| |
Hep C Maintenance Therapy Study in Coinfection: ACTG Study
|
| |
| |
Call your local ACTG site if you are interested in learning more about this study.
This is a phase II, open-label, two-arm, long-term maintenance therapy study at ACTG sites. This trial is designed to study whether long-term maintenance with pegylated-interferon alfa-2a (PEG-IFN) reduces the rate of liver fibrosis progression in subjects with HCV/HIV co-infection who have failed to achieve or maintain HCV loads below the limit of detection during treatment with PEG-IFN plus ribavirin. Both HCV treatment naïve and previously treated subjects will be enrolled.
STUDY RATIONALE: PEG-IFN will reduce the rate of liver fibrosis progression in subjects with HCV/HIV co-infection who have failed to achieve or maintain HCV loads below the limit of detection. An untreated observation group will be used for comparison. Poor sustained viral response rates, combined with evidence of accelerated fibrotic progression, strongly support the concept that interventions aimed at slowing rates of viral progression will be important, pending development of newer generations of treatment for HCV. There are limited data that directly address the issue of HCV viral suppression among subjects with HCV alone, and no published data in coinfected subjects. Trials using interferon monotherapy first suggested that long-term antiviral therapy with interferon might slow disease progression.
This study is designed to assess the efficacy of long-term IFN therapy in preventing progression of liver disease. To accomplish this, we will measure HCV-specific immune responses in intrahepatic lymphocytes at the time of study entry (Step 2) and end of treatment, using IFN gamma and interleukin-10 ELISPOT assays. The significance of the proposed immunologic studies is that they will address the immunologic factors that contribute to the progression of liver disease as well as the mechanisms underlying any beneficial effect of IFN in individuals with HCV and HIV coinfection. This information will be useful for defining the optimal patient populations for long-term IFN therapy and for designing new treatment strategies to prevent progression of liver disease.
"SUPPRESSIVE LONG-TERM ANTIVIRAL MANAGEMENT OF HEPATITIS C VIRUS (HCV) AND HIV-1 COINFECTED SUBJECTS (SLAM-C)"
A Multicenter Trial of the Adult AIDS Clinical Trials Group (AACTG)
Sponsored by:
The National Institute of Allergy and Infectious Diseases
Pharmaceutical Support Provided by:
Roche Laboratories, Inc.
SEVERAL KEY SITES WHERE STUDY IS HELD
UCLA Medical Center
UCLA CARE Center
Room BH 412 CHS
10833 Le Conte Avenue
Los Angeles, CA 90095-1793
Phone: 310-206-6414
University of Cincinnati ACTU
Holmes Hospital
Eden Ave & Sabin Way
Cincinnati, OH 45267-0535
Phone: 513-584-8373 (Ms Tammy Powell, RN, Clinical Trials Mgr)
Johns Hopkins University School of Medicine
Viral Hepatitis Center
1830 East Monument Street, Suite 448
Baltimore, MD 21287-0003
Phone: 410-614-6089
GI Unit, GRJ 724
Massachusetts General Hospital
Boston, MA 02114
Phone: 617-726-3766
STUDY OBJECTIVES
Primary Objective
To compare the reduction in the rate of histologic fibrosis progression from Step 2 entry to study termination between subjects treated with PEG-IFN and untreated controls in subjects with HIV-1 and hepatitis C virus (HCV) coinfection who have failed to respond to 12 weeks of PEG-IFN plus ribavirin.
Secondary Objectives
To evaluate the quality of life associated with long-term PEG-IFN in coinfecte subjects.
To determine the impact of long-term PEG-IFN use on HCV viral evolution and host immunologic response.
To investigate the role of noninvasive markers of fibrosis in predicting histologic response, including ALT, AST, bilirubin, albumin, and protein.
To determine the effect of long-term PEG-IFN versus untreated controls on CD4+ T-cell counts and AIDS-defining illnesses.
To determine the effect of long-term PEG-IFN therapy on metabolic parameters, including body weight, lipid profile, hyperglycemia, and insulin resistance.
To assess the impact of treatment on HIV disease control.
To determine the influence of host genetic polymorphisms on cytokine and fibrosis-related genes on baseline histology and on histologic progression of disease.
STUDY SCHEME
SUPPRESSIVE LONG-TERM ANTIVIRAL MANAGEMENT OF HEPATITIS C VIRUS (HCV) AND HIV-1 COINFECTED PATIENTS (SLAM-C)
DESIGN:
This is a phase II, open-label, two-arm, long-term maintenance therapy study. This trial is designed to study whether long-term maintenance with pegylated-interferon alfa-2a (PEG-IFN) reduces the rate of liver fibrosis progression in subjects with HCV/HIV co-infection who have failed to achieve or maintain HCV loads below the limit of detection during treatment with PEG-IFN plus ribavirin. Both HCV treatment naïve and previously treated subjects will be enrolled.
DURATION:
---Subjects on Steps 1 and 2 will be followed for a maximum of 90 weeks.
---Subjects on Steps 1 and 3 will be followed for a maximum of 96 weeks.
SAMPLE SIZE:
Depending on the proportion of naïve or previously treated vs. those rolled over from ongoing therapy, Step 1 may need to enroll up to 180 subjects. A total of 134 subjects will be randomized to Step 2: Arm B (67) and Arm C (67).
POPULATION:
Step 1: Two classes of HIV/HCV coinfected subjects will be admitted to Step 1:
A. Subjects who are HCV treatment naïve or
B. Subjects who failed to achieve HCV sustained virologic response to at least 12 weeks of any interferon-based therapy.
Subjects must be >= 18 years of age, with reasonable likelihood for long-term follow-up and evaluation.
Step 2: HIV/HCV coinfected subjects who have had exposure to and tolerated at least 12 weeks of PEG-IFN plus ribavirin and are currently:
A. In Step 1 or
B. On any other PEG-IFN plus ribavirin regimen.
All subjects must have had at least 12 weeks of active therapy and demonstrate less than a 2-log drop in HCV RNA and have HCV RNA detectable in serum/plasma.
Step 3: HCV/HIV coinfected subjects who achieved at least a 2-log drop in HCV RNA or have undetectable HCV RNA after 12 weeks on Step 1, and who tolerated Step 1 therapy.
STRATIFICATION:
Stratification applies to Step 2 entry. Subjects entering Step 2 will be stratified by pre HCV treatment CD4+ (when available). Strata include (a) <200 cell/mm3, (b) >=200 cells/mm3, (c) no available pretreatment baseline CD4+.
REGIMEN:
Step 1: All subjects will receive:
Arm A consisting of:
• PEG-IFN 180 µg Q weekly for a minimum of 12 weeks (maximum of 18 weeks while awaiting the results required for randomization to Step 2)
• Ribavirin 1-1.2 g/day (weight based) for 12 weeks (maximum of 18 weeks while awaiting the results required for randomization to Step 2).
Step 2: Subjects will be randomized to the following treatment arms:
• Arm B will be treated with PEG-IFN, given 180 µg once weekly for 72 weeks.
• Arm C is the observational control arm.
Step 3: Subjects registered to Step 3 will receive:
• Arm D, a continuation of Step 1 treatment (PEG-IFN 180 µg Q weekly plus ribavirin 1-1.2 g/day (weight based)) for a total of 72 weeks (12-18 weeks on Step 1 plus up to 60 weeks on Step 3). Subjects will then be followed for an additional 24 weeks.
Background
Hepatitis C virus (HCV) infection has emerged as a major etiology for liver disease related morbidity and mortality among HIV-infected subjects in the post-highly active antiretroviral therapy era. An accelerated course of progression to liver failure has been described (1-3). This observation appears to be directly related to an increase in the rate of fibrotic progression in the liver, relative to subjects infected with HCV alone.
Pegylated-Interferon Alfa 2a (PEG-IFN)
Pegylated-interferon alfa 2a (PEG-IFN) is chemically modified IFN in which a branched methoxy polyethylene glycol moiety has been covalently attached to IFN. PEG-IFN has a decreased systemic clearance and an approximately 10-fold increase in serum half-life compared with IFN, and thus circulates in the blood for a much longer time than the parent compound. The biological activity of PEG-IFN, as measured using serum 2'5'-oligoadenylate synthetase (OAS) activity, was similarly prolonged, resulting in a significantly improved pharmacodynamic (PD) response of PEG-IFN compared with IFN. Pharmacokinetic (PK) and PD data obtained from animals and healthy male volunteers suggested that PEG-IFN injected weekly would have the potential for superior efficacy compared with IFN injected 3 times per week.
Subsequent phase II and phase III trials have been performed. Zeuzem et al. (4) treated 531 subjects with either PEG-IFN 180 µg for 48 weeks or the standard IFN with an induction regimen. Sustained viral response was noted in 39% of PEG-IFN subjects and 19% of standard therapy (p<0.001). The regimens result in similar side-effect profiles to each other and to that typically observed with alpha interferons. A high rate of efficacy was also reported by Heathcote et al. (5) in subjects with bridging fibrosis or cirrhosis. A phase III trial has been completed and reported. (6). This multinational trial compared PEG-IFN plus ribavirin to standard IFN plus ribavirin. Subjects treated with the PEG-IFN and ribavirin achieved a 57% sustained viral response rate.
IFN and PEG-IFN side effects include predictable constitutional symptoms including fever, chills, myalgia, anorexia, and nausea. Some subjects may experience weight loss. Local skin reactions at injection sites are common. Subjects with underlying immune mediated disease including thyroiditis, inflammatory bowel disease, and autoimmune hepatitis may develop clinical manifestation or worsening of their clinical condition. Alopecia may be observed. There is a predictable and dose-dependent level of marrow suppression leading to a decrease in both platelet count and neutrophil count. Subjects at risk of depression may have exacerbation of this process.
Ribavirin
Ribavirin is a guanosine analogue that inhibits the in vitro replication of a wide range of RNA and DNA viruses. The mechanism of antiviral activity is not fully defined, although it may involve alteration of cellular nucleotide pools and inhibition of viral mRNA formation. Ribavirin treatment does not inhibit replication of HCV but may result in normalization of serum ALT activity and improve liver histology. However, relapse occurs in nearly all subjects treated with ribavirin. The role of ribavirin as an adjunctive therapy to IFN for treatment of chronic HCV infection was firmly established by McHutchison et al. (7) and confirmed in numerous other studies.
Ribavirin exhibits dose-dependent red cell hemolysis. At standard dosages, more than 85% of subjects will show evidence of hemolysis which can result in severe anemia. This can be controlled with programmed dose reduction and may be mitigated by use of red cell growth factors such as epoetin. Ribavirin is teratogenic and appropriate precautions in women of child-bearing potential must be exercised. Ribavirin may potentiate the toxicity of didanosine (ddI). Reports of symptomatic lactic acidosis in patients receiving ribavirin and ddI have prompted a FDA alert concerning the simultaneous use of these agents.
Rationale
Hypothesis: PEG-IFN will reduce the rate of liver fibrosis progression in subjects with HCV/HIV co-infection who have failed to achieve or maintain HCV loads below the limit of detection. An untreated observation group will be used for comparison.
Using the Metavir scoring schema, coinfected subjects' mean rate of progression was 0.181 units/year, while subjects with HCV alone progressed at a mean rate of 0.135 units/year (p<0.05). Concurrent alcohol use and low CD4+ T-cell count appear to be important factors associated with the alteration in fibrotic potential in coinfected subjects (8). Treatment of hepatitis C is currently centered on accomplishment of viral eradication. Sustained viral response is defined as viral clearance during treatment, with maintenance of viral clearance for 6 months following a course of therapy. Subjects who achieve this level of response appear to halt fibrotic progression, and some may even undergo remodeling of stainable collagen resulting in long-term improvement in liver fibrosis. Unfortunately, this level of treatment response has been demonstrated to occur in less than 50% of subjects with HCV alone, who are treated with the current standard therapy (interferon alfa + ribavirin).
Recent multinational studies have evaluated the use of pegylated interferons, which are long-acting formulations of previously approved agents. These appear to improve treatment response and may yield sustained viral response rates in excess of 60% (9). Among subjects with HCV/HIV co-infection, treatment data are limited. Landau et al. (10) reported that interferon plus ribavirin lead to an end treatment response rate of 50%, but sustained viral response was not reported. Zylberberg et al. (11) described treatment of 21 coinfected subjects with combination therapy, and achieved a 14.3% sustained response rate. These rates of response are far lower than rates observed in subjects with HCV infection alone. Indeed, interim analysis of A5071 has revealed that PEG-IFN plus ribavirin was associated with a 24-week response rate of 44%, appreciably lower than that seen in trials of mono-infected subjects (12). Thus it appears likely that more than 50% of coinfected subjects will fail to clear HCV, though many may have partial or complete suppression of virus during administration of the active drugs.
Therefore, poor sustained viral response rates, combined with evidence of accelerated fibrotic progression, strongly support the concept that interventions aimed at slowing rates of viral progression will be important, pending development of newer generations of treatment for HCV. There are limited data that directly address the issue of HCV viral suppression among subjects with HCV alone, and no published data in coinfected subjects. Trials using interferon monotherapy first suggested that long-term antiviral therapy with interferon might slow disease progression.
Poynard et al. (13) treated HIV-uninfected subjects with interferon alfa-2b for either 18 months or 6 months and reported rates of progression to cirrhosis among those with evidence of portal bridging on pretreatment liver biopsy. The follow-up liver biopsies were performed 2 years after initiation of either treatment arm. Among subjects treated for 6 months, 13% progressed to cirrhosis. The group treated for 18 months had a 6% rate of progression to the same endpoint. Shiffman et al. (14) reported the effect of 30 months of continuous interferon therapy among subjects classified as viral non-responders, compared with subjects who did not continue therapy after a month course of treatment. Subjects maintained on treatment demonstrated improvement in the mean fibrosis scores. The mean fibrosis score of 2.5 was reduced to 1.7 among long-term maintenance subjects. The control group mean fibrosis score went from 2.2 to 2.4, and 30% of subjects had worsening of the overall histologic score. A5071 described week 24 liver biopsies for coinfected subjects who failed to achieve virologic response, and found that 33% of those who underwent biopsy experienced histologic response, demonstrating that histologic response can indeed occur in persons who fail to respond virologically. Thus, the concept of maintenance therapy has merit in coinfection (12). Interferon-based therapy is associated with a number of clinically significant side effects, including marrow suppression, anorexia, persistent fatigue, flu-like syndrome, and increased risk of depression. Patient dropout for PEG-IFN monotherapy over a 1 year course has been reported to be 11% among cirrhotic/near-cirrhotic subjects, which suggests a high degree of tolerability for long-term treatment (15). There are no data that permit determination of dropout rates beyond one year of therapy for PEG-IFN.
Rationale for Metabolic Studies
Hepatitis C virus infection and advanced liver disease (e.g., cirrhosis) have been associated with the development of type 2 diabetes mellitus and insulin resistance. For example, HIV-uninfected persons 40 years of age or older with HCV infection were more than three times more likely than those without HCV infection to have type 2 diabetes (adjusted odds ratio, 3.77 [95% CI, 1.80 to 7.87]) (16). Similarly, among HIV-infected subjects followed in Baltimore, diabetes was significantly more common among those coinfected with hepatitis C (9.1%) compared with those with HIV alone (4.1%) (17). It is unknown whether the stage of liver disease (i.e., hepatic fibrosis) is related to the development of diabetes in HCV-infected persons. However, preliminary studies indicate that IFN therapy may improve glucose tolerance in non-diabetic and diabetic HCV-infected subjects independent of virologic eradication of HCV (18).
Similarly, several studies indicate that HCV coinfection is associated with lower serum cholesterol levels among HIV-infected subjects before and during HAART compared to those HCV uninfected (19-21). Conversely, Minuk and coworkers (21) found that responders to IFN had had significantly higher baseline serum cholesterol levels (5.29 ± 1.04 mmol/L [204 ± 40 mg/dL] compared with 4.15 ± 0.64 mmol/L [160 ± 25 mg/dL]; P = 0.006) and LDL cholesterol levels (3.39 ± 0.93 mmol/L [131± 36 mg/dL] compared with 2.16 ± 0.55 mmol/L [83 ± 21 mg/dL]; P = 0.003) than nonresponders (22). In addition, PEG-IFN monotherapy has been associated with significant increases in serum triglyceride levels among HCV-infected subjects. Furthermore, among HIV-infected subjects IFN therapy has been associated with significant weight loss; in one study, more than 70% of subjects lost > 10% of their baseline body weight.
The effect of long-term IFN therapy on measures of hyperglycemia and insulin resistance, lipid levels, and body weight and fat distribution is unknown.
Rationale for Virologic and Immunologic Studies
The pathogenesis of hepatocellular damage in HCV is poorly understood. HCV itself is not cytopathic, and the classic understanding of the pathogenesis of liver injury in viral hepatitis is that it is mediated by virus-specific cytotoxic T lymphocytes (CTL), which recognize infected cells and lyse them using a variety of effector molecules. This role of CTL is supported by studies in murine models of viral hepatitis (23). For example, adoptive transfer of HBV-specific CTL into mice transgenic for HBV proteins leads to liver injury (23). There have not been comparable studies in HCV. Nevertheless, there is some indirect evidence for the role of viral-specific and nonspecific cell-mediated immune responses in liver injury in HCV. First, transgenic mice expressing HCV proteins do not develop inflammatory liver injury characteristic of chronic HCV infection (24). Second, CD8+ T cells isolated from liver tissue are capable of lysing autologous hepatocytes (25). Third, treatment of an HCV-infected patient with anti-CD8 monoclonal antibodies resulted in improvement of serum transaminases, suggesting that the CD8+ T cells were the cause of liver damage (26). However, the relationship between the presence or absence of CTL responses in the chronic phase of HCV and liver injury is controversial, with some studies suggesting a protective role and others failing to demonstrate this (27, 28).
Thus, the finding that immunosuppressed hosts, who have global defects in T-cell number and function, have a more rapid disease progression is somewhat paradoxical. Despite their putative role in liver injury, there is little literature on CD8+ T-cell function in the setting of immunosuppression. CD8+ T cells are the major cells in the inflammatory hepatic infiltrate in hepatitis post orthotopic liver transplantation (OLT) as well as in immunocompetent hosts (29). It is not known, however, what proportion of these cells are truly HCV-specific versus non-specifically activated cells. There is no published data on the presence or absence of CTL in individuals with HIV and HCV coinfection and the relationship to histology. Thus, clearly more work needs to be performed in this area in order to understand the role of CD8+ T cells in immunocompetent as well as immunocompromised hosts.
In acute HCV, vigorous and multispecific peripheral blood mononuclear cell (PBMC) CD4+ T-cells response is correlated with viral clearance and self-limited disease (30). The role of these responses in chronic infection, however, is less clear; classically these cells are thought to be protective against progression of disease by limiting viral replication. A weaker CD4+ response during the initial phase of infection is associated with more rapid histological progression in the setting of HCV and schistosomiasis coinfection, but these data need to be confirmed in other groups of hosts (31). Certainly, the proliferative response of CD4+ T cells is decreased in immunosuppressed HCV subjects. Multispecific CD4+ T cells responses to at least one HCV antigen are found in the PBMC of 40% of liver transplant HCV patients with minimal to no histological abnormalities, whereas no proliferative response occurs in subjects with severe HCV recurrence, despite similar responses to control antigens in both patient populations (32). Similarly, the proliferative response of PBMC-derived CD4+ T cells in response to HCV proteins, but not to other recall antigens, is lower in HIV-HCV co-infected subjects than in HCV monoinfected individuals (33). Of note, HCV-specific responses appear to be maintained in long-term nonprogressors (34, 35) but are of lower frequency compared with HIV-specific responses in the same individuals (36). However, the presence or absence of a response has not been correlated with underlying liver histology or the likelihood of liver disease progression. One significant limitation of these studies is the fact that they were derived from PBMCs rather than intrahepatic T cells. In one study in liver transplant recipients, HCV-specific CD4+ T-cell responses were more vigorous in the liver compared with PBMC responses, but did not appear to be associated with disease severity (37). There is no comparable published data in individuals with HIV and HCV coinfection.
This study is designed to assess the efficacy of long-term IFN therapy in preventing progression of liver disease. To accomplish this, we will measure HCV-specific immune responses in intrahepatic lymphocytes at the time of study entry (Step 2) and end of treatment, using IFN gamma and interleukin-10 ELISPOT assays. The significance of the proposed immunologic studies is that they will address the immunologic factors that contribute to the progression of liver disease as well as the mechanisms underlying any beneficial effect of IFN in individuals with HCV and HIV coinfection. This information will be useful for defining the optimal patient populations for long-term IFN therapy and for designing new treatment strategies to prevent progression of liver disease.
Subjects who are coinfected with HIV appear to have increased HCV viral load in serum and more progressive liver disease (38). Studies of HCV quasispecies complexity suggest increased heterogeneity in coinfected subjects, possibly secondary to altered immunologic clearance (39). The role of quasispecies variability on the natural history and treatment outcome is not known in this patient population. In kidney and liver transplant recipients, more rapid disease progression is associated with a lower rate of quasispecies diversification, reflecting either the role of the immune response in limiting liver injury, as less immune pressure on the virus would lead to less diversification, or selection of more "pathogenic" strains of HCV (40, 41). Other RNA viruses experience antigenic drift, which is thought to represent an evolutionary mechanism, because of strong selection for antigenic novelty. This process is enhanced by the lack of repair mechanisms for RNA viruses (42). For the hepatitis C virus, viral evolution and antigenic selection may be linked (42).
Longitudinal evaluation of the emergence of variants will permit us to ascertain the relationship between the emergence of unique mutations and the antigenic selection process. Furthermore, we will try to determine whether quasispecies variability is a marker for degree of liver disease progression by comparison of quasispecies populations with liver biopsy findings.
Cloning and sequencing will be applied to serial samples from a group of treated and untreated coinfected subjects to determine the variability between pretreatment wild-type and post-treatment emergence of variants. While sequencing is essential to determine the specific nature of changes present, the overall level of quasispecies variability may be estimated by more rapid, less time-intensive procedures. We have developed a rapid semiquantitative methodology that has been validated by sequence comparison (43). However, in this protocol, we postulate that HCV RNA selection will be driven by long-term IFN use. Specific cloning and sequence analysis will be necessary to document phylogenetic changes in the treatment and control populations. The genetic diversity will be calculated using the Hamming distance, which is defined as the number of nucleotide differences between two sequences.
In view of increased diversity in HCV nonprogressors compared with those HCV-infected individuals with more severe disease (44), as well as data suggesting that successful antiviral therapy or clearance of acute HCV infection is associated with a narrowing or narrow quasispecies diversity in immunocompetent individuals (45,46), it is tempting to speculate that the observed increased diversity in HIV coinfected subjects may reflect an impaired endogenous antiviral response. HCV may be acting to circumvent the interferon pathway by inhibiting the kinase activity of PKR via the E2 and NS5A protein (47,48). This has not been observed in other studies. The 5'UTR represents an internal ribosomal binding site (IRES) for HCV replication. Though highly conserved, mutational variation is present. Since this site represents a key functional domain for replication, even minor modification may alter viral fitness. This concept has not been studied, particularly in the setting of long-term interferon suppressive therapy in nonresponder subjects.
E2/NS1 and 5'UTR interacting sequences will be amplified, cloned, and sequenced to determine at several time points before and during therapy whether evolution of these sequences predicts clinical treatment response.
Hypothesis -- Immunology and Virology studies
Suppression of HCV replication and decrease in fibrosis progression rate will be associated with an increase in HCV-specific immune responses.
The rate of mutation of HCV RNA will be increased by suppressive interferon therapy compared to that in untreated controls.
Rationale for DNA Polymorphisms
Genetic polymorphisms in HLA and various cytokines may account for differences in severity of liver disease and response to therapy. The ability of HLA Class I and II to bind peptide antigens partly determines immune responsiveness to viral pathogens. HLA alleles differ most in their antigen-binding regions and can confer either protection or susceptibility to a disease by specificity for bound antigen. HLA allelic differences have been shown to be important in HCV infection (48,49). Many of the known cytokines are products of T cells, which can be separated into T-helper (TH) subsets depending on the phenotype of cytokine expression (50). A predominant TH1 cytokine profile appears to correlate with hepatic damage and cirrhosis. Cytokines may kill virally infected cells, and interleukin (IL)-2 and TNF HCV viral replication (51-54). In addition, cells other than T lymphocytes can induce liver damage; e.g., Kupffer cells, produce cytokines that stimulate stellate cells to become fibrogenic. Polymorphisms in tumor necrosis factor and complement have also been shown to be associated with severity of disease and response to therapy (55-58). Thus genetic differences in host cytokines, cytokine receptors, and HLA alleles may play as important a role as the viral genotype.
Hypothesis -- DNA Polymorphisms
That selected genetic polymorphisms (e.g., HLA, other Class III molecules, hemochromatosis, cytokine, and cytokine receptors) correlate with fibrosis progression and with virologic, immunologic and histologic response to long-term therapy with interferon therapy.
|
|
| |
| |
|
|
|
