Abstract
Widespread use of immunosuppressive drugs, both conventional disease-modifying antirheumatic drugs (cDMARDs) and biologic disease-modifying antirheumatic drugs (bDMARDs), in autoimmune rheumatic diseases (ARDs) has been found to be associated with the reactivation of underlying latent viruses. The clinical features of virus reactivation can sometimes mimic flare of the underlying ARDs. The correct diagnosis and management of such reactivation is crucial, as increasing the dose of immunosuppressive drugs to treat a presumed flare of underlying ARDs would probably be of no benefit, and it could exert a detrimental effect on the host. This review focused on the effects of immunosuppressive drugs on underlying chronic viral infections, particularly hepatitis B virus, hepatitis C virus, human immunodeficiency virus, varicella zoster virus, Epstein–Barr virus, cytomegalovirus, John Cunningham (JC) virus, Kaposi sarcoma-associated herpesvirus, and human papillomavirus in patients with ARDs. It also covered the effect of interferon-α, which is used to treat chronic hepatitis infection, and the induction of autoimmunity.
Introduction
During the past three decades, understanding of the immunopathogenesis of various autoimmune rheumatic diseases (ARDs) has been increasing. The use of immunosuppressive drugs (IMDs), both conventional immunosuppressive drugs (cIMDs) or conventional disease-modifying antirheumatic drugs (cDMARDs) and biologic agents or biologic disease modifying-antirheumatic drugs (bDMARDs), as well as newer treatment strategies (e.g. treat-to-target and combination therapy), results in not only better control of disease activity but also improved quality of life. However, such advantages of this treatment are not without cost. Treatment with corticosteroids and IMDs has been found to be associated with an increased risk of infections . Furthermore, it is still not clear whether prolonged treatment with these compounds increases the risk of malignancies.
Infection is common among developing countries, and patients with ARDs occasionally have underlying chronic infections. Furthermore, infection involving more than one organism is not uncommon. Many viruses can infect the body latently after primary infection. Thus, the major concern is whether to use IMDs, particularly bDMARDs, in such cases, as they may exacerbate or reactivate underlying latent virus infection and be detrimental to the host. This review focuses on the effects of IMDs on underlying chronic viral infections, particularly hepatitis B virus (HBV), hepatitis C virus (HCV), human immunodeficiency virus (HIV), varicella zoster virus (VZV), Epstein–Barr virus (EBV), cytomegalovirus (CMV), John Cunningham virus (JCV), Kaposi sarcoma-associated herpesvirus, and human papillomavirus (HPV). It also covers the effect of medications used in the treatment of these infections, particularly that of interferon-α (IFNα), and induction of autoimmunity.
Hepatitis B virus
HBV is a double-stranded DNA (dsDNA) virus and belongs to the Hepadnaviridae family. It is one of the most common viral infections worldwide, and it is estimated to affect 400 million people. Its prevalence is low in the USA and northern Europe (<2%), but high in Southeast Asia, China, and Africa (>8%) . Most patients acquire the virus during the perinatal period through vertical transmission, or during early childhood. Primary HBV infection usually is asymptomatic in 30–50% of cases in individuals older than 5 years. Although the virus is cleared in most patients (95%), resulting in lifelong immunity, 30–50% of those who acquire it during childhood develop chronic hepatitis B (CHB) infection .
After infection with HBV, active viral replication in the liver occurs, with a minimal host immune response (high serum HBV DNA levels, presence of HBeAg (HBeAg+) and HBsAg (HBsAg+), normal liver transaminase enzymes, and minimal hepatocyte changes), or the immune tolerance phase is observed. The second or immune clearance phase, which occurs in adolescents and adults, is characterized by a vigorous immune response to HBV, resulting in symptoms of clinical hepatitis with a marked increase in transaminase enzymes, together with hepatocyte necrosis and fibrosis. During this phase, HBV replication declines, and in 90% of cases, HBsAg and HBeAg gradually disappear or HBe seroconversion occurs (HBsAg−/HBeAg−). This stage is accompanied usually by the appearance of antibodies to HBV antigens (anti-HBs+, anti-HBe+, and anti-HBc+). The third or low-replication phase is characterized by the presence of the abovementioned antibodies, and a very low or undetectable HBV DNA (<2000 IU/mL), normal transaminase enzymes, and minimal liver inflammation . The presence of the anti-HBs+ state indicates full immunity from HBV infection (resolved HBV infection). However, some patients might still have a very low or undetectable level of anti-HBs antibodies (anti-HBs−/anti-HBc+), and those with a very low level of HBV DNA (<2000 IU/mL) without evidence of liver inflammation are defined as having an “occult HBV infection.” Nevertheless, a certain percentage of “occult HBV-infected patients” have detectable HBV DNA, and they are at the risk of developing HBV reactivation (HBVr) . Besides, CHB can be seen in up to 30% of HBeAg− patients in this immune-inactive phase, which is due to HBV mutant strains that have lost their ability to secrete HBeAg, but can still replicate HBsAg.
Prevalence of HBV infection in patients with ARDs
The prevalence of HBV infection in patients with ARDs was studied mainly in Asian countries, where its prevalence was similar to that among the general population ( Table 1 ). However, a study from China in patients with ankylosing spondylitis found a significantly higher prevalence of HBsAg+ than rheumatoid arthritis (RA) and other rheumatic diseases, when compared with the general population . Two other studies found a significantly lower prevalence of HBsAg+ among patients with autoimmune disease, when compared with those with non-autoimmune diseases , and similarly in patients with systemic lupus erythematosus (SLE) when compared with non-SLE patients or the general population .
| Authors, year, [Ref.] | Country of study | Disease | No. of patients | HBsAg+ (%) | Anti-HBc+ (%) | Anti-HBs+ (%) | p -Value |
|---|---|---|---|---|---|---|---|
| Watanabe et al., 2014 | Japan | RA | 7650 | 1.1 | 25.2 | 16.7 | |
| SLE | 1031 | 0.4 | 13.7 | 10.1 | |||
| Sui et al., 2014, | China | Autoimmune conditions (AIH, PBC, SLE, and UC) | 938 | 2.24 | p = 0.0014 (autoimmune vs. non-autoimmune group) | ||
| Non-autoimmune disease | 3122 | 4.58 | |||||
| SLE | 155 | 2.58 | p = 0.24 (SLE vs, non-autoimmune group) | ||||
| Zou, et at., 2013, | China | RA | 223 | 11.2 (CHB1.7%) | p = NS | ||
| General population (Liang et al., ) | 8.7 (CHB 1.0%) | ||||||
| Tan et al., 2012, | China | RA | 476 | 6.5 | 51.1 | ||
| Zheng et al., 2012, | China | AS | 439 | 23.92 | p < 0.05 (AS vs. other groups) | ||
| General population | 606 | 12.87 | |||||
| Other-SpA | 172 | 14.53 | |||||
| RA | 698 | 9.60 | |||||
| OA | 200 | 8.18 | |||||
| Chen et al., 2012, | Taiwan | Primary Sjögren’s syndrome | 175 | 10.28 | |||
| Mori et al., 2011, | Japan | RA | 239 | 0.8 | 25.1 | ||
| Zhao et al., 2010, | China | SLE (hospitalization) | 859 | 2.33 | 67.52 | p < 0.01 (for HBsAg: SLE vs. general population). p < 0.001 (for HBsAg and anti-HBs: SLE vs. non-SLE) | |
| Non-SLE (hospitalization) | 78,046 | 12.75 | 57.25 | ||||
| General population | 20,000 | 9.57 | 58.78 | ||||
| Marcos et al., 2009, | Spain | Primary Sjögren’s syndrome | 603 | 0.83 | |||
| Guennoc et al., 2009, | France | Recent-onset arthritis | 808 | 0.12 | |||
| Permin et al., 1982, | Denmark | Autoimmune rheumatic diseases | 239 | 2.51 | 11.71 |
HBV reactivation
HBVr refers to the state of increased HBV DNA levels of >1 log 10 when compared with the baseline value, or a change in the status of HBV DNA detection from negative to positive . This can occur in HBsAg+ or HBsAg−/anti-HBc+ patients. The reactivation occurs as a result of lost host immune response (by either IMDs or acquired immune deficiency), and it causes unopposed HBV replication in the liver, resulting in increasing HBV DNA and expression of HBV-derived antigens. By discontinuing the immunosuppressive state and restoring the normal immune response, an immune reaction to the virus occurs, resulting in inflammation of the hepatocytes and liver. The clinical features of HBVr range from subclinical or asymptomatic to severe acute hepatitis, hepatic failure, and death.
Several risk factors for reactivation have been identified. Although the male sex has been reported as a risk factor from a series of cancer patients , a recent review of HBVr in 138 patients with immune-mediated inflammatory disease (ARDs in 76.81%) found an equal distribution in both sexes . HBsAg+ patients, and those with a high level of HBV DNA prior to immunosuppressive therapy, have a higher risk of HBVr when compared with those who are HBsAg−/anti-HBc+ or have a lower HBV DNA level . The risk of reactivation is lower in anti-HBs+ patients, as it indicates a full immune response to HBV infection. However, a very low level or loss of anti-HBs antibodies, during immunosuppressive therapy, might also increase the risk of reactivation . The type of underlying disease is another risk factor for HBVr. The frequency of HBVr is high among patients with hematologic malignancies, particularly those with lymphomas (27.8%) , organ transplants (16.66%) , or chemotherapy for breast cancer (41.16%) . This might be related to treating these conditions with high-dose corticosteroids and intense IMDs .
HBVr in patients with ARDs receiving antirheumatic therapy
With newer strategies for treating ARDs, the combination therapy with corticosteroids and cIMDs or cDMARDs, with or without bDMARDs, is often used to control disease activity. Therefore, HBVr in patients with occult HBV infections or HBV carriers is not unexpected. However, information regarding the prevalence, clinical features, and outcome of HBVr among patients with ARDs is scarce, when compared with the data on malignancies or organ transplantation.
Corticosteroids
Corticosteroids are among the common anti-inflammatory drugs used in patients with ARDs, with a dosage that can range from low (<10 mg/day) to very high (pulse corticosteroids) for treating life-threatening conditions. Corticosteroids have been shown to increase HBV transcription , and cases of HBVr after corticosteroid monotherapy for various conditions have been well described . Reactivation usually occurs in patients undergoing continuous treatment at a moderate to high dose (≥20 mg/day) for >3 months . Surprisingly, only sporadic cases of HBVr in patients with ARDs have been described, in which the incidence seems to be far less than the number of patients being treated with this agent alone. Further, the reactivation can be seen with low-dose corticosteroids (<10 mg/day of prednisolone) . This might be partially due to most of these patients also receiving IMDs as part of the therapy .
Conventional immunosuppressive agents
HBVr has been well recognized in patients receiving cIMDs or cDMARDs, and the risk is related to the intensity and use of combination cIMDs . Similar to those using corticosteroids, reports of HBVr among patients with ARDs receiving IMD monotherapy are scarce, as the majority of them also received corticosteroids . A recent prospective study found that the HBVr occurred in four of 211 (1.89%) patients with RA who were HBsAg+ or HBsAg−/anti-HBc+ and received cDMARD therapy without antiviral prophylaxis. The reactivation occurred between 1 and 15 months post cDMARD administration . A recent review found a prevalence of HBVr in 10 of 224 (4.46%) patients with rheumatic disease being treated with cDMARDs, eight of whom were given methotrexate (MTX) . Another study of HBVr was performed in 288 patients with SLE receiving corticosteroids and cIMDs. Eight of them were HBsAg+, and three had neither virology flares nor increased aminotransferase enzyme levels. An attempt to discontinue antiviral therapy (lamivudine) was made in three of five patients receiving it, and one of the three had a virological flare . Patients who have HBVr can be managed with generally good outcomes. Despite cIMDs or cDMARDs being considered as low risk for HBVr in patients with ARDs when used as monotherapy, acute fulminant hepatitis and fatal cases of HBVr with low-dose MTX have been described .
Biologic disease-modifying antirheumatic drugs
During the past decade, the use of bDMARDs has been increasing, particularly in patients with RA, spondyloarthropathies, and SLE. These agents have been shown to control disease activity very effectively. Unfortunately, information on HBVr among patients with ARDs receiving bDMARDs, besides anti-TNFα, is limited.
Anti-TNFα agents
As TNFα plays an important role in both innate and adaptive immunity against HBV infection, blockade of TNFα can result in HBV replication and reactivation. The mechanisms by which anti-TNFα induces reactivation of the HBV virus include the activation of complement, antibody-dependent cell-medicated cytotoxicity, complement-dependent cytotoxicity, B-cell depletion, and T-cell-dependent humoral response . Since approving anti-TNFα agents in the treatment of arthritic diseases, cases of HBVr occurring after their use have been reported ( Table 2 ). However, the risk of HBVr in anti-TNFα monotherapy is difficult to determine, as a majority of the patients also receive corticosteroids and cIMDs or cDMARDs as part of the combination therapy. In a recent review of 620 patients with rheumatic disease treated with anti-TNFα (416 with past HBV infection and 204 with chronic HBV infection), antiviral prophylaxis was administered in 36 of those with chronic HBV infection. HBVr occurred in 59 cases (9.52%), of which 13 and 46 belonged to the past HBV infection and chronic HBV infection groups, respectively. The risk of reactivation was higher among patients with CHB and inactive HBV carriers than patients with occult HBV infection and those who did not receive antiviral prophylaxis. The outcome of the HBVr treated with antiviral therapy was considered good, as only one patient suffered from liver failure and died 26 months later . Another recent meta-analysis, including 10 articles on HBVr in patients treated with anti-TNFα (nine with rheumatic diseases and one with psoriasis), found a pool estimate for the prevalence of HBVr at 4.2% (95% confidence interval (CI) 1.4–8.2), where the pool prevalence of HBVr was 3.0% and 15.4% among patients with occult HBV and overt HBV infection, respectively . The prevalence of reactivation was slightly lower in patients who received etanercept (3.9%) than those receiving adalimumab (4.6%). The pool estimated prevalence of reactivation was 4% in those who did not receive antiviral prophylaxis . The reason why patients treated with soluble receptor anti-TNFα have a lower risk of reactivation than those treated with monoclonal antibodies to anti-TNFα might be that the monoclonal antibody exhibits greater immunogenicity, and the frequency of administration at a certain interval results in a cytokine washout effect that does not occur in soluble receptor anti-TNFα .
| Authors, year, [Ref.] | Country of report | No. of patients | Disease | Anti-TNF treatment | HBV markers status | HBV reactivation | ||
|---|---|---|---|---|---|---|---|---|
| HBsAg+ | Anti-HBc+ | |||||||
| Anti-HBs− | Anti-HBs+ | |||||||
| Ye et al., 2014, | China | 87 | AS = 4, PsA = 3, RA = 10 | Not report | 37 (chronic hepatitis B = 6, antiviral prophylaxis = 13) | 50 | 8/24 (33.3%) in HBsAg+ without prophylaxis; none in resolved HBV infection | |
| Ryu et al., 2012, | South Korea | 49 | AS = 27, RA = 22 | A = 6, E = 38, I = 5 | 49 (antiviral prophylaxis = 20) | – | 1/20 (5.0%) with prophylaxis; 2/29 (7.7) without prophylaxis | |
| Lan et al., 2011, | Taiwan | 88 | RA | 18 (HBV DNA+ = 18, antiviral prophylaxis = 10) | 12 (HBV DNA+ = 4, no prophylaxis) | 58 (HBV DNA−) | 5/8 (62.5%) HBsAg+ without prophylaxis; 1/4 (25%) antiHBs−, HBV DNA+ without prophylaxis | |
| Mori 2011, | Japan | 239 | RA | A = 2, E = 18, I = 19 | 2 (under cDMARDs with antiviral prophylaxis) | 60 | – | 2/60 (3.3%) of anti-HBc+:
|
| Tamori et al., 2011, | Japan | 50 | RA | Not reported | 5 (HBV DNA+ in 3, antiviral prophylaxis) | 9 | 36 | 2/5 (40%) of HBsAg+ without prophylaxis; 1/45 (2.2%) of anti-HBc+, not under anti-TNF |
| Caporali et al., 2010, | Italy | 67 | AS = 4, PsA = 4, RA = 59 | A = 19, E = 23, I = 25 | – | 39 (HBV DNA-) | 28 | None |
| Vassilopoulos et al., 2010, | Greece | 131 | AS = 32, PsA = 21, RA = 66, Others = 12 | (A = 62, E = 64, I = 43)* | 14 (chronic hepatitis B = 3, antiviral prophylaxis = 14) | 9 | 10 | 1 chronic hepatitis B (with lamivudine resistant mutant); None in resolved HBV infection |
| Chung et al., 2009, | South Korea | 103 | AS = 59, JRA = 2, PsA = 1, RA = 41 | Not reported | 8 (carriers) | – | 1/8 (12.5%) | |
Rituximab
Currently, rituximab is indicated for rheumatoid arthritis, and antineutrophilic cytoplasmic antibody (ANCA)-associated vasculitis. Information regarding HBVr in patients with ARDs is far more limited than that of anti-TNFα. Experience with lymphomas clearly showed that HBVr was found more frequently in drug regimens containing rituximab than those without . However, cases of patients receiving rituximab for HBVr in RA and ANCA-associated vasculitis have been reported rarely . A recent prospective study of 14 HBV-infected patients with RA (HBsAg+ in two who received antiviral therapy, anti-HBs+/anti-HBc+ in nine, and anti-HBs−/anti-HBc+ in three) showed no HBVr during a follow-up period of 6–50 months (median 13 months) . The low prevalence of HBVr from rituximab in patients with ARDs might be due to the aggressive screening for HBV infection and preemptive antiviral therapy among those who are at risk.
Abatacept
Information on the use of abatacept and HBVr is also limited. However, cases of HBVr treated with abatacept have been reported in RA patients with resolved and occulted HBV infection, and antiviral therapy successfully treated the HBVr . In a retrospective study, eight RA patients with chronic HBV infection were treated with abatacept, and HBVr occurred in the four patients who did not receive antiviral prophylaxis, compared with no occurrence in the other four who did .
Tocilizumab
So far, three reports have addressed the use of tocilizumab in patients with HBV infection. The first was a patient with RA who was HBsAg+ with a high viral load, and who did not receive antiviral prophylaxis . The second was a patient with active RA who previously had HBVr with infliximab, but the virus was controlled successfully (undetectable HBV DNA) by antiviral therapy . The third was a patient with adult-onset Still’s disease who had active HBV infection (HBsAg+ with high viral load) together with antiviral prophylaxis . None of these patients developed HBVr.
Tofacitinib and belimumab are two recent IMDs approved for the treatment of RA and SLE, respectively. However, no case of HBVr from these two compounds has been reported at the time of this review.
Management and monitoring of HBV-infected patients requiring immunosuppressive therapy
It is clear that patients with CHB, HBsAg+, and past HBV infection, as well as active viral replication (high HBV DNA), are at a risk of HBVr after corticosteroids or immunosuppressive therapy; therefore, all patients with ARDs starting IMDs (either cIMDs or cDMARDs, or bDMARDs) should be screened for HBV infection (HBsAg, anti-HBs, and anti-HBc) . HBV vaccination should be given to nonexposed patients (HBsAg−/anti-HBs−/anti-HBc−). Those who have been vaccinated and are already immune to HBV (anti-HBs+/anti-HBc−) need not take further action. However, those with current HBV infection (CHB, HBsAg+ as a carrier, or past HBV infection (anti-HBc+/anti-HBs− or anti-HBs+)) should consult a hepatologist for a treatment plan and monitoring. HBsAg+ patients should receive preemptive antiviral therapy. It is less clear how anti-HBc+ patients, with or without anti-HBs antibodies, are managed. In general, anti-HBc+/anti-HBs+ patients are considered immune to HBV infection, and their risk of HBVr is minimal if they are receiving cDMARDs. However, a significant decrease in anti-HBsAb titer, of up to 70% of the baseline value, but not below 10 IU/L, has been observed in many cases of anti-TNFα therapy . This finding should be of concern among patients with a very low anti-HBsAb level, as it might become negative. The risk of HBVr in resolved HBV infection, with negative HBV DNA (<2000 U/mL), is lower than that with high HBV DNA (or occult infection); therefore, HBV DNA-positive patients should start preemptive antiviral therapy promptly before commencing IMDs. For those with HBV DNA at <2000 IU/mL, careful monitoring of the aspartate transaminase (AST), alanine transaminase (ALT), and HBV DNA level is recommended during IMD therapy, and antiviral therapy should be started as soon as the HBV DNA level increases. A marked increase in the serum HBV DNA level usually occurs prior to the elevation of the ALT with a median duration of 18.5 weeks (range, 12–28) . In countries where HBV DNA testing is costly or not widely available, preemptive therapy with antiviral therapy might be an option. Antiviral therapy should be given during IMD therapy and for at least 6–12 months after cessation, with careful monitoring of AST, ALT, HBeAg, HBeAb, and HBV DNA level (in those who were HBsAg+), and AST, ALT and HBV DNA level (in those with resolved HBV infection) . The currently approved antiviral medications for HBV infection are conventional IFNα, pegylated-IFNα, lamivudine, adefovir, entecavir, telbivudine, and tenofovir. The choice of antiviral therapy depends on the duration and intensity of the IMDs used, and the availability of antiviral drugs in each country, as well as the prevalence of resistant strains of the virus (e.g., lamivudine-resistant strain).
Elevation of AST or ALT levels during IMD therapy in HBV-infected patients is not always a case of HBVr. Differential diagnosis should also include drug-induced liver disease, hepatic involvement in rheumatic disease, alcoholic and nonalcoholic hepatitis, autoimmune hepatitis, thyroid disease, and other infectious causes of hepatitis (e.g., EBV, HIV, CMV, etc.).
Hepatitis B virus
HBV is a double-stranded DNA (dsDNA) virus and belongs to the Hepadnaviridae family. It is one of the most common viral infections worldwide, and it is estimated to affect 400 million people. Its prevalence is low in the USA and northern Europe (<2%), but high in Southeast Asia, China, and Africa (>8%) . Most patients acquire the virus during the perinatal period through vertical transmission, or during early childhood. Primary HBV infection usually is asymptomatic in 30–50% of cases in individuals older than 5 years. Although the virus is cleared in most patients (95%), resulting in lifelong immunity, 30–50% of those who acquire it during childhood develop chronic hepatitis B (CHB) infection .
After infection with HBV, active viral replication in the liver occurs, with a minimal host immune response (high serum HBV DNA levels, presence of HBeAg (HBeAg+) and HBsAg (HBsAg+), normal liver transaminase enzymes, and minimal hepatocyte changes), or the immune tolerance phase is observed. The second or immune clearance phase, which occurs in adolescents and adults, is characterized by a vigorous immune response to HBV, resulting in symptoms of clinical hepatitis with a marked increase in transaminase enzymes, together with hepatocyte necrosis and fibrosis. During this phase, HBV replication declines, and in 90% of cases, HBsAg and HBeAg gradually disappear or HBe seroconversion occurs (HBsAg−/HBeAg−). This stage is accompanied usually by the appearance of antibodies to HBV antigens (anti-HBs+, anti-HBe+, and anti-HBc+). The third or low-replication phase is characterized by the presence of the abovementioned antibodies, and a very low or undetectable HBV DNA (<2000 IU/mL), normal transaminase enzymes, and minimal liver inflammation . The presence of the anti-HBs+ state indicates full immunity from HBV infection (resolved HBV infection). However, some patients might still have a very low or undetectable level of anti-HBs antibodies (anti-HBs−/anti-HBc+), and those with a very low level of HBV DNA (<2000 IU/mL) without evidence of liver inflammation are defined as having an “occult HBV infection.” Nevertheless, a certain percentage of “occult HBV-infected patients” have detectable HBV DNA, and they are at the risk of developing HBV reactivation (HBVr) . Besides, CHB can be seen in up to 30% of HBeAg− patients in this immune-inactive phase, which is due to HBV mutant strains that have lost their ability to secrete HBeAg, but can still replicate HBsAg.
Prevalence of HBV infection in patients with ARDs
The prevalence of HBV infection in patients with ARDs was studied mainly in Asian countries, where its prevalence was similar to that among the general population ( Table 1 ). However, a study from China in patients with ankylosing spondylitis found a significantly higher prevalence of HBsAg+ than rheumatoid arthritis (RA) and other rheumatic diseases, when compared with the general population . Two other studies found a significantly lower prevalence of HBsAg+ among patients with autoimmune disease, when compared with those with non-autoimmune diseases , and similarly in patients with systemic lupus erythematosus (SLE) when compared with non-SLE patients or the general population .
| Authors, year, [Ref.] | Country of study | Disease | No. of patients | HBsAg+ (%) | Anti-HBc+ (%) | Anti-HBs+ (%) | p -Value |
|---|---|---|---|---|---|---|---|
| Watanabe et al., 2014 | Japan | RA | 7650 | 1.1 | 25.2 | 16.7 | |
| SLE | 1031 | 0.4 | 13.7 | 10.1 | |||
| Sui et al., 2014, | China | Autoimmune conditions (AIH, PBC, SLE, and UC) | 938 | 2.24 | p = 0.0014 (autoimmune vs. non-autoimmune group) | ||
| Non-autoimmune disease | 3122 | 4.58 | |||||
| SLE | 155 | 2.58 | p = 0.24 (SLE vs, non-autoimmune group) | ||||
| Zou, et at., 2013, | China | RA | 223 | 11.2 (CHB1.7%) | p = NS | ||
| General population (Liang et al., ) | 8.7 (CHB 1.0%) | ||||||
| Tan et al., 2012, | China | RA | 476 | 6.5 | 51.1 | ||
| Zheng et al., 2012, | China | AS | 439 | 23.92 | p < 0.05 (AS vs. other groups) | ||
| General population | 606 | 12.87 | |||||
| Other-SpA | 172 | 14.53 | |||||
| RA | 698 | 9.60 | |||||
| OA | 200 | 8.18 | |||||
| Chen et al., 2012, | Taiwan | Primary Sjögren’s syndrome | 175 | 10.28 | |||
| Mori et al., 2011, | Japan | RA | 239 | 0.8 | 25.1 | ||
| Zhao et al., 2010, | China | SLE (hospitalization) | 859 | 2.33 | 67.52 | p < 0.01 (for HBsAg: SLE vs. general population). p < 0.001 (for HBsAg and anti-HBs: SLE vs. non-SLE) | |
| Non-SLE (hospitalization) | 78,046 | 12.75 | 57.25 | ||||
| General population | 20,000 | 9.57 | 58.78 | ||||
| Marcos et al., 2009, | Spain | Primary Sjögren’s syndrome | 603 | 0.83 | |||
| Guennoc et al., 2009, | France | Recent-onset arthritis | 808 | 0.12 | |||
| Permin et al., 1982, | Denmark | Autoimmune rheumatic diseases | 239 | 2.51 | 11.71 |
HBV reactivation
HBVr refers to the state of increased HBV DNA levels of >1 log 10 when compared with the baseline value, or a change in the status of HBV DNA detection from negative to positive . This can occur in HBsAg+ or HBsAg−/anti-HBc+ patients. The reactivation occurs as a result of lost host immune response (by either IMDs or acquired immune deficiency), and it causes unopposed HBV replication in the liver, resulting in increasing HBV DNA and expression of HBV-derived antigens. By discontinuing the immunosuppressive state and restoring the normal immune response, an immune reaction to the virus occurs, resulting in inflammation of the hepatocytes and liver. The clinical features of HBVr range from subclinical or asymptomatic to severe acute hepatitis, hepatic failure, and death.
Several risk factors for reactivation have been identified. Although the male sex has been reported as a risk factor from a series of cancer patients , a recent review of HBVr in 138 patients with immune-mediated inflammatory disease (ARDs in 76.81%) found an equal distribution in both sexes . HBsAg+ patients, and those with a high level of HBV DNA prior to immunosuppressive therapy, have a higher risk of HBVr when compared with those who are HBsAg−/anti-HBc+ or have a lower HBV DNA level . The risk of reactivation is lower in anti-HBs+ patients, as it indicates a full immune response to HBV infection. However, a very low level or loss of anti-HBs antibodies, during immunosuppressive therapy, might also increase the risk of reactivation . The type of underlying disease is another risk factor for HBVr. The frequency of HBVr is high among patients with hematologic malignancies, particularly those with lymphomas (27.8%) , organ transplants (16.66%) , or chemotherapy for breast cancer (41.16%) . This might be related to treating these conditions with high-dose corticosteroids and intense IMDs .
HBVr in patients with ARDs receiving antirheumatic therapy
With newer strategies for treating ARDs, the combination therapy with corticosteroids and cIMDs or cDMARDs, with or without bDMARDs, is often used to control disease activity. Therefore, HBVr in patients with occult HBV infections or HBV carriers is not unexpected. However, information regarding the prevalence, clinical features, and outcome of HBVr among patients with ARDs is scarce, when compared with the data on malignancies or organ transplantation.
Corticosteroids
Corticosteroids are among the common anti-inflammatory drugs used in patients with ARDs, with a dosage that can range from low (<10 mg/day) to very high (pulse corticosteroids) for treating life-threatening conditions. Corticosteroids have been shown to increase HBV transcription , and cases of HBVr after corticosteroid monotherapy for various conditions have been well described . Reactivation usually occurs in patients undergoing continuous treatment at a moderate to high dose (≥20 mg/day) for >3 months . Surprisingly, only sporadic cases of HBVr in patients with ARDs have been described, in which the incidence seems to be far less than the number of patients being treated with this agent alone. Further, the reactivation can be seen with low-dose corticosteroids (<10 mg/day of prednisolone) . This might be partially due to most of these patients also receiving IMDs as part of the therapy .
Conventional immunosuppressive agents
HBVr has been well recognized in patients receiving cIMDs or cDMARDs, and the risk is related to the intensity and use of combination cIMDs . Similar to those using corticosteroids, reports of HBVr among patients with ARDs receiving IMD monotherapy are scarce, as the majority of them also received corticosteroids . A recent prospective study found that the HBVr occurred in four of 211 (1.89%) patients with RA who were HBsAg+ or HBsAg−/anti-HBc+ and received cDMARD therapy without antiviral prophylaxis. The reactivation occurred between 1 and 15 months post cDMARD administration . A recent review found a prevalence of HBVr in 10 of 224 (4.46%) patients with rheumatic disease being treated with cDMARDs, eight of whom were given methotrexate (MTX) . Another study of HBVr was performed in 288 patients with SLE receiving corticosteroids and cIMDs. Eight of them were HBsAg+, and three had neither virology flares nor increased aminotransferase enzyme levels. An attempt to discontinue antiviral therapy (lamivudine) was made in three of five patients receiving it, and one of the three had a virological flare . Patients who have HBVr can be managed with generally good outcomes. Despite cIMDs or cDMARDs being considered as low risk for HBVr in patients with ARDs when used as monotherapy, acute fulminant hepatitis and fatal cases of HBVr with low-dose MTX have been described .
Biologic disease-modifying antirheumatic drugs
During the past decade, the use of bDMARDs has been increasing, particularly in patients with RA, spondyloarthropathies, and SLE. These agents have been shown to control disease activity very effectively. Unfortunately, information on HBVr among patients with ARDs receiving bDMARDs, besides anti-TNFα, is limited.
Anti-TNFα agents
As TNFα plays an important role in both innate and adaptive immunity against HBV infection, blockade of TNFα can result in HBV replication and reactivation. The mechanisms by which anti-TNFα induces reactivation of the HBV virus include the activation of complement, antibody-dependent cell-medicated cytotoxicity, complement-dependent cytotoxicity, B-cell depletion, and T-cell-dependent humoral response . Since approving anti-TNFα agents in the treatment of arthritic diseases, cases of HBVr occurring after their use have been reported ( Table 2 ). However, the risk of HBVr in anti-TNFα monotherapy is difficult to determine, as a majority of the patients also receive corticosteroids and cIMDs or cDMARDs as part of the combination therapy. In a recent review of 620 patients with rheumatic disease treated with anti-TNFα (416 with past HBV infection and 204 with chronic HBV infection), antiviral prophylaxis was administered in 36 of those with chronic HBV infection. HBVr occurred in 59 cases (9.52%), of which 13 and 46 belonged to the past HBV infection and chronic HBV infection groups, respectively. The risk of reactivation was higher among patients with CHB and inactive HBV carriers than patients with occult HBV infection and those who did not receive antiviral prophylaxis. The outcome of the HBVr treated with antiviral therapy was considered good, as only one patient suffered from liver failure and died 26 months later . Another recent meta-analysis, including 10 articles on HBVr in patients treated with anti-TNFα (nine with rheumatic diseases and one with psoriasis), found a pool estimate for the prevalence of HBVr at 4.2% (95% confidence interval (CI) 1.4–8.2), where the pool prevalence of HBVr was 3.0% and 15.4% among patients with occult HBV and overt HBV infection, respectively . The prevalence of reactivation was slightly lower in patients who received etanercept (3.9%) than those receiving adalimumab (4.6%). The pool estimated prevalence of reactivation was 4% in those who did not receive antiviral prophylaxis . The reason why patients treated with soluble receptor anti-TNFα have a lower risk of reactivation than those treated with monoclonal antibodies to anti-TNFα might be that the monoclonal antibody exhibits greater immunogenicity, and the frequency of administration at a certain interval results in a cytokine washout effect that does not occur in soluble receptor anti-TNFα .
| Authors, year, [Ref.] | Country of report | No. of patients | Disease | Anti-TNF treatment | HBV markers status | HBV reactivation | ||
|---|---|---|---|---|---|---|---|---|
| HBsAg+ | Anti-HBc+ | |||||||
| Anti-HBs− | Anti-HBs+ | |||||||
| Ye et al., 2014, | China | 87 | AS = 4, PsA = 3, RA = 10 | Not report | 37 (chronic hepatitis B = 6, antiviral prophylaxis = 13) | 50 | 8/24 (33.3%) in HBsAg+ without prophylaxis; none in resolved HBV infection | |
| Ryu et al., 2012, | South Korea | 49 | AS = 27, RA = 22 | A = 6, E = 38, I = 5 | 49 (antiviral prophylaxis = 20) | – | 1/20 (5.0%) with prophylaxis; 2/29 (7.7) without prophylaxis | |
| Lan et al., 2011, | Taiwan | 88 | RA | 18 (HBV DNA+ = 18, antiviral prophylaxis = 10) | 12 (HBV DNA+ = 4, no prophylaxis) | 58 (HBV DNA−) | 5/8 (62.5%) HBsAg+ without prophylaxis; 1/4 (25%) antiHBs−, HBV DNA+ without prophylaxis | |
| Mori 2011, | Japan | 239 | RA | A = 2, E = 18, I = 19 | 2 (under cDMARDs with antiviral prophylaxis) | 60 | – | 2/60 (3.3%) of anti-HBc+:
|
| Tamori et al., 2011, | Japan | 50 | RA | Not reported | 5 (HBV DNA+ in 3, antiviral prophylaxis) | 9 | 36 | 2/5 (40%) of HBsAg+ without prophylaxis; 1/45 (2.2%) of anti-HBc+, not under anti-TNF |
| Caporali et al., 2010, | Italy | 67 | AS = 4, PsA = 4, RA = 59 | A = 19, E = 23, I = 25 | – | 39 (HBV DNA-) | 28 | None |
| Vassilopoulos et al., 2010, | Greece | 131 | AS = 32, PsA = 21, RA = 66, Others = 12 | (A = 62, E = 64, I = 43)* | 14 (chronic hepatitis B = 3, antiviral prophylaxis = 14) | 9 | 10 | 1 chronic hepatitis B (with lamivudine resistant mutant); None in resolved HBV infection |
| Chung et al., 2009, | South Korea | 103 | AS = 59, JRA = 2, PsA = 1, RA = 41 | Not reported | 8 (carriers) | – | 1/8 (12.5%) | |
Rituximab
Currently, rituximab is indicated for rheumatoid arthritis, and antineutrophilic cytoplasmic antibody (ANCA)-associated vasculitis. Information regarding HBVr in patients with ARDs is far more limited than that of anti-TNFα. Experience with lymphomas clearly showed that HBVr was found more frequently in drug regimens containing rituximab than those without . However, cases of patients receiving rituximab for HBVr in RA and ANCA-associated vasculitis have been reported rarely . A recent prospective study of 14 HBV-infected patients with RA (HBsAg+ in two who received antiviral therapy, anti-HBs+/anti-HBc+ in nine, and anti-HBs−/anti-HBc+ in three) showed no HBVr during a follow-up period of 6–50 months (median 13 months) . The low prevalence of HBVr from rituximab in patients with ARDs might be due to the aggressive screening for HBV infection and preemptive antiviral therapy among those who are at risk.
Abatacept
Information on the use of abatacept and HBVr is also limited. However, cases of HBVr treated with abatacept have been reported in RA patients with resolved and occulted HBV infection, and antiviral therapy successfully treated the HBVr . In a retrospective study, eight RA patients with chronic HBV infection were treated with abatacept, and HBVr occurred in the four patients who did not receive antiviral prophylaxis, compared with no occurrence in the other four who did .
Tocilizumab
So far, three reports have addressed the use of tocilizumab in patients with HBV infection. The first was a patient with RA who was HBsAg+ with a high viral load, and who did not receive antiviral prophylaxis . The second was a patient with active RA who previously had HBVr with infliximab, but the virus was controlled successfully (undetectable HBV DNA) by antiviral therapy . The third was a patient with adult-onset Still’s disease who had active HBV infection (HBsAg+ with high viral load) together with antiviral prophylaxis . None of these patients developed HBVr.
Tofacitinib and belimumab are two recent IMDs approved for the treatment of RA and SLE, respectively. However, no case of HBVr from these two compounds has been reported at the time of this review.
Management and monitoring of HBV-infected patients requiring immunosuppressive therapy
It is clear that patients with CHB, HBsAg+, and past HBV infection, as well as active viral replication (high HBV DNA), are at a risk of HBVr after corticosteroids or immunosuppressive therapy; therefore, all patients with ARDs starting IMDs (either cIMDs or cDMARDs, or bDMARDs) should be screened for HBV infection (HBsAg, anti-HBs, and anti-HBc) . HBV vaccination should be given to nonexposed patients (HBsAg−/anti-HBs−/anti-HBc−). Those who have been vaccinated and are already immune to HBV (anti-HBs+/anti-HBc−) need not take further action. However, those with current HBV infection (CHB, HBsAg+ as a carrier, or past HBV infection (anti-HBc+/anti-HBs− or anti-HBs+)) should consult a hepatologist for a treatment plan and monitoring. HBsAg+ patients should receive preemptive antiviral therapy. It is less clear how anti-HBc+ patients, with or without anti-HBs antibodies, are managed. In general, anti-HBc+/anti-HBs+ patients are considered immune to HBV infection, and their risk of HBVr is minimal if they are receiving cDMARDs. However, a significant decrease in anti-HBsAb titer, of up to 70% of the baseline value, but not below 10 IU/L, has been observed in many cases of anti-TNFα therapy . This finding should be of concern among patients with a very low anti-HBsAb level, as it might become negative. The risk of HBVr in resolved HBV infection, with negative HBV DNA (<2000 U/mL), is lower than that with high HBV DNA (or occult infection); therefore, HBV DNA-positive patients should start preemptive antiviral therapy promptly before commencing IMDs. For those with HBV DNA at <2000 IU/mL, careful monitoring of the aspartate transaminase (AST), alanine transaminase (ALT), and HBV DNA level is recommended during IMD therapy, and antiviral therapy should be started as soon as the HBV DNA level increases. A marked increase in the serum HBV DNA level usually occurs prior to the elevation of the ALT with a median duration of 18.5 weeks (range, 12–28) . In countries where HBV DNA testing is costly or not widely available, preemptive therapy with antiviral therapy might be an option. Antiviral therapy should be given during IMD therapy and for at least 6–12 months after cessation, with careful monitoring of AST, ALT, HBeAg, HBeAb, and HBV DNA level (in those who were HBsAg+), and AST, ALT and HBV DNA level (in those with resolved HBV infection) . The currently approved antiviral medications for HBV infection are conventional IFNα, pegylated-IFNα, lamivudine, adefovir, entecavir, telbivudine, and tenofovir. The choice of antiviral therapy depends on the duration and intensity of the IMDs used, and the availability of antiviral drugs in each country, as well as the prevalence of resistant strains of the virus (e.g., lamivudine-resistant strain).
Elevation of AST or ALT levels during IMD therapy in HBV-infected patients is not always a case of HBVr. Differential diagnosis should also include drug-induced liver disease, hepatic involvement in rheumatic disease, alcoholic and nonalcoholic hepatitis, autoimmune hepatitis, thyroid disease, and other infectious causes of hepatitis (e.g., EBV, HIV, CMV, etc.).
Related posts:
The intestinal microbiome in human disease and how it relates to arthritis and spondyloarthritis
Role of genetics in infection-associated arthritis
Vasculitis related to viral and other microbial agents
Evolving spectrum of HIV-associated rheumatic syndromes
Risk of infection with biologic antirheumatic therapies in patients with rheumatoid arthritis
Role of vaccinations and prophylaxis in rheumatic diseases
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
