Invented by Gregory Moore, Matthew Bernett, Rumana Rashid, John Desjarlais, Xencor Inc
The Xencor Inc invention works as followsThe present invention is directed at optimized anti-CD3 variables for use in various bispecific formats including those that employ scFv component. The invention also relates to the nucleic acid encoding the polypeptide as well as vectors containing the same, and host cells incorporating the vector. The invention also provides a pharmaceutical formulation containing the polypeptide mentioned and the medical uses of this polypeptide.
Background for Optimized anti CD3 variable regions
Antibody-based therapeutics have been used successfully to treat a variety of diseases, including cancer and autoimmune/inflammatory disorders. Despite this, there are still improvements that can be made to the class of drugs in order to improve their clinical effectiveness. The engineering of novel and additional antigen-binding sites into antibody-based drug molecules is one avenue that’s being explored. This allows a single immunoglobulin to co-engage two different antigens. Bispecifics are non-native, alternate or alternative antibody formats that engage with two antigens. The antibody variable region (Fv), which is so diverse, can be used to create an Fv to recognize virtually any molecule. This makes bispecific antibody generation a simple matter of adding new variable regions to the antibody.
Numerous alternate antibody formats were explored for bispecific targetting (Chames & Baty 2009, mAbs 1, 6:1-9, Holliger & Hudson 2005, Nature Biotechnology 23, 9:1126-1136, Kontermann mAbs 4, 2:182 2012, which are all expressly incorporated by reference). In the beginning, bispecific antibody was made by fusing cell lines that produced monoclonal antibodies (Milstein, et. al., 1983 Nature 305:537-554). The hybrid hybridoma and quadroma produced bispecific antibodies but they were a small population. It took extensive purification to isolate the desired antigen. This problem was solved by using antibody fragments. These fragments do not have the complex quaternary structures of full-length antibodies, so they can be used to link variable light and heavy chain in a single genetic construct. There have been many types of antibody fragments generated. These include diabodies and single-chain diabodies as well as tandem scFvs and Fab2 bispecifics. These formats are able to be expressed in high levels by bacteria, and their small size may provide favorable penetration. However, these formats clear quickly in vivo. They can also present manufacturing challenges due to their stability and production. These drawbacks are primarily due to the fact that most antibody fragments lack the constant region with its functional properties. This includes a larger size, higher stability, and the ability to bind various Fc receptors or ligands which maintain a long half-life (i.e. The neonatal FcRn receptor or other binding sites are used for purification. protein A and protein G).
Recent work has sought to overcome the limitations of fragment-based bispecifics through engineering dual-binding into full-length antibody-like formats. (Wu, et. al. 2007, Nature Biotechnology 25, 11: 1290-1297; U.S. No. 12/477,711; Michaelson et al., 2009, mAbs 1:128-141; PCT/US2008/074693; Zuo et al., 2000, Protein Engineering 13:361-367; U.S. Ser. No. No. These formats overcome many of the problems associated with bispecific antibody fragments, primarily because they contain an Fc-region. The fact that these formats build new antigen-binding sites on top the homodimeric constant chain means binding to the new antibody is always bivalent.
For many antigens, which are attractive in a bispecific therapeutic format as co-targets, the desired binding rather than bivalent. Cross-linking a monovalent interaction is the means by which cellular activation occurs for many immune receptors. Cross-linking can be mediated either by immune complexes containing antibody and antigen or through effector cell engagement with target cells. Low affinity Fc gamma (Fc?) receptors such as FcRIIa, FcRIIb and FcRIIIa can bind to the antibody Fc-region monovalently. Monovalent binding of these Fc?Rs does not activate the cells. However, upon immune contact or cell-to cell contact, receptors become cross-linked, clustered, and activated. When receptors are responsible for cellular killing (e.g. Fc?RIIIa, on NK cells), receptor cross-linking occurs and cellular activation when the effector engages the target in a highly avid manner (Bowles & Weiner 2005, J Immunol Methods 304 : 88-99; expressly incorporated as reference). On B cells, the inhibitory Fc?RIIb only inhibits B cell activation when it forms an immune complex with BCR, a mechanism mediated by immune fusion of soluble IgG with the antigen recognized by BCR (Heyman, 2003, Immunol Lett, 88 157-161; Smith & Clatworthy 2010, Nature Reviews Immunology, 10:328-343, expressly incorporated). Another example is that CD3 activation occurs only when the associated TCR engages antigen loaded MHC on antigen-presenting cells, in a highly avid synapse between cell to cell (Kuhns, et. al., 2006 Immunity 24:133-139). In fact, non-specific bivalent CD3 cross-linking using an anti-CD3 antibodies elicits a cytokine flood and toxicity. (Perruche et. al., J Immunol 183 :953-61; Chatenoud & Bluestone 2007, Nature Reviews Immunology 7,622-632, expressly incorporated). For practical clinical use, monovalent binding is preferred for CD3 coengagement to redirected kill of target cells. This results in activation upon engagement with co-engaged targets. There is also a need for anti-CD3 antibody and derivatives to cross-react against primate CD3 antigen in preclinical studies due to the possibility of unwanted CD3-mediated toxicity.
There are several multispecific antibodies or fragments of antibody that are being developed, which rely on a single binding site bind to CD3. Bispecific binding molecules are discussed, for instance, in US 20110262439. The present invention provides optimized anti-CD3 variables that can be used with bispecific formats including formats relying upon scFv format, scFv and Fc fusions and the like.
The present invention is directed at optimized anti-CD3 variables for use in various bispecific formats including those that employ scFv component. The invention also relates to the nucleic acid encoding the polypeptide as well as vectors containing the same, and host cells incorporating the vector. The invention also provides a pharmaceutical formulation containing the polypeptide mentioned and the medical uses of this polypeptide.
The present invention, therefore, provides compositions that comprise an anti-CD3 region with a variable region comprising a sequence of a VhCDR1 (SEQ ID No:411), a VhCDR2 (SEQ ID No:413), a VhCDR3 (SEQ ID no:416), a SEQ NO:420 for a SEQ NO:420CDR1, and SEQ NO:425CDR2 for a SEQ NO:425 In some cases the variable heavy region of the composition may not have the SEQ NO:1 sequence and the variable light area of the composition may not have SEQ NO:2 sequence.
In other aspects, the anti CD3 variable region is a scFv containing a sequence chosen from the following SEQ ID Nos: 4, 8, 16, 20 32, 40 52, 56 60 64, 72, 80, 90, 110, 128, 132 144, 162, 156, 160 168, 182, 180, 182, 184, 194, 186, 194, 192, 200,204,208,212,216,220,224,228,232,236,240,
In an additional aspect, compositions of invention also contain an Fc region including dual scFv, with a scFv and a scFv. The scFvs are a first anti CD30 scFV, followed by a second scFv. In certain aspects, the second scFv binds a target antigen chosen from the group consisting CD5, CD20 CD30 CD33 CD38 CD40 EGFR EpCAM Her2, HM1.24
In a Further Aspect, the anti CD3 variable region comprises an variable heavy region consisting of a sequence selected among SEQ ID Nos: 5, 9, 13 17, 21, 25, 29, 34, 41, 45, 49, 53, 56, 60, 65, 70, 75, 80, 85, 90, 95, 101, 105, 109, 112, 125, 130, 140, 150, 160, 173, 181, 185, 193, 193, 197,201, 209, 209
In another aspect, the anti CD3 variable region consists of a variable light area comprising a selection from the following SEQ ID Nos: 6, 10, 14 18, 22, 26, 30, 54, 56, 60, 64, 72, 80, 90, 112, 116, 132, 162, 166, 170, 175, 178, 182, 194, 198,202,206,210,214,218,226,230,234, 238,242, 246, 250. 254, 258,262, 262, 262,262, 266, 266, 266, 25, 260, 266, 24, 216,
In addition, the composition is bispecific IgG. The anti-CD30 variable area is a first single chain Fv, and the said composition contains a second single chain Fv.
In another aspect, the composition is MAb-Fv wherein the variable heavy region of the anti CD30 variable region attached to the first heavy chain mAbFv and variable light chain anti CD30 variable region attached to the second heavy chain mAbFv.
In addition, the composition has a multi-scFv.
The composition is also an-scFvCH3.
In a second aspect, this composition is a mAb scFv.
In a second aspect, this composition is a mAb scFv2.
In an additional aspect, this composition is a complete antibody.
In another aspect, the composition includes an Fc-domain and an amino acid replacement. In certain aspects, the Fc-domain has a changed binding to an Fc-R receptor such as Fc’RIIIa or Fc’RIIb. If the altered binding to Fc-RIIIa is the case, then the amino acid substitution selected is from the following group: 239D/239E, 236,R, 330L 332D/332E/330L; 239D/332E/330L; 239D/332E/330L; 267D/267E/328F/243L 298A.
In certain aspects, the composition includes an Fc-domain with altered binding to the FcRn receptor and comprises a substitution of an amino acid selected from the group consisting 434A, 334S/428L/308F/259I, 259I/428L/434S/259I/308F/436I/428L or 436V/434S/436V/428L.
In another aspect, the composition has an element selected from the group of FIGS. 7A-7M, FIGS. 8B, 8D and FIG. 9B, FIGS. 10A-10E, and FIGS. 11A-E. The Fc domain is a collection of variants chosen from FIGS. 12A and 12B. The composition can also be characterized by a selected structure from the group of FIGS. 7A-7M, FIGS. 8B, 8D and FIG. 9B, FIGS. 10A-10E, and FIGS. 11A-E wherein the Fc of the structure is a set of variants selected in FIG. 13. The composition can also be characterized by a selected structure from the group of FIGS. 7A-7M, FIGS. 8B, 8D and FIG. 9B, FIGS. 10A-10E, and FIGS. “The Fc domain comprises a selection of variants from FIGS. 14.
Click here to view the patent on Google Patents.