Is histological diagnosis of primary liver carcinomas with fibrous stroma reproducible among experts?

AIMS: In the era of targeted therapeutics, histological typing of hepatobiliary carcinomas has major clinical implications. Little is known about the reproducibility of the pathological diagnosis of primary liver carcinomas. Therefore, this study aimed to evaluate the worldwide variation in the pathological expert diagnoses of primary liver carcinomas with fibrous stroma in patients who did not have cirrhosis. METHODS: A single set of slides was selected from 25 tumours, and this set was reviewed independently by 12 pathologists who have worldwide expertise in liver tumours. Reproducibility of the diagnoses was evaluated by Light's kappa, and diagnoses were clustered by multidimensional scaling. Immunohistochemistry was performed after histological review. RESULTS: The interobserver reproducibility for diagnosis of hepatocellular carcinoma subtypes and cholangiocarcinomas was poor (kappa 0.23-0.52), even when the experts considered that the diagnosis required no additional stains or clinical information. Interestingly, multidimensional scaling revealed three main clusters of tumours: hepatocellular carcinoma with no other specifications (n = 13), fibrolamellar hepatocellular carcinoma (n = 3) and cholangiocarcinoma (n = 9). Using immunohistochemistry, these histological clusters correlated with expression of anti-hepatocyte and anti-cytokeratin 19 (p<0.001). CONCLUSIONS: The results demonstrate the poor reproducibility among experts of the pathological diagnosis of primary liver carcinomas with fibrous stroma in patients who did not have cirrhosis, and highlight that the systematic use of immunohistochemistry may improve the diagnostic accuracy.

Spreds, inhibitors of the Ras/ERK signal transduction, are dysregulated in human hepatocellular carcinoma and linked to the malignant phenotype of tumors.

Aberrant activation of the Ras/Raf-1/extracellular-regulated kinase (ERK) pathway has been shown to be involved in the progression of human hepatocellular carcinoma (HCC). However, the mechanism of dysregulation of ERK activation is poorly understood. Recently, we identified Sprouty-related protein with Ena/vasodilator-stimulated phosphoprotein homology-1 domain (Spred) as a physiological inhibitor of the Ras/Raf-1/ERK pathway. In this study, we found that the expression levels of Spred-1 and -2 in human HCC tissue were frequently decreased, comparing with those in adjacent non-tumorous tissue. Moreover, Spred expression levels in HCC tissue were inversely correlated with the incidence of tumor invasion and metastasis. Forced expression of Spred-1 inhibited HCC cell proliferation in vitro and in vivo, which was associated with reduced ERK activation. Spred-1 overexpression also reduced the secretion of matrix metalloproteinase-9 (MMP-9) and MMP-2, which play important roles in tumor invasion and metastasis. In addition, Spred-1 inhibited growth factor-mediated HCC cell motility. These data indicate that the reduction of Spred expression in HCC is one of the causes of the acquisition of malignant features. Thus, Spred could be not only a novel prognostic factor but also a new therapeutic target for human HCC.

SEN virus infection in patients with hepatocellular carcinoma.

Although most cases of hepatocellular carcinoma (HCC) are associated with either the hepatitis B or C viruses (HBV, HCV), about 10-20% of HCCs occur in patients with chronic hepatitis that is aetiologically undefined. The aim of the present study was to determine the prevalence of the transfusion-transmitted SEN virus (SEN-V) in patients with HCC, including those patients who do not otherwise appear to be infected with HBV or HCV. Fragments of SEN-V subtypes D and H were amplified separately by PCR from the sera of 50 patients with HCC (31 from Canada and 19 from Japan) as well as from HCC and adjacent nontumourous liver tissues from eight of the Canadian patients. SEN-V DNA was found in the serum of 10 of 31 (32%) Canadian patients and eight of 19 (42%) Japanese patients [overall, 18 of 50 (36%) HCC patients]. SEN-V DNA was detected in the serum of 10 of 23 (43%) HCC patients with antibody to HCV (anti-HCV), six of 11 (55%) with hepatitis B surface antigen (HBsAg), and two of 16 (12%) without detectable anti-HCV or HBsAg. Twenty-three HCC patients in this study had 'silent HBV,' characterized by the detection of HBV DNA in the absence of HBsAg; eight of these (35%) also had SEN-V infections. SEN-V DNA was detected in HCC patients most typically in those with coexistent HBV or HCV infection. SEN-V was found in only one of seven HCC patients without HBV (without HBsAg or HBV DNA) or HCV and thus does not appear to be an important cause of 'cryptogenic' HCC.

HBsAg-negative hepatitis B virus infections in hepatitis C virus-associated hepatocellular carcinoma.

This study was conducted to evaluate reports that hepatitis B virus (HBV) DNA sequences can be found in the serum and/or tumour tissue from some hepatocellular carcinoma (HCC) patients who have no detectable hepatitis B surface antigen (HBsAg) in their sera. Such HBV infections would be highly atypical, because prospective studies have shown a clear succession of specific serologic markers during and after most HBV infections. As most HBsAg-negative HCC patients in Japan have hepatitis C virus (HCV) infections, the present study was conducted to determine whether some of these patients actually have unrecognized HBV infections. Thirty newly diagnosed HCC patients from Kurume, Japan, with antibody to the hepatitis C virus (anti-HCV) were studied. None of the 30 had HBsAg detectable in their serum. Of 22 for whom test results for antibodies to the hepatitis B core antigen (anti-HBc) and antibodies to HBsAg (anti-HBs) were available, 14 (64%) had anti-HBc and anti-HBs, four (18%) had anti-HBc alone, and four (18%) had no HBV markers. Nested polymerase chain reaction was used to detect the HBV surface (S), core (C), polymerase (P) and core promoter gene sequences in the HCC tissues and in the adjacent nontumorous liver tissues. HBV DNA was detected in HCC and/or adjacent nontumorous liver in 22 of 30 (73%) patients [detected in both HCC and nontumorous liver in 19/30 patients (63%)]. Among the 22 patients with detectable HBV DNA, more than one HBV gene was detected in 10 (46%). Among the four patients whose sera were negative for all HBV markers, three had HBV DNA in either HCC and nontumorous liver (two cases) or only in the nontumorous liver (one case); HBV DNA could not be detected in tissues from the fourth patient. In 18 of 21 (86%) patients with detectable HBV core promoter sequences, mutations at both nucleotides 1762 (A-GT) and 1764 (G-A) in the core promoter region were found. No deletions were detected in the core promoter gene region of the type reported to be associated with some cases of HBsAg-negative HBV infection. Thus, HBV DNA was detectable in 22 (73%) HBsAg-negative, anti-HCV-positive HCCs, including three (10%) who were also negative for anti-HBc and anti-HBs. HBV mutations at both nucleotides 1762 (A-GT) and 1764 (G-A) in the core promoter region were found in the majority of cases, mutations that have previously been reported in HBV that is integrated in HCC DNA. In serologic surveys to determine etiologic associations of HCC, patients such as those in this study would have been incorrectly designated as having 'HCV-associated HCC,' whereas the data in this study suggest that HBV could have played a role in the development of their HCCs.

Integration of hepatitis B virus containing mutations in the core promoter/X gene in patients with hepatocellular carcinoma.

UNLABELLED: Integration of hepatitis B virus is thought to be an essential step in hepatitis B virus associated hepatocarcinogenesis. Mutations at nucleotides 1762 and 1764 in the hepatitis B virus, within a sequence encoding both the core promoter gene and the X gene, have been found frequently in patients with hepatocellular carcinoma. However, integration of these mutant sequences has not been reported to date. METHODS: A 228-base pair segment of the hepatitis B virus core promoter gene was amplified from hepatocellular carcinomas and adjacent non-tumourous liver tissue by nested PCR and sequenced. Integration of hepatitis B virus into human genomic DNA was investigated using the 'genome walking' method. RESULTS: Point mutations were found in both hepatitis B virus nucleotides 1762 and 1764 in 8 of 14 hepatocellular carcinoma tissues (57%) and in 11 of 14 adjacent non-tumourous liver tissues (79%). Three patients were evaluated using the 'genome walking' method; all were found to have hepatitis B virus DNA integrated in their hepatocellular carcinoma (two patients) and/or in their non-tumourous liver tissue (three patients). Integration occurred in all tissues near host genomic sites that are prone to integration. Hepatitis B virus was integrated at or near the hepatitis B virus DR1 site in all samples, and all contained truncated X gene sequences that have been reported to be capable of producing fusion transcripts with transactivation potential. CONCLUSIONS: Integrated hepatitis B virus DNA containing core promoter mutations at nucleotides 1762 and 1764 was found in hepatocellular carcinoma and/or adjacent non-tumourous liver tissue of three patients. These findings leave open the possibility that insertional mutagenesis or transactivation by fusion transcripts resulting from hepatitis B virus integration could play a role in hepatocarcinogenesis in some patients.

Immunohistologic study on the expressions of alpha-fetoprotein and protein induced by vitamin K absence or antagonist II in surgically resected small hepatocellular carcinoma.

Sixty-eight cases of single hepatocellular carcinoma (HCC) with less than 3 cm of diameter were immunohistochemically examined for the expressions of alpha-fetoprotein (AFP) and protein induced by vitamin K absence or antagonist II (PIVKA-II). In cancerous tissues, the expression rate was significantly higher for PIVKA-II (34 cases [50%]) than AFP (21 cases [31%]) (P <.05), suggesting a higher specificity of PIVKA-II to small HCC. Sixteen of the 68 cases (24%) were positive to both AFP and PIVKA-II, and in 8 of the 16 cases, AFP and PIVKA-II expressing areas within a nodule were clearly divided by a fibrous septum. According to histologic grades, PIVKA-II expression was confirmed in 2 of the 15 well-differentiated HCCs, and in the well-differentiated component of 6 of the 12 "nodule-in-nodule"-type well-differentiated HCCs. AFP expression was not found in well-differentiated HCCs, but found in 16 of the 40 moderately differentiated HCCs (40%) and in the moderately differentiated component of 3 of the 12 "nodule-in-nodule"-type well-differentiated HCCs. The positive rate in the tissues was correlated to the serum levels for both AFP and PIVKA-II. In addition, frequency of tissue-PIVKA-II expression was higher than tissue-AFP expression in the cases whose serum protein level was within the normal range. This indicates that AFP and PIVKA-II have different patterns of tissue expression and of secretion to the blood. In comparison with tissue-AFP-negative cases, tissue-AFP-positive HCCs had a larger tumor size, higher frequencies of portal vein invasion and intrahepatic metastasis, a high Ki-67 labeling index, and a lower rate of recurrence-free survival. Thus, tissue-AFP-positive HCCs are suggested to be biologically more malignant than those HCCs that are AFP-negative and PIVKA-II-positive.

Recurrence of hepatocellular carcinoma: multicentric occurrence or intrahepatic metastasis? A viewpoint in terms of pathology.

Recurrence after successful surgical or nonsurgical treatment of hepatocellular carcinoma (HCC) is caused either by intrahepatic metastasis or by metachronously multicentric occurrence. Intrahepatic metastasis is a major cause of recurrence of advanced HCCs with varying degrees of vascular invasion, and multicentric occurrence is a frequent cause of recurrence in small HCCs with no obvious vascular invasion. It is estimated that at least 20% of small HCCs have a high probability of recurrence due to multicentric occurrence, based on the finding that adenomatous hyperplasia (AH) and/or atypical adenomatous hyperplasia (AAH), which are considered premalignant lesions, are found in the vicinity of resected small HCCs with liver cirrhosis. However, because neither AH nor AAH occur in HCC cases without liver cirrhosis, most recurrence of HCC in noncirrhotic liver is considered to be due to intrahepatic metastasis or to de novo hepatocarcinogenesis. In a survey of autopsy cases of liver cirrhosis with small HCC, smaller HCC nodules were found in other liver slices in 50% of cases, and it is estimated that approximately 50% of HCC is already multicentric in the early stage.

Laminin deposition to type IV collagen enhances haptotaxis, chemokinesis, and adhesion of hepatoma cells through beta1-integrins.

BACKGROUND/AIMS: In hepatocellular carcinoma, laminin deposition to type IV collagen along the sinusoids is observed with the development of arterial network, coinciding with intrahepatic metastasis. We investigated the influence of laminin deposition to type IV collagen on hepatoma cell adhesion, motility and secretion of matrix metalloproteinases (MMPs), which are indispensable behaviors for tumor metastasis. METHODS: Hepatoma cell lines (KYN-1, -2 and -3) were used. The expression of integrin subunit mRNAs in hepatoma cells was confirmed by RT-PCR. The influence of laminin addition to type IV collagen on the adhesion, chemokinesis, and migration of KYN-1, -2 and -3 was evaluated by the haptotactic migration, phagokinetic track motility, and cell adhesion assays. The effects of integrin subunits on these activities were evaluated using the function-blocking antibodies for integrins. Phosphorylation of MEK1/2 and secretion of MMPs were investigated by Western blotting and gelatin zymography. RESULTS: Integrin alpha1, alpha2, alpha3, alpha6 and beta1 subunit mRNAs were detected. The combination of type IV collagen and laminin enhanced the migration, chemokinesis, and adhesion of hepatoma cells compared to that of type IV collagen when used alone. The enhanced activity was significantly suppressed by function-blocking antibodies for integrin alpha1, alpha2, alpha3, alpha6 and beta1 subunits. Hepatoma cells cultured on the combination of type IV collagen and laminin showed phosphorylation of MEK1/2 and increased secretion of MMPs. CONCLUSIONS: The addition of laminin to type IV collagen enhances hepatoma cell adhesion and motility through beta1-integrins.

Involvement of p21(WAF1/Cip1), p27(Kip1), and p18(INK4c) in troglitazone-induced cell-cycle arrest in human hepatoma cell lines.

Peroxisome proliferator-activated receptor gamma (PPARgamma) regulates cell growth and differentiation. Recent evidence has suggested that PPARgamma ligands had anti-tumor effects through inhibiting cell growth and inducing cell differentiation in several types of malignant neoplasm. In the present study, we investigated: 1) the expression of PPARgamma in both human hepatoma cell lines and 5 resected human hepatocellular carcinoma (HCC) tissues; 2) the growth-inhibitory effect of troglitazone, a PPARgamma ligand, on those hepatoma cells; and 3) the molecular mechanisms of troglitazone-induced cell-cycle arrest. Five hepatoma cell lines, HLF, HuH-7, HAK-1A, HAK-1B, and HAK-5, were used. The mRNA expression levels of PPARgamma, p21(WAF1/Cip1), and p27(Kip1) were determined by real-time quantitative reverse transcription-polymerase chain reaction. The expression of cell cycle-regulating proteins, such as p21, p27, p18(INK4c), cyclin E, and pRb, was examined using Western blotting. PPARgamma was constitutively expressed in all the cell lines and the HCC tissues used in this study. A cytostatic effect of troglitazone was found in those cell lines, and this inhibition of cell growth was dosage-dependent. G0/G1 arrest was apparently demonstrated in flow cytometric analysis in HLF, HAK-1A, HAK-1B, and HAK-5, all of which showed an increased expression of p21 protein. However, HuH-7, lacking p21 protein expression, did not demonstrate clear arrest in the cell-cycle analysis. HLF, which was deficient in the protein product of the retinoblastoma tumor-suppressor gene (pRb), responded most profoundly to troglitazone, showing an increased expression in not only p21, but also in p27 and in p18. These findings suggested that p21, p27, and p18 might be involved in troglitazone-induced cell-cycle arrest in human hepatoma cells.

Expression of angiogenic factors, basic fibroblast growth factor and vascular endothelial growth factor, in human biliary tract carcinoma cell lines.

In order to clarify angiogenic mechanism in biliary tract carcinoma, expressions and functions of basic fibroblast growth factor (bFGF) and its receptors (FGFR-1-4), and vascular endothelial growth factor (VEGF) and its receptors were investigated by using human biliary tract carcinoma cell lines (KMC-1, KMC-2, KMBC and KMG-C). Expression of bFGF was confirmed in KMC-1 and KMC-2, and that of FGFR-1-4 in all the cell lines except no FGFR-2 in KMC-2. Expression of VEGF was detected in all the cell lines, whereas the cell lines did not express VEGF receptors. Addition of anti-bFGF neutralizing antibody to the medium did not suppress cell proliferation, whereas exogenous bFGF with or without heparin accelerated cell proliferation in all cell lines. Addition of anti-bFGF neutralizing antibody or anti-VEGF neutralizing antibody to the co-culture of human umbilical vascular endothelial cells (HUVEC) and KMC-2 suppressed the proliferation of HUVEC. Surgically obtained cholangiocarcinoma tissues (n=7) were immunohistochemically negative to bFGF, while six of the seven were positive to VEGF. These findings suggested that human biliary tract carcinoma cells express both bFGF and VEGF not as autocrine growth factors but as angiogenic factors. On the other hand, expression of VEGF was found at a higher frequency than bFGF both in the cell lines and tissues.

Clinicopathologic study on cholangiolocellular carcinoma.

We clinicopathologically studied 6 resected cases of cholangiolocarcinoma (CLC) including 2 referred cases from other hospitals. The frequency of CLC was 0.56% of the 708 consecutively resected cases of primary liver cancer and the mean age of CLC cases was 66 years. Three of the 6 cases (50%) were hepatitis C virus antibody (HCVab) positive, one (17%) was hepatitis B virus surface antigen (HBsAg) positive, and 2 (33%) were negative to both HCVAb and HBsAg. Serum levels of alpha-fetoprotein were slightly elavated only in 1 case. Clinically, 4 cases were diagnosed as hepatocellular carcinoma (HCC) and 2 cases as cholangiocellular carcinoma (CCC). Grossly, CLCs were whitish in color and solid, not encapsulated, and resembled CCC. Histologically, the tumor cells had eosinophilic cytoplasm with ovoid nuclei, and mild atypia. The tumor proliferated in an anastomosing pattern of Hering's canal-like small glands with an abundant fibrous stroma. Four of the 6 tumors (83%) consisted of only CLC and other 2 tumors contained CCC-like area and HCC-like area in a part of the nodules, respectively. Immunohistochemically, all tumors were positive to cytokeratin (CK) 7. CK8 were also positive in all of 6 cases. These results revealed that CLC had the clinical features resembling HCC but the morphologic features resembling CCC. It is suggested that CLC cells might be derived from Hering's canal or stem cells which have the intermediate features between hepatocytes and bile duct epithelium.

Expression and function of interleukin-8 in human hepatocellular carcinoma.

Expression and functions of interleukin (IL)-8, a pro-inflammatory cytokine with angiogenesis action, was examined in 23 surgically resected hepatocellular carcinoma (HCC) specimens and 7 HCC cell lines. In all HCC tissues, IL-8 expression was confirmed with reverse-transcription polymerase chain reaction method and enzyme-linked immunosorbent assay, and immunohistochemistry showed HCC cells were the major producer of IL-8 in the tissues. Microvessel density was measured by the double immunohistochemical staining of muscular vessels in HCC tissues, but the density was not related to the level of IL-8 in the HCC tissues. On the other hand, in the co-culture of human umbilical vein endothelial cells (HUVEC) and a HCC cell line (KIM-1), IL-8 produced by KIM-1 significantly accelerated the proliferation of HUVEC. In addition, cases with a high IL-8 level in cancerous tissue had a significantly higher frequency of portal vein invasion, venous invasion and bile duct invasion (p<0.05). In the cultures of 7 HCC cell lines IL-8 secretion into culture medium increased with the treatment of IL-1beta or tumor necrosis factor-alpha. This showed IL-8 expression is regulated by inflammatory cytokines. IL-8 produced by HCC is an angiogenesis factor of HCC, but it could have a much more important role in the invasion and metastasis of HCC.

Results of surgical and nonsurgical treatment for small-sized hepatocellular carcinomas: a retrospective and nationwide survey in Japan. The Liver Cancer Study Group of Japan.

Hepatic resection (HX), percutaneous ethanol injection (PEI), and transcatheter arterial embolization (TCAE) have all been used in the treatment of patients with small-sized hepatocellular carcinomas (HCCs). However, the indications for these therapeutic modalities remain unclear. Therefore, the first step to minimize the debate on these indications is to review the standard results from each treatment based on an extensive survey. The participants in this study were patients with HCCs less than 5 cm in diameter who were enrolled in The Liver Cancer Study Group of Japan. The survival rates in the HX (n = 8,010), PEI (n = 4,037), and TCAE (n = 841) groups were calculated in relation to the number of tumors and the clinical stage. In the clinical stage I cases with a solitary tumor less than 2 cm in diameter and in all clinical stages with a solitary tumor greater than 2 cm and in the clinical stage II cases with 2 tumors greater than 2 cm, the HX group showed higher survival rates than the nonsurgical groups. The HX group had a higher male/female ratio and a younger mean age than the PEI or TCAE group. The ratio of HBs antigen-positive cases/hepatitis C virus antibody-positive cases in the PEI group was lower than that in the corresponding HX group. In contrast, the PIVKA-II values in the HX group tended to be higher than in the PEI group. In conclusion, these findings will provide useful information for selection of a therapeutic modality for small-sized HCCs.

Expression of Hu-IFN-alphaR2 chain of type I interferon receptor in human hepatocellular carcinoma and non-cancerous tissues.

Type I interferon (IFN) receptor consists of two chains (Hu-IFN-alphaR1 and Hu-IFN-alphaR2), and Hu-IFN-alphaR2 takes a soluble, short, or long form (Hu-IFN-alphaR2a, Hu-IFN-alphaR2b, or Hu-IFN-alphaR2c, respectively). We examined Hu-IFN-alphaR2 expression in hepatocellular carcinoma (HCC) tissues and their corresponding non-cancerous (non-HCC) tissues. Immunohistochemically, Hu-IFN-alphaR2 expression was positive in 53 (77%) of 69 HCC tissues and in 61 (88%) of 69 non-HCC tissues. Hu-IFN-alphaR2 protein in tissue homogenates of HCC and non-HCC tissues obtained from 29 patients was measured by using ELISA kits, and the amount was 12.7+/-10.9 pg/mg protein in HCC tissue and 10.5+/-5.0 pg/mg protein in non-HCC tissue. Number of specimens in which Hu-IFN-alphaR2 level was 3 pg/mg protein or lower, or 20 pg/mg protein or higher, was one each for non-HCC, while it was 7 (24%) and 6 (21%) for HCC. RT-PCR analysis was done in 7 of the 29 HCC cases. It revealed both Hu-IFN-alphaR2a and Hu-IFN-alphaR2c were expressed in all HCC tissues and in 6 of the 7 non-HCC tissues, and Hu-IFN-alphaR2b was expressed in all HCC tissues and in 4 of the 7 non-HCC tissues. Because immunostaining intensity of Hu-IFN-alphaR2 tended to be higher in the areas with active inflammation, effects of inflammatory cytokines (IL-1alpha, IL-1beta, and TNF-alpha) on Hu-IFN-alphaR2 expression were examined on 11 HCC cell lines. As a result, TNF-alpha up-regulated Hu-IFN-alphaR2 expression in 7 of the 11 cell lines. In 3 of the 7 cell lines, up-regulation of Hu-IFN-alphaR2 on cell surface, as well as of the soluble form of Hu-IFN-alphaR2, was induced not only by TNF-alpha, but also by IL-1alpha or IL-1beta. In conclusion, both HCC and non-HCC tissues frequently express Hu-IFN-alphaR2c that is necessary for Type I IFN response. Hu-IFN-alphaR2 expression in HCC tissues is often attenuated or enhanced, and may be regulated by inflammatory cytokines.

Clinicopathologic study on hepatocellular carcinoma negative for hepatitis B surface antigen and antibody to hepatitis C virus.

Among 326 consecutively resected cases of hepatocellular carcinoma (HCC), we studied clinicopathologic features of 13 cases (NBNC-HCC) negative for hepatitis B surface antigen (HBsAg), antibody to HBsAg (HBsAb), antibody to hepatitis B core antigen (HBcAb), hepatitis B envelope antigen (HBeAg), antibody to HBeAg (HBeAb), and antibody to hepatitis C virus (HCVAb). Forty-four HCC cases positive only for HBsAg (B-HCC) and 232 cases positive only for HCVAb (C-HCC) were studied as controls. Clinically, NBNC-HCC cases showed good liver function reserve. None of NBNC-HCC cases had periodic medical check ups and HCC was detected in 12 cases (92.3%) incidentally. Pathologically, the mean tumor diameter was significantly larger and the histologic grade was less differentiated in NBNC-HCC cases than in B-HCC and C-HCC cases. In the background liver, liver cirrhosis was associated in 15.4% of NBNC-HCC cases, 50.0% of B-HCC and 38.2% of C-HCC cases. The degree of inflammation and fibrosis was less in NBNC-HCC cases, and two cases (15.4%) had almost normal liver histology. In NBNC-HCC cases, synchronous and metachronous multicentric occurrence was not observed. In B-HCC cases, synchronous multicentric occurrence was found in 1 case (20.0%), but no metachronous multicentric occurrence. In C-HCC cases, synchronous and metachronous multicentric occurrence was found in 13 (43.3%) and 8 (30.8%) cases, respectively. However, the cumulative recurrence-free rate was not significantly different among the three groups. Accordingly, it was suggested that better prognosis could be expected in NBNC-HCC cases compared with B- and C-HCC cases, if the cancer could be detected in the early stage.

Bright loop appearance; a characteristic ultrasonography sign of early hepatocellular carcinoma.

Ultrasonography (US) and computed tomography (CT) are the most effective screening methodologies for hepatocellular carcinoma (HCC). In our US screening, 20% of small HCC nodules less than 20 mm in diameter were detected as hyperechoic tumors. Among these hyperechoic HCC nodules, we have often observed (BL) which is defined as hypoechoic nodules in the hyperechoic tumor. In this study, we report that the BL is a sign of dedifferentiation of early stage of HCC with fatty change by US. From 1994 to 1998, we performed tumor targeting needle biopsy in 938 hepatic nodular lesions. Among them, 284 nodules <20 mm in diameter, histologically diagnosed as HCC, were studied. BL is defined as a hyperechoic tumor containing a hypoechoic nodule >4 mm in diameter by US. Among 284 nodules, well, moderately and poorly differentiated HCC were 183 (64.4%), 100 (35.2%) and 1 (0.4%), respectively. On US, hypoechoic, isoechoic, and hyperechoic nodules were 188 (66.2%), 32 (11.3%) and 64 (22.5%), respectively. Forty-seven nodules of 64 hyperechoic HCC nodules <20 mm in diameter, 47 nodules (73.4%) showed fatty changes. Of 64 hyperechoic HCC nodules, we recognized 22 nodules (34.4%) as BL. The proportion of BL type hyperechoic nodules increased with the tumor size. Two hyperechoic nodules followed by US changed to BL with tumor enlargement. Histologic examination of a resected HCC with BL showed that hyperechoic HCC nodule represented well-differentiated HCC with fatty change and inner hypoechoic lesion represented moderately differentiated HCC without fatty change. In US screening for HCC, BL was often observed in HCC nodules from 11 to 20 mm in diameter. Histologic examination revealed that BL of HCC on US was associated with tumor progression and indicated dedifferentiation showing moderately differentiated HCC in well-differentiated HCC with fatty change.

Pathomorphologic study on the mechanism of fatty change in small hepatocellular carcinoma of humans.

BACKGROUND/AIMS: Fatty change is frequently observed in small hepatocellular carcinoma (HCC) of the early stage. However, the mechanism of fatty change and its pathomorphological features in small HCC are not yet fully understood. These issues are addressed here. METHODS: Histological examinations were conducted on 260 HCC nodules (< or =3 cm in diameter) which were surgically obtained from 249 patients. According to the distribution pattern, fatty changes were classified into two types: 'diffuse type' when the change was found throughout the cancerous nodule; and 'focal type' when the change was localized in part of the nodule. To study the pathogenesis of fatty change in HCC in relation to angioarchitecture, the number of arterial tumor vessels and intratumoral portal tracts in 104 of the 260 nodules was counted. RESULTS: Fatty change was found in 51 of the 260 nodules (19.6%), the frequency was highest (36.4%) in the nodules whose diameter was 1.1 to approximately 1.5 cm, and the frequency decreased with the increase in tumor diameter. Small well-differentiated HCCs were often associated with a diffuse type fatty change. With the increase in tumor diameter, moderately differentiated cancerous tissues without associated fatty change appeared, and the focal type was found more frequently. According to the angioarchitecture, in HCCs < or =1.5 cm, the number of intratumoral arteries was significantly smaller in HCCs with fatty change (p<0.05), though the number of intratumoral portal tracts was not significantly different compared with HCCs without fatty change. CONCLUSION: These findings suggest that fatty change of small HCC is closely related to the tumor size, the histological grade and insufficient development of the arterial tumor vessels.

Expression and localization of vascular endothelial growth factor receptors in human hepatocellular carcinoma and non-HCC tissues.

Flt-1 (VEGF receptor-1) and KDR/Flk-1 (VEGF receptor-2) are the high-affinity receptors for the angiogenesis factor, vascular endothelial growth factor (VEGF). VEGF expression has been confirmed in human hepatocellular carcinoma (HCC), and VEGF is thought to be involved in the angiogenesis within HCC tissues. However, expressions of VEGF receptors in HCC have not been reported. We immunohistochemically examined expressions and localizations of Flt-1 and KDR in 28 surgically resected HCC tissues. In non-cancerous area, Flt-1 and KDR were mainly found in macrophages including Kupffer cells; both receptors were found in vascular endothelial cells in the portal veins and arteries within portal tracts; and KDR was also found in some sinusoidal endothelial cells. In cancerous area, Flt-1 and KDR were found in some macrophages, and also in the endothelial cells of intratumoral blood vessels. In 25 moderately and/or poorly differentiated HCCs, KDR expression in the blood space endothelial cells was clear and continuous in 20 cases, and focal in 5 cases. These results suggest that there would be an angiogenesis mechanism via VEGF/Flt-1 or VEGF/KDR in HCC, and the VEGF/KDR system would take a more important role.