Corticosteroids induce HBV reactivation by two mechanisms: suppression of cytotoxic T cell function and direct stimulation of the HBV genomic sequence (3). A retrospective study of HBsAg-positive patients with asthma or chronic obstructive pulmonary disease found that HBV reactivation occurred in 15.8% of patients treated with systemic corticosteroids continuously for 3 months, substantially higher than the 3.2% of patients treated with inhaled corticosteroids (odds ratio [OR] 5.72; 95% CI 1.17–27.91) (37). Among patients receiving systemic corticosteroids, a higher risk of reactivation (14.0%) was reported in those receiving corticosteroid doses >20 mg/day prednisolone or equivalent compared to those receiving low-dose systemic corticosteroid (4.5%) (37). Patients receiving low-dose corticosteroids, containing <10 mg/day prednisolone or equivalent, showed low risk of HBV reactivation even after prolonged use (10).

Treatment with a dosage of >20 mg/day of prednisolone, or its equivalent, and a duration longer than 2 weeks is generally considered clinically significant immunosuppression (38). Accordingly, a short period of corticosteroid therapy of <1 week was believed to have a low risk of causing HBV reactivation. However, a recent study found that the risk of hepatitis flare began to increase in chronic hepatitis B patients who received high peak doses of corticosteroid of >40 mg of prednisolone or equivalent, even after <7 days of administration (39). In patients receiving high-dose corticosteroid (>40 mg of prednisolone or equivalent), the risk of hepatitis flare was close to 10% (9.3–11.6%) for which antiviral prophylaxis was recommended, regardless of treatment duration (39).

Anthracyclines such as doxorubicin and epirubucin are widely used chemotherapeutic agents either alone or as a combination therapy in the treatment of breast cancer, ovarian cancer, lymphoma, and sarcoma. This drug class is a representative chemotherapeutic agent known to be associated with a significant risk of HBV reactivation (10). A prospective study found that the HBV reactivation rate was as high as 41% in HBsAg-positive patients receiving conventional chemotherapy for breast cancer, which had a significant impact on early termination or delay of the treatment schedule (40). Based on the results of that study, surveillance and prophylaxis of HBV infection has been widely practiced.

Concomitant use of immunosuppressive or immunomodulatory agents may increase the intensity of immunosuppression, resulting in an increased risk of HBV reactivation. A randomized controlled trial of HBsAg-positive patients receiving ACE (epirubicin, cyclophosphamide, and etoposide) or PACE (prednisolone plus ACE) chemotherapy found that a corticosteroid-free regimen had a significantly lower risk of HBV reactivation, with cumulative incidences of HBV reactivation of 38% and 73% for the ACE and PACE groups, respectively (relative risk 2.36; 95% CI 1.05–5.29) (41).

The highest risk of HBV reactivation has been reported in patients undergoing hematopoietic stem cell transplantation (HSCT), who typically receive intensive chemotherapy for the underlying malignancy to induce remission, followed by additional chemotherapy or radiation therapy for myeloablation. The HBV reactivation rate in patients with resolved HBV infection who received HSCT was 19.7%, and the 1-, 2-, and 4-year cumulative probabilities of reactivation were 9.0, 21.7, and 42.9%, respectively (42). Similar reactivation rates were reported in another study, with 1- and 5-year cumulative rates of 13% and 22%, respectively (43).

Rituximab is a monoclonal antibody against CD20 expressed on the surface of B lymphocytes, which targets and destroys B cells. It is used to treat hematologic malignancies and various inflammatory rheumatic diseases. B cells contribute to HBV clearance by producing neutralizing antibodies, preventing the spread of viruses, and eliminating circulating viruses. Although T lymphocytes-mediated immune control is thought to be the predominant mechanism by which HBV replication is inhibited (44), the compelling evidence of B cell immunity suppression, leading to HBV reactivation, suggests that B cells perform additional functions in suppressing HBV replication in addition to producing neutralizing antibodies (2).

The reported incidence of HBV reactivation after rituximab use was 3–55%, and the overall mortality rate from reactivation was reported to be as high as 30–38% (45). A recent prospective study of patients receiving obinutuzumab- or rituximab-containing immunochemotherapy showed HBV reactivation occurred in 8.2% of the patients, a median 125 days after the first dose (46). In patients without prophylactic antiviral therapy, 10.8% showed HBV reactivation, whereas in patients receiving prophylactic antiviral therapy, 2.1% showed HBV reactivation, suggesting that prophylactic antiviral therapy significantly reduced the risk of reactivation (adjusted HR 0.08, 95% CI 0.02–0.41) (46).

TNF-α is a well-known proinflammatory agent, therefore biologics blocking TNF-α and related cytokine pathways have been used in various inflammatory and autoimmune diseases (47, 48). Its widespread use in various diseases has revealed the relevance of TNF-α to HBV reactivation. It was initially thought that these cytokines exert subtle regulation of the adaptive immune system responsible for HBV immune control, unlike other cytokines that induce non-specific inflammation. In particular, TNF-α can activate a unique host antiviral pathway, the APOBEC (apolipoprotein B mRNA editing enzyme, catalytic polypeptide-like) proteins, which cause the degradation of cccDNA in HBV-infected cells (49). Therefore, blocking this TNF-α-mediated antiviral pathway may cause HBV replication and reactivation.

A recent study of the incidence of HBV reactivation in patients treated with TNF-α inhibitor estimated a pooled incidence of HBV reactivation of 4.2% (95% CI 1.4–8.2%). The pooled incidence of reactivation in patients with resolved HBV infection was 3.0% (95% CI 0.6–7.2%) compared to 15.4% (95% CI 1.2–41.2%) in HBsAg-positive patients (50). Similarly, in a large retrospective study of patients receiving long-term TNF-α inhibitors, an HBV reactivation rate of 39% was reported in HBsAg-positive patients, whereas no reactivation occurred in patients with resolved HBV infection (6). In patients with resolved HBV infection, long-term biologic therapy using TNF-α inhibitors is not associated with an increased risk of HBV reactivation, and the risk of reactivation is negligibly low to none (51).

HBV reactivation have not so frequently reported with regard to transarterial chemoembolization (TACE) in patients with HBV-related HCC, as that of systemic chemotherapy. TACE is a kind of locoregional chemotherapy using Lipiodol which carry and localize chemotherapeutic agents inside the tumor (52). However, TACE and other local therapies for hepatocellular carcinoma can also contribute to HBV reactivation to different extents. Various results have been reported, showing a low risk of reactivation (53), to studies suggesting that reactivation occurs in 17.5–33.7% of patients with TACE, suggesting the association between TACE and reactivation of HBV (54, 55). The effect of TACE can be attributed mainly to the use of anthracyclines.

Activation of various kinase signaling pathways is essential for immune activation and lymphocyte proliferation (56). Many kinase inhibitors targeting these critical pathways have been developed to treat hematologic or other malignancies. Considering the importance of HBV-specific lymphocytes in the immune regulation of HBV replication, these kinase inhibitors may suppress the immune control of HBV replication, resulting in HBV reactivation. Several in vitro studies have shown that Bcr-Abl TKIs exert inhibitory effects on T cells (57), suggesting that T cell inhibition induced by Bcr-Abl TKIs may be a potential risk factor for HBV reactivation.

Several case reports and small-scale studies have suggested that tyrosine kinase inhibitors (TKIs), commonly used to treat chronic myeloid leukemia (CML) and gastrointestinal stromal tumors, may be associated with an increased risk of HBV reactivation (58–61). Two large-scale retrospective studies evaluating the risk of HBV reactivation in TKI-treated CML patients with positive HBsAg reported HBV reactivation rates of 26–34.8% (61, 62). More recently, a nationwide nested case-control study with 733,691 patients with HBV in Taiwan found that Bcr-Abl TKI use was independently associated with HBV reactivation (aHR, 1.56; 95% CI 1.11–2.20) (63).

Calcineurin inhibitors such as cyclosporine or tacrolimus suppress T cell function by inhibiting calcineurin required for signal transduction of T cell activation and inhibiting transcription of interleukin required for T cell proliferation (64). These immunosuppressive effects leading to broad immune dysfunctions could potentially lead to HBV reactivation. There have been several reports of HBV reactivation associated with the use of tacrolimus or cyclosporine in patients with rheumatic disease, some of which had serious consequences leading to death or liver transplantation (65, 66).

There is no convincing evidence of HBV reactivation caused by antimetabolites monotherapy such as azathioprine, 6-mercaptopurine or methotrexate. Despite long-standing clinical use of azathioprine and its active metabolite, 6-mercaptopurin, there is no documented cases in which azathioprine or 6-mercaptopurin leads to HBV reactivation. Regarding methotrexate, a prospective study reported that HBV reactivation occurred in two of five patients with HBsAg and in one of 45 patients without HBsAg after methotrexate treatment in patients with rheumatoid arthritis (67). It is not certain whether methotrexate induced HBV reactivation, as this study included patients additionally receiving TNF-α to methotrexate. On the contrary, another cross-sectional study reported that no HBV reactivation was occurred with long-term use of methotrexate in rheumatologic patients (68).

Although direct-acting antivirals (DAAs) are not considered immunosuppressive agents, DAAs targeting the hepatitis C virus (HCV) may result in HBV reactivation in patients co-infected with HBV and HCV. From 2013 to 2016, the U.S. Food and Drug Administration (FDA) identified 29 cases of H0BV reactivation in patients with HBV-HCV co-infection treated with DAAs, two of which resulted in death and one required liver transplantation (69). A recent meta-analysis of HBV reactivation in HBV-HCV co-infected patients showed that the estimated HBV reactivation rate in studies with interferon-free DAA-based therapy (12.2%) was not significantly different from interferon-based therapy (14.5%) (70). However, HBV reactivation was reported to occur much earlier with DAAs (4–12 weeks during treatment) than with interferon-based therapies (end of treatment or thereafter), and studies with DAA-based therapies were more likely to report hepatitis due to HBV reactivation (12.2% in DAAs vs. 0% in interferon) (70). Another meta-analysis demonstrated that HBV reactivation occurred in 24% of patients with positive HBsAg, compared to 1.4% in patients with resolved HBV infection after DAA treatment (8). Among the patients with resolved HBV infection, the titer of anti-HBs was reported to significantly decrease from the early stage of DAA treatment, suggesting patients with a negative or low titer of anti-HBs at baseline have potential risk of HBV reactivation (71).

In HBV-HCV co-infected patients, there are different viral dominance patterns, mostly HCV dominant, and the levels of HBsAg were found to be closely associated with viral dominance patterns (72). High levels of interferon γ-induced protein 10 (IP-10) were induced by HCV in HCV-dominant HBV-HCV co-infected patients, which is consistent with a low level of HBsAg (72). It can be hypothesized that HCV suppresses HBV DNA replication and HBsAg production by immune mechanisms such as the induction of IP-10. In these circumstances, HBV replication may be promoted when the immune mechanisms that suppress HBV replication are eliminated by DAA treatment for HCV infection.

Immune checkpoint inhibitors target key regulators of immune system to suppress various kinds of tumors. Programmed cell death 1 (PD-1) and programmed death-ligand 1 (PD-L1) inhibitors are representative immunotherapies that prevent the escape mechanism of tumor cells and promote restoration of T cell function. Cytotoxic T-lymphocyte antigen-4 (CTLA-4) is a negative regulator of T cell activation, and inhibition of CTLA-4 induces T cell activation and immune surveillance against cancer cells. There are rare case reports of HBV reactivation associated with immune checkpoint inhibitors, and only reports related to nivolumab (73), pembrolizumab (74) and ipilimumab-nivolumab sequential treatment (75). Theoretically, increased T cell activity induced by immune checkpoint inhibitors may provide durable control of HBV. A pilot study reported that nivolumab treatment was well-tolerated and reduced HBsAg titer in most HBeAg-negative patients with suppressed HBV viral load (76).

There is a risk of HBV reactivation in severe SARS-CoV-2 infected patients. Disruption of the balance between the host's immune status and viral replication contributes to HBV reactivation after SARS-CoV-2 infection, and the intensity of immunosuppressive therapies is a major risk factor for HBV reactivation. HBV reactivation in patients infected with SARS-CoV-2 is usually associated with immunosuppressive therapy such as IL-6 or IL-1 receptor antagonists (tocilizumab, anakinra), and high-dose corticosteroids (77, 78). There are several reports of HBV reactivation in SARS-CoV-2 infected patients. A retrospective study found that three out of 21 patients with SARS-CoV-2 and HBV coinfection developed HBV reactivation, of which two received corticosteroid therapy (79). The most recent prospective study evaluated the risk of HBV reactivation in HBsAg-/anti-HBc+ patients with severe SARS-CoV-2 infection receiving immunosuppressive therapy (80). At 1-year follow-up, there were no cases of HBsAg seroconversion and two out of 69 had detectable serum HBV DNA, suggesting a low risk of HBV reactivation in patients with severe SARS-CoV-2 infection and resolved HBV infection.