Home Translating Report News Physicians Diseases Body Sites Lab tests Search
Home Diseases and Health Information


This is the most common primary malignant tumor arising within the liver, accounting for >90% of all primary liver cancers. Metastatic tumors to the liver are actually more common, usually from the colon or prostate or breast. It is sometimes erroneously referred to as a hepatoma which should be avoided since this name implies a benign neoplasm. About 350,000 cases per year are diagnosed around the world. It is a global problem with 2-4 annual cases/100,000 in North and South America but 20 annual cases/100,000 around the Mediterranean and 150 annual cases/100,000 in China, Korea, and Taiwan. Males are favored ranging from 2:1 to 8:1.

There is a well-established link with hepatitis B viral infection (HBV) which accounts for the high prevalence throughout the world. In the West, hepatitis C viral infection (HCV) and alcoholism are the main culprits. Several mechanisms have been hypothesized to account for the neoplastic transformation. Because of the association of HBV infection, vaccination programs for children have dramatically decreased the infection rates.

The tumor may present with upper abdominal pain, malaise, weight loss, and fatigue. Depending upon the degree of liver involvement as well as the presence of cirrhosis, there may be jaundice, gastrointestinal bleeding, and fever. Serum alpha-feto protein is elevated in 60-75% of cases. The disease is progressive and death usually occurs within 10 months of initial diagnosis from cachexia, gastrointestinal bleeding, liver failure, or rupture of the tumor with massive hemorrhage.

The tumor may present as a solitary, multiple, or diffusely infiltrative cancer. HCCs have a marked propensity for vascular invasion, especially the portal vein or inferior vena cava. Under the microscope, the tumor grows in seveal patterns ranging from trabecular to an acinar pseudoglandular pattern.


Disease Associations  
Laboratory/Radiologic/Other Diagnostic Testing  
Gross Appearance and Clinical Variants  
Histopathological Features and Variants  
Special Stains/
Electron Microscopy
Differential Diagnosis  
Commonly Used Terms  
Internet Links  


Fibrolamellar hepatocellular carcinoma occurring 5 years after hepatocellular adenoma in a 14-year-old girl: a case report with comparative genomic hybridization analysis.

Terracciano LM, Tornillo L, Avoledo P, Von Schweinitz D, Kuhne T, Bruder E.

Institute of Pathology, University of Basel, Basel, Switzerland.
Arch Pathol Lab Med. 2004 Feb;128(2):222-6. Abstract quote  

An unusual case of fibrolamellar carcinoma of the liver developed 5 years after removal of a hepatocellular adenoma in a 14-year-old girl belonging to a family with Carney syndrome. Both tumors were studied by light and electron microscopy, flow cytometry, and comparative genomic hybridization.

The first tumor, removed at the age of 9, was a bulky well-circumscribed liver mass composed of large eosinophilic cells with a focal pseudoglandular pattern but without cytologic atypia or sclerosis. A diagnosis of hepatocellular adenoma was rendered. Five years later, another hepatic tumor was removed from the right lobe. Microscopic examination revealed polygonal cells, each with a large amount of eosinophilic cytoplasm and a round nucleus with a conspicuous nucleolus. These cells were arranged in nests and strands and separated by bands of dense fibrous tissue, leading to a diagnosis of fibrolamellar carcinoma. Comparative genomic hybridization analysis revealed no genetic alteration in the adenoma; however, several chromosomal aberrations (loss of chromosome regions 1p and 4p and gains of chromosome regions 6q, 13q, and Xq) were detected in the fibrolamellar carcinoma.

To our knowledge, this is the first report of an association between hepatocellular adenoma and fibrolamellar carcinoma.


Concurrent Evaluation of p53, b-Catenin, and a-Fetoprotein Expression in Human Hepatocellular Carcinoma

Michael Torbenson, MD, etal.
Am J Clin Pathol 2004;122:377-382 Abstract quote

Recent models suggest that hepatocellular carcinoma (HCC) develops through several independent pathways marked by key mutations in the b-catenin or p53 gene. An additional pathway potentially is marked by aberrant expression of a-fetoprotein (AFP).

To see whether these potential markers are expressed independently, we immunostained sequential sections from 55 HCCs. Of the cases, 30 (55%) were positive for 1 or more proteins: AFP, 19 cases (35%); p53, 12 cases (22%); and b-catenin, 9 cases (16%). Seven tumors (13%) were positive for more than 1 protein, with 4 of 7 positive in the same area of tumor and 3 of 7 positive in different areas of the carcinomas. By statistical analysis, expression of the markers was independent of one another and of tumor size.

Concurrent evaluation of p53, b-catenin, and AFP protein expression showed no associations, supporting models in which these proteins might serve as markers of independent pathways in the development of HCC.
AFLATOXINS Produced by molds and activated by hepatocytes
Products are intercalated into DNA to form mutagenic adducts with guanosine of p53 gene
Microsomal enzymes responsible for detoxifying aflatoxins are mutated in high risk areas
Acts synergistically with HBV infection

Caveolin and Thrombospondin Expression During Hepatocellular Carcinogenesis

Yerian, Lisa M MD; Anders, Robert A MD, PHD; Tretiakova, Maria MD, PHD; Hart, John MD

From the Department of Pathology, University of Chicago Hospitals, Chicago IL.

The American Journal of Surgical Pathology : Volume 28(3) March 2004 pp 357-364 Abstract quote

Macroregenerative and dysplastic nodules (MDNs) are hepatocellular carcinoma (HCC) precursor lesions and exhibit distinct vascular profiles relative to adjacent cirrhotic liver. Recent microarray analysis of MDN identified aberrant expression of caveolin-1 and thrombospondin-1, genes suspected to play a role in tumorigenesis at other sites.

We used immunohistochemistry to localize caveolin and thrombospondin expression in 14 MDNs from livers with hepatitis C cirrhosis and in tissue arrays that included samples of MDNs, HCC, and nonneoplastic liver. Hepatocytes were uniformly negative for caveolin. Sinusoidal endothelial cells exhibited increased caveolin expression in MDNs relative to adjacent cirrhotic liver in most (28 of 36, 78%) MDNs evaluated. However, few HCCs showed increased caveolin expression as compared with nonneoplastic liver (5 of 19, 26%). Unpaired arteries showed strong positive endothelial cell staining. Thrombospondin staining was weak or negative in hepatocytes in nearly all (77 of 92, 84%) MDNs and in 46 of 49 HCCs evaluated (94%). Sinusoidal endothelial cells were negative for thrombospondin, but hepatic arteries and MDNs showed positive mural staining; portal veins were positive both in vessel walls and in endothelial cells. The altered expression profiles of these genes identified in microarray analysis are not likely related directly to malignant transformation of hepatocytes but rather to an alteration in the vascular supply to these lesions.

The results illustrate the critical role of histologic techniques in interpretation of microarray data.

FHIT mRNA and protein expression in hepatocellular carcinoma.

Kannangai R, Sahin F, Adegbola O, Ashfaq R, Su GH, Torbenson M.

1Department of Pathology, The Johns Hopkins Hospital, Baltimore, MD, USA.
Mod Pathol. 2004 Jun;17(6):653-9. Abstract quote  

Loss of fragile histidine triad (FHIT) gene expression is seen in approximately 50% of hepatocellular carcinomas in China. However, little information is available on FHIT expression in hepatocellular carcinoma in the United States, where carcinogen exposure is generally lower.

Investigations of FHIT mRNA in hepatocellular neoplasms and paired non-neoplastic tissues demonstrated normal-sized FHIT transcripts in all non-neoplastic tissues and in all neoplasms including 11 hepatocellular carcinomas, two fibrolamellar carcinomas, and four benign proliferative lesions. In addition, all but one malignant and all benign neoplasms showed aberrant smaller transcripts. The smaller aberrant transcripts were overexpressed in 6/11 hepatocellular carcinomas and 1/2 fibrolamellar carcinomas. An additional 79 hepatocellular carcinomas, 12 fibrolamellar carcinomas and 15 hepatic adenomas were examined for FHIT expression by immunohistochemistry. No loss of immunostaining was seen in 67/79 (85%) of hepatocellular carcinomas, while a moderate or marked decrease was seen in 12/79 (15%). Fibrolamellar carcinomas and hepatic adenomas showed no loss of FHIT expression.

In conclusion, hepatocellular carcinomas retained expression of normal FHIT mRNA transcripts, but also showed universal expression of smaller sized aberrant transcripts and commonly overexpressed these aberrant transcripts. Loss of FHIT protein expression is relatively uncommon in this cohort from the United States, where exposure to hepatic carcinogens is generally low.

Hepatocellular carcinomas show abnormal expression of fibronectin protein.

Torbenson M, Wang J, Choti M, Ashfaq R, Maitra A, Wilentz RE, Boitnott J.

Departments of Pathology (MT, AM, REW, JB) and Surgery (MC), The Johns Hopkins Hospital, Baltimore, Maryland.


Mod Pathol 2002 Aug;15(8):826-30 Abstract quote

Fibronectin plays an important role in cell-to-cell adhesion, cell migration, and cell signaling. In the liver, fibronectin expression has been studied primarily as a component of the extracellular matrix, but little information is available on the expression of fibronectin protein in the neoplastic cells of hepatocellular carcinomas (HCCs).

Twenty-four surgically resected HCCs were immunostained with fibronectin. Tumor and normal liver tissues were concurrently analyzed in all cases, and expression in the tumor was evaluated in comparison to the nonneoplastic liver. The average age at resection was 54 +/- 18 years for the 18 men and 6 women. Twenty-one of the cases were classic HCCs including 6 cases that were well differentiated, 12 cases moderately differentiated, and 3 cases poorly differentiated. The remaining 3 cases were moderately differentiated fibrolamellar carcinomas. In the normal liver, fibronectin labeled the sinusoids and weakly to moderately stained the cytoplasm of hepatocytes. In HCCs, 15/24 showed overexpression of fibronectin in the cytoplasm, 8/24 showed no change from the nonneoplastic liver, and one case showed decreased cytoplasmic staining. In addition, an abnormal membranous staining pattern was noted in 16/24 HCCs. In contrast to the HCCs, none of the three fibrolamellar carcinomas showed increased cytoplasmic or membranous staining. Excluding fibrolamellar carcinoma, increased cytoplasmic staining and/or an abnormal membranous staining was noted in 19/21 (90%) of HCCs.

Fibronectin shows abnormal cytoplasmic and/or membranous staining in the majority of HCCs. The implications of fibronectin overexpression are uncertain but may reflect a critical step in tumor genesis.

HBV INFECTION Viral DNA integrated into the host cell genome
Tumors are clonal with respect to these insertions
Viral integration induces genomic instability
HBV X protein acts as a transactivator of cellular and viral promoters and binds to tumor suppressor gene p53

Full-length genomic analysis of hepatitis B virus isolates in a patient progressing from hepatitis to hepatocellular carcinoma.

Lin X, Qian GS, Lu PX, Wu L, Wen YM.

Department of Molecular Virology, Shanghai Medical University, Shanghai, China

Med Virol 2001 Jul;64(3):299-304 Abstract quote

In hepatitis B virus (HBV)-endemic countries, the majority of hepatocellular carcinoma (HCC) arises in HBV carriers. High frequency of mutations at nucleotides 1762(A-->T) and 1764(G-->A) in the core promoter region have been described in HCC. Due to the differences in genetic backgrounds, environmental risk factors and random cellular insertion sites, it is difficult to analyze the possible roles of HBV variants detected in different HCC patients.

In a follow-up cohort study, an HBsAg-positive asymptomatic carrier was diagnosed HCC within 4 years. Eleven full-length HBV isolates, three from the first serum sample obtained 4 years pre-HCC, and eight from serum sample, peri-tumor and tumor tissue post-HCC of this individual were sequenced and used to transfect HepG2 cells. When sequences were compared between pre- and post-HCC isolates, no single mutation common to all post-HCC isolates that differed from pre-HCC isolates was found.

Among all 11 isolates, there were 20 predicted amino acid substitutions shared by two or more post-HCC isolates. These were located in the S(5), X(4), core(4), polymerase(4), pre-S1(2) and pre-S2(1) proteins. Possible roles of amino acid substitutions and enhanced replication efficiency in cells transfected by post-HCC isolates are discussed.


Allelic alterations in nontumorous liver tissues and corresponding hepatocellular carcinomas from chinese patients.

Ho MK, Lee JM, Chan CK, Ng IO.

Hum Pathol. 2003 Jul;34(7):699-705. Abstract quote

Allelic imbalance may play an important in tumor progression in hepatocarcinogenesis, but the genetic background of the corresponding nontumorous liver in hepatocellular carcinoma (HCC) is not well defined.

We studied the incidence of loss of heterozygosity (LOH) and microsatellite instability (MSI) by microsatellite analysis in both nontumorous livers and the corresponding tumors, by comparing them with the normal DNA from Chinese patients with resected primary HCCs. We also evaluated the pathologic significance of the alterations. We used 18 highly polymorphic microsatellite markers on chromosomes 1, 3, 4, 7, 8, 9, 13, 16, 17, and 18.

Our results showed that 70.6% (24 of 34) of the HCCs exhibited LOH at 1 or more loci, and that the overall fractional allelic loss (FAL) was 0.169. MSI was observed in only 1 tumor. In contrast, the nontumorous livers of the HCCs showed a very low incidence of LOH, with only a single LOH detected in 1 of 34 (2.9%) of the nontumorous livers, with an overall FAL index of 0.005.

Tumors with LOH at 1 or more loci had significantly more frequent venous invasion (P = 0.019). Allelic loss at locus D9S199 (9p23) was seen more frequently in larger tumors (P = 0.031), and, less significantly, allelic loss at locus D16S516 (16q24.1) was seen more frequently in larger tumors (P = 0.059). LOH was common in predominantly hepatitis B virus-associated HCCs from Chinese patients. However, LOH or MSI in the corresponding cirrhotic or noncirrhotic livers was uncommon.

Genomic instability in chronic viral hepatitis and hepatocellular carcinoma

Maria Pina Dore, MD Giuseppe Realdi, MD Daniela Mura, MD Angela Onida, MD Giovanni Massarelli, MD Giuseppe Dettori, MD David Yates Graham, MD Antonia Rogado Sepulveda, MD, PhD

Hum Pathol 2001;32:698-703. Abstract quote

Chronic hepatitis may progress to cirrhosis and hepatocellular carcinoma (HCC). Progressive accumulation of mutations and genomic instability in chronic viral hepatitis might flag an increased risk of HCC development.

Genomic instability at dinucleotide microsatellite loci in chromosomes 2, 13, and 17 and at 2 mononucleotide repeat loci was examined in liver tissues from 41 patients, including 30 without HCC (18 patients with chronic hepatitis and 12 with cirrhosis) and 11 with HCC. Genomic instability was detected in 51% of the 41 cases. Allelic imbalance at informative dinucleotide loci occurred in 37% of the cases. In 14 cases (34%), allelic imbalance was detected in chronic hepatitis or cirrhosis without HCC. Allelic imbalance at the chromosome 13 locus was detected in 50% of the cases of chronic hepatitis C. Allelic imbalance at the TP53 chromosome locus and/or at the chromosome 13 locus was significantly more frequent than alterations at the chromosome 2 locus (P = .026). Low-level microsatellite instability was found in 20% of all cases examined and high-level microsatellite instability in 3 patients (7.5%), including 2 cases of chronic hepatitis and 1 case of cirrhosis.

Our results show that allelic imbalance occurs frequently in hepatitis-related HCC as well as in chronic hepatitis in patients without HCC. Allelic imbalance at the D13S170 chromosome 13 locus (13q31.2) occurs frequently in chronic hepatitis, suggesting that genomic alterations affecting the long arm of chromosome 13 might be used to monitor the natural progression of chronic hepatitis-associated liver carcinogenesis.

Detection of chromosomal aberrations in well-differentiated hepatocellular carcinoma by bright-field in situ hybridization.

Wilkens L, Bredt M, Flemming A, Mengel M, Klempnauer J, Kreipe H, Flemming P.

Department of Pathology, Medizinische Hochschule, Hannover, Germany.

Mod Pathol 2002 Apr;15(4):470-5 Abstract quote

Differentiation between well-differentiated hepatocellular carcinoma (HCC) and nonmalignant lesions with increased cellular proliferation may be difficult in needle biopsies. Based on recurrent chromosome aberrations known for HCC, we developed a nonfluorescent in situ hybridization technique that allows combination with morphological analysis in bright-field microscopy.

Fourteen biopsies of HCC and 31 samples of regenerative nodules (n = 10), chronic hepatitis (n = 10), fibrosis or cirrhosis of unknown origin (n = 5), focal nodular hyperplasia (n = 2), primary biliary cirrhosis (n = 2), steatosis (n = 1), and adenomatous hyperplasia (n = 1) were analyzed with probes specific for the centromeric regions of chromosomes 1, 6, 7, and 8. After microwave pretreatment and in situ hybridization, signals were detected using a tyramine-based system and AEC as substrate. Evaluation of signals was done by conventional bright-field microscopy. Using this approach, aberrant counts were seen for at least one chromosome in 12/14 cases of HCC. In contrast, none of the nonmalignant lesions revealed aberrant counts for any of the chromosomes analyzed.

In conclusion, this new combination of in situ hybridization and tyramine amplification allows fast and reliable evaluation of chromosome aberrations in a histomorphological context similar to paraffin immunohistochemistry. Registration of imbalances contributes to a reliable differentiation between malignant and nonmalignant lesions of the liver.

Identification of four distinct regions of allelic imbalances on chromosome 1 by the combined comparative genomic hybridization and microsatellite analysis on hepatocellular carcinoma.

Leung TH, Wong N, Lai PB, Chan A, To KF, Liew CT, Lau WY, Johnson PJ.

Department of Clinical Oncology, Sir Y.K. Pao Centre for Cancer, The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin, Hong Kong.


Mod Pathol 2002 Nov;15(11):1213-20 Abstract quote

Frequent chromosome 1 abnormalities detected in human hepatocellular carcinoma have been implicated in early genetic events of liver carcinogenesis. Recurrent loss of 1p with a common deleted region 1p36-p34 has been reported from microsatellite analysis, whereas common gain of the whole chromosome q-arm was described from several comparative genomic hybridization studies. The relationships between copy number changes and allelic status however remains unclear.

In this study, we have conducted a simultaneous comparative genomic hybridization and microsatellite analysis study on chromosome 1 in 31 hepatocellular carcinoma cases. Microsatellite analysis revealed frequent loss of heterozygosity on 1p at loci D1S468 (74%), D1S450 (67%), D1S2667 (65%), D1S2697 (75%), D1S199 (52%), and D1S234 (67%) corresponded to the distal 1p36 region and coincided with 12 cases (86%) that presented losses on 1p by comparative genomic hybridization analysis. Although comparative genomic hybridization indicated a common deleted region of 1p36-p35 in the current series, microsatellite analysis has refined the smallest overlapping region (SOR) to 1p36.13-p36.22. Gain of 1q as revealed by comparative genomic hybridization suggested low and high-level gains, and cases that displayed an amplicon below the heterochromatic region 1q21-q25.

Common allelic imbalances of polymorphic markers D1S2635 (64%), D1S484 (67%), D1S2878 (65%), D1S196 (70%), D1S249 (64%) D1S2785 (75%), D1S2842 (73%) and D1S2836 (74%) that corresponded to the regions 1q23.1-q24.2, 1q32.1 and 1q43-q44 were detected. Three distinct regions of allelic imbalances were thus suggested on recurring 1q gain found in hepatocellular carcinoma. Furthermore, microsatellite analysis has enabled a mapping of common overrepresented regions and suggested SOR on 1q23.1-q23.3 (D1S2635-D1S2878), 1q25.1-q31.1 (D1S452-D1S238), and 1q43 (D1S2785-D1S2842).

Our current study has refined chromosome 1 aberrations in hepatocellular carcinoma to four regions of allelic imbalances. The SORs delineated should provide basis for further molecular investigation in hepatocarcinogenesis on genes residing on these chromosomal regions.

CIRRHOSIS Stimulation of hepatocellular division with ongoing necrosis and inflammaiton

Occurrence of hepatocellular carcinoma and decompensation in western European patients with cirrhosis type B. The EUROHEP Study Group on Hepatitis B Virus and Cirrhosis.

Fattovich G, Giustina G, Schalm SW, Hadziyannis S, Sanchez-Tapias J, Almasio P, Christensen E, Krogsgaard K, Degos F, Carneiro de Moura M, et al.

Istituto di Semeiotica e Nefrologia Medica, University of Verona, Italy.

Hepatology 1995 Jan;21(1):77-82 Abstract quote

To examine the morbidity of compensated cirrhosis type B, a cohort of 349 Western European, white patients (86% men; mean age, 44 years) with biopsy-proven cirrhosis was followed up for a mean period of 73 months and was studied for occurrence of hepatocellular carcinoma (HCC) and decompensation.

At entry into the study all patients were tested for hepatitis B e antigen (HBeAg; 34% of patients were HBeAg-positive) and antibody to hepatitis delta virus (anti-HDV; 20% of patients were anti-HDV-positive); 48% of 252 patients tested were hepatitis B virus (HBV)-DNA-positive. During follow-up HCC developed in 32 (9%) of the 349 patients and decompensation was observed in 88 (28%) of 317 tumor-free patients. Five years after diagnosis, the probability of HCC appearance was 6% and the probability of decompensation was 23%. After the first episode of decompensation the probability of survival was 35% at 5 years. Cox's regression analysis identified three variables that independently correlated with HCC: age, serum levels of platelets, and liver firmness on physical examination. HBV (HBeAg or HBV-DNA) and HDV (anti-HDV) markers at presentation had no prognostic value for the development of HCC. In conclusion, a high proportion of patients with HBsAg-positive compensated cirrhosis do not experience worsening of their condition for several years, but once decompensation occurs life expectancy is poor. European, white patients with compensated cirrhosis type B are at consistent risk for HCC.

Prognostic factors for HCC reflect an advanced stage of cirrhosis and support the hypothesis that development of a tumor could be the likely consequence of long-standing hepatic disease.


Expression of Fas and Fas-related molecules in human hepatocellular carcinoma

Sug Hyung Lee, etal.

Hum Pathol 2001;32:250-256.

Many tumor cells, including hepatocellular carcinoma (HCC), express both Fas and its ligand on their surfaces, and it has remained a mystery why such cells do not spontaneously become apoptotic.

In the current study, we analyzed the alterations of Fas structure and the expression of Fas and Fas ligand (FasL) and of Fas pathway inhibitors, including soluble Fas (sFas), Fas-associated phosphatase-1 (FAP-1), and bcl-2, in 50 cases of human HCC.

Monoallelic loss of the Fas gene, as determined by loss of heterozygosity with intragenic polymorphisms, was observed in 5 of the 34 informative cases (15%), but none of the 50 cases showed Fas gene mutation. Expression of Fas and FasL was detected in 44 (88%) and 50 (100%) cases, respectively. sFas messenger RNA, as analyzed by in situ reverse-transcription polymerase chain reaction was expressed in 42 of the 50 cases (84%), and FAP-1 expression was observed in 40 of the 50 cases (80%).

In contrast, none of the 50 cases showed bcl-2 expression.

Our results showed that the majority of the HCCs (88%) coexpressed a death receptor, Fas and its cognate ligand, FasL, but all HCCs showed one or more alterations of the Fas pathway molecules known to inhibit Fas-mediated apoptosis.

These findings suggest that the expression of sFas and FAP-1 and, in part, loss of Fas expression, rather than Fas gene alteration or bcl-2 expression, may be involved in the Fas resistance of HCC in vivo and that these mechanisms may play important roles in the pathogenesis of human HCC.

Association of E-cadherin, matrix metalloproteinases, and tissue inhibitors of metalloproteinases with the progression and metastasis of hepatocellular carcinoma.

Gao ZH, Tretiakova MS, Liu WH, Gong C, Farris PD, Hart J.

1Department of Pathology and Laboratory Medicine, University of Calgary and Calgary Laboratory Services, Calgary, AB, Canada.

Mod Pathol. 2006 Apr;19(4):533-40. Abstract quote  

Molecular markers can provide additional information to traditional histomorphological evaluation for the assessment of tumor progression and predicting the likelihood of invasion and metastasis in various types of malignancies.

We studied the association of E-cadherin, matrix metalloproteinases (MMPs), and tissue inhibitors of metalloproteinase with the progression and metastasis of hepatocellular carcinoma. Tissue microarray including six normal livers, 14 cirrhotic livers, 39 macroregenerative nodules, 16 dysplastic nodules, 22 grade I hepatocellular carcinomas, 43 grade II hepatocellular carcinomas, seven grade III hepatocellular carcinomas, and 10 metastatic hepatocellular carcinomas were stained immunohistochemically with antibodies against MMPs -1, -2, -3, -7, -9, tissue inhibitors of metalloproteinase-1, -2, -3, and E-cadherin. The intensities of staining were scored manually by two pathologists and verified by the Chromavision Automated Cellular Imaging System. Compared with normal liver, cirrhotic liver had significantly lower E-cadherin and tissue inhibitors of metalloproteinase-1 but higher MMP-1 and -7, which suggest a more favorable environment for tumor invasion and metastasis. Grade I and grade II hepatocellular carcinomas demonstrated high E-cadherin and decreased MMP-3 and -9, which may explain the rarity of extrahepatic metastasis in low-grade hepatocellular carcinomas despite the high circulatory volume of the liver.

The histological progression from dysplastic nodule to well-differentiated hepatocellular carcinoma and to less differentiated tumors was associated with a gradual decrease in tissue expression of E-cadherin, tissue inhibitors of metalloproteinase-2 and -3. Metastatic hepatocellular carcinomas showed significantly lower level of tissue inhibitors of metalloproteinase-1, -2, -3 but higher level of MMP-7.

These data suggest that tissue expression of E-cadherin, certain MMPs, and tissue inhibitors of metalloproteinases could be useful markers to predict the progression and metastasis of hepatocellular carcinoma.

Overexpression of metastatic tumor antigen 1 in hepatocellular carcinoma: Relationship to vascular invasion and estrogen receptor-alpha.

Moon WS, Chang K, Tarnawski AS.

Department of Medicine, Veterans Administration Medical Center, Long Beach, California, USA.
Hum Pathol. 2004 Apr;35(4):424-9. Abstract quote  

The morbidity and mortality experienced by cancer patients is mainly due to the invasion and metastasis of the primary tumor.

Recently, a potential metastasis-associated gene and its product, the metastatic tumor antigen 1 (MTA1), were identified; this gene has been found to be overexpressed in a variety of cancers. MTA1 is also known as a potent co-repressor of estrogen receptor element transcription in breast cancer cells. The expression of MTA1 in hepatocellular carcinoma (HCC) and its potential relationship to metastasis and to estrogen receptor alpha (ER-alpha) expression has not been examined, forming the basis for this study. Paraffin sections of 45 HCC specimens, 4 different HCC cell lines, and normal hepatocyte cell line (h NHeps) were immunostained with MTA1 and ER-alpha antibodies. In addition, we examined, by Western blotting, the MTA1 and ER-alpha expression levels in 4 human HCC lines (HepG2 [wild p53], HLE, HLF, and HuH-7 [mutant p53]). MTA1 was overexpressed in HCC cells versus nonmalignant hepatocytes in 31 of 45 HCC specimens (69%). Its expression was predominantly localized to the nucleus or cytoplasm of HCC cells. Nineteen of 20 HCC (95%) specimens with vascular invasion displayed strong MTA1 expression.

Overexpression of MTA1 also significantly correlated with large tumor size. The cytoplasmic and nuclear immunoreactivity for ER-alpha was present in HCC specimens in 46% and 12%, respectively. Expression of MTA1 inversely correlated with the nuclear localization of ER-alpha. There was no marked difference in MTA1 and ER-alpha expression levels between HCC cell line expressing wild-type p53 and cell line with mutated p53 HCC.

In conclusion, these findings indicate that overexpression of MTA1 is associated with HCC growth and vascular invasion. Nuclear translocation of ER-alpha inversely correlated with MTA1 expression, suggesting negative regulatory mechanisms.

Expression of DNA methyltransferases in multistep hepatocarcinogenesis.

Choi MS, Shim YH, Hwa JY, Lee SK, Ro JY, Kim JS, Yu E.

Departments of Pathology, Internal Medicine, and Surgery, University of Ulsan College of Medicine, Asan Medical Center, Seoul, Korea and Department of Biological Sciences and Bio/Molecular Informatics Center, Konkuk University, Seoul, Korea.


Hum Pathol 2003 Jan;34(1):11-7 Abstract quote

Hypermethylation of cell cycle regulators and increased DNA methyltransferase 1 (Dnmt1) mRNA level have been reported in hepatocarcinogenesis. However, the expression of Dnmts has not yet been examined in hepatocellular carcinomas (HCCs).

We examined 13 cases of HCCs in dysplastic nodules (DNs) and 28 cases of advanced HCCs for Dnmt1 and Dnmt3a, and compared the results with those of 9 cases of low-grade DNs, 24 cases of high-grade DNs, and 59 cases of nonneoplastic liver tissues from 59 cases of surgically resected livers by immunohistochemical staining.

Nuclear expression of Dnmt1 was increased significantly in all HCCs in DNs and advanced HCCs compared with those of nonneoplastic livers, low-grade DNs, and high-grade DNs (P <0.05). Nuclear expression of Dnmt3a was not detectable in nonneoplastic liver and low-grade DN, whereas it was observed in high-grade DNs (7 of 24, 29.2%), HCCs in DNs (7 of 13, 53.8%), and advanced HCCs (11 of 28, 39.3%). Different from Dnmt1 immunostaining, cytoplasmic immunoreactivity for Dmnt3a was significantly decreased or absent in 13 of 24 cases of high-grade DNs (54.1%), 12 of 13 cases of HCCs in DNs (92.3%), and 22 of 28 cases of advanced HCCs (78.6%), compared with nonneoplastic livers and low-grade DNs (P <0.05).

Our data suggest that Dnmt1 and Dnmt3a play a role in the early stage of hepatocarcinogenesis and that dysregulation of Dnmt3a may be involved in the progression of HCC. Furthermore, the significantly decreased cytoplasmic immunoreactivity for Dnmt3a in high-grade DNs and HCCs can be used as a diagnostic adjunct.


HGF, MET, and Matrix-Related Proteases in Hepatocellular Carcinoma, Fibrolamellar Variant, Cirrhotic and Normal Liver.

Schoedel KE, Tyner VZ, Kim TH, Michalopoulos GK, Mars WM.

Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania.


Mod Pathol 2003 Jan;16(1):14-21 Abstract quote

Fibrolamellar variant is an uncommon subcategory of hepatocellular carcinoma with a better prognostic outcome. Proteinases and growth factors that are involved in the remodeling of extracellular matrix may influence the behavior of cancers.

To determine whether these factors contribute to the distinct etiologies of fibrolamellar hepatocellular carcinoma and traditional hepatocellular carcinoma, we assayed hepatocyte growth factor, the hepatocyte growth factor receptor, and two hepatocyte growth factor activators, hepatocyte growth factor activator and urokinase-type plasminogen activator, in hepatocellular carcinoma, fibrolamellar hepatocellular carcinoma, cirrhotic liver and normal liver. In addition, we examined the urokinase-type plasminogen activator receptor, the type 1 plasminogen activator inhibitor, plasmin, fibrinogen, and the type IV matrix metalloproteinases.

Eighteen hepatocellular carcinomas and 11 fibrolamellar hepatocellular carcinomas were obtained as paraffin embedded sections from the University of Pittsburgh Department of Pathology. Frozen tissues from a subset of cases (9 hepatocellular carcinomas, 4 fibrolamellar hepatocellular carcinomas, 12 cirrhotic livers and 2 normal livers) were also available for analysis. Antibodies against urokinase-type plasminogen activator, urokinase-type plasminogen activator receptor, hepatocyte growth factor and hepatocyte growth factor receptor were used to analyze immunoperoxidase stained slides from the paraffin blocks. Western blot analyses using antibodies against hepatocyte growth factor, hepatocyte growth factor receptor, phosphotyrosine, hepatocyte growth factor activator, urokinase-type plasminogen activator receptor, urokinase-type plasminogen activator, plasminogen activator inhibitor-1, fibrinogen and plasmin were performed on membrane-enriched fractions from the frozen tissue, as was collagen zymography for matrix metalloproteinase-2 and matrix metalloproteinase-9.

The most notable findings are as follows: hepatocyte growth factor activator was only detected in malignant tissue but not cirrhotic liver or normal liver. Although hepatocyte growth factor was detected in most samples, it was significantly elevated in 5/9 hepatocellular carcinomas. Furthermore, 8/9 fibrolamellar hepatocellular carcinomas demonstrated hepatocyte growth factor receptor levels similar to normal, whereas 8/9 hepatocellular carcinomas and 11/12 cirrhotic livers exhibited either an increase or decrease. In contrast, active matrix metalloproteinase-2, which was absent in normal liver, was elevated in fibrolamellar hepatocellular carcinoma as compared to cirrhotic liver and conventional hepatocellular carcinoma. Surprisingly, 10/12 cirrhotic livers and 2/4 fibrolamellar hepatocellular carcinomas but only 1/9 hepatocellular carcinomas were enriched for plasmin.

The combined data suggest that the hepatocyte growth factor and plasmin systems tend to be operative in hepatocellular carcinoma and cirrhotic liver, more than fibrolamellar hepatocellular carcinoma. Furthermore, matrix turnover appears to be a more prominent feature of fibrolamellar hepatocellular carcinoma.

These findings provide insight into the behavioral differences between hepatocellular carcinoma and the fibrolamellar variant.


High expression of p21 is correlated with human hepatocellular carcinoma in patients with hepatitis C virus-associated chronic liver diseases.

Wagayama H, Shiraki K, Sugimoto K, Ito T, Fujikawa K, Yamanaka T, Takase K, Nakano T.

First Department of Internal Medicine, Mie University School of Medicine, Tsu 514-8507, Japan.

Hum Pathol 2002 Apr;33(4):429-34 Abstract quote

p21(WAF1/CIP1) (p21) protein is a universal inhibitor of cyclin-dependent kinases and is regulated transcriptionally by p53, which is activated by DNA stress. Hepatocytes in chronic hepatitis receive several DNA stresses by lymphocytes and Kupffer cells.

Therefore, we analyzed p21 expression of hepatocytes in hepatitis C virus (HCV)-associated chronic liver diseases and investigated the possible involvement of p21 in hepatocarcinogenesis. We examined p21 expression in 35 cases of HCV-associated chronic hepatitis and 25 cases of HCV-associated liver cirrhosis by immunohistochemical analysis. The p21 labeling index (LI) was calculated as the ratio of positive cells to total cells. p21-positive hepatocytes were more numerous in areas of intense inflammation and spotty necrosis and areas close to fibrosis, and were increased according to the degrees of grading and staging. The p21 LI with liver cirrhosis was significantly higher than that with chronic hepatitis (14.4 +/- 5.9 versus 11.1 +/- 4.2, P = 0.014). The cumulative incidence of hepatocellular carcinoma (HCC) was significantly higher in the p21 LI >/=14% group than in the p21 LI <14% group (P = 0.0079). Multivariate analysis demonstrated that p21 expression can be recognized as an independent significant factor for HCC development (relative risk 5.00, P = 0.039). p21 LI decreased significantly after interferon therapy.

These results suggested that p21 is up-regulated by the stress of inflammation and fibrosis in HCV-associated chronic liver diseases and that high p21 expression might be related to hepatocarcinogenesis in cirrhotic patients.


Nuclear translocation of survivin in hepatocellular carcinoma: a key to cancer cell growth?

Moon WS, Tarnawski AS.

Department of Medicine, Veterans Administration Medical Center, Long Beach, CA 90822, USA.
Hum Pathol. 2003 Nov;34(11):1119-26. Abstract quote  

Survivin is a recently described anti-apoptosis protein and regulator of cell division. Its expression and localization in hepatocellular carcinoma (HCC) and in normal liver tissue has not been fully elucidated.

We examined the expression of survivin, Fas, proliferating cell nuclear antigen (PCNA), and apoptosis in 47 specimens of hepatocellular carcinoma (HCC) and surrounding nonmalignant hepatic tissues. To further determine the relationship between survivin expression and cell proliferation and apoptosis, we performed double immunostaining for survivin and PCNA TUNEL staining in the same HCC specimens. Positive immunostaining for survivin was present in 35 of 47 (74%) HCCs. Twenty-two of 35 survivin-positive HCCs (63%) showed punctate nuclear staining in HCC cells, and the remaining 13 showed predominant cytoplasmic staining. In contrast, nonmalignant hepatocytes showed only cytoplasmic staining. HCC cells had significantly higher PCNA-labeling and apoptotic indices compared with the case of nonmalignant hepatic tissue (P<0.001). Furthermore, nucleus-positive HCC specimens for survivin showed the highest PCNA labeling index. The nuclear localization of survivin in HCC cells correlated with tumor cell de-differentiation with the exception of the HepG2 cell line. Survivin expression was inversely associated with apoptosis and was strongly associated with Fas expression (P=0.01). All 4 HCC cell lines examined showed survivin expression and punctate nuclear localization.

Our results indicate that survivin is localized to the cytoplasm in quiescent nonmalignant liver cells to suppress apoptosis and translocates into the nucleus in HCC cells. In conclusion, translocation of survivin from the cytoplasm to the nucleus may constitute an important regulatory mechanism for cell proliferation and differentiation in HCC.

Expression of Transforming Growth Factor-beta1 and Transforming Growth Factor-beta Receptors in Hepatocellular Carcinoma and Dysplastic Nodules.

Paik SY, Park YN, Kim H, Park C.

Department of Pathology and Brian Korea 21 Project for Medical Science, Yonsei University, College of Medicine, Seoul, South Korea.


Mod Pathol 2003 Jan;16(1):86-96 Abstract quote

In this study we analyzed by immunohistochemistry the expression of TGF-beta1 protein and TGF-beta receptors I and II in 4 low-grade dysplastic nodules, 2 high-grade dysplastic nodules, 6 early, 22 small, and 62 advanced hepatocellular carcinomas.

The expression of TGF-beta1 protein by hepatocytes was decreased in advanced hepatocellular carcinoma compared with small or early hepatocellular carcinoma(P <.05). Frequent and intense staining of TGF-beta1 protein was noted in the sinusoidal endothelium of advanced hepatocellular carcinomas despite of its decreased staining in hepatocellular carcinoma cells. Reduced expression of TGF-beta receptors I and II compared with surrounding nontumorous tissue were noted from the early hepatocellular carcinoma stage suggesting that down-regulation of TGF-beta receptors is correlated with progression from premalignant to malignant phenotype. Reduced expression of both TGF-beta1 and TGF-beta receptor II in neoplastic hepatocytes were also significantly correlated with increased tumor size and increased proliferative activity(P <.05).

These findings suggest that during hepatocarcinogenesis, the inhibitory effects of TGF-beta1 protein on hepatocellular carcinoma cells is outweighed by its effects on stromal elements, which, overall, contributes indirectly to a tumor growth stimulatory environment. Also, the growth-inhibitory effects of TGF-beta1 may have been further negated by reduced TGF-beta receptors on hepatocellular carcinoma cells.

Smad4 overexpression in hepatocellular carcinoma is strongly associated with transforming growth factor beta II receptor immunolabeling.

Torbenson M, Marinopoulos S, Dang DT, Choti M, Ashfaq R, Maitra A, Boitnott J, Wilentz RE.

Departments of Pathology, Internal Medicine, and Surgery, The Johns Hopkins Hospital, Baltimore, MD, and Department of Pathology, University of Texas Southwestern Medical Center, Dallas, TX

Hum Pathol 2002 Sep;33(9):871-6 Abstract quote

In the normal liver, the transforming growth factor beta (TGF-beta) signaling pathway plays an important role in inhibiting hepatocyte growth. This effect is mediated through Smad4 (or Dpc4), a tumor-suppressor gene that affects gene transcription and controls cell growth. A loss of Smad4 is associated with carcinoma in a number of other organs, including the pancreas and colon. Despite these facts, several recent studies using cDNA microarrays have surprisingly shown overexpression of Smad4 in hepatocellular carcinoma (HCC). Because Smad4 plays a central role in the TGF-beta signaling pathway, we hypothesized that activation of the TGF-beta signaling pathway may explain Smad4 overexpression.

To investigate this, 21 surgically resected HCCs were immunostained with antibodies to Smad4 and TGF-beta receptor II. Tumor and normal liver tissues were stained in all cases, and expression in the tumor was scored in comparison to the nonneoplastic liver. Thirteen hepatic adenomas were also immunostained as a control group. The average age at resection was 58 +/- 16 years for the 17 men and 4 women with HCC. TGF-beta receptor II was weakly expressed in the hepatocyte cytoplasm of all normal livers and was overexpressed in 10 of 21 HCCs. Of these 10 HCCs increased Smad4 immunolabeling was also present in 10 of 10 cases.

In contrast, of the 11 of HCCs that did not show TGF-beta overexpression, only 1 showed increased Smad4 immunolabeling. Increased TGF-beta receptor II and Smad4 labeling was associated with a worse nuclear grade and increased mitotic activity. For the hepatic adenomas, the 13 women had an average age at resection of 36 +/- 10 years. Whereas 2 adenomas showed over expression of TGF-beta receptor II, there was no Smad4 overexpression in any case. In conclusion, increased Smad4 protein expression in HCC is tightly linked to overexpression of TGF-beta II receptors and is associated with increased mitoses and a worse nuclear grade. Hepatic adenomas only rarely show overexpression of TGF-beta II receptors and did not show increased Smad4 labeling.

The results from this study indicate that Smad4 protein overexpression is present in a subset of HCCs and is strongly correlated with immunostaining for TGF-beta II receptor, findings that may represent activation or dysregulation of the TGF-beta signaling pathway.


Overexpression of VEGF and angiopoietin 2: a key to high vascularity of hepatocellular carcinoma?

Moon WS, Rhyu KH, Kang MJ, Lee DG, Yu HC, Yeum JH, Koh GY, Tarnawski AS.

Department of Pathology and Institute for Medical Sciences, Chonbuk National University, Medical School, Chonju, Korea.

Mod Pathol. 2003 Jun;16(6):552-7. Abstract quote

Hepatocellular carcinoma (HCC) is becoming one of the most common malignant tumors worldwide and is characterized by a high vascularity. Angiogenesis, formation of new microvessels, is critical for the growth and progression of various human solid tumors. Vascular endothelial growth factor (VEGF) and angiopoietins (Ang1 and Ang2) are endothelial cell-specific vasculogenic and angiogenic growth factors, but their expression and roles in HCC have not been extensively explored.

The aim of this study was to determine the expression and cellular localization of VEGF, Ang1, and Ang2 in specimens of resected human HCC using in situ hybridization and immunohistochemical staining and to examine their relationship to microvessel density (MVD) and tumor size.

We also investigated whether mutation of p53 protein might affect the expression of the above angiogenic growth factors. VEGF and Ang2 were strongly expressed and localized predominantly to cancer cells, whereas Ang1 was detected in supportive cells of large blood vessels, stromal cells, endothelial cells, and tumor cells. Expression of the VEGF protein and the Ang2 (but not Ang1) mRNA were strongly correlated with MVD (P <.05, P =.001) and tumor size (P <.05). There was also a strong correlation between VEGF protein and Ang2 mRNA expression (P <.001). However, no significant correlation was found between overexpression of p53 and the expression of VEGF, angiopoietins, or MVD.

These findings suggest that overproduction of the angiogenic growth factors VEGF and Ang2 by HCC cells may increase vascularity and tumor growth in a paracrine manner. Our findings also suggest that interaction between VEGF and Ang2 may play a critical role in tumor angiogenesis in HCC.

Clonal relationship of tumor nodules in hepatocellular carcinoma: a hierarchical clustering analysis of comparative genomic hybridization data.

Wong QW, Wong N, Lai PB, To KF, Wong N.

Department of Anatomical and Cellular Pathology, The Chinese University of Hong Kong, Shatin, NT, SAR Hong Kong, China.

Hum Pathol. 2005 Aug;36(8):893-8. Abstract quote  

Information on the clonal relationship between tumor nodules in patients diagnosed with hepatocellular carcinoma (HCC) holds prognostic significance in the prediction of recurrence and postoperative treatment.

Here, we investigated the clonal relationships in 11 focal nodules from 5 patients with HCC by comparative genomic hybridization (CGH) and spectral karyotyping. CGH analysis indicated tumor nodules in 5 (100%) of 5 cases to share similar patterns of genomic imbalances, suggesting nodules to be clonally related. This clonal relationship was further substantiated in 2 cases, where spectral karyotyping analysis indicated identical structural rearrangements between focal tumors. Few studies have attempted to differentiate multicentric growth from intrahepatic dissemination in HCC using CGH analysis; however, data interpretation remained subjective. In this study, we have extended our data to include published CGH findings on 57 nodules derived from 19 HCC cases.

Simulation of CGH findings by unsupervised 2-way hierarchical clustering indicated tumor nodules in 20 (83.3%) of 24 cases to cluster concordantly, signifying a high incidence of clonal similarity.

In sum, bioinformatic analysis of genomic profiles may represent a reliable research tool in analyzing clonal relationships among tumor nodules.


The pathologist has the additional difficult task of differentiating several precursor lesions from hepatocellular carcinoma (HCC). These dysplastic nodules were once referred to as adenomatous hyperplasia. The most important message is fully developed malignant change even in a focus of less than 1 mm is diagnostic for HCC in the setting of a cirrhotic liver

Mallory bodies present in 20% of tumors, indistinguishable from alcoholic hyaline
Non-Mallory cytoplasmic globules present in 20%, usually PAS positive and diastase resistant (globules of alpha-1-antitrypsin)

Vascular invasion and herniation by hepatocellular carcinoma in cirrhosis: A wolf in sheep's clothing?

Quaglia A, Etessami N, Sim R, Difford J, Dhillon AP.

Department of Histopathology, Royal Free Hospital, London, United Kingdom.
Arch Pathol Lab Med. 2005 May;129(5):639-44. Abstract quote  

CONTEXT: Vascular invasion is an important diagnostic and prognostic feature of hepatocellular carcinoma (HCC) in cirrhosis. Intravascular free-floating tumor clusters (IvCs) of HCC are found histologically in the vicinity of HCC. Thrombus formation is not seen morphologically in association with these IvCs, which are usually covered by endothelium.

OBJECTIVE: Our hypothesis is that these IvCs are the result of a nondestructive form of vascular invasion by HCC, and we tried to define this aspect of microvascular invasion more accurately.

DESIGN: Tissue sections were stained with hematoxylin-eosin, and consecutive sections were stained for fibrin (Martius scarlet blue, fibrinogen), platelets (factor XIIIa), smooth muscle actin, and endothelium (CD34). We studied cirrhotic livers removed at transplantation between 1997 and 1999. Of the livers studied, 35 of 81 consecutive cirrhotic livers contained HCC, and 17 showed microscopic vascular invasion. Five of these 17 cases showed IvCs and were subjected to the study.

MAIN OUTCOME MEASURE: Presence or absence of thrombus formation in association with IvC.

RESULTS: Usually, IvCs were covered by endothelium, and no associated thrombus formation was seen. In 1 case of HCC, thrombus formation was seen focally in association with disruption of the endothelial coating.

CONCLUSIONS: We propose that the endothelial-lined trabecular structure of HCC everts, frondlike, via vascular structures within the tumor capsule into peritumoral vascular lumens without destruction of the endothelial coating. This may protect these HCC tumor projections from thrombus formation but may also act as a barrier to tumor extravasation, and this may be exploited from a therapeutic point of view.

Primary Liver Carcinoma in Genetic Hemochromatosis Reveals a Broad Histologic Spectrum

Mohib Morcos, MD
Sylvie Dubois, MSc
Marie-Pierre Bralet, MD
Jacques Belghiti, MD
Claude Degott, MD
Benoît Terris, MD

Am J Clin Pathol 2001;116:738-743 Abstract quote

Hepatocellular carcinoma (HCC) is a well-known complication of genetic hemochromatosis (GH). However, the frequency of primary liver carcinoma (PLC) with biliary differentiation, such as cholangiocarcinoma (CC) and combined hepatocholangiocarcinoma (CHCC), in GH remains unclear. We analyzed the histologic type of 20 PLCs occurring in the background of GH; all patients were homozygotic for the C282Y mutation. Ten were depleted of iron by successive phlebotomies, while the remaining 10 were untreated. Histologically, 13 cases were classified as HCC, 3 as CC, and 4 as CHCC. Immunohistochemical detection of Hep Par 1, cytokeratin 19 (CK19), and MUC1 supported this classification; PLC with biliary differentiation was immunoreactive for MUC1 in 86% (6/7) of cases and for CK19 in 100% (7/7) of cases. The nontumoral liver exhibited no cirrhosis or extensive fibrosis in 6 cases. Von Meyenburg complexes were present in 11 cases and intraparenchymal bile duct adenomas in 3.

These data suggest that PLCs in patients with GH present a wide histologic spectrum, with tumors showing frequent biliary differentiation; may arise on a nonfibrotic or a cirrhotic liver; and often are associated with Von Meyenburg complexes and to a lesser extent with bile duct adenomas.

Clear cell
Must distinguish from metatastatic renal cell carcinoma
Combined (HCC with cholangiocarcinoma)  

Combined Hepatocellular-Cholangiocarcinoma: A Histopathologic, Immunohistochemical, and In Situ Hybridization Study

Satish K. Tickoo, M.D.; Sui Y. Zee, M.D.; Sam Obiekwe, M.D.; Hong Xiao, B.S.; Jonathan Koea, M.D.; Christian Robiou, M.D.; Leslie H. Blumgart, M.D., F.A.C.S.; William Jarnagin, M.D.; Marc Ladanyi, M.D.; David S. Klimstra, M.D.

Am J Surg Pathol 2002; 26(8):989-997 Abstract quote

Combined hepatocellular-cholangiocarcinoma (CHC) forms a small but significant proportion of primary liver carcinomas. However, its diagnostic features are not well established, and this has possibly contributed to the variability in its reported clinical outcome in the literature. Many such tumors with features intermediate between hepatocellular carcinoma and cholangiocarcinoma (CC) may have been considered CC in the past based on positivity for "biliary differentiation" cytokeratins and the lack of availability of highly sensitive and specific hepatocellular markers. The utility of in situ hybridization for albumin mRNA, a recently available sensitive and specific hepatocellular marker, has not been reported in CHC.
We investigated 27 CHCs with regard to their histomorphologic spectrum and association of these morphologies with immunohistochemical staining for different cytokeratins (CK7, CK19, and CK20; AE1; Cam 5.2), epithelial membrane antigen, polyclonal carcinoembryonic antigen and alpha-fetoprotein, and in situ hybridization for albumin mRNA. All 27 tumors contained areas morphologically intermediate between hepatocellular carcinoma and CC (transitional-type tumors), and in each case such areas formed at least 25% of the tumor. Nine (33%) tumors showed areas with "antler-like" morphology, a feature not previously described in CHC. Twenty-two of 23 tumors (96%) showed positive signals on in situ hybridization for albumin mRNA. Positivity for both hepatocellular (albumin mRNA) and biliary (keratin immunohistochemical profile) markers confirmed the light microscopic impression of biphenotypic differentiation in these tumors. Immunohistochemical positivity for all cytokeratins (except CK7) and epithelial membrane antigen, as well as the expression of albumin mRNA by in situ hybridization, did not show significant differences between hepatocellular carcinoma and CC-like areas.

Based on the cytokeratin profile and results on polyclonal carcinoembryonic antigen/alpha-fetoprotein alone, many such tumors would be classified as CC. However, the positivity for albumin mRNA by in situ hybridization proves that such an interpretation would not have been accurate. Clinically, CHCs showed many differences from pure hepatocellular carcinoma, including the absence of cirrhosis (0 of 27), rarity of serum hepatitis B or C marker positivity (4 of 27), and normal to only mildly elevated serum alpha-fetoprotein levels (median 187 ng/mL). The tumor followed an aggressive clinical course, with overall 3-and 5-year survival rates of 30% and 18%, and in the resected cases of 38% and 24%, respectively.

Trabeculae grow together compressing sinusoids and forming sheets
Young adults (20-40 yrs)
No association with HBV or cirrhosis risk factors
Large tumor with fibrous bands coursing throughout
Well differentiated polygonal cells rowing in nests and cords separated by parallel lamellae of dense collagen bundles
Epidermal growth factor receptor expression and gene copy number in fibrolamellar hepatocellular carcinoma.

Buckley AF, Burgart LJ, Kakar S.

Department of Pathology, Veteran Affairs and UCSF Medical Center, San Francisco, CA 94143, USA.

Hum Pathol. 2006 Apr;37(4):410-4 Abstract quote.  

Increased expression of epidermal growth factor receptor (EGFR), a transmembrane tyrosine kinase, is associated with tumor progression in many carcinomas. Epidermal growth factor receptor inhibitors have shown promise in treating some of these tumors.

Fibrolamellar hepatocellular carcinoma (FL-HCC) is an aggressive neoplasm that occurs in young patients with no history of cirrhosis. This study examines the expression and gene copy number of EGFR in FL-HCC. Formalin-fixed, paraffin-embedded FL-HCC (n = 13) sections were stained with a monoclonal antibody against EGFR. Fluorescence in situ hybridization analysis was performed using probes against EGFR gene and centromeric region of chromosome 7 (CEP 7). Epidermal growth factor receptor and CEP 7 signals were counted in 50 tumor nuclei per case as well as 300 normal hepatocyte nuclei. The EGFR to CEP 7 signal ratio was calculated for each case. Most (92%, 12/13) of FL-HCC showed strong and diffuse staining with anti-EGFR antibody. Fluorescence in situ hybridization was informative in 11 cases, 10 of which showed extra EGFR gene copy numbers (mean, 3.69; range, 3.13-5.0). Epidermal growth factor receptor was overexpressed in all these cases. The mean number of EGFR signals per cell in FL-HCC was double that of normal hepatocytes (3.69 versus 1.80); the mean EGFR/CEP 7 ratio in tumor cells was 1.05.

In conclusion, EGFR is strongly overexpressed on the cell membrane in nearly all cases of FL-HCC. Similar gains of chromosome 7 are observed, indicating that the extra EGFR gene copies are due to polysomy rather than gene amplification. The strong expression of EGFR in FL-HCC tumors suggests that they may respond to treatment with EGFR antagonists.
Clinicopathologic features and survival in fibrolamellar carcinoma: comparison with conventional hepatocellular carcinoma with and without cirrhosis.

Kakar S, Burgart LJ, Batts KP, Garcia J, Jain D, Ferrell LD.

1Department of Anatomic Pathology, UCSF/VA Medical Center, San Francisco, CA, USA.

Mod Pathol. 2005 Nov;18(11):1417-23 Abstract quote.  

Fibrolamellar carcinoma arises in noncirrhotic livers of young individuals and has been considered to be less aggressive than conventional hepatocellular carcinoma.

This study compares survival and clinicopathologic features of fibrolamellar carcinoma with hepatocellular carcinoma arising in noncirrhotic and cirrhotic livers. Clinical and pathologic features including age, gender, tumor size, stage and survival were recorded in 20 resected cases of fibrolamellar carcinoma.

Survival was compared with resected hepatocellular carcinoma without (n=32) and with cirrhosis (n=30). Proliferative activity was determined by immunohistochemistry for Ki-67. In all, 12 (60%) patients with fibrolamellar carcinoma died during follow-up; the 5-year survival was 45%. Mortality in fibrolamellar carcinoma was higher with metastatic disease at presentation (6/7, 86% vs 5/13, 39%, P=0.06). Age, gender and tumor size did not correlate with survival. The 5-year (45 vs 56%, P=0.4) as well as overall survival (40 vs 56.3%, P=0.3) was similar in fibrolamellar carcinoma and hepatocellular carcinoma without cirrhosis. The 5-year and overall survival in hepatocellular carcinoma with cirrhosis was 27 and 23.3%, respectively, which was not significantly different compared to fibrolamellar carcinoma (P=0.2). Among the cases without metastases at presentation, 5-year survival in fibrolamellar carcinoma (62%) and hepatocellular carcinoma without cirrhosis (57%) was significantly better (P=0.03) than hepatocellular carcinoma with cirrhosis (27%). The mean Ki-67 index was similar in all three groups (P=0.1). In conclusion, fibrolamellar carcinoma is an aggressive neoplasm with 45% 5-year survival and overall mortality of 60%. Nearly half the patients develop lymph node or distant metastasis.

The prognosis of fibrolamellar carcinoma is similar to conventional hepatocellular carcinoma. Among nonmetastatic cases, the prognosis is better in fibrolamellar carcinoma and hepatocellular carcinoma without cirrhosis compared to hepatocellular carcinoma with cirrhosis. The better outcome in fibrolamellar carcinoma appears to be due to the absence of cirrhosis rather than its distinct clinicopathologic features.

Adult-type hepatocellular carcinoma in the center of a fibrolamellar hepatocellular carcinoma.
Seitz G, Zimmermann A, Friess H, Buchler MW.

Department of Visceral and Transplantation Surgery and the Institute of Pathology, University of Berne, Berne, Switzerland.

Hum Pathol 2002 Jul;33(7):765-9 Abstract quote

A large hepatic tumor was detected in the noncirrhotic liver of a 27-year-old female patient. The tumor was radiologically characterized by a peripheral mass encircling a central ovoid tumor, and was resected by an extended right hemihepatectomy. Histologic examination revealed that the peripheral and major component of the tumor represented a fibrolamellar hepatocellular carcinoma, whereas the central, well-demarcated tumor was a less well-differentiated adult-type hepatocellular carcinoma completely encircled by the former. Cells of the peripheral tumor mass abundantly expressed cytokeratin-7, typically present in the fibrolamellar variant, whereas no cytokeratin-7 immunoreactivity was found in the central tumor.

To our knowledge, this is the first reported case of a not admixed but clearly separated evolution of these 2 histologic patterns within the same tumor, and suggests that the 2 types of hepatocellular carcinoma may share a common pathogenic pathway.

Blood filled cystic spaces within the tumor
Centers of trabeculae with dilated canaliculus
Primary Carcinosarcoma of the Liver: Clinicopathologic Features of 5 Cases and a Review of the Literature.

Departments of *Hepatobilliary Surgery †Department of Pathology of Sun Yat-sen University and Department of Pathology of the First Affiliated Hospital of Sun Yat-sen University, Guangdong, China ‡Pathology, Cancer Center and State Key Laboratory of Oncology in South China, Sun Yat-sen University.


Am J Surg Pathol. 2007 Jun;31(6):817-826. Abstract quote

Carcinosarcoma of the liver is very rare worldwide. The terminology and pathogenesis of hepatic carcinosarcoma remain controversial issues. In this article, we studied the clinicopathologic features of 5 cases of hepatic carcinosarcomas (matching the World Health Organization definition), analyzed the clinical data, histologic and immunohistochemical (IHC) results, and discussed the terminology, pathologic differential diagnoses, pathogenesis, and prognosis.

The patients were 40 to 68 years old, and included 4 males and 1 female. All patients were Hepatitis B surface antigen positive with para-tumorous cirrhosis. The largest dimensions of the neoplasms ranged from 6.0 to 14.0 cm. Satellite nodules, portal vein tumor thrombi, direct invasion into local tissues (right diaphragm, right adrenal gland, and gastric wall) as well as metastatic foci in lungs and abdominal lymph nodes were identified.

Pathologically, the neoplasms consisted of carcinomatous and sarcomatous components. The carcinomatous components were exclusively conventional hepatocellular carcinomas in all 5 cases, whereas the sarcomatous components exhibited complex features. Confirmed by IHC studies, the sarcomatous elements in different cases included rhabdomyosarcomas, malignant fibrous histiocytomas, fibrosarcoma, and poorly differentiated spindle cells without distinctive differentiation. Furthermore, the sarcomatous elements in these 5 neoplasms stained negative for all the epithelial markers we applied for IHC staining, which support the pathologic diagnosis of carcinosarcoma rather than sarcomatoid carcinoma. The presence of transitional zones between carcinomatous and sarcomatous components may support the transformation theory. Four patients with palliative hepatectomy died within 6 months, whereas 1 patient is still alive 21 months after radical resection.

The poor prognosis of hepatic carcinosarcoma may be due to their highly invasive and metastatic features. Radical resection of early stage hepatic carcinosarcoma may contribute to a relatively optimistic prognosis.
Occurs in cirrhotic liver with considerable fibrous tissue
Lacks lamellar fibrosis and pleomorphic cytology of fibrolamellar type
Sclerosing hepatic carcinoma
Associated with hypercalcemia
Extensive fibrosis
Cirrhotic and non-cirrhotic livers
Synchronous (Hepatocellular and cholangiocarcinoma)  

Synchronous hepatocellular carcinoma and cholangiocarcinoma arising in two different dysplastic nodules.
Kwon Y, Lee SK, Kim JS, Ro JY, Yu E.Departments of Pathology and Surgery, University of Ulsan College of Medicine, Asan Medical Center, Seoul, Korea.


Mod Pathol 2002 Oct;15(10):1096-101 Abstract quote

We present the first reported case of explant cirrhotic liver that had synchronous cholangiocarcinoma and hepatocellular carcinoma arising in two different high-grade dysplastic nodules. The patient was a 55-year-old woman who had hepatitis B virus-associated liver cirrhosis for 3 years. The moderately differentiated cholangiocarcinoma occurred in high-grade dysplastic nodule with a 1.7-fold cell density compared with that of cirrhotic nodule. The hepatocellular carcinoma arose in a nodule-in-nodule pattern within a peripherally low-grade and centrally high-grade dysplastic nodule and had a 2.7-fold cell density compared with that of cirrhotic nodule. By immunohistochemistry, the tumor cells of the cholangiocarcinoma as well as bile ductular cells in dysplastic nodule were diffusely positive for cytokeratin 7, whereas hepatocellular carcinoma cells and dysplastic hepatocytes were negative for cytokeratin 7. The c-kit-positive hepatic progenitor cells were singly scattered between hepatocytes, and their number was highest in cirrhotic nodule and decreased in dysplastic nodule, whereas they were absent in cholangiocarcinoma and hepatocellular carcinoma arising in dysplastic nodules. Proliferation indices were progressively increased in cirrhotic nodule, dysplastic nodule, and cholangiocarcinoma or hepatocellular carcinoma, sequentially.

These observations indicate that cholangiocarcinoma as well as hepatocellular carcinoma can develop in dysplastic nodule and that hepatic progenitor cells might play a role in the early stage of cholangiocarcinogenesis and hepatocarcinogenesis.

Thickened trabeculae numbering up to 10-20 cells thick


Immunohistochemical Detection of Hepatocellular Carcinoma in the Setting of Ongoing Necrosis after Radiofrequency Ablation

Tomoo Itoh, M.D., Yasuko Orba, C.T., Hidehiro Takei, M.D., Yusuke Ishida, M.D., Makoto Saitoh, M.D., Hideaki Nakamura, M.D., Takashi Meguro, M.D., Shoichi Horita, M.D., Miri Fujita, M.D. and Kazuo Nagashima, M.D.

Laboratory of Molecular and Cellular Pathology (TI, YO, HT, YI, MS, KN), Hokkaido University School of Medicine, Hokkaido; Hokkaido Gastroenterology Hospital (HN, TM, SH), Hokkaido; Department of Pathology, Shin-Nittetsu Muroran General Hospital (MF), Muroran; and Core Research for Evolutional Science and Technology (KN), Tokyo, Japan

Mod Pathol 2002;15:110-115 Abstract quote

After radiofrequency ablation (RFA), hepatocellular carcinoma undergoes complete necrosis and an ongoing necrosis that is irreversible and characterized histologically by disrupted cell outlines, homogenous cytoplasmic eosinophilia, and preserved nuclear staining, with the cells appearing quite distinct from viable cancer cells. Antibody to detect single-stranded DNA (ssDNA) specifically labeled nuclei in the setting of ongoing necrosis, but not viable tumor cells, whereas human mitochondrial antibody labeled the cytoplasm of viable cells but not cells of ongoing necrosis.

The results demonstrate that RFA causes denaturation of both DNA and proteins and that the immunohistochemistry of ssDNA and mitochondrial protein is useful in detection of ongoing necrosis after RFA and provides pathological information on the validity of this procedure.

Comparison of Thyroid Transcription Factor-1 and Hepatocyte Antigen Immunohistochemical Analysis in the Differential Diagnosis of Hepatocellular Carcinoma, Metastatic Adenocarcinoma, Renal Cell Carcinoma, and Adrenal Cortical Carcinoma

Tad J. Wieczorek, MD, Jack L. Pinkus, PhD, Jonathan N. Glickman, MD, PhD, and Geraldine S. Pinkus, MD

Am J Clin Pathol 2002;118:911-921 Abstract quote

We compared the effectiveness of thyroid transcription factor-1 (TTF-1, cytoplasmic reactivity) and hepatocyte antigen (HPA) as markers for characterization of hepatocellular carcinoma (HCC) and as discriminators to distinguish HCC from its histologic and cytologic mimics. Formalin-fixed, paraffin-embedded sections of 258 specimens, including 76 HCCs, 85 metastatic adenocarcinomas, 75 renal cell carcinomas (RCCs), and 22 adrenal cortical carcinomas (ACCs), were evaluated. Specimens included tissue sections and cytologic material (cell blocks). Following heat-induced epitope retrieval, immunohistochemical studies were performed using an indirect immunoperoxidase technique. Cytoplasmic reactivity for TTF-1 was noted for 54 (71%) of 76 HCCs, 3 (4%) of 85 adenocarcinomas, none of 72 RCCs, and none of 22 ACCs. Cytoplasmic reactivity for HPA was observed for 50 (66%) of 76 HCCs, 1 (1%) of 83 adenocarcinomas, none of 74 RCCs, and none of 21 ACCs. Cytoplasmic reactivity for TTF-1 and HPA is highly specific for HCC, although a minority of HCCs, particularly poorly differentiated tumors, may be nonreactive.

Thus, these markers are useful for the characterization of HCC in tissue sections and cell blocks and are highly effective for distinguishing these tumors from other neoplasms included in the differential diagnosis.

Comparative immunohistochemical profile of hepatocellular carcinoma, cholangiocarcinoma, and metastatic adenocarcinoma.

Lau SK, Prakash S, Geller SA, Alsabeh R.

Department of Pathology and Laboratory Medicine, Cedars-Sinai Medical Center, Los Angeles, CA.

Hum Pathol 2002;33:1175-1181 Abstract quote

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.


Canalicular Immunostaining of Neprilysin (CD10) as a Diagnostic Marker for Hepatocellular Carcinomas Nicole Borscheri, M.D.; Albert Roessner, M.D., Ph.D.; Christoph Röcken, M.D., Ph.D.

From the Department of Pathology, Otto-von-Guericke-University, Magdeburg, Germany.

Am J Surg Pathol 2001;25:1297-1303 Abstract quoteNeprilysin (CD10) is expressed in both normal and neoplastic liver tissue, where it exhibits a characteristic canalicular pattern (CD10 can ) similar to the one observed when antibodies cross-react with biliary glycoprotein I (p-CEA). The aim of this retrospective study was to investigate the use of CD10 can in differentiating between hepatocellular carcinomas (HCCs; 63 specimens) and nonhepatocellular carcinomas (non-HCCs) metastatic to the liver (non-HCC; 25 specimens). Immunostaining was performed with antibodies directed against CD10, p-CEA, and -fetoprotein (AFP). Albumin mRNA was detected by nonradioactive in situ hybridization (ISH albumin ). In the HCC group a canalicular staining pattern for CD10 was found in 43 (68.3%) specimens, AFP was found in 15 (23.8%) specimens, and a canalicular staining pattern for p-CEA was present in 60 (95.2%) specimens. ISH albumin was performed in 35 HCC specimens and showed labeling of tumor cells in 30 (85.7%) specimens. In the non-HCC group, CD10 can , and p-CEA, immunostaining for AFP and labeling for ISH albumin were confined to non-neoplastic liver tissue. Sensitivity and specificity were, respectively, 68.3% and 100% for CD10 can , 23.8% and 100% for AFP, 95.2% and 100% for canalicular p-CEA, and 86.4% and 100% for ISH albumin .

Our results demonstrate that canalicular staining for CD10 is a highly specific marker of hepatocytic differentiation. Although it does not differentiate between benign and malignant lesions, CD10 can is clearly useful in differentiating between HCC and non-HCC.

CD117 (c-kit)  
Occurrence of c-kit+ Tumor Cells in Hepatitis B Virus–Associated Hepatocellular Carcinoma

Eung Seok Lee, MD, etal.
Am J Clin Pathol 2005;124:31-36 Abstract quote

Progenitor cells, termed oval cells, are involved in the pathogenesis of hepatocellular carcinoma (HCC) in animal models. By immunolabeling for c-kit and CD34 in human hepatitis B virus–associated cirrhosis with HCC (50 cases) and those with cirrhosis alone (10 cases), we found c-kit+ tumor cells in tumor tissue in 40 of 50 HCCs. The proportion was less than 0.1% of total tumor cell volume in most HCCs. Immunostaining for c-kit also was detected in sinusoidal endothelial cells in 43 of 50 HCCs.

The incidence of oval cell occurrence in the adjacent nonneoplastic tissue in cases of HCC was high (44/50). The occurrence of oval cells, c-kit+ tumor cells, and c-kit+ sinusoidal cells in cases of human hepatitis B virus–associated HCC suggests that oval cell proliferation might be associated with the development of human hepatitis B virus–associated HCC.

Furthermore, the c-kit+ sinusoidal cells might have a role in angiogenesis and progression of human hepatitis B virus–associated HCC.
Novel endothelial cell markers in hepatocellular carcinoma.

Chen X, Higgins J, Cheung ST, Li R, Mason V, Montgomery K, Fan ST, Rijn Mv M, So S.

1Department of Biopharmaceutical Sciences, University of California San Francisco, San Francisco, CA, USA.
Mod Pathol. 2004 Oct;17(10):1198-21 Abstract quote  

Hepatocellular carcinoma is characterized by hypervascularity and a propensity for vascular invasion. Detailed analysis of complementary DNA (cDNA) microarray global gene expression data and further validation on a smaller independent sample set by reverse transcription-polymerase chain reaction established the presence of two endothelial gene clusters in hepatocellular carcinoma.

Cluster I, consists of 20 cDNA clones, representing 15 unique genes. Cluster II consists of nine unique genes. The expression of the cluster I genes appeared to be significantly upregulated in hepatocellular carcinoma compared with normal liver, cirrhotic liver, or nontumor liver tissues adjacent to the hepatocellular carcinoma. The pattern of gene expression of cluster I genes correlated positively with the 'proliferation gene cluster' and 'stromal cells cluster 2'. Expression of cluster II genes, in contrast, was not significantly different between hepatocellular carcinoma and non-neoplastic liver tissues. Studies conducted to localize the protein products of these genes by immunohistochemical staining of tissue arrays with up to 350 cores of tissues, and by in situ hybridization led to the discovery of novel sinusoidal endothelial cell markers in hepatocellular carcinoma: podocalyxin-like and regulator of G protein signaling-5.

Our results underscore fundamental differences not only between neoplastic vs non-neoplastic liver cells but also between the hepatic sinusoidal endothelium of hepatocellular carcinoma and normal liver.
Immunoreactivity of Hep Par 1 in Hepatic and Extrahepatic Tumors and Its Correlation With Albumin In Situ Hybridization in Hepatocellular Carcinoma

Sanjay Kakar, MD, Trudie Muir, MD, Linda M. Murphy, HT(ASCP), Ricardo V. Lloyd, MD, PhD, and Lawrence J. Burgart, MD

Am J Clin Pathol 2003;119:361-366 Abstract quote

We evaluated the expression of Hep Par 1 (hepatocyte paraffin 1 monoclonal antibody) in 42 hepatocellular carcinomas (HCCs), 25 cholangiocarcinomas, 18 tumors metastatic to the liver, and 87 primary extrahepatic tumors. Albumin in situ hybridization (ISH) was performed in the HCC cases. Of 42 cases of HCC, 39 (93%) were positive for Hep Par 1. All cases of cholangiocarcinoma, renal cell carcinoma, adrenocortical carcinoma, and islet cell tumors were negative; 1 case each of primary urinary bladder (n = 10) and pancreatic (n = 10) adenocarcinoma and 3 of 11 cases of primary pulmonary adenocarcinoma showed focal positivity; 7 of 10 gastric and 6 of 8 esophageal adenocarcinomas were strongly positive. Albumin ISH was positive in 39 (93%) HCC cases. All cases of HCC were positive for Hep Par 1 or albumin ISH. Hep Par 1 immunoreactivity has high sensitivity in the diagnosis of HCC. Strong positive staining also occurs in gastroesophageal adenocarcinomas. Cholangiocarcinoma and carcinomas from most other sites are negative for Hep Par 1.

Hep Par 1 immunoreactivity shows high correlation with albumin ISH; their combined use for diagnosis of HCC had a sensitivity of 100% in this population.

Hep par 1 antibody stain for the differential diagnosis of hepatocellular carcinoma: 676 tumors tested using tissue microarrays and conventional tissue sections.
Fan Z, Van De Rijn M, Montgomery K, Rouse RV.Deparent of Pathology, Stanford University Medical Center, Stanford, California.


Mod Pathol 2003 Feb;16(2):137-44 Abstract quote

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.

Hepatocyte Antigen as a Marker of Hepatocellular Carcinoma: An Immunohistochemical Comparison to Carcinoembryonic Antigen, CD10, and Alpha-Fetoprotein

Peiguo G. Chu, M.D., Ph.D.; Shin Ishizawa, M.D.; Emerald Wu, B.S., HT (ASCP); Lawrence M. Weiss, M.D.

Am J Surg Pathol 2002; 26(8):978-988 Abstract quote

Hepatocyte monoclonal antibody (Hep) (alternatively Hep Par 1 for Hep paraffin 1) has been reported to stain normal hepatic tissue and hepatocellular carcinoma (HCC) with high specificity. We have studied the Hepatocyte expression in 96 cases of HCC and 311 cases of nonhepatic epithelial tumors. All cases of HCC were also stained with CEA-Gold 5, CD10, and alpha-fetoprotein. Hep, CEA-Gold 5, CD10, and alpha-fetoprotein immunostains were performed on formalin-fixed, paraffin-embedded tissue sections. Hep immunoreactivity was detected in 88 of 96 cases of HCC (92%), with a cytoplasmic and granular pattern of staining. The level of Hep expression in HCC corresponded to the nuclear grade and growth pattern. All 50 cases of nuclear grade 1 and nuclear grade 2 HCC were positive (100%), whereas 37 of 44 nuclear grade 3 (84%) and 1 of 2 nuclear grade 4 (50%) were positive. Sixty-seven of 68 cases of HCC with a trabecular, pseudoglandular, or scirrhous growth pattern were positive (98%), whereas 22 of 27 cases of HCC with a compact growth pattern were positive (81%). CEA-Gold 5, CD10, and alpha-fetoprotein immunoreactivity was detected in 76% (73 of 96), 52% (50 of 96), and 31% (30 of 96) cases of HCC, respectively. The positive predictive value of the combination of all four antibodies was 97%. Three cases of HCC were negative for all four antibodies; these cases had a high nuclear grade or a sarcomatoid or compact growth pattern. Twenty of 311 cases of nonhepatic tumors were positive for Hep (6%): 15 were adenocarcinomas and five were neuroendocrine tumors. The negative predictive value of Hep in HCC was 94%. The Hep-positive nonhepatic epithelial tumors were easily distinguished from HCC by the expression of keratin 7 or keratin 20 for adenocarcinoma and chromogranin and synaptophysin for neuroendocrine tumors because HCC does usually not express these markers.

With the exception of two cases of hepatoid gastric carcinoma, all Hep-positive nonhepatic epithelial tumors were negative for alpha-fetoprotein, CEA-Gold 5, and CD10. Our study demonstrates that Hep is a relatively specific marker for HCC. It is useful in differentiating HCC from primary hepatic cholangiocarcinoma and metastatic tumors when combined with other immunomarkers.

A Comparison of CD10 to pCEA, MOC-31, and Hepatocyte for the Distinction of Malignant Tumors in the Liver

Carl Morrison, M.D., D.V.M., William Marsh, Jr, M.D. and Wendy L. Frankel, M.D.
Department of Pathology, The Ohio State University, College of Medicine, Columbus, Ohio

Modern Pathology 2002;15:1279-1287 Abstract quote

The distinction of hepatocellular carcinoma (HCC) from metastatic adenocarcinoma (MA) and cholangiocarcinoma (CC) in some cases requires the use of immunohistochemistry. CD10 has recently been suggested as a useful stain for HCC. We directly compared CD10 with other immunohistochemical markers, Hepatocyte, pCEA, and MOC31, that have previously shown to be useful for the distinction between tumors in the liver to help define the current panel of stains that most readily distinguishes HCC from CC and MA. One hundred previously well-characterized tumors in the liver were evaluated and included 25 HCC, 15 CC, and 60 MAs (15 each from breast, esophageal/gastric, pancreatic, and colorectal origin). Tumors were immunostained with the commercially available antibodies Hepatocyte, pCEA, MOC31, and CD10. CD10 stained 13 of 25 HCC and was rarely positive in MA and CC (3/75). Hepatocyte stained 24 of 25 HCC and was negative in all 75 MA and CC. pCEA stained 24 of 25 HCC and 71 of 75 MA and CC with the proper pattern of immunoreactivity, but the pattern of staining was difficult to interpret in several cases. MOC31 stained 1 of 25 HCC and 65 of 75 MA and CC. Hepatocyte was the most sensitive and specific single marker for HCC. CD10 is not a useful addition or substitution to the panel of Hepatocyte, MOC31, and pCEA.

The combination of Hepatocyte, MOC31, and pCEA correctly classified 99 of 100 tumors in this study and is our proposed panel of immunostains for the initial workup of malignant tumors in the liver.


Loss of p53 transcriptional activity in hepatocellular carcinoma evaluated by yeast-based functional assay: Comparison with p53 immunohistochemistry.

Mitsumoto Y, Nakajima T, Marutani M, Kashiwazaki H, Moriguchi M, Kimura H, Okanoue T, Kagawa K, Tada M.

Hum Pathol 2004;35:350-356 Abstract quote

We studied the transcriptional activity of p53 protein in 50 tissues of hepatocellular carcinoma (HCC) using a yeast functional assay. In this assay, red yeast colonies indicate that p53 protein cannot bind to its specific domain and has lost its transcriptional activity.

We also clarified whether mutant p53 protein could inactivate wild-type p53 protein in a transdominant manner using a modified yeast assay. In addition, we examined whether immunohistochemically detectable p53 protein was functionally inactive. The incidence of p53 inactivation was significantly higher in tumors with capsular invasion. Out of 21 tumors diagnosed with p53 mutations, 11 exhibited >75% red colonies, and all contained missense mutations. In these tumors, p53 function was lost because there was supposedly no intact p53 gene on either allele. One missense mutant produced <60% red colonies, but it was also considered inactive as a p53 protein heterotetramer because of its transdominant activity. In 7 of the remaining 9 tumors, p53 was considered to be mutated on one allele and intact on the other. All of these 7 tumors contained nonsense or frameshift mutations and had no transdominant activity, which suggested that p53 function remained intact. Alternately, immunohistochemical analysis demonstrated that all of the tumors with missense mutations were positively immunostained, whereas those that contained nonsense or frameshift mutations were negatively stained. Consequently, positively immunostaining tumors mostly coincided with p53-inactive tumors.

These yeast-based assays suggested that p53 function was retained in some mutant cases. Immunohistochemistry was helpful in screening functionally inactive p53 protein in HCCs.
Cytoplasmic Staining of TTF-1 in the Differential Diagnosis of Hepatocellular Carcinoma vs Cholangiocarcinoma and Metastatic Carcinoma of the Liver

Jun-Yi Lei, MD, PhD, Patricia A. Bourne, P. Anthony diSant'Agnese, MD, and Jiaoti Huang, MD, PhD
Am J Clin Pathol 2006;125:519-525
Abstract quote

The cytoplasmic staining of thyroid transcription factor (TTF)-1 was analyzed in 86 liver resection specimens, including 40 hepatocellular carcinoma (HCC), 4 metastatic HCC, 20 cholangiocarcinoma, 2 combined hepatocellular-cholangiocarcinoma (CHC), and 20 metastatic carcinoma (MC), with immuno-histochemical stains to TTF-1, cytokeratin (CK)19, hepatocyte paraffin 1, a-fetoprotein, polyclonal carcinoembryonic antigen, CK7, and CK20. TTF-1 cytoplasmic staining was identified in 93% of HCCs (37/40), 100% of metastatic HCCs (4/4), 10% of cholangiocarcinomas (2/20), and 5% of MCs (1/20). CK19 was positive in all cholangiocarcinomas and MCs but only in 5% of HCCs (2/40) and none of the metastatic HCCs (0/4). TTF-1 cytoplasmic staining positively correlates with differentiation and the trabecular growth pattern of HCC.

The results suggest TTF-1 cytoplasmic staining, together with CK19, might serve as a useful marker for the diagnosis of primary and metastatic HCC and for the differential diagnosis of HCC from cholangiocarcinoma and MC. The mechanism of TTF-1 cytoplasmic staining is explored.
Cytoplasmic Immunoreactivity for Thyroid Transcription Factor-1 in Hepatocellular Carcinoma A Comparative Immunohistochemical Analysis of Four Commercial Antibodies Using a Tissue Array Technique

Chin-Chen Pan, MD, Paul Chih-Hsueh Chen, MD, PhD, Shyh-Haw Tsay, MD, and Hung Chiang, MD
Am J Clin Pathol 2004;121:343-349 Abstract quote

To evaluate the consistency of cytoplasmic immunoreactivity in hepatocellular carcinoma (HCC), we performed immunohistochemical stains for 4 commercial anti–thyroid transcription factor (TTF)-1 antibodies on 77 HCCs and 334 nonhepatic epithelial tumors. The HCC cases were submitted for hepatocyte antigen immunohistochemical stain.

Heat-induced epitope retrieval (HIER) methods were used: with Target Retrieval Solution, the positive rates of cytoplasmic TTF-1 in HCC for DAKO , Zymed , Santa Cruz , and Novocastra antibodies were 58% (45), 14% (11), 6% (5), and 0% (0), respectively; with EDTA buffer, the positive rates increased to 70% (54), 40% (31), 69% (53), and 0% (0), respectively. Immunoreactivity for the DAKO anti–TTF-1 antibody generally correlated with that for hepatocyte antigen. Among nonhepatic tumors, 2 of 6 ovarian mucinous carcinomas and 2 of 11 pancreatic adenocarcinomas showed cytoplasmic reactivity for the DAKO antibody; 28 cases showed nonspecific cytoplasmic staining for the Santa Cruz antibody with EDTA HIER. Zymed and Novocastra antibodies did not produce cytoplasmic staining in nonhepatic tumors.

Owing to the staining variation, we do not consider TTF-1 a reliable marker to distinguish HCC. In general, the Novocastra antibody with EDTA HIER is superior for its consistent nuclear positivity and absence of erratic cytoplasmic staining.


Distinguishing Nodules in a Cirrhotic Liver  
Low-grade dysplastic nodule
High-grade dysplastic nodule
Large cell change
Can be present
Can be present
Nuclear density>2x normal (small cell change)
No small cell change
Occasional small foci in borderline lesions
Common, large foci
Cell plates, zone >/= 3 cell thick (trabecular)
Rare in borderline lesions, no zones of trabeculae
Common pattern
Decrease or loss of reticulin
Reticulin intact
May see focal decrease or loss
Common extensive
Fibrous septa separating thick plates
Occasional pattern
Irregular, infiltrative edges
Not present
Sometimes present
Not present
Occasional pattern
Presence of portal zones
Usually present
Usually present
Increased iron stores
Almost always absent
Fatty change
Can be present
Can be present
Can be present
Bile production
Can be present
Can be present
Can be present
Mallory Hyaline
Can be present
Can be present
Can be present in clusters
Identification of a unique gene expression signature that differentiates hepatocellular adenoma from well-differentiated hepatocellular carcinoma.

Chen ZM, Crone KG, Watson MA, Pfeifer JD, Wang HL.

From Lauren V. Ackerman Laboratory of Surgical Pathology and Division of Laboratory Medicine, Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, MO.

Am J Surg Pathol. 2005 Dec;29(12):1600-8. Abstract quote  

It is often difficult to distinguish hepatocellular adenoma (HCA) from well-differentiated hepatocellular carcinoma (WDHCC) when limited tissue from a needle biopsy is evaluated.

The aim of this study was to identify gene expression patterns that can distinguish HCA from WDHCC, with the ultimate goal of discovering novel diagnostic markers. Gene expression profile analysis was performed using Affymetrix U133Plus2 GeneChip microarrays on RNA isolated from frozen tissue of 6 HCA and 8 WDHCC specimens. Statistical analysis of microarray data identified 63 genes whose expression levels were significantly different between HCA and WDHCC. These included 57 genes overexpressed by HCA and 6 overexpressed by WDHCC. Eight genes were chosen for further analysis by quantitative RT-PCR on RNA derived from archived, paraffin-embedded tissue blocks of an independent validation set comprising 9 HCAs and 9 HCCs. Seven of the 8 genes demonstrated average expression differences between HCA and HCC that were concordant with the microarray findings, and their expression pattern correctly classified the 18 tumors into HCA and HCC using unsupervised clustering analysis. Furthermore, immunohistochemical staining performed on a third, independent set of 27 HCAs and 33 HCCs confirmed the expression differences at protein levels for 5 of the genes.

Taken together, our data demonstrate significant molecular differences between HCA and WDHCC, despite their morphologic similarity. More importantly, we have identified a unique set of genes whose expression pattern can discriminate between these two types of hepatocellular neoplasms, suggesting the possibility of future development of ancillary molecular and immunohistochemical diagnostic methods.
Distinguishing well-differentiated hepatocellular carcinoma from benign liver by the physical features of fine-needle aspirates.

Yang GC, Yang GY, Tao LC.

Department of Pathology, New York University School of Medicine, New York, NY 10016, USA.
Mod Pathol. 2004 Jul;17(7):798-802. Abstract quote  

Distinguishing well-differentiated hepatocellular carcinoma (HCC) from benign hepatic lesions is challenging for pathologists in limited diagnostic material such as needle-core tissue biopsy and fine-needle aspiration (FNA) biopsy.

The objective of this study is to test a hypothesis that the fortification of liver by reticulin along single cell plates should protect benign hepatic lesions from breakdown by the force of aspiration and smearing, whereas the decreased reticulin in well-differentiated HCC would result in finely granular FNA smear.

The study involved FNA biopsies of 67 cases of well-differentiated HCC and 109 cases of benign hepatic lesions, including cirrhosis (22), liver cell adenoma (8), steatosis (7), focal nodular hyperplasia (6), liver with cholestasis (6), and unremarkable liver sampled from nodular hepatic lesions consistent with the regenerative nodules (60). A slide with the most sample from each case by gross inspection was mixed together. Two observers blinded to the diagnoses were asked to separate the slides into two groups based on smear characteristics by gross inspection. Fragments of rigid fine-needle cores was present in 109 out of 109 cases of benign hepatic lesions but absent in 61 out of 67 cases of well-differentiated HCC, which presented as finely granular smears. The difference is statistically significant. (P<0.001, df=1, chi2=149.3).

Using the physical characteristic of liver aspirates as the screening test for malignancy, the sensitivity is 91%, specificity is 100%, positive predictive value is 100%, negative predictive value is 94.8%, and efficiency is 96.6%.

In conclusion, the smear characteristics of liver samples in FNA biopsy correlate to their reticulin status on histology. This physical characteristic can be used as the first clue to distinguish malignant and benign liver aspirates prior to microscopic examination.

Hepatoid adenocarcinoma with liver metastasis mimicking hepatocellular carcinoma: an immunohistochemical and molecular study of eight cases.

Terracciano LM, Glatz K, Mhawech P, Vasei M, Lehmann FS, Vecchione R, Tornillo L.
Am J Surg Pathol. 2003 Oct;27(10):1302-12. Abstract quote  

SUMMARY: Hepatoid adenocarcinoma (HAC) is a special type of extrahepatic adenocarcinoma, which has a striking morphologic similarity to hepatocellular carcinoma. Seven HACs arising in the stomach and one in the lung, all with liver metastasis, were studied. They shared clinical features, such as old age, high serum alpha-fetoprotein level, aggressive behavior, and hepatic tumor in absence of risk factors for hepatocellular carcinoma (HCC).

Morphologically, tumors were characterized by an admixture of tubulo-and/or papillary adenocarcinoma with hepatoid foci. In six cases, liver metastases showed an exclusive hepatoid differentiation, virtually indistinguishable from HCC with solid growth pattern. As HAC and HCC cannot be differentiated on the basis of morphology alone, differences in immunohistochemical reaction patterns would be of considerable diagnostic help.

Immunostaining for CK7, CK8, CK18, CK19, CK20, alpha-fetoprotein, p-CEA, and HepPar1 revealed that hepatoid areas of both primary and metastatic HAC have a specific immunoprofile, distinctive of this entity. On the one hand, positivity of virtually all HACs for alpha-fetoprotein, CK8, CK18, and the membranous, canalicular staining for polyclonal carcinoembryonic antigen underline its hepatoid nature. On the other hand, positive staining for CK19 and CK20 and frequent negativity for HepPar1 in both primary tumors and their metastases were distinctive features of HAC. Furthermore, HAC differs from combined hepatocellular cholangiocarcinoma, being negative for CK7. In addition, for comparison of immunohistochemical results, we stained with the same antibody panel a tissue microarray of 121 HCCs.

Comparative genomic hybridization study of three HAC supports their hepatoid differentiation as aberrations found in HAC are common in HCC (4q-, 8p-), and hepatoblastoma (Xq+), respectively.
Fine-needle aspirates of hepatocellular carcinoma that are misclassified as adenocarcinoma: correlating cytologic features and performance in the College of American Pathologists Nongynecologic Cytology Program.

Renshaw AA, Haja J, Wilbur DC, Miller TR; Cytology Committee College of American Pathologists.

Department of Pathology, Baptist Hospital of Miami, Miami, FL 33156, USA.

Arch Pathol Lab Med. 2006 Jan;130(1):19-22. Abstract quote  

CONTEXT: The cytologic features of hepatocellular carcinoma in fine-needle aspirates are well described. While some cases are easily distinguished from adenocarcinoma, poorly differentiated tumors can be difficult to differentiate. We reviewed the cytologic findings for 9 aspirates from cases of hepatocellular carcinoma that were frequently misclassified as adenocarcinoma and compared them with another 10 cases of hepatocellular carcinoma that were rarely misclassified.

OBJECTIVE: To compare the cytologic features of cases of hepatocellular carcinoma in fine-needle aspirates that were both rarely and frequently misclassified as adenocarcinoma.

DESIGN: We reviewed a total of 762 interpretations from 19 different cases of hepatocellular carcinoma in liver fine-needle aspiration specimens in the College of American Pathologists Nongynecologic Cytology Program and correlated the cytologic features with performance in the program.

RESULTS: Overall, cases that were frequently misclassified as adenocarcinoma were misclassified 39% of the time (range, 18%-70%), while cases that were rarely misclassified were classified as adenocarcinoma 2% of the time (range, 0%-8%). The difference was statistically significant (P < .001). On review, 4 cytologic patterns were found. The most common pattern for cases that were rarely misclassified was prominent trabeculae of cells and endothelial cells wrapping the trabeculae (6/10 cases vs 2/9 cases that were frequently misclassified). The most common pattern among cases that were frequently misclassified was clusters of cells with granular cytoplasm and associated stripped nuclei (5/9 cases vs 2/10 cases that were rarely misclassified). However, the distribution of neither pattern was significantly different (P = .16 for both). One case with large atypical granular cells, as seen in the fibrolamellar variant, was rarely misclassified. The remaining 3 cases (2 frequently misclassified, 1 rarely misclassified) had a nonspecific a pattern of cells with granular cytoplasm without obvious trabeculae or stripped nuclei.

CONCLUSION: Correctly classifying hepatocellular carcinoma by cytology alone remains a significant challenge. While some patterns are more common in cases that performed well and other patterns are more common in cases that performed poorly, there was no significant difference in the distribution of these patterns. These results suggest that people should support their interpretations of aspirations with either immunologic evidence, biopsy evidence, or review by an experienced cytopathologist. Continued educational efforts in this area may be of value.
Fine-needle aspirates of adenocarcinoma/metastatic carcinoma that resemble hepatocellular carcinoma: correlating cytologic features and performance in the College of American Pathologists Nongynecologic Cytology Program.

Renshaw AA, Haja J, Wilbur DC, Miller TR; Cytology Committee, College of American Pathologists.

Department of Pathology, Baptist Hospital of Miami, Miami, FL 33156, USA.
Arch Pathol Lab Med. 2005 Oct;129(10):1217-21. Abstract quote  

CONTEXT: The cytologic features of adenocarcinoma/ metastatic carcinoma in liver fine-needle aspirates are well described. We review the cytologic findings from 16 aspirates of adenocarcinoma/metastatic carcinoma that were frequently misclassified as hepatocellular carcinomas and compare them with 17 cases that were rarely misclassified.

OBJECTIVE: To compare the cytologic features of adenocarcinoma/metastatic carcinoma in fine-needle aspiration specimens of the liver that were frequently misclassified as hepatocellular carcinoma with those of aspirates that were rarely misclassified.

DESIGN: We reviewed a total of 1712 interpretations from 33 different cases of adenocarcinoma/metastatic carcinoma tumor in liver fine-needle aspiration specimens in the College of American Pathologists Nongynecologic Cytology Program and correlated the cytologic features with performance in the program.

RESULTS: Overall, cases that were frequently misclassified as hepatocellular carcinoma were misclassified on average 26% of the time (range, 13%-54%), while infrequently misclassified cases were interpreted as hepatocellular carcinoma on average 0.7% of the time (range, 0%-3%). The difference was statistically significant (P < .001). On review, cases that were frequently misclassified most often had moderate amounts of granular cytoplasm (16/16 cases) and round nuclei with even chromatin (13/16 cases). Trabeculae (3/16 cases), bare nuclei (2/16 cases), and endothelial wrapping (1/16 cases) were also occasionally present. In contrast, cases that were rarely misclassified were more likely to have areas with cells showing scant cytoplasm that were crowded and overlapped or molded (13/17 cases) and contained dark hyperchromatic chromatin (13/17 cases) compared to cases that were frequently misclassified (P < .001 and P = .002, respectively). Trabeculae (2/17) and bare nuclei (2/17 cases) were also rarely present.

CONCLUSION: Cases of adenocarcinoma/metastatic carcinoma with moderate amounts of granular cytoplasm and round nuclei with even chromatin are frequently misclassified as hepatocellular carcinoma. Recognition of this problem, attention to cytologic criteria, and frequent use of immunohistochemical studies and core biopsy may help avoid this pitfall.



Ability to completely resect tumor is associated with best prognosis
In general, dismal prognosis, regardless of other factorsImproved postsurgical survival in patients with:

Lack of vascular invasion
Lack of underlying cirrhosis


Prognostic factors after hepatic resection for hepatocellular carcinoma with hepatitis C viral infection: univariate and multivariate analysis. Hanazaki K, Kajikawa S, Koide N, Adachi W, Amano J.

Second Department of Surgery, Shinshu University School of Medicine, Matsumoto, Japan.

Am J Gastroenterol 2001 Apr;96(4):1243-50 Abstract quoteOBJECTIVES: Although the incidence of hepatocellular carcinoma (HCC) with hepatitis C virus (HCV) infection is higher than in patients with hepatitis B virus (HBV)-related HCC in Japan, the long-term prognosis and prognostic factors of HCV-related HCC after hepatic resection are poorly understood. METHODS: The surgical outcome of HCV-related HCC in 172 consecutive patients who underwent hepatic resection between 1989 and 1997 was retrospectively clarified. Postresection prognostic factors were evaluated by univariate and multivariate analysis using Cox's proportional hazards model. RESULTS: The overall incidence of postoperative complications was 23.2%, and 11 patients among that group had hospital deaths (6.4%) including 9 (5.2%) operative deaths. The mean and median overall survivals including hospital death after surgery were 41 months and 33 months, respectively. The 3-, 5-, and 7-yr overall survival rates after hepatic resection were 63%, 52%, and 47%, respectively. The 3-, 5-, and 7-yr disease-free survival rates after hepatic resection were 33%, 20%, and 15%, respectively. Multivariate analysis revealed that serum alpha-fetoprotein (AFP) of > or = 1000 ng/ml and the presence of vascular invasion were independent unfavorable prognostic factors affecting overall survival and that AFP of > or = 1000 ng/ml was an independently significant factor of poor disease-free survival.

CONCLUSIONS: We found the postresection survival of patients with HCV-related HCC should be stratified by the high value of AFP and the presence of vascular invasion. AFP may be the most powerful predictor of the long-term prognosis and recurrence in such patients.

Prognostic Histologic Indicators of Curatively Resected Hepatocellular Carcinomas A Multi-institutional Analysis of 425 Patients With Definition of a Histologic Prognostic Index Gregory Y. Lauwers, M.D. ; Benoit Terris, M.D. , Ph.D. ; Ulysses J. Balis, M.D. ; Kenneth P. Batts, M.D. ; Jean-Marc Regimbeau, M.D. ; Yuchiao Chang, Ph.D. ; Fiona Graeme-Cook, M.B , B.Ch. ; Hirohiko Yamabe, M.D. ; Iwao Ikai, M.D. ; Karen R. Cleary, M.D. ; Shiro Fujita, M.D. ; Jean-Francois Flejou, M.D. ; Lawrence R. Zukerberg, M.D. ; David M. Nagorney, M.D. ; Jacques Belghiti, M.D. ; Yoshio Yamaoka, M.D. ; Jean-Nicolas Vauthey, M.D.

For the International Cooperative Study Group on Hepatocellular Carcinoma: Departments of: Pathology (G.Y.L., U.J.B., F.G.-C., L.R.Z.), Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts, U.S.A.; Anatomopathologie (B.T., J.-F.F.), Hôpital Beaujon, Paris, France; Pathology (K.P.B.), Mayo Clinic, Rochester, Minnesota, U.S.A.; Surgery (J.-M.R., J.B.), Hôpital Beaujon, Paris, France; Clinical Research Program (Y.C.), Massachusetts General Hospital, Boston, Massachusetts, U.S.A.; Pathology (H.Y.), Kyoto University Graduate School of Medicine, Kyoto, Japan; Surgery (I.I., S.F., Y.Y.), Kyoto University Graduate School of Medicine, Kyoto, Japan; Pathology (K.R.C.), University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.; Surgery (D.M.N.), Mayo Clinic, Rochester, Minnesota, U.S.A.; and Surgical Oncology (J.-N.V.), University of Texas M.D. Anderson Cancer Center, Houston, Texas, U.S.A.

Am J Surg Pathol 2002;26:25-34 Abstract quoteDespite growing information on the clinical behavior of hepatocellular carcinoma, the histologic features associated with survival are not well characterized. Clinical and pathologic data on 425 patients who underwent complete resection for hepatocellular carcinoma were reviewed. Six microscopic features, namely, microvascular invasion, nuclear pleomorphism, mitosis, tumor architecture, growth interface, and tumor necrosis, were examined. Independent predictors of survival were identified and combined into a simple prognostic index. By univariate analysis, microvascular invasion, seen in 51.3% of patients (p <0.001), nuclear grade 3, present in 42% of the cases (p <0.001), and mitosis (p <0.008) were significant predictors of poor survival. Hepatocellular carcinoma with a compact growth pattern had a better prognosis as compared with macrotrabecular (p = 0.014) and acinar (p = 0.051) patterns.By multiple regression analysis, only microvascular invasion (p <0.001) and nuclear grade 3 (p = 0.008) were independent predictors of poor survival. The predictive values of microvascular invasion and nuclear grade allowed the construction of a hepatocellular prognostic index (HPI) whereby HPI = (microvascular invasion status × 0.459) + (nuclear grade × 0.287), with microvascular invasion either absent (0) or present (1) and nuclear grade scored as 1, 2, or 3. Using a cut-off of 0.746 (corresponding to at least nuclear grade 2 with microvascular invasion), two groups could be segregated: fair prognosis (HPI 0.746), with a 50% survival of 5.06 years, and poor prognosis (HPI >0.746) with a 50% survival of 2.71 years (p <0.001). HPI was more discriminating than Edmondson grade, with Edmondson II hepatocellular carcinomas dispersed in both fair and poor prognosis groups. Microvascular invasion and nuclear grade 3 emerge as strong prognostic indicators, and their combination provides adequate prognostic stratification.

Practically, hepatocellular carcinoma can be stratified in two groups with regard to prognosis: 1) fair prognosis group (nuclear grade 1 with or without microvascular invasion and nuclear grade 2 without microvascular invasion), and 2) poor prognosis (nuclear grade 2 with microvascular invasion and nuclear grade 3 with or without microvascular invasion). The combination of these histologic parameters provides adequate prognostic stratification.

Simple tumor profile chart based on cell kinetic parameters and histologic grade is useful for estimating the natural growth rate of hepatocellular carcinoma Tomoki Nakajima, MD, PhD
Michihisa Moriguchi, MD
Yasuhide Mitsumoto, MD, PhD, etal


Hum Pathol 2002;33:92-99. Abstract quoteThirty-four untreated hepatocellular carcinomas (HCCs) with known growth rates were classified into 5 groups on a tumor profile chart based on their doubling time (DT), Ki-67–positive index (Ki-67-PI), apoptotic index (Apo-I), and histologic grade. The slow-growing HCCs (DT > 100 days) consisted of well-differentiated tumors with slight cell kinetic imbalance and were divided into groups A and B. Group A had Apo-I values <3%, and most tumors had Ki-67-PI values <10%, whereas group B had Apo-I values of 3‰ to 10‰ and Ki-67-PI values of 10% to 20%. The HCCs with intermediate growth rates, which had Ki-67-PI values similar to those of the tumors in group B, were divided into groups C and D based on differences in cell kinetics: group C consisted of well-differentiated tumors, most of which had Apo-I values <3‰, and group D consisted of moderately or poorly differentiated tumors with Apo-I values between 10‰ and 20‰. The rapidly growing tumors (DT < 50 days, group E) had higher Ki-67-PI values than other groups and a wide range of Apo-I values. Rapidly growing tumors were mostly moderately or poorly differentiated, with a large cell kinetic imbalance in favor of cell production.

This grouping system is useful for approximating the growth rate of HCCs in a clinical setting, even when only histologic parameters are available.


Clinicopathologic and prognostic significance of the histologic activity of noncancerous liver tissue in hepatitis B virus-associated hepatocellular carcinoma.Ng IO, Poon RT, Shek TW, Fan ST.Department of Pathology, University of Hong Kong, Queen Mary Hospital, People's Republic of China.


Am J Clin Pathol 2002 Mar;117(3):411-8 Abstract quote

We prospectively studied 66 patients infected with the hepatitis B virus who underwent liver resection for hepatocellular carcinoma (HCC) to evaluate the influence of the histologic activity of noncancerous liver tissue on clinicopathologic features and prognosis. Based on the histologic activity index (HAI) score of nontumorous liver tissue, patients were classified into 3 groups: mild, moderate, or severe hepatitis. Overall, higher HAI scores were more frequent in patients with poorer liver function: lower serum albumin levels and higher indocyanine green retention at 15 minutes. Moreover, patients with moderate hepatitis had more frequent venous invasion, and the tumor size decreased with increasing HAI scores. Similar results were observed when the fibrosis category was excluded in the calculation of HAI scores. The overall or disease-free survival rates did not differ significantly among the 3 groups of patients. However, higher fibrosis scores were associated significantly with shorter disease-free survival rates. HAI scores correlated significantly with certain clinicopathologic features.

In patients with hepatitis B-related HCC, a higher fibrosis score in the nontumorous liver tissue, but not histologic hepatitic activity, seems to be a significant factor predisposing to shorter survival.


Intrahepatic metastases commonly found in 60% of tumors <5 cm in diameter and >95% of tumors >5 cmVascular invasion with portal vein thrombosis in 65-75% and hepatic vein thrombosis in 20-25%
May extend into inferior vena cava and right atrium and right ventricle
Bile duct invasion in 5%
Invasion in advanced disease to the diaphragm, gallbladder, and peritoneal dissemination

Extrahepatic metastases foundin 50% of cases at autopsy:
Lung in 50%
Regional lymph nodes
Adrenal glands 15%

Recurrence Underlying cirrhosis is risk factor for development of new tumors

Most patients die of liver failure secondary to replacement of the liver by tumor

If surgically resectable, median survival of 20-45 months
5YRS of up to 75% for tumors <5 cm

Survivin expression in hepatocellular carcinoma: correlation with proliferation, prognostic parameters, and outcome.

Fields AC, Cotsonis G, Sexton D, Santoianni R, Cohen C.

1Department of Pathology and Laboratory Medicine, Emory University School of Medicine, Atlanta, GA, USA.
Mod Pathol. 2004 Nov;17(11):1378-85. Abstract quote  

Survivin is a novel inhibitor of apoptosis. It is detected in fetal and neoplastic adult tissue, but not in normal tissues. Several recent studies have shown that survivin not only inhibits apoptosis, but also accelerates cancer cell proliferative activity. Expression of the protein may be of prognostic significance and therapeutic relevance in many cancers.

We investigated survivin expression in hepatocellular carcinoma, correlating results with proliferation (MIB-1), prognostic factors, and outcome. Paraffin-embedded sections of 72 hepatocellular carcinoma were immunostained for survivin and MIB-1 using tissue microarray technology. Expression was evaluated in nuclei and cytoplasm as intensity (0-3+), and percentage of positive cells scored on a four-tiered system with less than 10%=negative; 10-25%=1; 26-50%=2; 51-75%=3; and 76-100%=4. Frequency of nuclear survivin expression was 43%. There was a significant correlation between nuclear survivin expression and nuclear grade (P=0.0271), microvascular invasion (P=0.0064), mitotic rate (P=0.0017), and MIB-1 (P=0.0001), as well as local recurrence (P=0.0487), and disease-free survival (P=0.0098). Histologic grade (P=0.0544) and stage (P=0.0548) tended to correlate with survivin expression, which did not correlate with cirrhosis, tumor necrosis, multiple tumors, metastatic disease, or overall survival.

Survivin expression correlates with poor prognostic parameters (high nuclear and histologic grade, microvascular invasion, increased proliferation (mitotic count, MIB-1)), local recurrence, and shorter disease-free survival, but does not correlate with overall survival. An important role is suggested for survivin in progression, recurrence, and treatment of hepatocellular carcinoma.

Complete resection if possible
Orthotopic liver transplantation

Chemotherapy, hormonal therapy, and radiotherapy have little effect

Angiographic embolization of hepatic artery
Chemoembolization combined with Lipiodol may produce tumor necrosis
Percutaneous ethanol injection into tumor
Has resulted in necrosis and palliation of tumors, usually <3cm

Orthotopic liver transplantation for hepatocellular carcinoma. Factors affecting long-term patient survival. Ojogho ON, So SK, Keeffe EB, Berquist W, Concepcion W, Garcia-Kennedy R, Imperial J, Esquivel CO.

Department of Surgery, Stanford University Medical Center, Calif., USA.

Arch Surg 1996 Sep;131(9):935-9 Abstract quote OBJECTIVE: To determine the influence of several clinicopathologic factors on the 3-year actuarial survival of patients with nonfibrolamellar hepatocellular carcinoma (HCC) following orthotopic liver transplantation (OLT). DESIGN: A case series of 26 consecutive patients with HCC treated with OLT, with a maximum follow-up of 90 months. SETTING: A tertiary care center. PATIENTS: Between March 1988 and December 1993, 521 OLTs were performed in 480 patients, 27 of whom had HCC. One patient was excluded because of donor-transmitted melanoma. Of the remaining 26 patients, there were 18 adults and 8 children, with a mean age of 41 years (range, 0.2-67.4 years). Fourteen patients (54%) had either hepatitis B (n = 6) or hepatitis C (n = 8), while 15 (58%) had coincidental tumor. INTERVENTION: OLT was performed using standard techniques. MAIN OUTCOME MEASURES: The effect of several clinicopathologic factors on 3-year actuarial patient survival. RESULTS: The overall actuarial survival rates for the 26 patients with HCC were 73%, 65.4%, and 65.4%, at 1, 2, and 3 years, respectively. Sixteen patients (62%) were alive at the time of this report, with 14 (54%) free of disease. None of the clinicopathologic factors significantly affected the 3-year patient survival rate. However, the rate of recurrent HCC was significantly higher in nonincidental vs coincidental tumors and in solitary vs multiple tumors.

CONCLUSION: Our results suggest that HCC should not contraindicate OLT, as long-term patient survival and cure can be achieved. While patient selection is important, survival in patients with HCC after OLT is not always predictable using the usual clinicopathologic prognostic factors.

Liver transplantation for hepatocellular carcinoma: a registry report of the impact of tumor characteristics on outcome. Klintmalm GB.

Department of Surgery, Baylor University Medical Center, Dallas, Texas 75246, USA.

Ann Surg 1998 Oct;228(4):479-90 Abstract quoteOBJECTIVE: The objective of this study from the International Registry of Hepatic Tumors in Liver Transplantation is to analyze the impact of tumor characteristics on tumor recurrence and patient survival. SUMMARY BACKGROUND DATA: Many attempts have been made to identify patients with hepatocellular carcinoma who can be treated successfully with liver transplantation. Studies presented to date lack enough patients to make reported findings universally accepted. In lieu of a prospective, randomized multicenter trial, in 1992 an International Registry of Hepatic Tumors in Liver Transplantation was established to collect data on these patients, their tumors, and their treatment. METHODS: The registry mails out new patient registration forms and patient follow-up forms twice yearly to all known liver transplant programs. Fifty-three programs in 21 countries have supplied information on 553 patients with tumors. Four hundred ten patients had hepatocellular carcinoma (HCC), and 12 had the fibrolamellar variant of HCC (FLL-HCC). These 422 patients were investigated for this study. One hundred sixty-nine of these (40.0%) were classified as "incidental tumors." For the remaining patients, the tumor was known before the transplant. Twenty-six and eight tenths percent of the patients had a history of hepatitis B and 32.7% had a history of hepatitis C. RESULTS: One hundred ninety patients (46.7%) have died, 99 free of tumor and 91 with tumor. Death was tumor related in 90 patients. Of the 232 patients now living, 215 are free of tumor and 17 have tumor. The most common sites for recurrence are the transplanted liver (41.7%) and the lungs (28.7%). The overall patient survival was 72.2% at 1 year, 63.4% at 2 years, 47.4% at 4 years, and 44.4% at 5 years. Using univariate analysis, incidental tumors (p = 0.3107), FLL-HCC (p = 0.0704), multifocal tumor (p = 0.5464), and bilobar tumor (p = 0.1024) were not found to have an influence on patient survival. Four factors, tumor size greater than 5 cm (p = 0.0221), vascular invasion (p = 0.0005), positive nodes (p = 0.0014), and histologic grade (p = 0.0001) had a profound impact on patient survival. Using Cox multiple regression analysis, only histologic grade had a negative impact on overall patient survival (p = 0.0009) and for patients with known tumors (p = 0.0003). For incidental tumors, patient survival was negatively influenced by multifocality (p = 0.0021) and an age older than 60 years (p = 0.0008). Tumor histologic grade (p = 0.0134) and size (>5 cm) (p = 0.0133) were significantly linked to recurrence-free patient survival.

CONCLUSIONS: This analysis has documented three tumor characteristics that strongly impact patient survival after transplantation for HCC. In addition to tumor size greater than 5 cm and the presence of vascular invasion (which confirm several, single-center studies), this registry notes that a poorly differentiated HCC may be a contraindication for transplantation. A liver tumor's histologic grade may be important information to have when these patients are considered for liver transplantation.

Resection or transplantation for hepatocellular carcinoma in cirrhotic patients: outcomes based on indicated treatment strategy. Figueras J, Jaurrieta E, Valls C, Ramos E, Serrano T, Rafecas A, Fabregat J, Torras J.

Liver Transplant Unit, Ciutat Sanitaria Universitaria Bellvitge, University of Barcelona, Spain.

J Am Coll Surg 2000 May;190(5):580-7 Abstract quote BACKGROUND: Surgical resection has been the treatment of choice for hepatocellular carcinoma (HCC), but the resection rate remains low in cirrhotic patients and recurrence is common. Unfavorable results compared with benign disease and the shortage of organ donors have led to a restricted indication for orthotopic liver transplantation (OLT) for HCC. STUDY DESIGN: The aim of this study was to analyze the results of our surgical approach to HCC in patients with cirrhosis. The first treatment strategy indicated in these patients was OLT. From January 1990 to May 1999, 85 patients underwent OLT and the remaining 35 had surgical resection. RESULTS: One-, 3-, and 5-year survival rates were 84%, 74%, and 60% versus 83%, 57%, and 51%, respectively, in the OLT and resection groups (p = 0.34). Hepatic tumor recurrence was much less frequent in the OLT group than in the resection group. The 1-, 3-, and 5-year disease-free survival rates were 83%, 72%, and 60% versus 70%, 44%, and 31%, respectively (p = 0.027). In the multivariate Cox regression analysis, macroscopic vascular invasion was the only factor independently associated with death or recurrence after OLT (p = 0.006). After partial liver resection, the tumors significantly associated with mortality and recurrence in the multivariate analysis were solitary or multiple tumors greater than 2cm with microscopic vascular invasion (pathologic pT3) (p = 0.01).

CONCLUSIONS: Our results confirm that in cirrhotic patients, OLT may provide better outcomes than liver resection in carefully selected HCC and that longterm survival is similar to the results of OLT in cirrhotic patients without tumors.

Origin of adenocarcinoma in a transplanted liver determined by microsatellite analysis.
Kakar S, Burgart LJ, Charlton MR, Saito Y, Halling K, Thibodeau SN.

Divisions of Pathology, Gastroenterology/Hepatology, and Molecular Pathology, Mayo Clinic, Rochester, MN.

Hum Pathol 2002 Apr;33(4):435-6 Abstract quote

Inadvertent transmission of neoplastic cells from an organ donor can occur at the time of transplantation. Determination of recipient versus donor origin of a tumor is crucial for patient management. This report illustrates the use of microsatellite (MS) analysis to determine the origin of adenocarcinoma arising in a liver transplant. The study patient was a 42-year-old male who had received a liver transplant for hepatitis C and alcohol-related cirrhosis. At the 1-year follow-up visit, a 1.5-cm liver mass was identified during routine ultrasound of the vascular anastamoses. A liver biopsy showed a moderately differentiated adenocarcinoma. Tumor, donor, and recipient DNA were isolated from the paraffin-embedded liver biopsy, pretransplant donor liver biopsy, and the explant liver tissue, respectively. MS analysis was performed by polymerase chain reaction using 5 markers: D5S346, ACTC, D2S123, D18S34, and TP53.

The allelic patterns of tumor DNA were identical to those of donor DNA and were distinct from the DNA profile of the recipient. The use of MS analysis clearly established that the adenocarcinoma was of donor origin.

Mod Pathol 2000;13:679-704
Macpherson and Pincus. Clinical Diagnosis and Management by Laboratory Methods. Twentyfirst Edition. WB Saunders. 2006.
Rosai J. Ackerman's Surgical Pathology. Ninth Edition. Mosby 2004.
Sternberg S. Diagnostic Surgical Pathology. Fourth Edition. Lipincott Williams and Wilkins 2004.
Robbins Pathologic Basis of Disease. Seventh Edition. WB Saunders 2005.
DeMay RM. The Art and Science of Cytopathology. Volume 1 and 2. ASCP Press. 1996.
Weedon D. Weedon's Skin Pathology Second Edition. Churchill Livingstone. 2002
Fitzpatrick's Dermatology in General Medicine. 6th Edition. McGraw-Hill. 2003.
Weiss SW and Goldblum JR. Enzinger and Weiss's Soft Tissue Tumors. Fourth Edition. Mosby 2001.

Commonly Used Terms


Large cell change-Term used in place of large cell dysplasia. Scattered foci within a cirrhotic liver consisting of enlarged hepatocytes with large, often irregularly shaped nuclei and nucleoli but with a normal nuclear:cytoplasmic ratio.

Small cell change-Term used in place of small cell dysplasia. Hepatocytes are smaller than normal liver cells and appear as a zone of nuclear crowding or increased nuclear density. Cytoplasm usually more basophilic than normal hepatocytes but lacking nuclear atypia or enlargement. Usually present as clusters.

Basic Principles of Disease
Learn the basic disease classifications of cancers, infections, and inflammation

Commonly Used Terms
This is a glossary of terms often found in a pathology report.

Diagnostic Process
Learn how a pathologist makes a diagnosis using a microscope

Surgical Pathology Report
Examine an actual biopsy report to understand what each section means

Special Stains
Understand the tools the pathologist utilizes to aid in the diagnosis

How Accurate is My Report?
Pathologists actively oversee every area of the laboratory to ensure your report is accurate

Got Path?
Recent teaching cases and lectures presented in conferences

Internet Links

Pathologists Who Make A Difference
Search for a Physician Specialist
Hepatocellular Carcinoma-Gross Photo
Hepatocellular Carcinoma-Gross Photo
Hepatocellular Carcinoma-Gross Photo
Hepatocellular Carcinoma-Microscopic Photo
Hepatocellular Carcinoma-Microscopic Photo

Last Updated June 7, 2007

Send mail to The Doctor's Doctor with questions or comments about this web site.
Read the Medical Disclaimer.

Copyright © The Doctor's Doctor