CEA is an oncofetal glycoprotein. It is expressed by fetal epithelial cells and in small amounts by normal adult epithelial cells and benign tumours. It is present in large amounts in adenocarcinomas of the gastrointestinal tract (including pancreas), lung and medullary carcinoma of the thyroid.
The rate of immunoreactivity in adenocarcinomas varies with different clones:
|
DAKO clone II-7 |
BioGenex clone THF 3H8-1 |
E-Z-EM clone D14 |
NKI, Parlam 1 |
||
Colonic carcinoma |
25/2542 |
25/2542 |
25/2542 |
93/9443 |
||
22/3542 |
32/3542 |
19/3542 |
|
|||
Breast: ductal carcinoma |
30/11942 |
86/11942 |
28/11942 |
43/10743 |
||
Breast: lobular carcinoma |
5/1042 |
7/1042 |
4/1042 |
|||
Gastric carcinomas |
29/3942 |
28/3942 |
26/3942 |
|
||
Pancreatic carcinoma |
23/2642 |
24/2642 |
21/2642 |
|
||
Ovarian carcinoma |
0/2942 |
6/2942 |
0/2942 |
25/8643 |
||
Renal carcinomas |
0/4542 |
3/4542 |
0/4542 |
|
||
It is also expressed by:
some myelomas
Diagnostic utility
Differentiation of mesotheliomas from adenocarcinomas
adenocarcinoma |
mesothelioma |
|
Wang 197912 (absorbed polyclonal antibody) |
12/12 |
0/9 |
Corson 198215 (polyclonal antibody) |
20/20 (all cases stained strongly or moderately) |
9/20 (1 case moderate staining, 8 cases weak or equivocal) |
Kwee 198216 (absorbed polyclonal antibody) |
16/25 (adenocarcinoma metastatic to pleura) |
0/37 |
Whitaker 198217 (absorbed polyclonal antibody) |
21/26 |
0/43 |
Holden 198418 (polyclonal antibody) |
18/18 (staining usually strong) |
8/22 (staining usually weak) |
Battifora 198519 (polyclonal antibody, Dako) |
65/100 (all 24 lung adenocarcinomas positive) |
2/12 (2 cases weakly positive) |
Gibbs 198520 (polyclonal antibody) |
24/27 |
0/29 |
Sheibani 198621 (absorbed polyclonal antibody) |
38/50 |
0/28 |
Sheibani 198621 (monoclonal antibody) |
36/50 |
0/28 |
Szpak 198622 (polyclonal antibody) |
19/20 |
4/20 |
Dewar 198723 (monoclonal antibody, Amersham) |
15/20 (intrapulmonary adenocarcinomas) |
0/12 (epithelioid type) |
Otis 198724 (absorbed polyclonal antibody, Dako) |
10/14 |
2/19 |
Otis 198724 (monoclonal, clone CEA-5) |
9/14 |
6/19 |
Pflatz 198725 (polyclonal antibody, Dako) |
22/22 |
2/47 |
Ordonez 198926 (monoclonal, Hybritech clone CEJ065) |
21/23 (all lung primaries) |
0/19 (all epithelioid type) |
O'Hara 199027 (absorbed polyclonal antibody, Dako) |
22/28 |
0/20 |
Wick 199028 (monoclonal antibody, Hybritech clone CEJ065) |
50/52 (peripheral adenocarcinomas of lung with pleural invasion) |
0/51 (all epithelioid type) |
Wick8 (Hybritech murine monoclonal, catalogue #0062) |
28/43 (22/23 peripheral pulmonary adenocarcinomas involving pleura, 1/10 serous surface papillary adenocarcinomas of peritoneum, 5/10 breast carcinomas metastatic to pleura. staining was predominantly membranous) |
0/41 (0/21 malignant epithelioid pleural mesotheliomas, 0/20 malignant epithelioid peritoneal mesotheliomas ) |
Delahaye 199129 (monoclonal antibody, on cytological smears) |
3/10 |
1/24 |
Wirth 199130 (monoclonal antibody, Zymed) |
19/20 (all lung primaries) |
0/50 |
Sheibani 19921 |
44/83 |
0/115 |
Collins 19922 |
38/48 |
0/31 |
Gaffey 199241 |
not done |
0/49 (32 epithelioid, 17 biphasic) |
Brown 199331 (monoclonal antibody, Zymed) |
100/103 (all lung primaries) |
1/34 (all epithelioid type) |
Moch 199332 (monoclonal antibody, Amersham) |
18/24 |
0/27 |
Dejmek 199433 (absorbed polyclonal antibody, Dako) |
20/20 |
13/61 |
Dejmek 199433 (monoclonal antibody, Dako, clone A5B7) |
15/20 |
0/61 |
Dejmek 199433 (monoclonal antibody, Boehringer-Mannheim, clone CEJ065) |
16/20 |
7/61 |
Dejmek 199433(monoclonal antibody, Biogenex, clone SP651) |
16/20 |
1/61 |
Doglioni 19963 (antibody not stated) |
20/22 |
2/20 |
Ordonez 19974 (polyclonal antibody, Incstar) |
92/107 |
0/139 |
Dejmek 199710 |
29/43 |
1/110 |
Bateman 199734 (polyclonal antibody, Dako |
10/14 |
2/17 |
Bateman 199734 (monoclonal antibody, Dako |
10/14 (primary and secondary adenocarcinomas involving pleura) |
0/17 |
Delahaye 199735 (monoclonal antibody, Euro-Diagnostics, clone PARLAM-4, on cytological preparations of pleural fluid) |
18/24 |
0/41 |
Leers 199813 |
11/21 ( 11 cases stained > 10% of cells. In addition, 4 cases showed staining of <10% of cells.) |
0/20 (2 cases showed staining of <10% of cells.) |
Garcia-Prats 199836 (monoclonal antibody, Dako, clone A5B7) |
12/18 (10/15 primary pulmonary adenocarcinomas and 2/3 extrapulmonary) |
1/40 (0/26 epithelioid, 0/10 sarcomatoid, 1/4 biphasic) |
Brockstedt 20005 |
37/57 |
2/119 |
Kayser 20016 |
112/146 (69/82 lung, 33/47 breast, 2/3 colon, 0/2 kidney, 8/12 site not known) |
11/118 (11/99 epithelioid, 0/12 mixed and 0/7 sarcomatoid) |
Gonzalez-Lois 20017(monoclonal antibody) |
11/15 (metastatic to pleura from various sites) |
3/44 (2/33 epithelioid, 1/5 sarcomatoid, 0/1 mixed and 0/1 lymphohistiocytoid mesotheliomas) |
Carella 20019 (polyclonal antibody, Dako) |
17/20 |
2/46 |
Comin 200111 (monoclonal antibody, Novocastra, clone 12.140.10) |
22/23 (Staining was strong and evenly distributed throughout the cytoplasm.) |
4/42 (The four positive cases showed staining of less than 5% of cells.) |
Roberts 200137 (monoclonal antibody, Novocastra, clone 12.140.10) |
15/18 |
1/112 |
Abutaily 200214 (monoclonal antibody, Dako) |
9/11 (35 primary ppulmonary adenocarcinomas were studied but CEA was applied only to the 11 E-cadherin-positive / TTF-1 negative tumours) |
1/41 (11 epithelioid, 7 sarcomatoid, and 23 mixed) |
Ordonez 200338 |
44/50 (all primary lung adenocarcinomas: 16 case >75% of cells stained, 17 cases 50-75% of cells, 8 cases 25-50% of cells and 3 cases 1-25% of cells. Staining was cytoplasmic with acentuation along the cell membrane.) |
0/60 (all epithelioid mesotheliomas) |
Overall |
78% (908/1160) |
4.4% (56/1283) |
Average for polyclonal antibodies, unabsorbed |
83% % (299/360) |
9% (31/361) |
Average for polyclonal antibodies, absorbed |
78% (127/163) |
7% (15/208) |
Average for monoclonal antibodies |
82% (432/524) |
2.3% (14/607) |
A systematic review of fifty one studies (consisting of 1524 pulmonary adenocarcinomas and 1818 epithelioid mesotheliomas) reported sensitivities and specificities for CEA of 83% and 95% respectively. For twenty four studies (consisting of 949 pulmonary adenocarcinomas and 1007 epithelioid mesotheliomas) using monoclonal CEA, the sensitivities and specificities were 81% and 97% respectively44.
Although CEA is very useful in the differentation of pulmonary adenocarcinoma from mesothelioma, it is less useful for serous carcinomas of the ovary and peritoneum, less than half of which are positive for CEA8,39,40.
Differentiation of reactive mesothelial cells (negative) from carcinoma (positive) in effusions.
differentiation of hepatocellular carcinoma from carcinoma metastatic to the liver: canalicular staining with polyclonal CEA is specific for hepatocellular carcinoma.
differentiation of endocervical from endometrial adenocarcinoma.
References
37 Roberts, F., C. M. Harper, et al. (2001). "Immunohistochemical analysis still has a limited role in the diagnosis of malignant mesothelioma. A study of thirteen antibodies." Am J Clin Pathol 116(2): 253-62. (Initial publication of data as abstract 8Harper CM. Evaluation of a commercially available immunohistochemical diagnostic panel for malignant mesothelioma. J Pathol 2001:193(suppl):39A.)
This page last revised 16.2.2006.
©SMUHT/PW Bishop