Engineered antibodies and antibody-drug conjugates comprising the engineered antibodies

Abstract

Provided herein is an engineered dimeric antibody comprising an antibody Fab as a first Fab domain, a second Fab domain comprising a TCR constant domain, an engineered hinge domain, and an Fc domain having a knob-hole mutation. Also provided is an antibody-drug conjugate and a method for preparing the same, a composition comprising the antibody-drug conjugate and its use in treating a disease, disorder or condition, wherein the antibody-drug conjugate comprises the engineered antibody coupled to one or more drug molecules via a linker.

Description

Technical Field

[0001] The present invention relates generally to the field of biopharmaceuticals, in particular to engineered antibodies and antibody-drug conjugates. Background Art

[0002] Antibodies are multifunctional immunoglobulins that have unique binding specificity for target antigens and a range of antigen-independent immune interactions, thereby playing an important role in the immune system. Many therapeutic biopharmaceuticals, diagnostic reagents, and research reagents currently in use are antibodies against antigens associated with pathologies, immune mechanisms, or biological mechanisms of interest.

[0003] In recent years, people have made a lot of efforts to develop antibody conjugates with drug loads. With regard to antibody drug conjugates (ADCs), ADCs include antibodies for targeting, linkers for drug attachment, and potent drug loads as effectors. Antibodies or their related forms bring cytotoxic drugs to cells expressing antigens or other target cells through antibody-antigen interactions. Meanwhile, toxicity significantly decreases after drug-antibody coupling. Therefore, ADCs expand the treatment window by reducing the minimum effective dose (MED) and increasing the maximum tolerated dose (MTD). FDA-approved ADC drugs include, for example, Mylotarg, Adcetris, Kadcyla, Besponsa, Polivy, Padcev, Enhertu, Trodelvy, and Blenrep.

[0004] The successful development of ADCs depends on the selection of antibodies, the selection of linker-drug payloads, the coupling method of linker-drug payloads, and the development of the coupling process. Cysteine ​​thiol groups in antibodies are ideal coupling reaction groups as strong nucleophiles. In the native form of antibodies, cysteine ​​residues exist in the form of disulfide bonds. Therefore, the reduction of disulfide bonds between the light chain and the heavy chain of antibodies provides ideal free cysteine ​​thiol groups for coupling. In order to cope with the opportunities and challenges associated with the preferred drug payload-antibody ratio (PAR) and the coupling position, a variety of coupling methods have been developed in the art. Ideally, an appropriate amount of drug payload should be connected to the antibody, resulting in a heterogeneous ADC product. Low PAR coupling products have insufficient efficacy, while high PAR products are highly toxic and unstable. Therefore, the heterogeneity of ADCs hinders the expansion of the therapeutic window. Therefore, efforts such as antibody engineering have been made to improve the homogeneity of ADC products.

[0005] One approach uses point mutations in antibodies to introduce amino acids with highly reactive residues for conjugation. TMThe technology was developed by Genentech, which induces cysteine ​​mutations on antibodies (Jagath R Junutula, et al., Site-specific conjugation of a cytotoxic drug to an antibody improves the therapeutic index, Nature Biotechnology, 2008, 26 (8): 925–932). The conjugation with Thiomab occurs on the modified cysteine ​​residue after reduction, producing a highly homogeneous conjugated product. Non-natural amino acid (NNAA) technology is also used to produce homogeneous conjugates. For example, unnatural amino acids are used to introduce keto groups or azide groups into antibodies as coupling sites (Jun Y. Axup, et al., Synthesis of site-specific antibody-drug conjugates using unnatural amino acids, PNAS, 2012, 109(40): 16101-16106; Michael P. Van Brunt, et al., Genetically Encoded Azide Containing Amino Acid in Mammalian Cells Enables Site-Specific Antibody–Drug Conjugates Using Click Cycloaddition Chemistry, Bioconjugate Chem., 2015, 26(11): 2249-2260). Due to the specific reaction, highly homogeneous products are also obtained.

[0006] There are some disadvantages to point mutation-based methods. First, the mutation site needs to be carefully selected, otherwise the stability and binding efficiency of the antibody will be affected. Second, the expression level of point mutation antibodies is usually low, which will become a problem in the chemical composition production and control (CMC) stage.

[0007] Another method is to introduce a short polypeptide tag that can be recognized by the enzyme as a conjugation site. Glutamine tag (LLQG) as mTG recognition motif (Pavel Strop et al., "Location Matters: Site of Conjugation Modulates Stability and Pharmacokinetics of Antibody Drug Conjugates", Chemistry & Biology, 2013, 20 (2): 161-167), LPETG as sortase A recognition motif (Roger R. Beerli et al., "Sortase Enzyme-Mediated Generation of Site-Specifically Conjugated Antibody Drug Conjugates with High In Vitro and In Vivo Potency", PLOS ONE, 2015, 10 (7): e0131177), and LCxPxR as formylglycine generator (FGE) recognition motif (Peng et al., "Sortase Enzyme-Mediated Generation of Site-Specifically Conjugated Antibody Drug Conjugates with High In Vitro and In Vivo Potency", PLOS ONE, 2015, 10 (7): e0131177). Wu et al., "Site-specific chemical modification of recombinant proteins produced in mammalian cells by using the genetically encoded aldehyde tag", PNAS, 2009, 106(9): 3000-3005) was used for conjugation to obtain a highly homogeneous product, in which the drug was linked to the polypeptide tag.

[0008] The disadvantages of short peptide tags are similar to those of point mutation-based methods. The insertion site of the peptide tag needs to be screened, but the available sites of peptide tags are limited. In addition, the expression titer of the tagged antibody is also a difficulty when using this strategy.

[0009] The most direct approach to producing antibody conjugates is to utilize the thiol groups of natural cysteines in antibody heavy and light chain polypeptides. As a strong nucleophile, thiol groups can achieve fast and efficient conjugation reactions in aqueous phase. In FDA-approved ADC drugs such as Adcetris and Polivy, MMAE is conjugated to cysteine ​​residues generated by partially reducing interchain disulfide bonds by reacting the thiol groups on the cysteine ​​residues with the maleimide groups in the linker used for monomethyl auristatin E (MMAE). Partial reduction is preferred over complete reduction because the hydrophobicity of the drug and the steric hindrance of all cysteine ​​residues during connection make the ADC drug unstable in plasma. However, the product obtained after partial reduction has poor homogeneity, which poses difficulties in CMC, especially the control of the conjugation process and QC compliance analysis. On the other hand, substances with different DARs in the ADC mixture show different stability and efficacy in vivo, which means that not all substances in the heterogeneous ADC product meet the expected structural type.

[0010] There are many similarities and differences in disulfide bond structures between IgG subclasses IgG1, IgG2, IgG3, and IgG4. Taking IgG1 and IgG4, which are the most commonly used therapeutic biopharmaceuticals, as examples, the two heavy chains of IgG1 and IgG4 are connected by two disulfide bonds and have a total of 12 intrachain disulfide bonds; however, the light chain of IgG1 is connected to the heavy chain through a disulfide bond between its last residue and the fifth cysteine ​​residue of the heavy chain, while the light chain of IgG4 is connected to the heavy chain through a disulfide bond between its last cysteine ​​residue and the third cysteine ​​residue of the heavy chain. In general, the solvent exposure levels of intrachain disulfide bonds and interchain disulfide bonds are different. Intrachain disulfide bonds are buried between the secondary structures of each domain and are not exposed to the solvent. Interchain disulfide bonds located in the hinge region are highly exposed to the solvent, including the heavy chain-heavy chain disulfide bonds of IgG1 and IgG4 and the heavy chain-light chain disulfide bonds of IgG1. The IgG4 heavy chain-light chain disulfide bond is located between the VH and CH1 domain interfaces, which are relatively inaccessible, and therefore has low contact with the solvent. The difference in solvent exposure between different disulfide bonds has an important impact on the bioconjugation of antibodies, because exposed cysteine ​​residues are considered to be more reactive than non-exposed cysteine ​​residues (Hongcheng Liu & Kimberly May, Disulfide bond structures of IgG molecules: Structural variations, chemical modifications and possible impacts to stability and biological function, Mabs, 2012, 4 (1): 17-23). ​​Experiments have shown that the heavy chain-light chain disulfide bond and the heavy chain-heavy chain disulfide bond of IgG1 are both highly reactive.

[0011] The hinge region is a flexible linker between the antibody Fab and Fc. The length and flexibility of the hinge region vary greatly between IgG subclasses IgG1, IgG2, IgG3 and IgG4. Taking IgG1 and IgG4, which are the most commonly used therapeutic biopharmaceuticals, as examples, the hinge region of IgG1 contains 15 amino acids and is very flexible, while the hinge region of IgG4 is shorter, with only 12 amino acids (Gestur Vidarsson, et al., IgG subclasses and allotypes: from structure to effector functions, Front. Immunol., 2014, 5: 520). Wild-type IgG1 and IgG4 differ by one amino acid in the core hinge region (EU number 226-229): Cys-Pro-Pro-Cys in IgG1 and Cys-Pro-Ser-Cys in IgG4. In the core hinge region of natural IgG4, there is a balance between interchain cysteine ​​disulfide bonds and intrachain cysteine ​​disulfide bonds, so the presence of IgG4 half-antibody molecules can be observed after heavy chain arm exchange and secretion. It has been demonstrated that the S228P mutation of IgG4 prevents natural arm exchange and thus significantly stabilizes the covalent interactions between IgG4 heavy chains (S. Angal et al., "A single amino acid substitution abolishes the heterogeneity of chimeric mouse / human (IgG4) antibody", Molecular Immunology, 1993, 30(1): 105-108; John-Paul Silva et al., "The S228P Mutation Prevents in Vivo and in Vitro IgG4 Fab-arm Exchange as Demonstrated using a Combination of Novel Quantitative Immunoassays and Physiological Matrix Preparation", Journal of Biological Chemistry, 2015, 290(9):5462-5469), and therefore has been widely used in the development and production of IgG4 antibodies.The S228P mutation forms a polyproline helix in the IgG4 hinge (5 Pro in the lower hinge region), which, combined with the shorter IgG4 hinge length, further limits its flexibility compared to the IgG1 hinge (3 Pro in the lower hinge region). The difference in flexibility between different hinges is important for the bioconjugation of antibodies, because cysteine ​​residues in flexible hinge fragments are considered to be more reactive than cysteine ​​residues in rigid hinges. Experiments have shown that both the heavy chain-light chain disulfide bonds and the heavy chain-heavy chain disulfide bonds of S228P IgG4 are weakly reactive.

[0012] The disadvantage of using native cysteines for antibody conjugation is that the similar reactivity between the four interchain disulfide bonds in IgG1 and IgG4 results in highly heterogeneous conjugated products. As mentioned above, this heterogeneity narrows the therapeutic window for conjugated drugs used in the clinic. For example, ADCs produced by partially reducing the native interchain disulfide bonds in IgG1 antibodies produce a product mixture with a normal distribution. The product heterogeneity of partially reduced IgG4 antibodies is even higher, with a lot of antibody not being reduced when the fully reduced antibody level is already high.

[0013] WuXiBody TM WuXiBody (hereinafter also referred to as "WuXiBody") is an innovative bispecific antibody (bsAb) platform developed by WuXi Biologics. Its main feature is the replacement of the CH1 / CL constant domain in the antibody Fab domain with the T cell receptor (TCR) constant domain, as described in PCT application PCT / CN2018 / 106766 (International Publication WO2019 / 057122). WuXiBody TM The design ensures homologous HC-LC pairing. WuXiBody-based BsAbs can be in an asymmetric format or in a symmetric format. Heterodimerization can be ensured by the "knob-in-hole" ("KIH") technology.

[0014] Nevertheless, there is still a need to improve antibody bioconjugated PARs, especially for therapeutic applications, thereby eliminating some or all of the above-mentioned deficiencies as much as possible. Summary of the invention

[0015] The present disclosure provides an engineered antibody comprising a WuXiBody TM The Fab domain of the TCR constant domain in the IgG1 antibody, the engineered hinge region and the Fc domain with KIH mutation. Surprisingly, the ADC produced with this engineered antibody has high homogeneity and well-controlled DAR. These ADCs are advantageously characterized by high stability and excellent therapeutic effect.

[0016] In a first aspect, the present invention provides an engineered dimeric antibody, wherein a first monomer comprises a first Fab domain and a first engineered hinge region operably linked thereto, and thereafter a first Fc region operably linked thereto, and a second monomer comprises a second Fab domain and a second hinge region operably linked thereto, and thereafter a second Fc region operably linked thereto;

[0017] Wherein, the first Fab domain is an antibody Fab domain, and the second Fab domain comprises a fusion

[0018] an antibody Fv domain conjugated to a TCR constant domain; and

[0019] wherein the first engineered hinge region is composed of a truncated IgG1 hinge region portion and a truncated IgG4 hinge region portion or is a modified IgG4 hinge region, so that the first engineered hinge region is coupled to the second hinge region, and the hinge domain composed thereof comprises at least two interchain disulfide bonds; and

[0020] Wherein, the first Fc region is coupled to the second Fc region, and the Fc domain constituted thereby comprises a knob-to-hole mutation.

[0021] In another aspect, provided herein is a nucleic acid molecule or a combination of nucleic acid molecules encoding an engineered antibody of the present invention.

[0022] In another aspect, provided herein is an antibody-drug conjugate comprising an engineered antibody of the invention coupled to one or more drug molecules via a linker.

[0023] In another aspect, provided herein is a composition comprising or consisting of a mixture of an antibody-drug conjugate of the invention, wherein at least about 65%, preferably at least about 70%, at least about 75%, at least about 80%, or at least about 90% of the antibody-drug conjugates have a drug / antibody ratio of 2 (DAR2).

[0024] On the other hand, the present invention also provides a product comprising a mixture of antibody-drug conjugates of the present invention or consisting of a mixture of antibody-drug conjugates of the present invention, wherein at least about 80%, preferably at least about 85%, or at least about 90% of the antibody-drug conjugates have a drug / antibody ratio of 6 (i.e., DAR6).

[0025] In another aspect, the present invention provides a pharmaceutical composition comprising the antibody-drug conjugate of the present invention and a pharmaceutically acceptable carrier.

[0026] In another aspect, the present invention provides a method for preparing an antibody-drug conjugate of the present invention, comprising coupling a partially reduced antibody of the present invention with a linker-drug payload compound having a maleimide or haloacetyl moiety via a Michael addition reaction.

[0027] On the other hand, the present invention provides an antibody-drug conjugate product obtained by the method of the present invention, which comprises a mixture of antibody-drug conjugates of the present invention or consists of a mixture of antibody-drug conjugates of the present invention, wherein at least about 65%, preferably at least about 70%, at least about 75%, at least about 80% or at least about 90% of the antibody-drug conjugates have a drug / antibody ratio of 2.

[0028] On the other hand, the present invention provides an antibody-drug conjugate product obtained by the method of the present invention, which comprises a mixture of antibody-drug conjugates of the present invention or consists of a mixture of antibody-drug conjugates of the present invention, wherein at least about 80%, preferably at least about 85%, at least about 90% of the antibody-drug conjugates have a drug / antibody ratio of 6.

[0029] In another aspect, provided herein is a method for treating a disease, disorder or condition in a subject in need thereof, comprising administering to the subject a therapeutically effective amount of an antibody drug conjugate of the invention. The disease may be cancer.

[0030] In yet another aspect, provided herein are antibody-drug conjugates of the invention for use in treating a disease, disorder or condition in a subject in need thereof. The disease may be cancer.

[0031] The present invention has a number of advantages. Since the natural immunoglobulin G hinge sequence is used and exchanged at its natural structural position without introducing any new amino acid sequence, the resulting engineered antibody will cause less immunogenicity in vivo. Moreover, the engineered antibody of the present invention can obtain protein expression titers comparable to its IgG1 or IgG4 corresponding reference. In addition, the engineered antibody of the present invention can obtain highly homogeneous ADC products with well-controlled DAR.

[0032] The production of the antibody-drug conjugates of the present invention can be significantly simplified, and can be a simple (one-pot) coupling process, including: first partial reduction with a mild reducing agent, and then coupling in the same buffer system. The ADC products of the present invention have highly homogeneous and well-controlled DAR, for example, the percentage of DAR2 type exceeds 90% or the percentage of DAR6 type exceeds 87%. In addition, the ADC of the present invention has excellent in vitro and in vivo stability.

[0033] Through the following detailed description, other purposes, features and advantages of the present invention will be apparent. However, it should be understood that the detailed description and specific examples are only exemplary descriptions while showing the preferred embodiments of the present invention, because those skilled in the art can understand various changes and adjustments within the concept and scope of the present invention from these detailed descriptions. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The drawings are part of the specification and are used to further illustrate the relevant contents of the present invention. The following one or more drawings and detailed description of specific embodiments are helpful for better understanding of the present invention.

[0035] Figure 1 : Schematic diagram of the engineered antibody of the present invention, and a schematic diagram of using the antibody to produce ADC with DAR2 or DAR6 as the main product.

[0036] Figure 2 The structure of antibody 886-39 and its HIC-HPLC results for coupling with MC-vc-PAB-MMAE are shown. As shown in the HIC-HPLC results, DAR2 is the major product.

[0037] Figure 3 The LC-MS characterization results of 886-39-MMAE-DAR2 are shown. Based on the measured abundance of drug molecules bound to the light chain and the measured abundance of drug molecules bound to the heavy chain, most of the drug molecules are loaded on the antibody Fab domain, and no drug is loaded on the TCR constant domain.

[0038] Figure 4 The structure of antibody 886-39 and its HIC-HPLC results for coupling with MC-vc-PAB-MMAE are shown. As shown in the HIC-HPLC results, DAR6 is the major product.

[0039] Figure 5 The LC-MS characterization results of 886-39-MMAE-DAR6 are shown. As shown in the LC-MS results, no drug was loaded on the TCR constant domain.

[0040] Figure 6 The structure of antibody 886-40 and its HIC-HPLC results for coupling with MC-vc-PAB-MMAE are shown. As shown in the HIC-HPLC results, DAR2 is the major product.

[0041] Figure 7 The LC-MS characterization results of 886-40-MMAE-DAR2 are shown. Based on the measured abundance of drug molecules bound to the light chain and the measured abundance of drug molecules bound to the heavy chain, most of the drug molecules are loaded on the antibody Fab domain, and no drug is loaded on the TCR constant domain.

[0042] Figure 8 The structure of antibody 886-40 and its HIC-HPLC results for coupling with MC-vc-PAB-MMAE are shown. As shown in the HIC-HPLC results, DAR6 is the major product.

[0043] Fig. 9 The LC-MS characterization results of 886-40-MMAE-DAR6 are shown. As shown in the LC-MS results, no drug was loaded on the TCR constant domain.

[0044] Fig.10 The structure of antibody 886-41 and its HIC-HPLC results for coupling with MC-vc-PAB-MMAE are shown. As shown in the HIC-HPLC results, DAR2 is the major product.

[0045] Fig.11 The LC-MS characterization results of 886-41-MMAE-DAR2 are shown. Based on the measured abundance of drug molecules bound to the light chain and the measured abundance of drug molecules bound to the heavy chain, most of the drug molecules are loaded on the antibody Fab domain, and no drug is loaded on the TCR constant domain.

[0046] Fig.12 The structure of antibody 886-41 and its HIC-HPLC results for coupling with MC-vc-PAB-MMAE are shown. As shown in the HIC-HPLC results, DAR6 is the major product.

[0047] Fig.13 The LC-MS characterization results of 886-41-MMAE-DAR6 are shown. As shown in the LC-MS results, no drug was loaded on the TCR constant domain.

[0048] Fig.14 : Example antibody light chain sequence and heavy chain sequence, wherein the variable region is bold, the KIH mutation residue is bold, and the engineered hinge region is underlined. Wherein, "LC" refers to the light chain (LC) of the monomer containing the antibody Fab domain, "TCR-LC" refers to the light chain (LC) of the monomer containing the TCR constant domain in the Fab domain, "TCR-HC-knob" refers to the heavy chain (HC) containing the knob mutation in the Fc region, and "HC-hole" refers to the heavy chain (HC) containing the hole mutation in the Fc region.

[0049] definition

[0050] Herein, the singular form introduced by "a", "an" and "the / said" includes the plural meaning unless otherwise specified. Moreover, the terms "one", "one or more" and "at least one" can be used interchangeably in this article.

[0051] Ordinal numbers such as "first", "second", "third", "fourth", etc. do not imply a certain order or relationship, but are simply used for convenience and clarity to refer to multiple independent entities or components, unless the context clearly indicates otherwise.

[0052] As used herein, the term "about" or "approximately" refers to an amount, level, value, number, frequency, percentage, dimension, size, quantity, weight, or length that differs by at most 30%, 25%, 20%, 25%, 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2% or 1% relative to a reference amount, level, value, number, frequency, percentage, dimension, size, quantity, weight, or length. In a specific embodiment, the term "about" or "approximately" when preceding a numerical value indicates a range of plus or minus 15%, 10%, 5% or 1% of that value.

[0053] Herein, the term "substantially free" refers to not only the absence (i.e., "none", "zero", etc.) of a certain situation or the existence of a certain substance, but also the presence or amount of the substance that is not significant or below the detection limit and thus cannot be detected. This is well known to those skilled in the art.

[0054] As used herein, the term “exemplary” means “serving as an example, instance, or illustration.” Anything described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other aspects.

[0055] In this document, the terms "comprising", "including", "characterized by", and "having", as well as their grammatical variations, are used interchangeably and should be understood to include the stated steps or elements without excluding any other steps or elements. Thus, they include an exclusive inclusion represented by the closed term "consisting of" and its grammatical variations, as well as a semi-closed inclusion represented by the term "consisting essentially of", which is open-ended only to elements of unimportant quality and / or quantity.

[0056] One or more features of one embodiment herein may be combined with one or more features of another embodiment without departing from the spirit and concept of the present invention.

[0057] Unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those of ordinary skill in the art to which the present invention belongs. The public documents and patent documents cited in the text are incorporated herein by reference and are applicable to all purposes. The documents cited in this specification should be regarded as the technical level of those of ordinary skill in the art, but this should not be construed as an admission that the present invention cannot be earlier than such disclosure due to prior invention.

[0058] As used herein, the term "antibody" encompasses any immunoglobulin, monoclonal antibody, polyclonal antibody, multispecific antibody or bispecific (bivalent) antibody that binds to one or more specific antigens. Typically, antibodies include two heavy chains and two light chains, just like natural intact antibodies. Each heavy chain includes a variable region ("VH") and the first, second, third constant regions (CH1, CH2, CH3) and a conditional fourth constant region (CH4), such as in the case of IgM and IgE antibodies, while each light chain consists of a variable region ("VL") and a constant region (CL). The heavy chains of mammals are divided into α, δ, ε, γ and μ types, while the light chains of mammals are divided into λ and κ types. The variable regions of light and heavy chains are responsible for antigen binding. Each variable region typically contains three highly variable loops, referred to as "complementarity determining regions (CDRs)". The boundaries of CDRs can be defined or identified by the rules of Kabat, Chothia or Al-Lazikani. The three CDRs are inserted between the framework regions (FRs) flanking them, which are highly conserved compared to the CDRs and form a scaffold that supports the hypervariable loops. The constant regions of the heavy and light chains do not participate in antigen binding, but exhibit a variety of effector functions. The five main classes of antibodies are IgA, IgD, IgE, IgG, and IgM, which are characterized by α, δ, ε, γ, and μ heavy chains, respectively. Several major antibody classes are divided into subclasses, such as IgG1 (γ1 heavy chain), IgG2 (γ2 heavy chain), IgG3 (γ3 heavy chain), IgG4 (γ4 heavy chain), IgA1 (α1 heavy chain) or IgA2 (α2 heavy chain). Therefore, in the present invention, a specific IgG type, such as "IgG1 type" or "IgG1 type", refers to each IgG isotype of a specified subclass, while different IgG types refer to IgG isotypes of different subclasses.

[0059] Herein, "variable region" for antibodies refers to an antibody variable region or fragment thereof comprising one or more CDRs. Although the variable region may include a complete variable region (such as VH or VL), it may also be an incomplete variable region that still retains the ability to bind to an antigen or form an antigen binding site.

[0060] Antibodies can be in a "Y" shape, with the two arms also called "antigen-binding fragments (Fab)", and the stem comprising the hinge and Fc domains of the antibody.

[0061] The terms "Fab", "Fab domain" and "Fab arm" are used interchangeably and refer to the domain formed by the coupling of the light chain and the heavy chain at the variable region and the first constant region in an immunoglobulin (such as an antibody). Typically, the Fab domain may contain one or more interchain disulfide bonds. In some embodiments, the constant regions of the light and heavy chains may be replaced by TCR constant regions. The Fab domain is responsible for a variety of antigen binding activities.

[0062] Herein, "Fc region" refers to a fragment consisting of the second constant region (CH2) of the heavy chain and the constant region thereafter, or refers to a fragment consisting of a part of the hinge region, the second constant region (CH2) of the heavy chain and the constant region thereafter. Also, herein, the "Fc domain" of a dimeric antibody refers to the part where the Fc region of each chain in the coupled heavy chain is located. The Fc region has a variety of effector functions, such as ADCC and CDC.

[0063] Herein, the "hinge" or "hinge region" of the heavy chain refers to the region connecting the C-terminus of the CH1 region and the N-terminus of the CH2 region of the heavy chain. The length of the hinge region can be about 12-62 amino acid residues. In human IgG1, the hinge region includes residues 216 to 230, and in human IgG4, it includes residues 219 to 230, as numbered by EU. Herein, the "hinge domain" of a dimeric antibody refers to the portion where the hinge regions of each chain in the coupled heavy chain are located. Typically, the hinge domain may contain one, two or more interchain disulfide bonds. The hinge region is flexible so as to allow the two Fab domains to move independently.

[0064] The hinge region is a flexible connection region between the antibody Fab and Fc. The length and flexibility of the hinge region vary greatly between IgG subclasses IgG1, IgG2, IgG3 and IgG4. Taking IgG1 and IgG4, which are most commonly used as therapeutic biologics, as examples, the hinge region of IgG1 contains 15 amino acids (e.g., EPKSCDKTHTCPPCP (SEQ ID NO: 9)), which is very flexible, and the hinge region of IgG4 is shorter, with only 12 amino acids. Wild-type IgG1 and IgG4 differ by one amino acid in the core hinge region (EU numbering 226-229): Cys-Pro-Pro-Cys in IgG1 and Cys-Pro-Ser-Cys in IgG4. The core hinge region of natural IgG4 has a balance between interchain and intrachain cysteine ​​disulfide bonds, so heavy chain arm exchange occurs and IgG4 half antibody molecules exist after secretion. It has been demonstrated that the S228P mutation of IgG4 (e.g., ESKYGPPCPPCP (SEQ ID NO: 10)) can significantly stabilize the covalent interactions between IgG4 heavy chains by preventing natural arm exchange, and has therefore been widely used in the development and production of IgG4 antibodies. The S228P mutation forms a polyproline helix (PPCPPCP) in the IgG4 hinge, which, combined with the shorter IgG4 hinge length, further limits its flexibility compared to the IgG1 hinge. The difference in flexibility between different hinges is of great significance for the bioconjugation of antibodies, as cysteine ​​residues located in flexible hinge fragments are considered to be more reactive than cysteine ​​residues located in rigid hinges. Experiments have shown that both the heavy chain-light chain disulfide bonds and the heavy chain-heavy chain disulfide bonds of S228P IgG4 are weakly reactive.

[0065] As used herein, "CH2 domain" refers to the portion of a heavy chain molecule extending from about amino acid 244 to amino acid 360 in an IgG antibody according to the conventional numbering scheme (amino acids 244 to 360 by Kabat numbering, amino acids 231 to 340 by EU numbering).

[0066] The "CH3 domain" extends from the CH2 domain of the IgG molecule to the C-terminus and contains approximately 108 amino acids. Some immunoglobulins (eg, IgM and IgE) also contain a CH4 region.

[0067] "Fv", "Fv fragment" and "Fv domain" are used interchangeably to refer to the smallest domain that contains the complete antigen binding site of an antibody. An Fv domain generally includes a light chain variable region (VL) and a heavy chain variable region (VH) coupled to each other.

[0068] The identity (identity) percentage (%) between biological sequences (including amino acid sequences and nucleic acid sequences) refers to the percentage of identical residues between the query sequence and the reference sequence when the two are aligned according to the maximum match. Sequence identity can be determined using public tools such as BLASTN, BLASTp (available on the website of the National Center for Biotechnology Information (NCBI) of the United States), ClustalW2 (available on the website of the European Bioinformatics Institute), and ALIGN or Megalign (DNASTAR) software.

[0069] As used herein, "specific binding" or "specifically binds" refers to a non-random binding reaction between two molecules (e.g., between an antibody and an antigen). In some embodiments, the engineered antibodies provided herein can specifically bind to an antigen with a binding affinity (K D )≤10 -6 M (e.g. ≤5x10 -7 M, ≤2x10 -7 M, ≤10 -7 M, ≤5x10 -8 M, ≤2x10 -8 M, ≤10 -8 M, ≤5x10 -9 M, ≤2x10 -9 M, ≤10 -9 M or ≤10 -10 M). K in this paper D The ratio of the dissociation rate to the association rate (k off / k on ).

[0070] "Operably linked / connected" refers to the juxtaposition of two or more target biological sequences, with or without spacers or linkers between them, in a manner that allows their relationship to function in the intended manner. When used to describe a polypeptide, it means that the polypeptide sequences are connected in such a way that the connection product has the intended biological function. For example, the variable region of an antibody can be operably linked to the constant region to obtain a stable product with antigen binding activity. The term can also be used to describe a polynucleotide. For example, when a polynucleotide encoding a polypeptide is operably linked to a regulatory sequence (such as a promoter, enhancer, silencer sequence, etc.), it means that these polynucleotide sequences are connected in such a way that the polynucleotide expresses the polypeptide in a regulated manner.

[0071] A sequence that is not 100% identical to a reference sequence may contain mutations at one or more positions, which may be substitutions, additions, deletions, or a combination thereof. The substitutions may be "conservative substitutions," which refer to replacements with different amino acids having similar side chain physicochemical properties, or substitutions at sites that are not important for the activity or function of the sequence. For example, conservative substitutions may be substitutions between amino acids with nonpolar side chains (such as Met, Ala, Val, Leu and Ile, Pro, Phe, Trp), between amino acids with uncharged polar side chains (such as Cys, Ser, Thr, Asn, Gly and Gln), between amino acids with acidic side chains (such as Asp, Glu), between amino acids with basic side chains (such as His, Lys and Arg), between amino acids with beta-branched side chains (such as Thr, Val and Ile), between amino acids with sulfur-containing side chains (such as Cys and Met), or between amino acids with aromatic side chains (such as Trp, Tyr, His and Phe). Conservative substitutions do not cause significant changes in the conformational structure and can retain the biological activity of the protein.

[0072] As used herein, a "subject" refers to a human or non-human animal. A non-human animal may be a mammal, such as a primate. Examples of non-human mammalian subjects include, but are not limited to, captive animals, livestock animals, and zoo animals, competitive animals, or pets, such as dogs, cats, guinea pigs, rabbits, rats, mice, horses, pigs, cows, and bears. Preferably, the subject is a human. A "subject in need" refers to a subject having a disease, disorder, or condition for which diagnosis, prognosis, alleviation, prevention, and / or treatment is desired. DETAILED DESCRIPTION

[0073] The following description is only for illustrating various embodiments of the present invention. Therefore, the specific modifications, changes, etc. discussed herein should not be interpreted as limiting the scope of the present invention. It is obvious to those skilled in the art that various equivalent forms, adjustments and changes can be obtained within the scope of the present invention, and it should be understood that these equivalent embodiments are included in the scope of the present invention. All references cited herein, including publications, patents and patent applications, are incorporated herein by reference in their entirety.

[0074] Engineered Antibodies

[0075] Content 1, the present invention provides an engineered dimeric antibody, wherein the first monomer comprises a first Fab domain, a first engineered hinge region operably connected thereto, and a first Fc region operably connected thereto, and the second monomer comprises a second Fab domain, a second hinge region operably connected thereto, and a second Fc region operably connected thereto;

[0076] Wherein, the first Fab domain is an antibody Fab domain, and the second Fab domain comprises an antibody

[0077] The Fv domain of the antibody and the TCR constant domain fused thereto; and

[0078] wherein the first engineered hinge region is composed of a truncated IgG1 hinge region portion and a truncated IgG4 hinge region portion or is a modified IgG4 hinge region, and the hinge domain formed by coupling the first engineered hinge region and the second hinge region comprises at least two interchain disulfide bonds; and

[0079] Wherein, the Fc domain formed by the coupling of the first Fc region and the second Fc region comprises a knob-to-hole mutation.

[0080] In some embodiments, the engineered antibodies of the present invention can be Figure 1 As shown in the schematic diagram of , wherein the TCR constant domain is represented by a pair of rectangles. The monomers are coupled to form a dimeric antibody through interchain binding forces (including interchain bonds and / or interchain interactions). Examples of interchain binding forces include, but are not limited to, disulfide bonds, hydrogen bonds, electrostatic interactions, salt bridges, hydrophobic-hydrophilic interactions, and knobs-into-holes mechanisms. The engineered antibodies of the present invention are heterodimers.

[0081] For the engineered antibodies of the present invention, "antibody Fab domain" conforms to the standard meaning commonly understood by those skilled in the art, and refers to a Fab domain having the typical structure and function of the Fab domain in an antibody immunoglobulin. TM The technology replaces the constant domain with the classical Fab domain of the TCR constant domain.

[0082] In the present invention, the antibody Fab domain can be a Fab domain derived from any antibody, particularly those clinical antibodies. In some embodiments, the antibody Fab domain is derived from an antibody that specifically binds to a tumor antigen (TA) such as a tumor-specific antigen (TSA) and a tumor-associated antigen (TAA). Examples of tumor antigens include, but are not limited to, CD20, CD38, CD123, ROR1, ROR2, BCMA, PSMA, SSTR2, SSTR5, CD19, FLT3, CD33, PSCA, ADAM 17, CEA, Her2, EGFR, EGFR-vIII, CD30, FOLR1, GD-2, CA-IX, Trop-2, CD70, CD38, mesothelin, EphA2, CD22, CD79b, GPNMB, CD56, CD138, CD52, CD74, CD30, CD123, RON, and ERBB2. Examples of TA-specific antibodies include, but are not limited to, Trastuzumab, Rituximab, Cetuximab, Bevacizumab, Panitumumab, Alemtuzumab, Matuzumab, Gemtuzumab, Polatuzumab, Inotuzumab, etc.

[0083] In the second Fab domain, the antibody Fv domain is based on the WuXiBody described in PCT / CN2018 / 106766 TM Fusion with TCR constant domain. Specifically, the heavy chain variable region (VH) in the heavy chain is fused with the TCR constant region, and the light chain variable region (VL) in the light chain is fused with another TCR constant region. For the engineered antibody of the present invention, the "antibody Fv" domain refers to the Fv domain of the above-mentioned antibody Fab. The first Fab domain and the second Fab domain of the engineered antibody of the present invention may have the same or different binding specificities.

[0084] In the second monomer, the TCR constant domain can be composed of a pair of Cα and Cβ regions of the TCR or a pair of Cγ and Cδ regions. In some embodiments, the TCR constant domain is a Cαβ constant domain. In some embodiments, the second monomer is a WuXiBody containing a TCR constant domain. TMAntibody half antibody. Although the TCR constant domain also has interchain disulfide bonds, their reduction reactivity is much lower, so it is more difficult to couple drugs than the interchain disulfide bonds of the antibody Fab domain. The present invention at least partially utilizes this difference in reduction reactivity, thereby achieving ADC production with high homogeneity and better DAR control.

[0085] In some embodiments, the antibody Fab domain is the Fab domain of an IgG1 antibody, i.e., the Fab domain of the IgG1 type. Preferably, the antibody Fab domain is the Fab domain of a human IgG1 antibody. In some embodiments, the Fv domain is the Fv domain of an IgG antibody, i.e., the Fv domain of the IgG class, preferably the Fv domain of human IgG.

[0086] The engineered antibody of the present invention characteristically contains an engineered hinge region, i.e., a "first engineered hinge region", in the first monomer, which is composed of a truncated IgG1 hinge region portion and a truncated IgG4 hinge region portion or is a modified IgG4 hinge region, so that the hinge structure formed by coupling the first engineered hinge region with the second hinge region contains at least two interchain disulfide bonds. The engineered hinge region is composed of natural amino acids, including cysteine ​​residues for forming at least two interchain disulfide bonds between heavy chains.

[0087] The wild-type hinge region of IgG1 includes 15 amino acids (e.g., EPKSCDKTHTCPPCP (SEQ ID NO: 9)) and is very flexible, while the hinge of IgG4 is shorter, with only 12 amino acids (same as above). Wild-type IgG1 and IgG4 differ by one amino acid in the core hinge region (EU numbering 226-229): Cys-Pro-Pro-Cys in IgG1 and Cys-Pro-Ser-Cys in IgG4. The IgG hinge with the S228P mutation can be represented by the sequence ESKYGPPCPPCP (SEQ ID NO: 10). The S228P mutation forms a polyproline helix (PPCPPCP) in the IgG4 hinge, which, combined with the shorter IgG4 hinge length, further limits its flexibility compared to the IgG1 hinge. The difference in flexibility between different hinges is of great significance for the bioconjugation of antibodies, because cysteine ​​residues located in flexible hinge fragments are considered to be more reactive than cysteine ​​residues located in rigid hinges. Experiments have shown that the heavy chain-light chain disulfide bonds and heavy chain-heavy chain disulfide bonds of S228P IgG4 are both weakly reactive.

[0088] The inventors surprisingly found that the engineered hinge region of the present invention provides an improvement in drug loading-antibody ratio (PAR, equivalent to DAR) during bioconjugation by taking advantage of the difference in accessibility of the interchain disulfide bonds of the hinge domain and the interchain disulfide bonds of the Fab domain to reducing agents. Specifically, as mentioned above, engineered antibodies with engineered hinge regions have multiple advantages for ADC production and ADC products themselves, such as high homogeneity, more controlled DAR, simplified production, and desirable pharmacokinetic and / or pharmacodynamic properties.

[0089] In some embodiments, the engineered hinge region comprises the sequence shown in the following formula (I):

[0090] EPKx1C x2 x3 x4 x5 x6 x7 x8 CPPCP(I)

[0091] Wherein, x1 = default or S, preferably S; x2 = default or E or S, preferably default; x3 = default or S; x4 = default or K or D; x5 = Y or K, preferably Y; x6 = G or T, preferably G; and / or x7x8 = PP, PT, HP or HT, preferably PP. In some embodiments, the engineered hinge region comprises a sequence as shown in the formula EPKx1C x2 x3 x4 x5 x6PPCPPPCP. In some embodiments, the engineered hinge region comprises a sequence as shown in the formula EPKSC x2 x3 x4 x5 x6PPCPPPCP.

[0092] Preferred examples of engineered hinge regions include:

[0093] EPKSCESKYGPPCPPCP(SEQ ID NO:1),

[0094] EPKSCSKYGPPCPPCP(SEQ ID NO:2),

[0095] EPKSCKYGPPCPPCP (SEQ ID NO: 3),

[0096] EPKSCYGPPCPPCP (SEQ ID NO: 4),

[0097] EPKCESKYGPPCPPCP(SEQ ID NO:5),

[0098] ESKYGHTCPPCP (SEQ ID NO: 6),

[0099] ESKYGHPCPPCP (SEQ ID NO: 7),

[0100] ESKYGPTCPPCP (SEQ ID NO:8); and

[0101] A sequence that is at least 85%, preferably at least 90% and more preferably at least 95% identical to any of the above sequences.

[0102] In some embodiments, the engineered hinge region may further include additional hinge segments (eg, upper hinge region segments) located on either or both sides of the hinge region.

[0103] For the engineered antibodies of the present invention, the term "hinge domain" refers to the portion of the coupled heavy chain where the two hinge regions are located. For example, as described in PCT / CN2018 / 106766, the first and second hinge regions in the two heavy chains are aligned with each other from the N-terminus. According to the present invention, the second hinge region in the second monomer may be natural, or may be an engineered TCR hinge region or an antibody hinge region or a hybrid thereof.

[0104] The engineered antibodies of the present invention comprise an Fc domain with a knob-in-hole mutation to promote heterodimerization. In some embodiments, the Fc domain is a human Fc domain. Preferably, the Fc domain is an IgG Fc domain, i.e., an Fc domain of the IgG class, more preferably an Fc domain of the IgG1 or IgG4 type. The Fc domain has a KIH mutation to achieve heterodimerization via the KIH mechanism.

[0105] The "Knobs-in-Holes" ("KIH") structure is a mechanism to ensure high heterodimerization. It refers to the fact that a steric effect can also be selectively produced by amino acid engineering to favor heterodimer formation and disfavor homodimer formation, which is sometimes also referred to as "knobs and holes". In some embodiments, the first Fc region, i.e., the Fc region in the first monomer comprising the antibody Fab domain, comprises a hole mutation, and the second Fc region, i.e., the Fc region in the second monomer comprising the second Fab domain replaced with a TCR constant domain, comprises a knob mutation. This further enhances heterodimerization and avoids mispairing between the two monomers, thereby providing the desired WuXiBody TM structure.

[0106] As an example of an embodiment, the engineered antibody of the present invention may comprise a set of four chains such as the following groups (a), (b) and (c):

[0107] (a) a first heavy chain comprising the amino acid sequence of SEQ ID NO: 11 or a sequence at least 85% identical thereto,

[0108] A first light chain comprising the amino acid sequence of SEQ ID NO: 15 or a sequence at least 85% identical thereto, and a second heavy chain comprising the amino acid sequence of SEQ ID NO: 14 or a sequence at least 85% identical thereto, and

[0109] a second light chain comprising the amino acid sequence of SEQ ID NO: 16 or a sequence at least 85% identical thereto;

[0110] (b) a first heavy chain comprising the amino acid sequence of SEQ ID NO: 12 or a sequence at least 85% identical thereto,

[0111] A first light chain comprising the amino acid sequence of SEQ ID NO: 15 or a sequence at least 85% identical thereto, and a second heavy chain comprising the amino acid sequence of SEQ ID NO: 14 or a sequence at least 85% identical thereto, and

[0112] a second light chain comprising the amino acid sequence of SEQ ID NO: 16 or a sequence at least 85% identical thereto;

[0113] or

[0114] (c) a first heavy chain comprising the amino acid sequence of SEQ ID NO: 13 or having a sequence similar to SEQ ID NO: 13

[0115] At least 85% identical sequence,

[0116] A first light chain comprising the amino acid sequence of SEQ ID NO: 15 or a sequence at least 85% identical thereto, and a second heavy chain comprising the amino acid sequence of SEQ ID NO: 14 or a sequence at least 85% identical thereto, and

[0117] a second light chain comprising the amino acid sequence of SEQ ID NO: 16 or a sequence at least 85% identical thereto;

[0118] The first heavy chain and the first light chain form the first monomer, and the second heavy chain and the second light chain form the second monomer.

[0119] On the other hand, the present invention provides a nucleic acid molecule or a combination of nucleic acid molecules encoding an engineered antibody of the present invention, wherein the nucleic acid molecule or the combination of nucleic acid molecules encodes: one or more sequences in SEQ ID NO: 11 and 14 to 16 or a sequence at least 85% identical thereto; one or more sequences in SEQ ID NO: 12 and 14 to 16 or a sequence at least 85% identical thereto; or one or more sequences in SEQ ID NO: 13 and 14 to 16 or a sequence at least 85% identical thereto. In some embodiments, the nucleic acid molecule may be one or more vectors, in particular expression vectors. It can be seen by those skilled in the art that nucleic acids encoding antibody heavy and light chains can be cloned in separate expression vectors and co-transfected into a host to recombinantly produce antibodies, or the coding sequences of these chains can be inserted into the same expression vector. Expression vectors and hosts known in the art can be used in the present invention. For example, but not limited to, plasmids, viral vectors, synthetic vectors, bacterial hosts, yeast, insect cells, and animal cells, such as CHO cells. In some embodiments, the nucleic acid molecule (eg, vector) or combination of nucleic acid molecules may be provided in the form of a kit, which may optionally contain instructions for using the nucleic acid molecule to recombinantly produce the antibody.

[0120] Antibody-drug conjugates

[0121] As one of the contents of the present invention, an antibody-drug conjugate comprising the engineered antibody of the present invention is provided herein, wherein the engineered antibody of the present invention is coupled to one or more drug molecules via a linker.

[0122] As described above, the engineered antibody of the present invention includes a first Fab domain and a second Fab domain, wherein the first Fab domain is an antibody Fab domain, and the second Fab domain comprises an Fv domain and a TCR constant domain fused thereto. The first Fab domain and the second Fab domain may have the same or different binding specificities, and the antibody Fab domain and the Fv domain may each be derived from any antibody, particularly those related antibodies used clinically, such as antibodies that specifically bind to tumor antigens (TA) such as tumor-specific antigens (TSA) and tumor-associated antigens (TAA).

[0123] The present invention has no particular limitation on the drugs that can be used (also referred to as "drug payload"). Drugs that can be used in the present invention include cytotoxic drugs, especially those used for cancer treatment. Such drugs include, but are not limited to, DNA damaging agents, DNA binding agents, antimetabolites, enzyme inhibitors (such as thymidylate synthase inhibitors and topoisomerase inhibitors), tubulin inhibitors and toxins (such as toxins of bacterial, fungal, plant or animal origin). Specific examples include, for example, paclitaxel, methotrexate, methotrexate, dichloromethotrexate, 5-fluorouracil, 6-mercaptopurine, cytarabine, melphalan, vinblastine, vinblastine, actinomycin, daunorubicin, doxorubicin, mitomycin C, mitomycin A, carminomycin, aminopterin, talithromycin, podophyllum and podophyllum derivatives such as etoposide or etoposide phosphate, vinblastine, vincristine, vindesine, taxanes including tretinoin, ... Taxol, docetaxel, retinoic acid, butyric acid, N8-acetyl spermidine, camptothecin, calicheamicin, esperamicin, ene-diyne, duocarmycin A, duocarmycin SA, calicheamicin, camptothecin, hemiesterin, maytansinoids (including DM1, DM2, DM3, DM4) and auristatins (including monomethyl auristatin E (MMAE), monomethyl auristatin F (MMAF), monomethyl auristatin D (MMAD)). In some embodiments, auristatins such as MMAE are preferred. The drug can be connected to the linker by any suitable method known in the art. In some embodiments, the drug participates in the coupling reaction in the form of an intermediate linker-drug payload compound, such as "MC-vc-PAB-MMAE".

[0124] The drug used in the present invention can be bound to the antibody through a linker. There are many linkers for ADC in the art. The present invention has no particular limitation on the linker that can be used, as long as it contains a portion that can react with the sulfhydryl group on the antibody and thus be connected to the antibody. Particularly suitable for the present invention are linkers functionalized with aminoimide (amleimido) or acyl halide. Examples include, but are not limited to -MC-vc-PAB- ("MC": maleimido-hexanoyl; "-vc-": "-Val-Cit-" dipeptide; "PAB": p-aminobenzyl), -MC-GGFG- ("-GGFG-": "-Gly-Gly-Phe-Gly-" tetrapeptide), -MC-vc-, -MC- and -SMCC- (succinimidyl-4-(N-maleimidomethyl) cyclohexane-1-carboxylate). In some embodiments, the linker is -MC-vc-PAB-.

[0125] In some embodiments, the linker-drug payload is connected to a cysteine ​​residue that participates in the reaction by opening a selected interchain disulfide bond by a reduction reaction. In some embodiments, the drug / antibody ratio (DAR) of the conjugate of the present invention can be about 2 to 6, preferably about 2 or about 6. The ratio can be an average value of a population of molecules, for example, the DAR average of an ADC population is DAR2 or DAR6. Moreover, in one of the preferred embodiments, the conjugate has two drug molecules, both connected to the antibody Fab domain, and the TCR constant domain has substantially no drug molecules connected thereto. In another preferred embodiment, the conjugate has 6 drug molecules, all connected to the antibody Fab domain and the hinge domain, and the TCR constant domain has substantially no drug molecules connected thereto.

[0126] On the other hand, the present invention provides a mixture comprising or consisting of an antibody-drug conjugate of the present invention, wherein at least about 65%, preferably at least about 70%, at least about 75%, at least about 80%, or at least about 90% of the DARs in the mixture are 2 (i.e., DAR2). Preferably, both drug molecules of the DAR2 conjugate are attached to the antibody Fab domain, and the TCR constant domain has substantially no drug molecules attached thereto. The present invention also provides a mixture comprising or consisting of an antibody-drug conjugate of the present invention, wherein at least about 80%, preferably at least about 85%, or at least about 90% of the antibody-drug conjugates have a drug / antibody ratio of 6 (i.e., DAR6). Preferably, the 6 drug molecules of the DAR6 conjugate are attached to the antibody Fab domain and the hinge domain, and the TCR constant domain has substantially no drug molecules attached thereto.

[0127] On the other hand, the present invention also provides a pharmaceutical composition comprising the above-mentioned antibody-drug conjugate or a mixture thereof and a pharmaceutically acceptable carrier.

[0128] Preparation of Antibody-Drug Conjugates

[0129] The ADC of the present invention can be prepared by any suitable method known in the art. In the present invention, the linker-drug load is coupled to a cysteine ​​residue released by a disulfide bond after reduction by a mild reducing agent. Specifically, the engineered hinge region of the present invention changes the reduction reactivity of the disulfide bond in the hinge domain, so that the partial reduction of the antibody by a mild reducing agent causes the disulfide bond to be selectively reduced. In addition, the introduction of a TCR constant domain in one of the Fab domains is more conducive to controlling the DAR of the ADC product. In general, the present invention uses an engineered hinge region and introduces a TCR constant domain to obtain a DAR-controlled, highly homogeneous ADC product, the ADC product comprising a conjugate of two molecules of linker-drug load and mainly connected to the antibody Fab domain, or a conjugate comprising six molecules of linker-drug load and mainly connected to the antibody Fab domain and the hinge domain, and the TCR constant domain is substantially free of linker-drug load connected thereto.

[0130] Provided herein are methods for preparing antibody-drug conjugates of the invention. Briefly, the methods may include a partial reduction reaction of an antibody and a coupling reaction of the partially reduced antibody with a linker-drug payload, such as Figure 1 As shown in the schematic diagram. Preferably, the coupling is carried out in a reducing buffer containing an organic solvent as an additive to promote the dissolution of the linker-drug payload. Specifically, the method may include coupling a partially reduced antibody of the present invention with a linker-drug payload compound having a maleimide or haloacetyl group by a Michael addition reaction. The partially reduced antibody can be produced by partially reducing the engineered antibody of the present invention with a mild reducing agent. In some embodiments, the method may include:

[0131] Partially reducing the engineered antibody of the invention with a mild reducing agent, and

[0132] The partially reduced antibody is coupled to a linker-drug payload compound bearing a maleimide or haloacetyl group via a Michael addition reaction.

[0133] In some embodiments, the mild reducing agent is TCEP or DTT. In some embodiments, the reducing agent / antibody ratio is about 1-20, preferably about 2-15, and more preferably about 3-10. In some embodiments, the partial reduction is carried out at a pH of about 4.0 to 8.0, preferably about 5 to 7. In some embodiments, the partial reduction is carried out for about 0.5 to 24 hours, preferably about 1 to 24 hours, about 1 to 16 hours, about 10 to 24 hours or about 16 to 24 hours. In some embodiments, the partial reduction is carried out at a temperature of about 4 to 37°C, preferably about 4 to 15°C or about 4 to 10°C. In some embodiments, the method provides a mixture comprising an antibody-drug conjugate of the present invention or a product consisting of a mixture of an antibody-drug conjugate of the present invention, wherein at least about 65%, preferably at least about 70%, at least about 75%, at least about 80% or at least about 90% of the DAR in the mixture is 2. Preferably, both drug molecules of the DAR2 conjugate are linked to the antibody Fab domain, and substantially no drug molecules are linked to the TCR constant domain.

[0134] In some other embodiments, the mild reducing agent is TCEP or DTT. In some embodiments, the reducing agent / antibody ratio is about 1-20, preferably about 3-20, about 3-10 or about 10-20. In some embodiments, the partial reduction is carried out under conditions of pH about 4.0 to 8.0, preferably about 6 to 8. In some embodiments, the partial reduction is carried out for about 0.5 to 24 hours, preferably about 3 to 24 hours, about 3 to 16 hours, about 10 to 24 hours or about 16 to 24 hours. In some embodiments, the partial reduction is carried out under conditions of a temperature of 4 to 37°C, preferably about 4 to 15°C or about 4 to 10°C. In some embodiments, the method provides a product comprising an antibody-drug conjugate mixture of the present invention or consisting of an antibody-drug conjugate mixture of the present invention, wherein at least about 80%, preferably at least about 85%, or at least about 90% of the antibody-drug conjugate has a drug / antibody ratio of 6 (i.e., DAR6). Preferably, the six drug molecules of the DAR6 conjugate are all linked to the antibody Fab domain and hinge domain, and there is substantially no drug molecule linked to the TCR constant domain.

[0135] In some embodiments, the coupling is carried out in a buffer having a pH of about 4.0 to 8.0, with an optional organic additive (e.g., an organic solvent or an organic cosolvent) in an amount of about 0.0% to 20.0% by weight, preferably about 5.0% to 15.0% by weight or about 10.0% to 15.0% by weight. In some embodiments, the drug / antibody ratio can be about 7-20, preferably about 7-10; the reaction temperature can be about 4 to 37° C., preferably about 4 to 20° C. or about 4 to 10° C.; and / or the reaction time can be about 0.5-4 hours, preferably about 1-3 hours.

[0136] treat

[0137] The antibody drug conjugates of the present invention can be used for subjects in need to treat their diseases, disorders or conditions, and the treatment includes giving the subject a therapeutically effective amount of antibody drug conjugates. The application also provides antibody drug conjugates of the present invention for subjects in need to treat their diseases, disorders or conditions. The diseases treated include, but are not limited to, cancers, including solid tumors and blood cancers. Examples of such cancers include, but are not limited to, breast cancer, gastric cancer, pancreatic cancer, liver cancer, lung cancer (e.g., NSCLC), head and neck cancer, colorectal cancer, B cell lymphoma (e.g., non-Hawkins lymphoma (NHL)) and leukemia.

[0138] abbreviation

[0139] ADC: Antibody-drug conjugate

[0140] CH: Heavy chain constant region

[0141] CMC: Chemical Composition Production and Control

[0142] DAR: drug / antibody ratio

[0143] DMA: N,N'-dimethylacetamide

[0144] DTT: 1,4-dithiothreitol

[0145] EGFR: epidermal growth factor receptor

[0146] Fab: antigen binding fragment

[0147] Fc: crystallizable fragment,

[0148] FDA: Food and Drug Administration

[0149] FGE: Formylglycine generating enzyme

[0150] HIC: Hydrophobic Interaction Chromatography

[0151] HPLC: High Performance Liquid Chromatography

[0152] IC50: half maximal inhibitory concentration

[0153] IgG: Immunoglobulin G

[0154] MC: Maleimide-hexanoyl

[0155] MED: Minimum effective dose

[0156] MMAE: Monomethyl auristatin E

[0157] MTD: Maximum tolerated dose

[0158] MWCO: Molecular Weight Cut-off

[0159] NaCl: Sodium Chloride

[0160] NNAA: Unnatural Amino Acid

[0161] mTG: microbial transglutaminase

[0162] PAB: p-aminobenzyl

[0163] PAR: drug load-antibody ratio

[0164] RP: Reverse

[0165] SEC: Size Exclusion Chromatography

[0166] TCEP: Tris(2-carboxyethyl)phosphine

[0167] VH: heavy chain variable region

[0168] eq: reducing agent / antibody molar ratio

[0169] Example

[0170] General Methods

[0171] Antibody preparation

[0172] All antibody molecules in this article were codon-optimized for gray hamsters (Cricetulus griseus), synthesized according to standard molecular biology methods and cloned into production vectors, and then amplified in large quantities by TOP10 Escherichia coli and obtained by plasmid extraction.

[0173] 72 hours before transfection, CHO K1 host cells were inoculated in CD CHO medium at 2-4E5 cells / mL. The cell density was calculated using a Vi-CELL counter, centrifuged at 290g for 7 minutes (min), and resuspended in pre-warmed fresh CD CHO medium before transfection. Before use, the resuspended host cells were incubated in a Kuhner shaker (36.5°C, 75% humidity, 6% CO2, 120rpm) for later use.

[0174] A total of 4 mg of plasmid encoding the antibody of interest was added to the host cell resuspension, followed by 12 mg of polyetherimide. The transfected culture was cultured in a Kuhner shaker at 36.5°C, 75% humidity, 6% CO2, 120 rpm for 4 hours. Self-feed medium was added, and the transfected culture was cultured in a Kuhner shaker at 31°C, 75% humidity, 6% CO2, 120 rpm for 9-10 days.

[0175] On the day of harvest, the transfected culture was clarified by centrifugation at 1,000 g for 10 minutes and then at 10,000 g for 40 minutes, and then sterilized by filtration through a 0.22 μm filter. The supernatant was titered and purified by ProA chromatography. The ProA eluate was neutralized by adding 1-2% neutralization buffer (1M Tris-HCl, pH 9.0) and then prepared in 20 mM histidine-acetate buffer at pH 5.5.

[0176] All proteins were subjected to quality control tests before coupling, including reducing and non-reducing SDS-PAGE, SEC-HPLC, endotoxin detection by LAL gel clot assay, and molecular identification by mass spectrometry.

[0177] Preparation of ADC

[0178] To a 1 mg / ml to 20 mg / ml antibody solution prepared in a buffer (pH 4.0-8.0, pH 6.0-8.0 is more suitable for preparing DAR6, such as histidine-acetate buffer), add 1 to 20 eq (preferably 2-15 eq for preparing DAR2, 3-20 eq for preparing DAR6) of a reducing agent (such as TCEP or DTT). Perform the reduction at 4-37°C with gentle shaking or stirring for 0.5 to 24 hours (preferably 1-24 hours for preparing DAR2, 3-24 hours for preparing DAR6). Without purification, add an organic cosolvent (such as DMA) to a concentration of 0% to 20% to the partially reduced antibody, and 7-20 eq of a maleimide- or haloacetyl-functionalized linker-drug payload. Perform the coupling reaction at 4-37°C with gentle shaking or stirring for 0.5 to 4 hours (preferably 1-24 hours for preparing DAR2, and 3-24 hours for preparing DAR6). Figure 1The final conjugated product characterization included UV-vis measurement of concentration, HIC-HPLC measurement of conjugate distribution and DAR, LC-MS measurement of drug loading on the light chain and heavy chain, RP-HPLC measurement of free drug residues, SEC-HPLC measurement of aggregation and purity, and kinetic turbidimetry measurement of endotoxin levels.

[0179] HIC-HPLC

[0180]

[0181] SEC-HPLC

[0182]

[0183]

[0184] RP-HPLC measurement of drug loading

[0185] Procedure: 20 μl of ADC sample was mixed with 75 μl of 8M guanidine hydrochloride and 5 μl of Tris-HCl, pH 8.0. 1 μl of 0.5M TCEP solution was added to the mixture. The mixture was reacted at 37°C for 30 minutes (min), and then the drug loading on the antibody was determined by RP-HPLC.

[0186]

[0187]

[0188] Determination of free drug by RP-HLPC

[0189] Procedure: 85 μl ADC solution was mixed with 15 μl DMA, and then the protein was precipitated with 100 μl precipitation buffer (37.5% v / v methanol / acetonitrile solution saturated with NaCl) and vortexed at 1400 rpm for 10 minutes (min) at 22°C.

[0190] The samples were centrifuged at 16,000 rpf for 10 minutes and the supernatant was subjected to RP-HPLC along with standard samples to determine free drug.

[0191]

[0192] LC-MS determination of DAR

[0193] Procedure: 85 μl of ADC solution was mixed with 15 μl of 50 mM TCEP and then incubated at 22°C for 30 minutes.

[0194] The samples were analyzed for DAR by LC-MS.

[0195]

[0196] The following examples are only used to illustrate the present invention and are not intended to limit the scope of the present invention.

[0197] Example 1

[0198] The anti-Her2 antibody 886-39 (IgG-TCR, IgG1-Fab, IgG4-Fc) was constructed using an engineered hinge region with the sequence EPKSCSKYGPPCPPCP (SEQ ID NO: 2). Fig.14 Also referred to as "WBP886-39" in the present invention. The first light chain (LC) sequence is SEQ ID NO: 15, the first heavy chain (HC) sequence is SEQ ID NO: 11, the second LC sequence is SEQ ID NO: 16, and the second HC sequence is SEQ ID NO: 14. The first LC and the first HC form a monomer having antibody Fab and Fc regions containing "hole" mutations, and the second LC and the second HC form a monomer having a TCR constant domain replaced in the Fab domain and a "knob" mutation in the Fc region. The antibody is recombinantly produced as described in the general method section.

[0199] The antibody was dissolved in 20 mM histidine buffer (150 mM NaCl, pH 5.5) at a concentration of 8.66 mg / ml. The formulation buffer was changed to PBS pH 7.0 and the concentration was 4 mg / ml. 2.0 eq of TCEP was added to the antibody solution, and the mixture was incubated at 4°C for 16 hours. DMA was added to the reduced antibody to a concentration of 10%, and then 7 eq of MC-vc-PAB-MMAE was added. The coupling reaction was carried out at 4°C for 1 hour. The coupling product was purified with a 40KD MWCO desalting column and stored in PBS at pH 7.0. The DAR and drug distribution were determined by HIC-HPLC, and the coupling position was determined by LC-MS, thereby characterizing the final product ( Figure 2 and Figure 3 ). The HIC-HPLC results of DAR and drug distribution are shown below:

[0200]

[0201] The results showed that the introduction of the TCR constant domain into the antibody and the engineering of the hinge domain unexpectedly improved the homogeneity of the ADC prepared using the antibody, and DAR2 was the main product. The product was highly homogeneous, and the percentage of the main product DAR2 type ADC reached 90.5%. In addition, LC-MS results showed that the drug molecules were mainly loaded in the antibody Fab domain, and there was no drug loading in the TCR constant domain.

[0202] Example 2

[0203] The anti-Her2 antibody 886-40 (IgG-TCR, IgG1-Fab, IgG4-Fc) was constructed using an engineered hinge region with the sequence EPKSCKYGPPCPPCP (SEQ ID NO: 3). Fig.14 Also referred to as "WBP886-40" in the present invention. The first light chain (LC) sequence is SEQ ID NO: 15, the first heavy chain (HC) sequence is SEQ ID NO: 12, the second LC sequence is SEQ ID NO: 16, and the second HC sequence is SEQ ID NO: 14. The first LC and the first HC form a monomer having antibody Fab and Fc regions containing "hole" mutations, and the second LC and the second HC form a monomer having a TCR constant domain replaced in the Fab domain and a "knob" mutation in the Fc region. The antibody is recombinantly produced as described in the general method section.

[0204] The antibody was dissolved in 20 mM histidine (150 mM NaCl, pH 5.5) at a concentration of 10.83 mg / ml. The formulation buffer was changed to PBS pH 7.0 and the concentration was 4 mg / ml. 2.0 eq of TCEP was added to the antibody solution, and the mixture was incubated at 4°C for 16 hours. DMA was added to the reduced antibody to a concentration of 10%, and then 7 eq of MC-vc-PAB-MMAE was added. The coupling reaction was carried out at 4°C for 1 hour. The coupling product was purified with a 40KD MWCO desalting column and stored in PBS at pH 7.0. The DAR and drug distribution were determined by HIC-HPLC, and the coupling position was determined by LC-MS, thereby characterizing the final product ( Figure 6 and Figure 7 ). The HIC-HPLC results of DAR and drug distribution are shown below:

[0205]

[0206] The results showed that the introduction of the TCR constant domain into the antibody and the engineering of the hinge domain unexpectedly improved the homogeneity of the ADC prepared using the antibody, and DAR2 was the main product. The product was highly homogeneous, with the percentage of the main product DAR2 type ADC reaching 73.2%. In addition, LC-MS results showed that the drug molecules were mainly loaded in the antibody Fab domain, and there was no drug loading in the TCR constant domain.

[0207] Example 3

[0208] The anti-Her2 antibody 886-41 (IgG-TCR, IgG1-Fab, IgG4-Fc) was constructed using an engineered hinge region with the sequence EPKSCYGPPCPPCP (SEQ ID NO: 4). Fig.14Also referred to as "WBP886-41" in the present invention. The first light chain (LC) sequence is SEQ ID NO: 15, the first heavy chain (HC) sequence is SEQ ID NO: 13, the second LC sequence is SEQ ID NO: 16, and the second HC sequence is SEQ ID NO: 14. The first LC and the first HC form a monomer having antibody Fab and Fc regions containing "hole" mutations, and the second LC and the second HC form a monomer having a TCR constant domain replaced in the Fab domain and a "knob" mutation in the Fc region. The antibody is recombinantly produced as described in the general method section.

[0209] The antibody was dissolved in 20 mM histidine (150 mM NaCl, pH 5.5) at a concentration of 4 mg / ml. 14.0 eq of TCEP was added to the antibody solution, and the mixture was incubated at 4°C for 16 hours. DMA was added to the reduced antibody to a concentration of 10%, and then 7 eq of MC-vc-PAB-MMAE was added. The coupling reaction was carried out at 4°C for 1 hour. The coupling product was purified using a 40KD MWCO desalting column and stored in a 20 mM histidine buffer at pH 5.5. The DAR and drug distribution were determined by HIC-HPLC, and the coupling position was determined by LC-MS, thereby characterizing the final product ( Fig.10 and Fig.11 ). The HIC-HPLC results of DAR and drug distribution are shown below:

[0210]

[0211]

[0212] The results showed that the introduction of the TCR constant domain into the antibody and the engineering of the hinge domain unexpectedly improved the homogeneity of the ADC prepared using the antibody, and DAR2 was the main product. The product was highly homogeneous, with the percentage of the main product DAR2 type ADC reaching 68.2%. In addition, LC-MS results showed that the drug molecules were mainly loaded in the antibody Fab domain, and there was no drug loading in the TCR constant domain.

[0213] Example 4

[0214] The anti-Her2 antibody 886-39 was dissolved in 20 mM histidine (150 mM NaCl, pH 5.5) at a concentration of 8.66 mg / ml. The formulation buffer was changed to PBS pH 8.0 and the concentration was 4 mg / ml. 12 eq of TCEP was added to the antibody solution, and the mixture was incubated at 4°C for 16 hours. DMA was added to the reduced antibody to a concentration of 10%, and then 13 eq of MC-vc-PAB-MMAE was added. The coupling reaction was carried out at 4°C for 1 hour. The coupling product was purified using a 40KD MWCO desalting column and stored in 20 mM histidine-acetate buffer at pH 5.5. The DAR and drug distribution were determined by HIC-HPLC, and the coupling position was determined by LC-MS, thereby characterizing the final product ( Figure 4 and Figure 5 ). The HIC-HPLC results of DAR and drug distribution are shown below:

[0215]

[0216] The results showed that the introduction of the TCR constant domain into the antibody and the engineering of the hinge domain unexpectedly improved the homogeneity of the ADC prepared using the antibody, and DAR6 was the main product. The product was highly homogeneous, with the percentage of the main product DAR6 type ADC reaching 87.1%. In addition, LC-MS results showed that the TCR constant domain had no drug loading.

[0217] Example 5

[0218] The anti-Her2 antibody 886-40 was dissolved in 20 mM histidine (150 mM NaCl, pH 5.5) at a concentration of 10.83 mg / ml. The formulation buffer was changed to PBS pH 8.0 and the concentration was 4 mg / ml. 14.0 eq of TCEP was added to the antibody solution, and the mixture was incubated at 4°C for 16 hours. DMA was added to the reduced antibody to a concentration of 10%, and then 17 eq of MC-vc-PAB-MMAE was added. The coupling reaction was carried out at 4°C for 1 hour. The coupling product was purified using a 40KD MWCO desalting column and stored in 20 mM histidine-acetate buffer at pH 5.5. The DAR and drug distribution were determined by HIC-HPLC, thereby characterizing the final product ( Figure 8 and Fig. 9 ). The HIC-HPLC results of DAR and drug distribution are shown below:

[0219]

[0220] The results showed that the introduction of the TCR constant domain into the antibody and the engineering of the hinge domain unexpectedly improved the homogeneity of the ADC prepared using the antibody, and DAR6 was the main product. The product was highly homogeneous, with the main product DAR6 material percentage reaching 84.4%. In addition, LC-MS results showed that the TCR constant domain had no drug loading.

[0221] Example 6

[0222] The anti-Her2 antibody 886-41 was dissolved in 20 mM histidine (150 mM NaCl, pH 5.5) at a concentration of 10.40 mg / ml. The formulation buffer was changed to PBS pH 8.0 and the concentration was 4 mg / ml. 16.0 eq of TCEP was added to the antibody solution, and the mixture was incubated at 4°C for 16 hours. DMA was added to the reduced antibody to a concentration of 10%, and then 19 eq of MC-vc-PAB-MMAE was added. The coupling reaction was carried out at 4°C for 1 hour. The coupling product was purified using a 40KD MWCO desalting column and stored in 20 mM histidine-acetate buffer at pH 5.5. The final product was characterized by measuring DAR and drug distribution using HIC-HPLC ( Fig.12 and Fig.13 ). The HIC-HPLC results of DAR and drug distribution are shown below:

[0223]

[0224] The results showed that the introduction of the TCR constant domain into the antibody and the engineering of the hinge domain unexpectedly improved the homogeneity of the ADC prepared using the antibody, and DAR6 was the main product. The product was highly homogeneous, with the percentage of the main product DAR6 type ADC reaching 87.5%. In addition, LC-MS results showed that the TCR constant domain had no drug loading.

Claims

1. An engineered dimeric antibody, wherein a first monomer comprises a first Fab domain and a first engineered hinge region operably linked thereto, and thereafter a first Fc region operably linked thereto, and a second monomer comprises a second Fab domain and a second hinge region operably linked thereto, and thereafter a second Fc region operably linked thereto; in, The first Fab domain is an antibody Fab domain, and the second Fab domain comprises an antibody Fv domain and a TCR constant domain fused thereto; and wherein the first engineered hinge region is composed of a truncated IgG1 hinge region portion and a truncated IgG4 hinge region portion or is a modified IgG4 hinge region, the hinge domain formed by coupling the first engineered hinge region and the second hinge region comprises at least two interchain disulfide bonds, and the engineered hinge region is composed of the sequence shown in SEQ ID NO: 3; and Wherein, the Fc domain formed by the coupling of the first Fc region and the second Fc region comprises a knob-to-hole mutation.

2. The engineered dimeric antibody according to claim 1, wherein The antibody Fab domain is of human IgG1 type; and / or the Fv domain is of human IgG class.

3. The engineered dimeric antibody of claim 1, wherein the Fc domain is of human IgG1 or human IgG4 type.

4. The engineered dimeric antibody according to claim 1, wherein The first Fc region comprises a hole mutation and the second Fc region comprises a knob mutation.

5. The engineered dimeric antibody according to claim 1, wherein The first heavy chain consists of the amino acid sequence SEQ ID NO: 12, The first light chain consists of the amino acid sequence SEQ ID NO: 15, The second heavy chain consists of the amino acid sequence SEQ ID NO: 14, The second light chain consists of the amino acid sequence of SEQ ID NO: 16; The first heavy chain and the first light chain form the first monomer, and the second heavy chain and the second light chain form the second monomer.

6. An antibody-drug conjugate, comprising the engineered dimeric antibody according to claim 1 coupled to one or more drug molecules via a linker.

7. The antibody-drug conjugate according to claim 6, characterized in that One or more of the following: Wherein, the linker is -MC-vc-PAB-; wherein the drug is MMAE; Wherein, the drug / antibody ratio of the conjugate is 2 or 6; Wherein, the conjugate has two drug molecules, both of which are connected to the antibody Fab domain of the engineered dimer antibody, and the TCR constant domain has no drug molecule connected thereto; and / or Wherein, the conjugate has six drug molecules, all of which are connected to the antibody Fab domain and the hinge domain of the engineered dimer antibody, and the TCR constant domain has no drug molecule connected to it.

8. A composition comprising or consisting of a mixture of the antibody-drug conjugates of claim 6, wherein: At least 65% of the antibody-drug conjugates have a drug / antibody ratio of 2; or wherein at least 80% of the antibody-drug conjugates have a drug / antibody ratio of 6.

9. A pharmaceutical composition comprising the antibody-drug conjugate according to claim 6 and a pharmaceutically acceptable carrier.

10. A method for preparing the antibody-drug conjugate of claim 6, comprising coupling the partially reduced antibody of claim 1 with a linker-drug payload compound having a maleimide or haloacetyl moiety via a Michael addition reaction.

11. The method of claim 10, further comprising partially reducing the engineered dimeric antibody of claim 1 with a mild reducing agent to provide the partially reduced antibody.

12. The method of claim 11, having one or more of the following features: The mild reducing agent is TCEP or DTT; The reducing agent / antibody ratio is 1-20; The partial reduction step is carried out at a pH of 4.0 to 8.0; The partial reduction step is performed for 0.5 to 24 hours; and / or The partial reduction step is performed at a temperature between 4 and 37°C.

13. The method of claim 12, having one or more of the following features: The reducing agent / antibody ratio is 2-15; and / or The partial reduction step is carried out for 1 to 24 hours.

14. The method of claim 12, having one or more of the following features: The reducing agent / antibody ratio is 3-20; The partial reduction step is performed at a pH of 6 to 8; and / or The partial reduction step is carried out for 3 to 24 hours.

15. An antibody-drug conjugate product obtained by the method of claim 12, comprising or consisting of a mixture of the antibody-drug conjugates of claim 6, wherein: At least 65% of the antibody-drug conjugates have a drug / antibody ratio of 2; or, the antibody-drug conjugate product comprises or consists of a mixture of the antibody-drug conjugates of claim 6, wherein at least 80% of the antibody-drug conjugates have a drug / antibody ratio of 6.

16. Use of an antibody-drug conjugate for the manufacture of a drug for treating breast cancer, wherein: The antibody-drug conjugate comprises the engineered dimeric antibody of claim 5 coupled to one or more drug molecules via a linker.

Images