GB Virus C Infection in Patients with Type II Mixed Cryoglobulinemia

Methods

We studied 58 patients with type II mixed cryoglobulinemia that was not associated with hematologic malignant conditions, autoimmune disorders, or infectious diseases other than chronic viral hepatitis. One hundred forty-five volunteer blood donors served as healthy controls. We also studied a second control group that included 20 noncryoglobulinemic blood donors who were age-matched with cryoglobulinemic patients and were positive for antibodies to HCV.

All patients were tested for liver and kidney function according to standard methods. The serum concentration of immunoglobulins, rheumatoid factor activity, and C3 and C4 components were measured by using routine nephelometric assays. Cryoglobulins were measured as the protein concentration of isolated cryoprecipitate and were characterized by using standard immunoelectrophoresis and immunofixation procedures.

Antibodies to HCV were measured with a second-generation enzyme-linked immunosorbent assay (Abbott Laboratories, North Chicago, Illinois). To search for enrichment of antibodies to HCV in cryoglobulins, the isolated and washed cryoprecipitate was resuspended in a volume of phosphate-buffered saline equal to the volume of the original serum. The activity of the antibodies to HCV per mg of IgG was subsequently measured, as reported elsewhere [12], in serial fivefold dilutions of both cryoprecipitate and supernatant.

Serum specimens were tested for HCV RNA with a reverse-transcription polymerase chain reaction (PCR) assay, as described elsewhere [5]. The HCV viral load was measured in both cryoprecipitate and supernatant by using a quantitative reverse-transcription PCR assay (Amplicor HCV Monitor, Roche Diagnostic Systems, Inc., Branchburg, New Jersey), according to the manufacturer's instructions.

The GBV-C genome was detected by using a reverse-transcription PCR method developed in our laboratory [13]. Total RNA was extracted from 100 µL of serum and was reverse transcribed by using random esanucleotides as primers. Complementary DNA was amplified through 45 cycles by using a pair of primers spanning a 161-base pair sequence in the conserved GBV-C helicase region. An aliquot of the PCR product was hybridized to an internal probe with a DNA enzyme immunoassay [14]. The optical densities of this immunoassay were used to measure GBV-C RNA levels in serial tenfold dilutions of cryoprecipitate-supernatant pairs.

Twenty-seven cryoglobulinemic patients received recombinant interferon-alpha2b, and 8 received human lymphoblastoid interferon-α, 4.5 to 6 million U three times weekly for 6 months and then 3 million U three times weekly for 6 to 12 months. For our purposes, virologic response was defined as the disappearance from serum of GBV-C RNA, HCV RNA, or both as a result of interferon-α therapy. A clear alleviation of the signs and symptoms of cryoglobulinemia, along with a reduction of 50% or more in serum cryoglobulin levels, was considered a clinical response.

Statistical analyses were done by using the two-sample Wilcoxon test, the chi-square test, or the Wilcoxon matched-pairs signed-rank test, as appropriate. All P values are two tailed, and statistical significance was set at the 0.05 level.

Results

We detected GBV-C RNA in 23 of the 58 patients with type II mixed cryoglobulinemia (40%); 20 of the 23 were also infected with HCV. We found HCV RNA alone in 28 patients.

We detected GBV-C RNA in 1 of 145 unselected blood donors; thus, the prevalence of this virus was much lower among unselected blood donors than among cryoglobulinemic patients (0.7% compared with 40%; P < 0.001). Similarly, only a small proportion of blood donors was positive for antibodies to HCV compared with cryoglobulinemic patients (2.7% compared with 81%; P < 0.001). In contrast, the prevalence of GBV-C infection was similar in cryoglobulinemic patients positive for antibodies to HCV and age-matched blood donors positive for these antibodies (43% compared with 35%; difference, 8% [95% CI, −18% to 33%]; P > 0.2).

As indicated above, 20 of 58 cryoglobulinemic patients (34%) were co-infected with GBV-C and HCV. No obvious difference was found when the demographic, clinical, and laboratory features of these patients were compared with those of patients infected with HCV alone. However, the patients with GBV-C and HCV co-infection tended to have slightly higher levels of cryoglobulins (P = 0.18) and rheumatoid factor activity (P > 0.2) and lower concentrations of C4 (P = 0.05).

We detected GBV-C RNA alone in the serum specimens of three cryoglobulinemic patients who seemed to be clinically similar in all respects to patients infected with HCV and patients without demonstrable HCV infection.

We measured the concentration of viral markers in samples of cryoprecipitate and supernatant from 10 patients. Results in patients who were co-infected with HCV and GBV-C are shown in the (Table 1). The concentration of antibodies to HCV, as expressed by the antibody activity per mg of IgG, was clearly higher in the cryoprecipitate than in the supernatant (P = 0.022). Similarly, the HCV RNA level was consistently greater in the cryoprecipitates than in the supernatants obtained from the same patients (P = 0.036). In contrast, GBV-C RNA levels seemed to be lower in cryoprecipitate than in supernatant in 4 of the 5 patients studied in this group (P > 0.2). Finally, GBV-C RNA was undetectable in all three cryoprecipitate specimens obtained from patients infected with GBV-C alone.

Twenty-eight of 35 patients receiving interferon-α responded by clearing circulating GBV-C, HCV, or both. The clinical condition did not improve in either of the two patients in whom GBV-C RNA but not HCV RNA disappeared from serum. In contrast, a clinical response was evident in 5 of 6 patients who cleared HCV RNA alone and 6 of 7 patients who cleared both HCV RNA and GBV-C RNA. Of the 18 patients with isolated HCV infection, 13 had both a virologic and a clinical response.

Discussion

Infection with GBV-C or its homologue, hepatitis G virus, has been demonstrated in many patients with such conditions as acute and chronic hepatitis, intravenous drug use, hemophilia, aplastic anemia, and maintenance hemodialysis [9-11, 15-18]. However, the pathogenic role of these viruses has not been established. We show that GBV-C infection, usually associated with HCV, is frequently found in patients with type II mixed cryoglobulinemia. It is interesting that the prevalence of GBV-C infection is equally high in healthy blood donors and cryoglobulinemic patients who are positive for antibodies to HCV. This suggests a common route of transmission for both viruses and may be taken as evidence against a causal role for GBV-C in type II mixed cryoglobulinemia. Moreover, because risk factors for parenteral exposure were uncommon in our patients with cryoglobulinemia, it is likely that most of these patients had community-acquired infection with cotransmitted GBV-C and HCV.

No clinical or biochemical characteristic clearly distinguished the cryoglobulinemic patients with GBV-C infection from those who were not infected. Nonetheless, the tendency of cryoglobulinemic patients with GBV-C and HCV co-infection to have more evident immunologic abnormalities than patients infected by HCV alone suggests a pathogenic role for GBV-C. This possibility is also supported by the fact that GBV-C was the only agent identified in three cryoglobulinemic patients.

Because mixed cryoglobulinemia is commonly considered to be a vasculitic process mediated by cryoprecipitable circulating immunocomplexes, several authors have sought markers of HCV infection in cryoprecipitate; some have found increased concentrations of antibodies to HCV, HCV RNA, or both [5-8]. In our patients with GBV-C and HCV co-infection, we found that concentrations of both antibodies to HCV and HCV RNA were consistently higher in cryoprecipitates than in supernatants, whereas levels of GBV-C RNA were lower in most cryoprecipitates obtained from the same patients. Moreover, GBV-C RNA was not detected in the cryoprecipitates of any of the three patients with isolated GBV-C infection.

Recent therapeutic trials [12, 19] indicate that the effectiveness of interferon-α in HCV-associated mixed cryoglobulinemia is related to the drug's antiviral activity. We were able to study the effects of interferon-α in patients co-infected with GBV-C and HCV. In these patients, we found that response with respect to GBV-C infection, expressed by the disappearance of viral genome from serum, was as frequent as response with respect to HCV infection. However, our most important finding was that clinical response was related to the virologic response with respect to HCV but not GBV-C infection.

Thus, our results suggest that GBV-C does not have a primary role in the pathogenesis of type II mixed cryoglobulinemia in patients co-infected with GBV-C and HCV. Although it seems unlikely, the situation could be different in cryoglobulinemic patients with isolated GBV-C infection. These patients formed a small group in our sample but may be more numerous in countries with lower prevalences of HCV-associated cryoglobulinemia [7, 8, 20]. A large trial studying the effects of interferon-α or other antiviral agents in patients with mixed type II cryoglobulinemia and isolated GBV-C infection may help to establish the clinical significance of GBV-C in cryoglobulinemia.

Drs. Mantero, Mori, and Primi: Consorzio per le Biotecnologie, Spedali Civili di Brescia, Piazzale Spedali Civili 1, 25123 Brescia, Italy.

Drs. Bellavita and Vicari: Servizio di Immunoematologia e Centro Trasfusionale, Ospedali Riuniti di Bergamo, Largo Barozzi 1, 24128 Bergamo, Italy.