AU6338898A - High affinity humanized anti-cea monoclonal antibodies - Google Patents

Description

WO 99/43817 PCT/US98/03680 HIGH AFFINITY HUMANIZED ANTI-CEA MONOCLONAL ANTIBODIES FIELD OF THE INVENTION The present invention relates to humanized monoclonal antibodies and fragments or 5 derivatives thereof which specifically bind carcinoembryonic antigen (CEA), which is an antigen expressed by various human carcinomas including breast, lung, and gastrointestinal carcinomas such as stomach and colon cancers. More specifically, the present invention relates to humanized monoclonal antibodies and humanized antibody fragments and derivatives thereof which are derived from murine monoclonal antibody COL-1, a high affinity 0 anti-CEA antibody. The present invention further relates to methods for producing such humanized monoclonal antibodies specific to CEA, pharmaceutical and diagnostic compositions containing such humanized monoclonal antibodies, and methods of use thereof for the treatment or diagnosis of cancer. 5 BACKGROUND OF THE INVENTION The identification of antigens expressed by tumor cells and the preparation of monoclonal antibodies which specifically bind such antigens is well known in the art. Anti tumor monoclonal antibodies exhibit potential application as both therapeutic and diagnostic agents. Such monoclonal antibodies have potential application as diagnostic agents 0 because they specifically bind tumor antigens and thereby can detect the presence of tumor cells or tumor antigen in an analyte. For example, use of monoclonal antibodies which bind tumor antigens for in vitro and in vivo imaging of tumor cells or tumors using a labeled form of such a monoclonal antibody is conventional in the art. Moreover, monoclonal antibodies which bind tumor antigens have well known 5 application as therapeutic agents. The usage of monoclonal antibodies themselves as therapeutic agents, or as conjugates wherein the monoclonal antibody is directly or indirectly attached to an effector moiety, e.g., a drug, cytokine, cytotoxin, etc., is well known. Essentially, if the monoclonal antibody is attached to an effector moiety, then the monoclonal antibody functions as a targeting moiety, i.e. it directs the effector moiety (which 0 typically possesses therapeutic activity) to the antibody's target, e.g., a tumor which expresses the antigen bound by the monoclonal antibody. In contrast, when the monoclonal antibody itself operates as a therapeutic agent, the antibody functions both as a targeting moiety -- i.e. it will specifically bind a cell which expresses the antigen -- and as an effector which mediates therapeutic activity, typically tumor cell lysis. A monoclonal antibody may -1 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 possess one or more of such effector functions, which include, e.g., antibody-dependent cellular cytotoxicity (ADCC) and complement dependent cytotoxicity (CDC), among others; these functions are effected by the portion of the antibody molecule generally referred to in the literature as the Fc portion. 5 One specific tumor antigen to which various monoclonal antibodies have been produced is the carcinoembryonic antigen (CEA). CEA is an antigen complex having a molecular weight of about 180,000 D, which is expressed by numerous carcinomas including gastrointestinal carcinomas, colorectal carcinomas, breast carcinomas, ovarian carcinomas, and lung carcinomas. See, e.g., Robbins et aL., Int'l J. Cancer, 53(6):892-897 (1993); Greiner .0 et aL., J. Clin. OncoL., 10(5):735-746 (1992); Obuchi et aL., Cancer Res., 47(13):3565-3571 (1987); Muraro et al., Cancer Res., 45(11 Pt. 2):5769-5780 (1985). The use of monoclonal antibodies to detect various, specific CEA epitopes differentially expressed on human carcinomas has been reported in the literature. See, e.g., Obuchi et al., Cancer Res., 47(13):3565-3571 (1987); Muraro et al., Cancer Res., 45(11 Pt. 5 2):5769-5780 (1985). In particular, Muraro et al. (id.) report generation of monoclonal antibodies designated COL-1 through COL-15, which exhibit a strong, selective reactivity for human colon carcinomas versus normal adult tissues. These antibodies react with distinct, restricted epitopes on CEA. Of these antibodies, the COL-1 antibody has been the focus of 0 considerable attention because of its high affinity for CEA (1.4x109 M') and also because it comprises no detectable reactivity for CEA-related antigens such as the nonspecific cross treating antigen (NCA) and the normal fecal antigen (NFA). Robbins et al., Int'I J. Canc., 53(6):892-897 (1993). Because of its binding properties, COL-1 is currently being evaluated for use as a 5 therapeutic agent. For example, Siler et al. (Biotech. Ther., 4(3-4):163-181 (1993)) report the administration of 1 3 1 'I-labeled COL-1 to LS-M4T human colon carcinoma xenograft containing athymic mice. They report that this treatment resulted in reduction of the rate of tumor growth, within little or no toxicity, and that their results demonstrate the potential therapeutic efficacy of radiolabeled COL-1 in clinical trials. Also, Yu etal. (J. Clin. Oncol., 0 14(6):1798-1809 (1996)) report that 13 1 'I-labeled COL-1 is now in phase 1 clinical trials in patients having gastrointestinal malignancies. They further indicate that the antibody conjugate is well tolerated, except for some hematologic toxicity. In addition, the use of conjugates of COL-1 and 3-galactosidase has been shown to specifically kill in vitro tumor cells from a variety of tumor cell lines. Abraham et al., Cell Biophys., 24-25:127-133 (1994). -2 qlIRSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 However, while murine antibodies, such as COL-1 and other anti-CEA murine antibodies, have applicability as therapeutic agents in humans, they are disadvantageous in some respects. Specifically, because murine antibodies are of foreign species origin, they may be immunogenic in humans. This may result in a neutralizing antibody response -- a 5 human anti-murine antibody (HAMA) response -- which is particularly problematic if the antibodies are desired to be administered repeatedly, e.g., for treatment of a chronic or recurrent disease condition. This is a significant drawback, as some cancer treatments are effected over a prolonged time period, e.g., over several years or longer. Also, because these antibody molecules contain murine constant domains they may not exhibit human 0 effector functions. In an effort to eliminate or reduce such problems, chimeric antibodies have been disclosed. Chimeric antibodies contain portions of two different antibodies, typically of two different species. Generally, such antibodies contain human constant regions attached to variable regions from another species, typically murine variable regions. For example, some 5 mouse/human chimeric antibodies have been reported which exhibit binding characteristics, of the parental mouse antibody and effector functions associated with the human constant region. See, e.g., U.S. Patent 4,816,567 to Cabilly et aL.; U.S. Patent 4,978,745 to Shoemaker et al.; U.S. Patent 4,975,369 to Beavers et aL.; and U.S. Patent 4,816,397 to Boss et al. Generally, these chimeric antibodies are constructed by preparing a genomic 0 gene library from DNA extracted from pre-existing murine hybridomas. Nishimura et aL., Cancer Res., 47:999 (1987). The library is then screened for variable region genes from both heavy and light chains exhibiting the correct antibody fragment rearrangement patterns. Alternatively, cDNA libraries are prepared from RNA extracted from the hybridomas and then screened, or the variable regions are obtained by polymerase chain reaction. The 5 cloned variable region genes are then ligated into an expression vector containing cloned cassettes of the appropriate heavy or light chain human constant region gene. The chimeric genes are then expressed in a cell line of choice, usually a murine myeloma line. Such chimeric antibodies have been used in human therapy. Moreover, the production of a chimeric mouse anti-human antibody derived from 0 COL-1, which specifically binds CEA, has been reported. See e.g., U.S. Patent 5,472,693 to Gourlie et al. (owned by The Dow Chemical Company). Also, Morrison et al. report the preparation of several anti-tumor chimeric monoclonal antibodies, in Important Advances in Oncology, Recombinant Chimeric Monoclonal Anti bodies, pp. 3-18 (S.A. Rosenberg, ed., 1990) (J.B. Lippincott, Philadelphia, PA). Results of -3 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 clinical trials with chimeric cMAb-17-1A in patients with metastatic colorectal carcinoma now show that this antibody has a 6-fold longer circulation time and significantly reduced immunogenicity as compared to the murine monoclonal antibody from which it was derived. LoBuglio et al., Proc. Natl. Acad. Sci. USA, 86:4220-4224 (1989); Meredith et al., J. Nucl. 5 Med., 32:1162-1168 (1991). However, while such chimerized monoclonal antibodies typically exhibit lesser immunogenicity, they are still potentially immunogenic in humans because they contain murine variable sequences which may elicit antibody responses. Thus, there is the possibility that these chimeric antibodies may elicit an anti-idiotypic response if administered 0 to patients. Saleh et al., Cancer Immunol. Immunother., 32:185-190 (1990). Because of the immunogenicity of chimeric antibodies, methods have been developed recently for the production of "humanized" antibodies. Ideally, "humanization" results in an antibody that is less immunogenic, with complete retention of the antigen binding properties of the original molecule. In order to retain all the antigen-binding 5 properties of the original antibody, the structure of its combining-site has to be faithfully reproduced in the "humanized" version. This can potentially be achieved by transplanting the combining site of the non-human antibody onto a human framework, either: (a) by grafting only the non-human complementarity determining regions (CDRs) onto human framework regions (FRs) and constant regions, with or without retention of critical framework 0 residues (see, Jones etal., Nature, 321:522 (1986) and Verhoeyen etal., Science, 239:1539 (1988); or (b) by transplanting the entire non-human variable domains (to preserve ligand binding properties) and also "cloaking" them with a human-like surface through judicious replacement of exposed residues (in order to reduce antigenicity) (see, Padlan, Molec. Immunol., 28:489 (1991)). 5 Essentially, humanization by CDR-grafting involves transplanting only the CDRs onto human framework and constant regions. Theoretically, this should substantially eliminate immunogenicity (except if allotypic or idiotypic differences exist). Jones et al., Nature, 321:522-525 (1986); Verhoeyen et al., Science, 239:1534-1536 (1988), Riechmann etal., Nature, 332:323-327 (1988). However, CDR-grafting by itself may not yield the desired 3 result. Rather, it has been reported that some framework residues of the original antibody may also need to be preserved in order to preserve antigen binding activity. Riechmann et aL, Nature, 332:323-327 (1988); Queen et al., Proc. Natl. Acad. Sci. USA, 86:10023-10029; Tempest et al., Biol. Technology, 9:266-271 (1991); Co et al., Nature, 351:501-502 (1991). -4 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 As discussed above, in order to preserve the antigen-binding properties of the original antibody, the structure of its combining site must be faithfully reproduced in the humanized molecule. X-ray crystallographic studies have shown that the antibody combining site is built primarily from CDR residues, although some neighboring framework 5 residues have been found to be involved in antigen binding. Amit et aL., Science, 233:747 753 (1986); Colman et aL., Nature, 326:358-363 (1987); Sheriff et al., Proc. Natl. Acad. Sci. USA, 84:8075-8079 (1987); Padlan et al., Proc. Natl. Acad. Sci. USA, 86:5938-5942 (1989); Fischmann et aL., J. Biol. Chem., 266:12915-12920 (1991); Tulip et al., J. Molec. Biol., 227:122-148 (1992). It has also been found that the structures of the CDR loops are 0 significantly influenced by surrounding framework structures. Chothia et al., J. Molec. Biol., 196:901-917 (1987); Chothia et aL., Nature, 342:877-883 (1989); Tramomonteno et aL., J. Molec. BioL., 215:175-182 (1990). In addition to the effect of the framework residues on the CDRs, small but significant differences from the relative disposition of the variable light chain (VL) and variable heavy 5 (V ) domains have been noted (Colman et aL., Nature, 326:358-363 (1987)) and those differences are ostensibly due to variations in the residues involved in the inter-domain contact (Padlan et aL., Molec. ImmunoL., 31:169-217 (1994)). Furthermore, structural studies on the effect of the mutation of interior residues, in which changes in side chain volume are involved, have shown that the resulting local 0 deformations are accommodated by shifts in side chain positions that are propagated to distant parts of the molecular interior. This suggests that during humanization the interior residues in the variable domains and in the interface between these domains, or at least the interior volumes, should also be maintained; a humanization protocol in which an interior residue is replaced by one of different properties, such as size, charge, or hydrophobicity, 5 could result in a significant modification of the antigen combining-site structure. One method of potentially identifying the framework residues which need to be preserved is by computer modeling. Alternatively, critical framework residues may potentially be identified by comparing known antibody combining site structures. Padlan, Molec. Immun., 31(3):169 217 (1994). 0 The residues which potentially affect antigen binding fall into several groups. The first group comprises residues that are contiguous with the combining site surface and which could therefore make direct contact with antigens. These residues include the amino terminal residues and those adjacent to the CDRs. The second group includes residues that could alter the structure or relative alignment of the CDRs by contacting either the CDRs or -5 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 the opposite chains. The third group comprises amino acids with buried side chains that could influence the structural integrity of the variable domains. The residues in these groups are usually found in the same positions (id.) according to the adopted numbering system. See Kabat et aL., Sequences of Proteins of Immunological Interest, NIH Pub. No. 91-3242 5 (5th ed., 1991) (U.S. Dept. Health & Human Services, Bethesda, MD) and Genbank. Given these effects of changes in amino acid residues, although humanized antibodies are desirable because of their potential low immunogenicity in humans, their production is unpredictable. For example, sequence modification of antibodies may result in substantial or even total loss of antigen binding affinity, or loss of binding specificity. 0 Alternatively, "humanized antibodies" may still exhibit immunogenicity in humans, irrespective of sequence modification. Thus, there still exists a significant need in the art for novel humanized antibodies to desired antigens. More specifically, there exists a need in the art for humanized antibodies specific to CEA, because of their potential as improved immunotherapeutic and 5 immunodiagnostic agents for treatment and diagnosis of cancers expressing CEA, e.g., gastrointestinal and colorectal cancers, breast cancers, lung cancers, and ovarian cancers, among others. OBJECTS OF THE INVENTION 0 Toward this end, it is an object of the invention to provide humanized antibodies which are specific to human carcinoembryonic antigen (CEA), i.e. anti-CEA ("aCEA") antibodies. More specifically, it is an object of the invention to provide humanized antibodies derived from murine aCEA antibodies and in particular from COL-1, a specific murine antibody of the IgG2a isotype having high affinity for CEA. 5 It is also an object of the invention to provide pharmaceutical compositions containing humanized xCEA antibodies. It is a more specific object of the invention to provide pharmaceutical compositions containing humanized antibodies derived from the high affinity murine aCEA antibody, COL-1. It is another specific object of the invention to provide methods of using humanized 0 uCEA antibodies for treatment of cancers which express CEA, in particular breast, lung, ovarian, gastrointestinal, and colorectal cancers, among others. It is another object of the invention to provide immunodiagnostic compositions for detecting cancer cells, the compositions containing a humanized XCEA antibody, preferably derived from COL-1, which antibody is in labeled or unlabeled form. It is another object of -6 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 the invention to provide a method of immunodiagnosis of cancer using compositions which contain a humanized xCEA antibody, preferably derived from COL-1, which is in labeled or unlabeled form. It is still another object of the invention to provide nucleic acid sequences which 5 encode humanized aCEA antibodies or fragments thereof. It is a more specific object of the invention to provide nucleic acid sequences which encode humanized antibodies derived from the high affinity murine aCEA antibody, COL-1. It is another object of the invention to provide vectors which provide for the expression of humanized cXCEA antibodies, in particular humanized antibodies derived from the high affinity murine aCEA antibody, COL-1. 0 BRIEF DESCRIPTION OF THE FIGURES Figure 1 contains an alignment of the amino acid sequences of murine COL-1 VH (COL1MuVH), a NEWM FR template, and a humanized NEWM-based VH (COL1NMVH or "HuVH"). The CDRs are boxed. Murine FR residues retained in the various humanized VHs 5 exemplified herein are indicated by the symbols (T) , (A) , (T) , (s) , (Tr) , (A), and (Y) , according to the table below. Murine Residues COL1NMVH Retained at 0 Version Position * 24 72 76 78 79 80 HuVH 1 HuVHA 1 A HuVHAT IT A T 5 HuVHAA IA A HuVHAY TA Y HuVHATAY IA Tr A Y HuVHASTAY TA s Tif A Y HuVHT I T 0 HuVHS I S HuVHSTAY T S Tf A Y ( -- Retained murine residues indicated by the symbol (1) are F-27, N-28, 1-29, K-30, N 97, and T-98.) HuVHSTAY is the version of COL1NMVH expressed from the deposited cell 5 line, ATCC CRL-12208. Figure 2 contains an alignment of the amino acid sequences of murine COL-1 VK (COL1MuVK or "HuVK"), an REI FR template, and a humanized REI-based VK (COL1 REVK). The CDRs are boxed. The murine FR residues retained in the humanized -7 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 sequence are indicated by the symbols: (F) for HuVKF, and (v) and (L) for HuVKVL. HuVKVL is the version of COL1 REVK expressed from the deposited cell line, ATCC CRL 12208. 5 Figure 3 shows the IgG1 expression vectors used to express the subject humanized antibodies in NSO myeloma cells. Figure 4 shows binding of different COL-1 antibodies to CEA, as measured by an ELISA assay. 0 Figure 5 contains results of ELISA assays with COL-1 antibodies including those produced according to the invention. Figure 6 contains results of ELISA assays with COL-1 antibodies including those 5 produced according to the invention. Figure 7 contains results of ELISA assays with COL-1 antibodies including those produced according to the invention. 0 Figure 8 contains results of ELISA assays with COL-1 antibodies including those produced according to the invention. Figure 9 contains results of ELISA assays with COL-1 antibodies including those produced according to the invention. 5 Figure 10 contains results of ELISA assays with COL-1 antibodies including those produced according to the invention. Figure 11 contains results of ELISA assays with COL-1 antibodies including those 0 produced according to the invention. Figure 12 contains results of ELISA assays with COL-1 antibodies including those produced according to the invention. -8 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 Figure 13 contains the amino acid sequences of the humanized VH expressedby the deposited cell line ATCC CRL-12208. Figure 14 contains the amino acid sequences of the humanized VK expressed by the 5 deposited cell line ATCC CRL-12208. Figure 15 presents the nucleotide sequence of the DNA template use to produce the initial humanized COL-1 heavy chain variable region, HuVH. .0 Figure 16 presents the nucleotide sequence of the DNA template used to produce a variety of HuVH derivatives. Figure 17 presents the nucleotide sequence of the DNA template used to produce the initial humanized COL-1 light chain variable region, HuVK. 5 Figure 18 presents the nucleotide sequence of the DNA template used to produce a the HuVKVL derivative of HuVK. DETAILED DESCRIPTION OF THE INVENTION 0 Prior to setting forth the invention, definitions of certain terms which are used in this disclosure are set forth below. Antibody - This refers to single chain, two-chain, and multi-chain proteins and glycoproteins belonging to the classes of polyclonal, monoclonal, chimeric, and hetero 5 immunoglobulins (monoclonal antibodies being preferred); it also includes synthetic and genetically engineered variants of these immunoglobulins. "Antibody fragment" includes Fab, Fab', F(ab') 2 , and Fv fragments, as well as any portion of an antibody having specificity toward a desired target epitope or epitopes. Humanized antibody - This will refer to an antibody derived from a non-human 0 antibody, typically murine, that retains or substantially retains the antigen-binding properties of the parent antibody but which is less immunogenic in humans. This may be achieved by various methods including (a) grafting only the non-human CDRs onto human framework and constant regions with or without retention of critical framework residues, or (b) transplanting the entire non-human variable domains, but "cloaking" them with a human-like -9 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 section by replacement of surface residues. Such methods as are useful in practicing the present invention include those disclosed in Jones et al., Morrison et al., Proc. Natl. Acad. Sci. USA, 81:6851-6855 (1984); Morrison and Oi, Adv. Immunol., 44:65-92 (1988); Verhoeyen et aL., Science, 239:1534-1536 (1988); Padlan, Molec. Immun., 28:489-498 5 (1991); Padlan, Molec. Immun., 31(3):169-217 (1994). Complementarity Determining Region, or CDR - The term CDR, as used herein, refers to amino acid sequences which together define the binding affinity and specificity of the natural Fv region of a native immunoglobulin binding site as delineated by Kabat et al (1991). 0 Framework Region - The term FR, as used herein, refers to amino acid sequences interposed between CDRs. These portions of the antibody serve to hold the CDRs in an appropriate orientation for antigen binding. Constant Region - The portion of the antibody molecule which confers effector functions. In the present invention, murine constant regions are substituted with human 5 constant regions. The constant regions of the subject chimeric or humanized antibodies are derived from human immunoglobulins. The heavy chain constant region can be selected from any of the five isotypes: alpha, delta, epsilon, gamma or mu. Further, heavy chains of various subclasses (such as the IgG subclasses of heavy chains) are responsible for different effector functions and thus, by choosing the desired heavy chain constant region 0 chimeric antibodies with desired effector function can be produced. Preferred constant regions are gamma 1 (IgG1), gamma 3 (lgG3) and gamma 4 (IgG4). More preferred is a constant region of the gamma 1 (IgG1) isotype. The light chain constant region can be of the kappa or lambda type, preferably of the kappa type. Chimeric antibody - This is an antibody containing sequences derived from two 5 different antibodies, which typically are of different species. Most typically chimeric antibodies comprise human and murine antibody fragments, generally human constant and murine variable regions. Mammals - Animals that nourish their young with milk secreted by mammary glands, preferably warm blooded mammals, more preferably humans. O Immunogenicity - A measure of the ability of a targeting protein or therapeutic moiety to elicit an immune response (humoral or cellular) when administered to a recipient. The present invention is concerned with the immunogenicity of the subject humanized antibodies or fragments thereof. -10 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 Humanized reduced immunogenicity - This refers to a humanized antibody exhibiting reduced immunogenicity relative to the parent antibody, typically a murine antibody such as COL-1. Humanized antibody substantially retaining the binding properties of the parent 5 antibody - This refers to a humanized antibody which retains the ability to specifically bind the antigen recognized by the parent antibody used to produce such humanized antibody. Preferably the humanized antibody will exhibit the same or substantially the same antigen binding affinity and avidity as the parent antibody, e.g., COL-1. Ideally, the affinity of the antibody will not be less than 5% of the parent antibody affinity, more preferably not less .0 than about 30%, and most preferably the affinity will not be less than 50% of the parent antibody. Methods for assaying antigen-binding affinity are well known in the art and include half-maximal binding assays, competition assays, and Scatchard analysis. Suitable antigen binding assays are described in this application. In its broadest embodiment, the present invention is directed to humanized 5 antibodies which specifically bind CEA, an antigen expressed by various human cancers, in particular gastrointestinal, colorectal, breast, lung, and ovarian cancers. Preferably, such humanized antibodies will be derived from antibodies having good binding affinity to CEA, such as COL-1 through COL-15 disclosed by Muraro et al., Cancer Res., 45(11 Pt. 2):5769 5780 (1985). 0 Most preferably, such humanized antibodies will be derived from COL-1, a murine antibody of the IgG2a isotype, which has been reported to bind to CEA with high affinity (1.4x109 M') with no detectable cross-reactivity for CEA-related antigens, such as the non specific cross-reacting antigen (NCA) and the normal fecal antigen (NFA). As discussed above, humanized antibodies afford potential advantages over murine 5 and also chimeric antibodies, e.g., reduced immunogenicity in humans. This is advanta geous because it should reduce and potentially eliminate the eliciting of a HAMA response when such humanized antibodies are administered in vivo, e.g., either for treatment of cancer or for diagnosis of cancer as by tumor imaging. Also, such antibodies may exhibit improved plasma clearance, pharmacokinetic, and tumor targeting properties. 0 However, as noted above, humanization may in some instances adversely affect antigen binding. Preferably, the humanized caCEA antibodies of the present invention will possess a binding affinity for CEA of not less than about 5% and more preferably not less than about 30%, and most preferably not less than 50% of the CEA binding antigen affinity of the parent murine antibody, preferably COL-1. Most preferably, the humanized antibodies -11 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 of the present invention will possess a binding affinity for CEA of not less than about 5% and more preferably not less than about 30% and most preferably not less than about 50% of the CEA binding affinity of COL-1, or a chimeric antibody derived therefrom. Preferably, the humanized antibodies of the present invention will bind the same 5 epitope as COL-1. Such antibodies can be identified based on their ability to compete with COL-1 for binding to CEA or to CEA-expressing cells. In general, the subject humanized antibodies are produced by obtaining nucleic acid sequences encoding the variable heavy (VH) and variable light chains (VL, e.g., VK) of an antibody which binds CEA (preferably COL-1), identifying the CDRs in said VH and VL 0 sequences, and grafting such CDR-encoding nucleic acid sequences onto selected human framework-encoding nucleic acid sequences. Preferably, the human framework amino acid sequences are selected such that the resulting antibody is likely to be suitable for in vivo administration in humans. This can be determined, e.g., based on previous usage of antibodies containing such human FRs. 5 Preferably, the human FRs will not themselves be significantly immunogenic. Examples of such human frameworks include NEWM and REI. Alternatively, the amino acid sequences of the FRs of the antibody to be humanized (e.g., COL-1) will be compared to those of known human FRs, and the human FRs to be used for CDR-grafting will be selected based on their comprising sequences highly similar to 0 those of the parent antibody, e.g., a murine antibody which binds CEA. Numerous human FRs have been isolated and their sequences reported in the literature. This enhances the likelihood that the resultant CDR-grafted "humanized" antibody, which contains the CDRs of the parent (e.g., murine) antibody grafted onto the selected human FRs, will substantially retain the antigen binding structure and thus retain the binding affinity of the parent antibody. 5 As a result of such studies, the FRs of REI and NEWM antibodies have been identified as having amino acid sequences which are likely to allow the CDRs of COL-1 To retain a significant degree of antigen binding affinity. As noted, the selected human framework regions will preferably be those that are expected to be suitable for in vivo administration, i.e., not immunogenic. Based on their amino acid sequences, REI and NEWM human 0 framework regions are expected to be substantially non-immunogenic. In either method, the DNA sequences encoding the VH and VL regions of the preferably murine oxCEA antibody must be obtained. Methods for cloning nucleic acid sequences encoding immunoglobulins are well known in the art. Such methods will generally involve the amplification of the immunoglobulin-encoding sequences to be cloned -12 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 using appropriate primers by polymerase chain reaction (PCR). Primers suitable for amplifying immunoglobulin nucleic acid sequences, and specifically murine variable heavy and variable light sequences, have been reported in the literature. After such immunoglobulin-encoding sequences have been cloned, they will be sequenced by methods 5 well known in the art. This will be effected in order to identify the VH- and V,-encoding sequences, and more specifically the portions thereof which encode the CDRs and FRs. This can be effected by well known methods which include, e.g., those disclosed in U.S. Patent No. 4,816,397 to Boss et al. and U.S. Patent No. 5, 225,539 to Winter et aL.. Once the DNA sequences encoding the CDRs and FRs of the antibody which is to be 0 humanized have been identified, the amino acid sequences encoding the CDRs are then identified (deduced based on the nucleic acid sequences and the genetic code and by comparison to previous antibody sequences) and the CDR-encoding nucleic acid sequences are grafted onto selected human FR-encoding sequences. This may be accomplished by use of appropriate primers and linkers. Methods for selecting suitable primers and linkers to 5 provide for ligation of desired nucleic acid sequences is well within the purview of the ordinary artisan. As discussed above, the selected human FRs used for humanization will preferably be those that are likely to be suitable for in vivo administration, i.e. they are not in themselves immunogenic in humans (e.g., because of allotypic differences); examples 0 thereof are REI and NEWM human FRs. Alternatively, the human FRs will be selected such that they comprise amino acid sequences which are highly similar to those of the parent antibody's FR sequences. This may be effected by comparing the amino acid sequences of the murine FRs to those of previously reported human FRs (see, e.g., Kabat et aL., id.). After the CDR-encoding sequences are grafted onto the selected human FR 5 encoding sequences, the resultant DNA sequences encoding the "humanized" variable heavy and variable light sequences will then be expressed to produce a humanized Fv or humanized antibody which binds CEA. Typically, the humanized VH and VL sequences will be expressed as part of a whole uCEA antibody molecule, i.e. as a fusion protein with human constant domain sequences whose encoding DNA sequences have been obtained 0 from a commercially available library or which have been obtained using, e.g., one of the above-described methods for obtaining DNA sequences. However, the V H and VL sequences can also be expressed in the absence of constant sequences to produce a -13 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 humanized ocCEA Fv. Nevertheless, fusion of human constant sequences is potentially desirable because the resultant humanized (cCEA antibody may possess human effector functions such as CDC and ADCC activity. Methods for synthesizing DNA encoding a protein of known sequence are well known 5 in the art. Using such methods, DNA sequences which encode the subject humanized VL and V, sequences (with or without constant regions) are synthesized, and then expressed in a vector system suitable for expression of recombinant antibodies. This may be effected in any vector system which provides for the subject humanized VL and V, sequences to be expressed as a fusion protein with human constant domain sequences and to associate to D produce functional (antigen binding) antibodies or antibody fragments. Useful methods are set forth, e.g., in U.S. Patent 4,816,397 to Boss et al. and U.S. Patent 5,225,539 to Winter et al. Expression vectors and host cells suitable for expression of recombinant antibodies and humanized antibodies in particular, are well known in the art. The following references 5 are representative of methods and vectors suitable for expression of recombinant immuno globulins which may be utilized in carrying out the present invention: Weidle et al., Gene, 51: 21-29 (1987); Dorai et al., J. Immunol., 13(12):4232-4241 (1987); De Waele et al., Eur. J. Biochem., 176:287-295 (1988); Colcher et al., Cancer Res., 49:1738-1745 (1989); Wood et al., J. Immunol., 145(a):3011-3016 (1990); Bulens et al., Eur. J. Biochem., 195:235-242 3 (1991); Beggington etal., Biol. Technology, 1i0:169 (1992); King etal., Biochem. J., 281:317 323 (1992); Page et al., Bio/. Technology, 9:64 (1991 ); King et al., Biochem. J., 290:723-729 (1993); Chaudary et al., Nature, 339:394-397 (1989); Jones et al., Nature, 321:522-525 (1986); Morrison and Oi, Adv. Immunol., 44:65-92 (1988); Benhar et al., Proc. Natl. Acad. Sci. USA, 91:12051-12055 (1994); Singer et al., J. Immunol., 150:2844-2857 (1993); Cooto 5 etal., Hybridoma, 13(3):215-219 (1994); Queen etal., Proc. Natl. Acad. Sci. USA, 86:10029 10033 (1989); Caron et al., Cancer Res., 32:6761-6767 (1992); Cotoma et al., J. Immunol. Meth., 152:89-109 (1992). Moreover, vectors suitable for expression of recombinant antibodies are commercially available. The vector may, e.g., be a bare nucleic acid segment, a carrier-associated nucleic acid segment, a nucleoprotein, a plasmid, a virus, a Sviroid, or a transposable element. Host cells known to be capable of expressing functional immunoglobulins include, e.g.: mammalian cells such as Chinese Hamster Ovary (CHO) cells; COS cells; myeloma cells, such as NSO and SP2/O cells; bacteria such as Escherichia coli; yeast cells such as Saccharomyces cerevisiae; and other host cells. Of these, CHO cells are used by many -14 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 researchers given their ability to effectively express and secrete immunoglobulins. NSO cells are one of the preferred types of host cells useful in the present invention. Essentially, recombinant expression of humanized antibodies is obtained by one of two general methods. In the first method, the host cells are transfected with a single vector 5 which provides for the expression of both VH and VL variable sequences optionally fused to selected constant regions. In the second method, host cells are transfected with two vectors, each of which provides for expression of either the VH or VL sequence, each optionally fused to a selected constant region. Human constant domain sequences are well known in the art, and have been 0 reported in the literature. Preferred human constant light chain sequences (CL) include the kappa and lambda constant light sequences. Preferred human constant heavy chain sequences include human gamma 1, human gamma 2, human gamma 3, human gamma 4, and mutated versions thereof which provide for altered effect or function, e.g., enhanced in vivo half-life, reduced Fc receptor binding, and the like. 5 After expression, the antigen binding affinity of the resultant humanized antibody will be assayed by known methods, e.g., Scatchard analysis. Ideally, the antigen-binding affinity of the humanized antibody will approximate that of the parent antibody, e.g., COL-1. As discussed above, ideally the affinity of the humanized antibody will not be less than 5% of the parent antibody, more preferably not less than 30%, and most preferably not less than 0 50% of that of the parent antibody, e.g., COL-1. In some instances, humanized antibodies produced by grafting CDRs (from an antibody which binds CEA) onto selected human FRs may provide humanized antibodies having the desired affinity to CEA. However, it may be necessary or desirable to further modify specific residues of the selected human FR in order to enhance antigen binding. This 5 may occur because it is believed that some framework residues are essential to or at least affect antigen binding. Preferably, those framework residues of the parent (e.g., murine) antibody which maintain or affect combining-site structures will be retained. These residues may be identified by X-ray crystallography of the parent antibody or Fab fragment, thereby identifying the three-dimensional structure of the antigen-binding site. 0 These residues may potentially be identified by X-ray crystallography of the parent Fab, thereby identifying the three-dimensional structure of the antigen-binding site. Also, framework residues which may be involved in antigen binding may be putatively selected based on previously reported humanized murine antibody sequences. Thus, it may be beneficial to retain these and other murine framework residues from the parent murine -15 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 antibody to optimize CEA binding. However, because of inherent unpredictability associated with amino acid modification of proteins, and antibodies in particular, the effects of such changes, if any, on antigen binding are unpredictable. Nevertheless, such methodology will ideally confer a "human-like" character to the resultant humanized antibody thus rendering it 5 less immunogenic while retaining the interior and contacting residues which affect antigen binding. The present invention further embraces variants and equivalents which are substantially homologous to the humanized antibodies and antibody fragments set forth herein. These may contain, e.g., conservative substitution mutations, i.e. the substitution of D one or more amino acids by similar amino acids. For example, conservative substitution refers to the substitution of an amino acid with another within the same general class, e.g., one acidic amino acid with another acidic amino acid, one basic amino acid with another basic amino acid, or one neutral amino acid by another neutral amino acid. What is intended by a conservative amino acid substitution is well known in the art. 5 The phrase "substantially homologous" is used in regard to the similarity of a subject amino acid sequence (of an oligo- or poly-peptide or protein) to a related, reference amino acid sequence. This phrase is defined as at least about 75% "correspondence" -- i.e. the state of identical amino acid residues being situated in parallel -- between the subject and reference sequences when those sequences are in "alignment," i.e. when a minimal number D of "null" bases have been inserted in the subject and/or reference sequences so as to maximize the number of existing bases in correspondence between the sequences. "Null" bases are not part of the subject and reference sequences; also, the minimal number of "null" bases inserted in the subject sequence may differ from the minimal number inserted in the reference sequence. In this definition, a reference sequence is considered "related" to a Subject sequence where both amino acid sequences make up proteins or portions of proteins which are either aCEA antibodies or antibody fragments with aCEA binding affinity. Each of the proteins comprising these ccCEA antibodies or antibody fragments may independently be antibodies or antibody fragments or bi- or multi-functional proteins, e.g., such as fusion proteins, bi- and multi-specific antibodies, single chain antibodies, and the ) like. The present invention is further directed to nucleic acid sequences from which such humanized antibodies and antibody fragments may be expressed, as well as expression vectors from which these humanized antibodies and antibody fragments may be expressed in transfected host cells. -16 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 In a preferred embodiment, such humanized antibodies and corresponding nucleic acid sequences will be derived from COL-1. Most preferably, the humanized V, sequence and the humanized VL sequence will have the sequences substantially as depicted in Figure 1 or 13 or in Figure 2 or 14, respectively, and as discussed in the Examples set forth below. 5 However, as discussed, the invention further contemplates other modifications of these humanized VH and VL sequences, e.g., sequences which further comprise one or more con servative amino acid substitutions or which retain one or more additional murine framework residues which affect or do not significantly reduce antigen binding. The subject humanized antibodies -- because they specifically bind CEA (an antigen 0 expressed on many different cancer cell types, e.g., lung carcinomas, breast carcinomas, gastrointestinal carcinomas such as stomach cancers, colorectal carcinomas such as colon cancers, ovarian carcinomas, etc.) and further because they will not be significantly immunogenic in humans -- should be suitable for use as: therapeutics for the treatment or prevention of cancers characterized by CEA expression; diagnostic agents, e.g., for 5 diagnosis and evaluating the prognosis of cancers characterized by CEA expression (based on levels of CEA expression); tumor imaging agents; or radiolabeled antibodies in the Radioimmunoguided Surgery@ System (RIGS®). See Hinkle et aL., Antibody, Immunoconjugates and Radiopharmaceuticals, 4(3):339-358 (1991). One skilled in the art would be able (by routine experimentation) to determine what 0 amount of antibody would be effective and non-toxic for the purpose of treating cancer. Generally, however, an effective dosage will be in the range of about 0.05 to 100 milligrams per kilogram body weight per day. The humanized antibodies or humanized antibody fragments of the invention may be administered to a human or other animal in accordance with the aforementioned methods of 5 treatment in an amount sufficient to produce a therapeutic or prophylactic effect. The antibodies of the subject invention can be administered to such human or other animal in a conventional dosage form prepared by combining the antibody of the invention with a conventional, pharmaceutically acceptable carrier, diluent, and/or excipient according to known techniques. It will be recognized by one of ordinary skill in the art that the form and 0 character of the pharmaceutically acceptable carrier, diluent, and/or excipient is dictated by the amount of active ingredient with which it is to be combined, the route of administration, and other well-known variables. Pharmaceutically acceptable formulations may include, e.g., a suitable solvent, preservatives such as benzyl alcohol if desired, and a buffer. Useful solvent may include, -17 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 e.g., water, aqueous alcohols, glycols, and phosphonate and carbonate esters. Such aqueous solutions contain no more than 50% by volume of organic solvent. Suspension type formulations may include a liquid suspending medium as a carrier, e.g., aqueous polyvinylpyrrolidone, inert oils such as vegetable oils or highly refined mineral oils, or 5 aqueous cellulose ethers such as aqueous carboxymethylcellulose. A thickener such as gelatin or an alginate may also be present, one or more natural or synthetic surfactants or antifoam agents may be used, and one or more suspending agents such as sorbitol or another sugar may be employed therein. Such formations may contain one or more adjuvants. O The route of administration of the antibodies (or fragment thereof) of the present invention may be oral, parenteral, by inhalation, or topical. The term parenteral as used herein includes intravenous, intramuscular, subcutaneous, rectal, vaginal, or intraperitoneal administration. The subcutaneous, intravenous, and intramuscular forms of parenteral administration are generally preferred. 5 The daily parenteral and oral dosage regimens for prophylactically or therapeutically employing humanized antibodies of the present invention will generally be in the range of about 0.005 to 100, but preferably about 0.5 to 10, milligrams per kilogram body weight per day. The antibodies of the present invention may also be administered by inhalation. By 0 "inhalation" is meant intranasal and oral inhalation administration. Appropriate dosage forms for such administration, such as an aerosol formulation or a metered dose inhaler, may be prepared by conventional techniques. The preferred dosage amount of a compound of the invention to be employed is generally within the range of about 0.1 to about 100, more preferably about 10 to 100, milligrams per kg body weight. 5 The antibody of the invention may also be administered topically. By topical administration is meant non-systemic administration. This includes the administration of a humanized antibody (or humanized antibody fragment) formulation of the invention externally to the epidermis or to the buccal cavity, and instillation of such an antibody into the ear, eye, or nose, and wherever it does not significantly enter the bloodstream. By systemic 0 administration is meant oral, intravenous, intraperitoneal, subcutaneous, and intramuscular administration. The amount of an antibody required for therapeutic, prophylactic, or diagnostic effect will, of course, vary with the antibody chosen, the nature and severity of the condition being treated and the animal undergoing treatment, and is ultimately at the -18 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 discretion of the physician. A suitable topical dose of an antibody of the invention will generally be within the range of about 1 to 100 milligrams per kilogram body weight daily. Formulations 5 While it is possible for an antibody or fragment thereof to be administered alone, it is preferable to present it as a pharmaceutical formulation. The active ingredient may comprise, for topical administration, from 0.001% to 10% w/w, e.g., from 1% to 2% by weight of the formulation, although it may comprise as much as 10% w/w but preferably not in excess of 5% w/w and more preferably from 0.1% to 1% w/w of the formulation. O The topical formulations of the present invention, comprise an active ingredient together with one or more acceptable carrier(s) therefor and optionally any other therapeutic ingredients(s). The carrier(s) must be "acceptable" in the sense of being compatible with the other ingredients of the formulation and not deleterious to the recipient thereof. Formulations suitable for topical administration include liquid or semi-liquid 5 preparations suitable for penetration through the skin to the site of where treatment is required, such as liniments, lotions, creams, ointments or pastes, and drops suitable for administration to the eye, ear, or nose. Drops according to the present invention may comprise sterile aqueous or oily solutions or suspensions and may be prepared by dissolving the active ingredient in a 0 suitable aqueous solution of a bactericidal and/or fungicidal agent and/or any other suitable preservative, and preferably including a surface active agent. The resulting solution may then be clarified and sterilized by filtration and transferred to the container by an aseptic technique. Examples of bactericidal and fungicidal agents suitable for inclusion in the drops are phenylmercuric nitrate or acetate (0.002%), benzalkonium chloride (0.01%) and 5 chlorhexidine acetate (0.01%). Suitable solvents for the preparation of an oily solution include glycerol, diluted alcohol and propylene glycol. Lotions according to the present invention include those suitable for application to the skin or eye. An eye lotion may comprise a sterile aqueous solution optionally containing a bactericide and may be prepared by methods similar to those for the preparation of drops. 0 Lotions or liniments for application to the skin may also include an agent to hasten drying and to cool the skin, such as an alcohol or acetone, and/or a moisturizer such as glycerol or an oil such as castor oil or arachis oil. Creams, ointments or pastes according to the present invention are semi-solid formulations of the active ingredient for external application. They may be made by mixing -19 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 the active ingredient in finely-divided or powdered form, alone or in solution or suspension in an aqueous or non-aqueous fluid, with the aid of suitable machinery, with a greasy or non greasy basis. The basis may comprise hydrocarbons such as hard, soft or liquid paraffin, glycerol, beeswax, a metallic soap; a mucilage; an oil of natural origin such as almond, corn, 5 arachis, castor or olive oil; wool fat or its derivatives, or a fatty acid such as stearic or oleic acid together with an alcohol such as propylene glycol or macrogels. The formulation may incorporate any suitable surface active agent such as an anionic, cationic or non-ionic surface active such as sorbitan esters or polyoxyethylene derivatives thereof. Suspending agents such as natural gums, cellulose derivatives or inorganic materials such as silicaceous 0 silicas, and other ingredients such as lanolin, may also be included. Kits according to the present invention include frozen or lyophilized humanized antibodies or humanized antibody fragments to be reconstituted, respectively, by thawing (optionally followed by further dilution) or by suspension in a (preferably buffered) liquid vehicle. The kits may also include buffer and/or excipient solutions (in liquid or frozen form) 5 - or buffer and/or excipient powder preparations to be reconstituted with water - for the purpose of mixing with the humanized antibodies or humanized antibody fragments to produce a formulation suitable for administration. Thus, preferably the kits containing the humanized antibodies or humanized antibody fragments are frozen, lyophilized, pre-diluted, or pre-mixed at such a concentration that the addition of a predetermined amount of heat, of 0 water, or of a solution provided in the kit will result in a formulation of sufficient concentration and pH as to be effective for in vivo or in vitro use in the treatment or diagnosis of cancer. Preferably, such a kit will also comprise instructions for reconstituting and using the humanized antibody or humanized antibody fragment composition to treat or detect cancer. The kit may also comprise two or more component parts for the reconstituted active 5 composition. For example, a second component part - in addition to the humanized antibodies or humanized antibody fragments - may be bifunctional chelant, bifunctional chelate, or a therapeutic agent such as a radionuclide, which when mixed with the humanized antibodies or humanized antibody fragments forms a conjugated system therewith. The above-noted buffers, excipients, and other component parts can be sold 0 separately or together with the kit. It will be recognized by one of skill in the art that the optimal quantity and spacing of individual dosages of a humanized antibody or humanized antibody fragment of the invention will be determined by the nature and extent of the condition being treated, the form, route and site of administration, and the particular animal being treated, and that such optima can -20 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 be determined by conventional techniques. It will also be appreciated by one of skill in the art that the optimal course of treatment, i.e., the number of doses of an antibody or fragment thereof of the invention given per day for a defined number of days, can be ascertained by those skilled in the art using conventional course of treatment determination tests. 5 The subject humanized antibodies may also be administered in combination with other anti-cancer agents, e.g., other antibodies or drugs. Also, the subject humanized antibodies or fragments may be directly or indirectly attached to effector moieties having therapeutic activity. Suitable effector moieties include by way of example cytokines (IL-2, TNF, interferons, colony stimulating factors, IL-1, etc.), cytotoxins (Pseudomonas exotoxin, 0 ricin, abrin, etc.), radionuclides, such as 90 Y, 1311 9 9 m Tc, 111 'n, 1251, among others, drugs (methotrexate, daunorubicin, doxorubicin, etc.), immunomodulators, therapeutic enzymes (e.g., beta-galactosidase), anti-proliferative agents, etc. The attachment of antibodies to desired effectors is well known. See, e.g., U.S. Patent No. 5,435,990 to Cheng et al. Moreover, bifunctional linkers for facilitating such attachment are well known and widely 5 available. Also, chelators (chelants and chelates) providing for attachment of radionuclides are well known and available. Alternatively, the subject humanized caCEA antibodies or fragments may be used as immunodiagnostic agents both in vivo and in vitro. A particularly preferred usage is for in vivo imaging of cancer cell lesions which express CEA. The subject antibodies are preferred 0 because they should elicit no significant HAMA or allergic response. Thus, they may be used repeatedly to monitor the disease status of a patient. As noted above, another preferred application of the subject humanized antibodies or fragments thereof is in the Radioimmunoguided System® (RIGS®). This technique involves the intravenous administration of a radiolabeled monoclonal antibody or its fragment 5 prior to surgery. After allowing for tumor uptake and blood clearance of radioactivity, the patient is taken to the operating room where surgical exploration is effected with the aid of a hand-held gamma activity probe, e.g., the Neoprobe® 1000. This helps the surgeon identify the tumor metastases and lessen the complications of excision. The RIGS® system is advantageous because it allows for the detection of tumors not 0 otherwise detectable by visual inspection and/or palpation. See, O'Dwyer et al., Arch. Surg., 121:1391-1394 (1986). This technique is described in detail in Hinkle et al., Antibody, Immunoconjugates and Radiopharmaceuticals, 4:(3)339-358 (1991). This reference discloses the use of this technique with aCEA antibodies including the COL-1 monoclonal antibody. This technique is particularly useful for cancers of the colon, breast, pancreas, -21 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 and ovaries. Thus, this technique should be applicable to the subject humanized antibodies which react with CEA expressed by colon, breast, and ovarian cancers. Also, Hinkle et al. (id.) cite numerous references describing this technique. The subject humanized antibodies or fragments thereof may be radiolabeled with radionuclides which are suitable for in vivo 5 administration, e.g., iodine radionuclides such as 1311 and 1251. Moreover, "'In and 9 9 m Tc are also suitable. The subject humanized antibodies may be used alone or in combination with other antibodies. Also, the subject humanized antibodies may be prepared in the form of a diagnostically effective composition. Generally, this will entail the incorporation of 0 diagnostically acceptable carriers and excipients, and labels which provide for detection. Suitable labels include diagnostic radionuclides, enzymes, etc. Methods for using antibodies for tumor imaging are well known in the art. Without further elaboration, it is believed that one skilled in the art can, using the preceding description, utilize the present invention to its fullest extent. The invention will be 5 further clarified by a consideration of the following examples, which are intended to be purely exemplary of the present invention and thus to be construed as merely illustrative examples and not limitations of the scope of the present invention an any way. EXAMPLES 0 Materials and Methods DNA Template Preparation All recombination work was performed upon DNA sequences in plasmid M13 vectors. The source of the NEWM framework regions for producing the initial humanized COL-1 VH 5 was an M13 construct bearing -- between the M13 BamH I and Hind III sites -- a DNA segment having the nucleotide sequence shown in figure 13. The source of REI framework regions for producing the initial humanized COL-1 VL was an M13 construct bearing between the M13 BamH I and Hind Ill sites -- a DNA segment encoding the REI amino acid sequence of Figure 2. 0 When overlap-extension procedures were used to introduce mutations into a given DNA sequence, double stranded M13 DNA was utilized. In contrast, when extension-ligation procedures were used instead, the oligonucleotides were designed to anneal to only one of the two DNA strands. In this latter procedure, the M13 DNA was first treated to substitute uridine for every thymidine base in the DNA, to produce uridinylated DNA. This was -22 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 accomplished by transfecting the M13 plasmid DNA into competent cells lacking dUTPase and uracil glycosylase, normally RZ1032 cells (though CJ236 cells available from Bio-Rad of Hercules, CA, are also suitable) by combining the following ingredients. 1 gL of M13 plasmid DNA 5 4 mL of LB broth 40 PL of competent RZ1032 cells. The culture was shaken for 5 hours at 370C and the resulting single-stranded plasmid DNA (ssDNA) was isolated and dissolved in 50 PL Tris-EDTA buffer. The DNA was then 0 treated with uracil glycosylase by mixing together: 1 pL uridinylated ssDNA 1 gL 10 Ox glycosylase buffer 1U uracil glycosylase (Gibco BRL, Gathersburg, MD) 40 pL 25mM MgCI 2 . 5 This mixture was then incubated at 370C for one hour and then 6.6 tL 25mM MgCI 2 and 9.9ljL 1M NaOH. The mixture was then further incubated for 5 minutes at 370C and 16.5 gL of 0.6M HCI was then added to neutralize the mixture. The DNA was then ethanol precipitated and dissolved in water. 0 M13 Oligonucleotide Primers The following oligonucleotide primers were used throughout the process of preparing the humanized COL-1 VHs and VLs exemplified below. 10. 5'-CTAAAACGACGGCCAGT-3' and 5 11. 5'-AACAGCTATGACCATG-3' (both for using in producing VHs); and 385. 5'-GCGGGCCTCTTCGCTATTACGC-3' and 391. 5'-CTCTCTCAGGGCCAGGCGGTGA-3' (both for use in producing VKs). These primers are complementary to regions of the plasmid M13 which are external both to the (NEWM or REI) target framework sequences and to the BamH I site -to- Hind Ill site 0 section of M13. Murine Variable Regions In order to compare the antibody binding characteristics of the antibodies produced according to the examples set forth below, antibodies having a chimeric heavy 5 chain (i.e. a heavy chain having a murine COL-1 VH region and a human IgG1 constant region) and/or a chimeric light chain (i.e. a light chain having a murine COL-1 VL and a human K constant region) were expressed. The source of these chimeric chains was the -23 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 ATCC-deposited cell line CRL 11217 (Budapest) which expresses a chimeric COL-1 antibody having both chains. The heavy chain of this antibody was termed "MuVH" and the light chain thereof was termed "MuVL." 5 Oligonucleotide Phosphorylation Protocol Mutating oligonucleotides used in non-overlap extensions were phosphorylated according to the procedure below. In a final volume of 25.L, the following ingredients were combined: 10 pmol of each oligonucleotide, .0 5gL of a 5x polynucleotide kinase buffer, and 5U of T4 polynucleotide kinase (Gibco BRL). The phosphorylation reaction was started with the addition of the enzyme and allowed to proceed for one hour at 370C. 5 Annealing Protocol for Non-Overlap Extension-Ligations The annealing step for non-overlap extension-ligations involved performing one annealing in which all mutation-carrying oligonucleotides and one primer oligonucleotide were annealed to a single stranded DNA template in which all thymidine bases had been 0 replaced with uridine bases. The mutating oligonucleotides were first phosphorylated according to the above oligonucleotide phosphorylation protocol. In a final volume of 20 RL, the following ingredients were combined: 1 pmol of each mutation-carrying phosphorylated oligonucleotide 1 pmol of a primer oligonucleotide 5 4 RL 5x annealing buffer 0.2 pmol ssU-DNA template The mixture was then heated to 900C for 30 sec., then quickly cooled to 700C, and finally allowed to slowly cool to 370C. 0 Extension-Ligation Protocol for Non-Overlap Extension-Ligations After completion of the annealing step in which the primer and phosphorylated mutating oligonucleotides were annealed to the ssU-DNA template, extension-ligation was performed as follows. In a final volume of 30 PL, the following ingredients were combined: 20 IL annealed ssU-DNA (i.e. the contents of the above annealing procedure) 5 2 RL 5x annealing buffer 2 RL 0.1 M dithiothreitol 0.3 IL 0.1M ATP 14L 6.25mM dNTP mixture of equimolar amounts of dATP, dTTP, dGTP, dCTP -24 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 2.5U T7 DNA polymerase (USB, now Amersham Life Sciences, Cleveland, OH) 0.5U T4 DNA ligase (Gibco BRL) Water to 30jL 5 This mixture was then incubated at room temperature for 1 hour. Standard PCR Protocols The following procedure was used, alternately, both to amplify the non-overlap extension-ligation DNA sequences and to perform extension of each overlap DNA sequence. LO In a final volume of 50gL, the following ingredients were combined: 2 gL template DNA (either annealed ssU-DNA or non-annealed ssDNA) 5 pL 10x Vent buffer (NEB, i.e. New England Biolabs, Beverly, MA) or 10Ox Thermalase buffer (IBI of New Haven, CT) 2 gL 6.25mM dNTP mixture of equimolar amounts of dATP, dTTP, dGTP, dCTP L5 25 pmol of one oligonucleotide primer 25 pmol of either a mutation-carrying oligonucleotide (for overlap-extension) or a second oligonucleotide primer 1U Vent DNA polymerase (NEB) or Thermalase DNA polymerase (IBI) S0 Reactions were initiated with the addition of the DNA polymerase and then treated with about 15 cycles of: (1) 94 0 C for 30 sec., (2) 50 0 C for 30 sec., and (3) 30-60 seconds at either 75 0 C (for Vent DNA polymerase) or 72 0 C (for Thermalase). Reactions were brought to completion with 5 minutes at a constant temperature of either 75 0 C (for Vent DNA polymerase) or 72oC (for Thermalase). '5 PCR Overlap-Extension Amplification Protocol After a pair of PCR reactions were performed - one for each of the two (partially complementary) overlapping DNA segments, the two resulting segments were joined according to the following PCR procedure. In a final volume of 50 pL, the following 0 ingredients were mixed: 1 gL of each overlap DNA (from the above overlap PCR extension reactions) 5 gL 10Ox Vent buffer (NEB) or Thermalase buffer (IBI) 2 p.L 6.25mM dNTP mixture of equimolar amounts of dATP, dTTP, dGTP, dCTP 25 pmol of each oligonucleotide primer used in the overlap PCR extensions 5 1U Vent DNA polymerase (NEB) or Thermalase DNA polymerase (IBI) Reactions were initiated with the addition of the DNA polymerase and then treated with about 15 cycles of: (1) 94 0 C for 30 sec., (2) 50 0 C for 30 sec., and (3) 30-60 seconds at either 750C (for Vent DNA polymerase) or 720C (for Thermalase). Reactions were brought -25 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 to completion with 5 minutes at a constant temperature of either 750C (for Vent DNA polymerase) or 720C (for Thermalase). Transfer of Humanized COL-1 Variable Region DNA Sequences from M13 to pSV Vectors 5 and Subsequent Antibody Expression Humanized antibodies were expressed in pSV vectors grown in NSO cells as follows. The humanized variable region constructs which were produced in the plasmid, M13, were obtained by ethanol precipitating the M13 plasmids (as cytosolic DNA), redissolving them in .0 aqueous solution, and digesting them with 10OU each of Hind Ill and BamH I (both from BRL, i.e. Gibco BRL) for 1 hour at 370C in a final volume of 100 gL with Tris-EDTA buffer. The resulting DNA fragments were then run on a low melting point agarose gel, the band containing the humanized construct DNA was cut out, and the DNA was purified using an ELUTIP 'd' column with 20 .L Tris-EDTA buffer. 10 giL of the purified DNA preparation was 5 then combined with 1 gL of a Hind III and BamH I-digested pSV preparation, 3 gL of 5x ligase buffer, and 1U of T4 DNA ligase (BRL), in order to insert the construct into a pSV plasmid. Humanized COL-1 VH constructs were inserted into pSVgpt vectors bearing a human IgG1 heavy chain constant region; the pSVgpt vector used is the "aLYS-30" shown in Figure 5. Humanized COL-1 VL constructs were inserted into pSVhyg vectors bearing a 0 human K light chain constant domain; the pSVhyg vector used in the "aLys-17" shown in Figure 5. Each humanized variable region construct was inserted adjacent to the respective constant region, i.e. so as to replace either the HuVHLYS or the HuVLLys segment illustrated in Figure 5. The resulting vectors were transfected into NSO cells as follows. About 3 gg of the 5 VH vector, or about 6ig of the VL vector, produced by the pSV-insertion procedures, was then linearized by digestion with 10OU Pvul (Gibco BRL). The digested DNA was then precipitated with ethanol and redissolved in 50 pL of water. NSO cells were collected by centrifugation and resuspended in 0.5mL Dulbecco's Modified Eagle's Medium (DMEM) and then transferred to a Gene Pulser cuvette (Bio-Rad). The DNA from both one VH and one 0 VL construct was gently mixed with the cells by pipetting and the cuvette was left on ice for 5 minutes. Next, the cuvette was inserted between the electrodes of the Bio-Rad Gene Pulser and a single pulse of 170V at 960 pF was applied. The contents of the cuvette were then transferred to a flask containing 20mL DMEM and the cells were allowed to rest for 1-2 days at 370C. Cells were again harvested by centrifugation and resuspended in 36mL selective 5 DMEM. 1.5mL aliquots of this resuspension were placed in each well of a 24-well plate and -26 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 incubated at 370C for 4 days, at which time the medium in each well was replaced with 1.5mL of fresh selective DMEM. After 6 more days of incubation at 370C, surviving cell colonies were visible to the naked eye and the supernatants of each well were assayed for antibody production. Both whole antibody production (i.e. without purification) and purified 5 antibody production were assayed. To obtain purified antibodies, the supernatants were passed through a protein A column. ELISA Assay Protocols Antibody concentrations and antibody binding characteristics were tested using 0 enzyme-linked immunosorbent assay (ELISA) procedures which are set forth as follows Measurement of IgG concentration The concentration of IgG secreted from transfected cells was measured using an enzyme-linked immunosorbent assay (ELISA) procedure which is set forth as follows. 5 Polyvinyl chloride (PVC) microtiter plates (Dynatech Laboratories, Chantilly, VA, catalog # 001-010-2101) were coated with goat anti-human IgG (10Omg/mL, GAHIG, Southern Biotechnology Associates, Inc., Birmingham, AL, catalog # 2010-01) diluted with Milli-Q® water and placed on the plates using 50mL/well. Plates were air-dried overnight at ambient temperature or at 370C for 3 hours. Prior to use, non-specific binding was blocked 0 the addition of 0.2 mL/well of 1% (w/v) bovine serum albumin (Sigma, St. Louis, MO catalog # A7888) in phosphate buffered saline (Sigma, catalog # 1000-3) (PBS/BSA). All incubations were carried out in a humidified container. Plates were incubated for 1-2 hours at 370C and the blocking solution removed prior to sample addition. Two-fold serial dilutions of samples or a standard IgG solution set at 500 ng/mL (50 il/well) were made in triplicate in 5 the PBS/BSA solution. The plate was incubated at 370C for 3 hours or overnight at 40C. The plate was washed 3-times with 0.025% Tween-20 (v/v, Sigma) using an automatic plate washer. 50 pl/well of 1:1000 dilution of a goat anti-human IgG conjugated to Horseradish Peroxidase (Southern Biotechnology Associates Inc.) was added and incubated at 370C for 1.5 hours. The wells were washed 3 times with 0.025% Tween-20 (v/v, Sigma) using an 0 automatic plate washer and 50 pl/well OPD substrate buffer added. The color was developed for 4 minutes, stopped with 12.5 pl 12.5% H 2

SO

4 and the absorbance at 492 nm read. The concentration of IgG in the test sample was estimated by comparison of the mean of the optical densities to a standard curve constructed from the standard IgG. -27 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 Determination of relative affinities of humanized antibodies Antibody binding characteristics were tested in an ELISA using partially purified CEA antigen immobilized on Polyvinyl chloride (PVC) microtiter plates (Dynatech Laboratories, Chantilly, VA, catalog # 001-010-2101) 5 PVC plates were coated with 50 pl/well CEA (Dow Chemical, lot #040191), diluted 1:300 in Milli-Q water. Plates were air-dried overnight at ambient temperature or at 37 0 C for 3 hours. Prior to use, non-specific binding was blocked by the addition of 0.2 mL/well of 1% (w/v) bovine serum albumin (Sigma, St. Louis, MO catalog # A7888) in phosphate buffered saline (Sigma, catalog # 1000-3) (PBS/BSA). Plates were incubated for 1-2 hours at 37 0 C 0 and the blocking solution removed prior to sample addition. All incubations were carried out in a humidified container. Two-fold serial dilutions (starting concentration range of 1.0p~g/ml 10pg/ml) of the samples to be tested in the PBS/BSA solution were added to triplicate wells of the TAG-coated plate (50 pl/well). The plate was incubated overnight at 4°C or 1-2 hours at 37 0 C. The plate was washed 3-times with 0.025% Tween-20 (v/v, Sigma) using an 5 automatic plate washer. 50 pl/well of 1:1000 dilution of a goat anti-human IgG conjugated to Horseradish Peroxidase (Southern Biotechnology Associates Inc.) was added and incubated at 37 0 C for 1.5 hours. The wells were washed 3 times with 0.025% Tween-20 (v/v, Sigma) using an automatic plate washer and 50 pl/well OPD substrate buffer added. The color was developed for 4 minutes, stopped with 12.5 p1 12.5% H 2

SO

4 and the absorbance at 492 nm 0 read. Determination of Affinity Constants for binding to CEA Two-fold dilutions of purified Hu-COL-1 were prepared in PBS/BSA over a range of 1.0gg/ml - 0.003 gg/ml and samples (20 pl/well) were applied in triplicate to TAG coated PVC prepared and blocked as described supra. Plates were incubated overnight at 4oC. 5 Following this incubation, samples were transferred from the plate to the corresponding wells on the GAHIG-coated trap plate. The original TAG plate was washed 3-times with 0.025% Tween-20 (v/v, Sigma, catalog # P1379) using an automatic plate washer. An 1 25 1-labeled goat anti-human IgG probe (ICN Biomedicals, Inc., catalog # 68088) was diluted to 75,000 cpm/25p1l in PBS/BSA and added (25pl/well) to all wells. This TAG plate was incubated for 1 0 hour at 37 0 C. After a 1 hour incubation at 37 0 C, the trap plate was washed as described above and 1 2 5 1-labeled GAHIG probe was added. This plate was incubated for 1 hour at 37 0 C. Both -28 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 plates (TAG and GAHIG-trap) containing probe were then washed with a microplate washer to remove the unbound probe. A plate cutter (D. Lee, Sunnyvale, CA, Model HWC-4) was used to separate the wells from the plate frame. The radioactivity in each well was quantified by a gamma counter. The resulting data was analyzed according to the method 5 of Scatchard (Ann. NYAcad., 51:600-672 (1946)). EXAMPLE 1 Synthesis of Initial CDR-Grafted (Humanized) Antibody from Murine COL-1 0 We describe in this Example the construction of humanized COL-1 monoclonal antibodies ("Mabs") -- COL-1 HuVH/HuVK -- using the VL and VH frameworks of human Mabs REI and NEWM, respectively. The CDRs for murine COL-1 were grafted onto human frame works according to known methods as discussed supra. In particular, human frameworks were selected from antibodies which, based on previous studies, were predicted to be 5 suitable, i.e. which should not adversely affect antigen binding and not exhibit significant immunogenicity in humans. The human frameworks selected for the variable heavy and variable light chains, respectively, were NEWM and REI. In the production of the initial version of the humanized VH, certain murine framework residues were also retained which, based on previous studies, might allow retention of 0 antigen binding properties. Specifically, residues F27, N28, 129, K30, N97, and T98 of the murine heavy chain were initially retained. The production of this NEWM-grafted humanized COL-1 VH was accomplished using two rounds of a dual PCR procedure: the standard PCR protocol (for overlap-extension) followed by PCR amplification with oligonucleotide primers 10 and 11. This procedure was 5 carried out as described above using a single-stranded M13 DNA template bearing, between the Hind III and BamH I sites thereof, a DNA segment having the nucleotide sequence shown in Figure 13. The mutating oligonucleotides used in the first round (with Vent DNA polymerase)were designed and synthesized with the following sequences: D 836.5'-TGAGAATGGTGATACTGAATATGCCCCGAAGT; and 837. 5'-TCGGGGCATATTCAGTATCACCATTCTCAGGATC-3'. Those for the second round (with Thermalase) were: 5 838.5'-ACTATGATTACGACGCGTTGGTTCTTCGATGTCTGGGGCCAAGGGTCCTT GGTCACCGTC-3'; and 839.5'-ACGCGTCGTAATCATAGTAGATAGACCCCGTGTATTACAGTAATAGACCGCG GTG-3'. -29 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 The resulting humanized VH was named COL1 NMVH or "HuVH." In the production of the initial version of the humanized VL, no uniquely murine framework residues were retained. The production of the REI-grafted humanized COL-1 VL 5 was accomplished by using two rounds of a procedure involving performing the annealing and extension-ligation protocols described above - followed by amplification by the standard Thermalase PCR protocol -- using a ssM13 template bearing, between the Hind III and BamH I sites thereof, a ssU-DNA segment produced by the above procedure using the REIVK sequence shown in Figure 17 (or in the second round, the mutated M13 template 0 resulting from the first round). The mutating oligonucleotides used in round number 1 were designed and synthesized with the following sequences: 842. 5'-TATAGCCAGATGCACTGACACTTTTGCTGGCCCTACAGGTGATG-3'; 844. 5'-GCTCTGGGTCATCTGGATGTCGG-3'; 5 849.5'-TTCTACTCACGTGTGATTTGCAGCTTGGTCCCTTGGCCGAACGTAGGAAG CTCCCTACTGTGCTGGCAGTAG-3'; 850.5'-ATGGTGAAGGTGTAGTCGGTACCGC-3'; and 851. 5'-GCCTTACCTGGCGTCTGCTGGTACC-3'. 0 The mutating oligonucleotide used in the second round was: 841. 5'-GCACACCAGATTGTAGGTTGGATGCAAGG-3'. The resulting humanized VL was called "COL1 REVK" or "HuVK." 5 Concurrently, a second humanized COL-1 VL was made by the same procedure using the following mutating oligonucleotides. For round 1: 0 842. 5'-TATAGCCAGATGCACTGACACTTTTGCTGGCCCTACAGGTGATG-3'; 844. 5'-GCTCTGGGTCATCTGGATGTCGG-3'; 849.5'-TTCTACTCACGTGTGATTTGCAGCTTGGTCCCTTGGCCGAACGTAGGAAG CTCCCTACTGTGCTGGCAGTAG-3'; and 5 851. 5'-GCCTTACCTGGCGTCTGCTGGTACC-3'. For round 2: 841. 5'-GCACACCAGATTGTAGGTTGGATGCAAGG-3'. 0 The resulting VL was termed "HuVKF." A number of amino acid substituted-versions of HuVH and HuVK were also constructed, using the above-described overlap extension technique employing oligonucleotide primers 5 10 and 11 or primers 385 and 391, followed by PCR amplification with Vent DNA -30 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 polymerase. The mutating oligonucleotides and DNA templates are described below under their resulting humanized variable region name: HuVHT (using the HuVH DNA shown in Figure 16 as a template) 5 954. 5'-GACAATGCTGACAGACACCAGCAA-3' and 955. 5'-TGCTGGTGTCTGTCAGCATTGTCA-3'; HuVHS (using the HuVH DNA as a template) 684. 5'-CACCAGCAGCAACCAGTTCAG-3' and 0 683. 5'-ACTGGTTGCTCGTGGTGTCTA-3', HuVHSTAY (using the HuVHS DNA as a template) 1026.5'-ACCAGCAGCAACACAGCCTACCTGAGACTCAGCAG-3' and 1028.5'-TGCTGAGTCTCAGGTAGGCTGTGTTGCTGCTGGTGT-3'; 5 HuVHA (using the HuVH DNA as a template) 745. 5'-TGACCTGCACCGCGTCTGGCTTCAAC-3' and 746.5'-TTGAAGCCAGACGCGGTGCAGGTCAG-3'; 0 HuVHAA (using the HuVHA DNA as a template) 1071. 5'-GAGACTCAGCAGCGTGACAG-3' and 1072.5'-CGCTGCTGAGTCTCAGGCTGAATGTGTTCTTGCTGGTGTC-3'; HuVHAT (using the HuVHA DNA as a template) 5 1071. 5'-CCTGAGACTCAGCAGCGTGACAG-3' and 1074.5'-CGCTGCTGAGTCTCAGGCTGGCCTGGTTCTTGCTGGTG-3'; HuVHAY (using the HuVHA DNA as a template) 1071. 5'-CCTGAGACTCAGCAGCGTGACAG-3' and 0 1073.5'-CGCTGCTGAGTCTCAGGTAGAACTGGTTCTTGC-3'; HuVHATAY (using the HuVHA DNA as a template) 1071. 5'-CCTGAGACTCAGCAGCGTGACAG-3' and 1075.5'-CGCTGCTGAGTCTCAGGTAGGCTGTGTTCTTGCTGGTGTC-3'; 5 HuVHASTAY (using the HuVHS DNA as a template) 745. 5'-TGACCTGCACCGCGTCTGGCTTCAAC-3' and 746.5'-TTGAAGCCAGACGCGGTGCAGGTCAG-3'; and 0 HuVKVL (using the HuVK DNA shown in Figure 18 as a template) 1010. 5'-ACTCCGACATCGTGCTGACCCAGAG-3' and 1011. 5'-CTCTGGGTCAGCACGATGTCGGAG-3'. The above-produced M13 DNA constructs were transferred to pSV vectors which were then 5 transfected into NSO host cells, according to the above-described procedures. The VH construct- and VL construct-containing pSV vectors were transfected into NSO cells in the following combinations (to produce antibodies having the indicated combinations of VH and -31 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 VL regions): MuVH/MuVK; HuVH/HuVK; HuVH/MuVK; 5 MuVH/HuVK; HuVHA/HuVK; HuVHT/HuVK; HuVHT/HuVKF; HuVH/HuVKVL; LO HuVHS/HuVKVL; HuVHSTAY/HuVK; HuVH/HuVKF; MuVH/HuVKF; and HuVHSTAY/HuVKVL. L5 Combinations of the other VHs with the various VKs are expected to behave in an analogous manner, though having an unpredictable variation in degree of binding. Final selection of the optimum combination will be a function of obtaining an antibody having the best, selective antigen binding properties with the fewest murine amino acid substitutions. The amino acid sequences of the initial humanized (CDR-grafted) COL-1 variable 0 heavy (VH) and variable light (VK) regions, HuVH and HuVK, are show in Figures 1 and 2, respectively. NSO transfectants were screened for whole antibody expression and the antigen binding characteristics of the antibodies produced thereby were measured by an ELISA test against CEA. Results are presented in Figures 4-9. This data shows that - of the graft-humanized and the mix-and-match clones - MuVH/HuVK outperformed both 25 HuVH/MuVK and HuVH/HuVK, but did not perform as well as the chimeric MuVH/MuVK antibodies. These results indicate that the initial fully humanized antibody (HuVH/HuVK) exhibits a 20-fold loss in CEA antigen binding affinity. Thus, it exhibits about 5% the binding affinity of murine COL-1. Based on analysis of CEA binding of mix-and-match antibodies 30 (HuVH/HuVK) and (MuVH/HuVK), it was determined that the reduction in CEA antigen binding was apparently largely attributable to the amino acid sequence of humanized heavy chain. However, an approximate 2-fold reduction in antigen binding also occurred apparently because of the amino acid sequence of the humanized kappa chain. This can be appreciated from the ELISA data in Figure 4. -32 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 As a result of this research, an antibody comprising the VHSTAY- and VKVL containing heavy and light chains was deposited on October 16, 1996 with the American Type Culture Collection, 12301 Parklawn Drive, Rockville, Maryland 20852 and accorded Accession Number ATCC CRL-12208 (this is a murine plasmacytoma cell line). This 5 deposit was made in accordance with the Budapest Treaty. This deposited cell line will be made irrevocably available, without restriction, upon issuance of a patent to this application, or any patent claiming benefit of priority to this application under 35 U.S.C. §120. Based on the foregoing, it will be appreciated that the humanized antibodies disclosed herein exhibit antigen-binding characteristics, i.e., CEA affinities comparable to the 0 parent monoclonal antibody, nCOL-1 (murine antibody), and to chimeric antibodies derived from nCOL-1, e.g., ChCOL-1yl (ATCC No. CRL 11217). Moreover, based on the foregoing results, these antibodies possess properties which will render them well suited for usage as in vivo diagnostics or therapeutics, e.g., improved serum clearance, metabolic properties, and little or no immunogenicity in humans. 5 These properties are highly significant because they will enable the subject humanized antibodies to be administered repeatedly, in large dosages, and over a prolonged period of time without significant adverse effects, e.g., a HAMA response or non-specific cytotoxicity. This is important for cancer treatment as well as for cancer diagnosis as it enables these antibodies to be used over prolonged time periods. Moreover, the clearance 0 properties of the subject human antibodies will enable these antibodies to effectively target desired target sites, e.g., CEA expressing carcinomas (because of the effects of serum clearance on targeting efficiency). Therefore, the humanized antibodies of the present invention comprise a substantial improvement in relation to previously disclosed antibodies specific to CEA. 5 Other embodiments of the invention will be apparent to those skilled in the art from a consideration of this specification or practice of the invention disclosed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the invention being indicated by the following claims. -33 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 SEQUENCE LISTING GENERAL INFORMATION: APPLICANT: NAME: The Dow Chemical Company STREET: 1790 Bldg. Washington Street CITY: Midland STATE: MI COUNTRY: U.S.A. POSTAL CODE: 48674 TELEPHONE: 517-636-1687 TELEFAX: 517-638-9786 TITLE OF INVENTION: High Affinity Humanized Anti-CEA Monoclonal Antibodies NUMBER OF SEQUENCES: 53 COMPUTER READABLE FORM: MEDIUM TYPE: 3-1/2" diskette COMPUTER: IBM PC Compatible OPERATING SYSTEM: MS-DOS SOFTWARE: PatentIn INFORMATION FOR SEQ ID NO:1 SEQUENCE CHARACTERISTICS: LENGTH: 124 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Murine COL-1 VH LOCATION: 1..124 SEQUENCE DESCRIPTION: SEQ ID NO: 1: Glu Val Gln Leu Gln Gln Ser Gly Ala Glu Leu Val Arg Ser Gly Ala 5 10 15 Ser Val Lys Met Ser Cys Thr Ala Ser Gly Phe Asn Ile Lys Asp Tyr 20 25 30 Tyr Met His Trp Val Lys Gln Arg Pro Glu Gln Gly Leu Glu Trp Ile 35 40 45 Gly Trp Ile Asp Pro Glu Asn Gly Asp Thr Glu Tyr Ala Pro Lys Phe 50 55 60 Gln Gly Lys Ala Thr Met Thr Thr Asp Tyr Ser Ser Asn Thr Ala Tyr 65 70 75 80 Leu Gln Leu Ser Ser Leu Thr Ser Glu Asp Thr Ala Val Tyr Tyr Cys 85 90 95 Asn Thr Arg Gly Leu Ser Thr Met Ile Thr Thr Arg Trp Phe Phe Asp 100 105 110 Val Trp Gly Ala Gly Thr Thr Val Ala Val Ser Ser 115 120 -34 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 INFORMATION FOR SEQ ID NO:2 SEQUENCE CHARACTERISTICS: LENGTH: 124 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: NEWM VH FR template LOCATION: 1..124 SEQUENCE DESCRIPTION: SEQ ID NO: 2: Gln Val Gln Leu Gln Glu Ser Gly Pro Gly Leu Val Arg Pro Ser Gln 5 10 15 Thr Leu Ser Leu Thr Cys Thr Val Ser Gly Phe Asn Ile Lys Asp Tyr 20 25 30 Tyr Met His Trp Val Arg Gln Pro Pro Gly Arg Gly Leu Glu Trp Ile 35 40 45 Gly Trp Ile Asp Pro Glu Asn Xaa Xaa Xaa Xaa Tyr Ala Pro Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Leu Val Asp Thr Ser Lys Asn Gln Phe Ser 65 70 75 80 Leu Arg Leu Ser Ser Val.Thr Ala Ala Asp Thr Ala Val Tyr Xaa Xaa 85 90 95 Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa 100 105 110 Xaa Xaa Xaa Xaa Xaa Xaa Xaa Val Thr Val Ser Ser 115 120 INFORMATION FOR SEQ ID NO:3: SEQUENCE CHARACTERISTICS: LENGTH: 124 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Humanized COL-1 VH, HuVH LOCATION: 1..124 SEQUENCE DESCRIPTION: SEQ ID NO: 3: Gln Val Gln Leu Gln Glu Ser Gly Pro Gly Leu Val Arg Pro Ser Gln 5 10 15 Thr Leu Ser Leu Thr Cys Thr Val Ser Gly Phe Asn Ile Lys Asp Tyr 20 25 30 Tyr Met His Trp Val Arg Gln Pro Pro Gly Arg Gly Leu Glu Trp Ile 35 40 45 -35 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 Gly Trp Ile Asp Pro Glu Asn Gly Asp Thr Glu Tyr Ala Pro Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Leu Val Asp Thr Ser Lys Asn Gln Phe Ser 65 70 75 80 Leu Arg Leu Ser Ser Val Thr Ala Ala Asp Thr Ala Val Tyr Tyr Cys 85 90 95 Asn Thr Arg Gly Leu Ser Thr Met Ile Thr Thr Arg Trp Phe Phe Asp 100 105 110 Val Trp Gly Gln Gly Ser Leu Val Thr Val Ser Ser 115 120 INFORMATION FOR SEQ ID NO:4: SEQUENCE CHARACTERISTICS: LENGTH: 124 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Humanized COL-1 VH, HuVHA LOCATION: 1..124 SEQUENCE DESCRIPTION: SEQ ID NO: 4: Gln Val Gln Leu Gln Glu Ser Gly Pro Gly Leu Val Arg Pro Ser Gln 5 10 15 Thr Leu Ser Leu Thr Cys Thr Ala Ser Gly Phe Asn Ile Lys Asp Tyr 20 25 30 Tyr Met His Trp Val Arg Gln Pro Pro Gly Arg Gly Leu Glu Trp Ile 35 40 45 Gly Trp Ile Asp Pro Glu Asn Gly Asp Thr Glu Tyr Ala Pro Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Leu Val Asp Thr Ser Lys Asn Gln Phe Ser 65 70 75 80 Leu Arg Leu Ser Ser Val Thr Ala Ala Asp Thr Ala Val Tyr Tyr Cys 85 90 95 Asn Thr Arg Gly Leu Ser Thr Met Ile Thr Thr Arg Trp Phe Phe Asp 100 105 110 Val Trp Gly Gln Gly Ser Leu Val Thr Val Ser Ser 115 120 INFORMATION FOR SEQ ID NO:5: SEQUENCE CHARACTERISTICS: LENGTH: 124 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear -36 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 FEATURE: NAME/KEY: Humanized COL-1 VH, HuVHAT LOCATION: 1..124 SEQUENCE DESCRIPTION: SEQ ID NO: 5: Gln Val Gln Leu Gln Glu Ser Gly Pro Gly Leu Val Arg Pro Ser Gln 5 10 15 Thr Leu Ser Leu Thr Cys Thr Ala Ser Gly Phe Asn Ile Lys Asp Tyr 20 25 30 Tyr Met His Trp Val Arg Gln Pro Pro Gly Arg Gly Leu Glu Trp Ile 35 40 45 Gly Trp Ile Asp Pro Glu Asn Gly Asp Thr Glu Tyr Ala Pro Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Leu Val Asp Thr Ser Lys Asn Thr Phe Ser 65 60 75 80 Leu Arg Leu Ser Ser Val Thr Ala Ala Asp Thr Ala Val Tyr Tyr Cys 85 90 95 Asn Thr Arg Gly Leu Ser Thr Met Ile Thr Thr Arg Trp Phe Phe Asp 100 105 110 Val Trp Gly Gln Gly Ser Leu Val Thr Val Ser Ser 115 120 INFORMATION FOR SEQ ID NO:6: SEQUENCE CHARACTERISTICS: LENGTH: 124 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Humanized COL-1 VH, HuVHAA LOCATION: 1..124 SEQUENCE DESCRIPTION: SEQ ID NO: 6: Gln Val Gln Leu Gln Glu Ser Gly Pro Gly Leu Val Arg Pro Ser Gln 5 10 15 Thr Leu Ser Leu Thr Cys Thr Ala Ser Gly Phe Asn Ile Lys Asp Tyr 20 25 30 Tyr Met His Trp Val Arg Gln Pro Pro Gly Arg Gly Leu Glu Trp Ile 35 40 45 Gly Trp Ile Asp Pro Glu Asn Gly Asp Thr Glu Tyr Ala Pro Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Leu Val Asp Thr Ser Lys Asn Gln Ala Ser 65 70 75 80 Leu Arg Leu Ser Ser Val Thr Ala Ala Asp Thr Ala Val Tyr Tyr Cys 85 90 95 -37 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 Asn Thr Arg Gly Leu Ser Thr Met Ile Thr Thr Arg Trp Phe Phe Asp 100 105 110 Val Trp Gly Gln Gly Ser Leu Val Thr Val Ser Ser 115 120 INFORMATION FOR SEQ ID NO:7: SEQUENCE CHARACTERISTICS: LENGTH: 124 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Humanized COL-1 VH, HuVHAY LOCATION: 1..124 SEQUENCE DESCRIPTION: SEQ ID NO: 7: Gln Val Gln Leu Gln Glu Ser Gly Pro Gly Leu Val Arg Pro Ser Gln 5 10 15 Thr Leu Ser Leu Thr Cys Thr Ala Ser Gly Phe Asn Ile Lys Asp Tyr 20 25 30 Tyr Met His Trp Val Arg Gln Pro Pro Gly Arg Gly Leu Glu Trp Ile 35 40 45 Gly Trp Ile Asp Pro Glu Asn Gly Asp Thr Glu Tyr Ala Pro Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Leu Val Asp Thr Ser Lys Asn Gln Phe Tyr 65 70 75 80 Leu Arg Leu Ser Ser Val Thr Ala Ala Asp Thr Ala Val Tyr Tyr Cys 85 90 95 Asn Thr Arg Gly Leu Ser Thr Met Ile Thr Thr Arg Trp Phe Phe Asp 100 105 110 Val Trp Gly Gln Gly Ser Leu Val Thr Val Ser Ser 115 120 INFORMATION FOR SEQ ID NO:8: SEQUENCE CHARACTERISTICS: LENGTH: 124 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Humanized COL-1 VH, HuVHATAY LOCATION: 1..124 SEQUENCE DESCRIPTION: SEQ ID NO: 8: Gln Val Gln Leu Gln Glu Ser Gly Pro Gly Leu Val Arg Pro Ser Gln 5 10 15 -38 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 Thr Leu Ser Leu Thr Cys Thr Ala Ser Gly Phe Asn Ile Lys Asp Tyr 20 25 30 Tyr Met His Trp Val Arg Gln Pro Pro Gly Arg Gly Leu Glu Trp Ile 35 40 45 Gly Trp Ile Asp Pro Glu Asn Gly Asp Thr Glu Tyr Ala Pro Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Leu Val Asp Thr Ser Lys Asn Thr Ala Tyr 65 70 75 80 Leu Arg Leu Ser Ser Val Thr Ala Ala Asp Thr Ala Val Tyr Tyr Cys 85 90 95 Asn Thr Arg Gly Leu Ser Thr Met Ile Thr Thr Arg Trp Phe Phe Asp 100 105 110 Val Trp Gly Gln Gly Ser Leu Val Thr Val Ser Ser 115 120 INFORMATION FOR SEQ ID NO:9: SEQUENCE CHARACTERISTICS: LENGTH: 124 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Humanized COL-1 VH, HuVHASTAY LOCATION: 1..124 SEQUENCE DESCRIPTION: SEQ ID NO: 9: Gln Val Gln Leu Gln Glu Ser Gly Pro Gly Leu Val Arg Pro Ser Gln 5 10 15 Thr Leu Ser Leu Thr Cys Thr Ala Ser Gly Phe Asn Ile Lys Asp Tyr 20 25 30 Tyr Met His Trp Val Arg Gln Pro Pro Gly Arg Gly Leu Glu Trp Ile 35 40 45 Gly Trp Ile Asp Pro Glu Asn Gly Asp Thr Glu Tyr Ala Pro Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Leu Val Asp Thr Ser Ser Asn Thr Ala Tyr 65 70 75 80 Leu Arg Leu Ser Ser Val Thr Ala Ala Asp Thr Ala Val Tyr Tyr Cys 85 90 95 Asn Thr Arg Gly Leu Ser Thr Met Ile Thr Thr Arg Trp Phe Phe Asp 100 105 110 Val Trp Gly Gln Gly Ser Leu Val Thr Val Ser Ser 115 120 -39 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 INFORMATION FOR SEQ ID NO:10 SEQUENCE CHARACTERISTICS: LENGTH: 124 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Humanized COL-1 VH, HuVHT LOCATION: 1..124 SEQUENCE DESCRIPTION: SEQ ID NO:10: Gln Val Gln Leu Gln Glu Ser Gly Pro Gly Leu Val Arg Pro Ser Gln 5 10 15 Thr Leu Ser Leu Thr Cys Thr Val Ser Gly Phe Asn Ile Lys Asp Tyr 20 25 30 Tyr Met His Trp Val Arg Gln Pro Pro Gly Arg Gly Leu Glu Trp Ile 35 40 45 Gly Trp Ile Asp Pro Glu Asn Gly Asp Thr Glu Tyr Ala Pro Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Leu Thr Asp Thr Ser Lys Asn Gln Phe Ser 65 70 75 80 Leu Arg Leu Ser Ser Val Thr Ala Ala Asp Thr Ala Val Tyr Tyr Cys 85 90 95 Asn Thr Arg Gly Leu Ser Thr Met Ile Thr Thr Arg Trp Phe Phe Asp 100 105 110 Val Trp Gly Gln Gly Ser Leu Val Thr Val Ser Ser 115 120 INFORMATION FOR SEQ ID NO:11: SEQUENCE CHARACTERISTICS: LENGTH: 124 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Humanized COL-1 VH, HuVHS LOCATION: 1..124 SEQUENCE DESCRIPTION: SEQ ID NO: 11ii: Gln Val Gln Leu Gln Glu Ser Gly Pro Gly Leu Val Arg Pro Ser Gln 5 10 15 Thr Leu Ser Leu Thr Cys Thr Val Ser Gly Phe Asn Ile Lys Asp Tyr 20 25 30 Tyr Met His Trp Val Arg Gln Pro Pro Gly Arg Gly Leu Glu Trp Ile 35 40 45 -40 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 Gly Trp Ile Asp Pro Glu Asn Gly Asp Thr Glu Tyr Ala Pro Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Leu Val Asp Thr Ser Ser Asn Gln Phe Ser 65 70 75 80 Leu Arg Leu Ser Ser Val Thr Ala Ala Asp Thr Ala Val Tyr Tyr Cys 85 90 95 Asn Thr Arg Gly Leu Ser Thr Met Ile Thr Thr Arg Trp Phe Phe Asp 100 105 110 Val Trp Gly Gln Gly Ser Leu Val Thr Val Ser Ser 115 120 INFORMATION FOR SEQ ID NO:12: SEQUENCE CHARACTERISTICS: LENGTH: 124 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Humanized COL-1 VH, HuVHSTAY LOCATION: 1..124 SEQUENCE DESCRIPTION: SEQ ID NO: 12: Gln Val Gln Leu Gln Glu Ser Gly Pro Gly Leu Val Arg Pro Ser Gln 5 10 15 Thr Leu Ser Leu Thr Cys Thr Val Ser Gly Phe Asn Ile Lys Asp Tyr 20 25 30 Tyr Met His Trp Val Arg Gln Pro Pro Gly Arg Gly Leu Glu Trp Ile 35 40 45 Gly Trp Ile Asp Pro Glu Asn Gly Asp Thr Glu Tyr Ala Pro Lys Phe 50 55 60 Gln Gly Arg Val Thr Met Leu Val Asp Thr Ser Ser Asn Thr Ala Tyr 65 70 75 80 Leu Arg Leu Ser Ser Val Thr Ala Ala Asp Thr Ala Val Tyr Tyr Cys 85 90 95 Asn Thr Arg Gly Leu Ser Thr Met Ile Thr Thr Arg Trp Phe Phe Asp 100 105 110 Val Trp Gly Gln Gly Ser Leu Val Thr Val Ser Ser 115 120 INFORMATION FOR SEQ ID NO:13: SEQUENCE CHARACTERISTICS: LENGTH: 110 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear -41 SUBSTITUTE SHEET (RULE 26) WO99/43817 PCTIUS98/03680 FEATURE: NAME/KEY: Murine COL-1 VK LOCATION: 1..110 SEQUENCE DESCRIPTION: SEQ ID NO: 13: Asp Ile Val Leu Thr Gln Ser Pro Ala Ser Leu Thr Val Ser Leu Gly 5 10 15 Leu Arg Ala Thr Ile Ser Cys Arg Ala Ser Lys Ser Val Ser Ala Ser 20 25 30 Gly Tyr Ser Tyr Met His Trp Tyr Gln Gln Arg Pro Gly Gln Pro Pro 35 40 45 Lys Leu Leu Ile Tyr Leu Ala Ser Asn Leu Gln Ser Gly Val Pro Ala 50 55 60 Arg Phe Ser Gly Ser Gly Ser Gly THr Asp Phe Thr Leu Asn Ile His 65 70 75 80 Pro Val Glu Glu Glu Asp Ala Ala Thr Tyr Tyr Cys Gln His Ser Arg 85 90 95 Glu Leu Pro Thr Phe Gly Gly Gly Thr Lys Leu Glu Ile Lys 100 105 110 INFORMATION FOR SEQ ID NO:14: SEQUENCE CHARACTERISTICS: LENGTH: 110 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: REI VK FR template LOCATION: 1..110 SEQUENCE DESCRIPTION: SEQ ID NO: 14: Asp Ile Gln Leu Thr Gln Ser Pro Ser Ser Leu Ser Ala Ser Val Gly 5 10 15 Asp Arg Val Thr Ile Thr Cys Arg Xaa Xaa Xaa Xaa Xaa Xaa Xaa Xaa 20 25 30 Gly Tyr Ser Tyr Met His Trp Tyr Gln Gln Lys Pro Gly Lys Ala Pro 35 40 45 Lys Leu Leu Ile Tyr Xaa Xaa Xaa Xaa Xaa Xaa Ser Gly Val Pro Ser 50 55 60 Arg Phe Ser Gly Ser Gly Ser Gly Thr Asp Phe Thr Phe Thr Ile Ser 65 70 75 80 Ser Leu Gln Pro Glu Asp Ile Ala Thr Tyr Tyr Cys Gln Xaa Xaa Xaa 85 90 95 Xaa Xaa Xaa Thr Phe Gly Gln Gly Thr Lys Xaa Xaa Xaa Xaa 100 105 110 -42 RUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 INFORMATION FOR SEQ ID NO:15: SEQUENCE CHARACTERISTICS: LENGTH: 110 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Humanized COL-1 VK, HuVK LOCATION: 1..110 SEQUENCE DESCRIPTION: SEQ ID NO: 15: Asp Ile Gln Met Thr Gln Ser Pro Ser Ser Leu Ser Ala Ser Val Gly 5 10 15 Asp Arg Val Thr Ile Thr Cys Arg Ala Ser Lys Ser Val Ser Ala Ser 20 25 30 Gly Tyr Ser Tyr Met His Trp Tyr Gln Gln Thr Pro Gly Lys Ala Pro 35 40 45 Lys Leu Leu Ile Tyr Leu Ala Ser Asn Leu Gln Ser Gly Val Pro Ser 50 55 60 Arg Phe Ser Gly Ser Gly Ser Gly Thr Asp Tyr Thr Phe Thr Ile Ser 65 70 75 80 Ser Leu Gln Pro Glu Asp Ile Ala Thr Tyr Tyr Cys Gln His Ser Arg 85 90 95 Glu Leu Pro Thr Phe Gly Gln Gly Thr Lys Leu Gln Ile Thr 100 105 110 INFORMATION FOR SEQ ID NO:16: SEQUENCE CHARACTERISTICS: LENGTH: 110 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Humanized COL-1 VK, HuVKVL LOCATION: 1..110 SEQUENCE DESCRIPTION: SEQ ID NO: 16: Asp Ile Val Leu Thr Gln Ser Pro Ser Ser Leu Ser Ala Ser Val Gly 5 10 15 Asp Arg Val Thr Ile Thr Cys Arg Ala Ser Lys Ser Val Ser Ala Ser 20 25 30 Gly Tyr Ser Tyr Met His Trp Tyr Gln Gln Thr Pro Gly Lys Ala Pro 35 40 45 Lys Leu Leu Ile Tyr Leu Ala Ser Asn Leu Gln Ser Gly Val Pro Ser 50 55 60 -43 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 Arg Phe Ser Gly Ser Gly Ser Gly Thr Asp Tyr Thr Phe Thr Ile Ser 65 70 75 80 Ser Leu Gln Pro Glu Asp Ile Ala Thr Tyr Tyr Cys Gln His Ser Arg 85 90 95 Glu Leu Pro Thr Phe Gly Gln Gly Thr Lys Leu Gln Ile Thr 100 105 110 INFORMATION FOR SEQ ID NO:17: SEQUENCE CHARACTERISTICS: LENGTH: 11ii0 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Humanized COL-1 VK, HuVKF LOCATION: 1..110 SEQUENCE DESCRIPTION: SEQ ID NO: 17: Asp Ile Gln Met Thr Gln Ser Pro Ser Ser Leu Ser Ala Ser Val Gly 5 10 15 Asp Arg Val Thr Ile Thr Cys Arg Ala Ser Lys Ser Val Ser Ala Ser 20 25 30 Gly Tyr Ser Tyr Met His Trp Tyr Gln Gln Thr Pro Gly Lys Ala Pro 35 40 45 Lys Leu Leu Ile Tyr Leu Ala Ser Asn Leu Gln Ser Gly Val Pro Ser 50 55 60 Arg Phe Ser Gly Ser Gly Ser Gly Thr Asp Phe Thr Phe Thr Ile Ser 65 70 75 80 Ser Leu Gln Pro Glu Asp Ile Ala Thr Tyr Tyr Cys Gln His Ser Arg 85 90 95 Glu Leu Pro Thr Phe Gly Gln Gly Thr Lys Leu Gln Ile Thr 100 105 110 INFORMATION FOR SEQ ID NO:18: SEQUENCE CHARACTERISTICS: LENGTH: 125 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Humanized COL-1 VH, ATCC CRL-12208 VH LOCATION: 1..125 SEQUENCE DESCRIPTION: SEQ ID NO: 18: Gln Val Gln Leu Gln Glu Ser Gly Pro Gly Leu Val Arg Pro Ser Gln 5 10 15 -44 SUBSTITUTE SHEET (RULE 26) WO99/43817 PCT/US98/03680 Thr Leu Ser Leu Thr Cys Thr Val Ser Gly Phe Asn Ile Lys Asp Tyr 20 25 30 Tyr Met His Trp Val Arg Gln Pro Pro Gly Arg Gly Leu Glu Trp Ile 35 40 45 Gly Trp Ile Asp Pro Glu Asn Gly Asp Thr Glu Tyr Ala Pro Gly Phe 50 55 60 Gln Gly Arg Val Thr Met Leu Val Asp Thr Ser Ser Asn Thr Ala Tyr 65 70 75 80 Leu Arg Leu Ser Ser Val Thr Ala Ala Asp Thr Al Val Tyr Tyr Cys 85 90 95 Asn Thr Arg Gly Leu Ser Thr Met Ile Thr Thr Arg Trp Phe Phe Asp 100 105 11ii0 Val Trp Gly Gln Gly Ser Leu Val Thr Val Ser Ser 115 120 125 INFORMATION FOR SEQ ID NO:19: SEQUENCE CHARACTERISTICS: LENGTH: 110 TYPE: amino acid STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Humanized COL-1 VK, ATCC CRL-12208 VK LOCATION: i..110 SEQUENCE DESCRIPTION: SEQ ID NO: 19: Asp Ile Val Leu Thr Gln Ser Pro Ser Ser Leu Ser Ala Ser Val Gly 5 10 15 Asp Arg Val Thr Ile Thr Cys Arg Ala Ser Lys Ser Val Ser Ala Ser 20 25 30 Gly Tyr Ser Tyr Met His Trp Tyr Gln Gln Thr Pro Gly Lys Ala Pro 35 40 45 Lys Leu Leu Ile Tyr Leu Ala Ser Asn Leu Gln Ser Gly Val Pro Ser 50 55 60 Arg Phe Ser Gly Ser Gly Ser Gly Thr Asp Tyr Thr Phe Thr Ile Ser 65 70 75 80 Ser Leu Gln Pro Glu Asp Ile Ala Thr Tyr Tyr Cys Gln His Ser Arg 85 90 95 Glu Leu Pro Thr Phe Gly Gln Gly Thr Lys Leu Gln Ile Thr 100 105 110 INFORMATION FOR SEQ ID NO:20: SEQUENCE CHARACTERISTICS: LENGTH: 348 TYPE: DNA STRANDEDNESS: double -45 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 TOPOLOGY: linear FEATURE: NAME/KEY: Template used to produce HuVH LOCATION: 1..348 SEQUENCE DESCRIPTION: SEQ ID NO: 20: CAG GTC CAA CTG CAG GAG AGC GGT CCA GGT CTT GTG AGA CCT AGC CAG 48 ACC CTG AGC CTG ACC TGC ACC GTG TCT GGC TTC AAC ATT AAA GAC TAC 96 TAT ATG CAC TGG GTG AGA CAG CCA CCT GGA CGA GGT CTT GAG TGG ATT 144 GGA TGG ATT GAT CCT GAG AAT NNN NNN NNN NNN TAT GCC CCG AAG TTC 192 CAG GGC AGA GTG ACA ATG CTG GTA GAC ACC AGC AAG AAC CAG TTC AGC 240 CTG AGA CTC AGC AGC GTG ACA GCC GCC GAC ACC GCG GTC TAT NNN NNN 288 NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN NNN GTC 336 ACC GTC TCC TCA 348 INFORMATION FOR SEQ ID NO:21: SEQUENCE CHARACTERISTICS: LENGTH: 372 TYPE: DNA STRANDEDNESS: double TOPOLOGY: linear FEATURE: NAME/KEY: Template used to produce HuVH variants LOCATION: 1..372 SEQUENCE DESCRIPTION: SEQ ID NO: 21: CAG GTC CAA CTG CAG GAG AGC GGT CCA GGT CTT GTG AGA CCT AGC CAG 48 ACC CTG AGC CTG ACC TGC ACC GTG TCT GGC TTC AAC ATT AAA GAC TAC 96 TAT ATG CAC TGG GTG AGA CAG CCA CCT GGA CGA GGT CTT GAG TGG ATT 144 GGA TGG ATT GAT CCT GAG AAT GGT GAT ACT GAA TAT GCC CCG AAG TTC 192 CAG GGC AGA GTG ACA ATG CTG GTA GAC ACC AGC AAG AAC CAG TTC AGC 240 CTG AGA CTC AGC AGC GTG ACA GCC GCC GAC ACC GCG GTC TAT TAC TGT 288 AAT ACA CGG GGT CTA TCT ACT ATG ATT ACG ACG GCT TGG TTC TTC GAT 336 GTC TGG GGC CAA GGG TCC TTG GTC ACC GTC TCC TCA 372 -46 EllRSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 INFORMATION FOR SEQ ID NO:22: SEQUENCE CHARACTERISTICS: LENGTH: 318 TYPE: DNA STRANDEDNESS: double TOPOLOGY: linear FEATURE: NAME/KEY: Template used to produce HuVK LOCATION: 1..318 SEQUENCE DESCRIPTION: SEQ ID NO: 22: GAC ATC CAR CTG ACC CAG AGC CCA AGC AGC CTG AGC GCC AGC GTG GGT 48 GAC AGA GTG ACC ATC ACC TGT AGG NNN NNN NNN NNN NNN NNN NNN NNN 96 GGC TAT AGT TAT ATG CAC TTG TAC CAG CAG ACG CCA GGT AAG GCT CCA 144 AAG CTG CTG ATC TAC NNN NNN NNN NNN NNN NNN TCT GGT GTG CCA AGC 192 AGA TTC AGC GGT AGC GGT AGC GGT ACC GAC TTY ACC TTC ACC ATC AGC 240 AGC CTC CAG CCA GAG GAC ATC GCC ACC TAC TAC TGC CAG NNN NNN NNN 288 NNN NNN NNN ACG TTC GGC CAA GGG ACC AAG 318 INFORMATION FOR SEQ ID NO:23: SEQUENCE CHARACTERISTICS: LENGTH: 330 TYPE: DNA STRANDEDNESS: double TOPOLOGY: linear FEATURE: NAME/KEY: Template used to produce HuVK variants LOCATION: 1..330 SEQUENCE DESCRIPTION: SEQ ID NO: 23: GAC ATC CAR ATG ACC CAG AGC CCA AGC AGC CTG AGC GCC AGC GTG GGT 48 GAC AGA GTG ACC ATC ACC TGT AGG GCC AGC AAA AGT GTC AGT GCA TGT 96 GGC TAT AGT TAT ATG CAC TGG TAC CAG CAG ACG CCA GGT AAG GCT CCA 144 AAG CTG CTG ATC TAC CTT GCA TCC AAC CTA CAA TCT GGT GTG CCA AGC 192 AGA TTC AGC GGT AGC GGT AGC GGT ACC GAC TAC ACC TTC ACC ATC AGC 240 AGC CTC CAG CCA GAG GAC ATC GCC ACC TAC TAC TGC CAG CAC AGT AGG 288 GAG CTT CCT ACG TTC GGC CAA GGG ACC AAG CTG CAA ATC ACA 330 -47 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCTIUS98/03680 INFORMATION FOR SEQ ID NO:24: SEQUENCE CHARACTERISTICS: LENGTH: 17 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 10 LOCATION: 1..17 SEQUENCE DESCRIPTION: SEQ ID NO: 24: CTAAAACGAC GGCCAGT 17 INFORMATION FOR SEQ ID NO:25: SEQUENCE CHARACTERISTICS: LENGTH: 16 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 11 LOCATION: 1..16 SEQUENCE DESCRIPTION: SEQ ID NO: 25: AACAGCTATG ACCATG 16 INFORMATION FOR SEQ ID NO:26: SEQUENCE CHARACTERISTICS: LENGTH: 22 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 385 LOCATION: 1..22 SEQUENCE DESCRIPTION: SEQ ID NO: 26: GCGGGCCTCT TCGCTATTAC GC 22 INFORMATION FOR SEQ ID NO:27: SEQUENCE CHARACTERISTICS: LENGTH: 22 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 391 LOCATION: 1..22 -48 SUBSTITUTE SHEET (RULE 26) WO99/43817 PCT/US98/03680 SEQUENCE DESCRIPTION: SEQ ID NO: 27: CTCTCTCAGG GCCAGGCGGT GA 22 INFORMATION FOR SEQ ID NO:28: SEQUENCE CHARACTERISTICS: LENGTH: 32 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 836 LOCATION: 1..32 SEQUENCE DESCRIPTION: SEQ ID NO: 28: TGAGAATGGT GATACTGAAT ATGCCCCGAA GT 32 INFORMATION FOR SEQ ID NO:29: SEQUENCE CHARACTERISTICS: LENGTH: 34 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 837 LOCATION: 1..34 SEQUENCE DESCRIPTION: SEQ ID NO: 29: TCGGGGCATA TTCAGTATCA CCATTCTCAG GATC 34 INFORMATION FOR SEQ ID NO:30: SEQUENCE CHARACTERISTICS: LENGTH: 59 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 838 LOCATION: 1..59 SEQUENCE DESCRIPTION: SEQ ID NO: 30: ACTATGATTA CGACGCGTTG GTTCTTCGAT GTCTGGGGCC AAGGGTCCTT GGTCACGTC 59 -49 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 INFORMATION FOR SEQ ID NO:31: SEQUENCE CHARACTERISTICS: LENGTH: 55 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 839 LOCATION: 1..55 SEQUENCE DESCRIPTION: SEQ ID NO: 31: ACGCGTCGTA ATCATAGTAG ATAGACCCCG TGTATTACAG TAATAGACCG CGGTG 55 INFORMATION FOR SEQ ID NO:32: SEQUENCE CHARACTERISTICS: LENGTH: 44 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 842 LOCATION: 1..44 SEQUENCE DESCRIPTION: SEQ ID NO: 32: TATAGCCAGA TGCACTGACA CTTTTGCTGG CCCTACAGGT GATG 44 INFORMATION FOR SEQ ID NO:33: SEQUENCE CHARACTERISTICS: LENGTH: 23 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 844 LOCATION: 1..23 SEQUENCE DESCRIPTION: SEQ ID NO: 33: GCTCTGGGTC ATCTGGATGT CGG 23 INFORMATION FOR SEQ ID NO:34: SEQUENCE CHARACTERISTICS: LENGTH: 72 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 849 LOCATION: 1..72 -50 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 SEQUENCE DESCRIPTION: SEQ ID NO: 34: TTCTACTCAC GTGTGATTTG CAGCTTGGTC CCTTGGCCGA ACGTAGGAAG CTCCCTACTG 60 TGCTGGCAGT AG 72 INFORMATION FOR SEQ ID NO:35: SEQUENCE CHARACTERISTICS: LENGTH: 25 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 850 LOCATION: 1..25 SEQUENCE DESCRIPTION: SEQ ID NO: 35: ATGGTGAAGG TGTAGTCGGT ACCGC 25 INFORMATION FOR SEQ ID NO:36: SEQUENCE CHARACTERISTICS: LENGTH: 25 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 851 LOCATION: 1..25 SEQUENCE DESCRIPTION: SEQ ID NO: 36: GCCTTACCTG GCGTCTGCTG GTACC 25 INFORMATION FOR SEQ ID NO:37: SEQUENCE CHARACTERISTICS: LENGTH: 29 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 841 LOCATION: 1..29 SEQUENCE DESCRIPTION: SEQ ID NO: 37: GCACACCAGA TTGTAGGTTG GATGCAAGG 29 -51 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 INFORMATION FOR SEQ ID NO:38: SEQUENCE CHARACTERISTICS: LENGTH: 44 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 842 LOCATION: 1..44 SEQUENCE DESCRIPTION: SEQ ID NO: 38: TATAGCCAGA TGCACTGACA CTTTTGCTGG CCCTACAGGT GATG 44 INFORMATION FOR SEQ ID NO:39: SEQUENCE CHARACTERISTICS: LENGTH: 24 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 954 LOCATION: 1..24 SEQUENCE DESCRIPTION: SEQ ID NO: 39: GACAATGCTG ACAGACACCA GCAA 24 INFORMATION FOR SEQ ID NO:40: SEQUENCE CHARACTERISTICS: LENGTH: 24 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 955 LOCATION: 1..24 SEQUENCE DESCRIPTION: SEQ ID NO: 40: TGCTGGTGTC TGTCAGCATT GTCA 24 INFORMATION FOR SEQ ID NO:41: SEQUENCE CHARACTERISTICS: LENGTH: 21 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 684 LOCATION: 1..21 -52 QlIRATITUTF RHFFT (RULE 26) WO 99/43817 PCT/US98/03680 SEQUENCE DESCRIPTION: SEQ ID NO: 41: CACCAGCAGC AACCAGTTCA G 21 INFORMATION FOR SEQ ID NO:42: SEQUENCE CHARACTERISTICS: LENGTH: 21 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 683 LOCATION: 1..21 SEQUENCE DESCRIPTION: SEQ ID NO: 42: ACTGGTTGCT CGTGGTCTCT A 21 INFORMATION FOR SEQ ID NO:43: SEQUENCE CHARACTERISTICS: LENGTH: 35 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 1026 LOCATION: 1..35 SEQUENCE DESCRIPTION: SEQ ID NO: 43: ACCAGCAGCA ACACAGCCTA CCTGAGACTC AGCAG 35 INFORMATION FOR SEQ ID NO:44: SEQUENCE CHARACTERISTICS: LENGTH: 36 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 1028 LOCATION: 1..36 SEQUENCE DESCRIPTION: SEQ ID NO: 44: TGCTGAGTCT CAGGTAGGCT GTGTTGCTGC TGGTGT 36 -53 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 INFORMATION FOR SEQ ID NO:45: SEQUENCE CHARACTERISTICS: LENGTH: 26 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 745 LOCATION: 1..26 SEQUENCE DESCRIPTION: SEQ ID NO: 45: TGACCTGCAC CGCGTCTGGC TTCAAC 26 INFORMATION FOR SEQ ID NO:46: SEQUENCE CHARACTERISTICS: LENGTH: 26 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 746 LOCATION: 1..26 SEQUENCE DESCRIPTION: SEQ ID NO: 46: TTGAAGCCAG ACGCGGTGCA GGTCAG 26 INFORMATION FOR SEQ ID NO:47: SEQUENCE CHARACTERISTICS: LENGTH: 20 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 1071 LOCATION: 1..20 SEQUENCE DESCRIPTION: SEQ ID NO: 47: GAGACTCAGC AGCGTGACAG 20 INFORMATION FOR SEQ ID NO:48: SEQUENCE CHARACTERISTICS: LENGTH: 40 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear -54 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 FEATURE: NAME/KEY: Oligonucleotide 1072 LOCATION: 1..40 SEQUENCE DESCRIPTION: SEQ ID NO: 48: CGCTGCTGAG TCTCAGGCTG AATGTGTTCT TGCTGGTGTC 40 INFORMATION FOR SEQ ID NO:49: SEQUENCE CHARACTERISTICS: LENGTH: 38 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 1074 LOCATION: 1..38 SEQUENCE DESCRIPTION: SEQ ID NO: 49: CGCTGCTGAG TCTCAGGCTG GCCTGGTTCT TGCTGGTG 38 INFORMATION FOR SEQ ID NO:50: SEQUENCE CHARACTERISTICS: LENGTH: 33 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 1073 LOCATION: 1..33 SEQUENCE DESCRIPTION: SEQ ID NO: 50: CGCTGCTGAG TCTCAGGTAG AACTGGTTCT TGC 33 INFORMATION FOR SEQ ID NO:51: SEQUENCE CHARACTERISTICS: LENGTH: 40 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 1075 LOCATION: 1..40 SEQUENCE DESCRIPTION: SEQ ID NO: 51: CGCTGCTGAG TCTCAGGTAG GCTGTGTTCT TGCTGGTGTC 40 -55 SUBSTITUTE SHEET (RULE 26) WO99/43817 PCT/US98/03680 INFORMATION FOR SEQ ID NO:52: SEQUENCE CHARACTERISTICS: LENGTH: 25 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 1010 LOCATION: 1..25 SEQUENCE DESCRIPTION: SEQ ID NO: 52: ACTCCGACAT CGTGCTGACC CAGAG 25 INFORMATION FOR SEQ ID NO:53: SEQUENCE CHARACTERISTICS: LENGTH: 24 TYPE: DNA STRANDEDNESS: single TOPOLOGY: linear FEATURE: NAME/KEY: Oligonucleotide 1011 LOCATION: 1..24 SEQUENCE DESCRIPTION: SEQ ID NO: 53: CTCTGGGTCA GCACGATGTC GGAG 24 -56 AllRETITUTE SHEET (RULE 26)

Claims (10)

1. A humanized antibody or humanized antibody fragment which specifically binds CEA which is characterized in that said humanized antibody or humanized antibody fragment is derived from a murine antibody that binds CEA. 5 2. The humanized antibody or humanized antibody fragment of Claim 1 wherein the CDRs of said humanized antibody or humanized antibody fragment are obtained from said murine antibody, the VH FRs of said humanized antibody or humanized antibody fragment have an amino acid sequence of the NEWM FRs or the humanized COL-1 FRs of Figure 1 or 13, and the VL FRs of said humanized antibody or humanized antibody fragment 0 have an amino acid sequence of the REI FRs or the humanized COL-1 FRs of Figure 2 or

14. 3. The humanized antibody or humanized antibody fragment of Claim 2 wherein said humanized antibody has an antigen binding affinity for CEA which is at least 10% that of COL-1 and said humanized antibody fragment has an amino acid sequence identical to that 5 of a constituent part of said humanized antibody. 4. The humanized antibody or humanized antibody fragment of Claim 2 wherein said humanized antibody has an antigen binding affinity for CEA which is at least 30% that of COL-1. 5. The humanized antibody or humanized antibody fragment of one of Claims 2 0 4 wherein said murine antibody is one of COL-1 through COL-15. 6. The humanized antibody or humanized antibody fragment of Claim 5 wherein said murine antibody is COL-1. 7. The humanized antibody or humanized antibody fragment of Claim 6 wherein said humanized antibody is expressed by ATCC CRL-12208 and said humanized antibody 5 fragment has an amino acid sequence identical to that of a constituent part of the antibody expressed by ATCC CRL-12208. 8. The humanized antibody or humanized antibody fragment of Claim 1 wherein said humanized antibody or humanized antibody fragment has a humanized variable heavy chain sequence of Figure 1 or 13 or a humanized variable light chain sequence of Figure 2 0 or 14, or has both said humanized variable heavy chain sequence and said humanized variable light chain sequence. 9. A nucleic acid sequence characterized in that a humanized antibody or humanized antibody fragment according to one of Claims 1-8 may be expressed from the nucleic acid sequence. -57 RUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 10. A vector characterized in that a humanized antibody or humanized antibody fragment according to one of Claims 1-8 may be expressed from the vector. 11. The vector according to Claim 10 wherein said vector is a bare nucleic acid segment, a carrier-associated nucleic acid segment, a nucleoprotein, a plasmid, a virus, a 5 viroid, or a transposable element. 12. A composition suitable for the treatment or in vivo or in vitro detection of cancer characterized in that it comprises, respectively, a therapeutically effective or a diagnostically effective amount of a humanized antibody or humanized antibody fragment according to one of Claims 1-8. ) 13. The composition of Claim 12 wherein said humanized antibody or humanized antibody fragment is, directly or indirectly, associated with or linked to an effector moiety having therapeutic activity, and the composition is suitable for the treatment of cancer. 14. The composition of Claim 13 wherein said effector moiety is a radionuclide, therapeutic enzyme, anti-cancer drug, cytokine, cytotoxin, or anti-proliferative agent. 5 15. The composition of Claim 12 wherein said humanized antibody or humanized antibody fragment is, directly or indirectly, associated with or linked to a a detectable label, and the composition is suitable for detection of cancer.

16. The composition of Claim 15 wherein the detectable label is a radionuclide or an enzyme. ) 17. A method for in vivo treatment of a mammal having a CEA-expressing cancer characterized in that the method is performed by administering to the mammal a therapeutically effective amount of a composition according to one of Claims 12-14.

18. A method of in vitro immunodetection of CEA-expressing cancer cells characterized in that the method is performed by contacting the cancer cells with a 5 composition according to any one of Claims 12, 15, and 16.

19. The method of Claim 18 wherein the humanized antibodies or humanized antibody fragments of the composition are bound to a solid support.

20. A method of immunodetection of CEA-expressing cancer cells in a mammal characterized in that the method is performed by administering to the mammal a ) diagnostically effective amount of a composition according to one of Claims 12, 15, and 16.

21. The method of Claim 20 wherein said immunodetection is in vivo tumor imaging.

22. A method of treatment of cancer by (i) intravenously administering a radionuclide-labeled antibody, (ii) thereafter detecting tumor cells using a radionuclide -58 SUBSTITUTE SHEET (RULE 26) WO 99/43817 PCT/US98/03680 activity probe, and (iii) thereafter removing the detected tumor cells by surgical excision, characterized in that the antibody is a humanized antibody or humanized antibody fragment according to one of Claims 1-8.

23. The method of claim 22, wherein the radionuclide is 1251 or 31. 5 24. A commercial package characterized in that it contains a composition according to any one of Claims 12-16 as active ingredient together with instructions for use thereof to treat or detect cancer wherein the composition is reconstituted prior to said use.

25. The use of a composition as claimed in any one of Claims 12-16 to treat or detect cancer. 0 26. The use of a humanized antibody or humanized antibody fragment according to any one of Claims 1-8 for the treatment or detection of cancer. -59 MUnSTITUTE SHEET (RULE 26)