Monday Parallel Session 24, October 30, 2006

HCV Diagnosis, Natural History and Epidemiology

 

 

163. HIV has a significant impact on the early HCV host-viral responses. 

M. Danta; N. Semmo; D. Brown; C. A. Sabin; P. Fabris; S. Bhagani; P. Klenerman; G. M. Dusheiko.

Background:

HCV is associated with accelerated liver injury in HIV co-infection. The outcome and natural history of HCV depends on early host-viral interactions. We studied the acute phase HCV host-viral interactions in a cohort of HIV co-infected individuals.

 

Methods:

Acute HCV was defined as seroconversion to anti-HCV ± ALT >10x ULN and a positive serum HCV RNA. Clinical parameters were compared before and after HCV infection. Serum and PBMCs samples were collected over the first 12 weeks from diagnosis of acute HCV. The E1/E2 region of the HCV genome was amplified and cloned using RT-PCR and TOPO TA cloning in 8 co-infected patients (mean 9 clones at 2-3 time-points). Diversity was calculated by pairwise comparison of the clones and complexity was described by determining the number of variants at each time-point, respectively. Selection pressure was assessed by calculating dN/dS ratios for this region. Cell-mediated responses were compared in 14 co-infected and 8 HCV mono-infected patients. HCV-specific T-cell responses were assessed using an IFN-γ ELISpot for HCV core derived peptides (20mers overlapping by 10aa) and HCV recombinant non-structural proteins (NS3-5).

 

Results:

A cohort of 55 individuals was identified (mean age 35.6, all male, median peak ALT 389, median HCV viral load 6.0 log IU/ml, median CD4 554 cells/µl, HAART exposure 64%). Mono-infected individuals (n=8; 5 males, mean age 32.5) were used for ELISpot comparison. HCV persisted in 52 (95%) co-infected individuals. HCV had no impact on CD4/CD8 counts or detectable HIV viral load. In co-infected patients, there was a significant increase in the HCV median genetic diversity (0.57% to 2.43%, p=0.04) and a trend to increased complexity (4.5 to 8 variants, p=0.1) between the first and subsequent time-points. All E1/E2 dN/dS ratios were <1 at each time-point. Comparison of IFN-γ ELISpots for NS3-5 proteins revealed significantly reduced responses in HIV co-infected versus HCV mono-infected individuals (1/10 vs. 5/6, p=0.008). No difference was seen for the core peptides (3/10 vs. 4/6, p=ns).

 

Conclusions:

HIV co-infection is associated with increased rates of HCV persistence. While there was evidence of increased HCV diversity and complexity over time, the dN/dS ratios remained <1, implying immune selection pressure was not driving the viral diversity. The paucity of detectable CD4 T-cell responses, which were infrequent and limited in breadth, supports a lack of immune pressure. While NS3-5 responses are particularly associated with resolving HCV infection, these were specifically lost in the co-infected group. HIV has a significant impact on the early HCV host-viral interaction.

 


164. Significant Liver Disease Progression among HIV/HCV coinfected Persons with Minimal Fibrosis on Initial Liver Biopsy. 

M. Sulkowski; S. Mehta; M. Torbenson ; Y. Higgins; R. Moore; D. Thomas.

Background:

Current guidelines suggest that HCV treatment may be deferred in persons with minimal fibrosis (Hepatology 2004;39:1147-71). To assess the applicability of this recommendation to HIV-infected persons, we prospectively examined the change in fibrosis stage between paired liver biopsies (bxs) among HIV/HCV coinfected pts attending an urban HIV clinic

 

Methods:

177 pts underwent two liver bxs. Paired bx were simultaneously evaluated by a single pathologist blinded to bx sequence and scored according to the Ishak criteria from F0 (no fibrosis) to F6 (cirrhosis). Logistic regression analysis was used to identify non-histological and histological correlates of progression of fibrosis (> 2 stages).

 

Results:

Of the 177 pts, 10 were excluded due to cirrhosis on the 1st bx. Characteristics at 1st bx included: median age, 44 years (yrs); male, 78%; Black, 78%; CD4 < 200/mm3, 23%; HIV RNA < 400 c/mL, 59%; median BMI, 24.8; alcohol abuse, 27%; persistently elevated AST, 31%; no steatosis, 68%. Between bxs, 82% and 12.5% received HIV and HCV treatment, respectively. Median time between bxs was 2.91 yrs (IQR, 2.47 – 3.39 yrs). Fibrosis increase ≥ 2 stages was observed in 37 (22%) of pts whereas a 1 stage decrease occurred in only 11 (6.5%) pts (8, F1 at 1st bx and 3 F2 at 1st bx). Among persons with mild fibrosis (≤ F1) at 1st bx, 22% had evidence of 2-stage progression. Compared to those with no progression, pts with a 2-stage fibrosis increase were more likely to have persistently elevated (> 100 IU/L)AST and/or ALT levels at bx 1 (p=0.01) and between bxs (P = .01). Additional analysis are planned to assess the relationship of fibrosis progression and HIV or HCV treatment.

 

Conclusions:

Unexpectedly, 22% of coinfected pts with mild fibrosis on initial liver bx had significant fibrosis on subsequent bx. If confirmed by others, these data do not support the application of current HCV treatment guidelines to HIV-infected pts on the basis of a single liver biopsy, and suggest such pts should be closely monitored for liver disease progression.

 

Comparison of Fibrosis scores at Biopsy 1 and Biopsy 2

Fibrosis at Biopsy 1 (Column)

 

 

 

 

 

 

 

 

Fibrosis at Biopsy 2 (Row)

F0

F1

F2

F3

F4

F5

F6

Total

F0

42

20

12

2

2

3

1

82

F1

8

20

12

6

2

0

2

50

F2

1

2

11

2

0

0

0

16

F3

0

0

1

7

3

5

2

18

F4

0

0

0

0

0

0

1

1

Total

51

42

36

17

7

8

6

167

 


165. Longitudinal analysis of fibrosis progression in chronic hepatitis C using analysis of multiple biopsies and Generalized Estimating Equations (GEE). 

B. B. Borg; D. E. Kleiner; J. Liang; J. H. Hoofnagle.

Background:

Liver biopsy is the gold standard for determination of stage of liver disease, but it cannot predict future progression. Attempts to develop predictive models for fibrosis in chronic hepatitis C have used various biochemical and clinical parameters. Since the linearity of progression of fibrosis cannot be assumed, multiple liver biopsies are needed to strengthen the predictive power of these models.

 

Aim:

To identify predictive factors of progression of fibrosis in pts with chronic hepatitis C using multiple liver biopsies in a longitudinal analysis using GEE models.

 

Methods:

1471 liver biopsies from 797 HCV-infected pts who were followed by the Liver Diseases Branch at the Clinical Center of the NIH were analyzed. Biopsies were scored for necroinflammatory activity and fibrosis by a single pathologist using Knodell and Ishak scoring systems. Preliminary analysis of variables from first biopsies were done using parametric and non-parametric tests. An exploratory analysis was performed by univariable logistic regression and logit plots. A GEE model was then applied using results from repeated measures. Multiple correlation matrices were used to evaluate the effects of the working correlation matrix. Optimal models were developed by backward, stepwise elimination. All analysis was done using SAS 9.0 and E Guide 2.1 software.

 

Results:

476 pts were treatment-naïve and 321 were enrolled in different therapeutic protocols. 396 pts had one biopsy, and 401 had 2 to 7 biopsies each. The mean ± SD of time interval between biopsies was 3.0 ± 2.7 yrs. In the analysis of the first biopsies, 494 (62%) patients had no or mild fibrosis, and 297 (37.2%) had moderate to severe fibrosis. Pts with severe fibrosis were more likely to be male and to be African American. They were also older, had higher BMI and longer duration of infection.

In the initial GEE model, increase in age, BMI, serum levels of AST, IgA, AFP, ferritin, and creatinine, and liver biopsy necro-inflammatory, fat, Fe and Cu scores were significantly associated with advanced fibrosis. A decreasing ALT and platelet count were also predictive of advanced fibrosis. In the final optimized model, increase in BMI (OR=1.05 p=0.024), serum AST (OR=1.2 p=0.006), HAI scores (OR=1.19 p<0.0001) and hepatic Fe scores (OR=1.56, p=0.008) and decrease in ALT (OR=1.08, p=0.025) and platelets (OR=1.04, p<0.0001) were predictive variables of progression of fibrosis.

 

Conclusion:

Biochemical markers of liver disease can be used reliably as predictors of progression of fibrosis. Prospective randomized controlled trials will be needed to confirm these effects and the effect of therapy on fibrosis.

 

 
166. Use of the State of Connecticut hepatitis C registry to hepatitis C-related patient care. 

A. N. Sofair; M. Fine; S. Huie-White; S. Speers; A. Roome; N. Forbes; N. Hulten; S. Bialek; B. P. Bell.

Background

The assessment of hepatitis C virus (HCV) infection requires confirmation of a positive anti-HCV test with a recombinant immunoblot assay (RIBA) or detection of HCV RNA. In Connecticut, all laboratory tests indicative of HCV infection are reportable to the Connecticut Department of Public Health. During 2004, the Yale University Emerging Infections Program, Waterbury Health Department and Connecticut Department of Public Health evaluated positive anti-HCV tests reported in the city of Waterbury, CT (2004 population 108,429).

 

Methods

Using information available on reported positive anti-HCV laboratory reports for Waterbury residents, we contacted medical providers who ordered these tests to determine the reason for testing, risk factors for infection including intravenous drug use (IDU), contact with an individual with hepatitis, or receipt of a blood transfusion prior to 1992. A provider data form was completed either by the provider or Emerging Infections Program staff.

 

Results

In 2004, 376 positive anti-HCV reports were received. Patient address or phone number were available for 350 (93%). The median age among those reported as having a positive anti-HCV test was 41 years (range 2-67 years); 237 (63%) were male. Provider data were obtained for 326 (88%). Using these data, the most commonly reported reasons for ordering anti-HCV testing were to screen asymptomatic patients with risk factors for HCV infection (30%) or evaluate patients with abnormal liver enzymes (21%). Information on risk factors for HCV infection was available through physician report or chart review in 205 (63%) patients and included IDU (41%), contact with a patient with hepatitis (10%) and blood transfusion before 1992 (6%). Confirmation of the initial positive anti-HCV tests was performed in 112/326 (34%) patients, of whom 17 were positive by RIBA and 66 by detection of HCV RNA. Liver enzyme test results were available for 168 (52%) patients in whom 62 (37%) were abnormal.

 

Conclusion

Over 65% of patients who tested positive for anti-HCV did not receive additional testing to confirm the diagnosis. The lack of confirmatory testing for HCV infection or evaluation of liver enzymes suggests that patients may not receiving adequate evaluation for HCV infection.


167. Injection drug users who resolve the HCV virus appear to be protected from reinfection. 

S. Currie; D. Tracy; J. Ryan; T. Belaye; M. Kim; A. Monto.

Introduction:

Injection drug use (IDU) is a primary and efficient route of infection for the HCV virus. Recent NHANES survey data suggest that at least 48.4% of all persons with HCV antibodies have a history of IDU. However, many persons with recent or active injection drug use who are chronically infected with the hepatitis C virus (HCV) do not receive HCV antiviral treatment due to a concern of possible HCV reinfection. Previously reported studies in Chimpanzees suggest that clearance of HCV is associated with immunologic resistance to reinfection. There are limited studies investigating reinfection in these patients.

 

Objective:  

To determine whether persons with a history of IDU can get reinfected with HCV.

 

Methods:  

Data were prospectively collected from 1997 to date on 427 persons with a current or history of drug use. Baseline and 6 month followup data were collected that included demographic, socioeconomic and risk behaviors such as ongoing IDU, alcohol and sexual behaviors. Serum samples were collected and tested for HCV antibody and HCV RNA positivity. Persons were included in this study if they were confirmed HCV antibody positive with RNA levels undetectable by qualitative assay (Roche Cobas Amplicor 2.0) in serum through spontaneous viral resolution or HCV antiviral treatment. Reinfection was defined by the presence of RNA positive samples at > 2 visits preceded by documented HCV clearance. 43 HCV resolvers were included in this analysis.

 

Results:  

The median age of the study group was 46 years (range of 30 – 71), 65% were male, 46.5% were Caucasian, 18.6% African American and 18.6% Latino. 26 (60.5%) of the 43 had baseline incomes < $10,000. 12 (28%) of the 43 persons continued to inject drugs for a total of 682 months or 56.83 person years of potential HCV exposure to reinfection. Of these persons, 0 (0%) were reinfected. Similarly, 0 (0%) were reinfected in the non-IDU group.

 

Conclusion:  

These data suggest that despite ongoing IDU risk and potential ongoing exposure to the HCV virus, reinfection is highly unlikely and should not be a barrier to initial HCV antiviral treatment. Further study should be conducted to evaluate possible factors associated with protective immunity


168. Chronic Hepatitis C in a High-Risk Pregnant Population: Screening, Perinatal Complications and Postnatal Follow-Up of Mother and Infant. 

E. Berkley; K. K. Leslie; C. R. Qualls; S. Arora; J. C. Dunkelberg.

Objective:

To assess perinatal complications, documentation of hepatitis C viral (HCV) antibody status, and postnatal referral of pregnant women and neonates at high risk for hepatitis C infection.

 

Methods:

A prospective historical study was performed of pregnant patients from Milagro, a drug dependence and treatment program at the University of New Mexico Hospital. Multivariate logistic regression statistical analysis was performed to assess the relative contributions of multiple factors (HCV antibody status, drug use and methadone use) to outcomes.

 

Results:

Three hundred and forty three women with a total of 351 pregnancies from the Milagro program received prenatal care between January 2000 and January 2006. One hundred and fifty five pregnant women (45.2%) were found to be HCV antibody positive, 137 (39.9%) tested negative for HCV antibody, and 51 (14.9%) were not evaluated for hepatitis C. Methadone use was highly associated with HCV antibody positivity (59.7% vs. 23.4%, p<0.001). The incidence of cholestasis of pregnancy was increased in mothers who were HCV antibody positive (10/159 vs. 0/141, p=0.002). In women who used methadone and were HCV antibody positive, there was a nine-fold increased risk of preterm delivery (p=0.05, OR 9.27, CI 1-85.95). Neonates born to HCV antibody positive mothers had a lower birthweight (2804 vs. 2943 g, p=0.047) and were more likely to weigh less than 2500 g (40% vs. 18.4%, p=0.003), an outcome associated with methadone use, not HCV antibody positivity. Neonates born to HCV antibody positive mothers had an increased incidence of NICU admission (20.8% vs. 12.1%, p=0.045). Only nine of 159 HCV antibody positive women were referred for evaluation for chronic hepatitis C in the peripartum period. Only 3 of 159 neonates born to HCV antibody positive mothers received referral for follow-up for chronic hepatitis C.

 

Conclusion:

HCV antibody positive pregnant women have an increased risk for cholestasis of pregnancy. In pregnant women involved in a drug dependence and treatment program, HCV antibody positivity is associated with methadone use, and neonates born to these women are at increased risk for preterm birth, low birthweight, and NICU admission. Postnatal referral of mother and infant for evaluation of hepatitis C is suboptimal.