Carcinoembryonic antigen (CEA) is a cell-surface glycoprotein serving as a drug target, diagnostic marker, and serum marker for cancer monitoring. However, prevalence data on CEA expression in cancer tissues vary considerably. This study was designed to determine CEA expression in normal and neoplastic tissues.

A tissue microarray containing 13,725 samples from 120 different tumor types, as well as 76 different normal tissue types, was analyzed by immunohistochemistry (IHC).

CEA was detectable in 65 (54.2%) of 120 tumor categories, including 49 (40.8%) tumor types with at least one strongly positive case. CEA positivity was most common in colorectal adenomas (100%) and carcinomas (98.7%), other gastrointestinal adenocarcinomas (61.1-80.3%), medullary carcinomas of the thyroid (96.3%), pulmonary adenocarcinoma (73.7%), mucinous carcinomas of the ovary (79.8%) and the breast (43.2%), small-cell carcinomas of the lung (64.3%), and urinary bladder (38.9%). CEA overexpression was linked to high tumor grade and invasive growth (p < 0.0001 each) in urinary bladder cancer, and estrogen and HER2 receptor positivity (p ≤ 0.0158) in invasive breast cancer of no special type. In colorectal adenocarcinomas, reduced CEA expression was associated with mismatch repair deficiency (p < 0.0001).

The comprehensive list of CEA-positive human tumor types demonstrates that CEA is expressed in a broad range of epithelial neoplasms, many of which might benefit from CEA serum monitoring and anti-CEA therapies.

We performed a retrospective examination of spontaneous hepatocellular carcinomas (HCCs) (primary and metastatic tumors) in 14 captive prosimians brought to the Veterinary Medical Diagnostic Laboratory in North Carolina State University over a period of 11 years (2003 to 2014) to characterize the tumors. These animals are endangered primates; a better understanding of the main fatal neoplasms is crucial. In addition to the histologic evaluation, an immunohistochemical study was also performed, using a hepatocyte marker (hepatocyte paraffin 1 [HepPar-1]) and 2 cholangiocyte markers (keratin 7 [K7] and keratin 19 [K19]), in an attempt to identify a specific profile for HCCs with metastatic behavior. Six of the 14 HCCs had pulmonary metastases. The most frequent histopathological findings were a trabecular pattern (14/14, 100%), presence of multinucleated cells (12/14, 85.7%), and foci of extramedullary hematopoiesis (9/14, 64.3%). The mitotic count was significantly higher in the metastatic HCCs (P < .05). HepPar-1 was detected in all primary and metastatic HCCs, with a strong intensity of staining. Labeling for K7 and K19 was positive in 12 HCCs (85.7%) and 1 HCC (7.1%), respectively. Contrary to the less aggressive HCCs, most of the metastatic HCCs (5/6) expressed K7 in more than 15% of cells. The percentage of K7-positive neoplastic hepatocytes was significantly higher in metastatic HCCs. This study suggests that K7 might be a prognostically relevant marker in HCCs of captive prosimians.

Despite recent advancements in therapeutic options for advanced hepatobiliary cancers, there remains an unmet need for innovative systemic treatments. Immunotherapy has shown an ability to provide prolonged clinical benefit, but this benefit remains limited to a small subset of patients. Numerous ongoing endeavors are investigating novel immunotherapy concepts. Immunotherapies that have demonstrated clinical efficacy in hepatobiliary cancers include PD-1 inhibitor therapy and CTLA-4 inhibitor therapy. Novel immunotherapy concepts include targeting emerging checkpoint proteins, bispecific T-cell engagers, combinatorial trials with checkpoint inhibitors, oncolytic virotherapy and chimeric antigen receptor T cells. The goal for these new treatment strategies is to achieve a meaningful expansion of patients deriving prolonged clinical benefit from immunotherapy.

To evaluate the significance of resection margin width in the management of hepatocholangiocarcinoma (HCC-CC).

Data of consecutive patients who underwent hepatectomy for hepatic malignancies in the period from 1995 to 2014 were reviewed. Patients with pathologically confirmed HCC-CC were included for analysis. Demographic, biochemical, operative and pathological data were analyzed against survival outcomes.

Forty-two patients were included for analysis. The median age was 53.5 years. There were 29 males. Hepatitis B virus was identified in 73.8% of the patients. Most patients had preserved liver function. The median preoperative indocyanine green retention rate at 15 min was 10.2%. The median tumor size was 6.5 cm. Major hepatectomy was required in over 70% of the patients. Hepaticojejunostomy was performed in 6 patients. No hospital death occurred. The median hospital stay was 13 d. The median follow-up period was 32 mo. The 5-year disease-free survival and overall survival were 23.6% and 35.4% respectively. Multifocality was the only independent factor associated with disease-free survival [P < 0.001, odds ratio 4, 95% confidence interval (CI): 1.9-8.0]. In patients with multifocal tumor (n = 20), resection margin of ≥ 1 cm was associated with improved 1-year disease-free survival (40% vs 0%; log-rank, P = 0.012).

HCC-CC is a rare disease with poor prognosis. Resection margin of 1 cm or above was associated with improved survival outcome in patients with multifocal HCC-CC.

Hepatocholangiocarcinoma (HCC-CC) is a rare primary liver cancer. Its long-term prognosis is still not well-defined. Results from the Eastern and Western literature have been conflicting and no conclusions can be drawn. The aim of the present study was to review the long-term outcome of curative hepatectomy for HCC-CC.

Prospectively collected data from December 1991 to 2006 recording patients with primary liver cancer receiving curative hepatectomy were reviewed. Twenty-five patients, 16 men and 9 women with a median age of 48 years, all ethnic Chinese, had HCC-CC. Their long-term outcome of resection was analyzed and compared to that of patients with cholangiocarcinoma (CC) or hepatocellular carcinoma (HCC).

The HCC-CC patients had a median tumor size of 7.5 cm. Five of them developed postoperative complications. The median follow-up period was 25 months. All of the patients developed recurrence. The median overall survival was 25.2 months. The HCC-CC and CC groups had significantly worse overall survival than the HCC group (HCC versus HCC-CC, p = 0.012; HCC versus CC, p = 0.001) whereas between them there was no significant difference (p = 0.822). As for disease-free survival, there was no significant difference between the three groups; the median disease-free survival for HCC-CC patients was 13.5 months; that for CC patients, 16.1 months; and that for HCC patients, 19.0 months. All HCC-CC patients died within 120 months of primary surgery.

Hepatocholangiocarcinoma entails poor long-term outcome after potentially curative hepatectomy. Other modalities of treatment should be explored in order to prolong survival of patients with this disease.

Combined hepatocellular-cholangiocarcinoma is a rare form of primary liver cancer showing features of both hepatocellular and biliary epithelial differentiation. We report here on a case with collision tumor, which apparently was the coincidental occurrence of both hepatocellular carcinoma and cholangiocarcinoma underlying schistosomiasis. A 39-year-old-Philippine female was transferred to our hospital for evaluation of a liver mass that was found on ultrasonography at a local hospital. HBsAg and Anti-HCV were negative and serum alpha-fetoprotein (AFP) level was normal. The tumor mass was histologically diagnosed as adenocarcinoma by sono-guided biopsy before the operation. Partial lobectomy was performed and we histologically identified the concurrent occurrence of hepatocellular carcinoma and cholangiocarcinoma, (a "collision type carcinoma").

Differential diagnosis of primary and metastatic carcinomas of the liver can be problematic and may require immunohistochemical work-up. Recently, new immunohistochemical markers have been introduced with promising results in this area. We studied three of these markers, human hepatocyte antibody (HepPar-1), human epithelial-related antigen (MOC-31), and thyroid transcription factor-1 (TTF-1), in cell-block sections of fine-needle aspirations from 30 hepatocellular carcinomas and 27 metastatic carcinomas from various sites. Of 27 metastatic carcinomas, all but one were positive for MOC-31 and, all but one were negative for HepPar-1. TTF-1 was positive only in the nuclei of metastatic poorly differentiated neuroendocrine carcinoma from the lung. Of 30 hepatocellular carcinomas, 26 were positive for HepPar-1 and all were negative for MOC-31, while TTF-1 showed cytoplasmic staining in 23 cases. HepPar-1, MOC-31, and TTF-1 are complementary markers in the differential diagnosis of primary and metastatic carcinomas of the liver, with high sensitivity and specificity.

A well-characterized positive marker for hepatocellular differentiation would be a useful tool for the diagnosis of hepatocellular carcinoma (HCC). The recently commercially available Hep Par 1 antibody (clone OCH1E5.2.10) has been reported to be a sensitive marker for HCC in paraffin embedded sections. Of non-hepatocellular tumors, occasional carcinomas have been reported to stain, most frequently gastric adenocarcinomas. This study further evaluated the staining of this antibody on a large number of neoplasms using tissue microarray technology as well as conventional tissue sections. Six hundred seventy-six tumors, including 19 cases of HCC, were tested. Eighteen of 19 cases of HCC were positive, 3 showing <5% staining. Two cases negative on the array showed focal staining when whole tissue sections from the same tumors were used. 16 of 34 cases of gastric carcinomas gave positive reactions, 4 of these showed less than 5% staining. Staining of gastric carcinomas was not limited to signet ring-type carcinomas or to areas of hepatoid differentiation. Only 1 of 11 cases of cholangiocarcinoma showed focal staining. We also noted several other tumors to stain occasionally, including adrenal cortical carcinoma (3/13), yolk sac tumor (2/9), colonic adenocarcinoma (8/106), lung carcinoma (3/52), ovarian carcinoma (5/48), and endocervical adenocarcinoma (1/5). We did not observe staining in pancreatic carcinoma (11), renal cell carcinoma (36), breast carcinoma (85), melanoma (25), or mesothelioma (5). This study supports Hep Par 1 as a useful marker in the differential diagnosis of HCC, but with significant limitations. Cautious use of this antibody in a panel with other positive (alpha fetoprotein, CD10, polyclonal carcinoembryonic antigen) and negative (epithelial membrane antigen, monoclonal carcinoembryonic antigen, CD15) markers of hepatocellular differentiation may aid in the accurate diagnosis of HCC.

Distinguishing hepatocellular carcinoma (HCC) from cholangiocarcinoma (CC) and metastatic adenocarcinoma (MA) involving the liver can be problematic, often requiring the use of immunohistochemistry to facilitate diagnosis. Hep Par 1, a monoclonal antibody with expression confined primarily to benign and malignant hepatocytes, has recently become commercially available. We evaluated Hep Par 1 along with other immunohistochemical markers used to differentiate HCC, CC, and MA, including AE1/AE3, CAM 5.2, B72.3, monoclonal carcinoembryonic antigen (mCEA), polyclonal CEA (pCEA), alpha-fetoprotein (AFP), factor XIIIa, inhibin, CD10, villin, MOC-31, cytokeratin (CK) 7, CK 19, and CK 20, to determine the markers most useful in differentiating these entities. Forty-two cases of HCC, 9 cases of CC, and 56 cases of MA (24 colon, 15 pancreas, 8 ovary, 5 breast, and 4 stomach) were studied. Hep Par 1 was sensitive and specific for HCC, with 38 of 42 (90%) cases staining positively, whereas reactivity was observed in only 8 of 56 (14%) MAs and 0 of 9 CCs. Though limited somewhat by poor sensitivity, a bile canalicular pattern of staining with pCEA, CD10, and villin was specific for HCC and was not observed in the other tumors. Lack of mCEA and MOC-31 immunoreactivity was also characteristic of HCCs. CK 19 positivity favored CC over HCC, but was not useful in differentiating CC from MA. Expression of AFP, although observed in only about one third of the cases, favored HCC over CC and MA. CK 7 and CK 20 were also useful in this differential diagnosis, particularly when dealing with MA of colonic origin. AE1/AE3, CAM 5.2, B72.3, inhibin, and factor XIIIa were noncontributory in differentiating these entities.

Humoral hypercalcemia caused by parathyroid hormone-related peptide (PTHrP), associated with cholangiocellular carcinoma (CCC), has rarely been documented. There have been no reports of CCC associated with extensive calcification of the tumor with psammoma body formation. A 66-year-old man was admitted with a large calcified tumor in the liver detected on an abdominal X-ray. An ultrasound-guided fine needle biopsy specimen of the liver tumor showed evidence of adenocarcinoma. He had hypercalcemia with an elevated PTHrP level. The patient died because of disseminated intravascular coagulation and progressive hepatic failure. A postmortem examination revealed a large poorly differentiated CCC in the liver. Immunohistochemical examination showed the presence of PTHrP-positive tumor cells. The calcified lesion consisted of a number of accumulated psammoma bodies. We present a case of PTHrP producing CCC with a marked psammoma formation.

Differentiating between primary tumors of the liver and metastatic lesions can, at times, be difficult. Various histochemical and immunohistochemical methods have been used in an effort to better delineate between hepatocellular carcinoma (HCC), especially the microglandular variant, primary cholangiocarcinoma, and metastatic adenocarcinoma; these ancillary studies can yield less than satisfactory results. Recently, anti-MOC31, a monoclonal antibody directed against a cell surface glycoprotein, has been shown to be helpful in distinguishing between adenocarcinoma and mesothelioma. This study addresses whether this antibody might be helpful in distinguishing between HCC, primary cholangiocarcinoma, and metastatic adenocarcinoma in the liver. Formalin-fixed, paraffin-embedded tissue sections from 15 HCC (including 10 microglandular variants), 14 primary cholangiocarcinomas, and 33 metastatic adenocarcinomas (7 colon, 1 lung, 8 breast, 4 GE jct/gastric, 9 pancreas, 2 small intestine, 1 renal, 1 ovary) were immunostained with anti-MOC 31 (1:40, Dako) after protease digestion and biotin block using a modified ABC technique. Positive staining was limited to membrane based reactivity; controls stained appropriately. Immunoreactivity for MOC31 was observed in 14 of 14 cholangiocarcinomas and 33 of 33 metastatic tumors. Staining was diffuse, intense, and readily interpretable, with rare exceptions. All 15 cases of HCC were negative. We conclude that cholangiocarcinoma and metastatic adenocarcinoma from a variety of sites express MOC31; HCC is uniformly negative for this marker. Anti-MOC31 may prove useful in the evaluation of liver neoplasms where primary hepatocellular and adenocarcinoma enter the differential diagnosis; it is not useful in separating primary cholangiocarcinoma from metastatic adenocarcinoma.

Eight hepatic atypical adenomatous hyperplasias (AH), 30 hepatocellular carcinomas (HCC) consisting of 11 well-, 13 moderately and six poorly differentiated HCC, and 10 intrahepatic cholangiocarcinomas (CC) were investigated immunohistochemically with anti-alpha-fetoprotein (AFP), carcinoembryonic antigen (CEA), CA19-9, epithelial membrane antigen (EMA), and cytokeratins (CK) 18 and 19 antibodies. Immunostaining was regarded as positive when more than 5% of cells were stained. Alpha-fetoprotein was positive, although focally, in five (17%) of 30 HCC but negative in all AH and CC. Carcinoembryonic antigen (polyclonal antibody) did not stain the cytoplasm of all AH and HCC, but stained two (25%) of eight AH and 10 (33%) of 30 HCC in a bile canalicular staining manner. Carcinoembryonic antigen showed intracytoplasmic or luminal border staining in six (60%) of 10 CC. CA19-9 was negative in all AH and HCC, while six (60%) of 10 CC were positive for CA19-9. Epithelial membrane antigen was positive in one (13%) of eight AH, seven (23%) of 30 HCC and in all 10 cases of CC. Cytokeratin 18 was positive in all AH, HCC and CC. Cytokeratin 19 was negative in both AH and HCC, whereas it stained the cytoplasm of tumor cells in all CC diffusely and intensely. These results suggest that immunostaining of AFP, CEA, CA19-9, EMA, CK18 and CK19 are not useful in the differential diagnosis between AH and well-differentiated HCC, and that CK19 is the most suitable reagent for the differential diagnosis between HCC and CC.

To examine the usefulness of Hep Par 1 together with selected antibodies in the separation of hepatocellular carcinoma (HCC) from cholangiocarcinoma (CC), combined tumours (HCC-CC) and metastatic carcinoma.

Antibodies to Hep Par 1, CK19, CK20 and factor XIIIa were applied to 32 HCCs, 27 CCs, five HCC-CCs and 19 metastatic carcinomas from a variety of sites. Hep Par 1 produced distinctive granular staining of all benign hepatocytes and stained 30 HCCs in a heterogeneous manner, irrespective of the degree of differentiation. While labelling all cases of combined HCC-CC, the antibody also stained the mucus-secreting cells of four cases of pure CC. Anti-CK19 produced distinctive staining of bile ducts and CC but also decorated four HCCs and 10 metastatic tumours. Factor XIIIa was not found in normal, reactive or neoplastic hepatocytes. CK20 was found in some cases of HCC and CC and in all cases of metastatic carcinomas from the colon.

Hep Par 1 was a sensitive marker of hepatocytes but its variable staining in HCC may produce false negative results in small biopsies and it was occasionally found in CC. The highest diagnostic yield was obtained when anti-Hep Par 1, CK19 and CK20 were used in a panel. Factor XIIIa staining has no role in the diagnosis of liver cancers.

Cholangiocarcinomas may be extrahepatic or intrahepatic; the latter are further divided into hilar and peripheral types. Peripheral cholangiocarcinomas often resemble adenocarcinomas arising in other organs. Although clear cell changes may occur in hepatocellular carcinoma and extrahepatic cholangiocarcinoma, peripheral cholangiocarcinomas with clear cell change are rare. In such cases, an extrahepatic primary carcinoma must be excluded. We present a patient with a large, clear cell papillary carcinoma in the liver. Extensive workup of the patient for other possible primary sites including kidneys, adrenals, thyroid, prostate, or urinary bladder failed to indicate any other neoplasm. The patient is alive without evidence of disease 30 months after complete resection. The histological, immunohistochemical, and electron microscopic results were most consistent with a neoplasm in the cholangiocarcinoma family. To the best of our knowledge, a clear cell papillary peripheral cholangio carcinoma has not been described previously. This neoplasm may be related to the recently described clear cell carcinomas of the gallbladder and extrahepatic bile ducts.

The immunohistochemical expression of the oncofetal proteins alpha-fetoprotein (AFP) and carcinoembryonic antigen (CEA), and the intermediate filament proteins keratin and vimentin was analysed in 18 canine liver carcinomas. All the tumours other than hepatocellular carcinomas, with the exception of one poorly differentiated carcinoma, were AFP negative, and only cholangiocarcinomas and mixed (hepatocellular and cholangiocellular) carcinomas were CEA positive. All the histological types of tumours expressed high and low molecular weight keratins, and keratin and vimentin were both expressed in three tumours (one moderately differentiated hepatocellular carcinoma, one mixed carcinoma and one poorly differentiated carcinoma). The findings demonstrate the use of immunohistochemical staining methods for analysing the expression of some tumour markers in routinely processed tissue samples of canine liver carcinomas, and suggest that some of the tumour markers are correlated with histological types of tumour.

Thirteen cases of combined hepatocellular (HCC) and cholangiocellular carcinoma (CCC) were examined. In addition to routine pathology, immunoreactivities for carcinoembryonic antigen, alpha-fetoprotein (AFP), cytokeratin (Cam 5.2 and AE1), epithelial membrane antigen (EMA) and tumor-associated glycoprotein 72 (B72.3) were also examined. The average age of the 13 cases was 64.8 years, which lay between the average ages of pure HCC and CCC cases. They were categorized as separate type (2), collision type (6), and intermingled type (5). AE1 and EMA were the best markers to differentiate the CCC from the HCC area. B72.3 immunoreactivity was detected only in CCC (46%). There were no transitional features between HCC and CCC in two cases of the separate type and two cases of the collision type. However, focal transitional features from HCC to CCC were observed in all cases of the intermingled type and in four of six cases of the collision type. In one case of the intermingled type, many cancer cells contained both bile and mucus simultaneously, and revealed dual immunoreactivities. The conclusions are: 1) the combined type is generated from two sources; one is the intrahepatic double cancer (thoroughly separate type and a part of the collision type) and another is the stem cell origin with diverse phenotypes (intermingled type and a part of the collision tumor); and 2) AE1 was the most helpful marker to differentiate the CCC area from HCC, and other markers, e.g. AFP for HCC and EMA, CEA, and B72.3 for CCC, were also supportive but somewhat limited in the differential diagnosis.

Expression of mucin core protein of mammary type (MUC-1 core protein) was investigated in primary hepatic carcinoma (25 cases of cholangiocarcinoma, 15 cases of combined hepatocellular-cholangiocellular carcinoma and 18 cases of hepatocellular carcinoma) with monoclonal antibody DF3 and a standard avidinbiotin complex method. MUC-1 core protein was almost always expressed in cholangiocarcinoma and in the cholangiocarcinoma area of hepatocellular-cholangio-cellular carcinoma to a varied degree, whereas such expression was virtually absent in hepatocellular carcinoma and the hepatocellular carcinoma areas of hepatocellular-cholangiocellular carcinoma. Nonneoplastic intrahepatic biliary epithelium, as well as hepatocytes, was virtually negative for this protein. In well-differentiated cholangiocarcinoma, this protein tended to be expressed on luminal surfaces, whereas poorly differentiated cholangiocarcinoma showed cell membranous or diffuse cytoplasmic staining patterns. Double staining with Alcian blue (pH 2.5) and immunostaining for MUC-1 core protein showed that although some parts of cancerous areas were positive for both stains, most cancerous areas were only positive for one. Alcian blue--positive areas were dominant over MUC-1 core protein--expressing areas in well-differentiated cholangiocarcinoma, whereas the reverse was the case in poorly differentiated cholangiocarcinomas and also in cholangiocarcinoma areas of hepatocellular-cholangiocellular carcinoma. This study is the first report to document that cholangiocarcinoma and cholangiocarcinoma areas of hepatocellular-cholangiocellular carcinoma express MUC-1 core protein.(ABSTRACT TRUNCATED AT 250 WORDS)

To determine if hyperplastic and neoplastic lesions from medaka showed similar immunoreactivity to intermediate filament antibodies as the tissues of origin, two week old medaka were exposed to 10 or 20 mg/L of methylazoxymethanol acetate for two hours and transferred to clean water for up to six months. Using a streptavidin peroxidase method, paraffin embedded Bouins fixed neoplasms were incubated with cytokeratin, vimentin, or neurofilament antibodies. Like their nonneoplastic cellular counterparts, hepatocellular carcinoma, pancreatic acinar carcinoma and mesenchymal neoplasms including hemangioma and hemangiopericytoma reacted negatively to cytokeratin antibodies. Cholangiocarcinoma, mesothelioma, and proliferative lesions containing biliary epithelial cells reacted positively to cytokeratin antibodies. All neoplasms and proliferative lesions were negative with vimentin and neurofilament antibodies. These data indicate that while some epithelial neoplasms showed cytokeratin reactivity similar to the parent tissues, additional markers are needed to identify mesenchymal tissues and neoplasms.

To develop useful markers for the gastrointestinal neoplasms, the authors have established a new monoclonal antibody (MoAb), FU-MK-1, directed to gastric adenocarcinoma.

The MoAb was produced by a mouse hybridoma technique by immunizing a BALB/c mouse with cancerous ascites derived from a poorly differentiated adenocarcinoma of the stomach.

FU-MK-1 did not react with the normal adult gastric mucosa, but did react with most carcinomas of the gastrointestinal tract. In addition, the MoAb recognized cholangiocarcinomas (CC), but it did not react with hepatocellular carcinomas (HCC). Furthermore, in the combined type tumor consisting of a mixture of HCC and CC, the MoAb react only with CC element, but not with pseudoglandular structures in the HCC areas. These results indicate that FU-MK-1 is a useful antigenic marker for distinguishing HCC from CC in the liver. Furthermore, because this MoAb retains its reactivity with formalin-fixed paraffin-embedded material, it may become a useful reagent for routine or retrospective immunohistologic studies. The molecular weight of the FU-MK-1 antigen was estimated to be ca. 41,000 dalton by the Western blot analysis. Periodic acid and trypsin treatment on the antigen suggested that the antigenic determinant is a glycoprotein.

This MoAb may contribute to the histopathologic and immunologic studies of the digestive system.

This study, aimed at elucidating the epidemiological features of primary liver carcinoma developing in non-cirrhotic livers, was based on 25,103 autopsies performed between 1975 and 1984 in Trieste, Italy. These autopsies correspond to approximately 70% of all deaths that occurred in this area. Various factors allegedly related to carcinomas were analysed in reference to our previous study on cirrhotic livers and in comparison with 5,603 autopsies in Kurume, Japan. There were 28 cases of hepatocellular carcinoma (HCC), 16 of cholangiocellular carcinoma (CCC) not associated with cirrhosis in Trieste, and 48 HCC and 19 CCC in Kurume. On the basis of our findings, it was concluded that cirrhosis, regardless of its cause, is the main pathogenetic factor in HCC; it is responsible for a much higher frequency (14.2:1) than in non-cirrhotic livers, as well as for early occurrence of tumours (an average of 6 years earlier in cirrhotic liver) in Trieste. Patients in Trieste were older than those in Japan, and the frequency of HCC among all autopsies was much greater in the latter. By contrast, the influence of cirrhosis on cholangiocellular carcinoma (CCC) was negligible, as such association appeared purely coincidental or absent. The incidence of CCC among autopsies was greater in Japan. Our data on CCC were not sufficient to demonstrate any clear aetiopathogenetic association between this tumour and alcohol abuse and hepatitis B virus (HBV) infection, except for a possible aetiological role of gallstones. The frequency of CCC relative to HCC was greater in Trieste than in Japan; the incidence of HCC was much less in Trieste, whereas CCC was more frequent in Japan.

One hundred benign and malignant primary liver tumours were screened immunocytochemically for alpha-fetoprotein (AFP), alpha 1-antitrypsin, alpha-human chorionic gonadotropin, carcinoembryonic antigen (CEA), keratin and vimentin. Alpha-fetoprotein was found in 16/63 (24%) hepatocellular carcinomas and in two hepatoblastomas. When comparing tissue positivity for AFP with tumour differentiation, grade 1 hepatocellular carcinomas were found to be negative, while 21% of grade 2, 36% of grade 3 and 16% of grade 4, respectively, stained positively. Alpha-fetoprotein positive cells were present in 9/10 hepatocellular carcinomas with serum levels exceeding 5000 ng/ml, but were absent in 17 tumours with serum AFP levels below 5000 ng/ml. All tumours other than hepatocellular carcinomas and hepatoblastomas were AFP negative. Carcinoembryonic antigen was present in 72% of cholangiocarcinomas, but was demonstrated in only one hepatocellular carcinoma. This exception was a combined hepatocellular-cholangiocarcinoma in which CEA expression was restricted to the cholangiocellular part. Alpha 1-antitrypsin was found in 4/63 hepatocellular carcinomas, in 2/2 fibrolamellar carcinomas and in 2/18 cholangiocarcinomas. Alpha-human chorionic gonadotropin was detected in one hepatocellular carcinoma and was strongly expressed in both fibrolamellar carcinomas. Weak staining for keratin was seen in most tumours with hepatocellular differentiation. All cholangiocarcinomas, in contrast, were strongly labelled with the keratin antibody. Co-expression of keratin and vimentin was observed in seven poorly differentiated hepatocellular carcinomas and three cholangiocarcinomas as well as in the two hepatoblastomas. The findings suggest that AFP is a diagnostic but rather insensitive immunocytochemical marker for hepatocellular differentiation in malignant liver tumours; CEA and keratin may help in discriminating cholangiocarcinomas from hepatocellular carcinomas.

An antiserum to carcinoembryonic antigen (CEA) and a monoclonal antibody to cytokeratin 19 (CK 19) were studied for their suitability as diagnostic reagents for the differential diagnosis of primary and secondary malignant epithelial tumours of the liver, on paraffin sections. With the antiserum to CEA, positive bile canalicular structures were found in 60 per cent of the hepatocellular carcinomas. All the cholangiocarcinomas and 66.6 per cent of the metastatic carcinomas were positive for CEA, without displaying a canalicular staining pattern. All the hepatocellular carcinomas were negative for CK 19. All the cholangiocellular carcinomas and the metastatic carcinomas were positive for CK 19. This staining profile may prove helpful in difficult diagnostic cases.

Eleven cases of hepatocellular carcinoma (HCC) in childhood were investigated by immunohistochemistry for association with hepatitis B virus (HBV) infection. Seven of 11 cases (64%) demonstrated positivity for hepatitis B surface antigen (HBsAG), whereas all 11 were negative for hepatitis B core antigen (HBcAG). Cirrhosis was absent in all cases, and other causes for HCC in childhood were not found. All children with HBV-associated HCC died within 6 months of diagnosis. The median survival time of these children was 2 months. Only one child with HCC of trabecular subtype without HBV association is still living after 18 months. However, this child has metastases and a local recurrence. Three other children with HCC of fibrolamellar subtype are free of disease after 2, 5, and 6 years, respectively. The high number of cases of HBV-associated HCC shows the important role of HBV infection as an etiologic factor for the development of childhood HCC in middle Europe.