Monoclonal antibodies against human colon carcinoma-associated antigens and uses therefor

Abstract

monoclonal antibodies, in particular 33.28 and 31.1, and chimeric antibodies, in particular mouse/humans chimeric chi #1 specific for glycoprotein antigens of colon carcinoma-associated antigens which are immunogenic in humans, are disclosed. Such antibodies, and fragments and derivatives thereof, are useful in immunodiagnosis and immunotherapy of human colon, breast, and ovarian cancer, and for purification of antigens which can serve as immunotherapeutic agents. Methods of detecting the colon carcinoma-associated antigen in a sample, and methods for treating subjects having colon, breast, and ovarian carcinomas are disclosed.

Description

This application is a continuation-in-part of U.S. application Ser. No. 08/159,836 filed Nov. 30, 1993, which is a continuation-in-part of U.S. application Ser. No. 08/117,430, filed Sep. 7, 1993, now abandoned, which is a continuation-in-part of U.S. application Ser. No. 07/670,816, filed Mar. 18, 1991, now abandoned, which is a continuation-in-part of U.S. application Ser. No. 07/176,337, filed Mar. 31, 1988, now abandoned.

BACKGROUND OF THE INVENTION

1. Field of the Invention

This invention, in the field of immunology and medicine, relates to new hybridoma lines and the monoclonal antibodies (mAbs) they secrete which are specific for clinically defined colon carcinoma-associated antigens. The mAbs are useful in vivo for immunodetection and immunotherapy of colon carcinoma as well as for the detection and purification of colon carcinoma-associated antigens.

2. Description of the Background Art

During the process of oncogenesis, a number of cell-surface molecules or markers appear on cells. Such tumor-related markers include oncofetoproteins, neoglycoproteins, sphignolipids, and modifications of existing surface proteins. Such new or altered structures are often shed from the tumor cell surface and appear in the serum or in other biological fluids. The detection of any of these substances or “tumor markers” serves as the basis for diagnosing or monitoring the progress of neoplastic disease.

Early animal studies demonstrated that, among these tumor markers, a subset of tumor membrane protein or glycoprotein antigens were immunogenic. Upon appropriate reintroduction into the tumor-bearing host, typically after surgical removal of the primary tumor, such antigens could effectively block the establishment of new tumor growth.

An attempt to use similarly derived tumor-associated antigens in humans was made by Hollinshead and Stewart using a relatively purified membrane preparation in patients with lung cancer (Stewart, T. H. M. et al., Ann. N.Y. Acad. Sci. 277:436 (1976)). These studies were later expanded to include patients with melanoma and colon carcinoma, wherein different pooled allogeneic tumor preparations were administered in combination with complete Freund's adjuvant (Hollinshead, A. C. et al., Cancer 4:9:1387 (1982); Hollinshead, A. C. et al., Cancer 56:480 (1985)).

The use of Freund's adjuvant was based on observations that normal tissue antigens with this adjuvant produced severe autoimmune responses in animal recipients, whereas in the absence of this adjuvant, no adverse reactions were seen. The adjuvant was thought to promote antigen processing by host macrophages as well as prolong the stimulatory action of the antigen at the site of its deposition (see, for example, Roitt, L. Essential Immunology, 6th Ed., Blackwell Scientific Publications, Oxford (1988)).

The above observations served as the basis for early clinical trials using specific human tumor membrane proteins and glycoproteins as tumor “vaccines.” Various of the tumor-associated antigens which had been isolated and characterized could prolong survival and, in some cases, produce regression of metastatic disease.

With the advent of monoclonal antibody (mAb) technology, it has become possible to obtain pure antibody populations which permit better purification and characterization of the various tumor markers and tumor-associated antigens that are useful for immunodiagnosis or immunotherapy. Many mAbs have been described that have varying degrees of selectivity for tumor antigens (versus normal cell surface markers); some of these tumor antigens are broadly represented across several or many tumor types, whereas others appear to be truly tumor-specific. A number of these mAb-tumor antigen systems are described below.

Herlyn et al., Proc. Natl. Acad. Sci. USA 76:1438 (1979), discloses two mAbs obtained by immunizing mice with human colorectal carcinoma (CRC) cells. The mAbs have selective reactivity with human CRC cells. One mAb, 1083-17 (the forerunner of 17-1A), is now known to react with a 41 kDa glycoprotein (see below).

Herlyn et al., J. Clin. Immunol. 2:135 (1982), described the detection of a circulating colorectal carcinoma (CRC)-associated antigen by a mAb developed against a membrane antigen of the SW116 cell line. MAbs 19-9 and 52a, which recognize a monosialoganglioside antigen (Magnani, J. L. et al., Science 212:55 (1981)), reacted with cells of 8 of 12 CRC lines as well as with the cells of one gastric carcinoma and one pancreatic carcinoma. MAb C₄14 reacted with four of six CRC cultures and with gastric tumor cells. The binding of mAbs 19-9 and 52a to tumor cells were inhibited by a CRC patient's serum. However, CRC sera inhibited binding less frequently than did sera from patients with pancreatic or gastric tumors.

Girardet et al., J. Immunol. 136:1497-1503 (1986), disclosed mAbs against human colon carcinomas. The L-D1 mAb reacted with a 41 kDa glycoprotein, believed to be the same antigen as that defined by mAb 1083-17-1A (Herlyn et al., 1979, supra). The L-C5 mAb precipitated proteins having molecular weights of 43, 45, 47 and 53 kDa from LoVo colon carcinoma cells. L-D1 did react with cervical carcinoma lines, while L-C5 reacted with breast carcinoma lines. Their binding to pancreatic carcinomas was not examined.

Greiner et al., Science 235:895-898 (1987), discloses mAb 06.2 which reacts with a 90 kDa glycoprotein allegedly found in 75-80% of breast carcinomas and more than 90% of colon carcinomas.

Sakamoto et al., U.S. Pat. No. 4,579,827 (Apr. 1, 1986), discloses a number of mAbs said to be useful for diagnosing or treating human colon cancer by a number of different approaches. None of these mAbs is shown to react with a human colon carcinoma-associated antigen that is a protein of or on human carcinoma cells other than colon, breast and ovarian carcinoma cellsand 31.1 mAbs antibody reacted with molecules having an apparent molecular weights weight of 61.1 kDaand 72 kDa, respectively , from both of these cells lines. These mAbs The 33.28 and 31.1 mAbs did not react with material from human PBMCs or from human tumor cell lines of other histologic types in Western blot analysis.

In order to define better the specificity of the mAbs of the present invention for the immunizing CCAA and to establish whether the mAbs reacted with an immunogenic component of the cell membrane preparation which has been used in clinical immunotherapy trials (Hollinshead et al., supra), the original immunogenic preparation described above was performed by high performance liquid chromatography (HPLC).

The analysis revealed 4 distinct peaks, each of which was tested for immune reactivity (elicitation of DH) in patients with colon carcinoma by skin test (FIG. 1). Among the 10 patients with colon carcinoma tested only the material in peak #4 induced a cutaneous DH reaction. The peak #4 antigen was found to react with mAb 33.28, while mAb 31.1 reacted with peak #3, the next most prominent peak.

The references cited above are all incorporated by reference herein.

EXAMPLE V

Affinity Purification of Colon Carcinoma-Associated Antigen

The mAb 33.28 was used in affinity chromatography to isolate antigen extracted from cells of the HT-29 line. Five mg of purified 33.28 IgG was coupled to CNBr-activated sepharose 4B. The column was pre-eluted with 0.05M diethylamine, pH 11.5, and then equilibrated with 0.14M NaCl/0.01M Tris (pH 8.0). CCAA preparation was applied to the column, and the column was eluted with 0.05M diethylamine, pH 11.5. The eluted fractions were neutralized by the addition of IM Tris-HCl, pH 8.0.

The material bound and eluted from the 33.28 affinity matrix was then subjected to HPLC. The eluted CCAA preparation was adjusted in starting buffer (0.01M sodium phosphate buffer, pH 7.0), applied to a Synchropak Wax weak anion exchange HPLC column (250×4.6 mm) and eluted with a gradient of 0 to 1M NaCl in 0.01M sodium phosphate buffer, pH 6.0, at a flow rate of 1 ml/min. Anion exchange chromatography was performed using a Hewlett-Packard HPLC (HP 1090, Hewlett-Packard, Arondale, Pa.).

Results appear in FIG. 2. The antigenic material derived from HT-29 cells isolated by mAb 33.28 gave a peak that matched peak #4 described above and had similar immunogenic activity in humans, indicating the utility of mAb 33.28 for isolation of a colon cancer preparation which is immunogenic for humans.

EXAMPLE VI

ADCC Activity of mAbs 33.28 and 31.1

In order to be therapeutically useful, a mAb specific for an immunogenic tumor antigen should have the following properties: (a) high tumor tissue specificity, (b) absence of cross-reactivity to normal human tissue, and (c) a biological activity associated with destruction of tumors, such as antibody-dependent cellular cytotoxicity (ADCC).

The ADCC activity of mAbs 33.28 and 31.1 was tested on the colon carcinoma line WiDR as target cell. The melanoma cell line, M-14, served as a specificity control. ADCC was assayed using a conventional 4 hr. ⁵¹Cr release assay using normal human PBMC as effector cells, and the results are shown as percent isotope release (% lysis) (Table 6). The background lysis was 8.3%. At an effector:target ratio of 100:1, mAb 33.28 caused 40.3% lysis of tumor cells, and 31.1 induced 51.8% lysis.

TABLE 6
ADCC Activity of mAbs 33.28 and 31.1
	% Lysis of Target Cells at E:T Ratios:
	Antibody	WiDR	M-14
	or Control	25	50	100	25	50	100
	33.28	23.1	40.3	45.3	6.9	8.4	9.0
	31.1	14.3	26.7	51.8	7.5	6.4	8.7
	OSA1	10.0	9.2	12.2	11.4	14.8	10.9
	NMS	12.2	11.7	13.1	14.2	15.0	11.1
	PBS	8.2	5.1	7.6	11.0	14.2	10.5
	ADCC was assayed by a 4 hour 51Cr release assay. Background 51Cr release was 8.3%. E:T Ratio indicates effector cell-to-target ratios. The mAb or serum was tested at a 1:100 dilution; OSA1 - mAb to-osteosarcoma associated antigens; NMS - normal mouse serum; WiDr - colon carcinoma cell line; M14 - melanoma cell line.

EXAMPLE VII

Detection of Circulating CCAA with mAbs 33.28 and 31.1

The mAbs of the present invention were tested for their ability to detect circulating CCAA in 79 unknown serum samples (Table 7). The assay was based on the ability of the serum samples to inhibit binding of the mAb to the CCAA in ELISA. None of the 50 normal serum samples gave false positive results. Nine of the ten serum samples from patients with active colon carcinoma were positive. None of the sera from disease-free colon cancer patients one year post-resection were positive.

TABLE 7
Detection of Circulating Colon Carcinoma-Associated Antigen
	No.
	of sera inhibiting binding of mAbs;
	33.28		31.1
	DONOR	No. of	<15%	>15%	<15%	>15%
	CONDITION	Samples	(Neg)	(Pos)	(Neg)	(Pos)
	Colon	10	3	7	2	8
	Carcinoma
	Colon	4	4	0	4	0
	Carcinoma
	(Resected)
	Breast	9	9	0	9	0
	Carcinoma
	Melanoma	5	5	0	5	0
	Prostate	1	1	0	1	0
	Cancer
	Normal Serum	50	50	0	50	0
	Colon carcinoma-associated antigens was detected by ELISA. 100 μl of serum were used in each assay.

EXAMPLE VIII

Comparison of Specificity with Other mAbs Reactive with Colon Tumors

Further studies of tumor specificity were conducted using ELISA (Table 8). The mAb 31.1 was compared with CC49, a colorectal carcinoma-specific mAb purified from B-72.3, and a control mouse myeloma protein. 31.1 was shown to react with a narrower range of colorectal carcinomas than did CC49. However, it had a higher degree of specificity, having lower or no cross-reactivity with stomach tumors or normal colon tissue.

TABLE 8
ELISA ON NORMAL AND TUMOR TISSUES USING
MAbs 31.1 CC49 AND MOPC-21
	Tissues	31.1	CC49	MOPC-21
	Colorectal Carcinomas
	1. COCA2A	−	+++	−
	2. COCA2	−	+++	−
	3. COCA3	+	++	−
	4. COCA4	+++	+++	−
	5. G820	±	++	−
	6. G853	+++	++	−
	7. G817	+++	+++	−
	8. G781	−	−	−
	Other Carcinomas
	1. Breast CA1	−	−	−
	2. Breast CA2	−	−	−
	3. Lung CA1	−	−	−
	4. Lung CA2	−	−	−
	5. Ovarian CAD106	−	−	−
	6. Ov CA5	−	−	−
	7. Ov CAV5	−	−	−
	8. Ov CAV45	−	−	−
	9. Ov CAV43	−	−	−
	10. Stomach CA14A	++	+++
	11. Stomach CA12A	−	+++	−
	12. Stomach CA15A	−	−	−
	Other Normal Tissues
	1. Endometrium E21	−	−	−
	2. Endometrium EC19	−	−	−
	3. Endometrium EC17	−	+	−
	4. Endometrium EC18	−	−	−
	(RBC)
	5. Red blood cells 1	−	−	−
	6. RBC 2	−	−	−
	7. RBC 3	−	−	−
	8. RBC 4	−	−	−
	9. RBC 5	−	−	−
	10. RBC 6	−	−	−
	11. RBC 7	−	−	−
	12. RBC 8	−	−	−
	13. RBC 9	−	−	−
	14. RBC 10	−	−	−
	15. RBC 11	−	−	−
	16. Granulocytes	−	−	−
	17. 385	−	−	−
	18. 386	−	−	−
	19. Normal spleen 3	−	−	−
	20. 392 (N. Spleen)	−	−	−
	21. 395 (N. Liver)	−	−	−
	22. 387 (N. Kidney)	−	−	−
	23. 398 (N. Spleen)	−	−	−
	24. 390 (N. Liver)	−	−	−
	25. N. Spleen #1	−	−	−
	26. N. Spleen #2	−	−	−
	27. 800 (N. Colon)	−	++	−
	28. N. Colon (GW)	−	++	−
	29. N. Colon (Meloy)	−	−	−
	30. N. Colon	−	−	−
	31. G1155B (N. Colon)	−	−	−
	32. G1164B (N. Colon)	−	−	−
	33. N. Colon	±	−	−
	34. Normal Stomach A	−	−	−
	35. N. Stomach B	−	−	−
	36. N. Stomach C	−	−	−
	37. Normal Lung	−	−	−
	38. Normal Liver	−	−	−
	CC49 - NCI monoclonal antibody to colorectal carcinomas
	MOPC-21 - negative control myeloma protein
	All monoclonal antibodies were used at 40 ng/well, POGAM at 1:3000 dilution

EXAMPLE IX

In Vivo Localization of mAbs 33 28 and 31.1 to Tumors

The in vivo behavior of the mAbs of the present invention was examined by pharmacokinetic studies using ¹²⁵I-labelled mAb and athymic nude mice bearing LS-174T colon tumor xenografts. Mice implanted with the A375 melanoma were employed as controls. The relative concentration of mAb in tumor as compared to adjacent normal tissue such as liver and spleen is represented by the radiolocalization index (concentration of radiolabelled material in tumor/concentration in surrounding tissue). The biodistribution of ¹²⁵I-labelled 31.1 and 33.28 mAbs is shown in Table 9. Both mAbs were able to significantly concentrate within the tumor, compared to localization in normal tissues (spleen and liver). This selective accumulation was six-fold at 96 hours and about 12-fold at 168 hrs. The radiolocalization indices for both mAbs to tumor as compared to blood, liver and spleen are shown in FIG. 3 and FIG. 4.

TABLE 9
Biodistribution of 125I-mAbs in Tumor-Bearing Athymic Nude Mice
	96 hours		168 hours
	Tissue	LS174T	A375	LS174T	A375
	A. mAb 31.1
	Blood	7.30	NA	4.67	4.42
	Tumor	21.92	NA	25.43	2.91
	Liver	3.74	NA	2.16	1.24
	Spleen	3.68	NA	2.41	1.32
	B. mAb 33.28
	Blood	7.81	NA	5.58	3.95
	Tumor	13.12	NA	15.50	2.24
	Liver	2.55	NA	1.74	1.68
	Spleen	2.31	NA	1.70	1.92
	Results are expressed in % injected dose/gram of tissue.
	LS174T = colon carcinoma; A375 = melanoma.

EXAMPLE X

Immunohistochemical Studies

The ability to detect tumor markers in the serum, image the related neoplastic process and define the cell population of that neoplasm by immunohistochemistry depends on the ability of specific mAbs to selectively characterize a tumor population.

The mAbs 31.1 and 33.28 were tested in more than 50 colon carcinomas by means of immunoperoxidase staining. They have been found to be highly reactive with the colon neoplasm and did not interact with the adjacent normal tissues. When polyps were evaluated, the wholly benign lesions such as villo-tubular adenomas showed no reactivity. Villous adenomas undergoing transformation reacted only at the site of malignancy. When similar tissues were evaluated using the more common carbohydrate-antigen derived mAbs, normal adjacent colon tissue reacted equally with the neoplastic portion of the specimen. With the 31.1 and 33.28 mAbs, each appeared to stain different cell populations within the tumor. This suggests that surface antigens representing different oncogene products were being defined.

EXAMPLE XI

Selective Binding to Subpopulations of Epithelial Cells in Paraffin-Embedded Benign and Malignant Mammary Lesions

Forty-one formalin-fixed, paraffin-embedded benign and malignant breast specimens were studied with mAbs 31.1 and 33.28, using the avidin-biotin staining method. Without enzymatic pretreatment, positive staining of epithelium was observed on the cell surface and in the cytoplasm with both antibodies. 7/21 (33%) duct carcinogens were positive with mAb 31.1 as were 5/20 (25%) samples of benign breast disease. 10/21 (48%) duct carcinomas were positive with mAb 33.28 together with 7/20 (35%) specimens of benign mammary disease. 10 to 75% of the cell population was positively stained. These results indicate that the antigens defined by mAbs 31.1 and 33.28 are expressed in a select group of women with breast disease and would be useful for diagnosis of said disease.

EXAMPLE XII

Selective Binding to Subpopulations of Epithelial Cells in Fresh Frozen Benign and Malignant Ovarian Tumors

Fresh frozen tissue biopsies obtained from twenty-one ovarian tumors subjected to immunocytochemical analysis were studied with mAbs 31.1 and 33.28 using the avidin biotin indirect immunoperoxidase assay. Focal positive staining was observed in 4/7 papillary mucous, 1/1 mucinous and ½ endometroid adenocarcinoma. None of the nonepithelial ovarian tumors stained positive using these monoclonal antibodies. These results indicate that the antigens as defined by mAbs 31.1 and 33.28 are expressed in a select group of woman with ovarian cancer and would be useful for diagnosis of said disease.

EXAMPLE XIII

Immunoreactivity of Human Carcinoma Tissues and Cell Lines with mAbs 31.1 and 33.28

The mAbs 31.1 and 33.28 were used to screen a panel of cell lines including colon adenocarcinoma, lymphoma, leukemia and neuroblastoma lines.

Using the avidin-biotin immunoperoxidase staining system, the mAbs 31.1 and 33.28 were shown to strongly bind to colon adenocarcinoma cell lines WIDR and HT-29. Immunoreactions were not observed with KG1-a, HL-60, Molt-3 and JUKRAT cell lines. Both antibodies reacted weakly with one lymphoma line (JY). The mAb 33.28 reacted weakly with one leukemic line (K562) and a neuroblastoma line (U87. MG). These results confirm and extend previous flow cytometry and immunofluorescent results in which it has been reported that strong binding reactions were observed with these mAbs with colon adenocarcinoma cell lines and reactions were not observed with other tumor cell lines. Using flow cytometry, mAb 31.1 reacted with 85% of HT-29 and WIDR colon carcinoma cells but not with SKBR-3 breast cancer cells.

Both antibodies were extensively shown to bind distinctively to colon carcinoma tissues (mAb 33.28-84%, mAb 31.1-64%), and not to normal tissues or malignant tissues including neuroblastoma tissues ( 0/3), lymphomas ( 0/3) and leukemic infiltrates ( 0/3) tested. These results suggest that these antibodies can serve as a useful research tool in evaluating tumor markers in cancer and cell biology research.

EXAMPLE XIV

Expression and Characterization of Chimeric Antibodies Against Human Colorectal Carcinoma-Associated Antigen

A chimeric mouse/human heavy chain gene was constructed by splicing the exon of the 31.1 antibody heavy chain variable region gene to the exon of the human gamma1 chain contstant region gene using the polymerase chain reaction. Subsequently, the 31.1 chimeric gene was cloned into a retroviral expression vector pLgptCXII and transfected into the packaging cell line PA317. The transfected cells (PA317H) were cultivated with another packaging cell line PA317L, which contained an irrelevant mouse/human chimeric light chain gene in retroviral expression vector pLneoCXII, and SP2/0-Ag14 cells. The transduced SP2/0-Ag14 cells yielded a complete chimeric antibody. Chi #1 which reacted with horseradish peroxidase-conjugated igG of goat anti-human IgG Fc in ELISA analyses, which indicated that the constant region of Chi #1 was human. Cytofluorometry analysis indicated that Chi #1 stained human colorectal carcinoma cell lines HT-29 and LS174T but not a human lung carcinoma cell line A-427. Antibody-dependent cell-mediated cytoxicity (ADCC) assay indicated that Chi #1 lysed Ls174T cells. These results shows that Chi #1 retained the antigen-binding specificity of the parental 31.1 mouse monoclonal antibody, suggesting the useful of this chimeric antibody in ascertaining prognosis of colon carcinoma.

Having now fully described this invention, it will be appreciated by those skilled in the art that the same can be performed within a wide range of equivalent parameters, concentrations, and conditions without departing from the spirit and scope of the invention and without undue experimentation.

While this invention has been described in connection with specific embodiments thereof, it will be understood that it is capable of further modifications. This application is intended to cover any variations, uses, or adaptations of the inventions following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains and as may be applied to the essential features hereinbefore set forth as follows in the scope of the appended claims.

Claims

1. A monoclonal antibody specific for a purified human colon carcinoma-associated protein antigen, wherein said antigen has the following characteristics:

(a) said antigen is purified to the extent that the membrane fractions are free of HL-A antigen and are substantially free from non-immunogenic glycoprotein fractions;

(b) said antigen is not detectable on normal colon cancer free human tissues;

(c) said antigen is not detectable on human carcinoma cells other than colon carcinoma cells;

(d) said antigen is specifically immunogenic in humans; and

(e) said antigen induces an immune response in humans having colon carcinoma which is expressed as cell mediated immunity which is murine monoclonal antibody 33.28 as produced by hybridoma pca 33.28, deposited with the American Type culture collection and assigned accession number PTA-5413.

2. An antibody according to claim 1 which is mouse monoclonal antibody 33.28 (ATCC HB-12315) or an antibody which binds specifically to a colon carcinoma-associated epitope that specifically binds to monoclonal antibody 3328.

3. An antibody according to claim 2 1 wherein said colon carcinoma-associated antigen is a protein having a molecular weight of about 61.1 kilodaltons as measured by gradient polyacrylamide gel electrophoresis.

4. An antibody according claim 1 A monoclonal antibody specific for a purified human colon carcinoma-associated protein antigen, which is mouse monoclonal antibody 31.1, as produced by hybridoma pca 31.1 (ATCC HB-12314) , deposited with the American Type culture collection and assigned accession number PTA-2497or an antibody which binds specifically to a colon carcinoma-associated epitope that specifically binds to monoclonal antibody 31.1 .

5. An A monoclonal antibody according to claim 4 wherein said colon carcinoma-associated antigen is a protein having a molecular weight of about 72 kilodaltons which comprises an antigen-binding region obtained from the H chain of a murine monoclonal antibody 33.28 as produced by hybridoma pca 33.28, deposited with the American Type culture collection and assigned action number PTA-5413.

6. An antibody according to claim 2 1 wherein said colon carcinoma-associated antigen is a glycoprotien, the protein component having a molecular weight of 61.1 kilodaltons as measured by gradient polyacrylamide gel electrophoresis.

7. An antibody according to claim 1, 4 or 5 immobilized on a solid phase.

8. An antibody according to claim 1, 4 or 5 which is detectably labeled.

9. An antibody according to claim 8 wherein said detectable label is a radiolabel.

10. An antibody according to claim 1, 4 or 5 conjugated to a cytotoxic radionuclide.

11. An antibody according to claim 1, 4 or 5 conjugated to a cytotoxic drug.

12. An antibody according to claim 1, 4 or 5 conjugated to a cytotoxic protein.

13. A composition comprising an antibody according to claim 10 in combination with a pharmaceutically acceptable carrier.

14. A composition comprising an antibody according to claim 11 in combination with a pharmaceutically acceptable carrier.

15. A composition comprising an antibody according to claim 12 in combination with a pharmaceutically acceptable carrier.

16. A monoclonal antibody against the monoclonal antibody of claim 1.

17. A monoclonal antibody against the monoclonal antibody of claim 2.

18. A monoclonal antibody against the monoclonal antibody of claim 3.

19. A monoclonal antibody against the monoclonal antibody of claim 4.

20. A monoclonal antibody against the monoclonal antibody of claim 5.

21. A monoclonal antibody against the monoclonal antibody of claim 6.

22. An immunoassay for detecting a colon carcinoma-associated antigen which binds to mouse monoclonal antibody 33.28 (ATCC HB-12315) as produced by hybridoma pca 33.28, deposited with the American Type culture collection and assigned accession number PTA-5413, in a sample comprising:

(a) contacting said sample with an effective binding amount of the antibody according to claim 1; and

(b) detecting said antigen by detecting the binding of the antibody to the purified colon carcinoma-associated protein antigen.

23. An immunoassay for detecting a colon carcinoma-associated antigen which binds to mouse monoclonal antibody 31.1 (ATCC HB-12314) , as produced by hybridoma pca 31.1, deposited with the American Type culture collection and assigned accession number PTA-2497, in a sample comprising:

(a) contacting said sample with an effective binding amount of the antibody according to claim 1 4 or claim 5; and

(b) detecting said antigen by detecting the binding of the antibody to the purified colon carcinoma-associated protein antigen.

24. A method for diagnosing colon cancer in humans comprising:

(a) removing a histological specimen from a patient suspected of having a colon cancer;

(b) contacting the specimen with monoclonal antibody 33.28 (ATCC HB-12315) , as produced by hybridoma pca 33.28, deposited with the American Type culture collection and assigned accession number PTA-5413;

(c) staining the specimen with an immunohistochemical stain; and

(d) detecting the presence of the antigen-antibody complex by the stain.

25. A method for diagnosing colon cancer in humans comprising:

(a) removing a histological specimen from a patient suspected of having colon- carcinoma;

(b) contacting the specimen with mouse monoclonal antibody 31.1 (ATCC HB-12314) , as produced by hybridoma pca 31.1, deposited with the American Type culture collection and assigned accession number PTA-2497);

(c) staining the specimen with an immunohistochemical stain; and

(d) detecting the presence of the antigen-antibody complex.

26. A method according to claim 24 wherein the stain is an avidin-biotin immunoperoxidase stain.

27. A method according to claim 25 wherein the stain is an avidin-biotin immunoperoxidase stain.

28. A kit for the immunohistochemical detection of colon carcinoma comprising:

(a) mouse monoclonal antibody 31.1 (ATCC HB-12314) , as produced by hybridoma pca 31.1, deposited with the American Type culture collection and assigned accession number PTA-2497;

(b) reagents for immunoperoxidase and secondary antibody;

(c) immunoperoxidase; and

(d) colorizing reagents.

29. A kit for the immunohistochemical detection of colon carcinoma comprising:

(a) mouse monoclonal antibody 33.28 (ATCC HB-12315 , as produced by hybridoma pca 33.28, deposited with the American Type culture collection and assigned accession number PTA-5413;

(b) reagents for immunoperoxidase and secondary antibody;

(c) immunoperoxidase; and

(d) colorizing reagents.

30. A compartmentalized kit for the detection of a human colon carcinoma-associated antigen, wherein the antigen has the following characteristics:

(a) said antigen is purified to the extent that the membrane fractions are free of HL-A antigen and are substantially free from non-immunogenic glycoprotein fractions;

(b) said antigen is not detectable on normal colon cancer free human tissues;

(c) said antigen is not detectable on human carcinoma cells other than colon carcinoma cells;

(d) said antigen is specifically immunogenic in humans; and

(e) said antigen induces an immune response in humans having colon carcinoma which is expressed as cell mediated immunity,

31. A kit according to claim 30 wherein the reporter molecule is a radioisotope, an enzyme, a fluorescent molecule, a chemiluminescent molecule or a bioluminescent molecule.

32. A kit according to claim 30 wherein the reporter molecule is an enzyme.

33. A kit according to claim 30 wherein the kit further comprises a third container adapted to contain a substrate for the enzyme.

34. A compartmentalized kit for the detection of a human colon carcinoma-associated antigen, wherein the antigen has the following characteristics:

(a) said antigen is purified to the extent that the membrane fractions are free of HL-A antigen and are substantially free from non-immunogenic glycoprotein fractions;

(b) said antigen is not detectable on normal colon cancer free human tissues;

(c) said antigen is not detectable on human carcinoma cells other than colon carcinoma cells;

(d) said antigen is specifically immunogenic in humans; and

(e) said antigen induces an immune response in humans having colon carcinoma which is expressed as cell mediated immunity,

35. A kit according to claim 34 wherein the reporter molecule is a radioisotope, an enzyme, a fluorescent molecule, a chemiluminescent molecule or a bioluminescent molecule.

36. A kit according to claim 32 wherein the reporter molecule is an enzyme.

37. A kit according to claim 33 wherein the kit further comprises a third container adapted to contain a substrate for the enzyme.

38. A compartmentalized kit for the detection of a human colon carcinoma-associated antigen, wherein the antigen has the following characteristics:

(a) said antigen is purified to the extent that the membrane fractions are free of HL-A antigen and are substantially free from non-immunogenic glycoprotein fractions;

(b) said antigen is not detectable on normal colon cancer free human tissues;

(c) said antigen is not detectable on human carcinoma cells other than colon carcinoma cells;

(d) said antigen is specifically immunogenic in humans; and

(e) said antigen induces an immune response in humans having colon carcinoma which is expressed as cell mediated immunity,

39. A kit according to claim 38 wherein the reporter molecule is a radioisotope, an enzyme, a fluorescent molecule, a chemiluminescent molecule or a bioluminescent molecule.

40. A kit according to claim 38 wherein the reporter molecule is an enzyme.

41. A kit according to claim 38 wherein the kit further comprises a third container adapted to contain a substrate for the enzyme.

42. The monoclonal antibody of claim 1 which is a chimeric antibody.

43. The A chimeric antibody according to claim 42 which is a chimeric mouse/human antibody chi #1 as produced by the cell line deposited with the American Type culture collection and assigned accession number (ATCC CRL-12316) .

44. The chimeric antibody according to claim 42 wherein said colon carcinoma-associated antigen is a protein having a molecular weight of 72 kilodalton.

45. A composition comprising the chimeric antibody according to claim 42 43 in combination with a pharmaceutically acceptable carrier.

46. A monoclonal antibody against the chimeric antibody of claim 42.

47. An immunoassay for detecting a colon carcinoma-associated antigen which binds to the mouse/human chimeric antibody chi #1 as produced by the cell line deposited with the American Type culture collection and assigned accession number (ATCC CRL-12316) of claim 42 in a sample comprising:

(a) contacting said sample with the chi #1 antibody according to claim 42 ; and

(b) detecting said antigen by detecting the binding of said antibody to the purified colon carcinoma-associated protein antigen.

48. A method for diagnosing colon cancer in humans comprising:

(a) removing a histological specimen from a patient suspected of having a colon carcinoma;

(b) contacting the specimen with a chimeric antibody which binds to an antigen according to claim 1 43;

(c) staining the specimen with an immunohistochemical stain; and

(d) detecting the presence of the antigen-antibody complex by the stain.

49. A method for diagnosing colon cancer in humans comprising:

(a) removing a histological specimen from a patient suspected of having a colon carcinoma;

(b) contacting the specimen with mouse/human chimeric antibody which binds to an antigen which binds to mouse/human chimeric antibody chi #1 (ATCC as produced by the cell line deposited with the American Type culture collection and assigned accession number CRL-12316) ;

(c) staining the specimen with an immunohistochemcial stain; and

(d) detecting the presence of the antigen-antibody complex by the stain.

50. A kit for the immunohistochemical detection of colon carcinoma comprising:

(a) mouse/human chimeric antibody chi #1 (ATCC CRL-12316);

(b) reagents for immunoperoxidase and secondary antibody;

(c) immunoperoxidase; and

(d) colorizing reagents.