icon star paper   Articles  
Back grey_arrow_rt.gif
 
 
Effect of GB Virus C Coinfection on Response to Antiretroviral Treatment in Human Immunodeficiency Virus Infected Patients
 
  ...our study indicates that patients with concomitant GBV-C and HIV infection exhibit a better response to HAART and confirms the findings of others regarding more-favorable immunologic and virological indicators of HIV disease course among coinfected individuals. Additional studies will be required to elucidate the mechanism of this beneficial effect. ... GBV-C infected patients exhibited a complete virological response to HAART more often than patients with GBV-C and had a greater increase in median CD4 cell count and a marginally greater median HIV RNA level decrease. This association was found to be independent of baseline CD4 cell count and plasma HIV RNA, which indicates that an association exists between GBV-C infection and response to HAART... GBV-C RNA positive patients had a significantly larger median CD4 cell count increase (240 cells/L vs. 150 cells/L and a marginally larger median HIV RNA level decrease (5.12 log10 copies/mL vs. 5.05 log10 copies/mL in response to HAART...
 
GB virus C (GBV-C) is a recently described member of the Flaviviridae family that was initially isolated from individuals with hepatitis that was not attributable to then-recognized hepatitis viruses. Despite reports that the virus is capable of replicating in hepatocytes, several studies have failed to demonstrate an association between GBV-C and clinical hepatitis or any other form of human disease.
 
GBV-C is most closely related to the hepatitis C virus (HCV), and, like HCV, it is most efficiently transmitted parenterally (IV or IM exposure), although sexual and vertical transmission have also been documented. Of healthy blood donors included in one study, 1.1%4% were found to have detectable GBV-C RNA in serum. Not surprisingly, given these viruses' common mechanisms of transmission, the prevalence of GBV-C infection is higher among HCV- or hepatitis B virus infected persons than it is in the general population, ranging from 10% to 35%. Among patients with human immunodeficiency virus (HIV) infection, the prevalence of GBV-C viremia ranges from 14% to 37%, with most series reporting figures >20%. We and others have also shown that, in this group, no association can be demonstrated between GBV-C infection and liver function test abnormalities or other signs of clinical disease.
 
GBV-C viremia has been shown to be associated with a more benign course of HIV infection. The mechanism for this beneficial effect of GBV-C infection is not completely understood, but patients infected with both viruses seem to have higher CD4 T cell counts, lower HIV RNA plasma levels, a slower clinical progression of HIV disease, and improved survival, compared with patients infected only with HIV. Little is known, however, about the effect of GBV-C coinfection on the response to antiretroviral therapy in HIV-infected patients, and most of the available information was obtained when zidovudine monotherapy or 2-drug regimens were considered the standard of care. The present study was designed to test the hypothesis that GBV-C coinfected patients have a more favorable response to highly active antiretroviral therapy (HAART) than do patients with HIV-1 infection alone.
 
The present study sought to investigate the effect of GBV-C coinfection on response to HAART in HIV-infected individuals. Tillmann et al. observed a similarly favorable course of HIV infection in their GBV-Ccoinfected patients before and after the introduction of HAART, as well as an inverse correlation between GBV-C and HIV plasma RNA levels that was independent of antiretroviral treatment, but they did not specifically address the impact of GBV-C infection on the effectiveness of HAART. Our results support the hypothesis that patients with GBV-C infection are more likely to respond to HAART than are patients without GBV-C infection.
 
The median baseline CD4 cell count and plasma HIV-1 RNA level were 30 cells/L and 5.25 log10 copies/mL. None of the patients had received PIs in the past, but 36.3% had received nucleoside reverse-transcriptase inhibitors (NRTIs).
 
The NRTIs most commonly used before the initiation of HAART were, in descending order of frequency, zidovudine, lamivudine, and stavudine, although the differences in frequency of use for each of those medications were small and did not reach statistical significance. Among the HAART regimens used in this group of subjects, the most common PIs were indinavir and saquinavir soft-gel capsules, followed by ritonavir and nelfinavir. There were no significant differences in the proportion of patients receiving each of these antiretroviral agents across categories of GBV-C RNA and E2Ab status. The mean duration of follow-up from the initiation of HAART to evaluation of treatment success was 328.81 days and was comparable among patients with or without GBV-C RNA or antibodies in plasma. The overall prevalence of GBV-C RNA in this population was 21.2%, and the prevalence of anti-E2 antibodies was 22.6%. No patient was positive for both E2Ab and GBV-C RNA.
 
When compared with all patients who did not have GBV-C RNA in plasma, patients positive for plasma GBV-C RNA had significantly higher baseline CD8 cell counts (median, 700 cells/L) vs. 610 cells/L, higher baseline CD4 cell counts (60 cells/L vs. 30 cells/L, and lower plasma HIV RNA levels (5.11 log10 copies/mL vs. 5.29 log10 copies/mL.
 
GBV-C RNA positive patients, compared with GBV-C RNA negative patients, also experienced a significantly larger median CD4 cell count increase (240 cells/L vs. 150 cells/L and a marginally larger median HIV RNA level decrease (5.12 log10 copies/mL vs. 5.05 log10 copies/mL) in response to HAART.
 
On univariate analysis, patients with detectable plasma GBV-C RNA, compared with patients in whom plasma GBV-C RNA was undetectable, were significantly more likely to be HBsAg positive and to achieve virological treatment success in response to HAART (46.4% vs. 28%; OR, 2.47). Age, sex, and frequency of HCV antibody were similar among GBV-C RNA positive and negative patients. In a logistic regression analysis that included all the studied variables, only presence of plasma GBV-C RNA and baseline HIV-1 RNA level remained independently associated with virological treatment success (adjusted OR for GBV-C positivity, 3.04).
 
Although this may seem to be a simple consequence of the lower baseline HIV RNA levels and higher baseline CD4 cell counts observed in the coinfected patients, we found that the magnitude of immunologic restoration by HAART, as reflected by the increase in CD4 cell count, was also significantly larger in the coinfected group. Moreover, in a rigorous multivariate analysis, the presence of GBV-C viremia remained independently associated with virological treatment success after HAART, which suggests that GBV-C coinfection itself exerts a beneficial effect on response to HAART that is not solely attributable to the lower HIV RNA levels and higher CD4 cell counts observed in coinfected patients.
 
The fact that HIV RNA levels were also independently associated with the outcome of HAART may reflect the previously demonstrated effect of GBV-C on HIV-1 replication and is consistent with a proposed interference between the 2 viruses. Although the mechanisms of this interaction remain unresolved, the data are consistent with a model in which GBV coinfection attenuates the magnitude of HIV replication either directly or indirectly through viral interference, by selecting for more-competitive but slower-growing HIV strains or, less likely, by inducing or preserving defenses that regulate HIV propagation.
 
Our study also confirms the findings of others and of one of our previous reports, in that the GBV-Ccoinfected patients in the present study had higher CD4 and CD8 cell counts and lower plasma HIV RNA levels at baseline. When baseline characteristics were compared according to both E2Ab and GBV-C RNA status, there seemed to be a gradation in the magnitude of the effect on baseline CD8 cell counts, which were higher in E2Ab-, GBV-C RNA negative patients than in E2Ab-positive, GBV-C RNA negative patients and higher in that group than in GBV-C RNA positive, E2Ab-negative patients. Because E2Ab is a marker of clearance of previously acquired GBV-C infection, these results suggest that, although even transient infection with GBV-C is associated with a measurable improvement in the clinical course of HIV infection, sustained infection, as manifested by the presence of GBV-C RNA in plasma, provides the greatest protective effect. This increasing effect with increasing levels of exposure to GBV-C is also consistent with a direct effect of GBV-C on HIV-1 replication. Additional evidence for such a direct effect has been reported by Xiang et al., who found decreased HIV replication in cultured peripheral blood mononuclear cells that were coinfected with GBV-C. Similarly, Tillmann et al. found an inverse correlation between GBV-C and plasma HIV RNA levels but not between GBV-C RNA levels and CD4 cell counts, which again suggests that an interaction between the viruses, rather than an immunologically mediated mechanism, is responsible for the observed beneficial effect of coinfection on HIV disease course. They also found that, in patients who start HAART, plasma GBV-C RNA levels increase as HIV RNA levels decrease, which indicates that the interaction may be bidirectional.
 
The Journal of Infectious Diseases 2003;187:504-507 Benigno Rodriguez,1 Ian Woolley,1,a Michael M. Lederman,1 Dietmar Zdunek,2 Georg Hess,2 and Hernan Valdez1 1University Hospitals of Cleveland and Case Western Reserve University Center for AIDS Research, Cleveland, Ohio; 2Roche Diagnostics, Penzberg, Germany
 
 
 
 
  icon paper stack View Older Articles   Back to Top   www.natap.org