Occult hepatitis B virus infection and hepatocellular carcinoma: a systematic review.

Occult hepatitis B (OHB) infection has been reported to play an important role in the development of hepatocellular carcinoma (HCC). In this systematic review, a significantly higher prevalence of OHB was observed in patients with HCC in the presence or absence of HCV infection when compared with control populations without HCC. Correspondingly, among adequately designed prospective studies, the cumulative probability of developing HCC was significantly greater among patients with OHB than among HBV DNA-negative patients in the presence or absence of HCV infection. Study design, inclusion criteria, treatment options, methodology and potential confounding variables were evaluated, and immunopathogenic mechanisms that could be involved in OHB as a risk factor in HCC were reviewed. From this analysis, we conclude that although OHB is an independent risk factor in HCC development in anti-HCV-negative patients, a synergistic or additive role in the occurrence of HCC in HCV-coinfected patients is more problematic due to the HCC risk attributable to HCV alone, especially in patients with advanced fibrosis and cirrhosis.

Occult hepatitis B virus infection: a covert operation.

Detection of occult hepatitis B requires assays of the highest sensitivity and specificity with a lower limit of detection of less than 10 IU/mL for hepatitis B virus (HBV) DNA and <0.1 ng/mL for hepatitis B surface antigen (HBsAg). This covert condition is relatively common in patients with chronic hepatitis C virus (HCV) that seems to exert some influence on the replicative capacity and latency of HBV. Detection of virus-specific nucleic acid does not always translate into infectivity, and the occurrence of primer-generated HBV DNA that is of partial genomic length in immunocompetent individuals who have significant levels of hepatitis B surface antibody (anti-HBs) may not be biologically relevant. Acute flares of alanine aminotransferase (ALT) that occur during the early phase of therapy for HCV or ALT levels that remain elevated at the end of therapy in biochemical nonresponders should prompt an assessment for occult hepatitis B. Similarly, the plasma from patients with chronic hepatitis C that is hepatitis B core antibody (anti-HBc) positive (+/-anti-HBs at levels of <100 mIU/mL) should be examined for HBV DNA with the most sensitive assay available. If a liver biopsy is available, immunostaining for hepatitis B surface antigen (HBsAg) and hepatitis B core antigen (HBcAg) should be contemplated and a portion of the sample tested for HBV DNA. This is another reason for optimal collection of a specimen (e.g. two passes with a 16-guage needle under ultrasound guidance). Transmission of HBV to immunosuppressed orthotopic liver transplant recipients by donors with occult hepatitis B (OHB) will continue to occupy the interests of the transplant hepatologist. As patients with OHB may have detectable HBV DNA in serum, peripheral blood mononuclear cells (PBMC) and/or liver that can be reactivated following immunosuppression or intensive cytotoxic chemotherapy, the patient needs to be either monitored or treated depending on the pretreatment serological results such as an isolated anti-HBc reaction or a detectable HBV DNA.

Hepatitis B virus genetic diversity and its impact on diagnostic assays.

Hepatitis B virus (HBV) circulates in blood as closely related, but genetically diverse molecules called quasispecies. During replication, HBV production may approach 10(11) molecules/day, although during peak activity this rate may increase 100-1000 times. Generally, DNA polymerases have excellent fidelity in reading DNA templates because they are associated with an exonuclease which removes incorrectly added nucleotides. However, the HBV-DNA polymerase lacks fidelity and proofreading function partly because exonuclease activity is either absent or deficient. Thus, the HBV genome and especially the envelope gene, is mutated with unusually high frequency. These mutations can affect more than one open reading frame because of overlapping genes. The S gene contains an exposed major hydrophilic region (residues 110-155), which encompasses the 'a' determinant that is important for inducing immunity. Nucleotide substitutions in this region are common and result in reduced binding or failure to detect hepatitis B surface antigen (HBsAg) in diagnostic assays. Adaptive immunity also depends on the recognition of HBsAg by specific antibody and variants pose a threat if they interfere with binding to antibody. Finally, genomic hypervariability allows HBV to escape selection pressures imposed by antiviral therapies, vaccines and the host immune system, and is responsible for creating genotypes, subgenotypes and subtypes.

Impact of GB virus type C infection on mother-to-child HIV transmission in the Women and Infants Transmission Study Cohort.

BACKGROUND: GB virus type C (GBV-C) viraemia is associated with a beneficial outcome in HIV-infected individuals in several though not all studies. GBV-C viraemia was examined in a matched case-control study of 133 HIV-infected pregnant women who transmitted HIV to their infants ('cases') and 266 non-transmitting controls. METHODS: HIV-infected children and controls were pair-matched for high-risk delivery, race and year of delivery. GBV-C status was determined in maternal plasma samples obtained at or within 3 months of delivery. RESULTS: Pregnant women with GBV-C viraemia (11% of those studied) had lower HIV RNA levels (P=0.01) and higher CD4 percentages (P=0.0006) [corrected] than women without GBV-C. A trend towards decreased mother-to-child transmission in the multivariate analysis was observed among GBV-C viraemic women delivering after highly active antiretroviral therapy (HAART) became available [odds ratio (OR) 0.30, 95% confidence interval (CI) 0.08-1.05; P=0.06], but not in women delivering prior to the widespread use of HAART. CONCLUSIONS: GBV-C viraemia was associated with a beneficial effect on CD4 percentage and HIV RNA level in these pregnant women, and was also associated with a trend towards reduced risk of mother-to-child HIV transmission among women after HAART became available. Further studies with larger or multiple cohorts are necessary to assess possible benefits in this population.

Occurrence of identical hypervariable region 1 sequences of hepatitis C virus in transfusion recipients and their respective blood donors: divergence over time.

A total of 240 stored serum specimens from 30 transfusion recipients and 120 blood donors from the Transfusion-Transmitted Viruses Study (TTVS) were evaluated with the objective of establishing transmission of hepatitis C virus (HCV) by specific blood donors. Phylogenetic analysis of hypervariable region 1 (HVR1) and HCV genotyping were performed on the genomic region encoding amino acids 329 to 410. Amino acid distances between HVR1 sequences were calculated by the Kimura formula. Bootstrap analysis of HVR1 sequences provided support for linking recipients to specific donors. Linear regression analysis showed no differences between donor and recipient HVR1 sequences 7.9 weeks posttransfusion, but donor and recipient sequences diverged thereafter (r = 0.690). The initial lag phase in the evolution of HVR1 in the infected recipient was attributed to the time required to mount host immunologic defenses against the virus. Within-recipient divergence in HVR1 was determined from analyses of serial specimens collected within 2 weeks after the alanine transaminase peak, at the end of the original study (1974-1979), and in the follow-up study (1987-present). HVR1 remained invariant over a period of 6.7 to 9.5 days (95% CI) during acute infection. Within-patient divergence in HVR1 increased over a period of 11 to 15 years (r = 0.771), reaching the degree of divergence observed between unlinked subjects. In cases in which transfusion involved more than one HCV subtype, only one of the HCV subtypes established infection in the recipient. Subtype-specific differences in HVR1 were shown.

Effect of coexisting HIV-1 infection on the diagnosis and evaluation of hepatitis C virus.

OBJECTIVES: To evaluate the diagnostic accuracy of the test for antibodies to hepatitis C virus by enzyme-linked immunosorbent assay (anti-HCV ELISA-2) in patients with and without HIV-1 infection. DESIGN: Cohort study. METHODS: In all, 369 patients were tested and grouped by available serologic tests. HCV RNA was quantified in these 369 patients using an Amplicor HCV (and/or HIV-1) Monitor, v1.0 test. Among 110 patients who were anti-HCV negative by ELISA-2, 39 were HIV/HBV coinfected and 71 had HIV alone. One hundred twelve patients were HIV/HCV coinfected and 147 patients had HCV infection alone. RESULTS: Six of 110 (5.5%) ELISA-2 anti-HCV-negative, HIV-infected patients had circulating serum HCV RNA. Their median CD4 count was 36 cells/mm(3), which was significantly lower than that observed in the HIV/HBV group (median CD4 = 109, p <.001) or the HIV/HCV cohort (CD4 = 235; p <.0001). The positive predictive value of the ELISA-2 test for diagnosing ongoing HCV infection in HIV-infected patients was 91%, which is significantly better than that determined for the HCV group, 76% (p =.002) presumably because HCV is less likely to resolve in the HIV patients. Mean alanine aminotransferase (ALT) levels were similar in the HIV/HCV (133 IU/L) and HCV (130 IU/L) cohorts. Median HCV RNA levels were higher in the HIV/HCV group (6.53 log(10) copies/ml) compared with the patients with HCV infection (5.62 log(10) copies/ml; p <.00001). There was no significant correlation between HCV RNA levels and ALT values, CD4 counts, or HIV RNA concentrations. CONCLUSIONS: The predictive value of the anti-HCV ELISA-2 test is better in HIV-coinfected patients than in patients infected only with HCV. False negative results, usually associated with acute infection or with low CD4 counts, are uncommon. These patients may be diagnosed with the ELISA-3 assay or by reverse transcriptase polymerase chain reaction (RT-PCR). Compared with patients with only HCV infection, HIV/HCV patients display similar ALT profiles, but a higher proportion of detectable serum HCV RNA.

The relationship of acute transfusion-associated hepatitis to the development of cirrhosis in the presence of alcohol abuse.

BACKGROUND: Although concomitant alcoholism is widely believed to enhance liver disease progression in persons with hepatitis C virus (HCV) infection, this relationship has not been well quantified. OBJECTIVE: To quantify the relationship of transfusion-associated HCV infection and history of heavy alcohol abuse to development of cirrhosis. DESIGN: Retrospective cohort study. SETTING: Liver clinics in university and government hospitals. PATIENTS: Extended follow-up of 1030 patients in prospective investigations of transfusion-associated viral hepatitis conducted in the United States between 1968 and 1980. MEASUREMENTS: Development of cirrhosis and history of heavy alcohol abuse were determined from review of interviews with patients or their proxies, medical records, death certificates, and autopsy and biopsy reports. Logistic regression was used to estimate the risk for cirrhosis associated with transfusion-associated HCV infection and history of heavy alcohol abuse. RESULTS: The absolute risk for cirrhosis was 17% among patients with transfusion-associated HCV; 3.2% among patients with transfusion-associated non-A, non-B, non-C hepatitis; and 2.8% among controls. Patients with transfusion-associated HCV were more likely than controls to develop cirrhosis (odds ratio, 7.8 [95% CI, 4.0 to 15.1]). A history of heavy alcohol abuse was associated with a fourfold increased risk for cirrhosis. Hepatitis C virus infection plus a history of heavy alcohol abuse led to a substantial increase in risk for cirrhosis (odds ratio, 31.1 [CI, 11.4 to 84.5]) compared with controls without such a history. CONCLUSIONS: Heavy alcohol abuse greatly exacerbates the risk for cirrhosis among patients with HCV infection. This finding emphasizes the need to counsel such patients about their drinking habits.

Long-term mortality and morbidity of transfusion-associated non-A, non-B, and type C hepatitis: A National Heart, Lung, and Blood Institute collaborative study.

Persons with non-A, non-B hepatitis (cases) identified in 5 transfusion studies in the early 1970s have been followed ever since and compared for outcome with matched, transfused, non-hepatitis controls from the same studies. Previously, we reported no difference in all-cause mortality but slightly increased liver-related mortality between these cohorts after 18 years follow-up. We now present mortality and morbidity data after approximately 25 years of follow-up, restricted to the 3 studies with archived original sera. All-cause mortality was 67% among 222 hepatitis C-related cases and 65% among 377 controls (P = NS). Liver-related mortality was 4.1% and 1.3%, respectively (P =.05). Of 129 living persons with previously diagnosed transfusion-associated hepatitis (TAH), 90 (70%) had proven TAH-C, and 39 (30%), non-A-G hepatitis. Follow-up of the 90 TAH-C cases revealed viremia with chronic hepatitis in 38%, viremia without chronic hepatitis in 39%, anti-HCV without viremia in 17%, and no residual HCV markers in 7%. Thirty-five percent of 20 TAH-C patients biopsied for biochemically defined chronic hepatitis displayed cirrhosis, representing 17% of all those originally HCV-infected. Clinically evident liver disease was observed in 86% with cirrhosis but in only 23% with chronic hepatitis alone. Thirty percent of non-A, non-B hepatitis cases were unrelated to hepatitis viruses A,B,C, and G, suggesting another unidentified agent. In conclusion, all-cause mortality approximately 25 years after acute TAH-C is high but is no different between cases and controls. Liver-related mortality attributable to chronic hepatitis C, though low (<3%), is significantly higher among the cases. Among living patients originally HCV-infected, 23% have spontaneously lost HCV RNA.

Long-term culture of human immunodeficiency virus type 1 resulting in loss of glycosylation sites.

Cultures of human immunodeficiency virus type 1 (HIV-1) provided a model for the study of mutations in the absence of host antibodies. Replicate cultures of biological and molecular clones of HIV-1 were passaged weekly for 30 or 34 weeks. Eight regions of HIV-1 genomic RNA were analyzed by means of single-strand conformation polymorphism analysis and nucleotide sequencing. Six mutations were detected in the biological clones. Two were G-->A substitutions. The frequency of mutations was higher in V1 compared to that in other regions (P = 0.01). Three mutations involved loss of potential glycosylation sites in V1. These results show that mutations in the viral genome may result from selection by factors other than host immune pressures.

A comparison of reverse transcription-polymerase chain reaction and branched-chain DNA assays for hepatitis C virus RNA in patients receiving interferon treatment. Consensus Interferon Study Group.

Measurement of hepatitis C virus (HCV) RNA may be beneficial in managing the treatment of patients with chronic HCV infection. In a phase 3 study comparing consensus interferon (IFN) and IFN-alpha2b treatment in patients with chronic HCV infection, serum samples were assayed for HCV RNA using two different assays: a quantitative multicycle reverse transcription-polymerase chain reaction (RT-PCR) method and the Quantiplex branched-chain DNA (bDNA) method. Lower and upper detection limits were 100 copies ml-1 and 5 x 10(6) copies ml-1, respectively, for the RT-PCR method, and 3.5 x 10(5) and 4 x 10(7) genome equivalents ml-1, respectively, for the bDNA method. The two assays were generally concordant over the common range of detectability. The major discrepancy was where PCR still indicated detectable virus in the sample but the bDNA result was negative. Assessment of serum samples during IFN treatment demonstrated that 37% of samples were negative for HCV RNA by bDNA but positive by RT-PCR. Differences were also noted in the quantification of baseline HCV RNA by genotype. These data suggest that HCV patients could be categorized as treatment responders by the bDNA assay when the more sensitive RT-PCR assay indicates lack of complete viral response.

Biochemical and viral response to consensus interferon (CIFN) therapy in chronic hepatitis C patients: effect of baseline viral concentration. Consensus Interferon Study Group.

OBJECTIVE: The effect of baseline viral concentration on response was assessed as part of a multicenter phase 3 trial evaluating the safety and efficacy of CIFN therapy for chronic HCV infection. METHODS: Patients (n = 472) received either CIFN 9 microg or IFN alpha-2b 3 MU subcutaneously t.i.w. for 24 wk, followed by 24 wk of observation. RESULTS: Efficacy was assessed by the percentage of patients who achieved normal ALT values or undetectable HCV RNA values (using RT-PCR with a sensitivity of 100 copies/ml). There was a clear relationship between baseline viral concentration and either ALT or HCV RNA response; patients with lower titer HCV RNA had better response rates. End-of-treatment HCV RNA responses were better for patients with low viral concentrations treated with CIFN (51%) than for patients treated with IFN a-2b (31%) (p = 0.03). ALT responses in patients with low viral concentrations were 60% for CIFN-treated patients and 27% for IFN alpha-2b-treated patients (p < 0.01) at the end of treatment. Patients with high titer HCV RNA were more likely to have a sustained HCV RNA response after treatment with CIFN 9 microg, compared with those treated with IFN alpha-2b (7% vs 0%, p = 0.03). CONCLUSIONS: Both genotype and baseline viral concentration were independent factors that affected response to interferon.

Factors contributing to the evolution and outcome of cirrhosis in hepatitis C.

A number of factors are important in the evolution of chronic hepatitis C to cirrhosis. These include the length of time that the patient has been infected, the age at exposure, dual infection with other viruses, alcohol use, and genotypes, especially in liver transplant recipients. The diagnosis of cirrhosis is made by clinical assessment of the patient, the use of imaging studies, a thoughtful laboratory evaluation, and histologic examination of liver tissue.

Genotype does not affect pattern of HCV RNA decrease among responders during interferon treatment of chronic hepatitis C. Consensus Interferon Study Group.

We assessed differences in the pattern of HCV RNA decrease for HCV genotypes 1, 2, and 3 during interferon treatment to determine if the lower response rates observed among genotype 1 patients were related to a slower decrease in HCV clearance. Serum HCV RNA values of 472 chronic hepatitis C patients treated with either consensus interferon (CIFN) or interferon alfa-2b (IFN alfa-2b) were evaluated. Neither virological sustained responders nor relapsers differed in the pattern of serum HCV RNA decrease based on genotype. Virological sustained responders infected with genotype 1 cleared HCV RNA as rapidly as sustained responders who were infected with genotype 2 or 3. Relapsers had a slower rate of serum HCV RNA decrease than did virological sustained responders. Nonresponders differed in the pattern of serum HCV RNA decrease based on genotype: HCV genotype 3 patients had the greatest decrease in serum HCV RNA; genotype 2 patients had an intermediate decrease; and genotype 1 patients had the least serum HCV RNA decrease. HCV genotype 1 patients treated with CIFN had a greater decrease in serum HCV RNA during therapy than did patients treated with IFN alfa-2b. However, there was no difference in the magnitude of serum HCV RNA decrease between the two interferon treatments for patients infected with genotype 2 or 3. In summary, both genotype and ultimate response to treatment are determinants of the pattern and rate of serum HCV RNA change during interferon therapy of chronic hepatitis C.

Early hepatitis C virus-RNA responses predict interferon treatment outcomes in chronic hepatitis C. The Consensus Interferon Study Group.

In previous studies employing interferons (IFNs) in the treatment of chronic hepatitis C, there have been few reliable predictors of sustained responses. We retrospectively evaluated the predictive value of hepatitis C virus (HCV)-RNA measurements in the first few months during consensus interferon (CIFN) treatment using a sensitive reverse-transcriptase polymerase chain reaction assay to determine sustained responses. Data from two large treatment trials, one of IFN-naive patients and one of retreated relapsers and nonresponders, were used, including serum samples at 2-week intervals in the naive study and 8-week intervals in the retreatment study. Patients received initial CIFN (9 microgram) treatment for 6 months and were assessed 6 months after treatment. There were 28 sustained viral responders of 232 CIFN-treated patients. Of the sustained responders, 48% had already cleared HCV RNA from serum (<100 copies/mL) by week 2, 78% by week 4, 81% by week 6, and 96% by week 12. Patients with early HCV-RNA clearance were more likely to have sustained responses than those who responded later. Early clearance of HCV from serum was also associated with greater likelihood of a sustained response to 48 weeks of retreatment with 15 microgram CIFN. Ninety-five percent of the sustained responders were HCV-RNA-negative by week 8 of retreatment. Early assessment of HCV RNA may help in the prediction of sustained responses to IFN and allow the value of continued treatment to be determined early in the course of IFN therapy.

Selection of appropriate HIV-1 genomic regions for single-strand conformation polymorphism analysis of the diversity, modification, and transmission of HIV-1 quasispecies.

Single-strand conformation polymorphism (SSCP) analysis is a useful tool for studying viral quasispecies. Four regions within the HIV-1 genome were studied by means of SSCP analysis with the aim of determining which regions were the most informative for the study of HIV-1 transmission or for detection of changes in HIV-1 quasispecies populations. Nested polymerase chain reaction (PCR) was used to amplify V1, V2, V3 of the env gene, and the p2 region in the gag gene. In total, 114 plasma specimens from 79 individuals were tested, including serial specimens from 10 mother-infant pairs that were provided by the Women and Infants Transmission Study (WITS). HIV-1 in specimens that were PCR-positive with primer pair SK38/SK39 showed different percentages of positive signals with primer pairs for the four regions: V1, 63%; V2, 83%; V3, 88%, and p2, 100%. HIV-1 sequences in the p2 target region displayed the greatest degree of polymorphism. Analysis of serial specimens showed that the V1 target region was the most variable of the four regions studied and was the most appropriate region for monitoring changes in quasispecies populations. Of the four regions studied, p2 was the most informative for the study of HIV transmission, as shown by analysis of samples from documented cases of mother to infant HIV-1 transmission.

Perinatal transmission of hepatitis C virus from human immunodeficiency virus type 1-infected mothers. Women and Infants Transmission Study.

Antepartum plasma hepatitis C virus (HCV) RNA was quantified in 155 mothers coinfected with HCV and human immunodeficiency virus type 1 (HIV-1), and HCV RNA was serially assessed in their infants. Of 155 singleton infants born to HCV antibody-positive mothers, 13 (8.4%) were HCV infected. The risk of HCV infection was 3.2-fold greater in HIV-1-infected infants compared with HIV-1-uninfected infants (17.1% of 41 vs. 5.4% of 112, P = .04). The median concentration of plasma HCV RNA was higher among the 13 mothers with HCV-infected infants (2.0 x 10(6) copies/mL) than among the 142 mothers with HCV-negative infants (3.5 x 10(5) copies/mL; P < .001), and there were no instances of HCV transmission from 40 mothers with HCV RNA concentrations of < 10(5) copies/mL. Women dually infected with HIV-1 and HCV but with little or no detectable HCV RNA should be reassured that the risk of perinatal transmission of HCV is exceedingly low.

Intra-assay performance characteristics of five assays for quantification of human immunodeficiency virus type 1 RNA in plasma.

Three kits (Roche AMPLICOR human immunodeficiency virus type 1 [HIV-1] Monitor, Chiron enhanced-sensitivity bDNA, and Organon Teknika NASBA HIV-1 QT) and two in-house assays (from National Genetics Institute and Baylor College of Medicine) were compared with a blinded panel. The results were evaluated as to intra-assay sensitivity, precision, and ability to detect differences in a dilution series.

Overview of viral hepatitis.

Viral hepatitis is a general term that is reserved for infections of the liver caused by one of at least six distinct hepatitis viruses, designated hepatitis virus A, B, C, D, E, and G/GB. The human hepatitis viruses are a group of diverse pathogens that share an ability to cause inflammation and necrosis of the liver. The most notable sign of this disease is jaundice, an orange-yellow discoloration of the scleroproteins of the skin and conjunctivae caused by the deposition of bilirubin in the blood resulting from faulty excretion of bile pigment by damaged hepatocytes.

Immunization of seronegative infants with hepatitis A vaccine (HAVRIX; SKB): a comparative study of two dosing schedules.

Hepatitis A virus (HAV) infection is of public health significance among infants and diapered children. Although two licensed HAV vaccines are available, they have not been assessed widely in children under the age of 2 years and are not currently licensed for this age group. The purpose of this study was to evaluate the immunogenicity and reactogenicity of HAV vaccine in seronegative infants. Fifty-three healthy infants were immunized with 360 ELISA Units (EL.U.) of an inactivated HAV vaccine at 2, 4, and 6 (Group 1) or 2, 4, and 15 months of age (Group 2). These injections were not received on the same day that participants received their routine childhood immunizations. HAV serum antibodies were detected using a modified radioimmunoassay procedure and concentrations were calculated using a World Health Organization serum anti-HAV reference standard. No serious-systemic or local reactions were noted among the immunized infants. Three months following the third immunization, seroconversion rates were 100% and 93% in groups 1 and 2, respectively. No significant differences were observed in the geometric mean anti-HAV concentrations between the two groups at comparable time points, i.e. 2 months after the second dose, 3 months after the third dose, and 19 months after the first dose. Three infants, not included in the data presented above, had preexisting maternal antibodies; one never responded to the vaccine and the other two did not respond until maternal antibody levels had become reduced. The results indicate that the inactivated HAV vaccine is highly immunogenic in seronegative infants and could be included in the routine harmonized infant immunization schedule.

Therapy of hepatitis C: consensus interferon trials. Consensus Interferon Study Group.

Chronic hepatitis C is an insidious disease associated with significant morbidity and mortality. Currently, the only approved therapies for chronic hepatitis C are the alpha interferons. Consensus interferon (CIFN) is a nonnatural, synthetic, recombinant type I interferon derived by assigning the most commonly observed amino acid in each position of several alpha interferon nonallelic subtypes to generate a consensus sequence. The efficacy and safety of CIFN in the treatment of chronic hepatitis C were assessed in two large phase 3 trials. The first trial was a multicenter, randomized, double-blind, controlled study of 704 patients who were treated with one of two doses of CIFN (3 microg and 9 microg) or interferon alfa-2b (3 million units [MU]) weekly for 24 weeks and then observed for an additional 24 weeks. Treatment with CIFN at a dose of 9 microg was safe and effective, with serum alanine aminotransferase (ALT) and hepatitis C virus (HCV) RNA sustained response rates of 20% and 12%, respectively. Responses to 3 MU interferon alfa-2b were comparable to 9 microg CIFN. The response rates were lower in the 3-microg CIFN cohort. At the end of the treatment and posttreatment observation periods, an undetectable serum HCV RNA was a better predictor of a normal ALT than the converse. Serum samples from early time points were available for HCV RNA quantitation from 27 of the 28 patients who experienced a sustained response with 9 microg CIFN. Of these, 13 patients (48%) had undetectable HCV RNA at 2 weeks, 21 patients (78%) at 4 weeks, and 26 patients (96%) at 12 weeks. CIFN (9 microg) induced a significantly greater reduction in the mean serum HCV RNA concentration than interferon alfa-2b during treatment (P < .01). In patients with high viral titers (> or = 4.75 x 10(6) copies/mL), the HCV RNA sustained response rate in patients treated with CIFN (9 microg) and interferon alfa-2b was 7% and 0%, respectively (P = .03). In patients infected with HCV genotype 1, the HCV RNA end-of-treatment (24% vs. 15%; P = .04) and sustained (8% vs. 4%; P = not significant) response rates were greater in patients treated with CIFN (9 microg) than with interferon alfa-2b (3 MU). In a subsequent multicenter trial, a higher dose of CIFN (15 microg) was reinstituted in patients who either had relapsed or were nonresponders to prior CIFN or interferon alfa-2b therapy. Patients were randomized to receive 24 or 48 weeks of retreatment followed by 24 weeks of observation. Patients who had relapsed after prior interferon therapy were more likely to have a serum HCV RNA end-of-retreatment and sustained response than patients who were nonresponders to prior interferon therapy. After patients from the 3-microg CIFN cohort were excluded, the HCV RNA sustained response rates were 28% in relapsers and 5% in nonresponders, respectively, in the 24-week retreatment cohort and 58% and 13%, respectively, in the 48-week retreatment cohort. The administration of 9 or 15 microg CIFN was well tolerated, and the adverse effects were similar to those for interferon alfa-2b. These data demonstrate that CIFN at a dose of 9 microg is effective initial therapy for patients with chronic hepatitis C, and that retreatment with a higher CIFN dose of 15 microg for 48 weeks provides meaningful responses in both relapsers and nonresponders.