In vivo and in vitro clinical applications of monoclonal antibodies against TAG-72

Nucl. Med. Eiol. Vol. 18, No. 4, pp. 39.5401, hr. J. Radial. Appl. Instrum. Part B Printed in Great Britain

0883-2897/91 53.00 + 0.00 Pergamon Press plc

1991

In Vivo and In Vitro Clinical Applications of Monoclonal Antibodies Against TAG-72 DAVID COLCHER*, DIANE E. MILENIC, MARIO ROSELLI and JEFFREY

PATRIZIA

FERRONI,

SCHLOM

National Cancer Institute, The National Institutes of Health, 9000 Rockville Pike, Bethesda, MD 20892. U.S.A. Monoclonal antibodies against a tumor-associated antigen (TAG-72) with mucin-like properties have been generated. MAb B72.3 was used to identify and help characterize this antigen. B72.3 has been successfully used for the localization of tumor metastases in situ after i.v. administration. MAb B72.3 has also been used in conjunction with CC49, another anti-TAG-72 MAb, to measure TAG-72 levels in sera and effusions. TAG-72 can be found in the fluids of patients with adenocarcinomas from many different sites. This CA 72-4 double determinant radioimmunoassay in conjunction with assays for carcinoembryonic antigen can identify patients with malignancies with greater sensitivity than either assay alone.

Introduction Monoclonal antibodies (MAb) that are reactive with tumor associated antigens (TAA) have led to many of the recent advances made in tumor immunology. MAbs have made it possible for investigators to identify and characterize TAAs that had previously been unknown. They have also enabled the characterization of TAAs that were still poorly defined. MAbs generated against TAAs have the potential of altering the management of cancer patients. In vivo applications of several MAbs have shown their utility in both immunodiagnostic and immunotherapeutic protocols (Goldenberg, 1990a, b; Carrasquillo et al., 1988; Colcher et al., 1987a, b). In vitro, serum radioimmunoassays for the detection of tumor-associated antigens (TAAs) have been used (e.g. CEA, CA 19-9, CA 15-3, CA 125, DU-PAN-2) to monitor the course of disease of several carcinomas (Thomson et al., 1969; Zamchek, 1974; Klug et al., 1984; Del Villano et al., 1983; Hayes et al., 1985; Colomer et al., 1986; Metzgar et al., 1984). MAb B72.3 was generated against membranes from a human carcinoma metastasis (Colcher et al., 1981). It has been shown to react with a high molecular weight glycoprotein that has many properties of a mucin (Johnson et al., 1986). Using immunohistochemical techniques, MAb B72.3 has been shown to react with 100% of the adenocarcinomas ____ *All correspondence should be addressed to: Dr David Colcher, University of Nebraska Medical Center, Department of Pathology/Microbiology, Omaha, NE 68198-3135, U.S.A.

of the ovary and endometrium, 96% of lung, 94% of colon, 84% of breast, 88% of pancreas and 88% of the carcinomas of the gastrointestinal tract (Stramignoni et al., 1983; Thor et al., 1986a; Ohuchi et al., 1986; Takasaki et al., 1988). No, or only trace amounts of, MAb B72.3 reactivity has been observed with a wide range of adult normal tissues, with limited reactivity in a few benign tissues such as those of the breast and colon tissue, as well as transitional epithelium (Thor et al., 1986b; Wolf et al., 1989). In addition, TAG-72 has been shown to be expressed at high levels in secretory endometrium; it is not found, however, in resting or post-menopausal endometrium (Thor et al., 1987). TAG-72 has also been detected in fetal colon, stomach and esophagus (Thor et al., 1986b). Experimental studies have shown the utility of 1311-B72.3IgG to localize and treat human colon carcinoma xenografts in athymic mice (Colcher et al., 1984; Esteban et nl., 1987). Following an i.v. injection of lz51-B72.3 IgG, over 10% of the injected dose per gram was observed in the tumor by 2 days. The radiolocalization index (RI) rose over a 7 day period, with tumor: liver, tumor: spleen and tumor: kidney ratios attaining approx. 18:l (Colcher et al., 1984). There was no specific uptake of “‘1-B72.3 in any of the normal organs, including brain, muscle, stomach, intestines, uterus and ovary. In vivo studies Studies were then initiated to define the specific tumor localization of i.v. administered “‘I-labeled MAb B72.3 IgG in colorectal cancer patients (Colcher et al., 1987a; Carrasquillo et al., 1988). 395

DAVIDCCILCHER

396

These data were obtained by direct analyses of biopsy materials (tumor and normal tissues removed for staging) from patients with metastatic colorectal cancer who had been given “‘I-MAb B72.3 approx. 1 week prior to surgery. Radiolocalization indices were obtained by direct analyses of “‘I cpm/g tumor vs cpm/g of normal tissues (Table 1). Using as an arbitrary “positive” localization a RI for carcinoma of 3 times greater than adjacent normal tissue (i.e. RI > 3.0), tumor lesions from 17 of 20 patients showed a positive localization. In 8 of these patients, moreover, all tumor lesions demonstrated RIs of > 3. In 5 patients, RIs of some of the lesions were > 10 and as high as 46. In total, 70% (99 of 142) of tumor lesions showed RIs of ~3. Only 12 of 210 histologically confirmed normal tissues examined showed RIs of >3; 10 of these tissues were either anatomically adjacent to tumor or draining tumor masses. In the case of 2 patients, elevated RIs appeared to be due to high levels of circulating MAb-antigen complexes (see below) that may be deposited in the spleen. RIs B 3 were defined in tumor masses located in the liver, small and large intestines, peritoneum as well as other sites. Planar imaging resulted in positive y scans (confirmed at surgery) in 14 of 27 patients. These scans accurately identified carcinoma in the liver, peritoneum, rectum and bone. An isotope identical control IgG, BL-3, coinjected into selected patients showed RIs considerably lower than that of B72.3. No toxicity of any type was observed using any of the doses of the 2 MAbs used. These studies thus define the delivery of radiolabeled MAb B72.3 to carcinoma lesions vs adjacent and distal normal tissues. These data also demonstrate the selective binding of MAb B72.3 to colon carcinoma lesions in uivo. This may lead the way for other diagnostic and potential therapeutic applications of this antibody and/or its fragments.

et al.

Positive y scans were confirmed at surgery and by histopathological analysis of tumor masses. The scan results, however, will most probably improve with the use of radionuclides, such as *Tc, “‘In and lz31,that are better suited for the y-cameras. While there is clearly more experimentation required to optimize B72.3 y scanning, preliminary studies clearly demonstrate the ability of this monoclonal antibody to selectively bind colorectal cancer lesions at many body sites. Indeed, the RIs of up to 46 are probably artifactually low, since, approx. 10% of the injected dose was still in the blood pool at the time of biopsy of the normal and carcinoma tissues. The optimal targeting of tumor lesions by MAbs for diagnostic or therapeutic applications requires that the MAb retain its integrity and immunoreactivity following administration in patients. Several investigators have presented pharmacokinetic studies of various MAbs (Carrasquillo et al., 1988; Hnatowich et al., 1985, 1987; Rosenblum et al., 1985; Bertram et al., 1986; Khazaeli et al., 1988; Schroff et al., 1987). These reports have shown that the circulating half-lives for individual antibodies vary and are dependent upon tissue reactivities and the dose of MAb administered. Serum samples from the 27 colorectal carcinoma patients who received i3’IB72.3 by i.v. administration were analyzed for TAG72 levels, the integrity and immunoreactivity of the injected MAb, the retention of the radionuclide on the B72.3 immunoglobulin molecule, the formation of immune complexes and the development of human anti-mouse immunoglobulin antibodies (HAMA) using solid-phase RIA, SDS-PAGE and HPLC techniques (Colcher et al., 1990). The amount of immunoreactive MAb B72.3 present in patients’ serum samples was quantitated using a solid-phase RIA. Circulating immunoreactive MAb B72.3 was measured in 19 of 27 patients. The other

Table 1. Localization of “‘1-B72.3 IgG to carcinoma and normal tissues Carcinoma

Tissue

c3

Liver Colon Small intestine Lymph node Lung Peritoneum Pelvis Soft tissue Spleen OthcrQ

4 5 3 4 3 18 1 2 3 0

24 10 1 7 6 18

c3

3-10

z 10

20 35 10 19 29 4 53 1 6 21

0 0

0

2 3 2 2 8 0 3 1 2

86 71 57 69 73 59 50 75 25 100

$%)

70

198 (92%)

>I0

(3Y%)

(3,‘,.,

1

3 0 5

Normal Percent positive RIt

RI’

I

Total

RI

7-n 211 2;ii,** 0 1 0 0 1x1 (48%)

0 0

Percent positive RI 0 0

0 0 0

17 9 6 0

:++ 0 0

8: 0 5

(2;)

6

‘The % ID/g of every biopsy specimen was divided by the % ID/g of the normal tissue(s) used as standard to determine the radiolocalization indices (RI). tRI > 3 are considered positive for MAb localization. fFor carcinoma lesions the additional tissues include ovaries (n = 3), fallopian tubes (n = 2) pouch of Douglas (n = 1) and vena cava (n = 1). $Additional normal organs include gall bladder (n = 5), vagina (n = 2), cervix (n = 2) uterus (n = 2). fallopian tube (n = 2). bone (n = 2), skin (n = 2) 1 each of: epicardium, ureter, vas deferens, seminiferous duct and portal vein. TTissues adjacent to tumor. /lAbundant lipogranulomas positive for TAG-72 by immunoperoxidase staining. **Granulation tissue. WPatients with circulating immune complexes. t$GaIl bladder from patient with liver metastasis. Taken from Esteban et al. (1987).

397

Clinical applications of MAbs against TAG-72

8 patients studied received less than 0.5 mg which resulted in insufficient circulating antibody to be quantitated using the existing methodologies. When the values are normalized to the level observed at 5 min post MAb administration, the pharmacokinetits of the patients are similar. The amount of antibody found in the sera follows a biphasic pattern of clearance (Fig. 1). There is an initia equilibration phase with a rapid removal of antibody from the sera within the first 4 h after antibody administration; this is followed by a more gradual clearance of the antibody. While there is a wide range of differences between individual patients, the clearance pattern of the immunoreactive B72.3 closely follows that of the ‘3’I-labeled B72.3 (Fig. 1, solid line), suggesting that the iodinated B72.3 can be used to follow the serum clearance of the circulating immunoreactive antibody. The pharmacokinetic patterns were similar for the doses of B72.3 ranging from 1 to 20 mg and was independent of whether there was any tumor-specific uptake of the radiolabeled MAb. SDS-PAGE analysis of the “‘I-labeled B72.3 in patient sera revealed that virtually all the radioactivity in the serum is associated with the IgG molecule as visualized by autoradiography. HPLC analysis at various time points post-injection of the antibody supported the results obtained by the SDS-PAGE analysis. In the majority of the serum samples tested, the radioactivity remains associated with the IgCi peak [Fig. 2(A-B)]. HPLC analysis also demonstrated the presence of immune complexes in the sera of 12 patients, radioactivity was associated with an entity having a higher molecular weight than IgG which was detected in the column void volume [Fig. 2(C)]. The alteration in the retention time of radiolabeled B72.3 on the columns was observed at the early time points and remained consistent throughout the time points analyzed. The amount of 13’1associated with the void volume ranged from 0 to 56%. The shift toward the higher molecular entities may

Fig. 1. Patient serum levels of MAb B72.3 following iv. administration. Serum levels of immunoreactive MAb B72.3 were detected using a solid-phase RIA described in Materials and Methods; HF (O), TR (a), CC (m) and PH (A). The patients received 19.24, 17.8, 19.24 and 19.7 mg of ‘%B72.3, respectively. The percentage of the injected dose of 13’1-B72.3IgG (solid line) is included for comparison. Taken from Colcher et al. (1990).

14 12 10 6 6 4 2

6 -3 n 's

4

23 t "2 1

(C)

TIME

(mln 1

Fig. 2. HPLC analysis of “‘IB72.3 IgG in patient sera. Serum samples drawn at various time points were applied to a TSK-2000 and TSK-3000 column in tandem arrangement. Fractions were collected and the “‘I measured in a y scintillation counter. Serum samples from patients BB (A), RC (B) and CP (C) were drawn at 30 mitt (m), 1h (0) 2 h (@), 4 h(A), 24 h (A) and 96 h (V). ‘3’I-B72.3IgG in 0.2 M sodium phosphate, pH 6.8 (0) or in pre-study serum (0) were used as standards. Taken from Colcher et al. (1990).

have been due to either the formation of immune complexes, either antigen-antibody or antibodyantibody, or aggregation of the IgG. There appears to be a correlation between the formation of immune complexes and the presence of elevated levels (i.e. > 16 U/ml) of serum TAG-72. The presence of 13’1B72.3 IgG in complexes, however, does not appear to lead to a sequestration of the complexes that results in the alterations of the clearance of the antibody from the plasma. In patients whose sera had high levels of the radiolabeled B72.3 in complexes, 70-80% of the ‘3’I-labeled B72.3 cleared the plasma by 96 h. This is similar to the average clearance of the MA\, from the blood of all the patients studied. The formation of immune complexes was also visualized by autoradiography of the SDS-polyacrylamide gels of serum samples. Formation of immune complexes was very rapid, as evidenced by the formation of a high molecular weight entity when ‘*‘I-B723 is added to the pre-study sera just prior to loading of the polyacrylamide gel.

398

DAVID COLCHERer al.

Twenty-five of the patients were evaluated for the development of anti-mouse IgG antibodies (HAMA) using a solid-phase double-determinant RIA (Colcher et al., 1990). The antibody titer was calculated by determining the titer obtained at a given number of counts bound and normalizing that value to a standard serum that is positive for HAMA in order to minimize interassay variability. Several different HAMA response patterns were detected in the 25 patients’ sera that were analyzed, some patients developed HAMA as early as 5-7 days post-MAb injection. Higher doses of administered MAb B72.3 correlation with the development of HAMA (P = 0.007). In the group of patients that received 0.8-l .5 and 3.7-20 mg, 75 and 7 I%, respectively, were positive for HAMA. In vitro studies Some of the major potential uses of MAbs will be their ability to detect specific carcinoma associated antigens in sera of asymptomatic individuals and to indicate the presence of occult metastatic lesions in patients whose primary ,carcinoma has been previously excised. A competitive radioimmunoassay (RIA) for the detection of TAG-72 in tissue extracts and serum was first developed (Paterson normal et al., 1986). Serum from apparently individuals contained an average of 2.2 U/mL of TAG-72 while 35% of carcinomas patients had elevated levels of serum TAG-72 distinct from CEA, GICA, 0C125 and CA 15-3. A double determinant RIA for the detection of TAG-72 was subsequently developed (Klug et al., 1986). In this assay, B72.3 was bound to a solid-phase matrix, and serum and “‘IB72.3 were added. In this assay, serum from healthy blood donors (1099 tested) had a mean level of 1.8 U/mL of TAG-72. Approximately 57% of patients with colorectal cancer exhibited elevated levels of TAG-72 using a reference level of 10 U/mL, with 1.3% of the blood donors having serum TAG-72 levels above 10 U/mL. Using a reference value of 20 U/mL, 37% of the colorectal cancer patient sera had elevated levels of TAG-72 while only 2 of the 1099 blood donor sera had elevated values. The TAG-72 antigen was then purified and used to prepare a second generation of anti-TAG-72 MAbs. One of these MAbs, designated CC49, was shown to react with an epitope on TAG-72 that could be distinguished from that recognized by B72.3. Moreover, while the K, of B72.3 was shown to be 2.5 x lo9 M-‘, that of CC49 was considerably higher at 16.2 x lo9 M-‘. These findings led to the development of a more efficient assay for TAG-72 using CC49 on a solid-phase matrix and ‘251-B72.3as the detecting antibody. This combination enabled the development of a sequential assay (designated CA 72-4) that showed optimal quantitative properties as demonstrated by such parameters as linear dose-response, high reproducibility and lack of

serum-matrix and “hook-back” effects (Gero et af., 1989). Only 2.5% of 1258 normal sera and 7.0% of 134 sera from patients with benign gastrointestinal diseases had TAG-72 levels greater than 6 U/mL in the CA72-4 assay (Gero et al., 1989). Approximately 40% of 504 patients with gastrointestinal malignancies had serum TAG-72 levels of greater than 6 U/mL (65% of the patients with advanced disease). Thirtysix percent of the patients with adenocarcinomas of the lung and 26% of patients with ovarian cancer (53% Stage IV patients) also had elevated serum TAG-72 levels (Table 2). In fact, the presence of elevated levels of TAG-72 in the patients’ serum prior to entry in radioimmunoscintigraphic studies with i.v. injected r3’I-B72.3 IgG was shown to correlate with the detection of lesions by y scanning. Serum TAG-72 therefore may be a potentially useful criteria for patient selection for immunodiagnostic or immunotherapeutic procedures using MAb B72.3. A poor correlation was found between the CEA and TAG-72 values of sera obtained from gastric cancer patients. Thirty-four percent of CEA negative cases were scored positive in the CA 72-4 assay (Fig. 3, top panel). Forty-one percent of the CEA negative effusions scored positive in the CA 72-4 assay, (Fig. 3, bottom panel) suggesting the complementarity of the CA 72-4 assay to CEA assays in the analysis of sera from patients with certain malignancies (Gero et af., 1989; Ohuchi et al., 1990, 1991). Tumor markers such as carcinoembryonic antigen (CEA) (Thomson et al., 1969; Zamchek, 1974), CA 125 (Klug et al., 1985) CA 19-9 (Del Villano et al., 1983), and CA 15-3 (Hayes et al., 1985; Colomer et al., 1986) have been also investigated as to their possible utility in discriminating between effusions of Table 2. Summary of serum TAG-72 levels Patient

Number of samples

Percentage with TAG-72 > 6 U/mL

12 291 93 177 74 13 4 7 165 28 60 61 16 97 15 118 50 25 29

25 40 57 43 60 31 50 57 16 36 17 II 19 26 53 16 20 4 3

Carcinomas

Esophagus Gastric Stage IV Colorectal Stage IV and V Pancreas Gall bladder Choledochus Lung Adenocarcinoma Squamous cell Small cell Non-small cell Ovarian Stage IV Breast Prostatic Squamous Sarcoma Benign diseases

Stomach Colorectal Misc. diseases Normal

75 59 17 1258

5 8 0 2.5

*Data taken from Gero er al. (1989) and Ohuchi el al. (1990, 1991).

399

Clinical applications of MAbs against TAG-72

b

0 a

501

n541

4 5

n=lO 10

20

50

.

>lOO

I

Fig. 3. Complementarity of TAG-72 and CEA values in patient fluids. Comparison of TAG-72 and CEA levels in the sera (n = 99; top panel) of patients with Stage III and IV gastric carcinoma and effusions of patients with adenocarcinomas (n = 49; bottom panel). The cutoff levels are indicated with the dashed lines. Taken from Gero et al. (1989) and Ferroni et al. (1991).

benign and malignant origin. The evaluation of TAA levels in effusions is starting to be studied to a greater extent. This is being used as another way to address the recurrent problem in clinical practice of whether an effusion is a manifestation of metastatic disease, infection or metabolic or cardiovascular imbalance. The cytological evaluation of the cells found in effusions is widely used for this determination, but the consistent rate of false negatives, approx. 38% (Dines et al., 1975; Dekker et al., 1978) has led several investigators to look for other useful diagnostic adjuncts. Many of these false negatives may be due to the presence of too few tumor cells to achieve a cytological diagnosis. The application of immunocytochemistry, using monoclonal antibodies reactive with different tumor associated antigens (TAAs) may also be useful in the detection of adenocarcinoma and its discrimination from mesothelial and lymphoid ceils in malignant effusions (Szpak et al., 1984; Mottolese et al., 1988). CEA levels were evaluated in effusions by several investigators, and elevated levels were found to correlate with malignancies with sensitivities ranging from 34% using a cutoff value of 12 ng/mL (Rittgers et al., 1978), to approx. 70% using cutoff values of 5 or 15 ng/mL (Whitaker ef al., 1986; Pinto et al., 1987). Nevertheless, some investigators have been unable to detect any significant difference between benign and

malignant effusions (Nystrom et al., 1977; Whiteside et al., 1979; Mezger et al., 1988). Fluids were obtained from 51 patients with adenocarcinomas, 27 with non epithelial malignancies and 68 with benign disorders as determined by pathological and/or cytological evaluation. The CA 72-4 radioimmunoassay (using a cutoff value of 8.5 U/mL) detected the TAG-72 antigen in 51% of adenocarcinoma patients’ effusions, while only 1 of 68 benign specimens had an elevated TAG-72 level (Fig. 4). Similarly, CEA levels above 5 ng/mL were found in 55% of the fluids from patients with adenocarcinoma and 3.2% of effusions from patients with benign disease (Ferroni et al., 1991). The effusions from patients with proven adenocarcinema were further analyzed to determine if the tumor markers correlated with the tumor site (Ferroni et al., 1991). Elevated CA 72-4 levels were found primarily in gastrointestinal, lung, ovarian and endometrial adenocarcinomas; CEA elevated values were present in gastrointestinal, lung and breast adenocarcinomas. Further analysis of the effusions failed to show any influence due to the site of the effusion on the detection of TAG-72, while CEA seems to show a much higher percentage of positivity in the pleural effusions due to adenocarcinomas when compared to ascites. The distribution of the various types of adenocarcinomas in both groups was similar with the exception of lung and breast adenocarcinomas which were most frequently associated with pleural effusions and were highly positive for CEA. Thus, the higher rate of positivity of pleural effusions found in CEA assays may be due to the malignancy’s origin and not to the site of the effusion.

1.cc.l

.

I ADENOCA

N.E.M.

BENIGN

Fig. 4. TAG-72 levels in effusions from patients with adenocarcinomas (n = 51), non epithelial malignancies (NEM; n = 27) and benign diseases (n = 68). The cutoff value of 8.5 U/mL is shown as a dashed line. Taken from Ferroni et al. (1991).

DAVIDCOLCHERet al.

400

A further comparison of the RIA results with the site of origin of the adenocarcinoma shows that with the use of a combination of the CA 72-4 and the CEA assay, elevated TAA levels found in 6 of 6 patients with gastrointestinal adenocarcinomas (Gero et al., 1989). This agreed with the clinical findings. Furthermore, the use of the CA 72-4 assay alone detected malignancy in 4 more patients with ovarian adenocarcinomas and 2 with endometrial adenocarcinemas, which were negative for the presence of CEA. On the other hand, CEA determination was successful in detecting malignancy in 4 patients with breast adenccarcinoma and 2 with adenocarcinoma of the lung, while those samples were negative for the presence of TAG-72. Using a combination of the CA 72-4 and CEA RIAs increased the sensitivity for malignant effusions to 73.5% (Fig. 3, bottom panel). No additional improvement in the overall sensitivity was observed using the CA 72-4 assay in combination with assays for the other markers (i.e. CA19-9, CA 15-3 and CA 125), with the exception of 1 effusion. We conclude that the CA 72-4 RIA, in combination with CEA assays may be useful in distinguishing adenocarcinomatous vs benign effusions (Ferroni et al., 1991) The high degree of selective binding of MAb B72.3 to a wide range of carcinoma tissues, the pre-clinical animal studies (both diagnostic and therapeutic), and recent clinical trials have all demonstrated the potential utility of MAb B72.3 for assaying biological fluids as well as for in oivo use for radioimmunodiagnosis and radioimmunotherapy. Thus, MAb B72.3, as well as the second generation MAb against TAG-72 and the recombinant/chimeric forms of these MAb, stand as viable candidates in various aspects of the management of cancer.

References Bertram J. H., Gill P. S., Levine A. M., Boquiren D., Hoffman F. M., Meyer P. and Mitchell M. S. (1986) Monoclonal antibody TlOl in T cell malignancies: a clinical, pharmacokinetic, and immunologic correlation. Blood 68, 752-161.

Carrasquillo J. A., Sugarbaker P., Colcher D., Reynolds J. C., Esteban J., Bryant G., Keenan A. M., Perentesis P., Yokoyama K., Simpson D. E., Ferroni P., Farkas R., Schlom J. and Larson S. M. (1988) Radioimmunoscintigraphy of colon cancer with iodine-l 3 l-labeled B72.3monoclonal antibody. J. Nucl. Med. 29,1022-1030. Colcher D., Horan Hand P., Nuti M. and Schlom J. (1981) A spe&um of mono&ma1 antibodies reactive‘ with human mammary tumor cells. Proc. Natl. Acad. Sci. U.S.A. 78, 3199-3203.

Colcher D., Keenan A. M., Larson S. M. and Schlom J. (1984) Prolonged binding of a radiolabeled monoclonal antibody (B72.3) used for the in situ radioimmunodetection of human colon carcinoma xenografts. Cancer Res. 44, 5744-5751. Colcher D., Esteban J. M., Carrasquillo J. A., Sugarbaker P., Reynolds J. C., Bryant G., Larson S. M. and Schlom J. (1987a) Quantitative analyses of selective radiolabeled monoclonal antibody localization in metastatic lesions of colorectal cancer patients. Cancer Res. 47, 1185-1189.

Colcher D., Esteban J., Carrasquillo J. A., Sugarbaker P., Reynolds J. C., Bryant G., Larson S. M. and Schlom J. (1987b) Complementation of intracavitary and intravenous administration of a monoclonal antibody (B72.3) in patients with carcinoma. Cuncer Res. 47, 4218-4224. Colcher D., Milenic D. E., Ferroni P., Carrasquillo J. A., Reynolds J. C., Roselli M., Larson S. M. and Schlom J. (1990) In vivo fate of monoclonal antibody B72.3 in patients with colon cancer. J. Nucl. Med. 31, 1133-l 142. Colomer R., Sole L. A., Navarro M., Encabo G., Ruibal A. and Salvador L. I. (1986) CA 15-3; early results of a new breast cancer marker. Anticancer Res. 6, 683-684. Dekker A. and Bupp P. A. (1978) Cytology of serious effusions, an investigation into the usefulness of cell blocks vs smear. Am. J. Clin. Pathol. 70, 8555860. Del Villano B. C., Brennan S., Brock P., Bucker C., Liu U., McClure M., Rake B., Space S., Westrick B., Schoemaker H. and Zurawski V. Jr (1983) Radioimmunometric assay for a monoclonal antibody-defined tumor marker, CA 19-9. Clin. Chem. 29, 549-552. Dines D. E., Pierre R. V. and Franzen S. J. (1975) The value of cells in the pleural fluid in the differential diagnosis. Mayo Clin. Proc. SO, 571-572.

Esteban J. M., Schlom J., Mornex F. and Colcher D. (1987) Radioimmunotherapy of athymic mice bearing human colon carcinomas with monoclonal antibody B72.3: histologic and autoradiographic study of effects on tumors and normal organs. Eur. J. Cancer Clin. Oncol. 23, 643-655.

Ferroni P., Colcher D., Szpak C., Greiner J. W., Simpson J. F., Johnston W. W. and Schlom J. (1990) CA 72-4 radioimmunoassay in the diagnosis of malignant effusions: comparison of various tumor markers. Int. J. Cancer. 46, 445451. Gero E. J., Colcher D., Ferroni P., Melsheimer R., Giani S., Schlom J. and Kaplan P. (1989) The CA 72-4 radioimmunoassay for the detection of the TAG-72 carcinoma associated antigen in serum of patients. J. Clin. Lab. Analysis 3, 360-369.

Goldenberg D. M. (Editor) (1990a) Cancer Imaging with Radiolabeled Antibodies. Kluwer Academic, Boston, Mass. Goldenberg D. M. (Editor) (1990b) Second Conference on Radioimmunodetection and Radioimmunotherapy of Cancer. Cancer Res. 50, 773s-1059s. Hayes D. F., Sekine K., Ohno T., Abe M., Keefe K. and Kufe D. W. (1985) Use of a murine monoclonal antibody for detection of circulating plasma DF3 antigen levels in breast cancer oatients. J. Clin. Invest. 75. 1671-1678. Hnatowich D. J.: Griffin T. W., Kosciuczyk C., Rusckowski M., Childs R. L., Mattis J. A., Shealy D. and Doherty P. W. (1985) Pharmacokinetics of an indium-1 11 labeled monoclonal antibody in cancer patients. J. Nucl. Med. 26, 849-858.

Hnatowich D. J., Gionet M., Rusckowski M., Siebecker D. A., Roche J., Shealy D., Mattis J. A., Wilson J., McGann J., Hunter R. E., Griffin T. and Doherty P. W. (1987) Pharmacokinetics of “‘In-labeled OC-125 antibody in cancer patients compared with the 19-9 antibody. Cancer Res. 47, 6111-6117. Johnson V. G., Schlom J., Paterson A. J., Bennett J., Magnani J. L. and Colcher D. (1986) Analysis of a human tumor-associated glycoprotein (TAG-72) identified by monoclonal antibody B72.3. Cancer Res. 46, 850-857. Khazaeli M. B., Saleh M. N., Wheeler R. H., Huster W. J., Holden H.. Carrano R. and LoBunlio A. F. (1988) Phase I trial of ‘multiple large doses cf murine ‘monbclonal antibody C017-1A. II. Pharmacokinetics and immune response. J. Nod Cancer Inst. SO, 937-942. Klug T. L., Sattler M. A., Colcher D. and Schlom J. (1986) Monoclonal antibody immunoradiometric assay for an antigenic determinant (CA 72) on a novel pancarcinoma antigen (TAG-72). Int. J. Cancer 38, 661-669.

Clinical applications of MAbs against TAG-72 Klug T. L., Bast R. C. Jr, Niloff J. M., Knapp R. C. and Zurawski V. R. Jr (1984) Monoclonal antibody immunoradiometric assay for an antigenic determinant (CA 125) associated with human epithelial ovarian carcinomas. Cancer Res. 44, 1048-1053. Metzgar R. S., Rodriquez N., Finn 0. J., Lan M. S., Daasch V. N., Femsten P. D., Meyers W. C., Sindelar W. F., Sandler R. S. and Seigler H. F. (1984) Detection of a pancreatic cancer-associated antigen (DU-PAN-2 antigen) in serum and ascites of patients with adenocarcinemas. Proc. Natl. Acad. Sci. U.S.A. 81, 5242-5246. Metzger J., Permanetter W., Gerbes A. L., Wilmanns W. and Lamerz R. (1988) Tumor associated antigens in diagnosis of serious effusions. J. Clin. Pathol. 41, 633-643. Mottolese M., Venturo I., Perrone Donnorso R., Gallo Curcio C., Rinaldi M. and Natali P. G.’ (1988) Use of selected combination of monoclonal antibodies to tumor associated antigens in the diagnosis of neoplastic effusions of unknown origin. Eur. J. Cancer Clin. Oncol. 24, 1277-1284. Nystrom J. S., Dyce B., Wada J., Bateman J. R. and Haverback B. (19771 Carcinoembrvonic antieen titers on effusion fluid. k diagnostic tool? krch. Intern. Med. 137, 875-879. Ohuchi N., Thor A., Nose M., Fujita J., Kyogoku M. and Schlom J. (1986) Tumor-associated glycoprotein (TAG-72) detected in adenocarcinomas benign lesions of ihe stomach. Znt. J. Cancer 38, 643-650. Ohuchi N., Gero E., Mori S., Akimoto M., Matoba N., Nishihira T., Hirayama K., Colcher D. and Schlom J. (1990) Clinical evaluation of CA72-4 immunoradiometric assay for serum TAG-72 in patients with carcinoma. J. Tumor Mark. Oncol. 5, l-10. Ohuchi N., Mori S., Gero E., Colcher D., Mochizuki F., Nishihira T., Akimoto M., Hirayama K., Matoba N., Kablan P. M. and Schlom J. (1991) Serum levels of t&or-associated glycoprotein (TAG-?2) in patients with carcinoma detected by CA72-4 radioimmunometric assay. Submitted for publication. Paterson A. J.. Schlom J.. Sears H. F.. Bennett J. and Colcher D. (i986) A radibimmunoassa; for the detection of a tumor-associated glycoprotein (TAG-72) using monoclonal antibody B72.3. Int. J. Cancer 37, 659-666. Pinto M. M., Bernstein L. H., Brogan D. A. and Criscuolo E. M. (1987) Carcinoembryonic antigen in effusions. A diagnostic adjunct to cytology. Acta Cytol. 31, 113-118. Rittgers R. A., Loewenstein M. S., Feinerman A. E., Kupchik H. Z., Marcel B. R., Koff R. S. and Zamcheck M. (1978) Carcinoembryonic antigen levels in benign and malignant pleural effusions. Ann. Intern. Med. 88, 63 l-634. Rosenblum M. G., Murray J. L., Haynie T. P., Glenn H. J., Jahns M. F., Benjamin R. S., Frincke J. M., Carlo D. J.

401

and Hersh E. M. (1985) Pharmacokinetics of “‘In-labeled anti-p97 monoclonal antibody in patients with metastatic malignant melanoma. Cancer Res. 45, 2382-2386. SchroffR. W., Morgan A. C. Jr, Woodhouse C. S. et al. (1987) Monoclonal antibody therapy in malignant melanoma: factors effecting in oiuo localization. J. Rio/. Res. Mod. 6, 457-472. Stramignoni D., Bowen R., Atkinson B. F. and Schlom J. (1983) Differential reactivity of monoclonal antibodies with human colon adenocarcinomas and adenomas. Znr. J. Cancer 31, 543-552. Szpak C. A., Johnston W. W., Lottich S. C., Kufe D., Thor A. and Schlom J. (1984) Patterns of reactivity of four novel monoclonal antibodies (B72.3, DF3, Bl. 1 and B6.2) with cells in human malignant and benign effusions. Acta Cytol. 28, 356-367. Takasaki H., Temper0 M. A., Uchida E., Buchler M., Ness M. J., Burnett D. A., Metzgar R. S., Colcher D., Schlom J. and Pour P. M. (1988) Comparative studies on the expression of tumor-associated glycoprotein (TAG-72), CA 19-9 and DU-PAN-2 in normal, benign and malignant pancreatic tissue. Int. J. Cancer 42, 681-686. Thomson D. M. P., Krupey J., Freedman S. 0. and Gold P. (1969) The radioimmunoassay of circulating carcinoembryonic antigen of the human digestive system. Proc. Natl. Acad. Sci. U.S.A. 64, 161-167. Thor A., Gorstein F., Ohuchi N., Szpak C. A., Johnston W. W. and Schlom J. (1986aj Tumor-associated alvcoprotein (TAG-72) in &aria; carcinomas definld- by monoclonal antibody B72.3 (1986) J. Natl. Cancer Inst. 76, 995-1006. Thor A., Ohuchi N., Szpak C., Johnston W. W. and Schlom J. (1986b) Distribution of oncofetal antigen TAG-72 defined by monoclonal antibody B72.3. Cancer Res. 46, 3118-312;1. Thor A., Viglione M. J., Muraro R., Ohuchi N., Schlom J. and Gorstein F. (1987) Monoclonal antibody B72.3 reactivity with human endometrium: A study of normal and malignant tissues. Inr. J. Gynecol. Path& 6, 235-247. Whitaker D., Shilkin K. B., Stuckey M. and Nieuwhof W. N. (1986) Pleural fluid CEA levels in the diagnosis of malignant mesothelioma. Pathology 18, 328-329. Whiteside T. L. and Dekker A. (1979) Diagnostic significance of carcinoembryonic antigen levels in serious effusions. Correlation with cytology. Acta Cytol. 23, 443-448. Wolf B. C., D’Emilia J. C., Salem R. R., DeCoste D., Sears H. F., Gottlieb L. S. and Steele G. D. Jr (1989) Detection of tumor-associated glycoprotein antigen (TAG-72) in pre-malignant lesions of the colon. J. Natl. Cancer Inst. 81, 1913-1917. Zamcheck N. (1974) Carcinoembryonic antigen: quantitative variations in circulating levels in benign and malignant digestive tract disease. Ado. Intern. Med. 19, 413-433.