Viral hepatitis and HIV coinfection q

Transcription

1 Journal of Hepatology 48 (2008) Review Viral hepatitis and HIV coinfection q Mark S. Sulkowski * Johns Hopkins University School of Medicine, 600 rth Wolfe Street, 1830 Building, Room 448, Baltimore, MD 21287, USA Persons at high risk for human immunodeficiency virus (HIV) infection are also likely to be at risk for other infectious pathogens, including hepatitis B virus (HBV) or hepatitis C virus (HCV). These are bloodborne pathogens transmitted through similar routes; for example, via injection drug use (IDU), sexual contact, or from mother to child during pregnancy or birth. In some settings, the prevalence of coinfection with HBV and/or HCV is high. In the context of effective antiretroviral therapy (ART), liver disease has emerged as a major cause of morbidity and mortality in HIV-infected persons. Further, coinfection with viral hepatitis may complicate the delivery of ART by increasing the risk of drug-related hepatoxicity and impacting the selection of specific agents (e.g., those dually active against HIV and HBV). Expert guidelines developed in the United States and Europe recommend screening of all HIV-infected persons for infection with HCV and HBV and appropriate management of those found to be chronically infected. Treatment strategies for HBV infection include the use of nucleos(t)ide analogues with or without anti-hiv activity and/or peginterferon alfa (PegIFN) whereas HCV treatment is limited to the combination of PegIFN and ribavirin (RBV). Current approaches to management of HIVinfected persons coinfected with HBV or HCV are discussed in this review. Ó 2007 Published by Elsevier B.V. on behalf of the European Association for the Study of the Liver. Keywords: HIV; HCV; HBV; Liver 1. Introduction Persons at high risk for human immunodeficiency virus (HIV) infection are also likely to be at risk for other infectious pathogens, including hepatitis B virus (HBV) or hepatitis C virus (HCV). These are bloodborne pathogens transmitted through similar routes; for example, via injection drug use (IDU), sexual contact, or from mother to child during pregnancy or birth Associate Editor: M. Colombo q Dr. Sulkowski declares that he receives research grants or funding/ lecture sponsorships/honoraria for continuing medical education (CME) programs from, or advisor or consultant for Boehringer Ingelheim, Merck, Human Genome Sciences, Valeant, Bristol-Myers- Squibb, Pfizer, Vertex, Roche and Schering, and government grants or research funding (DA and DA-13806) and General Clinical Research Center at the Johns Hopkins Medical Institutions (MOI- RR00052). * Tel.: ; fax: address: msulkowski@jhmi.edu [1]. In some settings, the prevalence of coinfection with HBV and/or HCV is high [2,3]. In the context of effective antiretroviral therapy (ART), liver disease has emerged as a major cause of morbidity and mortality in HIV-infected persons [4 6]. Further, coinfection with viral hepatitis may complicate the delivery of ART by increasing the risk of drug-related hepatoxicity and impacting the selection of specific agents (e.g., those dually active against HIV and HBV) [7]. Expert guidelines developed in the United States and Europe recommend screening of all HIV-infected persons for infection with HCV and HBV and appropriate management of those found to be chronically infected [8 11]. Treatment strategies for HBV infection include the use of nucleos(t)ide analogues with or without anti-hiv activity and/or peginterferon alfa (PegIFN) whereas HCV treatment is limited to the combination of PegIFN and ribavirin (RBV). Current approaches to management of HIIV-infected persons coinfected with HBV or HCV are discussed in this review /$32.00 Ó 2007 Published by Elsevier B.V. on behalf of the European Association for the Study of the Liver. doi: /j.jhep

2 354 M.S. Sulkowski / Journal of Hepatology 48 (2008) Hepatitis C virus infection 2.1. Epidemiology and natural history HCV and HIV have similar modes of transmission but the transmission efficiency of each virus differs. HCV is most efficiently spread through exposure to contaminated blood or blood products, particularly injection drug use (IDU). Rates of vertical and perinatal transmission are relatively low (3 6%), although increased 2-fold when the mother is HIV-infected [12,13]. Sexual transmission of HCV is inefficient and the exact risk related to different types of sexual activity is unknown. However, there is increasing evidence of sexually transmitted HCV in HIV-infected men who have sex with men (MSM). For example, in one cohort, the incidence of HCV infection among HIV seropositive MSMs increased 10-fold after 2000 [14]. Sexually acquired HCV infection has been associated with sexually transmitted diseases and traumatic anal receptive intercourse [15]. Based on the relative efficiency of transmission, the prevalence of HCV coinfection varies depending on the route of HIV transmission, ranging from 10% to 14% among persons reporting high-risk sexual exposure to approximately 85 90% among those reporting IDU [16]. In the United States and Europe, 33% of all HIV-infected persons are HCV infected [2,17,18]. HIV infection exacerbates the natural history of HCV infection [19 22]. HIV-infected patients are less likely to clear hepatitis C viremia following acute infection, have higher HCV RNA loads, and experience more rapid progression of HCV-related liver disease than those without HIV infection [23]. As early as 1993, Eyster and colleagues reported that HCV RNA levels were higher in people with hemophilia who became HIV infected than in those who remained HIV negative, and liver failure occurred exclusively in coinfected patients [19]. Among HCV-positive patients with hemophilia who were prospectively monitored, Goedert and colleagues estimated the 16-year cumulative incidence of end-stage liver disease (ESLD) among men with and without HIV to be 14.0% and 2.6%, respectively [24]. Among those men with coinfection, the ESLD risk increased 8.1-fold with HBV surface antigenemia, 2.1- fold with CD4 cell counts below 200 cells/mm 3, and 1.04-fold per additional year of age. Finally, the effect of HIV on HCV was summarized in a meta-analysis of multiple studies that assessed the correlation between HIV coinfection and the progression of HCV-related liver disease [6]. HIV coinfection was associated with a relative risk of ESLD of 6.14 and a relative risk of cirrhosis of 2.07 when compared with HCV monoinfection (Fig. 1). Studies on the effect of ART on the natural history of chronic HCV disease have been contradictory [25 28]. Fig. 1. Adjusted relative risk of decompensated liver disease or histological cirrhosis in patients with HIV/HCV coinfection compared with patients who have HCV infection alone (adapted from meta-analysis published by Graham and colleagues) [6]. Qurishi et al. reported a lower risk of liver mortality in persons who lived long enough to receive effective ART [29]. However, several prospective studies have not detected a beneficial effect of ART on HCV disease [4,30,31]. In other studies, ART has been associated with hepatic injury (e.g., hepatocellular necrosis and steatosis) [7,32,33]. Indeed, in a cohort of 23,441 HIVinfected patients, Weber et al. observed an increased risk of liver-related mortality with longer ART exposure [4]. Additional research is needed to determine the longterm effect of ART on HCV disease progression. netheless, as a consequence of the high HCV prevalence and accelerated disease progression, HCV-related morbidity and mortality is substantial in HIV-infected persons. In one study, Gebo and coworkers evaluated rates of admission at an urban hospital from 1995 to 2000 among HIV-infected patients and found that admissions for liver-related complications among HCV-positive patients increased nearly 5-fold from 5.4 to 26.7 admissions per 100 person-years during that time [5]. Similarly, among 23,441 HIV-infected rth American and European patients followed in the Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D) study, liver disease was the second leading cause of death, with an incidence of 0.23 cases per 100 personyears follow-up behind HIV/AIDS (0.59 cases per 100 person-years) and ahead of cardiovascular disease (0.14 cases per 100 person-years) (Fig. 2) [4]. These data suggest that HCV-related liver disease will continue to be a major cause of hospital admissions and deaths among HIV-infected persons. The effect of HCV infection on HIV disease progression is less clear. Some studies report impaired immune reconstitution in patients with HIV/HCV co-infection treated with ART compared to those with HIV alone [34]; however, this effect has not been observed in other studies [2,35] and is not likely to be clinically relevant [36]. Patients with underlying viral hepatitis are more likely to experience hepatotoxicity on ART [37]. How-

3 M.S. Sulkowski / Journal of Hepatology 48 (2008) % 35% 30% 25% 20% 15% 10% 5% 0% ever, for most persons, this does not impact ART delivery since the majority (90%) of coinfected patients do not develop severe hepatotoxicity. In some studies, the risk of ART-associated hepatotoxicity is greater in coinfected persons with advanced hepatic fibrosis, suggesting that liver disease staging prior to initiating ART may be useful to stratify the risk of hepatotoxicity [38]. Interestingly, effective treatment of HCV infection has been associated with reduced risk of ART-associated liver injury [10] Diagnosis AIDS Liver Cardiac Cancer Other Fig. 2. Causes of death in 23,441, HIV-infected adults followed prospectively in the Data Collection on Adverse Events of Anti-HIV Drugs (D:A:D) study [4]. All HIV-infected patients should be tested for evidence of chronic HCV infection [8,10]. Initial testing for HCV should be performed using sensitive immunoassays (3rd generation EIA) licensed for detection of anti-hcv. False negative anti-hcv EIA results may occur in HIV-infected persons with advanced immunosupression (CD4 < 100/mm 3 ) and true negative EIA are common in the setting of acute HCV infection (<12 weeks following acquisition) prior to seroconversion. If serologic test results are negative and HCV infection is suspected due to elevated liver enzyme levels or risk factors such as IDU or high-risk sex, HCV RNA testing should be performed. To confirm the presence of chronic infection, HCV seropositive persons should be tested for HCV RNA using a qualitative or quantitative assay. Quantitative HCV RNA level (i.e., viral load) does not correlate with degree of liver damage and does not serve as a surrogate for measuring disease severity, but it does provide important prognostic information about the response to antiviral therapy. Serial quantitative HCV RNA testing should be limited to persons receiving HCV treatment. While a single detectable HCV RNA result is sufficient to confirm the diagnosis of active HCV infection, a single negative result cannot exclude active viremia because RNA levels might transiently decline below the limit of detection; repeat testing should be performed. HCV genotype testing should be performed in all HIV-infected persons considering HCV treatment because it is best predictor of HCV response to treatment and may influence the decision to treat and/or perform liver biopsy. Repeat HCV genotype testing is rarely indicated. Serum alanine and aspartate aminotransferase (ALT and AST) levels often fluctuate and prolonged periods of normal serum liver enzyme levels may be observed. Although higher serum ALT and AST levels are clearly predictive of more rapid disease progression, significant liver disease may be present even in the case of persistently normal ALT levels [39]. Numerous imaging modalities are available to evaluate liver parenchymal changes including ultrasonography and CT or MRI scans. However, these tests have limited utility in staging liver disease as they are often abnormal only in advanced disease. Ultrasonography may be used as the initial test for evaluation of cirrhosis, especially if suspected based on physical examination or laboratory testing, or for preliminary detection of hepatic mass lesions suspicious for hepatocellular carcinoma. Due to cost, the use of CT with contrast or MRI scanning should generally be limited to evaluation of hepatic mass lesions among patients with cirrhosis. Liver biopsy remains the preferred test for evaluation of HCV-related disease (fibrosis stage) and is useful to assess prognosis and guide HCV treatment decisions. While ultrasound guidance reduces the risk, complications of percutaneous liver biopsy (i.e., hemorrhage, bile peritonitis, and pneumothorax) occur at rates of 1 3 per 1000 cases [40]. Further, disease staging by biopsy is expensive and subject to sampling error due to the heterogeneity of hepatic fibrosis. In some studies, disease progression has been observed over relatively short time periods in persons with minimal fibrosis. Therefore, while helpful in the evaluation of coinfected persons, liver biopsy is not required prior to the initiation of anti-hcv therapy, particularly in patients with high probability of responding to therapy. n-invasive testing strategies to evaluate liver fibrosis are an area of active research. Currently a number of tests are available which can reliably separate patients with minimal fibrosis from those with cirrhosis but fail to clearly distinguish intermediate stages of fibrotic disease progression [41 45]. While some tests are proprietary commercial assays (e.g., HCV FibroSUREä, Laboratory Corporation of America Ò Holdings), others are based on laboratory tests which are routinely obtained in most HIV-infected persons (FIB-4: age, ALT, AST, and platelet count; APRI: AST-Platelet Ratio Index) (Table 1). Additional studies indicate non-invasive liver stiffness measurements using transient elastography as a surrogate for liver fibrosis [46]. Prospective studies are ongoing to determine which test or combinations of

4 356 M.S. Sulkowski / Journal of Hepatology 48 (2008) Table 1 ninvasive markers of HCV-related liver disease Marker Components Model Sensitivity/specificity for advanced fibrosis (%) AST/ALT ratio AST/ALT Ratio > 1 53/100 Forns test APRI FibroSURE/ FibroTest Fibrospect FibroScan Platelets GGT Cholesterol AST Platelets GGT Haptoglobin Bilirubin Apolipoprotein A Alpha-2-macroglobulin Hyaluronic acid TIMP-1 Alpha-2-macroglobulin Transient elastography pulse-echo ultrasound acquisitions to measure liver stiffness , ln(platelet count [10 9 cells/l]) ln(ggt [UI/L]) ln(age [yrs]) (cholesterol [mg/dl]) (AST level upper limit of normal)/ (platelet count [10 9 cells/l]) /51 41/95 Proprietary 87/59 Proprietary 83/66 N/A 88/74 ALT, alanine aminotransferase; APRI, AST-to-platelet ratio index; AST, aspartate aminotransferase; GGT, c-glutamyl-transpeptidase; N/A, not applicable; TIMP-1, tissue inhibitor of metalloproteinase 1. tests provide the best predictive value for disease progression Treatment Treatment guidelines endorsed by the National Institutes of Health, the US Public Health Service, the American Association for the Study of Liver Diseases, the Infectious Diseases Society of America, and the European Consensus Conference Panel state that HCV should be treated in the HIV-infected person [8 10,47]. However, current HCV therapies are associated with significant toxicity and have limited efficacy; accordingly, treatment should be provided to patients whose risk of liver disease is judged to outweigh the risk of morbidity due to the adverse effects of therapy, and who are most likely to respond to treatment (Table 2). The goals of HCV therapy include eradication of HCV infection, prevention of hepatic fibrosis progression and, among persons with HCV-related cirrhosis, prevention of ESLD, hepatocellular carcinoma, and death. Although viral eradication is not anticipated in most treated persons, histologic and clinical benefits of therapy have been observed in the absence of virologic response [48]. On the basis of well-designed, randomized controlled trials, peginterferon plus ribavirin is the recommended treatment for hepatitis C in HIV-infected persons with observed sustained virologic response rates of 14 36% for HCV genotype 1 infection and 43 73% for HCV genotype 2 and 3 infection (see Table 3) [48 51]. Fixed-dose RBV (800 mg/day) is recommended for HIV-infected persons with genotype 2 or 3. However, among HIV seronegative persons with genotype 1, Peg- IFN plus weight-based RBV (1000 mg/day for persons weighing <75 kg; 1200 mg for persons weighing P75 kg) was more effective than fixed-dose ribavirin [52]. While the efficacy of weight-based ribavirin in HIV-infected persons has not been established in randomized controlled trials, studies indicate that this strategy is not associated with increased risk of adverse effects (e.g., anemia) in coinfected persons and may be associated with improved viral response [49,53]. Accordingly, most experts recommend the use of weight-based RBV in HIV-infected persons with HCV genotype 1. Because the efficacy of shorter treatment duration has not been adequately evaluated in HIV-infected persons, the recommended duration of treatment is 48 weeks duration for infections with all HIV-infected persons, including those with HCV genotype, 2 or 3. In the absence of contraindications (see below), HCV treatment should be routinely offered to persons in whom the potential benefits of therapy are judged to outweigh the potential risks including (but not limited) to persons with: HCV genotype 2 or 3 infection; HCV genotype 1 infection with low HCV RNA level (<400,000 IU/mL); significant hepatic fibrosis (bridging fibrosis or cirrhosis); stable HIV infection not requiring antiretroviral therapy who are motivated to undergo therapy; acute HCV infection (<6 months duration);

5 M.S. Sulkowski / Journal of Hepatology 48 (2008) Table 2 Recommendations for the management of HCV in HIV-infected patients [8] Recommendation Anti-HCV testing should be performed in all HIV-infected patients HCV RNA testing should be performed to confirm HCV infection in HIV-infected patients who are seropositive for anti-hcv, as well as in those who are seronegative and have evidence of unexplained liver disease HCV should be treated in the HIV/HCV-coinfected patient in whom the likelihood of serious liver disease and a treatment response are judged to outweigh the risk of morbidity from the adverse effects of therapy Initial treatment of HCV in most HIV-infected patients consists of peginterferon alfa plus ribavirin for 48 weeks Given the high likelihood of adverse events, HIV/HCV-coinfected patients on HCV treatment should be monitored closely Ribavirin should be used with caution in patients with limited myeloid reserves and in those taking ZDV or d4t. When possible, patients receiving ddi should be switched to an equivalent antiretroviral before beginning ribavirin HIV-infected patients with decompensated liver disease may be candidates for orthotopic liver transplantation cryoglobulinemic vasculitis; and/or cryoglobulinemic membranoproliferative glomerulonephritis. Conversely, treatment with PegIFN/RBV should be routinely withheld from persons in whom the potential risks of therapy are judged to outweigh the potential benefits including (but not limited to) persons with: pregnancy or who are not willing to use birth control; advanced AIDS with insufficient suppression of HIV replication. HCV treatment may be considered following the successful implementation of ART; hepatic decompensation (e.g., coagulopathy, hyperbilirubinemia, encephalopathy, ascites). Liver transplantation, where feasible, should be the primary treatment option for patients with decompensated liver disease; severe, uncontrolled comorbid medical conditions (e.g., cancer or cardiopulmonary disease); severe, active depression with suicidal ideation. HCV treatment may be considered following the successful implementation of psychiatric care; significant hematologic abnormality (e.g, hemoglobin <10.5 g/dl, absolute neutrophil count <1000/mm 3, platelet count <50,000/mm 3 ). HCV treatment may be considered following the correction of hematologic abnormalities (e.g., treatment of underlying causative conditions and/or use of hematopoietic growth factors); significant renal insufficiency. In such persons, HCV treatment with peginterferon alone may be considered; sarcoidosis due to increased risk of severe disease exacerbation with interferon therapy; active, uncontrolled autoimmune conditions (e.g., systemic lupus erythematosus or rheumatoid arthritis) due to increased risk of severe disease exacerbation with interferon therapy. Data from randomized controlled trials indicate that the pre-treatment CD4 cell count is not strongly associated with SVR. However, the efficacy and safety of Peg- IFN/RBV in persons with CD4 cell counts <200 cells/ ll has not been established. Therefore, for HIVinfected patients with CD4 cell counts <200 cells/ll, initiation of ART should be considered before HCV treatment. For persons who have had to terminate or could not initiate antiretroviral therapy because of hepatotoxicity, limited evidence supports HCV treatment in order to permit safe restart of ART, including in persons with CD4 cell count <200 cells/ll. As liver disease progression is likely to be slower among co-infected patients with no or minimal fibrosis or inflammatory changes on liver biopsy, these patients might not need HCV treatment and should be monitored periodically with serial determinations of ALT and repeat liver biopsy. In patients for whom HCV treatment is deferred, the most appropriate intervals to monitor such patients have not been determined, but because of unpredictable fibrosis progression, even among those with limited fibrosis, serial liver biopsy should be considered every 2 3 years [10]. Patients with contraindications to the use of ribavirin (e.g., unstable cardiopulmonary disease, pre-existing anemia unresponsive to erythropoietin, renal failure, or hemoglobinopathy) can be treated with PegIFN monotherapy. However, decreased SVR rates are expected among patients not receiving ribavirin. Additionally, contraindications to HCV treatment may be dynamic; persons with modifiable (e.g., active depression) contraindications to HCV treatment should be reassessed at regular intervals to assess candidacy for therapy. Active injection drug use does not represent a contraindication to HCV treatment; treatment of active IDUs should be considered on a case-by-case basis taking into account comorbid conditions, adherence to medical care and risk of re-infection. Management of HCV-infected IDUs is enhanced by linking IDUs to drug-treatment programs. Alcohol use negatively impacts HCV disease progression and treatment; therefore, alcohol abstinence is strongly recommended before and during antiviral therapy. A history of alcohol abuse is not a contraindication to therapy. Lack of early virologic response at Week 12 of therapy is highly predictive of virologic failure among HIV/HCV-coinfected patients. In the APRICOT study, SVR was observed in less than 2% of patients who did not achieve at least a 2 log 10 IU/mL reduction in HCV RNA from baseline or undetectable HCV RNA levels after 12 weeks of treatment (negative predictive value: %) [50]. Accordingly, current guidelines

6 358 M.S. Sulkowski / Journal of Hepatology 48 (2008) Table 3 Comparison of 5 randomized controlled trials of peginterferon plus ribavirin in HIV/HCV-coinfected patients Parameter APRICOT a [50] ACTG A5071 b [48] ANRS HC02 Laguno et al. [110] PRESCO d [49] RIBAVIC c [51] Number of subjects Country regimen Multinational United States France Spain Spain Peginterferon PegIFN alfa-2a 180 lg/wk PegIFN alfa-2a 180 lg/wk PegIFN alfa-2b 1.5 lg/kg/wk PegIFN alfa-2b 100 or150 lg/wk (weight-based) PegIFN alfa-2a 180 mcg/wk Ribavirin 800 mg/day Dose escalated from 600 to 800 to 1000 mg/day at 4-week intervals 800 mg/day mg/day (weight-based) 1000 mg/d (weight <75 kg) or 1200 mg/d (weight > 75 kg) Duration 48 wks 48 wks 48 wks 48 wks e 48 or 72 wks Baseline characteristics White (%) Mean or median CD4 cell count (cells/mm 3 ) (all subjects had CD4 >300) Undetectable HIV RNA (%) On ART f (%) Bridging fibrosis or cirrhosis (%) (cirrhosis) NA Genotype 1 (%) High level of HCV RNA (%) 72 (>800,000 IU/mL) 83 (>1,000,000 IU/mL) 47 (>800,000 IU/mL) 66.6 (>500,000 IU/mL) SVR by genotype (%) and and a APRICOT, AIDS Pegasys Ribavirin International Coinfection Trial. b ACTG, AIDS Clinical Trials Group. c ANRS; National AIDS Research Agency (France). d PRESCO, Peginterferon Ribavirin España Coinfection; ART, antiretroviral therapy. e 24 wks for HCV genotype 2 or 3 and HCV RNA <800,000 IU/mL. f Didanosine was excluded in the PRESCO trial but not in other studies. indicate that HCV treatment should be discontinued among HIV-infected patients who fail to achieve an early virologic response at 12 weeks of treatment to avoid exposing patients to the risks of treatment when they are very unlikely to achieve SVR. Recently, Payan and colleagues reported that the failure to suppress serum HCV RNA levels below 460,000 IU/mL after 4 weeks of therapy was associated with a negative predictive value of 100% among HIV-infected subjects enrolled in the RIBAVIC study [54]. These data suggest that clinical decisions regarding the continuation of therapy may be made earlier in some coinfected persons, particularly those experiencing significant toxicity. Adverse events are common among HCV/HIV-coinfected patients receiving peginterferon alfa plus ribavirin. Approximately 12 25% of coinfected patients in clinical trials discontinued therapy early because of an adverse event, and serious adverse events occurred in 17 29%. The most common adverse effects of HCV therapy include fatigue, depression, irritability, insomnia, and weight loss. Didanosine and zidovudine have been found to have important interactions with peginterferon and ribavirin. Didanosine has been associated with severe mitochondrial toxicity leading to pancreatitis, hepatic failure, and death, particularly among patients taking ribavirin [55]. As such, didanosine is contraindicated in patients receiving HCV treatment. Concomitant use of zidovudine and HCV therapy has been associated with higher rates of anemia, ribavirin dose reductions, and use of epoetin alfa compared with rates seen in patients not taking zidovudine [53,56]. If other antiretroviral regimens are available (based on tolerability and HIV resistance), discontinuation of zidovudine prior to PegIFN/RBV should be considered; if zidovudine is continued, hemoglobin levels should be monitored closely to detect significant anemia. t unexpectedly, in light of the relatively low HCV eradication rates and high rates of toxicity observed in carefully selected clinical trial subjects, the effectiveness of HCV therapy has been even more modest in clinical practice. In one urban setting with a high prevalence of HCV genotype 1 coinfection, African American ethnicity, and polysubstance abuse, the rate of referral for hepatitis C care and initiation of HCV treatment was

7 M.S. Sulkowski / Journal of Hepatology 48 (2008) relatively low [57]. Although the SVR rate among those initiating HCV treatment with standard IFN or PegIFN plus RBV was comparable (21%) to that observed in controlled trials, the overall effectiveness in this population was low, as less than 1% of the full cohort of HIV/ HCV-coinfected patients receiving HIV care were treated and achieved an SVR. These and similar data from other settings suggest that new paradigms for HCV care are urgently needed in the United States and beyond paradigms that incorporate patient and provider education, as well as case management to address competing patient needs, such as substance abuse, psychiatric illness, and HIV infection. These data also highlight the need for novel HCV therapies with improved efficacy and safety for coinfected patients Prevention vaccine is available for the prevention of HCV infection. The primary route of HCV transmission is drug injection via a syringe previously used by an infected person. An increased frequency of injection, a longer duration of injection drug use, and cocaine use are additional factors that increase the potential for HCV transmission [58]. Besides sharing syringes, other factors associated with injection, such as sharing drug solution containers, cookers, filters, cottons, and mixing water, also increase the likelihood of HCV transmission [59]. Considerable effort should be extended to encourage IDUs to stop using injection drugs preferably by entering a substance abuse treatment program. If IDUs are unwilling or unable to discontinue the use of injection drugs, elimination of sharing needles and drug preparation equipment should be emphasized to reduce the transmission of HCV infection [60]. Although efficiency of sexual transmission of HCV is relatively low, safe-sex practices should be encouraged for all HIVinfected persons, particularly men who have sex with men; barrier precautions (e.g., latex condoms) are recommended to reduce the risk for exposure to sexually transmitted pathogens, including HCV. 3. Hepatitis B virus infection 3.1. Epidemiology and natural history HBV can be transmitted by sexual intercourse, percutaneous exposure, or from mother to infant. Among persons coinfected with HIV in the United States and Europe, HBV is most often transmitted by sexual intercourse (both heterosexual and between men), followed by IDU [61,62]. In Asia and sub-saharan Africa, HBV is principally transmitted from mother to infant or during early childhood. Because the routes of transmission of HIV and HBV are similar, there is evidence of prior HBV infection in approximately 90% of HIV-infected persons whereas chronic HBV infection, indicated by reactive hepatitis B surface antigen (HBsAg), is detected in 5 15% of HIV-infected persons globally [63]. The outcome of HBV infection varies according to age at acquisition and the immune status of the host. HBV infection persists in 50 90% of persons infected at birth or early childhood. In contrast, among HIVuninfected adults, fewer than 5% of HBV infections become chronic [64]. HIV infection is associated with the failure to seroconvert following acute infection and a greater risk of developing chronic hepatitis B infection compared to HIV seronegative persons [65,66] Occult HBV infection Detection of HBV DNA in the serum in the absence of HBsAg has been reported in persons with HIV infection, particularly those with anti-hbc IgG alone [67]. However, estimates of the prevalence of this occult HBV infection vary considerably. Shire et al. reported that 10% of persons with isolated anti-hbc IgG had evidence of HBV DNA in the blood. On the other hand, Tsui and colleagues found that only 2% of 400 HIVinfected women with anti-hbc IgG had detectable HBV DNA and Rodriguez-Torres et al. found no occult HBV infection among HCV/HIV coinfected persons enrolled in the APRICOT study [68,69]. Further, in several studies, occult HBV infection has not been associated with elevation in serum ALT and the clinical significance in HIV-infected persons is uncertain. Based on current evidence, routine surveillance for HBV DNA in the absence of detectable HBsAg is not indicated. Among HIV-infected persons with chronic HBV infection (e.g., persistent detection of HBsAg), progressive liver fibrosis, cirrhosis, ESLD and hepatocellular carcinoma can occur. Among HIV-uninfected persons, progression to HCC and cirrhosis is more common among HBsAg positive persons with male gender, older age, cigarette smoking, alcohol consumption, elevated serum ALT level and higher levels of HBV DNA (>10,000 copies/ml). In a prospective cohort study of 3653 Taiwanese patients aged years who were HBsAg-positive (and HIV-negative), an elevated serum HBV DNA level (>10,000 copies/ml) was the strongest predictor for the development of cirrhosis and hepatocellular carcinoma independent of HBeAg status or serum ALT levels [70,71]. These data suggest that the level of HBV replication appears to be closely linked to the development of liver disease and represents an important prognostic marker. Interestingly, persons with HIV coinfection have higher levels of hepatitis B viremia compared to those with HBV infection alone [72]. The natural history of HBV infection is modified by HIV infection, which can result in higher rates of HBV persistence (HBsAg, HBeAg, and HBV DNA

8 360 M.S. Sulkowski / Journal of Hepatology 48 (2008) detection), HBV relapse (reemergence of HBsAg, HBeAg, or HBV DNA) and clinically significant disease [65,73 75]. Among those with persistent HBV infection, the risk of liver-related morbidity and mortality substantially increased in persons with HIV infection compared to those with HBV alone. Among HIV infected and uninfected persons followed a long-term cohort study, Thio and coworkers reported that the liverrelated mortality rate was higher in men with HIV-1 and HBsAg (14.2/1000) than in those with only HIV-1 infection (1.7/1000, P <.001) or only HBsAg (0.8/ 1000, P <.001) (Fig. 3) [76]. In coinfected individuals, the liver-related mortality rate was highest with lower nadir CD4 cell counts and was twice as high after 1996, when highly active antiretroviral therapy was introduced. The effect of antiretroviral-related immune restoration has been associated with spontaneous recovery from chronic HBV infection but, in other studies, with flares of hepatitis B. More recently, Puoti and coworkers observed that exposure to lamivudine containing ART was associated with decreased risk of liver-related death in 2000 HBV/HIV coinfected persons [77]. Similarly, Miailhes and colleagues reported that 12 of 82 coinfected persons who received ART experienced seroconversion to anti-hbe and/or anti-hbs [78]. While there are emerging data to suggest that antiretroviral regimens that contain drugs active against HBV infection (e.g., tenofovir, emtricitabine, and lamivudine) may modify the natural history of HBV disease in HIV-infected persons by slowing disease progression and, in some patients, leading to seroconversion, additional longterm follow-up is needed to evaluate the effect of dual treatment. The effects of HBV infection on HIV natural history are less apparent. Among 9802 European HIV-infected patients, chronic HBV infection was not associated with progression to AIDS or to viral or immunological Liver Mortality per 1000 person-years HIV -/HBsAg - HIV + HBsAg + HBsAg +/HIV + Fig. 3. Comparison of liver-related mortality by HIV and HBsAg status among 5293 men followed in the Multicenter AIDS Cohort Study (MACS) [76]. response to ART [79]. However, chronic HBV infection may complicate the administration of ART by increasing the risk of moderate or severe liver enzyme elevations [37,80]. In addition, the discontinuation of antiretroviral regimens containing anti-hbv active drugs has been associated with clinically significant liver disease due to the emergence of HBV replication. Accordingly, the discontinuation of anti-hbv active ART should be avoided in HBV coinfected persons; if discontinuation is necessary, coinfected persons should be closely monitored for HBV disease exacerbation [47] Diagnosis Ongoing HBV infection is diagnosed by the detection of HBsAg and HBV DNA in blood. Chronic hepatitis B infection is characterized by the presence of HBsAg with or without HBeAg. During the course of an infection, the loss of HBeAg and development of anti-hbe are usually associated with a decrease in serum HBV viral load, and are associated with a favorable prognosis. However, the loss of HBeAg may also reflect the emergence of HBeAg-negative HBV with precore and core promoter mutations that alter the normal HBeAg synthesis. In this situation, HBV replication remains active as indicated by detectable HBV DNA. Thus, all persons with detectable HBsAg should be assessed for evidence of active HBV replication by HBV DNA assays independent of HBeAg status. All persons with detectable HbsAg should undergo screening for hepatocellular carcinoma with serum AFP and liver ultrasound at regular intervals (6 12 months) [42,81]. Antibodies to the HBV core can be detected in some persons without HBsAg, HBeAg, or antibodies to these antigens. Isolated HBV core antibody serology occurs frequently in persons engaging in IDU (who are generally also HCV infected) and among both HIV-infected and HIV-uninfected persons [82]. The probability that isolated antibody to the HBV core represents HBV infection (versus a false-positive reaction) is related to the prevalence of HBV infection and the anticore antibody titer. For example, in low-prevalence settings, such as among volunteer blood donors, persons with low titers of antibody to the HBV core without other HBV markers rarely have anamnestic responses to HBV vaccination (detection of antibody to the surface antigen) after a single dose, thus suggesting that these may be false-positive anti-hbc reactions [83]. However, individuals with isolated anti-hbc who are at high risk for HBV infection are more likely to have evidence of prior HBV infection. Given the high risk for HBV infection, it is reasonable to assume that the finding of isolated HBV core antibody in HIV-infected persons represents evidence of past infection. However, controversy exists as to the need to assess HIV-infected persons with isolated antihbc for active HBV replication with assays to detect

9 M.S. Sulkowski / Journal of Hepatology 48 (2008) HBV DNA (i.e., occult HBV). In addition, the need for HBV vaccine in such persons is also unclear and some experts recommend the provision of HBV vaccine to all HIV-infected persons who are HBsAb seronegative independent of anti-hbc status Treatment Several medications are approved for the treatment of chronic HBV: interferon alfa, peginterferon alfa, adefovir, lamivudine, telbivudine and entecavir. In addition, tenofovir and emtricitabine are approved for the treatment of HIV infection and are dually active against HBV (Table 4) Standard or peginterferon alfa Few studies have addressed the efficacy of interferon alfa treatment in patients with HIV/HBV coinfection. The pegylated form of interferon appears to be more effective than standard interferon alfa for the treatment of chronic hepatitis B in HIV-uninfected persons but there are few published data on peginterferon efficacy in HIV/HBV-coinfected patients. One small study found that a sequential approach to HBV infection with the administration of PegIFN alfa-2a prior to starting ART containing tenofovir was tolerable but no data are available regarding efficacy [84]. However, limited data suggest that interferon therapy will be relatively ineffective for the treatment of HBV in HIV-infected persons [85]. Consensus guidelines recommend that Peg- IFN may be considered for the treatment of HBV infection in HIV-infected persons with high CD4 cell count (>350/mm 3 ), low HBV DNA level, HBV genotype A or B, no evidence of cirrhosis and no contraindications to its use [9] Lamivudine Lamivudine is a nucleoside analogue that, in its active triphosphate form, inhibits HBV DNA polymerase and HIV reverse transcriptase. Although HBV DNA levels decrease by an average of 2.7log 10 copies/ml in HIV/ HBV-coinfected persons taking lamivudine for 1 year, the incidence of lamivudine-resistant HBV is approximately 20% per year in HIV-infected persons [86,87]. When lamivudine-resistant variants emerge, HBV DNA levels increase, liver enzyme levels may rise, and the resulting hepatitis can be fatal in a minority of patients. In addition, the bulk of available data suggest that the benefit in preventing progression of liver disease is substantially diminished with the presence of lamivudine-resistant HBV [88]. Thus, the clinical effectiveness of lamivudine monotherapy is limited by the frequent emergence of resistant HBV variants Entecavir Entecavir is a nucleoside analogue and is licensed for the treatment of chronic HBV infection in both HIVinfected and -uninfected persons. Entecavir inhibits all three functions of the HBV polymerase, including base priming, reverse transcription of the negative strand, and synthesis of the positive strand of HBV DNA. The presence of lamivudine-resistant variants causes Table 4 Medications available for the treatment of chronic HBV infection Drug Dose and duration Indicated for chronic HBV in HIV-infected patients a Interferon alfa Peginterferon alfa-2a or alfa-2b Lamivudine Emtricitabine Adefovir Tenofovir Entecavir Telbivudine 5 MU daily or 10 MU three times per week by injection Duration weeks 180 lg weekly by injection or 1.5 mcg/kg weekly Duration 6 12 months 300 mg/day in HIV-positive individuals Optimal duration unknown 200 mg/day Optimal duration unknown 10 mg/day Optimal duration unknown 300 mg/day Optimal duration unknown 0.5 mg/day in lamivudine-naive patients 1.0 mg/day in lamivudine-experienced patients Optimal duration unknown 600 mg/day lamivudine naïve patients Optimal duration unknown MU, million units. a Indicated by FDA for treatment of chronic HBV in HIV-infected persons. b t considered active against HIV at doses used in HBV therapy. Yes Active against HIV and HBV Yes Yes Yes b Yes Yes

10 362 M.S. Sulkowski / Journal of Hepatology 48 (2008) decreased susceptibility to entecavir; thus, the recommended doses are 1 mg in lamivudine-experienced patients and 0.5 mg in lamivudine-naive patients [89]. In a randomized controlled trial of 68 HIV/HBV-coinfected persons with lamivudine-resistant HBV, 24 weeks of entecavir resulted in a 3.66log 10 copies/ml reduction in HBV DNA, which is similar to reductions observed in HBV monoinfection [90]. However, after 48 weeks of entecavir treatment, only 9% of patients achieved suppression of HBV replication below 300 copies/ml. To date, resistance to entecavir has not been reported among patients with wild-type HBV infection. However, entecavir resistance occurred in 7% of HBV-monoinfected patients with lamivudine-resistant HBV who received entecavir for 48 weeks and 39% of those treated for 4 years [91]. Thus, while active against lamivudine resistant variants, entecavir should not be used as monotherapy to treat such patients. Finally, although initial reports indicated that entecavir was not active against HIV, clinical observations of significant reduction of HIV RNA level in 3 coinfected patients receiving entecavir for the treatment of HBV in the absence of treatment for HIV infection led to additional in vitro experiments confirming anti-hiv activity of entecavir and the potency for the selection of drug resistant HIV variants [92]. Due to the risk of HIV resistance, entecavir should only be used in HIV-infected persons receiving concurrent effective antiretroviral therapy Adefovir Adefovir dipivoxil is a nucleotide analogue that, in its active diphosphate form, inhibits DNA polymerase and is licensed for the treatment of chronic hepatitis B in patients with HBV monoinfection and is active against lamivudine-resistant HBV [93 95]. In one study, a total of 35 HIV/HBV-coinfected persons were treated with adefovir for 192 weeks and achieved a substantial reduction in HBV DNA (>4log 10 ) [96]. It is clear from this study and from data in HBV-monoinfected patients that the incidence of clinically evident HBV resistance to adefovir is substantially lower than that to lamivudine. However, prolonged use (5 years) of adefovir in HIV seronegative persons with HBeAg negative HBV infection the cumulative probabilities of resistance were 0%, 3%, 11%, 18% and 29% at 1, 2, 3, 4 and 5 years, respectively [97,98]. Furthermore, the use of adefovir in HIV/ HBV-coinfected patients may theoretically involve a risk of selecting HIV cross-resistance to tenofovir, because adefovir is active against HIV at higher doses than those used in HBV management. netheless, HIV resistance following adefovir treatment has not been reported to date [99]. Accordingly, some experts recommend the use of adefovir in persons who are not receiving concurrent effective ART Tenofovir Tenofovir is a nucleotide analogue approved for the treatment of HIV. This agent is structurally related to adefovir, differing by one methyl group. In vitro, tenofovir s activity against HBV is at least equivalent to that of adefovir, with a similar IC 50 [100]. In a prospective, nonrandomized study involving 53 patients with lamivudine-resistant HBV and high HBV viral load, all 35 patients treated with tenofovir had HBV DNA <10 5 copies/ml at 48 weeks, compared with 44% of 18 patients treated with adefovir (P =.001) [101]. In the setting of HIV/HBV coinfection, where lamivudineresistant HBV is common, tenofovir s activity against HBV has been demonstrated to be non-inferior to that of adefovir. ACTG A5127 was a randomized, placebocontrolled trial in which 52 HIV HBV-coinfected patients, most (74 80%) of whom had previously used lamivudine, received either tenofovir or adefovir [95]. The time-weighted mean change in HBV DNA did not differ significantly between treatment arms at Week 48 by any analysis method used, and tenofovir reached the protocol-defined criteria for non-inferiority to adefovir (Table 5). Accordingly, expert guidelines recommend that HIV treatment regimens for HBV/HIV coinfected persons who require ART should include tenofovir plus emtricitabine or lamivudine [47,81]. However, long-term follow-up is not yet available regarding the safety and efficacy of this approach Emtricitabine Emtricitabine is a nucleoside analogue that, following intracellular phosphorylation, exerts potent inhibition of both HIV and HBV replication [ ]. Among HIV-uninfected person emtricitabine was studied in 248 HBV-infected patients who were naive to HBV nucleoside/nucleotide analogue therapy and received either emtricitabine or placebo [103]. At Week 48, 54% Table 5 Tenofovir vs adefovir for treatment of HBV in individuals coinfected with HIV [95] Study group n HBV DNA DAVG 48, log 10 copies/ml Adefovir arm Tenofovir arm Difference Intent-to-treat population Modified intent-to-treat population As-treated population HBV DNA DAVG 48, time-weighted mean change in HBV DNA at week 48.

11 M.S. Sulkowski / Journal of Hepatology 48 (2008) of patients in the emtricitabine group had HBV DNA <400 copies/ml compared with 2% in the placebo group (P <.001). Resistance mutations emerged in 13% of those treated with emtricitabine. HBV variants resistant to emtricitabine also display decreased sensitivity to lamivudine and entecavir. Due to the high risk of HIV and/or HBV resistance, expert guidelines indicate that, like lamivudine, emtricitabine should not be used in persons not receiving concurrent ART or as anti- HBV monotherapy; emtricitabine should be used in combination with tenofovir [47] Telbivudine Telbivudine is thymidine nucleoside analogue approved for chronic hepatitis B patients which has no activity against HIV [105]. After 52 weeks of therapy, HBeAg positive and negative patients, telbivudine 600 mg/day was associated with a HBV DNA decrease of 6.45log 10 in HBeAg positive and 4.45log 10 in HBeAg negative patients. Viral suppression to undetectable levels was achieved in 60% of HBeAg positive patients randomized to telbivudine compared to only 40% of those receiving lamivudine. Resistant variants emerged in 8.1% (HBeAg-negative) and 21% (HBeAg-positive) of patients taking telbivudine for 52 weeks. HBV variants resistant to telbivudine are also resistant to emtricitabine and lamivudine. Among HIV-infected persons, telbivudine may have a unique role in that current evidence suggests that it can be safely used in persons not taking fully suppressive antiretroviral drugs Combination therapy Current guidelines recommend that most coinfected patients should be treated with two drugs that are active against HBV infection to prevent emergence of drug resistant variants. However, data that support the use of combination HBV therapy in preference to the use of single agents in HIV/HBV-coinfected patients are limited. Dore et al. reported the HBV outcomes among a small number of coinfected clinical trial participants who received efavirenz and lamivudine plus either tenofovir or stavudine [106]. After 48 weeks of therapy, mean reduction in HBV DNA among 6 patients treated with lamivudine alone was 3.0log 10 copies/ml compared with 4.7log 10 copies/ml in 5 patients who received dual therapy (P =.055). Nelson and colleagues from the United Kingdom have reported results from an open-label, randomized trial comparing ART-containing lamivudine alone, tenofovir alone, and lamivudine plus tenofovir for treatment of HBV in HIV-infected individuals [107]. Among 27 patients who were naive to lamivudine at entry, the median reduction in HBV DNA at Week 24 was significantly greater with use of the combination regimen (5.03log 10 copies/ml vs 3.31log 10 copies/ml in the lamivudine monotherapy arm and 4.66log 10 copies/ml in the tenofovir monotherapy arm; P =.045 dual therapy arm vs lamivudine arm). In 32 lamivudine-experienced patients, switching to or adding tenofovir resulted in superior antiviral activity at Week 24 compared with remaining on lamivudine alone (P <.001) Treatment approach Because HIV infection can accelerate progression of HBV-related liver disease, treatment of chronic hepatitis B is generally recommended for most HIV-infected persons with active HBV infection (e.g., elevated serum ALT level and HBV DNA >10,000 IU/mL). However, the best strategy for management of HBV infection has not been fully defined. Among patients with chronic HBV infection with no current indication for HIV treatment (e.g., CD4 cell count >350 cells/mm 3 ), some experts recommend avoiding drugs active against HIV (e.g., emtricitabine, lamivudine, entecavir and tenofovir) and suggest using the use of adefovir, telbivudine and/or peginterferon. However, other experts recommend the institution of a fully suppressive antiretroviral regimen that includes the use of two agents active against HBV. This recommendation is based on the rationale that control of HIV infection may represent an important step in preventing HBV-related liver disease. Therapy for HIV/HBV-coinfected patients for whom HIV treatment is indicted is less controversial; most experts recommend the use of an antiretroviral regimen that includes the use of two agents that are active against HBV (e.g., tenofovir plus emtricitabine or lamivudine). consensus has been reached on the management of coinfected patients with lamivudine-resistant HBV. However, many experts recommend the use of tenofovir plus continuation emtricitabine or lamivudine (despite HBV resistance to these agents) whereas other experts recommend the use of two drugs active against lamivudine resistant HBV such as tenofovir and entecavir Prevention Vaccination and the observance of universal precautions are the chief public health measures for preventing HBV infection. HBV vaccination is indicated for all HIV-infected patients, including those with multiple sexual partners, men who have sex with men, and people engaging in IDU. The vaccine used in most well-resourced nations is a recombinant surface antigen expressed in yeast. When used as licensed (3 doses administered to the intradeltoid muscle), more than 95% of adults develop antibody responses that are considered protective. Post-vaccination antibody testing is recommended 1 2 months after the third vaccine dose for people with an increased risk of exposure. In HIV-infected patients, HBV vaccination appears to be safe, as measured by the change in HIV viral load or the subsequent progression of HIV infection. HBV