Humanized antibodies against PSMA

Humanized and deimmunized antibodies with modified VH and VL regions provide high affinity and low immunogenicity, addressing the limitations of existing anti-PSMA antibodies for effective treatment of prostate cancer and neurological diseases.

JP7586399B2Active Publication Date: 2024-11-19HEIDELBERG PHARMA RES GMBH
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
JP2021505326
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2018-07-31
Filing Date
2019-07-30
Publication Date
2024-11-19
Estimated Expiration
2039-07-30

AI Technical Summary

Technical Problem

Existing anti-PSMA antibodies are unsuitable for human therapy due to immunogenic potential, lack of target recognition, or insufficient binding affinity, limiting their effectiveness in treating prostate cancer and neurological diseases.

Method used

Development of humanized and deimmunized antibodies, antibody fragments, or derivatives that specifically bind to the extracellular domain of PSMA with high affinity, using modified VH and VL-based antigen binding regions to minimize immunogenicity and maintain target recognition.

Benefits of technology

The humanized antibodies exhibit improved binding characteristics and expression yields, reducing production costs and enhancing therapeutic efficacy in targeting prostate cancer and neurological disorders.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007586399000017
    Figure 0007586399000017
  • Figure 0007586399000018
    Figure 0007586399000018
  • Figure 0007586399000019
    Figure 0007586399000019
Patent Text Reader

Abstract

This application relates to humanized and / or deimmunized antibodies, antibody fragments, or antibody derivatives that bind to prostate-specific membrane antigen (PSMA), and methods for using said antibodies, antibody fragments, or antibody derivatives in the treatment of prostate cancer and other neoplastic and neurological diseases (Figure 1).
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present application relates to humanized and / or deimmunized antibodies, antibody fragments or antibody derivatives that bind to prostate-specific membrane antigen (PSMA) and methods for using said antibodies, antibody fragments or antibody derivatives to treat prostate cancer and other neoplastic and neurological diseases. [Background technology]

[0002] Prostate-specific membrane antigen (PSMA), also known as glutamate carboxypeptidase II (GCPII), N-acetyl-L-aspartyl-L-glutamic peptidase I (NAALADase I) or N-acetyl-aspartylglutamic acid (NAAG) peptidase, is an enzyme encoded by the human folate hydrolase (FOLH1) gene. PSMA is a membrane-bound cell surface peptidase with distinct physiological roles and is expressed in various tissues, including the prostate, kidney, small intestine, and central and peripheral nervous systems. It is highly expressed in malignant prostate epithelial cells and vascular endothelial cells in a number of solid tumor malignancies, including glioblastoma, breast cancer, and bladder cancer. The protein has also been implicated in various neurological disorders, including schizophrenia and ALS.

[0003] PSMA is a class II transmembrane glycoprotein with a short N-terminal cytoplasmic tail of 18 amino acids (aa 1-18), a single transmembrane helix (aa 19-43), and an extracellular portion consisting of aa 44-750 with a molecular weight of approximately 84 kDa (Barinka et al., 2004). The short N-terminal cytoplasmic tail has been shown to interact with membrane scaffolding proteins that govern the endocytosis of some of the PSMA-bound substrates, such as clathrin, clathrin adaptor protein 2, filamin A (FLNa), and caveolin-1. The extracellular portion is further divided into three domains: the protease (aa 57-116 and aa 352-590), the apical portion (aa 117-351), and the C-terminal or dimerization domain (aa 591-750). These moieties collectively perform the substrate / ligand recognition function (Barinka et al., 2012) (Figure 1). With regard to the aa sequence and domain organization, the 3D extracellular structure of PSMA resembles the transferrin receptor (Tfr). PSMA is enzymatically active only upon dimerization. Dimerization as well as glycosylation are essential for the correct conformation and enzymatic activity of the molecule.

[0004] In normal human prostate, PSMA expression and localization is associated with the cytoplasm and apical side of the epithelium surrounding the prostatic ducts. Dysplastic and neoplastic transformation of prostate tissue results in translocation of PSMA from the apical membrane to the luminal surface. Although PSMA was originally thought to be prostate specific, more recent studies have shown that it is expressed in small intestinal epithelial (brush border) cells, proximal tubule cells, and salivary glands, as well as in the nervous system.

[0005] In the small intestine, PSMA is located at the brush border of the proximal jejunum where it acts as a hydrolase on poly-γ-glutamate folates and actively transports monoglutamylated folates into the bloodstream (Navratil et al., 2014). Thus, it exhibits both hydrolytic and endocytic functions, at least in humans and pigs. Endocytic internalization occurs via clathrin-coated pits by receptor-mediated endocytosis. In the nervous system, PSMA, also called NAAG-hydrolase, catalyzes the hydrolysis of NAAG, a neurotransmitter widely distributed in the mammalian brain, into glutamate and N-acetylaspartate (NAA). The distinct roles of PSMA in different tissues have triggered extensive exploration of various therapeutic approaches that specifically target the delivery of drugs and small molecules to PSMA-expressing cells (Evans et al., 2016).

[0006] PSMA expression levels are higher in malignant tissues of different origins compared to normal healthy tissues (Evans et al., 2016; Haffner et al., 2009). This expression pattern directly implicates a role for PSMA in cancer progression and invasion, making this molecule a desirable target for the diagnosis and treatment of solid tumors.

[0007] PSMA is an ideal target in prostate cancer for several reasons. First, it is significantly overexpressed in almost all prostate cancer cells. The level of expression on prostate cancer cells is 100-1000-fold higher than in normal tissue. PSMA is expressed in nearly all prostate cancers, and only 5-10% of primary prostate cancers or prostate cancer lesions are PSMA negative by PET (Eiber et al., 2017).

[0008] Second, its expression is further increased in advanced stage, high-grade metastatic disease, hormone refractory prostate cancer and metastatic castration-resistant prostate cancer (mCRPC).

[0009] Third, in contrast to other prostate-associated antigens such as PSA, prostatic acid phosphatase, and prostate secretory protein, PSMA, as a type II integral membrane cell surface protein, is not secreted and is therefore an excellent target for, for example, antibody or other ligand-mediated therapy. After binding to the active center of the extracellular domain, the PSMA ligand is internalized into the cell. Subsequent endosomal recycling increases deposition, enhancing tumor uptake, retention, and high local doses for subsequent therapeutic applications.

[0010] In general, prostate cancer metastases frequently occur to the bone marrow and lymph nodes, locations that take up high levels of circulating antibodies and have responded well to monoclonal antibody therapy in other tumor types, such as lymphoma and breast cancer ( Nanus et al., 2003 ).

[0011] In prostate cancer, PSMA expression is negatively regulated by androgens (Israel et al., 1993). Gene transcription analysis demonstrated that androgens can repress the promoter of the PSMA gene. Thus, initiation of androgen deprivation therapy induces an early but transient upregulation of PSMA expression, downregulation under extended androgen deprivation therapy, and finally overexpression of PSMA in androgen-resistant tumors. These effects may be exploited to improve treatment and diagnosis (Eiber et al., 2017). Although the direct role of PSMA in prostate cancer metastasis has not yet been fully elucidated, hormonal androgen deprivation therapy for androgen-sensitive cells has been shown to increase PSMA levels, and increased expression of PSMA made the cells less invasive. Meanwhile, knocking down PSMA increased invasive activity five-fold.

[0012] Prostate cancer is the most commonly diagnosed cancer in men and the second leading cause of death in Western cultures. Due to the significant mortality and morbidity associated with disease progression, new targeted therapies are urgently needed. There are currently various approaches being evaluated for the treatment of prostate cancer. Small molecules, aptamers or antibodies have been used as specific ligands to exploit PSMA for targeted delivery of chemotherapeutic drugs (for review see Evans et al., 2016). PSMA-targeted gene therapy approaches based on, for example, RNAi, antisense RNA or aptamers are currently under evaluation, as are PSMA-mediated suicide gene therapies. In radioimmunotherapy, currently used in approximately 25% of patients with localized disease of prostate cancer (Cooperberg et al., 2010), the addition of targeting ligands such as monoclonal antibodies or aptamers to the radionuclides allows the targeting of cancer-associated cell surface antigens such as PSMA, thus reducing undesirable side effects and dose enhancement. Immunotherapies based on primary dendritic cells (DCs) or T cells bearing chimeric antigen receptors (CARs) against PSMA to induce CTL activity against PSMA-positive tumor cells are in the clinical stage.

[0013] Antibodies that bind to PSMA have been described in the prior art. PSMA was first characterized by the mouse monoclonal antibody 7E11, which was derived from mice immunized with partially purified cell membrane fractions isolated from human prostate adenocarcinoma (LNCap) cell line (Horoszewicz et al., 1986). This first published monoclonal antibody against PSMA recognizes the N-terminus (MWNLLH) of the intracellular domain of PSMA (Troyer et al., 1995).

[0014] Four mouse IgG monoclonal antibodies (J591, J415, J533, and E99) against the extracellular domain of PSMA have been prepared which, in contrast to 7E11, recognize two distinct, non-competing epitopes located on the outside of the cell ( Liu et al., 1997 ; Nanus et al., 2003 ).

[0015] Wolf et al. (2010a) generated three murine anti-PSMA monoclonal antibodies, 3 / A12, 3 / E7, and 3 / F11, which showed strong binding to the PSMA-positive prostate cancer LNCaP subcell line C4-2, with mean half-maximal saturation concentrations ranging from 9 to 17 nM (EP 1883698 B2). The antibodies stained epithelial cells in all normal and neoplastic prostate tissue specimens tested. As revealed by competitive binding studies, the three monoclonal antibodies bind to different extracellular PSMA epitopes. The epitope of 3 / A12 was found to be close to or partially overlapping with that of antibody J591.

[0016] An anti-PSMA single-chain antibody fragment (scFv), designated D7, isolated by phage display from monoclonal antibody 3 / F11, was used to construct an immunotoxin by C-terminal ligation to the D7 cDNA of Pseudomonas exotoxin A. This immunotoxin inhibited PSMA-positive prostate cancer growth in vitro and in vivo (Wolf et al., 2010b).

[0017] In contrast to the results obtained with monoclonal antibodies 3 / A12, 3 / E7, and 3 / F11, derived from mice immunized with crude LNCaP cell lysates, the same researchers obtained three monoclonal antibodies, K7, K12, and D20, from mice immunized with purified PSMA, which showed high reactivity only to isolated antigens but weak or no reaction with viable LNCaP cells (Elsasser-Beile et al., 2006).

[0018] Tykvart et al. (2014) characterized 13 PSMA-specific murine monoclonal antibodies in terms of their binding activity to native and denatured antigens.

[0019] WO2001009192 (Medarex) describes the development of human monoclonal antibodies against prostate specific membrane antigen. Human anti-PSMA monoclonal antibodies were generated by immunizing mice with purified PSMA or concentrated preparations of PSMA antigen. Such purified antigens are denatured PSMA, since they are purified by immunoadsorption after cell lysis with ionic detergents.

[0020] WO1997035616 (Pacific Northwest Cancer Foundation) describes a monoclonal antibody specific for the extracellular domain of prostate-specific membrane antigen. Immunization was performed with a C-terminal peptide or a PSMA-expressing tumor membrane specimen. The resulting monoclonal antibody did not specifically bind to PSMA-expressing cells and therefore cannot be used for diagnostic or therapeutic purposes.

[0021] Bander et al. (2003) disclosed monoclonal antibodies against prostate specific membrane antigen. Since immunization was performed with purified antigen, the monoclonal antibodies were not suitable for cell binding, and scFvs were not obtained from any of these monoclonal antibodies. WO1998 / 03873 describes the same antibody or a binding portion thereof that recognizes the extracellular domain of prostate specific membrane antigen, but could not show that the binding portion of the antibody actually binds to the antigen.

[0022] Immunogenicity is a crucial concern when developing antibody drugs (Almagro and Fransson, 2008). All exogenous proteins used for therapeutic purposes pose the risk of triggering anti-drug antibody production in the recipient (Schellekens 2002), and the use of therapeutic proteins and antibodies of non-human origin has been found to be usually associated with the generation of anti-drug antibodies, often leading to the generation of harmful immune complexes and / or neutralization of the therapeutic drug. In addition to the origin and nature of the antibody, other factors have been shown to affect immunogenicity, such as the type of disease, the route of administration, and the genetic background of the recipient.

[0023] Thus, for various reasons, the anti-PSMA antibodies described above from the prior art are not suitable for human therapy, in particular due to their immunogenic potential, lack of target recognition or insufficient binding affinity.

[0024] (Summary of the invention) In view of the prior art, one object of the present invention was therefore to provide a humanized antibody, antibody fragment or antibody derivative, as described in the present application, that binds to prostate-specific membrane antigen (PSMA) with high affinity, in particular at an epitope in the extracellular domain of PSMA.

[0025] The antibodies, antibody fragments or antibody derivatives disclosed herein contain preferred VH and VL-based humanized sequences of antigen-binding regions that maintain appropriate ligand affinity. Amino acid sequence modifications to obtain said humanized sequences may occur in the CDR regions and / or in the framework regions of the original antibody and / or in the antibody constant region sequences.

[0026] A further object of the present invention was to provide antibodies that specifically bind to native cell surface PSMA and thus have value in diagnostic and therapeutic applications focused on PSMA as a target antigen for prostate cancer.

[0027] A further object of the present invention was to provide compounds for use in a method for the treatment of prostate cancer or neurological disorders.

[0028] A further object of the present invention was to provide compounds that destroy prostate cancer cells that express PSMA.

[0029] Another object of the present invention was to provide an antibody, antibody fragment or antibody derivative thereof that is low or non-immunogenic in humans. Preferably, the antibody is a monoclonal antibody, antibody fragment or antibody derivative thereof.

[0030] Another object of the present invention was to provide antibody-drug conjugates (ADCs) for use in methods of treating prostate cancer and other diseases, preferably comprising amatoxins.

[0031] It was quite surprising that the humanized sequences according to the invention, in particular the CDR sequences of the VH and VL regions involved in binding to the target molecule, show specific and strong binding as demonstrated in the examples, retaining or even exceeding the binding characteristics of the original murine antibody 3F11 without any affinity maturation usually required after humanization of an antibody. A person skilled in the art would not have expected that the binding properties of the humanized variants would be similar, let alone better than those of the original murine antibody. Considering the sequence changes in the variable domains, in particular in the CDRs, the beneficial binding properties of the humanized sequences presented herein must be considered as a surprising technical effect.

[0032] Moreover, it was completely surprising that the humanized sequences according to the invention gave significantly better expression yields than the parent antibody sequences when expressed in eukaryotic host cells, thus making the biotechnological production of a humanized antibody according to the invention, or an antibody fragment or antibody derivative thereof, much cheaper than using the parent antibody sequences.

[0033] These and other objects are achieved by the methods and means according to the independent claims. The dependent claims relate to specific embodiments.

[0034] The general advantages of the invention and its features are explained in detail below. [Brief description of the drawings]

[0035] [Figure 1] Schematic diagram of the PSMA transmembrane protein as a homodimer (from Evans et al., 2016). [Diagram 2]Schematic diagram of an antibody. IgG structure showing two identical heavy chains (H chains) and two identical light chains (L chains). The H chain has four domains: three C domains (CH1-CH3) and one VH. The L chain has one CL domain and one VL domain. The Fv fragment (shown as a box) is the part of the molecule that interacts with the antigen. It is formed by the non-covalent pairing of VL and VH. The Fv fragment is shown in view of the antigen showing the arrangement of the hypervariable loops (from Almagro and Fransson, 2008). [Diagram 3] Sequence alignment of the variable heavy (VH) region of humanized / deimmunized variants of antibody 3F11. Amino acid substitutions relative to murine 3F11-wt are in bold. [Figure 4] Sequence alignment of the variable light (VL) region of humanized / deimmunized variants of antibody 3F11. Amino acid substitutions relative to murine 3F11-wt are in bold. [Diagram 5] SDS-PAGE and Coomassie staining for the analysis of humanized 3F11 antibody variants 13-16 under reducing conditions. 1 μg of samples and controls were loaded per well on the gel. [Figure 6] Binding dose-response curves of humanized 3F11 antibody variants. Symbols represent the mean ± SEM of replicates (n=2 for humanized samples, n=8 for chimeric 3F11-wt antibody) of normalized binding data for one experiment. MFI vs. log dose transformed data were normalized using maximum and minimum MFI values. R2 > 0.99 in all cases. [Figure 7] EC50 binding for humanized 3F11 antibody variants. Symbols represent the mean EC50 and bars show the 95% confidence interval (CI). Black dotted lines show the 95% CI for 3F11-wt antibody (n=8) and red dotted lines show the 95% CI for 3F11-var7 (n=1) from one experiment. The variants with the highest binding activity (3F11-var7) and the wt antibody are highlighted with filled symbols. Detailed Description of the Invention

[0036] Before describing the invention in detail, it is to be understood that the invention is not limited to the particular component parts of the devices described, or to the process steps of the methods described, as such devices and methods may vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting. It is to be noted that, as used in the specification and the appended claims, the singular forms "a," "an," and "the" include singular and / or plural references unless the context clearly dictates otherwise. Furthermore, when a parameter range is given that is bounded by numerical values, it is to be understood that the range is deemed to be inclusive of these limits.

[0037] Furthermore, it should be understood that the embodiments disclosed herein are not meant to be understood as separate embodiments unrelated to each other. Features discussed in one embodiment are meant to be disclosed in relation to other embodiments shown herein. In one case, if a particular feature is not disclosed in one embodiment but is disclosed in another embodiment, a person skilled in the art will understand that this does not necessarily mean that the feature is not meant to be disclosed in the other embodiment. A person skilled in the art will understand that it is the intent of the present application to disclose the features in other embodiments as well, but this has not been done merely for the sake of clarity and to keep the present specification to a manageable volume.

[0038] Furthermore, the contents of the prior art documents referred to in this specification are incorporated by reference. This refers in particular to prior art documents that disclose standard or conventional methods. In that case, incorporation by reference has the primary purpose of providing a full enabling disclosure and avoiding lengthy repetition.

[0039] As used herein, the term "antibody" refers to a protein consisting of one or more polypeptide chains encoded by immunoglobulin genes or fragments of immunoglobulin genes, or cDNA derived therefrom, including the light chain kappa, lambda, and heavy chain alpha, delta, epsilon, gamma, and mu constant region genes, as well as any of a number of different variable region genes.

[0040] The basic immunoglobulin (antibody) structural unit is usually a tetramer consisting of two identical polypeptide chain pairs, a light chain (L, molecular weight approximately 25 kDa) and a heavy chain (H, molecular weight approximately 50 to 70 kDa). Each heavy chain contains a heavy chain variable region (VH or V H ) and the heavy chain constant region (CH or C H The heavy chain constant region consists of three domains, namely CH1, CH2 and CH3. Each light chain contains a light chain variable region (VL or V L ) and the light chain constant region (CL or C L The VH and VL regions contain regions of hypervariability, also called complementarity determining regions (CDRs), interspersed between more conserved regions called framework regions (FRs). Each VH and VL region is composed of three CDRs and four FRs arranged from the amino terminus to the carboxy terminus in the following order: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. The variable regions of the heavy and light chains form the binding domain that interacts with the antigen.

[0041] CDRs are the most important for the binding of an antibody or its antigen-binding portion. FRs can be replaced with other sequences as long as the three-dimensional structure required for antigen binding is maintained. Structural changes in the construct most often lead to loss of sufficient binding to the antigen.

[0042] The term "antigen-binding portion" of a (monoclonal) antibody refers to one or more fragments of an antibody that retains the ability to specifically bind to prostate-specific membrane antigen in its native form. Examples of antigen-binding portions of an antibody include a Fab fragment, a monovalent fragment consisting of the VL, VH, CL and CH1 domains, a F(ab')2 fragment, a bivalent fragment containing two Fab fragments linked by a disulfide bridge at the hinge region, an Fd fragment consisting of the VH and CH1 domains, an Fv fragment consisting of the VL and VH domains of a single arm of an antibody, and a dAb fragment consisting of the VH domain and isolated complementarity determining regions (CDRs).

[0043] As used herein, the term "monoclonal antibody (mAb)" refers to an antibody composition having a homogenous antibody population, i.e., a homogenous population of whole immunoglobulins or fragments or derivatives thereof. Particularly preferred such antibodies are selected from the group consisting of IgG, IgD, IgE, IgA and / or IgM, or fragments or derivatives thereof.

[0044] The term "fragment" as used herein refers to fragments of such antibodies that retain target binding ability, such as the CDRs (complementarity determining regions), hypervariable regions, variable domains (Fv), IgG heavy chains (consisting of the VH, CH1, hinge, CH2 and CH3 regions), IgG light chains (consisting of the VL and CL regions), and / or Fab and / or F(ab)2.

[0045] The term "derivative" as used herein refers to protein constructs that are structurally different but still have some structural relatedness, such as scFv, Fab and / or F(ab)2, as well as bispecific, trispecific or higher specificity antibody constructs, all of which are explained below.

[0046] Other antibody derivatives known to those skilled in the art are diabodies, camelid antibodies, domain antibodies, bivalent homodimers with two chains consisting of scFv, IgA (two IgG structures linked by a J chain and a secretory component), shark antibodies, antibodies consisting of a New World primate framework plus non-New World primate CDRs, dimeric constructs containing CH3+VL+VH, other scaffold protein formats containing CDRs, and antibody conjugates (e.g. antibodies linked to drugs, toxins, cytokines, aptamers, nucleic acids such as desoxyribonucleic acid (DNA) or ribonucleic acid (RNA), therapeutic polypeptides, radioisotopes or labels, or fragments or derivatives thereof). Said scaffold protein formats may include, for example, ankyrin and affilin proteins, etc.

[0047] As used herein, the term "Fab" refers to an IgG fragment containing the antigen-binding region, said fragment being composed of one constant and one variable domain from each of the heavy and light chains of the antibody.

[0048] As used herein, the term "F(ab)2" refers to an IgG fragment consisting of two Fab fragments bound together by disulfide bonds.

[0049] As used herein, the term "scFv" refers to a single chain variable fragment that is a fusion of the variable regions of the heavy and light chains of an immunoglobulin linked together with a short linker that usually contains serine (S) and / or glycine (G) residues. This chimeric molecule retains the specificity of the original immunoglobulin despite the removal of the constant regions and the introduction of a linker peptide.

[0050] The modified antibody formats are, for example, bi- or tri-specific antibody constructs, antibody-based fusion proteins, immunoconjugates, and the like.

[0051] IgG, scFv, Fab and / or F(ab)2 are antibody formats well known to those skilled in the art. The relevant validation techniques are available from the respective textbooks.

[0052] Monoclonal antibodies (mAbs) from mice contain proteins from other species that may induce antibodies and therefore may cause undesirable immunological side effects. To overcome this problem, antibody humanization and maturation methods have been designed to generate antibody molecules that have minimal immunogenicity when applied to humans, while ideally still retaining the specificity and affinity of the non-human parent antibody (for a review see Almagro and Fransson 2008). These methods are used, for example, to replace mouse MAb framework regions with corresponding human framework regions (so-called CDR grafting). WO200907861 discloses the creation of humanized forms of mouse antibodies by linking the CDR regions of a non-human antibody to human constant regions by recombinant DNA techniques. US6548640 by the Medical Research Council describes CDR grafting techniques and US5859205 by Celltech describes the production of humanized antibodies.

[0053] However, humanization by CDR grafting often results in a significant reduction or even loss of binding affinity, since a set of supporting framework residues in the so-called Vernier zones are important for maintaining the conformation of the CDRs (Foote and Winters 1992). These residues are responsible for stabilizing the hypervariable loop structures and modifying their position, fine-tuning the antibody affinity. Reintroducing mouse residues into the human framework (Queen et al. 1989) can address this challenge. Such substitutions are commonly referred to as "back-mutations".

[0054] Immunogenicity is the ability to induce helper T (Th) cell responses triggered by the recognition of peptides bound to HLA (human leukocyte antigen) class II molecules presented on antigen-presenting cells (APCs) by specific T cell receptors. Peptides are produced from proteins internalized by antigen-presenting cells and processed via an endosomal cleavage pathway. Only peptides with sufficient affinity for HLA class II molecules can be presented on the cell surface of APCs and potentially induce Th responses.

[0055] A process known as deimmunization can be used to further reduce immunogenicity by removing Th epitopes (Chamberlain 2002; Baker and Jones 2007), which is achieved by predicting which peptides in each therapeutic protein can bind to HLA class II molecules and then introducing substitutions that eliminate or reduce peptide binding affinity for HLA class II molecules.

[0056] There are several HLA class II genes, HLA class II molecules consist of alpha and beta chains, each expressed from a different gene, with increasing mutations (DRA / DRB, DQA / DQB, and DPA / DPB), and almost all are highly polymorphic; only DRA is nonpolymorphic. The focus during deimmunization is on the DR allotypes, known to be expressed at higher levels than DQ and DP. The assessment of the relevance of individual epitopes is based on a criterion of promiscuity, i.e., the number of HLA allotypes to which a particular epitope binds, as well as a qualitative assessment of the importance (frequency of occurrence) of the allotype in the population and the strength of HLA-peptide complex binding. Since individual T cell populations have been selected not to recognize "self-peptides" of endogenous proteins like antibodies, proteins to be deimmunized can be screened against peptides corresponding to (known) self-peptides that should not normally induce a Th response.

[0057] The term "humanized antibody" as used herein refers to an antibody, fragment or derivative thereof in which at least a portion of the antibody constant and / or framework regions, and optionally some of the CDR regions, are derived from or aligned to human immunoglobulin sequences.

[0058] Sequence variants of the claimed nucleic acid or amino acid sequences, defined by the claimed % sequence identity (homology) and maintaining the above-mentioned properties of the antibodies, fragments or antibody derivatives according to the invention, are also included within the scope of the present invention. The term "sequence identity" relates to the percentage of identical nucleotides or amino acids when performing sequence alignment.

[0059] As used herein, the term "amino acid substitution" refers to the modification of the amino acid sequence of a protein, where one or more amino acids are replaced with the same number of different amino acids, producing a protein that contains a different amino acid sequence from that of the original protein.Conservative amino acid substitution is understood to refer to a substitution that does not significantly affect the structure and function of protein due to similar size, charge, polarity and / or conformation.Conservative amino acid groups in this sense include, for example, non-polar amino acids Gly, Ala, Val, Ile and Leu; aromatic amino acids Phe, Trp and Tyr; positively charged amino acids Lys, Arg and His; and negatively charged amino acids Asp and Glu.

[0060] The present invention relates to a humanized antibody or antibody fragment or derivative thereof that binds to the extracellular domain of prostate specific membrane antigen (PSMA).

[0061] Preferred CDR sequences are discussed below. It is important to understand that the CDRs are constructed from a suitable protein framework, such as the extracellular domain of prostate specific membrane antigen (PSMA).

[0062] The present invention also provides -GX1TX2TX3X4(CDR H1) [wherein X1, X2, X3 and X4 are Y, F and W, respectively], -GISPGDX5NX6NYX7X8X9FX 10 X5 is G, S, X6 is T, V, X7 is N, A, X8 is E, Q, X9 is N, K, and X 10 are K, Q], and -DGNX 11 PX 12 X 13 AMDS(CDR H3) [where X 11 ,X 12 ,X 13 are Y, F, and W, respectively. The present invention provides a humanized antibody, or an antibody fragment or derivative thereof, that binds to the extracellular domain of PSMA, comprising a VH domain comprising the CDR sequences of:

[0063] The present invention also provides the humanized antibody, or antibody fragment or antibody derivative thereof, wherein the VH domain comprises the CDR H1 sequence GYTFTYF (SEQ ID NO:1).

[0064] The invention further provides said humanized antibody, or antibody fragment or antibody derivative thereof, which binds to the extracellular domain of PSMA, wherein the VH domain comprises the CDR H2 sequence, SEQ ID NO:2 or SEQ ID NO:7 or SEQ ID NO:9 or SEQ ID NO:11.

[0065] The present invention further relates to said humanized antibody or antibody fragment or antibody derivative thereof, which binds to the extracellular domain of PSMA, wherein the VH domain comprises the CDR H3 sequence DGNFPYYAMDS (SEQ ID NO: 3).

[0066] In a preferred embodiment, the antibody, or antibody fragment or antibody derivative thereof, that binds to the extracellular domain of PSMA is SEQ ID NO:1 and SEQ ID NO:2 or SEQ ID NO:7 or SEQ ID NO:9 or SEQ ID NO:11 and SEQ ID NO:3 The VH domain comprises:

[0067] In a further preferred embodiment, the antibody or antibody fragment or antibody derivative thereof that binds to the extracellular domain of PSMA comprises a VH domain comprising any of SEQ ID NOs: 13, 15, 17 or 19.

[0068] The present invention also provides -RSSQSLVHSX1GX2TX3LH(CDR L1) [wherein X1 is N, S; X2 is N, Q; and X3 is Y, F, W], -TVSNRX4S(CDR L2) in which X4 is A, Y, F, W, and -SQSTHVPT(CDR L3) The present invention relates to a humanized antibody, or an antibody fragment or derivative thereof, which comprises a VL domain comprising the CDR sequences of:

[0069] In a preferred embodiment said antibody or antibody fragment or antibody derivative comprises a VL domain comprising the CDR L1 sequence SEQ ID NO:4 or SEQ ID NO:8 or SEQ ID NO:10.

[0070] In another preferred embodiment, said antibody or antibody fragment or derivative thereof comprises a VL domain comprising the CDR L2 sequence SEQ ID NO:5 or SEQ ID NO:12.

[0071] In a further preferred embodiment, the antibody or antibody fragment or antibody derivative thereof comprises SEQ ID NO: 4 or SEQ ID NO: 8 or SEQ ID NO: 10, and SEQ ID NO: 5 or SEQ ID NO: 12, and SEQ ID NO:6 The VL domain comprises a combination of:

[0072] The present invention particularly relates to an antibody or antibody fragment or derivative thereof comprising a VL domain comprising any of SEQ ID NOs: 14, 16, 18 or 20.

[0073] The present invention also relates to an antibody or an antibody fragment or derivative thereof comprising a combination of at least one of the above VH CDR sequences with at least one of the above VL CDR sequences.

[0074] The present invention also relates to said antibody or an antibody fragment or derivative thereof comprising a combination of at least one VH domain sequence selected from the group consisting of SEQ ID NOs: 13, 15, 17 and 19, and at least one VL domain sequence selected from the group consisting of SEQ ID NOs: 14, 16, 18 and 20.

[0075] The present invention also relates to an EC 50 Affinity of <0.4 μg / ml, preferably EC 50 Affinity <0.3 μg / ml, most preferably EC 50 Any such antibody as described above, or an antibody fragment or antibody derivative thereof, which binds to the extracellular domain of PSMA with an affinity of <0.4 μg / ml.

[0076] The humanized anti-PSMA antibody, or antibody fragment or derivative thereof, can be glycosylated. The glycan can be an N-linked oligosaccharide chain at asparagine 297 of the heavy chain.

[0077] The prostate specific membrane antigen may be a mammalian, non-primate, primate, particularly a human prostate specific membrane antigen.

[0078] The antibody or antibody fragment or derivative thereof according to the invention may be a monoclonal antibody. The antibody may be of the IgA, IgD, IgE, IgG or IgM isotype.

[0079] The present invention relates to an antibody fragment selected from the group consisting of a variable domain (Fv), a Fab fragment and a F(ab)2 fragment. The present invention also relates to an antibody derivative, which is preferably a single chain Fv (scFv).

[0080] In a preferred embodiment, the humanized antibody described in this application, or its fragment or derivative, such as scFv, specifically binds to native cell surface PSMA, and thus will be valuable in diagnostic and therapeutic applications, such as focusing on PSMA as a target antigen for prostate cancer.Since PSMA is expressed on prostate cancer cells with specific tertiary and quaternary structures, antibodies against this cellular conformation in particular can recognize and strongly bind to viable prostate cancer cells and PSMA-expressing tissues.Therefore, one of the aims of this study was to generate such antibodies, or their fragments or derivatives, that can be used for therapeutic and diagnostic targeting of prostate cancer.

[0081] In a preferred embodiment, the present invention further relates to an antibody-drug conjugate (ADC, also called an immunoconjugate) comprising a humanized antibody as described above, or a fragment or derivative thereof, optionally linked via a linker moiety to a therapeutic agent.

[0082] The therapeutic agent can be, for example, a drug, a chemotherapeutic agent, a cytotoxic agent, e.g., taxol, cytochalasin B, gramicidin D, a toxin, a growth inhibitory agent, an aptamer, a nucleic acid such as desoxyribonucleic acid (DNA) or ribonucleic acid (RNA), a therapeutic polypeptide, or a radioactive substance.

[0083] The ADC may include a linker region between the therapeutic agent and the anti-PSMA antibody, or antibody fragment or derivative thereof, the linker being cleavable, for example, but not limited to, by a cleaving agent or acidic pH conditions.

[0084] In another aspect of the invention, there is provided a humanized anti-PSMA antibody, or a fragment or derivative thereof, wherein the antibody is formulated in a pharmaceutical formulation.

[0085] In aqueous form, the pharmaceutical formulation may be ready for administration, whereas in lyophilized form, the formulation may be transferred to liquid form by adding water for injection which may or may not contain synthetic preservatives such as, for example, but not limited to, benzyl alcohol, antioxidants such as vitamin A, vitamin E, vitamin C, retinyl palmitate, and selenium, the amino acids cysteine ​​and methionine, citric acid and sodium citrate, and the parabens methylparaben and propylparaben.

[0086] The pharmaceutical formulation may further comprise one or more stabilizers, which may be, for example, amino acids, sugar polyols, disaccharides and / or polysaccharides. The pharmaceutical formulation may further comprise one or more surfactants, one or more isotonicity agents, and / or one or more metal ion chelators, and / or one or more preservatives.

[0087] The pharmaceutical formulations described herein are suitable for at least intravenous, intramuscular or subcutaneous administration. Alternatively, the antibodies according to the invention may be provided in a depot formulation allowing for sustained release of the biologically active agent over a period of time.

[0088] In yet another aspect of the invention, there is provided a primary packaging such as a pre-filled syringe or pen, a vial, or an infusion bag, comprising said formulation according to the previous aspect of the invention and / or said humanized anti-PSMA antibody, or a fragment or derivative thereof.

[0089] The pre-filled syringe or pen may contain the formulation either in lyophilized form (which must then be dissolved, e.g., with water for injection, prior to administration), or in aqueous form. The syringe or pen is often a disposable item for single use only, and may have a volume of 0.1-20 ml. However, the syringe or pen may also be a multi-use or multi-dose syringe or pen.

[0090] The vial may also contain the formulation in lyophilized or aqueous form and may serve as a single or multiple use device. As a multiple use device, the vial may have a larger capacity. The infusion bag usually contains the formulation in aqueous form and may have a capacity of 20-5000 ml.

[0091] The pharmaceutical formulation is intended for the treatment of a medical disorder associated with overexpression of PSMA and comprises a humanized antibody, or an antibody fragment or antibody derivative thereof, as described herein, together with at least a pharma- ceutical acceptable carrier.

[0092] The medical disorder associated with overexpression of PSMA may be, for example, prostate cancer or a neurological disease. The humanized anti-PSMA antibody or a fragment or derivative thereof may mediate an anti-tumor effect by activating humoral or cellular immune function, or by targeting tumors with conjugated cytotoxins or radioactivity as described above.

[0093] The present invention relates to an antibody, or an antibody fragment or derivative thereof, as described herein for use in the treatment of prostate cancer.

[0094] The present invention also relates to a method of treating a patient suffering from prostate cancer comprising administering an effective amount of an antibody, or an antibody fragment or derivative thereof, as described herein.

[0095] Another aspect of the invention relates to a diagnostic kit for the detection of tumor cells comprising the isolated monoclonal antibody, or an antigen-binding fragment or derivative thereof, which allows its detection after binding with suitable detection reagents and devices. The invention also provides a method for the in vitro identification of tumor cells, in which the tumor cells to be identified are contacted with the isolated monoclonal antibody, or an antigen-binding fragment or derivative thereof, which carries a label detectable by a suitable analytical device. This label allows the diagnostic identification of tumor cells, for example, in sections of human tissue obtained after surgery or biopsy.

[0096] The present invention also relates to host cells capable of producing said humanized anti-PSMA antibodies, or fragments or derivatives thereof, and to cell lines transfected with recombinant plasmids comprising the coding sequences for said antibodies of the invention.

[0097] According to another preferred embodiment of the present invention, the host cell is a mammalian cell. According to yet another preferred embodiment of the present invention, the host cell is at least one selected from the group including, for example, baby hamster kidney cell (e.g., BHK21), Chinese hamster ovary cell (CHO), mouse myeloma cell (e.g., SP2 / 0 or NS0), human embryonic kidney cell (e.g., HEK-293), human retina-derived cell (e.g., PER-C6) and amniotic cell (e.g., CAP).

[0098] In one embodiment, the mammalian cell is a CHO cell line (e.g., CHO-K1, CHO-DG44, CHO-DXB, or CHO-dhfr - ).

[0099] The present invention also relates to a nucleic acid encoding a humanized antibody, antibody fragment or antibody derivative thereof, that binds to PSMA as described in this application.

[0100] In one embodiment, the present invention provides a method for producing a medicament for the treatment of a cancer, comprising: CDR H1 sequence as set forth in SEQ ID NO: 1, A CDR H2 sequence selected from any of the sequences set forth in SEQ ID NO: 2, SEQ ID NO: 7, SEQ ID NO: 9, and SEQ ID NO: 11, and CDR H3 sequence as set forth in SEQ ID NO:3 and a variable heavy (VH) domain comprising A CDR L1 sequence selected from any of the sequences set forth in SEQ ID NO: 4, SEQ ID NO: 8 and SEQ ID NO: 10; a CDR L2 sequence selected from any of the sequences set forth in SEQ ID NO: 5 and SEQ ID NO: 12, and CDR L3 sequence as set forth in SEQ ID NO:6 a variable light chain (VL) domain comprising does not include a combination of the CDR H1 sequence set forth in SEQ ID NO: 1, the CDR H2 sequence set forth in SEQ ID NO: 2, the CDR H3 sequence set forth in SEQ ID NO: 3, the CDR L1 sequence set forth in SEQ ID NO: 4, the CDR L2 sequence set forth in SEQ ID NO: 5, and the CDR L3 sequence set forth in SEQ ID NO: 6; Binds to the extracellular domain of prostate-specific membrane antigen (PSMA), The present invention relates to a humanized antibody or an antibody fragment or derivative thereof.

[0101] For the avoidance of doubt, all 24 possible combinations of said CDRs (i.e., combinations of one CDR H1, one CDR H2, one CDR H3, one CDR L1, one CDR L2 and one CDR L3) are considered to be individually disclosed herein.

[0102] In one embodiment, the invention relates to a humanized antibody, or an antibody fragment or derivative thereof, comprising a VH domain having at least 90%, preferably 95%, sequence identity to a sequence selected from any of the following sequences: SEQ ID NO:13, SEQ ID NO:15, SEQ ID NO:17, and SEQ ID NO:19.

[0103] In one embodiment, the invention relates to a humanized antibody, or an antibody fragment or derivative thereof, comprising a VH domain sequence selected from any of the sequences set forth in SEQ ID NO:13, SEQ ID NO:15, SEQ ID NO:17, and SEQ ID NO:19.

[0104] In one embodiment, the present invention relates to a humanized antibody, or an antibody fragment or derivative thereof, comprising a VL domain having at least 90%, preferably 95%, sequence identity to a sequence selected from any of the following sequences: SEQ ID NO: 14, SEQ ID NO: 16, SEQ ID NO: 18 and SEQ ID NO: 20.

[0105] In one embodiment, the present invention relates to a humanized antibody, or an antibody fragment or derivative thereof, comprising a VL domain sequence selected from any of the sequences set forth in SEQ ID NO:14, SEQ ID NO:16, SEQ ID NO:18 and SEQ ID NO:20.

[0106] In one embodiment, the invention relates to a humanized antibody or an antibody fragment or antibody derivative thereof comprising a VH domain sequence set forth in SEQ ID NO: 13 and a VL domain sequence selected from any of the sequences set forth in SEQ ID NO: 14, SEQ ID NO: 16, SEQ ID NO: 18 and SEQ ID NO: 20, or a VH domain sequence set forth in SEQ ID NO: 15 and a VL domain sequence selected from any of the sequences set forth in SEQ ID NO: 14, SEQ ID NO: 16, SEQ ID NO: 18 and SEQ ID NO: 20, or a VH domain sequence set forth in SEQ ID NO: 17 and a VL domain sequence selected from any of the sequences set forth in SEQ ID NO: 14, SEQ ID NO: 16, SEQ ID NO: 18 and SEQ ID NO: 20, or a VH domain sequence set forth in SEQ ID NO: 19 and a VL domain sequence selected from any of the sequences set forth in SEQ ID NO: 14, SEQ ID NO: 16, SEQ ID NO: 18 and SEQ ID NO: 20.

[0107] In one embodiment, the invention relates to a humanized antibody or antibody fragment or derivative thereof, wherein the heavy chain comprises the sequence of any of SEQ ID NO:21, SEQ ID NO:23, SEQ ID NO:25, or SEQ ID NO:27.

[0108] In one embodiment, the invention relates to a humanized antibody, or an antibody fragment or derivative thereof, wherein the light chain comprises any of the sequences of SEQ ID NO:22, SEQ ID NO:24, SEQ ID NO:26, or SEQ ID NO:28.

[0109] In one embodiment, the present invention relates to a humanized antibody, or an antibody fragment or derivative thereof, comprising at least one heavy chain comprising the sequence of any of SEQ ID NOs: 21, 23, 25 or 27, and at least one light chain comprising the sequence of any of SEQ ID NOs: 22, 24, 26 or 28.

[0110] In a preferred embodiment, the present invention relates to a humanized antibody, or an antibody fragment or antibody derivative thereof, comprising at least one heavy chain comprising SEQ ID NO:27 and at least one light chain comprising SEQ ID NO:28.

[0111] In one embodiment, the present invention relates to an EC 50 The present invention relates to a humanized antibody, or an antibody fragment or antibody derivative thereof, which binds to the extracellular domain of PSMA with an affinity of <0.4 μg / ml.

[0112] In one embodiment, the present invention relates to a humanized antibody of the present invention, or an antibody fragment or derivative thereof, wherein the antibody is glycosylated. Said glycosylation may be associated with an N-linked oligosaccharide glycan at asparagine 297 of the heavy chain.

[0113] In one embodiment, the present invention relates to a humanized antibody, or antibody fragment or derivative thereof, wherein said prostate specific membrane antigen is human prostate specific membrane antigen.

[0114] The humanized antibody of the present invention, or an antibody fragment or derivative thereof, may be a monoclonal antibody, or an antibody fragment or derivative thereof. The monoclonal antibody may preferably be of the isotype IgG.

[0115] The humanized antibody fragment of the present invention may be selected from the group consisting of a variable domain (Fv), a Fab fragment, and a F(ab)2 fragment.

[0116] The humanized antibody derivative of the present invention can be a single chain Fv (scFv).

[0117] In one embodiment, the invention relates to an antibody-drug conjugate (ADC) comprising a humanized antibody, or a fragment or derivative thereof, conjugated, optionally via a linker moiety, to a therapeutic agent.

[0118] In one embodiment, the present invention relates to a pharmaceutical preparation for the treatment of a medical disorder associated with overexpression of PSMA, comprising the humanized antibody of the present invention or its antibody fragment or antibody derivative, together with at least a pharma- ceutically acceptable carrier.The medical disorder associated with overexpression of PSMA can be prostate cancer.The pharmaceutical preparation of the present invention can be a preparation suitable for at least intravenous, intramuscular or subcutaneous administration.

[0119] In one embodiment, the invention relates to a humanized antibody, or an antibody fragment or derivative thereof, of the invention for use in the treatment of prostate cancer.

[0120] In one embodiment, the invention relates to a method of treating a patient suffering from prostate cancer comprising administering to the patient an effective amount of a humanized antibody of the invention, or an antibody fragment or antibody derivative thereof.

[0121] In one embodiment, the invention relates to a nucleic acid encoding a humanized antibody, antibody fragment or antibody derivative thereof of the invention.

[0122] In one embodiment, the present invention relates to a host cell capable of producing the humanized antibody, antibody fragment or antibody derivative of the present invention. The host cell of the present invention may be a mammalian cell. The mammalian cell of the present invention is preferably a CHO cell line.

[0123] The antibody or fragment or derivative of the present invention can be produced by transfecting an expression vector containing the coding sequence of the antibody of the present invention into a host cell. The expression vector or recombinant plasmid is produced by placing the coding antibody sequence under the control of suitable regulatory genetic elements, including a promoter and enhancer sequence, such as the CMV promoter. The heavy and light chain sequences can be expressed from individual expression vectors, or dual expression vectors that are co-transfected. The transfection can be transient or stable transfection. The transfected cells are then cultured to produce the transfected antibody construct. If stable transfection is performed, then stable clones that secrete antibodies with appropriately related heavy and light chains are selected by screening with a suitable assay, such as, for example, ELISA, subcloning, and growing for future production. EXAMPLES

[0124] While the present invention has been illustrated and described in detail in the drawings and the foregoing description, such illustration and description are to be considered as illustrative or exemplary and not restrictive, and the present invention is not limited to the disclosed embodiments. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims should not be construed as limiting the scope.

[0125] All amino acid sequences disclosed herein are shown from N-terminus to C-terminus; all nucleic acid sequences disclosed herein are shown 5'->3'.

[0126] Example 1: Humanization and deimmunization of murine antibody 3F11 The murine anti-PSMA antibody 3 / F11 (Wolf et al., 2010a; EP 1883698 B2) was subjected to humanization by grafting of complementarity determining regions (CDRs) into acceptor frameworks selected from a set of human germline sequences.

[0127] Epibase® immunoprofiling of parent 3 / F11 against 85 HLA class II allotypes was performed to analyze the epitope or cluster of adjacent epitopes for substitutions that would eliminate or reduce binding to HLA allotypes, focusing on the HLA-DRB1 allotype. Critical positions that constitute the VH / VL interchain interface or participate in a distinct set of CDR canonical structures were identified. Based on sequence alignment of the parent antibody sequence against human germline sequences, the closest match, the optimal human germline sequence as the acceptor, was identified, taking into account in particular sequence identity across the framework, identical or compatible interchain interface residues, supporting loops with the parent CDR canonical conformation, and the combination of heavy and light chain germlines found in the expressed antibody. 3 / F11 was found to be most compatible with human light chain germline VK2-A17 and heavy chain germline VH1-1-69 of the VK2 light chain germline family VK2 and human heavy chain germline family VH1, respectively. For CDR grafting, amino acids of the parent framework that differ from the selected acceptor were replaced with the corresponding human amino acids.

[0128] The sequences were analyzed for potential post-translational modification sites (asparagine deamidation, aspartate isomerization, C-terminal lysine clipping, free Cys thiol groups, N- and O-glycosylation, N-terminal cyclization, oxidation and pyroglutamate formation) that may cause problems during the development of a therapeutic product.V A structural model of the domain was generated using an in silico modeling platform and a complete F V As well as for the framework and CDRs, candidate structural fragments were scored and ranked.

[0129] In order to evaluate the impact of all possible substitutions in an effective manner, substitutions were grouped where appropriate, and a total of four humanized / deimmunized light chains and four humanized / deimmunized heavy chains were selected. Table 2 lists the names of the engineered amino acid chains along with a description of the modifications. The amino acid sequences of the engineered chains are shown in Table 1. Sequence alignments of the humanized / deimmunized light and heavy chains, respectively, are shown in Figures 3 and 4.

[0130] The heavy chain variable domain variant 3F11-VH-1 contains all three CDRs of the parent mouse VH sequence. In variant 3F11-VH-2, two CDRs are identical to the parent mouse VH sequence, but CDR H2 differs from the parent mouse sequence at four positions (aa 61-63: NEN > AQK, aa 65: K > Q). In variant 3F11-VH-3, again two CDRs are identical to the parent mouse VH sequence, but CDR H2 differs from the parent mouse sequence at five positions (aa 58: T > V, aa 61-63: NEN > AQK, aa 65:K > Q). In variant 3F11-VH-4, two CDRs are again identical to the parental murine VH sequence, but CDR H2 differs at six positions from the parental murine sequence (aa 56: G > S, aa 58: T > V, aa 61-63: NEN > AQK, aa 65: K > Q).

[0131] The light chain variable domain variant 3F11-VL-1 contains all three CDRs of the parent mouse VL sequence. In variant 3F11-VL-2, the two CDRs are identical to the parent mouse VL sequence, but CDR L1 contains a substitution at amino acid number 35 (N>Q). In variant 3F11-VL-3, the two CDRs are identical to the parent mouse VL sequence, but CDR L1 differs from the parent mouse sequence at two positions (aa 33: N>S, aa 35: N>Q). In variant 3F11-VL-4, CDR L1 differs from the parent mouse sequence at two positions (aa 33: N>S, aa 35: N>Q), CDR L2 differs at one position (aa 60: F>A), and CDR L3 is identical to the parent mouse sequence.

[0132] (Table 2) JPEG0007586399000001.jpg183170

[0133] 3F11-VL-2 optionally has the following mutations: aa 41: F>Y. 3F11-VL-3 optionally has the following mutations: aa 41: F>Y. 3F11-VL-4 optionally has the following mutations: aa 41: F>Y; aa 42: L>Q, and / or aa 90: V>T.

[0134] To evaluate the impact of substitution combinations, a variant combination experimental design was set up as outlined in Table 3. The combination of four humanized / deimmunized heavy chains and four humanized / deimmunized light chains led to 16 different humanized / deimmunized full-length antibodies, which were named as shown in Table 3.

[0135] Table 3: 3F11 humanized / deimmunized variant VH and VL chain combinations form 16 different variants of humanized variant domains and full length antibodies, respectively. 3F11 wt is composed of murine VH and VL chains. (Table 3) JPEG0007586399000002.jpg76170

[0136] The predicted important epitopes of HLA DRB1, DRB3 / 4 / 5, DQ and DP for the parental and humanized / deimmunized sequences are shown in Table 4. The differences between the parental and variant antibodies in Table 4 take into account the removal of each potential epitope.

[0137] Table 4: Predicted number of important epitopes for HLA DRB1, DRB3 / 4 / 5, DQ and DP for parental and humanized / deimmunized sequences (Table 4) JPEG0007586399000003.jpg128170

[0138] Example 2: Expression and purification of humanized anti-PSMA antibody variants The heavy and light chain variable domains of the antibodies were synthesized and subcloned into the GS Xceed® vector (Lonza). The light chain variable domain coding region was transferred into the pXC-Kappa vector, and the heavy chain variable domain coding region was transferred into the pXC-IgG1zadeltaK vector.

[0139] Single gene vectors (SGV) were transiently co-transfected into Chinese hamster ovary GS knockout cells (CHOK1SV GS-KO) with reference mouse antibodies at a 200 ml scale. Transient transfections were performed by electroporation (Gene Pulse XCell, BioRad) using CHOK1SV GS-KO cells cultured for a minimum of 2 weeks. Cells were subcultured 24 h prior to transfection and cell viability was >99% at the time of transfection. Six days after transfection, cell viability and viable cell concentration were measured at harvest. Cells ranged from 4.4 to 7.5 × 10 6 viable cells / ml, and the viability was 89.0-96.1%.

[0140] The clarified supernatant was purified by Protein A chromatography using a 5ml HiTrap MabSelect SuRE column pre-packed on an AKTA purifier using standard methods. Upon elution with 10mM sodium formate, pH 3.5, the pH of the protein-containing fractions was quickly adjusted by adding 2x phosphate buffered saline (PBS) to give final buffer conditions of 1x PBS, 5mM sodium formate, pH 7.4. Table 5 below shows the titers and product yields for the 3F11 parental murine antibody and various humanized antibody variants.

[0141] As a surprising effect, we observed significantly higher expression titers for several humanized anti-PSMA antibody variants compared to the parental murine 3F11 antibody (see Table AA). Expression titers of 9.5-13 mg / liter were obtained with humanized variants 3F11_var2, 3F11_var6, 3F11_var9, 3F11_var11, 3F11_var13, 3F11_var15 and 3F11_var16 compared to 3.88 mg / liter for the parental murine 3F11 antibody. Due to the 3-4 fold increase in expression titers, the production costs of humanized / deimmunized variant-based therapeutics are expected to be significantly reduced.

[0142] Table 5: Characterization results of humanized 3F11 antibody variants (Table 5) JPEG0007586399000004.jpg123170

[0143] Example 3: Biochemical characterization of humanized anti-PSMA antibody variants Purified samples were analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and size exclusion (SE)-HPLC on an Agilent 1200 series system using a Zorbax GF-250 (9.4 mm ID x 25 cm) column (Agilent).

[0144] In SDS-PAA gels, under non-reducing conditions, a signal of >98 kDa was observed for all variants comparable to the control antibody (IgG1) and parental 3F11, consistent with full-length antibodies. Under reducing conditions, a signal of approximately 28 kDa and one signal of approximately 49 kDa were observed for all variants comparable to the control antibody and parental 3F11, respectively, consistent with the sizes of the antibody light and heavy chains. Representative results are shown in Figure 5.

[0145] The retention times of the humanized 3F11 variant monomeric antibodies (7.96-8.00 min) in SE-HPLC were comparable to the 3F11 murine parent antibody (7.968 min).SE-HPLC chromatographs of all samples showed an additional minor peak at a lower retention time (approximately 7.2 min) with an area of ​​up to 5.4% corresponding to higher molecular weight impurities such as soluble aggregates.

[0146] Example 4: Cell binding studies with humanized anti-PSMA antibody variants The quality and binding activity of all humanized 3F11 antibody variants were assessed in a cell-binding assay with the human prostate cancer cell line LNCAP (German Collection of Microorganisms and Cell Cultures, DSMZ no. ACC 256) in a 96-well plate-based assay format and in a cell-based FACS analysis. Pre-confluent LNCAP cells at 80-90% confluence were washed and detached from the plates with trypsin before being incubated for 1 h at 37 °C, 5% CO2 to recover cell surface proteins. Cells were washed and plated at 1 x 10 in staining medium (RPMI-1640, 25 mM HEPES, 3% hi FBS and 0.02% sodium azide). 6The concentration of each antibody was adjusted to 1000 μg / ml of cells. Monoclonal antibody samples (mouse wt and 16 humanized variants) at various concentrations ranging from 50 to 0.0032 μg / ml were prepared in staining medium and incubated with cells in pre-chilled 96-well plates on ice for 30 min to allow antibody binding to the LNCAP cell surface. Cells were then washed with PBS at 4°C to remove free antibody, and bound antibody was labeled with FITC goat anti-human IgG H&L serum adsorbed (Abcam) secondary antibody for 30 min on ice. Finally, cells were washed and fixed, and FITC fluorescence associated with cell-bound antibody was measured by flow cytometry (GUAVA 8HT cytometer). Data were analyzed using Incyte GuavaSoft 2.7 software (Merck-Millipore). The resulting dose-response data (plotted as log-transformed dose vs. response) typically yielded sigmoidal curves with variable slope, which were fitted to a four-parameter logistic dose-response equation using GraphPad Prism software. EC defined as the concentration of antibody that provides half-maximal target cell binding 50 Doses were used to compare the biological activity of the different antibody variants. Data points represent the mean and standard error of the mean (SEM) of duplicate wells. Binding curves of the parental 3F11 wt antibody and the 16 humanized variants are shown in Figure 6.

[0147] Quite surprisingly, cell-based binding assays revealed that all 16 3F11 humanized variants bound to PSMA-expressing cells with higher affinity than was obtained with the parental murine antibody 3F11. The binding activity (EC 50 ) ranged from 0.1784 to 0.3732 μg / ml, and in most cases, the parent antibody 3F11 (EC 50 = 0.4189 μg / ml) (based on 95% confidence interval, CI). 3F11-var7 had the highest binding activity (EC 50 = 0.1784 μg / ml), followed by 3F11-var8 and 3F11-var6 (EC 50 = 0.2040 μg / ml and EC 50= 0.2187 μg / ml, see Table 6). Results from one experiment are shown in FIG.

[0148] Table 6: EC of humanized 3F11 antibody variants 50 Binding activity and 95%CI (Table 6) JPEG0007586399000005.jpg120170

[0149] These findings indicate that most of the humanized variants (all of them except the heavy chain variant 3F11_VH_1) did not require back-mutations in the framework regions spanning between the CDRs, and in particular none of the variants underwent affinity maturation, which is known to be usually required after antibody humanization due to significant reduction or loss of affinity. Surprisingly, it was found that the humanized variant generation procedures and variants of the present invention do not require such additional efforts.

[0150] Example 5: Cell binding studies with anti-PSMA scFv constructs Combinations of 3F11-VH-1 and 3F11-VL-1 chains (3F11-var1), and 3F11-VH-4 and 3F11-VL-4 chains (3F11-var16) were also used to generate single-chain Fv (scFv) constructs. For both constructs, the respective variable domain sequences were cloned in VL-VH composition into the expression vector pHOG21 and expressed in bacterial cells. After affinity purification, both scFv constructs were tested for binding to PSMA-expressing cells of the LNCAP subline C4-2 and to PSMA-negative DU145 prostate cancer cells. Both scFv constructs showed strong binding to C4-2 cells but not to the negative control DU145 cells.

[0151] References: JPEG0007586399000006.jpg176170JPEG0007586399000007.jpg199170JPEG0007586399000008.jpg161170

[0152] array The following sequences form part of the disclosure of this application. A WIPO ST 25 compatible electronic sequence listing is also provided with this application. For the avoidance of doubt, in case of any discrepancy between the sequences in the table below and the sequences in the electronic sequence listing, the sequences in the table shall be deemed correct. h / d: humanized / deimmunized, CDR: complementarity determining region, FR: framework region, CR: constant region JPEG0007586399000009.jpg192170JPEG0007586399000010.jpg114170JPEG0007586399000011.jpg161170JPEG0007586399000012.jpg160170 JPEG0007586399000013.jpg163170JPEG0007586399000014.jpg154170JPEG0007586399000015.jpg161170JPEG0007586399000016.jpg102170

Claims

1. CDR H1 sequence as set forth in SEQ ID NO:1, a CDR H2 sequence selected from any of the sequences set forth in SEQ ID NO:2, SEQ ID NO:7, SEQ ID NO:9, and SEQ ID NO:11; and CDR H3 sequence as set forth in SEQ ID NO:3 and a variable heavy (VH) domain comprising A CDR L1 sequence selected from any of the sequences set forth in SEQ ID NO: 4, SEQ ID NO: 8 and SEQ ID NO: 10; A CDR L2 sequence selected from any of the sequences set forth in SEQ ID NO:5 and SEQ ID NO:12, and CDR L3 sequence as set forth in SEQ ID NO:6 wherein the variable light (VL) domain comprises does not include a combination of the CDR H1 sequence set forth in SEQ ID NO: 1, the CDR H2 sequence set forth in SEQ ID NO: 2, the CDR H3 sequence set forth in SEQ ID NO: 3, the CDR L1 sequence set forth in SEQ ID NO: 4, the CDR L2 sequence set forth in SEQ ID NO: 5, and the CDR L3 sequence set forth in SEQ ID NO: 6; A humanized antibody or antigen-binding fragment or portion thereof that binds to the extracellular domain of prostate-specific membrane antigen (PSMA).

2. The humanized antibody or antigen-binding fragment or portion thereof according to claim 1, comprising a VH domain having at least 90%, preferably 95%, sequence identity to a sequence selected from any of the sequences set forth in SEQ ID NO: 13, SEQ ID NO: 15, SEQ ID NO: 17 and SEQ ID NO:

19.

3. 2. The humanized antibody or antigen-binding fragment or portion thereof of claim 1, comprising a VH domain sequence selected from any of the sequences set forth in SEQ ID NO: 13, SEQ ID NO: 15, SEQ ID NO: 17 and SEQ ID NO:

19.

4. The humanized antibody or antigen-binding fragment or portion thereof described in claim 1, comprising a VL domain having at least 90%, preferably 95%, sequence identity to a sequence selected from any of the sequences set forth in SEQ ID NO: 14, SEQ ID NO: 16, SEQ ID NO: 18 and SEQ ID NO:

20.

5. The humanized antibody or antigen-binding fragment or portion thereof of claim 1, comprising a VL domain sequence selected from any of the sequences set forth in SEQ ID NO: 14, SEQ ID NO: 16, SEQ ID NO: 18 and SEQ ID NO:

20.

6. a VH domain sequence as set forth in SEQ ID NO: 13 and a VL domain sequence selected from any of the sequences as set forth in SEQ ID NO: 16, SEQ ID NO: 18 and SEQ ID NO: 20; or a VH domain sequence as set forth in SEQ ID NO: 15 and a VL domain sequence selected from any of the sequences as set forth in SEQ ID NO: 14, SEQ ID NO: 16, SEQ ID NO: 18 and SEQ ID NO: 20; or a VH domain sequence as set forth in SEQ ID NO: 17 and a VL domain sequence selected from any of the sequences as set forth in SEQ ID NO: 14, SEQ ID NO: 16, SEQ ID NO: 18 and SEQ ID NO: 20; or A VH domain sequence as set forth in SEQ ID NO: 19 and a VL domain sequence selected from any of the sequences set forth in SEQ ID NO: 14, SEQ ID NO: 16, SEQ ID NO: 18 and SEQ ID NO: 20, The humanized antibody or antigen-binding fragment or portion thereof according to claim 1.

7. A humanized antibody or antigen-binding fragment or portion thereof, which comprises the VH domain sequence set forth in SEQ ID NO: 13 and the VL domain sequence set forth in SEQ ID NO: 14 and binds to the extracellular domain of prostate-specific membrane antigen (PSMA).

8. 4. The humanized antibody or antigen-binding fragment or portion thereof of claim 3, wherein the heavy chain comprises any of the sequences set forth in SEQ ID NO:21, SEQ ID NO:23, SEQ ID NO:25, or SEQ ID NO:

27.

9. 6. The humanized antibody or antigen-binding fragment or portion thereof of claim 5, wherein the light chain comprises any of the sequences of SEQ ID NO:22, SEQ ID NO:24, SEQ ID NO:26 or SEQ ID NO:

28.

10. A humanized antibody or antigen-binding fragment or portion thereof that binds to the extracellular domain of prostate-specific membrane antigen (PSMA), comprising at least one heavy chain comprising the sequence of any of SEQ ID NOs: 21, 23, 25 or 27, and at least one light chain comprising the sequence of any of SEQ ID NOs: 24, 26 or 28.

11. A humanized antibody or antigen-binding fragment or portion thereof, which comprises the VH domain sequence set forth in SEQ ID NO:21 and the VL domain sequence set forth in SEQ ID NO:22 and binds to the extracellular domain of prostate-specific membrane antigen (PSMA).

12. A humanized antibody or antigen-binding fragment or portion thereof that binds to the extracellular domain of prostate-specific membrane antigen (PSMA), comprising at least one heavy chain comprising SEQ ID NO:27 and at least one light chain comprising SEQ ID NO:

28.

13. E.C. 50 The humanized antibody or antigen-binding fragment or portion thereof of any one of claims 1 to 10, which binds to the extracellular domain of PSMA with an affinity of <0.4 μg / ml.

14. 12. The humanized antibody or antigen-binding fragment or portion thereof of any one of claims 1 to 11, wherein the antibody is glycosylated.

15. 15. The humanized antibody or antigen-binding fragment or portion thereof of claim 14, wherein the glycan is an N-linked oligosaccharide chain at asparagine 297 of the heavy chain.

16. The humanized antibody or antigen-binding fragment or portion thereof according to any one of claims 1 to 15, wherein the prostate specific membrane antigen is human prostate specific membrane antigen.

17. 17. The humanized antibody or antigen-binding fragment or portion thereof of any one of claims 1 to 16, wherein the antibody is of isotype IgG.

18. The antigen-binding fragments of the humanized antibody include variable domains (Fv), Fab fragments and F(ab) fragments. 2 An antigen-binding fragment of the humanized antibody of any one of claims 1 to 17, selected from the group consisting of fragments thereof.

19. The antigen-binding fragment of a humanized antibody according to any one of claims 1 to 18, wherein the antigen-binding fragment of the humanized antibody is a single chain Fv (scFv).

20. 20. An antibody-drug conjugate (ADC) comprising the humanized antibody, or antigen-binding fragment or portion thereof, of any one of claims 1 to 19, conjugated to a therapeutic agent.

21. 21. The antibody-drug conjugate (ADC) of claim 20, wherein the humanized antibody, or antigen-binding fragment or portion thereof, is attached to the therapeutic agent via a linker moiety.

22. A pharmaceutical formulation for the treatment of a medical disorder associated with overexpression of PSMA comprising a humanized antibody or antigen-binding fragment or portion thereof according to any one of claims 1 to 19, together with at least a pharma- ceutical acceptable carrier.

23. 23. The pharmaceutical formulation of claim 22, wherein the formulation is suitable for at least intravenous, intramuscular or subcutaneous administration.

24. A humanized antibody or antigen-binding fragment or portion thereof according to any one of claims 1 to 19, or a pharmaceutical formulation according to claim 22 or 23, for use in the treatment of prostate cancer.

25. A pharmaceutical composition for treating a patient suffering from prostate cancer, comprising an effective amount of the humanized antibody or antigen-binding fragment or portion thereof according to any one of claims 1 to 19.

26. A nucleic acid encoding the humanized antibody or antigen-binding fragment or portion thereof according to any one of claims 1 to 19.

27. A host cell capable of producing the humanized antibody or antigen-binding fragment or portion thereof according to any one of claims 1 to 19.

28. 28. The host cell of claim 27, wherein the host cell is a mammalian cell.

29. 29. The host cell of claim 28, wherein the mammalian cell is a CHO cell line.

Citation Information

Patent Citations

  • Methods for antibody engineering

    JP2008518619A

  • Monoclonal Antibodies and Single-Chain Antibody Fragments Against Cell-Surface Prostate-Specific Membrane Antigens

    JP2008541711A

  • Method for humanizing and affinity maturing antibodies

    JP2012505654A