A case for treating high hepatitis B DNA levels before starting HIV therapy

Abstract

Approximately 6–10% of HIV-infected individuals are co-infected with hepatitis B virus (HBV) [1], conferring a 50% increased risk of mortality [2]. Currently, three licensed anti-HIV drugs, lamivudine, emtricitabine, and tenofovir, also possess anti-HBV activity, and are advocated for antiretroviral therapy (ART) in HIV/HBV co-infection [1]. Despite the overall effectiveness of ART, some patients experience paradoxical worsening, a complication termed ‘immune reconstitution inflammatory syndrome’ (IRIS) [3]. Although IRIS is classically associated with opportunistic infections, IRIS hepatitis associated with HBV has been reported [3–5]. Immune-mediated anti-HBV hepatocyte toxicity is often attenuated in patients with HIV-related immunosuppression [1], but the restoration of HBV-specific immune responses with ART can cause acute hepatitis [4]. We describe a case of immune reconstitution hepatitis after the initiation of therapy active against both HIV and HBV. We propose that the high HBV antigen burden present during immune reconstitution caused this complication, and that HBV treatment before initiating HIV therapy may have prevented IRIS. A 43-year-old man with HIV and chronic HBV infection, a CD4 cell count of 85 cells/μl (10%) and HIV viral load of 41 800 copies/ml, presented after being off ART for 4 years. Initial laboratory evaluation revealed: HBV surface antigen positive, core antibody positive, early antibody positive, albumin 4.3 g/dl, bilirubin 0.3 mg/dl, aspartate aminotransferase (AST) 59 U/l, alanine transaminase (ALT) 82 U/l, creatinine 0.95 mg/dl, and plasma HBV-DNA level of more than 1 000 000 000 copies/ml (Cobas TaqMan HBV analyte-specific reagent; Roche Diagnostic Systems, Branchburg, New Jersey, USA). Previous ART exposure included zidovudine, zalcitabine, nelfinavir, saquinavir, stavudine, lamivudine, and efavirenz. In 2005 upon reinitiating care, tenofovir DF, emtricitabine, abacavir, atazanavir and ritonavir were started. Four weeks after re-starting ART, he was doing well clinically. HIV RNA had dropped (193 copies/ml). Transaminases were unchanged (AST 57 U/l, ALT 84 U/l). At 8 weeks of ART, he presented to the clinic with jaundice, dark urine, and abdominal pain for one week. The laboratory evaluation revealed acute hepatitis (AST 1289 U/l, ALT 2409 U/l, bilirubin 2.2 mg/dl, alkaline phosphatase 292 U/l, gamma-glutamyl transferase 137 U/l, and international normalized ratio of 1.2). He had not taken other medications (acetaminophen level < 1 mg/l) or consumed alcohol. His HBV-DNA level dropped 3.5 logs or more to 334 000 copies/ml. His HIV RNA was undetectable (< 50 copies/ml), and the CD4 cell count had doubled to 152 cells/mcL (14%). Testing for other viral hepatitis aetiologies was negative (hepatitis A, hepatitis C, hepatitis D, cytomegalovirus, Epstein–Barr, herpes simplex). The laboratory results over time are presented (Fig. 1).Fig. 1: Alanine aminotransferase levels, log HBV-DNA level, and log HIV-RNA level presented over time. Treatment regimen and CD4 T-cell count at the time of therapy change are also indicated. Alanine aminotransferase (ALT; left axis), log HBV-DNA level (right axis), and log HIV-RNA level (right axis).HIV ART medications were discontinued, and lamivudine and entecavir were started for HBV treatment. On the basis of the ART history, lamivudine resistance of HIV and HBV was presumed [6], so entecavir was added to optimize HBV viral suppression. His transaminases normalized over subsequent months (AST and ALT both 43 U/l), and complete HBV suppression occurred (< 200 copies/ml). At 3 months, lamivudine and entecavir were stopped, and he resumed the previous ART regimen of tenofovir DF, emtricitabine, abacavir, atazanavir and ritonavir. Over the next 3 months, back on ART, his HIV RNA became undetectable (< 50 copies/ml) with the CD4 T-cell count rising to 143 cells/mcL (12%), whereas his HBV DNA remained suppressed (< 200 copies/ml) with persistently normal hepatic transaminases (AST and ALT both 30 U/l). This patient with HIV/HBV co-infection suffered an acute exacerbation of hepatitis 8 weeks after starting potent therapy against HIV and HBV, despite a rapid virological response. After full suppression of his HBV DNA, he had no further complications upon re-initiation of the identical ART regimen. The steady decline in HBV DNA, lack of evidence of drug toxicity, and timing with immune reconstitution all strongly suggest IRIS as the overriding explanation. The high HBV antigen burden in the setting of improved immunity was probably responsible for his hepatitis. Perhaps full HBV suppression before ART initiation could have prevented IRIS. IRIS risk factors include a decrease in the HIV-RNA load and a low CD4 cell count with an increase on therapy [7]. IRIS is hypothesized to arise from the restoration of pathogen-specific immune responses, so it follows that the antigen burden may predict the risk of IRIS. This is clearly the case with other opportunistic infections, such as tuberculosis and cryptococcus [8,9]. When reported with IRIS, HBV-DNA levels have been in the 80th percentile or greater before ART, and our patient was in the 90th percentile or greater [1,4,5,10]. Entecavir, an effective anti-HBV therapy without overlapping anti-HIV activity, can decrease the HBV antigen burden without developing HIV resistance. Among those with high HBV viral loads, controlling HBV infection with HBV-specific agents, such as entecavir, before initiating anti-HIV therapy may prevent IRIS-induced hepatitis. Acknowledgements The authors would like to thank Dr Winston Cavert, Alice Medley, Margaret Thornton, and Matt Larson for their assistance in the care of this patient.