INTRODUCTION — Two viral agents related to hepatitis have been isolated and designated hepatitis G virus (HGV) and GB virus type C (GBV-C). Molecular characterization of these flaviviruses has shown them to be virtually identical isolates of the same virus [1,2]. In comparison, HGV has only 29 percent amino acid homology with HCV [3].
HGV was initially cloned from a surgeon (whose initials were GB, hence the term GB virus or GBV). Plasma from this patient transmitted hepatitis to tamarins. Three different GBV agents were isolated and designed GB virus types A, B, and C; the first two were probably tamarin agents [3]. The agent designated GBV-C is virtually identical to HGV and can be spread by transfusion of contaminated blood and sexual contact, similar to hepatitis C virus and hepatitis B virus [2,4].
DISTRIBUTION — GBV-C has a global distribution, with a high prevalence in the United States donor population [5]. GBV-C can be diagnosed only by detecting its RNA in the serum by polymerase chain reaction. However, the sensitivity and specificity of this test are not known [3], and one study suggests that it may significantly underestimate the true incidence of infection [6]. In this report, serial serum samples were obtained from 116 injection drug users over a 6.5 year period; a marker of GBV-C infection (either RNA or antibody) was detected in 110, 70 of whom had anti-GBV-C antibodies but were never positive for GBV-C RNA. All eight patients who cleared GBV-C RNA were antibody positive and there were no new infections in 61 antibody-positive subjects. Thus, anti-GBV-C antibodies appear to be highly associated with viral clearance and protection from reinfection.
CLINICAL SIGNIFICANCE OF GBV-C INFECTION — The clinical significance of GBV-C infection with respect to acute or chronic hepatitis is not well understood, but the preponderance of other evidence suggests that GBV-C does not cause hepatitis in humans [7,8].
GBV-C RNA has been detected in patients with acute non-A to non-E viral hepatitis, in patients with chronic hepatitis of presumed viral etiology, in patients with cryptogenic cirrhosis, and in some patients with primary hepatocellular carcinoma. However, it is often difficult to tease out the direct role of GBV-C in these settings since coinfection with HCV is so common. Studies in patients with apparently isolated GBV-C infection suggest that the acute liver injury is similar to and may be less severe than that with HCV [3]. On the other hand, in patients with post-transfusion hepatitis in whom the presence of GBV-C can be documented, the plasma ALT peak and the peak viral titers may be discordant, suggesting that GBV-C may not be responsible for the elevation in ALT levels.
The following observations illustrate the range of findings in some of the largest studies:
●75 percent of individuals positive for GBV-C RNA have normal liver function tests [2].
●One report evaluated 189 patients with chronic HCV infection; 21 (11 percent) were positive for GBV-C RNA [9]. The course of the HCV infection and the response to interferon-alfa were not affected by coinfection with GBV-C. Interferon-alfa led to a decrease in GBV-C RNA titers that was not sustained after the cessation of therapy. Similar findings were noted in another study in which liver biopsies were performed [10]. Detailed histopathologic examination revealed no difference between the patients infected with HCV alone and the 15 percent who were coinfected with GBV-C.
●A study from Scotland investigated the prevalence of persistent infection with GBV-C or HCV in patients exposed to non-virus inactivated blood products [11]. Among non-remunerated blood donors, the incidence of GBV-C infection (by polymerase chain reaction) was much higher than HCV infection (3.2 versus 0.08 percent). However, among 95 hemophiliacs who had received multiple units of non-virus-inactivated concentrates, the incidence of GBV-C RNA was much lower than that of HCV RNA (14 versus 83 percent). Virus inactivation substantially reduced or eliminated contamination by GBV-C RNA in blood products.
●GBV-C infection is often present (25 percent in one report) in patients who undergo liver transplantation for HCV or nonviral chronic liver disease [12,13]. The patients with GBV-C infection have a clinical course similar to those who are GBV-C negative.
●The median liver/serum ratio of GBV-C RNA was less than one in one report, a finding in contrast to the known hepatotropic HCV (which had a high ratio of HCV RNA in the liver compared to the serum) [14]. This observation is consistent with serum contamination of liver tissue rather than the liver being the main site of GBV-C replication.
In contrast to these findings, an early report from Japan documented GBV-C RNA in three of six patients with acute fulminant hepatitis of uncertain etiology [15]. However, a later study, also from Japan, provided an explanation for how this might have occurred [16]. Nine patients with non-A-E fulminant hepatitis and one with late onset hepatic failure were all GBV-C-negative at presentation. Because of legal difficulties with organ donation in Japan, all were treated with plasmapheresis and plasma infusion; four became seropositive for GBV-C after transfusion. Other reports have confirmed that the presence of GBV-C RNA in patients with fulminant hepatitis of unknown etiology was related to the administration of blood products after the onset of disease [17].
In summary, the exact role, if any, of the GBV-C virus in producing disease in humans remains unclear [7]. There are no prospective studies that have documented histologically the progression from acute GBV-C infection through various stages of chronic liver disease, such as chronic hepatitis, cirrhosis, and primary hepatocellular carcinoma. In patients with chronic non-A to non-E hepatitis or cryptogenic cirrhosis, there is no proof that, if present, GBV-C is the cause of the disease, rather than being an innocent bystander.
HIV infection — Accumulating evidence suggests a protective effect of GBV-C on patients coinfected with HIV [18-22]. The protective effect of GBV-C may be related to maintenance of an intact T-helper 1 cytokine profile, the induction of HIV-inhibitory cytokines, and interference with HIV replication [23,24]. While the observations in these reports support further investigation into the mechanisms of interaction between GBV-C and HIV, the therapeutic implications remain unclear. The following summarizes results of the largest reports.
●A study of 158 HCV/HIV coinfected patients found 57 (26 percent) who had GBV-C RNA and 59 percent had evidence for past exposure based upon results of antibody testing [22]. Active GBV-C RNA infection was significantly associated with a reduced risk of cirrhosis and death from cirrhosis.
●GBV-C viremia was assessed 12 to 18 months after HIV seroconversion in 271 men participating in a longitudinal study [25]. A subgroup of 138 patients was evaluated again five to six years later. Since cohort enrollment occurred prior to 1990, fewer than six percent of subjects were receiving antiviral therapy and none were receiving combination antiviral therapy.
GBV-C infection was detected in 85 percent of patients. GBV-C status was not significantly associated with survival at 12 to 18 months following HIV seroconversion. In contrast, patients with persistent GBV-C viremia five to six years after HIV seroconversion were less likely to have died compared with those without GBV-C RNA. The poorest outcomes occurred among the nine percent of patients who lost GBV-C RNA between the two time points (relative hazard of death of 5.87 compared to those with persistent GBV-C RNA).
●An earlier study included 362 HIV-infected patients of whom 144 (40 percent) had GBV-C viremia [20]. Significantly fewer coinfected patients died during a mean follow-up of four years (56 versus 29 percent). GBV-C infection remained an independent predictor of survival even after adjusting for the effects of HIV treatment, baseline CD4 cell count, age, sex, race, and mode of transmission of HIV (relative risk 3.7, 95% CI 2.5 to 5.4).
In contrast to the possibility of a protective effect of GBV-C coinfection in the HIV-infected individual, GBV-C coinfection did not appear to prevent vertical transmission of HIV in a retrospective cohort study from Tanzania [26]. HIV transmission to their infants occurred among 28 percent of HGBV-C RNA positive women, 23 percent of those with antibodies to HGBV-C, and 26 percent of those without evidence of current or past HGBV-C infection.
There is little information on the natural history of HIV and GBV-C coinfection in patients who are receiving HAART, but the available data suggest that the protective effect of GBV-C coinfection is diminished. One report suggested that HIV viral load did not differ between HIV mono-infected patients and those coinfected with HCV and GBV-C who were followed for one year after initiating HAART [27]. Another report found that coinfection with GBV-C in HIV/HCV coinfected patients (who were receiving HAART) had no impact on the rates of sustained virologic response [28]. A third report found no protective effect from GBV-C coinfection on CD4 counts or HIV viral loads in HIV-infected women who were receiving HAART [29].
IMPLICATIONS OF GBV-C FOR THE BLOOD SUPPLY — A contentious issue is whether donor blood should be screened for GBV-C and other viruses that have been discovered in human blood but are not known to cause disease. Potential arguments in favor of such a policy include:
●The relatively high prevalence of GBV-C in the donor population.
●The transmissibility of GBV-C by transfusion.
●The possible causal role of GBV-C in producing acute hepatitis.
●The ability of GBV-C viremia to persist for several years.
●The possible role of GBV-C in some patients with chronic hepatitis or chronic liver disease of unexplained etiology.
Arguments against donor screening include:
●The potential for high donor loss.
●The apparently very low incidence of posttransfusion hepatitis [11].
●The apparently mild disease when acute hepatitis does occur.
●The lack of proof that chronic liver disease occurs with GBV-C infection.
●The apparently very low disease burden due to GBV-C.
●The difficulty in counseling healthy blood donors.
At present, routine screening for GBV-C is not performed in blood donors.
Unanswered questions — In addition to the above uncertainties, there are a number of other questions that are still unanswered:
●Does the virus replicate in human liver?
●Can GBV-C be transmitted to chimpanzees?
●What happens during the course of GBV-C in infection in humans in terms of changes in viral sequences and the immune response?
●Does coinfection of hepatitis C virus or hepatitis B virus alter the natural history of those disorders? As noted above, one report suggests coinfection has no effect on HCV infection [9].
SUMMARY AND RECOMMENDATIONS
●GBV-C has a global distribution, with a high prevalence in the United States donor population. Currently, GBV-C can be diagnosed only by detecting its RNA in the serum by polymerase chain reaction. However, the sensitivity and specificity of this test are not known. (See 'Introduction' above.)
●The significance of GBV-C infection with respect to acute or chronic hepatitis is not well understood, but the preponderance of other evidence suggests that GBV-C does not cause hepatitis in humans. (See 'Clinical significance of GBV-C infection' above.)
●There appears to be a protective effect of GBV-C on patients coinfected with HIV who are not receiving antiretroviral therapy, possibly related to inhibition of HIV replication. (See 'HIV infection' above.)
