolidone) polyethylene glycol, propylene glycol homopolymers, polypropylene oxide / ethylene oxide copolymers, polyoxyethyleneized polyols (e.g., glycerol), polyvinyl alcohol, and mixtures thereof. PEG-propionaldehyde may be advantageous in manufacturing due to its stability in water. Polymers can have any molecular weight and can be branched or unbranched. The number of polymers attached to the antibody can vary, and if more than one polymer is attached, they can be the same or different molecules. Typically, the number and / or type of polymers used for derivatization can be determined based on the following considerations, including but not limited to the specific properties or functions of the antibody to be improved, and whether the antibody derivative will be used for a limited therapy.

[0137] Recombination Method and Composition

[0138] Antibodies can be generated using recombinant methods and compositions, such as those described in US 4,816,567. For these methods, one or more isolated nucleic acids encoding the antibody are provided.

[0139] In the case of natural antibodies or fragments of natural antibodies, two nucleic acids are required: one for the light chain or a fragment thereof, and one for the heavy chain or a fragment thereof. These nucleic acids encode the amino acid sequence of the VL containing the antibody and / or the amino acid sequence of the VH containing the antibody (e.g., the light chain and / or heavy chain of the antibody). These nucleic acids can be expressed on the same expression vector or on different expression vectors.

[0140] In the case of bispecific antibodies with heterodimeric heavy chains, four nucleic acids are required: one for the first light chain, one for the first heavy chain containing the Fc region of the first heteromonomer polypeptide, one for the second light chain, and one for the second heavy chain containing the Fc region of the second heteromonomer polypeptide. These four nucleic acids can be contained in one or more nucleic acid molecules or expression vectors. These nucleic acids encode the amino acid sequence constituting the first VL of the antibody and / or the amino acid sequence constituting the first VH containing the Fc region of the first heteromonomer and / or the amino acid sequence constituting the second VL of the antibody and / or the amino acid sequence constituting the second VH containing the Fc region of the second heteromonomer (e.g., the first light chain and / or the second light chain and / or the first heavy chain and / or the second heavy chain of the antibody). These nucleic acids can be on the same expression vector or on different expression vectors; typically, these nucleic acids are located on two or three expression vectors, meaning that one vector can contain more than one of these nucleic acids. Examples of these bispecific antibodies are CrossMab (see, for example, Schaefer, W. et al., PNAS, 108 (2011) 11187-1191). For example, one of the heterologous single-chain heavy chains contains a so-called "palmock mutation" (T366W, and optionally one of S354C or Y349C), and the other contains a so-called "mortar mutation" (T366S, L368A, and Y407V, and optionally Y349C or S354C) (see, for example, Carter, P. et al., Immunotechnol. 2 (1996) 73), which are indexed according to EU index numbers.

[0141] In one aspect, isolated nucleic acids encoding antibodies used in the methods reported herein are provided.

[0142] In one aspect, a method for preparing the antibody of the present invention is provided, wherein the method includes culturing a host cell containing a nucleic acid encoding an antibody as provided above under conditions suitable for antibody expression, and optionally recovering the antibody from the host cell (or host cell culture medium).

[0143] In the recombinant production of the antibodies of the present invention, nucleic acids encoding the antibodies, such as those described above, are isolated and inserted into one or more vectors for further cloning and / or expression in host cells. Such nucleic acids can be readily isolated and sequenced using conventional procedures (e.g., by using oligonucleotide probes capable of specifically binding to genes encoding the heavy and light chains of the antibody), or obtained by recombinant methods or by chemical synthesis.

[0144] Suitable host cells for cloning or expressing vectors encoding antibodies include prokaryotic or eukaryotic cells as described herein. Antibodies can be generated in bacteria, for example, especially when glycosylation and Fc effector function are not required. For information on the expression of antibody fragments and peptides in bacteria, see, for example, US 5,648,237, US 5,789,199, and US 5,840,523. (See also Charlton, KA, in: Methods in Molecular Biology, Vol. 248, Lo, BKC (ed.), Humana Press, Totowa, NJ (2003), pp. 245-254, describing the expression of antibody fragments in *E. coli*.) Antibodies can be separated from the bacterial cell paste in a soluble fraction after expression and can be further purified.

[0145] Besides prokaryotes, eukaryotic microorganisms such as filamentous fungi or yeast are also suitable cloning or expression hosts for antibody-encoding vectors. These eukaryotic microorganisms, including fungal and yeast strains, have "humanized" glycosylation pathways, resulting in antibodies with partial or complete human glycosylation patterns. See Gerngross, TU, Nat. Biotech. 22 (2004) 1409-1414; and Li, H. et al., Nat. Biotech. 24 (2006) 210-215.

[0146] Suitable host cells for expressing (glycosylated) antibodies also originate from multicellular organisms (invertebrates and vertebrates). Examples of invertebrate cells include plant cells and insect cells. Many baculovirus strains have been identified that can be used in conjunction with insect cells, particularly for transfecting cells of the meadow armyworm (Spodoptera frugiperda).

[0147] Plant cell cultures can also be used as hosts. See, for example, US 5,959,177, US 6,040,498, US 6,420,548, US 7,125,978 and US 6,417,429 (which describe PLATNIBODIES™ technology for producing antibodies in transgenic plants).

[0148] Vertebrate cells can also be used as hosts. For example, mammalian cell lines adapted for growth in suspension may be useful. Other examples of available mammalian host cell lines include: monkey kidney CV1 line (COS-7) transformed from SV40; human embryonic kidney lines (e.g., 293 or 293T cells as described in Graham, FL et al., J. Gen Virol. 36 (1977) 59-74); young hamster kidney cells (BHK); mouse testicular supporting cells (e.g., TM4 cells as described in Mather, JP., Biol. Reprod. 23 (1980) 243-252); monkey kidney cells (CV1); African green monkey kidney cells (VERO-76); human cervical cancer cells (HeLa); canine kidney cells (MDCK); buffalo rat hepatocytes (BRL 3A); human lung cells (W138); human hepatocytes (Hep G2); mouse mammary tumor cells (MMT 060562); and TRI cells (e.g., Mather, JP et al., Annals NYAcad.). (As described in Sci. 383 (1982) 44-68); MRC 5 cells; and FS4 cells. Other useful mammalian host cell lines include Chinese hamster ovary (CHO) cells, including DHFR-CHO cells (Urlaub, G. et al., Proc. Natl. Acad. Sci. USA 77 (1980) 4216-4220); and myeloma cell lines such as Y0, NSO, and Sp2 / 0. For a review of certain mammalian host cell lines suitable for antibody production, see, for example, Yazaki, P. and Wu, AM, Methods in Molecular Biology, Vol. 248, Lo, BKC (ed.), Humana Press, Totowa, NJ (2004), pp. 255-268.

[0149] On the one hand, the host cells are eukaryotic cells, such as Chinese hamster ovary (CHO) cells or lymphocytes (e.g., Y0, NSO, Sp20 cells).

[0150] Pharmaceutical Composition

[0151] In another aspect, pharmaceutical compositions comprising any of the antibodies provided herein are provided, for example, in any of the following treatment methods. In one aspect, the pharmaceutical composition comprises any of the antibodies provided herein and a pharmaceutically acceptable carrier. In another aspect, the pharmaceutical composition comprises any of the antibodies provided herein and at least one additional therapeutic agent, such as those described below.

[0152] Pharmaceutical compositions (formulations) of the antibodies of the present invention as described herein can be prepared by combining the antibody with a pharmaceutically acceptable carrier or excipient known to those skilled in the art. See, for example, Remington's Pharmaceutical Sciences, 16th edition, Osol, A. (ed., 1980), Shire S., Monoclonal Antibodies: Meeting the Challenges in Manufacturing, Formulation, Delivery and Stability of Final Drug Product, 1st edition, Woodhead Publishing (2015), §4, and Falconer RJ, Biotechnology Advances (2019), 37, 107412. Exemplary pharmaceutical compositions of the antibodies of the present invention as described herein are lyophilized, aqueous, frozen, etc.

[0153] Pharmaceutically acceptable carriers are generally non-toxic to the treated individual at the doses and concentrations used, and include, but are not limited to: buffers such as histidine, phosphates, citrates, acetates, and other organic acids; antioxidants, including ascorbic acid and methionine; preservatives (such as octadecyl dimethyl benzyl ammonium chloride; hexamethyl diammonium chloride; benzalkonium chloride; benzyl chloride; phenol, butanol, or benzyl alcohol; alkyl esters of p-hydroxybenzoate, such as methylparaben or propylparaben; catechol; resorcinol; cyclohexanol; 3-pentanol; m-cresol); and low molecular weight (less than about 10). (1 residue) polypeptides; proteins, such as serum albumin, gelatin, or immunoglobulins; hydrophilic polymers, such as polyvinylpyrrolidone; amino acids, such as glycine, glutamine, asparagine, histidine, arginine, or lysine; monosaccharides, disaccharides, and other carbohydrates, including glucose, mannose, or dextrin; chelating agents, such as EDTA; sugars, such as sucrose, mannitol, trehalose, or sorbitol; salt-forming counterions, such as sodium; metal complexes (e.g., zinc protein complexes); and / or nonionic surfactants, such as polyethylene glycol (PEG).

[0154] The pharmaceutical compositions described herein may also contain more than one active ingredient essential for the specific indication being treated, preferably active ingredients having complementary activities that do not adversely affect each other. Such active ingredients are appropriately combined in amounts effective for the intended purpose.

[0155] Pharmaceutical compositions intended for internal administration are typically sterile. For example, sterility can be readily achieved through filtration using a sterile filter membrane.

[0156] Treatment

[0157] Any antibodies provided herein can be used for therapeutic purposes. In one aspect, antibodies of the present invention are provided for use as medicines. In another aspect, antibodies of the present invention are provided for treating neurological disorders.

[0158] In another aspect, the present invention provides the use of the antibody of the present invention in the manufacture or preparation of a medicament. In one aspect, the medicament is used to treat a neurological disease. In another aspect, a method of using the medicament to treat a neurological disease includes administering an effective amount of the medicament to an individual suffering from said disease. In one such aspect, as described below, the method further includes administering an effective amount of at least one other therapeutic agent to the individual. The "individual" according to any of the foregoing aspects can be a human being.

[0159] 3. Specific embodiments of the present invention

[0160] Specific embodiments of the present invention are listed below.

[0161] 1. A multispecific antibody, wherein the antibody binds to at least a first antigen and a second antigen, wherein the average concentration of the first antigen in a first human body fluid is at least one order of magnitude higher than the average concentration of the first antigen in a second human body fluid, wherein the apparent affinity of the antibody for the second antigen is higher in the presence of the average concentration of the first antigen in the first human body fluid than the affinity of the antibody for the second antigen in the presence of the average concentration of the first antigen in the second human body fluid.

[0162] 2. The antibody according to Example 1, wherein the first human fluid is human peripheral blood.

[0163] 3. The antibody according to Example 1 or 2, wherein the second human fluid is human cerebrospinal fluid.

[0164] 4. The antibody according to one of the foregoing embodiments, wherein the first antigen is selected from HSA, IgG and transferrin.

[0165] 5. A multispecific antibody that binds to human serum albumin (HSA) and a second antigen, wherein the apparent affinity of the antibody for the second antigen is higher in the presence of an average concentration of HSA in the first human body fluid than in the presence of an average concentration of HSA in the second human body fluid, wherein the average concentration of HSA in the first human body fluid is lower than the average concentration of HSA in the second human body fluid.

[0166] 6. Multispecific antibodies that bind to human serum albumin (HSA) and a second antigen, wherein the apparent affinity of the antibody for the second antigen is higher in the presence of an average concentration of HSA in human cerebrospinal fluid than in the presence of an average concentration of HSA in human peripheral blood.

[0167] 7. The multispecific antibody according to Example 1 or 2, wherein the second antigen is a therapeutic antigen.

[0168] 8. The multispecific antibody according to one of the foregoing embodiments, wherein the second antigen is a brain target.

[0169] 9. The multispecific antibody according to one of the foregoing embodiments, wherein the second antigen is selected from human VEGF-A, IL-1β, CD32A, CD32B, CD79b, 41BB, IFNα, and Ox40.

[0170] 10. The multispecific antibody according to any of the foregoing embodiments, wherein the antibody is bispecific.

[0171] 11. The multispecific antibody according to any of the foregoing embodiments, wherein the antibody is or comprises a bispecific Fab fragment.

[0172] 12. The multispecific antibody according to one of the foregoing embodiments, wherein the antibody comprises a double complementary Fv fragment.

[0173] 13. The multispecific antibody according to one of the foregoing embodiments, wherein the antibody comprises a double complementary Fv fragment having two overlapping complementary sites.

[0174] 14. The multispecific antibody according to one of the foregoing embodiments, wherein the antibody is DutaFab.

[0175] 15. The multispecific antibody according to any one of the foregoing embodiments, wherein the antibody is or comprises a bispecific Fab fragment, wherein the bispecific Fab fragment comprises a first complementary site capable of binding to a first antigen and a second complementary site capable of binding to a second antigen, wherein one of the complementary sites from the first and second complementary sites comprises amino acid residues from H-CDR2, L-CDR1, and L-CDR3, and the other complementary site comprises amino acid residues from H-CDR1, H-CDR3, and L-CDR2.

[0176] 16. The multispecific antibody according to any one of the foregoing embodiments, wherein the first antigen is HSA, and wherein the second antigen is a therapeutic target.

[0177] 17. The multispecific antibody according to any one of the foregoing embodiments, wherein the first antigen is HSA and wherein the second antigen is a brain target.

[0178] 18. The multispecific antibody according to any one of the foregoing embodiments, wherein the first antigen is HSA, and wherein the second antigen is selected from human VEGF-A, IL-1β, CD32A, CD32B, CD79b, 41BB, IFNα and Ox40.

[0179] 19. The multispecific antibody according to any one of the foregoing embodiments, wherein the first antigen is HSA and wherein the second antigen is human VEGF-A.

[0180] 20. The multispecific antibody according to any one of the foregoing embodiments, wherein the antibody is or comprises a bispecific Fab fragment, wherein the bispecific Fab fragment comprises a first complementary site capable of binding to HSA and a second complementary site capable of binding to human VEGF-A, wherein one of the complementary sites from the first and second complementary sites comprises amino acid residues from H-CDR2, L-CDR1, and L-CDR3, and the other complementary site comprises amino acid residues from H-CDR1, H-CDR3, and L-CDR-2.

[0181] 21. The multispecific antibody according to any one of the foregoing embodiments, wherein the antibody comprises

[0182] a) The VH domain of SEQ ID NO:01 and the VL domain of SEQ ID NO:02,

[0183] b) The VH domain of SEQ ID NO:03 and the VL domain of SEQ ID NO:04,

[0184] c) The VH domain of SEQ ID NO:05 and the VL domain of SEQ ID NO:06,

[0185] d) The VH domain of SEQ ID NO:07 and the VL domain of SEQ ID NO:08,

[0186] e) The VH domain of SEQ ID NO:09 and the VL domain of SEQ ID NO:10,

[0187] f) The VH domain of SEQ ID NO:19 and the VL domain of SEQ ID NO:20,

[0188] g) The VH domain of SEQ ID NO:21 and the VL domain of SEQ ID NO:22,

[0189] h) The VH domain of SEQ ID NO:23 and the VL domain of SEQ ID NO:24,

[0190] i) The VH domain of SEQ ID NO:25 and the VL domain of SEQ ID NO:26,

[0191] j) The VH domain of SEQ ID NO:27 and the VL domain of SEQ ID NO:28,

[0192] k) The VH domain of SEQ ID NO:29 and the VL domain of SEQ ID NO:30,

[0193] l) The VH domain of SEQ ID NO:31 and the VL domain of SEQ ID NO:32,

[0194] m) The VH domain of SEQ ID NO:33 and the VL domain of SEQ ID NO:34, or

[0195] n) The VH domain of SEQ ID NO:35 and the VL domain of SEQ ID NO:36.

[0196] 22. The multispecific antibody according to one of the foregoing embodiments, wherein the apparent affinity of the antibody for its second target antigen in a body fluid containing 3 µM HSA is at least an order of magnitude lower than the affinity of the antibody of the present invention for its second target antigen in a body fluid containing 600 µM HSA.

[0197] 23. A nucleic acid encoding a multispecific antibody according to one of the foregoing embodiments.

[0198] 24. A vector comprising the nucleic acid according to Example 18.

[0199] 25. A host cell comprising the nucleic acid as described in Example 18.

[0200] 26. A method for producing a multispecific antibody according to any one of Examples 1 to 18, the method comprising culturing a host cell according to Example 20 under conditions suitable for antibody expression.

[0201] 27. A pharmaceutical composition comprising an antibody according to any one of Examples 1 to 18 and a pharmaceutically acceptable carrier.

[0202] 28. The antibody according to any one of Examples 1 to 18 or the pharmaceutical composition according to Example 22, used as a medicine.

[0203] 29. The antibody according to any one of Examples 1 to 18 or the pharmaceutical composition according to Example 22, for the treatment of neurological disorders.

[0204] Description of amino acid sequence

[0205]

[0206] Example

[0207] The following examples are provided to aid in understanding the invention, the true scope of which is set forth in the appended claims. It should be understood that modifications may be made to the described procedures without departing from the spirit of the invention.

[0208] Example 1:

[0209] The formation of DutaFab conjugates that specifically bind to human serum albumin (HSA)

[0210] Bispecific anti-HSA / anti-VEGF-A Fab fragments were generated using phage display and methods similar to those previously described, for example, in WO2012 / 163520 and Beckmann, R. et al. Nat Commun. 2021;12(1):708.

[0211] To obtain bispecific Fab, several dedicated phage display libraries of synthetic Fab fragments were used. While half of the Fab CDRs were diverse, the remaining three CDR regions remained non-diversified. In contrast to the approach of WO2012 / 163520, which uses germline sequences (referred to as “virtual” sequences) in non-diversified CDRs, the invariant sequences in the libraries used in this work (with one exception) represent complementary sites capable of binding to VEGF-A. In all libraries, the CH1 domain of the Fab fragment was fused to a truncated gene III protein via a linker for phage display. Two basic designs can be distinguished among the libraries used: one type of library is designed to screen for bispecific Fab fragments, where the HSA complement mainly contains amino acid residues from CDR-H1, CDR-H3, and CDR-L2 (referred to as "H-side library" in this paper), and the other type of library is designed to screen for bispecific Fab fragments, where the HSA complement mainly contains amino acid residues from the remaining CDRs (i.e., CDR-H2, CDR-L1, and CDR-L3) (referred to as "L-side library" in this paper).

[0212] In the case of the L-side library, the complementary bit that can bind to VEGF-A is derived from the VEGF-A binding complementary bit described in WO 2021 / 198034. In the case of the H-side library, the VEGF-A binding complementary bit is obtained as described in PCT / EP2023 / 062545.

[0213] Phage library panning was used to enrich conjugates targeting human serum albumin or mouse serum albumin from each library. Following panning, plasmid mini-preparations were generated for the enriched phage vector pools. These mini-preparations were digested with restriction enzymes to remove the region encoding a truncated gene III protein, and then re-circularized by ligation to obtain expression vector pools encoding soluble Fab fragments encoding the enriched serum albumin conjugates. These vector pools were transformed into TG1 *E. coli* cells, and single colonies were selected and cultured for soluble expression of single Fab clones in microtiter plates. The supernatant containing the soluble Fab fragments was screened for binding to serum albumin and VEGF-A using standard ELISA methods.

[0214] Based on the screening data, the anti-serum albumin Fab fragment was selected, and DNA plasmids were prepared and sequenced on TG1 E. coli clones that produced specific bindings to obtain VH and VL sequence pairs. The sequence pairs together encode a bispecific Fab fragment that binds specifically to serum albumin or specifically to serum albumin and VEGF-A, respectively.

[0215] Based on the screening data shown in Table 1, five bispecific clones expected to bind to serum albumin and VEGF-A were used for further characterization. Fab P1AJ5895 (an antibody that binds to albumin only) was included as a monospecific control. This Fab was derived from a phage library containing a germline invariant CDR.

[0216] All Fabs are expressed using constant domains as provided in SEQ ID NO: 13 (CH1) and SEQ ID NO: 14 (Ckappa).

[0217] Table 1. Amino acid sequences of bispecific Fab clones binding to serum albumin and VEGF-A

[0218]

[0219] Example 2:

[0220] The binding of Fab to albumin from humans, cynomolgus monkeys, and mice was characterized by surface plasmon resonance.

[0221] To determine the binding kinetics of the generated bispecific Fab with serum albumin from humans (HSA), cynomolgus monkeys (CSA), and mice (MSA) as well as the second target VEGF-A, surface plasmon resonance (SPR) measurements were performed using an 8K system from Cytiva.

[0222] To test interactions with VEGF-A (glycosylated VEGF165, HZ-1153, Humanzyme), a CM3 chip (BR100536, Cytiva) was directly immobilized on VEGF-A at a concentration of 0.5 µg / ml via chemical coupling using an EDC / NHS, and four different concentrations of Fab were used as soluble analytes. A similar approach was used for a range of albumin molecules from humans, cynomolgus monkeys, and mice.

[0223] As a positive control expected to bind to albumin but not to VEGF-A, the monospecific Fab P1AJ5895 and Fab fragment “CA645” (VH of SEQ ID NO:15, VL of SEQ ID NO:16, described in R. Adams et al., Mabs 2016; 8(7):1336-1346) obtained as described above were used.

[0224] As a positive control for VEGF-A binding that does not bind to serum albumin and is therefore expected to bind to its target independently of the level of serum albumin in solution, the antibody Fab fragment P1AL9447 (VH of SEQ ID NO:17, VL of SEQ ID NO:18, disclosed in WO 2020 / 127873) was used.

[0225] Albumin binding

[0226] The albumin binding results of the tested Fab fragments are shown in Figure 1, where SPR traces are provided for serum albumin from humans, cynomolgus monkeys, and mice at concentrations of 0.25 nM, 2.5 nM, and 25 nM (control: 0 nM). Note that the same Y-axis range was plotted for a given target to allow for rapid visual differentiation between strong binding and weak binding or complete lack of binding.

[0227] Table 2 provides affinity data obtained from fitting individual data.

[0228] Table 2. Binding affinities of Fab to albumin from different species, determined by surface plasmon resonance (SPR). Association and dissociation rates are also provided for HSA. The numbers in parentheses indicate interactions that were clearly identifiable in the experiments, but due to the low affinity, the uncertainty of the affinity fit values ​​is high.

[0229]

[0230] It was demonstrated that two of the Fab fragments, P1AJ5849 and P1AJ5862, exhibited cross-reactivity across all three species, while the other three Fab fragments, P1AJ0400, P1AJ0293, and P1AJ0279, showed similar binding behavior to human and cynomolgus monkey albumin, but no related signal was shown for binding to mouse serum albumin.

[0231] The monospecific serum albumin-binding control antibody P1AJ5895 preferentially binds to human serum albumin, but shows only weak binding to albumin from cynomolgus monkeys, and no discernible binding to mouse serum albumin.

[0232] The second control antibody, CA645, showed high affinity binding to albumin from all three species tested, with values ​​in the low single-digit nanomolar range. The determined affinity was in complete agreement with literature values ​​of 0.82 nM for HSA and 2.9 nM for MSA (R. Adams et al., Mabs 2016; 8(7):1336-1346), demonstrating the effectiveness of the SPR setup used in the experiment.

[0233] VEGF-A binding

[0234] The corresponding data for SPR-based affinity measurements of VEGF-A by the bispecific Fabs are shown in Figure 1. For all bispecific Fabs, the affinity, association rate, and dissociation rate obtained by fitting the SPR data were comparable (Table 3), and the dissociation constants ranged from 78 pM to 236 pM. As expected, the monospecific serum albumin conjugate P1AJ5895 did not show measurable binding to VEGF-A.

[0235] Table 3. VEGF-A binding affinity of bispecific Fab determined by surface plasmon resonance (SPR).

[0236]

[0237] Example 3:

[0238] Evidence of the mutual repulsion binding of the bispecific Fab targets VEGF-A and HSA.

[0239] A key prerequisite for generating antibodies exhibiting conditional activity dependent on serum albumin concentration is that they bind to their targets in a mutually exclusive manner, i.e., binding to one of them but not both simultaneously. To determine whether the bispecific Fab generated in Example 1 met this requirement, a “bridging” SPR assay was used.

[0240] In this assay, VEGF-A was coated onto an SPR chip and subsequently used to immobilize bispecific Fab molecules. In a third step, human serum albumin was used as a soluble analyte and applied to the complex of the VEGF-A and Fab fragments. For comparison, human serum albumin was not added in the third step. Two possible outcomes are possible: if Fab is able to bind to both targets simultaneously, the albumin binding event will be visible as an additional signal in the SPR trace compared to the SPR trace without albumin. Conversely, if Fab cannot bind to both targets simultaneously and binds them in a mutually repulsive manner, this will result in no additional response compared to the control trace, and the two SPR traces will be identical. A plot of these possible outcomes is provided in Figure 2A for illustration.

[0241] For the experiments, a chip directly immobilized with VEGF-A (VEGF165, glycosylated, Humanzyme HZ-1153) was used. First, bispecific Fab (100 nM) or buffer was injected as a parallel control, followed by injection of 1000 nM human serum albumin, again using buffer only as a parallel control. To determine the ability to bind to both targets, sensor plots were compared with and without the second antigen (1000 nM). The results of the SPR run with serum albumin are shown in Figure 2A for the control, monospecific VEGF-A, and albumin conjugates, respectively. As expected, normal binding curves were shown, with little dissociation of VEGF-A alone and no signal at all for albumin alone.

[0242] The corresponding results for bispecific anti-VEGF-A / anti-HSA Fab are shown in Figure 2B.

[0243] In all cases, no significant albumin binding events were observed, demonstrating that no bispecific Fab fragment can bind to both targets simultaneously. Therefore, all bispecific anti-VEGF-A / anti-HSA Fabs used in this study are mutually exclusive antigen conjugates.

[0244] Example 4:

[0245] Characterization of HSA concentration-dependent binding of VEGF-A to bispecific anti-VEGF-A / anti-HSA antibodies using ELISA.

[0246] The following experiments were designed to demonstrate that the bispecific anti-VEGF-A / anti-HSA Fab, which binds to two targets in a mutually exclusive manner, does indeed exhibit conditional VEGF-A binding dependent on serum albumin concentration.

[0247] To characterize how different albumin concentrations affect the ability of bispecific Fab to bind VEGF-A, a series of competitive ELISA assays were performed. Briefly, a series of concentrations of Fab were incubated with a constant concentration of VEGF-A, and then the unbound VEGF-A was quantified in the ELISA setup described in more detail below. To determine the dependence of VEGF-A binding on the concentration of albumin in solution, experiments were performed either in the absence of albumin or by including a constant concentration of albumin in the mixture, varying the constant albumin concentration in parallel experiments to cover concentrations of 0.3 µM, 3 µM, 30 µM, 300 µM, and 600 µM. The VEGF-A conjugate P1AL9447, which does not bind albumin and is therefore not expected to show any activity dependence on albumin concentration, was used as a control. Importantly, the entire assay setup required the absence of albumin from any species as an inhibitor, such as commonly used bovine serum albumin or human serum albumin. The assays and results are described in more detail below.

[0248] To obtain VEGF receptor-coated plates for capturing unbound VEGF-A, Maxisorp (442404, Thermo Fisher Scientific) 96-well plates were coated with 50 µl / well of Fcγ fragment-specific goat anti-human IgG (Jackson Immuno research, 109-005-098) in sodium bicarbonate buffer, pH 9.4, and incubated at room temperature for 30 min. After washing twice with PBS-T (PBS, pH 7.2, supplemented with 0.1% v / v Tween-20), the plates were blocked with 200 µl / well of 2% casein-PBS-T and incubated at room temperature for 1 h. The plate was then washed twice more with PBS-T buffer and incubated for at least 30 minutes in the final step with a solution containing 0.3 µg / ml VEGFR1 / Flt-1Fc (321-FL-050 / CF, R&D Systems).

[0249] In individual 96-well polystyrene plates, with or without HSA diluent, pre-incubate at room temperature with a series of Fab concentrations (ranging from 600 nM to 0.29 nM) using a 1:2 dilution step, producing final concentrations of 0.3 µM, 3 µM, 30 µM, 300 µM, or 600 µM HSA. After Fab and HSA pre-incubation, add VEGF-A (VEGF165, glycosylated, Humanzyme HZ-1153) to a final concentration of 720 pM (=0.15 ng / ml) and incubate for 5 minutes.

[0250] The pre-incubation mixture of Fab, HSA, and VEGF-A was then transferred to each well of a coated, blocked, VEGF-R1-captured, and washed (2x) Maxisorp plate for a total incubation time of 5 minutes. After washing twice with PBS-T, 50 µl / well of biotinylated anti-VEGF165 (R&D systems, BAF293 c=0.2 µg / ml) and horseradish peroxidase-labeled streptavidin (KPL-SA-HRP; 1:2000 dilution of 0.5 mg / ml stock solution) were added and incubated for 30 minutes. The plate was thoroughly washed six times with PBS-T before adding the chromogenic horseradish peroxidase substrate 3,3',5,5'-tetramethylbenzidine (TMB) for detection. After approximately 3 minutes, the colorimetric reaction was stopped by adding 1 N H2SO4, and the absorbance of each well was measured at a wavelength of 450 nm using a Biotek ElX808 plate reader.

[0251] To make the data clearer, the measurement signals were normalized. This is advantageous, especially considering the non-specific effect of high HSA concentrations on the ELISA response, which can be attributed to the blocking activity of HSA. Normalization was performed by defining the average of the ELISA responses at the four lowest HSA concentrations for each individual Fab as the 100% response signal (maximum). The 0% response signal (minimum) was obtained by averaging the four highest concentrations measured for the positive control P1AL9447, which had never had HSA added. Since P1AL9447 was present on every plate, 0% was calculated plate-by-plate. The normalized value was then calculated as follows: (signal - minimum) / (maximum - minimum) * 100.

[0252] Figure 3A presents the plotted data, with one plot showing all experiments for a bispecific Fab covering various HSA concentrations. As expected, the control bound to serum albumin only (Figure 3B) did not show any concentration-dependent ability to inhibit VEGF-A binding to the coated receptor, regardless of the presence of HSA (as expected). The Fab that bound VEGF-A but not serum albumin showed the ability to bind VEGF-A by inhibiting its binding to the coated receptor, but the inhibition curves were very similar in the absence of HSA and in the presence of high concentrations of HSA (Figure 3B).

[0253] The behavior of bispecific Fab, capable of binding both albumin and VEGF-A, differs significantly (Figures 3A to 3E). Here, in all cases, a strong transition in the binding curve towards higher Fab concentrations is evident. The IC50 values ​​for the fitted data are provided in Table 4.

[0254] Table 4. Competitive ELISA. For the bispecific anti-VEGF-A / anti-HSA Fab and the non-albumin-bound VEGF-A specific control shown, the inhibition of VEGF-A binding to the VEGF receptor depended on albumin concentration. Values ​​are provided in nanomoles.

[0255]

[0256] By comparing the concentration of human serum albumin in peripheral circulation (approximately 600 µM) with that in cerebrospinal fluid (approximately 3 µM), an IC50 variation of approximately 18-fold can be obtained, for example, for P1AJ0293. This opportunity can be used to elicit differences in pharmacodynamic effects between the periphery and the brain, or differences in drug behavior in the periphery, such as reduced target-mediated drug disposition compared to drugs without albumin-dependent binding activity.

[0257] Example 5:

[0258] In vitro cell-based reporter gene assay

[0259] To investigate whether the presence of albumin elicits a corresponding biological effect on the biochemical activity of bispecific anti-VEGF-A / anti-HSA Fab, a reporter gene assay was employed. In this assay, the HEK293 cell line was stably transfected with an expression vector for human vascular endothelial growth factor receptor-2 (VEGFR2; also known as kinase insertion domain receptor [KDR]) and an NFAT-RE-luc2P luciferase reporter gene construct (NFAT: nuclear factor for activating T cells).

[0260] The report cell line was a stable monoclonal cell line purchased from Promega (GloResponse™ NFAT-RE-luc2P HEK293 cell line; catalog number E8510). Furthermore, the cell line was prepared by acCELLerate GmbH (Hamburg) for a "thaw and use" implementation, meaning the cells were applied directly after thawing without prior incubation. Assays were performed according to the manufacturer's protocol, with the important additional adjustment being the use of serum-free and therefore serum albumin-free medium to resuspend the cells for assays. For this purpose, we used Panserin 293S (catalog number P04-710609) from PAN-Biotech.

[0261] VEGF-mediated VEGFR2 activation induces NFAT-mediated luciferase reporter gene expression. The assay was performed in 384-well microplates of white polystyrene cell culture (Greiner Bio One, catalog number #781098) without any washing steps.

[0262] In this assay, serial dilutions of control and sample (15 µL each) were pre-incubated with 400 pM recombinant human VEGF165 (rhVEGF165, R&D Systems catalog number 293-VE). Pre-incubation was performed in the absence or with concentrations of 0.3 µM, 3 µM, 30 µM, 300 µM, and 600 µM human serum albumin (Millipore catalog number 126658-5GM). Panserin 293S was used as the culture medium. HSA was added directly to the medium and the mixture was incubated for 30 min. Simultaneously, NFAT-RE luc2P / KDR HEK293 cells were thawed, centrifuged at 130 xg for 5 min, and resuspended in Panserin 293S after discarding the supernatant. Cell viability was monitored again using Panserin 293S to achieve a yield of 75% or higher, and cell density was adjusted to 1 million cells per ml.

[0263] Subsequently, 15 µl of cell suspension was added to a mixture of Fab, VEGF-A, and HSA, and incubated for 4 hours in an incubator at 37°C, 95% humidity, and 5% CO2 atmosphere. After equilibration for 30 minutes, luciferase substrate was added using a ONE-Glo EX luciferase assay system from Promega catalog number E8120. The dose-dependent inhibition of NFAT-mediated luciferase reporter gene expression was quantified by measuring the luminescence signal, which was measured using an Envision instrument (Perkin Elmer). Each plate contained a minimum response reference (medium medium only) and a maximum response reference (400 pM VEGF in medium, without Fab). The corresponding luminescence signals were used for plate-by-plate normalization. Dosage-response results were plotted in GraphPad Prism as a percentage of normalized response. IC50 values ​​were calculated using a four-parameter logistic model in the presence of observed inhibition using GraphPad Prism 8.4.2 software.

[0264] The results are shown in Figure 4. While the monospecific anti-VEGF-A control antibody showed no dependence on albumin concentration (Figure 4G), VEGF-A signaling in the presence of bispecific anti-VEGF-A / anti-HSA Fab was dependent on the HSA concentration in the culture medium (Figures 4A to 4E). Note that data obtained at 600 µM HSA concentration are not depicted because high concentrations of HSA strongly affect assay performance, resulting in a very low signal-to-background ratio. Although the trend toward reduced HSA-dependent VEGF-binding inhibition was consistent, normalization and dose-response fitting were considered invalid due to the limited assay window and were therefore omitted.

[0265] The IC50 values ​​obtained from the fitted data are shown in Table 5. Comparing the data with the biochemical assay results (Example 4), we found qualitative consistency between the two datasets, demonstrating that the albumin dependence of VEGF-A binding is also relevant in assays expected to reflect biological activity.

[0266] Table 5. VEGF-A reporter gene assay. For the bispecific anti-VEGF-A / anti-HSA Fab, VEGF-A specific control, and albumin-binding control shown, inhibition of VEGF-A-induced NFAT signaling was albumin concentration-dependent. Values ​​are provided in nanomoles. na: Not applicable / Unable to fit inhibition curve.

[0267] Example 6:

[0268] This includes the generation of more bispecific Fab fragments from the serum albumin-binding complementary site of antibody P1AJ5849.

[0269] To demonstrate that the albumin-binding complementary site of the Fab provided in this application can be functionally combined with other complementary sites to generate bispecific antibodies that bind to albumin and a second target, serum albumin-binding complementary sites derived from Fabs with the highest binding affinity (i.e., P1AJ5849 and P1AJ0293) are used to provide further bispecific Fab fragments including complementary sites specific to other target antigens.

[0270] In the first set of experiments, the albumin-binding complementary site of P1AJ5849 was combined with the binding complementary sites of other Fabs generated based on DutaFab technology. Each complementary site is targeted at IL-1β, CD32A, CD32B, CD79b, 41BB, IFNα, and Ox40, and was obtained through a process similar to that described in Example 1.

[0271] The amino acid sequences of the derived bispecific Fab are shown in Table 6:

[0272] Table 6. Amino acid sequences of bispecific Fab clones binding to serum albumin and the indicated second target.

[0273]

[0274] DNA encoding the amino acid sequences of each of the novel bispecific antibodies was generated through gene synthesis and integrated into a suitable expression vector, expressed in *E. coli*, and characterized in SPR using methods similar to those described in Examples 2 (Target Binding) and 3 (Determination of Mutually Repulsive Binding). The following commercially available targets were used: HSA (Merck KGaA, A1887), IL1b (Peprotech, 200-01B), CD32A (R&D Systems, 1330-CD-050 / CF), CD32B (R&D Systems, 1875-CD), and 41BB (R&D Systems, 9220-4B). CD79b, IFNA, and Ox40 were produced in-house using standard methods and a mammalian expression host. For the target binding method, the target is directly immobilized on the surface at a concentration of 3 µg / ml (except for HSA, here: 5 µg / ml), and Fab flows at concentrations of 0, 0.5, 5.0, and 50 nM (for HSA: 0, 0.25, 2.5, and 25 nM). Experiments to determine mutual repulsion binding are performed on the same chip, using Fab (100 nM) or buffer for the first injection, or using a second target (1000 nM) or buffer for the second injection. Because some bispecific substances have weak affinity for albumin or the second target, mutual repulsion binding is not only set up in the manner described in Example 3 (i.e., using the second target to immobilize Fab in the second step of the assay), but also achieved by using human serum albumin coated on an SPR chip, followed by detection of mutual repulsion binding with the second target as a soluble analyte.

[0275] The results of the target binding analysis performed by the SPR assay are shown in Figure 5. We observed that in all cases, the bispecific Fab was generated by the combination of serum albumin alone and the complementary site of the second target. Figure 6 shows the results of the mutually repulsive binding of the bispecific Fab and its respective target. We did not observe a second binding event in any case, thus demonstrating that within the sensitivity limits of the assay, all bispecific antibodies bind albumin and their second target in a mutually repulsive manner.

[0276] In the second set of experiments, the albumin-binding complementary site of P1AJ0293 was combined with the binding complementary sites of other Fabs generated based on DutaFab technology. The respective complementary sites were targeted at 41BB and 0x40.

[0277] The amino acid sequences of the derived bispecific Fab are shown in Table 7:

[0278] Table 7. Amino acid sequences of bispecific Fab clones binding to serum albumin and the indicated second target.

[0279]

[0280] The target binding and mutual repulsion binding modes were determined as described above. The results of the target binding analysis performed by SPR assay are shown in Figure 7. Clearly, the bispecific Fab is able to bind to each of its intended targets. Figure 8 The results show the mutually repulsive binding of the bispecific Fab to its respective target. The data demonstrate that the bispecific Fab binds to its target in a mutually repulsive manner.

[0281] Therefore, we have demonstrated that the HSA binding complementary site can be combined with a large number of other complementary sites to generate bispecific Fab that binds to serum albumin and a second-selective target.

[0282] Example 7:

[0283] The HSA concentration-dependent inhibition of IL1β by the bispecific anti-HSA / anti-IL1β antibody P1AL6714 was characterized by ELISA.

[0284] To provide another example of target binding dependent on solution albumin concentration, we utilized the aforementioned anti-HSA / anti-IL1β bispecific Fab P1AL6714. To characterize how different albumin concentrations affected the ability of this bispecific antibody to bind IL1β, a series of competitive ELISA assays were performed. Briefly, a series of concentrations of Fab were incubated with a constant concentration of IL1β, and unbound IL1β was quantified in the ELISA setup described in more detail below. To determine the dependence of IL1β binding on solution albumin concentration, experiments were performed either in the absence of albumin or by including a constant concentration of albumin in the mixture, varying the constant albumin concentration in parallel experiments to cover concentrations of 0.3 µM, 3 µM, 30 µM, 300 µM, and 600 µM. The VEGF-A / IL1β conjugate P1AL9447, which does not bind albumin and is therefore not expected to exhibit any activity dependence on albumin concentration, was used as a control. Importantly, the entire assay setup required the absence of albumin from any species as a blocking agent. The measurements and results are described in more detail below.

[0285] The ELISA procedure is very similar to that described in Example 4, except that IL1β and its receptor are used instead of VEGF-A and its receptor: specifically, IL1R1-Fc (10126-H02H.Sino Biological) is used as the receptor for immobilizing the anti-Fc antibody, and recombinant human IL1b (200-01B, Peprotech) is added to the previously prepared mixture of Fab and HSA solutions at a final concentration of 720 pM. Biotinylated anti-IL1b (BAF201, R&D Systems) is used together with SA-HRP to detect IL1β binding to the IL1R1 receptor.

[0286] The results are shown in Figure 9 and Table 8.

[0287] We observed that the IC50 value of P1AL6714 was significantly dependent on serum albumin concentration, while the IC50 of the control P1AL9447 was almost the same in the absence of albumin and in the presence of 600 µM albumin.

[0288] Therefore, we provide another example of a conjugate designed to bind albumin and a second target in a mutually exclusive manner, demonstrating target binding activity that depends on the concentration of serum albumin in solution.

[0289] Table 8. Inhibition of IL1β receptor binding in anti-HSA / anti-IL1β P1AL6714 and control was dependent on albumin concentration. Values ​​are provided in nanomoles.

[0290]

Claims

1. A multispecific antibody, wherein the antibody binds to at least a first antigen and a second antigen. The average concentration of the first antigen in the first human body fluid is at least one order of magnitude higher than the average concentration of the first antigen in the second human body fluid. The apparent affinity of the antibody for the second antigen is higher when the first antigen is present in the average concentration of the first antigen in the first human body fluid compared to the affinity of the antibody for the second antigen when the first antigen is present in the average concentration of the first antigen in the second human body fluid.

2. The multispecific antibody according to claim 1, wherein the first human fluid is human peripheral blood, and wherein the second human fluid is human cerebrospinal fluid.

3. The antibody according to any one of the preceding claims, wherein the first antigen is selected from HSA, IgG and transferrin.

4. A multispecific antibody that binds to human serum albumin (HSA) and a second antigen, wherein the antibody has a higher apparent affinity for the second antigen in the presence of an average concentration of HSA in human cerebrospinal fluid than the antibody has an affinity for the second antigen in the presence of an average concentration of HSA in human peripheral blood.

5. The multispecific antibody according to any one of the preceding claims, wherein the second antigen is a therapeutic antigen.

6. The multispecific antibody according to any one of the preceding claims, wherein the second antigen is selected from human VEGF-A, IL-1β, CD32A, CD32B, CD79b, 41BB, IFNα, and Ox40.

7. The multispecific antibody according to any one of the preceding claims, wherein the antibody comprises a bispecific Fab fragment.

8. The multispecific antibody according to any one of the preceding claims, wherein the antibody is or comprises a bispecific Fab fragment, wherein the bispecific Fab fragment comprises a first complementary site capable of binding to the first antigen and a second complementary site capable of binding to the second antigen, wherein the first complementary site and the second complementary site are formed, one of the complementary sites comprises amino acid residues from H-CDR2, L-CDR1 and L-CDR3, and the other complementary site comprises amino acid residues from H-CDR1, H-CDR3 and L-CDR2.

9. The multispecific antibody of claim 5, wherein the antibody comprises a bicomplementary Fv fragment having two overlapping complementary sites.

10. The multispecific antibody according to any one of the preceding claims, wherein the first antigen is HSA, and wherein the second antigen is selected from human VEGF-A, IL-1β, CD32A, CD32B, CD79b, 41BB, IFNα, and Ox40.

11. The multispecific antibody according to any one of the preceding claims, wherein the antibody comprises a) The VH domain of SEQ ID NO:01 and the VL domain of SEQ ID NO:02, b) The VH domain of SEQ ID NO:03 and the VL domain of SEQ ID NO:04, c) The VH domain of SEQ ID NO:05 and the VL domain of SEQ ID NO:06, d) The VH domain of SEQ ID NO:07 and the VL domain of SEQ ID NO:08, e) The VH domain of SEQ ID NO:09 and the VL domain of SEQ ID NO:10, f) The VH domain of SEQ ID NO:19 and the VL domain of SEQ ID NO:20, g) The VH domain of SEQ ID NO:21 and the VL domain of SEQ ID NO:22, h) The VH domain of SEQ ID NO:23 and the VL domain of SEQ ID NO:24, i) The VH domain of SEQ ID NO:25 and the VL domain of SEQ ID NO:26, j) The VH domain of SEQ ID NO:27 and the VL domain of SEQ ID NO:28, k) The VH domain of SEQ ID NO:29 and the VL domain of SEQ ID NO:30, l) The VH domain of SEQ ID NO:31 and the VL domain of SEQ ID NO:32, m) The VH domain of SEQ ID NO:33 and the VL domain of SEQ ID NO:34, or n) The VH domain of SEQ ID NO:35 and the VL domain of SEQ ID NO:

36.

12. A nucleic acid encoding a multispecific antibody according to any one of the preceding claims.

13. A vector comprising the nucleic acid according to claim 12.

14. A host cell comprising the nucleic acid according to claim 12.

15. A method for producing a multispecific antibody according to any one of claims 1 to 12, the method comprising culturing a host cell according to claim 14 under conditions suitable for antibody expression.

16. A pharmaceutical composition comprising an antibody according to any one of claims 1 to 12 and a pharmaceutically acceptable carrier.

Images