The microwave coagulation therapy for nonresectable hepatocellular carcinoma

GASTROENTEROLOGY Vol. 114, No. 4

A1266 AASLD ABSTRACTS L0279

GANCICLOVIR AND LAMIVUDINE COMBINATION THERAPY AFTER ORTHOTOPIC LIVER TRANSPLANTATION. C. Jochum. G. Holtmann, P. Hoffmann, R. Lange, M. Roggendorf, F.W. Eigler, H. Goebell. University of Essen, Departments of Gastroenterology, General Surgery and Virology, Essen, Germany. Backeround: Orthotopic liver transplantation (OLT) in patients infected with hepatitis B virus (HBV) is known to be associated with a high recurrence rate and poor long term prognosis. While interferon treatment alone is not effective in patients after OLT; prolonged treatment with lamivudine may not be effective due to the development of resistant mutations of the virus. Combined treatment with different virostatic agents may overcome this problem and increase the efficacy. However, data are lacking. We report our experience in a patient with hepatitis B after OLT. Case report: In August 1996, OLT was performed in a 23 years old male patient l~ecause of liver cirrhosis due to hepatitis B. After OLT the patient was treated with imrnunoglobulins and HBsAg disappeared for 8 months. April 1997, recurrence of hepatitis B was noted with presence of HBV-DNA and elevated transaminases. Thus treatment with lamividine (100 mg o.d.) and immunoglobulins was initiated. However, this resulted only in a moderate reduction of the viral load and no change of transaminases. Consequently, treatment with immunoglobulins was terminated and Ganciclovir 250 nag (b.d.) added. After two months, the PCR became negative for virus DNA and the HBsAg disappeared. Treatment was continued for another 4 months. At follow-up visits, the patient's serum remained negative for virus DNA and HBsAg and transaminases were within the normal range. The virostatic therapy with lamivudine and ganciclovir was continued. The therapy was well tolerated Conclusion: In a patient with recurrent hepatitis B after OLT, initially unresponsive to treatment with lamivudine, adding ganciclovir to the regimen, yielded in a response with a disappearance of viral DNA and complete normalization of transaminases. Even though long term outcome is not yet available, this combination antiviral therapy deserves further evaluation. • L0280 INHIBITION OF CATHEPSIN B BY CYSTATIN A REDUCES BILE SALT- BUT NOT STAUROSPORINE-INDUCED APOPTOSIS IN A RAT HEPATOMA CELL LINE. B.A. Jones. P.J. Roberts, S.F. Bronk, E. Kominami, L.B. Agellon, G.J. Gores. Juntendo University, Tokyo, Japan; University of Edmonton, Alberta, Canada; and Mayo Clinic, Rochester, MN. We have previously demonstrated abrogation of bile salt-induced apoptosis by pharmacologic cathepsin B inhibitors. However, protease inhibitors are nonselective and molecular approaches are necessary to define the role of cathepsin B in apoptosis. Furthermore, if cathepsin B is to become an ideal pharmacologic target for bile salt-induced apoptosis it should initiate a selective apoptotic pathway. Thus, our AIM was to address the following question: Does expression of cystatin A, a peptide inhibitor of cathepsin B, prevent bile salt- and staurosporine-induced apoptosis. Staurosporine, a kinase inhibitor, was used as a general model of apoptosis. METHODS: For these studies, we employed the McNtcp.24 rat hepatoma cell line which is stably transfected with the bile salt transporter and undergoes bile salt-induced apoptosis. The cystatin A cDNA cloned into pBK-CMV was transfected into McNtcp.24 cells using lipofectamine. Cathepsin B activity was measured using the fluorogenic substrate Boc-Val-Leu-Lys-CMAC. Apoptosis was induced using either 50 pM glycochenodeoxycholic acid (GCDC), a toxic bile salt, or 1 ~M staurosporine, and assessed morphologically after 2 hrs of treatment. RESULTS: Immunoblot analysis demonstrated cystatin A protein expression in transfected cells but not untransfected cells. Cathepsin B activity did not increase in cystatin A transfected cells treated with GCDC although it increased 2.7-fold in untransfected cells (p<0.01). GCDC-induced apoptosis was reduced 2.7-fold in cystatin A treated cells compared to untransfected cells (33 _+ 1% vs. 12 _+2%, p<0.01). In contrast, staurosporine-induced apoptosis was actually increased in cystatin A transfected cells compared to controls (29 + 3% vs. 19-+ 1%, p<0.05). In CONCLUSION, our data demonstrate that bile saltbut not staurosporine-indnced apoptosis can be inhibited by blocking cathepsin B activity using the cystatin A transgene. These data suggest cathepsin B can be selectively inhibited to ameliorate pathologic hepatocyte apoptosis in cholestatic liver diseases without affecting other forms of apoptosis. This work was supported by NIH grant DK41876. L0281 SIMPLE E N Z Y M A T I C DETERMINATION OF TOTAL CHOLESTEROL IN GALLSTONES COMPARED $VITH INTRARED SPECTROPHOTOMETRY METHOD. M. S. Jot, D. H. Lee, W. Choi, P. S. Kim, Y. S. Kim Dept. of Internal Medicine, Inha Univ. Hospital. Inchon, Korea. Background: Methods used for gallstone analysis include X-ray diffraction analysis, infrared anaylsis and chemical analysis. We determined total cholesterol of gallstones with enzymatic method and this showed simple and rapid when infrared spectrophotometry was unable. Method: 1. Gallstones were obtained from 30 patients' gallbladder. 2. Gallstones were crushed into powder and then dissolved in N-N, dimethylformamide and dimethyl sulfoxide (8:2 by volume) mixture.

3. 20 pl of these solvent mixed with 2 ml of reagent mixtures and then incubated at 37°C for 10 minutes. Reagent mixture contains phosphate buffer 0.1 mol/L, 4-aminoantipyrin 0.8 mmol/L, p-hydroxy-benzen-sulfonate 20 mmol/L, esterase 150 U/L, peroxidase 25 KU/L and cholesterol oxidase 200 U/L. 4. The absorbances of these reaction mixtures were measured at 500 nm using spectrophotometer(Beckman-Du 650). 5. The cholesterol calibration curve was made by using standard cholesterols and reagent mixtures. 6. The cholesterol concentrations of gallstones were determined using cholesterol standard curve and each absorbances. 7. This concentrations were compared with the values using infrared spectrophotometry (Nicolet Inst. Corp. Omnic and Quick IR) method. Result: Total cholesterol concentration of 30 gallstones showed. (Fig.l) Correlation between cholesterol in gallstones as determined by enzymatic method and by IR spectrophotometry method.(Fig. 2) ................ ~ . % ! : _ m e h

. ~'.z..,...i,.~-~.-..

j .....~ ......

': r" "~r" - ~ ' ~ t " , - i-~ - -~i ................! ' ~ J ~

,...

- t~0~

............

e0~

...............................................

...............

Fig.l Cholesterol ~ of each stones

Fig.2 Correlation of Enzymatic meflmd and lit method

Conclusion: Enzymatic method for total cholesterol determination of gallstone showed useful, simple and rapid. [,0282

SOMATOSTATIN PLUS ISOSORBIDE -5- MONONITRATE VS SOMATOSTATIN IN THE CONTROL OF ACUTE GASTROESOPHAGEAL VARICEAL BLEEDING: A RANDOMIZED CONTROLLED TRIAL. F. Junquera, JC. L6pez-Talavera*, F. Mearin, E. Saperas, JR Armengol, R. Esteban*, JR. Malagelada. Digestive System Research Unit and * Liver Unit. Hospital General Universitari Vail d'Hebron. Barcelona. Spain. Background: Variceal bleeding is a common and severe complication in liver cirrhosis. Somatostatin (SMS) has been proven to be therapeutically effective in the control of variceal bleeding by decreasing splanchnic blood flow. However, bleeding may persist in about 20% of treated patients. Since nitrates decrease portal pressure by disminishing intrahepatic resistance, we hypothesized that combining nitrates and SMS would improve the results obtained with SMS alone. Aim: To compare the efficacy of intravenous infusion of SMS plus oral isosorbide 5 mononitrate (Is-5-Mn) versus SMS plus placebo in the control of acute variceal hemorrhage in cirrhotic patients in a double blind randomized controlled trial. Material and methods: Over 16 months, 60 cirrhotic patients with endoscopically proven active bleeding from esophageal or gastric varices were, on admission to our Bleeding Unit, classified into two groups according to Child-Pugh criteria (A+B and C), and randomly assigned to receive either a continous infusion of SMS 250 pg/hour plus Is-5-Mn (40 mg/12h p.o) (group I) or SMS infusion plus placebo (group II) for 72 hours. Results: Both patient groups were of similar age, sex distribution, clinical presentation, systemic hemodynamic parameters, biochemical and hematologic profile, endoscopic findings, and etiology of liver disease. Control of bleeding was achieved in 18 out of 30 patients (60%) receiving SMS+ Is-5-Mn and in 26 out of 30 patients (86%) receiving SMS+placebo (p<0.05). Mean packed red cells transfusions were similar in the two groups (2.6 .+ 0.4) group I and (1.8 .+ 0.3) in group II. Mortality was identical [2 out of 30 patients (6,6%)] in both groups. Side effects were not different between group I and II, but development of ascites was higher in SMS+Is-5-Mn (30%) than in SMS+placebo (6.6%) (p <0.05). Conclusion: In acute variceal bleeding in cirrhotics, addition of Is-5-Mn to SMS does not improve the therapeutic efficacy of SMS alone, may even diminish it and should not be used. L0283

THE MICROWAVE COAGULATION THERAPY FOR NONRESECTABLE HEPATOCELLULAR CARCINOMA. T. Kai, K. Kawano, Y. Morii, H. Yokoyama, T. Yoshida, S. Kitano Department of Surgery I, Oita Medical University, Oita, Japan. AIM: In spite of the development of techniques for liver resection, surgical resection is not always feasible in patients with hepatocellular carcinoma (HCC). The purpose of this study was to evaluate the efficacy of microwave coagulation therapy (MCT) which has been recently used as an alternative to hepatic resection. METHODS: Fourteen patients with HCC, from July 1994 to November 1997, were treated with MCT and followed for 1-39 months (average 15 months). A total of 19 nodules were coagulated. The patients underwent MCT through laparotomy (n=9), laparnscopy (n=3), or thoracoscopy (n=2) because of advanced liver cirrhosis, tumor location, and/or intrahepatic metastases. Under ultrasonography (US) guidance, the tumors were coagulated by microwave emitted from an electrode. The therapeutic effect

April 1998

AASLD A1267

after MCT was evaluated by US and contrast-enhanced computerized tomography. RESULTS: All patients had liver cirrhosis and poor hepatic reserve (Child A; n=5, Child B; n=5, Child C; n=4). Tumor size ranged from 10 to 50mm in diameter (average 24.2mm) and the number of electrode insertions for one nodule ranged from 2 to 33 times (output 70-80W, duration 40-45s). The operative time was 30-340min (average 156min). The blood loss was 10-980ml (average 139ml). Of the 19 nodules treated, all nodules underwent complete tumor ablation. In a patient, local recurrence was observed at 2 months after MCT. The operative mortality was 7% (1 of 14); the patient died of sepsis caused by intra-abdominal abscess which did not seem to be directly related to the MCT treatment. The other patients tolerated the operation without operative complication. Nine patients are alive either without tumor (n=3; 4-18 months) or with tumor (n=6; 4-39 months) at the time of this report. Four patients died of intrahepatic recurrence (n=2) or liver insufficiency (n=2). CONCLUSIONS: The present preliminary study suggests that MCT can be a curative treatment and an alternative to hepatic resection in selected patients with HCC. • L0284

GASTRIC VASCULAR ECTASIA DO NOT IMPROVE WITH REDUCTION OF PORTAL PRESSURE BY TRANSJUGULAR INTRAHEPATIC PORTOSYSTEMIC SHUNTS (TIPS). P.S. Kamath; M.A. Lacerda; D.A. Ahlquist; J.C.Andrewsl; M.A. McKusick 1 D.M.Nagorney2. Division of Gastroenterology and Hepatology, and Departments of Diagnostic Radiology 1, and Surgery2. Mayo Clinic, Rochester, MN. Portal hypertensive gastropathy (PHG) is characterized by a mosaic mucosal pattern, red spots, and diffuse hemorrhagic lesions. Some patients have aggregations of ectatic lesions in the stomach. In the antrum these lesions are termed gastric antral vascular ectasia (GAVE). There is some suggestion that GAVE lesions are distinct from portal hypertensive gastropathy (GE 1995;108:138). The effect of portal decompression on GAVE lesions is unclear. AIM: To study the role of TIPS in reducing gastric vascular ectasia (GVE) lesions. METHODS: In 40 patients undergoing TIPS, upper endoseopies carded out before the procedure and at 6 weeks, 3 months, and 6 months following the procedure were videotaped. The endoscopic findings were reanalyzed based on the following criteria: Mild PHG=mosaic pattern with pink center or a red speckling; severe PHG=red spots, brown spots; GVE=confluent cherry red spots, either in the antrum, body or fundus of the stomach (n=4), or linear lesions in the antrum termed "watermelon stomach" (n=4). Patients with GVE were transfusion-dependent and those with a "watermelon stomach" had invariably undergone laser treatment for their lesions without success. RESULTS: Resolution of the astric lesions followingTIPS is shown in the table.

PHG-Mild (n = 26) PHG-Severe (n = 6) GVE (n = 8)

Post TIPS Percentage Resolution 6 Weeks 3 Months 6 Months 69% 88% 80% 33% 33% 33% 0% 0% 0%

Transfusion requirements in patients with GVE did not decrease following placement of TIPS. Four other patients who had distal splenorenal shunts for bleeding from vascular ectasia (previously labeled severe portal hypertensive gastropathy) were also analyzed and no improvement, either clinically or endoscopically was observed. Of note, in all patients with vascular ectasia lesions, esophageal and gastric varices regressed post TIPS and shunt patency was confirmed on ultrasonography. SUMMARY: Following TIPS, mild changes of portal hypertensive gastropathy regress most often, while changes of severe portal hypertensive gastropathy regress in only about one-third of patients. GVE (vascular ectasia and watermelon stomach) associated with chronic liver disease show no change in response to TIPS. CONCLUSION: Gastric vascular ectasia are probably not related to elevated portal pressure. TIPS may not be an appropriate treatment for gastric vascular ectasia in patients of portal hypertension. TIPS should be recommended only with great caution as therapy for severe PHG. • L0285 ABERRANT CO-EXPRESSIONS OF CYTOKERATIN SUBCLASS AND ICAM-1 IN DESTRUCTIVE BILE DUCTS AND SERUM LEVELS OF ANTICYTOKERATIN SUBCLASS ANTIBODY AND SOLUBLE ICAM-I: A COMPARISON BETWEEN AUTOIMMUNE CHOLANGITIS AND PRIMARY BILIARY CIRRHOSIS. Y Kamegaya. M Oda, H Yokomori, S Miyaguchi, H Komatsu, N Tsukada, K Yonei, and H Ishii. Dept. of Internal Medicine, School of Medicine, Keio Univ., Tokyo, Japan.

It is still controversial whether anti-mitochondrial antibody (AMA) -negative and anti-nuclear antibody (ANA)-positive autoimmune cholangitis (AIC) is originally different from AMA-positive primary biliary cirrhosis (PBC). We have previously reported that cytokeratin subclass (MW 52kD) is aberrantly expressed in the destructive bile duct epithelial cells in PBC (Hepatology, 26: 439A, 1997). The aim of the present study is to clarify how immunocytochemical expressions of cytokeratin subclass, ICAM-1 and LFA-1 in the liver in AIC are different from those in AMA-positive PBC. Twelve cases of AMA- and pyruvate dehydrogenase(PDH)E2 component negative AIC (stage I; 6, stage II; 3, stage III; 2, stage IV; 1 cases) and twenty

cases of AMA-positive PBC (stage I; 5, stage II; 5, stage III; 7, stage IV; 3 cases) were subjected to this study. The surgical liver biopsy specimens obtained from these patients were observed by immunocytochemical and in situ hybridization method. Anti-cytokeratin subclass monoclonal antibody was produced against cytokeratin subclass 52kD derived from PtK2 cell line. The localization of cytokeratin subclass (MW 52kD), ICAM-1 and LFA-1 in liver tissue were examined by indirect immunoperoxidase method using anticytokeratin subclass (MW 52kD) antibody, anti-ICAM-l-monoclonal antibody (DAKO, JAPAN) and anti-LFA-l-monoclonal antibody (DAKO, JAPAN) as primary antibody. The localizations of ICAM-1 mRNA were observed by the in situ hybridization method using human ICAM-1 oligonucleotide eDNA probe (Oncogene Science, USA), DIG oligonucleotide tailing kit, DIG detection kit (Boehringer Mannheim). The titers of serum soluble (s) -ICAM-1 and anti-cytokeratin subclass antibody were determined by ELISA. Immunohistochemically ICAM-1 and cytoketain subclass(52kD) were aberrantly expressed on the epithelial cells of the interlobular and septal bile ducts characterized by CNSDC both in AIC and in PBC. ICAM-1 mRNA was also expressed on the interlobular bile duct epithelial cells. Most of the lymphocytes infiltrated particularly around the bile ducts in the portal area were LFA-1 positive. The serum titers of s-ICAM-1 and anti-cytokeratin subclass antibody were markedly elevated both in AIC and in PBC. The titer of s-ICAM-1 was higher in stage III and IV than stage I and II, while anticytokeratin subclass antibody titer are much higher in stage I and II. In conclusion, there are no significant diffemces between AIC and PBC in coexpressions of ICAM-1, LFA-1 and cytokeratin subclass and in serum levels of s-ICAM-1 and anti-cytokeratin subclass, suggesting that AIC belongs to the subspectrum of PBC. • L0286 EFFECTS OF LONG-TERM URSODEOXYCHOLATE ADMINISTRATION ON EXPRESSION LEVELS OF SECRETORY LOW MOLECULAR WEIGHT PHOSPHOLIPASES A 2 AND MUCIN GENES IN PATIENTS WITH MULTIPLE CHOLESTEROL STONF~. M. Kano 1, J. Shoda 1, T. Ueda 2, R. Iwasaki 2, T. Irimura3, T. Urasaki4, Y. Kawanchi4, T. Asano l, Y. Matsuzaki l, N. Tanaka 1. qnstitute of Clinical Medicine, The University of Tsukuba, Ibaraki; 2Mitsubishi Kagaku BCL, Tokyo; 3Department of Pharmacology, The University of Tokyo, Tokyo; 4yamanouchi Pharmaceutical Co., Ltd., Ibaraki, JAPAN. Background: Chronic cholecystitis on histology is common in cholesterol gallstone disease, which in turn may be of importance in causing inflammation-induced mucin hypersecretion by the gallbladder (GB) wall. Group IIA phospholipase A2 (PLA2-IIA), a secretory low molecular weight PLA2, may play a critical role in the process of GB mucosal inflammation in multiple cholesterol stones (CSs) (Gastroenterology 1997;112"2036). On the other hand, the decreased levels of biliary mucin and various pronucleating proteins upon long-term ursodeoxyeholate (UDC) administration as well as the reduction in cumulative occurrence rate of biliary colics may be attributed to the improvement of cholecystitis. Interest should be focused on whether long-term UDC administration may be influential on the expression levels of the secretory low molecular weight PLA2s and mucin genes in the GBs. Aim: To elucidate the beneficial effects of long-term UDC administration on the GB mucosal inflammmation, the expression levels of PLA2-IIA and group V PLA2 (PLA2-V), a novel PLA2 closely involved in prostanoid synthesis, and mucin genes in the GBs were studied for UDC-treated patients and untreated patients with multiple CSs. Methods: Forty-seven patients with multiple CSs were included in this study and 18 of 47 were randomly selected to receive UDC (600 rag/day) for 6-12 months. The remaining 29 patients served as untreated controls. PLA2-IIA levels in gallbladder bile (GBB) were determined by RIA. Expression levels of PLA2-IIA, PLA2-V and mucin gene mRNAs were determined by RT-PCR. Results: With long-term UDC administration, the PLA2-IIA protein mass (2.7 -+0.5 vs. 5.0 + 0.4 ng/mg.protein, mean-+SEM; P<0.01) and mRNA level (1.7+0.2 vs. 3.0+0.4; P<0.01) as well as the PLA2-V mRNA level (6.1-+1.1 vs. 11.9 -+ 1.9 x 10-2; P<0.01) were significantly decreased in the GBs, where the prostaglandin E2 level was concomitantly decreased (190.7-+27.9 vs. 393.6 -+ 55.3 ng/mg.protein; P<0.01). In the GBB, the PLA2-IIA levels were significantly decreased in the UDC-treated patients (43 -+4 ng/dL; P<0.01) than in the untreated patients (78 -+ 6). Significant decreases were similarly found for total protein (2.7-+0.4 vs. 5.1-+0.5 mg/mL; P<0.01), muein (0.53-+0.05 vs. 0.95-+0.09 mg/mL; P<0.01) and free arachidonate concentrations (13.2-+2.8 vs. 35.4-+3.4 [ag/mL; P<0.01), as well as nucleation activity (15.8 -+ 1.2 vs. 1.8 -+0.3 days; P<0.01). In contrast to the decreased mucin level, however, there were no significant changes in the expression levels of mucin genes (MUC1 - 6) between the UDC-treated and untreated patients. In their GBs, MUC3 (275 -+ 25 vs. 325 -+75% of stone-free controls), MUC5B (350-+70 vs. 260-+30%) and MUC6 (243-+ 29 vs. 257 -+29%) were predominantly expressed. Conclusions: Long-term UDC administration was observed to lower the increased PLA2-IIA protein mass and mRNA level as well as the PLA2-V mRNA level in the GBs with multiple CSs, which in turn may be of therapeutic importance in improving the GB mucosal inflammation and thereby decreasing the enhanced mucin secretion. This may be related to the reported efficacy of UDC treatment in cholesterol gallstone disease.