Conflict of interest: Dr Tsai has conducted clinical trials, received honoraria for serving as a consultant or received speaking fees from Wyeth Pharmaceuticals (now Pfizer), Serono International SA (now Merck Serono International), UniPharma/Biogen Idec, Janssen-Cilag Pharmaceuticals and Abbott Pharmaceuticals. Drs Cho, Chen and Chiu have no conflicts of interest to declare.
Use of anti-tumor necrosis factor-α therapy in hepatitis B virus carriers with psoriasis or psoriatic arthritis: A case series in Taiwan
Article first published online: 14 NOV 2011
© 2011 Japanese Dermatological Association
The Journal of Dermatology
Special Issue: Special Issue: Psoriasis (pages 211-289)
Volume 39, Issue 3, pages 269–273, March 2012
How to Cite
CHO, Y.-T., CHEN, C.-H., CHIU, H.-Y. and TSAI, T.-F. (2012), Use of anti-tumor necrosis factor-α therapy in hepatitis B virus carriers with psoriasis or psoriatic arthritis: A case series in Taiwan. The Journal of Dermatology, 39: 269–273. doi: 10.1111/j.1346-8138.2011.01434.x
- Issue published online: 21 FEB 2012
- Article first published online: 14 NOV 2011
- Received 2 August 2011; accepted 27 September 2011.
- hepatitis B virus;
- psoriatic arthritis;
- tumor necrosis factor
The use of anti-tumor necrosis factor (TNF)-α therapy in patients with psoriasis who are hepatitis B virus (HBV) carriers is usually not recommended, and routine antiviral prophylaxis is suggested for those who need the treatment. We report our experience on the safety of anti-TNF-α therapy in patients with psoriasis who are HBV carriers in our clinic using HBV viral load as a guide for HBV treatment. Between 2007 and 2011, seven HBV carriers receiving TNF-α inhibitors for psoriasis in our clinic were collected retrospectively. The HBV viral load and aminotransferase levels were regularly monitored. Two of the seven patients were inactive HBV carriers, and the other five patients had chronic hepatitis B. Only one patient received antiviral agents before the anti-TNF-α treatment. The mean duration of the anti-TNF-α treatment was 26.6 months (range, 14–45 months). These patients were followed up from the start of the anti-TNF-α therapy for a mean duration of 28.9 months (range, 14–45 months). HBV reactivation was observed in three patients, one of whom required antiviral treatment. No HBV reactivation-related hepatitis was observed. In conclusion, prevention of HBV reactivation by monitoring of HBV viral load is cost-effective and may decrease the risk of developing drug resistance from routine anti-HBV prophylaxis treatment. It can be considered as an alternative in psoriasis patients treated by TNF-α inhibitors, especially in areas with a high HBV burden and in hepatitis B e-antigen-negative patients who have a lower risk of viral reactivation.
Tumor necrosis factor (TNF)-α inhibitors have been used increasingly in the treatment of psoriasis and psoriatic arthritis. These drugs show selective immunosuppressive effects without direct hepatic or renal toxicities.1 However, TNF-α is an important pro-inflammatory cytokine in the host defense mechanism against many intracellular pathogens. Inhibition of the TNF-α pathway is thought to increase the susceptibility to various infections, including infections of Mycobacterium and HIV. The safety of TNF-α inhibitors in psoriatic patients with hepatitis B virus (HBV) infection is still controversial. Several case reports revealed HBV reactivation during treatment with TNF-α inhibitors in HBV carriers or chronically infected HBV patients.2–5 On the other hand, increasing evidences show that treatment with TNF-α inhibitors is safe in inactive HBV carriers who receive concomitant antiviral agents6–8 or in patients with resolved HBV infection.8–11 Nevertheless, safety data about anti-TNF-α therapy in HBV carriers are quite limited, especially in those who received anti-TNF-α therapy without concomitant antiviral prophylaxis. In Taiwan, up to 15% of the adult population is known to be HBV carriers. Herein, we summarize our experience of using TNF-α inhibitors in patients with psoriasis who were known to be HBV carriers before the initiation of TNF-α inhibitors.
Seven HBV carriers with psoriasis or psoriatic arthritis treated with anti-TNF-α agents (etanercept or adalimumab) between 2007 and 2011 were collected retrospectively in this study. Baseline data, including prior and concurrent treatment, aspartate aminotransferase (AST)/alanine aminotransferase (ALT) levels, HBV serological marker levels, and HBV DNA viral load were obtained. The patients were required to visit the outpatient clinic at least once a month during the treatment period. Serum AST/ALT levels and HBV viral load were regularly followed up as markers of hepatic injury and disease progression, respectively. Reactivation of HBV was defined as an increase in the HBV viral load to more than 10-fold of the baseline level or an absolute HBV viral load of more than 109 copies/mL during the period of anti-TNF-α therapy. HBV reactivation-related hepatitis was defined as more than threefold increase in the serum ALT level on two consecutive determinations, at least 5 days apart, along with evidence of HBV reactivation during the period of anti-TNF-α therapy.
If reactivation was encountered, patients were referred to a hepatologist for further evaluation and antiviral treatment.
The mean duration of anti-TNF-α treatment was 26.6 months (range, 14–45 months). The patients were followed up from the start of the anti-TNF-α therapy for a mean duration of 28.9 months (range, 14–45 months). All patients were known to be HBV carriers for decades; they were HBV surface antigen (HBsAg)-positive and HBV e-antigen (HBeAg)-negative prior to the beginning of the anti-TNF-α therapy (Table 1). Only one patient used adalimumab, while others received etanercept for their psoriasis. According to the definitions approved by the American Association for the Study of Liver Diseases (AASLD),12 five of the seven patients were HBeAg-negative chronic hepatitis B carriers and the other two patients were inactive HBV carriers (Table 2). One patient (case 6) was HBeAg-positive 2 years before initiation of the anti-TNF-α therapy and had several episodes of HBV reactivation in the past years. First, he received lamivudine treatment for 5 months. Because the results were unsatisfactory, his treatment was changed to entecavir. He had shown HBeAg seroconversion during entecavir treatment. After 13 months of entecavir treatment, he started receiving concomitant anti-TNF-α therapy (etanercept). At that time, he had normal AST/ALT levels and his HBV viral load was low (632 copies/mL). During the follow-up period, no HBV reactivation was detected and his HBV viral load remained constant (775 copies/mL).
|HBV serology||All patients: HBsAg-positive and HBeAg-negative|
|HBV DNA (copies/mL)||1.36 × 104||2100||115||400||Undetectable||632||8.2 × 104|
|HBV DNA (copies/mL)||3.05 × 105||2100||1320||600||2050||775||1.3 × 105|
|HBV carrier state||CHB||Inactive||CHB||CHB||Inactive||CHB||CHB|
|HBV reactivation-related hepatitis||−||−||−||−||−||−||−|
|Chronic hepatitis B|
|Definition||Chronic necroinflammatory disease of the liver caused by persistent infection with hepatitis B virus. Chronic hepatitis B can be subdivided into HBeAg-positive and HBeAg-negative chronic hepatitis B.|
|Diagnostic criteria||HBsAg-positive >6 months|
|Serum HBV DNA >105 copies/mL, lower values (104–105 copies/mL) are often seen in HBeAg-negative chronic hepatitis B|
|Persistent or intermittent elevation in ALT/AST levels|
|Liver biopsy showing chronic hepatitis with moderate or severe necroinflammation|
|Inactive HBsAg carrier state|
|Definition||Persistent HBV infection of the liver without significant ongoing necroinflammatory disease|
|Diagnostic criteria||HBsAg-positive >6 months|
|Serum HBV DNA <104 copies/mL|
|Persistently normal ALT/AST levels|
|Liver biopsy confirms absence of significant hepatitis|
In this case series, none of the patients received antiviral prophylaxis before the initiation of anti-TNF-α therapy except one patient (case 6) who received antiviral treatment as mentioned above. Three patients (cases 1, 3, and 5) fulfilled the criteria for HBV reactivation. Nevertheless, no patients experienced hepatitis flare-up during the anti-TNF-α therapy and the follow-up period. The elevation of the ALT level was usually minimal and acceptable. No other obvious adverse effects were observed among the patients.
Among the patients who showed HBV reactivation, two (cases 1 and 3) were HBeAg-negative chronic hepatitis B carriers and one (case 5) was an inactive HBV carrier. Case 1 had a higher baseline HBV viral load (1.36 × 104 copies/mL) with normal baseline AST/ALT levels. Elevation of the viral load was noticed 1 month after the initiation of etanercept treatment (the viral load increased to >105 copies/mL; Fig. 1). The patient was referred to a hepatologist and started receiving lamivudine treatment 2 months after the beginning of anti-TNF-α therapy. His HBV viral load decreased to 200 copies/mL after 2 months of lamivudine treatment and became undetectable with continued lamivudine and etanercept use for 45 months till now. In addition, his AST and ALT levels were within normal limits during the follow-up period. No evidence of acute exacerbation of viral hepatitis was noticed. The other two patients (cases 3 and 5) with HBV reactivation did not receive antiviral treatment because of persistently low HBV viral loads during the period of anti-TNF-α therapy.
Although case 7 did not meet the criteria for HBV reactivation, we referred the patient to a hepatologist for further evaluation because his HBV DNA level exceeded 105 copies/mL during the treatment period and he had slightly increased AST/ALT levels. He was monitored regularly and frequently because of the high probability of HBV reactivation. This patient will receive antiviral treatment if the criteria for HBV reactivation are met.
Hepatitis B virus infections are a worldwide health issue. Currently, more than 350 million people are chronic HBV carriers worldwide and 75% of them reside in the Asia–Pacific region. The prevalence rate is approximately 8.0–20% in Southeast Asia.13
Development of fulminant and even fatal hepatitis has been reported in HBV carriers treated with chemotherapeutic agents or immunosuppressants. However, the immunosuppressive effects are not equal among different chemotherapy regimens and immunosuppressive agents. The use of glucocorticoids in conjunction with chemotherapy for patients with hematological malignancies, including lymphoma, is believed to pose the greatest risk of reactivation (as high as 67%), either because of direct viral interaction or being secondary to immune modulation. A slightly lower risk of reactivation, approximately 40%, exists among patients receiving chemotherapy for solid tumors.14 For patients with malignancies in inactive HBV carrier state, prophylactic antiviral agents, such as lamivudine, have proven to be beneficial and are recommended before the beginning of chemotherapy; these agents may be used for up to 1 year after completion of chemotherapy.15 However, no such evidences exist in HBV carries receiving anti-TNF-α therapy for their psoriasis or psoriatic arthritis. Meanwhile, increasing evidences suggest that it is safe for patients with resolved hepatitis B infection to receive anti-TNF-α therapy for rheumatological diseases without antiviral prophylaxis.16 Besides, there is a lack of evidence for the benefits of prophylactic antiviral agents in patients with chronic HBV infection with rheumatic diseases, especially patients who are in an inactive HBV carrier state.17,18 For patients with psoriasis or psoriatic arthritis, only several studies examined the usage of anti-TNF-α agents in those with chronic hepatitis B infection.
An increased risk for HBV carriers has been found in patients with psoriasis in Taiwan.19 Nevertheless, the use of anti-TNF-α therapy for HBV carriers with psoriasis or psoriatic arthritis is still controversial. On the basis of several published case reports, some guidelines advise against the use of anti-TNF-α therapy for HBV patients with psoriasis or other rheumatic disease because of the risk of HBV reactivation.20,21 In the meantime, others do not restrict the use of anti-TNF-α therapy in HBV patients without advanced liver disease.22,23 Due to a lack of consensus in managing HBV carriers with psoriasis or psoriatic arthritis, antiviral prophylaxis is usually used in conjunction with anti-TNF-α therapy, based on the suggestions for HBV carriers receiving chemotherapeutic agents for their malignancies as mentioned previously. However, these patients may develop resistance to antiviral agents, especially lamivudine, during long-term use and show subsequent elevation of AST/ALT levels.14 In such cases, a change in the antiviral agent is usually required to control viral replication. The increased frequency of resistance to antiviral agents in treating HBV infection is a great concern, especially in HBV endemic areas. Mutations in HBV polymerase are associated with resistance to antiviral agents. The reported genotypic resistance to lamivudine is approximately 24% at 1 year and 70% at 5 years.24 Therefore, judicious use of these antiviral agents is necessary. In an attempt to find a potential resolution, it is important to select patients who are at greatest risk of HBV reactivation.
A multivariate risk-factor analysis showed that the most significant predictive factor for HBV reactivation is the HBV DNA level before the start of cytotoxic or immunosuppressive therapy.14 Elevation of HBV DNA levels to levels of more than 105 copies/mL is associated with a greater risk of HBV reactivation.25 Therefore, treatment with antiviral agents such as lamivudine is usually suggested in patients with a HBV DNA level of more than 105 copies/mL. Other identified risk factors for HBV reactivation include seropositivity for HBsAg and HBeAg, male sex, young age, precore- or core-promoter HBV mutations, and chemotherapy regimens containing highly myelosuppressive agents, glucocorticoids or rituximab.14
In regard to HBV clearance, two possible mechanisms have been proposed. One mechanism is the cytotoxic T-lymphocyte-mediated apoptosis of virus-infected hepatocytes via Fas/FasL interaction. The other mechanism is the non-cytolytic clearance of HBV from hepatocytes by T-cell-derived cytokines, including γ-interferon and TNF-α.26 The use of anti-TNF-α agents is relatively safer than chemotherapy with or without rituximab because anti-TNF-α agents block only one of the proposed anti-HBV mechanisms in the human body.
In addition, different anti-TNF-α agents may have different impacts on HBV reactivation. To date, most reported patients with HBV reactivation received infliximab.17 Only a few patients with HBV reactivation were treated with etanercept. Infliximab can neutralize either soluble or membrane-bound TNF-α, and as a result, cells bearing TNF-α can be destroyed through complement fixation or cytotoxic killing. Adalimumab functions similarly to infliximab but is a humanized immunoglobulin molecule administrated s.c. Etanercept is distinct from the other TNF-α inhibitors because it binds to soluble TNF-α without effects on membrane-bound TNF-α. These differences may explain the different impacts of these TNF-α inhibitors on HBV reactivation.27 Based on these observations and evidences, most of our patients in this study were chosen to receive etanercept rather than other anti-TNF-α agents.
In patients receiving anti-TNF-α therapy, measurements of baseline HBV DNA and AST/ALT levels as well as regular follow-up assessments of these clinical parameters are recommended for the determination of antiviral prophylaxis or early detection of HBV reactivation. Increased HBV DNA levels during the period of anti-TNF-α therapy indicate increased viral replication that precedes elevation of the AST/ALT levels. HBV viral load might be a more sensitive indicator to monitor HBV reactivation in HBV carriers receiving anti-TNF-α therapy.
In this case series, HBV reactivation occurred in three patients during anti-TNF-α therapy. No patient experienced HBV reactivation-related hepatitis. The rate of viral reactivation was lower than that observed with chemotherapy for treating malignant lymphoma. In addition, only one patient required concomitant antiviral treatment for effective control of HBV reactivation. Thus, we suggest that anti-TNF-α therapy may be used with caution for treating HBV carriers with psoriasis or psoriatic arthritis. Intimate collaborations with hepatologists to manage these patients are recommended. In areas with high HBV burden, monitoring of HBV viral load may be considered an alternative to routine use of HBV prophylactic drug treatment, considering the high costs of the anti-HBV agents and the possibility of development of antiviral resistance during long-term use. However, for patients with higher baseline viral load or positive HBeAg, further large scale studies or randomized trials are needed to confirm our suggestions and routine antiviral prophylaxis is still suggested.
- 6Safety of anti-tumor necrosis factor-alpha agents in psoriasis patients who were chronic hepatitis B carriers: a retrospective report of seven patients and brief review of the literature. J Eur Acad Dermatol Venereol 2011; 25: 471–474., , .
- 11Safety of tumor necrosis factor alpha blockers in hepatitis B virus occult carriers (hepatitis B surface antigen negative/anti-hepatitis B core antigen positive) with rheumatic diseases. Arthritis Care Res (Hoboken) 2010; 62: 749–754., , et al.
- 13Global epidemiology of Hepatitis B virus (HBV) infection. NAJMS 2011; 4: 7–13., .
- 22The 2008 American College of Rheumatology recommendations for the use of nonbiologic and biologic disease-modifying antirheumatic drugs in rheumatoid arthritis: where the rubber meets the road. Arthritis Rheum 2008; 59: 757–759., .