US00877 1966B2

(12) UnitedO States Patent (10) Patent No.: US 8,771,966 B2 Dennis et al. (45) Date of Patent: Jul. 8, 2014

(54) IMMUNO-PET IMAGING OF ANTIBODIES RE30,985 E 6/1982 Cartaya

AND IMMUNOCONUGATES AND USES 3. E. A 3. 3: RCai Ca.al THEREFOR 4,364,866 A 12/1982 Asai et al. 4,371.533. A 2, 1983 Akimoto et al. (71) Applicant: Genentech, Inc., South San Francisco, 4.419,446 A 12/1983 Howley et al. CA (US) 4.424,219 A 1/1984 Hashimoto et al. 4.450,254 A 5/1984 Isley et al. (72) Inventors: Mark Dennis, San Carlos, CA (US); Jan 3.288 A 12.8 s Marik, Hillsborough, CA (US); Paul 4657.866 A 4, 1987 RW, Polakis, Mill Valley, CA (US); Bonnee 4,676.980 A 6, 1987 Segal et al. Rubinfeld, Danville, CA (US); Simon 4,767,704 A 8/1988 Cleveland et al. Williams, Redwood City, CA (US) 4,816,567 A 3/1989 Cabilly et al. 4,927,762 A 5, 1990 Darfler (73) Assignee: Genentech, Inc., South San Francisco, 4,965,1994,933,294 A 10/19906, 1990 WaterfieldCapon et al. et al. CA (US) 4,970,198 A 1 1/1990 Lee et al. 4,975,278 A 12/1990 Senter et al. (*) Notice: Subject to any disclaimer, the term of this 5,053,394 A 10/1991 Ellestad et al. patent is extended or adjusted under 35 Sis A 3. 3: Fi trometal U.S.C. 154(b) by 0 days. 5,122,469 A 6/1992 MatherISO et al. 5,208,020 A 5/1993 Chari et al. (21) Appl. No.: 13/692,888 5,264.365 A 1 1/1993 Georgiou et al. 5,316,757 A 5/1994 Sherry et al. (22) Filed: Dec. 3, 2012 5,342,606 A 8/1994 Sherry et al. 5,362,852 A 11/1994 Geoghegan (65) Prior Publication Data 5,385,893 A 1, 1995 Kiefer 5,401,638 A 3, 1995 Carney et al. US 2013/O157289 A1 Jun. 20, 2013 5,416,064 A 5/1995 Chari et al. 5,428,139 A 6/1995 Kiefer et al. 5,428, 155 A 6/1995 Sherry et al. Related U.S. Application Data 5,462,725 A 10, 1995 Kiefer et al. (63) Continuation of application No. (Continued) PCT/US2011/038923, filed on Jun. 2, 2011. FOREIGN PATENT DOCUMENTS (60) Provisional application No. 61/351,195, filed on Jun. 3, 2010 EP OOO3O89 B1 8, 1981 s EP O404 097 6, 1990 (51) Int. Cl. (Continued) GOIN33/53 (2006.01) (52) U.S. Cl. OTHER PUBLICATIONS

USPC ...... 435/7.1; 435/7.2 (58) Field of Classification Search Aerts et al., “Disparity between in vivo EGFR expression and Zr None labeled cetuximab uptake assessed with PET'J Nucl Med. 50: 123 See application file for complete search history. 131 (2009). Albert et al., “Direct synthesis of DOTA-DPhe-octreotide and (56) References Cited (DOTA-DPhe'.Tyr-octreotide (SMT487): Two conjugates for sys temic delivery of radiotherapeutical nuclides to Somatostatin recep U.S. PATENT DOCUMENTS tor positive tumors in man” Bioorg Med Chem Lett 8:1207-1210 ( 1998). 3,720,760 A 3, 1973 Bennich et al. 3,773.919 A 11, 1973 Boswell et al. (Continued) 3,896,111 A 7/1975 Kupchan et al. 4,137,230 A 1/1979 Hashimoto et al. 4,151,0424.248,870 A 4/19792/1981 HigashideMiyashita etet al. Primary Examinerin a Jacob Cheu 4,256,746 A 3/1981 Miyashita et al. (74) Attorney, Agent, or Firm — Jessica L. Richardson 4,260,608 A 4, 1981 Miyashita et al. 4,265,814 A 5, 1981 Hashimoto et al. 4,275,149 A 6, 1981 Litman et al. (57) ABSTRACT 4.294,757 A 10, 1981 Asai 388 3. A 33. Siski, etal Anti-STEAP-1 antibodies and immunoconjugates thereofare 43.08.269 A 12/1981 Miyashita et al. provided. Methods of using anti-STEAP-1 antibodies and 4,309.428 A 1/1982 Miyashita et al. immunoconjugates thereof are provided. Methods of detect

1.33- K - A 3.18: EMSea ea.1 ing or determining the presence of STEAP-1 proteins are 4,317,821 A 3/1982 Miyashita et al. provided. 4,318,980 A 3/1982 Boguslaski et al. 4.322,348 A 3, 1982 Asai et al. 4.331,598 A 5/1982 Hasegawa et al. 9 Claims, 51 Drawing Sheets US 8,771,966 B2 Page 2

(56) References Cited FOREIGN PATENT DOCUMENTS U.S. PATENT DOCUMENTS EP O 425 235 B1 9, 1996 EP 1391213 A1 2, 2004 5,480,990 A 1/1996 Kiefer et al. WO 87/OO.195 A1 1, 1987 5,500,362 A 3, 1996 Robinson et al. WO 90,03430 A1 4f1990 5,508,192 A 4, 1996 Georgiou et al. WO 91/00360 A1 1, 1991 5,545,806 A 8/1996 Lonberg et al. WO 91705264 A1 4f1991 5,545,807 A 8, 1996 Surani et al. WO 91 10741 A1 7, 1991 5,569,825 A 10/1996 Lonberg et al. WO 92fOO373 A1 1, 1992 5,571,894 A 11/1996 Wells et al. WO 92.09690 A2 6, 1992 5,585,089 A 12/1996 Queen et al. WO 93,01161 A1 1, 1993 5,587,458 A 12/1996 King et al. WO 93,06213 A1 4f1993 5,606,040 A 2f1997 McGahren et al. WO 93,08829 A1 5, 1993 5,624,821 A 4/1997 Winter et al. WO 93.16.185 A2 8, 1993 5,625,126 A 4/1997 Lonberg et al. WO 93.21232 A1 10, 1993 5,633,425 A 5/1997 Lonberg et al. WO 94/11026 A2 5, 1994 5,635,483. A 6, 1997 Pettit et al. WO 94.29351 A2 12, 1994 5,641,870 A 6, 1997 Rinderknecht et al. WO 94.29351 A3 12?1994 5,648,237 A 7, 1997 Carter WO 96.O7754 A1 3, 1996 5,648,260 A 7, 1997 Winter et al. WO 96.O7754 R3 3, 1996 5,661,016 A 8/1997 Lonberg et al. WO 96.33735 A1 10, 1996 5,663,149 A 9, 1997 Pettit et al. WO 96.34096 A1 10, 1996 5,693,762 A 12/1997 Queen et al. WO 98.24893 A2 6, 1998 5,712.374. A 1/1998 Kuntsmann et al. WO 98.58964 12/1998 5,714,586 A 2, 1998 Kunstmann et al. WO 99.22764 A1 5, 1999 5,731, 168 A 3, 1998 Carter et al. WO 99,51642 A1 10, 1999 5,739,116 A 4, 1998 Hamann et al. WO OO/42O72 T 2000 5,739,277 A 4, 1998 Presta et al. WO 006 1739 A1 10, 2000 5,739,294 A 4, 1998 Kiefer et al. WO 01.00245 A2 1, 2001 5,750,660 A 5, 1998 Kiefer et al. WO 01.00245 A3 1, 2001 5,767,237 A 6, 1998 Sakakibara et al. WO O1/29246 A1 4/2001 5,767.285 A 6, 1998 Hamann et al. WO Of 45746 A2 6, 2001 5,770,701 A 6, 1998 McGahren et al. WO Of 45746 A3 6, 2001 5,770,710 A 6, 1998 McGahren et al. WO 02/088172 A2 11/2002 5,773,001 A 6, 1998 Hamann et al. WO 02/088172 A3 11, 2002 5,780,588 A 7, 1998 Pettit et al. WO O2/O88172 R1 11 2002 5,807,715 A 9, 1998 Morrison et al. WO O3,O11878 A2 2, 2003 5,821,337 A 10, 1998 Carter et al. WO O3,O11878 A3 2, 2003 5,834,456 A 11/1998 Kiefer et al. WO 03/043583 A2 5, 2003 5,834,597 A 11/1998 Tso et al. WO 03/043583 A3 5, 2003 5,840,523 A 11/1998 Simmons et al. WO 03/084570 A1 10, 2003 5,869,046 A 2f1999 Presta et al. WO O3,0851.19 A1 10, 2003 5,877,296 A 3, 1999 Hamann et al. WO 2004/OO 1993 A1 12/2003 6,027,888 A 2/2000 Georgiou et al. WO 2004/O10957 A2 2, 2004 6,083,715 A 7/2000 Georgiou et al. WO 2004/O10957 A3 2, 2004 6,124,431 A 9, 2000 Sakakibara et al. WO 2004/032828 A2 4/2004 6,165,745 A 12/2000 Ward et al. WO 2004/032828 A3 4/2004 6,194.551 B1 2/2001 Idusogie et al. WO 2004/043493 A1 5, 2004 6,329,503 B1 12/2001 Afar et al. WO 2004/056312 A2 T 2004 6,372,907 B1 4/2002 Lee et al. WO 2005/035586 A1 4/2005 6,441,163 B1 8, 2002 Chari et al. WO 2005/035778 A1 4/2005 6,528,624 B1 3/2003 Idusogie et al. WO 2005037992 A2 4/2005 6,602,677 B1 8, 2003 Wood et al. WO 2005/053742 A1 6, 2005 6,602,684 B1 8/2003 Umana et al. WO 2005/081711 A2 9, 2005 6,737,056 B1 5, 2004 Presta WO 2006/034488 A2 3, 2006 6,821,505 B2 11/2004 Ward WO 2006/034488 A3 3, 2006 6,824,780 B1 1 1/2004 Devaux et al. WO 2007/01 1968 A2 1, 2007 7,521,541 B2 4/2009 Eigenbrot et al. WO 2007/01 1968 A3 1, 2007 2002/0164328 A1 11/2002 Shinkawa et al. 2003/0115614 A1 6/2003 Kanda et al. OTHER PUBLICATIONS 2003. O157108 A1 8, 2003 Presta 2003/01903 11 A1 10/2003 Dall’Acqua et al. Alley, S.C. et al., “Controlling the location of drug attachment in 2004/0001827 A1 1/2004 Dennis antibody-drug conjugates, Abstract No. 627. American Association 39.83 A. 58. SE .. al. for Cancer Research, 2004 Annual Meeting, Mar. 27-31, 2004 Pro 2004.0109865 A1 6, 2004 Niwa et al. ceedings of the AACR. 45:52 (2004). 2004/0110282 A1 6/2004 Kanda et al. Amir et al., “Self-immmolative dendrimers' Angew. Chem. Int. Ed. 2004/O110704 A1 6/2004 Yamane et al. 42:4494-4499 (2003). 3. 36. A. 3. Sh et al. Amsberry et al., “The lactonization of 2'-hydroxyhydrocinnamic acid 2004/0253232 A1 12/2004 JakobovitsOS et al. amides:as A potential prodrug for amines'- - - - J Org Chem 55:5867-5877 2005/00 14934 A1 1/2005 Hinton et al. (1990). 2005/OO48572 A1 3/2005 Reilly et al. Arie et al., “Chaperone function of FkpA, a heat shock prolyl 2005, 0123546 A1 6, 2005 Umana et al. isomerase, in the periplasm of Escherichia coli' Mol Microbiol 2005, 0169933 A1 8/2005 Steeves et al. 39(1): 199-210 (2001). 2005/0238649 A1 10, 2005 Doronina et al. Axworthy et al., "Cure of human carcinoma xenografts by a single 2005/0256030 A1 1 1/2005 Feng dose of pretargeted yttirum-90 with negligible toxicity” PNatl Acad 2005/0276812 A1 12, 2005 Ebens et al. Sci USA97(4): 1802-1807 (2000). US 8,771,966 B2 Page 3

(56) References Cited Chen et al., “Selection and analysis of an optimized anti-VEGF antibody: crystal structure of an affinity-matured Fab in complex OTHER PUBLICATIONS with antigen” J Mol Biol 293(4):865-881 ( 1999). Chothia et al., “Canonical structures for the hypervariable regions of Baldwin et al., “Monoclonal antibodies in cancer treatment' Lancet immunoglobulins' J Mol Biol 196:901-917 (1987). 327(8481):603-605 (1986). Clackson et al., “Making antibody fragments using phage display Barbaset al., “Assembly of combinatorial antibody libraries on phage libraries' Nature 352:624-628 (Aug. 15, 1991). surfaces: the gene III site” PNatl AcadSci USA 88:7978-7982 (Sep. Clynes et al., “Fc receptors are required in passive and active immu 1991). Barbas et al., “In vitro evolution of a neutralizing human antibody to nity to melanoma P Natl AcadSci USA95:652-656 (Jan. 1998). human immunodeficiency virus type 1 to enhance affinity and Collaborative Computational Project, No. 4, “The CCP4 Suite: Pro broaden strain cross-reactivity” PNatl AcadSci USA91:3809-3813 grams for Protein Crystallography” Acta Cryst D50:760-763 (1994). (Apr. 1994). Crea et al., “Chemical sysnthesis of genes for human insulin' PNatl Barbas et al., “Semisynthetic combinatorial antibody libraries: a AcadSci USA 75:5765-5769 (Dec. 1978). chemical solution to the diversity problem” P Natl Acad Sci USA Cree et al., “Methotrexate chemosensitivity by ATP luminescence in 89:4457-4461 (May 1992). human leukemia cell lines and in breast cancer primary cultures: Barnes et al., “Methods for growth of cultured cells in serum-free comparison of the TCA-100 assay with a clonogenic assay Antican medium” Anal Biochem 102:255-270 (1980). cer Drugs 6:398-404 (1995). Bass et al., “Hormone Phage: An enrichment method for variant Crouch et al., “The use of ATP bioluminescence as a measure of cell proteins with altered binding properties' Proteins: Structure, Func proliferation and cytotoxicity J Immunol Methods 160:81-88 ( tion and Genetics 8:309-314 (1990). 1993). Bhaskar et al., “E-selectin up-regulation allows for targeted drug Cunningham et al., “High-resolution epitope mapping of hCGH-re delivery in prostate cancer Cancer Res 63:6387-6394 (Oct. 2003). ceptor interactions by alanine-scanning mutagenesis' Science Blend et al., “Labeling anti-HER2/neu monoclonal antibodies with 244: 1081-1085 (Jun. 2, 1989). 'In and 'Y using a bifunctional DTPA chelating agent” Cancer Daeron, “Fc receptor biology” Annu Rev Immunol 15:203-234 ( Biother Radio 18(3):355-363 (2003). 1997). Boerner et al., “Production of antigen-specific human monoclonal de Groot et al., "Cascade-Release Dendrimers' liberate all end antibodies from in vitro-primed human splenocytes' J Immunol groups upon a single triggering event in the dendritic core' Angew. 147(1):86-95 (Jul 1, 1991). Borjesson et al., “Performance of immuno-positron emission Chem. Int. Ed 42:4490-4494 (2003). tomography with Zirconium-89-labeled chimeric monoclonal anti De Haas et al., “Fcy receptors of phagocytes' J Lab Clin Med body U36 in the detection of lymph node metastases inhead and neck 126:330-341 (Oct. 1995). cancer patients' Clin Cancer Res 12(7):2133-2140 (2006). DeNardo et al., “Comparison of 1,4,7,10-tetraazacyclododecane Bothmann et al., “The periplasmic Escherichia coli peptidylprolyl N,N',N', N'-tetraacetic acid (DOTA)-Peptide-ChL6, a novel cis,trans-isomerase FkpA' J Biol Chem 275(22): 17100-17105 (Jun. immunoconjugate with catabolizable linker, to 2-iminothiolane-2- 2000). p-(bromoacetamido)benzyl-DOTA-ChL6 in breast cancer Brennan et al., “Preparation of bispecific antibodies by chemical xenografts” Clin Cancer Res 4:2483-2490 (Oct. 1998). recombination of monoclonal immunoglobulin G fragments' Sci Dennis et al., “Albumin binding as a general strategy for improving ence 229(4708):81-83 (Jul. 5, 1985). the pharmacokinetics of proteins' J Biol Chem 277(38):35035 Briggs et al., “Synthesis of functionalised fluorescent dyes and their 35043 (Sep. 20, 2002). coupling to amines and amino acids' J Chem Soc, Perkin Trans Dijkers et al., “Biodistribution of Zr-trastuzumab and PET imaging 1:1051-1058 (1997). of HER2-positive lesions in patients with metastatic breast cancer' Brinkley, M., “A brief Survey of methods for preparing protein con Clin Pharmacol Ther 87(5):586-592 (May 2010). jugates with dyes, haptens, and cross-linking reagents’ Bioconj Dijkers et al., “Development and characterization of clinical-grade Chem3:2-13 (1992). Zr-Trastuzumab for HER2/neu. ImmunoPET imaging” JNucl Med Bruggemann et al., “Designer mice: the production of human anti 50(6):974-981 (Jun. 2009). body repertoires in transgenic animals'Year in Immunology 7:33-40 Doronina et al., “Development of potent monoclonal antibody (1993). auristatin conjugates for cancer therapy” Nat Biotechnol 21(7):778 Cai et al., “PET Imaging of colorectal cancer in Xenograft-bearing 784 (Jul 2003). mice by use of an F-labeled T84.66 anti-carcinoembryonic antigen Doronina et al., “Enhanced activity of monomethylauristatin F diabody”J Nucl Med 48(2):304-310 (Feb. 2007). through monoclonal antibody delivery: effects of linker technology Camera et al., “Comparative biodistribution of indium- and yttrium on efficacy and toxicity” Bioconjug Chem 17(1): 114-124 (Jan. labeled B3 monoclonal antibody conjugated to either 2-(p-SCN-BZ)- 2006). 6-methyl-DTPA (1B4M-DTPA) or 2-(p-SCN-BZ)-1,4,7,10 Duncan et al., “The binding site for Cld on IgG” Nature 322:738-740 tetraazacyclododecane tetraacetic acid (2B-DOTA)” Eur JNuclMed (1988). 21(7):640-646 (1994). Eigenbrot et al., "X-Ray structures of the antigen-binding domains Camera et al., “Evaluation of a new DTPA-derivative chelator: Com from three variants of humanized anti-p185' antibody 4D.5 and parative biodistribution and imaging studies of 'in-labeled B3 comparison With molecular modeling” J Mol Biol 229:969-995 ( monoclonal antibody in athymic mice bearing human epidermoid 1993). carcinoma xenografts” Nucl Med Biol 20(8):955-962 ( 1993). Embleton et al., “In-cell PCT from mRNA:amplifying and linking Capel et al., “Heterogeneity of human IgG Fc receptors' the rearranged immunoglobulin heavy and light chain V-genes within Immunomethods 4:25-34 (1994). single cells' Nucl Acids Res 20015):3831-3837 (1992). Carlsson et al., “Protein thiolation and reversible protein—protein Erker et al., “Mono (m-cyclopentadienyl) Zirconium complexes: from conjugation” Biochem J 173:723-737 (1978) coordination chemistry to enantioselective catalysis' J Organometal Carter et al., “High level Escherichia coli expression and production lic Chem 400:185-203 (1990). of a bivalent humanized antibody fragment’ Bio/Technology Erker et al., “Stereochemistry and catalysis with Zirconium com 10(2):163-167 (Feb. 1992). plexes” Pure and Applied Chem. 63(6):797-806 (1991). Carter et al., “Humanization of an anti-p185' antibody for human Fellouse et al., “Synthetic antibodies from a four-amino-acid code: A cancer therapy” PNatl AcadSci USA 89:4285-4289 (May 1992). dominant role for tyrosine in antigen recognition' P Natl Acad Sci Chari et al., “Immunoconjugates containing novel maytansinoids: USA 101 (34): 12467-12472 (Aug. 24, 2004). Promising anticancer drugs' Cancer Res 52:127-131 ( 1992). Fishwild et al., “High-avidity human IgGK monoclonal antibodies Chen et al., “Chaperone activity of DsbC J Biol Chem from a novel strain of minilocus transgenic mice' Nat Biotechnol 274(28): 19601-19605 (Jul. 9, 1999). 14(7):845-851 (Jul 1996). US 8,771,966 B2 Page 4

(56) References Cited Holliger et al., “Diabodies': Small bivalent and bispecific antibody fragments' P Natl AcadSci USA90:6444-6448 (Jul. 1993). OTHER PUBLICATIONS Holliger et al., “Engineered antibody fragments and the rise of single domains' Nat Biotechnol 23(9): 1126-36 (Sep. 2005). Fraker et al., “Protein and cell membrane iodinations with a sparingly Hoogenboom et al., “By-passing immunisation. Human antibodies soluble chloroamide, 1,3,4,6-tetrachloro-3a,6a-diphenylglycoluril” from Synthetic repertoires of germline V gene segments rearranged Biochem Biophys Res Comm 80(4):849-857 (Feb. 28, 1978). in vitro” J Mol Biol 227(2):381-388 (Sep. 20, 1992). Francisco et al., “cAC10-vcMMAE, an anti-CD30 monomethyl Hubert etal. “STEAP: aprostate-specific cell-surface antigen highly auristatin E conjugate with potent and selective antitumor activity expressed in human prostate tumors' P Natl Acad Sci USA Blood 102(4): 1458-1465 (Aug. 15, 2003). Gazzano-Santoro et al., “A non-radioactive complement-dependent 96(25): 14523-8 (1999). cytotoxicity assay for anti-CD20 monoclonal antibody'J Immunol Hudson et al., “Engineered antibodies' Nature Medicine 9(1): 129 Methods 202:163-171 ( 1997). 134 (Jan. 2003). "Gemtuzumab Ozogamicin, Treatment of acute myeloid leukemia' Hurleet al., “Protein engineering techniques for antibody humaniza Drugs of the Future 25(7):686-692 (2000). tion” Curr Opin Biotechnol 5:428-433 (1994). Geoghegan et al., “Site-directed conjugation of nonpeptide groups to Idusogie et al., “Mapping of the C1q binding site on Rituxan, a peptides and proteins via periodate oxidation of a 2-amino alcohol. chimericantibody with a human IgG1 Fc'JImmunol 164:41.78-4184 Application to modification at N-terminal serine' Bioconjugate (2000). Chem3:138-146 (1992). Izardet al., “An improved method for labeling monoclonal antibodies Getz et al., “A comparison between the sylfhydryl reductants tris(2- with Samarium-153: Use of the bifunctional chelate 2-(p- carboxyethyl)phosphine and dithiothreitol for use in protein bio isothiocyanatobenzyl)-6-methyldiethylenetriamianepentaacetic chemistry” Anal Biochem 273:73-80 (1999). acid” Biocon Chem3:346-350 (1992). Graham et al., "Characteristics of a human cell line transformed by Jackson et al., “In vitro antibody maturation. Improvement of a high DNA from human adenovirus type 5” J Gen Virol 36(1):59-72 (Jul. affinity, neutralizing antibody against IL-1 " J Immunol 1977). 154(7):3310-3319 (1995). Gram et al., “In vitro selection and affinity maturation of antibodies Jacobovits et al., “Analysis of homozygous mutant chimeric mice: from a naive combinatorial immunoglobulin library” PNatl Acad Sci Deletion of the immunoglobulin heavy-chain joining region blocks USA 89:3576-3580 (Apr. 1992). B-cell development and antibody production” PNatl AcadSci USA Griffiths et al., “Human anti-self antibodies with high specificity 90:2551-2555 (1993). from phage display libraries” EMBOJ 12(2):725-734 (1993). Jakobovits et al., “Germ-line transmission and expression of a Gruber et al., “Efficient tumor cell lysis mediated by a bispecific human-derived yeast artificial chromosome” Nature 363:255-258 single chain antibody expressed in Escherichia coli J Immunol (Mar. 18, 1993). 152:5368-5374 (1994). Johnson et al. Methods in Molecular Biology “The Kabat Database Guss et al., "Structure of the IgG-binding regions of streptococcal and a Bioinformatics Example” Lo, Totowa, NJ: Human Press, vol. protein G” EMBOJ 5(7): 1567-1575 (1986). 248:11-25 (2003). Guyer et al., “Immunoglobulin binding by mouse intestinal epithelial Jones et al., “Materials and Methods/Rapid PCT-cloning of full cell receptors' J Immunol 117(2):587-593 (Aug. 1976). length mouse immunoglobulin variable regions' BioTechnology Ham et al., “Media and growth requirements” Method Enzymol 9:88-89 (Jan. 1991). 58:44-93 (1979). Jones et al., “Replacing the complementarity-determining regions in Hamblett et al., (Effect of drug loading on the pharmacology, a human antibody with those from a mouse” Nature 321:522-525 pharmacokinetics and toxicity of an anti-CD30 antibody-drug con (May 1986). jugate, Abstract No. 624, American Association for Cancer Research; Junutula et al., “Rapid identification of reactive cysteine residues for 2004 Annual Meeting, Mar. 27-31, 2004, Proceedings of the site-specific labeling of antibody-Fabs' JImmunol Methods 332:41 AACR45:52 (2004)). 52 (2008). Hamblett et al., “Effects of drug loading on the antitumor activity of Junutula et al., “Site-specific conjugation of a cytotoxic drug to an a monoclonal antibody drug conjugate” Clin Cancer Res 10:7063 antibody improves the therapeutic index' Nat Biotechnol 26(8):925 7070 (2004). 32 (Aug. 2008). Hamers-Casterman et al., “Naturally occurring antibodies devoid of Kam et al., “Carbon nanotubes as multifunctional biological trans light chains” Nature 363:446-448 (Jun. 3, 1993). porters and near-infrared agents for selective cancer cell destruction' Hara et al., “Overproduction of penicillin-binding protein 7 Sup P Natl AcadSci USA 102(33): 11600-11605 (Aug. 2005). presses thermosensitive growth defect at low osmolarity due to an spr Kerr et al., "Listeriolysin Opotentiates immunotoxin and bleomycin mutation of Escherichia coli Microbial Drug Resistance 201):63-72 cytotoxicity” Bioconjugate Chem 8(6):781-784 (1997). (1996). Kim et al., “Localization of the site of the murine IgG1 molecule that Harris, “Production of humanized monoclonal antibodies for in vivo is involved in binding to the murine intestinal Fc receptor' Eur J imaging and therapy” Biochem Soc Trans 23(4): 1035-1038 (Nov. Immunol 24(10):2429-2434 (Oct. 1994). 1995). King et al., “Facile synthesis of maleimide bifunctional linkers' Haugland Molecular Probes Handbook of Fluorescent Probes and Tetrahedron Lett 43:1987-1990 (2002). Research ChemicalsMolecular Probes, Inc., (2003). Kingsburg et al., “A novel peptide delivery system involving Hawkins et al., “Selection of phage antibodies by binding affinity peptidase activated prodrugs as antimicrobial agents. Synthesis and mimicking affinity maturation' J Mol Biol 226:889-896 (1992). biological activity of peptidyl derivatives of 5-fluorouracil' J Med Hay et al., “A 2-nitroimidazole carbamate prodrug of 5-amino-1- Chem 27:1447-1451 (1984). (chloromethyl)-3-(5,6,7-trimethoxyindol-2-YL)carbonyl-1,2- Klussman et al., “Secondary mAb-vcMMAE conjugates are highly dihydro-3H-benzEindole (amino-seco-DB1-TMI) for use with sensitive reporters of antibody internalization via the lysosome path ADEPT and GDEPT Bioorg Med Chem Lett 9:2237-2242 ( 1999). way’ Bioconjugate Chem 15:765-773 (2004). Hinman et al., “Preparation and characterization of monoclonal anti Kobayashi et al., “Evaluation of the in vivo biodistribution of indium body conjugates of the calicheamicins: A novel and potent family of 111 and yttrium-88 labeled dendrimer-1B4M-DTPA and its conju antitumor antibiotics' Cancer Res 53:3336-3342 ( 1993). gation with anti-Tac monoclonal antibody' Bioconj Chem 10:103 Hnatowich et al., “The preparation of DTPA-coupled antibodies 111 ( 1999). radiolabeled with metallic radionuclides: an improved method” J Kobayashi et al., “Evaluation of the in vivo biodistribution of yttium Immunol Methods 65:147-157 (1983). labeled isomers of CHX-CTPA-conjugated monoclonal antibodies' Hogrefeet al., “A bacteriophage lambda vector for the cloning and J. Nucl. Med 39:829-836 (1998). expression of immunoglobulin Fab fragments on the Surface of fila Kohler et al., "Continuous cultures of fused cells secreting antibody mentous phage' Gene 128:119-126 (1993). of predefined specificity” Nature 256:495-497 (Aug. 7, 1975). US 8,771,966 B2 Page 5

(56) References Cited Mao et al., “EphB2 as a therapeutic antibody drug target for the treatment of colorectal cancer Cancer Res 64:781-788 (Feb. 2004). OTHER PUBLICATIONS Mardirossian et al., “The stability in liver homogenates of indium 111 and yttrium-90 attached to antibody via two popular chelators' Kostelny et al., “Formation of a bispecific antibody by the use of Nucl Med Biol 20(1):65-74 (1993). leucine zippers' J Immunol 148(5):1547-1553 (Mar. 1, 1992). Marks et al., “By-passing immunization. Human antibodies from Kozbor et al., “A human hybrid myeloma for production of human V-gene libraries displayed on phage” J. Mol Biol 222:581-597 ( monoclonal antibodies' J Immunol 133(6):3001-3005 (Dec. 1984). 1991). Kukis et al., “Optimized conditions for chelation of yttrium-90 Marks et al., “By-passing immunization: building high affinity DOTA immunoconjugates' J Nucl Med 39:2105-2110 (1998). Kunkeletal., “Rapid and EfficientSite-specific Mutagenesis Without human antibodies by chain shuffling” Bio/Technology 10:779-783 Phenotypic Selection” Method Enzymol 154:367-382 ( 1987). (Jul 1992). Law et al., “CD70 is expressed on renal cell carcinoma and is a Mather, "Establishment and characterization of two distinct mouse potential target for tumor cell elimination by antibody-drug conju testicular epithelia cell lines' Biol Reprod 23:243-252 (1980). gates' P Am Assoc Canc Res 45: 144, #625 (2004). Matsuda et al., “Structure and physical map of 64 variable segments Lee et al., “Bivalent antibody phage display mimics natural in the 3' 0.8-megabase region of the human immunoglobulin heavy immunoglobulin' J Immunol Methods 284: 119-132 (2004). chain locus” Nat Genet 3:88-94 (Jan. 1993). Lee et al., "High-affinity human antibodies from phage-displayed Meares et al., “Conjugation of antibodies with bifunctional chelating synthetic Fab libraries with a single framework scaffold' J. Mol Biol agents: Isothiocyanate and bromoacetamide reagents, methods of 340(5):1073-1093 (Jul 23, 2004). analysis, and Subsequent addition of metal ions' Anal Biochem Lee et al., “Specific localization, gamma camera imaging, and 142:68-78 (1984). intracellular trafficking of radiolabelled chimeric anti-G, Meares et al., “Macrocyclic chelates of radiometals for diagnosis and ganglioside monoclonal antibody KM871 in SK-MEL-28 melanoma therapy” BrJ Cancer 62(Suppl X):21-26 (1990). xenografts' Cancer Res 61:4474-4482 (2001). Meijs et al., "A facile method for the labeling of proteins with zirco Lee et al., “The interpretation of protein structures: estimation of nium Isotopes' Nucl Med Biol 23:439-448 (1996). static accessibility” J Mol Biol 55(3):379-400 (1971). Meijs et al., "Zirconium-labeled monoclonal antibodies and their Leung et al., “A method for random mutagenesis of a defined DNA distribution in tumor-bearing nude mice” JNuclMed 38(1): 112-118 segment using a modified polymerase chain reaction' Technique-AJ (Jan. 1997). Methods Cell and Molec Biol 1(1):11-15 (1989). Miederer et al., “Pharmacokinetics, dosimetry, and toxicity of the Lewis et al., “An improved method for conjugating monoclonal anti targetable atomic generator, Ac-HuM195, in nonhuman primates” bodies with N-hydroxysulfosuccinimidyl DOTA Bioconjugate J Nucl Med 45:129-137 (2004). Chem 12:320-324 (2001). Mier et al., “Conjugation of DOTA using isolated phenolic active Lewis et al., “Maleimidocysteineamido-DOTA derivatives: New esters: The labeling and biodistribution of albumin as blood pool reagents for radiometal chelate conjugation to antibody sulfhydryl marker Bioconjugate Chem 16:237-240 (2005). groups undergo pH-dependent cleavage reactions' Bioconj Chem Miller, “Syntheses and therapeutic potential of hydroxamic acid 9:72-86 (1998). based siderophores and analogues' Chem Rev 89:1563-1579 (1989). Li et al., “Vinyl sulfone bifunctional derivatives of DOTA allow Milstein et al., "Hybrid hybridomas and their use in Sulfhydryl-or amino-directed coupling to antibodies. Conjugates immunohistochemistry” Nature 305:537-540 (Oct. 6, 1983). retain immunoreactivity and have similar biodistributions' Miraglia et al., “Homogenous cell- and bead-based assays for high Bioconjugate Chem 13:110-115 (2002). throughput screening using fluorometric microvolume assay technol Lindmark et al., “Binding of immunoglobulins to protein A and ogy” J Biomol Screening 4(4):193-204 (1999). immunoglobulin levels in mammalian sera” J Immunol Methods Mirzadeh et al., “Radiometal labeling of immunoproteins: Covalent 62: 1-13 (1983). linkage of 2-(4-isothiocyanatobenzyl)diethylenetriaminepentaacetic Liu et al., “Eradication of large colon tumor xenografts by targeted acid ligands to immunoglobulin' Bioconj Chem 1:59-65 (1990). delivery of maytansinoids' P Natl Acad Sci USA 93:8618-8623 Mitchell et al., “Targeting primary human Ph+ B-cell precursor leu (Aug. 1996). kemia-engrafted SCID mice using radiolabelied anti-CD 19 Liu, "Bifunctional coupling agents for radiolabeling of biomolecules monoclonal antibodies' J Nucl Med 44:1105-1112 (2003). and target-specific delivery of metallic radionuclides' Adv Drug Moore et al., “Apoptosis in CHO cell batch cultures: examination by Deliver Rev 60:1347-1370 (2008). flow cytometry” Cytotechnology 17:1-11 (1995). Lode et al., “Targeted therapy with a novel enediyene antibiotic Morrison et al., “Chimeric human antibody molecules: Mouse anti calicheamicin V11 effectively suppresses growth and dissemination gen-binding domains with human constant region domains' P Natl of liver metastases in a Syngeneic model of murine neuroblastoma' Acad Sci USA 81:6851-6855 (Nov. 1984). Cancer Res 58:2925-2928 (1998). Morrison, “Success in specification” Nature 368(6474):812-813 Lonberg et al., “Antigen-specific human antibodies from mice com (Apr. 28, 1994). prising four distinct genetic modifications' Nature 368(6474):856 Mosmann, “Rapid colorimetric assay for cellular growth and Sur 859 (Apr. 28, 1994). vival: Application to proliferation and cytotoxicity assays' J Lonberg et al., “Human antibodies from transgenic mice' Int Rev Immunol Methods 65:55-63 (1983). Immunol 13(1):65-93 (1995). Munson et al., “Ligand: A versatile computerized approach for char Lowman et al., “Mutational analysis and protein engineering of acterization of ligand-binding systems' Anal Biochem 107:220-239 receptor-binding determinants in human placental lactogen' J Biol (1980). Chem 266(17): 10982-10988 (Jun. 15, 1991). Nagengast et al., “In vivo VEGF imaging with radiolabeled MacCallum et al., “Antibody-antigen interactions: contact analysis bevacizumab in a human ovarian tumor xenograft”. J Nucl Med and binding site topography” J Mol Biol 262:732-745 (1996). 48: 1313-1319 (2007). Mandler et al., “Immunoconjugates of geldanamycin and Anti-HER2 Nakamura et al., “Stereochemistry and total synthesis of dolastatin monoclonal antibodies: antiproliferative activity on human breast E” Tetrahedron Letters 36(28):5059-5062 (Jul 10, 1995). carcinoma cell lines' J National Cancer Institute 92(19): 1573-1581 Neuberger, “Generating high-avidity human mabs in mice” Nat (Oct. 4, 2000). Biotechnol 14(7):826 (Jul 1996). Mandler et al., “Modifications in synthesis strategy improved the Nicolaou et al., "Calicheamicin 0: A rationally designed molecule yield and efficacy of geldanamycin-herceptin immunoconjugates' with extremely potent and selective DNA cleaving properties and Bioconjugate Chem 13:786-791 (2002). apoptosis Inducing activity.” Agnew Chem Intl ED Engl33(2): 183 Mandler et al., “Synthesis and evaluation of antiproliferative activity 186 (1994). of a Geldanamycin-HerceptionTM immunoconjugate” Bioorg Med Niculescu-Duvaz et al., “Antibody-Directed Enzyme Prodrug Chem Lett 10:1025-1028 (2000). Therapy (ADEPT): A Review” Adv Drg Del Rev 26:151-172 (1997). US 8,771,966 B2 Page 6

(56) References Cited Ravetch et al., “Fc receptors' Annu Rev Immunol 9:457-492 (1991). Reyes et al., “Expression of human B-interferon cDNA under the OTHER PUBLICATIONS control of a thymidine kinase promoter from herpes simplex virus' Nature 297:598-601 (Jun. 17, 1982). Nikula et al., “A rapid, single vessel method for preparation of clini Riechmann et al., “Reshaping human antibodies for therapy” Nature cal grade ligand conjugated monoclonal antibodies' Nucl Med Biol 332:323-327 (Mar. 24, 1988). 22:387-390 (1995). Ripka et al., “Two chinese hamster ovary glycosylation mutants Nikula et al., “Alpha-emitting bismuth cyclohexylbenzyl DTPA con affected in the conversion of GDP-mannose to GDP-fucose' Arch structs of recombinant humanized anti-CD33 antibodies: Biochem Biophys 249(2):533-545 (Sep. 1986). Pharmacokinetics, bioactivity, toxicity and chemistry” JNucl Med Rodeberg et al., “Recognition of six-transmembrane epithelial anti 40(1):166-176 (Jan. 1999). Oda et al., “Enrichment analysis of phosphorylated proteins as a tool gen of the prostate-expressing tumor cells by peptide antigen-In for probing the phosphoproteome” Nat Biotechnol 19:379-382 (Apr. duced cytotoxic T lymphocytes' Clin Cancer Res 11:45.45-4552 ( 2001). 2005). Okazaki et al., “Fucose depletion from human IgG1 oligosaccharide Rodrigues et al., “Synthesis and 3-lactamase-mediated activation of enhances binding enthalpy and association rate between IgG1 and a cephalosporin-taxol prodrug” Chem Biol 2:223-227 (Apr. 1995). FycyRIIIa J Molec Biol 336:1239-1249 (2004). Roselli et al., “In vivo comparison of CHX-DTPA ligand isomers in Orlandi et al., “Cloning immunoglobulin varible domains for expres athymic mice bearing carcinoma xenografts' Cancer Biother Radio sion by the polymerase chain reaction” P Natl Acad Sci USA 14(3):209-220 (1999). 86:3833-3837 (1989). Rowland et al., “Drug localisation and growth inhibition studies of Orum et al., “Efficient method for constructing comprehensive vindesine-monoclonal anti-CEA conjugates in a human tumour murine Fab antibody libraries displayed on phage” Nucl Acids Res xenograft' Cancer Immunol Immunother 21:183-187 (1986). 21(19):4491-4498 (1993). Ruegg et al., “Improved in vivo Stability and tumor targeting of Perk et al., “Zr as a PET surrogate radioisotope for scouting Bismuth-labeled antibody” Cancer Res 50:4221-4226 (Jul. 15. biodistribution of the gherapeutic radiometals 'Y and ''Lu in 1990). tumor-bearing nude mice after coupling to the internalizing antibody Sastry et al., “Cloning of the immunological repertoire in Escherichia cetuximab” JNucl Med 46(11): 1898-1906 (Nov. 2005). coli for generation of monoclonal catalytic antibodies: construction Perk et al., Facile radiolabeling of monoclonal antibodies and other of a heavy chain variable region-specific cDNA library” PNatl Acad proteins with zirconium-89 or gallium-68 for PET imaging using Sci USA 86:5728-5732 (Aug. 1989). p-isothiocyanatobenzyl-desferrioxamine, Nature Protocols, pub Schellenberger et al., “Optical imaging of apoptosis as a biomarker of lished online: DOI: 10.1038/NPROT 2008.22). tumor response to chemotherapy” Neoplasia 5(3): 187-192 (2003). Perk et al., “p-Isothiocyanatobenzyl-desferrioxamine: a new bifunc Schier et al., “Identification of functional and structural amino-acid tional chelate for facile radiolabeling of monoclonal antibodies with residues by parsimonious mutagenesis' Gene 169: 147-155 (1996). Zirconium-89 for immuno-PET imaging Eu J Nucl Med Molec Senter et al., “Immunoconjugates comprised of drugs with impaired Imaging 37:250-259 (2010). cellular permeability: A new approach to targeted therapy P Am Perk et al., “Preparation and evaluation of Zr-zevalin for monitor Assoc Cancer Res 45: 144 (2004). ing of 'Y-zevalin biodistribution with positron emission Shalaby et al., “Development of humanized bispecific antibodies tomography” Eur J. Nucl Med Mol Imaging 33: 1337-1345 (2006). reactive with cytotoxic lymphocytes and tumor cells overexpressing Perk et al., “Quantitative PET imaging of met-expressing human the HER2 protooncogene” J. Exp Med 175:217-225 (Jan. 1992). cancer xenografts with Zr-labeled monoclonal antibody DN30” Shamis et al., “Bioactivation of self-immolative dendritic prodrugs EurJ Nucl Med Mol Imaging 35:1857-1867 (2008). by catalytic antibody 38C2'JAm ChemSoc. 126:1726-1731 (2004). Pettit et al., “Antineoplastic agents 365. Dolastatin 10 SAR probes' Sheriff et al., “Redefining the minimal antigen-binding fragment” Anti-Cancer Drug Design 13:243-277 (1998). Nature Struct Biol3(9):733-736 (Sep. 1996). Pettit et al., “Dolastatins 24: synthesis of (-)-dolastatin 10. X-ray Shields et al., “High resolution mapping of the binding site on human molecular structure of N,N-dimethylvalyl-valyl-dolaisoleuine tert IgG1 for FcyRI. FcyRII, FcyRIII, and FcRn and design of IgG1 butyl ester J Chem Soc Perkins Trans 1:859-863 (1996). variants with improved binding to the FcyR J Biol Chem Pettit et al., “Isolation of dolastatins 10-15 from the marine molus 276(9):6591-6604 (Mar. 2, 2001). Dolabella auricularia” Tetrahedron 49(41):9151-9170 (1993). Shively, 'F labeling for immuno-PET: Where speed and contrast Pettit et al., “Specific activities of dolastatin 10 and peptide deriva meet”J Nucl Med 48(2): 170-172 (Feb. 2007). tives against Cryptococcus neoformans' Antimicrob Agents and Sias et al., “ELISA for quantitation of the extracellular domain of Chemotherapy 42(11):2961-2965 (Nov. 1998). p185'' in biological fluids.” J Immunol Methods 132(1):73-80 ( Pettit et al., “The absolute configuration and synthesis of natural 1990). (-)-Dolastatin 10' J Am ChemSoc. 111:5463-5465 (1989). Sidhu et al., “Phage-displayed antibody libraries of synthetic heavy Pettit et al., “The dolastatins; 18: Stereospecific synthesis of chain complementarity determining regions' J Mol Biol338(2):299 dollaproine' Synthesis:719-725 (Jun. 1996). 310 (2004). Pettit Progress in the Chemistry of Organic Natural Products W. Simmons et al., “Expression of full-length immunoglobulins in Hertz et al...SpringerWienNewYork, vol.70:1-79 ( 1997). Escherichia coli: Rapid and efficient production of aglycosylated Pluckthun, A., “Mono- and bivalent antibody fragments produced in antibodies' J Immunol Methods 263:133-147 (2002). Eschericia coli: engineering, folding and antigen binding” Immunol Sims et al., “A Humanized CD18 Antibody Can Block Function Reviews(130): 151-188 (1992). Without Cell Destruction” J Immunol 151(4): 2296-2308 (Aug. 15, Presta et al., “Humanization of an antibody directed against IgE” J 1993). Immunol 151(5):2623-2632 (Sep. 1993). Singh et al., "Labeling of antibodies by in situ modification of thiol Presta et al., “Humanization of an anti-vascular endothelial growth groups generated from Selenol-catalyzed reduction of native disulfide factor monoclonal antibody for the therapy of solid tumors and other bonds” Anal Biochem 304(2): 147-156 (May 15, 2002). disorders' Cancer Res 57:4593-4599 (Oct. 1997). Skerra, A., “Bacterial expression of immunoglobulin fragments' Presta, "Antibody engineering Curr Opin Struct Biol 2:593-596 ( Curr Opin Immunol 5:256-262 ( 1993). 1992). Sone et al., “Dolabellin, a cytotoxic bisthiazole metabolite from the Proba et al., “Functional antibody single-chain fragments from the Seahare Dolabella auricularia: Structural determination and synthe cytoplasm of Escherichia coli: influence of thioredoxin reductase sis' J Org Chem 60:4774-4781 (1995). (TrxB)” Gene 159:203-207 ( 1995). Storm et al., “Effect of small changes in orientation on reaction rate' Ramm et al., “The periplasmic Escherichia coli peptidylprolyl JAm ChemSoc.94:5815-5825 (1972). cis,trans isomerase FkpA' J Biol Chem 275(22): 17106-17113 (Jun. Sun et al., "Enabling ScFvs as multi-drug carriers: a dendritic 2000). approach” Bioorg Med Chem 11:1761-1768 (2003). US 8,771,966 B2 Page 7

(56) References Cited Waterhouse et al., "Combinatorial infection and in vivo recombina tion: a strategy for making large phage antibody repertoires' Nucl OTHER PUBLICATIONS Acids Res 21(9):2265-2266 (1993). Williams et al., “Cloning and sequencing of human immunoglobulin Sun et al., “Syntheses of dendritic linkers containing chlorambucil V gene segments' Eur J Immunol 23: 1456-161 (1993). residues for the preparation of antibody-multidrug Williams et al., “Numerical selection of optimal tumor imaging immunoconjugates’ Bioorg Med Chem Lett 12:2213-2215 (2002). agents with application to engineered antibodies' Cancer Biother Suresh et al., “Bispecific monoclonal antibodies from hybrid Radio 16(1):25-35 (2001). hybridomas” Method Enzymol 121:210-228 (1986). Winter et al., “Making antibodies by phage display technology” Swartzman et al., "A homogeneous and multiplexed immunoassay Annu Rev Immunol 12:433-455 (1994). for high-throughput screening using fluorometric microVolume assay Wiseman et al., “Ibritumomab tiuxetan radioimmunotherapy for technology” Anal Biochem 271: 143-151 (1999). patients with relapsed or refractory non-Hodgkin lymphoma and Syrigos et al., “Antibody directed enzyme prodrug therapy (ADEPT): mild thrombocytopenia: a phase II multicenter trial” Blood a review of the experimental and clinical considerations' Anticancer 99(12):4336-4342 (2002). Res 19:605-614 (1999). Wiseman et al., “Phase I/II 90Y-Zevalin (yttrium-90 ibritumomab Szalai et al., “Geometric disassembly of dendrimers: Dendritic tiuxetan, IDEC-Y2B8) radioimmunotherapy dosimetry results in amplification” JAm Chem Soc 125:15688-15689 (2003). relapsed or refractory non-Hodgkin's lymphoma' Eur J. Nucl Med Tait et al., “Improved detection of cell death in vivo with Annexin V 27(7):766-777 (Jul 2000). radiolabeled by site-specific methods” JNucl Med 47(9): 1546-1553 Witzig et al., “Randomized controlled trial of yttrium-90 labeled (Sep. 2006). ibritumomab tiuxetan radioimmunotherapy versus rituximab Thorpe, "Antibody carriers of cytotoxic agents in cancer therapy: a immunotherapy for patients with relapsed or refractory low-grade, review” Monoclonal Antibodies '84: Biological and Clinical Appli follicular, or transformed B-cell Non-Hodgkin's lymphoma' J Clin cations (A Pinchera, G. Doria, F. Dammacco and A. Bargellesi, eds. Oncol 20(10):2453-2463 (May 15, 2002). pp. 475-506 (1985)). Witzig et al., “Treatment with ibritumomab tiuxetan radioim Tomlinson et al., “The repertoire of human germline VH sequences munotherapy in patients with rituximab-refractory follicular non reveals about fifty groups of VH segments with different hyperVari Hodgkin's lymphoma” J. Clin Oncol 20(15):3262-3269 (Aug. 1. able loops” J Mol Biol 227(3):776-798 (Oct. 5, 1992). 2002). Traunecker et al., "Bispecific single chain molecules (Janusins) tar Woyke et al., “In vitro activities and postantifungal effects of the get cytotoxic lymphocytes on HIV infected cells' EMBO J potent dolastatin 10 derivative auristatin PHE” Antimicrob Agents 10(12):3655-3659 (1991). Chemother 45(12):3580-3584 (Dec. 2001). Tutt et al., “Trispecific F(ab') derivatives that use cooperative sig Wu et al., “Arming antibodies: prospects and challenges for naling via the TCR/CD3 complex and CD2 to activate and redirect immunoconjugates' Nat Biotechnol 23(9): 1137-1146 (Sep. 23. resting cytotoxic T cells' J Immunol 147(1):60-69 (Jul. 1991). 2005). Urlaub et al., “Isolation of chinese hamster cell mutants deficient in Xu et al., “Diversity in the CDR3 region of V is sufficient for most dihydrofolate reductase activity P Natl Acad Sci USA 77(7):4216 antibody specificities” Immunity 13:37-45 (Jul. 2000). (Jul. 1980). Yamane-Ohnuki et al., “Establishment of FUT8 knockout Chinese Van Dongen et al., “Immuno-PET. A navigator in monoclonal anti hamster ovary cells: an ideal host cell line for producing completely body development and applications” Oncologist 12:1379-1389 ( defucosylated antibodies with enhanced antibody-dependent cellular 2007). cytotoxicity” Biotechnol Bioeng 87(5):614-622 (Sep. 5, 2004). Vaswani et al., “Humanized antibodies as potential therapeutic Yaniv, “Enhancing elements for activation of eukaryotic promoters' drugs” Ann Allergy Asthma Immunol 1:105-115 (1998). Nature 297(6): 17-18 (May 1982). Verel et al., “Zr Immuno-PET: Comprehensive Procedures for the Yansura et al., "Nucleotide sequence selection for increased expres production of Zr-labeled monoclonal antibodies' J Nucl Med sion of heterologous genes in Escherichia coli'METHODS: A Com 44(8): 1271-1281 (Aug. 2003). panion to Methods in Enzymology 4(2): 151-158 (1992). Verel et al., “Long-lived positron emitters zirconium-89 and iodine Yelton et al., "Affinity maturation of the BR96 anti-carcinoma anti 124 for scouting of therapeutic radioimmunoconjugates' Cancer body by codon-based mutagenesis' J Immunol 155:1994-2004 ( Biother Radiopharm 18:655-661 (2003). 1995). Verel et al., “Quantitative 'Zr immuno-PET for in vivo scouting of Yu et al., “The biosynthetic gene cluster of the maytansinoid antitu 'Y-labeled monoclonal antibodies in xenograft-bearing nude mice' mor agent ansamitocin from Actinosynnema pretiosum” PNatl Acad J Nucl Med 44:1663-1670 (2003). Sci USA99(12):7968-7973 (Jun. 11, 2002). Verhoeyen et al., “Reshaping human antibodies: Grafting an Zhanget al., “Epidermal growth factor receptor-related protein inhib antilysozyme activity” Science 239:1534-1536 (Mar. 1988). its cell growth and induces apoptosis of BXPC3 pancreatic cancer Vermes et al., “A novel assay for apoptosis. Flow cytometric detec cells” Cancer Res 65(9):3877-3882 (May 1, 2005). tion of phosphatidylserine expression on early apoptotic cells using Zheng et al., "Caspase-3 controls both cytoplasmic and nuclear fluorescein labeled Annexin V' J Immunol Methods 184:39-51 ( events associated with Fas-mediated apoptosis in vivo” PNatl Acad 1995). Sci USA95:13618-13623 (Nov. 1998). Vieira et al., “Production of single-stranded plasmic DNA” Method Zoller et al., “Oligonucleotide-directed mutagenesis using M13-de Enzymol 153:3-11 (1987). rived vectors: An efficient and general procedure for the production Vitetta et al., “Redesigning nature's poisons to create anti-tumor of point mutations in any fragment of DNA” Nucl Acids Res reagents' Science 238: 1098-1104 (1987). 10:6487-6504 (1987). Wang et al., "Structural characterization of the maytansinoid Boswell et al. et al., “Development of radioimmunotherapeutic and monoclonal antibody immunoconjugate, hun901-DM1, by mass diagnostic antibodies: an inside-out view” Nucl Med Biol 34:757 spectrometry” Protein Sci 14:2436-2446 (2005). 778 (2007). Ward et al., “Binding activities of a repertoire of single Holland et al., “Zr-DFO-J591 for ImmunoPET of Prostate-Specific immunoglobulin variable domains secreted from Escherichia coli” Membrane Antigen Expression In Vivo” J. Nucl Med 51:1293-1300 Nature 341:544-546 (Oct. 12, 1989). (2010).

U.S. Patent Jul. 8, 2014 Sheet 4 of 51 US 8,771,966 B2

U.S. Patent Jul. 8, 2014 Sheet 5 Of 51 US 8,771,966 B2

U.S. Patent Jul. 8, 2014 Sheet 6 of 51 US 8,771,966 B2

ºººº)

U.S. Patent Jul. 8, 2014 Sheet 8 of 51 US 8,771,966 B2

LB50 120 y24 Phage (+) 120.graft O-120 i

i

FL2 LOG: FL2 LOG - PE U.S. Patent Jul. 8, 2014 Sheet 9 Of 51 US 8,771,966 B2

10H1 Phage

100 10 10? 10 FL2 LOG: FL2 LOG - PE

67 Phage 37.48 Phage 10H1 Phage Ch 120 Phage 120 graft Phage

100 10 10? 1O FL2 LOG: FL2 LOG - PE U.S. Patent Jul. 8, 2014 Sheet 10 of 51 US 8,771,966 B2

FAS: AN 100 ... 120 chi" : - - - hu 24-67 80 - if " on 120

60 i 40

20

100 10 10? 10 FL2 LOG: FL2 LOG - PE

FACS SE

i

109 10 102 10 FL2 LOG: FL2 LOG - PE U.S. Patent Jul. 8, 2014 Sheet 11 of 51 US 8,771,966 B2

FAS: CYN

80 i 60

100 10 10? 10 FL2 LOG: FL2 LOG - PE FAS: 293 Steap 350

80

60

i 40

20

1O O M 1O 10? 10 FL2 LOG: FL2 LOG - PE U.S. Patent Jul. 8, 2014 Sheet 12 of 51 US 8,771,966 B2

FACS: PC3 Steap PS5.4 100

80 - 60 s 40 20

10 O 10 10 2 10 FL2 LOG: FL2 LOG - PE

FAS: NCA 8 ------4-87 or 120 i

100 10 102 10 FL2 LOG: FL2 LOG - PE U.S. Patent US 8,771,966 B2

U.S. Patent US 8,771,966 B2

|Puºla U.S. Patent US 8,771,966 B2

U.S. Patent Jul. 8, 2014 Sheet 16 of 51 US 8,771,966 B2

|-0001 (cuu) eunOWJOLun LueeW U.S. Patent Jul. 8, 2014 Sheet 17 Of 51 US 8,771,966 B2

{}{}

{}{}{}osnow/ºska U.S. Patent Jul. 8, 2014 Sheet 18 of 51 US 8,771,966 B2

Schematic Diagra of SEAP-1 in the Ce: exrane

Cell vie faine

aa 4-28 (WKMPRRNLEEDDY.)

A 12

FIG 12 U.S. Patent Jul. 8, 2014 Sheet 19 Of 51 US 8,771,966 B2

Expression of SEAP Expression of SEAP if fostate are geis "cstate Car ster scies Exogenous Modeis Exogenous ivodeis 293 SEA 35 C3 SEA S5.4

Expression of SEAF Expression of SEAF in Prostate C3 cer scies Presists are is Endogenius viodeis Engioger Otis Riodels NC; CA 77

F.G. 13C U.S. Patent Jul. 8, 2014 Sheet 20 of 51 US 8,771,966 B2

s t

Citi ( , A

- - - Corto we fia

-0- 120 Chivc MMAE -- 120 Chinc MMAF ...A. 120 Chime MMAF 10x10-i- 120 Chivo MMAF 10-4 O-3 1-2 Ori 1g C O?

Ab 14A COrc, (agfm)

- - CCC is fiAF - - - Controlve MMAE -0- 120 Chivo MMAE -v. 120 Chirno MMAF 2.0x Of ...A. 120 Chirno MMAF -- 120 Chivo MMAF

5x 106

5,0x05

10-4 C-3 10-2 10- g3 10. O2

A. Conc, gin) U.S. Patent Jul. 8, 2014 Sheet 21 of 51 US 8,771,966 B2

4x6 3x6 m 2x06 - 11 x 06 x 6 9 x 5 8x8 or C. |-a- Critis C. WAF 5- 0 - Control vc MMAE 5-0-120 Chivo MMAE 30x105 -v- 120 Chime MMAF 2.0x105-...-a - 120 Chime MMAF 10x105-- - 120 Chivo MMAF LNCap im re-aara O-4 O-3 1-2 Ori 1CO i O2 A. FG. 14C Conc. (gfm)

1.3x106 2x16 x 6 x6 9. Ox O5. 8x5 7.0x105 6.0x105 -- - Control mc MMAF 5.0x105 -- - - Controlvo MMAE 4x15 -o- 2O C wi: A 3.0x105 -v- 120 Chimo MMAF 2. Ox OS ...A. 120 Chime MMAF 10x10 - . 120 Chivc MAAF 3 fest 10-4 C-3 1-2 1-1 10 4t 2

A. ConC. (agf) U.S. Patent Jul. 8, 2014 Sheet 22 of 51 US 8,771,966 B2

- - Control mc MMAF - O - Controivc MMAE -0-120 Chivc MMAE |-v- 120 Chime MMAF 1.8x10 ...A. 120 Chime MMAF 1.8x10-in. 120 Chivc MMAF 8x8 to

6x6 16x16 6x06 5x 6 4x (6 4x6

3x06 2x 6 1 x 16 O-4 1O-3 1-2 C-1 1) O Ai) FIG. 1. Conc. (g/ml) U.S. Patent Jul. 8, 2014 Sheet 23 Of 51 US 8,771,966 B2

U.S. Patent Jul. 8, 2014 Sheet 24 of 51 US 8,771,966 B2

s S Liu Liu Liu Liu Liu Liu Liu Liu Liu Liu Liu SPSP. SP. SP. SP. SP. SP. SP. SP. C2.9 4.4.4.4.4.4.4.4.4.4.4.

------coco cocococooooooooo.

CO CO CO.CO CO CO CO CO CO CO CO ------SPSP. SP. SP. SP. SP. SP. SP. SP. C2.9

HHHHHHHHHHH SSSSSSSSSS

( SS C.3 Y. S.N CN s is e O O O O - - - - O CN n CC H CfO U.S. Patent US 8,771,966 B2

$3.ÅÅ“ U.S. Patent Jul. 8, 2014 Sheet 26 of 51 US 8,771,966 B2

$1 sh; sh; U.S. Patent Jul. 8, 2014 Sheet 27 Of 51 US 8,771,966 B2

293 Steap c. 850 100

80 60

s 40

20

FL2 LOG: FL2 LOG - PE PC3 Steap c.S5.4

1OO

60

i 40

FL2 LOG: FL2 LOG - PE U.S. Patent Jul. 8, 2014 Sheet 28 of 51 US 8,771,966 B2

N

60

i 40

102 163 FL2 LOG: FL2 LOG - PE

as -huSteap TDC AMVV (CV205C) west of sma huSteap 1 TDC a? A's *- /RECA i8C) vicE ? w gir - ros.-- w s * (ADCs consolet) huSteap AIDC ?(t) vci, a rarer

160 10 O2 03 F.G. 1 FL2 LOG: FL2 LOG - PE U.S. Patent Jul. 8, 2014 Sheet 29 Of 51 US 8,771,966 B2

-hlsteap TDC i| (LCV205C) voE -huSieap1 TDC f (HCA1180) vcE a ** "w ow. new we we

husteap1 ADC ?essive

103 16 102 163 FL2 LOG: FL2 LOG - PE

TDC Control (VCE) husteap foc (LCV205C) vcE huSteap1 TDC (HCA118C) vCE

-busteap(std) woE ADC

100 102 103 FL2 LOG: FL2 LOG - PE U.S. Patent Jul. 8, 2014 Sheet 30 Of 51 US 8,771,966 B2

10x108 or E. K. --ADC sta 10x108 Control (veE) 9.5x (3- -o- DC Coffic S(x (3- (vcE) 8.5x1 (5. S. - i. -v- husteapi TDC 8x105- -steeS--ti- (HCA) i8C) woE 75x105- w --A-huSteapt TDC 7.0x105- S-en '. (CV205C vici 6.5x105- sa , -O-huSteapi ADC 8. Ox OS- '. - 5.5x5. 5.x1 (5- 4.5x1(5- 40x105 3.5x105-oporo- footo 10-4 102

9. (x1 (5 S. s. 2.fs - - ADC Sid Control (ycE) 8.0x1 (5 oxxon DC Contro:

oiro 1 -V-huSteap TDC

t S. a huSteap TDC 8. Ox OS (CV205C) vot: ---. -- huSteap1 ADC 5.x 05 -...-

0-4 10-3 -2 1O- 10t C 102 Ab Conc. (gfm) F.G. 19 U.S. Patent Jul. 8, 2014 Sheet 31 of 51 US 8,771,966 B2

18x6 3K: wo. ADC sid Cottro (VCE) 1.5x106 axxar DC Contil s (vcE) 1.2x106- SS -v-huSteap. TDC * (HCA205C) vCE O Ox6 -A-huSteapt TDC (LCVi8CycE 75x15- -- huSteap ADC (std) voE 5.0x105 2.5x105 0 1-4 O-3 1-2 1)- (0 10 102 Ab Conc. (gfm) FIG. 19C U.S. Patent US 8,771,966 B2

U.S. Patent US 8,771,966 B2

WS -f- (cuu) eunioA Jouni ueeW U.S. Patent US 8,771,966 B2

U.S. Patent Jul. 8, 2014 Sheet 36 of 51 US 8,771,966 B2

U.S. Patent Jul. 8, 2014 Sheet 37 Of 51 US 8,771,966 B2

3. s

s s

U.S. Patent Jul. 8, 2014 Sheet 38 Of 51 US 8,771,966 B2

U.S. Patent Jul. 8, 2014 Sheet 39 Of 51 US 8,771,966 B2

inited Aalysis - i), 2 r is Origina: Thio 2W24. a hio Mab O2 as A 739 f NCapSSay BR f $9.cells 0.015 K e rv a 0.010

{{O} n s-e- 1. 1) Oi {< 103 Total (Nanomolar)

titled Aalysis -

Origina: Thio 0.025 2W2á. 8 : it a OO2Ow : Assay 739: 3. NCaFBR Celis - 0.015 KD i? g 0.010

O. On U.S. Patent Jul. 8, 2014 Sheet 40 of 51 US 8,771,966 B2

titled Atalysis - O2 as

Original Thio 8. is 0.014 Assay 740: a 0.012 KD (; F O. O.8- 0.006 O4 m. O. 2n MV O, CO- 8 s--a- s 1 10) 11 12 13 Totai (Nanomoiar)

titled Aalysis - 0.020 Original Thio 8w 122: .8- = Thio Mab 8 O. : - ASSay 740. 0.012 NCa.3R Ces SPE a 00100.008 0.006 O, Cdr. O. On O. Or Bound (Piconoiar)

FIG.25D U.S. Patent Jul. 8, 2014 Sheet 41 of 51 US 8,771,966 B2

O8. intitiei Aaiysis - 8 2OW24 (16m his SGV O4- 8 ... his al-S 2. Assay 733: a .NCap BR Cells XD 23 ri s O8 as 0.006 4 {{2. (i. s s ra 18 O 12 1O-3 Total (Nanomciar

inted Araiysis - 8vo 12OV24. 6.016. " hio SGV 8 : hi Wai-S w Y 3 0.012- 8. Assay 733. i. NCap.BR Cells 0.010 KO 2.3 f S O8m OO6r O)4 r. OO2m

n U.S. Patent Jul. 8, 2014 Sheet 42 of 51 US 8,771,966 B2

institied Araiysis - On 8 2W22. it. SG are 8 : hio al-S ... .m. Assay 734: 2 as PBR Cls g ooo KD = 1.9nM S 0.006 3.04 as 2. 3. On r s-s-s 1. t 12 1. otai (Nanomolar)

rtied Araiysis - 1 ... 3" is V4. SG aw 8 : his fia-S

O Assay 734: 9,010 .NCaP.BRKD-10 inCeils 0.008 ar C 0.006 a 3. Cain r Ors 8 *- 8 : 8 8 . 32 a; Bound (Picomoiar)

U.S. Patent Jul. 8, 2014 Sheet 43 of 51 US 8,771,966 B2

{18or Jntitied Analysis - Oficia 8. N f E. 4- a = Thio Mab is .32 Assay 737: 0.010 293.850 Ceils K. : 7 i? 0.008 0.006 (O(4- :- G.O.O. - S 10 of O Oz to3 otai (Nanomiciar)

2. rtitied Araiysis - (, 8 N Original2,2a: Thio : 6 re = Thio Mab 3 ti Assay 737: i. 3.12 93.3 is res . On K. c. 7 if S 8 C O6 a - (2. m 8 : 8 Fa .3 2. 36 Bound (Picomolar)

U.S. Patent Jul. 8, 2014 Sheet 44 of 51 US 8,771,966 B2

irtitied Aaysis - O. : -

8 OriginalWa. Thio i-T ... Thia- Mah ( 0.010 293LE50Assay 738. cells e 0.008 KC-20 nM S 0906 f (.54 a CO2 as

O) t3Swsex8 Ot a s Totai (Nanomoiar)

(18- ratited Aaysis - w N orghio 0.14 - 8 : it a m Assay 738. a 93.5) is KO 3: 2. Of O. O.3a (8 a. COs a O. O. (OO 8 8 O 78 158 23. 32 39 Bound (Picoinciar)

U.S. Patent Jul. 8, 2014 Sheet 45 of 51 US 8,771,966 B2

initied Araiysis - i. i2. 2O2: 1. is SGW 0.1 O. 8 it to Wai-S 09 a ASSay 735. O.O.3a 293.850 Ceis K c 2.7 rif One O)5- O4s 3. Ose O.Or O. On cs: 1? 1) : t3 Total (Nanomolar)

3- ited Artaysis - 1

2O24. (12 his SGV s 0.10 c Assay 735: i. 0.08 293.850 Ceis res. KE) c 2 fi 0.06 f Cairo Ow COO 8 86 172 258 344. 43 Bound (Picomoiar)

U.S. Patent Jul. 8, 2014 Sheet 46 of 51 US 8,771,966 B2

i. i2.so in titled Analysis - O. : : - 2OV24. i). On

Assay 738. 293.35 Ces K} : 28

or "S 1) 12 O3 otai (Nanomoiar)

FIG 26G

titled Araiysis - 2. k T'Siy (Or 8 8 Wi-S

Assay 738:M i3 0.08 2.93.350 leis re. San K) as 2.3 M s {5 C O.O4.

(O2

CO r S. 78 52 28 34. 38. Bound (Piconoar)

U.S. Patent Jul. 8, 2014 Sheet 47 Of 51 US 8,771,966 B2

U.S. Patent Jul. 8, 2014 Sheet 48 of 51 US 8,771,966 B2

293-STEAP, .358

a X s

0- ri?ing fitti", Trierrrrrrn g O 102 s 04

G. 28A FL2-H: PE FL2-Height

293-SEAP, .35

s

irritor first rt is 1. O2 13 104

XY F.2-H: PE F 2-Height U.S. Patent Jul. 8, 2014 Sheet 49 Of 51 US 8,771,966 B2

293-SEAP, .353 1 on

80 m

x 6 m (S e

s

ex Sš 4 Oo

2 m

: ; : Fir riffiti riff O 102 13 104 F2

Na o

rur re-3X 8 m

x 6 a.

r

e S s & 40 or

w

0- a & i ; ; ; ; (0 10 O2 (3 14

FL2-H: PE FL2-Height

U.S. Patent Jul. 8, 2014 Sheet 51. Of 51 US 8,771,966 B2

Zirconium-83-SEAF. PE irraging

lica. 96. lucap 70 Cap 77 Lucap 35V %Dig is 2.65 %Dig - 4.68 %Dig - 7, 18 % Cigar 10.5 +/- 2.3 Animais Dosed OOuCi Z89-anti-Steap and imaged 4 Days later No-o- F.G. 30 US 8,771,966 B2 1. 2 MMUNO-PET IMAGING OF ANTIBODES member of cell Surface serpentine transmembrane antigens. It AND IMMUNOCONUGATES AND USES is expressed predominantly in the prostate cancer, and thus THEREFOR members of this family have been termed “STEAP' (Six Transmembrane Epithelial Antigens of the Prostate). Human RELATED APPLICATION STEAP proteins exhibit a high degree of structural conserva tion within the family but show no significant structural This application is a continuation of International Appli homology to any known human proteins. STEAP-1 appears cation No. PCT/US2011/038923, filed Jun. 2, 2011 claiming to be a type IIIa membrane protein expressed predominantly the benefit of U.S. Provisional Patent Application Ser. No. in prostate cells in normal human tissues. Structurally, 61/351,195, filed on Jun. 3, 2010, which application are fully 10 STEAP-1 is a 339 amino acid protein characterized by a incorporated herein by reference. molecular topology of six transmembrane domains and intra cellular N- and C-termini, Suggesting that it folds in a "ser SEQUENCE LISTING pentine' manner into three extracellular and two intracellular loops. STEAP-1 protein expression is maintained at high The instant application contains a Sequence Listing Sub 15 levels across various states of prostate cancer. STEAP-1 is mitted via EFS-Web and hereby incorporated by reference in highly over-expressed in other human cancers such as lung its entirety. Said ASCII copy, created on Nov. 29, 2012, is and colon. Murine antibodies have been raised to human named P4460R1 US Sequence listing.TXT, and is 66,025 STEAP-1 fragments and the antibodies were shown to bind bytes in size. STEAP-1 on the cell surface (see US Patent Application No. 20040253232A1). FIELD OF THE INVENTION Antibody-based therapy has proved very effective in the treatment of various cancers. For example, HERCEPTINR) The present invention relates to anti-STEAP-1 antibodies and RITUXANR) (both from Genentech, S. San Francisco), and immunconjugates thereof. The invention further relates have been used successfully to treat breast cancer and non to methods of using anti-STEAP-1 antibodies and immun 25 Hodgkin’s lymphoma, respectively. HERCEPTINR) is a conjugates thereof. recombinant DNA-derived humanized monoclonal antibody that selectively binds to the extracellular domain of the BACKGROUND human epidermal growth factor receptor 2 (HER2) proto oncogene. HER2 protein overexpression is observed in In humans, prostate cancer is one of the most commonly 30 25-30% of primary breast cancers. RITUXANR) is a geneti diagnosed malignancies in males and is the second leading cally engineered chimeric murine/human monoclonal anti cause of cancer related death in men. The American Cancer body directed against the CD20 antigen found on the surface Society estimates that for the year 2000, 180,400 new cases of of normal and malignant B lymphocytes. Both these antibod prostate cancer will be diagnosed with 31,900 deaths from the ies are produced in CHO cells. disease. In advanced stages, prostate cancer metastasizes to 35 The use of antibody-drug conjugates for the local delivery the bone. While advances in early diagnosis and treatment of of cytotoxic or cytostatic agents, i.e. drugs to kill or inhibit locally confined tumors have been achieved, prostate cancer tumor cells in the treatment of cancer (Syrigos and Epenetos is incurable once it has metastasized. Patients with metastatic (1999) Anticancer Research 19:605-614; Niculescu-Duvaz prostate cancer on hormonal therapy will eventually develop and Springer (1997) Adv. Drg Del. Rev. 26:151-172; U.S. Pat. anandrogen-refractory (androgen independent) state that will 40 No. 4,975.278) allows targeted delivery of the drug moiety to lead to disease progression and death. Currently, prostate tumors, and intracellular accumulation therein, where sys specific antigen (PSA) is the most widely used tumor marker temic administration of these unconjugated drug agents may for Screening, diagnosis, and monitoring prostate cancer. result in unacceptable levels of toxicity to normal cells as well However, widespread use of PSA as a tool for screening is as the tumor cells sought to be eliminated (Baldwin et al., controversial since PSA fails to discriminate accurately 45 (1986) Lancet pp. (Mar. 15, 1986):603-05; Thorpe, (1985) between benign and malignant prostate disease. Antibody Carriers Of Cytotoxic Agents. In Cancer Therapy: Depending on the stage of the cancer, prostate and bladder A Review.” in Monoclonal Antibodies 84: Biological And cancer treatment involves one or a combination of the follow Clinical Applications, A. Pinchera et al. (ed.s), pp. 475-506). ing therapies: Surgery to remove the cancerous tissue, radia Maximal efficacy with minimal toxicity is sought thereby. tion therapy, chemotherapy, androgen deprivation (e.g., hor 50 Both polyclonal antibodies and monoclonal antibodies have monal therapy) in the case of prostate cancer. While Surgical been reported as useful in these strategies (Rowland et al., or radiation therapy significantly improves Survival in (1986) Cancer Immunol. Immunother. 21:183-87). Drugs patients with early stages of the disease, the therapeutic used in these methods include daunomycin, doxorubicin, options are very limited for advanced cases, particularly for methotrexate, and vindesine (Rowland et al., Cancer Immu tumor recurrences following hormone ablation. The majority 55 nol. Immunother. 21:183-87 (1986)). Toxins used in anti of patients who undergo hormonetherapy progress to develop body-toxin conjugates include bacterial toxins such as diph androgen-independent disease. Currently, there is no effec theria toxin, plant toxins such as ricin, Small molecule toxins tive treatment for the 20-40% of prostate cancer patients who such as geldanamycin (Kerr etal (1997) Bioconjugate Chem. develop recurrent disease after Surgery or radiation therapy, 8(6):781-784; Mandleretal (2000) Journal of the Nat. Cancer or for those in whom the cancer has metastasized at the time 60 Inst. 92(19): 1573-1581; Mandler et al (2000) Bioorganic & of diagnosis. Chemotherapy has its toxic side effects, espe Med. Chem. Letters 10:1025-1028; Mandler et al (2002) cially in elderly patients. Development of new forms of Bioconjugate Chem. 13:786-791), maytansinoids (EP therapy especially for disease refractory to androgen depri 1391213: Liu et al., (1996) Proc. Natl. Acad. Sci. USA Vation is an urgent need in the management of prostatic car 93:8618-8623), and calicheamicin (Lode etal (1998) Cancer cinoma. 65 Res. 58:2928; Hinman et al (1993) Cancer Res. 53:3336 The identification of a novel cell surface antigen, STEAP-1 3342). The toxins may effect their cytotoxic and cytostatic has been described (see U.S. Pat. No. 6,329.503). STEAP-1 is effects by mechanisms including tubulin binding, DNA bind US 8,771,966 B2 3 4 ing, or topoisomerase inhibition (Meyer, D. L. and Senter, P. CD20-expressing cancers and immune disorders; (iv) anti D. “Recent Advances in Antibody Drug Conjugates for Can EphB2 antibodies 2H9 and anti-IL-8 for treatment of colorec cer Therapy” in Annual Reports in Medicinal Chemistry, Vol tal cancer (Mao, et al (2004) Cancer Research 64(3):781 38 (2003) Chapter 23, 229-237). Some cytotoxic drugs tend 788); (v) E-selectin antibody (Bhaskar et al (2003) Cancer to be inactive or less active when conjugated to large antibod Res. 63:6387-6394); and (vi) other anti-CD30 antibodies ies or protein receptor ligands. (WO 03/043583). Monomethylauristatin (MMAE) has also ZEVALINR) (ibritumomab tiuxetan, Biogen/Idec) is an been conjugated to 2H9, an antibody against EphB2R which antibody-radioisotope conjugate composed of a murine IgG1 is a type 1 TM tyrosine kinase receptor with close homology kappa monoclonal antibody directed against the CD20 anti between mouse and human, and is over-expressed in colorec gen found on the Surface of normal and malignant B lympho 10 cytes and '''In or 'Y radioisotope bound by a thiourea tal cancer cells (Mao et al (2004) Cancer Res. 64:781-788). linker-chelator (Wiseman et al (2000) Eur, Jour. Nucl. Med. Monomethylauristatin MMAF, a variant of auristatin E 27(7):766-77: Wiseman et al (2002) Blood 99(12):4336-42: (MMAE) with a phenylalanine at the C-terminus (U.S. Pat. Witzigetal (2002).J. Clin. Oncol. 20(10):2453-63; Witziget No. 5,767.237; U.S. Pat. No. 6,124,431), has been reported to al (2002) J. Clin. Oncol. 20(15):3262-69). Although ZEVA 15 be less potent than MMAE, but more potent when conjugated LIN has activity against B-cell non-Hodgkin’s Lymphoma to monoclonal antibodies (Senter et al. Proceedings of the (NHL), administration results in severe and prolonged American Association for Cancer Research, Volume 45, cytopenias in most patients. MYLOTARGTM (gemtuzumab Abstract Number 623, presented Mar. 28, 2004). Auristatin F oZogamicin, Wyeth Pharmaceuticals), an antibody drug con phenylene diamine (AFP); a phenylalanine variant of MMAE jugate composed ofahu CD33 antibody linked to calicheami was linked to an anti-CD70 mAb, 1F6, through the C-termi cin, was approved in 2000 for the treatment of acute myeloid nus of 1F6 via a phenylene diamine spacer (Law et al. Pro leukemia by injection (Drugs of the Future (2000) 25(7):686; ceedings of the American Association for Cancer Research, U.S. Pat. Nos. 4,970, 198; 5,079,233; 5,585,089; 5,606,040: Volume 45, Abstract Number 625, presented Mar. 28, 2004). 5,693,762; 5,739,116; 5,767.285; 5,773,001). Cantuzumab There exists a need in the art for additional drugs to treat mertansine (Immunogen, Inc.), an antibody drug conjugate 25 various cancers such as cancers and metatases of cancers in composed of the huC242 antibody linked via the disulfide the prostate, lung and colon. Particularly useful drugs for this linker SPP to the maytansinoid drug moiety, DM1, is being purpose include prostate, lung or colon cell targeted anti developed for the treatment of cancers that express CanAg STEAP-1 antibody-drug conjugates having a significantly antigen, Such as colon, pancreatic, gastric, and others. MLN lower toxicity, yet useful therapeutic efficiency. These and 2704 (Millennium Pharm. BZL Biologics, Immunogen Inc.), 30 other limitations and problems of the past are addressed by an antibody drug conjugate composed of the anti-prostate the present invention. specific membrane antigen (PSMA) monoclonal antibody Molecular imaging is an important tool in the development linked to the maytansinoid drug moiety, DM1, is under devel and evaluation of novel pharmaceuticals. Immuno-positron opment for the potential treatment of prostate tumors. The emission tomography (ImmunoPET) is a rapidly emerging same maytansinoid drug moiety, DM1, was linked through a 35 method for tracking and quantifying monoclonal antibodies non-disulfide linker, SMCC, to a mouse murine monoclonal (mAbs) in vivo as it efficiently combines the high sensitivity antibody, TA. 1 (Charietal. (1992) Cancer Research 52:127 of PET with the high specificity of mAbs. ImmunoPET 131). This conjugate was reported to be 200-fold less potent aspires to be the clinical method of choice for non-invasive than the corresponding disulfidelinker conjugate. The SMCC diagnosis providing “comprehensive immunohistochemical linker was considered therein to be "noncleavable.” 40 staining in Vivo” (van Dongen G. A. et al. “Immuno-PET: a Several short peptidic compounds have been isolated from navigator in monoclonal antibody development and applica the marine mollusk, Dolabella auricularia, and found to have tions” Oncologist 2007: 12:1379-89; Williams et al. (2001) biological activity (Pettituetal (1993) Tetrahedron 49:9151: Cancer Biother Radiopharm 16:25-35; Holliger et al (2005) Nakamura et al (1995) Tetrahedron Letters 36:5059-5062: Nat Biotechnol 23:1126–36)). The development of Zr Sone et al (1995) Journal Org Chem. 60:4474). Analogs of 45 based ImmunoPET has enabled the measurement of target these compounds have also been prepared, and some were expression and antibody biodistribution in mice and humans; found to have biological activity (for a review, see Pettit etal clinical ImmunoPET with Zr-Trastuzumab shows hetero (1998) Anti-Cancer Drug Design 13:243-277). For example, geneity in liver and bone Mets (Dijkers et al., Nature Vol. 87 auristatin E (U.S. Pat. No. 5,635,483) is a synthetic analogue Number 5 (May 2010). of the marine natural product Dolastatin 10, an agent that 50 PET imaging systems create images based on the distribu inhibits tubulin polymerization by binding to the same site on tion of positron-emitting isotopes in the tissue of a patient. tubulin as the anticancer drug Vincristine (G. R. Pettit, (1997) The isotopes are typically administered to a patient by injec Prog. Chem. Org. Nat. Prod. 70:1-79). Dolastatin 10, aurista tion of probe molecules that comprise a positron-emitting tin PE, and auristatin E are linear peptides having four amino isotope, such as F-18, C-11, N-13, or O-15, covalently acids, three of which are unique to the dolastatin class of 55 attached to a molecule that is readily metabolized or localized compounds, and a C-terminal amide. in the body (e.g., glucose) or that chemically binds to receptor The auristatin peptides, auristain E (AE) and monomethy sites within the body. In some cases, the isotope is adminis lauristatin (MMAE), synthetic analogs of dolastatin, were tered to the patient as an ionic solution or by inhalation. Small conjugated to: (i) chimeric monoclonal antibodies chBR96 immuno-PET imaging agents, such as Fab antibody frag (specific to Lewis Y on carcinomas); (ii) cAC10 which is 60 ments (50 kDa) or diabodies, paired dimers of the covalently specific to CD30 on hematological malignancies (Klussman, associated VH-VL region of Mab, 55 kDa (Shively et al et al (2004), Bioconjugate Chemistry 15(4):765-773; Dor (2007) J Nucl Med 48:170-2), may be particularly useful onina et al (2003) Nature Biotechnology 21 (7):778-784: since they exhibit a short circulation half-life, high tissue "Monomethylvaline Compounds Capable of Conjugation to permeability, and reach an optimal tumor to background ratio Ligands'; Francisco et al (2003) Blood 102(4): 1458-1465; 65 between two to four hours after injection facilitating the use US 2004/0018194; (iii) anti-CD20 antibodies such as Rit of short half-life isotopes such as the widely available 18F uxan R (rituximab) (WO 04/032828) for the treatment of (109.8 min). US 8,771,966 B2 5 6 The recitation of any reference in this application is not an cell under conditions suitable for expression of the polynucle admission that the reference is prior art to this application. All otide encoding the antibody, and isolating the antibody. references cited herein, including patents, patent applications In one aspect, an antibody that binds to STEAP-1 and publications, are incorporated by reference in their expressed on the surface of a cell is provided. In one embodi entirety. 5 ment, the antibody binds to an epitope within a region of human or mouse STEAP-1. In one embodiment, the cell is SUMMARY mammalian cell. In one embodiment, the cell is a human cell. In one embodiment, the cell is a cancer cell. In one embodi The invention provides anti-STEAP-1 antibodies and ment the cell is a prostate, lung or colon cell. In one embodi methods of using the same. 10 ment the cancer cell is a prostate cancer cell. In another In one aspect, an antibody that binds to STEAP-1 is pro embodiment, the cell from a metastasis of a primary prostate, vided, wherein the antibody comprises a light chain variable lung or colon cancer. In certain embodiments, any of the above antibodies is a domain comprising the amino acid sequence shown in FIG. monoclonal antibody. In one embodiment, the antibody is an 2A (SEQ ID NO:6) or a heavy chain variable domain com antibody fragment selected from a Fab, Fab'-SH, Fv, sclv, or prising the amino acid sequence shown in FIG. 2B (SEQ ID (Fab') fragment. In one embodiment, the antibody is human NO:9). In one aspect, an antibody that binds to STEAP-1 is ized. In one embodiment, the antibody is human. provided, wherein the antibody comprises a light chain vari In one aspect, a method of detecting the presence of able domain comprising the amino acid sequence shown in STEAP-1 in a biological sample is provided, the method FIG. 2A (SEQID NO:6) and a heavy chain variable domain comprising contacting the biological sample with any of the comprising the amino acid sequence shown in FIG. 2B (SEQ above antibodies under conditions permissive for binding of ID NO:9). the antibody to STEAP-1, and detecting whether a complex is In one aspect, an antibody that binds to STEAP-1 is pro formed between the antibody and STEAP-1. In one embodi vided, wherein the antibody comprises a heavy chain variable ment, the biological sample comprises prostate cells. In one domain having at least 90%, at least 91%, at least 92%, at least 25 embodiment, the biological sample is from a mammal expe 93%, at least 94%, at least 95%, at least 96%, at least 97%, at riencing or Suspected of experiencing a prostate cell disorder least 98%, or at least 99% sequence identity to an amino acid and/or a cell proliferative disorder of cells or tissues includ sequence of SEQ ID NOS:9 or 10. In one embodiment, the ing, but not limited to, prostate, lung, colon, bladder, and antibody comprises aheavy chain variable domain of SEQID ovarian cancer and Ewing's sarcoma as well as metastases of NOs:9 or 10. In one embodiment, the antibody comprises a 30 primary prostate, lung, colon, bladder, and ovarian cancers heavy chain variable domain framework region 1 of SEQID and Ewings sarcoma. See, for example, (see U.S. Pat. No. NO:25 or a heavy chain variable domain framework region 2 6,329,503; and Rodeberg, D. A. et al., Clin. Cancer Res. of SEQ ID NO:75 or 76 or 77 or a heavy chain variable 11(12):4545-4552 (2005)). domain framework region 3 of SEQID NO:78 or 79. In one aspect, a method of diagnosing a cell proliferative In one aspect, the antibody comprises a light chain variable 35 disorder associated with increased expression of STEAP-1 is domain having at least 90%, at least 91%, at least 92%, at least provided, the method comprising contacting a test cell with 93%, at least 94%, at least 95%, at least 96%, at least 97%, at any of the above antibodies; determining the level of expres least 98%, or at least 99% sequence identity to an amino acid sion of STEAP-1 by detecting binding of the antibody to sequence of SEQID NO:6. In one embodiment, the antibody STEAP-1; and comparing the level of expression of comprises a light chain variable domain of SEQID NO:6. 40 STEAP-1 by the test cell with the level of expression of In one aspect, an antibody that binds to STEAP-1 is pro STEAP-1 by a control cell, wherein a higher level of expres vided, wherein the antibody comprises a heavy chain variable sion of STEAP-1 by the test cell as compared to the control domain having at least 90%, at least 91%, at least 92%, at least cell indicates the presence of a cell proliferative disorder 93%, at least 94%, at least 95%, at least 96%, at least 97%, at associated with increased expression of STEAP-1. In one least 98%, or at least 99% sequence identity to an amino acid 45 embodiment, the test cell is a cell from a patient suspected of sequence of SEQ ID NOS:9 or 10. In one embodiment, the having a cell proliferative disorder, such as a prostate prolif antibody comprises aheavy chain variable domain of SEQID erative disorder. In one embodiment, the cell proliferative NOs:9 or 10. In one embodiment, the antibody comprises a disorder is selected from prostate cell disorders including but heavy chain variable domain framework region 1 of SEQID not limited to prostate cancer. In one embodiment, the method NO:25 or a heavy chain variable domain framework region 2 50 comprises determining the level of expression of STEAP-1 of SEQ ID NO:75 or 76 or 77 or a heavy chain variable on the Surface of the test cell (Such as, for example, a prostate domain framework region 3 of SEQ ID NO:78 or 79. In an cancer cell) and comparing the level of expression of embodiment, the antibody further comprises a light chain STEAP-1 on the Surface of the test cell with the level of variable domain having at least 90%, at least 91%, at least expression of STEAP-1 on the surface of the control cell 92%, at least 93%, at least 94%, at least 95%, at least 96%, at 55 (such, for example, as a normal prostate cell other than an least 97%, at least 98%, or at least 99% sequence identity to abnormally proliferating prostate cell). an amino acid sequence of SEQ ID NO:6. In one embodi In one aspect, a method of diagnosing a cell proliferative ment, the antibody comprises a light chain variable domain of disorder associated with an increase in cells, such as prostate SEQID NO:6. cells, expressing STEAP-1 is provided, the method compris In certain embodiments, a polynucleotide encoding any of 60 ing contacting a test cells in a biological sample with any of the above antibodies is provided. In one embodiment, a vector the above antibodies; determining the level of antibody bound comprising the polynucleotide is provided. In one embodi to test cells in the sample by detecting binding of the antibody ment, a host cell comprising the vector is provided. In one to STEAP-1; and comparing the level of antibody bound to embodiment, the host cell is eukaryotic. In one embodiment, cells in a control sample, wherein the level of antibody bound the host cell is a Chinese hamster ovary (CHO) cell. In one 65 is normalized to the number of STEAP-1-expressing cells in embodiment, a method of making an anti-STEAP-1 antibody the test and control samples, and wherein a higher level of is provided, wherein the method comprises culturing the host antibody bound in the test sample as compared to the control US 8,771,966 B2 7 8 sample indicates the presence of a cell proliferative disorder respectively, including the IgG subclasses: IgG1, IgG2, IgG3. associated with cells expressing STEAP-1. IgG4, IgA, and IgA2. Such antibodies and their uses are In one aspect, a method of detecting soluble STEAP-1 in disclosed in WO 2006/034488. blood or serum, the method comprising contacting a test Cysteine engineered antibodies of the invention preferably sample of blood or serum from a mammal Suspected of expe retain the antigenbinding capability of their wild type, parent riencing a prostate cell proliferative disorder with an anti antibody counterparts. Thus, cysteine engineered antibodies STEAP-1 antibody of the invention and detecting a increase are capable of binding, preferably specifically, to antigens. in soluble STEAP-1 in the test sample relative to a control Such antigens include, for example, tumor-associated anti sample of blood or serum from a normal mammal. In an gens (TAA), cell Surface receptor proteins and other cell embodiment, the method of detecting is useful as a method of 10 Surface molecules, transmembrane proteins, signalling pro diagnosing a prostate cell proliferative disorder associated teins, cell Survival regulatory factors, cell proliferation regu with an increase in soluble STEAP-1 in blood or serum of a latory factors, molecules associated with (for e.g., known or mammal. Suspected to contribute functionally to) tissue development or In one aspect, the antibodies of the invention include cys differentiation, lymphokines, cytokines, molecules involved teine engineered antibodies where one or more amino acids of 15 in cell cycle regulation, molecules involved in vasculogenesis a parent antibody are replaced with a free cysteine amino acid and molecules associated with (for e.g., known or Suspected as disclosed in WO2006/034488 (herein incorporated by ref to contribute functionally to) angiogenesis. erence in its entirety). A cysteine engineered antibody com An antibody of the invention may be conjugated to other prises one or more free cysteine amino acids having a thiol thiol-reactive agents in which the reactive group is, for reactivity value in the range of 0.6 to 1.0. A free cysteine example, a maleimide, an iodoacetamide, a pyridyl disulfide, amino acid is a cysteine residue which has been engineered or other thiol-reactive conjugation partner (Haugland, 2003, into the parent antibody and is not part of a disulfide bridge. Molecular Probes Handbook of Fluorescent Probes and Cysteine engineered antibodies are useful for attachment of Research Chemicals, Molecular Probes, Inc.; Brinkley, 1992, cytotoxic and/or imaging compounds at the site of the engi Bioconjugate Chem. 3:2; Garman, 1997, Non-Radioactive neered cysteine through, for example, a maleimide or halo 25 Labelling: A Practical Approach, Academic Press, London; acetyl. The nucleophilic reactivity of the thiol functionality of Means (1990) Bioconjugate Chem. 1:2: Hermanson, G. in a Cys residue to a maleimide group is about 1000 times higher Bioconjugate Techniques (1996) Academic Press, San Diego, compared to any other amino acid functionality in a protein, pp. 40-55, 643-671). The partner may be a cytotoxic agent Such as amino group of lysine residues or the N-terminal (e.g. a toxin Such as doxorubicin or pertussis toxin), a fluo amino group. Thiol specific functionality in iodoacetyl and 30 rophore Such as a fluorescent dye like fluorescein or maleimide reagents may react with amine groups, but higher rhodamine, a chelating agent for an imaging or radiothera pH (>9.0) and longer reaction times are required (Garman, peutic metal, a peptidyl or non-peptidyllabel or detection tag, 1997, Non-Radioactive Labelling: A Practical Approach, or a clearance-modifying agent such as various isomers of Academic Press, London). polyethylene glycol, a peptide that binds to a third compo Cysteine engineered antibodies may be useful in the treat 35 nent, or another carbohydrate or lipophilic agent. ment of cancer and include antibodies specific for cell Surface In one aspect, antibodies of the invention may be conju and transmembrane receptors, and tumor-associated antigens gated with any label moiety which can be covalently attached (TAA). Such antibodies may be used as naked antibodies to the antibody through a reactive moiety, an activated moiety, (unconjugated to a drug or label moiety) or as antibody-drug or a reactive cysteine thiol group (Singh et al (2002) Anal. conjugates (ADC). Cysteine engineered antibodies of the 40 Biochem. 304: 147-15; Harlow E. and Lane, D. (1999) Using invention may be site-specifically and efficiently coupled Antibodies: A Laboratory Manual, Cold Springs Harbor with a thiol-reactive reagent. The thiol-reactive reagent may Laboratory Press, Cold Spring Harbor, N.Y.; Lundblad R. L. be a multifunctional linker reagent, a capture label reagent, a (1991) Chemical Reagents for Protein Modification, 2nd ed. fluorophore reagent, or a drug-linker intermediate. The cys CRC Press, Boca Raton, Fla.). The attached label may func teine engineered antibody may be labeled with a detectable 45 tion to: (i) provide a detectable signal; (ii) interact with a label, immobilized on a solid phase Support and/or conju second label to modify the detectable signal provided by the gated with a drug moiety. Thiol reactivity may be generalized first or second label, e.g. to give FRET (fluorescence reso to any antibody where Substitution of amino acids with reac nance energy transfer); (iii) stabilize interactions or increase tive cysteine amino acids may be made within the ranges in affinity of binding, with antigen or ligand; (iv) affect mobility, the light chain selected from amino acid ranges: L-10 to L-20; 50 e.g. electrophoretic mobility or cell-permeability, by charge, L-38 to L-48; L-105 to L-115; L-139 to L-149; L-163 to hydrophobicity, shape, or other physical parameters, or (V) L-173; and within the ranges in the heavy chain selected from provide a capture moiety, to modulate ligand affinity, anti amino acid ranges: H-35 to H-45; H-83 to H-93; H-114 to body/antigen binding, or ionic complexation. H-127; and H-170 to H-184, and in the Fc region within the Labelled cysteine engineered antibodies may be useful in ranges selected from H-268 to H-291; H-319 to H-344; 55 diagnostic assays, e.g., for detecting expression of an antigen H-370 to H-380; and H-395 to H-405, where the numbering of interest in specific cells, tissues, or serum. For diagnostic of amino acid positions begins at position 1 of the Kabat applications, the antibody will typically be labeled with a numbering system (Kabat et al. (1991) Sequences of Proteins detectable moiety. Numerous labels are available which can of Immunological Interest, 5th Ed. Public Health Service, be generally grouped into the following categories: National Institutes of Health, Bethesda, Md.) and continues 60 Radioisotopes (radionuclides), such as H, C, C, F, sequentially thereafter as disclosed in WO 2006/034488. 32P 35S, (Cu, Ga, 86Y. 897r, 99Tc, 11 In, 123, 1241, 125I, 13 II, Thiol reactivity may also be generalized to certain domains of 'Xe, '77Lu, ''At, or 'Bi. Radioisotope labelled antibod an antibody, Such as the light chain constant domain (CL) and ies are useful in receptor targeted imaging experiments. The heavy chain constant domains, CH1, CH2 and CH3. Cysteine antibody can be labeled with ligand reagents that bind, che replacements resulting in thiol reactivity values of 0.6 and 65 late or otherwise complex a radioisotope metal where the higher may be made in the heavy chain constant domains C. reagent is reactive with the engineered cysteine thiol of the Ö, e, Y, and Lof intactantibodies: IgA, Ig), IgE, IgG, and IgM, antibody, using the techniques described in Current Protocols US 8,771,966 B2 9 10 in Immunology, Volumes 1 and 2, Coligen et al. Ed. Wiley enzyme may catalyze a color change in a Substrate, which can Interscience, New York, N.Y., Pubs. (1991). Chelating be measured spectrophotometrically. Alternatively, the ligands which may complex a metal ion include DOTA, enzyme may alter the fluorescence or chemiluminescence of DOTP, DOTMA, DTPA and TETA (Macrocyclics, Dallas, the Substrate. Techniques for quantifying a change in fluores Tex.). Radionuclides can be targetted via complexation with 5 cence are described above. The chemiluminescent substrate the antibody-drug conjugates of the invention (Wu et al becomes electronically excited by a chemical reaction and (2005) Nature Biotechnology 23(9): 1137-1146). may then emit light which can be measured (using a chemi Linker reagents such as DOTA-maleimide (4-maleimi luminometer, for example) or donates energy to a fluorescent dobutyramidobenzyl-DOTA) can be prepared by the reaction acceptor. Examples of enzymatic labels include luciferases of aminobenzyl-DOTA with 4-maleimidobutyric acid 10 (e.g., firefly luciferase and bacterial luciferase: U.S. Pat. No. (Fluka) activated with isopropylchloroformate (Aldrich), fol 4.737,456), luciferin, 2,3-dihydrophthalazinediones, malate lowing the procedure of Axworthy et al (2000) Proc. Natl. dehydrogenase, urease, peroxidase Such as horseradish per Acad. Sci. USA 97(4): 1802-1807). DOTA-maleimide oxidase (HRP), alkaline phosphatase (AP), B-galactosidase, reagents react with the free cysteine amino acids of the cys glucoamylase, lysozyme, Saccharide oxidases (e.g., glucose teine engineered antibodies and provide a metal complexing 15 oxidase, galactose oxidase, and glucose-6-phosphate dehy ligand on the antibody (Lewis et al (1998) Bioconj. Chem. drogenase), heterocyclic oxidases (such as uricase and Xan 9:72-86). Chelating linker labelling reagents such as DOTA thine oxidase), lactoperoxidase, microperoxidase, and the NHS (1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic like. Techniques for conjugating enzymes to antibodies are acid mono (N-hydroxysuccinimide ester) are commercially described in O'Sullivan etal (1981) “Methods for the Prepa available (Macrocyclics, Dallas, Tex.). Receptor target imag ration of Enzyme-Antibody Conjugates for use in Enzyme ing with radionuclide labelled antibodies can provide a Immunoassay, in Methods in Enzym. (ed J. Langone & H. marker of pathway activation by detection and quantitation of Van Vunakis), Academic Press, New York, 73:147-166. progressive accumulation of antibodies in tumor tissue (Al Examples of enzyme-substrate combinations include, for bert etal (1998) Bioorg. Med. Chem. Lett.8:1207-1210). The example: conjugated radio-metals may remain intracellular following 25 (i) Horseradish peroxidase (HRP) with hydrogen peroxi lysosomal degradation. dase as a Substrate, wherein the hydrogen peroxidase oxidizes Metal-chelate complexes suitable as antibody labels for a dye precursor (e.g., orthophenylene diamine (OPD) or 3.3', imaging experiments are disclosed: U.S. Pat. No. 5,342,606: 5,5'-tetramethylbenzidine hydrochloride (TMB)); U.S. Pat. No. 5,428, 155; U.S. Pat. No. 5,316,757; U.S. Pat. (ii) alkaline phosphatase (AP) with para-nitrophenylphos No. 5,480,990; U.S. Pat. No. 5,462,725; U.S. Pat. No. 5,428, 30 phate as chromogenic Substrate; and 139; U.S. Pat. No. 5,385,893; U.S. Pat. No. 5,739,294; U.S. (iii) B-D-galactosidase (B-D-Gal) with a chromogenic Sub Pat. No. 5,750,660; U.S. Pat. No. 5,834,456; Hnatowich etal strate (e.g., p-nitrophenyl-B-D-galactosidase) or fluorogenic (1983).J. Immunol. Methods 65:147-157: Meares etal (1984) substrate 4-methylumbelliferyl-3-D-galactosidase. Anal. Biochem. 142:68-78; Mirzadeh et al (1990) Bioconju Numerous other enzyme-substrate combinations are avail gate Chem. 1:59-65; Meares et al (1990) J. Cancer 1990, 35 able to those skilled in the art. For a general review, see U.S. Suppl. 10:21-26: Izard et al (1992) Bioconjugate Chem. Pat. No. 4,275,149 and U.S. Pat. No. 4,318,980. 3:346-350; Nikula et al (1995) Nucl. Med. Biol. 22:387-90; A label may be indirectly conjugated with an amino acid Camera et al (1993) Nucl. Med. Biol. 20:955-62: Kukis etal side chain, an activated amino acid side chain, a cysteine (1998).J. Nucl. Med. 39:2105-2110; Vereletal (2003).J. Nucl. engineered antibody, and the like. For example, the antibody Med. 44:1663-1670; Camera et al (1994) J. Nucl. Med. 40 can be conjugated with biotin and any of the three broad 21:640-646: Ruegget al (1990) Cancer Res. 50:4221-4226: categories of labels mentioned above can be conjugated with Verel et al (2003) J. Nucl. Med. 44:1663-1670; Lee et al avidin or streptavidin, or vice versa. Biotin binds selectively (2001) Cancer Res.61:4474-4482; Mitchell, et al (2003) J. to streptavidin and thus, the label can be conjugated with the Nucl. Med. 44:1105-11 12: Kobayashietal (1999) Bioconju antibody in this indirect manner. Alternatively, to achieve gate Chem. 10:103-111; Miederer et al (2004) J. Nucl. Med. 45 indirect conjugation of the label with the polypeptide variant, 45:129-137; DeNardo et al (1998) Clinical Cancer Research the polypeptide variant is conjugated with a small hapten 4:2483-90; Blend etal (2003) Cancer Biotherapy & Radiop (e.g., digoxin) and one of the different types of labels men harmaceuticals 18:355-363; Nikula et al.(1999).J. Nucl. Med. tioned above is conjugated with an anti-hapten polypeptide 40:166-76; Kobayashi et al (1998).J. Nucl. Med. 39:829-36: variant (e.g., anti-digoxin antibody). Thus, indirect conjuga Mardirossian etal (1993) Nucl. Med. Biol. 20:65-74; Roselli 50 tion of the label with the polypeptide variant can be achieved et al (1999) Cancer Biotherapy & Radiopharmaceuticals, (Hermanson, G. (1996) in Bioconjugate Techniques Aca 14:209-20. demic Press, San Diego). (b) Fluorescent labels such as rare earth chelates (europium The antibody of the present invention may be employed in chelates), fluorescein types including FITC, 5-carboxyfluo any known assay method, Such as ELISA, competitive bind rescein, 6-carboxy fluorescein, rhodamine types including 55 ing assays, direct and indirect Sandwich assays, and immu TAMRA; dansyl; Lissamine; cyanines; phycoerythrins; noprecipitation assays (Zola, (1987) Monoclonal Antibodies: Texas Red; and analogs thereof. The fluorescent labels can be A Manual of Techniques, pp. 147-158, CRC Press, Inc.). conjugated to antibodies using the techniques disclosed in A detection label may be useful for localizing, visualizing, Current Protocols in Immunology, Supra, for example. Fluo and quantitating a binding or recognition event. The labelled rescent dyes and fluorescent label reagents include those 60 antibodies of the invention can detect cell-surface receptors. which are commercially available from Invitrogen/Molecular Another use for detectably labelled antibodies is a method of Probes (Eugene, Oreg.) and Pierce Biotechnology, Inc. bead-based immunocapture comprising conjugating a bead (Rockford, Ill.). with a fluorescent labelled antibody and detecting a fluores (c) Various enzyme-substrate labels are available or dis cence signal upon binding of a ligand. Similar binding detec closed (U.S. Pat. No. 4,275,149). The enzyme generally cata 65 tion methodologies utilize the Surface plasmon resonance lyzes a chemical alteration of a chromogenic Substrate that (SPR) effect to measure and detect antibody-antigen interac can be measured using various techniques. For example, the tions. US 8,771,966 B2 11 12 Detection labels such as fluorescent dyes and chemilumi though the mechanism may not be associated with clinical nescent dyes (Briggs et al (1997) “Synthesis of Functiona outcome; Type II markers function as Surrogate endpoints lised Fluorescent Dyes and Their Coupling to Amines and where the change in, or signal from, the biomarker predicts a Amino Acids.” J. Chem. Soc., Perkin-Trans. 1:1051-1058) clinical benefit to “validate' the targeted response, such as provide a detectable signal and are generally applicable for measured bone erosion in rheumatoid arthritis by CT. Imag labelling antibodies, preferably with the following proper ing biomarkers thus can provide pharmacodynamic (PD) ties: (i) the labelled antibody should produce a very high therapeutic information about: (i) expression of a target pro signal with low background so that Small quantities of anti tein, (ii) binding of a therapeutic to the target protein, i.e. bodies can be sensitively detected in both cell-free and cell selectivity, and (iii) clearance and half-life pharmacokinetic based assays; and (ii) the labelled antibody should be photo 10 data. Advantages of in vivo imaging biomarkers relative to stable so that the fluorescent signal may be observed, lab-based biomarkers include: non-invasive treatment, quan monitored and recorded without significant photo bleaching. tifiable, whole body assessment, repetitive dosing and assess For applications involving cell surface binding of labelled ment, i.e. multiple time points, and potentially transferable antibody to membranes or cell Surfaces, especially live cells, effects from preclinical (Small animal) to clinical (human) the labels preferably (iii) have good water-solubility to 15 results. For Some applications, bioimaging Supplants or mini achieve effective conjugate concentration and detection sen mizes the number of animal experiments in preclinical stud sitivity and (iv) are nontoxic to living cells So as not to disrupt 1CS the normal metabolic processes of the cells or cause prema Peptide labelling methods are well known. See Haugland, ture cell death. 2003, Molecular Probes Handbook of Fluorescent Probes and Direct quantification of cellular fluorescence intensity and Research Chemicals, Molecular Probes, Inc.; Brinkley, 1992, enumeration of fluorescently labelled events, e.g. cell surface Bioconjugate Chem. 3:2; Garman, (1997) Non-Radioactive binding of peptide-dye conjugates may be conducted on an Labelling: A Practical Approach, Academic Press, London; system (FMATR 8100 HTS System, Applied Biosystems, Means (1990) Bioconjugate Chem. 1:2: Glazer et al (1975) Foster City, Calif.) that automates mix-and-read, non-radio Chemical Modification of Proteins. Laboratory Techniques in active assays with live cells or beads (Miraglia, “Homoge 25 Biochemistry and Molecular Biology (T. S. Work and E. neous cell- and bead-based assays for high throughput Work, Eds.) American Elsevier Publishing Co., New York; screening using fluorometric microVolume assay technol Lundblad, R. L. and Noyes, C.M. (1984) Chemical Reagents ogy’, (1999).J. of Biomolecular Screening 4:193-204). Uses for Protein Modification, Vols. I and II, CRC Press, New York; of labelled antibodies also include cell surface receptor bind Pfleiderer, G. (1985) “Chemical Modification of Proteins”, ing assays, immunocapture assays, fluorescence linked 30 Modern Methods in Protein Chemistry, H. Tschesche, Ed., immunosorbent assays (FLISA), caspase-cleavage (Zheng, Walter DeGryter, Berlin and New York; and Wong (1991) "Caspase-3 controls both cytoplasmic and nuclear events Chemistry of Protein Conjugation and Cross-linking, CRC associated with Fas-mediated apoptosis in Vivo”, (1998) Press, Boca Raton, Fla.); De Leon-Rodriguez et al (2004) Proc. Natl. Acad. Sci. USA 95:618-23; U.S. Pat. No. 6,372, Chem. Eur.J. 10:1149-1155; Lewis etal (2001) Bioconjugate 907), apoptosis (Vermes, “A novel assay for apoptosis. Flow 35 Chem. 12:320-324; Li et al (2002) Bioconjugate Chem. cytometric detection of phosphatidylserine expression on 13:110-115; Mier et al (2005) Bioconjugate Chem. 16:240 early apoptotic cells using fluorescein labelled Annexin V” 237. (1995) J. Immunol. Methods 184:39-51) and cytotoxicity Peptides and proteins labelled with two moieties, a fluo assays. Fluorometric microVolume assay technology can be rescent reporter and quencher in Sufficient proximity undergo used to identify the up or down regulation by a molecule that 40 fluorescence resonance energy transfer (FRET). Reporter is targeted to the cell Surface (Swartzman, 'A homogeneous groups are typically fluorescent dyes that are excited by light and multiplexed immunoassay for high-throughput Screening at a certain wavelength and transfer energy to an acceptor, or using fluorometric microVolume assay technology’, (1999) quencher, group, with the appropriate Stokes shift for emis Anal. Biochem. 271:143-51). sion at maximal brightness. Fluorescent dyes include mol Labelled antibodies of the invention are useful as imaging 45 ecules with extended aromaticity, Such as fluorescein and biomarkers and probes by the various methods and tech rhodamine, and their derivatives. The fluorescent reporter niques of biomedical and molecular imaging Such as: (i) MRI may be partially or significantly quenched by the quencher (magnetic resonance imaging); (ii) MicroCT (computerized moiety in an intact peptide. Upon cleavage of the peptide by tomography); (iii) SPECT (single photon emission computed a peptidase or protease, a detectable increase in fluorescence tomography); (iv) positron emission topography (PET) or 50 may be measured (Knight, C. (1995) “Fluorimetric Assays of Immuno-positron emission tomography (Immuno-PET) (see Proteolytic Enzymes’, Methods in Enzymology, Academic Van Dongen GA, et al "Immuno-PET: a navigator in mono Press, 248:18-34). clonal antibody development and applications' Oncologist The labelled antibodies of the invention may also be used 2007: 12:1379-89. (v) bioluminescence; (vi) fluorescence: as an affinity purification agent. In this process, the labelled and (vii) ultrasound. Immunoscintigraphy is an imaging pro 55 antibody is immobilized on a Solid phase such a Sephadex cedure in which antibodies labeled with radioactive sub resin or filter paper, using methods well known in the art. The stances are administered to an animal or human patient and a immobilized antibody is contacted with a sample containing picture is taken of sites in the body where the antibody local the antigen to be purified, and thereafter the support is washed izes (U.S. Pat. No. 6,528,624). Imaging biomarkers may be with a suitable solvent that will remove substantially all the objectively measured and evaluated as an indicator of normal 60 material in the sample except the antigento be purified, which biological processes, pathogenic processes, orpharmacologi is bound to the immobilized polypeptide variant. Finally, the cal responses to a therapeutic intervention. Biomarkers may Support is washed with another Suitable solvent, such as gly be of several types: Type 0 are natural history markers of a cine buffer, pH 5.0, that will release the antigen from the disease and correlate longitudinally with known clinical indi polypeptide variant. ces, e.g. MRI assessment of synovial inflammation in rheu 65 Labelling reagents typically bear reactive functionality matoid arthritis; Type I markers capture the effect of an inter which may react (i) directly with a cysteine thiol of a cysteine vention in accordance with a mechanism-of-action, even engineered antibody to form the labelled antibody, (ii) with a US 8,771,966 B2 13 14 linker reagent to form a linker-label intermediate, or (iii) with Exemplary albumin binding peptide sequences include, a linker antibody to form the labelled antibody. Reactive but are not limited to, the amino acid sequences listed in SEQ functionality of labelling reagents include: maleimide, halo ID NOS:80-84. acetyl, iodoacetamide succinimidyl ester (e.g. NHS, N-hy droxysuccinimide), isothiocyanate, Sulfonyl chloride, 2.6- dichlorotriazinyl, pentafluorophenyl ester, and SEO ID NO: 8O phosphoramidite, although other functional groups can also CDKTHTGGGSORLMEDICLPRWGCLWEDDF be used. SEO ID NO: 81 An exemplary reactive functional group is N-hydroxysuc ORLMEDICLPRWGCLWEDDF 10 cinimidyl ester (NHS) of a carboxyl group substituent of a SEO ID NO: 82 detectable label, e.g. biotin or a fluorescent dye. The NHS ORLIEDICLPRWGCLWEDDF ester of the label may be preformed, isolated, purified, and/or SEO ID NO: 83 characterized, or it may be formed in situ and reacted with a RLIEDICLPRWGCLWEDD nucleophilic group of an antibody. Typically, the carboxyl 15 form of the label is activated by reacting with some combi SEO ID NO: 84 nation of a carbodiimide reagent, e.g. dicyclohexylcarbodi DICLPRWGCLW. imide, diisopropylcarbodiimide, or a uronium reagent, e.g. Antibody-Drug Conjugates TSTU (O-(N-Succinimidyl)-N,N,N',N'-tetramethyluro In another aspect, the invention provides immunoconju nium tetrafluoroborate, HBTU (O-benzotriazol-1-yl)-N.N. gates, or antibody-drug conjugates (ADC), comprising an N',N'-tetramethyluronium hexafluorophosphate), or HATU antibody conjugated to a cytotoxic agent such as a chemo (O-(7-azabenzotriazol-1-yl)-N.N.N',N'-tetramethyluronium therapeutic agent, a drug, a growth inhibitory agent, a toxin hexafluorophosphate), an activator, Such as 1-hydroxybenzo (e.g., an enzymatically active toxin of bacterial, fungal, plant, triazole (HOBt), and N-hydroxysuccinimide to give the NHS or animal origin, or fragments thereof), or a radioactive iso ester of the label. In some cases, the label and the antibody 25 tope (i.e., a radioconjugate). In another aspect, the invention may be coupled by in situ activation of the label and reaction further provides methods of using the immunoconjugates. In with the antibody to form the label-antibody conjugate in one one aspect, an immunoconjugate comprises any of the above step. Other activating and coupling reagents include TBTU anti-STEAP-1 antibodies covalently attached to a cytotoxic (2-(1H-benzotriazo-1-yl)-1-1,3,3-tetramethyluronium agent or a detectable agent. hexafluorophosphate), TFFH (N,N',N',N'-tetramethyluro 30 The use of antibody-drug conjugates for the local delivery nium 2-fluoro-hexafluorophosphate), PyBOP (benzotriaz of cytotoxic or cytostatic agents, i.e. drugs to kill or inhibit ole-1-yl-oxy-tris-pyrrolidino-phosphonium hexafluorophos tumor cells in the treatment of cancer (Syrigos and Epenetos (1999) Anticancer Research 19:605-614; Niculescu-Duvaz phate, EEDQ (2-ethoxy-1-ethoxycarbonyl-1,2-dihydro and Springer (1997) Adv. Drg Del. Rev. 26:151-172; U.S. Pat. quinoline), DCC (dicyclohexylcarbodiimide); DIPCDI 35 No. 4,975.278) allows targeted delivery of the drug moiety to (diisopropylcarbodiimide), MSNT (1-(mesitylene-2-sulfo tumors, and intracellular accumulation therein, where sys nyl)-3-nitro-1H-1,2,4-triazole, and aryl Sulfonylhalides, e.g. temic administration of these unconjugated drug agents may triisopropylbenzenesulfonyl chloride. result in unacceptable levels of toxicity to normal cells as well Albumin Binding Peptide-Fab Compounds of the Inven as the tumor cells sought to be eliminated (Baldwin et al., tion: 40 (1986) Lancet pp. (Mar. 15, 1986):603-05; Thorpe, (1985) In one aspect, the antibody of the invention is fused to an Antibody Carriers Of Cytotoxic Agents. In Cancer Therapy: albumin binding protein. Plasma-protein binding can be an A Review.” in Monoclonal Antibodies 84: Biological And effective means of improving the pharmacokinetic properties Clinical Applications, A. Pinchera et al. (ed.s), pp. 475-506). of short lived molecules. Albumin is the most abundant pro Maximal efficacy with minimal toxicity is sought thereby. tein in plasma. Serum albumin binding peptides (ABP) can 45 Both polyclonal antibodies and monoclonal antibodies have alter the pharmacodynamics of fused active domain proteins, been reported as useful in these strategies (Rowland et al., including alteration of tissue uptake, penetration, and diffu (1986) Cancer Immunol. Immunother. 21:183-87). Drugs Sion. These pharmacodynamic parameters can be modulated used in these methods include daunomycin, doxorubicin, by specific selection of the appropriate serum albumin bind methotrexate, and vindesine (Rowland et al., (1986) supra). ing peptide sequence (US 2004.0001827). A series of albumin 50 Toxins used in antibody-toxin conjugates include bacterial binding peptides were identified by phage display Screening toxins such as diphtheria toxin, plant toxins such as ricin, (Dennis et al. (2002) “Albumin Binding. As A General Strat Small molecule toxins such as geldanamycin (Mandler et al egy For Improving The Pharmacokinetics Of Proteins' J Biol (2000) Jour of the Nat. Cancer Inst. 92(19): 1573-1581; Man Chem. 277:35035-35043; WO 01/45746). Compounds of the dler et al (2000) Bioorganic & Med. Chem. Letters 10:1025 invention include ABP sequences taught by: (i) Dennis et al 55 1028; Mandleretal (2002) Bioconjugate Chem. 13:786-791), (2002) J Biol Chem. 277:35035-35043 at Tables III and IV, maytansinoids (EP 1391213: Liu et al., (1996) Proc. Natl. page 35038; (ii) US 2004.0001827 at 0076 SEQ ID NOS: Acad. Sci. USA 93:8618-8623), and calicheamicin (Lode et 9-22; and (iii) WO 01/45746 at pages 12-13, all of which are all (1998) Cancer Res. 58:2928; Hinman et al (1993) Cancer incorporated herein by reference. Albumin Binding (ABP)- Res. 53:3336-3342). The toxins may effect their cytotoxic Fabs are engineered by fusing an albumin binding peptide to, 60 and cytostatic effects by mechanisms including tubulin bind for example, the C-terminus of Fab heavy chain in 1:1 sto ing, DNA binding, or topoisomerase inhibition. Some cyto ichiometric ratio (1 ABP/1 Fab). It was shown that association toxic drugs tend to be inactive or less active when conjugated of these ABP-Fabs with albumin increased antibody half life to large antibodies or protein receptor ligands. by more than 25 fold in rabbits and mice. The above described ZEVALINR) (ibritumomab tiuxetan, Biogen/Idec) is an reactive Cys residues can therefore be introduced in these 65 antibody-radioisotope conjugate composed of a murine IgG1 ABP-Fabs and used for site-specific conjugation with cyto kappa monoclonal antibody directed against the CD20 anti toxic drugs followed by in vivo animal studies. gen found on the Surface of normal and malignant B lympho US 8,771,966 B2 15 16 cytes and '''In or 'Y radioisotope bound by a thiourea auristatins, a trichothecene, and CC1065, and the derivatives linker-chelator (Wiseman et al (2000) Eur, Jour. Nucl. Med. of these toxins that have toxin activity, are also contemplated 27(7):766-77: Wiseman et al (2002) Blood 99(12):4336-42: herein. Witzigetal (2002).J. Clin. Oncol. 20(10):2453-63; Witziget al (2002) J. Clin. Oncol. 20(15):3262-69). Although ZEVA Maytansine and Maytansinoids LIN has activity against B-cell non-Hodgkin’s Lymphoma In Some embodiments, the immunoconjugate comprises an (NHL), administration results in severe and prolonged antibody (full length or fragments) of the invention conju cytopenias in most patients. MYLOTARGTM (gemtuzumab gated to one or more maytansinoid molecules. oZogamicin, Wyeth Pharmaceuticals), an antibody drug con jugate composed ofahu CD33 antibody linked to calicheami 10 Maytansinoids are mitototic inhibitors which act by inhib cin, was approved in 2000 for the treatment of acute myeloid iting tubulin polymerization. Maytansine was first isolated leukemia by injection (Drugs of the Future (2000) 25(7):686; from the east African shrub Maytenus serrata (U.S. Pat. No. U.S. Pat. Nos. 4,970, 198; 5,079,233; 5,585,089; 5,606,040: 3.896,111). Subsequently, it was discovered that certain 5,693,762; 5,739,116; 5,767.285; 5,773,001). Cantuzumab 15 microbes also produce maytansinoids, such as maytansinol mertansine (Immunogen, Inc.), an antibody drug conjugate and C-3 maytansinol esters (U.S. Pat. No. 4,151,042). Syn composed of the huC242 antibody linked via the disulfide thetic maytansinol and derivatives and analogues thereof are linker SPP to the maytansinoid drug moiety, DM1, is advanc disclosed, for example, in U.S. Pat. Nos. 4,137,230; 4.248, ing into Phase II trials for the treatment of cancers that express 870; 4,256,746; 4,260,608; 4,265,814; 4,294,757; 4,307,016: CanAg, such as colon, pancreatic, gastric, and others. MLN 2704 (Millennium Pharm. BZL Biologics, Immunogen Inc.), 4,308,268; 4,308,269; 4,309,428; 4.313,946; 4,315,929: an antibody drug conjugate composed of the anti-prostate 4,317,821; 4,322,348; 4,331,598; 4.361,650; 4,364,866; specific membrane antigen (PSMA) monoclonal antibody 4,424,219; 4,450,254; 4,362,663; and 4,371,533. linked to the maytansinoid drug moiety, DM1, is under devel Maytansinoid drug moieties are attractive drug moieties in opment for the potential treatment of prostate tumors. The 25 antibody drug conjugates because they are: (i) relatively auristatin peptides, auristatin E (AE) and monomethylau accessible to prepare by fermentation or chemical modifica ristatin (MMAE), synthetic analogs of dolastatin, were con jugated to chimeric monoclonal antibodies cBR96 (specific tion, derivatization offermentation products, (ii) amenable to to Lewis Y on carcinomas) and cAC10 (specific to CD30 on derivatization with functional groups suitable for conjugation hematological malignancies) (Doronina et al (2003) Nature 30 through the non-disulfide linkers to antibodies, (iii) stable in Biotechnology 21 (7):778-784) and are under therapeutic plasma, and (iv) effective against a variety of tumor cell lines. development. Maytansine compounds suitable for use as maytansinoid Chemotherapeutic agents useful in the generation of drug moieties are well known in the art, and can be isolated immunoconjugates are described herein. Enzymatically from natural Sources according to known methods, produced active toxins and fragments thereof that can be used include 35 using genetic engineering techniques (see Yu et al (2002) diphtheria Achain, nonbinding active fragments of diphtheria PNAS 99:7968-7973), or maytansinol and maytansinol ana toxin, exotoxin A chain (from Pseudomonas aeruginosa), logues prepared synthetically according to known methods. ricin. A chain, abrin A chain, modeccin A chain, alpha-sarcin, Aleurites fordii proteins, dianthin proteins, Phytolaca ameri Exemplary maytansinoid drug moieties include those hav 40 ing a modified aromatic ring, such as: C-19-dechloro (U.S. cana proteins (PAPI, PAPII, and PAP-S), momordica charan Pat. No. 4,256,746) (prepared by lithium aluminum hydride tia inhibitor, curcin, crotin, Sapaonaria officinalis inhibitor, reduction of ansamytocin P2); C-20-hydroxy (or C-20-dem gelonin, mitogellin, restrictocin, phenomycin, enomycin, and ethyl)+/-C-19-dechloro (U.S. Pat. Nos. 4.361,650 and 4.307, the tricothecenes. See, e.g., WO 93/21232 published Oct. 28, 016) (prepared by demethylation using Streptomyces or Acti 1993. A variety of radionuclides are available for the produc 45 tion of radioconjugated antibodies. Examples include 'Bi, nomyces or dechlorination using LAH); and C-20 'I, In, 'Y, and Re. Conjugates of the antibody and demethoxy, C-20-acyloxy (—OCOR), +/-dechloro (U.S. cytotoxic agent are made using a variety of bifunctional pro Pat. No. 4,294.757) (prepared by acylation using acyl chlo tein-coupling agents such as N-Succinimidyl-3-(2-py rides). and those having modifications at other positions ridyldithiol) propionate (SPDP), iminothiolane (IT), bifunc 50 Exemplary maytansinoid drug moieties also include those tional derivatives of imidoesters (such as dimethyl having modifications such as: C-9-SH (U.S. Pat. No. 4,424, adipimidate HCl), active esters (such as disuccinimidyl Sub 219) (prepared by the reaction of maytansinol with HS or erate), aldehydes (such as glutaraldehyde), bis-azido com PSs); C-14-alkoxymethyl(demethoxy/CHOR) (U.S. Pat. pounds (such as bis(p-azidobenzoyl) hexanediamine), bis 55 No. 4.331,598): C-14-hydroxymethyl or acyloxymethyl diazonium derivatives (such as bis-(p-diazoniumbenzoyl)- (CH-OH or CHOAc) (U.S. Pat. No. 4,450.254) (prepared ethylenediamine), diisocyanates (such as toluene 2,6- from Nocardia); C-15-hydroxy/acyloxy (U.S. Pat. No. 4.364, diisocyanate), and bis-active fluorine compounds (such as 866) (prepared by the conversion of maytansinol by Strepto 1,5-difluoro-2,4-dinitrobenzene). For example, a ricin immu myces); C-15-methoxy (U.S. Pat. Nos. 4.313,946 and 4,315, notoxin can be prepared as described in Vitetta et al (1987) 60 929) (isolated from Trewia nudiflora); C-18-N-demethyl Science, 238:1098. Carbon-14-labeled 1-isothiocyanatoben (U.S. Pat. Nos. 4.362,663 and 4,322,348) (prepared by the Zyl-3-methyldiethylene triaminepentaacetic acid (MX demethylation of maytansinol by Streptomyces); and 4.5- DTPA) is an exemplary chelating agent for conjugation of deoxy (U.S. Pat. No. 4,371.533) (prepared by the titanium radionucleotide to the antibody (WO94/11026). 65 trichloride/LAH reduction of maytansinol). Conjugates of an antibody and one or more Small molecule Exemplary embodiments of maytansinoid drug moieties toxins, such as a calicheamicin, maytansinoids, dolastatins, include: DM1; DM3; and DM4, having the structures: US 8,771,966 B2 17 18 -continued DM1 DM4

10

CHO CHO

15

DM3 2O

25 wherein the wavy line indicates the covalent attachment of the sulfur atom of the drug to a linker (L) of an antibody drug conjugate. HERCEPTINR) (trastuzumab) linked by SMCC to DM1 has been reported (WO 2005/037992, which is CHO 30 expressly incorporated herein by reference in its entirety). An antibody drug conjugate of the present invention may be prepared according to the procedures disclosed therein. Other exemplary maytansinoid antibody drug conjugates 35 have the following structures and abbreviations, (wherein Ab is antibody and p is 1 to about 8):

CHO

Ab-SPP-DM1 US 8,771,966 B2 19 20 -continued

CHO

Ab-SMCC-DM1

Exemplary antibody drug conjugates where DM1 is linked so directed against human colorectal cancer. The conjugate was through a BMPEO linker to a thiol group of the antibody have found to be highly cytotoxic towards cultured colon cancer the structure and abbreviation: cells, and showed antitumor activity in an in vivo tumor

CHO

where Ab is antibody; n is 0, 1, or 2; and p is 1, 2, 3, or 4. growth assay. Chari et al., Cancer Research 52:127-131 Immunoconjugates containing maytansinoids, methods of go (1992) describe immunoconjugates in which a maytansinoid making same, and their therapeutic use are disclosed, for was conjugated via a disulfide linker to the murine antibody example, in U.S. Pat. Nos. 5,208,020; 5.416,064; 6,441,163 A7 binding to an antigen on human colon cancer cell lines, or and European Patent EP 0 425 235 B1, the disclosures of to another murine monoclonal antibody TA.1 that binds the which are hereby expressly incorporated by reference. Liu et HER-2/neu oncogene. The cytotoxicity of the TA.1-maytan al., Proc. Natl. Acad. Sci. USA 93:8618-8623 (1996) 65 sonoid conjugate was tested in vitro on the human breast described immunoconjugates comprising a maytansinoid cancer cell line SK-BR-3, which expresses 3x10 HER-2 designated DM1 linked to the monoclonal antibody C242 Surface antigens per cell. The drug conjugate achieved a US 8,771,966 B2 21 22 degree of cytotoxicity similar to the free maytansinoid drug, compounds (such as 1,5-difluoro-2,4-dinitrobenzene). Par which could be increased by increasing the number of may ticularly preferred coupling agents include N-Succinimidyl tansinoid molecules per antibody molecule. The A7-may 3-(2-pyridyldithio)propionate (SPDP) (Carlsson et al., Bio tansinoid conjugate showed low systemic cytotoxicity in chem. J. 173:723-737 (1978)) and N-succinimidyl-4-(2- mice. 5 pyridylthio)pentanoate (SPP) to provide for a disulfide Anti-STEAP-1 antibody-maytansinoid conjugates are pre linkage. pared by chemically linking an antibody to a maytansinoid The linker may be attached to the maytansinoid molecule at molecule without significantly diminishing the biological various positions, depending on the type of the link. For activity of either the antibody or the maytansinoid molecule. example, an ester linkage may be formed by reaction with a See, e.g., U.S. Pat. No. 5,208,020 (the disclosure of which is hydroxyl group using conventional coupling techniques. The hereby expressly incorporated by reference). An average of reaction may occur at the C-3 position having a hydroxyl 3-4 maytansinoid molecules conjugated per antibody mol group, the C-14 position modified with hydroxymethyl, the ecule has shown efficacy in enhancing cytotoxicity of target C-15 position modified with a hydroxyl group, and the C-20 cells without negatively affecting the function or solubility of 15 position having a hydroxyl group. In a preferred embodiment, the antibody, although even one molecule of toxin/antibody the linkage is formed at the C-3 position of maytansinol or a would be expected to enhance cytotoxicity over the use of maytansinol analogue. naked antibody. Maytansinoids are well known in the art and In one embodiment, any of the antibodies of the invention can be synthesized by known techniques or isolated from 20 (full length or fragment) is conjugated to one or more may natural sources. Suitable maytansinoids are disclosed, for tansinoid molecules. In one embodiment of the immunocon example, in U.S. Pat. No. 5,208,020 and in the other patents jugate, the cytotoxic agent D, is a maytansinoid DM1. In one and nonpatent publications referred to hereinabove. Preferred embodiment of the immunoconjugate, the linker is selected maytansinoids are maytansinol and maytansinol analogues from the group consisting of SPDP, SMCC, IT, SPDP and modified in the aromatic ring or at other positions of the 2 SPP maytansinol molecule, such as various maytansinol esters. Auristatins and Dolostatins There are many linking groups known in the art for making In Some embodiments, the immunoconjugate comprises an antibody-maytansinoid conjugates, including, for example, antibody of the invention conjugated to dolastatins or dolosta those disclosed in U.S. Pat. Nos. 5,208,020, 6,441,163, or EP so tin peptidic analogs and derivatives, the auristatins (U.S. Pat. Patent 0425 235 B1, Chari et al., Cancer Research 52:127 Nos. 5,635,483; 5,780,588). Dolastatins and auristatins have 131 (1992), and US 2005/0169933 A1, the disclosures of been shown to interfere with microtubule dynamics, GTP which are hereby expressly incorporated by reference. Anti hydrolysis, and nuclear and cellular division (Woyke et al body-maytansinoid conjugates comprising the linker compo (2001) Antimicrob. Agents and Chemother. 45(12):3580 nent SMCC may be prepared as disclosed in U.S. patent 35 3584) and have anticancer (U.S. Pat. No. 5,663,149) and application Ser. No. 1 1/141,344, filed 31 May 2005, “Anti antifungal activity (Pettit et al (1998) Antimicrob. Agents body Drug Conjugates and Methods’. The linking groups Chemother. 42:2961-2965). The dolastatin or auristatin drug include disulfide groups, thioether groups, acid labile groups, moiety may be attached to the antibody through the N (amino) photolabile groups, peptidase labile groups, or esterase labile 40 terminus or the C (carboxyl) terminus of the peptidic drug groups, as disclosed in the above-identified patents. Addi moiety (WO 02/088172). tional linking groups are described and exemplified herein. Exemplary auristatin embodiments include the N-terminus Conjugates of the antibody and maytansinoid may be made linked monomethylauristatin drug moieties DE and DF, dis using a variety of bifunctional protein coupling agents such as closed in “Senteretal, Proceedings of the American Associa N-succinimidyl-3-(2-pyridyldithio)propionate (SPDP), suc 45 tion for Cancer Research, Volume 45, Abstract Number 623, cinimidyl-4-(N-maleimidomethyl)cyclohexane-1-carboxy presented Mar. 28, 2004, the disclosure of which is expressly late (SMCC), iminothiolane (IT), bifunctional derivatives of incorporated by reference in its entirety. imidoesters (such as dimethyl adipimidate HCl), active esters An exemplary auristatin embodiment is MMAE (wherein (such as disuccinimidyl Suberate), aldehydes (such as glut the wavy line indicates the covalent attachment to a linker (L) araldehyde), bis-azido compounds (such as bis(p-azidoben of an antibody drug conjugate).

O OH H H N, N N

O ON O On O MMAE

Zoyl) hexanediamine), bis-diazonium derivatives (such as as Another exemplary auristatin embodiment is MMAF, bis-(p-diazoniumbenzoyl)-ethylenediamine), diisocyanates wherein the wavy line indicates the covalent attachment to a (such as toluene 2,6-diisocyanate), and bis-active fluorine linker (L) of an antibody drug conjugate (US 2005/0238649): US 8,771,966 B2 23 24

O H H N, N N

O O O O O /r N N O ?oOH MMAF

Additional exemplary embodiments comprising MMAE herein) have the following structures and abbreviations or MMAF and various linker components (described further (wherein Ab means antibody and p is 1 to about 8):

O O Ab-S O l No,H H N---- OrrrrrrrlO O. O. Wal-Cit-NH n O. n O. O1 COOH p O Ab-MC-vic-PAB-MMAF

O O Ab-S O l H. "

N-N-- O Wal-Cit-?if- N Or O . N O. n O N On O 'O O p Ab-MC-vic-PAB-MMAE

Ab-S O O O H OH N N H N N N

O O O O N O O N

Ab-MC-MMAE

Ab-S O O O H N N, H N N N

O O O O N O O N O OH

Ab-MC-MMAF US 8,771,966 B2 25 26 Typically, peptide-based drug moieties can be prepared by imaging (also known as magnetic resonance imaging, mri). forming a peptide bond between two or more amino acids Such as iodine-123 again, iodine-131, indium-111, fluorine and/or peptide fragments. Such peptide bonds can be pre 19, carbon-13, nitrogen-15, oxygen-17, gadolinium, manga pared, for example, according to the liquid phase synthesis nese or iron. method (see E. Schröder and K. Lübke, “The Peptides”, Vol 5 The radio- or other labels may be incorporated in the con ume 1, pp 76-136, 1965, Academic Press) that is well known jugate in known ways. For example, the peptide may be in the field of peptide chemistry. The auristatin/dolastatin biosynthesized or may be synthesized by chemical amino drug moieties may be prepared according to the methods of acid synthesis using Suitable amino acid precursors involv U.S. Pat. No. 5,635,483: U.S. Pat. No. 5,780,588: Pettit etal 10 ing, for example, fluorine-19 in place of hydrogen. Labels (1989).J. Am. Chem. Soc. 111:5463-5465; Pettit etal (1998) such as tc" or I'', Re, Re and In can be attached Anti-Cancer Drug Design 13:243-277; Pettit, G. R., et al. via a cysteine residue in the peptide. Yttrium-90 can be Synthesis, 1996, 719-725; Pettit et al (1996) J. Chem. Soc. attached via alysine residue. The IODOGEN method (Fraker Perkin Trans. 15:859-863; and Doronina (2003) Nat Biotech etal (1978) Biochem. Biophys. Res. Commun. 80:49-57 can no121(7):778-784. 15 be used to incorporate iodine-123. "Monoclonal Antibodies Calicheamicin in Immunoscintigraphy” (Chatal, CRC Press 1989) describes In other embodiments, the immunoconjugate comprises an other methods in detail. antibody of the invention conjugated to one or more cali Conjugates of the antibody and cytotoxic agent may be cheamicin molecules. The calicheamicinfamily of antibiotics made using a variety of bifunctional protein coupling agents are capable of producing double-stranded DNA breaks at Such as N-Succinimidyl-3-(2-pyridyldithio)propionate Sub-picomolar concentrations. For the preparation of conju (SPDP), succinimidyl-4-(N-maleimidomethyl)cyclohexane gates of the calicheamicin family, see U.S. Pat. Nos. 5,712, 1-carboxylate (SMCC), iminothiolane (IT), bifunctional 374, 5,714,586, 5,739,116, 5,767,285, 5,770,701, 5,770,710, derivatives of imidoesters (such as dimethyl adipimidate 5,773,001, 5,877.296 (all to American Cyanamid Company). 25 HCl), active esters (such as disuccinimidyl suberate), alde Structural analogues of calicheamicin which may be used hydes (such as glutaraldehyde), bis-azido compounds (such include, but are not limited to, Yi', C.", Cls', N-acetyl-Y', as bis(p-azidobenzoyl) hexanediamine), bis-diazonium PSAG and 0 (Hinman et al., Cancer Research 53:3336-3342 derivatives (such as bis-(p-diazoniumbenzoyl)-ethylenedi (1993), Lode et al., Cancer Research 58:2925-2928 (1998) amine), diisocyanates (such as toluene 2,6-diisocyanate), and and the aforementioned U.S. patents to American Cyanamid). 30 bis-active fluorine compounds (such as 1.5-difluoro-2,4-dini Another anti-tumor drug that the antibody can be conjugated trobenzene). For example, a ricin immunotoxin can be pre is QFA which is an antifolate. Both calicheamicin and QFA pared as described in Vitetta et al., Science 238:1098 (1987). have intracellular sites of action and do not readily cross the Carbon-14-labeled 1-isothiocyanatobenzyl-3-methyldieth plasma membrane. Therefore, cellular uptake of these agents ylene triaminepentaacetic acid (MX-DTPA) is an exemplary through antibody mediated internalization greatly enhances 35 chelating agent for conjugation of radionucleotide to the anti their cytotoxic effects. body. See WO94/11026. The linker may be a “cleavable Other Cytotoxic Agents linker facilitating release of the cytotoxic drug in the cell. Other antitumor agents that can be conjugated to the anti For example, an acid-labile linker, peptidase-sensitive linker, bodies of the invention include BCNU, streptozoicin, Vinc 40 photolabile linker, dimethyl linker or disulfide-containing ristine and 5-fluorouracil, the family of agents known collec linker (Charietal., Cancer Research 52:127-131 (1992); U.S. tively LL-E33288 complex described in U.S. Pat. Nos. 5,053, Pat. No. 5,208,020) may be used. 394, 5,770,710, as well as esperamicins (U.S. Pat. No. 5,877, The compounds of the invention expressly contemplate, 296). but are not limited to, ADC prepared with cross-linker Enzymatically active toxins and fragments thereof which 45 reagents: BMPS, EMCS, GMBS, HBVS, LC-SMCC, MBS, can be used include diphtheria A chain, nonbinding active MPBH, SBAP, SIA, SIAB, SMCC, SMPB, SMPH, sulfo fragments of diphtheria toxin, exotoxin A chain (from EMCS, sulfo-GMBS, sulfo-KMUS, sulfo-MBS, Sulfo Pseudomonas aeruginosa), ricin A chain, abrin A chain, SIAB, sulfo-SMCC, and sulfo-SMPB, and SVSB (succinim modeccin. A chain, alpha-sarcin, Aleurites fordii proteins, idyl-(4-vinylsulfone)benzoate) which are commercially dianthin proteins, Phytolaca americana proteins (PAPI, 50 available (e.g., from Pierce Biotechnology, Inc., Rockford, PAPII, and PAP-S), momordica charantia inhibitor, curcin, Ill., U.S.A). See pages 467-498, 2003-2004 Applications crotin, Sapaonaria officinalis inhibitor, gelonin, mitogellin, Handbook and Catalog. restrictocin, phenomycin, enomycin and the tricothecenes. Preparation of Antibody Drug Conjugates: See, for example, WO 93/21232 published Oct. 28, 1993. In the antibody drug conjugates (ADC) of the invention, an The present invention further contemplates an immuno 55 antibody (Ab) is conjugated to one or more drug moieties (D), conjugate formed between an antibody and a compound with e.g. about 1 to about 20 drug moieties per antibody, through a nucleolytic activity (e.g., a ribonuclease or a DNA endonu linker (L). In one embodiment, the number of drug moieties clease Such as a deoxyribonuclease; DNase). (D) per antibody is from about 1 to about 5, alternatively, from For selective destruction of the tumor, the antibody may 60 about 2 to about 6, alternatively, from about 2 to about 5, comprise a highly radioactive atom. A variety of radioactive alternatively from about 3 to about 4 drug moieties per anti isotopes are available for the production of radioconjugated body. Because the number of drug moieties per antibody is antibodies. Examples include At?'', I'', I'', Y', Re", typically an average number over all conjugates in a popula Re'. Sm', Bi', P, Pb and radioactive isotopes of tion of an antibody drug conjugate, the number of drug moi Lu. When the conjugate is used for detection, it may comprise 65 eties per antibody may not be a whole number. The ADC of a radioactive atom for scintigraphic studies, for exampletc'" Formula I may be prepared by several routes, employing or I'', or a spin label for nuclear magnetic resonance (NMR) organic chemistry reactions, conditions, and reagents known US 8,771,966 B2 27 28 to those skilled in the art, including: (1) reaction of a nucleo particular enzymes, for example, a tumor-associated pro philic group of an antibody with a bivalent linker reagent, to tease, cathepsin B, C and D, or a plasmin protease. form Ab-L, via a covalent bond, followed by reaction with a Exemplary linker component structures are shown below drug moiety D; and (2) reaction of a nucleophilic group of a (wherein the wavy line indicates sites of covalent attachment drug moiety with a bivalent linker reagent, to form D-L via a 5 to other components of the ADC): covalent bond, followed by reaction with the nucleophilic group of an antibody. Additional methods for preparing ADC are described herein. O

Ab-(L-D), Formula I 10 The linker may be composed of one or more linker com ponents. Exemplary linker components include 6-maleimi or-aO docaproyl (“MC), maleimidopropanoyl (“MP), valine-cit O rulline (“val-cit”), alanine-phenylalanine (“ala-phe'), p-aminobenzyloxycarbonyl (PAB), N-Succinimidyl 4-(2- 15 MC pyridylthio)pentanoate (“SPP), N-Succinimidyl 4-(N-male O O imidomethyl)cyclohexane-1 carboxylate (“SMCC), and N-Succinimidyl (4-iodo-acetyl)aminobenzoate (“SIAB). In on embodiment, the linker is valine-citrullin-p-aminobenzy ht-y loxycaronyl (“vc-PAB). Additional linker components are known in the art and some are described herein. O In some embodiments, the linker may comprise amino acid MP residues. Exemplary amino acid linker components include a O O dipeptide, a tripeptide, a tetrapeptide or a pentapeptide. 25 Exemplary dipeptides include: valine-citrulline (vc or val cit), alanine-phenylalanine (aforala-phe). Exemplary tripep ~~~%. h O tides include: glycine-valine-citrulline (gly-val-cit) and gly O cine-glycine-glycine (gly-gly-gly). Amino acid residues MPEG which comprise an amino acid linker component include 30 those occurring naturally, as well as minor amino acids and non-naturally occurring amino acid analogs, such as citrul Additional exemplary linker components and abbrevia line. Amino acid linker components can be designed and tions include (wherein the antibody (Ab) and linker are optimized in their selectivity for enzymatic cleavage by a depicted, and p is 1 to about 8):

O su-y-O Ab A Ab H r p HN HN es NH2 1. NH2 Wal-cit MC-val-cit O

O s

Ab H H r HN 1. NH2 MC-val-cit-PAB US 8,771,966 B2 29 30 Nucleophilic groups on antibodies include, but are not and linker reagents including: (i) active esters such as NHS limited to: (i) N-terminal amine groups, (ii) side chain amine esters, HOB testers, haloformates, and acid halides; (ii) alkyl groups, e.g. lysine, (iii) side chain thiol groups, e.g. cysteine, and benzyl halides such as haloacetamides; (iii) aldehydes, and (iv) Sugar hydroxyl or amino groups where the antibody ketones, carboxyl, and maleimide groups. is glycosylated. Amine, thiol, and hydroxyl groups are Methods for the conjugation of linker-drug moieties to nucleophilic and capable of reacting to form covalent bonds cell-targeted proteins such as antibodies, immunoglobulins or with electrophilic groups on linker moieties and linker fragments thereof are found, for example, in U.S. Pat. No. reagents including: (i)active esters such as NHS esters, HOBt 5,208,020; U.S. Pat. No. 6,441,163; WO2005037992: esters, haloformates, and acid halides; (ii) alkyl and benzyl WO2005081711; and WO2006/034488, all of which are halides such as haloacetamides; (iii) aldehydes, ketones, car 10 hereby expressly incorporated by reference in their entirety. boxyl, and maleimide groups. Certain antibodies have reduc ible interchain disulfides, i.e. cysteine bridges. Antibodies Alternatively, a fusion protein comprising the antibody and may be made reactive for conjugation with linker reagents by cytotoxic agent may be made, e.g., by recombinant tech treatment with a reducing agent such as DTT (dithiothreitol). niques or peptide synthesis. The length of DNA may com Each cysteine bridge will thus form, theoretically, two reac 15 prise respective regions encoding the two portions of the tive thiol nucleophiles. Additional nucleophilic groups can be conjugate either adjacent one another or separated by a region introduced into antibodies through the reaction of lysines encoding a linker peptide which does not destroy the desired with 2-iminothiolane (Traut’s reagent) resulting in conver properties of the conjugate. sion of an amine into a thiol. Reactive thiol groups may be In yet another embodiment, the antibody may be conju introduced into the antibody (or fragment thereof) by intro gated to a “receptor (such streptavidin) for utilization in ducing one, two, three, four, or more cysteine residues (e.g., tumor pre-targeting wherein the antibody-receptor conjugate preparing mutant antibodies comprising one or more non is administered to the patient, followed by removal of native cysteine amino acid residues). unbound conjugate from the circulation using a clearing Antibody drug conjugates of the invention may also be agent and then administration of a “ligand’ (e.g., avidin) produced by modification of the antibody to introduce elec 25 which is conjugated to a cytotoxic agent (e.g., a radionucle trophilic moieties, which can react with nucleophilic sub otide). stituents on the linker reagent or drug. The Sugars of glyco In one embodiment of the immunoconjugate, the cytotoxic Sylated antibodies may be oxidized, e.g. with periodate agent, D, is an auristatin of formula D, or D,

R3 O

N A N H R2 O R.4 R.5

R3 O CH3 R9 O H A N N R2 O R.4 R.5

45 oxidizing reagents, to formaldehyde or ketone groups which and wherein R and Rare each methyl, Rand Rare each may react with the amine group of linker reagents or drug isopropyl, R is sec-butyl, each R is independently selected moieties. The resulting imine Schiffbase groups may form a from CH, O CH, OH, and H: R is H; R' is aryl; Z is stable linkage, or may be reduced, e.g. by borohydride 50 —O— or NH ; R'' is H. C-C alkyl, or—(CH), O— reagents to form stable amine linkages. In one embodiment, (CH.) O-(CH), O CH; and R' is C(R) C reaction of the carbohydrate portion of a glycosylated anti (R)-aryl; and body with either galactose oxidase or sodium meta-periodate (d) p ranges from about 1 to 8. may yield carbonyl (aldehyde and ketone) groups in the pro The following embodiments are further provided for any of tein that can react with appropriate groups on the drug (Her 55 the above immunoconjugates. In one embodiment, an immu manson, Bioconjugate Techniques). In another embodiment, noconjugate has in vitro or in vivo cell killing activity. In one proteins containing N-terminal serine or threonine residues embodiment, the linker is attached to the antibody through a can react with sodium meta-periodate, resulting in production thiol group on the antibody. In one embodiment, the linker is of an aldehyde in place of the first amino acid (Geoghegan & cleavable by a protease. In one embodiment, the linker com Stroh, (1992) Bioconjugate Chem. 3:138-146; U.S. Pat. No. 60 prises a val-cit dipeptide. In one embodiment, the linker com 5.362,852). Such aldehyde can be reacted with a drug moiety prises a p-aminobenzyl unit. In one embodiment, the p-ami or linker nucleophile. nobenzyl unit is disposed between the drug and a protease Likewise, nucleophilic groups on a drug moiety include, cleavage site in the linker. In one embodiment, the p-ami but are not limited to: amine, thiol, hydroxyl, hydrazide, nobenzyl unit is p-aminobenzyloxycarbonyl (PAB). In one oXime, hydrazine, thiosemicarbazone, hydrazine carboxy 65 embodiment, the linker comprises 6-maleimidocaproyl. In late, and arylhydrazide groups capable of reacting to form one embodiment, the 6-maleimidocaproyl is disposed covalent bonds with electrophilic groups on linker moieties between the antibody and a protease cleavage site in the US 8,771,966 B2 31 32 linker. The above embodiments may occur singly or in any Articles of Manufacture: combination with one another. In another embodiment of the invention, an article of In one embodiment, the drug is selected from MMAE and manufacture, or "kit, containing materials useful for the MMAF. In one embodiment, the immunoconjugate has the treatment of the disorders described above is provided. The formula article of manufacture comprises a container and a label or

O O Ab-S O H OH N, H O Or N N N---- Val-Cit-Nlff - O O. n O. ON O O p

15 wherein Ab is any of the above anti-STEAP-1 antibodies, S package insert on or associated with the container. Suitable is a Sulfur atom, and pranges from about 2 to about 5. In one containers include, for example, bottles, vials, Syringes, blis embodiment, the immunoconjugate has the formula ter pack, etc. The containers may be formed from a variety of

O O Ab-S O ul N.,H H N-a-N- Wal-Cit-N OrrrrrrlO O. n O. O. O O H n O1 OH p O

wherein Ab is any of the above anti-STEAP-1 antibodies, S materials such as glass or plastic. The container holds an is a Sulfur atom, and p ranges from about 1 to about 6, from 30 antibody-drug conjugate (ADC) composition which is effec about 2 to about 5, from about 2 to about 6, from about 2 to tive for treating the condition and may have a sterile access about 4, from about 2 to about 3, from about 3 to about 4, from port (for example the container may be an intravenous solu about 3 to about 5, from about 3 to about 6, or from about 4 to tion bag or a vial having a stopper pierceable by a hypodermic about 6. injection needle). At least one active agent in the composition 35 Labelled Antibody Imaging Methods: is an ADC. The label or package insert indicates that the composition is used for treating the condition of choice. Such In another embodiment of the invention, cysteine engi as cancer. Alternatively, or additionally, the article of manu neered antibodies may be labelled through the cysteine thiol with radionuclides, fluorescent dyes, bioluminescence-trig facture may further comprise a second (or third) container 40 comprising a pharmaceutically-acceptable buffer, Such as gering Substrate moieties, chemiluminescence-triggering bacteriostatic water for injection (BWFI), phosphate-buff Substrate moieties, enzymes, and other detection labels for imaging experiments with diagnostic, pharmacodynamic, ered Saline, Ringer's Solution and dextrose solution. It may and therapeutic applications. Generally, the labelled cysteine further include other materials desirable from a commercial engineered antibody, i.e. “biomarker” or “probe’, is admin and user standpoint, including other buffers, diluents, filters, 45 needles, and Syringes. istered by injection, perfusion, or oral ingestion to a living Pharmaceutical Compositions: organism, e.g. human, rodent, or other Small animal, a per In one aspect, a pharmaceutical composition is provided fused organ, or tissue sample. The distribution of the probe is comprising any of the above immunoconjugates and a phar detected over a time course and represented by an image. maceutically acceptable carrier. In one aspect, a method of Zirconium-Labelling Reagents 50 treating a prostate, lung, colon, bladder, or ovarian cell pro liferative disorder, or Ewings sarcoma is provided, wherein In another embodiment of the invention, the development the method comprises administering to an individual the of Zr-based Immuno-PET has enabled the measurement of pharmaceutical composition. In one embodiment, the pros target expression and antibody biodistribution in mice and tate, lung, colon, bladder, and ovarian cancer and Ewings humans: (Dijkers et al., Nature Vol. 87 Number 5 (May 55 sarcoma cell proliferative disorder is a metastasis of a primary 2010)). ImmunoPET as a measurement of target expression prostate, lung, colon, bladder, or ovarian cancer or Ewings and antibody biodistribution can have value in determining sarcoma. In one embodiment, the cell proliferative disorder is the distribution of ADC targets on primary vs. metastatic associated with increased expression of STEAP-1 on the lesions, particularly in prostate and other cancers. surface of a cell. The radioisotope or other labels may be incorporated in the 60 In one aspect, a method of inhibiting cell proliferation is conjugate in known ways (Fraker et al (1978) Biochem. Bio provided, wherein the method comprises exposing a cell to phys. Res. Commun. 80: 49-57; “Monoclonal Antibodies in any of the above immunoconjugates under conditions permis Immunoscintigraphy” Chatal, CRC Press 1989). Carbon-14 sive for binding of the immunoconjugate to STEAP-1. In one labeled 1-isothiocyanatobenzyl-3-methyldiethylene tri embodiment, the prostate, lung, colon, bladder, or ovary cell aminepentaacetic acid (MX-DTPA) is an exemplary chelat 65 or Ewings sarcoma is a tumor cell. In one embodiment, the ing agent for conjugation of a radionuclide to the antibody tumor cell is a prostate, lung, colon, bladder, or ovarian tumor (WO 94/11026). cell or Ewing's sarcoma cell of a mammal experiencing or US 8,771,966 B2 33 34 Suspected of experiencing prostate, lung, colon, bladder cell human VH acceptor 1 framework 'A' minus Kabat CDRs or Ewings sarcoma proliferative disorder including, but not (SEQID NOs:48, 42, 49, 29). limited to, a metastasis of a primary prostate, lung, colon, human VH acceptor frameworks “B” and “C” minus bladder cell cancer tumor or Ewings sarcoma tumor. In one extended hypervariable regions (SEQID NOs:44, 45, 49, 29: embodiment, the prostate, lung, colon, bladder cell or and SEQID NOs:44, 45, 50, 29). Ewing's sarcoma is a Xenograft. In one embodiment, the human VH acceptor 2 framework 'A' minus Kabat CDRs exposing takes place in vitro. In one embodiment, the expos (SEQID NOs:48, 42, 51, 29). ing takes place in vivo. human VHacceptor 2 framework “B,” “C.” and “D' minus In one aspect, a method of using the anti-STEAP-1 anti extended hypervariable regions (SEQID NOs:44, 45, 51, 29: body of the invention is provided to assay serum soluble 10 SEQID NOs:44, 45, 52,29; and SEQID NOs:44, 45, 53,29). STEAP-1 in a mammal experiencing prostate lung or colon FIGS. 4A and 4B show exemplary acceptor human vari cell proliferative disorder (or metastasis of a primary inci able light (VL) consensus framework sequences for use in dence of Such disorder), measuring clinical progression or practicing the instant invention with sequence identifiers as regression of the diseases, or assess tumor burden or relapse. follows: 15 human VL kappa Subgroup I-1 consensus framework (KV1 BRIEF DESCRIPTION OF THE FIGURES 1): SEQID NOs:54, 55, 56, 57 human VL kappa Subgroup I consensus framework (KV1): FIG. 1 depicts the amino acid sequence of human SEQID NOs:54,58, 56, 57 STEAP-1 (SEQ ID NO:1) aligned with STEAP-1 from human VL kappa Subgroup II consensus framework (KV2): mouse and cynomolgus monkey (cyno) (SEQID NOS:2 and SEQID NOs:69,59, 60,57 3, respectively). Extracellular domains 1, 2, and 3 are labeled human VL kappa Subgroup III consensus framework and marked by shaded boxes. (Kv3): SEQID NOs:61, 62, 63,57 FIGS. 2A-2B: FIG. 2A depicts the amino acid sequence of human VL kappa Subgroup IV consensus framework the light chain variable region of murine 120.545 anti (Kv4): SEQID NOs:64, 65, 66, 57. STEAP-1 antibody aligned with the chimera antibody (120 25 FIG. 5 depicts alignments of native sequence human IgG chimera) and humanized antibody (120 graft) and aligned Fc region sequences, humIgG1 (non-A allotype, SEQ ID with the human subgroup III sequence. The CDRs are boxed NO:85; and A allotype, where the amino acid sequence (CDR-L1, CDR-L2, and CDR-L3). The sequences bracket SREEM (SEQID NO:67) within SEQID NO:85 is changed ing the CDRs are the framework sequences (FR-L1 to FR to SRDEL (SEQ ID NO:68), humIgG2 (SEQ ID NO:86), L4). The sequences are numbered according to Kabat num 30 humIgG3 (SEQ ID NO:87) and humIgG4 (SEQ ID NO:88) bering. The Kabat, Chothia, and contact CDRs are indicated with differences between the sequences marked with aster about the boxed CDRs. FIG. 2B depicts the amino acid isks. Numbers above the sequences represent the EU num sequence of the heavy chain variable region of murine anti bering system. An exemplary kappa constant region is also STEAP-1 antibody (120.545) aligned with the chimera anti shown. body (120 chimera) and humanized antibody (120 graft) and 35 FIGS. 6A-6D depicts a FACS analysis normalized for the aligned with the human kappa I sequence. Humanized vari level of display of each antibody or variant on phage. FIG.6A ants 24, 37,48, 67, and 37/48, 67,71, and 78 were prepared by shows FACS shifts on STEAP-1 expressing cells (LB50) for making the following amino acid changes: A24V, V37I. four exemplary antibodies. FIG. 6B shows FACS shifts on V48M, F67I, and L78F in the heavy chain of the 120 graft STEAP-1 non-expressing cells (S408) for several antibodies antibody. The CDRs are boxed. The FR-H1, FR-H2, FR-3 40 as indicated in the figure and in Example 1. FIGS. 6C and 6D and FR-4 sequences bracket the CDRs (CDR-H1, CDR-H2, are FACS shift alignments after normalization for phage dis and CDR-H3). The sequences are numbered according to play levels. Kabat numbering. The Kabat, Chothia, and contact CDRs are FIGS. 7A-7F graphically depict FACS analyses showing indicated about the boxed CDRs. binding of anti-STEAP-1 murine, chimera and humanized FIGS. 3A and 3B show exemplary acceptor human vari 45 version 24 antibodies to human STEAP-1 expressed on the able heavy (VH) consensus framework sequences for use in cell Surface. FIGS. 7A-7C indicate that anti-STEAP-1 murine practicing the instant invention with sequence identifiers as 120, chimera 120 and humanized 120V.24 bind human and follows, where the FRSEQ ID NOs are listed in the order cynomolgus monkey STEAP-1, but not mouse STEAP-1. FR-1, FR-2, FR-3, FR-4: FIGS. 7D-7F are FACS plots showing binding of murine 120, human VH subgroup I consensus framework 'A' minus 50 120 chimera, and humanized 120V.24 (clone 67) to human Kabat CDRs (SEQID NOs:26, 27, 28, 29). STEAP-1 expressed on the cell surface. Exogenous STEAP-1 human VH subgroup I consensus frameworks “B,” “C.” was stably expressed in 293 cells (designated LB50 cells) and and “D' minus extended hypervariable regions (SEQ ID PC3 cells (designated PS5.4 cells) (FIGS. 7D and 7E), and NOs:30, 31, 28, 29; SEQID NOs:30, 31, 32,29; and SEQID endogenously expressed in LNCaP BR cells (FIG. 7F). NOs:30, 31, 33, 29). 55 FIGS. 8A and 8B. FIG. 8A is a graph showing that admin human VH subgroup II consensus framework 'A' minus istration of murine anti-STEAP-1 120-MC-vc-PAB-MMAE Kabat CDRs (SEQID NOS:34, 35, 36, 29). at 3 mg/kg was efficacious in a prostate tumor (LNCaP-Ner human VH subgroup II consensus frameworks “B,” “C.” cells) xenograft model. See Example 4. FIG. 8B is a graph and “D' minus extended hypervariable regions (SEQ ID showing that single dose administration of humanized anti NOs:37,38, 36,29; SEQID NOs:37,38, 39,29; and SEQID 60 STEAP-1 antibody 120V.24-MC-vc-PAB-MMAE (3 mg/kg), NOs:37, 38, 40, 29). 120V.24-MC-MMAF (6 mg/kg), 120V.24-MC-MMAF (12 human VH subgroup III consensus framework 'A' minus mg/kg), and anti-STEAP-1 120 chimera-MC-vc-PAB Kabat CDRs (SEQID NOs:41, 42, 43,29). MMAE (3 mg/kg) was shown to be efficacious in a LNCaP human VH subgroup III consensus frameworks “B,” “C.” cell Xenograft prostate tumor model. See Example 4. and “D' minus extended hypervariable regions (SEQ ID 65 FIG. 9 is a graph showing that administration of anti NOs:44, 45, 43,29; SEQID NOs:44, 45,46,29; and SEQID STEAP-1 antibody 120 chimera-MC-vc-PAB-MMAE (ab NOs:44, 45,46, 29). breviated anti-STEAP vcMMAE) at 3 mg/kg, or anti US 8,771,966 B2 35 36 STEAP-1 120 chimera-MC-MMAF (abbreviated anti bering and standardized numbering according to the EU sys STEAP mcMMAF) at 6 mg/kg, was shown to be efficacious tem. FIG. 17C discloses all sequences as SEQID NO:109. in a prostate cancer Xenograft model of castrated SCID-beige FIGS. 18A-F depict FACS analyses showing that the anti mice transplanted with LNCaP cells. See Example 4. STEAP-1 thio antibody drug conjugates (TDCs) retain the FIG. 10 is a graph showing that administration of anti ability to bind to STEAP-1 expressed on the cell surface. STEAP-1 antibody 120 chimera-MC-vc-PAB-MMAE (ab FIGS. 18A-18C are FACS plots showing binding of the anti breviated anti-STEAP vcMMAE (at 3 mg/kg) was shown to STEAP-1 TDCS thio-human120-vc-PAB-MMAE be efficacious in a prostate cancer xenograft model of SCID (LCV205C) (abbreviated as huSteap 1 TDC (L205C) vcE and beige male mice (androgen dependent) transplanted with thio-human120-vc-PAB-MMAE (HCA118C) (abbreviated Lucap 77 cells. See Example 4. 10 FIG. 11 is a graph showing that administration of human as huSteap 1 TDC (HCA118C) vcE) to human STEAP-1 ized anti-STEAP-1 antibody 120V.24-MC-vc-PAB-MMAE expressed on the cell surface. Exogenous STEAP-1 was sta at 3 mg/kg, humanized anti-STEAP-1 antibody 120V.24-MC bly expressed in 293 cells (designated LB50 cells) and PC3 MMAF at 6 mg/kg and 12 mg/kg to castrated SCID-beige cells (designated PS5.4 cells) (FIGS. 18A and 18B), and mice transplanted with Lucap35V prostate tumor was shown 15 endogenously expressed in LNCaP BR cells (FIG. 18C). to be efficacious relative to controls. See Example 4. FIGS. 18D, 18E. and 18F are alignments of the FACS shifts FIG. 12 is a diagram depicting STEAP-1 embedded in a shown in FIGS. 7A, 7B and 7C, respectively. cell membrane. Anti-STEAP-1 antibody 120 binding is con FIGS. 19 A-C show the relative effectiveness of the anti formation dependent and does not recognize a linear epitope STEAP-1 thio antibody drug conjugates (TDCs) thio-hu of STEAP-1. man120-vc-PAB-MMAE (LCV205C) (abbreviated as FIGS. 13 A-13D show STEAP-1 expressed on the surface huSteap 1 TDC (L205C) vcE) and thio-human120-vc-PAB of cells as detected by immunohistochemistry. FIG. 13A MMAE (HCA118C) (abbreviated as huSteap 1 TDC shows an immunohistochemical stain of 293 cells expressing (HCA118C) vcE) to kill STEAP-1 expressing cells in vitro. exogenous STEAP-1 on the cell surface. FIG. 13B shows an LB50 cells (FIG. 19A) are 293 cells transformed with a vector immunohistochemical stain of PC3 cells expressing exog 25 encoding STEAP-1 such that STEAP-1 is expressed on the enous STEAP-1 on the cell surface. FIG. 13C shows an cell surface. PS5.4 cells (FIG. 19B) are PC3 cells transformed immunohistochemical stain of LNCaP cells expressing with a vector encoding STEAP-1 such that STEAP-1 is endogenous STEAP-1 on the cell surface. FIG.13D shows an expressed on the cell surface. LNCaP cells (FIG. 19C) immunohistochemica stain of LucAP 77 cells expressing express STEAP-1 endogenously. endogenous STEAP-1 on the cell surface. 30 FIG. 20 is a graph showing that administration of anti FIGS. 14A-14E are graphs showing the relative effective STEAP-1 TDC thio-human120-vc-PAB-MMAE ness of anti-STEAP-1 antibody 120V.24-MCMMAF and (HCA118C) (abbreviated as hu Steap1 HC TDC veE) at 3 anti-STEAP-1 antibody 120V.24-MC-vc-PAB-MMAE to kill mg/kg was shown to be efficacious relative to controls in a STEAP-1 expressing cells in vitro. PS5.4 cells (FIG. 14A) are prostate cancer Xenograft model of male SCID-beige mice PC3 cells transformed with a vector encoding STEAP-1 such 35 (androgen-dependent) transplanted with LNCaP cells. See that STEAP-1 is expressed on the cell surface. LB50 cells Example 8. (FIG. 14B) are 293 cells transformed with a vector encoding FIG. 21 is a graph showing that administration of anti STEAP-1 such that STEAP-1 is expressed on the cell surface. STEAP-1 TDC thio-human120-vc-PAB-MMAE LNCaP cells (FIG. 14C) express STEAP-1 endogenously. (HCA118C) (abbreviated as hu Steap1 HC TDC veh) at 3 “PC3 vec” (FIG. 14D) and “293 vec” (FIG. 14E) refer to 293 40 mg/kg, or thio-human120-MC-MMAF (HCA118C) (abbre cells and PC3 cells, respectively, transformed with a vector viated as hu Steap1 HC TDC mcF) at 1, 3 or 6 mg/kg, was control. shown to be efficacious relative to controls in a prostate can FIG. 15 shows depictions of cysteine engineered anti cer Xenograft model of male SCID-beige mice (androgen STEAP-1 antibody drug conjugates (ADC) where a drug dependent) transplanted with LNCaP cells. See Example 8. moiety is attached to an engineered cysteine group in: the 45 FIG. 22 is a graph showing that administration of anti light chain (LC-ADC); the heavy chain (HC-ADC); and the STEAP-1 TDC thio-human120-vc-PAB-MMAE Fc region (Fc-ADC). (HCA118C) (abbreviated as hu Steap1 HC TDC veh) at 3 FIG. 16 shows the steps of: (i) reducing cysteine disulfide mg/kg, or thio-human120-MC-MMAF (HCA125C) (abbre adducts and interchain and intrachain disulfides in a cysteine viated ashu Steap1 HCTDC mcF) at 3 or 6 mg/kg, was shown engineered anti-STEAP-1 antibody (ThioMab) with reduc 50 to be efficacious relative to controls in a prostate cancer ing agent TCEP (tris(2-carboxyethyl)phosphine hydrochlo xenograft model of castrated SCID-beige mice transplanted ride); (ii) partially oxidizing, i.e. reoxidation to reform inter with LucaP 35V prostate tumor. See Example 8. chain and intrachain disulfides, with dha A (dehydroascorbic FIG. 23A-B shows the sites of amino acid substitutions acid); and (iii) conjugation of the reoxidized antibody with a made to generate the cysteine-engineered anti-STEAP-1 anti drug-linker intermediate to form a cysteine engineered anti 55 body (thio-mAb) designated “Simmons IV or simply STEAP-1 antibody drug conjugate (ADC). “SGIV.” The amino acid sequence of the light chain of SGIV FIGS. 17A-C show the sites of amino acid substitutions (SEQID NO:90) is shown in alignment with the light chain of made to generate cysteine-engineered anti-STEAP-1 anti mu 120 antibody (SEQ ID NO:85) and 120.v24 antibody bodies (thio-mAbs). FIG. 17A shows the thio-LC variant (SEQ ID NO:91). The thio-LC variant SGIV with corre V205C with corresponding sequential numbering and stan 60 sponding sequential numbering and standardized numbering dardized numbering according to the Kabat system. FIG.17A according to the Kabat system is shown aligned with the discloses all sequences as SEQ ID NO:70. FIG. 17B shows parent antibody mu 120 as well as the thio-LC variant the thio-HC variant A118C with corresponding sequential 120.V24 with corresponding sequential numbering and stan numbering and standardized numbering according to the EU dardized numbering according to the Kabat System. The system. FIG.17B discloses SEQID NOS 71-74 and 102-108, 65 CDRs are boxed (CDR-L1, CDR-L2, and CDR-L3). The respectively, in order of appearance. FIG. 17C shows the sequences bracketing the CDRS are the framework sequences thio-Fc variant S400C with corresponding sequential num (FR-L1 to FR-L4). The sequences are numbered according to US 8,771,966 B2 37 38 Kabat numbering. The Kabat, Chothia, and contact CDRs are useful for the detection of STEAP-1 expression on the surface indicated about the boxed CDRs. See Example 9. of normal and/or tumor or cancer cells of protate, lung or FIG. 24A-B shows the sites of framework amino acid colon tissue. Substitutions made to generate various cysteine-engineered Polynucleotides encoding anti-STEAP-1 antibodies are 5 provided. Vectors comprising polynucleotides encoding anti anti-STEAP-1 antibody (thio-mAb) variants of the SGIV and STEAP-1 antibodies are provided, and host cells comprising 120V.24 antibodies as disclosed in FIG. 23A. The amino acid Such vectors are provided. Compositions, including pharma sequence of the light chain of SGIV is shown with standard ceutical formulations, comprising any one or more of the ized numbering according to the Kabat System, in alignment polynucleotides, anti-STEAP-1 antibodies, or immunocon with variants LS.VLVH1 (SEQID NO:92); LS.VLVH2(SEQ jugates of the invention are also provided. ID NO:93); LS.Q (SEQ ID NO:94); and LS.CH1 (SEQ ID 10 Methods of treating a cell proliferative disorder, including NO: 95). The amino acid sequence of the light chain of but not limited to tumor or cancer, with an anti-STEAP-1 120.V24 with standardized numbering according to the Kabat antibody, antibody drug conjugate or immunoconjugate are system is shown in alignment with variants ED.FW1 (SEQID provided. Such methods include, but are not limited to, treat ment of tumor or cancer in prostate, lung or colon of a mam NO:96); EDFW2 (SEQ ID NO:97); EDFW3 (SEQ ID 15 mal. Methods of detecting STEAP-1 expression on a tissue NO:98); ED.all (SEQIDNO:99); ED.Pro (SEQID NO:100); cell using an anti-STEAP-1 antibody, antibody drug conju and ED. p1 (SEQ ID NO:101). The CDRs are boxed. The gate or immunoconjugate are provided. Such methods sequences are numbered according to Kabat numbering. See include, but are not limited to, detection of STEAP-1 expres Example 9. FIG. 24A discloses the “Simmons IV” and sion on, as a non-limiting example, a normal cell, tumor cell, “120V.24” variant sequences as SEQ ID NOs 110-111, or cancer cell of prostate, lung, or colon cell. respectively. General Techniques FIG. 25 shows Scatchard plots of antibody binding to The techniques and procedures described or referenced STEAP-1 expressed on the surface of LNCaPBR cells. herein are generally well understood and commonly Duplicate samples were measured using the 120.V24 anti employed using conventional methodology by those skilled body (FIGS. 25(A)-(D)) and the SGIV variant (FIGS. 25(E)- 25 in the art, such as, for example, the widely utilized method (H)). See Example 9. ologies described in Sambrook et al., Molecular Cloning. A FIG. 26 shows Scatchard plots of antibody binding to Laboratory Manual 3rd. edition (2001) Cold Spring Harbor STEAP-1 expressed on the surface of 293.LB50 cells. Dupli Laboratory Press, Cold Spring Harbor, N.Y.: Current Proto cate samples were measured using the 120.V24 antibody cols in Molecular Biology (F.M. Ausubel, et al. eds., (2003)); (FIGS. 26(A)-(D)) and the SGIV variant (FIGS. 26(E)-(H)). 30 the series Methods in Enzymology (Academic Press, Inc.): See Example 9. Pcr2: A Practical Approach (M.J. MacPherson, B.D. Hames FIG. 27 is a table comparing the average binding affinities, and G. R. Taylor eds. (1995)), Harlow and Lane, eds. (1988) as measured by Scatchard analysis, for the mu 1789, mu 120, Antibodies, A Laboratory Manual, and Animal Cell Culture Fc chimera, humanized 120.V24, thio-120.V24 and thio (R. I. Freshney, ed. (1987)); Oligonucleotide Synthesis (M. J. SGIV antibodies in PC-3-PS5.4, 293-LB50 and LNCaP-BR 35 Gait, ed., 1984); Methods in Molecular Biology, Humana cells, as well as in 293 cells transiently expressing STEAP-1. Press; Cell Biology: A Laboratory Notebook (J. E. Cellis, ed., See Example 9. 1998) Academic Press; Animal Cell Culture (R.I. Freshney), FIGS. 28A-D depicts a FACS analysis showing FACS ed., 1987); Introduction to Cell and Tissue Culture (J. P. shifts on cells stably transfected with STEAP-1 (293 Mather and P. E. Roberts, 1998) Plenum Press; Cell and STEAP-1 LB48, 293 STEAP-1 LB50 and 293 STEAP-1 40 Tissue Culture. Laboratory Procedures (A. Doyle, J. B. Grif LB53, respectively) with SGIV and 120.v24 antibody fiths, and D. G. Newell, eds., 1993-8) J. Wiley and Sons: samples. See Example 9. Handbook of Experimental Immunology (D. M. Weir and C. FIG. 29 shows the antibody titer observed in different C. Blackwell, eds.); Gene Transfer Vectors for Mammalian harvests from cells producing SGIV or 120.V24 antibody. Cells (J. M. Miller and M. P. Calos, eds., 1987); PCR. The FIG. 30 shows zirconium-89 PET imaging in animals 45 Polymerase Chain Reaction, (Mullis et al., eds., 1994); Cur dosed with 100 uCi ZR89-anti-STEAP-1 and imaged 4 days rent Protocols in Immunology (J. E. Coligan et al., eds., later wherein uptake of Zr89-STEAP-1 is highest in 1991); Short Protocols in Molecular Biology (Wiley and Lucap35V and lowest in Lucap 96.1 similar to in vitro char Sons, 1999); Immunobiology (C. A. Janeway and P. Travers, acterization. 1997); Antibodies (P. Finch, 1997); Antibodies: A Practical 50 Approach (D. Catty... ed., IRL Press, 1988-1989); Mono DETAILED DESCRIPTION OF EMBODIMENTS clonal Antibodies: A Practical Approach (P. Shepherd and C. OF THE INVENTION Dean, eds., Oxford University Press, 2000); Using Antibod ies: A Laboratory Manual (E. Harlow and D. Lane (Cold Isolated antibodies that bind to STEAP-1 are provided. Spring Harbor Laboratory Press, 1999); The Antibodies (M. Immunoconjugates comprising anti-STEAP-1 antibodies are 55 Zanetti and J. D. Capra, eds., Harwood Academic Publishers, further provided. Antibodies and immunoconjugates of the 1995); and Cancer. Principles and Practice of Oncology (V. invention are useful, e.g., for the diagnosis or treatment of T. DeVita et al., eds., J.B. Lippincott Company, 1993). Zr disorders associated with altered expression, e.g., increased Immuno-PET: Comphrehensive Procedures for the Produc expression, of STEAP-1. In certain embodiments, antibodies tion of Zr-Labeled Monoclonal Antibodies (Verel I., et al. J or immunoconjugates of the invention are useful for the diag 60 Nucl Med; 44:1271-1281, 2003). nosis or treatment of a cell proliferative disorder, Such as a tumor or cancer. In certain embodiments, STEAP-1 is DEFINITIONS AND ABBREVIATIONS expressed in tumor or cancer of prostate, lung, or colon tissue. In certain embodiments, antibodies or immunoconjugates of Definitions the invention are useful for the detection of STEAP-1, e.g., 65 STEAP-1 expressed on the cell surface. In certain embodi An "isolated antibody is one which has been identified ments, antibodies or immunoconjugates of the invention are and separated and/or recovered from a component of its natu US 8,771,966 B2 39 40 ral environment. Contaminant components of its natural envi into the genome of a host cell upon introduction into the host ronment are materials which would interfere with research, cell, and thereby are replicated along with the host genome. diagnostic or therapeutic uses for the antibody, and may Moreover, certain vectors are capable of directing the expres include enzymes, hormones, and other proteinaceous or non sion of genes to which they are operatively linked. Such proteinaceous solutes. In some embodiments, an antibody is 5 vectors are referred to herein as “recombinant expression purified (1) to greater than 95% by weight of antibody as vectors, or simply, 'expression vectors. In general, expres determined by, for example, the Lowry method, and in some sion vectors of utility in recombinant DNA techniques are embodiments, to greater than 99% by weight; (2) to a degree often in the form of plasmids. In the present specification, sufficient to obtain at least 15 residues of N-terminal or inter “plasmid' and “vector” may be used interchangeably as the nal amino acid sequence by use of for example, a spinning 10 plasmid is the most commonly used form of vector. cup sequenator, or (3) to homogeneity by SDS-PAGE under “Polynucleotide.” or “nucleic acid, as used interchange reducing or nonreducing conditions using, for example, Coo ably herein, refer to polymers of nucleotides of any length, massie blue or silver stain. Isolated antibody includes the and include DNA and RNA. The nucleotides can be deoxyri antibody in situ within recombinant cells since at least one bonucleotides, ribonucleotides, modified nucleotides or component of the antibody's natural environment will not be 15 bases, and/or their analogs, or any Substrate that can be incor present. Ordinarily, however, isolated antibody will be pre porated into a polymer by DNA or RNA polymerase or by a pared by at least one purification step. synthetic reaction. A polynucleotide may comprise modified An "isolated nucleic acid molecule is a nucleic acid mol nucleotides, such as methylated nucleotides and their ana ecule that is separated from at least one other nucleic acid logs. If present, modification to the nucleotide structure may molecule with which it is ordinarily associated, for example, be imparted before or after assembly of the polymer. The in its natural environment. An isolated nucleic acid molecule sequence of nucleotides may be interrupted by non-nucle further includes a nucleic acid molecule contained in cells otide components. A polynucleotide may comprise modifi that ordinarily express the nucleic acid molecule, but the cation(s) made after synthesis, Such as conjugation to a label. nucleic acid molecule is present extrachromosomally or at a Other types of modifications include, for example, "caps.” chromosomal location that is different from its natural chro 25 Substitution of one or more of the naturally occurring nucle mosomal location. otides with an analog, internucleotide modifications such as, “Purified” means that a molecule is present in a sample at for example, those with uncharged linkages (e.g., methyl a concentration of at least 95% by weight, or at least 98% by phosphonates, phosphotriesters, phosphoamidates, carbam weight of the sample in which it is contained. ates, etc.) and with charged linkages (e.g., phosphorothioates, The term “substantially similar or “substantially the 30 phosphorodithioates, etc.), those containing pendant moi same as used herein, denotes a Sufficiently high degree of eties, such as, for example, proteins (e.g., nucleases, toxins, similarity between two numeric values (for example, one antibodies, signal peptides, ply-L-lysine, etc.), those with associated with an antibody of the invention and the other intercalators (e.g., acridine, psoralen, etc.), those containing associated with a reference/comparator antibody). Such that chelators (e.g., metals, radioactive metals, boron, oxidative one of skill in the art would consider the difference between 35 metals, etc.), those containing alkylators, those with modified the two values to be of little or no biological and/or statistical linkages (e.g., alpha anomeric nucleic acids, etc.), as well as significance within the context of the biological characteristic unmodified forms of the polynucleotides(s). Further, any of measured by said values (e.g., Kd values). The difference the hydroxyl groups ordinarily present in the Sugars may be between said two values is, for example, less than about 50%, replaced, for example, by phosphonate groups, phosphate less than about 40%, less than about 30%, less than about 40 groups, protected by standard protecting groups, or activated 20%, and/or less than about 10% as a function of the refer to prepare additional linkages to additional nucleotides, or ence/comparator value. may be conjugated to Solid or semi-solid Supports. The 5' and The phrase “substantially reduced, or “substantially dif 3' terminal OH can be phosphorylated or substituted with ferent, as used herein, denotes a Sufficiently high degree of amines or organic capping group moieties of from 1 to 20 difference between two numeric values (generally one asso- 45 carbon atoms. Other hydroxyls may also be derivatized to ciated with a molecule and the other associated with a refer standard protecting groups. Polynucleotides can also contain ence/comparator molecule) Such that one of skill in the art analogous forms of ribose or deoxyribose Sugars that are would consider the difference between the two values to be of generally known in the art, including, for example, 2'-O- statistical significance within the context of the biological methyl-, 2-O-allyl-2'-fluoro- or 2-azido-ribose, carbocyclic characteristic measured by said values (e.g., Kd values). The 50 Sugar analogs, C.-anomeric Sugars, epimeric Sugars Such as difference between said two values is, for example, greater arabinose, Xyloses or lyxoses, pyranose Sugars, furanose Sug than about 10%, greater than about 20%, greater than about ars, Sedoheptuloses, acyclic analogs, and basic nucleoside 30%, greater than about 40%, and/or greater than about 50% analogs such as methyl riboside. One or more phosphodiester as a function of the value for the reference/comparator mol linkages may be replaced by alternative linking groups. These ecule. 55 alternative linking groups include, but are not limited to, The term “vector” as used herein, is intended to refer to a embodiments wherein phosphate is replaced by P(O)S (“thio nucleic acid molecule capable of transporting another nucleic ate'), P(S)S (“dithioate”), (O)NR (“amidate'), P(O)R, P(O) acid to which it has been linked. One type of vector is a OR, CO, or CH2 (“formacetal”), in which each R or R is “plasmid, which refers to a circular double stranded DNA independently H or substituted or unsubstituted alkyl (1-20 into which additional DNA segments may be ligated. Another 60 C) optionally containing an ether (—O—) linkage, aryl, alk type of vector is a phage vector. Another type of vector is a enyl, cycloalkyl, cycloalkenyl or araldyl. Notall linkages in a viral vector, wherein additional DNA segments may be polynucleotide need be identical. The preceding description ligated into the viral genome. Certain vectors are capable of applies to all polynucleotides referred to herein, including autonomous replication in a host cell into which they are RNA and DNA. introduced (e.g., bacterial vectors having a bacterial origin of 65 “Oligonucleotide, as used herein, generally refers to short, replication and episomal mammalian vectors). Other vectors generally single-stranded, generally synthetic polynucle (e.g., non-episomal mammalian vectors) can be integrated otides that are generally, but not necessarily, less than about US 8,771,966 B2 41 42 200 nucleotides in length. The terms "oligonucleotide' and acid sequence of STEAP-1 from mouse and cynomolgus "polynucleotide are not mutually exclusive. The description monkey (SEQID NOS:2 and 3, respectively). above for polynucleotides is equally and fully applicable to Antibodies’ (Abs) and “immunoglobulins’ (Igs) are gly oligonucleotides. coproteins having similar structural characteristics. While “Percent (%) amino acid sequence identity” with respect to antibodies exhibit binding specificity to a specific antigen, a reference polypeptide sequence is defined as the percentage immunoglobulins include both antibodies and other anti of amino acid residues in a candidate sequence that are iden body-like molecules which generally lack antigen specificity. tical with the amino acid residues in the reference polypeptide Polypeptides of the latter kind are, for example, produced at sequence, after aligning the sequences and introducing gaps, low levels by the lymph system and at increased levels by if necessary, to achieve the maximum percent sequence iden 10 myelomas. tity, and not considering any conservative Substitutions as part The terms “antibody” and “immunoglobulin' are used of the sequence identity. Alignment for purposes of determin interchangeably in the broadest sense and include mono ing percent amino acid sequence identity can be achieved in clonal antibodies (e.g., full length or intact monoclonal anti various ways that are within the skill in the art, for instance, bodies), polyclonal antibodies, monovalent antibodies, mul using publicly available computer software such as BLAST, 15 tivalent antibodies, multispecific antibodies (e.g., bispecific BLAST-2, ALIGN or Megalign (DNASTAR) software. antibodies so long as they exhibit the desired biological activ Those skilled in the art can determine appropriate parameters ity) and may also include certain antibody fragments (as for aligning sequences, including any algorithms needed to described in greater detail herein). An antibody can be chi achieve maximal alignment over the full length of the meric, human, humanized and/or affinity matured. sequences being compared. For purposes herein, however, 96 The term “anti-STEAP-1 antibody” or “an antibody that amino acid sequence identity values are generated using the binds to STEAP-1 refers to an antibody that is capable of sequence comparison computer program ALIGN-2. The binding STEAP-1 with sufficient affinity such that the anti ALIGN-2 sequence comparison computer program was body is useful as a diagnostic and/or therapeutic agent in authored by Genentech, Inc., and the source code has been targeting STEAP-1. Preferably, the extent of binding of an filed with user documentation in the U.S. Copyright Office, 25 anti-STEAP-1 antibody to an unrelated, non-STEAP-1 pro Washington D.C., 20559, where it is registered under U.S. tein is less than about 10% of the binding of the antibody to Copyright Registration No.TXU510087. The ALIGN-2 pro STEAP-1 as measured, e.g., by a radioimmunoassay (RIA). gram is publicly available from Genentech, Inc., South San In certain embodiments, an antibody that binds to STEAP-1 Francisco, Calif., or may be compiled from the source code. has a dissociation constant (Kd) of s1 uM, s100 nM, s10 nM, The ALIGN-2 program should be compiled for use on a 30 s1 nM, or s(). 1 nM. In certain embodiments, an anti UNIX operating system, preferably digital UNIXV4.0D. All STEAP-1 antibody binds to an epitope of STEAP-1 that is sequence comparison parameters are set by the ALIGN-2 conserved among STEAP-1 from different species. program and do not vary. The “variable region' or “variable domain of an antibody In situations where ALIGN-2 is employed for amino acid refers to the amino-terminal domains of the heavy or light sequence comparisons, the '% amino acid sequence identity of 35 chain of the antibody. The variable domain of the heavy chain a given amino acid sequence A to, with, or against a given may be referred to as “VH. The variable domain of the light amino acid sequence B (which can alternatively be phrased as chain may be referred to as “VL.” These domains are gener a given amino acid sequence A that has or comprises a certain ally the most variable parts of an antibody and contain the % amino acid sequence identity to, with, or against a given antigen-binding sites. amino acid sequence B) is calculated as follows: 40 The term “variable” refers to the fact that certain portions of the variable domains differ extensively in sequence among 100 times the fraction XY antibodies and are used in the binding and specificity of each where X is the number of amino acid residues scored as particular antibody for its particular antigen. However, the identical matches by the sequence alignment program variability is not evenly distributed throughout the variable ALIGN-2 in that programs alignment of A and B, and where 45 domains of antibodies. It is concentrated in three segments Y is the total number of amino acid residues in B. It will be called complementarity-determining regions (CDRs) or appreciated that where the length of amino acid sequence A is hypervariable regions (HVRs) both in the light-chain and the not equal to the length of amino acid sequence B, the 96 amino heavy-chain variable domains. The more highly conserved acid sequence identity of A to B will not equal the 96 amino portions of variable domains are called the framework acid sequence identity of B to A. Unless specifically stated 50 regions (FR). The variable domains of native heavy and light otherwise, all '% amino acid sequence identity values used chains each comprise four FR regions, largely adopting a hereinare obtained as described in the immediately preceding beta-sheet configuration, connected by three CDRs, which paragraph using the ALIGN-2 computer program. form loops connecting, and in Some cases forming part of the The term "STEAP-1, as used herein, refers to any native beta-sheet structure. The CDRs in each chain are held STEAP-1 from any vertebrate source, including mammals 55 together in close proximity by the FR regions and, with the Such as primates (e.g. humans, cynomolgus monkey (cyno)) CDRs from the other chain, contribute to the formation of the and rodents (e.g., mice and rats), unless otherwise indicated. antigen-binding site of antibodies (see Kabat et al., The term encompasses “full-length.” unprocessed STEAP-1 Sequences of Proteins of Immunological Interest, Fifth Edi as well as any form of STEAP-1 that results from processing tion, National Institute of Health, Bethesda, Md. (1991)). The in the cell. The term also encompasses naturally occurring 60 constant domains are not involved directly in the binding of variants of STEAP-1, e.g., splice variants, allelic variants, an antibody to an antigen, but exhibit various effector func and isoforms. The amino acid sequence of human STEAP-1 tions, such as participation of the antibody in antibody-de is depicted in FIG. 1 (SEQ ID NO:1). In one embodiment, pendent cellular toxicity. STEAP-1 is expressed on the cell surface, such as on the The “light chains of antibodies (immunoglobulins) from Surface of a normal prostate, lung or colon cell, and has 65 any vertebrate species can be assigned to one of two clearly increased expression in prostate, lung or colon cancer cells or distinct types, called kappa (K) and lambda (W), based on the metastases of such cancer cells. FIG. 1 also depicts the amino amino acid sequences of their constant domains. US 8,771,966 B2 43 44 Depending on the amino acid sequences of the constant bear a free thiol group. F(ab') antibody fragments originally domains of their heavy chains, antibodies (immunoglobulins) were produced as pairs of Fab' fragments which have hinge can be assigned to different classes. There are five major cysteines between them. Other chemical couplings of anti classes of immunoglobulins: IgA, Ig), IgE, IgG and IgM, and body fragments are also known. several of these may be further divided into subclasses (iso 5 “Single-chain Fv' or “scFv'antibody fragments comprise types), e.g., IgG, IgG, IgGs. IgG4, IgA1, and IgA2. The the VH and VL domains of antibody, wherein these domains heavy chain constant domains that correspond to the different are present in a single polypeptide chain. Generally, the Sclv classes of immunoglobulins are called C, 6, e, Y, and L, respec polypeptide further comprises a polypeptide linker between tively. The Subunit structures and three-dimensional configu the VH and VL domains which enables the scEv to form the rations of different classes of immunoglobulins are well 10 desired structure for antigenbinding. For a review of schv see known and described generally in, for example, Abbas et al. Pluckthun, in The Pharmacology of Monoclonal Antibodies, Cellular and Mol. Immunology, 4th ed. (2000). An antibody vol. 113, Rosenburg and Moore eds., Springer-Verlag, New may be part of a larger fusion molecule, formed by covalentor York, pp. 269-315 (1994). non-covalent association of the antibody with one or more The term "diabodies' refers to small antibody fragments other proteins or peptides. 15 with two antigen-binding sites, which fragments comprise a The terms “full length antibody.” “intact antibody' and heavy-chain variable domain (VH) connected to a light-chain “whole antibody are used herein interchangeably to refer to variable domain (VL) in the same polypeptide chain (VH an antibody in its substantially intact form, not antibody VL). By using a linker that is too short to allow pairing fragments as defined below. The terms particularly refer to an between the two domains on the same chain, the domains are antibody with heavy chains that contain the Fc region. forced to pair with the complementary domains of another Antibody fragments' comprise only a portion of an intact chain and create two antigen-binding sites. Diabodies may be antibody, wherein the portion retains at least one, and as many bivalent or bispecific. Diabodies are described more fully in, as most or all, of the functions normally associated with that for example, EP 404,097; WO93/1161; Hudson et al. (2003) portion when present in an intact antibody. In one embodi Nat. Med. 9:129-134; and Hollinger et al., Proc. Natl. Acad. ment, an antibody fragment comprises an antigenbinding site 25 Sci. USA 90: 6444-6448 (1993). Triabodies and tetrabodies of the intact antibody and thus retains the ability to bind are also described in Hudson et al. (2003) Nat. Med. 9:129 antigen. In another embodiment, an antibody fragment, for 134. example one that comprises the Fc region, retains at least one The term “monoclonal antibody” as used herein refers to of the biological functions normally associated with the Fc an antibody obtained from a population of Substantially region when present in an intact antibody, Such as FcRn 30 homogeneous antibodies, i.e., the individual antibodies com binding, antibody half life modulation, ADCC function and prising the population are identical except for possible muta complement binding. In one embodiment, an antibody frag tions, e.g., naturally occurring mutations, that may be present ment is a monovalent antibody that has an in vivo half life in minor amounts. Thus, the modifier"monoclonal indicates Substantially similar to an intact antibody. For example, Such the character of the antibody as not being a mixture of discrete an antibody fragment may comprise on antigen binding arm 35 antibodies. In certain embodiments, such a monoclonal anti linked to an Fc sequence capable of conferring in vivo Stabil body typically includes an antibody comprising a polypeptide ity to the fragment. sequence that binds a target, wherein the target-binding Papain digestion of antibodies produces two identical anti polypeptide sequence was obtained by a process that includes gen-binding fragments, called “Fab' fragments, each with a the selection of a single target binding polypeptide sequence single antigen-binding site, and a residual "Fo' fragment, 40 from a plurality of polypeptide sequences. For example, the whose name reflects its ability to crystallize readily. Pepsin selection process can be the selection of a unique clone from treatment yields an F(ab') fragment that has two antigen a plurality of clones, such as a pool of hybridoma clones, combining sites and is still capable of cross-linking antigen. phage clones, or recombinant DNA clones. It should be “Fv' is the minimum antibody fragment which contains a understood that a selected target binding sequence can be complete antigen-binding site. In one embodiment, a two 45 further altered, for example, to improve affinity for the target, chain Fv species consists of a dimer of one heavy- and one to humanize the target binding sequence, to improve its pro light-chain variable domain in tight, non-covalent associa duction in cell culture, to reduce its immunogenicity in vivo, tion. In a single-chain FV (ScPv) species, one heavy- and one to create a multispecific antibody, etc., and that an antibody light-chain variable domain can be covalently linked by a comprising the altered target binding sequence is also a flexible peptide linker such that the light and heavy chains can 50 monoclonal antibody of this invention. In contrast to poly associate in a “dimeric' structure analogous to that in a two clonal antibody preparations which typically include differ chain Fv species. It is in this configuration that the three ent antibodies directed against different determinants CDRs of each variable domain interact to define an antigen (epitopes), each monoclonal antibody of a monoclonal anti binding site on the surface of the VH-VL dimer. Collectively, body preparation is directed against a single determinant on the six CDRs confer antigen-binding specificity to the anti 55 an antigen. In addition to their specificity, monoclonal anti body. However, even a single variable domain (or half of an body preparations are advantageous in that they are typically Fv comprising only three CDRS specific for an antigen) has uncontaminated by other immunoglobulins. the ability to recognize and bind antigen, although at a lower The modifier "monoclonal indicates the character of the affinity than the entire binding site. antibody as being obtained from a Substantially homoge The Fab fragment contains the heavy- and light-chain vari 60 neous population of antibodies, and is not to be construed as able domains and also contains the constant domain of the requiring production of the antibody by any particular light chain and the first constant domain (CH1) of the heavy method. For example, the monoclonal antibodies to be used in chain. Fab' fragments differ from Fab fragments by the addi accordance with the present invention may be made by a tion of a few residues at the carboxy terminus of the heavy variety of techniques, including, for example, the hybridoma chain CH1 domain including one or more cysteines from the 65 method (e.g., Kohler et al., Nature, 256: 495 (1975); Harlow antibody hinge region. Fab'-SH is the designation herein for et al., Antibodies: A Laboratory Manual, (Cold Spring Harbor Fab' in which the cysteine residue(s) of the constant domains Laboratory Press, 2" ed. 1988); Hammerlinget al., in: Mono US 8,771,966 B2 45 46 clonal Antibodies and T-Cell Hybridomas 563-681 (Elsevier, lin constant region (Fc), typically that of a human immuno N.Y., 1981)), recombinant DNA methods (see, e.g., U.S. Pat. globulin. For further details, see Jones et al., Nature 321:522 No. 4,816,567), phage display technologies (see, e.g., Clack 525 (1986); Riechmann et al., Nature 332:323-329 (1988); and Presta, Curr. Op. Struct. Biol. 2:593-596 (1992). See also son et al., Nature, 352: 624-628 (1991); Marks et al., J. Mol. the following review articles and references cited therein: Biol. 222:581-597 (1992); Sidhu et al., J. Mol. Biol. 338(2): Vaswani and Hamilton, Ann. Allergy, Asthma & Immunol. 299-310 (2004); Lee et al., J. Mol. Biol. 340(5): 1073-1093 1:105-115 (1998); Harris, Biochem. Soc. Transactions (2004): Fellouse, Proc. Natl. Acad. Sci. USA 101 (34): 12467 23:1035-1038 (1995); Hurle and Gross, Curr. Op. Biotech. 12472 (2004); and Lee et al., J. Immunol. Methods 284(1-2): 5:428-433 (1994). 119-132 (2004), and technologies for producing human or A "human antibody' is one which possesses an amino acid human-like antibodies in animals that have parts or all of the 10 sequence which corresponds to that of an antibody produced human immunoglobulin loci or genes encoding human by a human and/or has been made using any of the techniques immunoglobulin sequences (see, e.g., WO98/24893; WO96/ for making human antibodies as disclosed herein. This defi 34096; WO96/33735; WO91/10741; Jakobovits et al., Proc. nition of a human antibody specifically excludes a humanized Natl. Acad. Sci. USA 90: 2551 (1993); Jakobovits et al., antibody comprising non-human antigen-binding residues. Nature 362: 255-258 (1993); Bruggemann et al., Year in 15 The term “hypervariable region.” “HVR,” or “HV,” when Immunol. 7:33 (1993); U.S. Pat. Nos. 5,545,807; 5,545,806; used herein refers to the regions of an antibody variable 5,569,825; 5,625,126; 5,633,425; 5,661,016; Marks et al., domain which are hyperVariable in sequence and/or form Bio.Technology 10: 779-783 (1992); Lonberg et al., Nature structurally defined loops. Generally, antibodies comprise six 368: 856-859 (1994); Morrison, Nature 368: 812-813 (1994): hypervariable regions; three in the VH (H1, H2, H3), and Fishwild et al., Nature Biotechnol. 14: 845-851 (1996); Neu three in the VL (L1, L2, L3). In native antibodies, H3 and L3 berger, Nature Biotechnol. 14: 826 (1996) and Lonberg and display the most diversity of the six hyperVariable regions, Huszar, Intern. Rev. Immunol. 13: 65-93 (1995). and H3 in particular is believed to play a unique role in The monoclonal antibodies herein specifically include conferring fine specificity to antibodies. Xu et al. (2000) "chimeric' antibodies in which a portion of the heavy and/or Immunity 13:37-45: Johnson and Wu (2003) in Methods in light chain is identical with or homologous to corresponding 25 Molecular Biology 248:1-25 (Lo, ed., Human Press, Totowa, sequences in antibodies derived from a particular species or N.J.). Indeed, naturally occurring camelid antibodies consist belonging to a particular antibody class or Subclass, while the ing of a heavy chain only are functional and stable in the remainder of the chain(s) is identical with or homologous to absence of light chain. Hamers-Casterman et al. (1993) corresponding sequences in antibodies derived from another Nature 363:446-448: Sheriffetal. (1996) Nature Struct. Biol. species or belonging to another antibody class or Subclass, as 30 3:733-736. well as fragments of such antibodies, so long as they exhibit A number of hypervariable region delineations are in use the desired biological activity (U.S. Pat. No. 4,816,567; and and are encompassed herein. The Kabat Complementarity Morrison et al., Proc. Natl. Acad. Sci. USA 81:6851-6855 Determining Regions (CDRS) are based on sequence variabil (1984)). ity and are the most commonly used (Kabat et al., Sequences “Humanized forms of non-human (e.g., murine) antibod 35 of Proteins of Immunological Interest, 5th Ed. Public Health ies are chimeric antibodies that contain minimal sequence Service, National Institutes of Health, Bethesda, Md. (1991)). derived from non-human immunoglobulin. In one embodi Chothia refers instead to the location of the structural loops ment, a humanized antibody is a human immunoglobulin (Chothia and Lesk J. Mol. Biol. 196:901-917 (1987)). The (recipient antibody) in which residues from a hypervariable AbM hypervariable regions representa compromise between region of the recipient are replaced by residues from a hyper 40 the Kabat CDRs and Chothia structural loops, and are used by variable region of a non-human species (donorantibody) Such Oxford Molecular's AbM antibody modeling software. The as mouse, rat, rabbit, or nonhuman primate having the desired “contact’ hyperVariable regions are based on an analysis of specificity, affinity, and/or capacity. In some instances, frame the available complex crystal structures. The residues from work region (FR) residues of the human immunoglobulin are each of these hypervariable regions are noted below.

Loop

replaced by corresponding non-human residues. Further HyperVariable regions may comprise “extended hyperVari more, humanized antibodies may comprise residues that are able regions” as follows: 24-36 or 24-34 (L1), 46-56 or 50-56 not found in the recipient antibody or in the donor antibody. (L2) and 89-97 or 89-96 (L3) in the VL and 26-35 or 26-35A These modifications may be made to further refine antibody 60 (H1), 50-65 or 49-65 (H2) and 93-102, 94-102, or 95-102 performance. In general, a humanized antibody will comprise (H3) in the VH. The variable domain residues are numbered substantially all of at least one, and typically two, variable according to Kabat et al., Supra, for each of these definitions. domains, in which all or substantially all of the hypervariable The HVR-H1, HVR-H2, and HVR-H3 hypervariable regions loops correspond to those of a non-human immunoglobulin, of the humanized anti-STEAP-1 120V.24 antibodies of the and all or substantially all the FRs are those of a human 65 invention are H26-H35A, H49-H6, and H95-H102 using immunoglobulin sequence. The humanized antibody option Kabat numbering. The HVR-L1, HVR-L2, and HVR-L3 ally will also comprise at least a portion of an immunoglobu hypervariable regions of the humanized anti-STEAP-1 US 8,771,966 B2 47 48 120V.24 antibodies of the invention are L24-34, L50-56, and based activation motif (ITAM) in its cytoplasmic domain L89-97 using Kabat numbering. As used herein, the terms Inhibiting receptor FcyRIIB contains an immunoreceptor “HVR'' and “CDR are used interchangeably. tyrosine-based inhibition motif (ITIM) in its cytoplasmic “Framework” or “FR residues are those variable domain domain. (see Daéron, Annu. Rev. Immunol. 15:203-234 residues other than the hyperVariable region residues as (1997)). FcRs are reviewed in Ravetch and Kinet, Annu. Rev. herein defined. Immunol 9:457-92 (1991); Capel et al., Immunomethods The term “variable domain residue numbering as in Kabat' 4:25-34 (1994); and de Haas et al., J. Lab. Clin. Med. 126: or "amino acid position numbering as in Kabat, and varia 330-41 (1995). Other FcRs, including those to be identified in tions thereof, refers to the numbering system used for heavy the future, are encompassed by the term “FcR' herein. chain variable domains or light chain variable domains of the 10 compilation of antibodies in Kabat et al., Sequences of Pro The term “Fc receptor or “FeRalso includes the neonatal teins of Immunological Interest, 5" Ed. Public Health Ser receptor, FcRn, which is responsible for the transfer of mater vice, National Institutes of Health, Bethesda, Md. (1991). nal IgGs to the fetus (Guyer et al., J. Immunol. 1 17:587 Using this numbering system, the actual linear amino acid (1976) and Kim et al., J. Immunol. 24:249 (1994)) and regu sequence may contain fewer or additional amino acids corre 15 lation of homeostasis of immunoglobulins. Methods of mea sponding to a shortening of, or insertion into, a FR or HVR of suring binding to FcRn are known (see, e.g., Ghetie 1997, the variable domain. For example, a heavy chain variable Hinton 2004). Binding to human FcRn in vivo and serum half domain may include a single amino acid insert (residue 52a life of human FcRn high affinity binding polypeptides can be according to Kabat) after residue 52 of H2 and inserted resi assayed, e.g., in transgenic mice or transfected human cell dues (e.g. residues 82a, 82b, and 82c, etc. according to Kabat) lines expressing human FcRn, or in primates administered after heavy chain FR residue 82. The Kabat numbering of with the Fc variant polypeptides. residues may be determined for a given antibody by align WO00/42072 (Presta) describes antibody variants with ment at regions of homology of the sequence of the antibody improved or diminished binding to FcRs. The content of that with a “standard Kabat numbered sequence. patent publication is specifically incorporated herein by ref An “affinity matured antibody is one with one or more 25 erence. See, also, Shields et al. J. Biol. Chem. 9(2): 6591 alterations in one or more HVRs thereof which result in an 6604 (2001). improvement in the affinity of the antibody for antigen, com “Human effector cells' are leukocytes which express one pared to a parent antibody which does not possess those or more FcRs and perform effector functions. In certain alteration(s). In one embodiment, an affinity matured anti embodiments, the cells express at least FcyRIII and perform body has nanomolar or even picomolar affinities for the target 30 ADCC effector function(s). Examples of human leukocytes antigen. Affinity matured antibodies are produced by proce which mediate ADCC include peripheral blood mononuclear dures known in the art. Marks et al. Bio/Technology 10:779 cells (PBMC), natural killer (NK) cells, monocytes, cytotoxic 783 (1992) describes affinity maturation by VH and VL T cells and neutrophils. The effector cells may be isolated domain shuffling. Random mutagenesis of HVR and/or from a native source, e.g., from blood. framework residues is described by: Barbas et al. Proc Nat. 35 Antibody-dependent cell-mediated cytotoxicity' or Acad. Sci. USA 91:3809-3813 (1994); Schier et al. Gene ADCC refers to a form of cytotoxicity in which secreted Ig 169: 147-155 (1995):Yeltonetal.J. Immunol. 155:1994-2004 bound onto Fc receptors (FcRs) present on certain cytotoxic (1995); Jackson et al., J. Immunol. 154(7):3310-9 (1995); and cells (e.g. Natural Killer (NK) cells, neutrophils, and mac Hawkins etal, J. Mol. Biol. 226:889-896 (1992). rophages) enable these cytotoxic effector cells to bind spe A “blocking antibody or an “antagonist' antibody is one 40 cifically to an antigen-bearing target cell and Subsequently which inhibits or reduces biological activity of the antigen it kill the target cell with cytotoxins. The primary cells for binds. Certain blocking antibodies or antagonist antibodies mediating ADCC, NK cells, express FcyRIII only, whereas substantially or completely inhibit the biological activity of monocytes express FcyRI, FcyRII and FcyRIII. FcR expres the antigen. sion on hematopoietic cells is Summarized in Table 3 on page An "agonistantibody, as used herein, is an antibody which 45 464 of Ravetch and Kinet, Annu. Rev. Immunol 9:457-92 mimics at least one of the functional activities of a polypep (1991). To assess ADCC activity of a molecule of interest, an tide of interest. in vitro ADCC assay, such as that described in U.S. Pat. No. Antibody “effector functions’ refer to those biological 5,500,362 or 5,821,337 or Presta U.S. Pat. No. 6,737,056 may activities attributable to the Fc region (a native sequence Fc be performed. Useful effector cells for such assays include region or amino acid sequence variant Fc region) of an anti 50 peripheral blood mononuclear cells (PBMC) and Natural body, and vary with the antibody isotype. Examples of anti Killer (NK) cells. Alternatively, or additionally, ADCC activ body effector functions include: C1q binding and comple ity of the molecule of interest may be assessed in Vivo, e.g., in ment dependent cytotoxicity; Fc receptor binding; antibody a animal model such as that disclosed in Clynes et al. PNAS dependent cell-mediated cytotoxicity (ADCC); (USA) 95:652-656 (1998). phagocytosis; down regulation of cell Surface receptors (e.g. 55 “Complement dependent cytotoxicity' or “CDC refers to B cell receptor); and B cell activation. the lysis of a target cell in the presence of complement. “Fc receptor” or “FcR” describes a receptor that binds to Activation of the classical complement pathway is initiated the Fc region of an antibody. In some embodiments, an FcRis by the binding of the first component of the complement a native human FcR. In some embodiments, an FcR is one system (C1q) to antibodies (of the appropriate Subclass) which binds an IgG antibody (a gamma receptor) and 60 which are bound to their cognate antigen. To assess comple includes receptors of the FcyRI. FcyRII, and FcyRIII sub ment activation, a CDC assay, e.g. as described in Gazzano classes, including allelic variants and alternatively spliced Santoro et al., J. Immunol. Methods 202:163 (1996), may be forms of those receptors. FeyRII receptors include FcyRIIA performed. (an “activating receptor) and FcyRIIB (an “inhibiting recep Polypeptide variants with altered Fc region amino acid tor”), which have similar amino acid sequences that differ 65 sequences and increased or decreased C1q binding capability primarily in the cytoplasmic domains thereof. Activating are described in U.S. Pat. No. 6,194.551B1 and WO99/ receptor FcyRIA contains an immunoreceptor tyrosine 51642. The contents of those patent publications are specifi US 8,771,966 B2 49 50 cally incorporated herein by reference. See, also, Idusogie et of interest through a cellular membrane, usually the inner al. J. Immunol. 164: 4178-4.184 (2000). membrane or bothinner and outer membranes of prokaryotes. The term “Fc region-comprising polypeptide' refers to a AS Such, the protein of interest Such as the immunoglobulin polypeptide. Such as an antibody or immunoadhesin, which light or heavy chain polypeptide is secreted into the periplasm comprises an Fc region. The C-terminal lysine (residue 447 of the prokaryotic host cells or into the culture medium. The according to the EU numbering system) of the Fc region may signal peptide encoded by the secretion signal sequence may be removed, for example, during purification of the polypep be endogenous to the host cells, or they may be exogenous, tide or by recombinant engineering the nucleic acid encoding including signal peptides native to the polypeptide to be the polypeptide. Accordingly, a composition comprising a expressed. Secretion signal sequences are typically present at polypeptide having an Fc region according to this invention 10 the amino terminus of a polypeptide to be expressed, and are can comprise polypeptides with K447, with all K447 typically removed enzymatically between biosynthesis and removed, or a mixture of polypeptides with and without the secretion of the polypeptide from the cytoplasm. Thus, the K447 residue. signal peptide is usually not present in a mature protein prod An “acceptor human framework” for the purposes herein is uct. a framework comprising the amino acid sequence of a VL or 15 A “free cysteine amino acid refers to a cysteine amino VH framework derived from a human immunoglobulin acid residue which has been engineered into a parent anti framework or a human consensus framework. An acceptor body, has a thiol functional group (-SH), and is not paired as, human framework “derived from a human immunoglobulin or otherwise part of an intramolecular or intermolecular dis framework or a human consensus framework may comprise ulfide bridge. the same amino acid sequence thereof, or it may contain The term “thiol reactivity value” is a quantitative charac pre-existing amino acid sequence changes. In some embodi terization of the reactivity of free cysteine amino acids. The ments, the number of pre-existing amino acid changes are 10 thiol reactivity value is the percentage of a free cysteine or less, 9 or less, 8 or less, 7 or less, 6 or less, 5 or less, 4 or less, amino acid in a cysteine engineered antibody which reacts 3 or less, or 2 or less. Where pre-existing amino acid changes with a thiol-reactive reagent, and converted to a maximum are present in a VH, preferably those changes occur at only 25 value of 1. For example, a free cysteine amino acid on a three, two, or one of positions 71H, 73H and 78H; for cysteine engineered antibody which reacts in 100% yield instance, the amino acid residues at those positions may be with a thiol-reactive reagent, such as a biotin-maleimide 71A, 73T and/or 78A. In one embodiment, the VL acceptor reagent, to form a biotin-labelled antibody has a thiol reac human framework is identical in sequence to the VL human tivity value of 1.0. Another cysteine amino acid engineered immunoglobulin framework sequence or human consensus 30 into the same or different parent antibody which reacts in 80% framework sequence. yield with a thiol-reactive reagent has a thiol reactivity value A "human consensus framework is a framework which of 0.8. Another cysteine amino acid engineered into the same represents the most commonly occurring amino acid residues or different parent antibody which fails totally to react with a in a selection of human immunoglobulin VL or VH frame thiol-reactive reagent has a thiol reactivity value of 0. Deter work sequences. Generally, the selection of human immuno 35 mination of the thiol reactivity value of a particular cysteine globulin VL or VH sequences is from a subgroup of variable may be conducted by ELISA assay, mass spectroscopy, liquid domain sequences. Generally, the Subgroup of sequences is a chromatography, autoradiography, or other quantitative ana Subgroup as in Kabat et al., Sequences of Proteins of Immu lytical tests. Thiol-reactive reagents which allow capture of nological Interest, 5" Ed. Public Health Service, National the cysteine engineered antibody and comparison and quan Institutes of Health, Bethesda, Md. (1991). In one embodi 40 titation of the cysteine reactivity include biotin-PEO-male ment, for the VL, the Subgroup is subgroup kappa I as in imide ((+)-biotinyl-3-maleimidopropionamidyl-3,6-diox Kabat et al., supra. In one embodiment, for the VH, the aoctainediamine, Oda et al (2001) Nature Biotechnology Subgroup is subgroup III as in Kabat et al., Supra. 19:379-382, Pierce Biotechnology, Inc.) Biotin-BMCC, A “VH subgroup III consensus framework' comprises the PEO-Iodoacetyl Biotin, Iodoacetyl-LC-Biotin, and Biotin consensus sequence obtained from the amino acid sequences 45 HPDP (Pierce Biotechnology, Inc.), and No.-(3-maleimidyl in variable heavy subgroup III of Kabat et al., supra. In one propionyl)biocytin (MPB, Molecular Probes, Eugene, embodiment, the VH subgroup III consensus framework Oreg.). Other commercial sources for biotinylation, bifunc amino acid sequence comprises at least a portion orall of each tional and multifunctional linker reagents include Molecular of the following sequences: EVOLVESGGGLVOPGGSL Probes, Eugene, Oreg., and Sigma, St. Louis, Mo. RLSCAAS (FR-1, SEQ ID NO:21)-HVR-H1 50 A "parent antibody' is an antibody comprising an amino WVRQAPGKGLEWV (FR-2, SEQ ID NO:22)-HVR-H2 acid sequence from which one or more amino acid residues RFTISRDTSKNTLYLQMNSLRAEDTAVYYCAR (FR-3, are replaced by one or more cysteine residues. The parent SEQ ID NO:23)-HVR-H3-WGQGTLVTVSS (FR-4, SEQ antibody may comprise a native or wild type sequence. The ID NO:24). parent antibody may have pre-existing amino acid sequence A “VL subgroup I consensus framework' comprises the 55 modifications (such as additions, deletions and/or substitu consensus sequence obtained from the amino acid sequences tions) relative to other native, wild type, or modified forms of in variable light kappa Subgroup I of Kabat et al., Supra. In one an antibody. A parent antibody may be directed against a embodiment, the VHSubgroup I consensus framework amino target antigen of interest, e.g. a biologically important acid sequence comprises at least a portion or all of each of the polypeptide. Antibodies directed against nonpolypeptide following sequences: DIQMTQSPSSLSASVGDRVTITC 60 antigens (such as tumor-associated glycolipid antigens; see (FR-L1, SEQ ID NO:17)-HVR-L1-WYQQKPGKAP U.S. Pat. No. 5,091,178) are also contemplated. KLLIY (FR-L2, SEQ ID NO:18)-HVR-L2-GVPSRF “Binding affinity” generally refers to the strength of the SGSGSGTDFTLTISSLQPEDFATYYC (FR-L3, SEQ ID Sum total of noncovalent interactions between a single bind NO:19)-HVR-L3-FGQGTKVEIKR (FR-L4, SEQ ID ing site of a molecule (e.g., an antibody) and its binding NO:20). “Secretion signal sequence' or “signal sequence' 65 partner (e.g., an antigen). Unless indicated otherwise, as used refers to a nucleic acid sequence encoding a short signal herein, “binding affinity” refers to intrinsic binding affinity peptide that can be used to direct a newly synthesized protein which reflects a 1:1 interaction between members of a bind US 8,771,966 B2 51 52 ing pair (e.g., antibody and antigen). The affinity of a mol rescence emission intensity (excitation 295 nmi; emis ecule X for its partner Y can generally be represented by the sion=340 nm, 16 nm band-pass) at 25°C. of a 20 nM anti dissociation constant (Kd). Affinity can be measured by com antigen antibody (Fab form) in PBS, pH 7.2, in the presence mon methods known in the art, including those described of increasing concentrations of antigen as measured in a herein. Low-affinity antibodies generally bind antigen slowly spectrometer. Such as a stop-flow equipped spectrophometer and tend to dissociate readily, whereas high-affinity antibod (Aviv Instruments) or a 8000-series SLM-Aminco spectro ies generally bind antigen faster and tend to remain bound photometer (ThermoSpectronic) with a stirred cuvette. longer. A variety of methods of measuring binding affinity are An “on-rate.” “rate of association,” “association rate,” or known in the art, any of which can be used for purposes of the “k according to this invention can also be determined as present invention. Specific illustrative embodiments are 10 described above using a BIAcoreTM-2000 or a BIAcoreTM described in the following. 3000 system (BIAcore(R), Inc., Piscataway, N.J.). In one embodiment, the “Kd' or “Kd value” according to A “disorder is any condition or disease that would benefit this invention is measured by a radiolabeled antigen binding from treatment with an substance/molecule or method of the assay (RIA) performed with the Fab version of an antibody of invention. This includes chronic and acute disorders includ interest and its antigen as described by the following assay. 15 ing those pathological conditions which predispose the mam Solution binding affinity of Fabs for antigen is measured by mal to the disorder in question. Non-limiting examples of equilibrating Fab with a minimal concentration of (125I)- disorders to be treated herein include cancerous conditions labeled antigen in the presence of a titration series of unla Such as cancers or metastases of prostate, lung, and colon. beled antigen, then capturing bound antigen with an anti-Fab The terms “cell proliferative disorder” and “proliferative antibody-coated plate (Chen, et al., (1999).J. Mol. Biol. 293: disorder” refer to disorders that are associated with some 865-881). To establish conditions for the assay, microtiter degree of abnormal cell proliferation. In one embodiment, the plates (Dynex) are coated overnight with 5ug/ml of a captur cell proliferative disorder is cancer. ing anti-Fab antibody (Cappel Labs) in 50 mM sodium car “Tumor, as used herein, refers to all neoplastic cell growth bonate (pH 9.6), and subsequently blocked with 2% (w/v) and proliferation, whether malignant or benign, and all pre bovine serum albumin in PBS for two to five hours at room 25 cancerous and cancerous cells and tissues. The terms 'can temperature (approximately 23°C.). In a non-adsorbent plate cer.” “cancerous.” “cell proliferative disorder,” “proliferative (Nunc #269620), 100 pM or 26 pM'I-antigen are mixed disorder and “tumor are not mutually exclusive as referred with serial dilutions of a Fab of interest (e.g., consistent with to herein. assessment of the anti-VEGF antibody, Fab-12, in Presta et The terms “cancer and "cancerous” refer to or describe the al., (1997) Cancer Res. 57:4593-4599). The Fab of interest is 30 physiological condition in mammals that is typically charac then incubated overnight; however, the incubation may con terized by unregulated cell growth. Examples of cancer tinue for a longer period (e.g., about 65 hours) to ensure that include, but are not limited to, carcinoma, lymphoma, blas equilibrium is reached. Thereafter, the mixtures are trans toma, sarcoma, and leukemia or lymphoid malignancies. ferred to the capture plate for incubation at room temperature More particular examples of such cancers include squamous (e.g., for one hour). The Solution is then removed and the plate 35 cell cancer (e.g. epithelial squamous cell cancer), lung cancer washed eight times with 0.1% Tween-20 in PBS. When the including Small-cell lung cancer, non-Small cell lung cancer, plates have dried, 150 ul/well of scintillant (MicroScint-20; adenocarcinoma of the lung and squamous carcinoma of the Packard) is added, and the plates are counted on a Topcount lung, cancer of the peritoneum, hepatocellular cancer, gastric gamma counter (Packard) for ten minutes. Concentrations of or stomach cancer including gastrointestinal cancer, pancre each Fab that give less than or equal to 20% of maximal 40 atic cancer, glioblastoma, cervical cancer, ovarian cancer, binding are chosen for use in competitive binding assays. liver cancer, bladder cancer, cancer of the urinary tract, According to another embodiment, the Kd or Kd value is hepatoma, breast cancer, colon cancer, rectal cancer, colorec measured by using Surface plasmon resonance assays using a tal cancer, endometrial or uterine carcinoma, salivary gland BIAcoreTM-2000 or a BIAcoreTM-3000 (BIAcore, Inc., Pis carcinoma, kidney or renal cancer, prostate cancer, Vulval cataway, N.J.) at 25°C. with immobilized antigen CM5 chips 45 cancer, thyroid cancer, hepatic carcinoma, anal carcinoma, at ~10 response units (RU). Briefly, carboxymethylated dex penile carcinoma, melanoma, multiple myeloma and B-cell tranbiosensor chips (CM5, BIAcore(RInc.) are activated with lymphoma, brain, as well as head and neck cancer, and asso N-ethyl-N'-(3-dimethylaminopropyl)-carbodiimide hydro ciated metastases. chloride (EDC) and N-hydroxysuccinimide (NHS) according A "STEAP-1-expressing cell is a cell which expresses to the supplier's instructions. Antigen is diluted with 10 mM 50 endogenous or transfected STEAP-1 on the cell surface. A sodium acetate, pH 4.8, to 5ug/ml (-0.2 uM) before injection "STEAP-1-expressing cancer is a cancer comprising cells at a flow rate of 5 ul/minute to achieve approximately 10 that have STEAP-1 protein present on the cell surface. A response units (RU) of coupled protein. Following the injec "STEAP-1-expressing cancer produces sufficient levels of tion of antigen, 1 M ethanolamine is injected to block unre STEAP-1 on the surface of cells thereof, such that an anti acted groups. For kinetics measurements, two-fold serial 55 STEAP-1 antibody can bind thereto and have a therapeutic dilutions of Fab (0.78 nM to 500 nM) are injected in PBS with effect with respect to the cancer. A cancer which “overex 0.05% Tween 20 (PBST) at 25° C. at a flow rate of approxi presses' STEAP-1 is one which has significantly higher lev mately 25 ul/min. Association rates (kon) and dissociation els of STEAP-1 at the cell surface thereof, compared to a rates (koff) are calculated using a simple one-to-one Lang noncancerous cell of the same tissue type. Such overexpres muir binding model (BIAcore(R) Evaluation Software version 60 sion may be caused by gene amplification or by increased 3.2) by simultaneously fitting the association and dissociation transcription or translation. STEAP-1 overexpression may be sensorgrams. The equilibrium dissociation constant (Kd) is determined in a diagnostic or prognostic assay by evaluating calculated as the ratio koff/kon. See, e.g., Chen, Y., et al., increased levels of the STEAP-1 protein present on the sur (1999).J. Mol. Biol. 293:865-881. If the on-rate exceeds 106 face of a cell (e.g. via an immunohistochemistry assay; FACS M-1 S-1 by the Surface plasmon resonance assay above, then 65 analysis). Alternatively, or additionally, one may measure the on-rate can be determined by using a fluorescent quench levels of STEAP-1-encoding nucleic acid or mRNA in the ing technique that measures the increase or decrease in fluo cell, e.g. via fluorescent in situ hybridization: (FISH; see US 8,771,966 B2 53 54 WO98/45479 published October, 1998), Southern blotting, molecule are outweighed by the therapeutically beneficial Northern blotting, or polymerase chain reaction (PCR) tech effects. A “prophylactically effective amount” refers to an niques, such as real time quantitative PCR (RT-PCR). One amount effective, at dosages and for periods of time neces may also study STEAP-1 overexpression by measuring shed sary, to achieve the desired prophylactic result. Typically, but antigen in a biological fluid Such as serum, e.g., using anti not necessarily, since a prophylactic dose is used in Subjects body-based assays (see also, e.g., U.S. Pat. No. 4,933,294 prior to or at an earlier stage of disease, the prophylactically issued Jun. 12, 1990; WO91/05264 published Apr. 18, 1991; effective amount would be less than the therapeutically effec U.S. Pat. No. 5,401,638 issued Mar. 28, 1995; and Sias et al. tive amount. In the case of cancer, the therapeutically effec J. Immunol. Methods 132: 73-80 (1990)). Aside from the tive amount of the drug may reduce the number of cancer above assays, various in vivo assays are available to the 10 skilled practitioner. For example, one may expose cells within cells; reduce the tumor size: inhibit (i.e., slow to some extent the body of the patient to an antibody which is optionally and preferably stop) cancer cell infiltration into peripheral labeled with a detectable label, e.g. a radioactive isotope, and organs; inhibit (i.e., slow to Some extent and preferably stop) binding of the antibody to cells in the patient can be evaluated, tumor metastasis; inhibit, to Some extent, tumor growth; and/ e.g. by external scanning for radioactivity or by analyzing a 15 or relieve to some extent one or more of the symptoms asso biopsy taken from a patient previously exposed to the anti ciated with the cancer. See preceding definition of “treating. body. A STEAP-1-expressing cancer includes prostate, lung, To the extent the drug may prevent growth and/or kill existing and colon cancer. cancer cells, it may be cytostatic and/or cytotoxic. As used herein, “treatment” (and variations such as “treat' “Chronic' administration refers to administration of the or “treating) refers to clinical intervention in an attempt to agent(s) in a continuous mode as opposed to an acute mode, alter the natural course of the individual or cell being treated, So as to maintain the initial therapeutic effect (activity) for an and can be performed either for prophylaxis or during the extended period of time. “Intermittent administration is course of clinical pathology. Desirable effects of treatment treatment that is not consecutively done without interruption, include preventing occurrence or recurrence of disease, alle but rather is cyclic in nature. viation of symptoms, diminishment of any direct or indirect 25 Administration “in combination with one or more further pathological consequences of the disease, preventing therapeutic agents includes simultaneous (concurrent) and metastasis, decreasing the rate of disease progression, ame consecutive administration in any order. lioration or palliation of the disease state, and remission or “Carriers' as used herein include pharmaceutically accept improved prognosis. In some embodiments, antibodies of the able carriers, excipients, or stabilizers which are nontoxic to invention are used to delay development of a disease or dis 30 the cell or mammal being exposed thereto at the dosages and order or to slow the progression of a disease or disorder. concentrations employed. Often the physiologically accept The above parameters for assessing successful treatment able carrier is an aqueous pH buffered solution. Examples of and improvement in the disease are readily measurable by physiologically acceptable carriers include buffers such as routine procedures familiar to a physician. For cancer phosphate, citrate, and other organic acids; antioxidants therapy, efficacy can be measured, for example, by assessing 35 including ascorbic acid; low molecular weight (less than the time to disease progression (TTP) and/or determining the about 10 residues) polypeptide; proteins, such as serum albu response rate (RR). For prostate cancer, the progress of min, gelatin, or immunoglobulins; hydrophilic polymers such therapy can be assessed by routine methods, usually by mea as polyvinylpyrrolidone; amino acids such as glycine, Suring serum PSA (prostate specific antigen) levels; the glutamine, asparagine, arginine or lysine; monosaccharides, higher the level of PSA in the blood, the more extensive the 40 disaccharides, and other carbohydrates including glucose, cancer. Commercial assays for detecting PSA are available, mannose, or dextrins; chelating agents such as EDTA: Sugar e.g., Hybitech Tandem-E and Tandem-R PSA assay kits, the alcohols such as mannitol or Sorbitol; salt-forming counteri Yang ProsCheck polyclonal assay (Yang Labs, Bellevue, ons such as sodium; and/or nonionic Surfactants such as Wash.), Abbott Imx (Abbott Labs, Abbott Park, Ill.), etc. TWEENTM, polyethylene glycol (PEG), and PLURON Metastasis can be determined by staging tests and by bone 45 ICSTM. scan and tests for calcium level and other enzymes to deter “Label as used herein refers to a detectable compound or mine spread to the bone. CT scans can also be done to look for composition which is conjugated directly or indirectly to the spread to the pelvis and lymph nodes in the area. Chest X-rays antibody so as to generate a “labeled antibody. The label may and measurement of liver enzyme levels by known methods be detectable by itself (e.g. radioisotope labels or fluorescent are used to look for metastasis to the lungs and liver, respec 50 labels) or, in the case of an enzymatic label, may catalyze tively. Other routine methods for monitoring the disease chemical alteration of a Substrate compound or composition include transrectal ultrasonography (TRUS) and transrectal which is detectable. needle biopsy (TRNB). The term “ImmunoPET is a term used for positron emis An “individual' is a vertebrate. In certain embodiments, sion tomography (PET) of radiolabelled antibodies including the vertebrate is a mammal. Mammals include, but are not 55 antibody drug conjugates such as anti-STEAP-1-vc-MMAE limited to, farm animals (such as cows), sport animals, pets ADC or thioMAb drug conjugates (TDC). The term “Zr (such as cats, dogs, and horses), primates, mice and rats. In STEAP-1 Immuno-PET refers to the Zirconium radiolabeled certain embodiments, a mammal is a human. PET imaging of STEAP-1. An “effective amount” refers to an amount effective, at The term “epitope tagged' used herein refers to a chimeric dosages and for periods of time necessary, to achieve the 60 polypeptide comprising an anti-PSCA antibody polypeptide desired therapeutic or prophylactic result. A “therapeutically fused to a "tagpolypeptide'. The tag polypeptide has enough effective amount of a substance/molecule of the invention residues to provide an epitope against which an antibody can may vary according to factors such as the disease State, age, be made, yet is short enough Such that it does not interfere sex, and weight of the individual, and the ability of the sub with activity of the Ig polypeptide to which it is fused. The tag stance/molecule, to elicit a desired response in the individual. 65 polypeptide is also preferably fairly unique so that the anti A therapeutically effective amount encompasses an amount body does not substantially cross-react with other epitopes. in which any toxic or detrimental effects of the substance/ Suitable tag polypeptides generally have at least six amino US 8,771,966 B2 55 56 acid residues and usually between about 8 and 50 amino acid include radioactive isotopes (e.g., At', I'', I',Y', Re", residues (preferably, between about 10 and 20 amino acid Re'. Sm', Bi', P, Pb and radioactive isotopes of residues). Lu), chemotherapeutic agents (e.g., methotrexate, adriami A “small molecule' is defined herein to have a molecular cin, Vinca alkaloids (Vincristine, vinblastine, etoposide), weight below about 500 Daltons. 5 doxorubicin, melphalan, mitomycin C, chlorambucil, dauno The term “package insert” is used to refer to instructions rubicin or other intercalating agents, enzymes and fragments customarily included in commercial packages of therapeutic thereof Such as nucleolytic enzymes, antibiotics, and toxins products, that contain information about the indications, Such as Small molecule toxins or enzymatically active toxins usage, dosage, administration, contraindications and/or of bacterial, fungal, plant or animal origin, including frag warnings concerning the use of Such therapeutic products. 10 ments and/or variants thereof, toxins, growth inhibitory An "isolated nucleic acid' is a nucleic acid, e.g., an RNA, agents, drug moieties, and the various antitumor oranticancer DNA, or a mixed polymer, which is substantially separated agents disclosed below. Other cytotoxic agents are described from other genome DNA sequences as well as proteins or below. A tumoricidal agent causes destruction of tumor cells. complexes such as ribosomes and polymerases, which natu A “toxin' is any Substance capable of having a detrimental rally accompany a native sequence. The term embraces a 15 effect on the growth or proliferation of a cell. nucleic acid sequence which has been removed from its natu A “chemotherapeutic agent' is a chemical compound use rally occurring environment, and includes recombinant or ful in the treatment of cancer. Examples of chemotherapeutic cloned DNA isolates and chemically synthesized analogues agents include alkylating agents such as thiotepa and or analogues biologically synthesized by heterologous sys CYTOXANR) cyclosphosphamide; alkyl sulfonates such as tems. A Substantially pure molecule includes isolated forms 20 buSulfan, improsulfan and piposulfan; aziridines Such as ben of the molecule. Zodopa, carboquone, meturedopa, and uredopa; ethylen “Vector includes shuttle and expression vectors. Typi imines and methylamelamines including altretamine, trieth cally, the plasmid construct will also include an origin of ylenemelamine, triethylenephosphoramide, replication (e.g., the ColE1 origin of replication) and a select triethylenethiophosphoramide and trimethylolomelamine; able marker (e.g., amplicillin or tetracycline resistance), for 25 acetogenins (especially bullatacin and bullatacinone); delta replication and selection, respectively, of the plasmids in 9-tetrahydrocannabinol (dronabinol, MARINOLR); beta bacteria. An "expression vector refers to a vector that con lapachone; lapachol; colchicines; betulinic acid; a camptoth tains the necessary control sequences or regulatory elements ecin (including the synthetic analogue topotecan for expression of the antibodies including antibody fragment (HYCAMTINR), CPT-11 (irinotecan, CAMPTOSARR), of the invention, in bacterial or eukaryotic cells. Suitable 30 acetylcamptothecin, Scopolectin, and 9-aminocamptoth vectors are disclosed below. ecin); bryostatin: cally statin: CC-1065 (including its adoze The cell that produces an anti-STEAP-1 antibody of the lesin, carzelesin and bizelesin synthetic analogues); podo invention will include the parent hybridoma cell e.g., the phyllotoxin: podophyllinic acid; teniposide; cryptophycins hybridomas that are deposited with the ATCC, as well as (particularly cryptophycin 1 and cryptophycin 8); dolastatin: bacterial and eukaryotic host cells into which nucleic acid 35 duocarmycin (including the synthetic analogues, KW-2189 encoding the antibodies have been introduced. Suitable host and CB1-TM1); eleutherobin: pancratistatin: a sarcodictyin: cells are disclosed below. spongistatin: nitrogen mustards such as chlorambucil, chlor A “growth inhibitory agent' when used herein refers to a naphazine, cholophosphamide, estramustine, ifosfamide, compound or composition which inhibits growth of a cell, mechlorethamine, mechlorethamine oxide hydrochloride, especially a PSCA expressing cancer cell, either in vitro or in 40 melphalan, novembichin, phenesterine, prednimustine, tro vivo. Thus, the growth inhibitory agent may be one which fosfamide, uracil mustard; nitrosoureas such as carmustine, significantly reduces the percentage of PSCA expressing chlorozotocin, fotemustine, lomustine, nimustine, and ranim cells in S phase. Examples of growth inhibitory agents nustine; antibiotics such as the enediyne antibiotics (e.g., include agents that block cell cycle progression (at a place calicheamicin, especially calicheamicin gammal I and cali other than S phase). Such as agents that induce G1 arrest and 45 cheamicin omegal (see, e.g., Agnew, Chem Intl. Ed. Engl. M-phase arrest. Classical M-phase blockers include the vin 33: 183-186 (1994)); dynemicin, including dynemicin A; an cas (Vincristine and vinblastine), taxanes, and topoisomerase esperamicin; as well as neocarzinostatin chromophore and II inhibitors such as doxorubicin, epirubicin, daunorubicin, related chromoprotein enediyne antiobiotic chromophores), etoposide, and bleomycin. Those agents that arrest G1 also aclacinomysins, actinomycin, authramycin, azaserine, bleo spill over into S-phase arrest, for example, DNA alkylating 50 mycins, cactinomycin, carabicin, caminomycin, carZinophi agents such as tamoxifen, prednisone, dacarbazine, mechlo lin, chromomycins, dactinomycin, daunorubicin, detorubi rethamine, cisplatin, methotrexate, 5-fluorouracil, and ara-C. cin, 6-diazo-5-oxo-L-norleucine, ADRIAMYCINCR) Further information can be found in The Molecular Basis of doxorubicin (including morpholino-doxorubicin, cyanomor Cancer, Mendelsohn and Israel, eds., Chapter 1, entitled pholino-doxorubicin, 2-pyrrolino-doxorubicin and deoxy “Cell cycle regulation, oncogenes, and antineoplastic drugs 55 doxorubicin), epirubicin, esorubicin, idarubicin, marcello by Murakami et al. (WB Saunders: Philadelphia, 1995), espe mycin, mitomycins such as mitomycin C, mycophenolic cially p. 13. The taxanes (paclitaxel and docetaxel) are anti acid, nogalamycin, olivomycins, peplomycin, porfiromycin, cancer drugs both derived from the yew tree. Docetaxel puromycin, quelamycin, rodorubicin, Streptonigrin, Strepto (TAXOTERE(R), Rhone-Poulenc Rorer), derived from the Zocin, tubercidin, ubenimex, Zinostatin, Zorubicin; anti-me European yew, is a semisynthetic analogue of paclitaxel 60 tabolites such as methotrexate and 5-fluorouracil (5-FU); (TAXOL(R), Bristol-Myers Squibb). Paclitaxel and docetaxel folic acid analogues such as denopterin, methotrexate, promote the assembly of microtubules from tubulin dimers pteropterin, trimetrexate; purine analogs such as fludarabine, and stabilize microtubules by preventing depolymerization, 6-mercaptopurine, thiamiprine, thioguanine; pyrimidine ana which results in the inhibition of mitosis in cells. logs such as ancitabine, azacitidine, 6-azauridine, carmofur, The term “cytotoxic agent” as used herein refers to a sub- 65 cytarabine, dideoxyuridine, doxifluridine, enocitabine, stance that inhibits or prevents a cellular function and/or floXuridine; androgens such as calusterone, dromostanolone causes cell death or destruction. The term is intended to propionate, epitioStanol, mepitiostane, testolactone; anti US 8,771,966 B2 57 58 adrenals such as aminoglutethimide, mitotane, triloStane; genes in signaling pathways implicated in abherant cell pro folic acid replenisher Such as frolinic acid; aceglatone; liferation, such as, for example, PKC-alpha, Raf, H-Ras, and aldophosphamide glycoside; aminolevulinic acid; eniluracil; epidermal growth factor receptor (EGF-R); vaccines such as amsacrine; bestrabucil; bisantrene; ediatraxate, defofamine; THERATOPE(R) vaccine and gene therapy vaccines, for demecolcine; diaziquone; elformithine; elliptinium acetate; 5 example, ALLOVECTINR vaccine, LEUVECTINR vac an epothilone; etoglucid: gallium nitrate; hydroxyurea; len cine, and VAXIDR vaccine; LURTOTECANR) topoi tinan; lonidainine; maytansinoids such as maytansine and somerase 1 inhibitor; ABARELIX(R) rmRH; lapatinib ditosy ansamitocins; mitoguaZone; mitoxantrone; mopidanmol; late (an ErbB-2 and EGFR dual tyrosine kinase small nitraerine; pentostatin: phenamet; pirarubicin; losoxantrone; molecule inhibitor also known as GW572016); and 2-ethylhydrazide; procarbazine; PSKR) polysaccharide com 10 pharmaceutically acceptable salts, acids or derivatives of any plex (JHS Natural Products, Eugene, Oreg.); razoxane: of the above. rhizoxin; sizofuran; Spirogermanium; tenuaZonic acid; triazi A “growth inhibitory agent' when used herein refers to a quone; 2.2.2"-trichlorotriethylamine; trichothecenes (espe compound or composition which inhibits growth of a cell cially T-2 toxin, Verracurin A, roridin A and anguidine); ure (such as a cell expressing STEAP-1) either in vitro or in vivo. than; vindesine (ELDISINE(R), FILDESINR); dacarbazine: 15 Thus, the growth inhibitory agent may be one which signifi mannomustine, mitobronitol; mitolactol; pipobroman, gacy cantly reduces the percentage of cells (such as a cell express tosine; arabinoside ("Ara-C); thiotepa; taxoids, e.g., ing STEAP-1) in S phase. Examples of growth inhibitory TAXOL(R) paclitaxel (Bristol-Myers Squibb Oncology, Prin agents include agents that block cell cycle progression (at a ceton, N.J.), ABRAXANETM Cremophor-free, albumin-en place other than S phase). Such as agents that induce G1 arrest gineered nanoparticle formulation of paclitaxel (American and M-phase arrest. Classical M-phase blockers include the Pharmaceutical Partners, Schaumberg, Ill.), and TAXO Vincas (Vincristine and vinblastine), taxanes, and topoi TERE(R) docetaxel (Rhone-Poulenc Rorer, Antony, France); Somerase II inhibitors such as doxorubicin, epirubicin, chloranbucil; gemcitabine (GEMZAR(R); 6-thioguanine: daunorubicin, etoposide, and bleomycin. Those agents that mercaptopurine; methotrexate; platinum analogs such as cis arrest G1 also spill over into S-phase arrest, for example, platin and carboplatin: vinblastine (VELBANR); platinum; 25 DNA alkylating agents such as tamoxifen, prednisone, dac etoposide (VP-16); ifosfamide; mitoxantrone; Vincristine arbazine, mechlorethamine, cisplatin, methotrexate, 5-fluo (ONCOVINR); oxaliplatin; leucovovin; vinorelbine (NA rouracil, and ara-C. Further information can be found in The VELBINE(R); novantrone; ediatrexate; daunomycin; aminop Molecular Basis of Cancer, Mendelsohn and Israel, eds., terin; ibandronate; topoisomerase inhibitor RFS 2000; dif Chapter 1, entitled "Cell cycle regulation, oncogenes, and luoromethylornithine (DMFO); retinoids such as retinoic 30 antineoplastic drugs” by Murakami et al. (WB Saunders: acid; capecitabine (XELODAR); pharmaceutically accept Philadelphia, 1995), especially p. 13. The taxanes (paclitaxel able salts, acids or derivatives of any of the above; as well as and docetaxel) are anticancer drugs both derived from the combinations of two or more of the above such as CHOP, an yew tree. Docetaxel (TAXOTERE(R), Rhone-Poulenc Rorer), abbreviation for a combined therapy of cyclophosphamide, derived from the European yew, is a semisynthetic analogue doxorubicin, Vincristine, and prednisolone, and FOLFOX, an 35 of paclitaxel (TAXOL(R), Bristol-Myers Squibb). Paclitaxel abbreviation for a treatment regimen with oxaliplatin and docetaxel promote the assembly of microtubules from (ELOXATINTM) combined with 5-FU and leucovovin. tubulin dimers and stabilize microtubules by preventing Also included in this definition are anti-hormonal agents depolymerization, which results in the inhibition of mitosis in that act to regulate, reduce, block, or inhibit the effects of cells. hormones that can promote the growth of cancer, and are 40 The term “intracellular metabolite' refers to a compound often in the form of systemic, or whole-body treatment. They resulting from a metabolic process or reaction inside a cell on may be hormones themselves. Examples include anti-estro an antibody-drug conjugate (ADC). The metabolic process or gens and selective estrogen receptor modulators (SERMs), reaction may be an enzymatic process, Such as proteolytic including, for example, tamoxifen (including NOLVADEXOR) cleavage of a peptide linker of the ADC, or hydrolysis of a tamoxifen), EVISTAR raloxifene, droloxifene, 4-hydroxyta 45 functional group Such as a hydraZone, ester, or amide. Intra moxifen, trioxifene, keoxifene, LY 1 17018, onapristone, and cellular metabolites include, but are not limited to, antibodies FARESTONR toremifene; anti-progesterones; estrogen and free drug which have undergone intracellular cleavage receptor down-regulators (ERDs); agents that function to after entry, diffusion, uptake or transport into a cell. Suppress or shut down the ovaries, for example, leutinizing The terms “intracellularly cleaved' and “intracellular hormone-releasing hormone (LHRH) agonists such as 50 cleavage” refer to a metabolic process or reaction inside a cell LUPRONR) and ELIGARDR) leuprolide acetate, goserelin on an antibody-drug conjugate (ADC) whereby the covalent acetate, buserelin acetate and tripterelin; other anti-andro attachment, i.e. linker, between the drug moiety (D) and the gens such as flutamide, nilutamide and bicalutamide; and antibody (Ab) is broken, resulting in the free drug dissociated aromatase inhibitors that inhibit the enzyme aromatase, from the antibody inside the cell. The cleaved moieties of the which regulates estrogen production in the adrenal glands, 55 ADC are thus intracellular metabolites. Such as, for example, 4(5)-imidazoles, aminoglutethimide, The term “bioavailability” refers to the systemic availabil MEGASE(R) megestrol acetate, AROMASINR exemestane, ity (i.e., blood/plasma levels) of a given amount of drug formestanie, fadrozole, RIVISOR(R) vorozole, FEMARAR administered to a patient. Bioavailability is an absolute term letrozole, and ARIMIDEXR anastrozole. In addition, such that indicates measurement of both the time (rate) and total definition of chemotherapeutic agents includes bisphospho 60 amount (extent) of drug that reaches the general circulation nates such as clodronate (for example, BONEFOSR or from an administered dosage form. OSTAC(R), DIDROCAL(R) etidronate, NE-58095, The term “cytotoxic activity” refers to a cell-killing, cyto ZOMETAR) Zoledronic acid/Zoledronate, FOSAMAX(R) static or growth inhibitory effect of an antibody-drug conju alendronate, AREDIAR pamidronate, SKELID(R) tiludr gate or an intracellular metabolite of an antibody-drug con onate, or ACTONELR risedronate; as well as troxacitabine (a 65 jugate. Cytotoxic activity may be expressed as the ICso value, 1,3-dioxolane nucleoside cytosine analog); antisense oligo which is the concentration (molar or mass) per unit volume at nucleotides, particularly those that inhibit expression of which half the cells survive. US 8,771,966 B2 59 60 Alkyl is C1-C18 hydrocarbon containing normal, sec include, but are not limited to: methylene (-CH ) 1.2- ondary, tertiary or cyclic carbon atoms. Examples are methyl ethyl (-CHCH ), 1,3-propyl (—CH2CHCH ), 1,4- (Me, —CH3), ethyl (Et, —CH2CH3), 1-propyl (n-Pr, n-pro butyl ( CHCH2CHCH ), and the like. pyl, —CH2CH2CH3), 2-propyl (i-Pr. i-propyl, —CH A "C-Clo alkylene' is a straight chain, Saturated hydro (CH3)), 1-butyl (n-Bu, n-butyl, -CH2CH2CH2CH3), carbon group of the formula —(CH) . Examples of a 2-methyl-1-propyl (i-Bu, i-butyl, —CH2CH(CH3)), 2-butyl C-Co alkylene include methylene, ethylene, propylene, (s-Bu, s-butyl, —CH(CH3)CH2CH3), 2-methyl-2-propyl butylene, pentylene, hexylene, heptylene, ocytylene, non (t-Bu, t-butyl, —C(CH3)), 1-pentyl (n-pentyl, ylene and decalene. CH2CH2CH2CH2CH3), 2-pentyl ( CH(CH3) Alkenylene' refers to an unsaturated, branched or straight CH2CH2CH3), 3-pentyl (-CH(CH2CH3)), 2-methyl-2- 10 butyl ( C(CH3)CH2CH3), 3-methyl-2-butyl (-CH(CH3) chain or cyclic hydrocarbon radical of 2-18 carbonatoms, and CH(CH3)), 3-methyl-1-butyl ( CH2CH2CH(CH3)), having two monovalent radical centers derived by the 2-methyl-1-butyl (-CH2CH(CH3)CH2CH3), 1-hexyl removal of two hydrogen atoms from the same or two differ ( CH2CH2CH2CH2CH2CH3), 2-hexyl ( CH(CH3) ent carbonatoms of a parent alkene. Typical alkenylene radi CH2CH2CH2CH3), 3-hexyl ( CH(CH2CH3) 15 cals include, but are not limited to: 1.2-ethylene (CH2CH2CH3)), 2-methyl-2-pentyl ( C(CH3), (-CH=CH-). CH2CH2CH3), 3-methyl-2-pentyl ( CH(CH3)CH(CH3) Alkynylene' refers to an unsaturated, branched or straight CH2CH3), 4-methyl-2-pentyl ( CH(CH3)CH2CH(CH3) chain or cyclic hydrocarbon radical of 2-18 carbonatoms, and 2), 3-methyl-3-pentyl ( C(CH3)(CH2CH3)2), 2-methyl-3- having two monovalent radical centers derived by the pentyl (-CH(CH2CH3)CH(CH3)2), 2,3-dimethyl-2-butyl removal of two hydrogen atoms from the same or two differ ( C(CH3)2CH(CH3)2),3,3-dimethyl-2-butyl (-CH(CH3) ent carbonatoms of a parent alkyne. Typical alkynylene radi C(CH3)3. cals include, but are not limited to: acetylene ( -C=C- ), The term “C1-C8 alkyl, as used herein refers to a straight propargyl (—CH2C=C ), and 4-pentynyl chain or branched, Saturated or unsaturated hydrocarbon hav (-CHCHCHC=C ). ing from 1 to 8 carbon atoms. Representative “C-C alkyl 25 Aryl refers to a carbocyclic aromatic group. Examples of groups include, but are not limited to, -methyl, -ethyl, -n- aryl groups include, but are not limited to, phenyl, naphthyl propyl, -n-butyl, -n-pentyl, -n-hexyl, -n-heptyl, -n-octyl, -n- and anthracenyl. A carbocyclic aromatic group or a hetero nonyl and -n-decyl; while branched C-C alkyls include, but cyclic aromatic group can be unsubstituted or Substituted are not limited to, -isopropyl, -sec-butyl, -isobutyl, -tert-bu with one or more groups including, but not limited to. —C- tyl, -isopentyl, 2-methylbutyl, unsaturated C-C alkyls 30 include, but are not limited to, -vinyl, -allyl, -1-butenyl, Cs alkyl, —O (C-C alkyl), -aryl, —C(O)R', —OC(O)R', -2-butenyl, -isobutylenyl, -1-pentenyl, -2-pentenyl, -3-me —C(O)CR', C(O)NH –C(O)NHR'. —C(O)N(R)- thyl-1-butenyl, -2-methyl-2-butenyl, -2,3-dimethyl-2-bute NHC(O)R', S(O).R', S(O)R', -OH, -halogen, N., nyl, 1-hexyl, 2-hexyl, 3-hexyl, -acetylenyl, -propynyl, -1-bu - NH, -NH(R'), N(R'), and –CN; wherein each R" is tynyl, -2-butynyl, -1-pentynyl, -2-pentynyl, -3-methyl-1 35 independently selected from H. —C-Cs alkyl and aryl. butynyl. methyl, ethyl, propyl, isopropyl. n-butyl, isobutyl, An “arylene' is an aryl group which has two covalent sec-butyl, tert-butyl, n-pentyl, isopentyl, neopentyl, n-hexyl, bonds and can be in the ortho, meta, or para configurations as isohexyl, 2-methylpentyl, 3-methylpenty1, 2,2-dimethylbu shown in the following structures: tyl, 2,3-dimethylbutyl, 2,2-dimethylpenty1, 2,3-dimethylpen tyl, 3.3-dimethylpenty1, 2,3,4-trimethylpentyl, 3-methyl 40 hexyl, 2,2-dimethylhexyl, 2,4-dimethylhexyl, 2.5- dimethylhexyl, 3,5-dimethylhexyl, 2,4-dimethylpentyl, 2-methylheptyl, 3-methylheptyl, n-heptyl, isoheptyl, n-octyl, and isooctyl. A C-C alkyl group can be unsubstituted or Substituted with one or more groups including, but not limited 45 OHC to. —C-Cs alkyl, —O—(C-C alkyl), -aryl, —C(O)R', –OC(O)R', C(O)OR, C(O)NH, C(O)NHR', C(O)N(R'). NHC(O)R', SOR', S(O).R', S(O)R', —OH, -halogen, N, NH, -NH(R'), N(R') and —CN; where each R" is independently selected from H. —C- 50 Cs alkyl and aryl. Alkenyl is C2-C18 hydrocarbon containing normal, sec ondary, tertiary or cyclic carbonatoms with at least one site of in which the phenyl group can be unsubstituted or Substi unsaturation, i.e. a carbon-carbon, sp' double bond. tuted with up to four groups including, but not limited to, Examples include, but are not limited to: ethylene or vinyl 55 —C-Cs alkyl, —O—(C-C alkyl), -aryl, —C(O)R'. - OC (—CH=CH), allyl ( CH-CH=CH-), cyclopentenyl (O)R', C(O)OR', C(O)NH, C(O)NHR', C(O) (—CH7), and 5-hexenyl ( CHCHCHCH-CH=CH-) N(R') NHC(O)R', S(O)R', S(O)R', -OH,-halogen, Alkynyl is C2-C18 hydrocarbon containing normal, sec —N, NH, NH(R'), —N(R') and —CN; wherein each ondary, tertiary or cyclic carbonatoms with at least one site of R" is independently selected from H. —C-C alkyl and aryl. unsaturation, i.e. a carbon-carbon, sp triple bond. Examples 60 Arylalkyl refers to an acyclic alkyl radical in which one include, but are not limited to: acetylenic ( -C=CH) and of the hydrogen atoms bonded to a carbon atom, typically a propargyl ( CH-C=CH), terminal or sp carbon atom, is replaced with an aryl radical. Alkylene' refers to a saturated, branched or straight chain Typical arylalkyl groups include, but are not limited to, ben or cyclic hydrocarbon radical of 1-18 carbon atoms, and Zyl. 2-phenylethan-1-yl, 2-phenylethen-1-yl, naphthylm having two monovalent radical centers derived by the 65 ethyl 2-naphthylethan-1-yl, 2-naphthylethen-1-yl, naph removal of two hydrogen atoms from the same or two differ thobenzyl, 2-naphthophenylethan-1-yl and the like. The ent carbonatoms of a parent alkane. Typical alkylene radicals arylalkyl group comprises 6 to 20 carbonatoms, e.g. the alkyl US 8,771,966 B2 61 62 moiety, including alkanyl, alkenyl or alkynyl groups, of the indolizinyl, isoindolyl, 3H-indolyl, 1H-indazolyl, purinyl, arylalkyl group is 1 to 6 carbon atoms and the aryl moiety is 4H-quinolizinyl, phthalazinyl, naphthyridinyl, quinoxalinyl, 5 to 14 carbon atoms. quinazolinyl, cinnolinyl, pteridinyl, 4aH-carbazolyl, carba “Heteroarylalkyl refers to an acyclic alkyl radical in Zolyl, B-carbolinyl, phenanthridinyl, acridinyl, pyrimidinyl, which one of the hydrogen atoms bonded to a carbon atom, 5 phenanthrolinyl, phenazinyl, phenothiazinyl, furazanyl, phe typically a terminal or sp carbon atom, is replaced with a noxazinyl, isochromanyl, chromanyl, imidazolidinyl, imida heteroaryl radical. Typical heteroarylalkyl groups include, Zolinyl, pyrazolidinyl, pyrazolinyl, piperazinyl, indolinyl, but are not limited to, 2-benzimidazolylmethyl 2-furylethyl, isoindolinyl, quinuclidinyl, morpholinyl, oxazolidinyl, ben and the like. The heteroarylalkyl group comprises 6 to 20 Zotriazolyl, benzisoxazolyl, OXindolyl, benzoxazolinyl, and carbon atoms, e.g. the alkyl moiety, including alkanyl, alk 10 isatinoyl. enyl or alkynyl groups, of the heteroarylalkyl group is 1 to 6 By way of example and not limitation, carbon bonded carbon atoms and the heteroaryl moiety is 5 to 14 carbon heterocycles are bonded at position 2, 3, 4, 5, or 6 of a atoms and 1 to 3 heteroatoms selected from N, O, P, and S. pyridine, position 3, 4, 5, or 6 of a pyridazine, position 2, 4, 5, The heteroaryl moiety of the heteroarylalkyl group may be a or 6 of a pyrimidine, position 2, 3, 5, or 6 of a pyrazine, monocycle having 3 to 7 ring members (2 to 6 carbon atoms 15 position 2, 3, 4, or 5 of a furan, tetrahydrofuran, thiofuran, or a bicycle having 7 to 10 ring members (4 to 9 carbonatoms thiophene, pyrrole or tetrahydropyrrole, position 2, 4, or 5 of and 1 to 3 heteroatoms selected from N, O, P, and S), for an oxazole, imidazole or thiazole, position 3, 4, or 5 of an example: a bicyclo 4.5, 5.5, 5.6, or 6.6 system. isoxazole, pyrazole, or isothiazole, position 2 or 3 of an “Substituted alkyl,” “substituted aryl, and “substituted aziridine, position 2, 3, or 4 of anaZetidine, position 2, 3, 4, 5, arylalkyl mean alkyl, aryl, and arylalkyl respectively, in 6, 7, or 8 of a quinoline or position 1, 3, 4, 5, 6, 7, or 8 of an which one or more hydrogen atoms are each independently isoquinoline. Still more typically, carbon bonded hetero replaced with a substituent. Typical substituents include, but cycles include 2-pyridyl, 3-pyridyl, 4-pyridyl, 5-pyridyl, are not limited to, —X, —R, —O, —OR, —SR, —S, 6-pyridyl, 3-pyridazinyl, 4-pyridaZinyl, 5-pyridazinyl, 6-py —NR - NR =NR, CX - CN, OCN, - SCN, ridazinyl, 2-pyrimidinyl, 4-pyrimidinyl, 5-pyrimidinyl, 6-py -N=C=O, NCS, NO, NO, —N, N NC(=O) 25 rimidinyl, 2-pyrazinyl, 3-pyrazinyl, 5-pyrazinyl, 6-pyrazinyl, R, C(=O)R, C(=O)NR, -SO, SOH, 2-thiazolyl, 4-thiazolyl, or 5-thiazolyl. —S(=O).R, OS(=O),OR, S(=O)NR, -S(=O)R, By way of example and not limitation, nitrogen bonded –OP(=O)(OR), -P(=O)(OR), PO, POH, heterocycles are bonded at position 1 of an aziridine, aZeti —C(=O)R, —C(=O)x, —C(=S)R, COR, CO., dine, pyrrole, pyrrolidine, 2-pyrroline, 3-pyrroline, imida —C(=S)OR, C(=O)SR, C(=S)SR, C(=O)NR, 30 Zole, imidazolidine, 2-imidazoline, 3-imidazoline, pyrazole, —C(=S)NR, —C(=NR)NR, where each X is indepen pyrazoline, 2-pyrazoline, 3-pyrazoline, piperidine, pipera dently a halogen: F, Cl, Br, or I; and each R is independently zine, indole, indoline, 1H-indazole, position 2 of a isoindole, —H, C-C alkyl, C-C aryl, C-C heterocycle, protect or isoindoline, position 4 of a morpholine, and position 9 of a ing group or prodrug moiety. Alkylene, alkenylene, and alky carbazole, or 3-carboline. Still more typically, nitrogen nylene groups as described above may also be similarly Sub 35 bonded heterocycles include 1-aziridyl, 1-azetedyl, 1-pyrro stituted. lyl, 1-imidazolyl, 1-pyrazolyl, and 1-piperidinyl. “Heteroaryl and "heterocycle” refer to a ring system in A "C-Cs heterocycle” refers to an aromatic or non-aro which one or more ring atoms is a heteroatom, e.g. nitrogen, matic C-Cs carbocycle in which one to four of the ring oxygen, and Sulfur. The heterocycle radical comprises 1 to 20 carbon atoms are independently replaced with a heteroatom carbon atoms and 1 to 3 heteroatoms selected from N, O, P, 40 from the group consisting of O, S and N. Representative and S. A heterocycle may be a monocycle having 3 to 7 ring examples of a C-Cs heterocycle include, but are not limited members (2 to 6 carbonatoms and 1 to 3 heteroatoms selected to, benzofuranyl, benzothiophene, indolyl, benzopyrazolyl, from N, O, P, and S) or a bicycle having 7 to 10 ring members coumarinyl, isoquinolinyl, pyrrolyl, thiophenyl, furanyl, thia (4 to 9 carbon atoms and 1 to 3 heteroatoms selected from N. Zolyl, imidazolyl pyrazolyl, triazolyl, quinolinyl, pyrimidi O, P, and S), for example: a bicyclo 4.5, 5.5, 5.6, or 6.6 45 nyl, pyridinyl, pyridonyl, pyrazinyl, pyridazinyl, isothiazolyl, system. isoxazolyl and tetrazolyl. A C-Cs heterocycle can be unsub Heterocycles are described in Paquette, Leo A.; “Principles stituted or Substituted with up to seven groups including, but of Modern Heterocyclic Chemistry” (W.A. Benjamin, New not limited to. —C-Cs alkyl, —O—(C-C alkyl), -aryl, York, 1968), particularly Chapters 1, 3, 4, 6, 7, and 9; “The C(O)R', OC(O)R', C(O)OR, C(O)NH, C(O) Chemistry of Heterocyclic Compounds. A series of Mono 50 NHR', C(O)N(R'). NHC(O)R', S(O).R', S(O)R', graphs” (John Wiley & Sons, New York, 1950 to present), in —OH, -halogen, —N, NH, NH(R'), —N(R') and particular Volumes 13, 14, 16, 19, and 28; and J. Am. Chem. —CN; wherein each R" is independently selected from H. Soc. (1960) 82:5566. —C-Cs alkyl and aryl. Examples of heterocycles include by way of example and "C-Cs heterocyclo” refers to a C-Cs heterocycle group not limitation pyridyl, dihydroypyridyl, tetrahydropyridyl 55 defined above wherein one of the heterocycle group’s hydro (piperidyl), thiazolyl, tetrahydrothiophenyl, sulfur oxidized gen atoms is replaced with a bond. A C-C heterocyclo can tetrahydrothiophenyl, pyrimidinyl, furanyl, thienyl, pyrrolyl, be unsubstituted or substituted with up to six groups includ pyrazolyl, imidazolyl, tetrazolyl, benzofuranyl, thianaphtha ing, but not limited to. —C-Cs alkyl, —O—(C-Cs alkyl). lenyl, indolyl, indolenyl, quinolinyl, isoquinolinyl, benzimi -aryl, -C(O)R', OC(O)R', C(O)OR', C(O)NH2, dazolyl, piperidinyl, 4-piperidonyl, pyrrolidinyl, 2-pyrrolido 60 C(O)NHR', C(O)N(R'). NHC(O)R', S(O).R', nyl, pyrrolinyl, tetrahydrofuranyl, bis-tetrahydrofuranyl, —S(O)R', —OH, -halogen, N, NH, -NH(R'), tetrahydropyranyl, bis-tetrahydropyranyl, tetrahydroquinoli —N(R') and —CN; wherein each R" is independently nyl, tetrahydroisoquinolinyl, decahydroquinolinyl, octahy selected from H. —C-C alkyl and aryl. droisoquinolinyl, azocinyl, triazinyl, 6H-1,2,5-thiadiazinyl, “Carbocycle” means a saturated or unsaturated ring having 2H,6H-1.5.2-dithiazinyl, thienyl, thianthrenyl, pyranyl, 65 3 to 7 carbon atoms as a monocycle or 7 to 12 carbon atoms isobenzofuranyl, chromenyl, Xanthenyl, phenoxathinyl, as a bicycle. Monocyclic carbocycles have 3 to 6 ring atoms, 2H-pyrrolyl, isothiazolyl, isoxazolyl pyrazinyl, pyridazinyl, still more typically 5 or 6 ring atoms. Bicyclic carbocycles US 8,771,966 B2 63 64 have 7 to 12 ring atoms, e.g. arranged as a bicyclo 4.5, 5.5, mixture of enantiomers is referred to as a racemic mixture or 5.6 or 6.6 system, or 9 or 10 ring atoms arranged as a a racemate, which may occur where there has been no stereo bicyclo5.6 or 6.6 system. Examples of monocyclic car selection or stereospecificity in a chemical reaction or pro bocycles include cyclopropyl, cyclobutyl, cyclopentyl, 1-cy cess. The terms “racemic mixture' and “racemate” refer to an clopent-1-enyl, 1-cyclopent-2-enyl, 1-cyclopent-3-enyl, equimolar mixture of two enantiomeric species, devoid of cyclohexyl, 1-cyclohex-1-enyl, 1-cyclohex-2-enyl, 1-cyclo optical activity. hex-3-enyl, cycloheptyl, and cyclooctyl. "Leaving group' refers to a functional group that can be A "C-C carbocycle' is a 3-, 4-, 5-, 6-, 7- or 8-membered Substituted by another functional group. Certain leaving saturated or unsaturated non-aromatic carbocyclic ring. Rep groups are well known in the art, and examples include, but resentative C-Cs carbocycles include, but are not limited to, 10 are not limited to, a halide (e.g., chloride, bromide, iodide), -cyclopropyl, -cyclobutyl, -cyclopentyl, -cyclopentadienyl, methanesulfonyl (mesyl), p-toluenesulfonyl (tosyl), trifluo -cyclohexyl, -cyclohexenyl, -1,3-cyclohexadienyl, -1,4-cy romethylsulfonyl (triflate), and trifluoromethylsulfonate. clohexadienyl, -cycloheptyl, -1,3-cycloheptadienyl, -1,3,5- cycloheptatrienyl, -cyclooctyl, and -cyclooctadienyl. A Abbreviations C-Cs carbocycle group can be unsubstituted or Substituted 15 with one or more groups including, but not limited to. —C- Linker Components: Cs alkyl, —O—(C-C alkyl), -aryl, —C(O)R', —OC(O)R', MC-6-maleimidocaproyl C(O)CR', C(O)NH, C(O)NHR', C(O)N(R') Val-Cit or “vc'-valine-citrulline (an exemplary dipeptide NHC(O)R', S(O).R', S(O)R', -OH, -halogen, N. in a protease cleavable linker) - NH, -NH(R'), N(R') and –CN; where each R' is Citrulline-2-amino-5-ureido pentanoic acid independently selected from H. —C-C alkyl and aryl. PAB p-aminobenzyloxycarbonyl (an example of a “self A "C-Cs carbocyclo” refers to a C-Cs carbocycle group immolative' linker component) defined above wherein one of the carbocycle groups hydro Me-Val-Cit=N-methyl-valine-citrulline (wherein the gen atoms is replaced with a bond. linker peptide bond has been modified to prevent its cleavage "Linker” refers to a chemical moiety comprising a covalent 25 by cathepsin B) bond or a chain of atoms that covalently attaches an antibody MC(PEG)6-OH-maleimidocaproyl-polyethylene glycol to a drug moiety. In various embodiments, linkers include a (can be attached to antibody cysteines). divalent radical Such as an alkyldiyl, an aryldiyl, a het SPP=N-succinimidyl-4-(2-pyridylthio)pentanoate eroaryldiyl, moieties such as: —(CR), O(CR), , repeating SPDP-N-succinimidyl-3-(2-pyridyldithio)propionate units of alkyloxy (e.g. polyethylenoxy, PEG, polymethyl 30 SMCC=Succinimidyl-4-(N-maleimidomethyl)cyclohex eneoxy) and alkylamino (e.g. polyethyleneamino, Jeffam ane-1-carboxylate ineTM); and diacid ester and amides including succinate, suc IT-iminothiolane cinamide, diglycolate, malonate, and caproamide. Cytotoxic Drugs: The term "chiral refers to molecules which have the prop MMAE-mono-methyl auristatin E (MW 718) erty of non-Superimposability of the mirror image partner, 35 MMAF-variant of auristatin E (MMAE) with a phenyla while the term “achiral” refers to molecules which are super lanine at the C-terminus of the drug (MW 731.5) imposable on their mirror image partner. MMAF-DMAEA=MMAF with DMAEA (dimethylami The term “stereoisomers' refers to compounds which have noethylamine) in an amide linkage to the C-terminal pheny identical chemical constitution, but differ with regard to the lalanine (MW 801.5) arrangement of the atoms or groups in space. 40 MMAF-TEG=MMAF with tetraethylene glycol esterified “Diastereomer' refers to a stereoisomer with two or more to the phenylalanine centers of chirality and whose molecules are not mirror MMAF-NtEu=N-t-butyl, attached as an amide to C-termi images of one another. Diastereomers have different physical nus of MMAF properties, e.g. melting points, boiling points, spectral prop DM1=N(2)-deacetyl-N(2)-(3-mercapto-1-oxopropyl)- erties, and reactivities. Mixtures of diastereomers may sepa 45 maytansine rate under high resolution analytical procedures such as elec DM3-N(2)-deacetyl-N2-(4-mercapto-1-oxopentyl)- trophoresis and chromatography. maytansine "Enantiomers' refer to two stereoisomers of a compound DM4=N(2)-deacetyl-N2-(4-mercapto-4-methyl-1-oxo which are non-Superimposable mirror images of one another. pentyl)-maytansine Stereochemical definitions and conventions used herein 50 Further abbreviations are as follows: AE is auristatin E, generally follow S.P. Parker, Ed., McGraw-Hill Dictionary of Boc is N-(t-butoxycarbonyl), cit is citrulline, dap is dola Chemical Terms (1984) McGraw-Hill Book Company, New proine, DCC is 1,3-dicyclohexylcarbodiimide, DCM is York; and Eliel, E. and Wilen, S., Stereochemistry of Organic dichloromethane, DEA is diethylamine, DEAD is diethyla Compounds (1994) John Wiley & Sons, Inc., New York. Zodicarboxylate, DEPC is diethylphosphorylcyanidate, Many organic compounds exist in optically active forms, i.e., 55 DIAD is diisopropylazodicarboxylate, DIEA is N,N-diiso they have the ability to rotate the plane of plane-polarized propylethylamine, dil is dolaisoleucine, DMA is dimethylac light. In describing an optically active compound, the prefixes etamide, DMAP is 4-dimethylaminopyridine, DME is ethyl D and L, or R and S. are used to denote the absolute configu eneglycol dimethyl ether (or 1,2-dimethoxyethane), DMF is ration of the molecule about its chiral center(s). The prefixes N,N-dimethylformamide, DMSO is dimethylsulfoxide, doe d and 1 or (+) and (-) are employed to designate the sign of 60 is dollaphenine, dov is N,N-dimethylvaline, DTNB is 5,5'- rotation of plane-polarized light by the compound, with (-) or dithiobis(2-nitrobenzoic acid), DTPA is diethylenetriamine 1 meaning that the compound is levorotatory. A compound pentaacetic acid, DTT is dithiothreitol, EDCI is 1-(3-dim prefixed with (+) or d is dextrorotatory. For a given chemical ethylaminopropyl)-3-ethylcarbodiimide hydrochloride, structure, these stereoisomers are identical except that they EEDQ is 2-ethoxy-1-ethoxycarbonyl-1,2-dihydroquinoline, are mirror images of one another. A specific stereoisomer may 65 ES-MS is electrospray mass spectrometry, EtOAc is ethyl also be referred to as an enantiomer, and a mixture of Such acetate, Fmoc is N-(9-fluorenylmethoxycarbonyl), gly is gly isomers is often called an enantiomeric mixture. A 50:50 cine, HATU is O-(7-azabenzotriazol-1-yl)-N.N.N',N'-tet US 8,771,966 B2 65 66 ramethyluronium hexafluorophosphate, HOBt is 1-hydroxy to STEAP-1 are provided. Monoclonal antibodies that bind to benzotriazole, HPLC is high pressure liquid chromatography, the same epitope as murine 120.545, 120 graft, and human ile is isoleucine, lys is lysine, MeCN(CHCN) is acetonitrile, ized 120V.24 are also provided. MeOH is methanol. Mtr is 4-anisyldiphenylmethyl (or In one aspect of the invention, polynucleotides encoding 4-methoxytrity1), nor is (1S,2R)-(+)-norephedrine, PBS is anti-STEAP-1 antibodies are provided. In certain embodi phosphate-buffered saline (pH 7.4), PEG is polyethylene gly ments, vectors comprising polynucleotides encoding anti col, Phis phenyl, Pnp is p-nitrophenyl, MC is 6-maleimidoca STEAP-1 antibodies are provided. In certain embodiments, proyl, phe is L-phenylalanine, PyBrop is bromo tris-pyrroli host cells comprising Such vectors are provided. In another dino phosphonium hexafluorophosphate, SEC is size aspect of the invention, compositions comprising anti exclusion chromatography, Su is succinimide, TFA is 10 STEAP-1 antibodies or polynucleotides encoding anti trifluoroacetic acid, TLC is thin layer chromatography, UV is STEAP-1 antibodies are provided. In certain embodiments, a ultraviolet, and val is valine. composition of the invention is a pharmaceutical formulation for the treatment of a cell proliferative disorder, such as those COMPOSITIONS AND METHODS OF MAKING enumerated herein. THE SAME 15 A detailed description of exemplary anti-STEAP-1 anti bodies is as follows: Antibodies that bind to STEAP-1 are provided. Immuno 1. Specific Embodiments of Anti-STEAP-1 Antibodies conjugates comprising anti-STEAP-1 antibodies are pro In one aspect, the invention provides an anti-STEAP-1 vided. Antibodies and immunoconjugates of the invention are antibody comprising a heavy chain variable domain compris useful, e.g., for the diagnosis or treatment of disorders asso ing SEQ ID NO:9 or 10 of FIG. 2B. In one aspect, the ciated with altered expression, e.g., increased expression, of invention provides an anti-STEAP-1 antibody comprising a STEAP-1. In certain embodiments, antibodies or immuno light chain variable domain comprising SEQID NO:6 of FIG. conjugates of the invention are useful for the diagnosis or 2A. treatment of a cell proliferative disorder. Such as cancer. In one aspect, the invention provides an anti-STEAP-1 Anti-STEAP-1 Antibodies 25 antibody comprising a heavy chain comprising SEQ ID In one aspect, the invention provides antibodies that bind to NO:9, having one or more of the following amino changes at STEAP-1. In some embodiments, antibodies are provided the indicated Kabat position: A24V, V37I, V48M, F67I, and that bind to a mature form of human and cynomolgus monkey L78F. In one embodiment, the heavy chain comprises a heavy (cyno) STEAP-1. In one such embodiment, a mature form of chain framework region selected from SEQID NOS:25, 75, human STEAP-1 has an amino acid sequence of SEQ ID 30 76, 77, 78, and 79. As used herein, heavy chain framework NO:1 (FIG. 1). The cyno STEAP-1 has an amino acid regions are designated “FR-1-H4” or “HC-FR1-FR4,” and sequence of SEQIDNO:3 (FIG.1). In some embodiments, an light chain framework regions are designated "FR-L1-L4” or antibody to STEAP-1 binds to a mature form of STEAP-1 “LC-FR1-FR4. In one aspect, the invention provides an anti expressed on the cell Surface. In some embodiments, an anti STEAP-1 antibody comprising a light chain comprising SEQ body that binds to a mature form of STEAP-1 expressed on 35 ID NO:6. the cell surface inhibits the growth of the cell. In some In one aspect, the invention provides an anti-STEAP-1 embodiments, an anti-STEAP-1 antibody binds to a mature antibody comprising 1, 2, 3, 4, 5, or 6 of the HVR sequences form of STEAP-1 expressed on the cell surface and inhibits of the antibody 120.v24 shown in FIGS. 2A and 2B. cell proliferation. In certain embodiments, an anti-STEAP-1 An anti-STEAP-1 antibody may comprise any suitable antibody binds to a mature form of STEAP-1 expressed on the 40 framework variable domain sequence, provided that the anti cell Surface and induces cell death. In some embodiments, an body retains the ability to bind STEAP-1. For example, in anti-STEAP-1 antibody binds to a mature form of STEAP-1 some embodiments, anti-STEAP-1 antibodies of the inven expressed on the Surface of cancer cells. In some embodi tion comprise a human subgroup III heavy chain framework ments, an anti-STEAP-1 antibody binds to a mature form of consensus sequence. In one embodiment of these antibodies, STEAP-1 that is overexpressed on the surface of cancer cells 45 the heavy chain framework consensus sequence comprises relative to normal cells of the same tissue origin. In some substitution(s) at position 24, 37, 48, 67, and/or 78. In one embodiments, an anti-STEAP-1 antibody is conjugated to a embodiment of these antibodies, position 24 is A or V. posi cytotoxin or a detectable label and binds to STEAP-1 on a cell tion 37 is I or V. position 48 is M or V. position 67 is I or F. Surface. In some embodiments, the antibody-toxin conjugate and/or position 78 is F or L. In one embodiment, these anti inhibits growth of the cell. In some embodiments, the anti 50 bodies comprise a heavy chain variable domain framework body-detectable label conjugate causes a cell expressing sequence of huMAb4D5-8, e.g., SEQID NO:21, 22, 23, and STEAP-1 on its surface to be detectable in vitro or in vivo. 24 (FR-1, FR-2, FR-3, FR-4, respectively). huMAb4D5-8 is In one aspect, an anti-STEAP-1 antibody is a monoclonal commercially known as HERCEPTINR) anti-HER2 anti antibody. In one aspect, an anti-STEAP-1 antibody is an body, Genentech, Inc., South San Francisco, Calif., USA; antibody fragment, e.g., a Fab, Fab'-SH, Fv, sclv, or (Fab') 55 also referred to in U.S. Pat. Nos. 6,407,213 & 5,821,337, and fragment. In one aspect, an anti-STEAP-1 antibody is a chi Lee et al., J. Mol. Biol. (2004), 340(5):1073-93. In one such meric, humanized, or human antibody. In one aspect, any of embodiment, these antibodies further comprise a human KI the anti-STEAP-1 antibodies described herein are purified. light chain framework consensus sequence. In one Such Exemplary monoclonal antibodies derived from a phage embodiment, these antibodies comprise a light chain variable library are provided herein. The antigen used for screening 60 domain framework sequence of huMAb4D5-8, e.g. SEQID the library was a polypeptide having the sequence of amino NO:17, 18, 19, and 20 (FR-L1, FR-L2, FR-L3, FR-L4, acid sequences of SEQID NO:28 or SEQID NO:30, corre respectively). sponding to the extracellular domains (ECDs) of STEAP-1 In one embodiment, an anti-STEAP-1 antibody comprises beta and alpha. The antibodies resulting from the library a heavy chain variable domain comprising a framework screen are affinity matured. 65 sequence and hyperVariable regions, wherein the framework In one aspect, monoclonal antibodies that compete with sequence comprises the FR-1-FR-4 sequences SEQ ID murine 120.545, 120 graft, and humanized 120V.24 binding NO:21 or 25 (FR-1), 22 (FR-2), 23 (FR-3), and 24 (FR-4), US 8,771,966 B2 67 68 respectively; the HVRH1 comprises the amino acid sequence In some embodiments, the light chain HVR sequences of SEQ ID NO:14; the HVR-H2 comprises the amino acid comprise the following: sequence of SEQID NO:15; and the HVR-H3 comprises an amino acid sequence of SEQID NO:16. In one embodiment, an anti-STEAP-1 antibody comprises a light chain variable HVR-L1 (KSSQSLLYRSNOKNYLA, SEQ ID NO: 11) domain comprising a framework sequence and hyperVariable HVR-L2 (WASTRES, SEQ ID NO: 12) regions, wherein the framework sequence comprises the FR-L1-FR-L4 sequences of SEQID NOS:17, 18, 19 and 20, HVR-L3 (OOYYNYPRT, SEO ID NO : 13). respectively; the HVR-L1 comprises the amino acid sequence In some embodiments, the light chain FR sequences com selected from SEQID NOS:11, 12, and 13. In one embodi 10 prise the following: ment of these antibodies, the heavy chain variable domain comprises SEQID NOs:9 or 10 and the light chain variable domain comprises SEQID NO:6. FR-L1 (DIOMTOSPSSLSASVGDRVTITC, SEO ID NO : 17); In some embodiments, the invention provides an anti STEAP-1 antibody comprising a heavy chain variable 15 FR-L2 (WYOOKPGKAPKLLIY, SEQ ID NO: 18); domain comprising an amino acid sequence having at least FR-L3 (GVPSRFSGSGSGTDFTLTISSLOPEDFATYYC, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% SEQ ID NO : 19) sequence identity to an amino acid sequence SEQID NO:9 or 10. In some embodiments, an amino acid sequence having at FR-L4 (FGOGTKVEIKR, SEQ ID NO: 2O). least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity contains Substitutions, insertions, or In one aspect, the invention provides an anti-STEAP-1 deletions relative to the reference sequence, but an antibody antibody comprising a light chain variable domain compris comprising that amino acid sequence retains the ability to ing an amino acid sequence having at least 90%, 91%, 92%, bind to STEAP-1. In some embodiments, a total of 1 to 10 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity amino acids have been Substituted, inserted, or deleted in a 25 to an amino acid sequence SEQID NO:6. In some embodi sequence SEQID NOs:9,10,14, 15, 16, 21, 22, 23, 24, 25.75, ments, an amino acid sequence having at least 90%, 91%, 76, 77, 78, and/or 79. In some embodiments, the substitu 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence tions, insertions, or deletions occur in regions outside the identity contains substitutions, additions, or deletions relative HVRs (i.e., in the FRs). In some embodiments, an anti to the reference sequence, but an antibody comprising that STEAP-1 antibody comprises a heavy chain variable domain 30 amino acid sequence retains the ability to bind to STEAP-1. comprising an amino acid sequence selected from SEQ ID In some embodiments, a total of 1 to 10 amino acids have NO:9 or 10. been substituted, inserted, or deleted in a sequence selected In some embodiments, the invention provides an anti from SEQID NOS: 6, 11, 12, 13, 17, 18, 19, and 20. In some STEAP-1 antibody comprising a heavy chain variable embodiments, the Substitutions, insertions, or deletions occur domain as depicted in FIG. 2B (SEQID NOs:9 or 10). 35 in regions outside the HVRs (i.e., in the FRs). In some In some embodiments, the heavy chain HVR and FR embodiments, an anti-STEAP-1 antibody comprises a light sequences comprise the following: chain variable domain comprising the amino acid sequence SEQID NO:6. In one aspect, the invention provides an anti-STEAP-1 HWR-H1 (GYSITSDYAWN, SEO ID NO : 14) 40 antibody comprising (a) a heavy chain variable domain com prising an amino acid sequence having at least 90%, 91%, HWR-H2 (GYISNSGSTSYNPSLKS, SEO ID NO: 15) 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity to an amino acid sequence selected from SEQ ID HWR-H3 (ERNYDYDDYYYAMDY SEO ID NO: 16) NO:9 and 10; and (b) a light chain variable domain compris FR-H1 (EVOLVESGGGLVOPGGSLRLSCAAS, 45 ing an amino acid sequence having at least 90%, 91%, 92%, SEQ ID NO: 21) 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence identity FR-H1 (EVOLVESGGGLVOPGGSLRLSCAVS, to an amino acid sequence SEQID NO:6. In some embodi SEQ ID NO: 25) ments, an amino acid sequence having at least 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% sequence FR-H2 (WWROAPGKGLEWW, SEQ ID NO: 22) 50 identity contains substitutions, additions, or deletions relative to the reference sequence, but an antibody comprising that FR-H2 (WIROAPGKGLEWW, SEO ID NO : 75) amino acid sequence retains the ability to bind to STEAP-1. FR-H2 (WVROAPGKGLEWM, SEO ID NO : 76) In some embodiments, a total of 1 to 10 amino acids have been substituted, inserted, or deleted in the reference FR-H2 (WIROAPGKGLEWM, SEO ID NO : 77) 55 sequence, including but not limited to a sequence selected FR-H3 (RFTISRDNSKNTLYLOMNSLRAEDTAVYYCAR, from SEQID NOs:9, 10, 14, 15, 16, 21, 22, 23, 24, 25, 75,76, SEQ ID NO: 23) 77, 78, 79. In some embodiments, the substitutions, inser tions, or deletions occur in regions outside the HVRS (i.e., in FR-H3 (RITISRDNSKNTLYLOMNSLRAEDTAVYYCAR, SEO ID NO: 78) the FRs). In some embodiments, an anti-STEAP-1 antibody 60 comprises a heavy chain variable domain comprising an FR-H3 (RFTISRDNSKNTFYLOMNSLRAEDTAVYYCAR, amino acid sequence of SEQID NO:9 or 10 and a light chain SEO ID NO: 79) variable domain comprising an amino acid sequence selected from SEQID NO:6. FR-H4 (WGOGTLWTVSS, SEQ ID NO: 24) In one aspect, the invention provides an anti-STEAP-1 In some embodiments, the invention provides an anti 65 antibody comprising (a) one, two, or three VHHVRs selected STEAP-1 antibody comprising a light chain variable domain from those shown in FIG. 2B and/or (b) one, two, or three VL as depicted in FIG. 2A (SEQID NO:6). HVRs selected from those shown in FIG. 2A. In one aspect, US 8,771,966 B2 69 70 the invention provides an anti-STEAP-1 antibody comprising The choice of human variable domains, both light and a heavy chain variable domain selected from those shown in heavy, to be used in making the humanized antibodies can be FIG. 2B and a light chain variable domain selected from those important to reduce antigenicity. According to the so-called shown in FIG. 2A. “best-fit’ method, the sequence of the variable domain of a 2. Antibody Fragments rodent antibody is screened against the entire library of The present invention encompasses antibody fragments. known human variable-domain sequences. The human Antibody fragments may be generated by traditional means, sequence which is closest to that of the rodent is then accepted Such as enzymatic digestion, or by recombinant techniques. as the human framework for the humanized antibody (Sims et In certain circumstances there are advantages of using anti al. (1993).J. Immunol. 151:2296; Chothia et al. (1987).J. Mol. 10 Biol. 196:901. Another method uses a particular framework body fragments, rather than whole antibodies. The smaller derived from the consensus sequence of all human antibodies size of the fragments allows for rapid clearance, and may lead of a particular Subgroup of light or heavy chains. The same to improved access to solid tumors. For a review of certain antibody fragments, see Hudson et al. (2003) Nat. Med. framework may be used for several different humanized anti 9:129-134. bodies (Carter et al. (1992) Proc. Natl. Acad. Sci. USA, 15 89:4285; Presta et al. (1993).J. Immunol., 151:2623. Various techniques have been developed for the production It is further generally desirable that antibodies be human of antibody fragments. Traditionally, these fragments were ized with retention of high affinity for the antigen and other derived via proteolytic digestion of intact antibodies (see, favorable biological properties. To achieve this goal, accord e.g., Morimoto et al., Journal of Biochemical and Biophysical ing to one method, humanized antibodies are prepared by a Methods 24:107-117 (1992); and Brennan et al., Science, process of analysis of the parental sequences and various 229:81 (1985)). However, these fragments can now be pro conceptual humanized products using three-dimensional duced directly by recombinant host cells. Fab, Fv and ScHv models of the parental and humanized sequences. Three antibody fragments can all be expressed in and secreted from dimensional immunoglobulin models are commonly avail E. coli, thus allowing the facile production of large amounts able and are familiar to those skilled in the art. Computer of these fragments. Antibody fragments can be isolated from 25 programs are available which illustrate and display probable the antibody phage libraries discussed above. Alternatively, three-dimensional conformational structures of selected can Fab'-SH fragments can be directly recovered from E. coli and didate immunoglobulin sequences. Inspection of these dis chemically coupled to form F(ab')2 fragments (Carter et al., plays permits analysis of the likely role of the residues in the Bio/Technology 10:163-167 (1992)). According to another functioning of the candidate immunoglobulin sequence, i.e., approach, F(ab') fragments can be isolated directly from 30 the analysis of residues that influence the ability of the can recombinant host cell culture. Fab and F(ab') fragment with didate immunoglobulin to bind its antigen. In this way, FR increased in vivo half-life comprising salvage receptor bind residues can be selected and combined from the recipient and ing epitope residues are described in U.S. Pat. No. 5,869,046. import sequences so that the desired antibody characteristic, Other techniques for the production of antibody fragments Such as increased affinity for the target antigen?s), is achieved. will be apparent to the skilled practitioner. In certain embodi 35 In general, the hyperVariable region residues are directly and ments, an antibody is a single chain Fv fragment (ScFV). See most Substantially involved in influencing antigen binding. WO 93/16185; U.S. Pat. Nos. 5,571,894; and 5,587,458. Fv 4. Human Antibodies and Schvare the only species with intact combining sites that Human anti-STEAP-1 antibodies of the invention can be are devoid of constant regions; thus, they may be suitable for constructed by combining Fv clone variable domain reduced nonspecific binding during in vivo use. ScFv fusion 40 sequence(s) selected from human-derived phage display proteins may be constructed to yield fusion of an effector libraries with known human constant domain sequences(s) as protein at either the amino or the carboxy terminus of an schv. described above. Alternatively, human monoclonal anti See Antibody Engineering, ed. Borrebaeck, Supra. The anti STEAP-1 antibodies of the invention can be made by the body fragment may also be a "linear antibody', e.g., as hybridoma method. Human myeloma and mouse-human het described in U.S. Pat. No. 5,641,870, for example. Such 45 eromyeloma cell lines for the production of human mono linear antibodies may be monospecific or bispecific. clonal antibodies have been described, for example, by Koz 3. Humanized Antibodies bor J. Immunol., 133: 3001 (1984); Brodeur et al., The invention encompasses humanized antibodies. Various Monoclonal Antibody Production Techniques and Applica methods for humanizing non-human antibodies are known in tions, pp. 51-63 (Marcel Dekker, Inc., New York, 1987); and the art. For example, a humanized antibody can have one or 50 Boerner et al., J. Immunol., 147: 86 (1991). more amino acid residues introduced into it from a source It is now possible to produce transgenic animals (e.g. mice) which is non-human. These non-human amino acid residues that are capable, upon immunization, of producing a full are often referred to as “import residues, which are typically repertoire of human antibodies in the absence of endogenous taken from an “import' variable domain. Humanization can immunoglobulin production. For example, it has been be essentially performed following the method of Winter and 55 described that the homozygous deletion of the antibody co-workers (Jones et al. (1986) Nature 321:522-525; Riech heavy-chain joining region (JH) gene in chimeric and germ mann et al. (1988) Nature 332:323-327; Verhoeyen et al. line mutant mice results in complete inhibition of endogenous (1988) Science 239:1534-1536), by substituting hypervari antibody production. Transfer of the human germ-line immu able region sequences for the corresponding sequences of a noglobulin gene array in Such germ-line mutant mice will human antibody. Accordingly, Such "humanized' antibodies 60 result in the production of human antibodies upon antigen are chimeric antibodies (U.S. Pat. No. 4,816,567) wherein challenge. See, e.g., Jakobovits et al., Proc. Natl. Acad. Sci Substantially less than an intact human variable domain has USA, 90: 2551 (1993); Jakobovits et al., Nature, 362: 255 been Substituted by the corresponding sequence from a non (1993); Bruggermann et al., Year in Immunol. 7: 33 (1993). human species. In practice, humanized antibodies are typi Gene shuffling can also be used to derive human antibodies cally human antibodies in which some hyperVariable region 65 from non-human, e.g. rodent, antibodies, where the human residues and possibly some FR residues are substituted by antibody has similar affinities and specificities to the starting residues from analogous sites in rodent antibodies. non-human antibody. According to this method, which is also US 8,771,966 B2 71 72 called "epitope imprinting, either the heavy or light chain In one embodiment of this approach, the bispecific anti variable region of a non-human antibody fragment obtained bodies are composed of a hybrid immunoglobulin heavy by phage display techniques as described herein is replaced chain with a first binding specificity in one arm, and a hybrid with a repertoire of human V domain genes, creating a popu immunoglobulin heavy chain-light chain pair (providing a lation of non-human chain/human chain ScFv or Fab chime second binding specificity) in the other arm. It was found that ras. Selection with antigen results in isolation of a non-human this asymmetric structure facilitates the separation of the chain/human chain chimeric scEv or Fab wherein the human desired bispecific compound from unwanted immunoglobu chain restores the antigen binding site destroyed upon lin chain combinations, as the presence of an immunoglobu removal of the corresponding non-human chain in the pri lin light chain in only one half of the bispecific molecule mary phage display clone, i.e. the epitope governs (imprints) 10 provides for a facile way of separation. This approach is disclosed in WO 94/04690. For further details of generating the choice of the human chain partner. When the process is bispecific antibodies see, for example, Suresh et al., Methods repeated in order to replace the remaining non-human chain, in Enzymology, 121:210 (1986). a human antibody is obtained (see PCTWO 93/06213 pub According to another approach, the interface between a lished Apr. 1, 1993). Unlike traditional humanization of non 15 pair of antibody molecules can be engineered to maximize the human antibodies by CDR grafting, this technique provides percentage of heterodimers which are recovered from recom completely human antibodies, which have no FR or CDR binant cell culture. The interface comprises at least a part of residues of non-human origin. the C3 domain of an antibody constant domain. In this 5. Bispecific Antibodies method, one or more Small amino acid side chains from the Bispecific antibodies are monoclonal antibodies that have interface of the first antibody molecule are replaced with binding specificities for at least two different antigens. In larger side chains (e.g. tyrosine or tryptophan). Compensa certain embodiments, bispecific antibodies are human or tory "cavities of identical or similar size to the large side humanized antibodies. In certain embodiments, one of the chain(s) are created on the interface of the second antibody binding specificities is for STEAP-1 and the other is for any molecule by replacing large amino acid side chains with other antigen. In certain embodiments, bispecific antibodies 25 Smaller ones (e.g. alanine or threonine). This provides a may bind to two different epitopes of STEAP-1. Bispecific mechanism for increasing the yield of the heterodimer over antibodies may also be used to localize cytotoxic agents to other unwanted end-products such as homodimers. cells which express STEAP-1. These antibodies possess a Bispecific antibodies include cross-linked or "heterocon STEAP-1-binding arm and an arm which binds a cytotoxic jugate' antibodies. For example, one of the antibodies in the agent, Such as, e.g., Saporin, anti-interferon-C. Vinca alkaloid, 30 heteroconjugate can be coupled to avidin, the other to biotin. ricin A chain, methotrexate or radioactive isotope hapten. Such antibodies have, for example, been proposed to target Bispecific antibodies can be prepared as full length antibodies immune system cells to unwanted cells (U.S. Pat. No. 4,676, or antibody fragments (e.g. F(ab') bispecific antibodies). 980), and for treatment of HIV infection (WO 91/00360, WO Methods for making bispecific antibodies are known in the 92/00373, and EP 03089). Heteroconjugate antibodies may art. Traditionally, the recombinant production of bispecific 35 be made using any convenient cross-linking method. Suitable antibodies is based on the co-expression of two immunoglo cross-linking agents are well known in the art, and are dis bulin heavy chain-light chain pairs, where the two heavy closed in U.S. Pat. No. 4,676.980, along with a number of chains have different specificities (Milstein and Cuello, cross-linking techniques. Nature, 305:537 (1983)). Because of the random assortment Techniques for generating bispecific antibodies from anti of immunoglobulin heavy and light chains, these hybridomas 40 body fragments have also been described in the literature. For (quadromas) produce a potential mixture of 10 different anti example, bispecific antibodies can be prepared using chemi body molecules, of which only one has the correct bispecific cal linkage. Brennan et al., Science, 229: 81 (1985) describe structure. The purification of the correct molecule, which is a procedure wherein intact antibodies are proteolytically usually done by affinity chromatography steps, is rather cum cleaved to generate F(ab')2 fragments. These fragments are bersome, and the product yields are low. Similar procedures 45 reduced in the presence of the dithiol complexing agent are disclosed in WO93/08829 published May 13, 1993, and sodium arsenite to stabilize vicinal dithiols and prevent inter in Traunecker et al., EMBO.J., 10: 3655 (1991). molecular disulfide formation. The Fab' fragments generated According to a different approach, antibody variable are then converted to thionitrobenzoate (TNB) derivatives. domains with the desired binding specificities (antibody-an One of the Fab'-TNB derivatives is then reconverted to the tigen combining sites) are fused to immunoglobulin constant 50 Fab'-thiol by reduction with mercaptoethylamine and is domain sequences. The fusion, for example, is with an immu mixed with an equimolar amount of the other Fab'-TNB noglobulin heavy chain constant domain, comprising at least derivative to form the bispecific antibody. The bispecific anti part of the hinge, CH2, and CH3 regions. In certain embodi bodies produced can be used as agents for the selective immo ments, the first heavy-chain constant region (CH1), contain bilization of enzymes. ing the site necessary for light chain binding, is present in at 55 Recent progress has facilitated the direct recovery of Fab'- least one of the fusions. DNAs encoding the immunoglobulin SH fragments from E. coli, which can be chemically coupled heavy chainfusions and, if desired, the immunoglobulin light to form bispecific antibodies. Shalaby et al., J. Exp. Med., chain, are inserted into separate expression vectors, and are 175: 217-225 (1992) describe the production of a fully co-transfected into a Suitable host organism. This provides for humanized bispecific antibody F(ab') molecule. Each Fab' great flexibility in adjusting the mutual proportions of the 60 fragment was separately secreted from E. coli and Subjected three polypeptide fragments in embodiments when unequal to directed chemical coupling in vitro to form the bispecific ratios of the three polypeptide chains used in the construction antibody. The bispecific antibody thus formed was able to provide the optimum yields. It is, however, possible to insert bind to cells overexpressing the HER2 receptor and normal the coding sequences for two or all three polypeptide chains human T cells, as well as trigger the lytic activity of human in one expression vector when the expression of at least two 65 cytotoxic lymphocytes against human breast tumor targets. polypeptide chains in equal ratios results in high yields or Various techniques for making and isolating bispecific when the ratios are of no particular significance. antibody fragments directly from recombinant cell culture US 8,771,966 B2 73 74 have also been described. For example, bispecific antibodies (Domantis, Inc., Waltham, Mass.: see, e.g., U.S. Pat. No. have been produced using leucine Zippers. Kostelny et al., J. 6.248,516 B1). In one embodiment, a single-domain antibody Immunol., 148(5):1547-1553 (1992). The leucine zipper pep consists of all or a portion of the heavy chain variable domain tides from the Fos and Jun proteins were linked to the Fab' of an antibody. portions of two different antibodies by gene fusion. The anti 8. Antibody Variants body homodimers were reduced at the hinge region to form In some embodiments, amino acid sequence monomers and then re-oxidized to form the antibody het modification(s) of the antibodies described herein are con erodimers. This method can also be utilized for the produc templated. For example, it may be desirable to improve the tion of antibody homodimers. The “diabody’ technology binding affinity and/or other biological properties of the anti described by Hollinger et al., Proc. Natl. Acad. Sci. USA, 10 body. Amino acid sequence variants of the antibody may be 90:6444-6448 (1993) has provided an alternative mechanism prepared by introducing appropriate changes into the nucle for making bispecific antibody fragments. The fragments otide sequence encoding the antibody, or by peptide synthe comprise a heavy-chain variable domain (VH) connected to a sis. Such modifications include, for example, deletions from, light-chain variable domain (VL) by a linker which is too and/or insertions into and/or Substitutions of residues within short to allow pairing between the two domains on the same 15 the amino acid sequences of the antibody. Any combination chain. Accordingly, the VH and VL domains of one fragment of deletion, insertion, and Substitution can be made to arrive at are forced to pair with the complementary VL and VH the final construct, provided that the final construct possesses domains of another fragment, thereby forming two antigen the desired characteristics. The amino acid alterations may be binding sites. Another strategy for making bispecificantibody introduced in the Subject antibody amino acid sequence at the fragments by the use of single-chain FV (SFV) dimers has also time that sequence is made. been reported. See Gruber et al., J. Immunol. 152:5368 A useful method for identification of certain residues or (1994). regions of the antibody that are preferred locations for Antibodies with more than two valencies are contem mutagenesis is called 'alanine Scanning mutagenesis' as plated. For example, trispecific antibodies can be prepared. described by Cunningham and Wells (1989) Science, 244: Tutt et al. J. Immunol. 147: 60 (1991). 25 1081-1085. Here, a residue or group of target residues are 6. Multivalent Antibodies identified (e.g., charged residues such as arg, asp, his, lys, and A multivalent antibody may be internalized (and/or catabo glu) and replaced by a neutral or negatively charged amino lized) faster than a bivalent antibody by a cell expressing an acid (e.g., alanine or polyalanine) to affect the interaction of antigen to which the antibodies bind. The antibodies of the the amino acids with antigen. Those amino acid locations present invention can be multivalent antibodies (which are 30 demonstrating functional sensitivity to the Substitutions then other than of the IgM class) with three or more antigen bind are refined by introducing further or other variants at, or for, ing sites (e.g. tetravalent antibodies), which can be readily the sites of substitution. Thus, while the site for introducing produced by recombinant expression of nucleic acid encod anamino acid sequence variation is predetermined, the nature ing the polypeptide chains of the antibody. The multivalent of the mutation per se need not be predetermined. For antibody can comprise a dimerization domain and three or 35 example, to analyze the performance of a mutation at a given more antigen binding sites. In certain embodiments, the site, ala Scanning or random mutagenesis is conducted at the dimerization domain comprises (or consists of) an Fc region target codon or region and the expressed immunoglobulins or a hinge region. In this scenario, the antibody will comprise are screened for the desired activity. an Fc region and three or more antigen binding sites amino Amino acid sequence insertions include amino- and/or terminal to the Fc region. In certain embodiments, a multiva 40 carboxyl-terminal fusions ranging in length from one residue lent antibody comprises (or consists of) three to about eight to polypeptides containing a hundred or more residues, as antigen binding sites. In one Such embodiment, a multivalent well as intrasequence insertions of single or multiple amino antibody comprises (or consists of) four antigenbinding sites. acid residues. Examples of terminal insertions include an The multivalent antibody comprises at least one polypeptide antibody with an N-terminal methionyl residue. Other inser chain (for example, two polypeptide chains), wherein the 45 tional variants of the antibody molecule include the fusion to polypeptide chain(s) comprise two or more variable domains. the N- or C-terminus of the antibody to an enzyme (e.g. for For instance, the polypeptide chain(s) may comprise VD1 ADEPT) or a polypeptide which increases the serum half-life (X1)n-VD2-(X2)n-Fc, wherein VD1 is a first variable of the antibody. domain, VD2 is a second variable domain, Fc is one polypep In certain embodiments, an antibody of the invention is tide chain of an Fc region, X1 and X2 representanamino acid 50 altered to increase or decrease the extent to which the anti or polypeptide, and n is 0 or 1. For instance, the polypeptide body is glycosylated. Glycosylation of polypeptides is typi chain(s) may comprise: VH-CH1-flexible linker-VH-CH1 cally either N-linked or O-linked. N-linked refers to the Fc region chain; or VH-CH1-VH-CH1-Fc region chain. The attachment of a carbohydrate moiety to the side chain of an multivalent antibody herein may further comprise at least two asparagine residue. The tripeptide sequences asparagine-X- (for example, four) light chain variable domain polypeptides. 55 serine and asparagine-X-threonine, where X is any amino The multivalent antibody herein may, for instance, comprise acid except proline, are the recognition sequences for enzy from about two to about eight light chain variable domain matic attachment of the carbohydrate moiety to the aspar polypeptides. The light chain variable domain polypeptides agine side chain. Thus, the presence of either of these tripep contemplated here comprise a light chain variable domain tide sequences in a polypeptide creates a potential and, optionally, further comprise a CL domain. 60 glycosylation site. O-linked glycosylation refers to the 7. Single-Domain Antibodies attachment of one of the Sugars N-aceylgalactosamine, galac In some embodiments, an antibody of the invention is a tose, or xylose to a hydroxyamino acid, most commonly single-domain antibody. A single-domain antibody is a single serine or threonine, although 5-hydroxyproline or 5-hydroxy polypeptide chain comprising all or a portion of the heavy lysine may also be used. chain variable domain or all or a portion of the light chain 65 Addition or deletion of glycosylation sites to the antibody variable domain of an antibody. In certain embodiments, a is conveniently accomplished by altering the amino acid single-domain antibody is a human single-domain antibody sequence Such that one or more of the above-described trip US 8,771,966 B2 75 76 eptide sequences (for N-linked glycosylation sites) is created or as further described below in reference to amino acid or removed. The alteration may also be made by the addition, classes, may be introduced and the products screened. deletion, or substitution of one or more serine or threonine residues to the sequence of the original antibody (for O-linked TABLE 1 5 glycosylation sites). Original Exemplary Preferred Where the antibody comprises an Fc region, the carbohy Residue Substitutions Substitutions drate attached thereto may be altered. For example, antibod Ala (A) Val; Leu: Ile Wall ies with a mature carbohydrate structure that lacks fucose Arg (R) Lys; Glin; ASn Lys attached to an Fc region of the antibody are described in US ASn (N) Gln; His; Asp, Lys; Arg Gln Pat Appl No US 2003/0157108 (Presta, L.). See also US 10 Asp (D) Glu: Asn Glu Cys (C) Ser: Ala Ser 2004/0093.621 (Kyowa Hakko Kogyo Co., Ltd). Antibodies Gln (Q) ASn: Glu ASn with a bisecting N-acetylglucosamine (GlcNAc) in the car Glu (E) Asp: Gln Asp bohydrate attached to an Fc region of the antibody are refer Gly (G) Ala Ala His (H) ASn; Glin; Lys; Arg Arg enced in WO 2003/01 1878, Jean-Mairet et al. and U.S. Pat. 15 Ile (I) Leu; Val: Met: Ala: Leu No. 6,602,684, Umana et al. Antibodies with at least one Phe: Norleucine galactose residue in the oligosaccharide attached to an Fc Leu (L) Norleucine: Ile: Val; Ile region of the antibody are reported in WO 1997/30087, Patel Met: Ala; Phe Lys (K) Arg: Gln; ASn Arg et al. See, also, WO 1998/58964 (Raju, S.) and WO 1999/ Met (M) Leu: Phe: Ile Leu 22764 (Raju, S.) concerning antibodies with altered carbohy Phe (F) Trp; Leu; Val: Ile: Ala: Tyr Tyr drate attached to the Fc region thereof. See also US 2005/ Pro (P) Ala Ala Ser (S) Thr Thr 0123546 (Umana et al.) on antigen-binding molecules with Thr (T) Val; Ser Ser modified glycosylation. Trp (W) Tyr; Phe Tyr In certain embodiments, a glycosylation variant comprises Tyr (Y) Trp; Phe: Thr; Ser Phe an Fc region, wherein a carbohydrate structure attached to the 25 Val (V) Ile: Leu: Met; Phe: Leu Fc region lacks fucose. Such variants have improved ADCC Ala: Norleucine function. Optionally, the Fc region further comprises one or more amino acid substitutions therein which further improve Substantial modifications in the biological properties of the ADCC, for example, substitutions at positions 298,333, and/ antibody are accomplished by selecting Substitutions that or 334 of the Fc region (Eu numbering of residues). Examples 30 differ significantly in their effect on maintaining (a) the struc of publications related to “defucosylated' or “fucose-defi ture of the polypeptide backbone in the area of the substitu cient antibodies include: US 2003/0157108: WO 2000/ tion, for example, as a sheet or helical conformation, (b) the 61739; WO 2001/29246; US 2003/0115614; US 2002/ charge or hydrophobicity of the molecule at the target site, or 0164328; US 2004/0093621; US 2004/0132140; US 2004/ (c) the bulk of the side chain. Amino acids may be grouped 0110704; US 2004/0110282: US 2004/0109865; WO 2003/ 35 according to similarities in the properties of their side chains 085119, WO 2003/084.570; WO 2005/035586: WO 2005/ (in A. L. Lehninger, in Biochemistry, second ed., pp. 73-75, 035778; WO2005/053742: Okazaki et al. J. Mol. Biol. 336: Worth Publishers, New York (1975)): 1239-1249 (2004): Yamane-Ohnuki et al. Biotech. Bioeng. (1) non-polar: Ala (A), Val (V), Leu (L), Ile (I), Pro (P), Phe 87: 614 (2004). Examples of cell lines producing defucosy (F), Trp (W), Met (M) lated antibodies include Lec13 CHO cells deficient in protein 40 (2) uncharged polar: Gly (G), Ser(S). Thr (T), Cys (C), Tyr fucosylation (Ripka et al. Arch. Biochem. Biophys. 249:533 (Y), Asn (N), Gln (Q) 545 (1986); US Pat Appl No US 2003/0157108A1, Presta, L; (3) acidic: Asp (D), Glu (E) and WO 2004/056312 A1, Adams et al., especially at (4) basic: Lys (K), Arg (R), His (H) Example 11), and knockout cell lines, such as alpha-1,6- Alternatively, naturally occurring residues may be divided fucosyltransferase gene, FUT8, knockout CHO cells (Ya 45 into groups based on common side-chain properties: mane-Ohnuki et al. Biotech. Bioeng. 87: 614 (2004)). (1) hydrophobic: Norleucine, Met, Ala, Val, Leu, Ile: In one embodiment, the antibody is altered to improve its (2) neutral hydrophilic: Cys, Ser, Thr, Asn., Gln; serum half-life. To increase the serum halflife of the antibody, (3) acidic: Asp, Glu; one may incorporate a salvage receptor binding epitope into (4) basic: His, Lys, Arg; the antibody (especially an antibody fragment) as described 50 (5) residues that influence chain orientation: Gly, Pro; in U.S. Pat. No. 5,739,277, for example. As used herein, the (6) aromatic: Trp, Tyr, Phe. term "salvage receptor binding epitope” refers to an epitope Non-conservative Substitutions will entail exchanging a of the Fc region of an IgG molecule (e.g., IgG1, IgG2, IgG3. member of one of these classes for another class. Such sub or IgG4) that is responsible for increasing the in vivo serum stituted residues also may be introduced into the conservative half-life of the IgG molecule (US 2003/0190311, U.S. Pat. 55 Substitution sites or, into the remaining (non-conserved) sites. No. 6,821,505; U.S. Pat. No. 6,165,745; U.S. Pat. No. 5,624, One type of substitutional variant involves substituting one 821; U.S. Pat. No. 5,648,260; U.S. Pat. No. 6,165,745; U.S. or more hyperVariable region residues of a parent antibody Pat. No. 5,834,597). (e.g. a humanized or human antibody). Generally, the result Another type of variant is an amino acid substitution vari ing variant(s) selected for further development will have ant. These variants have at least one amino acid residue in the 60 modified (e.g., improved) biological properties relative to the antibody molecule replaced by a different residue. Sites of parent antibody from which they are generated. A convenient interest for Substitutional mutagenesis include the hyperVari way for generating Such Substitutional variants involves affin able regions, but FR alterations are also contemplated. Con ity maturation using phage display. Briefly, several hyperVari servative substitutions are shown in Table 1 under the heading able region sites (e.g. 6-7 sites) are mutated to generate all of “preferred substitutions.” If such substitutions result in a 65 possible amino acid Substitutions at each site. The antibodies desirable change in biological activity, then more substantial thus generated are displayed from filamentous phage par changes, denominated “exemplary Substitutions' in Table 1, ticles as fusions to at least part of a phage coat protein (e.g., US 8,771,966 B2 77 78 the gene III product of M13) packaged within each particle. promote heterodimerization. These modifications comprise The phage-displayed variants are then screened for their bio introduction of a protuberance into a first Fc polypeptide and logical activity (e.g. binding affinity). In order to identify a cavity into a second Fc polypeptide, wherein the protuber candidate hyperVariable region sites for modification, Scan ance is positionable in the cavity so as to promote complexing ning mutagenesis (e.g., alanine scanning) can be performed to of the first and second Fc polypeptides. Methods of generat identify hyperVariable region residues contributing signifi ing antibodies with these modifications are known in the art, cantly to antigen binding. Alternatively, or additionally, it e.g., as described in U.S. Pat. No. 5,731,168. may be beneficial to analyze a crystal structure of the antigen 9. Antibody Derivatives antibody complex to identify contact points between the anti The antibodies of the present invention can be further body and antigen. Such contact residues and neighboring 10 modified to contain additional nonproteinaceous moieties residues are candidates for Substitution according to tech that are known in the art and readily available. Preferably, the niques known in the art, including those elaborated herein. moieties suitable for derivatization of the antibody are water Once Such variants are generated, the panel of variants is soluble polymers. Non-limiting examples of water soluble Subjected to screening using techniques known in the art, polymers include, but are not limited to, polyethylene glycol including those described herein, and antibodies with Supe 15 (PEG), copolymers of ethylene glycol/propylene glycol, car rior properties in one or more relevant assays may be selected boxymethylcellulose, dextran, polyvinyl alcohol, polyvinyl for further development. pyrrolidone, poly-1,3-dioxolane, poly-1,3,6-trioxane, ethyl Nucleic acid molecules encoding amino acid sequence ene/maleic anhydride copolymer, polyaminoacids (either variants of the antibody are prepared by a variety of methods homopolymers or random copolymers), and dextran or poly known in the art. These methods include, but are not limited (n-vinyl pyrrolidone)polyethylene glycol, propropylene gly to, isolation from a natural Source (in the case of naturally col homopolymers, prolypropylene oxide/ethylene oxide co occurring amino acid sequence variants) or preparation by polymers, polyoxyethylated polyols (e.g., glycerol), oligonucleotide-mediated (or site-directed) mutagenesis, polyvinyl alcohol, and mixtures thereof. Polyethylene glycol PCR mutagenesis, and cassette mutagenesis of an earlier propionaldehyde may have advantages in manufacturing due prepared variant or a non-variant version of the antibody. 25 to its stability in water. The polymer may be of any molecular It may be desirable to introduce one or more amino acid weight, and may be branched or unbranched. The number of modifications in an Fc region of antibodies of the invention, polymers attached to the antibody may vary, and if more than thereby generating an Fc region variant. The Fc region variant one polymer are attached, they can be the same or different may comprise a human Fc region sequence (e.g., a human molecules. In general, the number and/or type of polymers IgG1, IgG2, IgG3 or IgG4 Fc region) comprising an amino 30 used for derivatization can be determined based on consider acid modification (e.g. a substitution) at one or more amino ations including, but not limited to, the particular properties acid positions including that of a hinge cysteine. or functions of the antibody to be improved, whether the In accordance with this description and the teachings of the antibody derivative will be used in a therapy under defined art, it is contemplated that in Some embodiments, an antibody conditions, etc. of the invention may comprise one or more alterations as 35 In another embodiment, conjugates of an antibody and compared to the wild type counterpartantibody, e.g. in the Fc nonproteinaceous moiety that may be selectively heated by region. These antibodies would nonetheless retain Substan exposure to radiation are provided. In one embodiment, the tially the same characteristics required for therapeutic utility nonproteinaceous moiety is a carbon nanotube (Kam et al., as compared to their wild type counterpart. For example, it is Proc. Natl. Acad. Sci. 102: 11600-11605 (2005)). The radia thought that certain alterations can be made in the Fc region 40 tion may be of any wavelength, and includes, but is not that would result in altered (i.e., either improved or dimin limited to, wavelengths that do not harm ordinary cells, but ished) C1q binding and/or Complement Dependent Cytotox which heat the nonproteinaceous moiety to a temperature at icity (CDC), e.g., as described in WO99/51642. See also which cells proximal to the antibody-nonproteinaceous moi Duncan & Winter Nature 322:738-40 (1988): U.S. Pat. No. ety are killed. 5,648,260; U.S. Pat. No. 5,624,821; and WO94/29351 con 45 Certain Methods of Making Antibodies cerning other examples of Fc region variants. WO00/42072 1. Certain Hybridoma-Based Methods (Presta) and WO 2004/056312 (Lowman) describe antibody The anti-STEAP-1 monoclonal antibodies of the invention variants with improved or diminished binding to FcRs. The can be made using the hybridoma method first described by content of these patent publications are specifically incorpo Kohler et al., Nature, 256:495 (1975), or may be made by rated herein by reference. See, also, Shields et al. J. Biol. 50 recombinant DNA methods (U.S. Pat. No. 4,816,567). Chem. 9(2): 6591-6604 (2001). Antibodies with increased In the hybridoma method, a mouse or other appropriate half lives and improved binding to the neonatal Fc receptor host animal. Such as a hamster, is immunized to elicit lym (FcRn), which is responsible for the transfer of maternal IgGs phocytes that produce or are capable of producing antibodies to the fetus (Guyer et al., J. Immunol. 1 17:587 (1976) and that will specifically bind to the protein used for immuniza Kim et al., J. Immunol. 24:249 (1994)), are described in 55 tion. Antibodies to STEAP-1 generally are raised in animals US2005/0014934A1 (Hinton et al.). These antibodies com by multiple Subcutaneous (sc) or intraperitoneal (ip) injec prise an Fc region with one or more substitutions therein tions of STEAP-1 and an adjuvant. STEAP-1 may be pre which improve binding of the Fc region to FcRn. Polypeptide pared using methods well-known in the art, some of which are variants with altered Fc region amino acid sequences and further described herein. For example, STEAP-1 may be pro increased or decreased C1q binding capability are described 60 duced recombinantly. In one embodiment, animals are immu in U.S. Pat. No. 6,194,551 B1, WO99/51642. The contents of nized with a derivative of STEAP-1 that contains an extracel those patent publications are specifically incorporated herein lular portion of STEAP-1 fused to the Fc portion of an by reference. See, also, Idusogie et al. J. Immunol. 164: immunoglobulin heavy chain. In one embodiment, animals 4178-4.184 (2000). are immunized with an STEAP-1-IgG1 fusion protein. In one In one aspect, the invention provides antibodies compris 65 embodiment, animals are immunized with immunogenic ing modifications in the interface of Fc polypeptides compris derivatives of STEAP-1 in a solution with monophosphoryl ing the Fc region, wherein the modifications facilitate and/or lipid A (MPL)/trehalose dicrynomycolate (TDM) (Ribi US 8,771,966 B2 79 80 Immunochem. Research, Inc., Hamilton, Mont.), and the Press, Totowa, N.J.), and in certain embodiments, in Lee et al. solution is injected intradermally at multiple sites. Two weeks (2004) J. Mol. Biol. 340:1073-1093. later the animals are boosted. Seven to fourteen days later the In principle, synthetic antibody clones are selected by animals are bled, and the serum is assayed for anti-STEAP-1 screening phage libraries containing phage that display vari titer. Animals are boosted until titer plateaus. ous fragments of antibody variable region (FV) fused to phage Alternatively, lymphocytes may be immunized in vitro. coat protein. Such phage libraries are panned by affinity chro Lymphocytes then are fused with myeloma cells using a matography against the desired antigen. Clones expressing Suitable fusing agent, such as polyethylene glycol, to form a FV fragments capable of binding to the desired antigen are hybridoma cell (Goding, Monoclonal Antibodies: Principles adsorbed to the antigen and thus separated from the non 10 binding clones in the library. The binding clones are then and Practice, pp. 59-103 (Academic Press, 1986)). eluted from the antigen, and can be further enriched by addi The hybridoma cells thus prepared are seeded and grown in tional cycles of antigen adsorption/elution. Any of the anti a Suitable culture medium, e.g., a medium that contains one or STEAP-1 antibodies of the invention can be obtained by more substances that inhibit the growth or survival of the designing a Suitable antigen screening procedure to select for unfused, parental myeloma cells. For example, if the parental 15 the phage clone of interest followed by construction of a full myeloma cells lack the enzyme hypoxanthine guanine phos length anti-STEAP-1 antibody clone using the Fv sequences phoribosyl transferase (HGPRT or HPRT), the culture from the phage clone of interest and Suitable constant region medium for the hybridomas typically will include hypoxan (Fc) sequences described in Kabat et al., Sequences of Pro thine, aminopterin, and thymidine (HAT medium), which teins of Immunological Interest, Fifth Edition, NIH Publica substances prevent the growth of HGPRT-deficient cells. tion 91-3242, Bethesda Md. (1991), vols. 1-3. In certain embodiments, myeloma cells are those that fuse In certain embodiments, the antigen-binding domain of an efficiently, support stable high-level production of antibody antibody is formed from two variable (V) regions of about by the selected antibody-producing cells, and are sensitive to 110 amino acids, one each from the light (VL) and heavy a medium such as HAT medium. Exemplary myeloma cells (VH) chains, that both present three hypervariable loops include, but are not limited to, murine myeloma lines, such as 25 (HVRs) or complementarity-determining regions (CDRs). those derived from MOPC-21 and MPC-11 mouse tumors Variable domains can be displayed functionally on phage, available from the Salk Institute Cell Distribution Center, San either as single-chain Fv (scFv) fragments, in which VHand Diego, Calif. USA, and SP-2 or X63-Ag8-653 cells available VL are covalently linked through a short, flexible peptide, or from the American Type Culture Collection, Rockville, Md. as Fab fragments, in which they are each fused to a constant USA. Human myeloma and mouse-human heteromyeloma 30 domain and interact non-covalently, as described in Winter et cell lines also have been described for the production of al., Ann. Rev. Immunol. 12:433-455 (1994). As used herein, human monoclonal antibodies (Kozbor, J. Immunol. 133: ScFv encoding phage clones and Fab encoding phage clones 3001 (1984); Brodeur et al., Monoclonal Antibody Produc are collectively referred to as “Fv phage clones' or “FV tion Techniques and Applications, pp. 51-63 (Marcel Dekker, clones.” Inc., New York, 1987)). 35 Repertoires of VH and VL genes can be separately cloned Culture medium in which hybridoma cells are growing is by polymerase chain reaction (PCR) and recombined ran assayed for production of monoclonal antibodies that bind to domly in phage libraries, which can then be searched for STEAP-1. Preferably, the binding specificity of monoclonal antigen-binding clones as described in Winter et al., Ann. antibodies produced by hybridoma cells is determined by Rev. Immunol. 12: 433-455 (1994). Libraries from immu immunoprecipitation or by an in vitro binding assay, Such as 40 nized sources provide high-affinity antibodies to the immu radioimmunoassay (RIA) or enzyme-linked immunoadsor nogen without the requirement of constructing hybridomas. bent assay (ELISA). The binding affinity of the monoclonal Alternatively, the naive repertoire can be cloned to provide a antibody can, for example, be determined by the Scatchard single source of human antibodies to a wide range of non-self analysis of Munson et al., Anal. Biochem., 107:220 (1980). and also self antigens without any immunization as described After hybridoma cells are identified that produce antibod 45 by Griffiths et al., EMBOJ, 12:725-734 (1993). Finally, naive ies of the desired specificity, affinity, and/or activity, the libraries can also be made synthetically by cloning the unre clones may be subcloned by limiting dilution procedures and arranged V-gene segments from stem cells, and using PCR grown by standard methods (Goding, Monoclonal Antibod primers containing random sequence to encode the highly ies: Principles and Practice, pp. 59-103 (Academic Press, variable CDR3 regions and to accomplish rearrangement in 1986)). Suitable culture media for this purpose include, for 50 vitro as described by Hoogenboom and Winter, J. Mol. Biol. example, D-MEM or RPMI-1640 medium. In addition, the 227:381-388 (1992). hybridoma cells may be grown in vivo as ascites tumors in an In certain embodiments, filamentous phage is used to dis animal. Monoclonal antibodies secreted by the subclones are play antibody fragments by fusion to the minor coat protein Suitably separated from the culture medium, ascites fluid, or pIII. The antibody fragments can be displayed as single chain serum by conventional immunoglobulin purification proce 55 Fv fragments, in which VHandVL domains are connected on dures such as, for example, protein A-Sepharose, hydroxyla the same polypeptide chain by a flexible polypeptide spacer, patite chromatography, gel electrophoresis, dialysis, or affin e.g. as described by Marks et al., J. Mol. Biol., 222:581-597 ity chromatography. (1991), or as Fab fragments, in which one chain is fused to 2. Certain Library Screening Methods pIII and the other is secreted into the bacterial host cell peri Anti-STEAP-1 antibodies of the invention can be made by 60 plasm where assembly of a Fab-coat protein structure which using combinatorial libraries to screen for antibodies with the becomes displayed on the phage Surface by displacing some desired activity or activities. For example, a variety of meth of the wild type coat proteins, e.g. as described in Hoogen ods are known in the art for generating phage display libraries boom et al., Nucl. Acids Res., 19:4133-4137 (1991). and screening Such libraries for antibodies possessing the In general, nucleic acids encoding antibody gene frag desired binding characteristics. Such methods are described 65 ments are obtained from immune cells harvested from generally in Hoogenboom et al. (2001) in Methods in humans or animals. If a library biased in favor of anti Molecular Biology 178:1-37 (O'Brien et al., ed., Human STEAP-1 clones is desired, the subject is immunized with US 8,771,966 B2 81 82 STEAP-1 to generate an antibody response, and spleen cells (1992)), and mapped (reported in Matsuda et al., Nature and/or circulating B cells other peripheral blood lymphocytes Genet. 3: 88-94 (1993); these cloned segments (including all (PBLs) are recovered for library construction. In a preferred the major conformations of the H1 and H2 loop) can be used embodiment, a human antibody gene fragment library biased to generate diverse VH gene repertoires with PCR primers in favor of anti-STEAP-1 clones is obtained by generating an 5 encoding H3 loops of diverse sequence and length as anti-STEAP-1 antibody response in transgenic mice carrying described in Hoogenboom and Winter, J. Mol. Biol., 227: a functional human immunoglobulin gene array (and lacking 381-388 (1992).VH repertoires can also be made with all the a functional endogenous antibody production system) Such sequence diversity focused in a long H3 loop of a single that STEAP-1 immunization gives rise to B cells producing length as described in Barbas et al., Proc. Natl. Acad. Sci. human antibodies against STEAP-1. The generation of 10 USA, 89: 4457-4461 (1992). Human VK and V) segments human antibody-producing transgenic mice is described have been cloned and sequenced (reported in Williams and below. Winter, Eur. J. Immunol., 23: 1456-1461 (1993)) and can be Additional enrichment for anti-STEAP-1 reactive cell used to make synthetic light chain repertoires. Synthetic V populations can be obtained by using a suitable screening gene repertoires, based on a range of VHandVL folds, and L3 procedure to isolate B cells expressing STEAP-1-specific 15 and H3 lengths, will encode antibodies of considerable struc membrane bound antibody, e.g., by cell separation using tural diversity. Following amplification of V-gene encoding STEAP-1 affinity chromatography or adsorption of cells to DNAS, germline V-gene segments can be rearranged in vitro fluorochrome-labeled STEAP-1 followed by flow-activated according to the methods of Hoogenboom and Winter, J. Mol. cell sorting (FACS). Biol., 227:381-388 (1992). Alternatively, the use of spleen cells and/or B cells or other Repertoires of antibody fragments can be constructed by PBLs from an unimmunized donor provides a better repre combining VH and VL gene repertoires together in several sentation of the possible antibody repertoire, and also permits ways. Each repertoire can be created in different vectors, and the construction of an antibody library using any animal the vectors recombined in vitro, e.g., as described in Hogrefe (human or non-human) species in which STEAP-1 is not et al., Gene, 128: 119-126 (1993), or in vivo by combinatorial antigenic. For libraries incorporating in vitro antibody gene 25 infection, e.g., the loxP system described in Waterhouse et al., construction, stem cells are harvested from the Subject to Nucl. Acids Res., 21: 2265-2266 (1993). The in vivo recom provide nucleic acids encoding unrearranged antibody gene bination approach exploits the two-chain nature of Fab frag segments. The immune cells of interest can be obtained from ments to overcome the limit on library size imposed by E. coli a variety of animal species, such as human, mouse, rat, Lago transformation efficiency. Naive VH and VL repertoires are morpha, luprine, canine, feline, porcine, bovine, equine, and 30 cloned separately, one into a phagemid and the other into a avian species, etc. phage vector. The two libraries are then combined by phage Nucleic acid encoding antibody variable gene segments infection of phagemid-containing bacteria so that each cell (including VHandVL segments) are recovered from the cells contains a different combination and the library size is limited of interest and amplified. In the case ofrearranged VHandVL only by the number of cells present (about 10" clones). Both gene libraries, the desired DNA can be obtained by isolating 35 vectors contain in vivo recombination signals so that the VH genomic DNA or mRNA from lymphocytes followed by and VL genes are recombined onto a single replicon and are polymerase chain reaction (PCR) with primers matching the co-packaged into phage virions. These huge libraries provide 5' and 3' ends of rearranged VHand VL genes as described in large numbers of diverse antibodies of good affinity (Kl' of Orlandi et al., Proc. Natl. Acad. Sci. (USA), 86: 3833-3837 about 10 M). (1989), thereby making diverse V gene repertoires for expres 40 Alternatively, the repertoires may be cloned sequentially sion. The V genes can be amplified from cDNA and genomic into the same vector, e.g. as described in Barbas et al., Proc. DNA, with back primers at the 5' end of the exon encoding the Natl. Acad. Sci. USA, 88: 7978-7982 (1991), or assembled mature V-domain and forward primers based within the J-seg together by PCR and then cloned, e.g. as described in Clack ment as described in Orlandi et al. (1989) and in Ward et al., son et al., Nature, 352: 624-628 (1991). PCR assembly can Nature, 341: 544-546 (1989). However, for amplifying from 45 also be used to join VH and VL DNAs with DNA encoding a cDNA, back primers can also be based in the leader exon as flexible peptide spacer to form single chain Fv (sclv) reper described in Jones et al., Biotechnol. 9:88-89 (1991), and toires. In yet another technique, “in cell PCR assembly' is forward primers within the constant region as described in used to combine VH and VL genes within lymphocytes by Sastry et al., Proc. Natl. Acad. Sci. (USA), 86: 5728-5732 PCR and then clone repertoires of linked genes as described (1989). To maximize complementarity, degeneracy can be 50 in Embleton et al., Nucl. Acids Res., 20:3831-3837 (1992). incorporated in the primers as described in Orlandi et al. The antibodies produced by naive libraries (either natural (1989) or Sastry et al. (1989). In certain embodiments, library or synthetic) can be of moderate affinity (K, of about 10° to diversity is maximized by using PCR primers targeted to each 10 M'), but affinity maturation can also be mimicked in V-gene family in order to amplify all available VH and VL vitro by constructing and reselecting from secondary libraries arrangements present in the immune cell nucleic acid sample, 55 as described in Winter et al. (1994), supra. For example, e.g. as described in the method of Marks et al., J. Mol. Biol. mutation can be introduced at random in vitro by using error 222:581-597 (1991) or as described in the method of Orumet prone polymerase (reported in Leung et al., Technique, 1: al., Nucleic Acids Res., 21: 4491-4498 (1993). For cloning of 11-15 (1989)) in the method of Hawkins et al., J. Mol. Biol., the amplified DNA into expression vectors, rare restriction 226: 889-896 (1992) or in the method of Gram et al., Proc. sites can be introduced within the PCR primer as a tag at one 60 Natl. Acad. Sci USA, 89: 3576-3580 (1992). Additionally, end as described in Orlandi et al. (1989), or by further PCR affinity maturation can be performed by randomly mutating amplification with a tagged primeras described in Clackson one or more CDRs, e.g. using PCR with primers carrying et al., Nature, 352: 624-628 (1991). random sequence spanning the CDR of interest, in selected Repertoires of synthetically rearranged V genes can be individual Fv clones and Screening for higher affinity clones. derived in vitro from V gene segments. Most of the human 65 WO9607754 (published 14 Mar. 1996) described a method VH-gene segments have been cloned and sequenced (re for inducing mutagenesis in a complementarity determining ported in Tomlinson et al., J. Mol. Biol., 227: 776-798 region of an immunoglobulin light chain to create a library of US 8,771,966 B2 83 84 light chain genes. Another effective approach is to recombine STEAP-1. In one embodiment, the invention provides anti the VH or VL domains selected by phage display with reper STEAP-1 antibodies that block the binding between a toires of naturally occurring V domain variants obtained from STEAP-1 ligand and STEAP-1, but do not block the binding unimmunized donors and Screen for higher affinity in several between a STEAP-1 ligand and a second protein. Fv clones rounds of chain reshuffling as described in Marks et al., 5 corresponding to such anti-STEAP-1 antibodies can be Biotechnol., 10: 779-783 (1992). This technique allows the selected by (1) isolating anti-STEAP-1 clones from a phage production of antibodies and antibody fragments with affini library as described above, and optionally amplifying the ties of about 10 M or less. isolated population of phage clones by growing up the popu Screening of the libraries can be accomplished by various lation in a suitable bacterial host; (2) selecting STEAP-1 and techniques known in the art. For example, STEAP-1 can be 10 a second protein against which blocking and non-blocking used to coat the wells of adsorption plates, expressed on host activity, respectively, is desired; (3) adsorbing the anti cells affixed to adsorption plates or used in cell sorting, or STEAP-1 phage clones to immobilized STEAP-1; (4) using conjugated to biotin for capture with Streptavidin-coated an excess of the second protein to elute any undesired clones beads, or used in any other method for panning phage display that recognize STEAP-1-binding determinants which overlap libraries. 15 or are shared with the binding determinants of the second The phage library samples are contacted with immobilized protein; and (5) eluting the clones which remain adsorbed STEAP-1 under conditions suitable for binding at least a following step (4). Optionally, clones with the desired block portion of the phage particles with the adsorbent. Normally, ing/non-blocking properties can be further enriched by the conditions, including pH, ionic strength, temperature and repeating the selection procedures described herein one or the like are selected to mimic physiological conditions. The 20 more times. phages bound to the solid phase are washed and then eluted by DNA encoding hybridoma-derived monoclonal antibodies acid, e.g. as described in Barbas et al., Proc. Natl. Acad. Sci orphage display Fv clones of the invention is readily isolated USA, 88: 7978-7982 (1991), or by alkali, e.g. as described in and sequenced using conventional procedures (e.g. by using Marks et al., J. Mol. Biol., 222: 581-597 (1991), or by oligonucleotide primers designed to specifically amplify the STEAP-1 antigen competition, e.g. in a procedure similar to 25 heavy and light chain coding regions of interest from hybri the antigen competition method of Clackson et al., Nature, doma or phage DNA template). Once isolated, the DNA can 352: 624-628 (1991). Phages can be enriched 20-1,000-fold be placed into expression vectors, which are then transfected in a single round of selection. Moreover, the enriched phages into host cells such as E. coli cells, simian COS cells, Chinese can be grown in bacterial culture and subjected to further hamster ovary (CHO) cells, or myeloma cells that do not rounds of selection. 30 otherwise produce immunoglobulin protein, to obtain the The efficiency of selection depends on many factors, synthesis of the desired monoclonal antibodies in the recom including the kinetics of dissociation during washing, and binant host cells. Review articles on recombinant expression whether multiple antibody fragments on a single phage can in bacteria of antibody-encoding DNA include Skerra et al., simultaneously engage with antigen. Antibodies with fast Curr. Opinion in Immunol. 5: 256 (1993) and Pluckthun, dissociation kinetics (and weak binding affinities) can be 35 Immunol. Revs, 130: 151 (1992). retained by use of short washes, multivalent phage display DNA encoding the Fv clones of the invention can be com and high coating density of antigen in Solid phase. The high bined with known DNA sequences encoding heavy chain density not only stabilizes the phage through multivalent and/or light chain constant regions (e.g. the appropriate DNA interactions, but favors rebinding of phage that has dissoci sequences can be obtained from Kabat et al., Supra) to form ated. The selection of antibodies with slow dissociation kinet- 40 clones encoding full or partial length heavy and/or light ics (and good binding affinities) can be promoted by use of chains. It will be appreciated that constant regions of any long washes and monovalent phage display as described in isotype can be used for this purpose, including IgG, IgM, IgA, Bass et al., Proteins, 8:309-314 (1990) and in WO92/09690, IgD, and IgE constant regions, and that Such constant regions and a low coating density of antigen as described in Marks et can be obtained from any human or animal species. An Fv al., Biotechnol., 10: 779-783 (1992). 45 clone derived from the variable domain DNA of one animal It is possible to select between phage antibodies of differ (such as human) species and then fused to constant region ent affinities, even with affinities that differ slightly, for DNA of another animal species to form coding sequence(s) STEAP-1. However, random mutation of a selected antibody for “hybrid,” full length heavy chain and/or light chain is (e.g. as performed in Some affinity maturation techniques) is included in the definition of “chimeric' and “hybrid anti likely to give rise to many mutants, most binding to antigen, 50 body as used herein. In certain embodiments, an Fv clone and a few with higher affinity. With limiting STEAP-1, rare derived from human variable DNA is fused to human constant high affinity phage could be competed out. To retain all higher region DNA to form coding sequence(s) for full- or partial affinity mutants, phages can be incubated with excess bioti length human heavy and/or light chains. nylated STEAP-1, but with the biotinylated STEAP-1 at a DNA encoding anti-STEAP-1 antibody derived from a concentration of lower molarity than the target molar affinity 55 hybridoma of the invention can also be modified, for example, constant for STEAP-1. The high affinity-binding phages can by Substituting the coding sequence for human heavy- and then be captured by Streptavidin-coated paramagnetic beads. light-chain constant domains in place of homologous murine Such “equilibrium capture' allows the antibodies to be sequences derived from the hybridoma clone (e.g. as in the selected according to their affinities of binding, with sensi method of Morrison et al., Proc. Natl. Acad. Sci. USA, 81: tivity that permits isolation of mutant clones with as little as 60 6851-6855 (1984)). DNA encoding a hybridoma- or Fv two-fold higher affinity from a great excess of phages with clone-derived antibody or fragment can be further modified lower affinity. Conditions used in washing phages bound to a by covalently joining to the immunoglobulin coding Solid phase can also be manipulated to discriminate on the sequence all or part of the coding sequence for a non-immu basis of dissociation kinetics. noglobulin polypeptide. In this manner, "chimeric' or Anti-STEAP-1 clones may be selected based on activity. In 65 “hybrid antibodies are prepared that have the binding speci certain embodiments, the invention provides anti-STEAP-1 ficity of the Fv clone or hybridoma clone-derived antibodies antibodies that bind to living cells that naturally express of the invention.