Impact of Hepatic Steatosis on Fibrosis and Treatment Response in Chronic Hepatitis B: a Longitudinal Cohort Study

• There was high incidence of hepatic steatosis among patients with CHB.
• Patients with chronic hepatitis had significantly higher BMI compared to patients with chronic infection.
• The rates of CVR are higher in CHB patients without HS.

Because both NAFLD and CHB infection share the ability to cause progression of the hepatic inflammation and fibrosis, the interaction between these two diseases has become more and more interesting to research [6,7]. However, the effect of NAFLD on the outcome of CHB infection is not well established, and various studies showed conflicting results. Most of the previous studies concluded that hepatic steatosis (HS) was associated with hepatitis B surface antigen (HBsAg) seroclearance [8-12] and low HBV deoxyribonucleic acid (DNA) serum levels [13, 14]. However, other studies found that HS did not affect HBV viral load [15,16]. HS was found to be a contributing factor in liver fibrosis progression and HCC development [12, 17-19]. On the contrary, multiple studies have revealed that HS is not a risk factor for fibrosis progression or HCC development in CHB patients [6,20, 21]. According to a recent study, CHB cases with high controlled attenuation parameter (CAP) measures had incomplete response to antiviral medications [22]. Other studies, however, revealed no connection between NAFLD and long-term biochemical or virological response to antiviral therapy [20]. To our best knowledge, data about the effect of HS on CHB patients in the Middle East are scarce; therefore, we conducted this study aiming to assess the associations between biopsy-proven HS and HBV viral load, liver biochemistry, and liver fibrosis and inflammation in CHB patients. This study also aims to evaluate the effect of HS on the complete virological response (CVR) within 2 years of follow-up.

Statistical analysis

SPSS version 16 (SPSS Inc., Chicago, USA) was used to assess the data. Categorical data were expressed as numbers and percentages, whereas continuous data were expressed as mean ± standard deviation (SD) or median and interquartile range (IQR). For continuous data, examination of the differences between two groups was done using the Student t-test or Mann-Whitney U test, while the Chi-square test was used to analyze categorical data. The complete virological response over time was assessed using the Kaplan-Meier survival analysis test. Pie and bar charts were expressed by excel.

RESULTS

The current study was conducted on 91 CHB patients. Their mean age was 36±9.7years with an age range of 20-60 years. Of them,84 patients (92%) were HBeAg negative and only seven (8%) patients were HBeAg positive. Sixty patients (69%) had positive hepatitis B e antibody (Anti-HBe). When evaluating the studied population according to the CHB stage [23], we found that only one patient (1%) had HBeAg positive chronic HBV infection, seven patients (8%) had HBeAg positive chronic hepatitis, 29 patients (32%) had HBeAg negative chronic HBV infection, and 54 patients (59%) had HBeAg negative chronic hepatitis (Figure1). Patients with chronic hepatitis had significantly higher BMI and alanine transaminase (ALT) levels compared to patients with chronic infection (p = 0.003, 0.01, respectively). The prevalence of HS was higher in patients with chronic hepatitis compared to patients with chronic infection but without statistical significance. Patients with chronic hepatitis had significantly higher scores of lobular inflammations compared to those with chronic infection (p= 0.02) (Table 1). Forty-seven percent of the studied cohort (43 patients) showed HS on liver biopsy (Figure 2). A mild degree of HS was reported in 28 CHB patients (31%), a minimal degree of HS was reported in 11 patients (12%), and a moderate degree of HS was reported in 4 CHB patients (4%). However, a severe degree of HS was not reported (Figure 3). According to the presence of steatosis on liver biopsy, we divided our population into two groups: the first group included CHB patients with HS (n=43), while the second group included CHB patients without HS (n=48). The age and gender distribution had no significant difference between the two groups. The BMI, aspartate transaminase (AST), total bilirubin, and splenic size were significantly higher in CHB patients with HS than those without (p = 0.02, 0.03,0.001, respectively). Moreover, patients with HS had higher HBV DNA and ALT levels but without statistical significance. The severity of hepatic fibrosis was greater in patients with HS than those without steatosis but without a statistical significance (p = 0.08) (Table 2). High lobular inflammation scores had a strong significant association with advanced fibrosis stages (P=0.000) (Table3).

All patients with HBeAg positive and negative hepatitis (61 patients) received medical treatment in the form of lamivudine 100 mg tablets once daily. During follow up six patients were excluded due to poor compliance. Fifty- five patients were followed for 2 years to evaluate the virological response. CVR at 6 months, 12 months, and 24 months after treatment onset was significantly higher in CHB patients without HS compared to those with HS (p = 0.01. 0.000,0.06, respectively). As CVR at 6^th month of treatment was reported in 88% of CHB patients without HS versus 57%of those with HS. Moreover, after 1 years of follow up 100% of CHB without HS achieved CVR, while only 56.7% of those with HS achieved CVR with one patient reported virological breakthrough (Table 4).

Table 1: Baseline clinical, laboratory and histopathological characteristics of the studied population according to CHB stage (N=91).

The increasing global prevalence of NAFLD raised the interest of researchers to study the interaction between NAFLD and other causes of chronic liver diseases. Our biopsy-based longitudinal cohort study evaluated different aspects of the relationship between NAFLD and CHB infection, including HBV viral load, liver biochemistry, liver fibrosis, inflammation, and the effect of HS on virological response to antiviral therapy. We found a high prevalence of biopsy-proven HS (about 50%) in CHB patients. This is consistent with previously published data from Turkey [28,29], Italy [30], Taiwan [31], and China [32]. A low prevalence of HS in CHB patients was found by other investigators (5% to 34%) [33-35]. However, these studies were based on imaging modalities rather than liver biopsy, which is the gold standard for the diagnosis and grading of HS. The effect of HBV on lipid metabolism is complex. Animal studies showed that both lipid degradation and lipid synthesis were enhanced by HBV infection [36]. HBV X protein was found to be a key factor in regulating HS. HBV X protein enhances the gene expression of different lipogenic enzymes leading to fat deposition in hepatocytes and consequently HS [37-39]. Moreover, HBV DNA was found to be a contributing factor to HS by interfering with Sterol regulatory element binding proteins-1c (SREBP-1c) expression [40]. However, host factors such as metabolic syndrome have a predominant role in the induction of HS and cannot be neglected. In our study, we found that CHB patients with HS had a higher BMI, which is in agreement with many authors [28,31,33,41]. In addition, we found that CHB patients with active disease had a significantly higher BMI compared to patients with stable disease, while no significant difference was found between the two groups as regards HS frequency. This finding confirms the proposed role of obesity in the progression of fibrosis. Adiponectin, a powerful adipose-derived anti-inflammatory mediator, is frequently deficient in obesity. Adiponectin deficiency contributes to hepatic stellate cell (HSC) activation, fibrosis, and increased production of inflammatory mediators. Animal models showed that adiponectin knockout mice kept on a high-fat diet had more severe fibrosis than wild controls [42], whereas carbon tetrachloride 4 induced fibrosis is reduced in mice after receiving adiponectin [43].

Our data showed that CHB patients with HS tended to have higher HBV DNA levels than those without HS, but without statistical significance. Previous literature investigated the relationship between HBV DNA and HS with conflicting results. Two retrospective studies from China showed that HS was associated with metabolic factors rather than viral factors; namely HBV DNA and HBeAg status [44,45]. Moreover, many other studies did not find any relation between HBV DNA levels and NAFLD in CHB patients [33,46,47]. On the other hand, Xiaoyan et al [48].  found that HBV DNA levels were significantly higher in lean CHB patients with minimal HS compared to those without HS. However, the proposed protective effect of HS in CHB patients was attributed to the inverse relationship between HS and HBV DNA documented by many authors [28,49,50].  Nevertheless, the diagnosis of HS in many of these studies was imaging-based; using abdominal ultrasound or CAP measurements which is less accurate than liver biopsy; and this is a major strength in our study. Furthermore, the mechanism of the inverse relationship between HS and HBV DNA is still unclear. To our knowledge, few studies investigated this possible mechanism, and it was claimed that HS suppresses HBsAg and HBV DNA secretion directly through induction of hepatocyte endoplasmic reticulum (ER) stress [51], or indirectly through induction of hepatocyte apoptosis ¹⁰. Undoubtedly, more thorough research is required to uncover the relationship between HS and HBV DNA and its possible mechanisms.

The current study found that the frequency of advanced fibrosis (F3 and F4 METAVIR score) was higher in HS group but without a statistical significance. HS has been identified by many authors as a factor contributing to the burden of liver fibrosis in CHB patients. Two cross-sectional studies from China reported that severe steatosis measured by CAP score was associated with severe fibrosis in both treatment naïve CHB patients and those who received treatment [14,19]. A possible explanation of the association of HS with the progression of fibrosis is NAFLD-related lipotoxicity, with the production of excess free radicals through endoplasmic reticulum stress [52].  On the contrary, many authors did not find a relation between HS and the stage of fibrosis [4,53-55].  In addition, Bondini et al [56]. reported that liver fibrosis in CHB patients was related to viral and metabolic factors rather than HS. Moreover, we found that the severity of fibrosis was linked to biopsy-proven lobular inflammation rather than the presence of steatosis itself, and this is partially consistent with the results of Huang et al [57] who reported a significant association of the severity of fibrosis with both lobular inflammation and cytological ballooning rather than steatosis. An interesting finding was documented by a retrospective biopsy-based study from China, as they found that HS had an inverse relationship to the grade of inflammation and the stage of fibrosis [44]. Thus, we can say that convincing evidence linking HS to hepatic fibrosis is still lacking.

Another major finding of the current study is that HS had a negative impact on the virological response to oral antiviral therapy in CHB patients. We observed significantly higher rates of CVR in the non-steatotic group at 6 months, 12 months, and 24 months after the onset of lamivudine therapy. The negative impact of HS on the treatment response was also observed by Zhu et al [41]. They found that HBeAg positive patients with HS had a lower total virological response rate at 12 weeks of entecavir therapy compared to those without HS. This is consistent with Chen et al [57]. who observed a poor response to antiviral therapy in CHB patients with high CAP measurements at 12, 24, and 48 weeks of therapy compared to those with normal CAP measurements. Another prospective study observed that the rate of HBV-DNA clearance was significantly higher at 24, 48, and 96 weeks of entecavir therapy in CHB patients without HS [58]. On the contrary, Li et al [33]. observed comparable cumulative rates of complete viral suppression in CHB NAFLD and CHB non-NAFLD groups at 12, 24, 36, 48, and 60 months of oral antiviral therapy. Despite the advantage of long follow-up duration, the study of Li et al. used radiological studies as the main methods for diagnosis of HS, and only a few cases underwent liver biopsy. Moreover, DNA quantification was done using PCR testing with a relatively high detection limit (20-100 IU/ml) compared to that of the testing used in the current study (16 IU/ml), making the definition of CVR variable in the two studies. The possible negative effect of HS on treatment response could be attributed to hepatocellular fat droplet accumulation, which reduces the bioavailability of nucleoside analogues’ intrahepatic metabolites [59]. The effectiveness of therapy may also be impacted by decreased hepatic cytochrome activity in steatotic hepatocytes, insulin resistance, and obesity coexisting with HS, which results in dysfunction of cellular immunity [58].

Despite our best efforts, this study has some limitations. First, it is a single-center study with a small sample size. However, it was difficult to readily increase the sample size due to the biopsy-based nature of the study; which would be challenging to approach in future prospective studies given the emergence of noninvasive techniques. Undoubtedly, we performed a liver biopsy only when it was indicated. Secondly, a small proportion of our cohort had positive HbeAg. Thirdly, we evaluated the virological response to lamivudine only, which was the only available oral antiviral drug in our locality at the time of data collection. Lastly, we followed our cohort for 24 months only after the start of therapy, which is a relatively short period to detect treatment resistance.

To conclude, the results of this longitudinal cohort biopsy-based study of CHB patients confirmed the high prevalence of HS among CHB patients. We also detected that CHB activity is related to BMI rather than HS. We did not find a significant association between HS with HBV DNA and the severity of fibrosis. The rates of CVR were higher in the non-HS group, emphasizing the need for further studies with large sample sizes to confirm the results and to identify if combined CHB-HS necessitate any special considerations when choosing the suitable CHB treatment line.

Ethical statement: This study conformed to the ethical guidelines of the Declaration of Helsinki and was approved by the Scientific Research Ethical Committee, Faculty of Medicine, Sohag University (IRB Registration number: Soh-Med-22-12-26).

Conflict of interest: The authors declare that there are no conflicts of interest.

Funding:  There is no funding.

Authors^, contribution: Conceptualization, formal analysis, and original draft writing; Mona Mohammed Abdelrahman. Resources provision, data curation management, and supervision; Amira Maher, Nagwa Abd El-sadek Ahmed. Methodology, review, and editing; all authors.

Acknowledgments: We are grateful for our patients and our collages who help us to complete this research.

Clinical trial registration: NCT05678582, retrospectively registered, https://clinicaltrials.gov/study/NCT05678582.

Abbreviations: ALT: alanine transaminase; AST: aspartate transaminase; BMI: body mass index; CAP: controlled attenuation parameter; CHB: chronic hepatitis B; CVR: complete virological response; DNA: deoxyribonucleic acid; HBV: hepatitis B virus; HBsAg: hepatitis B surface antigen; HCC: hepatocellular carcinoma; HS: hepatic steatosis; INR: international randomized ratio; NAFLD: non-alcoholic fatty liver disease; PT: prothrombin time.