icon-folder.gif   Conference Reports for NATAP  
  10th Conference on Retroviruses and Opportunistic Infections
Boston, Mass, Feb 10-14, 2003
Back grey_arrow_rt.gif
HCV, GBV-C, and Hepatic Complications in HIV infected Persons
Written by David Thomas, MD, Johns Hopkins University College of Medicine
  Five papers were selected for oral presentations in Session 30, "HCV, GBV-C, and Hepatic Complications in HIV infected Persons"-- two dealing with very different aspects of HCV and three focused on the interaction between GB virus C and HIV.
Abstract 155. Margaret Ragni from the University of Pittsburgh provided an update on their experience with liver transplantation in persons with HIV.
Dr. Ragni compared the outcome of liver transplant in 23 HIV infected persons with what would be expected based on the UNOS database of 11,453 HIV uninfected persons who underwent liver biopsy. The 23 HIV infected patients undergoing liver biopsy had a median age of 47 years; 19 were male; 14 HCV infected; 3 had fulminant liver failure from acute hepatitis B infection, hepatitis A infection, or in association with neviripime ingestion. Before transplant, the median viral load was <400 copies/ml (range 400 - 179,000), median CD4 lymphocyte count was 200 cells/mm3 (range 76-506); and median MELD score was 15 (7-33).
The central finding of the study was the lack of difference in survival after orthotopic liver transplant comparing these HIV-infected persons with expectations from the UNOS database, after adjusting for age and race. In the HIV positive group, survival was 90.9%, 75.9%, and 75.9% at 12, 24, and 36 months, respectively, compared with expected survivals of 86.6%, 82.3%, and 79.2% of HIV negative persons in UNOS at the same intervals. HIV/HCV coinfected persons did worse and had greater mortality than those with HIV but not HCV infection (eg. patients with HIV/HBV) but had a similar mortality to HCV positive, HIV negative persons in UNOS. Adverse outcomes chiefly occurred due to drug-drug interactions (tacrolimus and nelfinivir) as well as due to initiation of interferon and ribavirin therapy for HCV infection in the early post transplant period. Interestingly, the patient's HIV stage pre-transplant (CD4 count and HIV load) was not associated with a poorer outcome.
Editorial note: Although small, this is an important study because it shows that HIV infected persons with liver failure can survive orthotopic liver transplant, which is virtually the only hope patients with end stage liver disease have of survival. It was pointed out that the average MELD scores, an index of severity, were low (low is better), raising the question of whether outcomes might be worse in other settings (edit note: suggesting that results might not be as good if patient's MELD score was higher). In addition, it would be helpful to know the liver biopsy results of surviving patients to evaluate the rate of fibrosis progression. Nonetheless, the study demonstrates the feasibility of doing liver transplantation in HIV infected persons and underscores the importance of clinical experience in managing these complex patients in the post-operative period. The research sets the stage for larger studies to more precisely characterize safety and the effect of various interventions to further enhance survival.
Abstract 158. Morphine enhances hepatitis C virus replicon expression. Presented by Dr. Li of the Children's Hospital of Philadelphia.
This study examined the hypothesis that opiates will affect replication of HCV infection in vitro. To test the hypothesis, the investigators first demonstrated that Huh 7 and Huh 8 cells express the opiate receptor. They then appeared to show that addition of morphine to Huh 8 cells resulted in increased production of HCV replicon mRNA, an effect that was blocked by the opiate receptor antagonists naltrexone and funaltrexamine. Morphine also appeared to diminish the inhibitory effect of interferon a on Huh 8 cells. It appeared that this effect was mediated through the NF-kb promotor since it was activated by morphine and the effect of morphine on HCV replicon transcription was abrogated by caffeic acid phenyl ester, an inhibitor of NF-kb activation. The investigators also showed that culture supernatants from CD8 positive cells taken from HCV infected persons and stimulated ex vivo would inhibit the replicon system, an effect that was partially blocked by morphine. Collectively, these data suggest that morphine increases expression of the HCV replicon system through activation of NF-kb and that, in vitro, it might interfere with effects of interferon a and other endogenously produced cytokines.
Editorial note. Although these well-controlled experiments were intriguing, few of the findings are supported by complimentary clinical data. In one study, elevated HCV RNA levels were found in persons actively using drugs (Thomas et al J Infect Dis). However, persons taking methadone appear to respond to interferon a as well as genotype matched persons not taking opiates (Sylvestre D et al). Clearly, more work is needed to corroborate these findings.
There were three papers examining the effect of GB virus C (GBV-C) on HIV progression. GBV-C is a virus from the same family as HCV that was discovered in persons who developed hepatitis after blood transfusions, and thus initially named "hepatitis G virus". Since then, it is fairly clear that the virus does not cause hepatitis. In fact, the only convincing clinical association with GBV-C is a protective effect on HIV progression.
Abstract 156. GBV-C infection inhibits CCR5 and CXCR4 HIV strains and alters chemokine and cytokine gene expression in PBMC cultures, presented by Jack Stapleton of University of Iowa.
PBMC and CD4 enriched T cell cultures were infected with either clinical isolates of GBV-C taken from patients or mock controls, then infected with well characterized HIV strains that utilize either CCR5 (R5) or CXCR4 (XR4) receptors and effects on HIV replication were examined chiefly by production of p24. GBV-C isolates inhibited both R5 and XR4 HIV strains. Inhibition of p24 was noted when cells were pre-infected with GBV-C and was greater when cells were GBV-C infected 48 vs 24 hours before HIV infection. IL-13 production was diminished, while IL2 and IL8 production were increased in GBV-C pre-infected cells. In addition, GBV-C infected cells had increased expression of RANTES, MIP-1a, MIP-1b, and SDF-1 compared to the controls and inhibition of R5 virus was diminished by antibodies to RANTES, MIP-1a, MIP-1b, and SDF-1. The investigators concluded that "GBV-C inhibits R5 and XR4 strains of HIV in vitro, with maximal inhibition occurring 48 hours post-infection. GBV-C infection of PBMCs appears to increase expression of CCR5 chemokines like RANTES, MIP-1a, MIP-1b, and SDF-1. Survival benefits associated with GBV-C coinfection may reflect a direct inhibitory effect of GBV-C on HIV replication."
Abstract 157. GB virus C viremia during the natural course of HIV-1 infection: viremia at diagnosis does not predict mortality, presented by Dr. Bjorkman from Malmo University in Sweden.
In this clinic-based investigation, survival among 230 HIV infected persons was compared according to GBV-C status at baseline and at a later time point in a subset of 163 patients. Baseline was within two years of diagnosis (not HIV seroconversion) of HIV infection and follow-up was for a median of 4.3 years. At baseline, the median age was 35 years, 89% were male, 86% Caucasian, and the median CD4 count was 465 cells/mm3. GBV-C viremia was found in 62 (27%) of patients, while anti-E2 (a marker of past GBV-C infection) was found in 69 (30%). No difference was detected in the incidence of AIDS or overall survival according to baseline GBV-C status. However, 11 of 44 patients who were GBV-C RNA positive at baseline lost GBV-C RNA without developing detectable anti-E2 (an antibody response that is believed to occur when persons clear GBV-C infection), and a marked increase risk of mortality was detected in these 11 patients. Eight of the 11 patients who resolved GBV-C RNA without developing anti-E2 died (P=0.007), 10 developed AIDS (P<0.001), and their average CD4 decline was greater (145 cells/mm3/year), compared with those who had repeated GBV-C testing and maintained marker status.
The authors concluded that "...the observed loss of GBV-C viremia without anti-E2 seroconversion in patients with progressive disease supports an interaction between these two viruses. However, the GBV-C status in HIV-1 infection is probably a secondary phenomenon during disease progression rather than an independent prognostic factor."
Abstract 159lb. Persistent GB virus type C (GBV-C) infection is associated with decreased risk of death in HIV-seroconverters in the Multicenter AIDS Cohort Study, presented by Dr. Williams.
In this investigation, stored sera collected either early (12-18 months after HIV seroconversion) or late (5-6 years after HIV seroconversion) from HIV seroconverters in the Multicenter AIDS Cohort Study (MACs) were tested for GBV-C markers (GBV-C RNA and anti-E2). At the early visit, 39% were GBV-C RNA positive, 46% were anti-E2 positive, and 14% were negative for both markers. GBV-C status at the early visit was NOT associated with survival. In contrast, GBV-C status at the late visit was associated with survival, which was greatest in those who were GBV-C RNA positive. Interestingly, those who cleared GBV-C RNA between early and late visits had the highest risk of death, almost six-fold higher than those who were GBV-C RNA positive at both visits (P<0.01). Likewise, the rate of CD4 decline was highest in those who lost GBV-C RNA, 107 cells/mm3/year versus 26 cells/mm3/year in the persistently RNA positive and 37 cells/mm3/year in those persistently negative. The median duration of follow-up, date of seroconversion, use of HIV treatment, median log HIV RNA levels, and CCR5 delta 32 mutation prevalence did not vary by GBV-C status.
Editorial note. The presentations on GBV-C and HIV contributed to our understanding of the interaction between these two viruses and raised new questions. The most important new finding was the strong positive association between loss of GBV-C RNA without anti-E2 and progression of HIV infection. This observation is important because it could potentially explain some of the protective effect previously attributed to being GBV-C RNA positive, since the comparison group (those who are negative for all markers) includes both those who lost GBV-C RNA without developing anti-E2 as well as those who were never infected with GBV-C (which is probably a small number). It remains unclear to what extent this could fully explain the apparent protective association of being GBV-C RNA positive since only a small fraction in the MACs was negative for all markers. In addition, this observation raises the question of why resolution of one virus in an atypical way (that is, without production of anti-E2 as occurs in HIV negative persons) would increase the risk of disease from another virus. In addition, although the in vitro work by Stapleton's lab continues to demonstrate clear GBV-C effects on HIV infection, the manner in which GBV-C affects HIV in vivo remains unknown.