Hepatitis B virus reactivation with corticosteroid therapy in patients with adrenal insufficiency.

Abstract Objective Whether or not reactivation of hepatitis B virus (HBV) might occur during corticosteroid therapy in hepatitis B surface antigen (HBsAg)‐negative patients with adrenal insufficiency was investigated. Patients and Methods We consecutively enrolled 66 patients with adrenal insufficiency undergoing physiological corticosteroid replacement therapy at Saitama Medical University Hospital between June 2013 and June 2014, and 220 patients with rheumatic disease receiving a pharmacologic dose of corticosteroids served as the positive control group. The latter group was separated into 101 patients treated only with corticosteroids, and 119 patients given corticosteroids plus immunosuppressants and/or disease‐modifying antirheumatic drugs (DMARDs). HBsAg and antibody (Ab) levels against HBs, and hepatitis B core (HBc) were determined in all the patients. In patients with positive HBsAb and/or HBcAb, real‐time PCR was performed for HBV‐DNA. The incidence rates of conversion to HBV‐DNA‐positive status were evaluated. Results Hepatitis B virus reactivation occurred in six patients with rheumatic disease, three of whom were receiving a pharmacological dose of corticosteroids only, and three who were receiving corticosteroids with immunosuppressants and/or DMARDs. However, no reactivation occurred in patients receiving corticosteroid replacements for adrenal insufficiency. Maintenance and maximum corticosteroid doses administered to patients with rheumatic disease were significantly greater than those in patients with adrenal insufficiency. Conclusion These results suggest that, although corticosteroid replacement therapy for adrenal insufficiency might be safe with respect to HBV reactivation, attention should be paid to HBV reactivation during corticosteroid therapy in rheumatic disease patients, since the dose of corticosteroids administered is usually large, and since other immunosuppressants are co‐administered.


| INTRODUC TI ON
Hepatitis B virus (HBV) reactivation occurring in patients undergoing cytotoxic chemotherapy and/or immunosuppressive therapy is a well-recognized complication of considerable clinical importance.
Reactivation of HBV is more often seen in hepatitis B carriers who are positive for hepatitis B surface (HBs) antigen (Ag) and for antibody (Ab) against hepatitis B core (HBc). Acute hepatitis, that is, de novo hepatitis, is likely to occur because of HBV reactivation during chemotherapy and/or immunosuppressive therapy given to patients with transplanted organs and leads to variable manifestations that range from subclinical serum aminotransferase elevation to fatal fulminant hepatitis.
However, a few cases of HBV reactivation in patients with resolved infection as defined by negative HBsAg and positive HBcAb with or without HBsAb have been reported during chemotherapy or immunosuppressive therapy. 1,2 The incidence of de novo hepatitis due to HBV reactivation in HBsAg-negative patients with malignant lymphoma treated with corticosteroid-containing cytotoxic chemotherapy was reported to be 2.7% over a 4-year period and 3.3% during a 12.4-month follow-up. 3,4 However, use of rituximab, an anti-CD20-directed monoclonal antibody, in combination with corticosteroid-containing chemotherapy, has been frequently associated with HBV reactivation and resultant de novo hepatitis. [4][5][6] Most of these reports come from the fields of oncology and transplantation, with a growing number of cases being reported in patients with rheumatic disease who are also undergoing immunosuppressive therapy. The incidence of HBV reactivation was reported to be 7/135 (5.2%) after a 1-year prospective observation in rheumatoid arthritis patients with resolved hepatitis B treated with corticosteroids, immunosuppressants and/or disease-modifying antirheumatic drugs (DMARDs). 7 A recent prospective observational study in Japan has reported that HBV reactivation among rheumatic disease patients with resolved HBV infections treated with prednisone, immunosuppressants and/or biological DMARDs, occurred in 35/1042 (3.4%) of individuals over a period of 2 years. 8 Estimated incidence rates of HBS reactivation and hence de novo hepatitis vary, likely due to differences among the patient populations studied, drugs administered and the duration of follow-up.
Of all the traditional immunosuppressive medications, corticosteroids have been most often implicated in the induction of HBV reactivation. The risk of HBV reactivation in patients with inflammatory bowel disease, vasculitis, sarcoidosis and autoimmune disease might differ, depending on whether the patients are HBsAg-positive/ HBcAb-positive or HBsAg-negative/HBcAb-positive, whether the dosage of chronic prednisone therapy is low, moderate or high, or whether the duration is short or longer than 4 weeks. 8,9 However, the exact incidence rate of HBV reactivation in HBsAg-negative/ HBcAb-positive patients with autoimmune disease such as rheumatoid arthritis treated only with corticosteroids has not been fully evaluated.
On the other hand, adrenal insufficiency is usually treated with optimal replacement dose corticosteroids, equivalent to 5 mg of prednisone/day, to achieve physiological plasma cortisol levels comparable to those in healthy individuals. The incidence of HBV reactivation in patients with adrenal insufficiency given a physiological dose of corticosteroids is also unknown. Therefore, we investigated the incidence of the reactivation of HBV in HBsAg-negative patients in adrenal insufficiency patients with corticosteroid therapy. And we also evaluated the incidence rate of HBV reactivation in patients with rheumatic disease treated with corticosteroids with or without immunosuppressants and/or DMARDs as the positive control. The levels of aspartate transaminase (AST), alanine transaminase (ALT), lactate dehydrogenase (LDH) and γ-GTP were also measured during the observation period to assess liver function.

| Statistical analysis
All the data are shown as mean ± SE. The chi-squared test and ANOVA followed by a Tukey-Kramer test were performed for comparisons, using JMP 9 statistical software. A P-value of less than 0.05 was considered to be statistically significant.   of corticosteroids (prednisone or equivalent), that is 6.6 ± 0.5 mg and 6.5 ± 0.5 mg in patients with rheumatic disease treated with corticosteroids alone, and corticosteroids plus other immunosuppressants and DMARDs, respectively, were significantly larger than those (4.2 ± 0.2 mg) in patients with adrenal insufficiency (P < 0.05).

| RE SULTS
Maximum daily doses of corticosteroids differed significantly among the three groups (P < 0.0047). The dose of corticosteroids in patients with rheumatic disease treated with only corticosteroids or corticosteroids plus other immunosuppressants and/or DMARDs, that is 24.1 ± 1.8 and 18.9 ± 1.8 mg, respectively, were significantly greater than 4.2 ± 0.2 mg in patients with adrenal insufficiency (P < 0.05).
The levels of AST, ALT, LDH and γ-GTP in the three groups are also shown in Table 2. No statistically significant difference was noted in these liver function tests. It should be noted that, among the patients who had converted to HBV-DNA-positive status, corticosteroids were administered as monotherapy in three patients and as combination therapy in three patients. Of these patients, one was co-administered with melpharan, while two were co-administered with methotrexate and either tacrolimus or abatacept (Table 4).

| D ISCUSS I ON
In this study, we investigated the incidence of HBV reactivation among HBsAg-negative patients treated with a physiological dose of corticosteroids as replacement therapy and a pharmacological dose for anti-inflammatory and/or immunosuppressive purposes. First of all, patients who showed negative for both HBsAb and HBcAb at the beginning of and/or during the study corresponded to about 85% of the patients included in the study. They were believed to have not been exposed to HBV, although we cannot completely rule out the possibility that several patients had been exposed to HBV but turned out to be seronegative, that is, HBsAg-negative. Those patients who had never been exposed to HBV during the study period provide no value in a study to elucidate the effects of corticosteroids on the reactivation of HBV. Still, it is meaningful to learn about how many patients were exposed to HBV, and how many were not during the study period. In the present study, 11 patients showed positive for HBsAb and/or HBcAb in the adrenal insufficiency group, com- There are several possible mechanisms for corticosteroid-induced HBV reactivation. The first mechanism might be that corticosteroids directly cause HBV to proliferate because HBV has a glucocorticoidresponsive element that has the same base sequence as the gluco- body. 13 In fact, a corticosteroid dose equivalent to 30 mg/d of prednisone results in the proliferation of HBV and imposes a risk of HBV reactivation associated with high-dose steroids.
In contrast, it should be noted that we had no cases of HBV reactivation among adrenal-insufficient patients on corticosteroid replacement therapy. The daily maintenance and maximum doses of corticosteroids were both 4.2 ± 0.2 mg, which were significantly lower than those prescribed to rheumatic disease patients. In recent reviews by Perrillo et al 14  are regarded to be at low risk with an incidence rate of <1.0%. Based on the current study, the incidence of HBV reactivation in rheumatic disease patients treated at low doses of corticosteroids (6.5 mg prednisone or equivalent) might be lower than those reported in these reviews. Furthermore, no HBV reactivation occurred in patients with adrenal insufficiency whose treatment was replaced with a physiological dose of corticosteroids, that is, 4.2 mg of prednisone or equivalent per day, indicating that replacement therapy might be safe with respect to HBV reactivation. However, it should be noted that, in an inactive HBV carrier, HBV reactivation had occurred during a very lowdose steroid treatment (2.5 mg of prednisone per day plus DMARDs such as sulfasalazine and hydroxychloroquine). 16 It has been reported that a greater number of HBcAb-positive and HBsAb-negative patients experience HBV reactivation, compared with patients who are HBcAb-negative and anti-HBs-positive. 6,17 In our study, all six HBV-positive patients were positive for HBcAb, suggesting that HBcAb has considerable potential as a marker and screening tool for patients with resolved HBV infection.
Our results suggest that special attention should be given to the possible reactivation of HBV in patients who are positive for HBcAb.
In our study, patients who experienced HBV reactivation were aged between 46 and 68 years, with five of the six patients aged over 60. Other studies have reported age to be significantly associated with HBV reactivation in HBsAg-negative patients with B-cell lymphoma or rheumatic disease treated with rituximab plus corticosteroid-containing chemotherapy or corticosteroids, immunosuppressive drugs and/ or DMARDs. 8,10 However, some rheumatic disease patients in their 20s and 30s treated with immunosuppressive and biological drugs were reported to show HBV reactivation. 18 Moreover, in one Japanese study that evaluated 135 resolved hepatitis B patients with rheumatoid arthritis, age was not a predictive factor for HBV reactivation. 7 Therefore, attention should be paid to not only the elderly but also to younger individuals with respect to HBV reactivation, especially when biological DMARDs are being co-administered with corticosteroids. In addition, although the male gender has been reported to be more associated with HBV reactivation in patients with chemotherapy-treated lymphoma, 3,5 five of the six patients in our present study were female. This is consistent with previous studies in rheumatic disease which report that the male gender was not necessarily associated with HBV reactivation on corticosteroids with or without biological agents. 7,8 Hui et al 4  This study has several limitations. First, it was designed as a cross-sectional observational study, not a randomly controlled prospective study. Second, the rate of incidence of HBV reactivation did not show a statistically significant difference among the three groups. In conclusion, our present study suggests that, although corticosteroid replacement therapy for adrenal insufficiency might be safe with respect to HBV reactivation, we should remain alert to the risk of HBV reactivation during corticosteroid therapy in patients with rheumatic disease, since the dose of corticosteroid administered is usually high, especially at the beginning, and since other immunosuppressive and molecular-targeted drugs are co-administered.

ACKOWLED G EM ENT
We would like to extend our special acknowledgement to Dr Nobuaki Nakayama, Department of Gastroenterology and Hepatology, Saitama Medical University Hospital for his valuable suggestion in interpreting data and writing the manuscript.

CO N FLI C T O F I NTE R E S T
Nothing to declare.

E TH I C A L S TATEM ENT
This study was approved by the Institutional Review Board of Saitama Medical University Hospital (No. 15-053-1) and the informed consent was obtained from all participants.

DATA ACCE SS I B I LIT Y
All data generated or analysed during this study are included in this published article.