11/13/2005 – Posters – Diagnosis and Natural History

Abstract ID: 63904

HCV: Diagnosis and Natural History

Histologic progress in paired biopsies between 15^th and 25^th years after a Hepatitis C Genotype 1B Single Source Outbreak in Germany.

Schiefke, University of Leipzig, Leipzig, Germany, M. Wiese, Muncipal Hospital St.Georg, Leipzig

Introduction

The cohort of German women infected with hepatitis C virus (HCV) genotype 1b via contaminated anti-D immunoglobulin in 1978 /79 represent a well characterized group to investigate the natural course of HCV infection. Studies of HCV lb infection with a high rate of cirrhosis face studies that showed slow progression of fibrosis.

Methods

Between 1994 and 2004 a total of 530 biopsies were performed. In our study only the paired biopsies between 15^th - 25^th year were analysed after the defined HCV 1b infection (from August 1978 until March 1979) due to HCV contaminated anti-D immunglobulin. In this interim analysis 76 patients were included.

Results

After 25 years, 61/76 (80%) women still tested positive for HCV RNA. This group had a significant higher histological inflammation score (grading) than those without HCV RNA in serum (responder to antiviral therapy; 15/76 (20%)) (p=0.001), whereas staging (fibrosis) was not significant different. Rates of fibrosis progression differ markedly between HCV RNA status between 15^th - 25^th year. Responder to therapy did not show a progress in staging (p=0.70). In the natural course of patients staging deteriorates during study period (p=0.001). In addition, sustained responder to pegylated interferon and ribavirin therapy had a lower grading compared with natural course patients (p=0.001). Repeated biopsy (interval >5 years) showed only minor changes in histologic signs of fibrosis in successful treated patients – 3/15 (20%) improved by 1 stage, 8/15 (53%) of paired biopsies remained unchanged and 2/15 (13%) had only a minor increase in fibrotic changes (+1 stage). None of the patients without successful treatment decreased in staging. In 30/61 (49%) increased by 1 stage, 24/61 (39%) had a marked increase in fibrotic changes (>+1 stage), and only 7/61 (11%) of paired biopsies remained unchanged. In 2/61 (3%) a cirrhosis was found.

Conclusions

A benign course of HCV infection with low disease progression was observed in this unique homogenous group of patients 25 years after infection. The progress of fibrosis was usually low in successful treated patients and significant higher in patients with natural course in paired biopsies between 15^th - 20^th year of infection. Patients with a natural course of infection had also a significant higher fibrotic score (with two cirrhotics) than patients with a sustained response to antiviral therapy. Patients with long term Hepatitis C infection may benefit also from late therapy.

Abstract ID: 62948

HCV: Diagnosis and Natural History

Spontaneous Fibrosis Regression in Untreated, Noncirrhotic, Persistently Viremic Irish Women with Chronic Hepatitis C Contracted Initially from Immunoglobulin Anti-D: Final Report.

R. Levine, SUNY Upstate Medical University, Syracuse, NY, S. Sanderson, Mayo Clinic, School of Medicine, Rochester, MN, R. Ploutz-Snyder, State University of New York, Syracuse, NY, F. Murray, Beaumont Hospital, Dublin, Ireland, D. Manning, Beaumnot Hospital, Dublin, Ireland, J. Hegarty, St. Vincent's Hospital, Dublin, Ireland, N. Nolan, St. Vincent's Hospital, Dublin, Ireland, D. Kelliher, St. James' Hospital, Dublin, Ireland, G. McDonald, St. James' Hospital, Dublin, Ireland, J. O'Keane, Mater Misericordiae University Hospital, Dublin, Ireland, J. Crowe, Mater Misercordiae University Hospital, Dublin, Ireland

Introduction

After 17 years of follow-up, Kenny Walsh (NEJM 1999;340:1228-33) described a benign course of 390 untreated Rh- women infected in 1977 with genotype 1 chronic hepatitis C (HCV) via contaminated immunoglobulin anti-D. The incidence of cirrhosis was only 1.9%, based on single liver biopsies in 1994. In 1996 we initiated a retrospective prospective study in 184 patients of the original Dublin cohort to provide further insight into the natural history of HCV. Our earlier abstracts of this cohort have been published previously (Gastroenterology 2000; 118: A944; Hepatology 2001; 34: 229A).

Methods:

Baseline 1994 biopsies were compared with sequential biopsies up to 8 years duration to document histological progression or regression. Using strict qualifying biopsy criteria (size > 15 mm, mean baseline biopsy length = 16.3 mm, portal areas ³ 5), paired biopsies, 33% of which had 3-5 sequential biopsies, were assessed by Ishak score by a pathologist unaware of the biopsy sequence. Grade worsening or improvement (increase or decrease of ³ 2 points) and stage progression or regression (increase or reduction of ³ 1 point) were correlated statistically, the latter also with digital image analysis for fibrosis %.

Results

At baseline, mild fibrosis (stages 0-3) was present in 82% and advanced fibrosis (stages 4-5) in only 18%. None had stage 6 cirrhosis, therefore all could potentially progress. Grade and stage scores decreased/increased significantly in 28/18% and 24/27% of patients, respectively, over a mean 3-year interval (range 1-8 years). A change of 2-3 stages was greater with fibrosis progression (31%) than with regression (18%). There was a significant positive correlation between baseline and sequential grade/stage scores (r = 0.39, P < .001), and between semi-quantitative Ishak scores and fibrosis % as measured by digital image analysis (Spearman’s rho = 0.85, P < .01), suggesting that sampling error was uncommon. Baseline ALT (mean 49, range 23-328 U/L) positively correlated with changes in grade (r = .41, P < .01) and stage (r = .39, P < .01) for up to 8 years, suggesting that baseline ALT is potentially a good predictor of inflammatory/fibrotic changes. Confounding variables (alcohol, smoking, herbal and paracetamol use) were not significantly correlated with histological outcomes.

Conclusions

Implications for this subcorhort of patients with minimally progressive disease suggest that even with elevated enzymes, 24% showed some degree of fibrosis regression. With inclusion of 2 cirrhotic biopsies at 6 and 8 years post-baseline, only 2.2% developed cirrhosis after 27 years. Given that the majority of our patients are in their fourth and fifth decades, we recognize that some may still be at risk for advanced liver disease, but for most we consider the outcome unlikely.

Abstract ID: 65581

HCV: Diagnosis and Natural History

Spontaneous HCV clearance in patients followed in a Haemophila Unit: Evaluation of cellular immunity against HCV recombinant proteins.

M. Bes, Blood Bank, BST, Barcelona, Spain, S. Sauleda, Blood Bank, Barcelona, Spain,

M. Martorell, Hemophilia Unit, Barcelona, Spain, J. I. Esteban, Liver Unit, Hospital

Vall d’Hebron, Barcelona, Spain, J. Quer, Liver Unit, Hospital Vall d’Hebron, Barcelona, Spain, R. Parra, Hemophilia Unit, Barcelona, Spain, J. Guardia, Liver Unit, HosptialVall d’Hebron, Barcelona, Spain, V. Vargas, Liver Unit, Hospital Vall d’Hebron,Barcelona, Spain

Aims:

To determine the rate of spontaneous HCV clearance in a population of

haemophiliacs followed in a specialized unit, to assess the influence of HIV

coinfection in this clearance and to caracterize the cellular immune response

in comparison with anti-HCV treated patients with sustained virological

response (SVR) and with non-responders (NR).

Patients and methods

We studied a cohort of 163 patients followed at a Haemophilia Unit, all of them with proven exposure to HCV infection (116 haemophilia A, 25 haemophilia B and 22 other bleeding disorders). 80 patients were HIVnegative and 88 HIV positive. In all patients we tested HCV-RNA in a plasma sample before any antiviral treatment (COBAS Amplicor v2.0, Roche Diagnostics) and recorded treatment outcome in treated patients. We collected PBMCs from patients with spontaneous clearance (SC) and from a subset of patients with SVR and NR. We also included 14 healthy blood donors as a negative controls. Cellular immune response against HCV core and NS3 recombinant proteins was evaluated by ELISPOT-IFNg (Mabtech). Results are expressed as median IFNg-producing cells/100.000 cells (IFNg-PC). Comparisons among groups were made with U-Mann Whitney.

Results

14 out of 163 patients were found to have spontaneously cleared (SC) HCV-RNA, 10 out of 79 (12.6 %) HIV (-) patients and four of 80 (5 %) HIV (+) (p: N.S.). All 10 HIV (-) who cleared the virus had persistently normal transaminase levels; conversely, two of four HIV (+) patients who cleared the virus presented elevated transaminase level probably due to antiretroviral therapy. Cellular immune response against HCV protein was evaluated in 12 patients with SC, 14 patients with NR and 12 patients with SVR. Patients that spontaneously cleared HCV infection showed significantly higher NS3 IFNg production than healthy controls (p=0.001), and similar to SVR patients (see figure). Non-responders did not have significant NS3 IFN-g production. As for HCV core protein, all patient groups had similar IFN-g response. Irrespectively of HCV outcome, HIV (+) patients showed weaker cellular response against HCV NS3 and core proteins than HIV (-) patients.

Conclusions

· Rate of spontaneous HCV-RNA clearance in haemophilic patients was lower than reported in general HCV population. In our study, spontaneous HCV RNA clearance was also detected in patients with HIV coinfection. HCV-specific cellular response was observed in spontaneously recovered patients with similar intensity to patients with sustained virological response to anti-HCV treatment.

· HCV specific cellular response was observed in patients who spontaneously cleared HCV infection after more than a decade of exposure and with similar intensity to patients with treatment mediated resolved infection.

· Patients with chronic HIV/HCV coinfection had significantly weaker cellular immune response than monoinfected patients.

Abstract ID: 66879

Category: JO3: HCV: Diagnosis and Natural History

Determination of hepatitis C Virus(HCV)infection status in screen-reactive dialysis patients: what is the best testing strategy?.

S. Rosenblum, Spectra Laboratories, Rockleigh, NJ, B. Lim, Spectra East Laboratories,

Rackleigh, NJ, C. Schaper, independent statical consultant, Philadelphia, PA, R. Belen,

Spectra East Laboratories, Rockleigh, NJ, R. Deats, Spectra East Laboratories,

Rockleigh, NJ, R. Kaamino, Spectra West Laboratories, Fremont, CA, G. Gusewtich,

Spectra West Laboratories, Fremont, CA, L. Comanor, indpendent research consultant,

Truckee, CA

Objectives

To determine an effective testing strategy for predicting antibody and viremic status of anti-HCV positive dialysis patients.

Materials and Methods

We selected dialysis patient specimens using a stratified design based on historical Abbott HCV EIA 2.0 (Abbott) (Abbott Laboratories, Abbott Park, Il) signal to cut-off ( s/co) ratios. Screening results for 424 patient specimens were obtained from 2 enzyme immunoassays, Abbott and Ortho HCV v. 3.0 ELISA, (Ortho Diagnostics, Raritin, N.J.) (Ortho), and one chemiluminescent assay, ADVIA Centaur HCV 3.0 (Centaur) (Bayer Healthcare LLC, Tarrytown, N.Y.). An additional 45 patient specimens were screened only by Abbott and Ortho. All specimens were then tested by VERSANT HCV Qualitative Assay (TMA)1 (Bayer Healthcare LLC) and RIBA HCV 3.0 SIA (Chiron Corp., Emeryville,CA). As testing all screen-reactive specimens by both TMA and RIBA is inefficient, we evaluated 2 algorithms for predicting the combined antibody/viremic status of screen-reactive specimens:

1) RIBA as the initial 2^nd line test followed by TMA only if the RIBA result is indeterminate (RIBA/TMA) and 2) TMA as the initial 2^nd line test followed by RIBA only if the TMA result is non-reactive (TMA/RIBA). Specimens with a single determinate result were predicted to belong to a patient with matching antibody/viremic status. A prediction was deemed correct if the algorithm-based predicted status matched the observed status. All estimates were weighted averages of stratum level estimates with the weights based on historical s/co values. Weighting was necessary because sampling proportions differed substantially across strata. We test ~240,000 specimens annually. We process 92 reportable samples by TMA in 5.2 hours compared to 27 samples by RIBA in 7.7 hours.

Results

The table shows the annual estimated number of screen- reactive patients by 3 anti-HCV assays, the estimated percent that would be correctly classified using 2 algorithms, and the annual testing hours associated with each scheme.

Testing with TMA followed by RIBA results in > 99.6% correct classification.

Conclusions

TMA/ RIBA algorithm classifies more patients correctly in less testing time than does RIBA/TMA. Screening with Centaur followed by TMA/RIBA appears to be the most efficient testing scheme.

¹approved indications include detection of HCV RNA as evidence of active infection in anti-HCV reactive individuals

Abstract ID: 62205

Category: JO3: HCV: Diagnosis and Natural History

Does Type and Duration of Antiretroviral Therapy Attenuate Liver Fibrosis in HIV/HCV Coinfected Patients?.

S. Verma, University of Southern California, Los Angeles, CA, C. Wang, University of Southern California, Los Angeles, CA, S. Govindarajan, Rancho Los Amigos Medical Center, Downey, CA, G. Kanel, University of Southern California, Los Angeles, CA, K. Squires, University of Southern California, Los Angeles, CA, M. Bonacini, Department of Transplantation, California Pacific Medical Center, Sanan Francisco, CA

To determine if type and duration of HIV therapy attenuates liver fibrosis in HIV and hepatitis C (HCV) coinfected patients.

Patients/methods:

Patients with HCV mono infection (Group 1) and HIV/HCV coinfection were retrospectively recruited, the latter classified into: received no therapy or only nucleoside reverse transcriptase inhibitors (NRTI) (Group 2), received highly active antiretroviral therapy (HAART) (Group 3), initially received NRTI followed by HAART after 1996, (Group 4). Fibrosis stage (0-6) and necroinflammatory (NI) score (0-18) were assessed according to Ishak system.

Results:

381 patients (HCV mono infected n=296, and HIV/HCV coinfected n=85) were recruited for the study. Mean duration of HIV therapy prior to liver biopsy in Groups 2, 3 and 4 was 3.8 + 2.8 yrs, 3.3 + 1.8 yrs and 6.6 + 2.2 yrs. Group 4 patients had received 2.7 + 1.1 years of NRTI followed by 3.9 + 1.8 years of HAART. Time from HIV diagnosis to HAART initiation was significantly shorter in Group 3 vs Group 4 (9.1 + 7.3 mths vs 34.1 + 13 mths, p<0.0001). Group 1 and 3 had similar fibrosis stage (3 .1 + 2 vs 3.4 + 2.4), rates of fibrosis progression ( 0.13 + 0.09/yr vs 0.16 + 0.11/yr), NI scores (6.1 + 1.8 vs 6.1 + 2.0), prevalence (33% vs 41%) and mean time to cirrhosis (27.4 + 6.8 yrs vs 25.1 + 7.1 yrs). Groups 2 and 4 had significantly more advanced HCV related liver disease as regards fibrosis stage (4.6 + 1.8 & 4.3 + 2.0), p<0.0009; fibrosis progression (0.24 + 0.11/yr & 0.20 + 0.10/yr), p<0.0001 and prevalence of cirrhosis (68% & 60%), p<0.0009 compared to Group 1. Probability of developing cirrhosis after 25 years was 16% in Groups 1 compared to 72% in Group 2 (p<0.0001), 24% Group 3 (p=0.02) and 38% Group 4 (p<0.0001).

Conclusion:

Coinfected subjects who receive HAART as soon as possible after HIV diagnosis (for a mean of 3.3 years) have HCV related disease severity comparable to HCV monoinfected subjects. Similar degree of benefit is not observed in those on no therapy/NRTI or HAART after NRTI, despite longer duration of therapy.

· It is not just the presence or duration of HAART, but the timing of its initiation that positively influences liver fibrosis in HIV/HCV subjects.

· The patients most likely to benefit are those whose HIV was diagnosed after 1996 and therefore only HAART rather than NNRTI or NNRTI sequentially followed by HAAER the severity of HCV related liver disease in this group is comparable with HCV monoinfected patients.

· This attenuation of HCV related liver disease appears to be chiefly related to immune restoration after HAART introduction.

· Along with reducing liver fibrosis and microinflamation HAART also appears to have an excellent hepatic safety profile.

· However, in presence of well preserved CD4 counts it would be advisable to treat the HCV infection first.

Abstract ID: 62443

Category: JO3: HCV: Diagnosis and Natural History

Elevated serum alphafetoprotein predicts long term prognosis in chronic hepatitis C: independent of deaths related to hepatocellular carcinoma.

S. Barclay, Station 9, Ayr, United Kingdom (Great Britain), R. Fox, Gartnavel General Hospital, Glasgow, United Kingdom (Great Britain), E. Spence, Gartnavel General Hospital, Glasgow, United Kingdom (Great Britain), E. McCruden, West of Scotland Regional Virus Laboratory, Galsgow, United Kingdom (Great Britain), J. McAllister, Gartnavel General Hospital, Glasgow, United Kingdom (Great Britain), S. Campbell, Hairmyres Hospital, East Kilbbride, United Kingdom (Great Britain), P. Mills, Gartnavel General Hospital, Glasgow, United Kingdom (Great Britain)

Introduction

Elevated serum alphafetoprotein (AFP) is associated with hepatocellular carcinoma (HCC) and as such should be associated with poorer overall survival. However, and elevated AFP is often seen in subjects with chronic hepatitis C (CHC) who do not ultimately develop HCC. It has been suggested that these subjects may have more severe liver disease and raised AFP might therefore be a useful non-invasive marker for non- HCC related mortality.

Aims

1) To determine if there is an independent association between elevated AFP and survival, when considering all causes of death, and also when HCC-related deaths are excluded. 2) To determine any association between elevated AFP and histological severity of CHC.

Method

A prospective follow-up study of 680 CHC subjects was undertaken at a single centre over the 10 year period 1993-2002. All patients had baseline AFP measured and were followed up for at least 18 months after the end of the study. Cause of death was determined in all cases. Survival estimates were determined by Kaplan Meier method and Cox proportional hazards. Other associations were determined using logistic regression.

Results

Patients were followed for a median of 227 weeks (range 100-584), with a time interval from AFP measurement to death of 192 weeks (100-440). Seven subjects developed HCC, all had a raised AFP and all died. The estimated survival at 500 weeks (death from all causes) in the normal AFP group was 93.2% and raised AFP group 56.8%. When HCC deaths were excluded, survival in normal AFP group was 93.2% and raised AFP group 76.1%. Raised AFP was significantly associated with non-HCC mortality after correction for age, genotype, IV drug use, and alcohol excess: hazard ratio 4.31 (95% CI 1.41-13.1), p=0.01. There was a positive association between raised AFP and patient age: odds ratio 1.06, 95% CI 1.01-1.12, p=0.04 and stage of liver fibrosis: odds ratio 22.4, 95% CI 5.2-96.5, p<0.001.

Conclusion

Elevated AFP can predict both HCC and non-HCC related survival over a 500 week period in CHC subjects. This may relate to the association between elevated AFP and advanced liver fibrosis.

Abstract ID: 62614

Category: JO3: HCV: Diagnosis and Natural History

HCV RNA QUANTIFICATION OF HCV GENOTYPES 1 TO 5: COMPARISON OF 4 DIFFERENT ASSAYS

C. Sarrazin, Saarland University Hospital, Homburg / Saar, Germany, B. Gaertner, Saarland University Hospital, Homburg / Saar, Germany, D. Sizmann, Roche Diagnostics, Penzberg, Germany, R. Babiel, Roche Diagnostics, Penzberg, Germany, U. Mihm, Saarland University Hospital, Homburg, Germany, P. Hofmann, Saarland University Hospital, Homburg, Germany, M. von Wagner, Saarland University Hospital, Homburg, Germany, S. Zeuzem, Saarland University Hospital, Homburg, Germany

Introduction

Diagnosis of hepatitis C virus (HCV) infection and management of therapy is based on qualitative and quantitative measurement of HCV-RNA. Standardization of HCV RNA assays to IU mainly are based on HCV genotype 1 panels. Little is known about the variability of commercially available HCV RNA assays for quantification of different HCV genotypes.

Methods

In the present study, 4 different HCV RNA assays (two real-time RT-PCR assays, High Pure System (HPS)- and Cobas Ampliprep (CAP)- Cobas TaqMan (CTM), Roche Diagnostics; one standard RT-PCR assay, Cobas Amplicor Monitor 2.0 (CAM); and one signal amplification assay, Versant Quantitative 3.0, Bayer, (bDNA)) were compared for quantification of HCV genotypes 1 to 5. Furthermore, reliability of the different assays for treatment decision at week 12 for patients infected with genotype 1 was investigated.

Results

For CAM assay as a reference assay in genotype 1 infected patients (n=40) the mean inter-assay differences compared with CAP/CTM, HPS/CTM and bDNA were 0.16, - 0.13 and -0.48 log IU/mL HCV RNA, respectively. For the remaining genotypes the results were as follows: genotype 2a/c (n=14), 0.24 (CAP/CTM), -0.78 (HPS/CTM), - 0.49 (bDNA); genotype 2b (n=11), -0.21 (CAP/CTM), -0.18 (HPS/CTM), -0.64 (bDNA); genotype 3a (n=24), 0.13 (CAP/CTM), -1.04 (HPS/CTM), -0.55 (bDNA); genotype 4 (n=9), -0.52 (CAP/CTM), -1.51 (HPS/CTM), -0.05 (bDNA); genotype 5 (n=10), -0.28 (CAP/CTM), -1.00 (HPS/CTM), -0.24 (bDNA) log IU/mL HCV RNA. For decision of treatment discontinuation on the basis of the 2 log decline rule at week 12 in genotype 1 infected patients, in all subsequent sustained responders or relapsers a correct decision was possible only when the same assay was used at baseline and week 12.

Conclusions

Comparison of CAM with the CAP/CTM assay in the present study showed an equal quantification of genotype 1, 2, 3, and 5, while genotype 4 samples were slightly underestimated. For HPS/CTM assay a significant underestimation of HCV RNA concentrations of genotypes 2a/c, 3, 4, and 5 was observed (0.78 to 1.51 log). For bDNA assay a constant lower quantification (approx. 0.5 log) of genotypes 1 to 3 was present.

Abstract ID: 64311

Category: JO3: HCV: Diagnosis and Natural History

Evidence of Central Nervous System (CNS) Involvement in Patients with Chronic Hepatitis C (HCV) Infection and Minimal Liver Disease.

S. Montagnese, Centre for Hepatology, Royal Free and UCL Medical School, London, United Kingdom (Great Britain), C. Jackson, Department of Neurophysiology, Royal Free Hampstead NHS Trust, London, United Kingdom (Great Britain), J. C. Ennen, Department of Neurology, Medizinische Hochschule Hannover, Hannover, Germany, J. Krause, Department of Neurology, Medizinische Hochschule Hannover, Hannover, Germany, H. L. Tillmann, Department of Gastroenterology, Hepatology and Endocrinology, Hannover, Germany, M. Y. Morgan, Centre for Hepatology, Royal Free and UCL Medical School, London, United Kingdom (Great Britain), K. Weissenborn, Department of Neurology, Medizinische Hochschule Hannover, Hannover, Germany

Introduction:

Individuals with chronic HCV infection are frequently fatigued, even in the absence of significant liver injury, and manifest a range of investigational abnormalities suggestive of CNS involvement in the disease process; the findings are, however, variously contended (Hilsabeck et al, 2002; Cordoba et al, 2003).

Aim:

To seek objective evidence of CNS involvement and 2) to characterize the EEG in HCV infected patients with minimal liver injury. The study population comprised 65 HCV-infected patients (24 men, 41 women, mean [range] age, 49 [30-75] yr); none had clinical, laboratory, radiological or, in 23 (35%), histological evidence of severe fibrosis/cirrhosis; none misused alcohol nor was on treatment with interferon or psychoactive medication.

Method:

All patients underwent: 1) neurological examination; 2) formal assessment of fatigue, depression and anxiety; 3) psychometric assessment using the PSE-syndrome test, the cancelling d test, and the TAP battery, and 4) an EEG. The presence of fatigue, depression and anxiety was determined by use of validated threshold scores. The results of the psychometric tests were age-and education-normalized. The EEGs were analysed visually and spectrally; reference data were obtained from 153 gender/age-matched healthy controls. The EEG was defined as 'slow' if the relative theta power was >35% and 'fast' if the relative beta power was >35% on the derivation P3-P4. Statistical analysis was performed using the Fisher exact test and ANOVA/ANCOVA. A very high prevalence of fatigue (81%), depression (68.5%) and anxiety (75%) was observed; depression and anxiety were more common in the fatigued patients (p<0.001 and p=0.059, respectively).

Results:

Alterations were observed in a number of psychometric tests reflecting changes in vigilance, working memory and higher executive function; these were generally more common in the fatigued and/or depressed patients. The EEG was slow in two (3%) patients and fast in 20 (34%). The incidence of fast EEG activity was significantly greater in the patients than in the controls (34% vs. 11%; p<0.05); its occurrence was independent of the presence of fatigue, depression or anxiety. The presence of fast EEG activity and impairment in vigilance and higher executive function were significantly correlated. Evidence of cognitive impairment was observed in patients with HCV infection and minimal liver disease, together with an excess of fast EEG activity, which is a novel finding not previously described. The pathophysiology of these abnormalities remains unclear but similar psychometric and EEG features have been described in patients infected with HIV (Sinha et al, 2003).

Conclusion

· Patients with HCV infection display abnormalities of both psychometric function and the EEG in the absence of significant liver disease

· The findings provide further evidence of significant evidence for cerebral involvement in HCV infection.Abstract ID: 64609

Abstract ID: 64609

Category: JO3: HCV: Diagnosis and Natural History

Impact of Hepatitis C Virus Infection and Other Comorbidities on Survival in Patients on Dialysis.

A. A. Butt, University of Pittsburgh, Pittsburgh, PA, M. Skanderson, VA Pittsburgh Healthcare System, Pittsburgh, PA, k. McGinnis, Universityof Pittsburgh, Pittsburgh, PA, T. Ahuja, University of Texas Medical Branch at Galveston, Galveston, TX, C. Bryce, University of Pittsburgh, Pittsburgh, PA, A. Barnato, University of Pittsburgh, Pittsburgh, PA, C. H. Chang, University of Pittsburgh, Pittsburgh, PA

Background:

Persons with renal disease face an increased risk of acquiring HCV

infection. We estimated the impact of HCV and other comorbid conditions upon survival in persons on dialysis.

Methods:

In secondary data analysis, we identified newly diagnosed HCV infected persons and uninfected controls in the United States Renal Data System (USRDS) using claims data in 1997-1998. Subjects with a renal transplant at any time were excluded. Because some variables violated the assumptions of proportional hazards, we used the Gray’s time-varying coefficients model to examine factors associated with survival among HCV infected and uninfected subjects.

Results:

A total of 5,737 HCV-infected subjects and 11,228 HCV-uninfected controls were identified. HCV infected subjects were younger (mean age 57.8 years vs. 65.3 years) and more likely to be male (57.6% vs. 49.6%) and black (54.0% vs. 36.4%). They were more likely to have a diagnosis of drug (16.5% vs. 4.6%) and alcohol use (14.0% vs. 3.1%), and were more likely to be HIV co-infected (7.4% vs.1.8%). (All comparisons, p<0.0005) HCV was associated with an increased risk of mortality (p<0.0005). The hazards were highest at time of HCV diagnosis, decreased and then stabilized two years after diagnosis. Other factors were also associated with increased risk of mortality (p<0.0005 unless stated): HIV coinfection; diagnosis of drug use (p=0.001); coronary artery disease (p=0.006); stroke; diabetes as the primary cause for renal failure; peripheral vascular disease; depression and presence of anemia. Hazards were constant over time for the following variables: Black race (HR 0.77; p<0.0001); female gender (HR 0.95, p=0.02); increasing age (HR 1.13 for each 5 years); time on dialysis prior to HCV (HR 1.01 per year, p=0.0004); and a diagnosis of cirrhosis (HR 1.51, p<0.0001).

Conclusions:

HCV is associated with higher risk of death in patients on dialysis, even after adjusting for concurrent comorbidities. The risk is highest at time of HCV diagnosis and stabilizes over time. HCV testing should be considered for all patients on dialysis. Earlier diagnosis and treatment strategies are needed to decrease HCV associated mortality in this population.

Abstract ID: 66298

Category: JO3: HCV: Diagnosis and Natural History

ROLE OF MANNOSE-BINDING LECTIN (MBL2) POLYMORPHISMS IN THE DEVELOPMENT AND PROGRESSION OF CHRONIC LIVER DISEASE.

J. Halangk, Charité, Berlin, Germany, H. Witt, Charité, Berlin, Germany, A. Hachfeld, Charité, Berlin, Germany, G. Puhl, Charité, Berlin, Germany, T. Mueller, Charité, Berlin, Germany, R. Nickel, Charité, Berlin, Germany, V. Weich, Charité, Berlin, Germany, A. Bergk, Charité, Berlin, Germany, B. Wiedenmann, Charité, Berlin, Germany, P. Neuhaus, Charité, Berlin, Germany, W. Luck, Charité, Berlin, Germany, T. Berg, Charité, Berlin, Germany

Introduction:

Mannose-binding lectin is a plasma protein belonging to the collectin subgroup of the C type lectin superfamily. Collectins bind specifically to microbial and viral pathogens initiating agglutination, opsonization and complement activation. It has been shown that MBL2 genotypes causing low MBL plasma concentrations are associated with an increased risk of different types of infections. Three missense mutations in the exon 1 of MBL2 (G54D, G57E and R52C) are associated with markedly diminished MBL plasma concentrations. Furthermore, single nucleotide polymorphisms in the promoter (H/L and X/Y) and 5’untranslated region (P/Q) influence the MBL plasma level. MBL2 polymorphisms have been associated with a variety of infectious and autoimmune diseases. Several studies in patients with viral and autoimmune liver disease have brought conflicting results.

Patients and methods:

We included 225 patients with alcoholic liver disease, 40 patients with acute liver failure, 629 patients with HCV, 203 patients with HBV, 98 patients with autoimmune hepatitis, 209 patients with PBC or PSC, 65 patients with cryptogenic cirrhosis and 48 patients with NAFLD. A birth cohort of healthy newborns served as a control population. Genotyping was performed by PCR amplification and melting curve analysis with FRET probes.

Results:

MBL2 genotype frequencies did not differ between the patient and control population. In chronically infected hepatitis C patients, MBL2 polymorphisms were not associated with disease severity as determined by fibrosis stage, development of cirrhosis and application of the non-invasive APRI-score. Patients with Exon 1 wildtype alleles were more likely to achieve a sustained virological response (SVR) after antiviral therapy with IFN/Peg-IFN and ribavirin (OR 1,72; p=0,031, 95% CI: 1,05-2,82). SVR rates according to MBL genotype (i.e..exon 1 wildtype homozygotes vs. heterozygotes, homozygotes and compound heterozygotes) in genotype 1 infected patients were 43,0% vs. 23,9%. Multivariate analysis including HCV genotype, level of viremia, and GGT confirmed MBL2 exon 1 as an independent predictor for SVR.

Conclusion:

In general, MBL2 polymorphisms are no major genetic risk factor for the development and progression of chronic liver disease. However, our study provides evidence that mutant MBL2 exon 1 alleles are associated with an unfavourable treatment outcome in HCV infection. In contrast, disease severity as determined by fibrosis stage and inflammation grade was not influenced by the MBL2 polymorphisms.

This work was in part supported by the German network of excellence (Hepnet) and the European Virgil network.

Abstract ID: 66587

Category: JO3: HCV: Diagnosis and Natural History

HCV RNA in peripheral blood mononuclear cells (PBMNCs) of Sustained Virological Responders to Interferon Based Therapies: Is it a risk for relapse?

M. S. El-Raziky, Tropical Medicine Department, Cairo University, Cairo, Egypt, W. A. El-Akel, Cairo University, Cairo, Egypt, M. A. Soliman, Tropical Medicine Department, Cairo University, Cairo, Egypt, S. El-Kafrawy , National Hepatology and Tropical Medicine Research Institute, Cairo, Egypt, M. Abdel-Hamid, National Hepatology and Tropical Medicine Research Institute, Cairo, Egypt, G. Esmat, Tropical Medicine Department, Cairo University, Cairo

Introduction:

Assessment of chronic hepatitis C outcome in sustained responders to interferon requires prolonged observation and close monitoring. Data on long term follow up of patients treated with genotype 4, are limited.

Aim:

We aimed to assess the occurrence of relapse among Sustained Virological Responders (SVR) to Pegylated Interferon or Interferon therapy over two years follow up and to study the characteristics of relapsers also to assess the possibility of persistence of HCV RNA in peripheral blood mononuclear cells (PBMNCs) of SVR as a risk for relapse.

Methods:

Two hundreds patients with chronic HCV (90% genotype IV) were included in a randomized controlled clinical trial for treatment of chronic HCV with either Pegylated Interferon or Interferon- alpha both with ribavirin for 48 weeks. Eighty-three (41.5%) subjects are SVR. We followed the responders for 48 weeks by clinical and ALT levels evaluation as well as HCV RNA testing in serum and PBMCs at 24 weeks interval.

Results:

We followed the responders for a mean follow up period of 30 months (min 22-max 41) after end of therapy. Most of the patients (84.3%) reported the disappearance of side effects developed while on treatment with significant increase in their Body Mass Index, which correlated with elevated ALT(P<0.05). Blood picture showed neutropenia or thrombocytopenia in 1.2% cases. Elevated ALT was found in 6% (max 1.85 folds) and 9% (max 1.51 folds) on 24 months and 30 months follow up respectively. Viraemia relapsed in only one patient (1.2%). Testing for HCV RNA in PBMCs has been done; 1.2% tested positive in absence of viraemia.

Conclusion

HCV relapse is uncommon in genotype 4 patients (as well as other genotypes) with SVR to interferon based combined therapies which proved to be safe on the long term. Persistence of HCV RNA in PBMNCs of SVR occurs in few subjects.

Abstract ID: 63830

Category: JO3: HCV: Diagnosis and Natural History

Proteomic analysis of serum proteins in patients with hepatocellular carcinoma associated with hepatitis C virus infection.

S. KANMURA, University of Miyazaki, Miyazaki, Japan, H. Uto, Miyazaki-university, Miyazaki, Japan, K. Kusumoto, University of Miyazaki, Miyazaki, Japan, Y. Takahama, Miyazaki Prefectual Industrial Support Foundation, Miyazaki, Japan, S. Hasuike, University of Miyazaki, Miyazaki, Japan, K. Nagata, University of Miyazaki, Miyazaki, Japan, K. Hayashi, University of Miyazaki, Miyazaki, Japan, A. Akio, Translational Research Center, Kyoto University Hospital, Kyoto, Japan, H. Tsubouchi, Kagoshima University Graduate School of Medical and Dental Sciences, KAGOSHIMA, Japan

Background/aim:

Hepatocellular carcinoma (HCC), associated with hepatitis C virus (HCV) infection, has been one of the major causes of cancer death in the world. Although there are many therapy modalities for early stage HCC, no method to diagnose early HCC has been established. To improve patient survival, further research on early diagnosis is needed. Proteomics has the potential to identify novel markers for HCC diagnosis. Surface-enhanced laser desorption ionization time-of-flight mass spectrometry(SELDI-TOF/MS) ProteinChip technology can provide a potentially powerful tool for discovery of new biomarker(s). The aim of this study is to identify novel HCC diagnostic marker(s) and tumor-specific protein(s) in order to both advance early HCC detection and discover potential therapeutic targets in the treatment of HCC.

Methods:

Serums were obtained from 57 patients with HCC and 51 patients without HCC, who all had chronic liver disease due to HCV infection. Serums of six healthy volunteers were also used. To identify proteins associated with HCC, all serum samples were applied to CM10 ProteinChip Arrays, and proteome alterations were analyzed using SELDI-TOF/MS, Ciphergen ProteinChip Software 3.0.2, Biomarker Patterns Software. To construct the decision tree, serum of 35 patients with HCC and 44 patients without HCC were used as the training set. This decision tree was then assessed with a first test set of serums from 22 patients with HCC, 7 patients without HCC and 6 healthy volunteers. A second test set composed of serums from 7 patients who initially were not diagnosed with HCC but who were found to have HCC one year later was also analyzed. The decision tree classification model was used to evaluate the sensitivity and the specificity of this proteomic analysis.

Results:

Using six protein peaks to construct the decision tree, 97% of the HCC positive samples in the training set were correctly identified as having HCC. In addition, the sensitivity and specificity of this decision tree using the first test set were 73% and 86%, respectively. Surprisingly, 6 of 7 (86 %) patients in the second test set were predicted to have HCC before HCC was apparent by ultrasonography.

Conclusions:

In patients with HCV infection, serum profiling by SELDI ProteinChip system is useful not only for separating HCC from chronic liver disease without HCC, but also for early detection of HCC. Furthermore, as we were able to identify proteome alterations in serum samples from HCC patients in this study, SELDI Protein Chip system may be a useful tool for screening the proteins associated with HCC to discover potential new therapeutic targets.

Abstract ID: 66676

Category: JO3: HCV: Diagnosis and Natural History

THE PERFORMANCE OF APRI FOR THE DIAGNOSIS OF SIGNIFICANT HEPATIC FIBROSIS IS IMPROVED BY A SIMPLE MODIFICATION.

O. K. Fix, Boston Medical Center, Boston, MA, C. R. Horsburgh, Boston University School of Public Health, Boston, MA, T. C. Heeren, Boston University School of Public Health, Boston, MA, J. F. Reinus, Montefiore Medical Center, Bronx, NY, E. Garcia, Montefiore Medical Center, Bronx, NY, D. S. Mishkin, Boston Medical Center, Boston, MA, C. S. Graham, Beth Israel Deaconess Medical Center, Boston, MA, M. J. Koziel, Beth Israel Deaconess Medical Center, Boston, MA, D. P. Nunes, Boston Medical Center, Boston, MA

BACKGROUND:

The AST-platelet ratio index (APRI) is a useful noninvasive marker of hepatic fibrosis in patients with chronic hepatitis C (CHC). The aim of this study was to develop a predictive model of clinically significant fibrosis using factors routinely measured in patients with CHC.

METHODS:

This was a retrospective analysis of consecutive liver biopsies for evaluation of CHC at Boston Medical Center (test cohort) and Montefiore Medical Center (validation cohort). Biopsies were excluded for HIV infection and biopsy length <1 cm. Biochemical and clinical factors were analyzed for their ability to predict significant fibrosis (Ishak fibrosis stage ≥ 3). A model was constructed in the test cohort combining the independent predictors from a multivariate analysis. Optimal cut-off values were identified and the diagnostic ability of the model was described in the validation cohort and compared with the APRI.

RESULTS:

There were 216 liver biopsies in the test cohort and 125 in the validation cohort. The prevalence of significant fibrosis was 30% in both cohorts. After multivariate analysis, age, albumin, AST, and platelets were identified as independent predictors of significant fibrosis. The proportional relationships of these variables to fibrosis were used to construct a model, (age*AST)/(albumin*platelets). The area under the ROC curve (AUC) was 0.835 (95% CI: 0.772-0.892) in the test cohort and 0.861 (0.791-0.930) in the validation cohort. Using a cut-off of 1.7, clinically significant fibrosis was excluded with a high degree of accuracy (negative predictive value of 96%) in the test cohort. A cut-off of 6.7 had a positive predictive value of 77% for the diagnosis of significant fibrosis. Applying the model to the validation cohort would have avoided biopsies in 51% of the cohort with a similar degree of certainty (see table). The model correctly identified 44% of the subjects in both cohorts combined. The addition of age and albumin increased the diagnostic performance of APRI significantly in both cohorts combined, with an AUC of 0.841 (0.795-0.885) for the model and 0.809 (0.759-0.859) for APRI (p<0.0001).

CONCLUSIONS:

Modification of APRI with the inclusion of age and albumin significantly improved its diagnostic performance in both the test and validation cohorts. This study shows that these simple noninvasive markers have sufficient diagnostic value to be useful in guiding clinical decision-making with respect to liver biopsy and treatment.

Abstract ID: 67046

Category: JO3: HCV: Diagnosis and Natural History

INCREASED SURVIVAL OF PATIENTS WITH HCV-RELATED CIRRHOSIS WITH A LONG-TERM RESPONSE TO INTERFERON THERAPY.

M. Viganò, IRCCS Maggiore Hospital Milan, University of Milan Italy, Milan, Italy, A. Aghemo, IRCCS Maggiore Hospital, University of Milan, Milan Italy, Milan, Italy, M. A. Iavarone, IRCCS Maggiore Hospital Milan, University of Milan, Milan, Italy, P Lampertico, IRCCS Maggiore Hospital Milan, University of Milan, Milan, Italy, M. Rumi, IRCCS Maggiore Hospital Milan, University of Milan Italy, Milan, Italy, A. Sangiovanni, IRCCS Maggiore Hospital Milan, University of Milan Italy, Milan, Italy, E. Del Ninno, IRCCS Maggiore Hospital Milan, University of Milan Italy, Milan, Italy, M. Colombo, IRCCS Maggiore Hospital Milan, University of Milan Italy, Milan, Italy

Background and Aims.

Whether interferon (IFN) therapy prevents hepatocellular carcinoma (HCC) and liver decompensation in patients with cirrhosis caused by hepatitis C virus (HCV), is unclear. To date, randomized controlled studies are unfeasible for ethical reasons whereas prospective cohort studies of patients followed for a sufficient time period may provide meaningful information.

Patients and Methods:

In 1997 a cohort of 323 patients (174 men, median age 61 years) with Child-Pugh A, HCC-free compensated cirrhosis and no history of previous gastrointestinal bleeding, jaundice or ascites, was generated and subjected to surveillance with liver chemistry, serum a-fetoprotein and abdominal ultrasound (US) performed every six months. One hundred and twelve received IFN at doses of 6 MU three times a week for 6-12 months, 36 patients received combination therapy with ribavirin (Rbv) at doses of 1000-1200 mg daily for 6-12 months according to viral genotype. End-points of the study were liver decompensation, gastrointestinal bleeding, porto-systemic encephalopathy, HCC, progression of Child-Pugh stage and death.

Results:

During 85 months (range 11-106) of follow-up, 131 (40%) patients developed at least one complication. Overall, the most frequent complication was HCC (81 patients, 25%), followed by liver decompensation (44, 13%) and gastrointestinal bleeding (6, 2%). Child-Pugh score progressed in 75 (23%) patients. 84 patients (26%) died, 40 of liver failure, 23 of HCC, 21 of extrahepatic complications. Ten patients underwent liver transplantation (8 for HCC, 2 for liver failure). Of 148 patients treated with IFN + Rbv, 42 (28%) were sustained virological responders (SVR). Decompensation rates were significantly lower in SVR than in non responders (NR) or untreated (NT) patients (0% vs. 1.6% vs. 3.0%, p< 0.004). The same was true for HCC (0.7% vs. 3.7% vs. 4.3%, p<0.004) and for death (0% vs. 3.7% vs. 5%, p<0.0003). By multivariate analysis, patients with sustained virological response and those with lower levels of bilirubin and higher levels of albumin at baseline, had significantly higher chances of survival.

Conclusion:

Sustained virological response to IFN + Rbv improved the survival of patients with compensated cirrhosis casued by hepatitis C.

Abstract ID: 67236

Category: JO3: HCV: Diagnosis and Natural History

DO SERUM AUTOANTIBODIES AFFECT THE SEVERITY OF CHRONIC HEPATITIS C.

V. Ozenne, Department of gastroenterology, CRETEIL, France, P. Chretien, Laboratory of Immunology, CRETEIL, France, H. Hagege, Department of gastroenterology, CRETEIL, France, G. Pileire, Department of gastroenterology, CRETEIL, France, I. Rosa, Department of gastroenterology, CRETEIL, France, T. Lons, Department of gastroenterology, CRETEIL, France, B. Elharrar, Department of gastroenterology, CRETEIL, France, M. Chousterman, Department of gastroenterology, CRETEIL, France

Background/Aim:

Non-organ specific autoantibodies (NOSA) are frequently found in patients with HCV infection. However, their relationship with the progression of the liver damage in such patients remains controversial. The aim of this study was to assess whether NOSA are associated with more severe liver disease.

Patients/Methods:

116 adult untreated patients with chronic hepatitis C were evaluated consecutively. Serum antinuclear (ANA), anti-smooth muscle (SMA), and anti-liverkidney microsomal (LMK-1) antibodies were studied. Significant serum dilutions for ANA, SMA, and LKM-1 were >1:40. Liver biopsies were scored according to the Metavir score and the components of this score.

Results:

1) Autoantibodies were detected in 37 patients (32%) (group 1). ANA, SMA, LKM-1 occurred in 23 (20%), 15 (13%), and 1 (1%) patients respectively. The concomitant positivity for 2 autoantibodies was observed in 3 cases. 2) Group 1 patients were more frequently associated with a route of infection other than drug injection or transfusion than patients without autoantibodies (group 2) (73% vs 51 %, p=0.05) 3) On the other hand, no difference was detected in age, sex, drinking and smoking habits, BMI, duration of infection, biological characteristics or in HCV genotype or viremia. 3) There was no difference in the necroinflammatory score in the two groups as a whole; on the contrary, the 23 patients with ANA autoantibodies showed a higher necroinflammatory activity than group 2 patients (A2-A3 : 17/23 = 74% vs 41/93 = 44%, p=0,01). 4) Fibrosis was more pronounced in group 1 than in group 2 (F3-F4 : 13/35 = 35% vs 12/79 = 15%, p=0.01).The same was observed when ANA-positive patients were compared with the group 2 patients.

Conclusion:

1) Chronic hepatitis C was more severe when autoantibodies were positive, mainly in ANA-positive patients. 2) The correlations observed between NOSA positivity and histological characteritics suggest that autoantibody occurence may not represent a fortuitous event in the course of HCV infection

Abstract ID: 67703

Category: JO3: HCV: Diagnosis and Natural History

THE SOURCE OF HCV DIAGNOSIS AFFECTS THE LONG-TERM COURSE OF CHRONIC HEPATITIS C (CHC).

V. Di Martino, CHU Besançon, Besancon, E. Naudet-Collin, CHU Besançon, Besançon, France, V. Jooste, Registre des Hépatites, CHU Dijon, Dijon, France, P. Evrard, Registre des Hépatites, CHU Besançon, Besançon, France, A. Minello, CHU Dijon, Dijon, France, J. Miguet, CHU Besançon, Besançon, France, E. Monnet, CHU Besançon, Besançon, France, P. Hillon, CHU Dijon, Dijon, France

Introduction:

The natural history of CHC is not yet fully established, since controversial data about the long-term progression to cirrhosis(CC), hepatocellular carcinoma(HCC), or liver-related death(LRD) were reported, depending on the population studied. The modelling of liver fibrosis progression did not allow definitive conclusions because it was built on cohorts of selected patients, referring to hepatology units.

Aim:

The aim of this work was to provide a more accurate determination of the natural history of CHC using a registry with prospective observation of a large and unselected population, and to assess the determinants of major outcome events through multivariate analyses with specific caution to the circumstances of HCV diagnosis.

Patients of methods:

Between 1994 and 2001, an exhaustive record of 1830 new cases of CHC without HIV or HBV coinfection, diagnosed in two French administrative areas was performed. The record of risk factors of HCV infection, source of HCV diagnosis, alcohol consumption, available virologic and histologic features, anti-HCV therapies, and outcome events was performed at the time of HCV diagnosis and updated yearly. The median follow-up from HCV diagnosis was 49 months. The Cox model was used for multivariate analyses.

Results:

Patients characteristics were as follows: 59% males, age at HCV diagnosis: 46yrs, 56% genotype 1, 61% IVDUs, 31% alcohol consumers. The source of HCV diagnosis was systematic screening in 69% of cases and involved a general practitioner (GP) in 33% of cases. The major outcome events were: 139 CC (8%), 51 HCC (3%), 291 death (15%), 57 LRD (3%). Through multivariate analyses, age at HCV diagnosis was an independent predictor of LRD (p<0.0001), HCC (p=0.0005), and CC (p=0.015), male gender was an independent predictor of death (RR=1.8, p=0.005), LRD (RR=2.83, p=0.051), CHC (RR=3.58, p=0.049), and CC (RR=2.02, p=0.013), alcohol consumption was an independent predictor of death (RR=2.71, p<0.0001), LRD (RR=5.87, p<0.0001), HCC (RR=4.62, p=0.0006) and CC (RR=2.50, p=0.0002), anti-HCV therapy was associated with lower mortality (RR=0.26, p<0.0001) and LRD (RR=0.26, p=0.036), systematic screening was associated with lower mortality (RR=0.53, p=0.0004), LRD (RR=0.08, p<0.0001), HCC (RR=0.40, p=0.036), and CC (RR=0.57, p=0.016), and HCV diagnosis from GP was associated with lower mortality (RR=0.49, p=0.002), but not LRD, HCC, and CC.

Conclusion:

In unselected population of CHC, the source of HCV diagnosis is associated with different outcome profiles of CHC. Better survival is observed in patients with HCV diagnosis given by systematic screening, independently of other prognosis factors.

Abstract ID: 67944

Category: JO3: HCV: Diagnosis and Natural History

Liver stiffness and Biomarkers: Correlation with cirrhosis,portal hypertension and hepatic synthetic function.

A. Khokhar, BIDMC, Boston, MA, R. Farnan, BIDMC, Boston, MA, C. MacFarlane, BIDMC, Boston, MA, B. Bacon, SLU, St. Louis, MO, J. McHutchison, Duke Univ Medical Center, Durham, NC, N. Afdhal, Beth Israel Deaconess Medical Center, Boston, MA

Intro:

Non-invasive biomarkers for chronic liver disease are a focus of research and development. Combinations of serum biomarkers and novel hepatic elastography (stiffness) may improve diagnosis of fibrosis. Currently the importance of staging fibrosis with a biopsy is to diagnose bridging fibrosis and particularly cirrhosis, since patients with cirrhosis require screening for portal hypertension and cancer. We hypothesize that combination studies will correctly stage liver disease and that in F3/F4 patients, stiffness and biomarkers may also correlate to important clinical variables such as portal hypertension or liver synthetic function. METHODS: We have studied 106 consecutive patients with Metavir F3 (20) and F4 (86) and 69 patients with F0-F2. The etiology of liver disease included HCV, alcohol, HBV and NASH. Patients had FibroScan, laboratory, clinical, radiological, endoscopic evaluations and ELISA for hyaluronic acid (HA; CORGENIX, Denver, CO) and YKL-40 (MetraBiosystems, CA).

RESULTS:

Median liver stiffness was 24.5kPa for F3/4 and 6.0kPa for F0-2. Receiver operator characteristics for FibroScan, HA, and YKL-40 to differentiate F3, F4 from F0-2 yielded an area under the curve (AUC) of 0.924, 0.823 and 0.766 respectively. Using a cutoff of 14kPa, sensitivity was 77% and specificity 90% for F3/4. We then substituted the FibroScan score for the Metavir stage of disease in F3/F4 patients and looked at correlations with predictors of cirrhosis. A significant correlation of stiffness in kPa was seen with platelets (p<0.02); APRI (p < 0.01), hyaluronic acid (p < 0.009) and albumin (p<0.0001). 46 cirrhotic patients had endoscopy within 6 months of biopsy and FibroScan. Median stiffness was 15.5 kPa in cirrhosis and no varices and 32.6 kPa in patients with varices. Correlation of stiffness to variceal size was highly significant (p <0.0002). AUC for the ROC for varices versus no varices was 0.767 and a cutoff of 20kPa gave a sensitivity of 68% and specificity of 84% for varices. Neither HA nor YKL-40 performed as well as stiffness for predicting varices (AUC 0.634 and AUC 0.621 respectively) or correlating to variceal size (HA p =ns; YKL-40 p< 0.01).

CONCLUSION:

Liver stiffness is not only able to diagnose F3/4 with great accuracy but also the degree of stiffness (in kPa) correlates with clinical parameters of progressive liver fibrosis such as portal hypertension and synthetic dysfunction.

Abstract ID: 61101

Category: JO3: HCV: Diagnosis and Natural History

PERSISTENCE OF HCV INFECTION IN APPARENTLY HEALTHY ANTI-HCV POSITIVE PATIENTS WITH CONSTANTLY SERUM HCV-RNA NEGATIVE AND NORMAL ALT LEVELS.

López-Alcorocho, Fundación para el Estudio de las Hepatitis Virales, Madrid, Spain, E. Rodríguez-Iñigo, Fundación para el Estudio de las Hepatitis Virales, Madrid, Spain, M. Pardo, Fundación para el Estudio de las Hepatitis Virales, Madrid, Spain, I. Castillo, Fundación para el Estudio de las Hepatitis Virales, Madrid, Spain, J. Quiroga, Fundación para el Estudio de las Hepatitis Virales, Madrid, Spain, V. Carreño, Fundación para el Estudio de las Hepatitis Virales, Madrid, Spain

Introduction

It is assumed that the presence of anti-HCV with undetectable serum HCV-RNA and persistently normal ALT levels reflects a resolved HCV infection. However, in some of these patients HCV-RNA is detected in their PBMC (1), although it is unknown if HCV is present in liver. So, we have studied the possible presence and replication of HCV in the liver of 12 anti-HCV positive patients who were persistently serum HCV-RNA negative and had normal ALT levels for 29.2 ± 19.8 months. These patients underwent abdominal surgical procedures for different reasons and they gave their written consent for the obtention of a liver sample during that procedure. PBMC were available in all cases. HCV-RNA-positive and -negative strands were detected by strand-specific realtime RT-PCR in liver and PBMC. The presence of both HCV-RNA strands in liver was confirmed by in situ hybridization. HCV-specific CD4+ T-cell responses were assessed by standard lymphoproliferative 3H-uptake and flow cytometric-based interferon-gamma secretion assays. HCV-RNA-positive strand was found in 10/12 (83%) liver samples (mean load: 2.6x10⁵ ± 1.9x10⁵ copies/mg total RNA). Positive cases had HCV-1b in liver. The HCV-RNA-negative strand was detected in 10/10 (100%) liver samples with HCV-RNA-positive strand (1.1x10 ± 1.0x10⁵ copies/mg total RNA). The existence of HCV-RNA of both polarities was confirmed in all cases by in situ hybridization. In 5 randomly selected patients HCV-RNA was also amplified with primers of the core region. Products were cloned and sequenced, demonstrating no cross-contamination between samples. Regarding PBMC, genomic HCV-RNA was found in 6/12 (50%) PBMC samples (2.2x10⁵ ± 1.9x10⁵ copies/mg total RNA) and 5/6 (83%) PBMC had also the HCV-RNA negative strand (1.5x10⁵ ± 8.5x10⁴ copies/mg total RNA). The two patients who did not have HCV-RNA in their liver were also negative for HCV in PBMC. 4/8 (50%) patients analyzed had a positive HCV-specific proliferative T-cell response. In addition, low frequencies of interferon-gamma secreting CD4+ T-cells were detected in response to HCV re-stimulation. Regarding liver histology, 8 patients (67%) had minimal changes, 1 had liver steatosis, 2 patients had non-alcoholic steatohepatitis, and 1 presented a chronic active hepatitis.

Conclusion:

In conclusion, HCV may persist and replicate in the liver and PBMC of anti-HCV positive patients who are serum HCV-RNA negative and have persistently normal ALT levels. The weak effector CD4+ T-cell responses may account for the virus persistence. In the light of these results, these patients must be followed-up in the long term as they present an ongoing viral infection. The epidemiological and clinical relevance of this finding should be studied in the future.

(1) J Infect Dis 2005;191:1730-3

Abstract ID: 62554

Category: JO3: HCV: Diagnosis and Natural History

Cryoglobulinemia is Associated with Steatosis and Fibrosis in Chronic Hepatitis C.

d. saadoun, Department of internal medicine, Pitie-salpetriere Hospital, Paris, Fr, paris, France, t. asselah, Department of hepatology, hôpital beaujon, Clichy, France, M. Resche-rigon, Department of Statistic, Hôpital Saint-Louis, paris, France, f. charlotte, Department of anatomo-pathologie, Hôpital Pitié-salpetriere, paris, p. bedossa, Department of anatomo-pathologie, Hôpital Beaujon, Clichy, d. valla, Department of Hepatology, Hôpital Beaujon, Clichy, j. Piette, Department of Internal medicine, Hôpital Pitié-Salpétrière, paris, p. marcellin, Department of hepatology, Hôpital Beaujon, Clichy, France, p. Cacoub, Department of Internal Medicine, Hôpital Pitié-Salpétrière, paris, France

Background and Aim :

Chronic hepatitis C virus (HCV) infection is associated with numerous extrahepatic manifestations, mostly mixed cryoglobulinemia (MC). The relationship between cryoglobulins and severity of liver lesions (necroinflammation and fibrosis) is debated. No study has focused on the relationship between presence of cryoglobulin and liver steatosis which is associated with progression of liver disease.

Aim:

The aim of this study was to determine the relationship of cryoglobulins with liver lesions (necroinflammation, fibrosis, and steatosis) in HCV infected patients.

Methods :

659 adults patients with untreated chronic hepatitis C admitted for liver biopsy were included in this study. Risk factors for fibrosis and steatosis were assessed.

Results :

Mean age was 49.0 ± 13.0 years and 49% were male. Cryoglobulin was present in 263 patients, 103 of whom had vasculitis. In multivariate analysis, cryoglobulin increased by nearly 3 folds the risk to have a high stage fibrosis (F3-F4) and a steatosis higher than 10%. There was no significant difference with regard to grade of fibrosis or steatosis between symptomatic and asymptomatic patients with cryoglobulins. Steatosis higher than 10% was associated with a higher body mass index (p<0.001), HCV genotype 3 (p=0.003), cryoglobulin (p=0.02), and a high grade of liver fibrosis (p=0.009). A high stage of fibrosis (F3-F4) was associated with cryoglobulin (p=0.008), high grade of necroinflammation (A2-A3) (p=0.009), and steatosis higher than 10% (p=0.01).

Conclusion :

Our study shows in a large population of patients with chronic hepatitis C, an independent association between cryoglobulin and steatosis as well as advanced fibrosis. Cryoglobulins with or without vasculitis may be a prognostic indicator associated with the severity of liver disease in patients with chronic hepatitis C.

Abstract ID: 63576

Category: JO3: HCV: Diagnosis and Natural History

Molecular Tracing of the Global Hepatitis C Virus Epidemic Predicts Regional Patterns of Hepatocellular Carcinoma Mortality.

Y. Tanaka, Nagoya City University Graduate School of Medical Sciences, Nagoya , Japan, F. KURBANOV, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan, V. VARGAS, Hospital General Universitari Vall d'Hebron, Barcelona, Spain, J. I. ESTEBAN, Hospital General Universitari Vall d'Hebron, Barcelona, Spain, M. YUEN, The University of Hong Kong, Hong Kong, Hong Kong, C. LAI, The University of Hong Kong, Hong Kong, Hong Kong, A. KRAMVIS, University of the Witwatersrand, Parktown , South Africa, M. C. KEW, University of the Witwatersrand, Parktown , South Africa, H. J. ALTER, Department of Transfusion Medicine, Bethesda, MD, M. MIZOKAMI, Nagoya City University Graduate School of Medical Sciences, Nagoya, Japan

Introduction:

Molecular Tracing of the Global Hepatitis C Virus Epidemic Predicts Regional Patterns of Hepatocellular Carcinoma Mortality

Background:

The molecular clock theory has successfully estimated the spread time of hepatitis C virus (HCV) in the 1930’s in Japan 30 years earlier than that in the US in the 1960’s, predicting that an increased hepatocellular carcinoma (HCC) prevalence will occur in the US over the next 2-3 decades (PNAS, 2002). The molecular evolutionary analysis based on coalescent theory can provide important insights into epidemiological processes worldwide. Aim: The aim is to estimate HCV spread time worldwide. This approach was combined with analyses of the HCV epidemiological-historical background and HCV-related HCC in different countries worldwide. Method: The HCV gene sequences of 131 genotype 1b (HCV-1b) strains from Japan, 38 HCV-1a from the USA, 33 HCV-1b from Spain, 27 HCV-3a from the former Soviet Union (FSU), 47 HCV-4a from Egypt, 25 HCV-5a from South Africa (SA), and 24 HCV-6a from Hong Kong (HKG) isolated in this and previous studies were analyzed. Sequences in the NS5B regions were determined directly and the molecular evolutionary analysis based on the coalescent theory was applied to HCV isolates of each genotype.

Results:

The coalescent analysis indicated that a transition from constant size to rapid exponential growth (spread time) occurred in Japan in the 1920s (HCV-1b), but not until the 1940s for the same genotype in Spain and other European countries. The spread time of HCV-1a in the USA was estimated to be in the 1960s, HCV-3a in the FSU, HCV-5a in SA, and HCV-6a in HKG in the 1960s, mid 1950s, and late 1970s, respectively. Three different linear progression curves were determined by analysis of the relationship between HCV seroprevalence and HCC mortality in different geographic regions; a steep ascent indicated the greatest progression to HCC in Japan, a near horizontal line indicated the least progression in the USA and FSU, and an intermediate slope was observed in Europe (figure). Conclusions: These findings strongly suggest that the initial spread time of HCV is associated with the progression dynamics of HCC in each area, irrespective of genotype.

Abstract ID: 63934

Category: JO3: HCV: Diagnosis and Natural History

CASP7 GENE POLIMORPHISMS AND FIBROSIS PROGRESSION IN CHRONIC HEPATITIS C.

A. Alvarez-Marquez, Hospitales Universitarios Virgen del Rocio, Seville, Spain, J. Aguilar-Reina, Hospitales Universitarios Virgen del Rocio, Seville, Spain, L. Gomez- Izquierdo, Hospitales Universitarios Virgen del Rocio, Seville, Spain, A. Nuñez-Roldan, Hospitales Universitarios Virgen del Rocio, Seville, Spain, M. F. Gonzalez-Escribano, Hospitales Universitarios Virgen del Rocio, Seville, Spain

Background and aims:

Hepatitis C virus (HCV) infection causes liver disease that is characterized by inflammation, cell damage, and progressive liver fibrosis leading to cirrhosis. Apoptosis and caspase activation are known factors that accelerated disease progression in HCV infection. In addition, host genetic factors could play key roles in the modulation of hepatic fibrosis and contribute to outcome of the disease. Caspase 7 (CASP7), an apoptosis-related cysteine protease, is involved in the activation cascade of caspases responsible for apoptosis execution. Two polymorphisms that causes amino acid change in the CASP7 gene have been identificated: 877C>G (S244T) polymorphism and 892A>G (K249R in the beta variant of the protein). The aim of this study was to evaluate the involvement of these allelic variants in the severity of liver damage and progression of fibrosis in chronic hepatitis C.

Patients and Methods:

A total of 225 Caucasian Spanish patients with biopsy proven chronic hepatitis C infection were genotyped for both polymorphisms using a PCR-based method. Liver biopsies obtained before treatment from all the patients were evaluated for fibrosis grades (according to the system of Scheuer), hepatocyte steatosis and iron staining were also investigated. Other variables recorded were: body mass index, HCV genotypes, viral load, presence of serum autoantibodies, thyroid alterations, biochemical parameters and disease duration.

Results:

Genotypes 892 AA and 877 GG were associated with higher degree of fibrosis (p=0.009) and with iron staining in liver biopsies (p= 0.003). Regarding the grade of necroinflammatory activity, 892 AA and 877 GG genotypes were related to higher levels of necroinflammatory and portal-periportal activity, although this comparison did not reach statistical significance (p= 0.077 and 0.088 respectively). Regarding the other parameters studied the frequency of 892-A and 877-G alleles was higher in patients with serum autoantibodies (42,8% vs. 25,8% p= 0.01; OR 2.09, 95% CI: 1.13-3.86 and 42,8% vs. 26.7% p=0.013; OR 2.06, 95% CI: 1.11-3.82 respectively) and in patients with tyroid alterations (40.6% vs. 26.55% p= 0.049 OR 2.04, 95% CI: 0.93-5-4.43, and 43.7% vs. 27.55% p= 0.05; OR 2.05, 95% CI: 0.93-4.50). Duration of the disease was not different among patients 892AA and 877GG vs. the rest.

Conclusions:

Our findings support an association between polymorphisms located in the CASP7 gene and the development of liver fibrosis and other factors that influence the outcome of chronic hepatitis C.

Abstract ID: 65020

Category: JO3: HCV: Diagnosis and Natural History

Monitoring liver stiffness: a new tool to measure liver fibrosis during therapy.

B. Coco, U.O.Gastroenterologia ed Epatologia - AOU Pisana, Pisa, Italy, F. Oliveri, U.O. Gastroenterologia ed Epatologia - AOU pisana, Pisa, Italy, P. Colombatto, U.O. Gastroenterologia ed Epatologia - AOU pisana, Pisa, Italy, P. Ciccorossi, U.O. Gastroenterologia ed Epatologia - AOU pisana, Pisa, Italy, R. Sacco, U.O. Gastroenterologia ed Epatologia - AOU pisana, Pisa, Italy, F. Bonino, Direzione Scientifica - Ospedale Maggiore IRCCS, Milano, Italy, M. R. Brunetto, U.O. Gastroenterologia ed Epatologia - AOU pisana, Pisa, Italy

Introduction:

Hepatic fibrosis is the major indicator of progressive liver disease. However histology, the gold standard tool to detect fibrosis, requires an invasive procedure, that is unsuitable for monitoring disease progression and may be affected by sampling errors.

Aim:

Our aims were: 1. to evaluate the diagnostic accuracy (DA) of liver stiffness by FIBROSCAN (Echosens, Paris, France) for detection of liver fibrosis and cirrhosis in patients (pts) with viral (HBV and HCV) chronic hepatitis; 2. to compare the DA of Fibroscan (FS) to other scores and surrogate markers of fibrosis: APRI and Forns scores, Fibrotest (FT) and Hyaluronic acid (HA). We evaluated 241 consecutive pts with a liver biopsy performed within 12 months from FS (159 pts) or with ultrasound signs of cirrhosis (Child A, 69 pts). Histological staging were scored according to METAVIR. 228 pts were suitable for the analysis: 7 pts were excluded for liver biopsy not adequate to stage fibrosis and 6 for technical limitations to FS. Stiffness mean values significantly correlated with fibrosis score: 6.3; 9.1; 20.8; 25.6 KPa for fibrosis F0-F1; F2-F3; F4 and clinical cirrhosis, respectively (p <0.001).By a ROC analysis we found that, in overall pts, FS values >/= 8.3 KPa identified fibrosis >/= F2 with 88.5% sensitivity and 74% specificity (DA 83%), whereas values >/=14 KPa detected cirrhosis with 78.3% sensitivity and 98.2% specificity (DA 88.2%). The DA of FS for detection of cirrhosis was higher (91.2%) considering only the 159 pts with liver biopsy. We observed as among cirrhotic pts, those with spontaneous or IFN/antiviral induced prolonged biochemical remission had FS values significantly lower than those with disease activity (mean 14.9 vs 25.6 KPa, p <0.001) and frequently lower than 14 KPa (52% vs 13.3% pts). FS had a better AUROCs than APRI and Forns scores (0,948 vs 0,813 and 0,901 respectively). FS showed an higher DA than FT and HA to identify both fibrosis >/=F2 (83.8% vs 73.9% and 77.5% respectively) and cirrhosis (86% vs 73.2% and 75.6% respectively). Combination with FT and HA did not improve the performance of FS, except in pts with disease remission (45 % sensitivity for FS vs 80% when using the 3 tests, with at least 1 positive).

Conclusions:

Liver stiffness measured by FS significantly correlates with fibrosis at histology. Lower stiffness values in pts with biochemical remission prompt its clinical usefulness to monitor treatment efficacy. In our cohort FS showed an higher DA compared to other scores and surrogate markers of liver fibrosis.

Abstract ID: 67099

Category: JO3: HCV: Diagnosis and Natural History

TIME IMPACT OF SERUM STORAGE TEMPERATURE ON STABILITY OF LIVER ENZYMES AND PROTEINS AND 0N FIBROTEST-ACTITEST RESULTS.

D. Messous, Groupe Hospitalier Pitié Salpêtrière, PARIS, France, M. Munteanu, Hôpital Pitié-Salpêtrière, PARIS, France, R. Morra, Hopital PitiéSalpetrière, paris, France, A. Piton, HOPITAL PITIE SALPETRIERE, PARIS, France, T. Poynard, Hop Pitié Salpetrière, Paris, B. Hainque, Hopital Pitié Salpêtrère, PARIS, France, F. Imbert- Bismut, Hop Pitié-Salpêtrière, paris, France

Background:

Combinations of alpha2macroglobulin (A2M), haptoglobin (HAPTO), apolipoprotein A1 (APOA1), total bilirubin (TB), gammaglutamyltransferase (GGT) [Fibrotest] and alanine aminotransferase (ALT) added [Actitest] are used as alternatives to liver biopsy in chronic hepatitis C. Little is known about ideal conditions for sera storage, but serum banks are widely used for retrospective analyses.

Aim:

To determine the time-dependent impact of different serum storage temperatures on the stability of Fibrotest-Actitest components and results.

Methods:

Three pools of sera were obtained from blood sampling hepatology department patients. Parameter stability was studied with serum sample storage at -20°C, -35°C and -80°C. Assays were performed on thawed samples every 15 days in the first four months and once a month thereafter, according to standardised international methods.

Results:

Except GGT, the parameters’ stability varied more than 5% with 8 months’ serum storage at -20°C (maximum variations: ALT -60%, A2M +24%, TB –17%, APOA1 +12% and HAPTO +6%). The results varied with a similar trend after 8 months’ -35°C serum storage. The important decrease in enzymatic activity of ALT was already noted after 15 days at -20°C or - 35°C. Parameter results did not vary more than 6% after 8 months’ storage at –80°C, except for APOA1 (+11%). The parameters’ variation had little influence (-5%) on FibroTest results after 8 months’storage. The impact of -20°C and -35°C storage temperature on ActiTest results is important (-50% variation) mostly because of very diminished ALT activity.

Conclusions:

Sera storage at –80°C is recommended for retrospective studies.

Abstract ID: 67581

Category: JO3: HCV: Diagnosis and Natural History

INDEPENDENT VALIDATION AND COMPARISON WITH FIBROSCAN,FIBROTEST AND LIVER BIOPSY OF CLINICAL GLYCOMICS FOR THE NON INVASIVE ASSESSMENT OF LIVER FIBROSIS IN CHRONIC HEPATITIS C.

L. Castera, Department of Hepatology, Hopital Haut Leveque, CHU Bordeaux, France, Pessac, A. Van Hecke, Department of Molecular Biomedical Research, Ghent University and VIB, Zwijnaarde, Belgium, P. Trimoulet, Department of Virology, Hopital Pellegrin, CHU Bordeaux, Bordeaux, France, N. Callewaert, Department of Molecular Biomedical Research, Ghent University and VIB,, Zwijnaarde, Belgium, P. Couzigou, Department of Hepatology, Hopital Haut Leveque, CHU Bordeaux, Pessac, France, V. de Ledinghen, Department of Hepatology, Hopital Haut Leveque, CHU Bordeaux, Pessac, France, R. Contreras, Department of Molecular Biomedical Research, Ghent University and VIB,, Zwijnaarde, Belgium, W. Laroy, Department of Molecular Biomedical Research, Ghent University and VIB,, Zwijnaarde , Belgium

Background & Aims:

Recently, novel and promising non invasive methods have ben proposed for the asesment of liver fibrosis in chronic hepatitis C (CHC): “clinical glycomics”, based on DNA sequencer/fragments analysers to profile serum protein N-glycans; liver stifnes measurement, using FibroScan (FS,

Conclusion:

Diagnostic performance of clinical glycomics appeared better for cirrhosis than for significant fibrosis (F≥2), being of the same order as that of FS and FT. The combination of FS with FT and FS with glycomics allowed to yield the best performances. The use of the combination of FS with FT or with glycomics could be useful in clinical practice for non invasive assessment of liver fibrosis in patients with CHC.

Abstract ID: 67806

Category: JO3: HCV: Diagnosis and Natural History

LIVER HISTOLOGY IN HEPATITIS C PATIENTS WITH NORMAL ALT LEVELS.

A. A. Mihas, Virginia Commonwealth University and McGuire DVA Medical Center, Richmond, VA, N. Chand, Virginia Commonwealth University, Richmond, VA, M. L. Shiffman, Virginia Commonwealth University, Richmond, VA, H. Lippman, Richmond DVA Medical Center, Richmond, VA, S. G. Abouassi, Virginia Commonwealth University, Richmond, VA, A. Habib, Virginia Commonwealth University, Richmond, VA, D. M. Heuman, Virginia Commonwealth University, Richmond,

BACKGROUND :

Elevated serum ALT (eALT) has traditionally been a criterion in selecting patients for liver biopsy and treatment in chronic hepatitis C. However, in recent years, it has become apparent that a small number of hepatitis C patients may be cirrhotic despite normal ALT. ALT levels commonly fluctuate. Normal ALT at a point in time may not accurately reflect the course of the hepatitis. We hypothesize that longitudinal evaluation of ALT over a period of many years, to distinguish patients with persistently normal ALT (pnALT) from those with intermittently normal ALT (inALT) may permit more accurate assessment of disease severity

AIMS:

The aim of this study was twofold : a) to compare the risk of cirrhosis between individuals with nALT and those with eALT at time of initial biopsy; and b) in patients with nALT, to compare the risk of cirrhosis between pnALT and inALT.

METHODS:

598 consecutive patients undergoing liver biopsy for evaluation of their chronic hepatitis C were included in this study. We utilized the VA electronic records to review ALT going back as much as 23 years, extracting the highest ALT value and the current ALT at time of liver biopsy. All patients had imaging of the liver (US or CT) as well as routine hematologic, biochemical and serologic tests by standard laboratory techniques. Hepatic fibrosis was graded on a 0-6 scale according to the Ishak scoring system by two blinded reviewers. Statistical analysis was carried out with ANOVA.

RESULTS :

The overwhelming majority of the 598 patients included in this study were male (M=95% ; F=15%) reflecting the US veterans population. More than two-thirds of the patients were African Americans (68%) and about one-third were white (35%). The mean patient age was 51, ranging from 41 to 71 years. Overall 44% of patients showed advanced fibrosis (ISHAK 3-6) in their liver biopsies and 15% were cirrhotic (ISHAK 5-6). Advanced fibrosis and cirrhosis were significantly more common in Caucasian patients than African Americans.

Only 11% of patients with pnALT had advanced fibrosis scores in contrast, advanced fibrosis was present in 51% of phALT and 62% in hALT group.

The prevalence of cirrhosis among the 3 groups was 1%, 17% and 23% respectively for pALT, phALT and hALT.

Among patients with pnALT, the prevalence of advanced fibrosis and cirrhosis were equally low in African Americans (10.7% and 1.7% respectively) as in Caucasians (10.6% and 0%, respectively).

CONCLUSIONS :

A substantial proportion of hepatitis C patients with normal ALT are found to have had previous ALT elevation on longitudinal review of liver tests. Compared to patients with persistently normal ALT, patients with intermittently normal ALT have a greater likelihood of cirrhosis. In absence of longitudinal ALT data, a normal ALT does not eliminate need for liver biopsy in hep C.

Abstract ID: 61116

Category: JO3: HCV: Diagnosis and Natural History

OCCULT HCV INFECTION IN HEMODIALYSIS PATIENTS.

I. Castillo, Fundación para el Estudio de las Hepatitis Virales, Madrid, Spain, E. Rodríguez-Iñigo, Fundación para el Estudio de las Hepatitis Virales, Madrid, Spain, G. Barril, Dept. Nephrology, Hospital U. La Princesa, Madrid, Spain, D. Arenas, Hemodialysis Unit, Hospital Perpetuo Socorro, Alicante, Spain, M. Espinosa, Dept. Nephrology, Hospital Reina Sofía, Cordoba, Spain, J. García-Valdecasas, Dept. Nephrology, Hospital Clínico San Cecilio, Granada, Spain, E. González-Parra, Dept. Nephrology, Hospital Central de la Defensa, Madrid, Spain, C. Sánchez, Dept. Nephrology, Hospital La Paz, Madrid, Spain, R. Selgas, Dept. Neprhrology, Hospital U. La Princesa, Madrid, Spain, V. Carreño, Fundación para el Estudio de las Hepatitis Virales, Madrid, Spain

Introduction:

Occult HCV infection is characterized by the presence of HCV-RNA in liver in the absence of anti-HCV and serum HCV-RNA (1). In addition, in a high percentage of these patients (70%), viral RNA is also detected in PBMC. Furthermore, it has been demonstrated that HCV replicates in the PBMC of patients with occult HCV infection (2). So, we have studied the existence of occult HCV infection in hemodialysis patients by detecting HCV-RNA in their PBMC.

Methods:

The inclusion criteria were: high ALT levels (normal values for hemodialysis patients: < 28 IU/l) and/or gamma-GT levels, persistently negative serological HCV markers (anti-HCV and serum HCV-RNA negative) and HBV markers (HBsAg,) and exclusion of other causes of liver damage. Six Spanish hemodialysis units participated in the study and 42 patients were enrolled (26 males) fulfilling the inclusion criteria. PBMC from 10 healthy donors who were repeatedly HCV-RNA negative were used as negative controls. Genomic HCV-RNA was detected by strand-specific RT-PCR and confirmed by fluorescent in situ hybridization in the PBMC of 26/42 (62%) hemodialysis patients.

Results:

Patients with occult HCV infection were younger (57.9 ± 13.4 years; p=0.01) than those without occult HCV (67.9 ± 13.2 years). No differences were found regarding gender, etiology of kidney disease or time on hemodialysis. Mean gamma-GT levels did not differ between both groups. ALT values were significantly higher (p=0.02) in patients with occult HCV infection than in negative ones (35.8 ± 19.6 vs 26.2 ± 24.3 IU/l, respectively). The antigenomic HCV-RNA strand was detected by strand-specific RT-PCR and by in situ fluorescent hybridization in 15/26 (58%) of the hemodialysis patients with occult HCV infection in PBMC. In 6 randomly selected patients with occult HCV infection in PBMC, HCV-RNA was also amplified with primers of the core region. Products were cloned and sequenced, demonstrating no cross-contamination between samples. A liver biopsy was performed in a hemodialysis patient with HCV infection in PBMC and the existence of an occult HCV infection was confirmed in this patient as the genomic and antigenomic HCV-RNA strands were detected in the liver cells by RT-PCR, real-time RT-PCR and by in situ hybridization.

Conclusion:

In conclusion, up to 62% of hemodialysis patients with persistently high values of ALT and/or gamma-GT of unknown etiology, present an occult HCV infection in their PBMC. Although these patients are serum HCV-RNA negative, they could be potentially infectious because HCV is replicating in their PBMC and so, preventive measures to avoid HCV spread must be considered.

(1) J Infect Dis 2004;189:7-14

(2) Gut 2005; 54:682-685

Abstract ID: 63730

Category: JO3: HCV: Diagnosis and Natural History

SYSTEMATIC REVIEW OF LENGTH AND NUMBER OF PORTAL TRACTS OBTAINED WITH PERCUTANEOUS LIVER BIOPSY.

E. Cholongitas, Royal Free Hospital, Liver Transplantation and Hepatobiliary Medicine, London, United Kingdom (Great Britain), A. Quaglia , Royal Free Hospital, Histopathology Department, London , United Kingdom (Great Britain), . Senzolo, Royal Free Hospital, Liver Transplantation and Hepatobiliary Medicine, London , United Kingdom (Great Britain), C. Triantos, Royal Free Hospital, Liver Transplantation and Hepatobiliary Medicine, London , United Kingdom (Great Britain), D. Patch , Royal Free Hospital, Liver Transplantation and Hepatobiliary Medicine, London , United Kingdom (Great Britain), A. Dhillon, Royal Free Hospital, Histopathology Department, London , United Kingdom (Great Britain), A. K. Burroughs, Royal Free Hospital, Liver Transplantation and Hepatobiliary Medicine, London , United Kingdom (Great Britain)

Background:

Percutaneous liver biopsy (PLB) is considered the ‘gold standard’ for evaluation of grading and staging in chronic hepatitis. However, recent data have shown that a liver biopsy should be 20-25 mm and/or contain >11 portal tracts (POT) (Colloredo et al, J Hepatol 2003; 39:239). This implies that more than one pass is required for PLB, which is known to increase complications. Aim: To evaluate the length and number of portal tracts in published series of PLB.

Patients/methods:

A Medline search (English/non-English) of articles using keys words: ‘percutaneous liver biopsy’, ‘needle’, ‘Menghini’, ‘Tru-cut’ ‘sample size’ and ‘length’. Published abstracts from European/American gastroenterology/hepatology conferences during the previous 10 years were also reviewed.

Results:

A total of 162 studies were retreived. Only 32 (27 full-paper and 5 abstracts) detailed length; 12 also provided information for POT with none reporting POT alone. 8430 patients underwent 9219 PLB [Menghini-PLB (MB), Tru-cut PLB (TrB), Usguided PLB (USB) and blind PLB (BB)]. Mean length was 17.5±5.8 mm and mean number of POT was 7.5±3.4. PLB were longer in studies with ≥ 100 PLB, compared to <100 PLB (20.4 mm vs 16 mm, p=0.026). MB were significantly longer, compared to TrB (19.5 mm vs 13.8 mm, p=0.01). USB were also longer, compared to BB (20.5 mm vs 14.4 mm, p=0.021). There was no significant difference in mean POT between MB and TrB (7.6 vs 7, p=0.6) and between USB and BB (8.3 vs 5.3, p=0.13). MB under Usguidance were 26.9 mm, significantly longer than TrB-US guided 13.3 mm (p<0.001), and blind MB 15.3 mm (p=0.004). There was no difference in PLB length (range: 16.3- 20.7 mm) according to needle size.

Conclusions:

This systematic review demonstrates that in well documented series of PLB, the average biopsy length and number of portal tracts is well below the currently set gold standard in over half the patients. This confirms the suspicion that many patients would require more than one pass and thus risk more complications from a PLB if an optimal biopsy is needed.

Abstract ID: 64869

Category: JO3: HCV: Diagnosis and Natural History

Combination of non invasive markers to identify liver fibrosis in HCV patients with persistently normal ALT (PNALT).

g. sebastiani, department of clinical and experimental medicine, padova, Italy, a. vario, department of clinical and experimental medicine, padova, Italy, a. ferrari, department of clinical and experimental medicine, padova, Italy, r. pistis, department of clinical and experimental medicine, padova, Italy, f. noventa, department of clinical and experimental medicine, padova, Italy, a. alberti, department of clinical and experimental medicine, padova, Italy

Background:

About 30-40% of HCV carriers present with persistently normal transaminases. Around 15-20% of these cases show significant fibrosis (F>2 according to METAVIR) in their liver and have therefore clear indication to antiviral therapy. Liver biopsy is currently used to identify these patients. Recently non-invasive methods have been proposed in chronic hepatitis C as surrogate markers of liver fibrosis but their diagnostic accuracy in HCV patients with PNALT has not been assessed and might be less adequate compared to patients with elevated ALT. Aim of this study was to develop a diagnostic algorithm for liver fibrosis in patients with PNALT using a combination of non invasive markers and minimising the need of invasive liver biopsy.

Methods:

We have investigated a consecutive, prospective series of 65 HCV positive patients with PNALT (28 males and 37 females, mean age: 44.9 + 14.3) who underwent a liver biopsy for diagnostic purposes. Liver fibrosis stages according to METAVIR were distributed as follows: F0-F1=52.3%, F2=35.4%, F3=7.7%, F4=4.6%. Fibrotest, AST to Platelets Ratio Index (APRI) and Forns’ score were measured in all patients at the same time of liver biopsy, taken as the gold standard. On the basis of the statistical performance of each of these markers, a diagnostic algorithm was developed by their sequential use. The end-points were: 1) to reduce the number of liver biopsies needed to correctly distinguish between absent-minimal fibrosis (F0-F1) and moderate-advance fibrosis (F>2) and 2) to avoid underestimation and to minimise overestimation of significant fibrosis for practical application in a pre-treatment scenario. The statistical performance was analysed as specificity, sensitivity, positive predictive value (PPV), negative predictive value (NPV) and accuracy.

Results:

Overall, Fibrotest had the best accuracy while APRI and Forns had better NPV. Forns also showed excellent PPV. Fibrotest, APRI and Forns classify 100%, 74% and 56% of patients, respectively. The best predictive model is shown in the figure. It led to save 55.4% liver biopsies with excellent statistical performance, no under-diagnosis and only 3% over-diagnosed cases.

Conclusions:

In the subgroup of HCV patients with PNALT, significant liver fibrosis, representing a definitive indication to antiviral therapy, can be diagnosed with high accuracy using an algorithm that combines non invasive fibrosis markers and reduces by more than 50% the need of taking a liver biopsy.

Abstract ID: 66368

Category: JO3: HCV: Diagnosis and Natural History

TRANSJUGULAR LIVER BIOPSY: HOW GOOD IS IT FOR ACCURATE HISTOLOGICAL INTERPRETATION?.

E. Cholongitas, Royal Free Hospital, Liver Transplantation and Hepatobiliary Unit, London, United Kingdom (Great Britain), A. Quaglia, Royal Free Hospital, Histopathology Department, London , United Kingdom (Great Britain), D. Samonakis, Royal Free Hospital, Liver Transplantation and Hepatobiliary Medicine, London, United Kingdom (Great Britain), M. Senzolo, Royal Free Hospital, Liver Transplantation and Hepatobiliary Medicine, London, United Kingdom (Great Britain), C. Triantos, Royal Free Hospital, Liver Transplantation and Hepatobiliary Medicine, London, United Kingdom (Great Britain), D. Patch, Royal Free Hospital, Liver Transplantation and Hepatobiliary Medicine, London, United Kingdom (Great Britain), G. Leandro, Royal Free Hospital, Liver Transplantation and Hepatobiliary Medicine, London, United Kingdom (Great Britain), A. Dhillon , Royal Free Hospital, Histopathology Department, London , United Kingdom (Great Britain), A. K. Burroughs, Royal Free Hospital, Liver Transplantation and Hepatobiliary Medicine, London, United Kingdom (Great Britain)

Background:

Currently an adequate percutaneous liver biopsy to evaluate chronic hepatitis should be ≥ 20-25 mm long and/or containing ≥ 11 complete portal tracts (CP) based on a selected series ≥ 30mm (Colloredo et al, J Hepatol 2003; 39:239). However, even with 17G needles only 42% of biopsies have ≥ 10 CP (Rocken et al, Liver 2001; 21:391). These data imply that more than one pass is necessary for adequate samples. In contrast, transjugular liver biopsy (TJLB), despite smaller diameter, allows multiple passes without increasing complications.

Aim:

To document length of TJLB and number of portal tracts/biopsy and to evaluate diagnostic efficacy.

Patients/methods:

326 consecutive TJLB in 274 patients, always using 3 passes (19G Trucut biopsy needle). For each TJLB, the number of fragments, length of each fragment and number of CP contained in each fragment were evaluated in a specific review.

Results:

There were no technical failures. 161 TJLB were performed post liver transplant (OLT). Histological diagnosis was possible in 322 (98.8%). The median number of fragments: 5 (1-13), with a median total length of 22 (3-46) mm with 65% being 20 mm or more. The total length was similar between pre-OLT and post-OLT TJLB (23 vs 22 mm, p=0.45) and between cirrhotics and non-cirrhotics (23 vs 22 mm, p=0.07). In 132 (40.5%) with severe parenchymal disturbance (cirrhosis and necrosis) the number of CP was not evaluated. In 194 (59.5%) the median number of CP was 8 (0- 26), with ≥ 6 in 146 (76%). Fifty (26%) had ≥ 11 CP; these biopsies were significantly longer than those <11 CP (28 mm vs 20 mm, p<0.0001). All TJLB with ≥ 11 CP were longer than 15 mm and 50% of those ≥ 25 mm length had ≥ 11 CP. There was only a moderate correlation between total length and number of CP (r=0.49, p<0.001). 60% of TJLB ≥ 28 mm long had ≥ 11 CP.

Conclusions:

TJLB with 3 passes yields adequate biopsy samples for diagnosis and 65% were 20mm or more in length. However only 26% had ≥ 11 CP. An increase to 4 passes should increase this. TJLB is an ideal technique to obtain sufficiently longer liver biopsies particularly when multiple biopsies/patient are being evaluated, but the minimum number of passes will need to be at least 4 to reliably evaluate chronic hepatitis.

Abstract ID: 67719

Category: JO3: HCV: Diagnosis and Natural History

ANALYSIS OF ACUTE HCV INFECTION IN HIV-POSITIVE WOMEN: EVIDENCE OF INFECTION IN PBMC AND GENITAL COMPARTMENTS.

T. Laskus, St. Joseph Hospital, Phoenix, AZ, M. Radkowski, Warsaw Medical University, Warsaw, M. Augenbraun , SUNY-Downstate Medical Center Brooklyn, Brooklyn, NY, J. Wilkinson, St. Joseph Hospital, Scottsdale , AZ, M. Nowicki, University of Southern California, Los Angeles, CA

Introduction:

Studies on HCV dynamics in the natural course of infection are rare due to difficulties in obtaining samples during the early phase of primary HCV infection. Although there is mounting evidence that the virus may be present in various compartments, no studies addressed this issue in primary infection. The aim of the study was to characterize HCV in the natural course of primary infection in plasma, PBMC and genital tract.

Patients:

Among 1517 patients in The Women’s Interagency HIV Study (WIHS) seroconversion to anti-HCV was found in 22 (1.5%). We studied 9 of these patients: the number of analyzed samples, which were collected every 6 months, ranged from 6 to 15.\ In each case samples were available both before and after seroconversion. 5’UTR sequences was analyzed in plasma, PBMC and cervical lavages (CVL). Single-strand conformational polymorphism (SSCP) was used to compare virus composition between the compartments and to analyze changes over time.

Results:

In 1 patient all analyzed samples were HCV RNA negative, in another patient a single plasma sample was positive and in a third patient viral sequences could be amplified only from plasma samples. The remaining 6 patients were analyzed in at least 2 different compartments. In patients 1-3 (Group I) HCV RNA sequences were amplified from all 3 compartments and SSCP analysis revealed that they were identical. In 1 case from this group viral sequences appeared in all compartments at the same time, in the second case plasma positivity preceded the detection of virus in other compartments, while in the third patient HCV RNA was detectable first in the CVL. In patients 4-6 (Group II) there were initial differences in viral sequences between PBMC and serum compartments but eventually plasma-derived virus was supplanted by the one present in PBMC. In 1 case virus appeared in serum and PBMC at the same time, while CVL was negative. In two other cases PBMC positivity preceded plasma positivity, while CVL compartment was identical to plasma compartment in 1 patient and negative in the other patient.

Conclusions:

In summary, in de novo infection virus was usually present in all 3 compartments. PBMC may be the initial site of amplification as its positivity preceded plasma positivity in 2 out of 6 cases. Furthermore, in 3 out of 6 cases where sequences in PBMC and serum were initially different, the PBMC variant eventually supplanted the plasma variant. Common presence of viral sequences in the genital tract and occasional detection of virus in CVL earlier than in other compartments suggests that it may be an initial site of HCV amplification, which would be compatible with sexual transmission.

Abstract ID: 66686

Category: JO3: HCV: Diagnosis and Natural History

Comparison of HCV Genotype and Subtype Determination Using Global Clinical Samples with the VERSANT HCV Genotyping Kit, TRUGENE HCV 5’NC Genotyping Kit and a New NS5b Sequencing Tool.

T. Guettouche, Bayer Institute for Clinical Investigation, Berkeley, CA, C. Elkin, Bayer Reference Testing Laboratory, Berkeley, CA, K. Leung, Bayer Reference Testing Laboratory, Berkeley, CA, H. Hnatyszyn, Bayer Institute for Clinical Investigation, Berkeley, CA Disclosures: H. James Hnatyszyn - Employee of: Bayer Heaqlthcare, LLC, Diagnostics, Berkeley, CA., Discussion will include off-label / investigative use of medicine(s), medical devices or procedure(s); Toumy Guettouche - Employee of: Bayer HealthCare, LLC; Claudia Elkin - Employee of: Bayer HealthCare, LLC; Kimmy Leung – Employee of: Bayer HealthCare, LLC

Introduction:

HCV genotype and subtype determination for clinical samples using the VERSANT® HCV Genotyping Kit (LiPA), the TRUGENE® HCV 5’NC Genotyping Kit and a new, laboratory-developed HCV NS5b Sequencing Tool using the OpenGene® DNA Sequencing System were compared. Methods:

Clinical samples (n=72) were collected globally to obtain a cross-representation of HCV genotypes. The HCV viral load was quantified for each of these samples using the VERSANT® HCV RNA 3.0 Assay (bDNA). HCV RNA extracted from the clinical samples and commercial HCV panels were analyzed with the VERSANT® HCV Genotyping Assay (LiPA) and the TRUGENE® HCV 5’NC Genotyping Kit using the manufacturer’s protocols as well as the prototype HCV NS5b Sequencing Tool using a laboratory developed protocol. Concordance for clade and subtype determination was compared between the commercial HCV genotyping assays and the HCV NS5b Sequencing Tool. The library of the prototype HCV NS5b Sequencing Tool contains HCV clade and subtype classifications from both the conventional nomenclature and the proposed six clade nomenclature. Representatives of all six HCV clades were identified across the clinical sample population. The TRUGENE® HCV 5’NC Genotyping Kit was observed to be the most sensitive of the three methods, generating genotypes for 97% of all samples (70/72) compared to the VERSANT® HCV Genotyping Assay (LiPA) (85%) or the NS5b Sequencing Tool (89%).

Results:

The majority of indeterminate samples not genotyped by the three assays were observed to be below the limit of detection (615 IU/mL) of the VERSANT® HCV RNA 3.0 Assay (bDNA). 100% concordance for genotype determination was observed for samples analyzed using the 5’NC and LiPA assays. However, concordance for genotype determination was reduced to 95% for both LiPA and 5’NC sequencing when compared to results obtained using the HCV NS5b Sequencing Tool. The discrepant results occurred for samples misclassified as Clade 1 by LiPA and 5’NC versus Clade 6 by the NS5b Tool. The HCV NS5b Tool provided subtype determination for greater than 98% of all clinical samples compared to 5’NC (94%) and LiPA (90%). Concordance for subtype determination for clinical samples was 83% when comparing results from the 5’NC and LiPA assays. Subtype concordance was 76% and 84% between the HCV NS5b Sequencing Tool and LiPA or 5’NC sequencing, respectively.

Results:

Both the VERSANT® HCV Genotyping Kit (LiPA) and the TRUGENE® HCV 5’NC Genotyping Kit is suitable for determining genotypes of viral species from HCV clades 1 to 5. The HCV NS5b Sequencing Tool can determine both genotype and subtype for HCV species from Clades 1 to 6 using clinical samples from individuals with chronic hepatitis C.

Abstract ID: 62850

Category: JO3: HCV: Diagnosis and Natural History

A prospective analysis of FibroTest-ActiTest-FibroSure prognostic value in patients with chronic hepatitis C.

Y. Ngo, GHPS, Paris, France, M. Munteanu, Biopredictive, Paris, France, D. Messous, GHPS, Paris, France, F. Charlotte, GHPS, Paris, France, F. Imbert-Bismut, GHPS, Paris, France, D. Thabut, GHPS, Paris, France, P. Lebray, GHPS, Paris, France, V. Thibault, GHPS, Paris, France, Y. Benhamou, GHPS, Paris, France, J. Moussalli, GHPS, Paris, France, V. Ratziu, GHPS, Paris, France, T. Poynard, GHPS, Paris, France

Background:

FibroTest-ActiTest-FibroSURE (FT-AT) is a non-invasive biochemical marker of liver fibrosis used as alternative to liver biopsy in chronic hepatitis C.

Aim:

The aims were to assess the 5 year prognostic value of Fibrotest, for cirrhosis complications and survival, in comparison with biopsy staging.

Methods:

Fibrosis stage and activity grade were assessed on the same day by a liver biopsy and by markers, in a prospective cohort of 537 patients, 157 classified at baseline as severe fibrosis (FT greater than 0.58), 137 as moderate fibrosis (0.32-0.58) and 243 as no or minimal fibrosis (FT lower than 0.32).

Results:

In patients with severe fibrosis, survival without HCV complications was 78.5% (95% Confidence Interval= 71.2-95.7%; 28 complications), and survival without HCV related death was 92.7% (88.0-97.3%; 9 HCV deaths). In patients with moderate fibrosis, these were 98.8% (CI=96.6-100%; 1 complication; P<0.001) and 100% (no HCV death; P<0.001) respectively, and in patients with minimal fibrosis, 100% (no complication; P<0.001) and 100% (no HCV death; P<0.001), respectively. FT was a better predictor than histological staging for HCV complications with area under the ROC curves (se) =0.96 (0.01) vs 0.91 (0.02); P=0.01, and HCV deaths: AUROC= 0.96 (0.02) vs 0.87 (0.03) P=0.046). The prognostic value of FibroTest was also significantly higher than fibrosis staging and other indexes (APRI, Forns, Pugh) in uni (P<0.001) and multivariate regression analyses (P<0.01) taking into account treatment response and known prognostic factors (HIV coinfection, alcohol consumption).

Conclusion:

FibroTest is a surrogate marker of Hepatitis C severity with a very significant 5 years prognostic value, better than histology, for severe complications and death related to HCV.

Abstract ID: 63957

Category: JO3: HCV: Diagnosis and Natural History

PROSPECTIVE ANALYSIS OF DISCORDANCE BETWEEN FIBROSCAN AND FIBROTEST WHEN USED IN COMBINATION AS FIRST-LINE ASSESSMENT OF LIVER FIBROSIS IN CHRONIC HEPATITIS C.

L. Castera, Department of Hepatology, Hopital Haut Leveque, CHU Bordeaux, France, Pessac, France, B. Le Bail, Department of Pathology, Hopital Pellegrin, CHU Bordeaux, Bordeaux, France, J. Foucher, Department of Hepatology, Hopital Haut Leveque, CHU Bordeaux, Pessac, France, J. Bertet, Department of Hepatology, Hopital Haut Leveque, CHU Bordeaux, Pessace, France, M. Darriet, Department of Biochemistry, Hopital Pellegrin, CHU Bordeaux, Bordeaux, France, P. Couzigou, Department of Hepatology, Hopital Haut Leveque, CHU Bordeaux, Pessac, France, V. de Ledinghen, Department of Hepatology, Hopital Haut Leveque, CHU Bordeaux, Pessac, France

Background:

An algorithm combining 2 non invasive methods, FibroScan (FS) and Fibrotest (FT), has been recently proposed as first-line assessment of liver fibrosis in patients with chronic hepatitis C (CHC). Based on this algorithm (agreement between FS and FT), liver biopsy (LB) could have been avoided in more than 75% of patients for the diagnosis of significant fibrosis. The aim of the present study was to prospectively determine the prevalence and causes of discordance between FS and FT (as compared with LB), using this algorithm for the assessment of fibrosis in CHC patients.

Methods:

The fibrosis stage was prospectively assessed the same day by LB, FS, and FT. Fibrosis was scored F0 to F4 according to METAVIR scoring system by a single pathologist blinded to FS and FT results. FT was performed in a single laboratory applying the preanalytical and analytical recommendations required. The definition of a significant discordance between FS and FT was a difference of 2 stages of fibrosis or more. Risk factors for FT failure were defined as : hemolysis, acute inflammation or Gilbert’s disease. Risk factors for FS failure were:validated measurements <10, success rate <60%. Factors of poor interpretability for liver biopsy were: small size and fragmentation.

Results:

219 consecutive CHC patients (127 males, mean age 51±12 yrs) were included. At histology, the distribution of METAVIR fibrosis score was as follows: F1 24% ; F2 34%; F3 21% ; F4 21%. LB size was >15 mm in 70% of cases and >25 mm in 25% of case. Overall, significant discordance between FS and FT was observed in 57 cases (26%), mostly for the diagnosis of F2 (47%) and F3 (28%) (F1 12%; F4 13%). LB was considered not reliable in 6 discordant cases. Discordance was attributable to FS failure

in 15 (6.8%) patients and to FT failure in 27 (12.4%) patients, not attributable in 15 patients (6.8%). The most frequent failure of FS consisted of overestimation of fibrosis (60%), whereas the most frequent failure for FT was underestimation (60%). Taking into account the identification of discordance in 42 cases, use of the algorithm could have avoided LB in 204 (93%) patients (162 with agreement plus 42).

Conclusion:

Significant discordance between FS and FT was observed in 26% of cases, and a reason could be identified in more than 70% of these cases. FT failure was twice more common than FS failure. Our results suggest that first-line assessment of liver fibrosis in CHC, using the combination of FS and FT is implementable in clinical practice, allowing to avoid liver biopsy in more than 90% of patients.

Abstract ID: 65444

Category: JO3: HCV: Diagnosis and Natural History

Greater Prevalence of Cirrhosis in Hispanics Than in Caucasians or African Americans.

A. Hoyumpa, University of Texas Health Science Center, San Antonio, TX, F. E. Sharkey, University of Texas Health Science Center, San Antonio, TX, P. S. Brock, University of Texas Health Science Center, San Antonio, TX, R. T. Page, University of Texas Health Science Center, San Antonio, TX, P. S. Brock, University of Texas Health Science Center, San Antonio, TX, M. R. Kashi, University of Texas Health Science Center, San Antonio, TX, A. M. Herrera, University Physicians Group, San Antonio, TX, W. Smith, University of Texas Health Science Center, San Antonio, TX, I. G. Poy, University of Texas Health Science Center, San Antonio, TX, A. L. Webb, University of Texas Health Science Center, San Antonio, TX, O. Ali, South Texas Veterans Health Care System, Audie Murphy Division, San Antonio, TX

Background:

There are racial differences in epidemiology, histology and treatment response of patients with hepatitis C. The present study focuses on the Hispanic patients.

AIM:

To determine the prevalence of cirrhosis in Hispanics, Caucasians and African Americans with hepatitis C.

METHOD:

Pretreatment liver biopsies of hepatitis C patients seen between 2000 and 2004 were reviewed (those with less than 5 portal tracts to reduce the chance of sampling error). Biopsies were assessed by a single pathologist for inflammation, fibrosis, fatty change, iron deposits and were correlated with clinical parameters, including age, sex, race, diabetes, BMI, genotype, viral load, lipids, alcohol, Fe levels and comorbidities. There were 222 patients with single biopsies. RESULTS:

Cirrhosis (grade 4 fibrosis) was found in 30.2% of 144 Hispanics, 12.1% of 62 Caucasians (p=0.007) and 7.1% of 16 African Americans (p=0.05). Grades of inflammation were: Hispanics: 1.28 ± 0.4, Caucasian: 1.15 ± 0.4 (p<0.05) and African Americans: 1.25 ± 0.04 (p>0.05). Stages of fibrosis were Hispanics: 2.4 ± 1.2, Caucasians: 1.7 ± 1.1 (p=0.01) and African Americans: 1.7 ± 1.1 (p=0.06). These differences were associated with an increased incidence of steatosis in Hispanics (70.8%) compared to 45.2% in Caucasians (p<.001) and 37.5% in African Americans (p=0.007). Diabetes mellitus was also more common in Hispanics (28.9%) than in Caucasians (10.2%, p=0.003), but different in African Americans (40.0%, p=0.38). Serum Fe level was higher in Hispanics, 118.2 ± 52 µ g/dl than in Caucasians, 102.8 ± 43 (p = 0.05), as was Fe saturation, 36.2 ± 18% vs 30.5 ± 14%, respectfully (p = 0.03). Data were insufficient for African Americans. Ferritin levels were similar in Hispanics and Caucasians. No statistical differences were noted in all three groups with respect to age, sex distribution, cholesterol, genotype I distribution, IV drug use and co-infection with HIV or HBV.

CONCLUSION:

1) Cirrhosis was 2.5 times more common in Hispanics than in Caucasians and 4 times more than in African Americans with hepatitis C.

2) The greater prevalence of cirrhosis in Hispanics was associated with higher incidence of steatosis and diabetes than in Caucasians and African-Americans, and with slightly or marginally higher serum Fe levels and Fe saturation than in Caucasians.

3) Steatosis and diabetes may have played predominate roles while iron abnormalities were contributing factors.

$) There is a need to study other factors that may be involved in racial differences in hepatitis C progression.

Abstract ID: 66760

Category: JO3: HCV: Diagnosis and Natural History

INTERNAL MEDICINE RESIDENTS KNOWLEDGE AND MANAGEMENT OF HEPATITIS C VIRUS (HCV) INFECTION IMPROVE WITH A TARGETED EDUCATIONAL INTERVENTION.

B. Pearlman, Center for Hepatitis C, Atlanta, GA, A. Ravi, Atlanta Medical center, Atlanta, GA, C. Ehleben, Atlanta Medical Center, Atlanta, GA

Background:

Despite the enormous current and projected burden of HCV-related illness, primary care physicians’ knowledge about testing and management of this clinical entity is inadequate. We examined the results of a targeted yet comprehensive intervention dministered to internal medicine residents aimed at improving their facility with the diagnosis and treatment of HCV.

Methods:

Internal medicine residents in three post-graduate years of training in a community-based teaching program participated in our study. The educational intervention consisted of several hours of HCV-related formal didactic lecture, handouts and self-evaluation questionnaires covering the natural history, epidemiology, screening, diagnosis, prevention and treatment of HCV over a ten-month period. Residents were also exposed to a primary care continuity clinic with many chronically infected HCV outpatients, both treatment-naïve and on interferon-based combination therapy. Ten pages of questionnaires evaluating knowledge in the aforementioned areas were administered pre- and post-intervention. Comparative data were analyzed using student’s t-test and ANOVA (analysis of variance).

Results:

Resident participants were equally distributed among three post-graduate years of training. A total of 29 residents participated in the intervention and pre-intervention testing, and 28 residents completed the post-intervention questionnaire. Results were partitioned into three groups including A: HCV natural history, epidemiology, screening, diagnosis, and prevention, B: HCV treatment and side-effect management and C: HCV comprehensive, which included both A and B. The mean percentage of correctly answered questions at baseline was 13.14%; however, post-intervention, the overall improvement in mean test scores was 66.1%. Residents knowledge significantly improved in all areas tested (A, p=0.00004; B, p<0.0000001; C, p<0.0000001). Knowledge differed significantly by post-graduate level of training before the intervention (A, p=0.005; B, p=0.003; C, p= 0.001), but not after the intervention (A, p=0.78; B, p=0.12; C, p=0.19), even when treatment experience was held constant. This disparity further supports the effectiveness of the teaching intervention.

Conclusion:

We confirmed that hepatitis C knowledge is inadequate among internal medicine physicians in training. A comprehensive educational intervention administered to internal medicine residents over a ten month period was successful in improving physicians’ knowledge in all HCV-related areas tested. Similar educational interventions should be studied in more diverse primary care training settings.

Abstract ID: 67124

Category: JO3: HCV: Diagnosis and Natural History

Hispanic Race/Ethnicity is Associated with an Aggressive Course of Chronic Hepatitis C Infection: Role of Patient Demographics, Hepatic Steatosis and Necroinflammation.

s. Verma, University of Southern California, Los Angeles, CA, M. Bonacini, California pacific Medical Center, San Francisco, CA, S. Govindarajan, Rancho Los Amigos Center, Downey, CA, G. Kanel, University of Southern California, Los Angeles, CA, K. Lindsay, University of Southern California, Los Angeles, CA, A. Redeker, Rancho Los Amigos MEdical Center, Downey, CA

Background and aims:

There is very little data available on patient demographics and liver histology in Hispanic patients with chronic hepatitis C (HCV) infection. The aim of this study was to analyze the severity of HCV related liver disease amongst the Hispanics compared to other race/ethnic groups. Patients and methods: patients were recruited fromthe Los Angeles County Hepatitis Clinic. Race/ethnicity was classified as: Non Hispanic Whites (NHW), Hispanics and Others (Blacks and Asians). Liver fibrosis and necroinflammatory score were assessed by the Ishak scoring system. Hepatic steatosis was graded as 0-4.

Results:

296 patients were found suitable for the study (NHW =63, Hispanic =169 and Others =64). Hispanics were older at exposure due to higher prevalence of blood transfusion as a risk factor (p<0.007 vs NHW). They also had significantly higher fibrosis stage (3.3 + 2 vs 2.3 + 6.9, p=0.001), necroinflammatory score (6.4 +1.8 vs 5.6 + 1.6, p=0.002) and faster fibrosis progression (0.14 + 0.09/yr vs 0.09 + 0.07/yr), (p=0.0002 vs NHW). Prevalence of obesity (52%) and hepatic steatosis (77%) was highest in Hispanics vs NHW (22% and 47%) and Others (31% and 47%), p<0.004. Mean steatosis grade (1.1 vs 0.6) was also higher in Hispanics vs NHW and Others, p<0.0003. Non-NHW ethnicity, OR 1.8 (95% CI 1.0-3.5), p=0.08, and necroinflammatory score, OR 1.6 (95% CI 1.3-1.8), p<0.0001 were independent predictors of fibrosis stage > 4. Hispanic ethnicity was an independent predictor of both NI score > 7 (OR 1.7 (95% CI 1.0-2.8), p=0.05) and hepatic steatosis (OR 3.0 (95% CI 1.6-5.6), p=0.0007. A significant correlation was observed between fibrosis stage and both NI score (r=0.45, p<0.001) and hepatic steatosis (r=0.21, p=0.001).

Conclusion:

Hispanics have an aggressive course of HCV infection compared to NHW. This may be related to older age at exposure, higher necroinflammatory scores and prevalence of hepatic steatosis. Race/ethnicity is an important variable that should be weighed in when deciding treatment options for patients with HCV infection.

Abstract ID: 67232

Category: JO3: HCV: Diagnosis and Natural History

Is liver biopsy justified in hepatitis C genotype 2 and 3?.

p. rizzi, liver unit, kings' college hospital, london, london, United Kingdom (Great Britain), p. harrison, Department of Liver Studies and Transplantation, Division of Gene and, London, United Kingdom (Great Britain)

Background.

It has been argued that liver biopsy (LB) may not be justified in patients with HCV genotypes 2 and 3, particularly if they have normal transaminase levels, because of their high rate of sustained response (SVR) to antiviral treatment. However, without a LB, cirrhosis may go unrecognised, leading to inappropriate discharge from follow-up and surveillance for hepatocellular carcinoma (HCC).

Aim.

Aim of our study was to determine the prevalence of cirrhosis in patients with HCV genotype 2 or 3, with normal or abnormal AST.

Methods.

We reviewed the reports of all histological examinations of liver carried out at our institution between 01-01-01 and 31-12-04. These included examination of LB and livers explanted at transplantation (OLT livers). We selected all reports of examinations performed in patients with HCV, and observed in these the rate of cirrhosis in relation to genotype and AST. Results.

We reviewed 9831 reports: HCV accounted for 658/9831 (6.7%) cases. Cirrhosis (Ishak fibrosis stage =6) was diagnosed in 198/658 (30%) cases and only 87/198 (44%) of these reports were from examination of OLT livers. Genotype was available in 440/658 of HCV patients; genotypes 2 or 3 accounted for 179/440 (40,6%) cases. Of these 179 patients, 60 (33.5%) had cirrhosis; this was diagnosed in 39/60 (65%) cases with LB and 21/60 (35%) from examination of OLT livers. HCC was diagnosed in 9/60 (15%) of patients with genotype 2 or 3 and cirrhosis, and 7 of these were transplanted. 25 of the total number of 198 cirrhotic patients (12.6%) patients had normal AST. Of the subgroup of 60 patients with cirrhosis and genotype 2 or 3, 8 had normal AST: of these, 2 had HCC and underwent OLT.

Conclusions.

In the cohort of 179 patients with HCV genotypes 2 or 3, 39 (22%) were cirrhotic on percutaneous LB, of these 8 (4.5%) had normal LFT. Although patients with HCV genotype 2 or 3 have a high response rate to antiviral treatment, our data indicate that a significant proportion of these patients have cirrhosis. A finding of infection with HCV genotype 2 or 3 does not diminish the need for LB.

Abstract ID: 67535

Category: JO3: HCV: Diagnosis and Natural History

IMPACT OF CAFFEINE CONSUMPTION ON SERUM ALANINE AMINOTRANSFERASE LEVEL AND HISTOLOGICAL ACTIVITY IN PATIENTS WITH CHRONIC ACTIVE HEPATITIS C.

C. Costentin, Department of Hepatology - Hopital Henri Mondor, Creteil, France, C. Hezode, Department of Hepatology - INSERM U635 - Hopital Henri Mondor, Creteil, France, F. Roudot-Thoraval, Department of Public Health - Hopital Henri Mondor, Creteil, France, F. Medkour, Department of Hepatology, Creteil, France, E. Zafrani, Department of Pathology -Hopital Henri Mondor, Creteil, France, J. Pawlotsky, INSERM U635 - Department of Virology - Hopital Henri Mondor, Creteil, France, D. Dhumeaux, Department of Hepatology - Hopital Henri Mondor, Creteil, France, A. Mallat, Department of Hepatology - INSERM U581- Hopital Henri Mondor, Creteil, France

Introduction:

It has recently been suggested that caffeine consumption is associated to a lower risk of elevated serum alanine aminotransferase (ALT) activity in patients at high risk for liver disease (Ruhl et al. Gastroenterology 2005). This effect might be attributed to antioxydative properties of caffeine. Thus, the aim of this study was to evaluate the impact of caffeine consumption on ALT level and histological activity in patients with chronic hepatitis C. 183 consecutive naïve patients with histologically proven chronic hepatitis C were included. Data collected included demographics, route of transmission, daily consumptions of alcohol, tobacco, caffeine during the 6 months preceding liver biopsy, body mass index, genotype, steatosis, activity and fibrosis (METAVIR) and ALT level at the time of liver biopsy. Daily caffeine consumption was estimated as the sum of mean intakes of coffee, tea and caffeine-containing sodas. Patients (119 men, 64 women, mean age: 44.6*10.9 years) were classified into 4 groups according to caffeine consumption. The relationship between caffeine consumption, ALT level and histological activity is shown below : Caffeine consumption (mg/day) Mean ALT level x normal range Elevated ALT level n (%) Activity A2-A3

n (%) < 260 n=45 2.1*1.8 29 (64.4) 31 (68.9) 261-400 n=39 1.9*1.6 29 (74.4) 26 (66.7) 401-675 n=51 2.0*1.3 39 (76.5) 26 (51.0) >675 n=48 1.9*1.5 35 (72.9) 26 (54.2) There was no relationship between daily caffeine consumption and ALT level or histological activity. Steatosis (OR=3.8 (1.6-9.2)) and elevated ALT (OR=3.3 (1.5-7.1)) level were the only two predictors of moderate-marked (A2-A3) activity by multivariate analysis. In summary, caffeine consumption does not appear to affect ALT level and histological activity in patients with chronic hepatitis C.

Abstract ID: 61536

Category: JO3: HCV: Diagnosis and Natural History

Hepatitis C Virus(HCV) Genotype and Subtype Determination Using a New NS5b Sequencing Tool.

M. Beld, Bayer Reference Testing Laboratory, Mijdrecht, Netherlands, R. Gouw, Bayer Reference Testing Laboratory, Mijdrecht, Netherlands, C. van der Meer, Bayer Reference Testing Laboratory, Mijdrecht, Netherlands, C. Elkin, Bayer Reference Testing Laboratory, Berkeley, CA, K. Leung, Bayer Reference Testing Laboratory, Berkeley, CA, T. Guettouche, Bayer Institute for Clinical Investigation, Berkeley, CA, H. Hnatyszyn, Bayer Institute for Clinical Investigation, Berkeley, CA

Introduction:

DNA sequencing of the NS5b region of the HCV genome has been proposed to determine both viral clade and subtype.

Methods:

The performance of a prototype HCV NS5b sequencing tool using the OpenGene® DNA Sequencing System is described. Commercial HCV plasma panels (n=18) and clinical samples with known (n=38) and unknown (n=48) genotypes were extracted using the QIAamp Viral RNA Kit. An additional set of clinical samples with known HCV genotypes (n=92) was extracted using the MagNA Pure System. Samples were analyzed with the VERSANT HCV Genotyping Assay (LiPA) and the TRUGENE® HCV 5’NC Genotyping Kit using the manufacturer’s protocols as well as the prototype HCV NS5b sequencing tool using a laboratory developed protocol. Concordance for clade and subtype determination was compared between the commercial HCV genotyping assays and the HCV NS5b sequencing tool. Dilutions of the HCV panel members with known concentrations of HCV RNA (IU/ml) were used to determine the sensitivity of the HCV NS5b sequencing tool. The library of the prototype HCV NS5b sequencing tool contains HCV clade and subtype classifications from both the conventional nomenclature and the proposed six clade nomenclature.

Results:

The HCV NS5b sequencing tool was observed to reliably resolve the HCV clade and subtype in samples with HCV RNA concentrations at 193 IU/ml for Clades 1 through 5 and their subtypes, and 962 IU/ml for Clade 6 and related subtypes. Specificity for this tool using HCV negative samples was 100%. Analysis of panel members representing Clades 1 through 6 with associated subtypes using the HCV NS5b sequencing tool resulted in 100% concordance for clade determination and 94.5% agreement with subtype (n=17/18). In comparison, concordance for clade determination by LiPA (78.5%) and 5’NC analysis (69%) was lower, most noticeably due to indeterminant results for panel members representing Clade 6. Regardless of extraction method used, the HCV NS5b sequencing tool generated both a clade and subtype determination for greater than 95% of all clinical samples (169/178 samples). There was 99.2% and 97.7% concordance for clade determination between the HCV NS5b sequencing tool, LiPA (126/127) and 5’NC sequencing (43/44). HCV subtype determinations were achieved in 99% of all samples where the HCV NS5b sequencing tool was used to identify the clade compared to LiPA (73.6%) and 5’NC sequencing (66%) analysis. Concordance for subtype determination for clinical samples was 84% and 70% between the HCV NS5b sequencing tool and LiPA or 5’NC sequencing, respectively.

Conclusion:

The HCV NS5b sequencing tool can determine both clade and subtype using clinical samples from individuals with chronic hepatitis C.

Abstract ID: 66240

Category: JO3: HCV: Diagnosis and Natural History

Serum proteome profile to predict virological response in patients with chronic hepatitis C treated by pegylated interferon plus ribavirin.

V. Paradis, Hôpital Beaujon, Clichy, France, T. Asselah, Hopital Beaujon, Clichy, France, D. Dargere, CNRS UMR 8149, Paris, France, M. Ripault, Hopital Beaujon, Clichy, France, M. Martinot, Hôpital Beaujon, Clichy, France, N. Boyer, Hôpital Beaujon, Clichy, France, D. Valla, Hôpital Beaujon, Clichy, France, P. Marcellin, Hôpital Beaujon, Clichy, France, P. Bedossa, Hôpital Beaujon, Clichy, France

Background and aim :

Surface Enhanced Laser Desorption Ionisation-Time-of-Flight (SELDI-TOF) mass spectrometry is a proteomic technique that enables the global profiling of proteins present in serum. We used this approach to monitor the kinetic of serum proteome in patients with HCV chronic infection receiving standard bitherapy regimen in order to predict treatment response.

Patients and methods :

96 patients with chronic hepatitis C were restrospectively selected. All patients received pegylated-interferon (PEG-IFN) alpha-2b in combination with ribavirin at a dose adjusted to bodyweight. Patients had serum sampling before starting treatment, at the end of treatment and at the end of the follow-up.

Results :

Comparison of protein profiles in pretreatment and after treatment serum allowed to characterize 50 protein peaks the level of which significantly varied. There was significantly less longitudinal variation of the serum proteome during treatment in patients with genotype 1 or in patients with clinically significant fibrosis at time of inclusion. In the group of patients that obtained sustained virological response to treatment, 37 peaks displayed significant variation during treatment whereas only one peak differed in non responders. In pretreatment sera, we identified 6 peaks whose level significantly differed between responders and non-responders patients. A logistic regression analysis allowed to define a predictive algorithm composed of a combination of 2 peaks, fibrosis stage and genotype allowing to correctly predict in pretreatment serum, response to treatment in 86% of all patients with an AUROC of 0.84 in an independent validating set of patients,

Conclusion :

This study suggests that the kinetics of proteome are significantly different in serum of patients according to treatment response. Serum protein profiling allows to predict response to antiviral treatment in a significant proportion of patients.

Abstract ID: 59540

Category: JO3: HCV: Diagnosis and Natural History

Cutaneous signs of liver disease: value for prognosis of severe fibrosis and cirrhosis.

C. Niederau, St. Josef Hospital Oberhausen, Germany, 46045 Oberhausen, Germany, S. Lange, Department of Medical Informatics, Biometry and Epidemiology, Ruhr-Uni, Bochum, Germany, M. Frühauf, Katholische Kliniken Oberhausen gGmbH, St. Josef Hospital Oberhausen, Oberhausen, Germany, A. Thiel, Department of Pathology, Evangelische Kliniken Duisburg Nord, Bethesda, Duisburg, Germany

Aims:

Although physicians have looked for cutaneous signs of liver disease for more than a century, their prognostic value has never systematically been evaluated. As yet, liver biopsy is the standard for evaluation of liver fibrosis. Methods:

Various cutaneous changes were prospectively recorded in all patients referred for liver biopsy from June 2000 to May 2004. Liver fibrosis was staged from F0 to F4 according to Desmet and Scheuer by the apthologist without knowing skin data.

Results:

The analysis included 744 patients (379 men, 365 women); 520 of the 744 patients (69,9%) had chronic hepatitis C while the remaining 30,1% had other liver diseases. By univariate analysis the frequency of several skin changes was significantly associated with the degree of fibrosis. In general at fibrosis stages F0-1 cutaneous changes such as spider naevi, palmar erythema, nail/hair changes and glossy tongue/lips were infrequent; these changes became slightly more frequent at F2 and were frequent at F3-4. To analyse the predictive value of skin changes, patients with F0-2 were compared to those with F3-4. Multivariate regression included only variables found significant in univariate analysis (P<0,01). Logistic regression was then performed by stepwise procedure. Final calculation included spider naevi, palmar erythema, teleangiectasia, bleeding signs, and dry skin as well as age and sex. When routine laboratory values were added by this procedure platelets, prothrombine time, gamma-glutamyl-transferase and albumin were included. ROC showed a good discrimination of F0-2 from F3-4 by the modelled score when combining skin changes and laboratory tests: at a score of 8 sensitvity and specificity were 90% and at a score of 7 only 2% of patients with F3-4 were misdiagnosed (sensitivity 98%) while specificity was still 75%. At the cost of 2% of non-diagnosed patients with F3-4 one might have saved 60% of biopsies. Exclusion of laboratory tests (only including skin changes) resulted in a slightly less valuable ROC which still had good discriminative power. ROC was less valuable discriminating fibrosis F0-1 from F2-4. The discriminative power of skin changes was better than the ASAT/platelet ratio (APRI). Conclusions:

The present results prove that it is useful to look for skin changes in patients with liver disease. Several skin changes are signs of severe liver fibrosis and cirrhosis. Their prognostic value is better than that of routine laboratory tests or the APRI ratio. In the absence of most of such skin changes, severe liver fibrosis is very unlikely; on the other hand, even patients with cirrhosis do not simultaneously have all of such skin changes.

Abstract ID: 65023

Category: JO3: HCV: Diagnosis and Natural History

Diagnosis of hepatic Fibrosis and cirrhosis by liver stiffness measurement in HIV-HCV co-infected patients: a prospective study.

V. de Ledinghen, Hopital Haut-Leveque, Pessac, France, C. Douvin, Hopital Henri Mondor, Creteil, France, A. Kettaneh, Hopital Saint-Antoine, Paris, France, M. Ziol, Hopital jean Verdier, Bondy, France, D. Roulot, Hopital Avicenne, Bobigny, France, P. Marcellin, Hopital Beaujon, Clichy, France, D. Dhumeaux, Hopital Henri Mondor, Creteil, France, M. Beaugrand, Hopital Jean Verdier, Bondy, France

Background.

Complications of cirrhosis are already the main cause of death in HIV-HCV co infected patients. In these patients the assessment of liver fibrosis is considered mandatory but the use of liver biopsy is often precluded by a high rate of refusal and hemostasis disorders. Liver stiffness measurement, a new non-invasive rapid (<5min) and reproducible method allowing to evaluate liver fibrosis at bedside, is already validated in HCV mono-infected patients. The purpose of this prospective study was to evaluate liver stiffness measurement using FibroScan® in co-infected patients and to compare it with blood tests. Methods. We studied 72 consecutive HIV patients (52 males, mean age 42.4 ± 5.9 years, BMI 22.4 ± 3.8) with chronic hepatitis C referred for liver biopsy who had simultaneously liver stiffness measurement. Liver fibrosis was assessed on biopsy sample by two experienced pathologists using METAVIR score. Areas under ROC curves (AUROCs) for the prediction of liver fibrosis were compared between liver stiffness measurement, and other indexes (platelet count < 140 G/L, AST/ALT ratio > 1 and APRI > 2). Results. Liver stiffness values ranged from 3 to 46.4 kPa (median: 6.6 kPa). METAVIR fibrosis stage was F0F1 n=28, F2 n=22, F3 n=5, F4 n=17. Liver stiffness was significantly correlated to the fibrosis stage evaluated by METAVIR score (Kendall’s tau-b=0.48, p < 0.0001). AUROCs of liver stiffness measurement, were 0.72 (95%CI 0.60-0.84) for F&#61619;2, and 0.97 (95%CI 0.94-1.0) for F=4. For the diagnosis of cirrhosis, optimal cut-off value of liver stiffness was 11.8 kPa. For the diagnosis of cirrhosis, AUROCs of liver stiffness measurement were significantly higher than those for platelet count (p=0.02), AST/ALT ratio (0.0001), and APRI (0.01).

Conclusion.

In conclusion, liver stiffness measurement is a promising non invasive method for the detection of fibrosis and cirrhosis in HIV-patients with chronic HCV infection.The use of FibroScan® for the follow-up and management of HCV-HIV co-infected patients could be of great interest and should be further evaluated.

Abstract ID: 67646

Category: JO3: HCV: Diagnosis and Natural History

SEVERE IMPAIRMENT IN QUALITY OF LIFE IN HIV POSITIVES WITH HEPATITIS C BUT NOT WITH HEPATITIS B.

H. L. Tillmann, University Leipzig, Leipzig, Germany, T. Kaiser, Universität Leipzig, Leipzig, Germany, M. P. Manns, Medizinische Hochschule Hannover, Hannover, Germany, R. E. Schmidt, Medizinische Hochschule Hannover, Hannover, Germany, M. Stoll, Medizinische Hochschule Hannover, Hannover, Germany

Introduction:

HIV-infected individuals are frequently infected with different hepatitis viruses. HCV has been associated with impaired quality of life in non-HIV infected patients. Little is known concerning the quality of life in HIV-infected individuals in relation to these different hepatitis viruses.

Patients and Methods:

We investigated a cohort of 250 patients who had answered “HIVSELT” and “EQ-5D” questionnaires assessing quality of life. Data on HBsAg, anti-HBc, anti-HCV, and GBV-C-RNA were available for 191, 188, 189, 98 patients, respectively. HCV-RNA was tested in 33 of 35 anti-HCV positive patients.

Results:

There was no difference in quality of life in relation to active or past HBVinfection defined by HBsAg (n=15) and anti-HBc in the absence of HBsAg (n=84), respectively, for both overall HIV-SELT (p=0.66, and p=0.43, respectively) and visual “EQ-5D” (p=0.93 and p=0.64, respectively). However, anti-HCV positivity (n=35) was associated with significantly impaired quality of life (HIV.SELT overall p<0.001). Importantly, no difference was found in relation to HCV-viraemia in anti-HCV positive patients (p=0.77). In multivariate analysis anti-HCV positivity, employment status, HIV viral load and GBV-C status were relevant to quality of life, with GBV-C being beneficial and HCV being negative.

Conclusions:

While HBV seems to play no role concerning quality of live in HIVinfected patients, the flavi-viruses HCV and GBV-C display differential influence on quality of life. As quality of life was similar impaired in HCV-viraemic and HCV-nonviraemic anti-HCV positive patients but better in GBV-C viraemic patients, this should be taken into account in case of planed interferon therapy, which might fail HCV clearance but clear the beneficial GBV-C.

Abstract ID: 65163

Category: JO3: HCV: Diagnosis and Natural History

The Natural History of Hepatitis C Virus (HCV) Recurrence in Pediatric Liver Transplant

Recipients: An Analysis of National Data. N. R. Barshes, Michael E. DeBakey Department of Surgery, Baylor College of Medicine, Houston, TX, I. W. Udell, Baylor College of Medicine, Houston, TX, T. C. Lee, Baylor College of Medicine, Houston, TX, C. A. O'Mahony, Baylor College of Medicine, Houston, TX, S. J. Karpen, Texas Children's Liver Center, Houston, TX, B. A. Carter, Texas Children's Liver Center, Houston, TX, J. M. Vierling, Baylor College of Medicine, Houston, TX, J. A. Goss, Baylor College of Medicine, Houston, TX

Introduction:

The natural history of HCV infections in pediatric patients differs from the natural history in adults, as pediatric patients generally have a more benign course and occasional spontaneous clearance of the virus. The natural history of the disease following liver transplantation has never been described, however, and only extrapolations of adult studies have given guidance to clinicians offering liver transplantation to pediatric patients with HCV.

Method:

All pediatric patients (age <17 y.o.) that have undergone liver transplantation for HCV were enrolled in this study through the United Network for Organ Sharing liver transplant database. Survival analysis was performed by the Kaplan-Meier product limit estimate.

Results:

A total of 75 pediatric patients have undergone liver transplantation for HCV in the United States since 1988. Median age was 14 y.o. (range <1 y.o. to 17 y.o.). Approximately 75% were 12 y.o. or older. 85% represented de novo cases of HCV, and the remaining 15% occurred after liver transplantation done for a non-HCV indication (including biliary atresia and metabolic liver disease). Fifty-five (73%) represented primary transplants, while 20 (27%) represented retransplants (including two patients who had received 3 transplants each). Patient survival at 5 years was 81% for primary transplants and 53% for retransplants (p=0.003). Recurrent hepatitis or hepatoma accounted for 25% of patient deaths. Graft survival at 5 years posttransplant was 59% for primary transplants and 37% for retransplants (p=0.02). Overall, 20% of patients required retransplantation a median of 308 days (range 1 to 5,186 days) after initial liver transplantation. Of these, 86% were due to HCV recurrence or other complications of HCV. Only 6 (8%) received liver allografts from living donors; the survival of these grafts and their respective donors did not differ significantly from those of deceased donors.

Abstract ID: 64191

Category: JO3: HCV: Diagnosis and Natural History

Long-term outcome of patients with HCV-related, Child' s class A cirrhosis treated with Interferon–alpha (IFN): the impact of sustained virologic response (SVR) on hepatocellular carcinoma (HCC) occurence and mortality.

S. Bruno, Unità di Epatologia , A O Fatebenefratelli e Oftalmico, Milan, Italy, T. Stroffolini, A O S. Giacomo, Roma, Italy, S. Bollani, A O Fatebenefratelli e Oftalmico, Milan, Italy, L. Benvegnù, Dipartimento di Medica Clinica e Sperimentale, Università di Padova, Padova, Italy, G. B. Gaeta, Clinica delle Malattie Infettive, Ospedale Gesù e Maria, Napoli, Italy, M. Persico, Unità di Medicina e Epatologia , seconda Università di Napoli, Napoli, Italy, G. Mazzella, Divisione di Gastroenterologia, Policlinico S. Orsola, Bologna, Italy, A. Ascione, U.O. di Epatologia, A .O. A. Cardarelli, Napoli, A. Mangia, U.O. di Gastroenterologia, IRCCS Casa Sollievo della Sofferenza, San Giovanni Rotondo (FG), Italy, T. Santantonio, Clinica delle Malattie Infettive, Università di Bari, Bari, Italy, P. Andreone, Dipartimento di Medicina Interna, Policlinico S' Orsola, Bologna, Italy, P. . Almasio, Divisione di Gastroenterologia, Università di Palermo, Palermo, Italy, A. Studio Fegato (AISF), AISF, Roma, Italy

Introduction

The relationship between persistence of HCV replication and the development of HCC in patients with coexisting cirrhosis is still not fully understood especially in Western population. A limited amount of data from clinical trials originally designed to assess the efficacy of IFN-based therapies in inducing HCV clearance, suggests a slight protective effect limited to the patients achieving SVR.

Aim

We aimed to re-evaluate the impact of SVR on HCC development and mortality in a large Western population-based group of patients with HCV-related, Child’s A cirrhosis treated with IFN.

Methods

Twenty three Italian referral centres concurred to develop an exhaustive database including all consecutive patients with clinical or histological diagnosis of cirrhosis treated with IFN monotherapy between January 1992 and December 1997. 1214 patients were included in the study (mean age 59.9 ± 8.9, males 62%, 89.3% with histological diagnosis, 55.3% infected by genotype 1, 20.4% by other genotypes and 24.3% with missing data, none with HBV or HIV coinfection).

Results

Overall SVR, based on HCV-RNA undetectability 24 weeks after discontinuation of IFN, was certified in 199 subjects (16.4%).

Younger age, female sex, higher IFN dose and genotype other than 1 were associated with SVR. During follow-up (mean 92.5 ± 39.9 months) 183 HCC (incidence: 1.9% person/years) were diagnosed, 12 in the SVR group and 171 in non responder (no SVR) group (incidence: 0.7% person/years vs. 2.2% person/years, crude H.R.= 3.1, figure 1.).

By Cox regression analysis, three variables were significantly linked to HCC occurrence: male sex (H.R. 2.3, C.I.:1.6-2.3), age older than 58 years at baseline (H.R. 2.4, C.I.:1.7-3.3) and no SVR (H.R. 2.2, C.I.:1.2-4.2). One hundred eighty-one patients died including 11 in SVR group and 170 in the no SVR group (5.5% vs. 16.7%, p<0.001, figure 2). The risk factors associated with mortality were lack of IFN response (H.R. 2.8, C.I.: 1.4-4.1) and diagnosis of HCC (H.R. 5.1, C.I.: 2.8-7.6).

Conclusion:

In conclusion, our data demonstrate that the achievement of SVR after IFN in patients with HCV- related, Child’s A class cirrhosis is independently associated, in the long term, either with a lower risk of HCC development and with a significant increase of life expectancy. Antiviral treatment of cirrhosis because of HCV has hence a sound clinical justification, if aimed at viral eradication. In addition, the author noted that with the higher SVR rates associated with pegylated interferon plus ribavirin would help to increase HCV patient survival.

Abstract ID: 66241

Category: JO3: HCV: Diagnosis and Natural History

Natural History of Hepatitis C Virus Infection Among Injection Users in Vancouver, Canada.

J. Grebely, University of British Columbia, Vancouver, Canada, B. Conway, University of British Columbia, Vancouver, Canada, J. Raffa, University of British Columbia, Vancouver, Canada, C. Lai, BC Centre for Excellence in HIV/AIDS, Vancouver, Canada, M. Krajden, BC Centre for Disease Control, Vancouver, T. Kerr, BC Centre for Excellence in HIV/AIDS, Vancouver, Canada, M. Tyndall, BC Centre for Excellence in HIV/AIDS, Vancouver

Objective:

To measure the rate of virologic clearance in HCV-infected individuals and to compare the rate of new infection to that observed in previously uninfected members of the cohort.

Methods:

The CHASE Project is a prospective cohort study of an inner city population (consisting mainly of IDUs) designed to monitor the uptake of health services and to estimate the incidence of communicable infections. The cohort data was linked with a longitudinal (1992-2005) laboratory database at the BC Centre for Disease Control that includes HCV antibody and PCR assay results. We identified viremic (HCV antibody and PCR positive) and uninfected subjects, as well as new infections, virologic clearance, and subsequent re-infections. Factors associated with HCV infection, clearance, and re-infection were identified by multivariate logistic regression.

Results:

We identified 940 uninfected (633 males, 307 females) and 437 (284 males, 153 females) infected viremic subjects at baseline, with 91/437 (20.8%) spontaneously clearing viremia over a median follow-up of 5.1 [range 0-13.6] years. HCV clearance was associated with female gender (Adjusted OR 2.1; 1.3-3.3; P=0.003), Aboriginal race (AOR 2.6; 1.6-4.3; P<0.001) and a history of HBV co-infection (AOR 5.2; 2.2-17.4, P<0.001) and inversely associated with HIV infection (AOR 0.41; 0.21-0.80 P=0.01). HCV re-infection was documented in 4/91 (4.4%) individuals or an incidence rate of 0.8 cases/100 person-years. Uninfected individuals were more often male (67%), Caucasian (58%) and HIV-negative (95.3%). Over a median follow-up of 2.4 [range 0.01-16.8] years, new infections were documented in 187/940 (19.9%) individuals (9.4 cases/100 person-years).

Conclusions:

In this observational cohort, the rate of spontaneous clearance of HCV infection was consistent with rates reported in non-IDU populations. Clearance occurred more often among women, Aboriginals and those with a history of HBV infection. Spontaneous clearance may confer some protection against re-infection. If such protection extends to those who have cleared their viremia following antiviral therapy, it could provide a stronger rationale for expanding treatment programs to IDUs who continue to be at risk for HCV re-exposure.

Abstract ID: 62567

Category: JO3: HCV: Diagnosis and Natural History

Acute Hepatitis C: Clinical Presentation, Laboratory Findings, and Treatment Outcomes.

R. Loomba, National Institute of Diabetes and Digestive and Kidney Diseases, NIH, Bethesda, MD, R. McBurney, Clinical Center, National Institutes of Health, Bethesda, MD, Y. Park, National Institute of Diabetes and Digestive and Kidney Diseases, NIH, Bethesda, MD, V. Haynes-William, National Institute of Diabetes and Digestive and Kidney Dseases, NIH, Bethesda, MD, G. A. Lutchman, National Institute od Diabetes and Digestive and kidney Diseases, NIH, Bethesda, MD, B. B. Borg, National Institute of Diabetes and Digestive and Kidney Diseases, NIH, Bethesda, MD, J. J. Feld, National institute of Diabetes and Digestive and Kidney Diseases, NIH, Bethesda, MD, B. Rehermann, National Institute of Diabetes and Digestive and Kidney Diseases, NIH, Bethesda, MD, S. K. Herrine, Jefferson Medical College, Thomas Jefferson University, Philadelphia, PA, H. Alter, Department of Transfusion Medicine, Clinical Center, NIH, Bethesda, MD, J. T. Liang, National Institute of Diabetes and Digestive and Kidney Diseases, NIH, Bethesda, MD, J. H. Hoofnagle, National Institute of Diabetes and Digestive and Kidney Diseases, NIH, Bethesda, MD, T. Heller, National Institute of Diabetes and Digestive and Kidney Diseases, NIH, Bethesda, MD

Background:

Acute HCV is often asymptomatic and may escape medical attention. The clinical presentation and course of acute HCV have not been well described.

Aim:

To describe the clinical presentation, outcomes and results of treatment in pts with acute HCV infection.

Methods:

We conducted a retrospective-prospective study, involving pts with acute HCV infection who were seen in the Clinical Center of the NIH. Case definition required a known exposure and/or documented seroconversion. Pts were followed with regular physician visits, serial HCV RNA levels and anti-HCV antibody testing.

Results:

25 pts (14F/11M) diagnosed with acute HCV were followed: 14 Caucasians, 6 African Americans, 3 Hispanics, 2 Asians. Mean age at exposure was 40.8 (range 23-71) yrs. Exposures included: needlestick 40%, occupational 12%, medical exposure 12%, razor 8%, injection drug use 8%, sexual 8% and manicure 4%. 18 pts (72%) were symptomatic and 10 (40%) jaundiced; 2 developed ascites and 1 became encephalopathic. Clinical symptoms developed a mean of 14 wks after exposure. Seroconversion occurred at an average of 9 wks. Genotype distribution was: 62% gt 1, 4% gt 2, 4% gt 3 and 30% unknown. HCV RNA titer fluctuated by >1 log in 62% and was intermittently negative by PCR in 14% of pts. 5 pts spontaneously cleared virus within 17 wks, and the remaining 20 were treated with either interferon (IFN) or peginterferon (PEG) with or without ribavirin (Rbv). 14/20 pts were treated with PEG and Rbv for 24 (9-48) wks after awaiting 20 (4-37) wks post exposure. In HIV-negative pts who were treated within 24 wks of exposure, the response rate was 100% (n =13); 10 with a sustained virologic response (SVR), 3 with an end of treatment response (ETR) but less than 24 wk follow-up. The only treatment failures occurred in HIV +ve pts (2/3 with viral breakthrough, 1 with ETR) and those in whom treatment was started >24 wks after exposure (1/4 non-responder, 2 SVR, 1 ETR). Side effects of therapy were problematic with psychiatric side effects occurring in 90%, depression and/or anxiety in 55% (despite SSRI prophylaxis in some), weight loss in 20%, inability to work during therapy in 15% and autoimmune diseases in 25% of pts.

Summary:

Acute HCV has a variable clinical course and is often associated with marked fluctuations in HCV RNA levels. Treatment with Peginterferon and Ribavirin is highly effective but is associated with significant rates of side effects.

Abstract ID: 63703

Category: JO3: HCV: Diagnosis and Natural History

CHANGES IN VIRAL LOAD AND DISEASE PROGRESSION AT 5 YEARS IN A COHORT OF UNTREATED PATIENTS INFECTED WITH THE SAME STRAIN OF HCV: THE O'BRIEN PROJECT.

V. I. Descalzi, Fundacion Favaloro, Buenos Aires, Argentina, S. M. Soria, Fundacion Favaloro, Buenos Aires, Argentina, A. E. Ruf, Fundacion Favaloro, Buenos Aires, Argentina, N. Massenzio, Fundacion Favaloro, Buenos Aires, Argentina, S. Munne, Laboratorio Infectologia de Avanzada (InAv), Buenos Aires, G. Picchio, Virco Lab, Inc (J&J company), Raritan, NJ, P. Baré, Academia Nacional de Medicina, Buenos Aires, Argentina, J. E. Gonzalez, Laboratorio Infectologia de Avanzada (InAv), Buenos Aires, Argentina, F. G. Villamil, Fundacion Favaloro, Buenos Aires, Argentina

Backgroound:

O'Brien is a small rural town of 2200 inhabitants with a 5.6% prevalence of HCV infection (102/1832) by EIA-3 (12.6% in individuals aged >40 yrs and 23.4% >60 yrs) in 1999 ( AASLD 2000, 426A). RIBA-3 was (+) in 92/102 (90%) and indeterminate for c22-c33 in 10. HCV RNA was detectable in 83 (81%). All viremic patients (pts) were infected with genotype 1b and showed 90.4%-97.5% homology in viral nucleotide sequence suggesting a common source of infection (unsafe injections administered an average of 35 yrs before the study, AASLD 2002, 1600A). Liver biopsy was indicated in viremic pts aged <65 yrs. The overall prevalence of cirrhosis in O’Brien was 28%, 13/52 (25%) on biopsy and 16/50 (32%) on clinical grounds (AASLD 2001, 186A).

Aim:

The goal is to investigate changes in viremia and disease progression after a follow-up period of 5 yrs (1999-2004). Methods: In 2004, HCV RNA levels were investigated in 67/83 (81%) viremic pts and in the 19 non-viremic pts studied in 1999. High viral load (VL) was defined as >5.93 log IU. No pt received antivirals until 01/05.

Results:

During the 5-yr study period, 1/67 viremic pts (1.5%) cleared serum HCV RNA and 2/9 non-viremic pts (10.5%) became HCV RNA (+). Mean VL (log IU) were 6.1±0.6 in 1999 and 6.3±0.9 in 2004 (p=0.13). Significant changes in VL (>1 log) occurred in 14 (21%) pts (increase in 7 and decrease in 7). High VL was observed in 44 pts (66%) in 1999 and 46 (69%) in 2004. Eight pts (12%) converted from low-to-high and 6 (9%) from high-to-low VL. Among the 29 pts with cirrhosis, 1 had decompensated disease and 3 advanced hepatocellular carcinoma (HCC) at the time of the initial diagnosis in 1999 all of whom died within a few months. During the 5-yr follow-up, 5/25 (20%) initially compensated cirrhotics developed decompensation and another 5 (20%) HCC. Two of these pts died and 3 underwent OLT. None of the 73 non-cirrhotic pts developed decompensated disease, HCC or died. In 01/05, 32 pts were started on PEG-IFN alfa-2A + ribavirin of which 30 completed 12 wks of therapy. Mean pre-treatment VL was 6.4±0.8 (high VL in 76%) and mean fibrotic stage (METAVIR) 2.0±1.0 (1999). Early virological response (12 wks) was obtained in 29 (97%) pts (HCV RNA neg in 24 and >2 log decrease in 5).

Conclusions:

1) Over a 5-yr period significant changes in viremia occurred in approximately 20% of pts; 2) Cirrhotic pts with long standing HCV infection had a high risk of decompensation (21%), HCC (28%) and death or need for OLT (31%); 3) The remarkably homogeneous virological response observed at 12 wks suggests that the HCV strains circulating may share common genetic determinants of PEG-IFN/ribavirin susceptibility

Abstract ID: 62224

Category: JO3: HCV: Diagnosis and Natural History

Virological and Histologgical features of hepatitis C virus patitents with normal aminotransferase levels: results of a ten year propsective follow-up study.

C. Vandelli, Modena and Reggio Emilia University, Modena, Italy, Modena, Italy, F. Renzo, Modena and Reggio Emilia University, Modena, Italy, A. Bagni, Azienda Ospedaliera Policlinico, MOdena, Modena, Italy, M. Vandelli, Modena and reggio Emilia University, Modena, Italy, M. Ponz de Leon, University of Modena and Reggio Emilia, Modena, Italy

Introduction:

The natural history of patients (pts) HCVAb+ve and normal ALT (nALT) is not well defined. The aims of our investigation were: to assess the presence of viral replication in subjects HCVAb+ve with nALT and the relationship between viremia, nALT, histologic liver damage; to characterize its natural history in ten years follow-up study. 207 subjects (mean age 49.7 ± 11.6 ys) were selected based on their anti HCV positivity and normality of ALT value. They were enrolled into the study and observed for 10 years. Normal ALT values were defined as < 30 IU obtained at serial monthly evaluation during 6 months. These data were obtained at 0, 5 and 10 years. ALT value was also collected four times every year during the study. HCV RNA was detected by PCR yearly. Liver histology was examined in all pts at baseline and in 75% of the cases at the end of the study. 189 pts were HCV RNA+ve at baseline, 23 tested HCV RNA-ve at the end of follow-up. Mean viral load was 1.320.500 IU/ml. 101 (48.8%) pts had persistently nALT values; in 106 (21.2%) sporadic ALT elevations of 2-3xUNL occurred. The majority of pts was infected by genotype 1. At baseline, liver histology showed no lesions in 1.4% of pts, minimal changes in 24.2%, chronic moderate hepatitis in 40.6%, severe hepatitis with or without cirrhosis in 33.8%. In 96.3% of pts with nALT value a variable degree of grading and staging was observed. Of 101 pts with nALT, 5 had ALT elevations after 5 years followup. 72/96 pts with persistently nALT underwent to a 2nd liver biopsy. A deterioration of liver histology was detected in 15 pts (20.83%). Most patients remained asymptomatic. The majority of anti HCV+ve pts with nALT is HCV-RNA+ve; persistently nALT is observed in 45.4%. Despite absence of biochemical abnormality, liver histology worsened in 20.8% of anti HCV+ve pts with persistently nALT, with some showing significant fibrosis. Longlasting 5 years of follow-up seems to be the required period of time to define normal ALT value carrier. Neither serum HCV-RNA nor ALT level can reliably predict activity or degree of fibrosis, therefore, more sensitive tests for diagnosis of liver damage in anti-HCV+ve pts are required.

Intrahepatic CD8-mediated killing is the probable cause of progressive liver damage in HCV/HIV coinfected patients.

C. Almerighi, Institute of Liver Studies, London, United Kingdom (Great Britain), M. J. Hussain, Institute of Liver Studies, London, United Kingdom (Great Britain), A. H. Mohsen, Department of Gastroenterology, London, United Kingdom (Great Britain), M. Morsy, Institute of Liver Studies, London, United Kingdom (Great Britain), D. P. Bogdanos, Institute of Liver Studies, London, United Kingdom (Great Britain), S. Norris, Hepatology Centre, Dublin, Ireland, B. Portmann, Insitute of Liver Studies, London, United Kingdom (Great Britain), D. Vergani, Institute of Liver Studies, London, United Kingdom (Great Britain)

Aim

To compare the intrahepatic cell infiltrate in HCV monoinfected and HCV/HIV coinfected patients to search for an explanation for the different disease evolutions.

Materials and methods

Formalin-fixed liver biopsies from 34 interferon-naive noncirrhotic patients (16 HCV/HIV; 18 HCV) were scored according to Ishak and analysed for CD3 (pan-T), CD8 (T-cytotoxic), CD4 (T-helper), CD79a (B cells), CD68 (macrophages), CD56 (NK cells), and perforin/granzyme (cytolytic molecules) using a semiquantitative immunological scoring (IS) system (J Pediatr Surg, 2001). Unpaired t test, Mann-Whitney and Spearman’s correlation tests were used as appropriate.

Results

IS for CD4 and CD79a positive cells in the portal-periportal areas was lower in HCV/HIV coinfected than in HCV monoinfected patients (p= 0.001; p<0.001 respectively). Within the HCV/HIV coinfected group, the Ishak grading correlated in descending order of significance with the IS for CD3 (r=0.80; p<0.001), CD8 (r=0.73; p=0.001), granzyme (r=0.70; p=0.002), CD79a (r=0.59; p=0.01), CD4 (r=0.55; p=0.02), CD68 (r=0.55; p=0.02), and perforin (r=0.52; p=0.03) in the portal-periportal areas, while in the HCV monoinfected group, the grading correlated with the IS for CD68 (r=0.58; p=0.01), CD3 (r=0.54; p=0.02), and CD8 (r=0.47; p=0.04) positive cells in the parenchyma.

Conclusion

CD4 depletion in HCV/HIV coinfected patients may affect immunoregulatory cells leading to uncontrolled CD8 T cell mediated hepatocyte damage, as suggested by the strong correlation between CD8/granzyme positive T-cells and severity of histological score.

Abstract ID: 67351

Category: JO2: HCV: Virology

HIV INCREASES THE FORMATION AND ACTIVATION OF MACROPHAGE RESERVOIRS OF HCV IN COINFECTED PATIENTS.

I. Dichamp, Service de Virologie, EA3186 et IFR133, CHU Besançon, Besançon, France, C. Drobacheff, Service de Dermatologie, CHU Besançon, Besançon, France, S. Bresson-Hadni, Service d'Hépatologie, CHU besançon, Besançon, France, J. Miguet, Service d'Hépatologie, CHU besançon, Besançon, France, G. Herbein, Service de Virologie, EA3186, et IFR 133, CHU besançon, Besançon, France, V. Di Martino, Service d'Hépatologie CHU Jean Minjoz, Besancon

Background

HIV accelerates the progression of chronic hepatitis C and increases HCV viral load in coinfected patients, suggesting that HIV directly facilitates the replication of HCV. Such phenomenon may involve extrahepatic sites of HCV replication. Among PBMC, monocyte-derived macrophages (MO) have been reported to harbor both viruses and to act potentially as reservoirs of virions. Replication of both viruses has been reported to depend on the activation of transcription factors such as nuclear factor-kappa B (NF-kB) and activator protein-1 (AP-1).

Aim

The Aim of this study was to assess the role of the HCV core and HIV Nef proteins on NF-kB and AP-1 activation in coinfected patients versus monoinfected patients, by studying PBMC, peripheral blood lymphocytes (PBL) and MO.

Patients and Methods

BMC and purified PBL and MO (> 94% CD14+ by flow cytometric analysis) were prepared from peripheral blood of HCV-, HIV-, and coinfected subjects using Ficoll-Hypaque. In parallel, PBMC, PBL and MO were also prepared from peripheral blood of healthy donors, and treated with purified recombinant HCV core and HIV Nef proteins (100 ng/ml), used alone or together, for different periods of time (0, 0.5, 1 3, 5 hours). Nuclear extracts were prepared from all the cell types and NF-kB and AP-1 activation was measured using electromobility shift assay (EMSA).

Results

Ex vivo, HCV-RNA load vas 5-fold higher in MO from coinfected subjects than in MO from HCV monoinfected subjects ; moreover, high levels of both NF-kB and AP-1 activation were measured in MO from coinfected subjects versus subjects infected only with one of the two viruses. In agreement with the data obtained, ex vivo, both the HCV core protein and the HIV Nef protein activated synergistically in MO in vitro. Such synergistic activation, assessed by a sustained increase of NF-kappaB and AP-1 in MO, was not further observed in autologous PBL, suggesting that gene activation triggered by HCV core and HIV Nef proteins was mainly restricted to the MO cell type.

Conclusion

Our results suggest that both HCV core protein and HIV Nef protein might play a critical role in the formation of macrophage reservoirs in coinfected patients via activation of both NF-kB and AP-1. Moreover a better understanding of the molecular mechanisms involved in HIV/HCV coinfection may ultimately lead to the development of new anti-HIV and anti-HCV treatments, resulting ultimately in the clearance of the extrahepatic reservoirs of virions in coinfected patients.

Abstract ID: 65574

Category: JO1: HCV: Pathogenesis

Analysis of CD1d-depndent NKT Function in Subjects with HCV versus HIV/HCV infection.

R. Baden Herman, Beth Israel Deaconess Medical Center/Harvard Medical School, Boston, MA, M. Kress, Beth Israel Deaconess Medical Center, Boston, MA, J. Murray, Boston Medical Center, Boston, MA, R. Horsburgh, Boston Medical Center, Boston, MA, Q. He, Beth Israel Deaconess Medical Center/Harvard Medical School, Boston, MA, A. Wells, Beth Israel Deaconess Medical Center, Boston, MA, M. Exley, Beth Israel Deaconess Medical Center/Harvard Medical School, Boston, MA, C. S. Graham, Beth Israel Deaconess Medical Center/Harvard Medical School, Boston, MA, M. J. Koziel, Beth Israel Deaconess Medical Center/Harvard Medical School, Boston, MA

Background

Natural Killer T cells (NKT) demonstrate characteristics of both the cellular and innate arms of the immune system and have been implicated in fighting viral infections, although little is known about their role in HCV and HIV. We examined peripheral NKT effector function as it pertained to HIV status, liver histology and antigen-specific CD4 responses from a cohort of subjects with HCV mono-infection versus HCV/HIV co-infection. To our knowledge, this is the first such comparative analysis of NKT function in HCV and HCV/HIV co-infection.

Methods

Blood samples from a cohort with HCV (n=15) or HCV/HIV infection (n=16) were analyzed. Subjects had liver histology and were dichotomized into mild (0-2; n=20) or severe (4-6; n=11) fibrosis. We used ELISpot assays to quantify the CD1d-dependent NKT and antigen-specific CD4 T cells and their cytokine secretion. PBMCs were used as effector cells. For NKT ELISpot, PBMCs were incubated with CD1d transfectants to

stimulate the NKTs. Averaged numbers of interferon gamma (IFNg) spot-forming cells (SFC) in control wells with mock transfectants were subtracted from CD1d-stimulated wells to correct for background cytokine production and reported per 106 PBMC. For CD4 cells, ELISpot was performed using recombinant HCV proteins or recall antigens, and secretion of IFNg, IL-10 and TNF alpha was quantified as the number of SFC after subtraction of buffer control.

Results

Numbers of IFNg secreting NKT cells were correlated with histology score and CD4 T cell IFNg, IL-10, and TNF responses. Subjects were initially stratified based on HIV status and Fibrosis score. There was no significant difference in NKT function between HCV and HIV/HCV groups, nor in subjects with mild fibrosis compared to those with severe fibrosis. (p=0.40 and p=1.00 respectively). There was an overall trend of fewer NKT SFC in subjects with severe fibrosis compared to mild fibrosis in both the HCV (median 140 SFC; IQR 40, 580 versus 360 SFC; IQR 120, 540) and HIV/HCV groups (90 SFC; IQR 40, 600 versus 340 SFC; IQR 240, 420). Overall, there was no correlation between HCV-specific CD4 T cell responses and NKT function in either group. Interestingly, there was a significant positive correlation between TNF responses to Candida, and NKT function in HCV and HCV/HIV subjects with severe fibrosis. (r=0.77, p=0.07 and r=0.9, p=0.04 respectively)

Conclusions: In this cohort with either HCV or HCV/HIV co-infection, there were no significant associations between peripheral NKT effector function, HIV status, disease progression or HCV specific immunity. However, there were interesting trends that warrant further study in a larger cohort.

Abstract ID: 67514

Category: JO1: HCV: Pathogenesis

HIV/HCV coinfected patients display increased perisinusoidal hepatic stellate cell activation and less inflammatory activity.

R. V. Iyer, Stroger Hospital of Cook County, Chicago, IL, M. Atten, Stroger Hospital of Cook County, Chicago, IL, B. M. Attar, Stroger Hospital of Cook County, Chicago, IL,

F. Shi, Stroger Hospital of Cook County, Chicago, IL, A. Orucevic, Stroger Hospital of Cook County, Chicago, IL

Introduction

Persistent hepatic stellate cell (HSC) activation plays a major role in the development of liver fibrosis. The presence of HIV in HIV/HCV coinfected patients accelerates the progression to cirrhosis; however, the mechanism of this accelerated fibrogenesis is unclear.

Aim

Our aim was to explore the difference in HSC activation in HIV/HCV coinfected with HCV monoinfected patients.

Methods

We examined liver tissue from 30 patients with HIV/HCV coinfection and 15 patients with HCV monoinfection. Routinely processed sections were scored for grade of inflammation and stage of fibrosis. α–SMA expression in activated HSCs was evaluated using a monoclonal antibody to α – SMA, streptavidin-avidin-biotin labeled and signal visualized with diaminobenzidine.

The presence of α–SMA immunoreactive HSCs was semiquantitatively scored (ASMA score) in portal tracts/fibrous septa and hepatic lobules (perisinusoidal) as described by Schmitt-Graf. Statistical analysis was done using the t test and logistic regression.

Results

Serum ALT levels and HCV viral loads were similar in each group, and most patients had genotype 1. Seventy-seven percent of coinfected patients had CD4 counts >200. Gender distribution was similar between the two groups with the majority of each group composed of males. Coinfected patients were younger (47±1.6 years) than HCV patients (52.2±1.7 years, p<0.05). The total ASMA score was similar between the two groups (3.33±0.34 vs 2.47±0.40, p=0.13) and the portal/septal ASMA score was similar between the two groups (1.67±0.18 vs 1.53±0.22, p=0.65). However, the perisinusoidal ASMA score was significantly higher in HIV/HCV coinfected patients (1.73±0.18) compared to HIV monoinfected patients (0.93±0.24, p=0.02). Coinfection with HIV was associated with a higher perisinusoidal ASMA score with an OR of 2.1. In both groups, a higher ASMA score was associated with a lower prevalence of inflammation (lower grade) with an OR of 0.46, and an OR of 0.24 when adjusted for HIV coinfection.

Discussion

HIV/HCV coinfected patients progress more rapidly to cirrhosis than HCV monoinfected patients suggesting that HIV-induced immunosuppression contributes to disease progression. These coinfected patients were found to have less hepatic inflammation but increased perisinusoidal HSC activation (lower grade with higher ASMA score). Thus, immunosuppression or immune mechanisms within the liver may be associated with increased HSC activation and more rapid fibrogenesis.

Abstract ID: 66233

Category: JO1: HCV: Pathogenesis

HCV-specific T-cell responses of acutely HCV infected individuals with and without HIV.

M. Danta, Centre for Hepatology, London, United Kingdom (Great Britain), N. Semmo, Nuffeild Department of Clinical Medicine, Oxford, United Kingdom (Great Britain), J. Northfield, Nuffeild Department of Clinical Medicine, Oxford, United Kingdom (Great Britain), D. Brown, Centre for Hepatology, London, United Kingdom (Great Britain), G. Dusheiko, Centre for Hepatology, London, United Kingdom (Great Britain), P. Fabris, Department of Infectious Disease, Vincenza, Italy, S. Bhagani, Department of HIV medicine, London, United Kingdom (Great Britain), P. Klenerman, Nuffeild Department of Clinical Medicine, Oxford, United Kingdom (Great Britain)

Aim

Control of HCV requires a sustained cellular immune response. We assessed HCVspecific T-cell responses and serum cytokines in HIV-positive and HIV-negative individuals with acute HCV infection (16% vs 46% spontaneous HCV clearance).

Methods

Acute HCV was defined by seroconversion to anti-HCV within six months of a negative result and/or a positive HCV RNA. Frozen PBMCs and serum from multiple time points in the acute phase of HCV infection from a cohort of HIV-positive individuals in London and an Italian cohort of HCV mono-infected individuals were used. HCV-specific T-cell responses were assessed using an IFN-g ELISpot for HCV core derived peptides (20mers overlapping by 10aa) and HCV recombinant non-structural proteins (NS3-5). Proliferative T-cell responses were assessed with flow cytometric CFSE dye dilution using the same HCV-specific antigens and p24 protein. Serum cytokines were analysed using cytokine bead array and FACS analysis.

Results:

HIVpositive individuals (n=18; all male, mean age 33, mean CD4 714 cells/ml) were compared with HCV mono-infected (n=13; 6 male, mean age 40) individuals. Comparison of IFN-g ELISpots for NS3-5 proteins revealed significantly reduced responses in HIV-positive vs. HIV-negative individuals (1/10 vs. 5/6, p=0.008). No difference was seen for the core peptides (3/10 vs. 4/6, p=ns). There was no significant CD4 proliferation to either core peptides, non-structural proteins or p24 antigen on the CFSE assays in the co-infected samples (n=11). These responses did not change over time. The cytokine patterns were distinctly different between the co-infected (n=18) and mono-infected (n=13) patients. Co-infected patients had significantly lower IL-2 levels (3.30 + 0.06 vs. 8.57 + 0.47 pg/ml, p<0.001) and higher IFN-g concentrations (130.5 + 5.89 vs. 80.91 + 4.05 pg/ml, p<0.001). While there was a strong relationship between IFN-g and ALT in HCV mono-infected persons (R=0.5, p=<0.001), there was no relationship between IFN-g with ALT in co-infected patients (R=-0.23, p=ns).

Conclusions

Failure of early immunological control of HCV in HIV-positive individuals is supported by the finding of weak antigen-specific T-cell responses to HCV antigens. Responses to NS3-5 are particularly associated with resolving HCV infection, and these were specifically lost in the co-infected group. The low IL-2 levels are consistent with the lack of significant CD4 proliferation, implying a deficient T-helper response. In addition, the loss of the normal relationship between inflammatory cytokine release and ALT suggests that non-specific responses are mounted which may fail to efficiently control HCV in the acute phase of infection.
Abstract ID: 63895

Category: JO5: HCV: Clinical Trials and Therapeutic Developments

Predictive Factors in the Treatment of Acute HCV-Infection in HIV-positive individuals – Interim Analysis of a Large German Multicenter Study.

M. Vogel, Medizinische Klinik und Poliklinik I, Bonn University, Bonn, Germany, A. Baumgarten, Practice Dupke/Carganico/Baumgarten, Berlin, Germany, G. Klausen, Praxiszentrum Kaiserdamm, Berlin, Germany, T. Lutz, Practice Gute/Locher/Lutz, Frankfurt/M, Germany, S. Mauss, Practice Mauss/Schmutz/Carls, Düsseldorf, Germany, A. Theisen, Practice Wiesel/Theisen, Köln, Germany, P. Gute, Practice Gute/Locher/Lutz, Frankfurt/M, Germany, M. Rausch, Practice Freiwald/Rausch, Berlin, Germany, D. Schranz, Practice Schranz, Berlin, Germany, B. Bieniek, Pracice Bieniek/Cordes, Berlin, Germany, C. Hoffmann, HIV-Ambulanz der Universität im Städtischen Krankenhaus, Kiel, Germany, A. Trein, Practice Trein, Stuttgart, Germany, K. Schewe, Practice St. Georg, Hamburg, Germany, K. Ummard, Praxiszentrum Kaiserdamm, Berlin, Germany, A. Carganico, Practice Dupke/Carganico/Baumgarten, Berlin, Germany, U. Spengler, Medizinische Klinik und Poliklinik I, Bonn University, Bonn, Germany, J. K. Rockstroh, Medizinische Klinik und Poliklinik I, Bonn University, Bonn, Germany

Background

Previous data from retrospective analysis of acute hepatitis C in HIVcoinfected patients suggested high sustained virological response rates in 90% of the patients. In this prospective trial the efficacy and safety of early treatment of acute hepatitis C in HIV-coinfected patients is further examined.

Methods

Multicenter, prospective, non-randomized clinical trial. Main inclusion criteria were HIV-positive patients with a CD4-count > 300/μl and an acute HCV-infection defined by the simultaneous presence of two of the following three criteria within four months prior to diagnosis.

1) known or suspected exposure to HCV,

2) documented seroconversion to positivity for antibodies against HCV,

3) a serum ALT level of more than 350 IU/l with documented normal levels during the year before infection.

Patients were treated with pegylated interferon over 24 weeks.

Results

End of treatment results at 24 weeks are available in 24 patients (all male, median age 38 years). At baseline median ALT was 248 IU/l. Median HCV-RNA was 534 568 IU/ml, median CD4-count 447/μl. HCV-genotypes were 1 (n=14), 2 (n=2), 3(n=4), 4 (n=2), 1 and 2 (n=1) and other (n=1). 63 % of the treated patients (15/24) showed an end of treatment response with a negative HCV-NA at week 24. No difference in response was observed between HCV-genotype 1 or 4 (10/17 patients, 59%) and 2 or 3 (4/6 patients, 63%). It was found that early ribavirin significantly improved long-term virological outcomes compared to those who started ribavirin later (63% vs. 23%).

Predictive factors for an end of treatment response were a negative HCV-RNA at week 4, 8 or 12, high maximum ALT-elevation and a documented anti-hcv seroconversion within 4 months prior to diagnosis. On the contrary, dose reductions of interferon or ribavirin were significantly associated with treatment failure, defined as detectable HCV-RNA at week 24.

Conclusions

In this prospective study we show that early interferon therapy of acute HCV-infection achieved high virological response rates in patients with HIV-coinfection.
Abstract ID: 67610

Category: JO5: HCV: Clinical Trials and Therapeutic Developments

Sustained Virologic Response (SVR) to Retreatment with IFN alfacon 1 + ribavirin in HCV/HIV coinfected patients who had failed 12 weeks of PEG IFN alpha 2 + ribavirin therapy.

C. Leevy, New Jersey Medical School Liver Center, Newark, NJ, C. Chalmers, InterMune, Brisbane, CA, L. M. Blatt, InterMune, Brisbane, CA

Background

Patients who are coinfected with HCV and HIV are at significant risk for progression to liver fibrosis, cirrhosis and HCC and thus HCV viral eradication is an essential goal of therapy. SVR rates in treatment naive HCV/HIV coinfected patients who are treated with PEG IFN alfa 2a + ribavirin (RBV) are ~40%. We have recently reported that IFN-alfacon 1 + RBV retreatment resulted in a SVR of 37% in monoinfected Chronic Hepatitis C patients who failed therapy with PEG IFN alpha 2 + RBV.

Methods

All patients (N=61) received PEG IFN alfa 2a + RBV for 12 wks, and did not achieve > 2-log10 drop in HCV RNA. With no washout, patients were retreated with IFN alfacon-1, 15 mg SQ daily, and weight-based RBV for 72 wks. 57(93%) patients had HCV genotype 1, mean baseline viral load was 6.1 ± 0.4 log10 IU/mL and 32 (53%) patients were African American. Mean baseline CD4 counts were 516 ± 28 cell/mL.

We now report herein retreatment of HCV/HIV coinfected patients with IFN-alfacon 1 + RBV who failed to achieve a ³ 2 log10 reduction in HCV RNA following 12.

Results

Therapy was well tolerated by most patients and flu-like symptoms were the most commonly reported side effect. Hematological toxicity required the use of Epoetin alfa in 32 (53%) patients and Fligrastim in 27 (44%) patients. 10 patients (16%) required both Epoetin alfa and Fligrastim. No patient was withdrawn from therapy.

Conclusions

These results suggest that a significant number of HCV/HIV coinfected patients who have failed treatment with PEG IFN alfa 2a + RBV can achieve a SVR when retreated with IFN alfacon-1, 15 mg SQ daily, and weight-based RBV for 72 wks. Further study is warranted to confirm these initial observations.

Abstract ID: 63917

Category: IO1: Clinical: Hepatocellular Carcinoma
and Cholangiocarcinoma

Body mass index influences hepatic function and survival prognosis in patients with HCV-related hepatocellular carcinoma.

T. Mizuta, Saga Medical School, Saga, Japan, I. Ozaki, Saga Medical School, Saga, Japan, Y. Ide, Saga Medical School, Saga, Japan, K. Yamamoto, Saga Medical School, Saga, Japan

Aim

Obesity is reported to influence the clinical course of chronic hepatitis C, but little is understood about the influence of BMI in the patients with HCV-related hepatocellular carcinoma (HCC). We analyzed the relationships between BMI and hepatic function or survival prognosis.

Methods

One hundred fifty HCV-Ab positive patients with HCC were classified into four groups; lean (L)( <22, n=57), normal (N)(22-25, n=56), overweight (OW)(25-28, n=29), and obesity (OB)(>28, n=8) according to their BMI at first the occurrence of HCC. Differences of age, gender, alcohol consumption, smoking (pack/year), diabetes, hepatic function (ALT, albumin, prothrombin activity, and Child-Pugh class) among the four groups were examined. Cumulative survival rates after treatment were compared among the four groups by Kaplan-Meier method in respect to early stage patients by TNM stage I or II (n=80). Moreover, we analyzed what kinds of host-related factors were associated with the survival period using a multivariate Cox proportional hazards model.

Results

The age of the L group was significantly higher than that of the N and OW groups (L: 68.2, N:64.7, OW: 63.3, OB: 66.6). Serum ALT level of the L group was significantly lower than that of N group (L: 64.0, N: 80.9, OW: 74.6, OB: 70.6). Serum albumin concentration and prothrombin activity levels gradually decreased as BMI increased (ALB (g/dl) L: 3.6, N: 3.5, OW: 3.3, OB: 3.2, PT (%) L: 74, N: 67, OW: 61, OB: 66). The ratio of Child-Pugh C class in each group increased with BMI (L: 5%, N: 18%, OW: 31%, OB: 38%). There were no differences in gender, alcohol, smoking and diabetes among the four groups. The cumulative survival rate of the OB group was significantly lower than that of the other groups. Multivariate analysis showed that BMI>28 was the only host-related factor associated with poor prognosis of early stage HCC.

Conclusions

The authors concluded that Body Mass Index was the only factor clearly associated with early poor prognosis of early stage HCC. In addition, the reduction of body weight may improve HCV disease progression by delaying the development of liver cancer.