GB virus C infection: Clinical significance

In 1995 and 1996, respectively, Simons et al (1) reported GB virus C (GBV-C) and Linnen et al (2) reported hepatitis G virus as new hepatitis viruses capable of causing non-A through -E hepatitis. These two viruses were concluded to be identical because they belong to the species Flaviviridae and show 86% homology at the nucleotide level and 95% at the amino acid level (3). This virus infects humans, inducing acute hepatic injury and sustaining infection. However, its clinical significance remains largely unclear. To address this uncertainty, we examined rates of GBV-C RNA positivity in patients with fulminant hepatitis, acute hepatitis, chronic hepatitis, liver cirrhosis and hepatocellular carcinoma.


PATIENTS AND METHODS Patients:
A total of 231 patients (151 males, 80 females; mean age 51.8±13.9 years) with viral hepatitis were selected randomly. They were hospitalized in the authors' department between January 1987 and December 1996: 12 with fulminant hepatitis, eight with acute hepatitis, 127 with chronic hepatitis, 40 with liver cirrhosis and 44 with hepatocellular carcinoma. All patients with chronic hepatitis and 28 with cirrhosis were diagnosed histologically. Patients with fulminant hepatitis and acute hepatitis, and 12 patients with liver cirrhosis were diagnosed by blood chemistry tests and hepatic imaging (ultrasound, computed tomography or magnetic resonance imaging). Patients with hepatocellular carcinoma were diagnosed histologically or by hepatic imaging including angiography.
For interferon (IFN) therapy, IFN-a was used in six cases, and a combination of IFN-b and IFN-a was used in five others. In all cases, therapy continued for six months. The total IFN dose was 702 to 780 MU. The trial of IFN therapy conformed to the ethical guidelines of the Declaration of Helsinki; treated patients gave informed consent.
The effect of IFN therapy was evaluated virologically as follows. Patients in whom GBV-C RNA or hepatitis C (HCV) RNA was negative six months after cessation of IFN therapy were defined as having a sustained response; others were considered nonresponders. In a similar fashion, patients in whom alanine aminotransferase (ALT) levels were normal for more than six months after cessation of IFN therapy were defined as having a sustained response; all others were categorized as nonresponders. Detection of GBV-C RNA: Serum RNA was extracted from 100 µL serum using the Sepa Gene RV-R kit (Sanko, Tokyo, Japan). cDNA was synthesized from the RNA sample at 37°C for 1 h using Moloney murine leukemia virus reverse transcriptase (GIBCO BRL, Gaithersberg, Maryland) and a random primer. GBV-C RNA was detected by reverse transcription-nested polymerase chain reaction (PCR) with specific primers derived from the 5¢ untranslated subgenomic region (4,5). First round PCR was performed with the sense primer 5gf2: 5¢-GGTTGGTAGGTCGTAAATCCCGGT CA-3¢ and the antisense primer 5gr6: 5¢-GACATTGAAG GGCGTRGACCGTAC-3¢, for 35 cycles consisting of denaturation for 1 min at 94°C, annealing for 45 s at 55°C and extension for 1 min at 72°C. Second round PCR was performed with the sense primer 5gf3: 5¢-TGGTAGCCACTA TAGGTGGGT-3¢, and the antisense primer 5gr4: 5¢GCGA CGTGGACCGTACRTGGG CGT-3¢, for 35 cycles under the same conditions as for first round PCR. Amplicons were analyzed by electrophoresis on 3% agarose gels stained with ethidium bromide. Nucleotide sequences in PCR-positive subjects were determined by the dideoxy chain termination method using the ABI PRISM Dye Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems Division, Foster City, California).
In each case, serum was collected aseptically, immediately frozen at -80°C, stored and thawed before examination.

RESULTS
The incidence of GBV-C RNA in patients with acute hepatic disease is shown in Table 1. Among patients with fulminant hepatitis, GBV-C RNA was not found in the serum initially in any case, but the serum became positive after plasma exchange in one patient. Among patients with acute hepatitis, one of two positive for hepatitis B surface antigen (HbsAg) and one of four with non-A through -C hepatitis were positive for GBV-C RNA.
Frequency of GBV-C RNA among patients with chronic hepatic disease was 18% for HBsAg-positive chronic hepatitis patients, 20% for cirrhosis patients and 16.7% for hepatitis B patients. Among HCV RNA-positive cases, prevalence of GBV-C RNA was 9.8% for chronic hepatitis, 4% for cirrhosis, 11% for hepatocellular carcinoma and 9.3% for hepatitis C. Among non-B, non-C hepatitis patients, positive cases accounted for 6.3% (Table 2).
Positivity for GBV-C RNA in blood and clinical features of patients with chronic hepatitis C are shown in Table 3. With respect to sex, mean age, history of transfusion, ALT, histology and therapeutic effect of IFN against HCV, no significant difference was observed among the 11 patients positive for GBV-C RNA and the 101 negative patients.
Eleven patients with chronic hepatitis showing combined infection with HCV and GBV-C were followed for longer than one year after IFN treatment. GBV-C RNA, HCV RNA and ALT were determined serially over time to assess the effect of IFN therapy. In three patients in whom GBV-C RNA showed a sustained response, ALT demonstrated a sustained response in one and no response in two. In eight patients in whom GBV-C RNA did not respond, ALT demonstrated a sustained response in two and no response in six. Meanwhile, in three patients in whom HCV RNA showed a sustained response, ALT demonstrated a sustained

DISCUSSION
In patients with chronic hepatic diseases, GBV-C frequently has been found to coinfect with HCV. Linnen et al (2) reported the incidence of coinfection with GBV-C in chronic hepatitis C to be 18.8%, Dawson et al (6) about 20% and Schleicher et al (7) 19%. In Japan, coinfection has been re-ported to range from 3% to 10% (4,8,9). In the present study, the incidence of coinfection was 9.8%.
In 11 patients who were GBV-C RNA-positive and 101 who were GBV-C RNA-negative, among those infected with chronic hepatitis C, no difference was observed in clinical findings, ALT or liver histology. Moreover, changes in ALT after IFN therapy frequently coincided with the changes in HCV RNA positivity but correlated less with positivity for GBV-C RNA. Therefore, we believe that HCV played the important role in hepatocellular injury in cases with coinfection. Antigenomic GBV-C RNA has recently been detected in the liver, revealing that the virus had proliferated there. However, in that report, no clinicopathological difference was found between GBV-C RNA-positive and -negative cases of chronic hepatitis (10). Thus, involvement of the virus in chronic hepatic diseases was felt to be minimal, as we also concluded.
All hepatocellular carcinomas associated with serum GBV-C RNA positivity were also associated with positivity for HBV-or HCV-related markers. In hepatocellular carcinoma associated with chronic non-B, non-C hepatitis, we found no patient to be positive for GBV-C RNA; Linnen et al (2) reported similar findings. Recently, an immunoassay for antibodies against a protein molecule encoded by the envelope 2 segment of the GBV-C genome was developed. Detection of these antibodies seems to exclude the viral genome from serum (11). Our study detects the presence of GBV-C; we did not perform the immunoassay for antibodies.
Existing reports on hepatocellular injury by GBV-C can be divided into two groups -one suggesting that the virus induces serious hepatic disorders and the other group concluding that resulting hepatic injury is mild. The former view was first reported by Yoshiba et al (12) with respect to fulminant hepatitis of unknown etiology, three of six such patients were GBV-C RNA-positive. However, Kuroki et al (13) suggested that GBV-C could not be considered the cause of fulminant hepatitis because, of seven patients with fulminant hepatitis studied, no patient's serum was GBV-C RNA-positive before plasma exchange. In the present correlative survey, GBV-C RNA was negative before plasma exchange in all eight patients with fulminant non-A through -C hepatitis. Only one patient became positive after plasma exchange. These results suggest that infection with GBV-C and onset of fulminant non-A through -C hepatitis show little correlation.

Figure1) Efficacy of interferon (IFN) therapy against GB virus C (GBV-C) in chronic hepatitis patients with GBV-C and hepatitis C virus (HCV) coinfection. Panel A Serial determinations of alanine aminotransferase (ALT) in three patients with virologic sustained responses according to GBV-C RNA. Panel B ALT in eight patients who were virologically nonresponsive according to GBV-C RNA. Correlation is relatively poor between GBV-C virologic response and the changes in ALT after cessation of IFN therapy. Panel C Serial determinations of ALT in three patients showing a sustained virologic response with respect to HCV. Panel D Serial profile of ALT concentration in eight patients with nonresponding HCV RNA positivity. A good correlation is seen between HCV virologic response and changes in ALT after cessation of IFN therapy
Infection with GBV-C was found in one patient with acute non-A through -C hepatitis. Hepatitis D virus coinfects with hepatitis B virus, and hepatitis E virus is not present in Japan. These findings indicate that the above patient was considered to have acute non-A through -E hepatitis. The patient complained of severe symptoms and demonstrated marked jaundice. This patient seemed to be infected with GBV-C alone, but viral RNA was still detected after clinical hepatitis had resolved. Alter et al (14) studied 10 patients with acute hepatitis in whom GBV-C RNA was positive, reporting that total bilirubin concentrations showed no elevation and that any changes associated with GBV-C were mild, unlike the apparent situation of the patient we report here.
GBV-C is classified into three types by molecular analysis: one found predominantly in Africa, another dominant in America and the third most prevalent in Asia. The strain infecting the patient reported here was identified as the Asian type (4). Heringlake et al (15) reported characteristic base-substitution mutations commonly seen in the NS 3 region in patients with fulminant hepatitis and GBV-C RNA positivity. Accordingly, further study of the relationship between acute hepatic disorders and genetic variation among viruses is necessary.