Carcinoembryonic Antigen in Histopathology: Immunoperoxidase Staining of Conventional Tissue Sections1. 2 David M. Goldenberg,3 Robert M. Sharkey,3 and F. James Primus 4 ABSTRACT -A triple-bridge, indirect peroxidase-antiperoxidase method for demonstrating carcinoembryonic antigen (CEA) in frozen, ethanol-fixed or formalin-fixed, paraffin-embedded specimens was evaluated. Examination of 359 tissue specimens-234 malignant tumors, 37 benign neoplasms, 41 nonneoplastic diseased tissues, and 47 normal specimens-showed that CEA could usually be demonstrated in a group of cancers. We could detect CEA in carcinomas of the stomach, colon, rectum, pancreas, lung, and cervix. However, malignant tumors of the breast, prostate, kidney, larynx, brain, Iymphoreticular system, soft tissues, and skin proved negative for CEA by the immunoperoxidase test. CEA could be detected in ethanol- or formalin-fixed sections. The only nonmalignant specimens showing CEA staining were a few benign tumors, the mucosae of some cases of colitis, and the resection margins of 2 cases of colon cancer; however, these were commonly very weak reactions. Measurement of tumor CEA content by radioimmunoassay revealed two causes for this relative specificity of the immunoperoxidase test for CEA: 1) a quantitative difference existed in tissue CEA among the various specimens, and 2) the threshold for CEA staining in malignant specimens was usually above that in nonmalignant specimens. An analysis of the formalin-paraffin-treated sections showed that immunoperoxidase-tested CEA positivity reflected CEA levels in tissue of at least 3.0-5.0 "gIg; this permitted retrospective estimates of minimal tissue CEA concentrations in older histopathologic specimens by the immunoperoxidase reaction method. Formalin-paraffin-treated sections as old as 10 years still had demonstrable CEA. Although tumor CEA concentration correlated well with immunoperoxidase staining for CEA, plasma CEA titer did not necessarily reflect tumor CEA content. CEA positivity in primary and secondary tumors was strongly correlated; it was less strongly correlated with level of tumor dlfferentiation.-J Natl Cancer Inst 57: 11-22, 1976.
CEA (1, 2) is one of the most widely studied tumorassociated substances because increased quantities of this antigen are found in the blood of patients with diverse malignant tumors (3). Although an increased plasma CEA titer is by itself not indicative of cancer (3, 4), the CEA plasma test currently appears valuable for following cancer patients whose pretreatment levels were elevated (3-5). However, the correlation between tumor and circulating CEA concentrations has not been established; blood CEA values can be normal in patients with progressively growing tumors, whereas in other cases the CEA plasma level is high when only very small tumors are identifiable. These discrepancies may reflect differential rates of tumor CEA biosynthesis and release, differences in host metabolism and degradation of circulating CEA, or both. It would, therefore, be useful to determine CEA content in tumor specimens for comparison with circulating CEA concentration. Although this could be achieved by extracting fresh tumor and measuring CEA content by radioimmunoassay, such an approach is not feasible in a conventional histopathology setting. A significant advance in the localization of antigen or VoL.
57, NO. I, JULY 1976
antibody in tissue was the development of immunocytochemical techniques based upon enzyme-coupled antigen-antibody reactions, such as with horseradish peroxidase (6, 7). These methods have proved particularly valuable for localizing synthetic products of cells by electron microscopy and for demonstrating antigens in cells more conveniently without the use of a fluorescence microscope. We have investigated use of the immunoperoxidase method to detect CEA in tissue sections processed for histopathology. We found that CEA could be detected in both ethanol- and formalin-fixed specimens, which permitted immunologic evaluation of histology sections. Minimal CEA present in such tissues could also be estimated. This allowed a comparison between extracellular levels of CEA and CEA levels in tissue. Our findings suggest that the immunocytochemical identification of tumor-associated antigens such as CEA in tissue sections might have diagnostic implications which would make the evaluation of immunologic markers in malignant tumors of interest to the histopathologist. This identification method might also help elucidate the function of CEA in oncogenesis. MATERIALS AND METHODS
Histologic specimens.-Surgical, biopsy, and autopsy specimens from a variety of normal, diseased, or neoplastic tissues were either fixed in 10% buffered formalin or "snap-frozen" without prefixation in tissueembedding medium (Tissue-TEK O.C.T. Compound; Ames Co., Inc., Elkhart, Ind.). Then the frozen sections were fixed in -700 C absolute ethanol and kept at 40 C until used. The tissues directly fixed in formalin were embedded in paraffin by the usual procedures. Sections 5-7 J.t thick were cut from each specimen and placed on glass slides. The paraffin sections were rehydrated in xylene and decreasing concentrations of ethanol. Both the ethanol and formalin-paraffintreated tissues were finally hydrated in 0.01 M PBS (pH 7.2) for two 5-minute changes. Representative sections were also obtained from each specimen for the usual hematoxylin and eosin staining. Since formalin-paraffin treatment did not destroy the CEA demonstrable by immunocytochemistry, paraffin blocks of various tissues ABBREVIATIONS USED: CEA=carcinoembryonic antigen; PBS= phosphate-buffered saline; CCA-III =colon carcinoma antigen III; H_BFP=hematoxylin-basic fuchsin-picric acid. Received September 22, 1975; accepted January 6, 1976. Supported in part by a grant from the University of Kentucky Tobacco and Health Research Institute, and by Public Health Service grant CAI5799 from the National Large Bowel Cancer Project of the National Cancer Institute. 3 Division of Experimental Pathology, Department of Pathology, University of Kentucky College of Medicine, Lexington, Ky. 40506. 4 Department of Immunology, Hoffmann-La Roche, Nutley, N.J. 07110. 1
2
11
J
NATL CANCER INST
12
GOLDENBERG, SHARKEY, AND PRIMUS
dating back as far as 10 years were also evaluated for CEA content by the immunoperoxidase reaction method. Antisera. -The preparation of goat antiserum against CEA was described earlier (8). The anti-CEA antiserum cross-reacted with CEA and CCA-I1I (9). Anti-CCA-I1I immunoreactivity was removed by adsorption onto CCA-III coupled to CNBr-reactivated Sepharose, as described in (10). The same procedure was employed to produce a specificity control in which the anti-CEA antiserum, after removal of cross-reacting CCA-III antibody, was adsorbed onto a CEA-containing immunoadsorbent (although in this instance the CEA was not from the same patient source as the immunogen). Both procedures completely removed CCA-I1I cross-reacting antibody and, when desired for the control reaction, antiCEA antibody from the anti-CEA antiserum, as measured by radioimmunoassay (9, 11) and immunodiffusion. Furthermore, the anti-CEA antiserum used for the immunoperoxidase staining gave equivalent plasma CEA titers when it was substituted for the antiserum used in the Hoffmann-La Roche CEA kit (Hoffmann-La Roche Inc., Nutley, N.J.) (Hansen HJ: Personal communication, 1975). Although the anti-CEA antiserum was free of detectable hemagglutinating antibodies, it was initially absorbed with a mixture of major blood group erythrocytes. Antiserum against goat IgG was prepared by injecting rabbits sc biweekly over a 2-month period with 2 mg IgG (Miles Laboratories, Kankakee, Ill.) emulsified in an equal volume of complete Freund's adjuvant (Difco, Detroit, Mich.). Goat antiserum against horseradish peroxidase (type VI; Sigma Chemical Co., St. Louis, Mo.) was prepared similarly. Both antisera were used, without further treatment, in the immunoperoxidase triple-bridge reaction. Immunoperoxidase reaction.-Our immunoperoxidase technique was described in detail in (8). We added 1-2 drops of goat anti-CEA antiserum (diluted 1:300) to the tissue sections. The same serum, neutralized with CEA, was used at a dilution of 1:300 in an adjacent identical section as a control. The sections were incubated in a moist chamber at 37° C for 10 minutes and washed in two 5-minute changes of 0.01 M PBS. Then the sections were treated with rabbit antiserum against goat IgG (1 :50 dilution), followed by goat antiserum against peroxidase (1: 100 dilution) under the same conditions as for the primary antiserum. We bound the antiserum to horseradish peroxidase (type VI, Sigma) by adding the peroxidase at a concentration of 100 ILg/ml for 20 minutes at 37° C. Then we removed it by washing. Thus our technique was a triple-bridge, indirect, peroxidase-antiperoxidase procedure. The reaction was then developed with Karnovsky's solution (12) [75 mg 3,3' -diaminobenzidine/lOO ml of 0.05 M Tris buffer (pH 7.6) and 0.05% H 2 0 2 ], which was permitted to incubate at room temperature for 20 minutes. The sections were finally washed in PBS, dehydrated in graded ethanol solutions and xylene, and mounted with Permount. As required, the immunoperoxidase-stained sections were subsequently counterstained with HBFP. J
NATL CANCER INST
Tissue and plasma CEA content.-Fresh tissue specimens were homogenized as a 10% (wt/vol) suspension in cold H 2 0 and then spun at 12,000 Xg for 20 minutes at 4° C in a Sorvall RC2-B centrifuge. The supernatants were extracted with equal volumes of 1.2 M perchloric acid and dialyzed against several changes of deionized H 2 0 for 24 hours. For the last 3 hours, the dialyzate was equilibrated with 0.1 M ammonium acetate; the final pH was adjusted to 6.5. The dialyzates were then analyzed for CEA by the radioimmunoassay procedure of Hansen et al. (11). Similarly, plasma CEA was determined by radioimmunoassay with Hoffmann-La Roche reagents. Each CEA value represented the average of two samples analyzed simultaneously. Interpretation of staining reaction.-The immunoperoxidase staining reaction for CEA was interpreted only in reference to the adjacent control section, in which the primary anti-CEA antiserum was completely neutralized by exhaustive adsorption with CEA. Staining was considered positive for CEA only when a distinction could be made between the test and control sections; the controls were devoid of brownish pigment in the tissue except where blood cells showed reactivity. Although malignant tumors did show differences in the intensity of staining, no attempt was made to grade them in reference to one another, since each specimen had its own control section. RESULTS Malignant Tumors
We assayed 234 malignant tumors for CEA by the immunoperoxidase reaction method (table 1). Fresh tissue specimens were prepared as described above. Older, preserved specimens were also evaluated, but unlike the fresh tissues, they had not been processed for CEA extraction. A portion of each fresh tumor tissue was also submitted for radioimmunoassay of CEA after perchloric acid extraction (see table 1: ethanol-fixed sections, CEA-positive specimens). Table 1 (columns 1,2) shows that a specific immunoperoxidase-induced stain for CEA was observed in a wide variety of malignant tumors, including carcinomas of the stomach, colon, rectum, pancreas, lung, and uterine cervix; only single cases of ovarian and urinary bladder cancers showed CEA staining. Moreover, the immunoperoxidase test induced CEA staining in a number of formalin-paraffin-processed tissues that were originally submitted for histopathology 1-10 years earlier. Only 4 cases revealed a discrepancy between the frozen, ethanol-fixed and the formalin-paraffin-treated sections: In 1 colon, 1 bronchogenic, and 2 cervical carcinomas the former sections were positive and the latter were negative for CEA. Since the results indicate that morphology frequently determined CEA positivity in tumors, each major tumor type is discussed separately. Digestive tract cancers.-None of 4 squamous cell carcinomas of the esophagus was positive for CEA. Of 12 adenocarcinomas of the stomach, 1 was positive after ethanol or formalin fixation; another 4, which had been processed by formalin-paraffin 1-4 years earlier, were VOL. 57, NO.1, JULY 1976
13
IMMUNOPEROXIDASE STAINING OF TISSUE CEA
TABLE
1.--lncidence ofCEA staining by the indirect immunoperoxidase reaction test in ethanol- and formalin-fixed malignant tumor specimens; tissue CEA levels as determined by radioimmunoassay Ethanol-fixed sections'
Tissue histopathology
Total No. testedt.l
Skin: Basal cell carcinoma Squamous cell carcinoma Brain: Glioblastoma Neuroblastoma Parotid gland: Squamous cell carcinoma Larynx: Squamous cell carcinoma Esophagus: Squamous cell carcinoma Stomach: Adenocarcinoma Colon: Adenocarcinoma
Rectum: Adenocarcinoma Anus: Squamous cell carcinoma Liver: Hepatocellular carcinoma Pancreas: Adenocarcinoma Bronchus: Squamous cell carcinoma
Lung: Adenocarcinoma Breast: Scirrhous adenocarcinoma Adenocarcinoma Medullary carcinoma Urinary bladder: Transi tional cell carcinoma Squamous cell carcinoma In situ (papillary) carcinoma Kidney: Renal cell carcinoma Cervix: Squamous cell carcinoma Carcinoma in situ
Total No. positive b
Total
Number positive
CEA-positive specimens iJ.g/g
Number
Formalin-fixed sections Positive
CEA-negative specimens d
negative
Total'
iJ.g/g
Number positive
fresh specimens
Positive
old sections
Number negative
o
o
o
3
o
o
o
3
5
o o
o
o
o
5
o
o
o
5
5
o
3
o
3
5
o
o
o
5
8
o
1
o
8
o
o
o
8
1
o
o
o
o
o
o
o
1
4
o
3
o
3
4
o
o
o
4
4
o
o
o
o
4
o
o
o
4
12
5
3
27
18
12
10
5
1
o
1
o
9
o
o
o
7
2
o
o
17
6
10
3
11
1
1
o
9
o o o
o o
o o o
o
10
1
o
o
3
o
1
o
o o
o o
o
3
o
3
o
3
0.002, 0.005, 0.007
19
8
9
5
4
11
o
o
o
0.2, 0.2, 0.3, 1.4
3
8 1
VOL. 57, NO.1, JULY 1976
2
9
0.0004, 0.002, 0.07 0.008
0.01, NA
NA,
18.6
2
0.002,0.2
12
5
1
4
7
5.4, 6.5, 10.1, 11.2, 23.1, 24.0, 26.2, 39.3, 177.6
3
0.1, 0.2, NA
27
17
8
9
10
3.0
o
10
5
4
5
1
o
o
o
1
o
9
o
o
o
9
o
7
2
o
2
5
0.008, 0.03, 0.06, 0.3, 1.5, 0.4, NA
17
5
2
3
12
0.1
11
0.03
5.5, 5.7, 9.2
7
10
9
o
o
o
9
2 1
o o
o o
o o
2
o
10
1
o
3
o o
o o o
o o
1
3
o
o
o
3
19
6
3
3
13
11
o
o
o
11
o 2
0.7, 0.9, 4.8, 7.7, 239.1
o
o
0.2,0.2
J
1
9 3
NATL CANCER INST
GOLDENBERG, SHARKEY, AND PRIMUS
14
TABLE l.-Continued Ethanol-fixed sections" Tissue histopathology
Total No. tested a
Ovary: Mucinous cystadenocarcinoma Serous cystadenocarcinoma Squamous cell carcinoma Adenocarcinoma Endometrium: Adenocarcinoma Prostate: Adenocarcinoma
a
c d
CEA (1, 2) is one of the most widely studied tumorassociated substances because increased quantities of this antigen are found in the blood of patients with diverse malignant tumors (3). Although an increased plasma CEA titer is by itself not indicative of cancer (3, 4), the CEA plasma test currently appears valuable for following cancer patients whose pretreatment levels were elevated (3-5). However, the correlation between tumor and circulating CEA concentrations has not been established; blood CEA values can be normal in patients with progressively growing tumors, whereas in other cases the CEA plasma level is high when only very small tumors are identifiable. These discrepancies may reflect differential rates of tumor CEA biosynthesis and release, differences in host metabolism and degradation of circulating CEA, or both. It would, therefore, be useful to determine CEA content in tumor specimens for comparison with circulating CEA concentration. Although this could be achieved by extracting fresh tumor and measuring CEA content by radioimmunoassay, such an approach is not feasible in a conventional histopathology setting. A significant advance in the localization of antigen or VoL.
57, NO. I, JULY 1976
antibody in tissue was the development of immunocytochemical techniques based upon enzyme-coupled antigen-antibody reactions, such as with horseradish peroxidase (6, 7). These methods have proved particularly valuable for localizing synthetic products of cells by electron microscopy and for demonstrating antigens in cells more conveniently without the use of a fluorescence microscope. We have investigated use of the immunoperoxidase method to detect CEA in tissue sections processed for histopathology. We found that CEA could be detected in both ethanol- and formalin-fixed specimens, which permitted immunologic evaluation of histology sections. Minimal CEA present in such tissues could also be estimated. This allowed a comparison between extracellular levels of CEA and CEA levels in tissue. Our findings suggest that the immunocytochemical identification of tumor-associated antigens such as CEA in tissue sections might have diagnostic implications which would make the evaluation of immunologic markers in malignant tumors of interest to the histopathologist. This identification method might also help elucidate the function of CEA in oncogenesis. MATERIALS AND METHODS
Histologic specimens.-Surgical, biopsy, and autopsy specimens from a variety of normal, diseased, or neoplastic tissues were either fixed in 10% buffered formalin or "snap-frozen" without prefixation in tissueembedding medium (Tissue-TEK O.C.T. Compound; Ames Co., Inc., Elkhart, Ind.). Then the frozen sections were fixed in -700 C absolute ethanol and kept at 40 C until used. The tissues directly fixed in formalin were embedded in paraffin by the usual procedures. Sections 5-7 J.t thick were cut from each specimen and placed on glass slides. The paraffin sections were rehydrated in xylene and decreasing concentrations of ethanol. Both the ethanol and formalin-paraffintreated tissues were finally hydrated in 0.01 M PBS (pH 7.2) for two 5-minute changes. Representative sections were also obtained from each specimen for the usual hematoxylin and eosin staining. Since formalin-paraffin treatment did not destroy the CEA demonstrable by immunocytochemistry, paraffin blocks of various tissues ABBREVIATIONS USED: CEA=carcinoembryonic antigen; PBS= phosphate-buffered saline; CCA-III =colon carcinoma antigen III; H_BFP=hematoxylin-basic fuchsin-picric acid. Received September 22, 1975; accepted January 6, 1976. Supported in part by a grant from the University of Kentucky Tobacco and Health Research Institute, and by Public Health Service grant CAI5799 from the National Large Bowel Cancer Project of the National Cancer Institute. 3 Division of Experimental Pathology, Department of Pathology, University of Kentucky College of Medicine, Lexington, Ky. 40506. 4 Department of Immunology, Hoffmann-La Roche, Nutley, N.J. 07110. 1
2
11
J
NATL CANCER INST
12
GOLDENBERG, SHARKEY, AND PRIMUS
dating back as far as 10 years were also evaluated for CEA content by the immunoperoxidase reaction method. Antisera. -The preparation of goat antiserum against CEA was described earlier (8). The anti-CEA antiserum cross-reacted with CEA and CCA-I1I (9). Anti-CCA-I1I immunoreactivity was removed by adsorption onto CCA-III coupled to CNBr-reactivated Sepharose, as described in (10). The same procedure was employed to produce a specificity control in which the anti-CEA antiserum, after removal of cross-reacting CCA-III antibody, was adsorbed onto a CEA-containing immunoadsorbent (although in this instance the CEA was not from the same patient source as the immunogen). Both procedures completely removed CCA-I1I cross-reacting antibody and, when desired for the control reaction, antiCEA antibody from the anti-CEA antiserum, as measured by radioimmunoassay (9, 11) and immunodiffusion. Furthermore, the anti-CEA antiserum used for the immunoperoxidase staining gave equivalent plasma CEA titers when it was substituted for the antiserum used in the Hoffmann-La Roche CEA kit (Hoffmann-La Roche Inc., Nutley, N.J.) (Hansen HJ: Personal communication, 1975). Although the anti-CEA antiserum was free of detectable hemagglutinating antibodies, it was initially absorbed with a mixture of major blood group erythrocytes. Antiserum against goat IgG was prepared by injecting rabbits sc biweekly over a 2-month period with 2 mg IgG (Miles Laboratories, Kankakee, Ill.) emulsified in an equal volume of complete Freund's adjuvant (Difco, Detroit, Mich.). Goat antiserum against horseradish peroxidase (type VI; Sigma Chemical Co., St. Louis, Mo.) was prepared similarly. Both antisera were used, without further treatment, in the immunoperoxidase triple-bridge reaction. Immunoperoxidase reaction.-Our immunoperoxidase technique was described in detail in (8). We added 1-2 drops of goat anti-CEA antiserum (diluted 1:300) to the tissue sections. The same serum, neutralized with CEA, was used at a dilution of 1:300 in an adjacent identical section as a control. The sections were incubated in a moist chamber at 37° C for 10 minutes and washed in two 5-minute changes of 0.01 M PBS. Then the sections were treated with rabbit antiserum against goat IgG (1 :50 dilution), followed by goat antiserum against peroxidase (1: 100 dilution) under the same conditions as for the primary antiserum. We bound the antiserum to horseradish peroxidase (type VI, Sigma) by adding the peroxidase at a concentration of 100 ILg/ml for 20 minutes at 37° C. Then we removed it by washing. Thus our technique was a triple-bridge, indirect, peroxidase-antiperoxidase procedure. The reaction was then developed with Karnovsky's solution (12) [75 mg 3,3' -diaminobenzidine/lOO ml of 0.05 M Tris buffer (pH 7.6) and 0.05% H 2 0 2 ], which was permitted to incubate at room temperature for 20 minutes. The sections were finally washed in PBS, dehydrated in graded ethanol solutions and xylene, and mounted with Permount. As required, the immunoperoxidase-stained sections were subsequently counterstained with HBFP. J
NATL CANCER INST
Tissue and plasma CEA content.-Fresh tissue specimens were homogenized as a 10% (wt/vol) suspension in cold H 2 0 and then spun at 12,000 Xg for 20 minutes at 4° C in a Sorvall RC2-B centrifuge. The supernatants were extracted with equal volumes of 1.2 M perchloric acid and dialyzed against several changes of deionized H 2 0 for 24 hours. For the last 3 hours, the dialyzate was equilibrated with 0.1 M ammonium acetate; the final pH was adjusted to 6.5. The dialyzates were then analyzed for CEA by the radioimmunoassay procedure of Hansen et al. (11). Similarly, plasma CEA was determined by radioimmunoassay with Hoffmann-La Roche reagents. Each CEA value represented the average of two samples analyzed simultaneously. Interpretation of staining reaction.-The immunoperoxidase staining reaction for CEA was interpreted only in reference to the adjacent control section, in which the primary anti-CEA antiserum was completely neutralized by exhaustive adsorption with CEA. Staining was considered positive for CEA only when a distinction could be made between the test and control sections; the controls were devoid of brownish pigment in the tissue except where blood cells showed reactivity. Although malignant tumors did show differences in the intensity of staining, no attempt was made to grade them in reference to one another, since each specimen had its own control section. RESULTS Malignant Tumors
We assayed 234 malignant tumors for CEA by the immunoperoxidase reaction method (table 1). Fresh tissue specimens were prepared as described above. Older, preserved specimens were also evaluated, but unlike the fresh tissues, they had not been processed for CEA extraction. A portion of each fresh tumor tissue was also submitted for radioimmunoassay of CEA after perchloric acid extraction (see table 1: ethanol-fixed sections, CEA-positive specimens). Table 1 (columns 1,2) shows that a specific immunoperoxidase-induced stain for CEA was observed in a wide variety of malignant tumors, including carcinomas of the stomach, colon, rectum, pancreas, lung, and uterine cervix; only single cases of ovarian and urinary bladder cancers showed CEA staining. Moreover, the immunoperoxidase test induced CEA staining in a number of formalin-paraffin-processed tissues that were originally submitted for histopathology 1-10 years earlier. Only 4 cases revealed a discrepancy between the frozen, ethanol-fixed and the formalin-paraffin-treated sections: In 1 colon, 1 bronchogenic, and 2 cervical carcinomas the former sections were positive and the latter were negative for CEA. Since the results indicate that morphology frequently determined CEA positivity in tumors, each major tumor type is discussed separately. Digestive tract cancers.-None of 4 squamous cell carcinomas of the esophagus was positive for CEA. Of 12 adenocarcinomas of the stomach, 1 was positive after ethanol or formalin fixation; another 4, which had been processed by formalin-paraffin 1-4 years earlier, were VOL. 57, NO.1, JULY 1976
13
IMMUNOPEROXIDASE STAINING OF TISSUE CEA
TABLE
1.--lncidence ofCEA staining by the indirect immunoperoxidase reaction test in ethanol- and formalin-fixed malignant tumor specimens; tissue CEA levels as determined by radioimmunoassay Ethanol-fixed sections'
Tissue histopathology
Total No. testedt.l
Skin: Basal cell carcinoma Squamous cell carcinoma Brain: Glioblastoma Neuroblastoma Parotid gland: Squamous cell carcinoma Larynx: Squamous cell carcinoma Esophagus: Squamous cell carcinoma Stomach: Adenocarcinoma Colon: Adenocarcinoma
Rectum: Adenocarcinoma Anus: Squamous cell carcinoma Liver: Hepatocellular carcinoma Pancreas: Adenocarcinoma Bronchus: Squamous cell carcinoma
Lung: Adenocarcinoma Breast: Scirrhous adenocarcinoma Adenocarcinoma Medullary carcinoma Urinary bladder: Transi tional cell carcinoma Squamous cell carcinoma In situ (papillary) carcinoma Kidney: Renal cell carcinoma Cervix: Squamous cell carcinoma Carcinoma in situ
Total No. positive b
Total
Number positive
CEA-positive specimens iJ.g/g
Number
Formalin-fixed sections Positive
CEA-negative specimens d
negative
Total'
iJ.g/g
Number positive
fresh specimens
Positive
old sections
Number negative
o
o
o
3
o
o
o
3
5
o o
o
o
o
5
o
o
o
5
5
o
3
o
3
5
o
o
o
5
8
o
1
o
8
o
o
o
8
1
o
o
o
o
o
o
o
1
4
o
3
o
3
4
o
o
o
4
4
o
o
o
o
4
o
o
o
4
12
5
3
27
18
12
10
5
1
o
1
o
9
o
o
o
7
2
o
o
17
6
10
3
11
1
1
o
9
o o o
o o
o o o
o
10
1
o
o
3
o
1
o
o o
o o
o
3
o
3
o
3
0.002, 0.005, 0.007
19
8
9
5
4
11
o
o
o
0.2, 0.2, 0.3, 1.4
3
8 1
VOL. 57, NO.1, JULY 1976
2
9
0.0004, 0.002, 0.07 0.008
0.01, NA
NA,
18.6
2
0.002,0.2
12
5
1
4
7
5.4, 6.5, 10.1, 11.2, 23.1, 24.0, 26.2, 39.3, 177.6
3
0.1, 0.2, NA
27
17
8
9
10
3.0
o
10
5
4
5
1
o
o
o
1
o
9
o
o
o
9
o
7
2
o
2
5
0.008, 0.03, 0.06, 0.3, 1.5, 0.4, NA
17
5
2
3
12
0.1
11
0.03
5.5, 5.7, 9.2
7
10
9
o
o
o
9
2 1
o o
o o
o o
2
o
10
1
o
3
o o
o o o
o o
1
3
o
o
o
3
19
6
3
3
13
11
o
o
o
11
o 2
0.7, 0.9, 4.8, 7.7, 239.1
o
o
0.2,0.2
J
1
9 3
NATL CANCER INST
GOLDENBERG, SHARKEY, AND PRIMUS
14
TABLE l.-Continued Ethanol-fixed sections" Tissue histopathology
Total No. tested a
Ovary: Mucinous cystadenocarcinoma Serous cystadenocarcinoma Squamous cell carcinoma Adenocarcinoma Endometrium: Adenocarcinoma Prostate: Adenocarcinoma
a
c d
