Association between hepatitis B virus markers and liver fibrosis staging in patients with chronic hepatitis B

Abstract

BACKGROUND

Existing assessment systems for chronic hepatitis B (CHB)-associated liver fibrosis (LF) exhibit insufficient accuracy, thereby requiring further improvements.

AIM

To investigate the association of LF staging with hepatitis B core antibody (HBcAb), hepatitis B virus DNA (HBV-DNA), and hepatitis B surface antigen (HBsAg) in patients with CHB.

METHODS

We selected 120 patients with CHB receiving treatment in Hangzhou Linping District First People’s Hospital from January 2020 to June 2024. Participants were allocated into the mild (F0-F1, n = 52) and moderate-to-severe groups (F2-F4, n = 68) following the rigorous LF staging criteria. HBcAb, HBV-DNA, and HBsAg concentrations were measured. Pearson correlations were employed to examine the correlations of HBcAb with HBV-DNA and HBsAg, whereas Spearman correlation analysis was conducted to identify the associations of the three with LF staging. Receiver operating characteristic (ROC) curves were further used to analyze the performance of these biomarkers in diagnosing LF stages. Furthermore, binary logistic regression analysis was conducted to determine the association of these three with LF progression in CHB.

RESULTS

Markedly increased HBcAb and notably decreased HBV-DNA and HBsAg were observed in moderate-to-severe cases vs their mild counterparts. A positive correlation was observed between HBV-DNA and HBsAg, whereas both markers were inversely associated with HBcAb. Moreover, LF staging exhibited a significant positive correlation with HBcAb and an inverse connection with HBV-DNA and HBsAg. The receiver operating characteristic analysis revealed area under the curve values of 0.715, 0.799, and 0.662 for HBcAb, HBV-DNA, and HBsAg in diagnosing LF staging, respectively. Combining these markers improved the area under the curve to 0.851. The final analysis identified HBcAb as promoting fibrosis advancement (odds ratio = 2.765), whereas HBV-DNA demonstrated protective properties (odds ratio = 0.247).

CONCLUSION

HBcAb is negatively correlated with HBV-DNA and HBsAg but positively associated with LF staging. All three markers are valuable in assessing LF staging, and their combined use presents the highest diagnostic efficacy. Importantly, a high HBcAb/low HBV-DNA profile markedly increased fibrosis progression risks in CHB-affected individuals.

Key Words: Hepatitis B core antibody; Hepatitis B virus DNA; Hepatitis B surface antigen; Chronic hepatitis B; Liver fibrosis staging

Core Tip: By combining three serum biomarkers, hepatitis B core antibody, hepatitis B virus-DNA, and hepatitis B surface antigen, this study proposes a cost-effective, non-invasive diagnostic approach for assessing chronic hepatitis B-related liver fibrosis (LF) staging, providing advantages in operational convenience and repeatability. The multi-parameter model demonstrated superior diagnostic performance without compromising either sensitivity or specificity. Further, a strong correlation of hepatitis B core antibody and hepatitis B virus-DNA with LF progression was determined. Altogether, a promising non-invasive serological tool for LF staging is proposed in this study, providing a cost-effective solution to improve clinical diagnostics and patient care.

INTRODUCTION

Chronic hepatitis B (CHB) is a clinical condition that is associated with long-term hepatitis B virus (HBV) infection, posing a grave global health burden by its potential progression to terminal liver disease and hepatocellular carcinoma (HCC), and endangering patient survival and well-being[1,2]. Globally, chronic HBV infection prevalence stands at 3.61%, with its complications, including cirrhosis, liver failure, and HCC, contributing to approximately 900000 deaths annually[3,4]. Liver fibrosis (LF) is a key driver of CHB progression; thus, its precise staging is vital to inform therapy and boost patient survival[5]. Currently, liver biopsy remains the benchmark for assessing LF staging; however, its invasiveness, limited suitability for longitudinal monitoring, and lack of repeatability hinder its widespread application[6,7]. Consequently, developing cost-effective, non-invasive, and reproducible diagnostic tools are urgently warranted. Such advancements are of paramount clinical significance for optimizing CHB treatment.

Hepatitis B core antibody (HBcAb) is widely recognized as a serological marker of previous HBV infection and has been implicated in occult HBV infection[8]. Emerging evidence indicates that HBcAb positivity correlates with high fibrosis-4 scores, demonstrating a potential positive connection between HBcAb and LF degrees[9]. HBV-DNA, a crucial marker that reflects viral replication, is associated with advanced LF in immunotolerant individuals and is a known disease progression driver in CHB[10,11]. Meanwhile, hepatitis B surface antigen (HBsAg) loss, or a “functional cure” or “resolved hepatitis B”, is considered the optimal therapeutic endpoint for patients with CHB[12]. However, HBV-infected individuals with reduced HBsAg levels in the presence of cirrhosis demonstrated an increased risk of HCC and poorer outcomes[13]. Research that investigated the association of HBcAb, HBV-DNA, and HBsAg concentrations with LF staging in patients with CHB remains limited despite the clinical relevance of these biomarkers. To fill this void, this research aims to investigate the associations between these markers, thereby contributing to the establishment of non-invasive approaches that can be used to assess LF staging in patients with CHB.

MATERIALS AND METHODS

Study population

This study recruited 120 patients with CHB who presented to Hangzhou Linping District First People’s Hospital from January 2020 to June 2024. Based on LF staging, participants were categorized into the mild (F0-F1, n = 52) and moderate-to-severe LF groups (F2-F4, n = 68). Inclusion criteria were confirmed diagnosis of CHB, with positive HBsAg and/or HBV DNA for more than 6 months[14]; absence of jaundice, irrespective of alanine aminotransferase and aspartate aminotransferase levels; availability of complete clinical records and routine laboratory test results; no history of antiviral therapy or use of liver enzyme-lowering medications within the preceding 3 months; normal communication and cognitive abilities, along with demonstrated compliance. The Metavir scoring system was used to assess LF staging, as recommended by the 2015 Guidelines for the Prevention and Treatment of CHB[15].

Exclusion criteria were presence of extrahepatic fibrotic diseases, such as systemic lupus erythematosus, rheumatic diseases, or diabetes; coinfection with hepatitis A, C, D, or E viruses; diagnosis of alcoholic/autoimmune/cholestatic hepatopathy, or non-alcoholic fatty liver; history of drug-induced hepatotoxicity, hereditary metabolic liver diseases, or advanced cirrhosis complicated by malignant tumors; concurrent hematological disorders, endocrine diseases, or metabolic conditions.

Detection methods

Serum HBcAb and HbsAg levels were quantitatively measured using a fully automated time-resolved immunofluorescence analyzer. Serial dilutions (100-fold, 500-fold, and 1000-fold) were performed for samples exceeding the linear detection range to ensure accurate concentration determination. This experiment used the Bio-Rad iCycler PCR system for HBV-DNA quantification, with a linear detection range set at 20-1 × 10⁸ IU/mL and a lower limit of detection of 20 IU/mL. The assay demonstrated a > 95% detection rate when HBV-DNA levels were ≥ 20 IU/mL. Experimental validation confirmed that this detection system demonstrates no cross-reactivity with viruses such as hepatitis C virus or human immunodeficiency virus. To ensure assay reliability, World Health Organization-certified HBV-DNA standard samples were incorporated as quality controls throughout the entire experimental procedure. Regarding precision performance, the intra-assay coefficient of variation was below 5%. Shanghai Jianglai Biotechnology Co., Ltd. supplied all reagent kits, with corresponding product numbers 1534692054, 1533627249, and 1534774710.

Statistical analysis

The inter-group differences of continuous data (reported as mean ± SD were established utilizing the independent sample’s t-test. Categorical data, presented as n (%), were comparatively analyzed with χ² tests. Pearson’s correlation coefficients were used to analyze the correlations among HBcAb, HBsAg, and HBV-DNA. Spearman’s correlation coefficients were employed to assess the associations of these markers with LF staging. Receiver operating characteristic (ROC) curves were used to examine the predictive accuracy of these markers for LF staging. ROC curve analysis identified the optimal cut-offs based on the Youden index. Sensitivity (Sen) was calculated as [true positives/(true positives + false negatives)], whereas specificity (Spe) was identified as [true negatives/(true negatives + false positives)]. Furthermore, binary logistic regression explored potential associations between HBcAb, HBsAg, HBV-DNA, and LF progression. Statistical package for the Social Sciences version 19.0 was used for all data analyses, and P < 0.05 indicated significance throughout.

RESULTS

Baseline characteristics of the study groups

The baseline characteristics, including age, sex, body mass index, white blood cell count, hemoglobin, and alpha-fetoprotein, were compared between the mild (F0-F1) and moderate-to-severe (F2-F4) groups. No significant inter-group differences in any of these parameters were observed (P > 0.05), indicating that the two patient cohorts were well-matched at baseline (Table 1).

Table 1 Inter-group comparison of baseline characteristics, mean ± SD/n (%).

Baseline characteristics	Mild group (n = 52)	Moderate-to-severe group (n = 68)	χ2/t	P value
Age (years old)	45.35 ± 9.19	47.06 ± 10.49	0.884	0.379
Sex			0.067	0.796
Male	28 (53.85)	35 (51.47)
Female	24 (46.15)	33 (48.53)
BMI (kg/m2)	23.73 ± 2.32	23.59 ± 2.36	0.324	0.746
WBC (× 109/L)	5.02 ± 0.87	4.85 ± 0.87	1.061	0.291
Hb (g/L)	109.69 ± 7.75	112.0 ± 11.88	1.217	0.226
AFP (μg/L)	5.71 ± 1.30	6.06 ± 1.90	1.139	0.257

BMI: Body mass index; WBC: White blood cell count; Hb: Hemoglobin; AFP: Alpha-fetoprotein.

HBcAb, HBV-DNA, and HBsAg concentrations in two patient groups

The moderate-to-severe group demonstrated markedly higher HBcAb levels and notably lower HBV-DNA and HBsAg concentrations than the mild group (P < 0.001 for all comparisons; Figure 1).

Figure 1 Hepatitis B core antibody, hepatitis B virus-DNA, and hepatitis B surface antigen concentrations in the two groups. A: Expression of hepatitis B core antibody; B: Expression of hepatitis B virus-DNA; C: Expression of hepatitis B surface antigen. HBcAb: Hepatitis B core antibody; HBV-DNA: Hepatitis B virus DNA; HBsAg: Hepatitis B surface antigen. ^cP < 0.001.

Correlation among HBcAb, HBV-DNA, and HBsAg

Significant associations among the three biomarkers were identified with Pearson correlation analysis: HBcAb demonstrated an inverse connection with HBV-DNA (r = -0.329, P < 0.001) and HBsAg (r = -0.319, P < 0.001), whereas HBV-DNA and HBsAg were positively correlated (r = 0.387, P < 0.001; Table 2).

Table 2 Correlation among hepatitis B core antibody, hepatitis B virus-DNA, and hepatitis B surface antigen.

Marker pair	r	P value
HBcAb vs HBV-DNA	-0.329	< 0.001
HBcAb vs HBsAg	-0.319	< 0.001
HBV-DNA vs HBsAg	0.387	< 0.001

HBcAb: Hepatitis B core antibody; HBV-DNA: Hepatitis B virus DNA; HBsAg: Hepatitis B surface antigen.

Correlation of HBcAb, HBV-DNA, and HBsAg with LF staging

Spearman correlation analysis revealed significant associations between the three biomarkers and LF staging. HBcAb demonstrated a positive correlation with fibrosis staging (r = 0.369, P < 0.001), whereas HBV-DNA (r = -0.512, P < 0.001) and HBsAg (r = -0.278, P < 0.001) exhibited negative correlations with fibrosis staging (Table 3).

Table 3 Correlation of hepatitis B core antibody, hepatitis B virus-DNA, and hepatitis B surface antigen with liver fibrosis staging.

Correlation	Liver fibrosis staging
	r	P value
HBcAb	0.369	< 0.001
HBV-DNA	-0.512	< 0.001
HBsAg	-0.278	0.002

HBcAb: Hepatitis B core antibody; HBV-DNA: Hepatitis B virus DNA; HBsAg: Hepatitis B surface antigen.

ROC analysis of HBcAb, HBV-DNA, and HBsAg for diagnosing LF staging

The ROC curve analysis was conducted for the diagnostic efficacy assessment of HBcAb, HBV-DNA, and HBsAg in evaluating LF staging. The area under the curve (AUC), Sen, and Spe for HBcAb were 0.715 [95% confidence interval (CI): 0.623-0.807], 69.12%, and 69.23%, respectively, with an optimal cut-off of 4.29 Log10IU/mL (P < 0.001). These values were 0.799 (95%CI: 0.719-0.878), 67.65%, and 76.92%, with an optimal cut-off of 5.21 Log10IU/mL for HBV-DNA (P < 0.001) and 0.662 (95%CI: 0.562-0.762), 85.29%, and 44.23%, with an optimal cut-off of 4.57 Log10IU/mL for HbsAg, respectively (P < 0.001). The AUC significantly increased to 0.851 (95%CI: 0.779-0.923) when a combined diagnostic approach was adopted for LF stages, with the optimal cut-off as 0.51 and Sen and Spe of 85.29% and 73.08%, respectively (P < 0.001; Figure 2, Table 4).

Figure 2 Receiver operating characteristic curves for hepatitis B core antibody, hepatitis B virus-DNA, and hepatitis B surface antigen in diagnosing liver fibrosis staging. AUC: Area under the curve; HBcAb: Hepatitis B core antibody; HBsAg: Hepatitis B surface antigen.

Table 4 Receiver operating characteristic analysis results for hepatitis B core antibody, hepatitis B virus-DNA, and hepatitis B surface antigen in diagnosing liver fibrosis staging.

ROC analysis	AUC	95%CI	SE	Sensitivity (%)	Specificity (%)	P value
HBcAb	0.715	0.623-0.807	0.047	69.12	69.23	< 0.001
HBV-DNA	0.799	0.719-0.878	0.041	67.65	76.92	< 0.001
HBsAg	0.662	0.562-0.762	0.051	85.29	44.23	< 0.001
Combined prediction	0.851	0.779-0.923	0.037	85.29	73.08	< 0.001

ROC: Receiver operating characteristic; AUC: Area under the curve; CI: Confidence interval; HBcAb: Hepatitis B core antibody; HBV-DNA: Hepatitis B virus DNA; HBsAg: Hepatitis B surface antigen.

Association of HBcAb, HBV-DNA, and HBsAg with LF progression by multivariate analysis

Multivariate logistic regression identified HBcAb as a significant predictor of LF progression in patients with CHB (odds ratio = 2.765). In contrast, HBV-DNA demonstrated a protective effect against fibrosis advancement (odds ratio = 0.247). Table 5 summarizes the complete ROC analysis.

Table 5 Receiver operating characteristic evaluation of hepatitis B core antibody, hepatitis B virus-DNA, and hepatitis B surface antigen in diagnosing liver fibrosis stages.

Factor	B	SE	Wald	P value	OR	95%CI
HBcAb	1.017	0.446	5.197	0.023	2.765	1.153-6.630
HBV-DNA	-1.400	0.457	9.373	0.002	0.247	0.101-0.604
HBsAg	-0.931	0.489	3.624	0.057	0.394	0.151-1.028

HBcAb: Hepatitis B core antibody; HBV-DNA: Hepatitis B virus DNA; HBsAg: Hepatitis B surface antigen; OR: Odds ratio; CI: Confidence interval.

DISCUSSION

This research categorized patients with CHB based on LF staging into the mild (F0-F1) and moderate-to-severe (F2-F4) groups. Laboratory tests revealed abnormal HBcAb, HBV-DNA, and HBsAg levels in both cohorts. Subsequent analysis revealed that these markers not only demonstrated significant intercorrelation but also exhibited strong associations with fibrosis severity. ROC curve assessment of diagnostic performance yielded AUC values of 0.662-0.799 for individual markers, indicating moderate diagnostic capability, with improved accuracy observed when combined. The multivariate analysis ultimately identified HBcAb as positively associated with fibrosis progression, whereas HBV-DNA demonstrated an inverse association with advancing fibrosis stages.

A growing body of research has investigated the non-invasive diagnostic methods for assessing LF progression in patients with CHB. Song et al[16] identified chemokine (C-C motif) ligand 20 and CD8 antigen, alpha polypeptide as potential biomarkers and therapeutic targets for LF in patients with CHB, indicating their role in regulating chronic inflammatory responses. Besides, Wang et al[17] revealed that serum interleukin-34 Levels of ≥ 15.83 pg/mL could effectively identify severe fibrosis in patients with chronic HBV infection, boasting high Sen (86.6%) and Spe (78.7%). Furthermore, Tsuji et al[18] reported that Mac-2 binding protein glycosylation isomer levels could accurately diagnose advanced LF, achieving an AUC of 0.902. Our study aims to determine the correlation and diagnostic performance of serum HBcAb, HBV-DNA, and HBsAg levels in association with LF staging in CHB-afflicted individuals. In our results, HBcAb levels were statistically increased in patients with moderate-to-severe fibrosis compared with their mild counterparts, whereas HBV-DNA and HBsAg concentrations demonstrated an inverse trend, being markedly lower in the moderate-to-severe group. Thus, HBcAb, HBV-DNA, and HBsAg all contribute to LF progression in patients with CHB, with HBcAb potentially exerting a pro-fibrotic effect and HBV-DNA and HBsAg playing inhibitory roles. Similarly, Gao et al[19] reported gradual HBsAg and HBV-DNA reductions as LF advanced in patients with hepatitis B e antigen (HBeAg)-positive CHB. HBcAb, produced as part of the host’s immune response against HBV, typically seroconverts early in HBV infection[20]. In addition to mediating HBV infection progression across various stages, HBcAb implicates itself in HBV-associated non-specific immune responses and may be associated with systemic inflammation[21]. HBcAb’s influence on CHB-associated LF may be related to its involvement in modulating the activity of hepatitis B core antigen-specific memory B cells[22]. HBsAg, one of HBV’s envelope proteins, is crucial in monitoring the natural course of chronic hepatitis and patient outcome prediction[23,24]. By identifying the correlations of these biomarkers with LF progression, this study intends to pioneer non-invasive diagnostic approaches to ultimately optimize CHB management.

Subsequent correlation analysis revealed a significant positive association between HBV-DNA and HBsAg, whereas both of them displayed an inverse connection with HBcAb, indicating a potential interplay among them in LF pathogenesis and progression in CHB-afflicted individuals. Of them, HBcAb demonstrated a positive association with LF staging, whereas HBV-DNA and HBsAg exhibited significant inverse correlations with LF progression. Supporting our results, Yıldız Kaya et al[25] reported the ability of HBsAg quantitation to identify whether patients with negative HBeAg have an inactive infection or active liver inflammation while identifying a positive connection between HBsAg and HBV-DNA. Further corroborating our results, Goyal et al[26] emphasized the intimate inverse correlation of HBsAg with LF staging in HBeAg-positive CHB cases. Moreover, prior evidence has strongly associated down-regulated HBsAg to moderate-to-severe LF in patients with HBeAg-positive CHB, similar to our results[27]. In ROC assessment of diagnostic performance, HBV-DNA exhibited the best discrimination for LF staging (AUC of 0.799). Among individual parameters, HBsAg achieved optimal Sen (85.29%), whereas HBV-DNA displayed maximal Spe (76.92%). Notably, their combination contributed to great diagnostic performance enhancement (AUC = 0.851, Sen = 85.29%, Spe = 73.08%), highlighting how these biomarkers complement each other to optimize non-invasive LF staging reliability in CHB. Further, specific cut-off values, HbcAb > 4.29 Log10IU/mL, HBV-DNA < 5.21 Log10IU/mL, or HBsAg < 4.57 Log10IU/mL, can effectively predict significant LF in patients with CHB. Closely aligning with our results, Zhang et al[28] reported the significance of HBcAb (AUC = 0.640), HBV-DNA (AUC = 0.723), and HBsAg (AUC = 0.617) in predicting significant LF in HBeAg-positive CHB. Similarly, Praneenararat et al[29] identified an HBV-DNA threshold (> 5.5 Log IU/mL) predicting significant LF in HBeAg-negative CHB (Sen = 71.4%, Spe = 93.3%), further validating our observations. Furthermore, multiple studies have emphasized the clinically significant added value of these biomarkers. For instance, Zhang et al[30] demonstrated the potential of HBsAg and HBV-DNA measurements in predicting significant hepatitis activity in HBeAg-positive chronic HBV-infected patients. Moreover, Gong et al[31] reported the great value of serum HBsAg and HBeAg levels in predicting antiviral outcomes in HBV-induced cirrhosis managed by entecavir therapy. The final analysis indicated that high-titer HBcAb (≥ 4.29 Log10IU/mL) served as a significant predictor for worsening LF in patients with CHB. Conversely, high-load HBV-DNA (≥ 5.21 Log10IU/mL) became a protective factor against fibrosis stage advancements.

This research can further improve methodological design and investigative depth. First, the homogeneity of sample origins presents certain constraints. All specimens were obtained from a single institution; thus, geographical factors may affect the generalizability of results. Subsequent investigations could broaden sample diversity by implementing multicenter and cross-regional cooperation to strengthen the validity and practical relevance of conclusions. Second, mechanistic exploration of disease progression warrants deeper investigation. Identifying the molecular interactions between HBcAb and HBV-DNA during hepatic fibrogenesis in patients with CHB would significantly advance current theoretical frameworks. Third, the prognostic value of HBcAb, HBV-DNA, and HBsAg for long-term clinical outcomes remains unknown. Incorporating such assessments would maximize the clinical utility of these biomarkers. Finally, the Bio-Rad iCycler PCR system employed in this study meets standard clinical Sen requirements with its 20 IU/mL detection threshold; however, it may overlook cases with subthreshold viral loads. Future studies could employ ultrasensitive detection platforms (≤ 10 IU/mL Sen) to characterize more precisely the association between HBV-DNA levels and LF.

CONCLUSION

Conclusively, serum HBcAb, HBV-DNA, and HBsAg are all intricately associated with LF in patients with CHB. Specifically, HBcAb demonstrates a positive connection with LF, whereas HBV-DNA and HBsAg exhibit inverse correlations. The integration of these three biomarkers significantly improves diagnostic accuracy for identifying significant fibrosis, providing both high Sen and Spe. In HBeAg-negative CHB cases, as well as those undergoing varied treatment protocols, integrating multiple diagnostic markers could improve LF staging precision. Notably, both HBcAb titers and HBV-DNA viral load demonstrate strong associations with fibrotic liver damage advancement in CHB cases. Our results provide a non-invasive serological diagnostic approach for LF staging in patients with CHB while simultaneously establishing a risk assessment framework for individuals susceptible to rapid fibrosis progression. This advancement demonstrates substantial positive clinical implications, as it not only reduces diagnostic costs but also enables timely intervention, thereby ultimately improving patient prognosis.