The use of combination antiretroviral therapy (ART) that includes a HIV protease inhibitor to treat HIV-infected patients has resulted in unprecedented reductions in plasma HIV RNA and increased blood CD4 T-cell counts  and appears to restore pathogen-specific immune responses . However, it is being increasingly recognized that some patients develop hepatitis after commencing protease inhibitors. It has been argued that the hepatitis results from drug hepatotoxicity [3–5], increased hepatitis C virus (HCV) replication in the case of patients with HCV–HIV coinfection , or restoration of anti-hepatitis B virus (HBV) immune responses in a report of a single patient with HBV infection . We describe three HIV-infected patients who experienced hepatitis following treatment with potent ART and provide evidence that the hepatic inflammation in these cases may be due to the restoration of previously impaired HCV-specific immune responses.
Patients with symptomatic hepatitis associated with an increase of the serum alanine aminotransferase (ALT) concentration to more than 5× upper limit of normal (ULN) were identified from a group of 133 patients with greater than 1 log10 copies/ml decrease in plasma HIV RNA after ART (ART responders) in a retrospective survey of all patients commencing potent ART through the pharmacy of our hospital prior to 1 July 1997 (manuscript submitted). Data on serum ALT concentrations, clinical events, HCV antibodies, and liver biopsies were collected from medical records. Serum antibodies to HCV were detected by Abbott Axsym system HCV version 3 (Abbott Diagnostika, Weisbaden-Delkenheim, Germany) and positive samples confirmed using Murex anti-HCV enzyme immunoassay (version III) (Murex Diagnostics, Temple Hill, Dartford, Kent). HCV RNA assays were undertaken on plasma samples stored at −20°C. The Roche Amplicor HCV assay or Roche Amplicor HCV Monitor assay (Roche Diagnostic Systems Inc., Branchburg, New Jersey, USA) were used for qualitative or quantitative determinations of HCV RNA, respectively.
Three (2%) of the 133 ART responders developed symptomatic hepatitis and an increase in serum ALT to above 5× ULN.
A 36-year-old HIV-positive man with a CD4 T-cell count of 32 × 106/l and plasma HIV RNA concentration of 5.45 log10 copies/ml was started on stavudine 40 mg twice daily, lamivudine 150 mg twice daily, and ritonavir 600 mg twice daily. He had a history of Pneumocystis carinii pneumonia (PCP), microsporidial enteritis, recurrent herpes simplex virus type 2 eruptions and molluscum contagiosum since diagnosis of HIV infection in 1986. There was no previous evidence of HCV or HBV infection by routine serological screening (HCV IgG antibody, HBV surface antigen, HBV surface and core antibody) on three separate occasions between 1993 and 1997. There was however asymptomatic, low-grade elevation of the serum ALT concentration to 58 IU/l (reference range, 1–36 IU/l). Within 4 weeks of commencing ART, there was a greater than 2.8 log10 reduction in HIV RNA to below 2.6 log10 copies/ml and a rise in the CD4 T-cell count to 138 × 106/l. Over the ensuing weeks the patient experienced night sweats, anorexia, jaundice and tender hepatomegaly. Serum ALT began to rise at 8 weeks, reaching a peak at 24 weeks of 301 IU/l, representing a fourfold increase from baseline (Fig. 1). In addition, serum γ-glutamyl transpeptidase concentration increased 10-fold with mild elevations of alkaline phosphatase and bilirubin. Ultrasound imaging of the liver showed mild diffuse hepatomegaly, and histopathological examination of a liver biopsy specimen demonstrated chronic hepatitis with periportal inflammation, fatty change and apoptotic hepatocytes. Eosinophilia and granulomata were notably absent, arguing against a drug-related hypersensitivity. Microbiological investigations for hepatitis A virus, HBV, cytomegalovirus, Epstein–Barr virus, Mycobacterium tuberculosis, Mycobacterium avium complex (MAC) and Cryptococcus infections were negative, although repeat testing for HCV IgG antibody proved positive for the first time, indicating seroconversion for HCV antibody. The patient denied any risk behaviours that suggested acute infection with HCV. Analysis of stored historical and current plasma samples in parallel showed that HCV RNA was present up to 2 years prior to the episode of clinical hepatitis, although the patient remained HCV antibody-negative throughout that time (Fig. 1). HCV antibody seroconversion, preceded by hepatitis, only occurred after commencing potent ART and the resultant reduction in HIV RNA and increase in CD4 T-cell count. Retrospective analysis of stored plasma showed a 0.5 log10 copies/ml increase in plasma HCV RNA concentration in the first month after ART (Fig. 1).
Ritonavir, stavudine and lamivudine were ceased with improvement but not complete resolution of transaminitis. Subsequently, plasma HIV RNA concentration increased to 5.76 log10 copies/ml and an alternative ART regimen comprising stavudine 40 mg twice daily, didanosine 200 mg twice daily and indinavir 800 mg three times daily was initiated. However, immunological improvement (CD4 T-cell count rise from 9 to 144 × 106/l) and a greater than 2.8 log10 reduction in HIV RNA to below 2.6 log10 copies/ml was once again accompanied by clinical hepatitis and worsening of transaminitis (Fig. 1). HCV RNA increased by 0.4 log10 copies/ml. The patient was then treated with 3×106 U interferon (IFN)-α three times weekly while continuing ART with some improvement in symptoms and transaminitis in the first 8 weeks. HCV RNA also decreased by 0.5 log10 copies/ml in this time and HIV RNA remained undetectable. Serum ALT subsequently rose again to levels between 400 and 600 IU/l and was associated with symptoms of vomiting, hepatic tenderness and weight loss. Prednisolone 35 mg once daily was then added in an attempt to suppress hepatic inflammation while retaining control of HIV replication by ART and of HCV replication by IFN-α. However, prednisolone failed to control the hepatitis sufficiently and together with indinavir and IFN-α was discontinued after a further 4 weeks. Thereafter, there was improvement of the hepatitis but a concurrent rise in HIV RNA to 5.53 log10 copies/ml (Fig. 1).
A 38-year-old man who had acquired HIV infection from a blood transfusion in India in 1989 had indinavir added to existing ART of lamivudine and zidovudine. He had no past opportunistic infections and was HCV antibody-negative with normal baseline liver function tests. An increase in CD4 T-cell count from 266 to 416 × 106/l and a reduction in HIV RNA of 2 log10 copies/ml was achieved in the first 4 weeks after introduction of indinavir. Serum ALT concentration was measured 4 months later and found to be 188 IU/l. HCV IgG antibody was again undetected. ALT continued to rise reaching a peak of 333 IU/l, and indinavir was therefore ceased 8 months after it had been commenced. Transaminitis improved but persisted for a further 3–4 months. Serum was tested again for HCV antibody and found to be positive for the first time. A liver biopsy undertaken 4 months after ceasing ART demonstrated severe chronic active hepatic inflammation with a lymphocytic infiltrate, bridging necrosis and Councilman bodies. Eosinophilia was absent.
The patient denied exposure to HCV other than the blood transfusion in 1989. Retrospective analysis of the oldest stored plasma available (taken 15 months prior to eventual seroconversion and clinical hepatitis) demonstrated HCV RNA. Thus, this patient had ‘subclinical’ HCV infection, and remained HCV-seronegative for at least 15 months and possibly 7 years until a decline in HIV RNA and increase in CD4 T-cell count was induced by potent ART. The plasma HCV RNA concentration 2 weeks before liver biopsy was 4.8 log10 copies/ml compared with 5.04 log10 copies/ml 14 weeks before commencing ART. There was no significant decrease in the serum ALT or plasma HCV RNA concentration following treatment with 5 × 106 U IFN-α three times weekly for 3 months.
A 30-year-old woman who had acquired HIV and HCV infection from past injecting drug use started indinavir 800 mg three times daily, stavudine 30 mg twice daily and lamivudine 150 mg twice daily. Past opportunistic infections included PCP and oral candidiasis. Prior to starting ART she had positive HCV IgG, positive HCV RNA and chronic low-grade asymptomatic elevation of the serum ALT concentration between 40 and 60 IU/l. She was severely immunodeficient with a CD4 T-cell count of 14 × 106/l and absent cutaneous delayed-type hypersensitivity responses to all antigens on a Cell-Mediated Immunity Multitest panel (Institut Merieux, Lyon, France) as well as to tuberculin and M. avium purified protein derivative (PPD) when tested using the Mantoux method. Within 2 weeks of starting therapy the patient developed high fevers, sweats and palpable splenomegaly. A gallium scan demonstrated increased uptake in mediastinal lymph nodes. Repeat Mantoux testing now showed a strongly positive response of 26 mm induration to M. avium PPD. Mycobacteria were not cultured from multiple blood cultures. The patient stopped ART and was then started on rifabutin, clarithromycin and ethambutol for ART-induced immune restoration to subclinical MAC infection . ART was restarted 1 month later with no recurrence of fever. After 2 months of therapy, the HIV viral load fell from 5.79 log10 copies/ml to below 2.6 log10 copies/ml and her CD4 T-cell count rose from 14 to 154 × 106/l. During this time she developed vomiting, tender hepatomegaly, and her serum ALT concentration rose to 460 IU/l and fluctuated thereafter between 400 and 500 IU/l. A pruritic blistering eruption of cutaneous porphyria was noted on her hands (confirmed by a dermatologist) and red blood cell porphyrins were markedly elevated at 4.6 µmol/l (reference range, 0–1.5 µmol/l), in keeping with this diagnosis. This patient was HCV antibody-positive prior to starting ART, but hepatitis (associated with porphyria) occurred only after ART, at a time of marked immunological recovery.
We have demonstrated in a cohort of unselected HIV-infected patients responding to potent ART that cases of acute symptomatic hepatitis associated with the use of protease inhibitors can occur in patients with HCV infection. However, in two of the three patients described, HCV infection was only recognized after HCV antibody seroconversion (or seroreversion), which presumably resulted from the use of protease inhibitors. It is noteworthy that two of the five previously reported cases of acute hepatitis after taking protease inhibitors [3–5] had HCV infection and one had HBV infection. In the other two cases, HCV infection was excluded by demonstrating absence of HCV antibody in serum which, as demonstrated here, may be an unreliable means of excluding infection.
HCV coinfection occurs in 8–23% of HIV-infected patients . There is a higher incidence of HCV viraemia, higher HCV RNA plasma concentration, greater histological damage and faster progression to cirrhosis in HCV–HIV-coinfected patients than in those with HCV infection alone [10,11]. This suggests that HIV-induced immunodeficiency is associated with impairment of HCV-specific cell-mediated immune responses and hence greater HCV replication and more severe clinical disease. Persistent or ‘false’ HCV seronegativity (as in the first two cases described here) as well as loss of HCV antibody (seroreversion) has been documented in HIV-infected patients, reflecting impairment of immunity .
The pathogenesis of acute hepatitis following the use of protease inhibitors in HCV–HIV-coinfected patients is unclear. The liver histopathology in the first two cases described here and in the patients reported by Vento et al. , and the occurrence of hepatitis with two different protease inhibitors in the first of our cases argue against direct drug hepatotoxicity. An increase in plasma HCV RNA following ART  may reflect increased HCV replication but could also indicate increased turnover of blood mononuclear leukocytes in which HCV can replicate, or increased release of HCV from immune-damaged HCV-infected hepatocytes. The increase in plasma HCV RNA in the first of our patients was relatively small (below 0.5 log10 copies/ml) and HCV RNA was not increased at a time of active hepatitis in the second case. Furthermore, HCV RNA does not increase in patients with a decrease in HIV RNA following treatment with two nucleoside analogues .
Antibody seroconversion associated with chronic asymptomatic HCV infection, as described in these cases, indicates that an HCV-specific immune response can be restored by ART. The generation of specific antibody may serve as a marker for functional restoration of CD4 and CD8 T-cell responses that specifically target HCV-infected hepatocytes and mediate hepatitis. We propose that the same process occurred in the third patient given the temporal coincidence of hepatitis with ART response, the onset of cutaneous porphyria (which has been noted to occur with greater frequency in HCV–HIV-coinfected patients treated with zidovudine ) and evidence of ‘immune restoration’ to MAC infection . The restored immune responses in all three patients are clearly not protective but immunopathological leading to tissue inflammation, perhaps by a similar mechanism to that described in a patient with HBV infection reported by Carr and Cooper .
We suggest that the hepatitis occurring in the HCV–HIV-coinfected patients reported here represented HCV-associated ‘immune restoration disease’ and that this might be the cause of hepatitis following the use of potent ART in other patients with HCV–HIV coinfection. Our observations highlight the fallibility of immunological tests for diagnosing HCV infection in HIV-infected patients and the need to develop and use tests of pathogen-specific immunity in the evaluation of all pathogen-associated disease following ART-induced immune reconstitution.
1. Hammer SM, Squires KE, Hughes MD, et al.
: A controlled trial of two nucleoside analogues plus indinavir in persons with human immunodeficiency virus infection and CD4 cell counts of 200 per cubic millimeter or less
. N Engl J Med
2. Jacobson MA, French M: Altered natural history of AIDS-related opportunistic infections in the era of potent combination antiretroviral therapy
3. Brau N, Leaf HL, Wieczorek RL, et al.
: Severe hepatitis in three AIDS patients treated with indinavir
4. Matsuda J, Gohchi K, Yamanaka M: Severe hepatitis in patients with AIDS and haemophilia B treated with indinavir [letter]
5. Vergis E, Patterson D, Singh N: Indinavir-associated hepatitis in patients with advanced HIV infection
. Int J STD AIDS
6. Vento S, Garofano T, Renzini C, et al.
: Enhancement of hepatitis C virus replication and liver damage in HIV-coinfected patients on antiretroviral combination therapy [letter]
7. Carr A, Cooper D: Restoration of immunity to chronic hepatitis B infection in HIV-infected patient on protease inhibitor
8. French MAH, Mallal SA, Dawkins RL: Zidovudine-induced restoration of cell-mediated immunity to mycobacteria in immunodeficient HIV-infected patients
9. Ockenga J, Tillmann HL, Trautwein C, et al.
: Hepatitis B and C in HIV-infected patients. Prevalence and prognostic value
. J Hepatol
10. Garcia-Samaniego J, Soriano V, Castilla J, et al.
: Influence of hepatitis C virus genotypes and HIV infection on histological severity of chronic hepatitis C. The Hepatitis/HIV Spanish Study Group
. Am J Gastroenterol
11. Soto B, Sanchez-Quijano A, Rodrigo L, et al.
: Human immunodeficiency virus infection modifies the natural history of chronic parenterally-acquired hepatitis C with an unusually rapid progression to cirrhosis
. J Hepatol
12. Chamot E, Hirschel B, Wintsch J, et al.
: Loss of antibodies against hepatitis C virus in HIV-seropositive intravenous drug users
13. Rockstroh K, Albers-Busch U, Schliefer K, et al.
: Antiviral combination therapy decreases HIV load but does not alter hepatitis C replication [abstract]
14. O'Connor WJ, Murphy GM, Darby C, et al.
: Porphyrin abnormalities in acquired immunodeficiency syndrome
. Arch Dermatol