- Email: [email protected]
Viral kinetic study of induction dosing with interferon alpha and ribavirin in the treatment of chronic hepatitis C
A1360 AASLD ABSTRACTS
• L0662 DETERMINING THE MOST EFFICACIOUS DOSE OF URSODEOXYCHOLIC ACID IN PRIMARY BILIARY CIRRHOSIS. A Verma*, RP Jazrawi*, HA Ahmed*, T Davis*, M Bland @, M Bensnn #, RT Orchard #, A Theodossit, JD Maxwell*, TC Northfield*. *Division of Medicine & @Statistics, St George's Hospital Medical School, #St Helier Hospital & tMayday University Hospital, UK. Ursodeoxycholic acid (UDCA) improves liver function tests and prolongs survival in primary biliary cirrhosis (PBC). The dose used in the large trials of 13-15mg/kg/day has largely been based on that used for gallstone dissolution. The only dose response study of UDCA in PBC suggested that a dose of 8mg/kg/day was the most efficacious. However, disease stage of the patients was not known, higher doses of UDCA were not tried and there was no "washout period" between the different doses 1. The aim of this study was to determine the most efficacious dose of UDCA in early stage PBC (stage 1 and 2). A power calculation estimated that 20-25 patients were needed to show a 20% difference between doses with a power of 90% at the 5% level, using alkaline phosphatase (ALP), "/ glutamyl transferase (~'GT) and alanine transaminase (ALT) as variables. 24 biopsy proven early stage PBC patients (M:F, 1:23) had five doses of UDCA (0, 300, 600, 900, 1200 mg/day) each for 2 months with a 1 month washout period between doses. Symptoms (pruritus, fatigue, diarrhoea) were assessed on a four point scale. Statistical analysis was carried on the covariance of variables at the end of the doses, with dose and patients as cofactors, and the baseline variable as the covariate. For all the variables UDCA treatment, irrespective of dose, was significantly better than placebo. 900 and 1200rag were better: than both 300 and 600mg using "tGT and total bilirubin as a variable; than 300mg using ALP, protein, IgG and IgM as varaibles; than 600rag using albumin as a variable. No variables showed a significant difference between the UDCA dose of 900 and 1200rag. We therefore conclude that the most efficacious dose of UDCA is 900rag/day. [1] Podda M et al. Effect of different doses of ursodeoxycholic acid in chronic liver disease. Dig Dis Sci 1989; 34, 12, Suppl 59S-65S. L0663 CORRELATION BETWEEN SERUM QUANTITATIVE HEPATITIS-C VIRUS RIBONUCLEIC ACID LEVELS BY POLYMERASE CHAIN REACTION METHOD AND LIVER HISTOLOGIC ACTIVITIES. V. Viriyautsahakul,P. Suwannakul, T. Tirawatnapong,S. Israsena, V. Mahachai. Department of Medicine, Pathology and Microbiology, Chulalongkom Hospital, Bangkok, Thailand. Background & Aim: The aim of this study was to establish whether the serum Hepatitis-C ribonucleic acid ( HCV-RNA ) levels measured by semiquantitative polymerase chain reaction ( PCR ) method have a correlation with the severity of hepatic inflammation graded by standard histologic scores. Methods: Chronic hepatitis-C patients were included in this study on the basis of positive anti-HCV test by second generation enzyme immunoassay and HCV-RNA test by PCR method. Patients with alcohol abuse, coinfected with other viruses and those who were taking immunomodulatory therapy were excluded. Serum samples for HCV-RNA levels and transaminases were taken within 3 months of liver biopsies. Serum samples were measured for HCV-RNA levels by semiquantitative PCR method using Amplicor ® kit. Liver histologic activities were evaluated and scored using Knodell and Scheuer systems by the only one pathologist who was blinded to HCV-RNA and transaminases levels. Results: A total of 38 patients were included in this study. There were 26 male and 12 female patients. 27 (71.1%) patients had previous transfusion history and 11 (28.9%) patients were sporadic. The correlation coefficients between HCV-RNA levels and Knodell and Scheuer scores were 0.2458 (p=0.175) and 0.2370 (p=0.192), respectively. But in 7 patients with HCV-RNA levels above 0.6 x 106 copies/ml, Knodell and Scheuer scores were higher than 6 and 4, respectively. Both histologic scores had a strong correlation with a coefficient of 0.9097 (p <0.001). The correlation coefficients between serum ALT and AST vs Knodell scores were 0.3746 (p=0.027) and 0.5278 (p=0.001), respectively. There was no difference in histologic activities between transfusion and sporadic groups with the mean Knodell scores of 7.7500 and 7.6364, respectively. Conclusions: Our results showed that the correlation of the serum HCV-RNA levels and liver histologic activities are very poor. But HCV-RNA levels above 0.6 x 106 copies/ml may indicate severe histologic abnormalities. Serum transaminases levels especially AST have a better correlation with histologic activities but clinical implication of this finding needs to be further determined. This study was partially supported by Roche Thailand Ltd. L0664 ANTI-HBE AND HBVo~A POSITIVE CHRONIC HEPATITIS B: SUCCESSFUL FAMCICLOVIR-INTERFERON THERAPY. A. yon Aretin, U. Oesen*, U. G. Liebert, A. Tannapfel, J. MSssner, F. Berr, Leipzig and *Chemnitz, Germany.
Chronic hepatitis with HBe antigen defective (HBe minus mutant) hepatitis B virus (HBV) (and Anti-HBe & HBVDNA positive serologic profile) is characterized by a progressive clinical course and very poor long-term response to interferon-alpha (IFN) therapy. Reduction of viral replication by
GASTROENTEROLOGY Vol. 114, No. 4
an antiviral drug such as famciclovir (FAM) and subsequent combination therapy with FAM & IFN may enhance the therapeutic response. Case Report: A 34 years old Asian male with long-standing, highly replicative, chronic hepatitis B (HBVDNA 413 pg/ml, HBe negative, amino transferase ALAT 410 to 785 U/L) had failed to respond to a 6 months treatment course with high-dose IFN (3 x 6-9 MU IFN per week). Liver histology showed severe chronic aggressive hepatitis (hepatitis activity index grading score l0 [of maximum 18]) with increased fibrosis of portal tracts and few discrete portal septa (staging score 2 [of maximum 6]). After 2 weeks treatment with FAM (500 mg t.i.d.) serum ALT had decreased to 40% of pretreatment activity and HBVDNA to less than 10 pg/ml, but had remained positive by PCR. Subsequently, on combination therapy with FAM and IFN-alpha (3 x 5 MU per week) serum ALT activity remained 1.5-2fold elevated and HBVDNA positive by PCR. After 7 months a flare of serum transaminases occured and combination therapy was stopped. Subsequently ALT remained below 100 U/I, PCR for HBVDNA consistently negative and HBs antigen positive in serum for meanwhile 17 months consistent with the transition to the nonreplicative phase of the HBV infection. At present liver histology shows minimal chronic aggressive hepatitis (score 5) and unchanged stage of fibrotic changes (score 2) with moderate expression of HBs antigen in approximately 20% of hepatocytes. Combination therapy had been well tolerated and constitutional complaints have faded after HBV clearance from serum. Conclusion: This case of chronic hepatits B with an unfavourable HBe-minus mutant profile shows a surprising long-term response to combination therapy with FAM and IFN. This combination therapy may be a promising approach to chronic hepatitis B nonresponsive to IFN monotherapy, in particular with HBe-minus mutant virus, and is presently under evaluation in an open pilot trial. • L0665 LOCALIZATION OF THE REGION IN THE 5'-FLANKING SEQUENCE OF RAT GLUTATHIONE S-TRANSFERASE (rGST) A2 GENE THAT MEDIATES THE DECREASED EXPRESSION CAUSED BY INTERLEUKIN 6 (IL-6). Susan H. Voss, Richard Whalen, Thomas D. Boyer. Dept of Medicine, Emory University School of Medicine, Atlanta, GA. Introduction: Cytokines, such as IL-6, are known to affect the expression of various hepatic genes by well-characterized response elements in their 5'-flanking sequences. IL-6 in the presence of dexamethasone has been shown to decrease the expression of GSTs in primary cultured rat hepatocytes (Biochem. J. 317:627-632; 1996) and the effect was suggested to be mediated by a known overlapping GRE/NF-IL-6 site in the gene. Aim: To define the region in the 5'-fianking sequence of the rGSTA2 gene that mediates IL-6 suppression. Method: Deletion and mutation constructs of a 1.6 kb region of the 5'-flanking sequence (full promoter) of the rGSTA2 gene cloned into a CAT reporter plasmid were transfected into primary cultured rat hepatocytes in the presence or absence of IL-6. Relative CAT activity was expressed as the ratio of activity in IL-6 treated cells to the activity in untreated cells. Results: CAT activity was reduced by 43% when ceils were transfected with the 1.6 kb CAT construct and grown in the presence of IL-6. Mutations that destroyed the GRE and/or NF-IL-6 sites (bp -1596 to -1615) did not alter the response to IL-6. Deletion of base pairs from -722 to -892 eliminated the suppressive effect of IL-6 on rGSTA2 expression. Construct 1 to -1600 1 to -892, -663 to -892 and 1 to -190 1 to -722 -660 to -722 and 1 to -164
Relative CAT activity + SE 0.57 + 0.06 (n = 8) 0.83 -+ 0.19 (n = 4) 0.64 + 0,23 (n = 4) 1.61 +-0.38 (n = 4) 1.72 __0.68 (n = 4)
Conclusion: The suppressive effect of IL-6 on rGSTA2 gene expression is mediated by response elements between bp -722 to -892 in the 5'-flanking sequence of the gene. This region contains consensus binding sequences for several transcription factors that may mediate this response. The expression of other enzymes of detoxification, such as P450s and other GSTs, are also reduced by IL-6 and the response element which accounts for the effect may be common to the different genes. The exact location of the response element and the identity of factor(s) that bind to it are currently being determined. • L0666 VIRAL KINETIC STUDY OF INDUCTION DOSING WITH INTERFERON ALPHA AND RIBAVIRIN IN THE TREATMENT OF CHRONIC HEPATITIS C. B. Vucelic, R. Ostojic, I. Hrstic, B. Grahovac, M. Scukanec-Spoljar, Zagreb, Croatia.
The viral kinetics was studied in naive hepatitis C patients treated with a combination of high dose interferon alpha and ribavirin for 4 weeks. Therapy was then continued with interferon 3MU TIW and ribavirin daily. The study group included 19 HCV-RNA positive naive patients (9 males and 10 females, mean age 40.21, age range 18-55 years) with a biopsy proven CHC. They were treated daily with combination of interferon alpha (Intron A, Schering-Plough) and ribavirin (Schering-Plough). Group A (10 patients): interferon 10 MU and ribavirin 1000-1200 mg. Group B (9 patients): interferon 5 MU and ribavirin 1000-1200 mg (group B). Serum HCV-RNA
AASLD A1361
April 1998 levels were measured 4 times during the first week (24,48,72,120,168 hours) and twice a week in the remaining 3 weeks (National Genetics Institute, Culver City, CA). Clearance of HCV-RNA was considered as primary virological response. Baseline HCV-RNA levels ranged from 100 to 35.000.000 eq/ml (mean 4.513.689 eq/ml). Genotypes: la - 1 (5.8%), lb - 13 (76.5%), 3a - 2 (11.8%), 6a - 1 (5.9%), not defined in 2 patients due to low serum level of HCV-RNA. Primary virological response at 4 weeks was observed in 9/19 (47.4%) patients, i.e. in 6/10 (60%) patients in group A and in 3/9 (33.3%) patients in group B. Among those who remained positive, the HCV-RNA serum level were reduced at 4 weeks by 99.8% (from mean level of 8.010.000 eq/ml to mean level of 13.760 eq/ml). No serious side effects were observed with exception of severe depression with suicidal ideas in 1 patient who was permanently excluded from the study. 100000oo0
"101:11111011 1OO0O NOof eopl~~
10 1 0
1
2
3
5
7
10
14
17
21
24
28
Days
Daily therapy with interferon alpha and ribavirin was more efficaceous with higher interferon dose regarding primary virological response. Rebounding of viral titre was observed in 9 patients on the 4th day of therapy. • 1,0667 PORTAL VENOUS PRESSURE AND PORTAL VENOUS PRESSURE GRADIENT A R E S I M I L A R IN PATIENTS WITH REFRACTORY ASCITES AND RECURRENT VARICEAL BLEEDING. K Wadwa. SW Paik, A Venbrux, F Osterman, S Mitchell, S Savader, G Lund, PJ Thuluvath. Department of Medicine & Radiology, The Johns Hopkins University, Baltimore, MD. It is not known why certain patients with cirrhosis develop refractory ascites whereas other patients have recurrent variceal bleeding without significant ascites. We speculated that the portal venous pressure threshold for refractory ascites and recurrent variceal bleeding may be different. Methods: In this study we compared direct portal venous pressure (PVP), portal venous pressure gradient (PVPG) (gradient between portal vein and right atrium) and right atrial pressure (RAP) of 119 patients (Child's A 8%, B 56%, C 36%) who were undergoing transjugular intrahepatic porto-systemic shunting (TIPS) for either refractory ascites or recurrent variceal bleeding. TIPS was performed for refractory ascites in 51 patients and for recurrent variceal bleeding in 68 patients. Results: Mean age and Child-Pugh score (9.6 - 1.6 vs 8.6 ± 2.1, P=0.07) of patients with ascites and recurrent bleeding were similar. PVP, PVPG and RAP were similar in Child's B and Child's C cirrhosis. PVP (32.2 ± 8.4 vs 31.3 ± 8.3 mm of Hg)) and PVPG (24.2 _+7.2 vs 23.0 ± 6.7 mm of Hg) were similar between patients with ascites and recurrent bleeding and also when patients were stratified according to their Child's status. RAP (9.3 ± 8.0 vs 7.7 _+5.1) was higher (P = 0.035) in patients with recurrent bleeding compared to patients with ascites. Conclusion: Our findings suggest that portal venous pressure and portal venous pressure gradient are similar in patients with recurrent bleeding and refractory ascites. Higher right atrial pressure observed in patients with recurrent bleeding merits further attention.
• L0668
BASEMENT MEMBRANE PEPTIDES AS MARKERS OF LIVER DISEASE IN CHRONIC HEPATITIS C. KM Walsh. A Fletcher, *RNM MacSween, AJ Morris. *Department of Pathology, Western Infirmary and Departments of Human Nutrition and Gastroenterology, Royal Infirmary, Glasgow, UK. In the normal liver, hepatic sinusoids lack a basement membrane. In liver fibrosis, basement membrane components have been identified around the vessels and in the portal tracts. Serum assays are available which measure the two major components of the basement membrane, type IV collagen and laminin. Aim: To determine if serum levels of type IV collagen or laminin are related to severity of liver disease in chronic hepatitis C. Methods: 37 patients with chronic hepatitis C (CHC) and 5 healthy controls were included in the study. Serum type IV collagen was measured by a one-step sandwich EIA kit (Fuji, Japan) and serum laminin was measured by RIA (CIS, UK). Liver biopsies in patients with CHC were scored using a previously described grading and staging system (Ishak 1994), Liver biopsy scores were related to serum levels of laminin, type IV collagen and alanine aminotransferase (ALT) using Spearman's rank correlation test. Results: Serum type IV collagen was median 127.1 ng/ml (17.7 to 317.4) in CHC patients compared to a median of 61.3 ng/ml (11.5 to 102.3) in controls, p 0.006. Serum laminin was median 1.12 U/ml (0.74 to 2.46) in CHC compared to a median of 0.87 U/ml (0.83 to 1.06) in controls, p 0.07. Serum laminin was correlated with stage of liver fibrosis (13 0.0001) and also with indices of hepatic inflammation such as histological activity index (p 0.002), portal inflammation (p 0.004), periportal necrosis (p 0.015) and confluent necrosis (p 0.04). Serum type 1V collagen was related to liver fibrosis (p 0.007) and to the inflammatory indices, histological activity index (p 0.001), portal inflammation (p 0.004), periportal necrosis (p 0.03) and confluent necrosis (p 0.05). Serum ALT was only related to portal inflammation (p 0.006). Conclusions: Serum assays of basement membrane peptides are accurate non-invasive markers of liver fibrosis and liver inflammation in chronic hepatitis C. These markers are superior to serum ALT in reflecting liver injury in chronic hepatitis C. • L0669 THE MAJOR GALLSTONE GENE (LITH 1) IS RESPONSIBLE FOR BILIARY CHOLESTEROL (CH) HYPERSECRETION AND CH GALLSTONE FORMATION IN CONGENIC MICE (AKR.L-Lithls). David Q-H Wang, Frank Lammert, Beverly Paigen*, Martin C Carey. Brigham and Women's Hospital, Harvard Medical School and Harvard Digestive Diseases Center, Boston, MA; and *The Jackson Laboratory, Bar Harbor, ME. Ch gallstone formation in inbred mice is determined by multiple Lith genes (PNAS, 1995:92:7729). Compared to gallstone-resistant AKR mice, susceptible C57L mice with multiple Lith genes display high bile flow, biliary lipid secretion and Ch to phospholipid (PL) molar ratios as well as rapid Ch crystallization, early onset and high gallstone prevalence (JLR, 1997:38:1395). To explore the contributions of individual genes to gallstone phenotypes, we constructed a congenic strain, AKR.L-Lithl s (A-Lithl), by transferring the gallstone predisposing allele of Lith 1 from the C57L strain into the AKR genetic background. Our aim was to determine which of the phenotypic differences between C57L and AKR mice was transferred along with the Lith 1 allele from C57L mice. Methods: Before and during feeding male C57L, A-Lithl and AKR mice the lithogenic diet containing 15% dairy fat, 1% Ch and 0.5% cholic acid for 56 days, we employed microscopy to quantify crystallization and gallstones in gallbladder biles. In addition, the common bile duct was cannulated for measuring bile flow and biliary lipid secretion rates. Lipid compositions were determined by standard methodology. Results: Table I, gallstone prevalence and hepatic biles (n=5; values expressed as means -+ SD); (Abbr.: BS, bile salts; CSI, Ch saturation index). Conclusions: 1) Since the gallstone and biliary phenotypes in the A-Lithl strain are identical to the C57L parent, Lith 1 plays a predominant role in determining biliary Ch hypersecretion and Ch gallstone formation. 2) Lith 1 does not influence BS-independent bile flow. Soeculation: Since the level of biliary lipid secretion in A-Lithl lies between C57L and AKR, it is likely that other Lith genes affect biliary lipid secretion in mice. Secretion Rates of Biliarv Lioids (umol/hr/k~BW) Gallstones Ch PL BS Mice 56d 0d 56d 0d 56d 0d 56d C57L 80% 12±5 29±7* 30±10" 54±7 163±41" 211±23" A-Lithl 80%** 7±I** 20_+4** 19±2"* 47_+4** 96±13 156±26"* AKR 15% 5_+1 8_+3 14_+3 23_+8 100_+16 66_+27 HowRates(~lJhr/100gBVO CSI Ch/PLRatios 0d 56d 0d 56d 0d 56d C57L 555±125" 648±136" 1.6±0.4 2.0_+0.3 0.37_+0.136 0.55±0.11 A-Lilhl 310_+58 381_+52 1.5±0.3"* 1.6±0.3 0.39_+0.06 0.44±0.09** AKR 276±82 257_+128 1.1_+0.2 1.4±0.2 0.33_+0.08 0.33_+0.04 *P < 0.05, C57L vs. A-Lithl strain; **P < 0.05, A-Lithl vs. AKR mice.
• L0662 DETERMINING THE MOST EFFICACIOUS DOSE OF URSODEOXYCHOLIC ACID IN PRIMARY BILIARY CIRRHOSIS. A Verma*, RP Jazrawi*, HA Ahmed*, T Davis*, M Bland @, M Bensnn #, RT Orchard #, A Theodossit, JD Maxwell*, TC Northfield*. *Division of Medicine & @Statistics, St George's Hospital Medical School, #St Helier Hospital & tMayday University Hospital, UK. Ursodeoxycholic acid (UDCA) improves liver function tests and prolongs survival in primary biliary cirrhosis (PBC). The dose used in the large trials of 13-15mg/kg/day has largely been based on that used for gallstone dissolution. The only dose response study of UDCA in PBC suggested that a dose of 8mg/kg/day was the most efficacious. However, disease stage of the patients was not known, higher doses of UDCA were not tried and there was no "washout period" between the different doses 1. The aim of this study was to determine the most efficacious dose of UDCA in early stage PBC (stage 1 and 2). A power calculation estimated that 20-25 patients were needed to show a 20% difference between doses with a power of 90% at the 5% level, using alkaline phosphatase (ALP), "/ glutamyl transferase (~'GT) and alanine transaminase (ALT) as variables. 24 biopsy proven early stage PBC patients (M:F, 1:23) had five doses of UDCA (0, 300, 600, 900, 1200 mg/day) each for 2 months with a 1 month washout period between doses. Symptoms (pruritus, fatigue, diarrhoea) were assessed on a four point scale. Statistical analysis was carried on the covariance of variables at the end of the doses, with dose and patients as cofactors, and the baseline variable as the covariate. For all the variables UDCA treatment, irrespective of dose, was significantly better than placebo. 900 and 1200rag were better: than both 300 and 600mg using "tGT and total bilirubin as a variable; than 300mg using ALP, protein, IgG and IgM as varaibles; than 600rag using albumin as a variable. No variables showed a significant difference between the UDCA dose of 900 and 1200rag. We therefore conclude that the most efficacious dose of UDCA is 900rag/day. [1] Podda M et al. Effect of different doses of ursodeoxycholic acid in chronic liver disease. Dig Dis Sci 1989; 34, 12, Suppl 59S-65S. L0663 CORRELATION BETWEEN SERUM QUANTITATIVE HEPATITIS-C VIRUS RIBONUCLEIC ACID LEVELS BY POLYMERASE CHAIN REACTION METHOD AND LIVER HISTOLOGIC ACTIVITIES. V. Viriyautsahakul,P. Suwannakul, T. Tirawatnapong,S. Israsena, V. Mahachai. Department of Medicine, Pathology and Microbiology, Chulalongkom Hospital, Bangkok, Thailand. Background & Aim: The aim of this study was to establish whether the serum Hepatitis-C ribonucleic acid ( HCV-RNA ) levels measured by semiquantitative polymerase chain reaction ( PCR ) method have a correlation with the severity of hepatic inflammation graded by standard histologic scores. Methods: Chronic hepatitis-C patients were included in this study on the basis of positive anti-HCV test by second generation enzyme immunoassay and HCV-RNA test by PCR method. Patients with alcohol abuse, coinfected with other viruses and those who were taking immunomodulatory therapy were excluded. Serum samples for HCV-RNA levels and transaminases were taken within 3 months of liver biopsies. Serum samples were measured for HCV-RNA levels by semiquantitative PCR method using Amplicor ® kit. Liver histologic activities were evaluated and scored using Knodell and Scheuer systems by the only one pathologist who was blinded to HCV-RNA and transaminases levels. Results: A total of 38 patients were included in this study. There were 26 male and 12 female patients. 27 (71.1%) patients had previous transfusion history and 11 (28.9%) patients were sporadic. The correlation coefficients between HCV-RNA levels and Knodell and Scheuer scores were 0.2458 (p=0.175) and 0.2370 (p=0.192), respectively. But in 7 patients with HCV-RNA levels above 0.6 x 106 copies/ml, Knodell and Scheuer scores were higher than 6 and 4, respectively. Both histologic scores had a strong correlation with a coefficient of 0.9097 (p <0.001). The correlation coefficients between serum ALT and AST vs Knodell scores were 0.3746 (p=0.027) and 0.5278 (p=0.001), respectively. There was no difference in histologic activities between transfusion and sporadic groups with the mean Knodell scores of 7.7500 and 7.6364, respectively. Conclusions: Our results showed that the correlation of the serum HCV-RNA levels and liver histologic activities are very poor. But HCV-RNA levels above 0.6 x 106 copies/ml may indicate severe histologic abnormalities. Serum transaminases levels especially AST have a better correlation with histologic activities but clinical implication of this finding needs to be further determined. This study was partially supported by Roche Thailand Ltd. L0664 ANTI-HBE AND HBVo~A POSITIVE CHRONIC HEPATITIS B: SUCCESSFUL FAMCICLOVIR-INTERFERON THERAPY. A. yon Aretin, U. Oesen*, U. G. Liebert, A. Tannapfel, J. MSssner, F. Berr, Leipzig and *Chemnitz, Germany.
Chronic hepatitis with HBe antigen defective (HBe minus mutant) hepatitis B virus (HBV) (and Anti-HBe & HBVDNA positive serologic profile) is characterized by a progressive clinical course and very poor long-term response to interferon-alpha (IFN) therapy. Reduction of viral replication by
GASTROENTEROLOGY Vol. 114, No. 4
an antiviral drug such as famciclovir (FAM) and subsequent combination therapy with FAM & IFN may enhance the therapeutic response. Case Report: A 34 years old Asian male with long-standing, highly replicative, chronic hepatitis B (HBVDNA 413 pg/ml, HBe negative, amino transferase ALAT 410 to 785 U/L) had failed to respond to a 6 months treatment course with high-dose IFN (3 x 6-9 MU IFN per week). Liver histology showed severe chronic aggressive hepatitis (hepatitis activity index grading score l0 [of maximum 18]) with increased fibrosis of portal tracts and few discrete portal septa (staging score 2 [of maximum 6]). After 2 weeks treatment with FAM (500 mg t.i.d.) serum ALT had decreased to 40% of pretreatment activity and HBVDNA to less than 10 pg/ml, but had remained positive by PCR. Subsequently, on combination therapy with FAM and IFN-alpha (3 x 5 MU per week) serum ALT activity remained 1.5-2fold elevated and HBVDNA positive by PCR. After 7 months a flare of serum transaminases occured and combination therapy was stopped. Subsequently ALT remained below 100 U/I, PCR for HBVDNA consistently negative and HBs antigen positive in serum for meanwhile 17 months consistent with the transition to the nonreplicative phase of the HBV infection. At present liver histology shows minimal chronic aggressive hepatitis (score 5) and unchanged stage of fibrotic changes (score 2) with moderate expression of HBs antigen in approximately 20% of hepatocytes. Combination therapy had been well tolerated and constitutional complaints have faded after HBV clearance from serum. Conclusion: This case of chronic hepatits B with an unfavourable HBe-minus mutant profile shows a surprising long-term response to combination therapy with FAM and IFN. This combination therapy may be a promising approach to chronic hepatitis B nonresponsive to IFN monotherapy, in particular with HBe-minus mutant virus, and is presently under evaluation in an open pilot trial. • L0665 LOCALIZATION OF THE REGION IN THE 5'-FLANKING SEQUENCE OF RAT GLUTATHIONE S-TRANSFERASE (rGST) A2 GENE THAT MEDIATES THE DECREASED EXPRESSION CAUSED BY INTERLEUKIN 6 (IL-6). Susan H. Voss, Richard Whalen, Thomas D. Boyer. Dept of Medicine, Emory University School of Medicine, Atlanta, GA. Introduction: Cytokines, such as IL-6, are known to affect the expression of various hepatic genes by well-characterized response elements in their 5'-flanking sequences. IL-6 in the presence of dexamethasone has been shown to decrease the expression of GSTs in primary cultured rat hepatocytes (Biochem. J. 317:627-632; 1996) and the effect was suggested to be mediated by a known overlapping GRE/NF-IL-6 site in the gene. Aim: To define the region in the 5'-fianking sequence of the rGSTA2 gene that mediates IL-6 suppression. Method: Deletion and mutation constructs of a 1.6 kb region of the 5'-flanking sequence (full promoter) of the rGSTA2 gene cloned into a CAT reporter plasmid were transfected into primary cultured rat hepatocytes in the presence or absence of IL-6. Relative CAT activity was expressed as the ratio of activity in IL-6 treated cells to the activity in untreated cells. Results: CAT activity was reduced by 43% when ceils were transfected with the 1.6 kb CAT construct and grown in the presence of IL-6. Mutations that destroyed the GRE and/or NF-IL-6 sites (bp -1596 to -1615) did not alter the response to IL-6. Deletion of base pairs from -722 to -892 eliminated the suppressive effect of IL-6 on rGSTA2 expression. Construct 1 to -1600 1 to -892, -663 to -892 and 1 to -190 1 to -722 -660 to -722 and 1 to -164
Relative CAT activity + SE 0.57 + 0.06 (n = 8) 0.83 -+ 0.19 (n = 4) 0.64 + 0,23 (n = 4) 1.61 +-0.38 (n = 4) 1.72 __0.68 (n = 4)
Conclusion: The suppressive effect of IL-6 on rGSTA2 gene expression is mediated by response elements between bp -722 to -892 in the 5'-flanking sequence of the gene. This region contains consensus binding sequences for several transcription factors that may mediate this response. The expression of other enzymes of detoxification, such as P450s and other GSTs, are also reduced by IL-6 and the response element which accounts for the effect may be common to the different genes. The exact location of the response element and the identity of factor(s) that bind to it are currently being determined. • L0666 VIRAL KINETIC STUDY OF INDUCTION DOSING WITH INTERFERON ALPHA AND RIBAVIRIN IN THE TREATMENT OF CHRONIC HEPATITIS C. B. Vucelic, R. Ostojic, I. Hrstic, B. Grahovac, M. Scukanec-Spoljar, Zagreb, Croatia.
The viral kinetics was studied in naive hepatitis C patients treated with a combination of high dose interferon alpha and ribavirin for 4 weeks. Therapy was then continued with interferon 3MU TIW and ribavirin daily. The study group included 19 HCV-RNA positive naive patients (9 males and 10 females, mean age 40.21, age range 18-55 years) with a biopsy proven CHC. They were treated daily with combination of interferon alpha (Intron A, Schering-Plough) and ribavirin (Schering-Plough). Group A (10 patients): interferon 10 MU and ribavirin 1000-1200 mg. Group B (9 patients): interferon 5 MU and ribavirin 1000-1200 mg (group B). Serum HCV-RNA
AASLD A1361
April 1998 levels were measured 4 times during the first week (24,48,72,120,168 hours) and twice a week in the remaining 3 weeks (National Genetics Institute, Culver City, CA). Clearance of HCV-RNA was considered as primary virological response. Baseline HCV-RNA levels ranged from 100 to 35.000.000 eq/ml (mean 4.513.689 eq/ml). Genotypes: la - 1 (5.8%), lb - 13 (76.5%), 3a - 2 (11.8%), 6a - 1 (5.9%), not defined in 2 patients due to low serum level of HCV-RNA. Primary virological response at 4 weeks was observed in 9/19 (47.4%) patients, i.e. in 6/10 (60%) patients in group A and in 3/9 (33.3%) patients in group B. Among those who remained positive, the HCV-RNA serum level were reduced at 4 weeks by 99.8% (from mean level of 8.010.000 eq/ml to mean level of 13.760 eq/ml). No serious side effects were observed with exception of severe depression with suicidal ideas in 1 patient who was permanently excluded from the study. 100000oo0
"101:11111011 1OO0O NOof eopl~~
10 1 0
1
2
3
5
7
10
14
17
21
24
28
Days
Daily therapy with interferon alpha and ribavirin was more efficaceous with higher interferon dose regarding primary virological response. Rebounding of viral titre was observed in 9 patients on the 4th day of therapy. • 1,0667 PORTAL VENOUS PRESSURE AND PORTAL VENOUS PRESSURE GRADIENT A R E S I M I L A R IN PATIENTS WITH REFRACTORY ASCITES AND RECURRENT VARICEAL BLEEDING. K Wadwa. SW Paik, A Venbrux, F Osterman, S Mitchell, S Savader, G Lund, PJ Thuluvath. Department of Medicine & Radiology, The Johns Hopkins University, Baltimore, MD. It is not known why certain patients with cirrhosis develop refractory ascites whereas other patients have recurrent variceal bleeding without significant ascites. We speculated that the portal venous pressure threshold for refractory ascites and recurrent variceal bleeding may be different. Methods: In this study we compared direct portal venous pressure (PVP), portal venous pressure gradient (PVPG) (gradient between portal vein and right atrium) and right atrial pressure (RAP) of 119 patients (Child's A 8%, B 56%, C 36%) who were undergoing transjugular intrahepatic porto-systemic shunting (TIPS) for either refractory ascites or recurrent variceal bleeding. TIPS was performed for refractory ascites in 51 patients and for recurrent variceal bleeding in 68 patients. Results: Mean age and Child-Pugh score (9.6 - 1.6 vs 8.6 ± 2.1, P=0.07) of patients with ascites and recurrent bleeding were similar. PVP, PVPG and RAP were similar in Child's B and Child's C cirrhosis. PVP (32.2 ± 8.4 vs 31.3 ± 8.3 mm of Hg)) and PVPG (24.2 _+7.2 vs 23.0 ± 6.7 mm of Hg) were similar between patients with ascites and recurrent bleeding and also when patients were stratified according to their Child's status. RAP (9.3 ± 8.0 vs 7.7 _+5.1) was higher (P = 0.035) in patients with recurrent bleeding compared to patients with ascites. Conclusion: Our findings suggest that portal venous pressure and portal venous pressure gradient are similar in patients with recurrent bleeding and refractory ascites. Higher right atrial pressure observed in patients with recurrent bleeding merits further attention.
• L0668
BASEMENT MEMBRANE PEPTIDES AS MARKERS OF LIVER DISEASE IN CHRONIC HEPATITIS C. KM Walsh. A Fletcher, *RNM MacSween, AJ Morris. *Department of Pathology, Western Infirmary and Departments of Human Nutrition and Gastroenterology, Royal Infirmary, Glasgow, UK. In the normal liver, hepatic sinusoids lack a basement membrane. In liver fibrosis, basement membrane components have been identified around the vessels and in the portal tracts. Serum assays are available which measure the two major components of the basement membrane, type IV collagen and laminin. Aim: To determine if serum levels of type IV collagen or laminin are related to severity of liver disease in chronic hepatitis C. Methods: 37 patients with chronic hepatitis C (CHC) and 5 healthy controls were included in the study. Serum type IV collagen was measured by a one-step sandwich EIA kit (Fuji, Japan) and serum laminin was measured by RIA (CIS, UK). Liver biopsies in patients with CHC were scored using a previously described grading and staging system (Ishak 1994), Liver biopsy scores were related to serum levels of laminin, type IV collagen and alanine aminotransferase (ALT) using Spearman's rank correlation test. Results: Serum type IV collagen was median 127.1 ng/ml (17.7 to 317.4) in CHC patients compared to a median of 61.3 ng/ml (11.5 to 102.3) in controls, p 0.006. Serum laminin was median 1.12 U/ml (0.74 to 2.46) in CHC compared to a median of 0.87 U/ml (0.83 to 1.06) in controls, p 0.07. Serum laminin was correlated with stage of liver fibrosis (13 0.0001) and also with indices of hepatic inflammation such as histological activity index (p 0.002), portal inflammation (p 0.004), periportal necrosis (p 0.015) and confluent necrosis (p 0.04). Serum type 1V collagen was related to liver fibrosis (p 0.007) and to the inflammatory indices, histological activity index (p 0.001), portal inflammation (p 0.004), periportal necrosis (p 0.03) and confluent necrosis (p 0.05). Serum ALT was only related to portal inflammation (p 0.006). Conclusions: Serum assays of basement membrane peptides are accurate non-invasive markers of liver fibrosis and liver inflammation in chronic hepatitis C. These markers are superior to serum ALT in reflecting liver injury in chronic hepatitis C. • L0669 THE MAJOR GALLSTONE GENE (LITH 1) IS RESPONSIBLE FOR BILIARY CHOLESTEROL (CH) HYPERSECRETION AND CH GALLSTONE FORMATION IN CONGENIC MICE (AKR.L-Lithls). David Q-H Wang, Frank Lammert, Beverly Paigen*, Martin C Carey. Brigham and Women's Hospital, Harvard Medical School and Harvard Digestive Diseases Center, Boston, MA; and *The Jackson Laboratory, Bar Harbor, ME. Ch gallstone formation in inbred mice is determined by multiple Lith genes (PNAS, 1995:92:7729). Compared to gallstone-resistant AKR mice, susceptible C57L mice with multiple Lith genes display high bile flow, biliary lipid secretion and Ch to phospholipid (PL) molar ratios as well as rapid Ch crystallization, early onset and high gallstone prevalence (JLR, 1997:38:1395). To explore the contributions of individual genes to gallstone phenotypes, we constructed a congenic strain, AKR.L-Lithl s (A-Lithl), by transferring the gallstone predisposing allele of Lith 1 from the C57L strain into the AKR genetic background. Our aim was to determine which of the phenotypic differences between C57L and AKR mice was transferred along with the Lith 1 allele from C57L mice. Methods: Before and during feeding male C57L, A-Lithl and AKR mice the lithogenic diet containing 15% dairy fat, 1% Ch and 0.5% cholic acid for 56 days, we employed microscopy to quantify crystallization and gallstones in gallbladder biles. In addition, the common bile duct was cannulated for measuring bile flow and biliary lipid secretion rates. Lipid compositions were determined by standard methodology. Results: Table I, gallstone prevalence and hepatic biles (n=5; values expressed as means -+ SD); (Abbr.: BS, bile salts; CSI, Ch saturation index). Conclusions: 1) Since the gallstone and biliary phenotypes in the A-Lithl strain are identical to the C57L parent, Lith 1 plays a predominant role in determining biliary Ch hypersecretion and Ch gallstone formation. 2) Lith 1 does not influence BS-independent bile flow. Soeculation: Since the level of biliary lipid secretion in A-Lithl lies between C57L and AKR, it is likely that other Lith genes affect biliary lipid secretion in mice. Secretion Rates of Biliarv Lioids (umol/hr/k~BW) Gallstones Ch PL BS Mice 56d 0d 56d 0d 56d 0d 56d C57L 80% 12±5 29±7* 30±10" 54±7 163±41" 211±23" A-Lithl 80%** 7±I** 20_+4** 19±2"* 47_+4** 96±13 156±26"* AKR 15% 5_+1 8_+3 14_+3 23_+8 100_+16 66_+27 HowRates(~lJhr/100gBVO CSI Ch/PLRatios 0d 56d 0d 56d 0d 56d C57L 555±125" 648±136" 1.6±0.4 2.0_+0.3 0.37_+0.136 0.55±0.11 A-Lilhl 310_+58 381_+52 1.5±0.3"* 1.6±0.3 0.39_+0.06 0.44±0.09** AKR 276±82 257_+128 1.1_+0.2 1.4±0.2 0.33_+0.08 0.33_+0.04 *P < 0.05, C57L vs. A-Lithl strain; **P < 0.05, A-Lithl vs. AKR mice.