Hepatitis B reactivation with TNF-α inhibitors: assessing antiviral prophylaxis efficacy — protocol for systematic review and meta-analysis

Abstract

Background

The widespread use of TNF-α inhibitors (infliximab, adalimumab, golimumab, certolizumab pegol, and etanercept) in gastrointestinal, rheumatologic, and dermatologic disorders raises concerns about hepatitis B reactivation (HBVr). The exact risk remains unclear due to the variability in previous meta-analyses. This study aims to assess the effectiveness of antiviral prophylaxis in preventing HBV reactivation in patients on TNF-α therapy and to investigate the associated reactivation risk.

Method

A systematic review adhering to PRISMA guidelines will be conducted. Comprehensive searches of electronic databases (MEDLINE via PubMed, Google Scholar, CENTRAL, and ClinicalTrials.gov) will identify relevant studies. Eligible studies will include patients with a hepatitis B infection treated with anti-TNF-α therapy. The efficacy of antiviral prophylaxis will be assessed using risk ratios (RR) with 95% confidence intervals (CI) in a random-effects model that controls for variation among trials. Heterogeneity will be evaluated using the I² statistic and Cochran’s Q test, with I² > 50% or p < 0.10 indicating significant heterogeneity. Subgroup analyses will explore sources of heterogeneity, such as the type of antiviral medication, HBV serostatus, and the type of anti-TNF agent. Publication bias will be assessed using funnel plots and Egger’s test. The incidence of HBV reactivation will be estimated using pooled estimates and 95% CIs in a random-effects model, excluding patients receiving antiviral prophylaxis.

Discussion

The rate of hepatitis B virus (HBV) reactivation varies greatly between chronic and occult carrier states. Despite established practice standards, high-risk individuals still get insufficient preventative antiviral treatment. Current recommendations propose prophylactic nucleoside/nucleotide analogue (NA) prophylaxis for immunosuppressive treatment that has a high risk of reactivating HBV. For low-risk treatments, on-demand NA treatment is recommended. For intermediate-risk drugs, either strategy might be suitable. However, there is an urgent need for agreement on standardized criteria and reporting protocols for HBV reactivation in the setting of immunosuppressive treatments.

Trial registration

Systematic review registration:

PROSPERO CRD42024548106.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13643-025-02923-y.

Background

Tumor necrosis factor-α inhibitor (anti-TNF-α) have emerged as a transformative therapy for chronic inflammatory diseases, demonstrably improving outcomes across diverse conditions like inflammatory bowel disease (IBD), rheumatologic disorders, and dermatologic manifestations [1]. However, its usage raises concerns about the reactivation of a hepatitis B virus (HBVr). The research on HBV reactivation with TNF-α inhibitors has primarily focused on rheumatologic and dermatologic investigations, as they have larger patient populations than IBD research [2–5]. The pivotal role of TNF-α in regulating inflammation, immune responses, and immune complex clearance underscores the potential risk of HBV reactivation despite the therapeutic success of TNF-α inhibitors [6]. This risk is further influenced by factors such as the host’s immune response, the phase of HBV infection, and the specific immunosuppressive agent used [7–9]. Although current evidence suggests that antiviral therapy, with or without discontinuation of biological agents, can effectively resolve acute reactivation [7, 9–11], reports of subfulminant hepatic failure necessitate further investigation [12, 13]. This meta-analysis aims to determine the risk of hepatitis B reactivation with TNF-α inhibitors, stratified by HBV status (carrier, occult, resolved), while evaluating factors such as antiviral prophylaxis and concurrent immunosuppressants. A key innovation is addressing the variability in prior meta-analyses [5, 7, 14, 15], which often reported inconsistent findings due to differences in study designs and reactivation definitions. By harmonizing diverse criteria and incorporating studies with varying treatment durations, this analysis provides a standardized assessment. With an estimated 316 million people globally affected by chronic hepatitis B, the findings will guide treatment decisions and inform strategies to mitigate reactivation risks in immunocompromised patients.

Methods/design

This meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) 2020 guidelines [16, 17] and the Meta-analysis of Observational Studies in Epidemiology (MOOSE) standards [18]. To enhance transparency and facilitate the assessment of reporting completeness, a checklist systematically mapping the protocol’s content to the PRISMA 2020 reporting criteria has been developed and included as a supplementary table (or appendix) in the revised manuscript. A comprehensive literature search was performed to identify relevant publications, ensuring methodological rigor and adherence to best practices.

Criteria for eligibility

Participants

This study will evaluate the efficacy of antiviral prophylaxis in preventing HBV reactivation in immunosuppressed patients with chronic or occult HBV infection, defined by detectable HBV DNA, positive HBsAg, or positive anti-HBc, but without clinical evidence of active HBV replication or liver disease. The study population will be derived from studies including patients diagnosed with inflammatory bowel disease (IBD), rheumatologic disorders, or dermatologic conditions (according to established diagnostic criteria) who have received anti-TNF-α therapy for a minimum of 12 weeks. Inclusion criteria will not discriminate based on age, gender, ethnicity, or nationality; HBV status will be determined by either a positive HBsAg or anti-HBc test. Studies will be excluded if they involve pediatric populations, are case reports, include patients receiving multiple concurrent biologic agents, are review articles, or have sample sizes less than 3 [5, 7], or an anti-TNF-α treatment duration shorter than 12 weeks was not considered.

Intervention

This systematic review will evaluate studies assessing the efficacy of antiviral prophylaxis for HBV in patients with inflammatory bowel disease (IBD) and rheumatologic and dermatologic diseases undergoing anti-TNF-α therapy. Data extraction will encompass the type and duration of antiviral prophylaxis and its correlation with the initiation and discontinuation of anti-TNF-α therapy. The interventions of interest include standard antiviral prophylactic agents such as tenofovir, entecavir, and lamivudine. Studies including a comparator group of patients not receiving prophylactic antiviral treatment will be considered.

Comparators

Participants with comparable baseline characteristics who did not receive antiviral prophylaxis for HBV served as a control group to evaluate the effectiveness of the intervention.

Outcome measures

Primary outcomes

This systematic review and meta-analysis will look into the risk of HBV reactivation in individuals undergoing anti-TNF-α therapy. The research will assess the effectiveness of hepatitis B antiviral prophylaxis and compare reactivation rates with different TNF-α medications. The findings aim to improve clinical practice recommendations and management of HBV-infected individuals requiring anti-TNF-α medication. The primary outcome measure will be the incidence of HBV reactivation, defined by an elevation in HBV DNA levels, the reappearance of HBsAg, or the development of clinical hepatitis, as assessed through HBV DNA quantification and liver function tests.

Secondary outcomes

Secondary endpoints will include the incidence of adverse events associated with HBV reactivation episodes and any potential side effects related to the treatment regimen. Additionally, secondary outcomes will evaluate adverse events linked to anti-TNF-α therapy and the development of other liver-related complications. These will encompass liver-related complications (such as ascites, hepatic encephalopathy, and liver failure), the effectiveness of antiviral prophylaxis in preventing reactivation, mortality rates (both HBV related and all-cause), adverse events related to prophylaxis, and the timing of reactivation in relation to treatment and follow-up. Each secondary outcome will be assessed using relevant clinical and laboratory data, followed by statistical analysis to offer comprehensive insights into the study’s results.

Search strategy

A comprehensive systematic search will be conducted across multiple electronic databases, including Cochrane Central Register of Controlled Trials, PubMed, ClinicalTrials.gov, and Google Scholar, utilizing database-specific syntax with detailed field codes (e.g., [tiab], [MeSH]), Boolean operators (AND/OR), and truncation. For PubMed, the strategy combines MeSH terms (e.g.,"Hepatitis B/drug therapy"[Mesh],"Tumor Necrosis Factor-alpha/antagonists & inhibitors"[Mesh]) and free-text keywords (e.g.,"HBV reactivation,""HBsAg positive,""viral reactivation,""recurrent infection,""infliximab,""adalimumab,""golimumab,""certolizumab pegol") tailored to capture studies on hepatitis B reactivation, anti-TNF-α therapies, and associated complications (e.g., hepatic encephalopathy, liver failure). Similar strategies will be explicitly designed for other databases (e.g., Embase via Ovid, Scopus) with full syntax—including operators, proximity commands, and filters—provided in supplementary materials to ensure reproducibility. Non-English literature will be included through searches in databases covering multilingual content (e.g., LILACS, SciELO) and translation protocols using professional services or validated tools for critical articles, with non-English studies restricted to languages with institutional translation capacity (e.g., Chinese, Spanish, French). Exclusion of untranslatable studies will be documented. All search strategies will undergo peer review via the PRESS checklist to optimize sensitivity and precision.

Supplementary search strategies

To enhance the comprehensiveness of the review, backward citation searches will be conducted on all included studies and relevant reviews to identify additional references not captured in the initial database search. Similarly, forward citation searches will be performed using Google Scholar to identify recently published studies citing the included articles. This approach will help incorporate the most up-to-date findings into our analysis.

Search time frame and article retrieval

The search will cover studies published up to [31-7-2025]. If full-text access to any key articles is unavailable, we will contact the corresponding authors directly to request the necessary information.

Hepatitis B reactivation (HBVr) definitions

Hepatitis B virus (HBV) infection is classified into chronic and occult carrier states based on hepatitis B surface antigen (HBsAg) status, with chronic carriers being HBsAg positive and occult carriers being HBsAg negative but having detectable serum HBV DNA, typically < 200 IU/mL, regardless of hepatitis B core antibody (anti-HBc) status, as per EASL guidelines [10]. Hepatitis B reactivation (HBVr) is defined as the resurgence of viral replication, characterized by the reappearance of HBsAg or HBV DNA in occult carriers, a ≥ 2 log (100-fold) increase in HBV DNA from baseline in chronic carriers, or de novo detection of HBV DNA > 20,000 IU/mL if baseline levels were undetectable [19]. Definitions of HBVr vary across guidelines: AASLD (2018) defines HBVr as HBV DNA > 10,000 IU/mL, particularly when baseline viral load is unknown, while APASL requires a ≥ 2 log increase in HBV DNA or HBV DNA > 100 IU/mL in patients with previously detectable virus or > 20,000 IU/mL in those with undetectable baseline levels [19, 20]. Given this heterogeneity, this systematic review will categorize studies based on the criteria used (AASLD, APASL, EASL, or author-defined), conduct sensitivity analyses to assess the impact of differing definitions on pooled outcomes, and carefully extract and analyze outcomes separately to avoid double-counting, particularly in studies reporting both antiviral prophylaxis efficacy and HBVr incidence. Studies with unclear or overlapping definitions will undergo subgroup scrutiny to ensure rigorous evidence synthesis, accounting for historical variations, especially in studies published before 2015 when author-defined criteria were commonly used.

Study selection

We will employ a systematic and transparent search strategy to identify eligible studies. Two independent reviewers will conduct a two-stage screening process: initially, titles and abstracts from comprehensive searches of electronic databases (PubMed, Cochrane Library, ClinicalTrials.gov, and Google Scholar) will be assessed, followed by a full-text analysis of possibly suitable studies. Two independent reviewers will screen study titles and abstracts, and any differences will be handled by a third reviewer. Discrepancies during full-text screening will be handled by consensus or consultation with a third reviewer. Inclusion criteria include observational studies (in any language) on patients with gastroenterological, dermatological, or rheumatic diseases who received anti-TNF-α medication for at least 12 weeks. Studies will be excluded if they involve pediatric populations, case reports, patients receiving concurrent biological agents, or review articles. Additionally, studies with a sample size of fewer than three patients will be excluded" [5, 7, 15]. Studies with treatment durations shorter than 12 weeks will be analyzed separately to assess the risk and timing of early HBV reactivation. This subgroup analysis will compare reactivation incidence between studies with < 12 weeks of treatment and those meeting the original ≥ 12-week criterion, identifying differences in reactivation patterns and associated risk factors. To ensure methodological rigor, we will assess baseline HBV characteristics, antiviral prophylaxis use, and monitoring protocols within these studies. Additionally, a meta-regression analysis will be conducted, if data allow, to evaluate the influence of treatment duration on reactivation risk. A sensitivity analysis will determine the impact of integrating shorter-length studies on the total pooled estimates, ensuring that findings are robust and not overly impacted by study duration differences.

Data extraction

A standardized, pre-tested data extraction form will be developed to systematically collect detailed information across several domains. The form will include specific fields for study characteristics (e.g., author, publication year, study design, sample size, data collection period, and geographic location), patient demographics (e.g., age, gender, HBV status [HBsAg positive, anti-HBc positive, or resolved infection], and history of prior antiviral treatment), anti-TNF-α therapy details (e.g., drug type [infliximab, adalimumab, golimumab, certolizumab pegol], dosage, treatment duration, and timing relative to HBV reactivation), and primary outcomes (e.g., occurrence and severity of HBV reactivation, liver-related complications [ascites, hepatic encephalopathy, liver failure], and mortality). Operational definitions will be clearly outlined; for example, HBV reactivation will be defined as detectable HBV DNA in previously undetectable cases or a significant increase in HBV DNA levels in patients with chronic HBV. The form will also document antiviral prophylaxis details, including the specific agent used (e.g., tenofovir, entecavir), duration of prophylaxis, and timing relative to anti-TNF-α therapy initiation and discontinuation. To ensure consistency, each field will include clear instructions and definitions to guide reviewers during data extraction. Data extraction will be done separately by two reviewers, with any disagreements handled by discussion or contact with a third reviewer. To handle missing or incomplete data, a structured process will be used: research authors will be contacted via email (with up to two follow-up attempts) to request clarification or more information. If no answer is received, the missing data will be openly recorded, and sensitivity analysis will be done to determine how these gaps may affect the results. When numerous publications report overlapping data from the same study, the most extensive or latest dataset is selected, and sensitivity tests are performed to assess the impact of probable duplication. These measures, including the detailed extraction form structure, operational definitions, and robust procedures for addressing missing data, aim to ensure data integrity, minimize bias, and enhance the reliability of pooled estimates in the systematic review.

Methodological quality assessment and reviewer training

The two reviewers will independently review the methodological quality of the selected articles, with discrepancies handled by consensus or judgment by a third reviewer. To ensure methodological rigor and consistency, all team members underwent training prior to commencing these tasks, including the application of standardized guidelines and participation in pilot exercises. Additionally, a statement addressing potential conflicts of interest has been included in the protocol to enhance transparency throughout the review process.

Quality control

To ensure inter-rater reliability during the data extraction process, we have incorporated a detailed quality control procedure in the protocol. Specifically, two reviewers will independently extract data from each study to minimize bias and enhance consistency. Any discrepancies between the reviewers will be resolved through discussion, and if consensus cannot be reached, a third reviewer will adjudicate. Additionally, inter-rater reliability will be assessed using the kappa statistic or, if deemed more appropriate, another measure of agreement such as the intraclass correlation coefficient (ICC), percent agreement, Cohen’s weighted kappa, or Krippendorff’s alpha. This flexibility allows the selection of the most suitable metric based on the type of data and context of the study, ensuring a robust validation of the consistency and interchangeability of the reviewers.

Risk of bias

The risk of bias will be evaluated using the ROBINS-I tool for observational studies [21], complemented by the Cochrane Risk of Bias Tool and the Newcastle–Ottawa Scale (NOS). Assessment will focus on key domains, including participant selection, confounding control, group comparability, and measurement accuracy, while incorporating additional domain-specific criteria to ensure methodological rigor such as the following: (1) laboratory monitoring protocols, where studies will be evaluated for standardized HBV DNA quantification methods (e.g., PCR assays), testing frequency (e.g., monthly vs. sporadic), and adherence to guideline-recommended thresholds (e.g., EASL/AASLD cutoffs), with inconsistent or unreported protocols flagged as high risk; (2) HBV reactivation definition consistency, where studies will be assessed for alignment with guideline criteria (AASLD, APASL, EASL) or use of author-defined definitions, with deviations (e.g., unreported DNA thresholds or failure to distinguish occult vs. chronic carriers) noted as potential misclassification bias; and (3) follow-up completeness, where loss-to-follow-up rates and monitoring duration (e.g., < 6 months vs. > 12 months) will be recorded, with studies exhibiting > 20% attrition or inadequate follow-up periods to capture delayed HBVr (e.g., post-immunosuppression) deemed high risk. All authors will independently score studies using pre-agreed criteria, resolving discrepancies by consensus, and sensitivity analyses will stratify results by risk levels (low/moderate/high) to quantify bias impact, with high-risk studies explicitly discussed in the final synthesis and their influence on outcome validity highlighted.

Detection of heterogeneity

Heterogeneity will be assessed using Cochran’s Q statistic (p < 0.10 indicating significance) and quantified by the I² statistic, which categorizes heterogeneity as low (0–25%), moderate (25–75%), or high (> 75%) [22]. To address heterogeneity, a multifaceted approach will be employed. Sensitivity analyses will assess the robustness of findings by systematically excluding studies with a high risk of bias or disproportionate influence on heterogeneity, ensuring consistency across different scenarios. By stratifying data according to pertinent factors, such as patient demographics (e.g., age, sex), hepatitis B virus infection status (e.g., chronic vs. occult), the type of anti-TNF-α therapy administered, and the use of antiviral prophylaxis, subgroup analyses will investigate potential sources of heterogeneity. Additionally, meta-regression will be applied, where appropriate, to identify and quantify the impact of continuous and categorical covariates contributing to heterogeneity. For cases of significant heterogeneity, a random-effects model will be utilized to provide more conservative effect size estimates while accounting for inter-study variability. The magnitude and sources of heterogeneity will be transparently reported, with a thorough discussion of the clinical and methodological factors contributing to these findings included in the final manuscript. This comprehensive approach ensures the rigorous evaluation and management of heterogeneity throughout the analysis and interpretation process.

The protocol would benefit from addressing domain-specific challenges unique to studies on HBV reactivation, particularly in the risk-of-bias assessment. While the ROBINS-I tool provides a general framework, additional criteria should be developed to evaluate key aspects such as laboratory monitoring protocols, consistency in defining HBV reactivation, and completeness of follow-up data, all of which are critical in reactivation studies. Furthermore, the subgroup analysis plan should be expanded to include additional effect modifiers, such as concurrent immunosuppressive medications, different disease indications (e.g., IBD, rheumatologic, and dermatologic conditions), and geographic regions with varying HBV prevalence. Lastly, the data extraction process requires more detailed protocols for handling missing or unclear data, including contacting study authors for additional information, imputing unreported statistics like standard deviations, and systematically documenting decisions to ensure transparency and consistency in the analysis.

Assessment of publication bias

To assess publication bias, a multi-faceted approach will be employed. Funnel plots will be generated to visually inspect asymmetry, with specific criteria for interpretation: asymmetry suggesting potential bias will be evaluated based on the distribution of effect sizes relative to study precision, particularly focusing on smaller studies with disproportionately large effects. Egger’s test (unweighted) will be used to statistically assess funnel plot asymmetry, with a two-sided p-value < 0.05 considered significant only if more than 10 studies are included [23]. If publication bias is detected, adjustments to the results will be made using the trim-and-fill method, which imputes missing studies to estimate a corrected effect size [23], and the influence of these adjustments on the overall findings will be explicitly reported. Sensitivity analyses will be conducted by sequentially excluding smaller studies or those with outlier effect sizes to evaluate the robustness of the results [24]. Stratified analyses may also be performed based on study characteristics such as sample size, funding sources, and methodological quality to identify potential sources of bias [25]. All findings, including the results of funnel plot interpretations, Egger’s test, trim-and-fill adjustments, and sensitivity analyses, will be transparently reported. The impact of detected bias on the final conclusions will be clearly discussed, ensuring the reliability and validity of the systematic review’s findings. Forest plots will be used to visually support the interpretation of results and any adjustments made.

Statistical analyses

The primary outcome measure of this meta-analysis will be the risk difference (RD) for HBV reactivation rates. This will be calculated by subtracting the proportion of patients experiencing HBV reactivation in the non-prophylaxis group from the proportion in the antiviral prophylaxis group. Secondary outcomes will encompass time-to-event data, analyzed using hazard ratios (HRs). This will assess the impact of prophylaxis on the time to HBV reactivation between the two groups. Additionally, risk ratios (RRs) will be employed to evaluate the risk ratios for binary outcomes, such as the occurrence of complications associated with HBV reactivation episodes. Furthermore, we will perform subgroup analyses to estimate the hepatitis B reactivation rate for each anti-TNF-α agent included in the review. This will evaluate the possible variability in reactivation risk among various TNF-α inhibitors. A random-effects model will be used for the meta-analysis, as it assumes that the true effect size may vary across studies due to differences in populations, designs, or interventions. This model accounts for between-study heterogeneity, providing a more comprehensive and generalizable estimate of the effect size. The choice of this model is supported by the expectation of variability in the included studies, as evidenced by metrics such as an I² statistic exceeding 50% or a p-value for heterogeneity below 0.10, indicating significant heterogeneity. The heterogeneity among studies will be assessed using the I² statistic, which describes the percentage of variation across studies that is due to heterogeneity rather than chance. An I² value greater than 50% will be considered indicative of substantial heterogeneity. The publication bias will be evaluated using funnel plots and Egger’s test. Asymmetry in the funnel plot or a significant Egger’s test may indicate the presence of publication bias. The sensitivity analyses will be conducted to assess the robustness of the results. This may include excluding studies with high risk of bias, using alternative statistical models, or analyzing subgroups based on study characteristics. We will perform a proportional meta-analysis to synthesize the data and derive a pooled estimate of the outcome. To investigate heterogeneity and examine the influence of study-level factors, we will conduct subgroup analyses by stratifying studies according to their key characteristics (e.g., patient population, study design) and by specific outcomes. Given the low frequency of HBVr, we calculated weighted proportions using the Freeman-Tukey transformation under both fixed and random-effects models [26]. Case series were defined as reports involving at least three patients were included [5, 7], recognizing possible heterogeneity due to case selection, demographics, study designs, and different dosing regimens for induction and maintenance in psoriasis compared to inflammatory bowel disease. We employed a random-effects model for analysis, utilizing prevalence rates (event rates) with 95% confidence intervals (CI) [25]. Data analysis will be performed using R software, specifically utilizing the “meta” package for conducting the meta-analysis. Statistical tests, including random-effects models, will be applied to assess the overall effect sizes. Heterogeneity will be evaluated using the I² statistic, and sensitivity analyses will be conducted to examine the robustness of the findings. All analyses will be reproducible, and the specific R scripts used will be available upon request to ensure transparency and replicability of the analysis.

Subgroup analyses

We will conduct subgroup analyses to examine variations in HBV reactivation outcomes based on antiviral type, HBV serostatus, anti-TNF agent, concurrent immunosuppressive therapy, underlying disease indication, and geographic region. Subgroups will be defined as follows: (1) Antiviral type (e.g., tenofovir, entecavir, or lamivudine); (2) HBV serostatus (HBsAg positive vs. anti-HBc positive but HBsAg negative); (3) anti-TNF agent (e.g., infliximab, adalimumab, golimumab, certolizumab pegol, or etanercept); (4) concurrent immunosuppressive therapy, including corticosteroids, methotrexate, azathioprine, and other immunomodulators; (5) underlying disease indication, categorized as inflammatory bowel disease (IBD), rheumatologic diseases (e.g., rheumatoid arthritis, ankylosing spondylitis), or dermatologic conditions (e.g., psoriasis); and (6) geographic region, considering regional variations in HBV endemicity and healthcare access. Studies will be classified into subgroups based on reported characteristics, with mixed populations analyzed separately if data allow. Random-effects meta-regression models and interaction tests will assess differences across subgroups, with qualitative comparisons conducted where meta-regression is not feasible. Recognizing potential limitations, such as small sample sizes and unmeasured confounders, findings will be interpreted cautiously as exploratory analyses. This approach aims to provide a nuanced understanding of factors influencing HBV reactivation risk in patients receiving anti-TNF therapy.

Discussion

Treatment recommendations for biologic therapy in hepatitis B patients show some alignment across major liver disease societies, including the European Association for the Study of the Liver (EASL), the American Association for the Study of Liver Diseases (AASLD), and the Asian Pacific Association for the Study of the Liver (APASL). However, these guidelines primarily address immunosuppressants, while supporting evidence leans heavily on research with oncology drugs. Additionally, a standardized definition for hepatitis B reactivation following immunosuppressive therapy remains lacking [10, 19, 20]. Clinical practice guidelines for hepatitis B reactivation associated with immunosuppressive therapy exhibit convergence on a risk stratification system (low, intermediate, and high) based on the overall reactivation risk. However, significant discord remains regarding the classification of anti-TNF-α therapy. Specifically, the guidelines lack consensus on whether anti-TNF-α therapy poses a high, intermediate, or low risk for reactivation in patients who are hepatitis B surface antigen (HBsAg) negative but hepatitis B core antibody (anti-HBc) positive [9, 11, 27, 28]. Papatheodoridis et al. categorized the risk of hepatitis B virus reactivation (HBVr) associated with anti-TNF-α therapy as unknown for hepatitis B surface antigen (HBsAg)-negative patients and low for hepatitis B core antibody (anti-HBc)-positive patients. However, their valuable meta-analysis did not evaluate the risk for individual anti-TNF-α agents [7]. Aljamali et al. conducted a meta-analysis that estimated the risk of hepatitis B virus reactivation (HBVr) associated with different anti-TNF-α agents. This work addressed the crucial question of heterogeneity in HBVr among anti-TNF-α therapies. However, a potential limitation of their study is the inclusion of patients receiving antiviral prophylaxis. Since prophylaxis can lower the true incidence of HBVr, its presence might introduce confounding bias in the risk estimates for individual anti-TNF-α agents [5]. To address this knowledge gap, we propose a meta-analysis that estimates the risk of hepatitis B virus (HBV) reactivation specifically associated with each anti-TNF-α therapy. By rigorously excluding patients receiving antiviral prophylaxis, this analysis will provide a more precise assessment of the inherent HBV reactivation risk posed by individual anti-TNF-α agents. This information can directly translate into improved clinical decision-making, ultimately enhancing patient care.

Abbreviations

HBV

Hepatitis B virus

TNF-α

Tumor necrosis factor-α

Anti-TNF-α

Tumor necrosis factor-α inhibitor

IBD

Inflammatory bowel disease

HbsAg

Hepatitis B surface antigen

Anti-HBc

Hepatitis B core antibody

Authors’ contributions

MH is the chief investigator and guarantor of this work. He conceived the study, led the protocol development, and designed the research methods (conceptualization, methodology). SS wrote the first draft of the manuscript (writing — original draft). All authors (MH, SS) participated in revising the manuscript (writing — review and editing) and approved the final written manuscript.

MH acted as the guarantor of the review.

Data availability

N/A

Declarations

Ethics approval and consent to participate

This study does not require ethical review or clearance as it involves conducting a meta-analysis based solely on data from previously published studies. No patients or members of the public were involved in the design or planning of this study.

Consent for publication

Not applicable. We have not included any individual’s data in our study protocol.

Competing interests

The authors declare that they have no competing interests. To enhance openness while minimizing possible biases, the research team has taken several measures: all members have disclosed their affiliations, funding sources, and potential competing interests prior to the review, any study authored by a team member will be evaluated by independent reviewers within the team to avoid bias, this review is not funded by any external organization or entity with a vested interest in its outcomes, and the protocol adheres to PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analysis.