Update on chronic hepatitis C in HIV/HCV-coinfected patients: viral interactions and therapy
AIDS 2003; 17(16):2279-2290
Norbert Bräu Department of Medicine, Division of Infectious Diseases, Mount Sinai School of Medicine, New York, NY, and the Veterans Affairs Medical Center, Bronx, NY, USA.
- Significance of HCV in the HIV-infected population
- Prevalence of HIV/HCV-coinfection
- Interactions between HIV and HCV infection Influence of HIV infection on the course of chronic hepatitis Influence of HCV infection on the course of HIV disease
- Treatment of chronic hepatitis C in HIV/HCV-coinfected patients
- Lactic acidosis
- Special considerations in HCV therapy for HIV-infected persons
- Which HIV/HCV-coinfected patients should be treated for chronic hepatitis C and how?
- Management of HIV/HCV-coinfected patients with end-stage liver disease Treatment for decompensated liver disease Liver transplantation
- Future research questions
With highly active antiretroviral therapy (HAART), HIV-infected patients can now live longer and healthier lives, and other comorbid diseases, such as chronic hepatitis C, have emerged as a significant health concern. Coinfection with the hepatitis C virus (HCV) may limit life expectancy because it can lead to serious liver disease including decompensated liver cirrhosis and hepatocellular carcinoma. HCV-induced fibrosis progresses faster in HIV/HCV-coinfected persons, although HAART may be able to decrease this disease acceleration.
Combination therapy for HCV with interferon and ribavirin can achieve a sustained viral response, although at a lower rate than in HCV-monoinfected patients. Combination treatment with pegylated interferon and ribavirin will probably emerge as the next HCV therapy of choice for HIV/HCV-coinfected patients. HCV combination therapy is generally safe, but serious adverse reactions, like lactic acidosis, may occur. Cytopenia may present a problem leading to dose reductions, but the role of growth factors is under study.
All HIV/HCV-coinfected patients should be evaluated for therapy against the hepatitis C virus. A sustained viral load will probably lead to regression of liver disease, and even interferon-based treatment without viral clearance may slow down progression of liver disease.
HIV/HCV-coinfected patients who have progressed to end-stage liver disease have few therapeutic options other than palliative care, since liver transplants are generally unavailable. The mortality post-transplant may be higher than in HCV-monoinfected patients. We are entering an era where safe and effective HCV therapy is being defined for HIV/HCV-coinfected patients, and all eligible patients should be offered treatment.
With the introduction of protease inhibitors and highly active antiretroviral treatment (HAART) in 1996, HIV-infected patients have experienced dramatically prolonged survival and decreased morbidity. With longer survival, patients with HIV/AIDS now increasingly get ill from comorbidities not related to HIV, chief among them coinfection with the hepatitis C virus (HCV).
Significance of HCV in the HIV-infected population
In a recent 5-year study of a Canadian HIV-infected cohort that covered the HAART era (1996-2000), Klein and coworkers found increased mortality in HIV/HCV-coinfected individuals compared to HIV patients who were HCV negative [risk ratio (RR) 11.7; P = 0.05]. Mortality per 100 patient-years was 6.7 in HCV+ patients compared to 2.3 in HCV- patients. In addition, HIV/HCV-coinfected patients had a higher rate of hospitalization of 15.0 versus 6.8 per 100 patient-years (HCV+ versus HCV-). While overall mortality of HIV-infected persons has declined dramatically since the introduction of HAART, the relative contribution of liver disease to death is rising. A large French cohort study compared 17 487 HIV-infected outpatients during 1995 with 26,500 patients during 1997, and found that the overall mortality declined from 2.0% in 1995 to 0.9% in 1997, but relative mortality from cirrhosis or hepatocellular carcinoma increased from 6.5% to 10%. In another French study of hospitalized HIV-infected patients during the first 6 months of 2000, 535 of 62 000 (0.9%) died. The cause of death was related to complications from HIV disease in 51% of deaths, and the most frequent non-AIDS related causes of death were HCV infection (10%) and non-AIDS defining cancer (9%). In two small studies from Boston (n = 22) and Madrid (n = 20), relative mortality from liver disease has reached 50.
Prevalence of HIV/HCV-coinfection
The prevalence of HCV infection in HIV-infected individuals varies widely and depends largely on the mode of transmission of HIV itself. In a large French cohort of 1935 HIV+ patients, the prevalence of HCV seropositivity was high in injecting drug users (IDU; 91%) and in recipients of blood or blood products (71%), but low in sexually transmitted HIV (7.3%). In smaller HIV+ cohorts of 150-300 patients in Germany, France, Switzerland, and Greece, a similar pattern of HCV coinfection rates according to HIV risk factor was found. This confirms that in HIV/HCV-coinfected patients, as in HCV-monoinfected persons, parenteral transmission of HCV is very common (especially in IDU), but sexual transmission is rare. In two large US HIV trials (N = 1687), the overall rate of HIV/HCV-coinfection was 16% (ACTG study). There are fewer data available on the prevalence of HIV infection among patients with hepatitis C. (edit note: others estimate co-infection rates of 25-30% in the USA). Among US veterans in the New York City metropolitan area who tested HCV+ during a 1-day cross-sectional survey, the rate of HIV coinfection was 25%, with a higher rate in New York City (29%) than in the suburbs (8.7%). One study showed a higher rate of HIV coinfection of 52%, but in that survey there was a bias toward a higher HIV coinfection rate since many patients were referred from Infectious Diseases specialists.
Interactions between HIV and HCV infection
Influence of HIV infection on the course of chronic hepatitis C
Levels of HCV RNA are significantly higher in HIV/HCV-coinfected patients than in HCV infection alone, both in plasma and in liver tissue. This may be due to the fact that in HIV-infected patients, HCV replicates not only in hepatocytes but also in the lymphoid cells. Laskus and colleagues demonstrated replication of HCV in peripheral monocytes/macrophages and in CD4 and CD8 lymphocytes, as well as in lymph node tissue of HIV/HCV-coinfected patients. The observation that HCV-infected persons have a strong increase of HCV viral load following HIV seroconversion is explained by the fact that lymphoid cells serve as an extrahepatic HCV replication reservoir when these cells are also infected with HIV. In multiple small studies (n = 10-65), initiation of HAART did not change HCV RNA levels in HIV/HCV-coinfected individuals, but to date no large study has been conducted.
In the clinical setting, HIV/HCV coinfection has a negative impact on HCV-induced liver disease. This has been confirmed in numerous studies. In a Spanish study of 547 HCV-infected patients, time from initial HCV infection to cirrhosis was much shorter in HIV/HCV-coinfected patients (7 years) than in patients with HCV only (23 years; P <0.001). A French study made similar observations and found the fibrosis progression rate (METAVIR fibrosis score 0-4) to be faster in HIV/HCV-coinfected patients (0.153 units per year) than in HCV-monoinfected individuals (0.106 units per year; P < 0.001). In HIV/HCV-coinfected patients, a low CD4 cell count, high alcohol consumption, and higher age at HCV infection were associated with a higher liver fibrosis progression rate. An independent association between low CD4 cell count and increase in advanced liver fibrosis was also found in an Italian study.
An autopsy series found a significantly higher rate of centrilobular fibrosis and cirrhosis in HIV/HCV-coinfected compared to HCV-monoinfected patients. A British cohort of 4865 hemophilic men found a liver-related death rate (presumably from HCV infection in most cases) within the first 25 years of treatment with blood products of 6.5% in HIV+ patients compared to 1.5% in HIV- individuals. In a French cohort of 373 patients with hemophilia and HCV-related liver cirrhosis, decompensation developed more rapidly in HIV/HCV-coinfected patients compared to HCV infection alone: 50% versus 13% at 2 years; 70% versus 40% at 5 years; P = 0.005. Furthermore, in that study the mortality rate was increased in HIV/HCV-coinfected patients (21% versus 4.5% at 2 years; 43% versus 12% at 5 years; P = 0.03). Multivariate analysis identified three independent predictors of death: HIV/HCV coinfection (RR, 5.7), no anti-HCV therapy (RR, 2.4), and increased age.
Graham and colleagues performed a meta-analysis of all studies that examined the influence of HIV on HCV-related liver disease. Comparing HIV/HCV-coinfected with HCV-monoinfected patients, they found a pooled relative risk for development of cirrhosis of 2.07 [95% confidence interval (CI), 1.40-3.07] and for liver decompensation of 6.14 (95% CI, 2.86-13.20) .
Development of fatal hepatocellular carcinoma was described in seven HIV/HCV-coinfected patients from Madrid. Compared to hepatocellular carcinoma patients with HCV-monoinfection, the HIV/HCV-coinfected patients were younger at diagnosis and had a shorter duration of HCV infection.
All of the above studies that confirm accelerated progression to HCV cirrhosis in HIV/HCV-coinfected patients, especially in those with low CD4 cell counts, were performed in the pre-HAART era (edit note: other studies conducted in the post-HAART era and presented at conferences have found that HIV accelerates HCV). There is one preliminary study that suggests that HAART may be able to slow down the accelerated fibrosis progression in HIV/HCV coinfection. In 182 HIV/HCV-coinfected patients, fibrosis progression rate was lower in patients on combination therapy with a protease inhibitor compared to patients who had never received a protease inhibitor. The estimated 15-year rate of progression from initial HCV infection to cirrhosis was 18% versus 5% (P < 0.001) in patients with or without protease inhibitor treatment.
Influence of HCV infection on the course of HIV disease
While it is apparent that HIV accelerates the course of HCV-induced liver damage, the effects of HCV coinfection on HIV disease progression are less clear. Studies show conflicting results, with most data showing no effect but some data indicating that HCV may increase the progression of HIV disease. Studies from the pre-HAART era consistently show that HIV/HCV coinfection does not influence progression of HIV-induced immunodeficiency or death.
Prospective cohort studies of HIV-infected patients in Edinburgh (N = 240), Italy (N = 416), and the Atlanta Veterans Affairs Medical Center (N = 350) all showed no difference in survival, clinical progression to first HIV symptoms and first AIDS-defining illness, as well as immunological progression to less than 200 x 106 CD4 cells/l between HCV+ and HCV- patients with HIV infection. In one prospective HIV cohort in Los Angeles, an independent association (controlled for CD4 cells and HIV viral load) was found between HCV viral load and progression to clinical AIDS (RR, 1.66; P = 0.016) and to AIDS-related mortality (RR, 1.54; P = 0.034). However, this study made no comparison to clinical progression and death with HIV-infected patients without hepatitis C.
Since HAART was introduced, four large studies on the influence of HCV on HIV disease progression have been published. Two of them demonstrate a faster progression of HIV disease in the setting of HIV/HCV confection, and the other two show no influence of HCV on HIV disease. In the Swiss HIV Cohort Study (N = 3111), Greub and coworkers reported that the risk of progression to clinical AIDS or death was independently associated with HCV seropositivity (hazard ratio [HR], 1.7; 95% CI, 1.26-2.30) and with active IDU (HR, 1.38; 95% CI, 1.02-1.88). They also found that HCV seropositivity was associated with a slower CD4 cell recovery with HAART.
The Italian antiretroviral-naive cohort study prospectively followed 1350 HIV-infected patients were after starting HAART. In this study, De Luca and colleagues showed that seropositivity to HCV was associated with an increased risk of progression to AIDS or death (HR, 1.55; 95% CI, 1.00-2.41; P = 0.05). Recovery of CD4 cell count was slower in HIV/HCV-coinfected patients, but HCV had no influence on HIV RNA response to HAART. By contrast, in the Johns Hopkins HIV cohort (n = 1955), Sulkowski and colleagues showed no influence of HCV coinfection on progression to clinical AIDS (HR, 1.03; 95% CI, 0.86-1.23) or to death (HR, 1.05; 95% CI, 0.85-1.30). The study found no difference in CD4 cell recovery after initiation of HAART between HCV+ and HCV- patients. Similarly, in the HIV cohort (n = 995) of Aquitaine (the region around Bordeaux), Rancinan and coworkers found no influence of HIV/HCV coinfection on overall mortality (HR, 1.20; 95% CI, 0.75-1.92; P = non-significant).
Chung and colleagues also found no influence of HCV coinfection on CD4 cell recovery with HAART. Thus, the notion that HCV does not alter the course of HIV disease has been shown in multiple studies, but two studies did show a faster progression to AIDS or death in the setting of HIV/HCV coinfection. Similarly, CD4 cell recovery was unchanged in coinfection in two studies, but impaired in two other studies. In subsequent debates, this discrepancy could not be resolved other than by pointing out the fact that each study had a different patient population.
Antiretroviral drugs can cause severe but reversible hepatotoxicity. In a cohort of 298 HIV-infected patients in an HIV clinic in Baltimore, it occurred in 10.4% of patients. The rate was highest with the use of ritonavir (30%). Coinfection with HCV was associated with a higher rate of hepatotoxicity (RR, 3.7; 95% CI, 1.0-11.8), but 88% of HIV/HCV-coinfected patients tolerated HAART well and without liver toxicity. Increased hepatotoxicity of HAART in HIV/HCV coinfection was also observed in a study from Madrid (HIV+/HCV+, 21% versus HIV+, 7%; P= 0.03). Melvin and colleagues found more frequent and earlier discontinuation of HAART in HIV/HCV-coinfected patients than in HIV patients who were HCV-. It is conceivable that effective treatment of HCV infection in HIV+ patients could improve tolerance for antiretroviral therapies, but this has not been studied.
Treatment of chronic hepatitis C in HCV-monoinfected patients
The primary goal of HCV treatment is permanent clearance of the virus or sustained viral response (SVR), defined as undetectable HCV RNA 6 months after the end of therapy. An SVR has a lasting impact on HCV-induced liver disease. It prevents clinical complications from liver disease and liver-related death, liver inflammation resolves, and over 5-10 years, HCV-induced liver fibrosis either resolves completely or regresses to a very mild level.
Reversal of cirrhosis is observed in 49% of patients who have received 48 weeks of treatment with an interferon-based regimen. However, even in the absence of sustained clearance of HCV, interferon-based therapy can slow down progression of liver fibrosis and reduce the risk of progression to liver decompensation or hepatocellular carcinoma.
Within a matter of 12 years (1989-2001), there has been great progress in the development of therapies against chronic hepatitis C. Monotherapy with interferon [alpha]-2b at 3 million international units (MIU) three times weekly for 6 months had a low sustained response rate of 10%. Combination therapy with interferon [alpha]-2b and ribavirin improved the SVR rate to 38% (US) and 43% (Europe and Canada), with an SVR rate in HCV genotype 1 of 29% and in HCV genotypes 2 and 3 of 66%. Following SVR with interferon [alpha]-2b plus ribavirin therapy, HCV RNA remained undetectable after 4 years in 98.2% of patients.
Recently, the introduction of once-weekly pegylated interferon that sustains interferon blood levels through the week has further improved response rates. Two products have been approved: 12KD peginterferon [alpha]-2b and 40KD peginterferon [alpha]-2a. Manns and coworkers reported that combination therapy of 12KD peginterferon [alpha]-2b at 1.5 [mu]g/kg per week and ribavirin 800 mg daily led to an improved SVR rate of 54% compared to the prior therapy of choice, standard interferon [alpha]-2b plus ribavirin (47%; P = 0.01). With peginterferon [alpha]-2b plus ribavirin, SVR rate was 42% in HCV genotype 1 versus 33% with interferon [alpha]-2b plus ribavirin (P = 0.02) and 82% in HCV genotypes 2 and 3 (no difference with interferon [alpha]-2b plus ribavirin).
The combination of 40KD Peginterferon [alpha]-2a at 180 [mu]g per week and ribavirin 1000/1200 mg per day was compared to interferon [alpha]-2b plus ribavirin in a study by Fried and colleagues. Peginterferon [alpha]-2a plus ribavirin achieved an SVR rate of 56% compared to 45% with interferon [alpha]-2b plus ribavirin (P = 0.001). Peginterferon [alpha]-2a plus ribavirin had an improved SVR rate in both HCV genotype 1 (46% versus 36%; P = 0.016) and in HCV genotypes 2 and 3 (76% versus 61%; P = 0.008).
In both studies, influenza-like symptoms, cytopenia and neuropsychiatric symptoms were the main side effects with either pegylated or standard interferon plus ribavirin. A recent study by Hadziyannis and coworkers found that peginterferon [alpha]-2a plus ribavirin had a better SVR rate in HCV genotype 1 when given for 48 weeks with a ribavirin dose of 1000-1200 mg per day (51%) versus 800 mg/day (40%, P = 0.03). By contrast, in HCV genotypes 2 and 3, the SVR rate was 78% with 24 weeks of treatment and a low ribavirin dose of 800 mg/day with no difference to 48 weeks of treatment and a ribavirin dose of 1000-1200 mg/day. Thus, with the new combination of pegylated interferon and ribavirin, rates of SVR can be achieved in 42-51% of patients with HCV genotype 1 and in about 80% of patients with genotype 2 or 3.
To study the effect of pegylated interferon on liver fibrosis progression and clinical progression of liver disease after viral non-response to peginterferon and ribavirin, two large randomized, controlled trials are currently ongoing. One study uses low-dose peginterferon [alpha]-2a at 90 [mu]g per week versus no treatment for 4 years (HALT-C) and the other trial gives low-dose peginterferon [alpha]-2b at 0.5 [mu]g/kg per week versus daily colchicine for 4 years (COPILOT).
Treatment of chronic hepatitis C in HIV/HCV-coinfected patients
Interferon [alpha]-2b monotherapy
There are still relatively few data on treatment of chronic hepatitis C in HIV/HCV-coinfected patients, and most are from small, uncontrolled observational studies. An uncontrolled study of monotherapy with interferon [alpha]-2b at 3 MIU three times weekly for 12 months showed an SVR rate of 23% in 80 HIV/HCV-coinfected patients that was similar to the SVR rate of 26% in 27 HCV-monoinfected patients (P = 0.72). Independent predictors of SVR in HIV/HCV-coinfected patients were baseline HCV viral load above 10 million copies per ml and CD4 cell count above 500 x 106/l. For interferon monotherapy, the SVR rates in this study were relatively high (usually 10-12%), probably due to the fact that only 60% of patients were infected with HCV genotype 1.
Interferon [alpha]-2b and ribavirin combination therapy
The American Foundation for AIDS Research (amfAR) study DCRI 010, a placebo-controlled multi-center study, used therapy with interferon [alpha]-2b at 3 MIU three times a week combined with either initial or 16-week delayed ribavirin at 800 mg/day in 110 patients (76% with HCV genotype 1). Viral response at week 12 was 23% in the interferon [alpha]-2b plus ribavirin combination group compared to 5% in the interferon [alpha]-2b plus placebo group (P =0.016), and the SVR rate was 11% in the initial interferon [alpha]-2b plus ribavirin group compared to 5% in the delayed ribavirin group (P = non-significant). Early discontinuation rate was high at 50%, with 21% of patients stopping treatment due to an adverse event, 11% due to viral non-response at week 24 and 4% due to recurring substance abuse. The most common adverse events that led to treatment discontinuation were depression (6%), anxiety, anemia, and fatigue (each 3%). In the first 16 weeks (double-blind phase), anemia occurred in 28% of patients on interferon and ribavirin compared to 5% in the interferon + placebo arm (P = 0.002). Other adverse events were influenza-like symptoms that have been reported in a similar fashion in treatment of HCV-monoinfected patients. No new types of side effects were noted.
A second randomized, controlled trial in HIV/HCV-coinfection, Hepatitis Resource Network study HRN-002, compared daily versus three times per week (tiw) interferon [alpha]-2b at 3 MIU, each combined with ribavirin 800 mg/day (n = 155). The frequency of HCV genotype 1 was 77%. The early viral response rate at week 12 was 45% (daily) and 16% (tiw; P = 0.0001). The SVR rate was 20.3% in the daily group and 8.6% in the tiw group (P < 0.001). In this study, too, the early discontinuation rate was high at 69% in the daily interferon group and 87% in the tiw group (P = 0.008), due mainly to lack of viral response at week 24. Main adverse events were depression, fatigue, and anemia, similar to the amfAR trial.
The safety profile of interferon [alpha]-2b plus ribavirin for 48 weeks in these two randomized, controlled coinfection studies are summarized in Table 1 and are compared to the safety data from the two large trials in HCV monoinfection reported by Poynard and McHutchison. Early discontinuation rates were 50% (Amfar) and 86% (HRN-002) for co-infection studies vs 21% (McHutchison) and 19% (Poynard) for monoinfection studies. These data clearly indicate more difficulty tolerating therapy for co-infected individuals, and the discontinuation rates effect the sustained viral response rates. Three small observational studies, in Paris (n = 51, SVR in 21%), Barcelona (n = 20, SVR in n = 8), and Aviano, Italy (n = 19, SVR in 19%) found similar toxicity. An observational study in New York found a similar SVR rate in 32 HIV/HCV-infected patients of 22% compared with 64 matched HCV-monoinfected patients (27%; P = non-significant). None of the studies noted a significant change in HIV viral load or CD4 percentage. A drop in CD4 absolute cell count was explained by interferon-induced leukopenia. In summary, the SVR rate in HIV/HCV-coinfected patients treated with interferon [alpha]-2b plus ribavirin is around 15-20%.
Peginterferon [alpha] (2a or 2b) and ribavirin combination therapy
Several randomized, controlled trials of pegylated interferon plus ribavirin combination therapy in HIV/HCV-coinfected patients are currently ongoing, and only preliminary data have been reported so far. In the French RIBAVIC study, peginterferon [alpha]-2b at 1.5 [mu]g/kg per week is compared to standard interferon [alpha]-2b at 3 MIU tiw, each combined with ribavirin 800 mg/day (N = 418). While efficacy data are still incomplete, Perronne and coworkers have reported serious safety issues in this study, with a rate of treatment discontinuation of 30% and severe side effects in 24% of patients. Twenty patients (5.7%) required hospitalization, and eight (1.9%) were hospitalized with symptomatic lactic acidosis. Of the eight patients with lactic acidosis, all received concurrent therapy with didanosine (ddI), and seven were also receiving stavudine (d4T).
The AIDS Clinical Trials Group study 5071 is a pilot study of 134 patients comparing safety and efficacy of peginterferon [alpha]-2a at 180 [mu]g per week with standard interferon [alpha]-2b at 3 MIU tiw, each combined with increasing doses of ribavirin from 600 to 1000 mg/day, as tolerated. The viral response rate at week 24 (the study's primary endpoint) was 44% in peginterferon [alpha]-2a plus ribavirin compared to 15% in interferon [alpha]-2b plus ribavirin (P < 0.001). Treatment discontinuation was the same in both groups at 12%. There was a higher rate of grade 4 (life threatening) adverse events in the pegylated interferon group (17%) compared to the standard interferon group (4%, P = 0.0012), but most of these events were laboratory abnormalities.
In the first 24 weeks, there was a small increase in percent CD4 cells (+3.5% in peginterferon plus ribavirin versus +2.5% in interferon plus ribavirin), but a drop in absolute CD4 cell count (-194 x 106/l versus -112 x 106/l) due to interferon-induced leukopenia. A third study (APRICOT), a large phase III trial (N = 868) that compares peginterferon [alpha]-2a plus ribavirin versus peginterferon [alpha]-2a (plus placebo) versus interferon [alpha]-2a plus ribavirin is currently ongoing.
(edit note from Jules Levin: preliminary data presented at conferences from studies exploring pegylated interferon plus ribavirin in co-infected patients suggest reduced SVRs of about 20-35%, which compares to 54% seen in monoinfection studies. It is difficult to evaluate the SVR results from these co-infection studies. Clearly, discontinuation rates are high due to difficulty in tolerating medications. It is hard to evaluate response rates in co-infected patients who are able to tolerate therapy and complete the full course. Several basic science studies suggest HIV impairs the immune response to HCV and HCV therapy. My feeling is that co-infected patients can achieve better SVRs than indicated by these studies, if the patients complete the full course of therapy. The SVRs may not be comparable to those seen in monoinfected but they may be considerably better than that seen in the studies reported. Support services including adherence programs, education, and access to Procrit are important to prevent discontinuations. Dose reductions or temporary stopping of ribavirin due to anemia in the first 1-2 months of treatment can reduce SVRs. Use of Procrit can increase hemoglobin and improve quality of life, and perhaps prevent discontinuation and improve SVR. As well, use of anti-depressants can prevent discontinuation and improve tolerability of therapy).
Development of symptomatic lactic acidosis and other mitochondrial toxicity has emerged as a serious safety issue of HCV treatment with interferon and ribavirin in HIV/HCV-coinfected patients, especially in those patients whose HAART regimen includes a dideoxy-nucleoside like ddI or d4T. In January 2001, Lafeuillade reported the first two cases of symptomatic lactic acidosis during therapy with interferon and ribavirin, one receiving ddI and the other d4T. Since then, 36 more cases of either lactic acidosis or pancreatitis in patients concurrently receiving ribavirin and ddI have been reported to the US Food and Drug Administration (FDA). As a result, the product information of didanosine (Videx) was revised in September 2002 and now contains a warning to use ribavirin with caution in patients receiving ddI.
While lactic acidosis and other mitochondrial toxicity are rare side effects of ddI, the FDA reports that ribavirin treatment increases exposure to ddI and its triphosphorylated active metabolite and thus increases its toxicity. In addition to Lafeuillade's first report of lactic acidosis in a patient receiving ribavirin and d4T, other cases have been linked to this combination, including the above mentioned preliminary safety report of the RIBAVIC trial. As yet, the FDA has not instructed the manufacturer of d4T to include a warning in the FDA-approved product information, but in February 2002, the manufacturer sent a letter to clinicians warning them about the possibility of lactic acidosis in this setting.
Based on these data, combination therapy with interferon and ribavirin should be used only with caution in patients who are receiving ddI or d4T. It should be noted, however, that the FDA warning is based on cumulative case reports. Lactic acidosis had previously been reported in HIV-infected patients being treated with ddI or d4T without the concurrent use of ribavirin. This is the result of the higher affinity of these drugs to the mitochondrial DNA polymerase gamma compared to other nucleoside analogs like zidovudine or lamivudine with rare to no incidence of lactic acidosis. To date, no comparative study has been performed on the incidence of symptomatic lactic acidosis in patients taking ddI or d4T, with and without co-administration of ribavirin.
Special considerations in HCV therapy for HIV-infected persons
There are other areas of concern in HCV treatment that are specific to HIV/HCV-coinfected patients, namely interactions between ribavirin and other nucleoside analogs, cytopenia, and CD4 cell count. In vitro, ribavirin can inhibit phosphorylation of zidovudine, d4T and zalcitabine, which is required for intracellular antiviral activity. In the preliminary data presented so far, breakthrough of HIV RNA or decreased response to HAART while on ribavirin-based combination therapy have not been observed; however, complete data of all ongoing randomized, controlled studies have not yet been analyzed.
In patients with advanced HIV disease, cytopenia, especially anemia and neutropenia, is common. Therefore, all current HCV treatment protocols have used a reduced ribavirin dose of 800 mg/day to minimize anemia induced by HCV therapy. However, over time, HAART can normalize HIV-induced anemia, and today many HIV-infected persons on HAART have normal hemoglobin levels. It remains to be seen whether the degree of anemia induced by interferon plus ribavirin treatment is more pronounced in HIV/HCV-coinfected patients than in comparable HCV-monoinfected patients. The optimal ribavirin dose in HIV/HCV coinfection remains to be defined, and it is probably higher than the currently used dose of 800 mg per day. A decrease in absolute CD4 cell count is almost always observed during treatment with interferon. However, this is not due to worsening of immunodeficiency, since CD4 percentage is typically stable, but it is caused by leukopenia due to interferon treatment.
The reasons for the worse response to HCV therapy in HIV/HCV coinfection are not clear, but several theories are being discussed. As mentioned earlier, HIV/HCV-coinfected patients have a higher HCV viral load. A high viral load has been associated with a lower SVR rate in all studies of HCV-monoinfected patients, and the same phenomenon has been observed in the above-mentioned HCV treatment studies in HIV/HCV-coinfected patients.
Another possibility is the increased number of HCV quasispecies with increased variability in the envelope-coding region that have been shown in HIV/HCV-coinfected patients compared to HCV-monoinfected patients. These HCV variants may respond less well to interferon. Another explanation would be the lower dose of ribavirin used in HIV/HCV-coinfected patients (800 mg/day) and the greater incidence of ribavirin dose discontinuations due to anemia.
In the amfAR trial, it occurred in 5.7% of HIV/HCV-coinfected patients compared to 0.9% and 0.4% in the two large HCV-monoinfection studies. Reduced duration and dosing of ribavirin therapy has been associated with a decreased rate of SVR. A decreased rate of SVR in HIV/HCV-coinfection may also be due to a greater need for dose reductions of interferon due to neutropenia, but there are no prospective data. Ongoing trials will give answers to these questions.
The National Institutes of Health Consensus Conference Statement of June 2002 has recognized the important issue of chronic hepatitis C in HIV-coinfected patients. As already recommended by the Centers for Disease Control and Prevention since 1998, the statement re-emphasizes the need for all HIV-infected persons to be tested for HCV. It also recommends that all HIV/HCV-coinfected patients should be considered for possible HCV therapy, with individualized therapy and monitoring of side effects such as lactic acidosis.
Which HIV/HCV-coinfected patients should be treated for chronic hepatitis C and how?
Generally, all HIV-infected patients with HCV viremia (HCV RNA positive) should be evaluated for therapy. Regardless of alanine aminotransferase (ALT) values, a liver biopsy should be done in all patients without a clear contraindication to evaluate the grade of necroinflammation and the stage of liver fibrosis. The more advanced the fibrosis, the more therapy against HCV is indicated. Generally, HCV treatment is indicated for patients with bridging fibrosis or cirrhosis, or in lesser disease, when clinician and patient aim at achieving a viral cure (which is more likely with HCV genotypes 2 or 3 and a low HCV viral load).
Patients with normal ALT values (52% in HIV/HCV-coinfection) should not be excluded from HCV therapy, since they have similar rates of SVR as patients with elevated ALT levels. In patients with advanced fibrosis or cirrhosis where HCV therapy is most indicated, an effort should be made to prepare the patient for treatment by removing or improving conditions that are deemed relative or absolute contraindications to treatment.
Preexisting anemia or neutropenia can be addressed with appropriate growth factors, and where active psychiatric disease (including substance and alcohol abuse) precludes interferon treatment, close collaboration with a psychiatric team should aim at stabilizing the HIV/HCV-coinfected patient for HCV therapy.
Several studies have shown that, under close psychiatric supervision, patients with a wide range of psychiatric disease (including schizophrenia and bipolar disorder) can complete a course of interferon plus ribavirin therapy, and they have similar rates of treatment discontinuation (15%) and of SVR (37%) as non-psychiatric patients with chronic hepatitis C.
However, patients with former drug addiction who are not on methadone maintenance have a higher rate of treatment discontinuation (42%). In these patients, sufficient time should be allowed between end of active substance abuse and start of interferon therapy, typically at least 1 year. Active management of side effects of HCV therapy like initial influenza-like symptoms, cytopenia, neuropsychiatric symptoms (such as depression, mood swings, irritability, insomnia, etc.), ribavirin-induced rash, and others, should aim at avoiding early treatment discontinuation and minimizing dose reductions of both interferon and ribavirin where possible.
While in patients monoinfected with HCV genotypes 2 or 3, treatment with interferon (standard or pegylated) and ribavirin can be limited to 24 weeks without decrease in SVR, the same phenomenon has not yet been shown for HIV/HCV-coinfection. Until ongoing large trials clarify the issue, treatment for HIV/HCV-coinfected patients should be for 48 weeks. Similarly, HCV-monoinfected patients who do not achieve a viral response (undetectable HCV RNA) or at least a 100-fold drop in HCV RNA level on peginterferon plus ribavirin at week 12, have a chance of SVR of only 3%, and in these patients treatment is typically discontinued. Currently, there are no data on whether the same applies to HIV/HCV-coinfected patients, and again the results of current trials will provide data to support recommendations on early stop rules. Until then, however, it appears reasonable to consider HCV treatment discontinuation in coinfected patients who have side effects and do not have a week 24 viral response.
Management of HIV/HCV-coinfected patients with end-stage liver disease
Treatment of HCV infection in the setting of decompensated liver disease
The worst outcome of HCV-related liver disease is end-stage liver disease (ESLD) with complications of portal hypertension and bleeding esophageal varices, ascites, hepatic encephalopathy, or hepatocellular carcinoma. At this advanced stage, therapy directed against HCV remains largely untested, and management focuses on symptomatic treatment of end-stage liver disease and placement of patients on a liver transplant list.
In a single pilot study, Crippin and colleagues treated 15 patients on a liver transplant list with either interferon [alpha]-2b alone (n = 9) or interferon plus ribavirin (n = 6). An on-treatment viral response during therapy was seen in five patients (33%), which is similar to treatment response in compensated cirrhosis (32-43%). Toxicity was high with 50% cytopenia and 20% hepatic encephalopathy. Two of 15 patients developed serious infections, one of them fatal. As a result, the study was terminated before a rate of SVR could be determined. These very preliminary results show that a viral response of HCV in decompensated liver disease is possible, but treatment toxicity needs to be better defined. Future trials need to include an untreated control group to determine whether hepatic encephalopathy and bacterial infections (both common in ESLD itself) are more frequent with HCV therapy than without. HCV therapy in ESLD remains experimental and should not be performed outside a research study. This leaves transplantation as the only option for patients with HCV-induced ESLD.
Liver transplantation in HIV-infected patients
Liver transplantation, though, has not generally been available to the HIV/HCV-coinfected patient with ESLD. Until recently, no HIV-infected patient would be considered for any transplantation, mostly because of the limited life expectancy and the uncertainty of the influence of post-transplant immunosuppressive therapy on HIV disease. Indeed, in the pre-HAART era, HIV-infected patients receiving liver transplants had a poor outcome.
However, as HAART has significantly prolonged survival of HIV-infected patients, the issue of transplantation is being reconsidered. Currently, a few centers perform transplantations with liver or kidney in HIV-infected persons, though, in many cases, each center operates under its own criteria and protocol.
Among the first four liver transplants in HIV/HCV-coinfected patients at King's College in London, all four died between 3 and 22 months after transplantation, whereas three HIV/HBV-coinfected patients with ESLD survived transplantation up to 33 months.
In a first interim report of retrospective data pooling from five HIV transplant centers in the USA and the UK, Ragni and colleagues reported outcome on 23 liver transplants in HIV-infected patients with a median follow-up of 15 months. The overall death rate (no timeline) was 30% (7/23) for all liver transplants, all seven of them occurring in HIV/HCV-coinfected patients, leading to an overall mortality rate of 44% (7/16) for the HIV/HCV-coinfected liver transplant recipients.
Better survival in the HIV-infected liver transplant recipients was associated with better post-transplant tolerance of HAART, higher age, not being hemophiliac, and a higher CD4 cell count in the post-transplant phase. In a subsequent analysis, the mortality at months 12, 24, and 36 was 9.1%, 24.1%, and 24.1%, respectively, which was found to be comparable to United Network for Organ Sharing (UNOS) mortality statistics of HIV-negative liver transplant recipients (P = 0.643).
The latter statistics included both HCV+ and HCV- transplant recipients. In a prospective pilot study of 19 HIV-infected liver transplant recipients (HCV+ or HCV-), Roland and colleagues found a 1-year mortality of 8.0% which compared favorably to the UNOS statistics of HIV- liver transplant recipients whose 1-year mortality is 12.1%. No mortality data on the subgroup of HIV/HCV-coinfected patients were provided. One of 19 patients progressed to clinical AIDS with cytomegalovirus esophagitis. The above post-transplant mortality statistics indicate that HIV-infected liver transplant recipients who are not coinfected with HCV seem to survive the transplantation equally well as HIV-negative patients. However, mortality appears to be increased in HIV/HCV-coinfected patients. The data presented to date are too small in number to allow full conclusions on the safety and survival rate of liver transplantations in HIV/HCV-coinfected patients.
A large prospective transplantation study with a unified protocol sponsored by the National Institutes of Health is now being initiated and will eventually provide valuable data. Given the small number of patients who have received a liver transplant, the procedure remains generally unavailable for HIV/HCV-coinfected patients with ESLD, but efforts are underway to increase access to liver transplantation. Therefore, all the more effort should be put into treating HCV infection before it leads to ESLD.
Future research questions
Peginterferon and ribavirin will probably emerge as the next therapy of choice for chronic hepatitis C in HIV/HCV-coinfected patients. Currently ongoing trials will provide more data on this combination therapy in the near future. However, optimal dosing, especially of ribavirin, is an area that requires further investigation. Also undefined is the role of growth factors like granulocyte-colony stimulating factor (G-CSF, filgrastim) or recombinant human erythropoietin (rhEPO, epoetin-[alpha]). In HCV-monoinfected patients, treatment with rhEPO has been shown to improve interferon plus ribavirin-induced anemia and to leave the ribavirin dose unchanged. A similar trial is currently ongoing for HIV/HCV-coinfected patients with HCV treatment-induced anemia. Maintenance therapy with low-dose peginterferon-[alpha] to reduce fibrosis progression after viral non-response to optimal HCV therapy is currently being evaluated in two trials, as mentioned earlier. An ongoing third randomized, controlled trial is investigating the same concept in HIV/HCV-coinfected patients, who will receive low dose of peginterferon [alpha]-2a versus no treatment after viral non-response to peginterferon [alpha]-2a plus ribavirin.
Another controversial issue is whether HIV or HCV should be treated first in patients whose HIV status would not require initiation of HAART. There is a general consensus that in HIV/HCV-coinfected patients with symptoms or low CD4 cell count, HAART should be initiated first, and HCV therapy only after stabilization of HIV disease. However, whether otherwise stable HIV-infected persons need to be started on HAART prior to beginning HCV treatment needs further study.
Chronic hepatitis C can lead to progressive liver disease in HIV/HCV-coinfected patients, and today accounts for significant morbidity and mortality in HIV-infected individuals who under HAART are living longer and healthier lives. Combination treatment with interferon and ribavirin can lead to a sustained viral response, but treatment success is less pronounced than in HCV-monoinfected patients. HCV therapy with peginterferon plus ribavirin, already the treatment of first choice for HCV-monoinfection, is likely to soon become the future therapy for HIV/HCV-coinfection as well. Side effects to HCV therapy are common and predictable, but with active management of these side effects most patients can complete 12 months of treatment. Any HIV-infected patient who is also viremic with HCV is at risk for progressive liver disease and should have a liver biopsy to assess the degree of liver inflammation and fibrosis. HCV therapy should be offered to all eligible patients, especially those with advanced fibrosis.