Downstream processing of bispecific antibody constructs

Abstract

The present invention provides improved manufacturing methods for the production of bispecific antibody products comprising at least two binding domains, including at least one method comprising the step of refolding a construct, wherein one or more of the domains comprise a disulfide bond, and the construct is contacted with a refolding buffer that refolds the construct to its native form, the refolding buffer comprising (i) a redox and / or oxidizing agent having a concentration of at least 0.1 mM and (ii) a chaotropic agent having a concentration >1 M, and the pH of the refolding buffer corresponds to the pI value of the first domain + / - 4.5 or the pI value of the entire construct + / - 4.0.

Description

[Technical field]

[0001] The present invention relates to biotechnology methods, in particular downstream manufacturing processes for the production of bispecific antibody constructs. [Background technology]

[0002] Among the most rapidly and promisingly developed therapeutics in recent years are protein-based drugs, which have already played an important role in almost all medical fields and are among the fastest growing therapeutics in (pre)clinical development and on the market (Leader, Nature Reviews Drug Discovery 2008 Jan 7, 21-39). Compared with small molecule chemical drugs, protein drugs have high specificity and activity at relatively low concentrations, and usually provide treatment for diseases with high impact, such as various cancers, autoimmune diseases and metabolic disorders (Roberts, Trends Biotechnol. 2014 Jul; 32(7): 372-80, Wang, Int J Pharm. 1999 Aug 20; 185(2): 129-88).

[0003] With the advancement of commercial-scale purification processes, protein-based pharmaceuticals, such as recombinant proteins, can now be obtained with high purity at the time of initial production. However, proteins are only marginally stable and are highly susceptible to chemical and physical degradation even during upstream production. Chemical degradation refers to modifications involving covalent bonds, such as deamidation, oxidation, cleavage or formation of new disulfide bridges, hydrolysis, isomerization, or deglycosylation. Physical degradation includes protein unfolding, undesirable adsorption to surfaces, and aggregation. Addressing these physical and chemical instabilities is one of the most difficult challenges in the development of protein pharmaceuticals (Chi et al., Pharm Res, Vol. 20, No. 9, Sept 2003, pp. 1325-1336; Roberts, Trends Biotechnol. 2014 Jul; 32(7): 372-80).

[0004] Therefore, despite advances in manufacturing, novel protein-based pharmaceuticals require new and innovative manufacturing processes to avoid effects on product quality such as protein aggregation, which impacts upstream manufacturing, downstream production, storage and application.

[0005] Such novel protein-based pharmaceuticals include, for example, bispecific (monoclonal) antibodies. Bispecific antibodies are artificial proteins that can bind to two different types of antigens simultaneously. They are known in several structural forms and are currently being explored for applications in cancer immunotherapy and drug delivery (Fan, Gaowei; Wang, Zujian; Hao, Mingju; Li, Jinming (2015). "Bispecific antibodies and their applications". Journal of Hematology & Oncology. 8:130).

[0006] In general, bispecific antibodies may be IgG-like, i.e., full-length bispecific antibodies, or non-IgG-like bispecific antibodies that are not full-length antibody constructs. Full-length bispecific antibodies usually retain the normal monoclonal antibody (mAb) structure of two Fab arms and one Fc region, except that the two Fab sites bind different antigens. Non-full-length bispecific antibodies completely lack the Fc region. These include chemically linked Fabs consisting of only the Fab region, as well as various types of bivalent and trivalent single-chain variable fragments (scFv). There are also fusion proteins that mimic the variable domains of two antibodies. The most developed of these new formats is likely the bispecific T cell engager (BiTE®) (Yang, Fa; Wen, Weihong; Qin, Weijun (2016). "Bispecific Antibodies as a Development Platform for New Concepts and Treatment Strategies". International Journal of Molecular Sciences. 18(1):48).

[0007] Bispecific molecules, such as BiTE® antibody constructs, are recombinant protein constructs composed of two flexibly linked antibody-derived binding domains. One binding domain of the BiTE® antibody construct is specific for a selected tumor-associated surface antigen on a target cell; the second binding domain is specific for CD3, a subunit of the T cell receptor complex on a T cell. Due to their special design, BiTE® antibody constructs are uniquely suited to transiently link T cells to target cells and simultaneously potently activate the intrinsic cytolytic potential of T cells against the target cells. An important further development of the first generation BiTE® antibody constructs deployed in the clinic as AMG 103 and AMG 110 (see WO 99 / 54440 and WO 2005 / 040220) provided bispecific antibody constructs that bind to context-independent epitopes at the N-terminus of the CD3 epsilon chain (WO 2008 / 119567). BiTE antibody constructs binding to this selected epitope not only do not show cross-species specificity for human and marmoset (Callithrix jacchus), cotton-top tamarin (Saguinus oedipus) or squirrel monkey (Saimiri sciureus) CD3ε chains, but also do not non-specifically activate T cells to the same extent as observed with previous generation T cell engaging antibodies, because they recognize this specific epitope instead of the epitope of the CD3-binding molecule previously described in the bispecific T cell engaging molecule. This reduced T cell activation is associated with less or reduced T cell redistribution in patients, which was deemed to be at risk for side effects.

[0008] Currently, about 25% (m / m) or more of bispecific antibody constructs such as bispecific T cell engager (BiTE®) molecules conjugated to single chain Fc (ScFc-BiTE®) produced by industrial mammalian CHO cell cultures (expressed, secreted, harvested from bioreactors and purified by, for example, Protein A columns) can be obtained in aggregated form, reducing yields and increasing costs of production. The aggregates are removed by the next downstream purification step, which is usually cation exchange chromatography (CEX). Thus, there is a need to reduce downstream aggregation.

[0009] Redox refolding has traditionally been applied after the production of aggregated proteins in inclusion bodies by microbial systems (E. coli). High concentrations of denaturants (6M guanidine, urea, etc.) are usually required in those cases. Oxidants (copper, dissolved oxygen, etc.) are used to form disulfide bonds, or redox reagents (cysteine / cystine, cysteine / cystamine, etc.) are used to reorganize disordered disulfide bonds. In general, industrial mammalian cells are expected to express proteins with the desired correct conformation (folding) and disulfide bonding properties, but in practice this is not always the case.

[0010] Several cases have been described in the literature, where proteins and antibodies produced in mammalian cells were further treated with the aim of refolding the proteins and creating or rearranging disulfide bonds: TNF-Fc fusion protein (etanercept) was oxido-reduction treated with cysteine / cystine to form and rearrange disordered disulfides to minimize the later eluting HIC fraction (Sassenfeld, HM, Remmele, RL, & McCoy, RE 2007, Increased recovery of active proteins, U.S. Pat. No. 7,157,557; Sassenfeld, HM, Remmele, RL, & McCoy, RE 2009, Increased recovery of active proteins (antibodies), U.S. Pat. No. 7,544,874 B2). In another case, a high percentage (16%) of open disulfides was observed in the Fab region of an IgG1 antibody (Chaderjian, WB, Chin, ET, Harris, RJ, & Etcheverry, TM 2005. Effect of copper sulfate on performance of a serum-free CHO cell culture process and the level of free thiol in the recombinant antibody expressed. Biotechnol. Prog., 21, (2) 550-553). Disulfide bond formation was completed by oxidation by adding 50-100 uM copper sulfate to the bioreactor (Chaderjian, Chin, Harris, & Etcheverry 2005).IgG2 antibodies were subjected to redox refolding to enrich for the desired IgG2A or IgG2B disulfide isoforms (Dillon, T.M., Speed-Ricci, M., Vezina, C., Flynn, G.C., Liu, Y.D., Rehder, D.S., Plant, M., Henkle, B., Li, Y., Varnum, B., Wypych, J., Balland, A., & Bondarenko, P.V. 2008. Structural and functional characterization of disulfide isoforms of the human IgG2 subclass. J. Biol. Chem., 283, 16206-16215; Dillon, T.M., Rehder, D.S., Bondarenko, P.V., Ricci, M., Gadgil, H.S., Banks, D., Zhou, J., Lu, Y., Goetze, A., & Zhang, Y. 2011. Methods for refolding of recombinant antibodies. U.S. Patent No. 7,928,205. Redox refolding and redox treatment to enrich for the desired IgG2 disulfide isoforms were also performed on IgG2 molecules immobilized on a Protein A column (Zhou, J. & Hong, T. 2006, Methods for refolding polypeptides, WO 2006 / 060083). Weaker denaturing conditions, including 1M urea, were used for single chain Fv antibodies (Chen, LH, Huang, Q., Wan, L., Zeng, LY, Li, SF, Li, YP, Lu, XF, & Cheng, JQ 2006. Expression, purification, and in vitro refolding of a humanized single-chain Fv antibody against human CTLA4 (CD152). Protein Expr. Purif., 46, (2) 495-502).However, applicable refolding conditions for refolding aggregated bispecific antibody constructs, such as BiTE® antibody constructs, have not been previously described that allow for rapid, versatile and resource-efficient steps that can be included in a routine downstream sequence to increase overall product yield. Furthermore, larger bispecific antibody constructs, for example those that contain half-life-extending domains and are typically at least 100 kDa in size, become more difficult to handle in downstream processing. For example, single-chain Fc (ScFc) BiTE® antibody constructs are usually large in size (about 105 kDa) and contain long linkers designed by protein engineering efforts. In comparison, there are very few single-chain native mammalian proteins (less than 5%) that are larger than 100 kDa. In this regard, those skilled in the art believe that a standard IgG antibody is composed of two heavy chains of about 50 kDa and two light chains of about 25 kDa that are largely folded separately and then linked together. In comparison, single chain antibody constructs such as the scFc BiTE® antibody construct fold as only one heavily engineered chain, and thus previous knowledge of downstream processing needs to be further refined for heavily engineered single chain antibody constructs, even those of unnaturally high molecular weight. [Prior art documents] [Patent documents]

[0011] [Patent Document 1] WO 99 / 54440 [Patent Document 2] International Publication No. 2005 / 040220 [Patent Document 3] International Publication No. 2008 / 119567 [Patent Document 4] U.S. Patent No. 7,157,557 [Patent Document 5] U.S. Patent No. 7,544,874 [Patent Document 6] U.S. Patent No. 7,928,205 [Patent Document 7] International Publication No. 2006 / 060083 [Non-patent literature]

[0012] [Non-Patent Document 1] Leader,Nature Reviews Drug Discovery 2008 Jan 7,21-39 [Non-Patent Document 2] Roberts,Trends Biotechnol.2014 Jul;32(7):372-80,Wang,Int J Pharm.1999 Aug 20;185(2):129-88 [Non-Patent Document 3] Chi et al.,Pharm Res,Vol.20,No.9,Sept 2003,pp.1325-1336,Roberts,Trends Biotechnol.2014 Jul;32(7):372-80 [Non-Patent Document 4] Fan,Gaowei;Wang,Zujian;Hao,Mingju;Li,Jinming(2015).“Bispecific antibodies and their applications”.Journal of Hematology & Oncology.8:130 [Non-Patent Document 5] Yang, Fa; Wen, Weihong; Qin, Weijun (2016). “Bispecific Antibodies as a Development Platform for New Concepts and Treatment Strategies”. International Journal of Molecular Sciences. 18(1):48 [Non-Patent Document 6] Chaderjian,WB,Chin,ET,Harris,RJ,& Etcheverry,TM2005.Effect of copper sulfate on performance of a serum-free CHO cell culture process and the level of free thiol in the recombinant antibody expressed.Biotechnol.Prog.,21,(2)550-553 [Non-Patent Document 7] Dillon, TM, Speed-Ricci, M., Vezina, C., Flynn, GC, Liu, YD, Rehder, DS, Plant, M., Henkle, B., Li, Y., Varnum, B., Wypych, J., Balland, A., & Bondarenko, PV2008. [Non-Patent Document 8] Structural and functional characterization of disulfide isoforms of the human IgG2 subclass.J.Biol.Chem.,283,16206-16215 [Non-Patent Document 9] Chen,LH,Huang,Q.,Wan,L.,Zeng,LY,Li,SF,Li,YP,Lu,XF,& Cheng,JQ2006.Expression, purification, and in vitro refolding of a humanized single-chain Fv antibody against human CTLA4(CD152).Protein Expr.Purif.,46,(2)495-502 Summary of the Invention [Means for solving the problem]

[0013] Surprisingly, it was possible to provide a downstream method that ensures improved quality of the bispecific antibody construct product, thus increasing product yield, and that can be integrated into existing downstream processing steps.

[0014] Thus, in one aspect, there is provided a bispecific antibody construct, the construct comprising: A first domain having a pI of 5.0 to 9.5 that binds to a target, preferably a tumor target; a second domain that binds to an extracellular epitope of the human and macaque CD3ε chain; and a downstream method of purifying a bispecific antibody construct, comprising two polypeptide monomers, each comprising a hinge, a CH2 domain and a CH3 domain, the two polypeptide monomers comprising a third domain fused to each other via a peptide linker, the construct comprising at least one open disulfide bridge, the method comprising refolding the construct, one or more of the domains comprising a disulfide bond; and It is envisaged in accordance with the present invention to provide a method wherein the construct is contacted with a refolding buffer that refolds the construct into its native form, the refolding buffer comprising (i) a redox and / or oxidizing agent having a concentration of at least 0.1 mM, and (ii) a chaotropic agent having a concentration of >1 M, and the pH of the refolding buffer corresponds to + / - 4.5 of the pI value of the first domain or a pI value of 4.0 of the entire construct.

[0015] According to the above aspect, it is also envisaged that the oxidizing agent is selected from the group consisting of copper(II), dissolved oxygen and (L)-dehydroascorbic acid (DHA), preferably copper(II).

[0016] According to the above aspect, it is also envisaged that the oxidizing agent is copper (II) sulfate.

[0017] According to the above embodiment, it is also envisaged that the oxidizing agent is present in a concentration of 0.1 to 10 mM, preferably 0.1 to 1 mM.

[0018] According to the above aspect, it is also envisaged that the redox agent is selected from the group consisting of glutathione-reduced / glutathione-oxidized, cysteine / cystine, cysteine / cysteamine, cysteine / cystamine, dithiothreitol (DTT), beta-mercaptoethanol and glutathione.

[0019] According to the above embodiment, it is also envisaged that the redox agent is present in a concentration of 0.1 to 10 mM, preferably 0.1 to 1 mM.

[0020] According to the above aspect, it is also envisaged that the refolding buffer comprises an oxidizing agent but does not comprise a redox agent.

[0021] According to said embodiment, it is also envisaged that the chaotrope is selected from the group consisting of arginine, urea, dimethylurea, methylurea, ethylurea, organic solvents, detergents, elevated temperature (preferably above 40, 50, 60, 70 or 80° C.) and guanidine (or its protonated form guanidinium), preferably guanidinium.

[0022] According to the above embodiment, it is also contemplated that the chaotropic agent is present at a concentration in the range of 1.2-2M.

[0023] According to said embodiment, it is also envisaged that the pH value is at least 4.9, preferably in the range of pH 5.0-9.5, preferably about 5.0-6.0 or 6.5-9.0, more preferably 5.0 or 7.0-8.0.

[0024] According to said aspect, it is also envisioned that the first domain binds to CD33, CDH19, MSLN, FLT3, BCMA, CD19, MUC17, EpCAM, EGFRviii, DLL3, CLDN 18, CDH3, and / or PSMA.

[0025] According to said embodiment, it is also envisaged that the monomer content of the construct after application of the refolding buffer (percent monomer peak), as determined, for example, by size exclusion chromatography, is increased by at least 25% compared to the monomer content before redox refolding, preferably reaching an amount of at least 40%, more preferably at least 60%, 65%, at least 70% or at least 85% monomer content.

[0026] According to the above embodiment, it is also envisaged that the method is carried out at room temperature (about 25° C.).

[0027] According to said aspect, it is also envisaged that the method has an incubation time of 1 to 24 hours, preferably 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10 hours, more preferably 4 hours.

[0028] According to the above aspect, it is also envisaged that contacting the construct with a refolding buffer occurs (i) in the process liquid (pool) in a downstream purification step or (ii) on a chromatography column in a downstream purification step.

[0029] According to the above aspect, it is also contemplated that (i) the construct concentration in the liquid pool is 0.5-15 mg / ml, preferably 0.7-12 mg / ml, or (ii) 5-15, preferably 10-15 mg / ml on the chromatography column.

[0030] According to the above aspects, it is also envisaged that the liquid pool is harvested cell culture fluid (HCCF), a chromatography eluate pool, a Protein A, Protein L or Protein G affinity chromatography eluate pool, a filtered viral inactivation pool, a cation exchange chromatography (CEX) monomer pool and a CEX post high molecular weight peak pool.

[0031] According to said aspect, the method of the present invention preferably comprises: (a) providing a harvested cell culture fluid (HCCF) containing a bispecific antibody construct secreted by mammalian cells; (b) contacting the HCCF with a first separation matrix for chromatographic purification under conditions suitable for the construct to associate with the separation matrix, the separation matrix being an affinity resin selected from the group consisting of Protein A, Protein G, Protein L and synthetic mimetic affinity resins; (c) washing the first separation matrix; (d) eluting the construct from the first separation matrix; (e) inactivating viruses in the eluate containing the construct; (f) contacting the eluate containing the constructs with a second separation matrix for chromatographic polishing under conditions suitable for the constructs associated with the separation matrix, the separation matrix being a cation exchange chromatography matrix; (g) washing the second separation matrix; (h) eluting the constructs from the second separation matrix, where the constructs are separated into two fractions: (i) a monomer pool comprising construct monomers and (ii) a high molecular weight pool comprising construct aggregates; (i) separating the virus from the fraction (i) containing the eluate by filtration; and (j) optionally subjecting the eluate to a further ultrafiltration and / or diafiltration step; It is also envisioned that the construct is contacted with a refolding buffer that refolds the construct into its native form in the cell culture medium prior to harvesting or in the HCCF of step (a).

[0032] According to said embodiment, instead, the construct is contacted with a refolding buffer that refolds the construct into its native form on the separation matrix (on the column) in at least one of steps (b) or (f) as described above.

[0033] According to said embodiment, instead, the construct is contacted with a refolding buffer that refolds the construct into its native form in the process liquid (pool) in at least one of steps (d), (e) or (h) as described above.

[0034] According to said aspect, the first binding domain of the construct comprises: (a) CDR-H1 as set forth in SEQ ID NO: 4, CDR-H2 as set forth in SEQ ID NO: 5, CDR-H3 as set forth in SEQ ID NO: 6, CDR-L1 as set forth in SEQ ID NO: 1, CDR-L2 as set forth in SEQ ID NO: 2 and CDR-L3 as set forth in SEQ ID NO: 3; (b) CDR-H1 as set forth in SEQ ID NO:29, CDR-H2 as set forth in SEQ ID NO:30, CDR-H3 as set forth in SEQ ID NO:31, CDR-L1 as set forth in SEQ ID NO:34, CDR-L2 as set forth in SEQ ID NO:35 and CDR-L3 as set forth in SEQ ID NO:36; (c) CDR-H1 as set forth in SEQ ID NO: 42, CDR-H2 as set forth in SEQ ID NO: 43, CDR-H3 as set forth in SEQ ID NO: 44, CDR-L1 as set forth in SEQ ID NO: 45, CDR-L2 as set forth in SEQ ID NO: 46 and CDR-L3 as set forth in SEQ ID NO: 47; (d) CDR-H1 as set forth in SEQ ID NO: 53, CDR-H2 as set forth in SEQ ID NO: 54, CDR-H3 as set forth in SEQ ID NO: 55, CDR-L1 as set forth in SEQ ID NO: 56, CDR-L2 as set forth in SEQ ID NO: 57 and CDR-L3 as set forth in SEQ ID NO: 58; (e) CDR-H1 as set forth in SEQ ID NO: 65, CDR-H2 as set forth in SEQ ID NO: 66, CDR-H3 as set forth in SEQ ID NO: 67, CDR-L1 as set forth in SEQ ID NO: 68, CDR-L2 as set forth in SEQ ID NO: 69 and CDR-L3 as set forth in SEQ ID NO: 70; (f) CDR-H1 as set forth in SEQ ID NO: 83, CDR-H2 as set forth in SEQ ID NO: 84, CDR-H3 as set forth in SEQ ID NO: 85, CDR-L1 as set forth in SEQ ID NO: 86, CDR-L2 as set forth in SEQ ID NO: 87 and CDR-L3 as set forth in SEQ ID NO: 88; (g) CDR-H1 as set forth in SEQ ID NO: 94, CDR-H2 as set forth in SEQ ID NO: 95, CDR-H3 as set forth in SEQ ID NO: 96, CDR-L1 as set forth in SEQ ID NO: 97, CDR-L2 as set forth in SEQ ID NO: 98 and CDR-L3 as set forth in SEQ ID NO: 99; (h) CDR-H1 as set forth in SEQ ID NO: 105, CDR-H2 as set forth in SEQ ID NO: 106, CDR-H3 as set forth in SEQ ID NO: 107, CDR-L1 as set forth in SEQ ID NO: 109, CDR-L2 as set forth in SEQ ID NO: 110 and CDR-L3 as set forth in SEQ ID NO: 111; (i) CDR-H1 as set forth in SEQ ID NO: 115, CDR-H2 as set forth in SEQ ID NO: 116, CDR-H3 as set forth in SEQ ID NO: 117, CDR-L1 as set forth in SEQ ID NO: 118, CDR-L2 as set forth in SEQ ID NO: 119, and CDR-L3 as set forth in SEQ ID NO: 120; (j) CDR-H1 as set forth in SEQ ID NO: 126, CDR-H2 as set forth in SEQ ID NO: 127, CDR-H3 as set forth in SEQ ID NO: 128, CDR-L1 as set forth in SEQ ID NO: 129, CDR-L2 as set forth in SEQ ID NO: 130 and CDR-L3 as set forth in SEQ ID NO: 131; (k) CDR-H1 as set forth in SEQ ID NO: 137, CDR-H2 as set forth in SEQ ID NO: 138, CDR-H3 as set forth in SEQ ID NO: 139, CDR-L1 as set forth in SEQ ID NO: 140, CDR-L2 as set forth in SEQ ID NO: 141 and CDR-L3 as set forth in SEQ ID NO: 142; (l) CDR-H1 as set forth in SEQ ID NO: 152, CDR-H2 as set forth in SEQ ID NO: 153, CDR-H3 as set forth in SEQ ID NO: 154, CDR-L1 as set forth in SEQ ID NO: 155, CDR-L2 as set forth in SEQ ID NO: 156 and CDR-L3 as set forth in SEQ ID NO: 157; (m) CDR-H1 as set forth in SEQ ID NO: 167, CDR-H2 as set forth in SEQ ID NO: 168, CDR-H3 as set forth in SEQ ID NO: 169, CDR-L1 as set forth in SEQ ID NO: 170, CDR-L2 as set forth in SEQ ID NO: 171 and CDR-L3 as set forth in SEQ ID NO: 172; (n) CDR-H1 as set forth in SEQ ID NO: 203, CDR-H2 as set forth in SEQ ID NO: 204, CDR-H3 as set forth in SEQ ID NO: 205, CDR-L1 as set forth in SEQ ID NO: 206, CDR-L2 as set forth in SEQ ID NO: 207 and CDR-L3 as set forth in SEQ ID NO: 208; (o) CDR-H1 as set forth in SEQ ID NO: 214, CDR-H2 as set forth in SEQ ID NO: 215, CDR-H3 as set forth in SEQ ID NO: 216, CDR-L1 as set forth in SEQ ID NO: 217, CDR-L2 as set forth in SEQ ID NO: 218 and CDR-L3 as set forth in SEQ ID NO: 219; (p) CDR-H1 as set forth in SEQ ID NO: 226, CDR-H2 as set forth in SEQ ID NO: 227, CDR-H3 as set forth in SEQ ID NO: 228, CDR-L1 as set forth in SEQ ID NO: 229, CDR-L2 as set forth in SEQ ID NO: 230, and CDR-L3 as set forth in SEQ ID NO: 231; and (q) CDR-H1 as set forth in SEQ ID NO: 238, CDR-H2 as set forth in SEQ ID NO: 239, CDR-H3 as set forth in SEQ ID NO: 240, CDR-L1 as set forth in SEQ ID NO: 241, CDR-L2 as set forth in SEQ ID NO: 242, and CDR-L3 as set forth in SEQ ID NO: 243. It is also envisaged that the antibody comprises a VH region comprising CDR-H1, CDR-H2 and CDR-H3, and a VL region comprising CDR-L1, CDR-L2 and CDR-L3 selected from the group consisting of:

[0035] According to said aspect, it is also envisaged that the method is applied to constructs produced by upstream continuous manufacturing processes.

[0036] According to another aspect, an apparatus is also envisioned for carrying out the methods of the present invention.

[0037] According to another aspect, it is also envisaged that there is provided a bispecific antibody construct purified by the method of the present invention. [Brief description of the drawings]

[0038] [Figure 1] Figure 1 shows an example of a downstream purification method step configuration for purifying a bispecific antibody construct according to the invention. The names of the generated materials correspond to the process liquids (pools) after each downstream purification step, samples of which are collected for size exclusion chromatography (SEC) analysis. In addition to the percentage of monomers determined by SEC, the remaining percentage of the material is aggregates. Low molecular weight (LMW) species only make up a very small proportion. Protein A filtered virus inactivated pool (FVIP) and cation exchange chromatography (CEX) HMW pool samples were used for redox refolding. [Diagram 2] FIG. 2 shows a size-exclusion chromatogram of aggregated exemplary CD33xCD3 bispecific antibody construct CEX HMW (post-peak) pool following treatment with guanidinium only (GuHCl) at a pH of 8. [Diagram 3] FIG. 3 shows size-exclusion chromatograms of aggregated exemplary CD33xCD3 bispecific antibody construct HMW (post-peak) pools following treatment with GuHCl and copper(II) or dehydroascorbic acid. [Figure 4] FIG. 4 shows a size exclusion chromatogram of an aggregated exemplary CD33xCD3 bispecific antibody construct Protein A FVIP pool after treatment with GuHCl. [Diagram 5] FIG. 5 shows a size-exclusion chromatogram of aggregated exemplary CD19xCD3 bispecific antibody construct CEX HMW (post-peak) pool following treatment with GuHCl and copper(II). [Figure 6] FIG. 6 shows a size-exclusion chromatogram of aggregated exemplary CLDNxCD3 bispecific antibody construct CEX HMW (post-peak) pool after treatment with GuHCl and copper(II). DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0039] Described herein are (preferably downstream) methods that provide optimized redox refolding conditions for bispecific antibody constructs, preferably single chain antibody constructs such as bispecific T cell engager (BiTE®) antibody constructs, in particular scFc BiTE® antibody constructs, which advantageously improves the quality of the antibody construct product, including reducing aggregation, increasing the yield of monomeric antibody constructs, and / or reducing the occurrence of isoforms, thereby improving the overall process economics. The methods can be versatile added to several standard downstream processing steps such as chromatographic purification, viral inactivation, cation ion exchange chromatography, viral filtration and / or ultrafiltration / diafiltration. Certain redox refolding conditions have surprisingly been found to be effective in reducing aggregation of bispecific antibody constructs and increasing the yield of their monomeric product, including higher than might be expected from the literature, for example, chaotropic agent concentrations of guanidine, also referred to herein as protonated form guanidinium, and low concentrations of redox agents, preferably oxidizing agents such as copper(II) and / or low concentrations of redox reagents such as cysteine / cystine. In certain embodiments, oxidizing agents are preferred.

[0040] It has been found that usually, a high percentage, such as more than 25% or even more than 50 or 70%, of bispecific antibody constructs such as BiTE® antibody constructs conjugated to single-chain Fc (ScFc-BiTE®) produced by industrial mammalian CHO cells (expressed, secreted, harvested from a bioreactor and purified by a Protein A column) is disadvantageously obtained in downstream aggregated form, which significantly reduces the yield and increases the cost of production. The aggregates are usually removed by the next downstream purification step, which is usually cation exchange chromatography (CEX). However, the redox refolding method as an additional downstream processing step in the method of bispecific antibody construct (e.g., ScFc-BiTE®) production by mammalian cells (or microbial systems), including the redox refolding step as described herein, increases the yield of the desired monomeric protein product (bispecific antibody construct) and reduces the aggregates. A wide range of refolding conditions, with or without reducing / oxidizing (redox) reagents, have been evaluated for bispecific antibody constructs.

[0041] Described herein is a (preferably downstream) method of processing an upstream produced antibody construct, preferably a bispecific antibody construct such as an ScFc-BiTE® antibody construct, comprising contacting the protein produced by a mammalian cell or microbial system with a chaotropic agent and a redox agent at a preferred pH that depends on the pI of the bispecific antibody construct. In one embodiment, the preferred pH depends on the pI of the target domain of the bispecific antibody construct and is typically in the range of pH 5-8. Such pH values ​​correspond to typical pH values ​​of the pooling environment or column conditions in downstream processing of the bispecific antibody construct. Several chaotrope (denaturing, unfolding) agents were evaluated, including guanidine, arginine, and increased temperature. Guanidine and arginine were preferred as chaotropic agents, with guanidine being the most preferred. Exemplary high concentrations of guanidine are, for example, in the range of 0.5-2.5 M, with a significant increase in monomer content being observed at concentrations of chaotropic agent above 1 M, preferably in the range of 1.2-2 M. Higher concentrations of chaotropic agent usually carry the risk of increasing the chance of product aggregation. In contrast, amounts below 1 M usually do not show significant refolding effects.

[0042] According to the present invention, a chaotropic agent such as guanidinium, especially when combined with a redox agent such as copper II, surprisingly reduces the occurrence of downstream bispecific antibody construct aggregates, and preferably also reduces the occurrence of isoforms and / or heterogeneous antibody constructs. This is usually evidenced by fewer shoulder peaks in size exclusion chromatography, which are understood as deviations from the typical Gaussian-like shape of the peak, e.g., by partial broadening or tailing of the peak. Similarly, according to the present invention, the occurrence of non-aggregated, i.e., monomeric, bispecific antibody constructs can be significantly increased in downstream processing resulting in an overall more efficient manufacturing process of the antibody constructs according to the present invention. Usually, the monomer content, which represents the proportion of the desired product, is increased by at least 25%, preferably at least 40, 50, 60, 70, 80, 90, 100% or more than 100% compared to the monomer percentage before applying the present invention. For example, the monomer content (monomer peak per size exclusion chromatography) is increased from about 25 to about 50%, and an increase in monomer content of about 100% has been achieved compared to the monomer content before applying the present invention. The actual final monomer content (percentage) that can be achieved may depend on the particular molecule. However, a percentage increase in monomer content of at least 25%, typically at least about 50%, is usually observed for the antibody construct as a whole as described herein. In this regard, it is preferred that a redox reagent is added to the refolding buffer along with a chaotropic agent such as guanidine to increase the percentage of monomer. When cysteine ​​and cystine form redox reagents in a refolding buffer, a buffer having a total redox reagent concentration in the range of at least 1 M, e.g., 1-2.5 M, preferably 1.2-2 M, e.g., 6:1, 14:1, and 33:1 (Cys:Cyss) molar ratios, the percentage of monomer increased can be from 39% (no redox reagent) to 52%, 61%, and 69%, respectively, i.e., up to about 76%.When using an oxidizing agent such as copper(II) sulfate or dehydroascorbic acid, e.g., 0.5 mM copper(II) sulfate, together with an oxidizing agent such as copper(II) sulfate or dehydroascorbic acid, e.g., 1.2 M to 2.0 M guanidine, a surprisingly high refolding is observed resulting in an increase in monomer content of up to about 90%. The percentage of monomer is also increased by chaotropic agents in the range of about 1.2-2 M, e.g., 1.2 M to 2.0 M guanidine, together with a concentration of dehydroascorbic acid in the range of 0.1-1 M, preferably 0.5, up to about 73%. However, the refolding procedure described herein, e.g., for scFc-BiTE® CEX post-peak pool, is also suitable for earlier downstream purification. Such earlier downstream product streams to be efficiently purified may include Protein A pools, where a Protein A FVIP sample contains, for example, approximately 73% monomer and 27% aggregated antibody constructs. For example, when subjected to refolding according to the present invention with 2.0 M guanidine, the percentage of monomer can be increased up to about 94% at a pH ranging from about pH 5.0 to pH 8.0.

[0043] In the context of the present invention, it has been surprisingly achieved that a pH range is preferably effective for a refolding step integrated into the method of downstream processing. Said pH range depends on the pI of the first binding domain that binds to the (tumor) target and / or the pI of the whole antibody construct. The first domain of the antibody construct as described herein usually has a pI of at least 4.9. For example, compared to the first binding domain, i.e. the target binding domain, the CD33xCD3 antibody construct as described herein has a pI of about 4.9. Other pI values ​​are usually about 6.0 for the CD19xCD3 antibody construct as described herein, about 6.4 for the BCMAxCD3 antibody construct as described herein, about 8.7 for the CD19xCD3 antibody construct as described herein, about 9.4 for the MSLNxCD3 antibody construct as described herein, and about 9.2 for the FLT3xCD3 antibody construct as described herein. Thus, the pH of the refolding buffer should at least correspond to the typically lowest pI of the first binding domain of the antibody construct of the invention, which is typically but not necessarily 4.9. Thus, for example, a refolding buffer of about pH 5 is generally suitable in the context of the present invention.

[0044] In the context of the present invention, the pI of the second binding domain, i.e. the CD3 epsilon binding domain, of the antibody construct according to the present invention will usually have a higher pI than the target binding domain, e.g. a pI of at least 9, typically a pI of 9.3. If a third domain is present according to the present invention, the pI will typically be at least 7, typically 7.2. Thus, the pI of the whole antibody construct will be higher than that of the first binding domain. Usually, the pI of the whole antibody construct will be in the range of about 7.0-9.0, i.e. neutral. Thus, a neutral pH of the refolding buffer or the environment in which refolding occurs, such as pH 7.0-9.0, e.g. 8.0, is preferred in the context of the present invention. For example, the CD33xCD3 antibody construct as described herein has a pI of about 7.2. Other pI values ​​are typically about 8.0 for a CD19xCD3 antibody construct as described herein, about 8.0 for a BCMAxCD3 antibody construct as described herein, about 8.6 for a CD19xCD3 antibody construct as described herein, about 8.8 for a MSLNxCD3 antibody construct as described herein, and about 8.8 for a FLT3xCD3 antibody construct as described herein.

[0045] A "pool" is understood in the present invention as a downstream process liquid that contains the process product (i.e., bispecific antibody construct) and is not associated with a matrix (e.g., a column). According to the method of the present invention, the refolding buffer can be applied to several different pools or columns. Pools include, for example, the filtered viral inactivation pool (FVIP) after Protein A purification or the high molecular weight (HMW) pool after CEX with buffer exchange or coincident with the filtered viral inactivation step.

[0046] The bispecific antibody construct concentration may be in the range of 0.5 to 10 mg / ml, preferably around 1 mg / ml, and the period may be several hours, for example 2 to 48 hours, preferably up to 24 hours, more preferably about 10, 9, 8, 7, 6, 5, 4, 3 or 2 hours, for example 4 hours.

[0047] According to the method of the present invention, the temperature may be in the range of 4° C. to 40° C., for example, 10 to 30° C., preferably about 20 to about 25° C., for example, room temperature of about 25° C. Such a temperature selection avoids both uncontrolled and undesired aggregation and ensures a suitable reaction time.

[0048] In the context of the present invention, several oxidizing and redox agents may be utilized. An "oxidizing agent" including copper (i.e., copper(II)), dehydroascorbic acid, and dissolved oxygen is preferably used in the refolding buffer. The preferred concentration range is 0.1-10 mM, preferably 0.5-1 mM, which is higher than what has been taught in the art for closing open disulfide bonds. Lower oxidizing agent concentrations (e.g., copper(II)) usually increase the time required for oxidation, i.e., reconstitution of disulfide bonds. Higher oxidizing agent (e.g., copper(II)) concentrations may result in unfavorable oxidation and clipping of methionine residues. Methionine oxidation and clipping should be minimized if they are in protein regions important for function. In general, the oxidizing agent (e.g., copper(II)) concentration according to the present invention is selected as a compromise between remaining open disulfide bonds and clipping. The copper concentration should also be determined by the protein concentration. Lower copper concentrations have been described previously in the art, for example 0.005-0.1 mM copper sulfate by Harris et al. 2005 in US 6,808,902 and US 7,723,490, respectively, and about 0.01 mM by Treuheit et al. However, the present invention refers to methods including 0.1-10 mM oxidizing agent, preferably 0.5-1 mM oxidizing agent such as copper(II) (preferably copper(II) sulfate), which leads to a more rapid reaction to reconstitute disulfide bonds made accessible by the chaotropic agent utilized in the present invention. Without being bound by theory, it is believed that without the use of a chaotropic agent, only open, i.e., reduced, disulfides exposed on the surface would be accessible to an oxidizing or redox agent to close (i.e., oxidize) the disulfides and restore bonds and tertiary structure, advantageously reducing the amount of aggregates.

[0049] Several "redox agents" may be used additionally or alternatively, including the following pairs: cysteine / cystine, cysteine / cystamine, reduced glutathione / oxidized glutathione. Cysteine ​​or reduced glutathione may be used alone, for example, as they generate oxidized forms at near-neutral pH. Surprisingly, it was found that the use of cysteine ​​and cystine significantly reduced aggregates. However, cysteinylation of two cysteines (one open disulfide bond) was observed when using cysteine ​​and cystine. Preferably, the cystine concentration is about 1 mM, and the cysteine ​​concentration is n-fold higher compared to the cysteine ​​concentration. For example, the cystine concentration may range from 0.1 to 10 mM, preferably 1 mM. The cysteine:cystine ratio may be any value between 1:1 to 33:1, preferably 20:1. Cystamine instead of cystine may be utilized at comparable concentrations and ratios. Reduced and oxidized glutathione may also be utilized in similar concentrations and ratios.

[0050] Addition of an oxidizing agent according to the invention, such as copper or dehydroascorbic acid, together with a chaotropic agent advantageously reduces the amount of unwanted product aggregates. Typically, a significant reduction in aggregation can be achieved, for example, from 26% to 6% for CD33xCD3 bispecific antibody constructs in Protein A FVIP pools, by refolding. Preferred conditions can include, for example, contacting the aggregates with a refolding buffer at pH 5.1 (acetate buffer) and 2.0 M guanidine for 4 hours at room temperature. A significant reduction in aggregation can also be achieved in CEX HMW pools by applying a refolding buffer comprising, for example, 2.0 M guanidine and 0.5 mM copper(II) at pH 8.0, or, for example, 2.0 M guanidine and 0.5 mM dehydroascorbic acid at pH 8.0.

[0051] The present invention is also directed to a refolding buffer that may be advantageously utilized in the method according to the present invention.

[0052] The refolding buffers according to the invention are also suitable for refolding on a downstream chromatography column, such as a Protein A column. In general, the methods of redox refolding described herein can be carried out by adding the above-mentioned redox and chaotrope agents in the vessel, on the column, in the cell culture medium after harvest or in the cell culture medium in the bioreactor before harvest.

[0053] Bispecific antibody constructs according to the invention, such as ScFc-BiTE® molecules, usually contain an Fc region that is of similar size, pI and hydrophobicity to the corresponding region of an IgG antibody. General downstream processing of IgG antibodies from host cell proteins (HCPs), other bioreactor impurities and reagents, high molecular weight (HMW) species or aggregates, low molecular weight species or clips is known in the art (Shukla et al. 2006). Such downstream processing has been adapted for the purification of bispecific antibody constructs, such as scFc-BiTE®, and usually includes the following steps: harvesting of cell culture; protein A chromatography; viral inactivation by filters; a second column chromatography polishing step or two, e.g., CEX; viral filtration; UF / DF. Bispecific antibody construct material obtained upstream, e.g., in a continuous manufacturing process, is usually purified by applying the steps shown in Figure 1. The method of performing redox refolding according to the invention can be carried out at any of the above listed harvest and purification steps, which often contain a large percentage of unwanted aggregates. For example, redox refolding can be performed in a post-Protein A chromatography vessel that is purified from host cell proteins and reagents but still contains aggregates. Denaturing reagents can be added to the Protein A pool or the Protein A pool can be buffer exchanged into a denaturing, refolding buffer. Refolding can also be performed on a Protein A column by rinsing and incubating the protein immobilized on the column with a chaotropic agent such as guanidine, arginine, urea, etc. An oxidizing agent (e.g., copper or dehydroascorbic acid) or a redox reagent can also be added to perform on-column refolding.

[0054] In the context of the present invention, a "chaotrope" or "chaotropic agent" is understood as, for example, a denaturing or unfolding agent such as guanidine, arginine, urea, extreme pH, organic solvents such as alcohols and / or detergents such as SDS, which disrupt the hydrogen bond network and van der Waals interactions of macromolecules such as bispecific antibody constructs. This increases the entropy of the macromolecule, which correlates with an increase in the kinetic energy or movement of the macromolecule. At low concentrations, which are preferably applied according to the present invention, chaotropes may partially or temporarily disturb the three-dimensional surface structure of proteins. For full-length monoclonal antibodies (mAbs), Mehta et al. 2014 tested by using various spectroscopic techniques, size-exclusion chromatography (SEC) and analytical ultracentrifugation, prior to the action of the chaotrope guanidine hydrochloride (GdnHCl) at concentrations between 0 and 2 M. At day 0, a moderate disturbance in the tertiary structure of the mAb was detected. At day 1 (after 24 h at 37°C), significant disruption of the tertiary structure was observed at chaotrope concentrations above 1.2 M and especially at chaotrope concentrations above 1.8 M. This resulted in an increase in high molecular weight (HMW) species after 6 and 11 days of incubation. Aggregation of mAb was only detected in solutions that induced partial unfolding of guanidine above 1.2 M. Mehta et al. 2014 showed that "Differential scanning calorimetry studies showed that the CH2 domain of the antibody was unfolded in antibody molecules incubated in 1.2 M and higher concentrations of GuHCl. These results suggest that unfolding of the CH2 domain causes aggregation."

[0055] Based on teachings in the art, a person skilled in the art would hesitate to treat an antibody or bispecific antibody construct with a chaotrope at a guanidine concentration of more than 1 M, e.g. 1.2 M, when aggregation should be prevented. It is therefore surprising that in the method of the invention, a bispecific antibody of the invention can be treated with a chaotrope concentration of more than 1 M for the purpose of refolding and still obtain a non-aggregated product. However, the bispecific antibody construct of the invention comprises at least one extra disulfide bond in any one of the three binding domains, i.e. at least one of the target binding domain, the CD3 binding domain or the scFc binding domain, respectively. Preferably, there is an extra disulfide bond in the CH2 of the third domain, which makes the bispecific antibody construct more stable, i.e. more stable than the CH2 of a glycosylated mAb. As a supporting feature, the method of the invention preferably applies milder reaction conditions, i.e. a reaction time shorter than 24 hours, e.g. 4 hours, and a temperature lower than 37° C., e.g. 25° C. Without wishing to be bound by theory, according to the properties as claimed herein, applying the method according to the invention does not cause significant perturbations of the tertiary structure that would ultimately lead to aggregation. Instead, the partial and temporary unfolding by chaotropes of the aggregated bispecific antibody constructs subjected to the method of the invention allows oxidizing and / or redox agents to reach the "open" disulfide bonds and "close" them again.

[0056] "Closing" in the context of the present invention means the re-establishment of the -SS-disulfide covalent bond, thereby re-establishing the tertiary structure of the bispecific antibody construct and eliminating or at least reducing previous aggregation, resulting in an increased product yield in downstream processes to which the method of the present invention is applied.

[0057] In the context of the present invention, "cell culture" or "culture" refers to the growth and propagation of cells outside a multicellular organism or tissue. Culture conditions suitable for mammalian cells are known in the art. See, for example, Animal cell culture: A Practical Approach, D. Rickwood, ed., Oxford University Press, New York (1992). Mammalian cells can be cultured in suspension or attached to a solid culture medium.

[0058] The term "mammalian cell" refers to any cell from or derived from any mammal (e.g., human, hamster, mouse, green monkey, rat, pig, cow, or rabbit). For example, the mammalian cell may be an immortalized cell. In some embodiments, the mammalian cell is a differentiated cell. In some embodiments, the mammalian cell is an undifferentiated cell. Non-limiting examples of mammalian cells are described herein. Further examples of mammalian cells are known in the art.

[0059] As used herein, the term "cell culturing medium" (also referred to as "culture medium", "cell culture media", "tissue culture medium") refers to any nutrient liquid used for growing cells, e.g., animal or mammalian cells, and generally provides at least one or more components from the following: an energy source (usually in the form of a carbohydrate such as glucose); all essential amino acids, and generally one or more of the 20 basic amino acids and cysteine; vitamins and / or other organic compounds, which are typically required in low concentrations; lipids or free fatty acids; and trace elements, e.g., inorganic compounds or naturally occurring elements, which are typically required in very low concentrations, usually in the micromolar range.

[0060] Cell culture media include, but are not limited to, those commonly utilized and / or known for use in any cell culture process, such as batch, expanded batch, fed-batch and / or perfusion or continuous culture of cells.

[0061] "Growth" cell culture medium or feed medium refers to a cell culture medium that is used in cell culture during the period of exponential growth, the "growth phase," and is complete enough to support cell culture during this phase. Growth cell culture medium may also contain a selection agent that confers resistance or survival to a selectable marker incorporated into the host cell line. Such selection agents include, but are not limited to, geneticin (G4118), neomycin, hygromycin B, puromycin, zeocin, methionine sulfoximine, methotrexate, glutamine-free cell culture medium, cell culture medium lacking glycine, hypoxanthine and thymidine, or cell culture medium lacking only thymidine.

[0062] "Production" cell culture medium or feed medium refers to a cell culture medium that is typically used for cell culture during the transition phase when exponential growth ends and during the subsequent transition and / or production phase when protein production becomes predominant. Such a cell culture medium is sufficiently complete to maintain a desired cell density, viability and / or product titer during this phase.

[0063] "Perfusion" cell culture medium or feed medium refers to cell culture medium typically used in cell culture that is maintained by perfusion or continuous culture methods and is complete enough to support the cell culture during this process. Perfusion cell culture medium formulations may be richer or more concentrated than basic cell culture medium formulations to accommodate the methods used to remove spent medium. Perfusion cell culture medium may be used in both the growth and production phases.

[0064] The term "culture" or "cell culture" refers to the maintenance or growth of mammalian cells under a controlled set of physical conditions.

[0065] The term "mammalian cell culture" means a liquid culture medium containing a plurality of mammalian cells maintained or grown under a controlled set of physical conditions.

[0066] The term "liquid culture medium" refers to a liquid containing sufficient nutrients to grow or proliferate cells (e.g., mammalian cells) in vitro. For example, the liquid culture medium may contain one or more of amino acids (e.g., the 20 amino acids), purines (e.g., hypoxanthine), pyrimidines (e.g., thymidine), choline, inositol, thiamine, folic acid, biotin, calcium, niacinamide, pyridoxine, riboflavin, thymidine, cyanocobalamin, pyruvic acid, lipoic acid, magnesium, glucose, sodium, potassium, iron, copper, zinc, and sodium bicarbonate. In some embodiments, the liquid culture medium may contain serum from a mammal. In some embodiments, the liquid culture medium does not contain serum or another extract from a mammal (defined liquid culture medium). In some embodiments, the liquid culture medium may contain trace metals, mammalian growth hormones, and / or mammalian growth factors. Another example of a liquid culture medium is a minimal medium (e.g., a medium containing only inorganic salts, a carbon source, and water). Non-limiting examples of liquid culture media are described herein. Further examples of liquid culture media are known in the art and are commercially available. The liquid culture medium may contain any density of mammalian cells. For example, as used herein, the volume of liquid culture medium removed from the bioreactor may be substantially free of mammalian cells.

[0067] The second liquid culture medium may be the same as the first liquid culture medium. In some examples of fed-batch culture, the second liquid culture medium is a concentrated form of the first liquid culture medium. In some examples of fed-batch culture, the second liquid culture medium is added as a dry powder.

[0068] The term "antibody product" refers to a "secreted protein" or a "secreted recombinant protein," meaning a protein (e.g., a recombinant protein) that originally contains at least one secretory signal sequence when translated in a mammalian cell, and that is secreted, at least in part, into the extracellular space (e.g., liquid culture medium) via enzymatic cleavage of the secretory signal sequence within the mammalian cell. One of skill in the art will understand that a "secreted" protein need not completely dissociate from the cell to be considered a secreted protein.

[0069] The term bispecific antibody product encompasses bispecific antibodies, such as full-length, e.g., IgG-based antibodies and fragments thereof, which are generally referred to herein as bispecific antibody constructs.

[0070] The term "antibody construct" refers to a molecule whose structure and / or function is based on the structure and / or function of an antibody, e.g. a full-length or complete immunoglobulin molecule, and / or derived from the variable heavy (VH) and / or variable light (VL) domains of an antibody or a fragment thereof. The antibody construct is thus capable of binding to its specific target or antigen. Furthermore, the binding domain of the antibody construct according to the invention comprises the minimum structural requirement of an antibody that allows target binding. This minimum requirement may for example be defined by the presence of at least three light chain CDRs (i.e. CDR1, CDR2 and CDR3 of the VL region) and / or three heavy chain CDRs (i.e. CDR1, CDR2 and CDR3 of the VH region), preferably all six CDRs. An alternative way of defining the minimum structural requirement of an antibody is the definition of the epitope of the antibody within the structure of a specific target, the protein domains (epitope clusters) of the target protein that respectively constitute the epitope region, or by reference to specific antibodies that compete with the epitope of the defined antibody. Antibodies on which the constructs according to the invention are based include, for example, monoclonal antibodies, recombinant antibodies, chimeric antibodies, deimmunized antibodies, humanized antibodies and human antibodies.

[0071] The binding domain of the antibody construct according to the invention may, for example, comprise the CDRs of the above-referenced groups. Preferably, those CDRs are comprised within the framework of an antibody light chain variable region (VL) and an antibody heavy chain variable region (VH); however, it is not necessary that it comprises both. An Fd fragment, for example, has two VH regions and often retains some antigen-binding function of an intact antigen-binding domain. Further examples of formats of antibody fragments, antibody variants or binding domains include (1) a Fab fragment, which is a monovalent fragment having the VL, VH, CL and CH1 domains; (2) a F(ab')2 fragment, which is a bivalent fragment having two Fab fragments linked by a disulfide bridge at the hinge region; (3) an Fd fragment having two VH and CH1 domains; (4) an Fv fragment having the VL and VH domains of one arm of an antibody, (5) a dAb fragment having a VH domain (Ward et al., (1989) Nature 341:544-546); (6) isolated complementarity determining regions (CDRs), and (7) single chain Fvs (scFvs), the latter of which are preferred (e.g., from an scFV library). Exemplary embodiments of antibody constructs according to the invention are described, for example, in WO 00 / 006605, WO 2005 / 040220, WO 2008 / 119567, WO 2010 / 037838, WO 2013 / 026837, WO 2013 / 026833, US 2014 / 0308285, US 2014 / 0302037, WO 2014 / 144722, WO 2014 / 151910 and WO 2015 / 048272.

[0072] The definition of "binding domain" or "domain that binds to" also includes fragments of full-length antibodies such as VH, VHH, VL, (s)dAb, Fv, Fd, Fab, Fab', F(ab')2 or "r IgG" ("half antibodies"). Antibody constructs according to the invention may also include modified fragments of antibodies, also called antibody variants, such as scFv, di-scFv or bi(s)-scFv, scFv-Fc, scFv-zipper, scFab, Fab2, Fab3, diabodies, single chain diabodies, tandem diabodies (Tandab's), tandem di-scFv, tandem tri-scFv, "multibodies" such as triabodies or tetrabodies, and single domain antibodies, such as nanobodies, or single variable domain antibodies comprising only one variable domain which may be a VHH, VH or VL that specifically binds to an antigen or epitope independent of other V regions or domains.

[0073] As used herein, the term "single-chain Fv", "single-chain antibody" or "scFv" refers to a single polypeptide chain antibody fragment that contains the variable regions from both the heavy and light chains but lacks the constant region. Typically, single-chain antibodies further comprise a polypeptide linker between the VH and VL domains that allows them to form the desired structure that allows for antigen binding. Single-chain antibodies are discussed in detail by Pluckthun in The Pharmacology of Monoclonal Antibodies, vol. 113, Rosenburg and Moore eds. Springer-Verlag, New York, pp. 269-315 (1994). Various methods of making single chain antibodies are known, including those described in U.S. Patent Nos. 4,694,778 and 5,260,203; International Patent Publication WO 88 / 01649; Bird (1988) Science 242:423-442; Huston et al. (1988) Proc. Natl. Acad. Sci. USA 85:5879-5883; Ward et al. (1989) Nature 334:54454; Skerra et al. (1988) Science 242:1038-1041. In certain embodiments, single chain antibodies can be bispecific, multispecific, human and / or humanized, and / or synthetic.

[0074] Furthermore, the definition of the term "antibody construct" includes monovalent, bivalent and polyvalent / multivalent constructs, thus bispecific constructs that specifically bind to only two antigenic structures, as well as polyspecific / multispecific constructs that specifically bind to more than two antigenic structures, e.g. three, four or more, through different binding domains. Furthermore, the definition of the term "antibody construct" includes molecules consisting of only one polypeptide chain, as well as molecules consisting of two or more polypeptide chains, where the chains can be either identical (homodimers, homotrimers or homooligomers) or different (heterodimers, heterotrimers or heterooligomers). Examples of the above identified antibodies and variants or derivatives thereof are described, inter alia, in Harlow and Lane, Antibodies a laboratory manual, CSHL Press (1988) and Using Antibodies: a laboratory manual, CSHL Press (1999), Kontermann and Duebel, Antibody Engineering, Springer, 2nd ed. 2010 and Little, Recombinant Antibodies for Immunotherapy, Cambridge University Press 2009.

[0075] As used herein, the term "bispecific" refers to an antibody construct that is "at least bispecific", i.e., it comprises at least a first binding domain and a second binding domain, where the first binding domain binds to one antigen or target (here a surface antigen of a target cell) and the second binding domain binds to another antigen or target (e.g. CD3). Thus, an antibody construct according to the invention comprises specificity for at least two different antigens or targets. For example, the first domain preferably does not bind to one or more extracellular epitopes of CD3ε of the species described herein. The term "surface antigen of a target cell" refers to an antigenic structure that is expressed by a cell and is present on its cell surface so that it can be accessed by the antibody construct described herein. It may be a protein, preferably an extracellular portion of a protein, or a carbohydrate structure, preferably a carbohydrate structure of a protein such as a glycoprotein. It is preferably a tumor antigen. The term "bispecific antibody construct" according to the present invention also encompasses multispecific antibody constructs, such as trispecific antibody constructs comprising three binding domains or constructs with four or more (e.g. four, five...) specificities.

[0076] When the antibody constructs according to the invention are (at least) bispecific, they do not occur in nature and they differ significantly from naturally occurring products. Thus, a "bispecific" antibody construct or immunoglobulin is an artificial hybrid antibody or immunoglobulin having at least two different binding sites with different specificities. Bispecific antibody constructs can be produced by a variety of methods including fusion of hybridomas or linking of Fab' fragments. See, for example, Songsivilai & Lachmann, Clin. Exp. Immunol. 79:315-321 (1990).

[0077] At least two binding domains and a variable domain (VH / VL) of the antibody construct of the present invention may or may not comprise a peptide linker (spacer peptide). The term "peptide linker" according to the present invention comprises an amino acid sequence that interconnects the amino acid sequences of one (variable and / or binding) domain and the other (variable and / or binding) domain of the antibody construct of the present invention. A peptide linker may also be used to fuse a third domain to another domain of the antibody construct of the present invention. An essential technical feature of such a peptide linker is that it does not contain any polymerization activity. Suitable peptide linkers include those described in US Pat. Nos. 4,751,180 and 4,935,233 or WO 88 / 09344. A peptide linker may also be used to attach other domains or modules or regions (such as half-life extending domains) to the antibody construct of the present invention.

[0078] The antibody construct of the present invention is preferably an "in vitro generated antibody construct". This term refers to an antibody construct according to the above definition, in which all or part of the variable region (e.g. at least one CDR) is generated in non-immune cell selection, such as in vitro phage display, protein chip or any other method that allows testing candidate sequences for antigen binding capacity. Thus, this term preferably excludes sequences that are generated only by genome rearrangement in immune cells of an animal. A "recombinant antibody" is an antibody generated by the use of recombinant DNA technology or genetic engineering.

[0079] As used herein, the term "monoclonal antibody" (mAb) or monoclonal antibody construct refers to an antibody obtained from a population of substantially homogeneous antibodies, i.e., individual antibodies comprising the population that are identical except for possible naturally occurring mutations and / or post-translational modifications (e.g., isomerization, amidation) that may be present in minor amounts. Monoclonal antibodies are highly specific and are directed against a single antigenic site or determinant on the antigen, in contrast to conventional (polyclonal) antibody preparations that typically contain different antibodies directed against different determinants (or epitopes). In addition to their specificity, monoclonal antibodies are advantageous in that they are synthesized by hybridoma culture and are therefore uncontaminated by other immunoglobulins. The modifier "monoclonal" indicates the character of the antibody as being obtained from a substantially homogeneous population of antibodies and is not to be construed as requiring production of the antibody by any particular method.

[0080] For the preparation of monoclonal antibodies, any technique that results in antibodies produced by continuous cell line culture can be used. For example, the monoclonal antibodies used can be made by the hybridoma method first described by Koehler et al., Nature, 256:495 (1975), or by recombinant DNA methods (see, for example, U.S. Pat. No. 4,816,567). Additional examples of techniques for producing human monoclonal antibodies include the trioma technique, the human B-cell hybridoma technique (Kozbor, Immunology Today 4 (1983), 72), and the EBV-hybridoma technique (Cole et al., Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, Inc. (1985), 77-96).

[0081] The hybridomas can then be screened using standard methods, such as enzyme-linked immunosorbent assay (ELISA) and surface plasmon resonance (BIACORE™) analysis, to identify one or more hybridomas that produce antibodies that specifically bind to the designated antigen. Any form of the relevant antigen can be used as the immunogen, for example, recombinant antigen, naturally occurring form, any variant or fragment thereof, as well as antigenic peptides thereof. Surface plasmon resonance, as employed in the BIAcore system, can be used to increase the efficiency of binding of phage antibodies to epitopes of surface antigens of target cells (Schier, Human Antibodies Hybridomas 7 (1996), 97-105; Malmborg, J. Immunol. Methods 183 (1995), 7-13).

[0082] Another exemplary method for generating monoclonal antibodies includes screening protein expression libraries, such as phage display or ribosome display libraries. Phage display is described, for example, in U.S. Patent No. 5,223,409 to Ladner et al.; Smith (1985) Science 228:1315-1317; Clackson et al., Nature, 352:624-628 (1991); and Marks et al., J. Mol. Biol., 222:581-597 (1991).

[0083] In addition to using display libraries, the relevant antigen can be used to immunize a non-human animal (e.g., a rodent, such as a mouse, hamster, rabbit, or rat). In one embodiment, the non-human animal comprises at least a portion of a human immunoglobulin gene. For example, mouse strains deficient in mouse antibody production can be engineered with large fragments of the human Ig (immunoglobulin) locus. Hybridoma technology can be used to generate and select antigen-specific monoclonal antibodies derived from genes with the desired specificity. See, e.g., XENOMOUSE™, Green et al. (1994) Nature Genetics 7:13-21, U.S. Patent Application Publication No. 2003-0070185, WO 96 / 34096, and WO 96 / 33735.

[0084] Monoclonal antibodies can also be obtained from non-human animals and subsequently modified, e.g., humanized, deimmunized, chimerized, etc., using recombinant DNA techniques known in the art. Examples of modified antibody constructs include humanized variants of non-human antibodies, "affinity matured" antibodies (see, e.g., Hawkins et al. J. Mol. Biol. 254, 889-896 (1992) and Lowman et al., Biochemistry 30, 10832-10837 (1991)), and antibody variants with altered effector functions (see, e.g., U.S. Pat. No. 5,648,260; Kontermann and Duebel (2010), supra; and Little (2009), supra).

[0085] In immunology, affinity maturation is the process by which B cells produce antibodies with increased antigen affinity during an immune response. Repeated exposure to the same antigen will cause the host to produce antibodies with successively higher affinities. Similar to natural prototypes, in vitro affinity maturation is based on the principle of mutation and selection. In vitro affinity maturation has been successfully used to optimize antibodies, antibody constructs, and antibody fragments. Random mutations are introduced within the CDRs using radiation, chemical mutagens, or error-prone PCR. In addition, chain shuffling can increase genetic diversity. Two or three rounds of mutation and selection using display methods such as phage display usually result in antibody fragments with affinities in the low nanomolar range.

[0086] A preferred type of amino acid substitution variant of an antibody construct is one that involves substitution of one or more hypervariable region residues of a parent antibody (e.g., a humanized or human antibody). Generally, the resulting variants selected for further development will have improved biological properties relative to the parent antibody from which they were generated. A convenient method for generating such substitution variants involves affinity maturation using phage display. Briefly, several hypervariable region sites (e.g., 6-7 sites) are mutated to generate all possible amino acid substitutions at each site. The antibody variants so generated are displayed in a monovalent form from filamentous phage particles as fusions with the gene III product of M13 packaged within each particle. The phage-displayed variants are then screened for their biological activity (e.g., binding affinity) as disclosed herein. To identify candidate hypervariable region sites for modification, alanine scanning mutagenesis can be performed to identify hypervariable region residues that contribute significantly to antigen binding. Alternatively or additionally, it may be beneficial to analyze a crystal structure of the antigen-antibody complex to identify contact points between the binding domain and, for example, a surface antigen of a human target cell. Such contact and adjacent residues are candidates for substitution by the techniques detailed herein. After generating such variants, a panel of variants may be subjected to screening as described herein, and antibodies with superior properties in one or more relevant assays may be selected for further development.

[0087] The monoclonal antibodies and antibody constructs of the invention include "chimeric" antibodies (immunoglobulins) in which a portion of the heavy and / or light chain is identical or homologous to corresponding sequences in antibodies derived from a particular species or belonging to a particular antibody class or subclass, while the remainder of the chain is identical or homologous to corresponding sequences in antibodies derived from another species or belonging to another antibody class or subclass, as well as fragments of such antibodies, so long as they exhibit the desired biological activity (U.S. Pat. No. 4,816,567; Morrison et al., Proc. Natl. Acad. Sci. USA, 81:6851-6855 (1984)). Chimeric antibodies of interest herein include "primatized" antibodies that contain variable domain antigen-binding sequences derived from a non-human primate (e.g., Old World Monkey, Ape, etc.) and human constant region sequences. Various techniques for producing chimeric antibodies have been described. See, e.g., Morrison et al., Proc. Natl. Acad. Sci USA 81:6851, 1985; Takeda et al., Nature 314:452, 1985; Cabilly et al., U.S. Pat. No. 4,816,567; Boss et al., U.S. Pat. No. 4,816,397; Tanaguchi et al., EP 0171496; EP 0173494; and GB 2177096.

[0088] Antibodies, antibody constructs, antibody fragments or antibody variants may also be modified by specific deletion of human T-cell epitopes (a method called "deimmunization") by the methods disclosed in the examples of WO 98 / 52976 or WO 00 / 34317. Briefly, the heavy and light chain variable domains of the antibody can be analyzed for peptides that bind to MHC class II. These peptides represent potential T-cell epitopes (as defined in WO 98 / 52976 and WO 00 / 34317). For the detection of potential T-cell epitopes, a computer modeling approach called "peptide threading" can be applied, as described in WO 98 / 52976 and WO 00 / 34317, and in addition, a database of human MHC class II binding peptides can be searched for motifs present in the VH and VL sequences. These motifs bind to any of the 18 major MHC class II DR allotypes, and therefore represent potential T cell epitopes. Potential T cell epitopes detected can be eliminated by substituting a small number of amino acid residues in the variable domains, or preferably by single amino acid substitutions. Conservative substitutions are usually made. In many, but not all, amino acids common to positions in human germline antibody sequences can be used. Human germline sequences are disclosed, for example, in Tomlinson, et al. (1992) J. Mol. Biol. 227:776-798; Cook, GP et al. (1995) Immunol. Today Vol. 16(5):237-242; and Tomlinson et al. (1995) EMBO J. 14:14:4628-4638. The V BASE directory provides a comprehensive directory of human immunoglobulin variable region sequences (edited by Tomlinson, LA. et al. MRC Centre for Protein Engineering, Cambridge, UK). These sequences can be used as a source of human sequences, e.g., for the framework regions and CDRs.For example, the consensus human framework regions described in US Pat. No. 6,300,064 can be used.

[0089] A "humanized" antibody, antibody construct, variant or fragment thereof (Fv, Fab, Fab', F(ab')2 or other antigen-binding subsequence of an antibody) is an antibody or immunoglobulin of largely human sequence that contains minimal sequence derived from non-human immunoglobulin. In most cases, humanized antibodies are human immunoglobulins (recipient antibody) in which residues from a hypervariable region (also referred to as CDR) of the recipient are replaced by residues from a hypervariable region of a non-human (e.g. rodent) species (donor antibody) such as mouse, rat, hamster or rabbit having the desired specificity, affinity, and capacity. In some cases, Fv framework region (FR) residues of the human immunoglobulin are replaced by corresponding non-human residues. Furthermore, as used herein, "humanized antibodies" may also comprise residues that are not found in the recipient antibody or in the donor antibody. These modifications are made to further refine and optimize antibody performance. A humanized antibody may also comprise at least a portion of an immunoglobulin constant region (Fc), typically that of a human immunoglobulin. For further details, see Jones et al., Nature, 321:522-525 (1986); Reichmann et al., Nature, 332:323-329 (1988); and Presta, Curr. Op. Struct. Biol., 2:593-596 (1992).

[0090] Humanized antibodies or fragments thereof can be generated by replacing sequences of Fv variable domains that are not directly involved in antigen binding with equivalent sequences from human Fv variable domains. Exemplary methods for generating humanized antibodies or fragments thereof are provided by Morrison (1985) Science 229:1202-1207; Oi et al. (1986) BioTechniques 4:214; and U.S. Pat. Nos. 5,585,089; 5,693,761; 5,693,762; 5,859,205; and 6,407,213. These methods include isolating, manipulating, and expressing nucleic acid sequences that encode all or a portion of an immunoglobulin Fv variable domain from at least one of the heavy or light chains. Such nucleic acids can be obtained from hybridomas and other sources that produce antibodies against a predetermined target as described above. The recombinant DNA encoding the humanized antibody molecule can then be cloned into an appropriate expression vector.

[0091] Humanized antibodies can also be produced using transgenic animals, such as mice, that express human heavy and light chain genes but are unable to express endogenous mouse immunoglobulin heavy and light chain genes. Winter describes an exemplary CDR grafting method that can be used to prepare the humanized antibodies described herein (U.S. Pat. No. 5,225,539). All of the CDRs of a particular human antibody can be replaced with at least a portion of a non-human CDR, or only a portion of the CDRs can be replaced with a non-human CDR. It is only necessary to replace as many CDRs as are required for the binding of the humanized antibody to a given antigen.

[0092] Humanized antibodies can be optimized by introducing conservative substitutions, consensus sequence substitutions, germline substitutions and / or back mutations. Such modified immunoglobulin molecules can be produced by any of several techniques known in the art (e.g., Teng et al., Proc. Natl. Acad. Sci. USA, 80:7308-7312, 1983; Kozbor et al., Immunology Today, 4:7279, 1983; Olsson et al., Meth. Enzymol., 92:3-16, 1982, and EP 239400).

[0093] The terms "human antibody," "human antibody construct," and "human binding domain" include antibodies, antibody constructs, and binding domains having antibody regions, such as variable and constant regions or domains, that substantially correspond to human germline immunoglobulin sequences known in the art, including, for example, those described by Kabat et al. (1991) supra. A human antibody, antibody construct, or binding domain of the invention may include amino acid residues not encoded by human germline immunoglobulin sequences (e.g., mutations introduced by random or site-specific mutagenesis in vitro or by somatic mutation in vivo), for example in the CDRs, particularly CDR3. A human antibody, antibody construct, or binding domain may have at least one, two, three, four, five, or more positions substituted with an amino acid residue not encoded by human germline immunoglobulin sequences. However, as used herein, the definitions of human antibody, antibody constructs and binding domains also contemplate "fully human antibodies" which comprise only non-artificially and / or genetically altered human sequences of antibodies, which may be obtained by using techniques or systems such as Xenomouse. Preferably, a "fully human antibody" does not include amino acid residues that are not encoded by human germline immunoglobulin sequences.

[0094] In some embodiments, the antibody construct of the present invention is an "isolated" or "substantially pure" antibody construct. "Isolated" or "substantially pure" as used to describe the antibody construct disclosed herein means an antibody construct that has been identified, separated and / or recovered from components of its production environment. Preferably, the antibody construct is free or substantially free from association with all other components from its production environment. Contaminating components of its production environment, such as components arising from recombinant transfected cells, are materials that would normally interfere with diagnostic or therapeutic uses for the polypeptide, which may include enzymes, hormones and other proteinaceous or non-proteinaceous solutes. The antibody construct may, for example, comprise at least about 5% or at least about 50% by weight of the total protein in a given sample. It is understood that an isolated protein may comprise 5% to 99.9% by weight of the total protein content, depending on the circumstances. The polypeptide may be produced at significantly higher concentrations by the use of inducible promoters or high expression promoters so that the polypeptide is produced at high concentration levels. This definition includes production of antibody constructs in a variety of organisms and / or host cells known in the art. In a preferred embodiment, the antibody construct will be purified (1) to a degree sufficient to obtain at least 15 residues of N-terminal or internal amino acid sequence using a spinning cup sequenator, or (2) to homogeneity by SDS-PAGE under non-reducing or reducing conditions with Coomassie blue or, preferably, silver staining. Typically, however, an isolated antibody construct will be prepared by at least one purification step.

[0095] The term "binding domain" is considered in the context of the present invention as a domain that (specifically) binds to / interacts with / recognizes a given target epitope or a given target site on a target molecule (antigen), e.g. CD33 and CD3, respectively. The structure and function of the first binding domain (e.g. recognizing CD33), and preferably also the structure and / or function of the second binding domain (e.g. recognizing CD3), are based on the structure and / or function of an antibody, e.g. a full-length or complete immunoglobulin molecule, and / or are derived from the variable heavy (VH) and / or variable light (VL) domains of an antibody or a fragment thereof. Preferably, the first binding domain is characterized by the presence of three light chain CDRs (i.e. CDR1, CDR2 and CDR3 of the VL region) and / or three heavy chain CDRs (i.e. CDR1, CDR2 and CDR3 of the VH region). The second binding domain preferably also comprises the minimal structural requirements of an antibody that allow target binding. More preferably, the second binding domain comprises at least three light chain CDRs (i.e. CDR1, CDR2 and CDR3 of the VL region) and / or three heavy chain CDRs (i.e. CDR1, CDR2 and CDR3 of the VH region). It is envisaged that the first and / or second binding domains are generated or obtained by phage display or library screening methods other than grafting CDR sequences from an existing (monoclonal) antibody onto a scaffold.

[0096] According to the present invention, the binding domain is in the form of one or more polypeptides. Such polypeptides may contain proteinaceous and non-proteinaceous portions (e.g. chemical linkers or chemical cross-linking agents such as glutaraldehyde). Proteins (including fragments thereof, preferably biologically active fragments and peptides, usually having less than 30 amino acids) comprise two or more amino acids linked together through covalent peptide bonds (resulting in a chain of amino acids).

[0097] As used herein, the term "polypeptide" refers to a group of molecules that usually consists of more than 30 amino acids. Polypeptides may further form multimers, such as dimers, trimers and higher oligomers, i.e., consisting of two or more polypeptide molecules. The polypeptide molecules forming such dimers, trimers, etc. may be identical or non-identical. The corresponding higher order structures of such multimers are therefore referred to as homo- or heterodimers, homo- or heterotrimers, etc. An example of a heteromultimer is an antibody molecule, which in its native form consists of two identical polypeptide light chains and two identical polypeptide heavy chains. The terms "peptide", "polypeptide" and "protein" also refer to naturally modified peptides / polypeptides / proteins, which are modified, for example, by post-translational modifications such as glycosylation, acetylation, phosphorylation, etc. As referred to herein, "peptide", "polypeptide" or "protein" may also be chemically modified, such as pegylation. Such modifications are well known in the art and are described herein below.

[0098] Preferably, the binding domain that binds to the surface antigen of the target cell and / or the binding domain that binds to CD3ε are human binding domains. Antibodies and antibody constructs that include at least one human binding domain avoid some of the problems associated with antibodies or antibody constructs with non-human variable and / or constant regions, such as from rodents (e.g., mice, rats, hamsters or rabbits). The presence of such rodent-derived proteins may result in rapid clearance of the antibody or antibody construct or may generate an immune response against the antibody or antibody construct by the patient. To avoid the use of rodent-derived antibodies or antibody constructs, human or fully human antibodies / antibody constructs can be generated by introducing human antibody functions into rodents such that the rodent produces fully human antibodies.

[0099] The ability to clone and reconstruct megabase-sized human loci in YACs and introduce them into the mouse germline provides a powerful approach for elucidating the functional elements of very large or coarsely mapped loci and for generating useful models of human disease. Furthermore, the use of such techniques to replace mouse loci with their human equivalents will provide unique insights into the expression and regulation of nascent human gene products, their transfer to other systems, and their involvement in disease induction and progression.

[0100] An important practical application of such a strategy is the "humanization" of the mouse humoral immune system. The introduction of human immunoglobulin (Ig) loci into mice in which the endogenous Ig genes have been inactivated provides an opportunity to study the mechanisms underlying the programmed expression and assembly of antibodies and their role in B-cell development. Furthermore, such a strategy would provide an ideal source for the generation of fully human monoclonal antibodies (mAbs), which would represent a major milestone in realizing the potential of antibody therapy in human diseases. Fully human antibodies or antibody constructs are expected to minimize the immunogenic and allergic reactions inherent to mouse or mouse-derived mAbs, thereby increasing the efficacy and safety of the administered antibodies / antibody constructs. The use of fully human antibodies or antibody constructs can be expected to provide significant advantages in the treatment of chronic and recurrent human diseases that require repeated administration of compounds, such as inflammation, autoimmunity and cancer.

[0101] One approach towards this goal has been to engineer mouse strains deficient in mouse antibody production with large fragments of the human Ig loci, with the expectation that such mice would produce a broad repertoire of human antibodies without producing mouse antibodies. The large human Ig fragments would preserve the broad diversity of variable genes and the proper regulation of antibody production and expression. By utilizing the mouse machinery for antibody diversification and selection, as well as the lack of immune tolerance to human proteins, the human antibody repertoires recapitulated in these mouse strains should produce high affinity antibodies against any antigen of interest, including human antigens. Using hybridoma technology, antigen-specific human mAbs with the desired specificity could be easily generated and selected. This general strategy was demonstrated in conjunction with the generation of the first XenoMouse mouse strains (see Green et al. Nature Genetics 7:13-21 (1994)). This XenoMouse strain was engineered with yeast artificial chromosomes (YACs) containing 245 kb and 190 kb sized germline-configured fragments of the human heavy and kappa light chain loci, respectively, that contained the core sequences of the variable and constant regions. The human Ig-containing YACs proved compatible with the mouse system for both antibody rearrangement and expression, and were able to replace inactivated mouse Ig genes. This was demonstrated by their ability to induce B cell development to produce an adult-like human repertoire of fully human antibodies and to produce antigen-specific human mAbs. These results also suggested that the introduction of a large portion of the human Ig locus, containing multiple V genes, additional regulatory elements, and human Ig constant regions, could recapitulate a virtually complete repertoire characterized by the human humoral response to infection and immunization. Recently, extending the work of Green et al., the introduction of megabase-sized germline-configured YAC fragments of the human heavy and kappa light chain loci, respectively, introduced approximately 80% of the human antibody repertoire.See Mendez et al. Nature Genetics 15:146-156 (1997) and U.S. Patent Application Publication No. 08 / 759,620.

[0102] The generation of XenoMouse mice was performed in accordance with U.S. Patent Application Publication Nos. 07 / 466,008, 07 / 610,515, 07 / 919,297, 07 / 922,649, 08 / 031,801, 08 / 112,848, 08 / 234,145, 08 / 376,279, 08 / 430,938, 08 / 464,584, 08 / 464,582, 08 / 463,191, 08 / 462,837, 08 / 463,192, 08 / 463,281, 08 / 463,306, 08 / 463,193, 08 / 463,282, 08 / 463,306, 08 / 463,283, 08 / 463,308, 08 / 463,309, 08 / 463,284, 08 / 463,309, 08 / 463,285, 08 / 463,286, 08 / 463,287, 08 / 463,288, 08 / 463,289, 08 / 463,309, 08 / 463,289, 08 / 463,289, 08 / 463,286, 08 / 463,287, 08 / 463,289 ... and 5,939,598, and Japanese Patent Publications Nos. 3068180B2, 3068506B2, and 3068507B2. See also Mendez et al. Nature Genetics 15:146-156 (1997) and Green and Jakobovits J. Exp. Med. 188:483-495 (1998), EP 0463151 B1, WO 94 / 02602, WO 96 / 34096, WO 98 / 24893, WO 00 / 76310 and WO 03 / 47336.

[0103] In another approach, others including GenPharm International, Inc. have utilized a "minilocus" approach. In the minilocus approach, an exogenous Ig locus is mimicked by including fragments (individual genes) from the Ig locus. Thus, one or more VH genes, one or more DH genes, one or more JH genes, a mu constant region and a second constant region (preferably a gamma constant region) form a construct that is inserted into an animal. This approach is disclosed in U.S. Pat. No. 5,545,807 to Surani et al., and U.S. Pat. Nos. 5,545,806; 5,625,825; 5,625,126; 5,633,425; 5,661,016; 5,770,429; 5,789,650; 5,814,318; and 5,877,397 to Lonberg and Kay, respectively. Nos. 5,874,299; and 6,255,458, U.S. Pat. Nos. 5,591,669 and 6,023.010 to Krimpenfort and Berns, U.S. Pat. Nos. 5,612,205; 5,721,367; and 5,789,215 to Berns et al., and U.S. Pat. No. 5,643,763 to Choi and Dunn, and GenPharm and U.S. Patent Application Nos. 07 / 574,748, 07 / 575,962, 07 / 810,279, 07 / 853,408, 07 / 904,068, 07 / 990,860, 08 / 053,131, 08 / 096,762, 08 / 155,301, 08 / 161,739, 08 / 165,699, and 08 / 209,741 to International.See also EP 0546073B1, WO 92 / 03918, WO 92 / 22645, WO 92 / 22647, WO 92 / 22670, WO 93 / 12227, WO 94 / 00569, WO 94 / 25585, WO 96 / 14436, WO 97 / 13852 and WO 98 / 24884, and U.S. Pat. No. 5,981,175. See also Taylor et al. (1992), Chen et al. (1993), Tuaillon et al. (1993), Choi et al. (1993), Lonberg et al. (1994), Taylor et al. (1994), and Tuaillon et al. (1995), Fishwild et al. (1996).

[0104] Kirin has also demonstrated the production of human antibodies from mice into which large chromosome pieces or entire chromosomes have been introduced by microcell fusion. See European Patent Applications Nos. 773288 and 843961. Xenerex Biosciences is developing a technology for the potential production of human antibodies, in which SCID mice are reconstituted with human lymphoid cells, e.g., B cells and / or T cells. The mice are then immunized with an antigen and allowed to generate an immune response against that antigen. See U.S. Patent Nos. 5,476,996; 5,698,767; and 5,958,765.

[0105] Human anti-mouse antibody (HAMA) responses have led the industry to generate chimeric or otherwise humanized antibodies. However, it is expected that some human anti-chimeric antibody (HACA) responses will be observed, especially in chronic or multi-dose antibody applications. It would therefore be desirable to provide an antibody construct that includes a human binding domain for a target cell surface antigen and a human binding domain for CD3ε to eliminate the concerns and / or impact of HAMA or HACA responses.

[0106] The terms "(specifically) bind", "(specifically) recognize", "(specifically) induce" and "(specifically) react" mean according to the present invention that a binding domain interacts or specifically interacts with a target molecule (antigen), here a surface antigen of a target cell and a given epitope or a given target site on CD3ε, respectively.

[0107] The term "epitope" refers to a site on an antigen to which a binding domain, such as an antibody or immunoglobulin or a derivative, fragment or variant of an antibody or immunoglobulin, specifically binds. An "epitope" is antigenic, and thus the term epitope is also sometimes referred to herein as an "antigenic structure" or "antigenic determinant". The binding domain is therefore the "antigen interaction site". Said binding / interaction is understood to also define "specific recognition".

[0108] An "epitope" can be formed by both contiguous amino acids or non-contiguous amino acids juxtaposed by tertiary folding of a protein. A "linear epitope" is an epitope that comprises an epitope in which a primary amino acid sequence is recognized. A linear epitope usually comprises at least 3 or at least 4, and more usually at least 5, or at least 6, or at least 7, e.g., about 8 to about 10 amino acids in a unique sequence.

[0109] A "conformational epitope", in contrast to a linear epitope, is an epitope in which the primary sequence of amino acids that comprise the epitope is not the only element that defines the recognized epitope (e.g., an epitope in which the primary sequence of amino acids is not necessarily recognized by the binding domain). In general, a conformational epitope comprises a larger number of amino acids compared to a linear epitope. With respect to the recognition of a conformational epitope, the binding domain recognizes the three-dimensional structure of an antigen, preferably a peptide or protein or a fragment thereof (in the context of the present invention, the antigenic structure for one of the binding domains is contained within the surface antigen protein of a target cell). For example, when a protein molecule folds to form a three-dimensional structure, certain amino acids and / or polypeptide backbones that form a conformational epitope are juxtaposed, thereby allowing the antibody to recognize the epitope. Methods for determining the conformational structure of an epitope include, but are not limited to, x-ray crystallography, two-dimensional nuclear magnetic resonance (2D-NMR) spectroscopy, and site-specific spin labeling and electron paramagnetic resonance (EPR) spectroscopy.

[0110] The method of epitope mapping is described below. When a region (a contiguous stretch of amino acids) of a human target cell surface antigen protein is exchanged / substituted with a corresponding region of a non-human and non-primate target cell surface antigen (e.g., a mouse target cell surface antigen, but also chicken, rat, hamster, rabbit, etc.), a reduction in the binding property of the binding domain is expected to occur, as long as the binding domain is not cross-reactive with the non-human, non-primate target cell surface antigen used. The reduction is preferably at least 10%, 20%, 30%, 40% or 50%; more preferably at least 60%, 70% or 80%, and most preferably 90%, 95% or even 100% compared to the binding to the corresponding region in the human target cell surface antigen protein, assuming that the binding to the corresponding region of the human target cell surface antigen protein is 100%. It is envisaged that the above human target cell surface antigen / non-human target cell surface antigen chimera is expressed in CHO cells. It is also envisaged that the human target cell surface antigen / non-human target cell surface antigen chimeras are fused to the transmembrane and / or cytoplasmic domains of different membrane-associated proteins, such as EpCAM.

[0111] In an alternative or additional method of epitope mapping, several truncations of the human target cell surface antigen extracellular domain can be made to determine the specific region recognized by the binding domain. In these truncations, different extracellular target cell surface antigen domains / subdomains or regions are deleted stepwise starting from the N-terminus. It is envisaged that the truncated target cell surface antigen can be expressed in CHO cells. It is also envisaged that the truncated target cell surface antigen may be fused with the transmembrane and / or cytoplasmic domains of different membrane-bound proteins such as EpCAM. It is also envisaged that the truncated target cell surface antigen may include a signal peptide domain at their N-terminus, for example a signal peptide derived from the mouse IgG heavy chain signal peptide. It is further envisaged that the truncated target cell surface antigen may include a v5 domain at the N-terminus (following the signal peptide) that can confirm their correct expression on the cell surface. It is expected that reduced or lost binding occurs for truncated target cell surface antigens that no longer include the target cell surface antigen region recognized by the binding domain. The reduction in binding is preferably at least 10%, 20%, 30%, 40% or 50%; more preferably at least 60%, 70%, 80%, and most preferably 90%, 95% or even 100%, relative to binding to the entire surface antigen protein (or extracellular region or domain thereof) of a human target cell taken as 100.

[0112] A further method for determining the contribution of specific residues of target cell surface antigens to recognition by antibody constructs or binding domains is alanine scanning (see, for example, Morrison KL & Weiss GA.Cur Opin Chem Biol.2001 Jun;5(3):302-7), in which each residue to be analyzed is replaced with alanine, for example by site-directed mutagenesis. Alanine is used due to its non-bulky and chemically inert methyl functional group, even though it mimics the secondary structure criteria that many other amino acids have. If it is desirable to preserve the size of the residue to be mutated, bulky amino acids such as valine or leucine may sometimes be used. Alanine scanning is a mature technique that has been used for a long time.

[0113] The interaction between a binding domain and an epitope or an epitope-containing region means that the binding domain exhibits a measurable affinity for the epitope / epitope-containing region on a particular protein or antigen (herein, target cell surface antigen and CD3, respectively) and generally does not exhibit significant reactivity with proteins or antigens other than the target cell surface antigen or CD3. "Measurable affinity" refers to a binding domain that exhibits a measurable affinity of about 10 -6 M(KD) or stronger. Preferably, the binding affinity is about 10 -12 ~10 -8 M, 10 -12 ~10 -9 M, 10 -12 ~10 -10 M, 10 -11 ~10 -8 M, preferably about 10 -11 ~10 -9M, the binding is considered specific. Whether a binding domain specifically reacts with or binds to a target can be easily tested, inter alia, by comparing the reaction of said binding domain to a target protein or antigen with the reaction of said binding domain to a protein or antigen other than the surface antigen of the target cell or CD3. Preferably, the binding domain of the invention essentially or substantially does not bind to a protein or antigen other than the surface antigen of the target cell or CD3 (i.e., the first binding domain cannot bind to a protein other than the surface antigen of the target cell and the second binding domain cannot bind to a protein other than CD3). It is an envisaged feature of the antibody construct according to the invention to have superior affinity properties compared to other HLE formats. Such superior affinity consequently suggests an extended half-life in vivo. The longer half-life of the antibody construct according to the invention may reduce the duration and frequency of administration, which usually contributes to improved patient compliance. This is particularly important, since the antibody construct of the invention is particularly beneficial for cancer patients who are highly debilitated or even multi-disease.

[0114] The terms "does not essentially / substantially bind" or "cannot bind" mean that the binding domain of the invention does not bind to proteins or antigens other than the target cell surface antigen or CD3, i.e., when the binding to the target cell surface antigen or CD3, respectively, is taken as 100%, it does not show a reactivity of more than 30%, preferably 20% or less, more preferably 10% or less, and particularly preferably 9%, 8%, 7%, 6% or 5% or less to proteins or antigens other than the target cell surface antigen or CD3.

[0115] Specific binding is believed to be brought about by specific motifs in the amino acid sequences of the binding domain and the antigen. Thus, binding occurs as a result of their primary, secondary and / or tertiary structure, as well as secondary modifications of said structure. Specific interaction of the antigen interaction site with its specific antigen may result in simple binding of said site to the antigen. Furthermore, specific interaction of the antigen interaction site with its specific antigen may alternatively or additionally result in the initiation of a signal, for example by inducing a conformational change in the antigen, oligomerization of the antigen, etc.

[0116] The term "variable" refers to that portion of an antibody or immunoglobulin domain (i.e., the "variable domain") that exhibits variability in sequence and is involved in determining the specificity and binding affinity of a particular antibody. The pairing of a variable heavy chain (VH) and a variable light chain (VL) together forms a single antigen-binding site.

[0117] The variability is not uniformly distributed throughout the variable domain of an antibody; it is concentrated in subdomains of each of the heavy and light chain variable regions. These subdomains are called "hypervariable regions" or "complementarity determining regions" (CDRs). The more conserved (i.e. non-hypervariable) parts of the variable domains are called "framework" regions (FRMs or FRs) and provide a scaffold for the six CDRs in the three-dimensional space that forms the antigen-binding surface. Naturally occurring heavy and light chain variable domains each contain four FRM regions (FR1, FR2, FR3 and FR4) that largely adopt a β-sheet configuration, which are connected by three hypervariable regions that form loop connections and, in some cases, form part of the β-sheet structure. The hypervariable regions of each chain are held together in close proximity by the FRMs and contribute to the formation of the antigen-binding site with the hypervariable regions of the other chain (see Kabat et al., supra).

[0118] The term "CDR" and its plural "CDRs" refer to the complementarity determining regions, three of which constitute the binding properties of the light chain variable region (CDR-L1, CDR-L2 and CDR-L3) and three of which constitute the binding properties of the heavy chain variable region (CDR-H1, CDR-H2 and CDR-H3). The CDRs contain most of the residues responsible for the specific interaction of the antibody with the antigen and thus contribute to the functional activity of the antibody molecule. The CDRs are the primary determinants of antigen specificity.

[0119] The exact definition of the boundaries and lengths of CDRs are subject to various classifications and numbering systems. Thus, CDRs may be represented by Kabat, Chothia, contact, or any other boundary definition, including the numbering systems described herein. Although the boundaries are different, each of these systems has some overlap in the parts that constitute the so-called "hypervariable regions" in the variable sequences. Thus, the definitions of CDRs according to these systems may differ in length and in the boundary regions with respect to the adjacent framework regions. See, for example, Kabat (an approach based on interspecies sequence variability), Chothia (an approach based on crystallographic studies of antigen-antibody complexes), and / or MacCallum (Kabat et al., supra; Chothia et al., J.MoI.Biol, 1987, 196:901-917; and MacCallum et al., J.MoI.Biol, 1996, 262:732). Yet another standard for characterizing antigen-binding site is the AbM definition used by Oxford Molecular's AbM antibody modeling software.See, for example, Protein Sequence and Structure Analysis of Antibody Variable Domains.In:Antibody Engineering Lab Manual (Ed.:Duebel, S. and Kontermann, R., Springer-Verlag, Heidelberg).As long as two residue identification techniques define overlapping but not identical regions, they can be combined to define hybrid CDRs.However, the so-called Kabat numbering system is preferred.

[0120] Typically, CDRs form loop structures that can be classified as canonical structures. The term "canonical structure" refers to the main chain conformation that antigen-binding (CDR) loops adopt. Comparative structural studies have found that five of the six antigen-binding loops have only a limited repertoire of available conformations. Each canonical structure can be characterized by the torsion angle of the polypeptide backbone. Thus, corresponding loops between antibodies can have very similar three-dimensional structures, despite the high amino acid sequence variability found in the majority of the loops (Chothia and Lesk, J. Mol. Biol., 1987, 196:901; Chothia et al., Nature, 1989, 342:877; Martin and Thornton, J. MoI. Biol, 1996, 263:800). Furthermore, there is a relationship between the loop structure adopted and the amino acid sequence surrounding it. The conformation of a particular canonical class is determined by the length of the loop and the amino acid residues present at key positions within the loop and within the conserved framework (i.e., outside the loop), and therefore assignment to a particular canonical class can be made based on the presence of these key amino acid residues.

[0121] The term "canonical structure" may also include considerations of the linear sequence of an antibody, for example, as classified by Kabat (Kabat et al., supra). The Kabat numbering scheme is a widely adopted standard for numbering the amino acid residues of antibody variable domains in a consistent manner, and is the preferred scheme applied in the present invention, as mentioned elsewhere herein. Additional structural considerations may be used to determine the canonical structure of an antibody. For example, differences not fully reflected by the Kabat numbering system can be described by the Chothia et al. numbering system and / or revealed by other techniques, such as crystallography and two- or three-dimensional computer modeling. Thus, a given antibody sequence may be classified into a canonical class that allows, among other things, the identification of a suitable chassis sequence (e.g., based on the desire to include various canonical structures in a library). The Kabat numbering system of antibody amino acid sequences and the structural considerations described in Chothia et al., supra, and their significance for interpreting the canonical aspects of antibody structure are described in the literature. The subunit structures and three-dimensional configurations of different classes of immunoglobulins are well known in the art. For a general overview of antibody structure, see Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, eds. Harlow et al., 1988.

[0122] The CDR3 of the light chain and especially the CDR3 of the heavy chain may constitute the most important determinant in antigen binding within the variable regions of the light and heavy chains. In some antibody constructs, the heavy chain CDR3 is likely to be the main contact area between antigen and antibody. An in vitro selection scheme that changes only CDR3 can be used to change the binding properties of the antibody or to determine which residues contribute to antigen binding. Therefore, CDR3 is usually the greatest source of molecular diversity within the antibody binding site. For example, H3 can be as short as 2 amino acid residues or more than 26 amino acids.

[0123] In classical full-length antibodies or immunoglobulins, each light (L) chain is linked to a heavy (H) chain by one covalent disulfide bond, while the two H chains are linked to each other by one or more disulfide bonds depending on the H chain isotype. The CH domain closest to the VH is usually referred to as CH1. The constant ("C") domains are not directly involved in antigen binding, but exhibit various effector functions, such as antibody-dependent, cell-mediated cytotoxicity, and complement activation. The Fc region of an antibody is contained within the heavy chain constant domain and can interact with Fc receptors located on the cell surface, for example.

[0124] The sequences of antibody genes after construction and somatic mutation are highly diverse, and these diversified genes are 10 It is predicted that these genes encode different antibody molecules (Immunoglobulin Genes, 2 nd ed., eds. Jonio et al., Academic Press, San Diego, CA, 1995). Thus, the immune system provides a repertoire of immunoglobulins. The term "repertoire" refers to at least one nucleotide sequence derived in whole or in part from at least one sequence encoding at least one immunoglobulin. The sequence may be generated by in vivo rearrangement of the V, D and J segments of the heavy chain and the V and J segments of the light chain. Alternatively, the sequence may be generated from a cell in response to, for example, an in vitro stimulus that causes the rearrangement. Alternatively, some or all of the sequence may be obtained by DNA splicing, nucleotide synthesis, mutagenesis and other methods (see, for example, U.S. Pat. No. 5,565,332). A repertoire may include only one sequence or may include multiple sequences, including those in a genetically diverse collection.

[0125] The term "Fc portion" or "Fc monomer" in the context of the present invention means a polypeptide comprising at least one domain with the function of a CH2 domain and at least one domain with the function of a CH3 domain of an immunoglobulin molecule. As is evident from the term "Fc monomer", a polypeptide comprising these CH domains is a "polypeptide monomer". An Fc monomer may be a polypeptide comprising a fragment of an immunoglobulin constant region excluding at least the first constant region immunoglobulin domain (CH1) of the heavy chain, but maintaining at least a functional part of one CH2 domain and a functional part of one CH3 domain, the CH2 domain being amino-terminal to the CH3 domain. In a preferred embodiment of this definition, an Fc monomer may be a polypeptide constant region comprising a part of an Ig-Fc hinge region, a CH2 region and a CH3 region, the hinge region being amino-terminal to the CH2 domain. It is envisaged that the hinge region of the present invention promotes dimerization. Such Fc polypeptide molecules can be obtained, for example, but not limited to, by papain digestion of an immunoglobulin region (which of course results in a dimer of two Fc polypeptides). In another aspect of this definition, an Fc monomer can be a polypeptide region that includes a portion of a CH2 region and a CH3 region. Such Fc polypeptide molecules can be obtained, for example, but not limited to, by pepsin digestion of an immunoglobulin molecule. In one embodiment, the polypeptide sequence of an Fc monomer is substantially similar to the Fc polypeptide sequence of an IgG1 Fc region, an IgG2 Fc region, an IgG3 Fc region, an IgG4 Fc region, an IgM Fc region, an IgA Fc region, an IgD Fc region, and an IgE Fc region. (See, for example, Padlan, Molecular Immunology, 31(3), 169-217 (1993)). Because some variation exists between immunoglobulins, and simply for clarity, the Fc monomer refers to the last two heavy chain constant region immunoglobulin domains of IgA, IgD and IgG, and the last three heavy chain constant region immunoglobulin domains of IgE and IgM. As mentioned, the Fc monomer may also include a flexible hinge N-terminal to these domains.In the case of IgA and IgM, the Fc monomer may comprise a J chain. In the case of IgG, the Fc portion comprises the immunoglobulin domains CH2 and CH3, and the hinge between the first two domains and CH2. Although the boundaries of the Fc portion may vary, an example of a human IgG heavy chain Fc portion comprising functional hinge, CH2 and CH3 domains may be defined, for example, to include residues D231 (of the hinge domain - corresponding to D234 in Table 1 below) to the carboxyl terminus of the CH3 domain, P476, L476 (for IgG4), respectively, according to the numbering according to Kabat. Two Fc portions or Fc monomers fused to each other via a peptide linker define the third domain of the antibody construct of the invention, which may also be defined as a scFc domain.

[0126] In one embodiment of the invention, it is envisaged that the scFc domains disclosed herein, the Fc monomers each fused to one another, are comprised only in the third domain of the antibody construct.

[0127] In accordance with the present invention, IgG hinge regions may be identified by similarity with the Kabat numbering set out in Table 1. In accordance with the above, it is envisaged that the hinge domain / region of the present invention comprises amino acid residues corresponding to the stretch of IgG1 sequence from D234 to P243 according to the Kabat numbering. Similarly, it is envisaged that the hinge domain / region of the present invention comprises or consists of the IgG1 hinge sequence DKTHTCPPCP (SEQ ID NO: 182) (corresponding to the D234 to P243 section as shown in Table 1 below - variants of said sequence are also envisaged, provided that the hinge region still promotes dimerization). In a preferred embodiment of the present invention, the glycosylation site at Kabat position 314 of the CH2 domain within the third domain of the antibody construct is removed by a N314X substitution, where X is any amino acid other than Q. Said substitution is preferably a N314G substitution. In a more preferred embodiment, said CH2 domain further comprises the following substitutions (positions according to Kabat): V321C and R309C (these substitutions introduce intradomain cysteine ​​disulfide bridges at Kabat positions 309 and 321).

[0128] It is also envisaged that the third domain of the antibody construct of the invention comprises or consists of, in amino to carboxyl order: DKTHTCPPCP (SEQ ID NO: 182) (i.e. hinge)-CH2-CH3-linker-DKTHTCPPCP (SEQ ID NO: 182) (i.e. hinge)-CH2-CH3. The peptide linker of the aforementioned antibody construct is in a preferred embodiment characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e. Gly4Ser (SEQ ID NO: 187), or a polymer thereof, i.e. (Gly4Ser)x, where x is an integer equal to or greater than 5 (e.g. 5, 6, 7, 8, etc. or more), with 6 being preferred ((Gly4Ser)6). The construct may further comprise the substitution N314X, preferably N314G, as described above, and / or the further substitutions V321C and R309C. In a preferred embodiment of the inventive antibody construct as defined herein above, it is envisaged that the second domain binds to an extracellular epitope of the human and / or Macaca CD3 epsilon chain.

[0129] [Table 1]

[0130] In further embodiments of the invention, the hinge domain / region comprises or consists of the IgG2 subtype hinge sequence ERKCCVECPPCP (SEQ ID NO: 183), the IgG3 subtype hinge sequence ELKTPLDTTHTCPRCP (SEQ ID NO: 184) or ELKTPLGDTTHTCPRCP (SEQ ID NO: 185), and / or the IgG4 subtype hinge sequence ESKYGPPCPSCP (SEQ ID NO: 186). The IgG1 subtype hinge sequence may be the following sequence EPKSCDKTHTCPPCP (as shown in Table 1 and in SEQ ID NO: 183). Thus, these core hinge regions are also envisaged in the context of the present invention.

[0131] The locations and sequences of IgG CH2 and IgG CD3 domains can be identified by similarity using the Kabat numbering set forth in Table 2.

[0132] [Table 2]

[0133] In one embodiment of the invention, the amino acid residues highlighted in bold in the CH3 domain of the first or both Fc monomers are deleted.

[0134] The peptide linker by which the polypeptide monomers of the third domain ("Fc portion" or "Fc monomers") are fused to each other preferably comprises at least 25 amino acid residues (25, 26, 27, 28, 29, 30, etc.). More preferably, the peptide linker comprises at least 30 amino acid residues (30, 31, 32, 33, 34, 35, etc.). It is also preferred that the linker comprises up to 40 amino acid residues, more preferably up to 35 amino acid residues, most preferably exactly 30 amino acid residues. A preferred embodiment of such a peptide linker is characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e. Gly4Ser (SEQ ID NO: 187), or a polymer thereof, i.e. (Gly4Ser)x, where x is an integer equal to or greater than 5 (e.g. 6, 7 or 8). Preferably, the integer is 6 or 7, more preferably, the integer is 6.

[0135] When a linker is used to fuse the first domain with the second domain, or the first or second domain with the third domain, this linker is preferably of sufficient length and sequence to ensure that each of the first and second domains retains their different binding specificities independently of each other. In the case of peptide linkers connecting at least two binding domains (or two variable domains) in the antibody construct of the present invention, these peptide linkers are preferably those that contain only a few amino acid residues, for example those that contain 12 or less amino acid residues. Thus, peptide linkers of 12, 11, 10, 9, 8, 7, 6 or 5 amino acid residues are preferred. Conceivable peptide linkers of less than 5 amino acids contain 4, 3, 2 or 1 amino acid, where Gly-rich linkers are preferred. A preferred embodiment of a peptide linker for the fusion of the first and second domains is shown in SEQ ID NO:1. A preferred linker embodiment of the peptide linker for fusing the second and third domains is a (Gly)4-linker and a G4-linker, respectively.

[0136] A particularly preferred "single" amino acid in connection with one of the above "peptide linkers" is Gly. Thus, the above peptide linker may consist of a single amino acid Gly. In a preferred embodiment of the invention, the peptide linker is characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e., Gly4Ser (SEQ ID NO: 187), or a polymer thereof, i.e., (Gly4Ser)x, where x is an integer equal to or greater than 1 (e.g., 2 or 3). Preferred linkers are shown in SEQ ID NOs: 1 to 12. The characteristics of said peptide linkers, including not promoting secondary structures, are known in the art and are described, for example, in Dall'Acqua et al. (Biochem. (1998) 37, 9266-9273), Cheadle et al. (Mol Immunol (1992) 29, 21-30) and Raag and Whitlow (FASEB (1995) 9(1), 73-80). Moreover, peptide linkers that do not promote any secondary structure are preferred. The mutual linkage of the domains can be provided, for example, by genetic engineering as described in the Examples. Methods for preparing fused and operably linked bispecific single chain constructs and expressing them in mammalian cells or bacteria are well known in the art (e.g., WO 99 / 54440, or Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 2001).

[0137] In an antibody construct or preferred embodiment of the invention, the first and second domains form an antibody construct of a format selected from the group consisting of (scFv)2, scFv-single domain mAb, diabody and oligomers of any of these formats.

[0138] According to a particularly preferred embodiment and as described in the accompanying examples, the first and second domains of the antibody construct of the invention are "bispecific single chain antibody constructs", more preferably bispecific "single chain Fv" (scFv). Although the two domains of the Fv fragment, VL and VH, are encoded by separate genes, they can be linked by a synthetic linker as described herein above, which allows them to be produced using recombinant methods as a single protein chain in which the VL and VH regions pair to form a monovalent molecule; see, for example, Huston et al. (1988) Proc. Natl. Acad. Sci USA 85:5879-5883. These antibody fragments are obtained using conventional techniques known to those skilled in the art, and the fragments are evaluated for function in the same manner as complete or full-length antibodies. Thus, a single chain variable fragment (scFv) is a fusion protein of the variable region of the heavy chain (VH) and the variable region of the light chain (VL) of an immunoglobulin, connected by a short linker peptide, usually about 10 to about 25 amino acids, preferably about 15 to 20 amino acids. The linker is usually rich in glycine for flexibility and serine or threonine for solubility, and can either link the N-terminus of VH to the C-terminus of VL or vice versa. The protein retains the specificity of the original immunoglobulin despite the removal of the constant region and the introduction of the linker.

[0139] Bispecific single chain antibody constructs are known in the art and are described in WO 99 / 54440; Mack, J. Immunol. (1997), 158, 3965-3970; Mack, PNAS, (1995), 92, 7021-7025; Kufer, Cancer Immunol. Immunother., (1997), 45, 193-197; Loeffler, Blood, (2000), 95, 6, 2098-2103; Bruehl, Immunol., (2001), 166, 2420-2426; Kipriyanov, J. Mol. Biol., (1999), 293, 41-56. Techniques described for the production of single chain antibodies (see, inter alia, U.S. Pat. No. 4,946,778; Kontermann and Duebel (2010), supra; and Little (2009), supra) can be adapted to produce single chain antibody constructs that specifically recognize a target of choice.

[0140] Bivalent (also called divalent) or bispecific single chain variable fragments (bi-scFv or di-scFv with the format (scFv)2) can be created by linking two scFv molecules (e.g., using a linker as described herein above). If these two scFv molecules have the same binding specificity, the resulting (scFv)2 molecule is preferably called bivalent (i.e., it has a valency of two for the same target epitope). If these two scFv molecules have different binding specificities, the resulting (scFv)2 molecule is preferably called bispecific. Linking can be done by creating a single peptide chain with two VH and two VL regions to generate tandem scFvs (see, for example, Kufer P. et al., (2004) Trends in Biotechnology 22(5):238-244). Another possibility is to create scFv molecules with a linker peptide that is too short (e.g., about 5 amino acids) to allow the two variable regions to fold together, allowing the scFvs to dimerize, a form known as a diabody (see, e.g., Hollinger, Philipp et al., (July 1993) Proceedings of the National Academy of Sciences of the United States of America 90(14):6444-8).

[0141] In accordance with the present invention, either the first domain, the second domain or the first and second domains may comprise a single domain antibody, a variable domain of a single domain antibody or at least the CDRs of a single domain antibody, respectively. Single domain antibodies contain only one (monomeric) antibody variable domain that is capable of selectively binding to a specific antigen independent of other V regions or domains. The first single domain antibodies were generated from heavy chain antibodies found in camels, which consist of V H H fragment. Cartilaginous fish also have V fragments. NARIt has heavy chain antibodies (IgNAR) from which single domain antibodies called fragments can be obtained. An alternative approach is to split the dimeric variable domain from a common immunoglobulin, for example from human or rodent origin, into monomers, thereby obtaining VH or VL as single domain Abs. Most research on single domain antibodies is currently based on heavy chain variable domains, but nanobodies derived from light chains have also been shown to specifically bind target epitopes. Examples of single domain antibodies are called sdAbs, nanobodies or single variable domain antibodies.

[0142] Therefore, (single domain mAb)2 is V H , V L , V H H and V NAR A monoclonal antibody construct is a monoclonal antibody construct composed of (at least) two single domain monoclonal antibodies individually selected from the group comprising: 1) a single domain antibody selected from the group comprising: 2) a single domain antibody selected from the group comprising: 3) a single domain antibody selected from the group comprising: 4) a single domain antibody selected from the group comprising: 5) a single domain antibody selected from the group comprising: 6) a single domain antibody selected from the group comprising: 7) a single domain antibody selected from the group comprising: 8) a single domain antibody selected from the group comprising: 9) a single domain antibody selected from the group comprising: 10) a single domain antibody selected from the group comprising: 11) a single domain antibody selected from the group comprising: 12) a single domain antibody selected from the group comprising: 13) a single domain antibody selected from the group comprising: 14) a single domain antibody selected from the group comprising: 15) a single domain antibody selected from the group comprising: 16) a single domain antibody selected from the group comprising: 17) a single domain antibody selected from the group comprising: 18) a single domain antibody selected from the group comprising: 19) a single domain antibody selected from the group comprising: 20) a single domain antibody selected from the group comprising: 21) a single domain antibody selected from the group comprising: 22) a single domain antibody selected from the group comprising:

[0143] Whether an antibody construct binds in competition with another given antibody construct can be measured by a competitive assay, such as a competitive ELISA or a competitive assay using cells. Avidin-conjugated microparticles (beads) can also be used. Similar to an avidin-coated ELISA plate, each of these beads can be used as a substrate on which an assay can be performed when reacted with a biotinylated protein. Antigen is coated on the beads, and then pre-coated with the first antibody. A second antibody is added to confirm any further binding. Possible means for reading include flow cytometry.

[0144] T cells or T lymphocytes are a type of lymphocyte (itself a type of white blood cell) that plays a central role in cell-mediated immunity. There are several subsets of T cells, each with different functions. T cells can be distinguished from other lymphocytes, such as B cells and NK cells, by the presence of T cell receptors (TCRs) on their cell surface. The TCR is responsible for the recognition of antigens bound to major histocompatibility complex (MHC) molecules and is composed of two different protein chains. In 95% of T cells, the TCR consists of an alpha (α) chain and a beta (β) chain. When the TCR binds to an antigen peptide and MHC (peptide / MHC complex), the T lymphocyte is activated through a series of biochemical events mediated by associated enzymes, co-receptors, specialized adaptor molecules, and activated or released transcription factors.

[0145] The CD3 receptor complex is a protein complex and is composed of four chains. In mammals, this complex contains the CD3γ (gamma) chain, the CD3δ (delta) chain and two CD3ε (epsilon) chains. These chains associate with the T cell receptor (TCR) and the so-called ζ (zeta) chain to form the T cell receptor CD3 complex, which generates activation signals in T lymphocytes. The CD3γ (gamma), CD3δ (delta) and CD3ε (epsilon) chains are highly related cell surface proteins of the immunoglobulin superfamily that contain a single extracellular immunoglobulin domain. The intracellular tail of the CD3 molecule contains a single conserved motif known as the immunoreceptor tyrosine-based activation motif, or ITAM for short, which is essential for the signaling ability of the TCR. The CD3 epsilon molecule is a polypeptide encoded by the CD3E gene present on chromosome 11 in humans. The most preferred CD3 epsilon epitope is included in the range of amino acid residues 1 to 27 of the human CD3 epsilon extracellular domain. The antibody constructs according to the invention are typically and advantageously expected to exhibit only low levels of undesired non-specific T cell activation in specific immunotherapy, which translates into a low risk of side effects.

[0146] Lysis of redirected target cells via recruitment of T cells with multispecific, at least bispecific, antibody constructs involves the formation of a cytolytic synapse and the delivery of perforin and granzymes. Bound T cells are capable of continuous target cell lysis and are not subject to immune evasion mechanisms that prevent peptide antigen processing and presentation or clonal T cell differentiation (see, e.g., WO 2007 / 042261).

[0147] The cytotoxicity mediated by the antibody constructs of the invention can be measured in various ways. The effector cells can be, for example, stimulated enriched (human) CD8 positive T cells or unstimulated (human) peripheral blood mononuclear cells (PBMCs). If the target cells are of macaque origin or express or are transfected with the surface antigen of the macaque target cell bound by the first domain, the effector cells should also be of macaque origin, such as a macaque T cell line, e.g. 4119LnPx. The target cells should express (at least the extracellular domain of) the surface antigen of the target cell, e.g. the surface antigen of a human or macaque target cell. The target cells can be a cell line (such as CHO) that is stably or transiently transfected with the surface antigen of the target cell, e.g. the surface antigen of a human or macaque target cell. Alternatively, the target cells can be a cell line positively expressing the surface antigen of the natural target cell. Usually, EC 50 Values ​​are expected to be lower for target cell lines that express high levels of the target cell surface antigen on the cell surface. The effector to target cell (E:T) ratio is usually about 10:1, but this can be varied. The cytotoxic activity of the target cell surface antigen x CD3 bispecific antibody construct is 51It can be measured in a Cr release assay (incubation time of about 18 hours) or a cytotoxicity assay using FACS (incubation time of about 48 hours). The incubation time (cytotoxicity response) of the assay can be changed. Other methods for measuring cytotoxicity are well known to those skilled in the art and include MTT or MTS assays, ATP-based assays including bioluminescence assays, sulforhodamine B (SRB) assays, WST assays, clonogenic assays and ECIS techniques.

[0148] The cytotoxic activity mediated by the target cell surface antigen x CD3 bispecific antibody construct of the present invention is preferably measured in a cell-based cytotoxicity assay. 51 Cytotoxicity can also be measured by EC 50 The EC50 value corresponds to the half maximal effective concentration (the concentration of the antibody construct that induces a cytotoxic response halfway between the baseline and maximum). Preferably, the EC50 value corresponds to the half maximal effective concentration (the concentration of the antibody construct that induces a cytotoxic response halfway between the baseline and maximum). 50 The values ​​are ≦5000 pM or ≦4000 pM, more preferably ≦3000 pM or ≦2000 pM, even more preferably ≦1000 pM or ≦500 pM, even more preferably ≦400 pM or ≦300 pM, even more preferably ≦200 pM, even more preferably ≦100 pM, even more preferably ≦50 pM, even more preferably ≦20 pM or ≦10 pM, and most preferably ≦5 pM.

[0149] In various assays, the EC 50 Values ​​can be measured. Stimulated / enriched CD8 + When T cells were used as effector cells, ECs were significantly higher than unstimulated PBMCs. 50 Those skilled in the art will recognize that the values ​​are expected to be lower. 50 Values ​​can be expected to be lower if the target cells express a large number of target cell surface antigens compared to rats with fewer target expressions. For example, stimulated / enriched human CD8 +When T cells are used as effector cells (and either target cell surface antigen-transfected cells such as CHO cells or target cell surface antigen-positive human cell lines are used as target cells), EC of the target cell surface antigen x CD3 bispecific antibody construct 50 The value is preferably ≦1000 pM, more preferably ≦500 pM, even more preferably ≦250 pM, even more preferably ≦100 pM, even more preferably ≦50 pM, even more preferably ≦10 pM, and most preferably ≦5 pM. When human PBMCs are used as effector cells, the EC 50 The value is preferably ≦5000 pM or ≦4000 pM (especially when the target cells are target cell surface antigen positive human cell lines), more preferably ≦2000 pM (especially when the target cells are cells transfected with the target cell surface antigen such as CHO cells), more preferably ≦1000 pM or ≦500 pM, even more preferably ≦200 pM, even more preferably ≦150 pM, even more preferably ≦100 pM, and most preferably ≦50 pM or less. When a macaque T cell line such as LnPx4119 is used as the effector cell and a cell line transfected with a macaque target cell surface antigen such as CHO cells is used as the target cell line, the EC of the target cell surface antigen x CD3 bispecific antibody construct is 50 The value is preferably ≦2000 pM or ≦1500 pM, more preferably ≦1000 pM or ≦500 pM, even more preferably ≦300 pM or ≦250 pM, even more preferably ≦100 pM, and most preferably ≦50 pM.

[0150] Preferably, the target cell surface antigen x CD3 bispecific antibody construct of the present invention does not induce / mediate lysis or essentially does not induce / mediate lysis of target cell surface antigen negative cells, such as CHO cells. The terms "does not induce lysis", "essentially does not induce lysis", "does not mediate lysis" or "essentially does not mediate lysis" mean that the antibody construct of the present invention does not induce or mediate lysis of more than 30%, preferably not more than 20%, more preferably not more than 10%, particularly preferably not more than 9%, 8%, 7%, 6% or 5% of target cell surface antigen negative cells, taking the lysis of a target cell surface antigen positive human cell line as 100%. This is usually the case at concentrations of the antibody construct up to 500 nM. The skilled person knows how to measure cell lysis without further effort. Moreover, specific instructions on how to measure cell lysis are taught herein.

[0151] The difference in cytotoxic activity between the monomeric and dimeric isoforms of a particular target cell surface antigen x CD3 bispecific antibody construct is referred to as the "potency gap." This potency gap can be measured, for example, by comparing the EC 50 Values ​​and EC of dimeric form 50 The potency gap of the target cell surface antigen x CD3 bispecific antibody construct of the invention is preferably ≦5, more preferably ≦4, even more preferably ≦3, even more preferably ≦2, and most preferably ≦1.

[0152] The first and / or second (or any further) binding domain of the antibody construct of the invention is preferably cross-species specific in members of the mammalian order of primates. Cross-species specific CD3 binding domains are described, for example, in WO 2008 / 119567. According to one embodiment, the first and / or second binding domain, in addition to binding to surface antigens of human target cells and human CD3, will also bind to surface antigens / CD3 of target cells of primates, including, but not limited to, New World primates (such as marmosets (Callithrix jacchus), cotton-top tamarins (Saguinus Oedipus) or squirrel monkeys (Saimiri sciureus)), Old World primates (such as baboons and macaques), gibbons and non-human homininae, respectively.

[0153] In one embodiment of the antibody construct of the invention, the first domain binds to a surface antigen of a human target cell and further binds to a surface antigen of a macaque target cell, such as a surface antigen of a cynomolgus monkey (Macaca fascicularis) target cell, more preferably a surface antigen of a macaque target cell expressed on the surface of a macaque cell. The affinity of the first binding domain for the surface antigen of a macaque target cell is preferably ≦15 nM, more preferably ≦10 nM, even more preferably ≦5 nM, even more preferably ≦1 nM, even more preferably ≦0.5 nM, even more preferably ≦0.1 nM, and most preferably ≦0.05 nM or even ≦0.01 nM.

[0154] Preferably, the binding affinity gap of the antibody construct according to the present invention for the surface antigen of a macaque target cell to the surface antigen of a human target cell [surface antigen of a ma target cell:surface antigen of a hu target cell] (determined, for example, by BiaCore or Scatchard analysis) is <100, preferably <20, more preferably <15, even more preferably <10, even more preferably <8, more preferably <6, and most preferably <2. The preferred range of the binding affinity gap of the antibody construct according to the present invention for the surface antigen of a macaque target cell to the surface antigen of a human target cell is 0.1-20, more preferably 0.2-10, even more preferably 0.3-6, even more preferably 0.5-3 or 0.5-2.5, and most preferably 0.5-2 or 0.6-2.

[0155] The second (binding) domain of the antibody construct of the invention binds to human CD3 epsilon and / or Macaca CD3 epsilon. In a preferred embodiment, the second domain further binds to Callithrix jacchus, Saguinus Oedipus or Saimiri sciureus CD3 epsilon. Both Callithrix jacchus and Saguinus oedipus are New World primates belonging to the family Callitrichidae, whereas Saimiri sciureus is a New World primate belonging to the family Cebidae.

[0156] With respect to the antibody construct of the invention, the second binding domain that binds to an extracellular epitope of human and / or macaque CD3 comprises: (a) CDR-L1 as set forth in SEQ ID NO: 27 of WO 2008 / 119567, CDR-L2 as set forth in SEQ ID NO: 28 of WO 2008 / 119567, and CDR-L3 as set forth in SEQ ID NO: 29 of WO 2008 / 119567; (b) CDR-L1 as set forth in SEQ ID NO: 117 of WO 2008 / 119567, CDR-L2 as set forth in SEQ ID NO: 118 of WO 2008 / 119567, and CDR-L3 as set forth in SEQ ID NO: 119 of WO 2008 / 119567; and (c) preferably comprises a VL region comprising CDR-L1, CDR-L2 and CDR-L3 selected from CDR-L1 as set forth in SEQ ID NO: 153 of WO 2008 / 119567, CDR-L2 as set forth in SEQ ID NO: 154 of WO 2008 / 119567 and CDR-L3 as set forth in SEQ ID NO: 155 of WO 2008 / 119567.

[0157] In a further preferred embodiment of the antibody construct of the invention, the second domain binding to an extracellular epitope of the human and / or Macaca CD3 epsilon chain comprises (a) CDR-H1 as set forth in SEQ ID NO: 12 of WO 2008 / 119567, CDR-H2 as set forth in SEQ ID NO: 13 of WO 2008 / 119567, and CDR-H3 as set forth in SEQ ID NO: 14 of WO 2008 / 119567; (b) CDR-H1 as set forth in SEQ ID NO: 30 of WO 2008 / 119567, CDR-H2 as set forth in SEQ ID NO: 31 of WO 2008 / 119567, and CDR-H3 as set forth in SEQ ID NO: 32 of WO 2008 / 119567; (c) CDR-H1 as set forth in SEQ ID NO: 48 of WO 2008 / 119567, CDR-H2 as set forth in SEQ ID NO: 49 of WO 2008 / 119567, and CDR-H3 as set forth in SEQ ID NO: 50 of WO 2008 / 119567; (d) CDR-H1 as set forth in SEQ ID NO: 66 of WO 2008 / 119567, CDR-H2 as set forth in SEQ ID NO: 67 of WO 2008 / 119567, and CDR-H3 as set forth in SEQ ID NO: 68 of WO 2008 / 119567; (e) CDR-H1 as set forth in SEQ ID NO: 84 of WO 2008 / 119567, CDR-H2 as set forth in SEQ ID NO: 85 of WO 2008 / 119567, and CDR-H3 as set forth in SEQ ID NO: 86 of WO 2008 / 119567; (f) CDR-H1 as set forth in SEQ ID NO: 102 of WO 2008 / 119567, CDR-H2 as set forth in SEQ ID NO: 103 of WO 2008 / 119567, and CDR-H3 as set forth in SEQ ID NO: 104 of WO 2008 / 119567; (g) CDR-H1 as set forth in SEQ ID NO: 120 of WO 2008 / 119567, CDR-H2 as set forth in SEQ ID NO: 121 of WO 2008 / 119567, and CDR-H3 as set forth in SEQ ID NO: 122 of WO 2008 / 119567; (h) CDR-H1 as set forth in SEQ ID NO: 138 of WO 2008 / 119567, CDR-H2 as set forth in SEQ ID NO: 139 of WO 2008 / 119567, and CDR-H3 as set forth in SEQ ID NO: 140 of WO 2008 / 119567; (i) CDR-H1 as set forth in SEQ ID NO: 156 of WO 2008 / 119567, CDR-H2 as set forth in SEQ ID NO: 157 of WO 2008 / 119567, and CDR-H3 as set forth in SEQ ID NO: 158 of WO 2008 / 119567; and (j) a VH region comprising CDR-H1, CDR-H2 and CDR-H3 selected from CDR-H1 as set forth in SEQ ID NO: 174 of WO 2008 / 119567, CDR-H2 as set forth in SEQ ID NO: 175 of WO 2008 / 119567 and CDR-H3 as set forth in SEQ ID NO: 176 of WO 2008 / 119567.

[0158] In a preferred embodiment of the antibody construct of the invention, the above three sets of VL CDRs are combined with the above ten sets of VH CDRs in the second binding domain to form a set (30) comprising CDR-L1-3 and CDR-H1-3, respectively.

[0159] With regard to the antibody construct of the invention, it is preferred that the second domain that binds to CD3 comprises a VL region selected from the group consisting of the VL regions as set out in SEQ ID NOs: 17, 21, 35, 39, 53, 57, 71, 75, 89, 93, 107, 111, 125, 129, 143, 147, 161, 165, 179 or 183 of WO 2008 / 119567 or as set out in SEQ ID NO: 200.

[0160] It is also preferred that the second domain that binds to CD3 comprises a VH region selected from the group consisting of the VH regions as set forth in SEQ ID NOs: 15, 19, 33, 37, 51, 55, 69, 73, 87, 91, 105, 109, 123, 127, 141, 145, 159, 163, 177 or 181 of WO 2008 / 119567 or as set forth in SEQ ID NO: 201.

[0161] More preferably, the antibody construct of the invention comprises: (a) a VL region as set forth in SEQ ID NO: 17 or 21 of WO 2008 / 119567 and a VH region as set forth in SEQ ID NO: 15 or 19 of WO 2008 / 119567; (b) a VL region as set forth in SEQ ID NO: 35 or 39 of WO 2008 / 119567 and a VH region as set forth in SEQ ID NO: 33 or 37 of WO 2008 / 119567; (c) a VL region as set forth in SEQ ID NO: 53 or 57 of WO 2008 / 119567 and a VH region as set forth in SEQ ID NO: 51 or 55 of WO 2008 / 119567; (d) a VL region as set forth in SEQ ID NO: 71 or 75 of WO 2008 / 119567 and a VH region as set forth in SEQ ID NO: 69 or 73 of WO 2008 / 119567; (e) a VL region as set forth in SEQ ID NO: 89 or 93 of WO 2008 / 119567 and a VH region as set forth in SEQ ID NO: 87 or 91 of WO 2008 / 119567; (f) a VL region as set forth in SEQ ID NO: 107 or 111 of WO 2008 / 119567 and a VH region as set forth in SEQ ID NO: 105 or 109 of WO 2008 / 119567; (g) a VL region as set forth in SEQ ID NO: 125 or 129 of WO 2008 / 119567 and a VH region as set forth in SEQ ID NO: 123 or 127 of WO 2008 / 119567; (h) a VL region as set forth in SEQ ID NO: 143 or 147 of WO 2008 / 119567 and a VH region as set forth in SEQ ID NO: 141 or 145 of WO 2008 / 119567; (i) a VL region as set forth in SEQ ID NO: 161 or 165 of WO 2008 / 119567 and a VH region as set forth in SEQ ID NO: 159 or 163 of WO 2008 / 119567; and (j) a second domain that binds to CD3 comprising a VL region and a VH region selected from the group consisting of a VL region as set out in SEQ ID NO: 179 or 183 of WO 2008 / 119567 and a VH region as set out in SEQ ID NO: 177 or 181 of WO 2008 / 119567.

[0162] A second domain that binds to CD3 comprising a VL region as set forth in SEQ ID NO: 200 and a VH region as set forth in SEQ ID NO: 201 is also preferred in the context of an antibody construct of the invention.

[0163] According to a preferred embodiment of the antibody construct of the invention, the first and / or second domain has the following format: the pair of VH and VL domains is in the format of a single chain antibody (scFv). The VH and VL domains are arranged in the order VH-VL or VL-VH. It is preferred that the VH domain is arranged at the N-terminus of the linker sequence and the VL domain is arranged at the C-terminus of the linker sequence.

[0164] A preferred embodiment of the above-mentioned antibody construct of the present invention is characterized by a CD3-binding second domain comprising an amino acid sequence selected from the group consisting of SEQ ID NOs: 23, 25, 41, 43, 59, 61, 77, 79, 95, 97, 113, 115, 131, 133, 149, 151, 167, 169, 185 or 187 of WO2008 / 119567 or as set forth in SEQ ID NO: 202.

[0165] Covalent modifications of antibody constructs are also included within the scope of the present invention, which are generally, but not necessarily, performed post-translationally. For example, some types of covalent modifications of antibody constructs are introduced into the molecule by reacting specific amino acid residues of the antibody construct with organic derivatizing agents capable of reacting with selected side chains or N- or C-terminal residues.

[0166] Cysteinyl residues most commonly are reacted with α-haloacetates (and corresponding amines), such as chloroacetic acid or chloroacetamide to give carboxymethyl or carboxyamidomethyl derivatives.Cysteinyl residues are also derivatized by reaction with bromotrifluoroacetone, α-bromo-β-(5-imidozoyl)propionic acid, chloroacetyl phosphate, N-alkylmaleimides, 3-nitro-2-pyridyl disulfide, methyl 2-pyridyl disulfide, p-chloromercuribenzoic acid, 2-chloromercuri-4-nitrophenol, or chloro-7-nitrobenzo-2-oxa-1,3-diazole.

[0167] Histidyl residues are derivatized by reaction with diethylpyrocarbonate at pH 5.5-7.0 because this agent is relatively specific for the histidyl side chain. Para-bromophenacyl bromide is also useful; the reaction is preferably carried out in 0.1 M sodium cacodylate at pH 6.0. Lysinyl and amino terminal residues react with succinic or other carboxylic acid anhydrides. Derivatization with these agents has the effect of reversing the charge of lysinyl residues. Other suitable reagents for derivatizing alpha-amino-containing residues include imidoesters such as methyl picolinimidate; pyridoxal phosphate; pyridoxal; chloroborohydrides; trinitrobenzenesulfonic acid; O-methylisourea; 2,4-pentanedione; and transaminase-catalyzed reactions with glyoxylate.

[0168] Arginyl residues are modified by reaction with one or more conventional reagents, among them phenylglyoxal, 2,3-butanedione, 1,2-cyclohexanedione, and ninhydrin. Due to the high pKa of the guanidine functional group, derivatization of arginine residues requires that the reaction be carried out under alkaline conditions. Furthermore, these reagents can react with lysine groups and the arginine epsilon-amino group.

[0169] The specific modification of tyrosyl residues may be undertaken for the purpose of introducing spectral labels into tyrosyl residues by reaction with aromatic diazonium compounds or tetranitromethane. Most commonly, N-acetylimidizole and tetranitromethane are used to form O-acetyltyrosyl species and 3-nitro derivatives, respectively. 125 I or 131 The chloramine T method described above, in which tyrosyl residues are iodinated with I to prepare labeled proteins for use in radioimmunoassay, is preferred.

[0170] Carboxyl side groups (aspartyl or glutamyl) are selectively modified by reaction with carbodiimides (R'-N=C=N--R'), where R and R' are optionally different alkyl groups, such as 1-cyclohexyl-3-(2-morpholinyl-4-ethyl)carbodiimide or 1-ethyl-3-(4-azonia-4,4-dimethylpentyl)carbodiimide. Furthermore, aspartyl and glutamyl residues are converted to asparaginyl and glutaminyl residues by reaction with ammonium ions.

[0171] Derivatization with bifunctional agents is useful for crosslinking the antibody constructs of the present invention to water-insoluble support matrices or surfaces for use in a variety of methods. Commonly used crosslinking agents include, for example, homobifunctional imidoesters including 1,1-bis(diazoacetyl)-2-phenylethane, glutaraldehyde, N-hydroxysuccinimide esters such as esters with 4-azidosalicylic acid, disuccinimidyl esters such as 3,3'-dithiobis(succinimidyl propionate), and bifunctional maleimides such as bis-N-maleimido-1,8-octane. Derivatizing agents such as methyl-3-[(p-azidophenyl)dithio]propioimidate provide photoactivatable intermediates capable of forming crosslinks in the presence of light. Alternatively, reactive water-insoluble matrices such as cyanogen bromide-activated carbohydrates and reactive substrates described in U.S. Pat. Nos. 3,969,287; 3,691,016; 4,195,128; 4,247,642; 4,229,537; and 4,330,440 are used for protein immobilization.

[0172] Glutaminyl and asparaginyl residues are frequently deamidated to the corresponding glutamyl and aspartyl residues, respectively. Alternatively, these residues are deamidated under mildly acidic conditions. Either form of these residues is included within the scope of the invention.

[0173] Other modifications include hydroxylation of proline and lysine, phosphorylation of the hydroxyl group of seryl or threonyl residues, methylation of the α-amino groups of lysine, arginine and histidine side chains (TECreighton, Proteins: Structure and Molecular Properties, W.H. Freeman & Co., San Francisco, 1983, pp. 79-86), acetylation of the N-terminal amine and amidation of any C-terminal carboxyl group.

[0174] Another type of covalent modification of an antibody construct included within the scope of the present invention includes altering the glycosylation pattern of the protein. As is known in the art, the glycosylation pattern can depend both on the sequence of the protein (e.g., the presence or absence of particular glycosylated amino acid residues, discussed below) or on the host cell or organism in which the protein is produced. Particular expression systems are discussed below.

[0175] Glycosylation of polypeptides is usually either N-linked or O-linked. N-linked refers to the attachment of a sugar chain to the side chain of an asparagine residue. The tripeptide sequences asparagine-X-serine and asparagine-X-threonine, where X is any amino acid except proline, are the recognition sequences for enzymatic attachment of a sugar chain to the asparagine side chain. Thus, the presence of either of these tripeptide sequences in a polypeptide creates a potential glycosylation site. O-linked glycosylation refers to the attachment of one of the sugars N-acetylgalactosamine, galactose, or xylose to a hydroxyamino acid, most commonly serine or threonine, although 5-hydroxyproline or 5-hydroxylysine may also be used.

[0176] Addition of glycosylation sites to an antibody construct is conveniently accomplished by altering the amino acid sequence to include one or more of the above-mentioned tripeptide sequences (for N-linked glycosylation sites). The alteration may also be accomplished by the addition of, or substitution by, one or more serine or threonine residues to the starting sequence (for O-linked glycosylation sites). Briefly, it is preferred to alter the amino acid sequence of the antibody construct by alterations at the DNA level, in particular by mutating the DNA encoding the polypeptide at preselected bases to generate codons that will be translated into the desired amino acid.

[0177] Another means of increasing the number of carbohydrate moieties on an antibody construct is by chemical or enzymatic coupling of glycosides to the protein. These procedures are advantageous in that they do not require production of the protein in a host cell that has glycosylation capabilities for N-linked and O-linked glycosylation. Depending on the mode of attachment used, sugars can be added to (a) arginines and histidines, (b) free carboxyl groups, (c) free sulfhydryl groups such as those of cysteine, (d) free hydroxyl groups such as those of serine, threonine or hydroxyproline, (e) aromatic residues such as those of phenylalanine, tyrosine or tryptophan, or (f) the amide group of glutamine. These methods are described in WO 87 / 05330 and in Aplin and Wriston, 1981, CRC Crit. Rev. Biochem., pp. 259-306.

[0178] Removal of carbohydrate moieties present on the starting antibody construct can be accomplished chemically or enzymatically. Chemical deglycosylation requires exposure of the protein to the compound trifluoromethanesulfonic acid, or an equivalent compound. This treatment cleaves most or all sugars except the linking sugar (N-acetylglucosamine or N-acetylgalactosamine), while leaving the polypeptide intact. Chemical deglycosylation is described by Hakimuddin et al., 1987, Arch. Biochem. Biophys. 259:52 and Edge et al., 1981, Anal. Biochem. 118:131. Enzymatic cleavage of carbohydrate moieties on polypeptides can be achieved by the use of various endo- and exo-glycosidases as described by Thotakura et al., 1987, Meth. Enzymol. 138:350. Glycosylation at potential glycosylation sites can be prevented by the use of the compound tunicamycin as described by Duskin et al., 1982, J. Biol. Chem. 257:3105. Tunicamycin blocks the formation of protein-N-glycosidic bonds.

[0179] Other modifications of the antibody construct are also contemplated herein.For example, another type of covalent modification of the antibody construct includes linking the antibody construct to various non-proteinaceous polymers, including but not limited to various polyols such as polyethylene glycol, polypropylene glycol, polyoxyalkylenes or copolymers of polyethylene glycol and polypropylene glycol in the manner shown in U.S. Patent Nos. 4,640,835; 4,496,689; 4,301,144; 4,670,417; 4,791,192 or 4,179,337.In addition, as known in the art, amino acid substitutions can be made at various positions in the antibody construct to facilitate the addition of polymers such as PEG.

[0180] In some embodiments, the covalent modification of the antibody construct of the present invention comprises the addition of one or more labels. To reduce potential steric hindrance, the labeling group can be attached to the antibody construct via a spacer arm of various lengths. Various methods for labeling proteins are known in the art and can be used in carrying out the present invention. The term "label" or "labeling group" refers to any detectable label. In general, labels are divided into various classes depending on the assay for detecting them, including but not limited to the following examples: a) an isotopic label, which may be a radioactive isotope or a heavy isotope, such as a radioisotope or a radionuclide (e.g., 3 H, 14 C. 15 N, 35 S, 89 Zr, 90 Y, 99 Tc, 111 In, 125 I, 131 I) b) Magnetic labels (e.g., magnetic particles) c) redox-active moieties d) optical dyes (including but not limited to chromophores, fluorophores and fluorophores), such as fluorescent groups (e.g., FITC, rhodamine, lanthanide fluorophores), chemiluminescent groups and fluorophores, which can be either "small molecule" fluorophores or proteinaceous fluorophores; e) Enzymes (e.g., horseradish peroxidase, β-galactosidase, luciferase, alkaline phosphatase) f) Biotinylation Group g) A predetermined polypeptide epitope recognized by a secondary reporter (eg, a leucine zipper pair sequence, a binding site for a secondary antibody, a metal binding domain, an epitope tag, etc.).

[0181] "Fluorescent label" refers to any molecule that can be detected by its inherent fluorescent properties. Suitable fluorescent labels include fluorescein, rhodamine, tetramethylrhodamine, eosin, erythrosine, coumarin, methyl-coumarin, pyrene, malachite green, stilbene, Lucifer Yellow, Cascade Blue J, Texas Red, IAEDANS, EDANS, BODIPY FL, LC Red 640, Cy5, Cy5.5, LC Red 705, Oregon Green, Alexa-Fluor dyes (Alexa Fluor 350, Alexa Fluor 430, Alexa Fluor 488, Alexa Fluor 546, Alexa Fluor 568, Alexa Fluor 594, Alexa Fluor 633, Alexa Fluor 660, Alexa Fluor 680), Cascade Blue, Cascade Yellow, and R-Phycoerythrin (PE) (Molecular Suitable optical dyes, including fluorophores, include, but are not limited to, fluorophores, such as fluorochromes, fluorochromes, and fluorochromes that are suitable for use in the detection of cancer. Suitable optical dyes, including fluorochromes, include, but are not limited to, fluorochromes, fluorochromes that are suitable for use in the detection of cancer. Suitable optical dyes, including fluorochromes, include, but are not limited to, fluorochromes, fluorochromes that are suitable for use in the detection of cancer. Suitable optical dyes, including fluorochromes, include, but are not limited to, fluorochromes that are suitable for use in the detection of cancer.

[0182] Suitable proteinaceous fluorescent labels include GFP from Renilla, Ptilosarcus or Aequorea species (Chalfie et al., 1994, Science 263:802-805), green fluorescent protein including EGFP (Clontech Laboratories, Inc., Genbank Accession No. U55762), blue fluorescent protein (BFP, Quantum Biotechnologies, Inc. 1801 de Maisonneuve Blvd. West, 8 th Floor, Montreal, Quebec, Canada H3H 1J9; Stauber, 1998, Biotechniques 24:462-471; Heim et al., 1996, Curr. Biol. 6:178-182), enhanced yellow fluorescent protein (EYFP, Clontech Laboratories, Inc.), luciferase (Ichiki et al., 1993, J. Immunol. 150:5408-5417), β-galactosidase (Nolan et al., 1997, J. Immunol. 150:5408-5417), and β-galactosidase (Nolan et al., 1998, J. Immunol. 150:5408-5417). al., 1988, Proc. Natl. Acad. Sci. USA 85:2603-2607), and Renilla (WO 92 / 15673, WO 95 / 07463, WO 98 / 14605, WO 98 / 26277, WO 99 / 49019, U.S. Pat. Nos. 5,292,658; 5,418,155; 5,683,888; 5,741,668; 5,777,079; 5,804,387; 5,874,304; 5,876,995; 5,925,558).

[0183] The antibody construct of the present invention may also include additional domains, for example, useful for the isolation of the molecule or relevant for adapting the pharmacokinetic profile of the molecule. The domains useful for the isolation of the antibody construct may be selected from peptide motifs or secondarily introduced moieties that can be captured by an isolation method, for example, an isolation column. Non-limiting examples of such additional domains include peptide motifs known as Myc tags, HAT tags, HA tags, TAP tags, GST tags, chitin binding domains (CBD tags), maltose binding protein (MBP tags), Flag tags, Strep tags and variants thereof (e.g., StrepII tags), and His tags. All of the antibody constructs disclosed herein characterized by identified CDRs may include a His tag domain, commonly known as a repeat of consecutive His residues, preferably five, more preferably six His residues (hexahistidine) in the amino acid sequence of the molecule. The His tag may, for example, be located at the N-terminus or C-terminus of the antibody construct, but is preferably located at the C-terminus. Most preferably, a hexahistidine tag (HHHHHH) (SEQ ID NO: 199) is attached to the C-terminus of the antibody construct according to the invention via a peptide bond. In addition, the conjugate system of PLGA-PEG-PLGA can be combined with a polyhistidine tag for sustained release applications and improved pharmacokinetic profile.

[0184] Amino acid sequence modifications of the antibody constructs described herein are also contemplated. For example, improving the binding affinity and / or other biological properties of the antibody construct may be desired. Amino acid sequence variants of the antibody construct are generated by introducing appropriate nucleotide changes into the nucleic acid of the antibody construct or by peptide synthesis. All of the amino acid sequence modifications described below should result in an antibody construct that still retains the desired biological activity (binding to surface antigens and CD3 on target cells) of the unmodified parent molecule.

[0185] The term "amino acid" or "amino acid residue" refers to amino acids having an art-recognized definition, such as an amino acid selected from the group consisting of: alanine (Ala or A); arginine (Arg or R); asparagine (Asn or N); aspartic acid (Asp or D); cysteine ​​(Cys or C); glutamine (Gln or Q); glutamic acid (Glu or E); glycine (Gly or G); histidine (His or H); isoleucine (Ile or I); leucine (Leu or L); lysine (Lys or K); methionine (Met or M); phenylalanine (Phe or F); proline (Pro or P); serine (Ser or S); threonine (Thr or T); tryptophan (Trp or W); tyrosine (Tyr or Y); and valine (Val or V), although modified, synthetic, or rare amino acids may be used as appropriate. In general, amino acids can be classified according to the presence of a nonpolar side chain (e.g., Ala, Cys, Ile, Leu, Met, Phe, Pro, Val); a negatively charged side chain (e.g., Asp, Glu); a positively charged side chain (e.g., Arg, His, Lys); or an uncharged polar side chain (e.g., Asn, Cys, Gln, Gly, His, Met, Phe, Ser, Thr, Trp and Tyr).

[0186] Amino acid modifications include, for example, deletions from, and / or insertions into, and / or substitutions of residues within the amino acid sequence of the antibody construct. Any combination of deletions, insertions, and substitutions may be made to arrive at the final construct, provided that the final construct retains the desired characteristics. Amino acid changes may also alter post-translational processes of the antibody construct, such as changing the number or position of glycosylation sites.

[0187] For example, 1, 2, 3, 4, 5 or 6 amino acids may be inserted, substituted or deleted in each of the CDRs (depending, of course, on their length), while 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 or 25 amino acids may be inserted, substituted or deleted in each of the FRs. Preferably, insertions of amino acid sequences into the antibody constructs include amino- and / or carboxyl-terminal fusions ranging in length from 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10 residues to polypeptides containing 100 or more residues, as well as intrasequence insertions of single or multiple amino acid residues. Corresponding modifications may be made within the third domain of the antibody constructs of the invention. Insertion variants of the antibody constructs of the invention include fusions of enzymes or fusions to polypeptides to the N-terminus or C-terminus of the antibody construct.

[0188] The most important sites for substitution mutagenesis include (but are not limited to) the CDRs of heavy and / or light chains, especially the hypervariable regions, but also contemplate modification of FRs in heavy and / or light chains. Substitutions are preferably conservative substitutions as described herein. Preferably, depending on the length of the CDR or FR, 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10 amino acids may be substituted in the CDRs, while 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 or 25 amino acids may be substituted in the framework regions (FRs). For example, if the CDR sequence comprises 6 amino acids, it is envisaged that 1, 2 or 3 of these amino acids are substituted. Similarly, if the CDR sequence comprises 15 amino acids, it is envisaged that 1, 2, 3, 4, 5 or 6 of these amino acids are substituted.

[0189] A useful method for identifying specific residues or regions of an antibody construct that are preferred locations for mutagenesis is called "alanine scanning mutagenesis" as described by Cunningham and Wells in Science, 244:1081:1085 (1989). In this method, a residue or target residues within the antibody construct are identified (e.g., charged residues such as arg, asp, his, lys, and glu) that affect the interaction of the amino acid with the epitope and are replaced with neutral or negatively charged amino acids (most preferably alanine or polyalanine).

[0190] Next, further or other variants are introduced at or in place of the substitution site to select those amino acid positions that show functional sensitivity to the substitution. Thus, the site or region where the amino acid sequence variant is introduced is predetermined, but the nature of the mutation itself need not be predetermined. For example, to analyze or optimize the performance of a mutation at a given site, alanine scanning or random mutagenesis may be performed at the target codon or target region, and the variants of the expressed antibody construct are screened for the optimal combination of desired activity. Techniques for making substitution mutations at a predetermined site in DNA with a known sequence are well known, such as M13 primer mutagenesis and PCR mutagenesis. Screening of mutants is performed using an assay of antigen binding activity, such as surface antigen or CD3 binding of target cells.

[0191] Generally, when amino acids are replaced in one or more or all of the CDRs of the heavy and / or light chain, it is preferred that the resulting "replaced" sequence is at least 60% or 65%, more preferably 70% or 75%, even more preferably 80% or 85% and especially preferably 90% or 95% identical to the "original" CDR sequence. This means that how identical it is to the "replaced" sequence depends on the length of the CDR. For example, a CDR with 5 amino acids is preferably 80% identical to its replaced sequence, since it has at least one replaced amino acid. Thus, the CDRs of an antibody construct may have different degrees of identity to their replaced sequences, for example CDRL1 may have 80%, while CDRL3 may have 90%.

[0192] Preferred substitutions (or replacements) are conservative substitutions. However, any substitution (including non-conservative substitutions or one or more of the "exemplary substitutions" listed in Table 3 below) is envisaged, as long as the antibody construct retains the ability to bind to the target cell surface antigen via the first domain and to bind to CD3, CD3 epsilon via the second domain, respectively, and / or its CDRs have identity to the replaced sequence (at least 60% or 65%, more preferably 70% or 75%, even more preferably 80% or 85% and particularly preferably 90% or 95% identical to the "original" CDR sequence).

[0193] Conservative substitutions are shown under the heading of "preferred substitutions" in Table 3. If such substitutions alter biological activity, they are referred to as "exemplary substitutions" in Table 3, or more substantial changes as described further below in connection with amino acid classes can be introduced and the products screened for desired characteristics.

[0194] [Table 3]

[0195] Substantial alterations in the biological properties of the antibody constructs of the invention are achieved by selecting substitutions that differ significantly in their impact on (a) the structure of the polypeptide backbone in the substituted region, e.g., as a sheet or helical conformation, (b) the charge or hydrophobicity of the molecule at the target site, or (c) maintaining the bulk of the side chain. Naturally occurring residues are classified into the following groups based on common side chain properties: (1) hydrophobic: norleucine, met, ala, val, leu, ile; (2) neutral hydrophobic: cys, ser, thr, asn, gln; (3) acidic: asp, glu; (4) basic: his, lys, arg; (5) residues that affect chain orientation: gly, pro; and (6) aromatic: trp, tyr, phe.

[0196] Non-conservative substitutions would entail exchanging a member of one of these classes for another. Any cysteine ​​residue not involved in maintaining the proper conformation of the antibody construct may be substituted, generally with serine, to improve the oxidative stability of the molecule and avoid aberrant crosslinking. Conversely, the addition of cysteine ​​bond(s) to the antibody may improve its stability, particularly where the antibody is an antibody fragment such as an Fv fragment.

[0197] For amino acid sequences, sequence identity and / or similarity are determined by standard techniques known in the art, such as, but not limited to, the local sequence identity algorithm of Smith and Waterman, 1981, Adv. Appl. Math. 2:482, the sequence identity alignment algorithm of Needleman and Wunsch, 1970, J. Mol. Biol. 48:443, the search for similarity method of Pearson and Lipman, 1988, Proc. Nat. Acad. Sci. USA 85:2444, computerized implementations of these algorithms (GAP, BESTFIT, FASTA and TFASTA in the Wisconsin Genetics Software Package, Genetics Computer Group, 575 Science Drive, Madison, Wis.), the Best Fit sequence program described by Devereux et al., 1984, Nucl. Acid Res. 12:387-395, preferably using default settings, or by visual inspection. Preferably, the percent identity is calculated by FastDB based on the following parameters: mismatch penalty of 1; gap penalty of 1; gap size penalty of 0.33; and joining penalty of 30, “Current Methods in Sequence Comparison and Analysis”, Macromolecule Sequencing and Synthesis, Selected Methods and Applications, pp 127-149 (1988), Alan R. Liss, Inc.

[0198] One example of a useful algorithm is PILEUP. PILEUP generates a multiple sequence alignment from a group of related sequences using a progressive pairwise alignment method. It can also plot a tree showing the clustering relationships used to generate the alignment. PILEUP uses a simplified version of the progressive alignment method of Feng & Doolittle, 1987, J. Mol. Evol. 35:351-360. This method is similar to that described by Higgins and Sharp, 1989, CABIOS 5:151-153. Useful PILEUP parameters include a default gap weight of 3.00, a default gap length weight of 0.10, and weighted end gaps.

[0199] Another example of a useful algorithm is the BLAST algorithm described in Altschul et al., 1990, J. Mol. Biol. 215:403-410; Altschul et al., 1997, Nucleic Acids Res. 25:3389-3402; and Karin et al., 1993, Proc. Natl. Acad. Sci. USA 90:5873-5787. A particularly useful BLAST program is the WU-BLAST-2 program, obtained from Altschul et al., 1996, Methods in Enzymology 266:460-480. WU-BLAST-2 uses several search parameters, most of which are set to default values. Adjustable parameters are set to the following values: overlap span=1, overlap fraction=0.125, word threshold (T)=II. The HSP S and HSP S2 parameters are dynamic values ​​and are established by the program itself depending on the composition of the particular sequence and the composition of the particular database in which the sequence of interest is searched; however, the values ​​can be adjusted to increase sensitivity.

[0200] An additional useful algorithm is Gapped BLAST, reported in Altschul et al., 1993, Nucl. Acids Res. 25:3389-3402. Gapped BLAST uses the BLOSUM-62 substitution score; the threshold T parameter is set to 9; the two-hit method that produces ungapped extensions has a cost of 10+k for a gap length k; Xu is set to 16, and Xg is set to 40 for the database search stage and 67 for the output stage of the algorithm. Gapped alignments are produced with a score equivalent to approximately 22 bits.

[0201] Generally, the amino acid homology, similarity or identity between the individual variant CDR or VH / VL sequences is at least 60% relative to the sequences shown herein, and more typically, it is preferred that the homology or identity is increased to at least 65% or 70%, more preferably at least 75% or 80%, even more preferably at least 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99% and even near 100%. Similarly, "percent (%) nucleic acid sequence identity" with respect to the nucleic acid sequences of the binding proteins identified herein is defined as the percentage of nucleotide residues in a candidate sequence that are identical to nucleotide residues in the coding sequence of the antibody construct. A specific method utilizes the BLASTN module of WU-BLAST-2 set to the default parameters with overlap span and overlap fraction of 1 and 0.125, respectively.

[0202] Generally, the nucleic acid sequence homology, similarity or identity between the nucleotide sequences encoding the individual variant CDR or VH / VL sequences and the nucleotide sequences set forth herein will be at least 60%, more typically the homology or identity will be at least 65%, 70%, 75%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99%, and preferably increasing to nearly 100%. Thus, a "variant CDR" or "variant VH / VL region" is one that has a particular homology, similarity or identity to a parent CDR / VH / VL of the invention and shares a biological function, including but not limited to, at least 60%, 65%, 70%, 75%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% of the specificity and / or activity of the parent CDR or VH / VL.

[0203] In one embodiment, the percentage of identity to the human germline of the antibody construct according to the invention is ≧70% or ≧75%, more preferably ≧80% or ≧85%, even more preferably ≧90%, and most preferably ≧91%, ≧92%, ≧93%, ≧94%, ≧95% or even ≧96%. Identity to the human antibody germline gene product is considered to be an important feature to reduce the risk of a therapeutic protein eliciting an immune response against the drug in the patient being treated. Hwang & Foote ("Immunogenicity of engineered antibodies"; Methods 36 (2005) 3-10) demonstrate that the reduction of the non-human portion of the drug-antibody construct results in a reduced risk of eliciting anti-drug antibodies in the patient being treated. By comparing a large number of clinically evaluated antibody drugs and the corresponding immunogenicity data, humanization of the V region of an antibody shows a tendency for the protein to be less immunogenic (5.1% of patients on average) than antibodies carrying unmodified non-human V regions (23.59% of patients on average). Therefore, for protein therapeutics in the form of V region-based antibody constructs, a high identity to the human sequence is desirable. To determine this germline identity, the V region of the VL can be aligned with the amino acid sequence of human germline V and J segments (http: / / vbase.mrc-cpe.cam.ac.uk / ) using Vector NTI software, and the amino acid sequence can be calculated in percentage by dividing the number of identical amino acid residues by the total number of amino acid residues of the VL. A similar method is possible for the VH segment (http: / / vbase.mrc-cpe.cam.ac.uk / ), with the exception that the VH CDR3 can be excluded due to its high diversity and lack of existing human germline VH CDR3 alignment partners. Recombinant techniques can then be used to increase sequence identity to human antibody germline genes.

[0204] In a further embodiment, the bispecific antibody constructs of the invention show high monomer yields under standard laboratory scale conditions, e.g. in a standard two-step purification process. Preferably, the monomer yield of the antibody constructs according to the invention is ≧0.25 mg / L supernatant, more preferably ≧0.5 mg / L, even more preferably ≧1 mg / L, and most preferably ≧3 mg / L supernatant.

[0205] Similarly, the yield of dimeric antibody construct isoforms and therefore the monomer percentage (i.e. monomer:(monomer+dimer)) of the antibody construct can be determined. The productivity of monomeric and dimeric antibody constructs and the calculated monomer percentage can be obtained, for example, by SEC purification step of culture supernatants from standardized research scale production in roller bottles. In one embodiment, the monomer percentage of the antibody construct is ≧80%, more preferably ≧85%, even more preferably ≧90%, and most preferably ≧95%.

[0206] In one embodiment, the antibody construct preferably has a plasma stability (ratio of EC50 in the presence of plasma to EC50 in the absence of plasma) of ≦5 or ≦4, more preferably ≦3.5 or ≦3, even more preferably ≦2.5 or ≦2, and most preferably ≦1.5 or ≦1. The plasma stability of the antibody construct is determined by incubating the construct in human plasma at 37° C. for 24 hours, followed by 51The antibody can be tested by determining the EC50 in a chromium release cytotoxicity assay. The effector cells in the cytotoxicity assay can be stimulated enriched human CD8 positive T cells. The target cells can be, for example, CHO cells transfected with surface antigens of human target cells. The effector cell to target cell (E:T) ratio can be selected to be 10:1. The human plasma pool used for this purpose is derived from blood of healthy donors collected by EDTA-coated syringes. The cellular components are removed by centrifugation and the upper plasma phase is collected and then pooled. As a control, the antibody construct is diluted in RPMI-1640 medium immediately before the cytotoxicity assay. The plasma stability is calculated as the ratio of EC50 (after plasma incubation) to EC50 (control).

[0207] It is further preferred that the antibody construct of the present invention has a low monomer to dimer conversion rate. The conversion rate can be measured under different conditions and analyzed by high-performance size exclusion chromatography. For example, incubation of the monomeric isoform of the antibody construct can be performed for 7 days at 37°C in an incubator at a concentration of, for example, 100 μg / ml or 250 μg / ml. Under these conditions, it is preferred that the antibody construct of the present invention exhibits a dimer percentage of ≦5%, more preferably ≦4%, even more preferably ≦3%, even more preferably ≦2.5%, even more preferably ≦2%, even more preferably ≦1.5%, and most preferably ≦1% or ≦0.5% or even 0%.

[0208] It is also preferred that the bispecific antibody construct of the present invention exhibits a very low dimer conversion rate after several freeze / thaw cycles. For example, the monomer of the antibody construct is adjusted to a concentration of, for example, 250 μg / ml in a universal formulation buffer and subjected to three freeze / thaw cycles (freezing at -80°C for 30 minutes followed by thawing at room temperature for 30 minutes) followed by rapid SEC to determine the percentage of the initial monomeric antibody construct that has been converted to a dimeric antibody construct. Preferably, the percentage of dimers in the bispecific antibody construct is, for example, ≦5%, more preferably ≦4%, even more preferably ≦3%, even more preferably ≦2.5%, even more preferably ≦2%, even more preferably ≦1.5%, and most preferably ≦1% or even ≦0.5% after three freeze / thaw cycles.

[0209] The bispecific antibody constructs of the present invention preferably exhibit good thermal stability with an aggregation temperature of ≧45° C. or ≧50° C., more preferably ≧52° C. or ≧54° C., even more preferably ≧56° C. or ≧57° C., and most preferably ≧58° C. or ≧59° C. The thermal stability parameter in terms of the aggregation temperature of the antibody can be determined as follows: An antibody solution at a concentration of 250 μg / ml is transferred into a single-use cuvette and placed in a dynamic light scattering (DLS) instrument. The sample is heated from 40° C. to 70° C. at a heating rate of 0.5° C. / min while the measurement radius is constantly being acquired. The increase in radius, which indicates melting and aggregation of the protein, is used to calculate the aggregation temperature of the antibody.

[0210] Alternatively, melting temperature curves can be determined by differential scanning calorimetry (DSC) to determine the intrinsic biophysical protein stability of the antibody construct. These experiments are performed using a MicroCal LLC (Northampton, MA, USA) VP-DSC instrument. The energy uptake of samples containing the antibody construct is recorded from 20°C to 90°C and compared to samples containing only the formulation buffer. The antibody construct is adjusted to a final concentration of, for example, 250 μg / ml in SEC running buffer. The overall temperature of the samples is increased stepwise to record each melting curve. The energy uptake of the samples and formulation buffer standards at each temperature T is recorded. The difference in energy uptake Cp (kcal / mole / °C) of the samples minus the standards is plotted against each temperature. The melting temperature is defined as the temperature at which the energy uptake first reaches a maximum.

[0211] It is also envisaged that the target cell surface antigen x CD3 bispecific antibody constructs of the invention have a turbidity (measured by OD340 after concentrating the purified monomeric antibody construct to 2.5 mg / ml and incubating overnight) of ≦0.2, preferably ≦0.15, more preferably ≦0.12, even more preferably ≦0.1, and most preferably ≦0.08.

[0212] It is further envisaged that the bispecific antibody constructs of the invention exhibit therapeutic efficacy or anti-tumour activity, which can be assessed, for example, in the test disclosed in the Examples below in advanced stage human tumour xenograft models.

[0213] The skilled artisan knows how to vary or adapt certain parameters of this test, such as the number of tumor cells injected, the site of injection, the number of human T cells implanted, the amount of bispecific antibody construct administered and the timeline, and still obtain meaningful and reproducible results. Preferably, the tumor growth inhibition T / C[%] is ≦70 or ≦60, more preferably ≦50 or ≦40, even more preferably ≦30 or ≦20, and most preferably ≦10 or ≦5 or even ≦2.5.

[0214] In a preferred embodiment of the antibody construct of the present invention, the antibody construct is a single chain antibody construct.

[0215] In a preferred embodiment of the antibody construct of the present invention, the third domain comprises, in order from amino to carboxyl: Hinge-CH2-CH3-Linker-Hinge-CH2-CH3 Includes.

[0216] Also, in one embodiment of the invention, the CH2 domain of one or preferably each (both) polypeptide monomer of the third domain comprises an intradomain cysteine ​​disulfide bridge. As known in the art, the term "cysteine ​​disulfide bridge" refers to a functional group with the general structure RSSR. This link, also called SS bond or disulfide bridge, is obtained by coupling of two thiol groups of cysteine ​​residues. With regard to the antibody construct of the invention, it is particularly preferred that the cysteines which form the cysteine ​​disulfide bridge in the mature antibody construct are introduced into the amino acid sequence of the CH2 domain corresponding to 309 and 321 (Kabat numbering).

[0217] In one embodiment of the invention, the glycosylation site at Kabat position 314 of the CH2 domain is removed. This removal of glycosylation site is preferably achieved by an N314X substitution, where X is any amino acid other than Q. Said substitution is preferably an N314G substitution. In a more preferred embodiment, said CH2 domain further comprises the following substitutions (positions according to Kabat): V321C and R309C (these substitutions introduce intradomain cysteine ​​disulfide bridges at Kabat positions 309 and 321).

[0218] For example, it is believed that the preferred features of the antibody constructs of the invention compared to bispecific hetero-Fc antibody constructs known in the art (Figure 1b) may relate inter alia to the introduction of the above mentioned modifications in the CH2 domain. Thus, with regard to the constructs of the invention, it is preferred that the CH2 domain within the third domain of the antibody construct of the invention comprises intradomain cysteine ​​disulfide bridges at Kabat positions 309 and 321 and / or the glycosylation site at Kabat position 314 is removed by an N314X substitution, preferably an N314G substitution, as described above.

[0219] In a further preferred embodiment of the invention, the CH2 domain within the third domain of an antibody construct of the invention comprises intradomain cysteine ​​disulfide bridges at Kabat positions 309 and 321, and the glycosylation site at Kabat position 314 is eliminated by a N314G substitution.

[0220] In one embodiment, the present invention provides an antibody construct comprising: (i) the first domain comprises two antibody variable domains and the second domain comprises two antibody variable domains; (ii) the first domain comprises one antibody variable domain and the second domain comprises two antibody variable domains; (iii) the first domain comprises two antibody variable domains and the second domain comprises one antibody variable domain; or (iv) An antibody construct, wherein the first domain comprises one antibody variable domain and the second domain comprises one antibody variable domain.

[0221] Thus, the first and second domains may each be binding domains comprising two antibody variable domains, such as a VH and a VL domain. Examples of such binding domains comprising two antibody variable domains as described herein above include, for example, the Fv fragment, scFv fragment or Fab fragment as described herein above. Alternatively, either or both of the binding domains may comprise only a single variable domain. Examples of such single domain binding domains as described herein above include, for example, nanobodies or single variable domain antibodies comprising only one variable domain, which may be a VHH, VH or VL, that specifically binds to an antigen or epitope independently of other V regions or domains.

[0222] In a preferred embodiment of the antibody construct of the invention, the first and second domains are fused to the third domain via a peptide linker. Preferred peptide linkers are described herein above and are characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e. Gly4Ser (SEQ ID NO: 187) or a polymer thereof, i.e. (Gly4Ser)x, where x is an integer equal to or greater than 1 (e.g., 2 or 3). A particularly preferred linker for the fusion of the first and second domains to the third domain is shown in SEQ ID NO: 1.

[0223] In a preferred embodiment, the antibody construct of the invention comprises, in amino to carboxyl order: (a) the first domain; (b) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 187 to 189; (c) the second domain; (d) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 187, 188, 189, 195, 196, 197 and 198; (e) a first polypeptide monomer of the third domain; (f) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 191, 192, 193 and 194; and (g) a second polypeptide monomer of the third domain. The present invention is characterized by comprising:

[0224] In one aspect of the present invention, the surface antigen of the target cell bound by the first domain is a tumor antigen, an antigen specific to an immune disorder, or a viral antigen. As used herein, the term "tumor antigen" can be understood as those antigens presented on tumor cells. These antigens can be presented on the cell surface with an extracellular portion, and often combine the transmembrane and cytoplasmic portions of the molecule. These antigens can sometimes only be presented by tumor cells, and never by normal cells. Tumor antigens can be expressed exclusively on tumor cells, or can show tumor-specific mutations compared to normal cells. In this case, they are called tumor-specific antigens. More common antigens are antigens presented by tumor cells and normal cells, and they are called tumor-associated antigens. These tumor-associated antigens can be overexpressed compared to normal cells, or are available for antibody binding in tumor cells due to the less compact structure of tumor tissue compared to normal cells. Non-limiting examples of tumor antigens used herein are CDH19, MSLN, DLL3, FLT3, EGFRvIII, CD33, CD19, MUC17, CLDN18.2, CDH3, CD70, BCMA and PSMA.

[0225] Further target cell surface antigens specific for immunological disorders in the context of the present invention include, for example, TL1A and TNF-alpha. Said targets are preferably addressed by the bispecific antibodies of the present invention, which are preferably full-length antibodies. In a highly preferred embodiment, the antibodies of the present invention are hetero-IgG antibodies.

[0226] In a preferred embodiment of the antibody construct of the invention, the tumor antigen is preferably selected from the group consisting of CDH19, MSLN, DLL3, FLT3, EGFRvIII, CD33, CD19, MUC17, CLDN18.2, CDH3, CD70, BCMA and PSMA.

[0227] In one aspect of the invention, the antibody construct comprises, in amino to carboxyl order: (a) SEQ ID NOs: 7, 8, 17, 27, 28, 37, 38, 39, 40, 41, 48, 49, 50, 51, 52, 59, 60, 61, 62, 63, 64, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 89, 90, 91, 92, 93, 100, 101, 102, 103, 104, 113, 114, 121, 122, 123, 124, 125, 131, 1 a first domain having an amino acid sequence selected from the group consisting of 32, 133, 134, 135, 136, 143, 144, 145, 146, 147, 148, 149, 150, 151, 158, 159, 160, 161, 162, 163, 164, 165, 166, 173, 174, 175, 176, 177, 178, 179, 180, 181, 223, 235, and 246; (b) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 187 to 189; (c) a second domain having an amino acid sequence selected from the group consisting of SEQ ID NOs: 23, 25, 41, 43, 59, 61, 77, 79, 95, 97, 113, 115, 131, 133, 149, 151, 167, 169, 185 or 187 of WO 2008 / 119567 or SEQ ID NO: 202; (d) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 187, 188, 189, 195, 196, 197 and 198; (e) a first polypeptide monomer of a third domain having a polypeptide sequence selected from the group consisting of SEQ ID NOs: 17-24 of WO 2017 / 134140; (f) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 191, 192, 193 and 194; and (g) a second polypeptide monomer of a third domain having a polypeptide sequence selected from the group consisting of SEQ ID NOs: 17 to 24 of WO 2017 / 134140.

[0228] In one aspect, the bispecific antibody construct of the invention binds to the surface antigens of the respective target cells: (a) SEQ ID NOs: 27, 28, 37-41; CD33 (b) each of SEQ ID NOs: 48 to 52; EGFRvIII (c) each of SEQ ID NOs: 59 to 64; MSLN (d) each of SEQ ID NOs: 71 to 82; CDH19 (e) each of SEQ ID NOs: 100 to 104; DLL3 (f) SEQ ID NOs: 7, 8, 17, 113 and 114; CD19 (g) each of SEQ ID NOs: 89 to 93; FLT3 (h) each of SEQ ID NOs: 121 to 125; CDH3 (i) each of SEQ ID NOs: 132 to 136; BCMA (j) each of SEQ ID NOs: 143 to 151, 158 to 166, and 173 to 181; PSMA (k) SEQ ID NO: 213; MUC17 (l) each of SEQ ID NOs: 225 and 237; CLDN18.2, and (m) SEQ ID NO: 248; CD70 and characterized in having an amino acid sequence directed thereto.

[0229] The present invention further provides a polynucleotide / nucleic acid molecule encoding the antibody construct of the present invention. A polynucleotide is a biopolymer composed of 13 or more nucleotide monomers covalently linked in a chain. DNA (such as cDNA) and RNA (such as mRNA) are examples of polynucleotides with different biological functions. Nucleotides are organic molecules that function as monomers or subunits of nucleic acid molecules such as DNA or RNA. A nucleic acid molecule or polynucleotide can be double-stranded and single-stranded, linear and circular. It is preferably contained within a vector contained within a host cell. The host cell is capable of expressing the antibody construct, for example after transformation or transfection with a vector or polynucleotide of the present invention. For that purpose, the polynucleotide or nucleic acid molecule is operably linked to a control sequence.

[0230] The genetic code is the set of rules that translate the information coded in genetic material (nucleic acids) into proteins. Biological decoding in living cells is performed by ribosomes, which carry amino acids and attach amino acids in the order specified by the mRNA using tRNA molecules that read the mRNA three nucleotides at a time. The code defines how the sequence of these nucleotide triplets, called codons, specifies which amino acid is added next during protein synthesis. With some exceptions, a three-nucleotide codon in a nucleic acid sequence specifies a single amino acid. Because the vast majority of genes are coded with the exact same code, this particular code is often referred to as the canonical or standard genetic code. The genetic code determines the protein sequence of a given coding region, but other genomic regions can influence when and where these proteins are produced.

[0231] Furthermore, the present invention provides vectors comprising the polynucleotide / nucleic acid molecules of the present invention. A vector is a nucleic acid molecule used as a vehicle to transfer (foreign) genetic material into a cell. The term "vector" includes, but is not limited to, plasmids, viruses, cosmids and artificial chromosomes. In general, genetically engineered vectors contain an origin of replication, a multiple cloning site and a selection marker. The vector itself is generally a nucleotide sequence, generally a DNA sequence, that contains an insert (transgene) and a larger sequence that serves as the "backbone" of the vector. Modern vectors may contain additional features in addition to the transgene insert and backbone: promoters, genetic markers, antibiotic resistance, reporter genes, targeting sequences, protein purification tags. Vectors called expression vectors (expression constructs) are specifically intended for the expression of a transgene in a target cell and generally have regulatory sequences.

[0232] The term "control sequences" refers to DNA sequences necessary for the expression of an operably linked coding sequence in a particular host organism. Control sequences suitable for prokaryotes include, for example, a promoter, optionally an operator sequence, and a ribosome binding site. Eukaryotic cells are known to utilize promoters, polyadenylation signals, and enhancers.

[0233] A nucleic acid is "operably linked" when it is placed into a functional relationship with another nucleic acid sequence. For example, DNA for a presequence or secretory leader is operably linked to DNA for a polypeptide if it is expressed as a protein precursor that participates in the secretion of the polypeptide; a promoter or enhancer is operably linked to a coding sequence if it affects the transcription of the sequence; or a ribosome binding site is operably linked to a coding sequence if it is positioned so as to facilitate translation. Generally, "operably linked" means that the DNA sequences being linked are contiguous, and, in the case of a secretory leader, contiguous and in reading frame. Enhancers, however, need not be contiguous. Linkage is accomplished by ligation at convenient restriction sites. If such sites do not exist, synthetic oligonucleotide adaptors or linkers are used in accordance with conventional practice.

[0234] "Transfection" is the process of deliberately introducing a nucleic acid molecule or polynucleotide (including a vector) into a target cell. The term is primarily used for non-viral methods in eukaryotic cells. Transduction is often used to describe the viral-mediated transfer of a nucleic acid molecule or polynucleotide. Transfection of animal cells typically involves opening transient pores or "holes" in the cell membrane to allow uptake of materials. Transfection can be performed using calcium phosphate, by electroporation, by cell compression, or by mixing cationic lipids with substances to create liposomes, which fuse with the cell membrane and accumulate the cargo inside.

[0235] The term "transformation" is used to describe the non-viral transfer of nucleic acid molecules or polynucleotides (including vectors) into bacteria and into non-animal eukaryotic cells, including plant cells. Transformation is thus the genetic modification of bacteria or non-animal eukaryotic cells resulting from the direct uptake and subsequent incorporation of exogenous genetic material (nucleic acid molecules) from its surroundings through the cell membrane. Transformation can occur by artificial means. For transformation to occur, cells or bacteria must be in a competent state in which transformation can occur as a timed response to environmental conditions such as starvation and cell density.

[0236] Furthermore, the present invention provides host cells transformed or transfected with the polynucleotide / nucleic acid molecule or vector of the present invention. As used herein, the term "host cell" or "recipient cell" is intended to include any individual cell or cell culture that can be or has been a recipient of vectors encoding the antibody construct of the present invention, exogenous nucleic acid molecules and polynucleotides; and / or the antibody construct itself. Introduction of the respective substance into the cell is performed by transformation, transfection, etc. The term "host cell" is also intended to include the progeny or potential progeny of a single cell. Since certain modifications may occur in successive generations, either due to spontaneous, accidental, or deliberate mutations, or due to environmental influences, such progeny may not in fact be completely identical (morphologically, or in terms of genome or total DNA set) to the parent cell, but still fall within the scope of the term as used herein. Suitable host cells include prokaryotic or eukaryotic cells, and include, but are not limited to, bacteria, yeast cells, fungal cells, plant cells and animal cells, such as insect cells and mammalian cells, such as mouse, rat, macaque or human cells.

[0237] The antibody construct of the present invention can be produced in bacteria. After expression, the antibody construct of the present invention can be isolated from the E. coli cell paste in a soluble fraction and purified, for example, by affinity chromatography and / or size exclusion. Final purification can be performed, for example, similar to the purification method for antibodies expressed in CHO cells.

[0238] In addition to prokaryotes, eukaryotic microbes such as filamentous fungi or yeast are suitable cloning or expression hosts for the antibody constructs of the present invention. Saccharomyces cerevisiae, or common baker's yeast, is the most commonly used among lower eukaryotic host microorganisms. However, many other genera, species, and strains are commonly available and useful in the present invention, such as Schizosaccharomyces pombe, K. lactis, K. fragilis (ATCC 12424), K. bulgaricus (ATCC 16045), K. wickeramii (ATCC 24178), K. waltii (ATCC 56500), K. drosophilarum (ATCC 16045), K. saccharomyces pombe, K. lactis, K. fragilis (ATCC 12424), K. bulgaricus (ATCC 16045), K. saccharomyces pombe, K. lactis, K. fragilis (ATCC 124 ... Kluyveromyces hosts such as Kluyveromyces spp., K. thermotolerans, and K. marxianus; Yarrowia spp. (EP 402226); Pichia pastoris (EP 183070); Candida spp.; Trichoderma reesia (EP 244234); Neurospora crassa; Schwanniomyces occidentalis; occidentalis; and filamentous fungi such as Neurospora, Penicillium, Tolypocladium, and Aspergillus hosts such as A. nidulans and A. niger.

[0239] Suitable host cells for the expression of the glycosylated antibody constructs of the present invention are derived from multicellular organisms. Examples of invertebrate cells include plants and insect cells. Many strains and variants of baculoviruses and corresponding permissive insect host cells from hosts (e.g., Spodoptera frugiperda (caterpillar), Aedes aegypti (mosquito), Aedes albopictus (mosquito), Drosophila melanogaster (fruit fly) and Bombyx mori) have been identified. Various virus strains for transfection are publicly available, such as the L-1 variant of Autographa californica NPV and the Bm-5 strain of Bombyx mori NPV, and such viruses may be used as viruses herein according to the invention, particularly for transfection of Spodoptera frugiperda cells.

[0240] Plant cell cultures of cotton, corn, potato, soybean, petunia, tomato, Arabidopsis and tobacco can also be used as hosts. Cloning and expression vectors useful for producing proteins in plant cell cultures are known to those skilled in the art. See, for example, Hiatt et al., Nature (1989) 342:76-78, Owen et al. (1992) Bio / Technology 10:790-794, Artsaenko et al. (1995) The Plant J 8:745-750 and Fecker et al. (1996) Plant Mol Biol 32:979-986.

[0241] However, interest has been greatest in vertebrate cells, and propagation of vertebrate cells in culture (tissue culture) has become routine procedure. Examples of useful mammalian host cell lines are the SV40 transformed monkey kidney CV1 line (COS-7, ATCC CRL 1651); human embryonic kidney line (293 cells or 293 cells subcloned for growth in suspension culture, Graham et al., J. Gen Virol. 36:59 (1977)); baby hamster kidney cells (BHK, ATCC CCL 10); Chinese hamster ovary cells / -DHFR (CHO, Urlaub et al., Proc. Natl. Acad. Sci. USA 77:4216 (1980)); mouse Sertoli cells (TM4, Mather, Biol. Reprod. 23:243-251 (1980)); monkey kidney cells (CVI ATCC CCL 70); African green monkey kidney cells (VERO-76, ATCC CRL 1587); human cervical carcinoma cells (HELA, ATCC CCL 2); canine kidney cells (MDCK, ATCC CCL 34); buffalo rat liver cells (BRL 3A, ATCC CRL 1442); human lung cells (W138, ATCC CCL 75); human liver cells (Hep G2, 1413 8065); mouse mammary tumor (MMT 060562, ATCC CCL5 1); TRI cells (Mather et al., Annals NY Acad. Sci. (1982) 383:44-68); MRC 5 cells; FS4 cells; and a human hepatoma line (Hep G2).

[0242] In a further embodiment, the present invention provides a method for the production of an antibody construct of the present invention, the method comprising culturing a host cell of the present invention under conditions allowing expression of the antibody construct of the present invention and recovering the produced antibody construct from the culture.

[0243] As used herein, the term "culturing" refers to the maintenance, differentiation, growth, proliferation and / or propagation of cells in vitro under suitable conditions in a culture medium. The term "expression" includes any step involved in the production of an antibody construct of the invention, including but not limited to transcription, post-transcriptional modification, translation, post-translational modification and secretion.

[0244] When using recombinant techniques, the antibody constructs can be produced intracellularly in the periplasmic space or directly secreted into the medium. If the antibody constructs are produced intracellularly, the first step is to remove particulate debris of host cells or lysed fragments, for example by centrifugation or ultrafiltration. Carter et al., Bio / Technology 10:163-167 (1992) describes a procedure for isolating antibodies secreted into the periplasmic space of E. coli. Briefly, cell paste is thawed in the presence of sodium acetate (pH 3.5), EDTA, and phenylmethylsulfonyl fluoride (PMSF) for about 30 minutes. Cell debris can be removed by centrifugation. If the antibody is secreted into the medium, the supernatant from such expression systems is generally first concentrated using a commercially available protein concentration filter, such as an Amicon or Millipore Pellicon ultrafiltration unit. Protease inhibitors such as PMSF may be included in any of the foregoing steps to inhibit proteolysis, and antibiotics may be included to prevent the growth of adventitious contaminating bacteria.

[0245] The antibody construct of the present invention prepared from the host cell can be recovered or purified using, for example, hydroxyapatite chromatography, gel electrophoresis, dialysis and affinity chromatography. Depending on the antibody to be recovered, other protein purification techniques, such as fractionation on an ion exchange column, ethanol precipitation, reverse phase HPLC, chromatography on silica, chromatography on heparin SEPHAROSE™, chromatography on anion or cation exchange resins (e.g., polyaspartic acid columns), chromatofocusing SDS-PAGE, and ammonium sulfate precipitation, are also available. When the antibody construct of the present invention contains a CH3 domain, Bakerbond ABX resin (JT Baker, Phillipsburg, NJ) is useful for purification.

[0246] Affinity chromatography is the preferred purification technique. The matrix to which the affinity ligand is attached is most often agarose, although other matrices are available. Mechanically stable matrices such as controlled pore glass or poly(styrenedivinyl)benzene allow for faster flow rates and shorter processing times than can be achieved with agarose.

[0247] Furthermore, the present invention provides a pharmaceutical composition comprising the antibody construct of the present invention or the antibody construct produced according to the method of the present invention. In the pharmaceutical composition of the present invention, the homogeneity of the antibody construct is preferably ≧80%, more preferably ≧81%, ≧82%, ≧83%, ≧84% or ≧85%, even more preferably ≧86%, ≧87%, ≧88%, ≧89% or ≧90%, even more preferably ≧91%, ≧92%, ≧93%, ≧94% or ≧95%, and most preferably ≧96%, ≧97%, ≧98% or ≧99%.

[0248] As used herein, the term "pharmaceutical composition" relates to a composition suitable for administration to a patient, preferably a human patient. Particularly preferred pharmaceutical compositions of the present invention comprise one or more antibody constructs of the present invention, preferably in a therapeutically effective amount. Preferably, the pharmaceutical composition further comprises a suitable formulation of one or more (pharmaceutical effective) carriers, stabilizers, excipients, diluents, solubilizers, surfactants, emulsifiers, preservatives and / or adjuvants. Components of acceptable compositions are preferably non-toxic to recipients at the dosages and concentrations employed. Pharmaceutical compositions of the present invention include, but are not limited to, liquid, frozen and lyophilized compositions.

[0249] The composition of the present invention may contain a pharmaceutically acceptable carrier.Generally, as used herein, "pharmaceutically acceptable carrier" refers to any aqueous and non-aqueous solution, sterile solution, solvent, buffer solution, such as phosphate buffered saline (PBS) solution, water, suspension, emulsion such as oil / water emulsion, various types of wetting agents, liposome, dispersion medium and coating that are suitable for pharmaceutical administration, especially parenteral administration.The use of such media and agents in pharmaceutical compositions is well known in the art, and the composition containing such carriers can be formulated by well known conventional methods.

[0250] Certain embodiments provide pharmaceutical compositions comprising an antibody construct of the invention and further one or more excipients, such as those exemplarily described in this section and elsewhere herein. Excipients can be used in the invention for a wide range of purposes, including methods of the invention for adjusting the physical, chemical or biological properties of the formulations, such as adjusting the viscosity, and / or improving the effectiveness and / or stabilizing such formulations, as well as against degradation and damage due to stresses encountered during and after manufacture, transportation, storage, preparation prior to use, administration.

[0251] In certain embodiments, pharmaceutical compositions may contain formulating materials intended to, for example, modify, sustain, or protect the pH, osmolality, viscosity, clarity, color, isotonicity, odor, sterility, stability, dissolution or release rate, adsorption, or penetration of the composition (see REMINGTON'S PHARMACEUTICAL SCIENCES, 18" Edition, (ARGenrmo, ed.), 1990, Mack Publishing Company). In such embodiments, suitable formulating materials may include, but are not limited to, the following: Charged amino acids, preferably containing lysine, lysine acetate, arginine, glutamate and / or histidine, e.g. glycine, alanine, glutamine, asparagine, threonine, proline, 2-phenylalanine Antibacterial and antifungal agents ·Antioxidants such as ascorbic acid, methionine, sodium sulfite or sodium bisulfite; Buffers, buffer systems and buffering agents used to maintain the composition at or slightly below physiological pH; examples of buffers are borate, bicarbonate, Tris-HCl, citrate, phosphate or other organic acids, succinate, phosphate, and histidine; for example, Tris buffer at about pH 7.0-8.5; non-aqueous solvents, for example propylene glycol, polyethylene glycol, vegetable oils such as olive oil, and injectable organic esters such as ethyl oleate; Aqueous carriers, including water, alcoholic / aqueous solutions, emulsions or suspensions, including saline and buffered media; · Biodegradable polymers such as polyester; ·Bulking agents such as mannitol or glycine; · Chelating agents such as ethylenediaminetetraacetic acid (EDTA); ·Isotonicity and absorption retarding agents; complexing agents, e.g. caffeine, polyvinylpyrrolidone, β-cyclodextrin or hydroxypropyl-β-cyclodextrin) Injectables; monosaccharides; disaccharides; and other carbohydrates (such as glucose, mannose or dextrins); the carbohydrates may be non-reducing sugars, preferably trehalose, sucrose, octasulfate, sorbitol or xylitol; (low molecular weight) proteins, polypeptides or proteinaceous carriers, such as human or bovine serum albumin, gelatin or immunoglobulins, preferably of human origin; Colouring and flavouring agents; Sulfur-containing reducing agents, such as glutathione, thioctic acid, sodium thioglycolate, thioglycerol, [α]-monothioglycerol and sodium thiosulfate · Diluents; ·emulsifier; -Hydrophilic polymers such as polyvinylpyrrolidone) · Salt-forming counterions such as sodium; preservatives, such as, for example, antimicrobials, antioxidants, chelating agents, inert gases, and the like; examples are benzalkonium chloride, benzoic acid, salicylic acid, thimerosal, phenethyl alcohol, methylparaben, propylparaben, chlorhexidine, sorbic acid or hydrogen peroxide); Metal complexes such as Zn-protein complexes; Solvents and co-solvents (such as glycine, propylene glycol or polyethylene glycol); sugars and sugar alcohols, such as trehalose, sucrose, octasulfate, mannitol, sorbitol or xylitol, stachyose, mannose, sorbose, xylose, ribose, myoinisitose, galactose, lactitol, ribitol, myoinisitol, galactitol, glycerol, cyclitols (e.g. inositol), polyethylene glycol; and polyhydric sugar alcohols; ·Suspending agents; Surfactants or wetting agents, such as pluronics, PEG, sorbitan esters, polysorbates, such as polysorbate 20, polysorbate, triton, tromethamine, lecithin, cholesterol, tyloxapal; surfactants can be detergents, preferably with a molecular weight of >1.2 KD and / or polyethers, preferably with a molecular weight of >3 KD; non-limiting examples of preferred detergents are Tween 20, Tween 40, Tween 60, Tween 80 and Tween 85; non-limiting examples of preferred polyethers are PEG 3000, PEG 3350, PEG 4000 and PEG 5000; ·Stability enhancers such as sucrose or sorbitol; isotonicity enhancing agents, such as alkali metal halides, preferably sodium chloride or potassium chloride, mannitol, sorbitol; Parenteral delivery vehicles including sodium chloride solution, Ringer's dextrose, dextrose and sodium chloride, lactated Ringer's solution, or fixed oils; Intravenous delivery vehicles, including fluid and nutrient replenishers, electrolyte replenishers (such as those based on Ringer's dextrose).

[0252] It will be apparent to one skilled in the art that different components of a pharmaceutical composition (e.g., those listed above) may have different effects, e.g., amino acids can act as buffers, stabilizers and / or antioxidants; mannitol can act as a bulking agent and / or tonicity enhancer; sodium chloride can act as a delivery vehicle and / or tonicity enhancer, etc.

[0253] It is envisaged that the compositions of the invention may contain, in addition to the polypeptides of the invention as defined herein, further biologically active agents depending on the intended use of the composition. Such agents may be drugs acting on the gastrointestinal system, drugs acting as cytostatic agents, drugs preventing hyperurikemia, drugs inhibiting immune responses (e.g. corticosteroids), drugs modulating inflammatory responses, drugs acting on the circulatory system and / or cytokines, etc., known in the art. It is also envisaged that the antibody constructs of the invention are applied in combination therapy, i.e. in combination with another anti-cancer drug.

[0254] In certain embodiments, the optimal pharmaceutical composition will be determined by those skilled in the art depending, for example, on the intended route of administration, delivery format, and desired dosage. See, for example, REMINGTON'S PHARMACEUTICAL SCIENCES, supra. In certain embodiments, such compositions may affect the physical state, stability, in vivo release rate, and in vitro clearance rate of the antibody construct of the present invention. In certain embodiments, the primary vehicle or carrier in the pharmaceutical composition may be aqueous or non-aqueous in nature. For example, a suitable vehicle or carrier may be water for injection, saline solution, or artificial cerebrospinal fluid, optionally supplemented with other materials common in compositions for parenteral administration. Neutral buffered saline or saline mixed with serum albumin are further exemplary vehicles. In certain embodiments, the antibody construct of the composition of the present invention may be prepared for storage in the form of a lyophilized cake or aqueous solution by mixing the selected composition having the desired degree of purity, optionally with a compounding agent (REMINGTON'S PHARMACEUTICAL SCIENCES, supra). Further, in certain embodiments, the antibody constructs of the present invention may be formulated as a lyophilizate using appropriate excipients such as sucrose.

[0255] When parenteral administration is intended, the therapeutic composition for use in the present invention may be provided in the form of a pyrogen-free, parenterally acceptable aqueous solution containing the desired antibody construct of the present invention in a pharma- ceutically acceptable vehicle. A particularly suitable vehicle for parenteral injection is sterile distilled water, in which the antibody construct of the present invention is formulated as a sterile, isotonic solution, appropriately preserved. In certain embodiments, the formulation may include a formulation of the desired molecule with an agent that can provide controlled or sustained release of a product that can be delivered via depot injection, such as injectable microspheres, bioerodible particles, polymeric compounds (such as polylactic acid or polyglycolic acid), beads, or liposomes. In certain embodiments, hyaluronic acid, which has the effect of enhancing duration in the circulation, may also be used. In certain embodiments, an implantable drug delivery device may be used to introduce the desired antibody construct.

[0256] Further pharmaceutical compositions will be evident to those skilled in the art, including formulations comprising the antibody constructs of the present invention in sustained or controlled delivery / release formulations. Techniques for formulating various other sustained or controlled delivery means, such as liposome carriers, biodegradable microparticles or porous beads, and depot injections, are also known to those skilled in the art. See, for example, International Patent Application No. PCT / US 93 / 00829, which describes controlled release of porous polymeric microparticles for delivery of pharmaceutical compositions. Sustained release formulations may include semipermeable polymer matrices in the form of shaped articles, such as films or microcapsules. The sustained release matrix may comprise polyesters, hydrogels, polylactides (as disclosed in U.S. Pat. No. 3,773,919, EP 058481), copolymers of L-glutamic acid and gamma-ethyl-L-glutamate (Sidman et al., 1983, Biopolymers 2:547-556), poly(2-hydroxyethyl-methacrylate) (Langer et al., 1981, J. Biomed. Mater. Res. 15:167-277 and Langer, 1982, Chem. Tech. 12:98-105), ethylene vinyl acetate (Langer et al., 1981, supra), or poly-D(-)-3-hydroxybutyric acid (EP 133,988). The sustained release composition may also comprise liposomes, which may be prepared by any of several methods known in the art. See, e.g., Eppstein et al., 1985, Proc. Natl. Acad. Sci. USA 82:3688-3692; European Patent Applications Publication Nos. 036,676; 088,046 and 143,949, which are incorporated by reference.

[0257] The antibody constructs may also be encapsulated in microcapsules (e.g., hydroxymethylcellulose or gelatin microcapsules and poly(methyl methacrylate) microcapsules, respectively), colloidal drug delivery systems (e.g., liposomes, albumin microspheres, microemulsions, nanoparticles and nanocapsules) or macroemulsions prepared, for example, by coacervation techniques or by interfacial polymerization. Such techniques are described in Remington's Pharmaceutical Sciences, 16 th edition, Oslo, A. Ed. (1980).

[0258] Pharmaceutical compositions used for in vivo administration are generally provided as sterile preparations. Sterilization can be achieved by filtration through sterile filtration membranes. When the composition is lyophilized, sterilization using this method can be carried out either before or after lyophilization and reconstitution. Compositions for parenteral administration can be stored in lyophilized form or in solution. Parenteral compositions are generally filled into containers with sterile access ports, such as intravenous solution bags or vials with stoppers that can be pierced by hypodermic injection needles.

[0259] Another aspect of the invention includes the self-buffering antibody constructs of the formulations of the invention that can be used as pharmaceutical compositions, as described in International Patent Application WO 06138181 A2 (PCT / US 2006 / 022599). Various descriptions are available on protein stabilization and formulation materials and methods useful in this regard, see for example Arakawa et al., "Solvent interactions in pharmaceutical formulations", Pharm Res. 8(3):285-91 (1991); Kendrick et al. "Physical stabilization of proteins in aqueous solution" in: RATIONAL DESIGN OF STABLE PROTEIN FORMULATIONS: THEORY AND PRACTICE, Carpenter and Manning, eds. Pharmaceutical Biotechnology. 13:61-84 (2002), and Randolph et al., "Surfactant-protein interactions", Pharm Biotechnol. 13:159-75 (2002), particularly with regard to protein pharmaceuticals and processes for veterinary and / or human medical use, in particular the parts relating to the same excipients and processes as the self-buffering protein formulations according to the present invention.

[0260] Salts may be used according to certain embodiments of the invention, for example, to adjust the ionic strength and / or tonicity of the formulation and / or to improve the solubility and / or physical stability of proteins or other components of the composition according to the invention. As is well known, ions can stabilize proteins in the native state by binding to charged residues on the surface of the protein and by shielding charged and polar groups in the protein, reducing the strength of their electrostatic, attractive and repulsive interactions. Ions can also stabilize proteins in the denatured state, in particular by binding to modified peptide bonds (--CONH) of the protein. Furthermore, ionic interactions with charged and polar groups in the protein can also reduce intermolecular electrostatic interactions, thereby preventing or reducing protein aggregation and insolubilization.

[0261] Ionic species vary significantly in their effects on proteins. A number of categorical rankings of ions and their effects on proteins have been developed and can be used in formulating pharmaceutical compositions according to the invention. One example is the Hofmeister series, which ranks ionic and polar nonionic solutes by their effect on the conformational stability of proteins in solution. Stabilizing solutes are called "kosmotropics". Destabilizing solutes are called "chaotropics". Kosmotropes are commonly used at high concentrations (e.g., >1 molar ammonium sulfate) to precipitate ("salt out") proteins from solution. Chaotropes are commonly used to denature and / or solubilize ("salt in") proteins. The relative effects of an ion on "salting in" and "salting out" define the ion's position on the Hofmeister series.

[0262] Free amino acids can be used in the antibody constructs of the formulations of the invention according to various embodiments of the invention as bulking agents, stabilizers and antioxidants, as well as other standard uses. Lysine, proline, serine and alanine can be used to stabilize proteins in the formulation. Glycine is useful for ensuring proper cake structure and properties in lyophilization. Arginine can be useful for inhibiting protein aggregation in both liquid and lyophilized formulations. Methionine is useful as an antioxidant.

[0263] Polyols include sugars such as mannitol, sucrose and sorbitol, and polyhydric alcohols such as glycerol and propylene glycol, and for purposes of discussion herein, polyethylene glycol (PEG) and related substances. Polyols are kosmotropic. They are useful stabilizers to protect proteins from physical and chemical degradation processes in both liquid and lyophilized formulations. Polyols are also useful to adjust the isotonicity of the formulation. Among the polyols useful in selected embodiments of the present invention is mannitol, which is commonly used in lyophilized formulations to ensure structural stability of the cake. Mannitol ensures structural stability of the cake. Generally, it is used in conjunction with a lyoprotectant such as sucrose. Sorbitol and sucrose are among the preferred agents for adjusting isotonicity and stabilizers for protection from freeze-thaw stress during transportation or bulk preparation in the manufacturing process. Reducing sugars (containing free aldehyde or ketone groups), such as glucose and lactose, can glycate surface lysine and arginine residues. Therefore, they are generally not included among the preferred polyols for use in the present invention.In addition, sugars that form such reactive species, such as sucrose, are also not included among the preferred polyols of the present invention, in that they are hydrolyzed to fructose and glucose under acidic conditions, resulting in glycation.PEG is useful for stabilizing proteins and as a cryoprotectant, and in this regard can be used in the present invention.

[0264] The antibody construct embodiment of the formulation of the present invention further comprises a surfactant. Protein molecules can be prone to surface adsorption and denaturation and resulting aggregation at air-liquid, solid-liquid and liquid-liquid interfaces. These effects are generally inversely proportional to protein concentration. These adverse interactions are generally inversely proportional to protein concentration and are typically exacerbated by physical agitation, such as that encountered during product transportation and handling. Surfactants are routinely used to prevent, minimize or reduce surface adsorption. Surfactants useful in the present invention in this regard include polysorbate 20, polysorbate 80, other fatty acid esters of sorbitan polyethoxylate and poloxamer 188. Surfactants are also commonly used to control protein conformational stability. In this regard, the use of surfactants is protein specific, since any given surfactant typically stabilizes some proteins and destabilizes others.

[0265] Polysorbates are prone to oxidative degradation and often contain sufficient peroxide content as supplied to cause oxidation of the side chains of protein residues, especially methionine. Polysorbates should therefore be used with caution and, when used, at the lowest possible concentrations. In this respect, polysorbates exemplify the general rule that excipients should be used at the lowest possible concentrations.

[0266] The antibody construct embodiment of the formulation of the present invention further comprises one or more antioxidants. By maintaining an appropriate level of ambient oxygen and temperature, and by avoiding exposure to light, harmful oxidation of proteins in pharmaceutical formulations can be prevented to some extent. Antioxidant excipients can be used to prevent oxidative degradation of proteins as well. Antioxidants that are particularly useful in this regard include reducing agents, oxygen / free radical scavengers and chelating agents. Antioxidants for use in therapeutic protein formulations according to the present invention are preferably water-soluble and maintain activity over the shelf life of the product. In this regard, EDTA is a preferred antioxidant according to the present invention. Antioxidants can damage proteins. For example, reducing agents such as glutathione can, inter alia, disrupt intramolecular disulfide bonds. Thus, antioxidants for use in the present invention are selected, inter alia, to eliminate or sufficiently reduce the possibility of damaging proteins in the formulation themselves.

[0267] The formulations according to the invention may contain metal ions, which are cofactors for proteins and are required to form protein coordination compounds, such as zinc, which is required to form certain insulin suspensions. Metal ions can also inhibit some processes that break down proteins. However, metal ions also catalyze physical and chemical processes that break down proteins. Magnesium ions (10-120 mM) can be used to inhibit the isomerization of aspartic acid to isoaspartic acid. Ca +2 ions (up to 100 mM) can increase the stability of human deoxyribonuclease. +2 , Mn +2 and Zn +2 can destabilize rhDNase. +2 and Sr +2 can stabilize factor VIII, which is +2 , Mn +2 and Zn +2 , Cu +2 and Fe +2The agglomeration can be destabilized by Al +3 It can be increased by ions.

[0268] The antibody construct embodiment of the formulation of the present invention further comprises one or more preservatives. Preservatives are necessary when developing a multi-dose parenteral formulation with more than one withdrawal from the same container. Its main function is to inhibit microbial growth and ensure sterility of the product over the shelf life or use period of the formulation. Commonly used preservatives include benzyl alcohol, phenol, and m-cresol. Although preservatives have a long history of use with small molecule parenteral drugs, developing protein formulations containing preservatives can be difficult. Preservatives almost always have a destabilizing effect on proteins (aggregation), which is the main factor limiting their use in multi-dose protein formulations. To date, most protein drugs have only been formulated for single use. However, if a multi-dose formulation were possible, it would add the benefits of patient convenience and increased marketability. A good example is human growth hormone (hGH), where the development of a preserved formulation led to the development of a more convenient multi-use injection pen for commercialization. At least four such pen devices containing preserved formulations of hGH are currently available on the market: Norditropin (liquid, Novo Nordisk), Nutropin AQ (liquid, Genentech), and Genotropin (lyophilized-dual chamber cartridge, Pharmacia & Upjohn) contain phenol, while Somatrope (Eli Lilly) is formulated with m-cresol. During the formulation and development of preserved dosage forms, several aspects must be considered: The effective preservative concentration in the drug product must be optimized. This requires testing a given preservative in the dosage form for a concentration range that confers antimicrobial efficacy without compromising protein stability.

[0269] As expected, developing liquid formulations containing preservatives is more challenging than lyophilized formulations. Freeze-dried products can be lyophilized without preservatives and reconstituted with a diluent containing the preservative at the time of use. This reduces the time the preservative is in contact with the protein and greatly minimizes the associated stability risks. For liquid formulations, the efficacy and stability of the preservative should be maintained throughout the entire product shelf life (approximately 18-24 months). It is important to note that the efficacy of the preservative needs to be demonstrated in the final formulation containing the active drug and all excipient components.

[0270] The antibody constructs disclosed herein may also be formulated as liposomes. "Liposomes" are small vesicles composed of various types of lipids, phospholipids and / or surfactants that are useful for drug delivery to mammals. The components of the liposome are generally arranged in a bilayer structure similar to the lipid arrangement of biological membranes. Liposomes containing antibody constructs are prepared by methods known in the art, for example, as described in Epstein et al., Proc. Natl. Acad. Sci. USA, 82:3688 (1985); Hwang et al., Proc. Natl Acad. Sci. USA, 77:4030 (1980); U.S. Pat. Nos. 4,485,045 and 4,544,545; and WO 97 / 38731. Liposomes with enhanced circulation time are disclosed in U.S. Pat. No. 5,013,556. Particularly useful liposomes can be made by reverse phase evaporation using a lipid composition comprising phosphatidylcholine, cholesterol and PEG-derivatized phosphatidylethanolamine (PEG-PE). Liposomes are extruded through filters of defined pore size to obtain liposomes with the desired diameter. Fab' fragments of the antibody constructs of the present invention can be conjugated to liposomes by disulfide exchange reaction as described in Martin et al. J. Biol. Chem. 257:286-288 (1982). Optionally, a chemotherapeutic agent is contained within the liposome. See Gabizon et al. J. National Cancer Inst. 81(19)1484 (1989).

[0271] After the pharmaceutical composition has been formulated, it may be stored in sterile vials as a solution, suspension, gel, emulsion, solid, crystal, or as a dehydrated or lyophilized powder. Such formulations may be stored in a ready-to-use form or in a form (e.g., lyophilized) that is reconstituted prior to administration.

[0272] The biological activity of the pharmaceutical compositions defined herein may be determined, for example, by cytotoxicity assays as described in the Examples below, in WO 99 / 54440, or in Schlereth et al. (Cancer Immunol. Immunother. 20 (2005), 1-12). As used herein, "efficacy" or "in vivo efficacy" refers to the response to therapy with the pharmaceutical composition of the present invention, for example, using standardized NCI response criteria. The success of therapy with the pharmaceutical composition of the present invention or in vivo efficacy refers to the efficacy of the composition for its intended purpose, i.e., the ability of the composition to cause its desired effect, i.e., depletion of pathological cells, e.g., tumor cells. In vivo efficacy may be monitored by established standard methods for the respective disease entity, including, but not limited to, white blood cell counts, differential, fluorescence activated cell sorting, bone marrow aspiration. In addition, various disease-specific clinical chemistry parameters and other established standard methods may be used.In addition, computed tomography, x-ray, and nuclear magnetic resonance imaging (e.g., response assessment based on the National Cancer Institute criteria [Cheson BD, Horning SJ, Coiffier B, Shipp MA, Fisher RI, Connors JM, Lister TA, Vose J, Grillo-Lopez A, Hagenbeek A, Cabanillas F, Klippensten D, Hiddemann W, Castellino R, Harris NL, Armitage JO, Carter W, Hoppe R, Canellos GP. Report of an international workshop to standardize response criteria for non-Hodgkin's lymphomas. NCI Sponsored International Working Group. J Clin Oncol. 1999 Apr;17(4):1244]), positron emission tomography scanning, white blood cell count, differential, fluorescence activated cell sorting, bone marrow aspiration, lymph node biopsy / histology and various lymphoma-specific clinical chemistry parameters (e.g., lactate dehydrogenase), as well as other established standard methods may be used.

[0273] Another major challenge in the development of drugs, such as the pharmaceutical compositions of the present invention, is the predictable modulation of pharmacokinetic properties. To this end, a pharmacokinetic profile of a drug candidate can be established, i.e., a profile of pharmacokinetic parameters that affect the ability of a particular drug to treat a given pathology. Pharmacokinetic parameters of drugs that affect the ability of a drug to treat a certain disease include, but are not limited to, half-life, volume of distribution, hepatic first-pass metabolism, and degree of serum binding. The efficacy of a given drug can be influenced by each of the above parameters. A presumed feature of the antibody constructs of the present invention provided by a particular FC format is that they include differences, for example, in pharmacokinetic behavior. The half-life extended targeting antibody constructs of the present invention preferably exhibit a surprisingly increased residence time in vivo compared to the "canonical" non-HLE version of said antibody construct.

[0274] "Half-life" refers to the time it takes for 50% of an administered drug to be eliminated through biological processes, such as metabolism, excretion, etc. "Hepatic first-pass metabolism" refers to the tendency of a drug to be metabolized upon first contact with, i.e., during its first passage through, the liver. "Volume of distribution" refers to the degree of retention of a drug across various compartments of the body, such as, for example, intracellular and extracellular spaces, tissues and organs, and the distribution of the drug within these compartments. "Extent of serum binding" refers to the tendency of a drug to interact with and bind to serum proteins, such as albumin, resulting in a reduction or elimination of the drug's biological activity.

[0275] Pharmacokinetic parameters also include bioavailability, lag time (Tlag), Tmax, absorption rate, onset of more action and / or Cmax for a given amount of drug administered. "Bioavailability" refers to the amount of drug in the blood compartment. "Lag time" refers to the time delay from administration of a drug until it is detectable and measurable in the blood or plasma. "Tmax" is the time after which the maximum blood concentration of the drug is reached, and "Cmax" is the maximum blood concentration obtained with a given drug. All parameters affect the time it takes for a drug to reach the blood or tissue concentration required for a biological effect. Pharmacokinetic parameters of bispecific antibody constructs exhibiting cross-species specificity that can be determined in preclinical animal studies in non-chimpanzee primates as outlined above are also provided, for example, in the publication by Schlereth et al. (Cancer Immunol. Immunother. 20 (2005), 1-12).

[0276] In a preferred embodiment of the invention, the pharmaceutical composition is stable for at least 4 weeks at about -20°C. As is evident from the accompanying examples, the quality of the antibody constructs of the invention relative to the quality of the corresponding state-of-the-art antibody constructs can be tested using different systems. These tests are understood to be in accordance with the "ICH Harmonised Tripartite Guideline: Stability Testing of Biotechnological / Biological Products Q5C and Specifications: Test procedures and Acceptance Criteria for Biotech Biotechnological / Biological Products Q6B" and are selected to provide a stability indicator profile that allows reliable detection of changes in the identity, purity and potency of the product. It is well accepted that the term purity is a relative term. Due to the effects of glycosylation, deamidation or other heterogeneity, the absolute purity of a biotechnological / biological product should usually be assessed by more than one method and the purity values ​​obtained are method-dependent. For the purpose of stability testing, the tests of purity should be aligned with the method of determination of degradation products.

[0277] To assess the quality of a pharmaceutical composition comprising an antibody construct of the present invention, it can be analyzed, for example, by analyzing the content of soluble aggregates in solution (HMWS by size exclusion). Stability at about -20°C for at least 4 weeks is preferably characterized by a content of less than 1.5% HMWS, preferably less than 1% HMWS.

[0278] A preferred formulation for the antibody construct as a pharmaceutical composition may, for example, include the components of the formulation as described below. ·formulation: Potassium phosphate pH 6.0, L-arginine hydrochloride, trehalose, polysorbate 80

[0279] Other examples of the evaluation of the stability of the antibody construct of the invention in the form of a pharmaceutical composition are provided in the accompanying Examples 4-12. In those examples, embodiments of the antibody construct of the invention are tested for different stress conditions in different pharmaceutical formulations and the results are compared with other bispecific T cell engaging antibody constructs in half-life extension (HLE) formats known in the art. In general, it is assumed that the antibody constructs provided in a particular FC format according to the invention are more stable against a wide range of stress conditions, such as temperature and light stress, compared with both antibody constructs provided with different HLE formats and antibody constructs without any HLE format (e.g., "canonical" antibody constructs). Said temperature stability may relate to both low temperatures (below room temperature, including freezing temperatures) and high temperatures (above room temperature, including temperatures up to or above body temperature). As the skilled person will appreciate, such improved stability against stresses that are difficult to avoid in clinical practice makes the antibody construct safer in clinical practice, since it produces fewer degradation products. As a result, said improved stability means improved safety.

[0280] One embodiment provides an antibody construct of the invention or an antibody construct produced according to the method of the invention for use in the prevention, treatment or amelioration of a proliferative disease, a neoplastic disease, a viral disease or an immune disorder.

[0281] The formulations described herein are useful as pharmaceutical compositions for treating, ameliorating and / or preventing the pathological medical conditions described herein in patients in need thereof. The term "treatment" refers to both therapeutic treatment and prophylactic or preventative measures. Treatment includes application or administration of the formulation to the body, isolated tissues or cells of a patient having a disease / disorder, a symptom of a disease / disorder or a predisposition to a disease / disorder with the intent to cure, remedy, relieve, alleviate, alter, correct, ameliorate, improve or affect the disease, symptom of a disease or predisposition to a disease.

[0282] As used herein, the term "amelioration" refers to any improvement in the disease state of a patient having a tumor or cancer or metastatic cancer as defined herein below, by administration of an antibody construct according to the present invention to a subject in need thereof. Such improvement may be seen as a slowing or halting of progression of the patient's tumor or cancer or metastatic cancer. As used herein, the term "prevention" refers to the avoidance of the onset or recurrence of a patient having a tumor or cancer or metastatic cancer as defined herein below, by administration of an antibody construct according to the present invention to a subject in need thereof.

[0283] The term "disease" refers to any condition that would benefit from treatment with an antibody construct or pharmaceutical composition described herein, including chronic and acute disorders or diseases, including pathological conditions that predispose a mammal to the disease in question.

[0284] A "neoplasm" is an abnormal growth of tissue, usually, but not necessarily, forming a mass. When it forms a mass, it is also commonly called a "tumor". A neoplasm or tumor can be benign, potentially malignant (precancerous) or malignant. Malignant neoplasms are generally called cancers. They usually invade and destroy surrounding tissues and can form metastases, i.e., they spread to other parts, tissues or organs of the body. Thus, the term "metastatic cancer" encompasses metastases to other tissues or organs other than that of the primary tumor. Lymphomas and leukemias are lymphatic neoplasms. For the purposes of the present invention, they are also encompassed by the terms "tumor" or "cancer".

[0285] The term "viral disease" refers to a disease that is the result of infection of a subject with a virus.

[0286] As used herein, the term "immune disorder" refers to immune disorders such as autoimmune diseases, hypersensitivity disorders, and immune deficiencies in accordance with the general definition of the term.

[0287] In one embodiment, the present invention provides a method for the treatment or amelioration of a proliferative disease, a neoplastic disease, a viral disease or an immune disorder, comprising the step of administering to a subject in need thereof an antibody construct of the present invention or an antibody construct produced according to a method of the present invention.

[0288] The term "subject in need" or "subject in need of treatment" includes subjects already with the disorder as well as subjects in which the disorder is to be prevented. A subject in need or "patient" includes human and other mammalian subjects receiving either prophylactic or therapeutic treatment.

[0289] The antibody constructs of the present invention will generally be designed for a particular route and method of administration, a particular dosage and frequency of administration, and a particular treatment of a particular disease, particularly in the areas of bioavailability and duration. The materials of the composition are preferably formulated in concentrations that are acceptable to the site of administration.

[0290] Thus, formulations and compositions may be designed for delivery by any suitable route of administration in accordance with the present invention. Topical route (e.g. on the skin, inhalation, nose, eyes, auricle / ear, vagina, mucous membranes); Enteral routes (e.g., oral, gastrointestinal, sublingual, sublabial, buccal, rectal); and Parenteral routes (e.g., intravenous, intraarterial, intraosseous, intramuscular, intracerebral, intraventricular, epidural, intrathecal, subcutaneous, intraperitoneal, extra-amniotic, intra-articular, intracardiac, intradermal, intralesional, intrauterine, intravesical, intravitreal, transdermal, intranasal, transmucosal, intrasynovial, intraluminal).

[0291] The pharmaceutical compositions and antibody constructs of the invention are particularly useful for parenteral administration, e.g., subcutaneous or intravenous delivery, e.g., by injection, such as a bolus injection, or by infusion, such as continuous infusion. The pharmaceutical composition may be administered using a medical device. Examples of medical devices for administering pharmaceutical compositions are described in U.S. Pat. Nos. 4,475,196; 4,439,196; 4,447,224; 4,447,233; 4,486,194; 4,487,603; 4,596,556; 4,790,824; 4,941,880; 5,064,413; 5,312,335; 5,312,335; 5,383,851; and 5,399,163.

[0292] In particular, the present invention provides for uninterrupted administration of suitable compositions. As a non-limiting example, uninterrupted or substantially uninterrupted, i.e. continuous administration, may be achieved by a miniature pump system worn by the patient to regulate the influx of therapeutic agent into the patient's body. The pharmaceutical composition comprising the antibody construct of the present invention may be administered by using said pump system. Such pump systems are generally known in the art and typically rely on periodic replacement of a cartridge containing the therapeutic agent to be infused. When replacing the cartridge in such a pump system, a temporary interruption in the otherwise uninterrupted influx of therapeutic agent into the patient's body may result. Even in such a case, the administration step before the cartridge replacement and the administration step after the cartridge replacement would still be considered within the meaning of the pharmaceutical means and methods of the present invention, which together constitute the "uninterrupted administration" of such therapeutic agent.

[0293] The continuous or uninterrupted administration of the antibody construct of the present invention may be intravenous or subcutaneous administration by a fluid delivery device or a miniature pump system, which includes a fluid delivery mechanism for pumping fluid from a reservoir and a drive mechanism for driving the delivery mechanism. The pump system for subcutaneous administration may include a needle or cannula for penetrating the patient's skin and delivering the suitable composition into the patient's body. The pump system may be fixed or attached directly to the patient's skin, whether vein, artery or blood vessel, allowing direct contact between the pump system and the patient's skin. The pump system may be attached to the patient's skin for 24 hours to several days. In some cases, the pump system may be a miniature pump system with a small reservoir volume. As a non-limiting example, the reservoir volume for the suitable pharmaceutical composition to be administered may be 0.1 to 50 ml.

[0294] Continuous administration can also be transdermal, by a patch that is attached to the skin and replaced from time to time.Those skilled in the art are aware of suitable patch systems for drug delivery for this purpose.It is noted that transdermal administration is particularly suitable for uninterrupted administration, since it has the advantage that, for example, a new second patch can be attached to the skin surface directly next to the first used patch, and the replacement of the first used patch can be completed at the same time just before removing the first used patch.There is no problem of interruption of inflow or battery failure.

[0295] If the pharmaceutical composition is lyophilized, the lyophilized material is first reconstituted with an appropriate liquid prior to administration. The lyophilized material can be reconstituted, for example, with bacteriostatic water for injection (BWFI), saline, phosphate buffered saline (PBS), or the same formulation the protein was in before lyophilization.

[0296] The composition of the present invention can be administered to a subject at a suitable dose, which can be determined, for example, by a dose escalation study, by administering increasing doses of the antibody construct of the present invention, which exhibits cross-species specificity as described herein, to a non-chimpanzee primate, such as a macaque. As mentioned above, the antibody construct of the present invention, which exhibits cross-species specificity as described herein, has the advantage that it can be used in the same form in preclinical trials in non-chimpanzee primates and as a drug in humans. The administration schedule will be determined by the attending physician based on clinical factors. As is well known in the medical field, the dosage for a given patient depends on many factors, including the patient's size, body surface area, age, the individual compound administered, sex, time and route of administration, general health, and other drugs administered at the same time.

[0297] The term "effective dose" or "effective dosage" is defined as an amount sufficient to achieve or at least partially achieve a desired effect. The term "therapeutically effective dose" is defined as an amount sufficient to cure or at least partially arrest the disease and its complications in a patient already suffering from the disease. The amount or dosage effective for this use will depend on the condition (indication) being treated, the antibody construct being delivered, the nature and purpose of the treatment, the severity of the disease, previous treatments, the patient's medical history and responsiveness to the therapeutic agent, the route of administration, the size (weight, body surface area or organ size) and / or condition (age and general health) of the patient, and the general condition of the patient's own immune system. The appropriate dosage can be administered to the patient in a single dose or multiple doses and can be adjusted according to the judgment of the attending physician to obtain the optimal therapeutic effect.

[0298] Typical dosages can range from about 0.1 μg / kg up to about 30 mg / kg or more, depending on the factors mentioned above. In certain embodiments, dosages can range from 1.0 μg / kg up to about 20 mg / kg, optionally 10 μg / kg up to about 10 mg / kg, or 100 μg / kg up to about 5 mg / kg.

[0299] A therapeutically effective amount of the antibody construct of the invention preferably reduces the severity of disease symptoms, increases the frequency or duration of disease symptom-free periods, or prevents impaired function or disability due to disease affliction. For the treatment of antigen-expressing tumors of target cells, a therapeutically effective amount of the antibody construct of the invention, e.g., an anti-target cell antigen / anti-CD3 antibody construct, preferably inhibits cell proliferation or tumor growth by at least about 20%, at least about 40%, at least about 50%, at least about 60%, at least about 70%, at least about 80%, or at least about 90% compared to untreated patients. The ability of a compound to inhibit tumor growth can be assessed in animal models predictive of efficacy.

[0300] The pharmaceutical composition can be administered in a single treatment or, if necessary, in combination with additional treatments, such as anti-cancer therapies, e.g., other proteinaceous and non-proteinaceous drugs, which can be administered simultaneously with the composition comprising the inventive antibody construct as defined herein, or can be administered separately at defined time intervals and dosages before or after administration of said antibody construct.

[0301] As used herein, the term "effective and non-toxic dose" refers to a tolerated dose of the antibody construct of the present invention that is sufficient to cause depletion of pathological cells, tumor elimination, tumor regression or disease stabilization without or essentially without significant toxic effects. Such an effective and non-toxic dose can be determined, for example, by dose escalation studies as described in the art, and should be below the dose that induces serious adverse side events (dose-limiting toxicity, DLT).

[0302] As used herein, the term "toxicity" refers to the toxic effects of a drug that manifest as adverse events or serious adverse events. These side events may refer to a lack of systemic drug tolerance and / or a lack of local tolerance after administration. Toxicity may also include teratogenic or carcinogenic effects caused by the drug.

[0303] As used herein, the term "safety", "in vivo safety" or "tolerability" is defined as the administration of a drug that does not induce serious adverse events immediately after administration (local tolerance) and during longer drug application periods. "Safety", "in vivo safety" or "tolerability" can be evaluated, for example, periodically during treatment and follow-up periods. Measurements include clinical evaluations, such as organ findings and screening for laboratory abnormalities. Clinical evaluations can be performed and deviations from normal findings can be recorded / coded according to NCI-CTC and / or MedDRA standards. Organ findings can include criteria such as allergy / immunology, blood / bone marrow, cardiac arrhythmias, coagulation, etc., as set forth in the Common Terminology Criteria for adverse events v3.0 (CTCAE). Laboratory parameters that can be tested include, for example, hematology, clinical chemistry, coagulation profile and urinalysis, as well as tests of other body fluids, such as serum, plasma, lymph or spinal fluid. Thus, safety can be assessed, for example, by physical examination, imaging techniques (i.e. ultrasound, x-ray, CT scan, magnetic resonance imaging (MRI), other measurements using technical devices (i.e. electrocardiogram), vital signs, measuring examination parameters and recording adverse events. For example, in the uses and methods according to the invention, adverse events in non-chimpanzee primates can be examined by histopathological and / or histochemical methods.

[0304] The above terms are also referenced, for example, in Preclinical safety evaluation of biotechnology-derived pharmaceuticals S6; ICH Harmonised Tripartite Guideline; ICH Steering Committee meeting of July 16, 1997.

[0305] Finally, the present invention provides kits comprising the antibody construct of the invention, the antibody construct produced according to the method of the invention, the pharmaceutical composition of the invention, the polynucleotide of the invention, the vector of the invention and / or the host cell of the invention.

[0306] In the context of the present invention, the term "kit" refers to two or more components packaged together in a container, vessel or other package, one of the components corresponding to an antibody construct, pharmaceutical composition, vector or host cell of the present invention. A kit can thus be described as a set of products and / or implements sufficient to accomplish a particular purpose that can be sold as a single item.

[0307] The kit may comprise one or more containers (e.g., vials, ampoules, containers, syringes, bottles, bags) of any suitable shape, size and material (preferably waterproof, e.g., plastic or glass) containing an antibody construct or pharmaceutical composition of the invention in a dosage amount suitable for administration (see above). The kit may further comprise instructions for use (e.g., in the form of a leaflet or instruction manual), a means for administering the antibody construct of the invention, e.g., a syringe, pump, infuser, etc., a means for reconstituting the antibody construct of the invention and / or a means for diluting the antibody construct of the invention.

[0308] The present invention also provides kits for single-dose administration units. The kits of the present invention may also include a first container containing a dried / lyophilized antibody construct and a second container containing an aqueous formulation. In certain embodiments of the present invention, kits are provided that include single-chamber and multi-chamber pre-filled syringes (e.g., liquid syringes and lyosyringes).

[0309] It should be noted that, as used herein, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to "reagents" includes one or more of such different reagents, and reference to "methods" includes reference to equivalent steps and methods known to those of skill in the art that may be modified for or substituted for the methods described herein.

[0310] Unless otherwise indicated, the term "at least" preceding a series of elements should be understood to refer to every element in the series. Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein. Such equivalents are intended to be encompassed by this invention.

[0311] The term "and / or", wherever used in this specification, includes the meanings "and", "or" and "any other combination of the elements connected by said term".

[0312] As used herein, the term "about" or "approximately" means within 20%, preferably within 10%, and more preferably within 5% of a given value or range, however, the term also includes specific numbers, for example, about 20 includes 20.

[0313] The terms "less than" or "greater than" include a specific number. For example, less than 20 means less than or equal to. Similarly, greater than or greater than means greater than or equal to, or greater than or equal to, respectively.

[0314] Throughout this specification and the claims which follow, unless the context requires otherwise, the word "comprise", and variations such as "comprises" and "comprising", should be understood to mean the inclusion of a stated integer or step or group of integers or steps, but not the exclusion of any other integer or step or group of integers or steps. As used herein, the term "comprise" can also be replaced with the terms "containing" or "including" or, when used herein, the term "having".

[0315] As used herein, "consisting of" excludes any component, step, or ingredient not specified in the claim component. As used herein, "consisting essentially of" does not exclude materials or steps that do not materially affect the basic and novel characteristics of the claim.

[0316] In each instance herein, any of the terms "comprising," "consisting essentially of," and "consisting of" may be replaced with one of the other two terms.

[0317] It is to be understood that this invention is not limited to the particular methodology, protocols, materials, reagents, and substances, etc. described herein and as such may vary. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of the invention, which is defined solely by the claims.

[0318] All publications and patents cited throughout the text of this specification, whether supra or infra, including all patents, patent applications, scientific publications, manufacturer's specifications, instructions, etc., are hereby incorporated by reference in their entirety. Nothing herein should be construed as an admission that the invention is not entitled to antedate such disclosure by virtue of prior invention. To the extent that material incorporated by reference is inconsistent or inconsistent with the present specification, the present specification will take precedence over any such material.

[0319] A better understanding of the present invention and its advantages will be obtained from the following examples, which are provided for illustrative purposes only and are not intended to limit the scope of the invention in any way. EXAMPLES

[0320] Example 1: Determination of optimal refolding conditions for CD33xCD3 BiTE® antibody construct Materials and Methods: CD33xCD3xscFc bispecific antibody construct (ScFc-BiTE®; SEQ ID NO: 41) was used in this study. ScFc-BiTE® material was purified according to the steps in Figure 1. Protein A filtered viral inactivated pool (FVIP) and CEX HMW pool samples were used for redox refolding.

[0321] Size exclusion chromatography (SEC) was used to measure the percentage of monomeric and aggregate species of ScFc-BiTE® and included the following conditions: Column: G3000SWxl TOSOH Bioscience, 7.8 mm IDx30 cm column (catalog #08541), room temperature or 25°C. Mobile phase: 100 mM sodium phosphate, 250 mM NaCl, pH 6.8. Flow rate: 0.5 ml / min. Analysis time: 35 min. Unless otherwise stated, the refolding procedure was carried out at room temperature by stepwise dialysis using Thermo Scientific Slide-A-Lyzer™ dialysis cassettes (20k MWCO). After treatment with refolding buffer, samples were dialyzed overnight back into 100 mM NaOAc, pH 5.1, 4°C.

[0322] Results: A series of pH and guanidine concentrations were applied to increase the concentration of monomeric species in the ScFc-BiTE® CEX post-peak pool, which contains only 25-27% monomeric species. When a sample of the ScFc-BiTE® CEX post-peak pool was buffer exchanged with control buffer (100 mM NaOAc, pH 5.1), the percentage of monomer remained near 25%-27%. However, when guanidine was included in the refolding buffer at concentrations higher than 1 M and up to 2 M, the percentage of ScFc-BiTE® monomer increased in a concentration-dependent manner up to 60% of monomer. A detailed list of refolding conditions and the resulting percentage of monomer is reported in Table 4, and the effect of guanidine concentration during refolding is shown in the SEC chromatograms in Figure 2. To evaluate the effect of pH on refolding, refolding buffers containing 2.0 M and 2.5 M guanidine were adjusted to pH 4.2, 5.1, or 8.0 with acetate or Tris base. No significant differences were observed between pH 5.1 and pH 8.0. However, post-CEX peak ScFc-BiTE® samples aggregated when incubated at pH 4.2 (Table 4).

[0323] To further increase the percentage of monomer, a redox reagent was added to the refolding buffer along with guanidine. When cysteine ​​and cystine were added to the 1.2 M guanidine refolding buffer in molar ratios of 6:1, 14:1, and 33:1 (Cys:Cyss), the percentage of monomer increased from 39% (without redox reagent) to 52%, 61%, and 69%, respectively (Table 4). Therefore, the refolding process with Cys / Cyss deserves further investigation.

[0324] Oxidizing agents such as copper(II) sulfate and dehydroascorbic acid were evaluated. 0.5 mM copper(II) sulfate with 1.2 M or 2.0 M guanidine was found to have the greatest refolding effect increasing the monomer percentage to 90% and 88%, respectively (see Table 4, Figure 3). The monomer percentage was also increased by 0.5 mM dehydroascorbic acid to 66% and 73% at 1.2 M and 2.0 M guanidine, respectively.

[0325] To demonstrate that the refolding procedure reported above for the post-peak pool of scFc-BiTE® CEX is also suitable for initial downstream purification such as the Protein A pool, a Protein A FVIP sample containing approximately 73% monomer and 27% aggregates was subjected to refolding with 2.0 M guanidine. The percentage of monomer increased to 94% and 89% at pH 5.1 and pH 8.0, respectively (Table 5 and Figure 4).

[0326] [Table 4]

[0327] [Table 5]

[0328] Example 2: Determination of optimal refolding conditions for DLL3xCD3 antibody constructs on column The DLL3xCD3xscFc (SEQ ID NO: 104) bispecific antibody construct was tested for redox refolding effect on column. For this purpose, a ProA column was used with isocratic elution. The load is 10 g / Lr (per DLL3xCD3 bispecific antibody construct process) HCCF. Wash buffer conditions are 25 mM Tris+1.2 M guanidine, 2 M guanidine, copper(II) sulfate in 1.2 M guanidine*; control at 25 mM Tris+Arg condition*, all at pH 8.0. Wash with 2 column volumes of buffer; hold for 2 hours; collect wash pool separately. Elute protein at pH 3.7 (per DLL3xCD3 scFc-BiTE® process); sample. Viral inactivation: titrate to pH 3.6, hold for 60 minutes, neutralize with 2 M Tris base to pH 5.1; sample. Pools: wash pool; elution with pH 3.9 pool; neutralized viral inactivation pool. Analytical tools used were: SEC, CHOP, leached ProA.

[0329] Example 3: Determination of successful refolding conditions for CD19 and CLDN bispecific antibody constructs The CD19xCD3xscFc (SEQ ID NO: 114) and CLNDxCD3xscFc antibody constructs CEX peak post pools were refolded with 2.0 M guanidine and 0.5 mM copper(II) at pH 8.0 for 4 hours at room temperature and then buffer exchanged back into acetate buffer at pH 5.1. The percentage of monomer increased from 24.0% to 48.7% (i.e., an increase of about 100%) for the CD19xCD3xscFc antibody construct (see FIG. 5 and Table 6) and from 23.4% to 38.5% (i.e., an increase of about 64%) for the CLNDxCD3xscFc antibody construct (see FIG. 6 and Table 7). In some cases, the applied method of the invention significantly reduced aggregate formation and increased the monomer percentage. Also, improved homogeneity of the monomer structure (reduction of monomeric isoforms) was observed.

[0330] [Table 6]

[0331] [Table 7]

[0332] In addition, refolding experiments were also performed at pH 7.0. When incubated with 2.0 M guanidine and 0.5 mM copper(II) at pH 7.0, the monomeric isoform was reduced from 10.2% to 0% (see Table 8).

[0333] [Table 8]

[0334] [Table 9]

[0335] [Table 10]

[0336] [Table 11]

[0337] [Table 12]

[0338] [Table 13]

[0339] [Table 14]

[0340]

Table 15

[0341]

Table 16

[0342]

Table 17

[0343]

Table 18

[0344]

Table 19

[0345]

Table 20

[0346]

Table 21

[0347]

Table 22

[0348]

Table 23

[0349]

Table 24

[0350]

Table 25

[0351]

Table 26

[0352]

Table 27

[0353]

Table 28

[0354]

Table 29

[0355]

Table 30

[0356]

Table 31

[0357]

Table 32

[0358]

Table 33

[0359]

Table 34

[0360]

Table 35

[0361]

Table 36

[0362]

Table 37

[0363]

Table 38

[0364]

Table 39

[0365]

Table 40

[0366]

Table 41

[0367]

Table 42

[0368]

Table 43

[0369]

Table 44

[0370]

Table 45

[0371]

Table 46

Claims

1. A method for purifying a bispecific single chain antibody construct produced by a mammalian cell, said construct comprising at least three domains, in amino to carboxyl order: - a first domain with a pI of 5.0 to 9.5 that binds to CD33, CD19, DLL3, or CLDN18.2; - a second domain that binds to an extracellular epitope of the human and macaque (Macaca) CD3 epsilon chain, wherein said first domain and said second domain have the following format: the pair of VH and VL domains is in the format of a single chain antibody (scFv); and a third domain comprising two polypeptide monomers each comprising a hinge, a CH2 domain and a CH3 domain, wherein said two polypeptide monomers are fused to each other via a peptide linker; wherein the construct comprises at least one open disulfide bridge, and one or more of the domains comprises a disulfide bond; The method includes the step of refolding the construct, wherein the construct is contacted with a refolding buffer that refolds the construct to its native form, the buffer comprising: (i) an oxidizing agent having a concentration of 0.1 to 10 mM, wherein the oxidizing agent is selected from the group consisting of copper (II), dissolved oxygen, and (L)-dehydroascorbic acid (DHA); and (ii) a chaotropic agent having a concentration ranging from 1 M to 2.5 M, wherein said chaotropic agent is guanidinium; Including, The method of claim 1, wherein the pH of the refolding buffer corresponds to + / - 4.5 of the pI value of the first domain or + / - 4.0 of the pI value of the entire construct.

2. The method of claim 1 , wherein the oxidizing agent is copper (II).

3. 2. The method of claim 1, wherein the oxidizing agent is copper (II) sulfate.

4. The method of claim 1, wherein the oxidizing agent is present at a concentration of 0.1 to 1 mM.

5. 2. The method of claim 1, wherein the refolding buffer comprises an oxidizing agent but does not comprise an oxidation-reducing agent.

6. The method of claim 1, wherein the chaotropic agent is present at a concentration of 1.2 to 2M.

7. The method of claim 1, wherein the pH value is in the range of 5.0 to 6.0 or 7.0 to 9.

0.

8. 2. The method of claim 1, wherein the monomer content (percent monomer peak) of the construct after applying the refolding buffer is at least 60%, at least 65%, at least 70% or at least 85%, as determined, for example, by size exclusion chromatography.

9. The method of claim 1, wherein the method is carried out at a temperature of from 4 to 30°C.

10. The method of claim 1, wherein the method has an incubation time of 1 to 24 hours.

11. 2. The method of claim 1, wherein contacting the construct with the refolding buffer occurs (i) in a process liquid (pool) in a downstream purification step or (ii) on a chromatography column in a downstream purification step.

12. 12. The method of claim 11, wherein the construct concentration is (i) 0.5-15 mg / ml in the liquid pool, or (ii) 5-15 mg / ml on the chromatography column.

13. 12. The method of claim 11, wherein the liquid pool is selected from the group consisting of harvested cell culture fluid (HCCF), chromatography eluate pool, Protein A, Protein L, or Protein G affinity chromatography eluate pool, filtered viral inactivation pool, cation exchange chromatography (CEX) monomer pool, and CEX post-high molecular weight peak pool.

14. (a) providing a harvested cell culture fluid (HCCF) comprising the bispecific antibody construct secreted by mammalian cells; (b) contacting the HCCF with a first separation matrix for chromatographic purification under conditions suitable for the construct to associate with the separation matrix, wherein the first separation matrix is ​​an affinity resin selected from the group consisting of Protein A, Protein G, Protein L and synthetic mimetic affinity resins; (c) washing the first separation matrix; (d) eluting the constructs from the first separation matrix; (e) inactivating viruses in the eluate containing the construct; (f) contacting the eluate containing the constructs with a second separation matrix for chromatographic polishing under conditions suitable for the association of the constructs with the second separation matrix, wherein the second separation matrix is ​​a cation exchange chromatography matrix; (g) washing the second separation matrix; (h) eluting the constructs from the second separation matrix, wherein the constructs are separated into two fractions: (i) a monomer pool comprising construct monomers and (ii) a high molecular weight pool comprising construct aggregates. (i) separating the virus from the fraction (i) containing the eluate by filtration; (j) subjecting the eluate to a further ultrafiltration and / or diafiltration step; Including, (1) in the cell culture medium prior to harvest or in the HCCF of step (a), and / or (2) on the separation matrix (on the column) in at least one of steps (b) or (f), and / or (3) In the process liquid (pool) in at least one of steps (d), (e) or (h). The method of claim 1, wherein the construct of is contacted with the refolding buffer, which refolds the construct into its native form.

15. The first binding domain of the construct comprises: (a) CDR-H1 shown in SEQ ID NO: 29, CDR-H2 shown in SEQ ID NO: 30, CDR-H3 shown in SEQ ID NO: 31, CDR-L1 shown in SEQ ID NO: 34, CDR-L2 shown in SEQ ID NO: 35, and CDR-L3 shown in SEQ ID NO: 36; (b) CDR-H1 shown in SEQ ID NO: 94, CDR-H2 shown in SEQ ID NO: 95, CDR-H3 shown in SEQ ID NO: 96, CDR-L1 shown in SEQ ID NO: 97, CDR-L2 shown in SEQ ID NO: 98, and CDR-L3 shown in SEQ ID NO: 99; (c) CDR-H1 as set forth in SEQ ID NO: 105, CDR-H2 as set forth in SEQ ID NO: 106, CDR-H3 as set forth in SEQ ID NO: 107, CDR-L1 as set forth in SEQ ID NO: 109, CDR-L2 as set forth in SEQ ID NO: 110, and CDR-L3 as set forth in SEQ ID NO: 111; and (d) CDR-H1 shown in SEQ ID NO: 214, CDR-H2 shown in SEQ ID NO: 215, CDR-H3 shown in SEQ ID NO: 216, CDR-L1 shown in SEQ ID NO: 217, CDR-L2 shown in SEQ ID NO: 218, and CDR-L3 shown in SEQ ID NO:

219.

2. The method of claim 1, comprising a VH region comprising CDR-H1, CDR-H2 and CDR-H3, and a VL region comprising CDR-L1, CDR-L2 and CDR-L3 selected from the group consisting of:

16. The method of claim 1 , wherein the method is applied to a construct produced by an upstream continuous manufacturing process.

Images