Hepatitis B is one of the most frequent forms of viral hepatitis globally and a significant cause of liver-related mortality. It has been estimated that in 2019, 295.9 million people worldwide were chronically infected with hepatitis B virus (HBV), leading to an estimated 821,100 deaths from liver cirrhosis, liver failure, and hepatocellular carcinoma (1).

Patients with chronic HBV (cHBV) are unable to develop an immune response that would be sufficiently robust, functional, and sustained to eliminate the infection (2, 3). Nucleotide or nucleoside analogs (NUCs) are currently the frequently applied therapeutics for cHBV. Antiviral agents like tenofovir and entecavir effectively inhibit the replication of HBV in almost all chronic patients, but hepatitis B surface antigen (HBsAg) loss, which reflects the functional clearance of the virus, is rarely achieved (4). Therefore, long-term or even life-long NUC therapy is required to maintain HBV suppression (4, 5).

Immunological control of HBV infection is critical for the functioning and permanent elimination of the virus. Thus, restoring adequate immune responses to the virus represents a major challenge that can lead to a promising therapeutic strategy (6). CD4⁺ T cells, CD8⁺ T cells, and natural killer (NK) cells play critical roles in the immune response to HBV infection (7). CD4⁺ T cells are essential for orchestrating the immune response, helping activate CD8⁺ T cells, and providing help for B-cell antibody production. Intrahepatic CD8⁺ T cells, mainly the population of HBV-specific CD8⁺ T cells, play a major role in the resolution of HBV chronic infection via their cytolytic and non-cytolytic effector functions. Similarly, NK cells, which are enriched in the liver, are crucial for antiviral immune response, as they can identify and directly kill HBV virus-infected cells with the release of cytotoxic granules and antiviral cytokines, without major histocompatibility complex restriction (8). However, in cHBV infection, both T cells and NK cells have shown an impairment in their function and activity (7, 9, 10).

The antiviral capacity of both T cells and NK cells depends on complex crosstalk within the liver microenvironment, and the expression of immune checkpoint molecules (ICMs) plays a critical role. The upregulation of inhibitory ICMs, such are programmed cell death 1 (PD-1) or T-cell immunoglobulin domain and mucin domain-3 (Tim-3) receptor, contributes to the exhaustion of effector cells that become unable to effectively eliminate the virus and thus favors the chronicity of the disease (11) but may also be important in avoiding excessive immune activation and thereby limiting immune-mediated liver damage. However, stimulatory ICMs, such as 4-1BB and OX40, offer potential avenues to enhance immune responses by restoring T-cell functionality and promoting antiviral activity. Overall, the interplay between inhibitory and stimulatory checkpoint molecules in cHBV infection plays a crucial role in shaping immune responses (10).

Several studies demonstrated that the expression of the inhibitory receptor PD-1 is significantly upregulated on HBV-specific T cells in cHBV infection (12–14), highlighting the role of immune exhaustion in the dysfunction of these cells. In fact, PD-1 expression is associated with diminished proliferative capacity, cytokine production, and cytotoxic activity of HBV-specific CD8⁺ T cells (12, 13). This immune dysfunction is one of the key factors in the pathogenesis of cHBV infection, allowing the virus to evade immune clearance and sustain long-term infection. Blocking the PD-1/PD-L1 axis could restore the reactivity of T cells, mainly the capacity of HBV-specific T-cell populations to expand and to produce cytokines, which could lead to the elimination of HBV (12, 13, 15–17).

The role of PD-1 expression on intrahepatic and circulating T cells in the context of HBV infection is well recognized, and several studies have also explored the modulation of other ICMs on lymphocytes of cHBV-infected patients (10). While much of this research has focused on circulating lymphocytes, there is less information about ICM expression in the intrahepatic environment, despite its importance. Intrahepatic immune cells are unique, with CD8⁺ T cells and NK cells being more prevalent in the liver than in circulation. Their distinct capacity to express specific ICMs could significantly influence antiviral responses. Recent advances, including single-cell RNA sequencing, have shed light on ICM expression in intrahepatic T cells, offering valuable insights into liver-specific immune regulation during HBV infection (18, 19). Although this costly technique can only be currently used on a very limited number of samples, the results of Zhang et al. indicate that each phase of HBV infection may be associated with a unique intrahepatic immune signature (19). In addition, a recent study by Andreata et al. underscores the importance of stimulatory ICMs in cHBV infection. Their research using HBV-transgenic mice demonstrated that 4-1BB agonism is a promising strategy for converting dysfunctional effector cells into antiviral effectors for treating cHBV infections (20).

For developing new therapies to eradicate HBV, it is essential to describe the expression of ICMs during different phases of HBV in patients, as well as whether and how the NUC treatment modulates the frequency, activity, and ICM expression of intrahepatic NK and T cells. HBV infection is a dynamic disease that reflects the ongoing interaction between HBV and the host’s immune system over time. Patients with cHBV infection can be classified based on viral activity and liver disease markers. HBV patients in the chronic infection phase usually exhibit normal alanine aminotransferase (ALT) levels, minimal or no liver inflammation, and low fibrosis. In contrast, the chronic hepatitis phase is marked by intermediate to high levels of HBsAg, elevated ALT, and moderate-to-severe liver disease (21). In this study, we present a multiparametric analysis of immune checkpoint expression on intrahepatic and circulating lymphocyte subsets in both untreated HBV patients across different phases of infection and treated HBV-infected patients. By comparing these findings with normal liver tissue, we highlight the link between intrahepatic immune checkpoint expression and the clinical stage of the patients. Our findings provide novel insights into the immune landscape of intrahepatic T and NK cells in the setting of chronic HBV infection, offering contributions to the understanding of immune regulation in this disease.

Both T cells and NK cells are considered to play an irreplaceable role in the elimination of HBV and determine the outcome of HBV infection. ICMs seem to be at the center of the pathogenesis of HBV infection and have a promising potential for clinical translations as targets for therapies. Indeed, the dysregulation of certain ICMs in cHBV patients was previously described on circulating T and NK cells. However, their expression on lymphocytes of the liver, which is the main site of the disease, has been less studied, thus limiting the objective view of their importance in the actual pathology (10).

In this study, we characterized the expression of ICMs (PD-1, TIM-3, LAG-3, CTLA-4, ICOS, 4-1BB, and OX40) in fresh liver biopsies and revealed major alterations of intrahepatic lymphocytes in cHBV patients according to the clinical stage. Our findings bring new insights into the landscape of immune checkpoints on intrahepatic T and NK cells in the context of cHBV infection.

First, we showed that untreated cHBV patients, in both the chronic infection and chronic hepatitis phases, are characterized by an accumulation of CD8⁺ T cells that highly express the activation marker CD69. In accordance with previous observations (15, 16, 27–29), both groups of untreated HBV patients displayed high frequencies of intrahepatic PD-1⁺CD8⁺ T cells. In contrast, TIM-3, ICOS, and 4-1BB-positive CD8⁺ T cells as well as PD-1, TIM-3, and 4-1BB-positive CD4⁺ T cells were enriched mainly in the liver of chronic hepatitis patients. This observation aligns with previous studies demonstrating that TIM-3 expression is elevated on both CD4⁺ and CD8⁺ circulating T cells during the active chronic hepatitis phase, compared to other phases (30, 31). This increased TIM-3 expression is associated with T-cell exhaustion, contributing to impaired antiviral immune responses during active disease. Interestingly, the liver of chronic hepatitis patients contained also the highest frequency of 4-1BB-positive NK cells.

Importantly, we also highlighted that the patients’ HBeAg status is associated with the specific immune phenotype of intrahepatic CD8⁺ T cells and NK cells, with high expression of ICMs, mainly the co-stimulatory molecule 4-1BB. 4-1BB is known to be expressed by T cells and NK cells upon their activation (32). It binds to a single confirmed ligand (4-1BBL) found on the surface of activated antigen-presenting cells (32), thus providing stimulatory signals leading to IL-2 production and T-cell activation, proliferation, and survival and the generation of memory cells (32–34). A previous study revealed that the combination of PD-1 blockade and 4-1BB activation increased responses of human liver T cells against HBV but not HCV (35). Thus, the 4-1BB/4-1BBL pathway may be a crucial protective mechanism engaged in effector cell responses in the HBV context, and both NK cells and T cells should be considered as targeted immune cells in possible 4-1BB activation therapy. Importantly, a recent study by Andreata et al. demonstrated in HBV-transgenic mice that 4-1BB agonism is the most promising strategy to convert dysfunctional effector cells into antiviral effectors for treating cHBV infections. This finding places 4-1BB at the center of current research interest, with our study being crucial, as it describes 4-1BB expression on intrahepatic lymphocytes in patients across different HBV phases.

In general, limited information exists about the expression of ICMs on NK cells in the context of HBV infections. A recent study by Marotel et al. described slightly upregulated PD-1 expression in peripheral NK cells from HBV patients compared to healthy individuals (36). This is consistent with our results showing that the expression of ICMs on the largest population of circulating NK cells, CD56^dimCD16⁺ circulating NK cell subset, is negligible and that the increased PD-1 expression is observed rather on CD56^brightCD16⁻ NK cells. Concerning the intrahepatic NK cells, the main population of NK cells—CD56^brightCD16⁻ NK cells—can express PD-1 and is responsible for the increased frequency of intrahepatic PD-1⁺ NK cells in untreated HBeAg⁺ patients.

The information about the stimulatory molecule ICOS on intrahepatic NK cells in HBV patients was never reported. Previous studies highlighted the important role of ICOS in NK cell development and effector function, but the exact role of ICOS on human NK cells remains undetermined (10). Our study revealed that ICOS is mainly expressed in intrahepatic NK cells of HBeAg⁺ chronic hepatitis patients and that the expression is linked to the CD56^brightCD16⁻ NK cells, while the CD56^dimCD16⁺ NK cell subset seems to be ICOS negative. In that regard, the specific features of intrahepatic NK subsets, which cannot be sampled in blood, confirm the importance of research done directly on core liver biopsies or fine-needle aspirates (37).

NUC therapy induces prolonged suppression of viral replication in patients with cHBV, as evidenced by a reduction in serum HBV DNA and HBsAg levels. It has been reported that PD-1 expression on circulating T cells is normalized in patients with NUC therapy compared to untreated cHBV patients, and this reduction is accompanied by an improvement in the ability of T lymphocytes to secrete IL-12 and IFN-γ (38–43). A recent study revealed that the PD-1 expression was reduced in the liver biopsies of patients 96 weeks after the start of the NUC treatment compared to baseline (44). In addition, Nkongolo et al. described via single-cell sequencing of fine-needle aspiration samples from five untreated cHBV patients a specific hepatotoxic signature of CD8⁺ T cells that is not virus specific but associated with liver damage (18). This cell population expresses, among others, PD-1, LAG-3, TIM-3, and 4-1BB, and disappears after NUC treatment. Similarly, our data showed that the number and the distribution of intrahepatic lymphocytes, mainly CD8⁺ T cells, seem to be normalized in NUC-treated HBV patients. Still, even if our results from NUC-treated HBV patients suggest certain trends, they should be interpreted with caution due to the small sample size and the persistence of unnormalized ALT levels and activity scores in some patients despite treatment. In general, we observed that while the frequency of intrahepatic PD-1⁺, ICOS⁺, and 4-1BB⁺ NK cells was increased in untreated HBeAg⁺ patients compared to controls, NUC treatment partially normalized their frequencies. Thus, such modulation of ICMs on intrahepatic lymphocytes during NUC treatment could be associated with HBV suppression. However, we observed no association between the frequency of 4-1BB⁺ lymphocytes and liver injury.

Previous studies demonstrated that the quantity and function of HBV-specific T cells are associated with the outcome of HBV infection. Unfortunately, due to the small size of the liver biopsies, it was not possible to evaluate HBV-specific CD8⁺ T cells. Likewise, the limited sample size of the fresh core liver biopsy prevented us from conducting additional functional studies of immune cells. The scarcity of cHBV patients undergoing core needle liver biopsies further complicates the ability to recruit larger sample sizes for subgroup analyses, including studies focusing on age or sex/gender differences in HBV infection. This limitation is particularly significant, as prior research has shown that age and sex differences have a substantial impact on lymphocyte phenotype and function in HBV (23, 45–47). In our study, NUC-treated patients were older, as age is the key factor in the decision to initiate antiviral therapy.

In summary, analysis of cHBV patient-derived liver tissue revealed the altered expression of ICMs according to the patients’ clinical stage and the link with the HBeAg status of patients. By analyzing a large panel of intrahepatic ICMs expressed by different immune cells, we show that the liver of untreated cHBV patients is characterized by a high expression of immune checkpoint molecules, mainly PD-1, TIM-3, ICOS, and 4-1BB, providing an attractive approach to investigate their potential role as targets for effective therapeutic interventions.