Treatment of Hepatitis C in HIV-Infected Patients: Title and subTitle BreakSignificant Progress But Not the Final Step

In the modern era of highly active antiretroviral therapy (HAART), hepatitis C virus (HCV) has emerged as a major cause of morbidity and death in human immunodeficiency virus (HIV) infection. An estimated 15% to 30% of HIV-infected individuals are coinfected with HCV, representing 150 000 to 300 000 patients in the United States alone.¹ The prevalence of anti-HCV antibodies in HIV-infected persons differs significantly according to the HIV-exposure risk category, ranging from less than 10% in homosexual men to more than 85% in injecting drug users.¹ In the vast majority of immunocompetent non–HIV-infected persons, chronic hepatitis C usually takes a relatively mild course, leading to liver cirrhosis in 2% to 25% after 20 to 25 years depending on the individual risk factor profile.^{2 - 3} Liver-related death is only slightly more frequent in HCV-infected individuals.⁴ In contrast, hepatitis C may take a much more severe course in HIV-infected patients.⁵ In men with hemophilia, the prevalence of cirrhosis as well as liver-related deaths is significantly higher in HCV-HIV–coinfected individuals than in patients infected with HIV alone.⁶ However, these findings may not necessarily reflect the situation in other cohorts in which HCV infection has not been shown to increase the risk of death.⁷

Since antiretroviral therapy can be associated with significant hepatotoxicity, there has been some concern that HAART, specifically protease inhibitors, may promote progression of fibrosis in HCV-RNA–positive patients. Fortunately, this has not been the case,⁸ and antiretroviral therapy has even been shown to reduce long-term liver-related mortality in HIV-HCV–coinfected patients.⁹ In contrast, HCV infection appears to influence the natural history of HIV disease. In the Swiss HIV cohort, among more than 3000 HIV-infected individuals, those also infected with HCV had a modestly increased risk for developing AIDS and a less robust increase in CD4 cell counts after initiation of antiretroviral therapy compared with patients infected with HIV alone.¹⁰ On the other hand, a study from Baltimore found that HCV coinfection had no influence on developing AIDS or response to HAART.⁷ These differences highlight the problem that different cohorts may be difficult to compare because both HIV and HCV disease may have different trajectories depending on the route of infection (eg, blood transfusions vs community acquired) and other factors.

Interactions between HCV and HIV are difficult to investigate. Although the primary site of replication differs (for HCV, the liver; for HIV, T cells, monocytes, and dendritic cells), HCV may also replicate in dendritic cells and monocytes, while HIV can cause hepatitis.¹ Thus, both viruses may directly interact with the other’s replication and influence the host’s immune response to the other virus. Hepatitis C virus replication is much more rapid in HIV-infected patients than in patients infected with HCV alone, indicating decreased immune function.¹ On the other hand, HCV proteins may directly inhibit the function of dendritic cells and natural killer cells and thus alter the anti-HIV–specific immune response.^{11 - 12} Thus, coinfection with HIV and HCV may influence viral load and immune responses and thereby affect efficacy of antiviral treatment against both HIV and HCV.

Treatment options for patients infected with hepatitis C have significantly improved only in recent years. Acute hepatitis C^{13 - 14} as well as chronic HCV genotype 2 or 3 infection^{15 - 16} is now considered a curable disease in immunocompetent patients. Even patients infected with the more difficult-to-treat HCV genotypes 1 and 4 can achieve sustained clearance rates of HCV-RNA in about 50%.^{15 - 17} Three major steps led to this significant improvement. First, the introduction of therapy with interferon alfa plus ribavirin increased response rates from less than 20% to more than 40%.¹⁸ Second, the development of pegylated interferons, permitting therapy to be administered weekly instead of 3 times per week, further increased responses above 50%. Third, treatment adherence increased due to better understanding of and response to adverse events by patients and their treating physicians.

Currently, 2 pegylated interferons are available: peginterferon alfa-2a, with a branched 40-kDa methoxypolyethylene glycol moeity attached covalently to interferon alfa-2a,¹⁹ and peginterferon alfa-2b, with a single 12-kDa polyethylene glycol molecule attached to interferon alfa-2b.²⁰ Peginterferon alfa-2a is predominantly metabolized in the liver, whereas peginterferon alfa-2b is predominantly cleared by the kidney.^{21 - 22} The pegylation process also restricts the volume of distribution, such that the 40-kDa peginterferon alfa-2a is limited to the blood and interstitial fluid, whereas the 12-kDa peginterferon alfa-2b has a wider distribution. As a result, peginterferon alfa-2a does not need to be adjusted for body weight. Although no large studies have compared the 2 drugs directly, when they are administered in combination with ribavirin to patients infected with HCV alone, the differences in sustained virologic response rates are small and both peginterferons can be recommended equally.^{15 - 17}

Several studies have explored the efficacy of interferon alfa plus ribavirin combination therapy in HCV-HIV–coinfected patients, but they showed disappointing results.^{23 - 24} Thus, the question arose whether pegylated interferons, established as more effective in non–HIV-infected patients, could enhance virologic responses in coinfected patients. Two trials of peginterferon alfa-2a plus ribavirin in HCV-HIV–positive individuals were published recently. Torriani et al conducted the larger trial in 19 countries, with most patients recruited in Europe.²⁵ Chung et al²⁶ conducted a smaller study in the United States. In this issue of JAMA, Carrat and colleagues²⁷ report the third trial investigating pegylated interferon and the first investigating peginterferon alfa-2b in combination with ribavirin for the treatment of HCV in HIV-positive patients. The study was performed in France in 71 centers and included more than 400 individuals. Study design, baseline characteristics, and virologic response rates of all 3 trials are summarized in the Table.

Comparing the 3 trials yields several important and some hard-to-explain differences. First, the trial by Torriani et al showed the highest response rates in the peginterferon-plus-ribavirin group. However, the conventional interferon-plus-ribavirin group had surprisingly low responses, with a sustained virologic response rate of only 12%.²⁵ The unblinded injection schedule may explain in part the higher discontinuation rates in the conventional interferon group, assuming patients believed they were receiving the less effective treatment. On the other hand, in the study by Carrat et al, responses in the standard interferon alfa-2b-plus-ribavirin group were rather high at 20%.²⁷ Neither report convincingly explains these differences. Even the 2 trials that used peginterferon alfa-2a showed significant differences in viral responses. For peginterferon alfa-2a monotherapy, differences in treatment responses between US patients²⁸ and patients from Europe, Canada, and Australia²⁹ have already been reported and may be explained by a higher proportion of black patients and higher baseline body mass index values in US trials, both of which are associated with lower response rates.

Carrat et al achieved sustained viral response rates of 17% for patients with an HCV genotype of 1 or 4. This rate, as well as the sustained virologic response rates of patients infected with genotype 1 in the other 2 studies, is especially low compared with immunocompetent patients infected with genotype 1. Whether prolonging therapy beyond 48 weeks could improve the poor outcome in genotype 1 patients remains to be investigated; in immunocompetent patients, individuals with a high body mass index and high viral load may benefit from prolonged treatment.³⁰ Importantly, Carrat et al reported that all but 1 patient who did not respond by 12 weeks also had not responded at the 72-week end point, confirming the usefulness of the 12-week stopping rule for HIV-HCV–coinfected patients.

The study by Carrat et al also had significantly lower response rates in patients with HCV genotype 2 or 3 compared with the 2 other trials. Recently, genotype 2 was shown to respond better to peginterferon-based therapies than was genotype 3.³¹ The study by Carrat et al and the previous 2 studies do not report the distribution of genotype 2 and 3. Regardless, the low response rate for the “good” genotypes 2 and 3 in the study by Carrat et al highlights the need for 48 weeks of therapy in HIV-positive patients vs only 24 weeks for genotypes 2 and 3 in patients who are not infected with HIV. Carrat et al found that for patients infected with HCV genotypes 2 and 3, peginterferon alfa-2b was not superior to conventional interferon alfa-2b, a finding that replicates the results for HIV-negative patients infected with the same virus genotypes.¹⁵

Ribavirin causes hemolysis in a dose dependent manner, a significant problem particularly for HIV-infected patients receiving HAART. Thus, compared with doses used for patients infected with HCV alone, ribavirin was underdosed in all 3 trials on HIV-infected patients. The importance of a sufficient ribavirin dose has been shown now by several investigators.^{15 - 16 ,18} Future studies have to address this important issue. Whether drugs such as viramidine, currently under investigation, can induce less hemolysis and still retain the immunomodulatory function of ribavirin remains to be seen. Epoetin alfa may be an alternative to maintain ribavirin dose in some patients.³²

Another interesting point of the article by Carrat et al is that successful antiviral therapy was associated with decreased hepatic steatosis in patients infected with genotype 3. Similar findings have been reported for patients not infected with HIV.³³ However, the importance of reduced fat in the liver may be much more significant for HIV-positive persons. Steatosis indicates mitochondrial toxicity and thus can limit the administration of appropriate antiretroviral therapies.

Therapy with interferon-based therapies can cause significant adverse effects including flulike symptoms, autoimmune disorders such as thyroiditis, and psychiatric diseases.³⁴ Carrat et al reported only slightly higher frequencies of the major adverse effects compared with previous studies of patients infected with HCV alone.²⁷ Treatment was completed according to protocol in about three-fifths of the patients. Psychiatric adverse effects including depression were reported with low frequency, suggesting that the presence of psychiatric diseases may not necessarily require exclusion from interferon alfa-based therapies. Patients with more severe psychiatric disease might benefit from consultation with a psychiatrist before and during treatment to ensure optimal adherence with therapy.³⁵ Antidepressant medication can significantly reduce the frequency of depression during interferon therapy.³⁶ Importantly and again consistent with findings in patients infected with hepatitis C alone,³⁷ adherence to therapy was a major factor determining treatment response in the study by Carrat et al.²⁷ Therefore, treating patients in an optimal setting of experts in different fields may help to achieve this goal.

Pancreatitis is an important consideration if peginterferon-plus-ribavirin therapy is started in HIV-infected patients receiving HAART. Some HIV patients receiving HAART treated with ribavirin for hepatitis C experience hyperlactatemia.³⁸ Ribavirin, a guanosine nucleoside analogue, may interact with nucleoside reverse transcriptase inhibitors and antagonizes anti-HIV drugs such as zidovudine and stavudine in vitro.³⁹ In addition, through inhibition of inosine-5′-monophosphate dehydrogenase, ribavirin facilitates conversion of didanosine to its active metabolite and thus potentially causes mitochondrial toxicity. Three cases of pancreatitis were reported in the ribavirin groups in the trial by Torriani et al while 1 case of pancreatitis was reported in the trial by Chung et al. Carrat et al reported 11 cases of pancreatitis, all in patients treated with didanosine-containing antiretroviral regimens. Thus, careful evaluation of antiretroviral therapy is mandatory before ribavirin therapy is started in patients infected with HIV. If possible, the HAART regimen should be changed if didanosine is included.

Patients with cirrhosis may experience hepatic decompensation during interferon therapy. In the trial by Torriani et al, 10% of patients with cirrhosis experienced decompensation while receiving treatment⁴⁰ and 6 of these 14 patients died, highlighting that close monitoring of patients with cirrhosis and HCV infection is warranted, whether or not they also have HIV infection. In the study by Carrat et al, 2 of the 80 patients with significant fibrosis or cirrhosis taking peginterferon alfa-2b plus ribavirin experienced liver failure.

Although CD4 cell counts of patients in the study by Carrat et al decreased during antiviral therapy, CD4 cell counts normalized after the end of therapy or even increased compared with baseline levels. However, patients selected for all 3 trials had relatively high baseline CD4 cell counts with a mean of about 500 × 10⁶/L. Patients with CD4 counts below 200 × 10⁶/L should not be considered for interferon-based therapies. Human immunodeficiency virus–viral load was not affected by treatment in the studies by Chung et al and Carrat et al. In the trial by Torriani et al, patients receiving peginterferon alfa-2a experienced a 0.7 log decline of HIV-RNA from week 0 to week 48. Overall, all 3 studies showed the safety of peginterferon-plus-ribavirin therapy in for patients infected with HIV if patients are selected carefully based on their individual risk profile.

A final issue that has not been considered by any of the 3 studies is the presence or absence of GBV-C (hepatitis G) coinfection. In 1998, Heringlake et al⁴¹ reported an improved course of HIV infection if patients were also coinfected with the GBV-C virus, a phenomenon that has been confirmed by several groups in different parts of the world.^{41 - 44} This also held true for patients receiving HAART. However, interferon alfa not only induces clearance of the flavivirus HCV but also clearance of the flavivirus GBV-C. A recent study by Williams et al⁴⁵ showed that the patients with the worst outcomes were those who lost GBV-C during follow-up. None of the 3 studies in coinfected patients reported prevalence of GBV-C infection in their cohorts or eradication rates of GBV-C due to peginterferon plus ribavirin anti-HCV treatment. Careful follow-up studies are needed to address whether clearance of GBV-C by interferon-based therapies may worsen the outcome of HIV infection in these patients.

In summary, the study by Carrat et al adds significantly to current knowledge of treatment of hepatitis C in HIV-infected patients. Drugs are now available that can lead to a reasonable proportion of patients being able to clear HCV. However, responses are still low and most patients infected with HCV genotype 1 do not show long-term benefits from treatment. In addition, many patients have contraindications to treatment. If treatment is initiated, patients should be selected carefully and must be monitored closely. Thus, alternative treatment options are still urgently needed for this difficult-to-treat cohort. The introduction of peginterferon alfa plus ribavirin for hepatitis C in HIV-infected patients represents a major step—but it is not the final goal by far. Additional therapies for HCV in HIV-infected patients will be needed, but likely will remain interferon-based. Approaches to improve adherence to the full dose of treatment will be essential to improve outcomes in coinfected patients.