Anti-TNF-α therapy for rheumatoid arthritis among patients with chronic hepatitis B infection
- 1Brighton and Sussex University Hospitals NHS Trust, Brighton, East Sussex and 2Brighton and Sussex Medical School, Euan Keat Education Centre, Princess Royal Hospital, Haywards Heath, West Sussex, UK
- Correspondence to: Karen Walker-Bone, Euan Keat Education Centre, Princess Royal Hospital, Lewes Road, Haywards Heath, West Sussex RH16 4EX, UK. E-mail: k.walker-bone{at}bsms.ac.uk
- Accepted January 6, 2009.
Sir, The safety and efficacy of anti-TNF-α therapy for inflammatory arthritis among patients infected with HBV are not established. There is evidence from experimental models that TNF-α has a critical role in the clearance of HBV from infected hepatocytes [1]. High-dose immunosuppression in transplant and oncology patients causes HBV reactivation in up to 72% of chronically infected individuals. This reactivation may be asymptomatic occurring only after withdrawal of immunosuppression (immune reconstitution). There have been no randomized controlled trials using anti-viral therapies to prevent active HBV replication during or after immunosuppression but widespread use of anti-viral drugs could result in the emergence of resistant HBV strains. Published British Society of Rheumatology guidance advocates avoidance of anti-TNF-α ‘until more definitive data are available’. In contrast, however, British Society of Gastroenterology guidelines advise use with caution and frequent monitoring of aminotransferases and viral load during and for 3 months after cessation of therapy and consideration of prophylactic or early intervention strategies with nucleoside analogues if reactivation of viral replication occurs [2].
We present a case of severe RA in a chronic HBV-infected individual treated successfully and safely with anti-TNF-α alone using an early intervention monitoring schedule and not requiring anti-viral treatment.
A 63-year-old Caucasian female with a 4-year history of seropositive erosive RA and chronic HBV infection failed adequate therapeutic trials of SSZ and MTX and developed abnormal liver function tests (LFTs) with AZA. She had persistently high disease activity scores (DAS28 7.82 and 8.17 1 month apart) and was requiring oral prednisolone 10 mg daily. HBV serology confirmed positive HBsAg, anti-HBc and anti-HBe, negative HBeAg and an HBV DNA <100 IU/ml. Anti-CCP antibodies were positive (titres of 7420 IU/ml, normal range <15 IU/ml). Following gastroenterology advice and careful counselling, anti-TNF-α treatment was commenced with etanercept 25 mg twice weekly, MTX 10 mg weekly, prednisolone 10 mg daily and intensive (monthly) monitoring of LFTs (including serum transaminases) and HBV DNA. An initial good response to etanercept, with suppression of inflammatory markers, reduction in steroid requirement and improvement in clinically evident synovitis was not maintained at 6 months. Switching to adalimumab 40 mg fortnightly resulted in excellent clinical response, improvement in disease activity scores and reduction in steroid requirement to zero. At follow-up, after 26 months of anti-TNF-α therapy in total (20 months of adalimumab), the RA remains well-controlled (clinically and serologically) and LFTs completely normal. Viral DNA has been elevated transiently on four occasions (greatest being 285 IU/ml) but settled spontaneously within a month without change in the anti-TNF-α therapy or addition of anti-viral therapy.
To date, 15 cases of patients treated with anti-TNF-α treatment in the context of chronic HBV infection have been published [3–12]. In one case, a patient with adult-onset Still's disease and HBV required liver transplantation for fulminant hepatic failure that developed 10 days after his second infusion of infliximab [9]. However, during this episode, there were no changes in HBV serology or HBV DNA titres so that hepatic failure secondary to the Still's disease or an idiopathic drug reaction could not be excluded as the cause.
Figure 1 summarizes the remaining 14 cases. Overall, 2/14 cases had unfavourable outcomes (in one, active viral replication necessitating cessation of anti-TNF-α, and the other fulminant hepatic failure and death), neither of whom were co-prescribed anti-viral agents. Three patients who were not initially co-prescribed anti-virals developed evidence of viral replication necessitating commencement of lamivudine (GSK, GlaxoSmithKline, Stockley Park West, UK). Nine patients had good outcomes [four of these were co-prescribed lamivudine (GSK) and five were not]. All patients co-prescribed lamivudine 100 mg daily (GSK) had good outcomes but an equal number of patients with good outcomes received anti-TNF-α alone. Infliximab appeared to be associated with the least favourable outcomes.
Published cases of patients receiving anti-TNF therapy in the context of chronic HBV infection.
Successful treatment of active RA with anti-TNF-α in patients chronically infected with HBV is possible and can be safe. Our findings, together with those of others, suggest that chronic HBV infection need not preclude therapy with anti-TNF-α where clinically indicated but an individualized approach is required, including careful counselling, collaboration with colleagues in gastroenterology and intensive monitoring. To date, however, data are only available from case reports and more information is needed from controlled trials about: whether any one anti-TNF-α agent is safer than any other; what risk is presented after cessation of immunosuppression when immune reconstitution may occur; if the risk of emergence of lamivudine resistance in patients receiving prophylactic treatment is at a level that renders this approach unsafe and whether all potential recipients of anti-TNF-α therapy should be screened for evidence of HBV infection, past or present. Finally, it remains to be seen as to what effect anti-TNF-α therapy might have on the risk of hepatic fibrosis, cirrhosis and hepatocellular carcinoma in this high-risk group.
Disclosure statement: The authors have declared no conflicts of interest.
- © The Author 2009. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org
