Posters – Sunday May 21, 2006  8:00AM  Hepatitis C

 

Epidemiology

 

Abstract S1039 – Emerging Importance of Chronic Hepatitis C in Asian Americans

J. T. CHENG; C. Hsien; H. J. Sun; M. J. Tong

 

Background

Studies on hepatitis C in the United States included only small numbers of patients with Asian heritage and revealed that Asian Americans tend to have acquired hepatitis C early in their life. We examined the epidemiology and natural history of chronic hepatitis C in the Asian American population at our tertiary referral center.

 

Patients and Methods

A retrospective survey was conducted on 254 chronic hepatitis C patients of Asian descent, who presented to the Liver Center, Huntington Medical Research Institutes.

 

Results

The mean follow-up period was 62.4 ± 54.2 SD months. There were 135 males (53.1%) and 119 females. Mean age at presentation was 57.3 ± 13.5 SD years. Among these 254 Asian American patients, 7.87% were born in the United States while the remainder were born in Asia (27.95% in Vietnam, 17.72% in China, and 14.57% in Taiwan). The majority were ethnic Chinese (60.6%), followed by Korean (10.2%), Vietnamese (10.2%), and Japanese (9.4%). A majority of patients (51.2%) reported a history of unsanitized medical injections in Asia, 40.6% reported a history of transfusion, 4.3% a history of intravenous drug abuse, 2.0% a history of multiple needle sticks from healthcare-related work, and 1.2% reported history of tattoo as the only risk factor. Based on available history, 31.5% of patients reported Vietnam as the probable place of HCV transmission, 17.7% reported Taiwan, 17.7% United States, and 7.9% Korea. Of 208 patients tested, 63.9% had genotype 1. A total of 129 patients received interferon-based therapy. Based on their response to therapy, 44.2% were sustained virologic responders (SVR), 15.5% responder-relapsers (RR), and 24.0% non-responders (NR). In terms of genotypes and treatment response, 26.8% of patients with genotype 1 treated were SVR. In patients with non-genotype 1, 70% were SVR. There were 37 patients who presented with HCC while 25 developed HCC during follow-up. Mean age at the time of HCC diagnosis was 68.2 ± 9.0 SD years. Only 2 HCC patients had HBV co-infection. Of 51 deaths that occurred during the follow-up, 82.4% were due to HCC, 7.8% due to cirrhosis, and the remainder due to non-liver-related deaths. Mean age of death from HCC was 70.6 ± 8.9 years. The annual rate of developing HCC in chronic hepatitis C patients was 4.89%.

 

Conclusions

A majority of Asian Americans with chronic hepatitis C reported a history of unsanitized medical injection as the only risk factor for mode of transmission. Genotype 1 was the most prevalent HCV genotype detected and appeared to have a lower rate of SVRs compared to other genotypes. HCC developed frequently in our chronic hepatitis C patients of Asian descent.


Abstract  S1040 – Initial Content Validation of a Disease-Targeted Health-Related Quality of Life (HRQOL) Instrument in Chronic Hepatitis B Virus (HBV) Infection: The HBQOL v1.0

B. M. Spiegel; P. Martin; R. Bolus; S. Han; E. Esrailian; J. Talley; F. Kanwal

 

Background:

Despite the increasing realization that HRQOL is an important outcome in chronic HBV, there are no validated, disease-targeted instruments currently available. Investigators and clinicians must therefore rely on either generic HRQOL instruments or traditional biological outcomes in order to assess their patients’ health status. However, generic instruments may fail to capture the most important components of HRQOL in HBV, and biological outcomes fail to acknowledge patient perception. We therefore sought to develop and validate a disease-targeted HRQOL instrument in HBV: the HBQOL v1.0. This abstract describes the candidate items and scales resulting from our initial content validation process.

 

Methods:

“Content validity” is the degree to which an instrument contains a representative range of items and scales relevant to the disease under study. If an instrument fails to measure attributes of importance to patients with the targeted disease, then it will likely fail to accurately measure HRQOL. In order to establish content validity for our instrument, we initially conducted a search of MEDLINE for previously published HRQOL instruments in both chronic viral hepatitis and other forms of chronic liver disease. We then compiled a pool of items with relevance to HBV based on a priori hypotheses and clinical judgment. We next convened a panel of 5 expert hepatologists experienced in HBV. Using a semi-structured protocol, we first elicited the domains perceived by the panel as most relevant in HBV. We then presented the items identified by the systematic review, and asked the panel to comment on the content, breadth, and relevancy of the list. Finally, in concert with 3 psychometricians, we developed a conceptual model of the scales in HBV.

 

Results:

We selected 4 scales on the basis of their content validity and potential responsiveness (i.e. ability to detect HRQOL change after successful treatment): (1) Anticipation Anxiety/Psychological Well Being (e.g. fear of developing cancer or cirrhosis), (2) Sexual Well Being/Intimacy (e.g. transmission concern, impact on sexuality), (3) Disease Stigma/Social Well Being (e.g. embarrassment, concern someone influential may find out, shame), and (4) Daily Functioning (e.g. impact on diet or medication use).

 

Conclusions:

Four scales may capture HRQOL in HBV across a range of psychological and social factors. The breadth and depth of biopsychosocial symptoms in HBV highlights the significant HRQOL burden of this condition. Ongoing research includes testing these scales with patient focus groups, and prospectively measuring the construct validity and reliability of this evolving instrument.

 


Abstract S1041 – Sero-prevalence of hepatitis B and C in hospital personnel

A. Hila; R. Bouali; R. Belaaj; A. Belhadj; F. Khediri

 

Aim:

To assess the seroprevalence of hepatitis B and C in hospital personnel during a hepatitis B vaccination update campaign.

 

Methods:

A prospective study of 1895 personnel working at our hospital Tunisa). These represent 95.2% of our total personnel, which are 1992 people. The study was performed from June 2004 to August 2004. Subjects included in this study were all personnel of the hospital, including doctors, nurses, technicians, administrative and environmental personnel. All study subjects had blood drawn for: antigen (Ag) HBs, antibody (Ab) anti-HBc, Ab anti-HBs and Ab anti-HCV.

Mean age = 41 yo, range 20-60 years old;  male to female ratio = 1.9 (990/905).

 

Results:

Prevalence of HBs Ag and isolated Ab anti-HBc was respectively 1.95% (37) and 4.6% (88). 14 subjects (0.74%) were positive for Ab anti-HCV.

860 personnel (45.3%) had no HBV markers, and were thus vaccinated. Only 528 personnel (27.8%) were immune to HBV due to prior immunization, and 240 of these had Ab anti-HBs titers < 100mU/ml, requiring a repeat dose of vaccine. Of 65 cases with isolated Ab anti-HBc +, 50 (77%) developed Ab anti-HBs after only 1 dose of vaccine. 6 out of the 15 non-responders (9.2 % of the total) were HBV DNA positive. Of the 14 cases with Ab anti-HCV +, 3 had known chronic hepatitis C and were previously treated with a sustained virologic response, 7 were RNA PCR negative and 4 were positive.

 

Conclusion:

Prevalence of both hepatitis B and C in our hospital are not higher than in the general population in our country. Most of cases with isolated Ab anti-HBc positivity were immunized by disease (occult hepatitis: blood DNA positivity < 10%). 50% of cases with + Ab anti-HCV had + RNA. This vaccination campaign allowed us to vaccinate 45% of the personnel who were not immunized for hepatitis B.


Abstract S1042 – Prevalence and predictors of high body mass index among native born and immigrant patients with chronic hepatitis C in Canada

W. Chen1; T. Wong; J. Heathcote; M. Krahn

 

Background:

High body mass index (BMI) among patients with chronic hepatitis C (HCV) is common as is hepatic steatosis on liver biopsy.

 

Objective:

To compare the prevalence of high BMI between patients with HCV with that in the general population and to investigate predictors of BMI in patients with HCV.

 

Methods:

HCV patients with age 25-64 years at first visit between 1990 and 2004 were included. Medical records was reviewed retrospectively to extract: age, gender, ethnicity, place of birth (for immigration status), height, weight, alcohol abuse (> 50 g/d over 10 years), heavy smoking (1 ppd > 10 years), mode of transmission of HCV, estimated duration of infection (infection assumed to start from birth if transmission mode unknown), genotype, liver biopsy (fibrosis stage, presence of severe steatosis over 30%), and co-morbidity.

 

Data analysis:

Unpaired student t test was applied for continuous outcomes; chi square test was applied for frequency outcomes; univariate and multivariable linear regression were conducted to explore predictors for BMI.

 

Results:

1502 patients met inclusion criteria. Our patients with HCV had a significantly higher prevalence of high BMI above 29.9 (23.2% vs. 16.1%; P<0.001) than the general population (Statistics Canada 2001). Among 446 patients with a documented liver biopsy, 301 patients were native born and 145 patients were immigrants to Canada. Compared to native born, immigrants were older (49.4 years vs. 47.3 years, P=0.011), less often Caucasian (64.1% vs. 99.0%, P<0.001), consumed less alcohol and were less often heavy smokers (9.7% vs.23.6%, P<0.001), they had longer duration of HCV (35.2 years vs. 27.7 years, P<0.001), less often use injection drugs (16.6% vs. 57.1%, P<0.001), and were more like to have no identified HCV transmission mode (43.5% vs. 10.3%, P<0.001); they had less depression (4.83% vs. 10.96%, P=0.034), were less often hemophiliic (1.38% vs. 8.31%, P=0.004), and were less often co-infected with HIV (0% vs. 3.32%, P=0.02). Multivariable regression analyses indicated that hypertension (coefficients 3.796, P<0.001) and diabetes (coefficients 2.622, P=0.013) were significantly associated with BMI among the native-born, while advanced hepatic fibrosis (coefficients 1.766, P=0.026), severe hepatic steatosis (coefficients 4.069, P=0.001), and hypertension (coefficients 5.595, P<0.001) were significantly associated with BMI among immigrant patients.

 

Conclusion:

High BMI (>29.9) is more prevalent among patients with HCV than the general population in Canada. Data demonstrate a relationship between BMI and fibrosis progression in immigrants to Canada with HCV, but not in those who were native-born.  

 


Abstract S1043 – Polymorphisms in TLR7 and 8 affect severity of hepatic fibrosis and inflammation in individuals infected with the hepatitis C virus from a single source.

C. A. Goulding; R. Mc Manus; A. Murphy; G. S. Mac Donald; M. Dring; J. Hegarty; S. Mc Kiernan; D. Kelleher

 

Background and Aims:

Toll-like receptors (TLRs) are type I transmembrane proteins, which are highly conserved through evolution. They activate innate immunity by recognizing and binding to a wide variety of pathogenic substances. TLR7 and 8 both respond to the synthetic imidazoquinoline compounds, known to have anti-viral properties. TLR7 and probably TLR8 have also been demonstrated to recognize single stranded viral RNA. This study aimed to assess the involvement of inherited variations in TLR7 and 8 in determining disease outcome in HCV infected individuals.

 

Methods:

223 women from Ireland were all exposed to HCV genotype 1b from a single donor, and including 85 who had spontaneously cleared the virus and 138 chronically infected, were genotyped for TLR 7 and 8 polymorphisms and haplotype tagging was performed. The frequencies of these polymorphisms were then compared with disease activity and severity.

 

Results:

TLR 7 and 8 genotypes were compared with HCV PCR status, ALT levels and liver histology. There was no association between HCV PCR status and the TLR polymorphisms. Of the 5 SNPs examined in TLR8, wild types were associated with significantly worse fibrosis for 2 SNP loci (TLR8C (C401T); 1.38 vs 0.78, p = 0.03, TLR8D (C426T); 1.39 vs. 0.70, p = 0.018). Degree of inflammation was worse in TLR8B (A796C) wild type; 5.1 vs. 4.2, p = 0.008, and this group also had higher ALT levels; 66 vs. 41, p = 0.018. TLR7B (C149T) was associated with lower inflammation and 4.4 vs. 5.7, p = 0.015 ALT levels; 47.8 vs. 73.4, p = 0.014. On analysis of haplotypes, those containing the minor allele for TLR8A or TLR8B had significantly less fibrosis (p = 0.025, p < 0.01, respectively). Haplotypes containing the minor allele for TLT7B had significantly lower inflammatory scores, p< 0.025. No haplotype was associated with viral clearance.

 

Conclusion:

Consistent with prior functional data regarding viral clearance, HCV viral clearance was not associated with any of the TLR 7 or 8 SNPs examined in this study. However, SNPs in TLR 7 and 8 were associated with a lesser degree of hepatic inflammation and fibrosis. Further functional characterisation of these SNPs could provide important information on the response of the host to HCV infection.

 


Abstract S1044 – Discordant IRF-3 Activation and Hepatic Interferon-Stimulated Gene Expression Associates with Immune Cell Infiltration in Chronic Hepatitis C

D. T. Lau; P. Fish; S. M. Lemon; G. Michael

 

Alpha/beta Interferons (IFNs) play important role in viral infections. HCV encodes proteins that block IFN production and antagonize IFN actions to support viral persistence. The NS3/4A protease of HCV disrupts viral activation of IRF-3, thereby attenuating IFN production and IFN-stimulated gene (ISG) expression. [Foy et al, Science 2003; 300:1145-48].

 

In this study, we compared the transcriptional profiles of ISGs among liver biopsy samples from normal controls, HCV genotype-1 patients and patients with nonalcoholic fatty liver disease (NAFLD) using Affymetrix HG-U95A Human GeneChips. Liver biopsy slides were also examined by immunofluorescence staining and confocal microscopy analysis to define the subcellular distribution of IRF-3, the extent of viral protein expression in hepatocytes, and the composition of hepatic infiltrating immune cells.

 

HCV samples had significantly higher transcriptional levels of ISGs associated with IFN such as OAS2, MX1,ISG56 compared to controls and NAFLD (p<0.005). The level of IRF-3, however, was similar in HCV and normal controls. HCV NS3 or NS5A proteins were observed within focal areas of biopsy specimen and distributed in a perinuclear context within hepatocytes. IRF-3, though abundant, was typically found in an inactive state as defined by its cytoplasmic-bound distribution in liver cells. Furthermore, HCV infection was associated with specific CD3-positive T cell and plasmacytoid dendritic cell (PDC) infiltrates in biopsy specimens.

 

Our results indicated that in chronic HCV, IRF-3 activation within infected hepatocytes is limited, which lends further evidence of NS3/4A disruption of IRF-3 activation. Despite the lack of active IRF-3, increased hepatic ISG expression in HCV infection was observed and was correlated with the presence of immune cell infiltrates in liver samples. Infiltrating immune cells could be an important source of alpha/beta IFNs that potentially influence hepatic ISG expression and the outcome of HCV infection.

 


Abstract S1045 – Can we predict advanced fibrosis in daily practice based on common blood tests?

R. Cheung; S. Currie; H. Shen; T. Morgan; K. Hu; S. Ho; N. Brau; E. Bini; T. Wright

 

Background:

Several indexes based on simple lab tests correlated with liver biopsies read by expert pathologists in highly selected settings; applicability to less stringent, community-based practice is unclear.

 

Aim:

To compare the specificity, sensitivity and ROC of these indexes with liver biopsies in a cohort of 548 veterans with chronic hepatitis C from 24 centers nationwide enrolled in a previous study.

 

Method:

All lab tests including interpretation of the liver biopsy were done locally. The following indexes were calculated and correlated with a 5-point fibrosis stage (F0-F4): platelet <100 x109/L, AAR (AST/ALT), Pohl score (positive if AAR>1 and platelet <150x109/L), APRI ([AST/ULN]/platelet [x109/L]x100, and “Model 3” (log odds [predicting cirrhosis]=-5.56-0.0089 x platelet (x103/mm3)+1.26xAAR+5.27xINR) (Lok et al, 2005).

 

Results:

This cohort was predominately male with 24% blacks, and distribution fibrosis stage of 0,1,2,3,4 were 11%,24%,28%,24%,13%, respectively. When patients with mild fibrosis (F0-2) were compared to those with advanced fibrosis (F3-4), the area under the ROC were 0.531 for platelet count alone, 0.527 for AAR, 0.537 for Pohl score, 0.746 for APRI and 0.754 for model 3. Recent alcohol use within 12 months did not affect the performance of these models.

 

Conclusions:

AAR, Pohl and platelet counts have limited ability to predict advanced fibrosis when compared to liver biopsy. However, APRI > 1.5 has a high negative predictive and positive predictive value (0.71 and 0.70). In addition, the newly described model 3 with cutoff value of <0.2 has a high sensitivity for excluding and >0.5 has high specificity for predicting advanced fibrosis. These scoring systems may be useful in daily clinical practice and deserve further prospective validation studies.

 

Variable Name

Sensitivity

95% CI of Sensitivity

Specificity

95% CI of specificity

Positive predictive value

95% CI of PPV

Negative predictive value

95% CI of NPPV

AAR>=1.0

0.217

0.163-0.282

0.798

0.749-0.839

0.391

0.301-0.489

0.630

0.582-0.676

Pohl score

0.096

0.060-0.148

0.979

0.955-0.991

0.731

0.520-0.877

0.644

0.600-0.686

APRI<0.5

0.123

0.082-0.177

0.631

0.577-0.682

0.165

0.111-0.235

0.548

0.498-0.598

APRI>=1.5

0.376

0.307-0.443

0.904

0.867-0.932

0.697

0.601-0.780

0.708

0.663-0.750

APRI<1.0

0.397

0.330-0.468

0.204

0.163-0.251

0.228

0.186-0.276

0.363

0.296-0.435

APRI>=2.0

0.235

0.180-0.300

0.939

0.907-0.961

0.696

0.572-0.798

0.674

0.630-0.716

Model 3

<0.2*

0.931

0.884-0.962

0.321

0.270-0.376

0.454

0.404-0.505

0.886

0.810-0.936

Model 3

>0.5

0.508

0.435-0.581

0.848

0.802-0.885

0.669

0.585-0.744

0.740

0.691-0.784

 


Abstract S1046 – C-Methacetin Breath Test as Quantitative Liver Function Test in Patients with Chronic Hepatitis C: Continuous Automatic Molecular Correlation Spectroscopy Compared to Isotopic Ratio Mass Spectrometry

O. Goetze; N. Selzner; M. A. Kwiatek; M. Fried; T. Gerlach; B. Muellhaupt

 

Background and aims:

The 13C-methacetin breath test (MBT) has been proposed for the non-invasive evaluation of hepatic microsomal activity. Up to now, “gold standard” of stable isotope analysis is isotopic ratio mass spectrometry (IRMS). The aim of this study was to test a new continuous online automatic breath collection and analysis system in comparison to IRMS in patients with chronic hepatitis C.

 

Methods: Sixteen patients with chronic hepatitis C infection at different METAVIR fibrosis stages (F0, n=2; F1, n=6; F2, n=0; F3, n=1; F4, n=7) were studied. After an overnight fast each subject received 75 mg of 13C-methacetin dissolved in 100 ml of water. The 13C/12C ratio of breath samples was analyzed over 120 min both by molecular correlation spectroscopy with approximately one sample/3min (Oridion, BreathID LTD, Israel) and by IRMS (Analytical Precision Limited, AP2003, UK) every 10 min. Results were expressed as delta over baseline (DOB [‰]) at each time interval and maximal DOB (DOBpeak[‰]). The association between both methods was tested by linear regression analysis and by Bland-Altman analysis after spline interpolation of the 13CO2 exhalation curves and for the detected peak height.

 

Results: A high positive linear association between both analytical methods was observed (DOB: R2=0.95, p<0.001, DOBpeak: R2=0.96, p<0.001). For all DOB values the bias was -0.27 ‰ with a standard deviation (SD) of 1.7 ‰ and limits of agreement of -3.7 ‰ and 3.2 ‰. For DOBpeak the bias was -0.97 ‰ with a SD of 2.6 ‰ and greater limits of agreement (-6.2 ‰ and 4.2 ‰), which were caused by a poor peak detection with IRMS in 4 cases (figure).

 

Conclusions: The MBT obtained by molecular correlation spectroscopy with continuous online automatic breath collection and analysis system is an easy to operate method in patients with chronic hepatitis C infection. It provides results comparable to the “gold standard” isotopic ratio mass spectrometry and a better peak detection due to the higher sampling frequency.

 

 


Abstract S1047 – 13C-Methacetin Breath Test by Online Molecular Correlation Spectroscopy Compared to APRI and Liver Biopsy for the Assessment of Fibrosis in Chronic Hepatitis C

O. Goetze; N. Selzner; A. Grau; M. Fried; T. Gerlach; B. Muellhaupt

 

Background:

Stable isotope breath tests have been developed for the non-invasive assessment of microsomal liver function in patients with chronic liver disease. Among different tests investigated, the 13C-methacetin breath test (MBT) appears to be particularly suitable for the rapid assessment of hepatic functional reserve.

 

Aims:

To assess prospectively the performance of MBT in patients with chronic hepatitis C infection using molecular correlation spectroscopy with a continuous, online, automatic breath collection and analysis system (Oridion BreathID, LTD Israel) and to compare MBT outcomes with aspartate transaminase to platelets ratio index (APRI) and with the METAVIR fibrosis score of a liver biopsy specimen as a “gold standard”.

 

Methods:

60 patients (37 M, 48.1 ± 9.8 y., BMI 24.2 ± 3.6 kg/m2, AST 1.5 ± 1.2, ALT 2.2 ± 2.0 x upper limit of normal, platelet count 173 ± 81 x 103/mm3) with chronic hepatitis C were studied (fibrosis stage F0, n=8; F1, n=22; F2, n=11; F3, n=6; F4, n=13; obtained within six months of MBT). After an overnight fast each patient received 75 mg of 13C-methacetin dissolved in 100 ml of water. The 13C/12C ratio was determined in each breath sample over 90 minutes by molecular correlation spectroscopy (1sample/3min) as delta over baseline (DOB[‰]) and was expressed as maximal (PDRpeak[%/h]) as well as cumulative percentage dose of 13C recovered at 30 min (cPDR30[%]).

 

Results:

Both PDRpeak and cPDR30 in patients with F≤1 were higher than in patients with F≥2 (PDRpeak: 29.6 ± 9.8 vs 18.7 ± 11.9%/h, cPDR30: 9.7 ± 3.3 vs 5.7 ± 3.9%, p<0.001). Mean areas under the receiver operating characteristic (ROC) curve of PDRpeak, cPDR30 and APRI values were similar for F≥2 (0.82, 0.87, 0.77) and for F≥3 (0.93, 0.95, 0.88). For F=4 area under ROC was higher for cPDR30 than APRI (0.94 vs 0.81, p<0.05) and similar for PDRpeak (0.91).

 

Conclusions:

MBT by continuous automatic molecular correlation spectroscopy is an easy to use and effective method for assessing liver fibrosis, with a better performance than APRI score for detection of severe stages of fibrosis and with at least a similar performance to other reported non-invasive surrogate marker of liver fibrosis.

 


Abstract S1048 – Increased serum gammaglutamyltranspeptidase activity: a surrogate marker of non alcoholic hepatic steatosis in chronic hepatitis C?

F. Benini; L. Bercich; M. G. Pigozzi; L. Romanini; A. Reggiani; P. Donati; A. Pozzi; F. Lanzarotto; C. Ricci; A. Lanzini

 

Predictors of poor response to antiviral treatment in chronic hepatitis C include increased serum gammaglutamyltranspeptidase activity (γGT). Histopathology, constitutional and viral factors, or alcohol consumption may be involved in this effect, but little information is available. The aim of our study was to assess factors affecting γGT activity in patient with chronic hepatitis C by carefully selecting patients with no present or past history of alcohol intake.

 

We selected 63 consecutive patients with biopsy proven chronic hepatitis C and no history of alcohol consumption. We measured pretreatment anthropometric parameters and insuline resistance (HOMA IR) in addition to conventional virological and serological liver tests. Forty patients were also tested for small intestinal bacterial overgrowth using glucose H2 breath test. Liver histology was classified according to Knodell and hepatic steatosis according to Brunt. All patients were treated with PEG-interferon alfa 2-b (1.5 μg/kg/weekly) plus ribavirin (800 to 1200 mg/day according to body weight).

 

Thirty-eight patients had pretreatment γGT > 1 the upper limit of normal, and 25 had values within the normal range. There was no difference in pretreatment viral load and genotype distribution among the 2 groups. Peptide-C (mean + SD: 2.98 ± 1.66 ng/mL vs 2.04 ± 0.90 ng/mL, p=0.0175), insuline resistance (2.83 ± 1.9 vs 1.79 ± 1.12, p=0.023) and hepatic steatosis score (0.78 ± 0.5 vs 0.22 ± 0.43, p=0.001) were significantly higher in patients with high than in those with normal γGT. No patient tested positive at glucose H2 breath test Insuline resistance (r=0.467, p<001), hepatic staging (r=0.313, p <0.05) and steatosis (r=0.399, p<0.007) were significantly related (Pearson correlation) to serum γGT. Hepatic steatosis was the only parameter independently related to serum GGT (r=0.510, p <0.007) at multiple regression analysis.

 

In conclusion, a substantial proportion of patients with chronic hepatitis C that are not alcohol consumers have serum γGT activity above the upper limit of normal. This phenomenon is independent of constitutional or virological characteristics and is associated with hepatic steatosis at histopathology, suggesting that increased serum GGT activity may represent a surrogate marker of hepatic non alcoholic steatosis.


Abstract S1049 – Sinusoidal Lymphocytosis as a Marker for Cryoglobulinemia in Hepatitis C Liver Biopsies

S. Carmack; W. Ahrens; P. Ravichandran; P. Hui; M. Robert; T. Taddei; P. Mistry; D. Jain

 

Background:

Cryoglobulinemia (CryoG) is commonly seen with hepatitis C virus (HCV) and is associated with a higher grade of fibrosis, increased incidence of low-grade non-Hodgkin lymphoma, and poorer response to therapy. The development of CryoG may be a marker of chronic antigenic stimulation by HCV that in some cases leads to clonal B-cell proliferation and lymphoma. Currently no histological findings in liver biopsy are known to correlate with the presence of CryoG in HCV infected patients. We have anecdotally noted prominent sinusoidal lymphocytosis in CryoG liver biopsies. The goal of this study is to determine whether this could be used as a marker of CryoG in liver biopsies from chronic HCV patients.

 

Design:

10 chronic HCV patients with CryoG who underwent biopsy from 1998-2005 were identified from the liver clinic database. 10 HCV CryoG-negative cases matched for age and stage of fibrosis were included as controls. Histological features (sinusoidal lymphocytes, inflammatory activity, acidophil bodies, fibrosis stage), and clinical and laboratory data (SPEP, LFT's, HCV viral load, EBV status, treatment), were evaluated. Formalin-fixed paraffin embedded sections were stained for CD3, CD20 and CD68. Sinusoidal lymphocytes were counted in 5 HPF on H&E, CD3 and CD20 immunostains. CD68+ Kuppfer cells were counted in a similar fashion.

 

Results:

The mean and standard deviation (SD) of fibrosis stage, inflammatory grade, and lymphocyte/Kuppfer cell counts are shown in the table. CryoG-positive cases were significantly correlated with increased sinusoidal T-cell lymphocytosis (P=0.028) as compared to CryoG-negative cases. There were no differences in the two groups with other histological parameters.

 

Conclusion:

CryoG-positive liver biopsies show a prominent sinusoidal T-cell lymphocytosis compared to CryoG-negative biopsies. These T-cells probably imply increased antigenic stimulation and may play a role in the pathogenesis of CryoG. However, this needs further investigation. Evaluation of this feature on the diagnostic work-up of liver biopsies may have important implications with regard to further work-up, treatment, and follow-up of these patients.

 

 

Stage

Grade

H&E Lymphocytes

CD3+ cells

CD20+ cells

CD68+ cells

CryoG + (mean +/- SD)

2.4 +/- 1.2

1.8 +/- 0.9

247 +/- 183

318 +/- 244

42 +/- 54

239 +/- 46

CryoG- (mean +/- SD)

2.4 +/- 0.5

2.1 +/- 0.7

180 +/- 42

113 +/- 50

15 +/- 7

220 +/- 51

 

 


Abstract S1050 – Liver fibrosis and necroinflammatory activity in chronic hepatitis C patients with persistently normal aminotransferases

A. M. Loaeza; F. Sanchez-Avila; J. Gallegos-Orozco; E. Oviedo; M. Weimersheimer; M. Sanchez; A. Meixueiro; J. Garcia; G. Castro; A. Montano; F. Vargas-Vorackova; M. Uribe

 

 

Introduction:

Liver fibrosis (LF) progression in chronic hepatitis C (CHC) patients is considered to be linear and accelerated at advanced stages. Natural history of liver fibrosis in patients with normal ALT levels is uncertain, in liver biopsy minimal inflammation and LF are often found, so a less aggressive form of disease has been speculated, nonetheless advanced lessions such as bridging fibrosis and cirrhosis have been reported.

 

Aim:

To compare LF stage and degree of necroinflammatory activity (IA) in CHC patients with normal ALT versus CHC patients with elevated ALT.

 

Method:

CHC treatment-naive patients with a liver biopsy and known duration of infection were included. CHC with normal ALT was defined as HCV-RNA detectable in serum and at least 3 ALT values in the normal range within a 6 months period. All liver biopsies were evaluated according to METAVIR score by one blinded pathologist. LF, IA and LF progression were compared between both groups. Advanced LF was defined as a METAVIR >= F2.

 

Results:

158 patients were included, mean age 50 +/- 12 years-old, 101 (64%) were female. 96 (61%) had elevated ALT and 62 (39%) had normal ALT, the mean ALT value in both groups was 139 UI/L and 47.8 IU/L (P < 0.0001) respectively. The LF progression rate was 0.102 and 0.125 METAVIR points/year (P=0.352), with a mean duration of infection of 22.4 vs 26 years (P=0.07) in patients with normal and elevated ALT respectively. 31 patients with normal ALT (50%) and 58 patients with elevated ALT (60.4%) had advanced LF (P=0.25). Differences in LF and IA between groups are shown in the table.There was an association between elevated ALT levels and the presence of A3 (OR 4.6, 95% CI, 1.02-21.5; p<0.05) and F4 (OR: 5.9, 95% CI, 2.9-12.1; p<0.001), conversely normal ALT levels were associated with F0 (OR:0.4, 95% CI, 0.17–0.94; p < 0.05).

 

Conclusions.

In the studied population, severe IA and cirrhosis were more frequent in patients with elevated ALT. Although the absence of LF was more frequent in patients with normal ALT, half of them showed significant fibrosis. In the abscense of validated non-invasive LF markers in patients with persistently normal ALT, a liver biopsy should be considered, specially in those with a long term infection.

 

Fibrosis

METAVIR

F0

(n=28)

F1

(n=41)

F2

(n=23)

F3

(n=17)

F4

(n=48)

Normal ALT

n (%)

16 (26)

15 (24)

10 (16)

9(15)

12 (19)

Elevated

ALT

n (%)

12 (12.5)

26 (27)

13 (13.5)

8 (8.5)

36 (37.5)

P

<0.05

0.7

0.5

0.2

<0.001

Necroinflammation

 

Necroinflammatory

activity

METAVIR

A0

(n=17)

A1

(n=81)

A2

(n=45)

A3

(n=15)

-

Normal ALT

n (%)

10 (16)

34 (55)

16 (26)

2 (3)

-

Elevated

ALT

n (%)

7 (7)

47 (49)

29 (30)

13 (14)

-

P

0.08

0.5

0.5

<0.05

-

 


Abstract S1051 – High accuracy pre-treatment prediction of Early Viral Response to combined pegIFN/ribavirin therapy in HCV-infected patients by a novel method of gene expression analysis: a preliminary study.

J. Alsobrook; P. Hraber; C. Harris; L. Davis; P. Doherty; B. Griffith; T. Williams; S. Arora

 

Background

Current predictors of Sustained Viral Response (SVR) to HCV therapy have limited power. Accurate prediction of treatment response prior to therapy would have great clinical utility. Exagen’s goal is to discover small sets of genomic biomarkers for prognostic testing. This pilot study was designed to discover a small suite of genes whose combined expression pattern distinguishes patients with a 12-week Early Viral Response (EVR) from non-responders.

 

Patients

Protocols were IRB approved; informed consent was obtained from all patients prior to enrollment. Clinical data and blood samples were collected from 47 HCV-infected patients.

 

Analysis

Total mRNAs were isolated, and cDNAs were prepared, from pre-treatment blood samples and a reference sample derived from a commercial source of normal tissues. All cDNAs were assayed with a custom microarray representing 24,000 human genes. Array data were analyzed with Exagen’s proprietary software which uses an empirical, data-driven computational methodology with no prior assumptions about biological relationships, whereby the predictive classification accuracy of gene combinations are evaluated directly without regard to (or assessment of) the accuracy of individual genes. The significance of any resultant combinatorial classifier is evaluated by an iterative random permutation process. Patients were classified as responder or non-responder by EVR status (undetectable viral RNA / zero viral titer or detectable viral titer, respectively). Of the 47 cases enrolled, 3 had incomplete data. The remaining 44 cases were randomly assigned to a training set and a test set. The training set contained 25 responders & 8 non-responders for use in a classifier search, and the test set contained 8 responders & 3 non-responders for use in classifier assessment.

 

Results

Multiple classifier sets derived from pre-treatment blood samples gave high overall accuracy for predicting EVR; one gene appeared consistently among the best classifiers. A 2-gene classifier achieved an overall 97% accuracy in the training set (32 out of 33 correctly classified), and an overall 81.8% accuracy in the test set (9 out of 11 correctly classified). This successful pilot study for pre-treatment prediction of HCV EVR suggests that our method may be successfully applied to HCV SVR, particularly in light of the high negative predictive value of EVR for SVR outcome.

 

Conclusion

This successful pilot study for pre-treatment prediction of HCV Early Virological Response will be extended in a larger clinical sample.  The same discovery method may also be successfully applied to HCV Sustained Viral Response, particularly in light of the high negative predictive value of EVR for SVR.

 

 

 

Genotype 1

Genotype 2

Genotype 3

EVR

+

20

7

6

EVR

-

11

0

0

 

 

Training EVR

Test EVR

Classifier Prediction

 

+

-

+

-

 

+

24

0

7

1

 

-

1

8

1

2

 

 

 


Abstract S1052MDM2 Promoter SNP309 is Associated with Hepatocellular Carcinoma in Japanese Patients with Chronic Hepatitis C

N. Dharel; N. Kato; R. Muroyama; M. Moriyama; R. Shao; T. Kawabe; M. Omata

 

Background:

Hepatitis C virus (HCV) infection remains a major cause of chronic hepatitis, liver cirrhosis and hepatocellular carcinoma (HCC) worldwide. Over 80% of all HCC in Japan are caused by chronic HCV infection. We have previously reported several gene polymorphisms associated with HCC development in patients with chronic hepatitis C (Wang et al. Hepatology 2003, Wang et al. Clin Cancer Res 2004, Kato et al. Hepatology 2005). Recently, a single nucleotide polymorphism (SNP309) in the promoter region of MDM2 gene, which negatively regulates tumor suppressor p53, has been shown to be associated with accelerated tumor formation in both hereditary and sporadic cancers in humans (Bond et al. Cell 2004). Here, we evaluated the association of MDM2 SNP309 with HCC in patients with chronic hepatitis C.

 

Patients and Methods:

We genotyped the SNP309 at the MDM2 promoter in 435 Japanese patients with chronic HCV infection including 187 patients with HCC, as well as 48 healthy volunteers, using a fluorogenic polymerase chain reaction (TaqMan SNP genotyping assay). The result of the SNP assays were also verified by direct sequencing of some randomly selected samples.

 

Results:

The genotype distribution of SNP309 among Japanese population including 435 patients with chronic hepatitis C and 48 healthy controls were as follows: (T/T - 22%, T/G - 51% and G/G - 27%) and were different from the original report in Caucasian population (48%, 40% and 12%, respectively) (Bond et al. Cell 2004). Among the chronic hepatitis C patients, the proportion of G/G genotype of the SNP309 in 187 patients with HCC (33%) was significantly higher than in the 248 patients without HCC (23%), with an odds ratio of 2.28 (95% confidence interval 1.30 – 3.98). A multivariate analysis revealed that MDM2 SNP309 (G/G vs. T/T), age over 60, male gender, presence of cirrhosis, serum alpha-fetoprotein >20 μg/L, and serum albumin <3.2 gm/dL were independently associated with the HCC development at odds ratio of 2.27, 2.46, 3.08, 4.15, 4.87, and 6.33, respectively.

Conclusions: MDM2 promoter SNP309 is associated with development of HCC in Japanese patients with chronic hepatitis C. The G allele of MDM2 SNP309 could serve as an important genetic marker for the risk of HCC among patients with chronic hepatitis C.


Abstract S1054Comparative study concerning the severity of hepatic lesions at the moment of diagnosis in patients with Chronic B Viral Hepatitis vs. patients with Chronic C Viral Hepatitis

R. Sirli; I. Sporea; A. Popescu; M. Danila; M. Cornianu

 

Aim:

The aim of this paper is to compare the severity of hepatic lesions evaluated by means of liver biopsy (LB) at the moment of diagnosis, in patients with Chronic B viral hepatitis vs. patients with Chronic C viral hepatitis.

 

Material and method:

We evaluated 846 consecutive patients diagnosed in our department with chronic B and C viral hepatitis. In all these patients we performed LB that was evaluated by means of Knodell score. We compared the demographic data of the two groups, as well as the severity of hepatic lesions (histology activity index – HAI, and the fibrosis score).

 

Results:

From the total of 846 patients, 247 (29.2%) had chronic B hepatitis and 599 (70.8%) chronic C hepatitis. The mean age at diagnosis was 39.9+/-12.2 in HBV patients and 47.7 +/-10.7 in HCV patients (p<0.0001). From the 247 HBV patients, 39.7% (98) were female, while from the 599 HCV patients, 59.3% (355) were female (p<0.0001).

 

Regarding the severity of liver damage on LB, the mean HAI score was 7.14+/-3.35 in HBV patients and 8.11 +/- 2.86 in HCV patients (p<0.0001). The mean fibrosis score was 1.10+/-1.15 in HBV patients and 1.3 +/- 1.18 in HCV patients (p=0.02 S).

 

In HBV patients there were no statistically significant differences between men and women regarding the severity of hepatic lesions: mean HAI score 7.13 +/- 3.2 in women and 7.14 +/- 3.4 in men (p=0.93); mean fibrosis score 1.09+/-0.12 in women and 1.12+/-0.09 in men (p=0.66).

 

Also, in HCV patients there were no statistically significant differences between men and women regarding the severity of hepatic lesions: mean HAI score 8.06 +/- 2.9 in women and 8.13 +/- 2.72 in men (p=0.68); mean fibrosis score 1.27+/-1.16 in women and 1.34+/-0.22 in men (p=0.56).

 

Conclusions:

1. In our group chronic C viral hepatitis was 2.5 times more frequent than B chronic hepatitis (599 vs. 247 cases).

2. The age at diagnosis was significantly lower in HBV patients as compared to HCV patients (39.9 vs. 47.7, p<0.0001).

3. C chronic hepatitis was significantly more frequent in women than in men (female-male ratio = 1.45:1), as compared to B chronic hepatitis (female-male ratio=1:1.52) (p<0.0001)

4. There were no statistically significant differences between men and women regarding the severity of hepatic lesions, in both HBV and HCV groups of patients.

5. The hepatic lesions were significantly more severe in patients with chronic C hepatitis than in those with chronic B hepatitis, regarding both the mean HAI score (8.11 vs. 7.14, p<0.0001), and the mean fibrosis score (1.3 vs. 1.1, p=0.02).


Abstract S1055 – Non-invasive Biomarkers of Liver Fibrosis in Hemophilia Patients with Hepatitis C: Can You Avoid Liver Biopsy?

Y. Maor; D. Bashari; G. Kenet; A. Lubetsky; J. Luboshitz ; J. M. Schapiro ; G. Penaranda ; S. Bar-Meir; U. Martinowitz; P. Halfon

 

Introduction:

Liver biopsy remains the gold standard for the evaluation of fibrosis despite its risks and limitations, especially in hemophilia patients. Recently, non-invasive biomarkers were used to assess histological features. The most thoroughly evaluated biomarker is the Fibrotest (FT) (AUROC 0.80 for fibrosis stages F2F3F4 vs. F0F1).

 

Aim:

To assess liver fibrosis in hemophilia patients with HCV using non-invasive biomarkers without liver biopsy. Methods: One-hundred and thirty two hemophilia patients (124 males, mean age 39 ± 14 years) with anti-HCV antibodies were evaluated. These patients were stratified into several groups: patients with features of advanced liver disease- 7, persistently HCV RNA-negative- 21, persistently normal LFT's- 24, HCV/HIV co-infected- 27. The following biomarkers of fibrosis were used: FT, AST-to-platelet ratio index (APRI), Forns index, age-platelet index and hyaluronic acid. The obtained scores were correlated with the clinical features of the patients.

 

Results:

Estimated by the FT, the distribution of the stage of fibrosis in the 132 patients was: F0F1=65% (86/132), F2=5% (7/132), F3=13% (17/132) and F4=17% (22/132). Using FT, all patients with clinical suspicion of advanced liver disease were classified as F3F4, whereas patients with persistently HCV RNA-negative were all classified as F0F1. Twenty-one percent (5/24) of the patients with persistently normal LFT's had fibrosis stage F3F4. The proportion of F3F4 among HCV/HIV co-infected patients was significantly higher than among HCV mono-infected (52% vs. 33%; p=0.05). Concordance of 3 or more biomarkers was present in 43% (57/132) of the patients. Liver biopsy could be avoided in 70% of the patients using a practical assumption that if FT is in concordance with APRI and/or Forns than there is a concordance with liver biopsy. The concordance rate for patients with presumably advanced or minimal liver disease was excellent (100% and 95%, respectively).

 

Conclusions:

In our hemophilia patients infected with HCV, FT identified correctly advanced or minimal liver disease. Discordance among the various biomarkers of fibrosis was considerate; nevertheless, practical combination of FT, APRI, and Forns may predict stage of fibrosis with accuracy, avoiding liver biopsy in 70% of the patients.

 


Abstract S1056 Steatosis associated with more severe fibrosis in in chronic hepatitis C.

K. Corey; A. K. Bhan; R. T. Chung

 

Background and Aims

Recent work has suggested that steatosis contributes to more advanced fibrosis in patients with chronic hepatitis C (CHC). The etiology of this steatosis varies by genotype. In genotype 3 steatosis the hepatitis C virus (HCV) is believed to initiate steatosis while in genotype 1 the steatosis is likely metabolic in origin. These etiologies of steatosis may have important implications for adjunctive therapy in HCV. We sought to assess the relationship between steatosis and fibrosis in a cohort of patients with CHC who underwent liver biopsy and further establish the relationship between fibrosis and steatosis when limited to genotype 1.

 

Methods

A retrospective chart review of 223 patients with CHC and liver biopsy was undertaken to assess relationship between steatosis and fibrosis. HCV RNA, genotype, AST, ALT, presence of diabetes mellitus, hypertension and hyperlipidemia were recorded.

 

Biopsies were analyzed by a single pathologist (AKB) and graded for necroinflammatory activity and fibrosis staging according to Ishak et al. as well as Brunt steatosis score. Steatosis was scored on a scale of 0-4, with 0 = no steatosis, 1= <5%, 2=5-33%, 3=34-66% and 4=>66% of hepatocytes with steatosis.

 

Results

Steatosis was observed in 66% of CHC patients in this group. Fibrosis was found on liver biopsy in 77% of patients. Mean fibrosis score was 2.51+/- 1.47 and the mean steatosis score 0.99 +/-0.9. Twenty-seven percent and 66% of genotype 1 pts had grade 2+ and 1+ steatosis, respectively. Thirty percent and 78% of genotype 3 patients had grade 2+ and 1+ steatosis, respectively. While an absolute correlation between steatosis score and fibrosis stage was not observed, when fibrosis was dichotomized to low fibrosis (stage 0-2) and severe fibrosis (grade 3-6) a significant relationship with steatosis was seen. (Pearson Chi2 = 11.9, p = 0.008). A significant relationship between steatosis and fibrosis was also observed when the data was confined to genotype 1 (p=0.05).

 

Conclusion

This finding that increased steatosis is associated with worsening fibrosis suggests a possible role for steatosis in the acceleration of liver disease in HCV patients, especially in genotype 1 patients. Efforts to control steatosis may therefore have an important role in halting HCV liver disease progression, particularly in persons who are nonresponders to antiviral therapy.