Biological interpretation of baseline hepatitis B surface antigen and cirrhosis in determining interferon treatment duration

Abstract

The recent identification of baseline hepatitis B surface antigen and cirrhosis as independent predictors of extended interferon therapy provides a valuable clinical framework, yet the biological mechanisms underlying these associations require further elucidation. In this opinion review, we interpret these factors not merely as statistical covariates but as pathological surrogates for the virus-host equilibrium. We propose that high baseline hepatitis B surface antigen may reflect a deeply entrenched intrahepatic viral reservoir and may create an antigen-rich environment that impairs effective antiviral immunity. Concurrently, cirrhosis may be better understood as a spectrum of architectural remodeling in which vascular changes and sinusoidal defenestration could limit immune surveillance and affect local therapeutic access. Consequently, treatment durations exceeding 48 weeks may be required, at least in part, to overcome these profound immunological and structural barriers. We advocate shifting future predictive modeling from binary clinical definitions to a granular integration of viral kinetics and histological morphology to refine precision medicine in chronic hepatitis B.

Key Words: Chronic hepatitis B; Hepatitis B surface antigen; Sinusoidal capillarization; Liver sinusoidal endothelial cells; Cirrhosis

Core Tip: We interpret baseline hepatitis B surface antigen and cirrhosis as pathological surrogates for the virus-host equilibrium. High hepatitis B surface antigen may indicate an entrenched viral reservoir and may be associated with persistent antigenic pressure that contributes to antiviral immune dysfunction, while cirrhotic vascular remodeling may reduce the efficiency of immune surveillance. Consequently, extended interferon therapy may be more frequently required in patients with these biological and structural constraints.

INTRODUCTION

Recent clinical evidence has identified baseline hepatitis B surface antigen (HBsAg) levels and cirrhosis as independent predictors of the need for extended pegylated interferon alpha-2b therapy in chronic hepatitis B[1]. By specifically addressing treatment courses exceeding 48 weeks, this work provides a valuable framework for individualized therapeutic decision-making and highlights an important gap in current predictive models[2]. Beyond their statistical utility, these variables may also serve as biological surrogates for the underlying virus-host equilibrium. From a pathology-based perspective, a more detailed interpretation of baseline HBsAg and cirrhosis may clarify how intrahepatic viral persistence, immune dysfunction, and hepatic architectural remodeling together influence the need for prolonged therapy.

WHY TREATMENT DURATION IN CHRONIC HEPATITIS B NEEDS BIOLOGICAL INTERPRETATION?

Treatment duration in chronic hepatitis B should not be regarded solely as a practical aspect of therapeutic administration, but rather as a clinically meaningful determinant of efficacy, tolerability, and individualized treatment planning. This consideration is particularly pertinent in the context of pegylated interferon-α (Peg-IFN-α)-based therapy, where the value of a finite treatment course must be carefully weighed against variable response rates and treatment-related adverse effects[3]. Accordingly, the relevant clinical question extends beyond whether treatment can induce a response to include how long antiviral and immunomodulatory pressure must be sustained before a meaningful endpoint, particularly HBsAg decline or loss, can be achieved.

Although current clinical models may help identify patients who are more likely to require prolonged therapy, these associations remain incomplete without biological interpretation. Meta-analytic evidence has shown that both baseline and on-treatment factors are associated with HBsAg seroclearance during Peg-IFN-α-based therapy, supporting the clinical relevance of treatment-duration stratification[2]. In particular, early on-treatment HBsAg kinetics have been reported to correlate with subsequent treatment response, supporting the view that treatment duration is biologically linked to the tempo of antiviral and immunologic control rather than only to a preset treatment schedule[4]. However, a purely statistical framework may describe correlation without adequately explaining the mechanisms underlying differences in treatment duration.

In chronic hepatitis B virus (HBV) infection, the need for extended therapy may reflect the difficulty of overcoming persistent viral antigen production, a challenge that is central to the broader effort to achieve a functional cure[5]. Integrated HBV DNA can continue to contribute to HBsAg production even in the absence of active viral replication, providing one mechanistic explanation for persistent antigenemia despite virological suppression[6]. In parallel, chronic HBV infection is characterized by virus-specific T-cell dysfunction and immune exhaustion, which are closely linked to sustained antigen exposure and impaired antiviral immune control[7,8].

Importantly, treatment response in chronic HBV infection occurs within a structurally altered hepatic microenvironment that may itself influence therapeutic responsiveness. Cirrhosis-associated immune dysfunction is characterized by progressive immune dysregulation and immune paralysis as liver disease advances[9]. Moreover, liver sinusoidal endothelial cells (LSECs) are essential regulators of hepatic immune homeostasis, and sinusoidal remodeling or capillarization may disrupt intrahepatic cellular crosstalk and immune regulation in chronic liver disease[10,11]. Viewed together, these observations support the concept that treatment duration may reflect the combined influence of persistent antigen burden, impaired immune restoration, and adverse hepatic structural remodeling. Within this framework, baseline HBsAg and cirrhosis are not merely predictive variables but clinically accessible surrogates for the biological barriers that hinder functional cure. This perspective provides the rationale for examining each factor in greater mechanistic detail.

WHAT DOES BASELINE HBSAG BIOLOGICALLY REPRESENT?

Contrary to the misconception that HBsAg is merely a diagnostic serum marker, it serves as a clinically useful surrogate for intrahepatic viral persistence and antigen burden[12]. Serum HBsAg quantification fundamentally reflects the transcriptional activity from covalently closed circular DNA and, substantially, from integrated HBV DNA sequences within the host genome[6,12,13]. This distinction is critical because integrated viral fragments typically remain transcriptionally active for HBsAg production even when viral replication is suppressed by nucleos(t)ide analogues[6,14]. From a histopathological perspective, serum HBsAg levels demonstrate a strong positive correlation with the total intrahepatic viral burden and with the percentage of HBsAg-positive hepatocytes in liver tissue[15-17]. Consequently, high baseline HBsAg levels indicate broader hepatocyte involvement, in which a larger proportion of the liver parenchyma is actively engaged in viral protein production. Furthermore, because HBV DNA integration begins early in the viral life cycle and accumulates throughout the natural progression of the disease, elevated HBsAg titers often signify a long-standing infection history[6,14,15].

It is essential to emphasize that HBsAg levels do not directly reflect necroinflammation or active liver damage, which are better indicated by markers such as alanine aminotransferase or hepatitis B core antigen expression[17,18]. Instead, HBsAg reflects the viral antigen load, which has been proposed to act as an “immunological decoy” and may hinder the host’s ability to mount an effective immune response. Consistent with this interpretation, higher viral antigen exposure has been linked to more impaired HBV-specific immune function, although serum HBsAg alone remains an incomplete surrogate of the overall antiviral immune state[19,20]. Although persistent antigen exposure has been associated with HBV-specific T-cell dysfunction, serum HBsAg alone does not fully capture the complexity of antiviral immune status in individual patients[6,7,12,13].

Ultimately, a high baseline HBsAg level is more consistent with a larger and more biologically entrenched viral reservoir, which may help explain why some patients require a longer duration of immunomodulatory therapy such as interferon. This pathological framework supports the clinical finding that patients with high baseline titers are considerably more likely to require extended therapy to achieve the desired endpoint of functional cure. Nevertheless, viral antigen burden alone is unlikely to fully determine treatment responsiveness because antiviral immunity operates within a hepatic microenvironment whose structural integrity may be profoundly altered in cirrhosis.

SHOULD CIRRHOSIS BE VIEWED AS A BINARY PREDICTOR OR A HISTOLOGICAL CONTINUUM?

In clinical guidelines and predictive models, cirrhosis is often treated as a binary variable, despite important pathological heterogeneity. This variation may partly explain why patients categorized under the same label of “cirrhosis” do not respond uniformly to interferon therapy. Patients with thinner septa, less sinusoidal capillarization, and less parenchymal compartmentalization may retain more effective immune cell access and greater treatment responsiveness than those with advanced nodular and vascular remodeling[21-23]. This concept is reflected in the Laennec substaging system, which ranges from 4A (mild, thin septa) to 4C (severe, broad septa with marked nodular transformation) and offers a more refined structural framework for prognostic and therapeutic stratification (Figure 1A-C)[21,23].

Figure 1 Laennec cirrhosis substaging (4A-4C) and sinusoidal remodeling illustrate structural changes potentially relevant to immune surveillance and interferon bio-distribution in chronic hepatitis B. A-C: Representative Masson trichrome-stained liver sections demonstrating progressive architectural distortion across Laennec cirrhosis substages (collagen in blue; hepatocytes in pink/red). Laennec 4A: Mild cirrhosis with relatively thin septa and early nodular remodeling (A). Laennec 4B: Moderate cirrhosis with broader septa and more evident nodule formation (B). Laennec 4C: Severe cirrhosis with thick, widely distributed septa and pronounced regenerative nodules, reflecting advanced compartmentalization of the parenchyma (C); D: CD34 immunohistochemistry showing sinusoidal endothelial cell capillarization (CD34-positive sinusoidal staining; brown), consistent with cirrhosis-associated vascular remodeling. Arrows denote a regenerative nodule; E: Conceptual schematic contrasting the healthy sinusoid - with fenestrated liver sinusoidal endothelial cells facilitating solute transfer and immunosurveillance - vs the cirrhotic sinusoid, where capillarization/defenestration with basement membrane deposition creates a physical and functional barrier that may limit interferon penetration and immune effector cell access to infected hepatocytes. The pathology images were captured by Zou LF; The schematic in panel E was created by Wang CL using BioRender. Scale bars = 100 μm. LSECs: Liver sinusoidal endothelial cells; BM: Basement membrane.

We contend that the requirement for extended interferon therapy in “cirrhotic” patients may be influenced by biological barriers inherent in advanced histological remodeling. First, extracellular matrix accumulation and the transition from thin, regressive septa to thick, cross-linked progressive septa result in marked architectural subversion[22,24,25]. In cirrhosis, microvascular remodeling is reflected by CD34-positive sinusoidal staining, which is consistent with LSEC capillarization (Figure 1D)[26]. Defenestration and basement membrane deposition are also evident (Figure 1E)[27-29]. Together, these changes likely reduce sinusoidal permeability and may create a local microanatomical barrier to immune cell-hepatocyte interaction. However, direct evidence that these changes materially impair interferon biodistribution in chronic hepatitis B remains limited.

Second, and perhaps more critically, this histological remodeling profoundly disrupts hepatocyte-immune cell interactions. Under physiological conditions, intravascular effector CD8+ T cells perform immunosurveillance by extending cytoplasmic protrusions through endothelial fenestrae to probe subsinusoidal hepatocytes[26]. In advanced cirrhosis, however, sinusoidal capillarization and defenestration create a physical barrier that decreases liver porosity[26]. This vascular remodeling may reduce the efficiency with which infected hepatocytes are surveyed and recognized by effector T cells[26]. Recent digital pathology work also supports the idea that cirrhosis is structurally heterogeneous even after viral response, reinforcing that patients grouped under the same clinical label may still harbor biologically distinct intrahepatic architectures[23]. When combined with a compromised immune microenvironment, including reported alterations in immune cell composition and persistent inflammatory dysregulation in cirrhosis, the “probing” efficiency of the host immune response may be further diminished[9,30,31].

Taken together, these findings suggest that baseline HBsAg and cirrhosis should be viewed as complementary dimensions of a persistent virus-host equilibrium. Whereas high HBsAg titers may reflect sustained antigenic burden and ongoing pressure on antiviral immunity, cirrhotic remodeling may reduce the efficiency of immune surveillance within the liver[6,23]. Their coexistence may therefore define a biologically entrenched state that is less amenable to rapid clearance, providing a plausible mechanistic explanation for why these patients more often require treatment durations exceeding 48 weeks. At the same time, translating this pathology-based interpretation into routine clinical decision-making remains difficult because cirrhosis is usually defined by non-invasive criteria rather than histological architecture.

WHAT ARE THE LIMITATIONS OF GUIDELINE-BASED ASSESSMENT WITHOUT HISTOLOGICAL CORRELATION?

While liver biopsy remains the recommended reference standard for diagnosing and staging hepatic fibrosis, it is often unavailable in routine real-world clinical practice because of its invasive nature, associated costs, and potential for complications[32,33]. Consequently, current clinical guidelines rely heavily on non-invasive tests and imaging-based definitions to identify cirrhosis[34-36]. However, this reliance introduces significant biological uncertainties that may affect the interpretation of treatment outcomes.

Specifically, non-invasive definitions of cirrhosis may overlook the substantial histological heterogeneity inherent in advanced chronic liver disease. Studies have demonstrated that liver stiffness measurements and biochemical scores can vary according to spatial distribution, meaning that a single-point assessment may not capture the diverse architectural patterns, ranging from early fibrotic remodeling to advanced nodular transformation, that characterize the cirrhotic spectrum[23,37,38]. This limitation is especially relevant when the clinical question concerns treatment responsiveness rather than only fibrosis stage, because structurally distinct livers may share similar non-invasive category assignments[23].

Furthermore, imaging and stiffness-based modalities frequently miss ongoing inflammatory activity, which is a major confounder in liver stiffness measurements[39-43]. This is particularly relevant in the context of interferon response, because necroinflammation significantly increases liver stiffness regardless of the actual fibrosis stage. In patients who achieve an early decline in stiffness during therapy, this change often reflects the amelioration of hepatic inflammation rather than true fibrosis regression[40,42-44].

By treating cirrhosis as a binary variable based on imaging thresholds, some patients may be misclassified because substantial differences in architectural remodeling can exist within the same clinical category. Recognizing this heterogeneity is essential to a more nuanced understanding of why certain patients with “cirrhosis” respond more favorably to extended interferon therapy than others. Accordingly, non-invasive definitions of cirrhosis are clinically useful but may not fully capture the architectural and inflammatory heterogeneity most relevant to treatment responsiveness. When liver biopsy is not feasible, longitudinal non-invasive assessment, such as dynamic liver stiffness measurement interpreted together with biochemical activity markers, may still provide indirect support for treatment monitoring and clinical stratification.

WHAT ARE THE CURRENT CONTROVERSIES AND UNRESOLVED QUESTIONS?

Despite increasing interest in biologically guided interferon treatment stratification, the interpretation of baseline HBsAg remains controversial. Serum HBsAg is not a simple measure of replicative viral burden, because it may derive from both transcriptionally active covalently closed circular DNA and integrated HBV DNA. This biological complexity may help explain why HBsAg has clinical predictive value but does not function as a complete surrogate for the intrahepatic viral reservoir or immune status in every patient[2,6,12].

A second unresolved issue is whether cirrhosis should be modeled as a binary clinical variable or as a histological continuum. Although most treatment algorithms classify cirrhosis as present or absent, the Laennec system and recent digital pathology studies suggest that structural heterogeneity may carry biological relevance beyond conventional staging. This raises the possibility that patients grouped together as “cirrhotic” may in fact have distinct intrahepatic microenvironments and different probabilities of responding to prolonged interferon therapy[21,23].

A third controversy concerns whether currently available non-invasive or mechanistic markers can adequately capture the barriers most relevant to interferon responsiveness. Reviews of cirrhosis-associated immune dysfunction and LSEC biology support the concept that architectural remodeling is also immunologically meaningful, but direct evidence linking these changes to interferon treatment duration in chronic hepatitis B remains limited. In particular, most currently available markers capture only selected viral, inflammatory, or structural dimensions rather than the full biologic interplay among antigen burden, immune dysfunction, and hepatic microarchitectural remodeling[9,10,12]. Whether these domains can be integrated into a clinically robust framework for treatment-duration stratification remains unresolved. The principal controversies and unresolved questions relevant to biologically guided interferon treatment stratification are summarized in Table 1.

Table 1 Key unresolved issues in biologically guided interferon treatment stratification.

Issue	Why unresolved
Baseline HBsAg	Reflects both cccDNA activity and integrated HBV DNA, so it is not a complete surrogate of viral burden or immune status
Cirrhosis classification	Binary clinical definitions do not capture histological heterogeneity
Non-invasive markers	Existing markers reflect only part of the viral, inflammatory, and structural landscape
Hepatic remodeling	Mechanistic plausibility exists, but direct evidence remains limited
Integrated models	A robust framework combining viral, structural, and immune factors is still lacking

HBsAg: Hepatitis B surface antigen; HBV: Hepatitis B virus.

WHERE SHOULD FUTURE RESEARCH AND CLINICAL PRACTICE GO NEXT?

The findings by Yan et al[1] provide a crucial foundation for personalizing interferon therapy; however, they also underscore the need to incorporate pathological and biological context more deeply into clinical decision-making[1]. Future research should prioritize moving beyond a binary “yes/no” classification of cirrhosis toward a stratified architectural approach. Using systems such as the Laennec staging system allows clinicians to categorize cirrhosis into stages 4A (mild), 4B (moderate), and 4C (severe) based on septal thickness and nodule size[21]. Such structural differences are likely to represent distinct biological barriers, with advanced septal cross-linking and sinusoidal capillarization potentially conferring poorer immune cell access and lower treatment responsiveness than less advanced remodeling[23].

Consequently, future studies should directly correlate fibrosis morphology with treatment duration. Evidence suggests that, although short-term therapy may be insufficient for those with severe architectural subversion, prolonged interferon administration is associated with significant fibrosis regression, particularly in patients with higher baseline collagen scores[45]. Incorporating advanced quantitative morphometric tools such as qFibrosis could further refine this approach by identifying subtle regressive changes in septal width that are often imperceptible in “stable” Ishak scores, thereby providing a more precise benchmark for determining when therapy can be safely discontinued[23,46-49]. Quantitative collagen-feature analysis may be especially useful because automated morphometric methods provide repeatable and reproducible assessment of liver fibrosis and can also detect spatial and architectural changes in fibrosis that are not fully captured by conventional ordinal staging systems[23,50-52].

In parallel, future research should establish machine-learning and deep-learning models as a complementary extension of computational pathology. In chronic hepatitis B, these approaches have been applied to fibrosis classification and multimodal risk assessment, including deep-learning models for fibrosis staging from histology and data-integration networks combining imaging and clinical variables[53-55]. More broadly, recent AI-based liver studies support the feasibility of AI-assisted whole-slide analysis, improved scoring reproducibility, and treatment-response modeling[56-61]. Future HBV-specific models should therefore move beyond fibrosis quantification alone and integrate histological architecture, vascular remodeling, virological markers, and non-invasive clinical data to identify patients most likely to require therapy beyond 48 weeks.

Finally, integrated predictive models should be developed to combine virological, structural, and non-invasive markers within a single framework. Such models may incorporate baseline HBsAg levels as a surrogate of intrahepatic antigen burden and integrated HBV transcriptional activity, histological activity and fibrosis patterns as indicators of architectural constraint, and non-invasive markers such as liver stiffness measurement and enhanced liver fibrosis scores as complementary readouts of the dynamic components of liver injury. In this way, clinicians may move toward a multidimensional virus-host profile that more accurately identifies patients who require extended therapy to overcome entrenched structural and immunological barriers to functional cure.

CONCLUSION

In our view, the clinical predictors identified by Yan et al[1] gain greater significance when interpreted through a pathology-based lens. High baseline HBsAg and cirrhosis should not be regarded merely as statistical correlates of treatment extension but as clinically accessible indicators of the underlying virus-host equilibrium. Elevated baseline HBsAg may reflect persistent viral antigen burden and a more deeply entrenched viral reservoir, whereas cirrhosis may denote a structurally remodeled and immunologically dysregulated hepatic microenvironment that constrains effective immune surveillance and viral clearance. From this perspective, the need for interferon therapy beyond 48 weeks may represent the biological effort required to overcome these virological, immunological, and architectural barriers to functional cure. A more integrated understanding of these processes may help move the field beyond binary clinical definitions toward biologically informed risk stratification and more precise therapeutic individualization in chronic hepatitis B.