UC Berkeley

GB virus C (GBV-C), an RNA virus closely related to hepatitis C virus (HCV), is transmitted through sexual, parenteral, and vertical routes. GBV-C is highly prevalent among patients receiving blood products and those at high risk of sexual or parenteral exposure. Unlike HCV, GBV-C replicates mainly in lymphocytes; many attempts to find an association between GBV-C infection and human disease have been unsuccessful. Therefore, donated blood is not routinely screened for GBV-C infection. In vitro and clinical studies have suggested that GBV-C co-infection may inhibit human immunodeficiency virus (HIV) replication by several different biological mechanisms. Some previous studies, but not all, have shown an association between GBV-C infection and both lower HIV viral load (VL) and better survival among HIV-infected patients. Few studies describe predictors of acute GBV-C infection following transfusion in HIV-infected patients. Reports on survival benefits associated with co-infection after advent of highly active retroviral therapy (HAART) are inconclusive. An open question in many previous reports is the temporal relationship between GBV-C infection and HIV disease markers.

To address some of the currently unanswered questions concerning GBV-C and HIV co-infection, we used a limited access database obtained from the National Heart, Lung, and Blood Institute. The Viral Activation Transfusion Study (VATS) was a randomized controlled trial comparing leukoreduced (LR) vs. non-LR transfusions given to anemic HIV-infected transfusion-naïve patients. Pre- and post-transfusion samples from 489 subjects were tested for GBV-C markers. We used the VATS dataset and the results of GBV-C testing to examine two hypotheses.

First, we tested the hypothesis that GBV-C is transmitted to HIV-infected VATS subjects (n=294) via transfusion. We estimated the risk of acquiring GBV-C RNA per unit of blood transfused and examined the predictors of GBV-C acquisition. We found an incidence of 39 GBV-C infections per 100 person-years during follow-up in this population and an 8% increased risk of acquiring GBV-C associated with each additional unit of blood transfused, controlling for HAART status and baseline HIV VL. A lower HIV VL, use of HAART and white race were associated with an increased risk of subsequent GBV-C acquisition.

Second, we examined the hypothesis that GBV-C co-infection is associated with lower mortality and lower HIV VL in 489 HIV-infected VATS subjects and in two VATS sub-cohorts. GBV-C viremia was associated with significantly lower mortality and HIV VL in unadjusted analyses. We found a non-significant trend towards lower mortality and lower HIV VL among HIV-infected VATS subjects, after adjusting for HIV risk behavior and time-updated E2 antibody, HAART status, HIV VL, and CD4 cell count. Acquisition of GBV-C was associated with lower mortality in the sub-cohort of individuals who were GBV-C RNA and antibody negative at baseline (n=294), adjusting for time-updated covariates (HR= 0.31, 95% CI 0.11, 0.86).

Our results suggest high rates of GBV-C transmission by transfusion among HIV-infected subjects and an increased hazard of GBV-C acquisition with lower pre-transfusion HIV VL and current use of HAART. Our results also indicate that GBV-C viremia is associated with a trend towards lower mortality and lower HIV VL, and GBV-C acquisition via transfusion is associated with a significant reduction in mortality in HIV-infected individuals, after adjusting for HIV disease markers. These findings confirm previous reports that GBV-C infection inhibits HIV replication in vitro and in vivo.