Abstract
Objective Hepatitis B virus (HBV) infection remains a significant public health challenge, particularly for immunocompromised patients. Our aim was to evaluate the serologic immunity in immunocompromised rheumatology and inflammatory bowel disease (IBD) patients, assess factors for serologic nonimmunity, and evaluate their response to 1 HBV booster dose.
Methods Immunocompromised rheumatology and IBD patients with completed HBV screening were identified. A chart review was performed to collect demographics, clinical information, baseline HBV serology results, and serologic response to booster vaccination. Serologic nonimmunity was defined as a negative/indeterminate hepatitis B surface antibody (anti-HBs) level.
Results Among 580 patients, 71% were nonimmune. The highest portion of nonimmune patients were 11–18 years old (P = 0.004). There was no significant difference between immune and nonimmune patients with regards to diagnosis (P = 0.34), age at diagnosis (P = 0.64), duration of treatment (P = 0.07), or type of medications (P = 0.08). Sixty-two percent of those who received a booster vaccine were rescreened, and most (68%) seroconverted. In those 18 years or older, only half seroconverted.
Conclusion Results of this study support the benefit of HBV screening in immunosuppressed patients. Beginning at age 11 years, most patients lacked serologic immunity to HBV. Seroconversion for most patients 11–18 years occurred after 1 booster vaccine. Thus, for immunocompromised patients without recent HBV serologic data, obtaining the HBV serology beginning at age 11 years might be considered. Those 18 years and older were least likely to seroconvert after 1 booster, indicating that they may benefit from receiving the 3-dose HBV vaccine series.
Advances in the understanding of the mechanisms of inflammation involved in rheumatic disorders and inflammatory bowel disease (IBD) have led to the invention of new target-specific medications referred to as “biologics”. Biologic agents are designed to inhibit specific components of the immune system such as cytokines and their complex interactions. The use of biologics alone or in combination with other immunosuppressive medications, such as corticosteroids, methotrexate, 6-mercaptopurine, have dramatically improved the outcome for children with rheumatic disorders and IBD, allowing for improved disease control and maintenance of remission. However, enthusiasm for improved outcomes must be tempered by the fact that chronic immunosuppression may lead to an increase in vaccine-preventable illnesses and reactivation of previous infections.1,2,3 Further, biologic agents have been associated with a higher risk of serious infections when compared with the nonbiologic disease-modifying antirheumatic medications.4
In particular, hepatitis B virus (HBV) is a significant concern for patients considered immunocompromised by either their medications or underlying disease condition. Although universal vaccination against HBV was implemented in the United States in 1991, evidence shows that 10% of HBV vaccine recipients fail to demonstrate a serologic response to the vaccine.5 In addition, use of biologic agents has been shown to accelerate the natural decline in the protective antibody levels against HBV.6 For example, when evaluating previously vaccinated pediatric patients with IBD, 50% were found to no longer have serologic immunity to HBV.6
Acute HBV infection as well as reactivation of HBV in immunocompromised patients entail considerable risk for the patient. In addition to elevated transaminase levels, and clinical signs of hepatitis, patients may suffer serious complications, such as liver failure, liver cirrhosis, or hepatocellular carcinoma.7,8 Reactivation of HBV is defined as abrupt increase in HBV replication in a patient with current or past HBV infection.9 Hepatitis B reactivation while receiving biologic therapy has been known to result in 5–25% mortality.7,10 Data on the rate of past or chronic HBV infections in immunocompromised children are not available. However, there are data that when screened prior to starting biologic treatment, up to 6% of adult rheumatology patients have evidence of past HBV infection.8 Therefore, the Advisory Committee on Immunization Practices recommends that all patients receiving immunosuppressive therapy for rheumatology or gastroenterology disorders should be screened for HBV infection and/or immunity and receive repeat vaccinations as needed.11 This screening recommendation is endorsed by multiple professional societies.10
Therefore, our study aimed to describe the prevalence of serologic immunity against HBV among a single-center cohort of immunocompromised patients with pediatric-onset IBD and rheumatologic diseases, investigate the potential factors associated with lack of immunity, and evaluate the response to a single HBV vaccine booster.
METHODS
Population and study design. We conducted a retrospective chart review on patients seen in the rheumatology and IBD clinics at Cincinnati Children’s Hospital Medical Center (CCHMC). Patients included in this study were identified as immunocompromised through the electronic medical chart using a unique logic defined by the following: (Meds: Current, by grouper OR Problems by grouper OR HM: Modifier). This logic has been validated elsewhere.12 The time range for the data collected was between January 1, 2011, and January 25, 2017. Patients who lacked hepatitis B surface antibody (anti-HBs) were excluded. The primary outcome measure was anti-HBs positivity (serologic immunity). The following data items were collected: sex, age at diagnosis, age at start of immunosuppressive medications, age at HBV screening, immunosuppressive medications at the time of HBV screening, initial HBV serology results and time, time and type of HBV booster doses, and time and results of HBV serology after booster doses. Patients were divided into the following categories as per their diagnosis: IBD (including Crohn disease and ulcerative colitis), autoimmune arthritis (mainly juvenile idiopathic arthritis [JIA]), autoinflammatory disease (including systemic JIA, periodic fever syndrome, Behçet disease, chronic recurrent multifocal osteomyelitis), connective tissue disease/vasculitis (including systemic lupus erythematosus, systemic sclerosis, and vasculitides), and other (including uveitis and myositis). Immunosuppressant medications taken at the time of HBV screening were recorded and categorized as biologics, nonbiologic disease-modifying antirheumatic drugs (DMARDs; including hydroxychloroquine, methotrexate, mycophenolate mofetil, azathioprine, tacrolimus), or as none if the patient was not treated with any immunosuppressants at the time of HBV screening or the 3 months prior. Patients who received intravenous or subcutaneous immunoglobulins were eliminated from this study. This study was approved by the Cincinnati Children’s Institutional Review Board (2016-6462).
Hepatitis B serology. At CCHMC, the rheumatology and IBD divisions implemented a quality improvement initiative to standardize HBV screening and repeat vaccinations for their immunocompromised patients in 2015.
The recommended serologic screening for HBV includes hepatitis B surface antigen (HbsAg), hepatitis B core antibody (anti-HBc), and anti-HBs. In our laboratory, these serologies are analyzed by chemiluminescent microparticle immunoassay and reported as positive, negative, or indeterminate. Negative values were those < 10 mIU/mL. Indeterminate for anti-HBs was considered negative. Different patterns of HBsAg or anti-HBc can represent active infection, chronic inactive carrier, or previous infection. If HBsAg or anti-HBc are positive, they are referred to the department of hepatology and monitored closely. If anti-HB is positive, then the patient is considered immune and no further action is needed. If anti-HB is negative, then the patient is vaccinated per the 2013 Infectious Disease Society of America guidelines. Initial attempts at gathering primary vaccine records were extremely time consuming and largely unsuccessful, and therefore, patients were assumed to have completed the primary HBV vaccine series.
Data analysis. Descriptive statistics were calculated for all variables. R software (R Foundation for Statistical Computing) was used to perform analyses. For continuous variables, mean and SD are reported, and comparisons were calculated using 2-sample t tests. For categorical variables, frequencies and percentages are reported, and comparisons were calculated using chi-square tests.
RESULTS
Demographic and clinical characteristics of patients. A total of 580 patient charts were reviewed for this study. Of those, 354 (61%) were from the rheumatology clinic, whereas 226 were from the IBD clinic (Table 1). There were 237 (41%) males and 343 (59%) females in the cohort. The age range at the time of screening was 4–29 years of age. The mean age at screening was 16 years for rheumatology patients and 15 years for patients with IBD. The majority of patients from either clinic were exposed to an antitumor necrosis factor (anti-TNF) biologic agent, chiefly infliximab (IFX). Of the rheumatology patients, 213 (60%) were diagnosed with juvenile arthritis, 93 (26%) had a connective tissue disease or vasculitis, 18 (5%) had an autoinflammatory disease, and 30 (9%) had other rheumatologic diagnoses.
Demographics and clinical characteristics of patients by clinic.
Hepatitis B serology results and seroconversion rates. The majority of patients (71%) were serologically nonimmune to HBV upon screening. The highest portion of nonimmune to immune patients (75%) were in patients between the ages of 11 and 18 years of age (P = 0.004; Figure 1A). The percent of nonimmunity was approximately equal between the rheumatology and IBD clinics (73% and 67%, respectively). After providing 1 booster dose of HBV vaccine to 291 out of 409 serologically nonimmune patients (71%), we found that 68% of patients seroconverted. This response varied by age, where patients > 18 years had the highest portion of nonimmunity (47%, P = 0.01) after the booster dose (Figure 1B). A subset of those patients in rheumatology clinic received the rest of the HBV vaccination series as per the clinical algorithm followed at the rheumatology clinic (Figure 2). The serologic data are collected during 1 visit and the booster is given to those that need it at the next visit. The majority of the 118 who had not received HBV booster for lack of serologic immunity had not yet returned for a follow-up clinic visit at the time the data were analyzed. Further data about their response to the 3-dose series were not available.
Serologic immunity to HBV by age group. Y-axis indicates percentage levels. Absolute numbers of immune and nonimmune patients are outlined within bars. Panel (A) shows the initial HBV serologic immunity across age groups. Panel (B) shows the serologic immunity response to 1 dose of HBV vaccine across age groups. HBV: hepatitis B virus.
Flowchart of rheumatology and IBD patients in this cohort through screening and immunization for HBV. a For a total of 3 doses. IBD: inflammatory bowel disease; HBV: hepatitis B virus.
Factors associated with lack of serologic immunity to hepatitis B. Several factors were assessed for their effect on the status of serologic immunity for HBV (Table 2). Being between the ages of 11 and 18 years was significantly associated with serologic nonimmunity (P = 0.004). Neither age at diagnosis (P = 0.64) nor duration of treatment (P = 0.07) had an effect on HBV serology upon screening. The risk of lacking serologic immunity was not different between the different disease categories (P = 0.34) or between the different classes of immunosuppressant medications (P = 0.08). In addition, the same factors were taken into account to assess seroconversion and only age showed statistical significance (Table 3).
Demographics and clinical characteristics by initial serologic immunity status.a
Demographics and clinical characteristics by serologic immunity status after 1 booster dose of HBV vaccine.a
Among the patients screened during this period of time, 1 patient was found to have evidence of exposure to HBV infection indicated by the presence of anti-HBc. The patient was reported to have completed HBV primary vaccination. It was thought that HBV was transmitted perinatally since the mother has history of HBV. This patient was treated with IFX upon screening and continues to be closely monitored by the departments of rheumatology and hepatology for any clinical or laboratory evidence of HBV reactivation while continuing to receive IFX.
DISCUSSION
This study revealed that the majority (71%) of patients classified as immunocompromised in our pediatric rheumatology and IBD clinics lacked serologic immunity against HBV. Other cohorts with pediatric patients with IBD also reported serologic nonimmunity in more than half of their patients.5,13 Upon stratification by age in our cohort, patients between the ages of 11 and 18 years stood out as the group most likely to lack immunity, as well as the age range most likely to respond to a single booster vaccine. Thus, preemptive administration of HBV booster vaccine may be appropriate for those between 11–18 years of age, in particular when recent HBV serologic data are not available.
The use of anti-TNF agents was not associated with lack of protective anti-HBs in this analysis. This was also the case in a pediatric IBD cohort by Moses, et al.5 However, our analysis is limited, since only the current use of medications was documented, and recent history of medications was not accounted for. The lack of TNF response in this population is underscored since TNF plays an important role in the suppression of viral replication and the proliferation of HBV-specific cytotoxic T lymphocytes.14,15 As a result, several studies reported cases of HBV reactivation in patients on anti-TNF therapy.7,16,17,18
Most of our patients who returned for repeat testing (68%) seroconverted after a single dose of HBV vaccine. This rate of seroconversion is comparable to other pediatric rheumatology and IBD cohorts as well as patients with rheumatoid arthritis.5,19,20,21 In one of the pediatric IBD cohorts, Urganci, et al compared seroconversion after HBV booster in patients with IBD and healthy controls and found rates of 70.2% and 90%, respectively.19 This implies the presence of IBD-related factors that contribute to lack of immunogenicity of the vaccine when compared to healthy subjects. These can be attributed to the disease itself or the use of immunosuppressant medications. Exploring factors contributing to seroconversion rates in the immunocompromised population can be of great value. Previously, the use of prednisone, DMARDs, or biologic medications had no influence on HBV vaccine immunogenicity in rheumatology patients.20,22 However, the use of anti-TNF agents in IBD cohorts was associated with a significantly lower seroconversion rate.23,24 It is important to note that with regards to HBV reactivation, the type of immunosuppressive therapy effects the risk. This has led to risk-stratification per the type of immunosuppressive therapy.10,25
In our cohort, age was identified as the only relevant factor. Several studies in healthy populations have also reported age as a factor that negatively affects HBV seroconversion rates.26,27,28 Accordingly, we propose that future work determine if those 18 years and older should all receive the complete 3-dose series as opposed to an initial booster with repeat serology per the algorithm (Figure 2). The overall rate of seroconversion after a 3-dose HBV vaccine series has yielded a higher seroconversion rate of 93–97% in rheumatology patients.22,29 At the time of analysis of this study, only a minority of patients had completed the 3-dose series. Therefore, an overall assessment of the rate of seroconversion after the full series is not available.
Patients with rheumatic diseases are particularly at high risk for HBV reactivation. Changes in T cell lymphocyte dynamics and homeostasis seen in patients with RA30,31 could point to a potential inherent susceptibility to HBV reactivation in rheumatic patients. In addition, the use of immunosuppression in patients with rheumatic diseases is largely linked to HBV reactivation, as it influences cellular and humoral responses essential for fighting HBV.32
Although HBV reactivation has been reported in a patient with adequate anti-HBs levels on immunosuppressive treatment,33 lack of anti-HBs has been linked to HBV reactivation.34,35,36 While this highlights the importance of positive surface antibody serology for protection against HBV, further studies are needed to understand the long-term effect of serologic immunity in immunocompromised patients. This is distinctly important because immunocompetent individuals may continue to have immunologic memory despite declining anti-HBs levels.37 The Center for Disease Control and Prevention’s recommendations to screen patients for HBV include those who are about to commence immunosuppressive treatment.38 Despite this, the American College of Rheumatology (ACR) and the European Alliance of Associations for Rheumatology (EULAR) do not include screening for HBV as part of their current guidelines.39,40 The guidelines note that HBV vaccination can be safely administered to patients who are at general risk of contracting HBV (intravenous drug abuse, multiple sex partners in the previous 6 months, healthcare personnel, or living in endemic areas).39,40 In clinical practice, the rate of HBV universal screening for rheumatology patients undergoing immunosuppressive treatment varies. In a survey administered to members of the ACR, 69% reported performing universal screening for HBV prior to starting biologics, and 42% prior to DMARDs.41 The lack of consensus on HBV screening for this patient population is likely behind the low numbers reported. It is ideal to screen and vaccinate patients prior to initiation of immunosuppression therapy. This is encouraged and outlined in several other societies’ guidelines.10,25,42 However, this may not be possible due to timing, and so screening can still be sought after therapy has started. Screening of oncology patients has been shown to be cost-effective in those starting rituximab, among other immunosuppressive agents.43 A decision model developed by Eckman, et al highlighted the cost-effectiveness of universal screening for HBV in the general population even when the prevalence is 0.3%.44 This further emphasizes the ultimate need for comprehensive HBV screening in immunosuppressed patients. We demonstrated that the rate of past infection is 0.2% in our pediatric cohort. This low rate is presumed to be a result of the universal vaccination that started in 1991. However, thorough screening with history and laboratory investigation should not be undermined specially in immunocompromised populations.
In summary, this study emphasizes the need for comprehensive screening with HBsAg, anti-HBc, as well as anti-HBs to thoroughly assess for immunity and risk of HBV. Ideally this HBV serologic screening would be performed prior to the initiation of immunosuppressive therapy. However, we have demonstrated in this study that the useful results are obtained during immunosuppressive therapy as well. Beginning at age 11 years, most patients lacked serologic immunity to HBV. Regardless of current immunosuppressive medications, the administration of HBV vaccine resulted in seroconversion after 1 booster dose in the majority of those between the age of 11 and 18 years who had follow-up testing. Thus, for those immunocompromised rheumatology and IBD patients who do not have recent HBV serologic data, one might consider obtaining this serologic screening at age 11 years. Because only half of patients who were 18 years and older seroconverted following 1 booster vaccine, this age group may benefit from receiving the complete HBV vaccine series. Efforts should be directed toward expanding and then sustaining the screening process for HBV into routine clinical care in this immunocompromised population.
ACKNOWLEDGMENT
The authors of this paper would like to acknowledge Dr. Marie Raphaelle Jean and Pam Morgan from the Department of Pediatric Gastroenterology for provision of the data on patients with IBD.
Footnotes
The authors have no conflicts of interest to disclose.
- Accepted for publication July 16, 2020.
- © 2021 The Journal of Rheumatology
