Serum ApoB/ApoA1 ratio in patients with CHB and the occurrence of HBV related cirrhosis and HBV related hepatocellular carcinoma

Clinical research has suggested that chronic HBV infection exerts a certain effect on the occurrence of cardiovascular disease by regulating cholesterol metabolism in liver cells. High serum apolipoprotein B/apolipoprotein A1 (ApoB/ApoA1) ratio plays a certain role in the above regulation, and it serves as a risk factor for cardiovascular disease. However, whether the ApoB/ApoA1 ratio is correlated with chronic HBV infection and its disease progression remains unclear. In accordance with the inclusion and exclusion criteria, all 378 participants administrated at Renmin Hospital of Wuhan University from March 2021 to March 2022, fell into Healthy Control (HC) group (50 participants), Hepatocellular carcinoma (HCC) group (107 patients), liver cirrhosis (LC) group (64 patients), chronic hepatitis B (CHB) group (62 patients), chronic hepatitis C (CHC) group (46 patients) and Hepatitis E Virus (HEV) group (49 patients). Serum ApoA1 and ApoB concentrations were measured at admission, and the ApoB/ApoA1 ratio was determined. The levels of laboratory parameters in the respective group were compared and ApoB/ApoA1 ratios in HCC patients and LC patients with different severity were further analyzed. ROC curves were plotted to analyze the early diagnostic ability of ApoB/ApoA1 ratio for HBV-associated HCC. Logistic regression and restricted cubic spline analysis were used to explore the correlation between ApoB/ApoA1 ratio and LC and HCC risk. A comparison was drawn in terms of ApoB/ApoA1 ratio between the groups, and the result was expressed in descending sequence: HEV group > CHB group > LC group > HCC group > CHC group > HC group, early-stage HCC < middle-stage HCC < advanced-stage HCC, Class A LC < Class B LC < Class C LC. Serum ApoB/ApoA1 ratio combined diagnosis with AFP exhibited the capability of increasing the detection efficacy and specificity of AFP for HCC and AFP-negative HCC. The incidence of LC and HCC in the respective logistic regression model showed a negative correlation with the serum ApoB/ApoA1 ratio in CHB patients (P < 0.05). After all confounding factors covered in this study were regulated, the result of the restricted cubic spline analysis suggested that in a certain range, serum ApoB/ApoA1 ratio showed an inverse correlation with the prevalence of LC or HCC in CHB patients. Serum ApoB/ApoA1 ratio in CHB patients may be conducive to identifying high-risk patients for HCC or LC, such that LC and HCC can be early diagnosed and treated.

standard for diagnosis and treatment of primary liver cancer (2022 edition) 26 , deliberation on the diagnostic criteria for hepatitis E 27 .HCC and LC patients covered were correlated with CHB only, never received any therapy for HCC or LC, and negative in hepatitis B e antigen (HBeAg).All HCC patients had LC.Exclusion criteria for included patients are elucidated as follows: (1) Malignant tumors other than primary liver cancer.(2) Other hepatitis virus infections other than HBV, HCV, HEV.(3) Incomplete basic information such as medication history and previous disease history.(4) Severe diabetes, hyperthyroidism, and cardiovascular disease.(5) Pregnant patients.Figure 1 presents the specific research object inclusion and exclusion details. Figure 1 illustrates the specific research object inclusion and exclusion details.
This study was reviewed and approved by the Medical Ethics Review Committee of Renmin Hospital of Wuhan University (WDRY2022-K117) and approved to exempt patients from informed consent.

Assessment of liver cirrhosis severity
In all LC patients the severity of illness was assessed using a modified Child-Pugh's classification.As depicted in Table 1.

Clinic and laboratory data
Points for increasing abnormality

Statistical analysis
The experimental data were analyzed with SPSS version 20.0 (IBM Corp, NY), MedCal 15.2.2 (Ostend, Belgium) and GraphPad Prism 6.0 (GraphPad Software, Inc, La Jolla, CA).In accordance with the measurement data, whether the data of the respective group conform to normality was examined using the single-sample Kolmogorov-Smirnov method, the normal distribution data was expressed as χ ± s, the comparison between multiple groups were performed through analysis of variance, and the further pairwise comparison was drawn through LSD-t test.The non-normal distribution data were expressed as M (P 25 -P 75 ).The comparison between multiple groups was drawn through the Kruskal-Wallis H test, the Mann-Whitney U test was performed to draw the comparison between the two groups, and the pairwise comparison was conducted based on the Bonferroni adjustment test level method.The diagnostic value was calculated using the stepwise logistic regression model.P < 0.05 indicated a difference that achieved statistical significance.Simple and multiple logistic regression was used to investigate the correlation between ApoB/ApoA1 ratio and LC prevalence in CHB and HCC prevalence in LC.Restricted cubic spline analysis was used to investigate the non-linear correlation between ApoB/ApoA1 ratio and LC prevalence in CHB and HCC prevalence in LC.

Ethical approval and consent to participate
This study was approved by the Medical Ethics Review Committee of Renmin Hospital of Wuhan University, China, and complies with the Declaration of Helsinki and all participants signed informed consent.As depicted in Fig. 2B, serum ApoB/ApoA1 ratio in patients suffering from advanced-stage hepatocellular carcinoma [0.89(0.71,1.26)] were notably higher than those in patients subjected to early-stage hepatocellular carcinoma [0.47(0.35,0.66)] and those in patients carrying middle-stage hepatocellular carcinoma [0.61(0.49,0.74)] (P < 0.05); serum ApoB/ApoA1 ratio in patients with middle-stage hepatocellular carcinoma were significantly higher than those in patients suffering from early-stage hepatocellular carcinoma (P < 0.05).As depicted in Fig. 2C, serum ApoB/ApoA1 ratio in patients with Class C liver cirrhosis [1.40 (0.80, 2.06)] were significantly higher than those in patients with Class A liver cirrhosis [0.69 (0.53, 0.95)] (P < 0.05), serum ApoB/ApoA1 ratio in patients with Class B liver cirrhosis [0.82 (0.68, 1.31)] had no statistically significant difference with patients subjected to Class A and Class C liver cirrhosis (P > 0.05).
As depicted in Fig. 2F and G, there is further analysis of the diagnostic role of ApoB/ApoA1 ratio in differentiating HCC from LC or CHB and in combination with AFP, a common diagnostic marker for hepatocellular carcinoma.After combining the ApoB/ApoA1 ratio and AFP, the area under the curve for the combined index was significantly higher compared with the AUC for ApoB/ApoA1 ratio alone (P < 0.05).The diagnostic specificity of the combined index was higher than that of ApoB/ApoA1 ratio and that of AFP.As depicted in Tables 3  and 4. As shown in Fig. 2H and Table 5, the ApoB/ApoA1 ratio could reach an AUC of 0.773 (0.703-0.844) and Specificity of 83.87% in the diagnosis of HCC with HC.After combining AFP, the AUC of the combined index was elevated to 0.782 (0.749-0.862), and Specificity was elevated to 95.16%.We analyzed HCC patients with AFP < 10 ng/ml as a separate subgroup in order to further explore the diagnostic value of ApoB/ApoA1 ratio, and tentatively named this group as AFP-negative HCC subgroup.As shown in Fig. 2I and Table 5, the diagnostic AUC of the indicator after the combination of ApoB/ApoA1 ratio and AFP was 0.736 (0.632-0.824), and Specificity could reach 98.00%.

Logistic analysis of the prevalence of HCC in LC patients and prevalence of LC in CHB patients and ApoB/ApoA1 ratio
To analyze the correlation between ApoB/ApoA1 ratio and HCC in the LC patients in depth, and the correlation between ApoB/ApoA1 ratio and LC in the CHB patients, the data fell into quartiles of ApoB/ApoA1 ratio, with www.nature.com/scientificreports/ the first quartiles as the reference to determine the odds ratio (OR) for HCC or LC, and the results are listed in Tables 6 and 7.The prevalence of HCC in LC patients and prevalence of LC in CHB patients dispalyed a negative correlation with the ApoB/ApoA1 ratio in the Crude model (P < 0.01).After age and gender regulated in Model 1, the results were similar to those of Crude model (P < 0.01).The difference continued to achieve statistical significance after further control of HBV infection time, anti-HBV therapy, smoking, drinking, Fn, CHE, FFA, Urea, TCh, TG, HDL-C, LDL-C, sdLDL-C, T/HDL-C, LP(a), LDH, ALT, AST, ALP and GGT.The fully regulated OR in Model 2 was 0.043(95% CI: 0.009, 0.201) for quartile 4 of circulating ApoB/ApoA1 ratio in the LC patients (the highest) versus quartile 1 (the lowest); it reached 0.196 (95% CI: 0.037, 1.033) for quartile 4 of ApoB/ApoA1 ratio in the CHB patients (the highest) versus quartile 1 (the lowest).

Restricted cubic spline analysis
Restricted cubic spline analyses found that after adjusting for all confounding factors included in this study, the inflection point occurs when the ApoB/ApoA1 ratio is 0.7, which means that when the ratio of ApoB/ApoA1 was less than 0.70, the prevalence of HCC in LC patients decreased with ratio of ApoB/ApoA1.When the ratio of ApoB/ApoA1 was greater than 0.7, the prevalence of HCC reached a plateau.When the ratio of ApoB/ApoA1 was between 0.51 and 0.86, there was negatively correlation between the prevalence of LC and the ApoB/ApoA1 ratio; when the ratio was greater than 0.86, it was a plateau for the prevalence of LC.

Discussion
This is the first study to investigate the correlation between ApoB/ApoA1 ratio and the CHB disease progression.Chronic HBV infection refers to a major public health concern.In chronic HBV infection, there may be persistent low-grade liver inflammation with transient high-grade episodes of liver inflammation and activation of fibrotic processes that trigger liver fibrosis and LC, which may eventually lead to decompensated liver disease and HCC in 25-40% of HBV carriers 28 .HBeAg and HBV DNA serve as independent and vital factors for the development of disease complications.In patients subjected to HBV infection in early childhood, most   29 .Some existing research has suggested that a direct proportion exists between HBV DNA levels and the risk of LC and HCC development, and the incidence of LC and HCC is tenfold lower in people with persistently controlled low levels of HBV DNA than in those with high levels of HBV DNA 30,31 .LC and HCC outcomes after HBV infection were still not effectively controlled at this stage, such that continuous effective antiviral therapy and early recognition and diagnosis turned out to be the focus of clinical treatment.www.nature.com/scientificreports/Considerable risk assessment models have been adopted to predict LC and HCC in CHB patients (e.g., GAG-HCC score 32 , REACH-B score 33 , as well as the PAGE-B score 34 ).GAG-HCC score predicts the occurrence of HCC in patients with hepatitis B within 5 and 10 years based on age, gender, albumin, bilirubin, HBV DNA and LC.REACH-B score predicts the occurrence of HCC in hepatitis B patients within 3, 5 and 10 years based on age, gender, HBeAg, HBV DNA and ALT.PAGE-B score predicts the occurrence of HCC in patients with hepatitis B within 5 years based on age, gender and platelet count.The above-mentioned risk scores tend to require patients to have complete and real medical record information, have some difficulties in practical clinical application, and predict less well than expected.ApoB/ApoA1 ratio refers to a common and easily obtainable indicator, showing a correlation with the prevalence of LC in CHB and the prevalence of HCC in LC.After various confounding factors were regulated, ApoB/ApoA1 ratio remained negatively correlated with the prevalence of both.In this study, ROC curves were generated.As indicated by the curves, combined diagnosis of ApoB/ApoA1 ratio and AFP exhibited the capability of improving the diagnostic efficacy and specificity of AFP for HCC, thus suggesting the clinical application value of combined diagnosis of ApoB/ApoA1 ratio and AFP.Studies have shown that not all hepatocellular carcinomas secrete high levels of AFP, and that high levels of AFP can be detected in the serum of some patients with cirrhosis and chronic hepatitis 35 , so in this paper, we set up an AFP-negative HCC group for differential diagnosis, and we found that the combined index of ApoB/ApoA1 ratio and AFP has a good diagnostic value.Thus, detection of ApoB/ApoA1 ratio takes on critical significance in identifying high-risk patients with HCC versus LC and effectively preventing CHB progression.
The role played by ApoB/ApoA1 ratio in the disease progression of CHB remains unclear, whereas multiple important mechanisms may be involved.Existing research has confirmed that HBV infection is capable of protecting infected subjects from the development of metabolic syndrome, and it shows a negative correlation with lipid metabolism 36 .The incidence of hypertriglyceridemia, hypercholesterolemia, and fatty liver in HBV infected individuals is relatively low.HBV infection increased fatty acid oxidation and reduced lipid droplets (LD) formation.The content of intracellular TGs and the mean size of individual LDs were significantly lower in HBV-infected or transfected cells compared to control cells due to reduced levels of proteins involved in LD expansion and lipid storage.HBV replication also increased adiponectin expression.Adiponectin alleviated lipid accumulation by increasing carnitine palmitoyl transferase 1 (CPT1) activity, enhancing hepatic fatty acid oxidation, and reducing ACC and FAS activity 20 .
Serum ApoB and ApoA1 decreased in CHB, HBV-associated LC, and HBV-associated HCC groups compared with healthy controls, while there were few comparative studies between the groups 21,37,38 .HBV can modulate ApoA1 function through HBx, leading to ApoA1 dysfunction (e.g., decreased self-association ability, elevated carbonyl levels, as well as impaired lipid binding ability 20,21 ).Epigenetic silencing of ApoA1 gene expression by CpG island DNA hypermethylation induced by HBV can also reduce serum ApoA1 level 39 .HBV suppresses ApoB production by inhibiting microsomal triglyceride transfer protein expression 24 , overexpression of HBx leads to negative regulation of microsomal TG transfer proteins, which can increase the expression of β-d-mannoside-1, 4-N-acetylglucosaminyl transferase-III, and cause inhibition of ApoB secretion and intracellular accumulation of TG and TCh 40 .
The higher the ApoB/ApoA1 ratio, the more cholesterol will be deposited on the arterial wall, such that atherogenesis can be triggered, and cardiovascular disease risk will be elevated 16 .In terms of CHB patients, there are evidences of a relatively high risk of platelet activation and atherothrombosis 18,19 .Perhaps in the progression of CHB disease, HBV reduces the risk of atherosclerosis through the incompletely verified molecular mechanism, resulting in a negative correlation between ApoB/ApoA1 ratio and the prevalence of HCC and LC.It has been demonstrated that HBV-infected humanized mice show a significant increase in human genes correlated with cholesterol uptake, biosynthesis and transcriptional regulation, such as low-density lipoprotein receptor (LDLR), hydroxymethylglutaryl coenzyme A reductase (HMGCR), and HBV exacerbates hepatic cholesterol accumulation by up-regulating LDLR and HMGCR in HepG2 cells 41,42 .Sodium taurocholate cotransporting polypeptide (NTCP) is capable of mediating the passage of bile salts from the hepatic portal vein blood into hepatocytes.Besides, NTCP has been identified as a receptor mediating species-specific entry of HBV into hepatocytes.HBV infection plays a certain role in limiting NTCP function.Blocking bile acid uptake can facilitate compensatory up-regulation of cholesterol 7α-hydroxylase, such that cholesterol synthesis can be promoted in cells 43,44 .HBV infection promotes hepatic cholesterol accumulation and may lower the risk of atherosclerosis.Accordingly, with the progression of CHB disease, the risk of atherosclerosis can be lowered, and the ApoB/ApoA1 ratio can be down-regulated.ApoB/ApoA1 ratio may serve as a potential marker to lay a novel basis for early identification of LC and HCC in CHB patients.

Study strength and limitations
This study first aimed to verifying that the ApoB/ApoA1 ratio is strongly correlated with CHB disease progression, and its predictive potential is not correlated with potential risk factors (e.g., age, anti-hepatitis treatment, and duration of chronic hepatitis).Thus, the above-mentioned findings have high clinical utility for the identification of high-risk patients with HCC or LC.
However, several limitations remained in this study.First, all serum samples were from Chinese, suggesting that the generalizability of the findings in different regions or ethnic groups should be validated in depth.Second, the statistical power of this study was limited by the small sample size and may be biased.Third, prognostic and survival studies were hindered by the lack of long-term follow-up data in this study.Fourth, the confounding factors regulated for in this study may not be comprehensive.

Conclusion
In brief, low serum ApoB/ApoA1 ratio was correlated with an elevated risk of LC and HCC in CHB patients.Notable difference was reported in serum ApoB/ApoA1 ratio between LC and HCC patients at different stages, and combined diagnosis with AFP exhibited the capability of increasing the detection efficacy and specificity of AFP for HCC and AFP-negative HCC.As revealed by the above-mentioned findings, the serum ApoB/ApoA1 ratio in CHB patients may be conducive to identifying high-risk patients for HCC or LC, such that LC and HCC can be early diagnosed and treated.

Figure 2 .
Figure 2. Analysis of ApoB/ApoA1 ratio in different liver diseases and different stages of HCC, LC and HC.(A) The ratio of ApoB/ApoA1 in patients with different types of liver diseases and healthy people.^P < 0.05, compared with HC group; * P < 0.05, compared with HCC group; # P < 0.05, compared with LC group; & P < 0.05, compared with CHB group; α P < 0.05, compared with CHC group.(B) The ratios of serum ApoB/ApoA1 of patients with different stages of HCC.*P < 0.05, compared with early-stage HCC; # P < 0.05, compared with middle-stage HCC.(C) Ratios of serum ApoB/ApoA1 of patients with different Child-Pugh grades of LC. * P < 0.05, compared with Class A liver cirrhosis.(D) Ratios of serum ApoB/ApoA1 before and after treatment in patients with CHB.^P < 0.05, compared with HC group; # P < 0.05, compared with CHB with treatment group.(E) Ratios of serum ApoB/ApoA1 before and after treatment in patients with HCC.* P < 0.05, compared with HCC group before treatment.(F) ROC curve of ApoB/ApoA1 ratio for diagnosis of HCC from LC. (G) ROC curve of ApoB/ApoA1 ratio for diagnosis of HCC from CHB. (H) ROC curve of ApoB/ApoA1 ratio for diagnosis of HCC from HC. (I) ROC curve of ApoB/ApoA1 ratio for diagnosis of AFP-negative HCC from HC.

Table 2
showed the characteristics of 378 hepatitis participants and the variability of each indicator between groups as superscripts.Among the 233 CHB participants, 64 participants developed LC (27.47%), and 107 participants developed comorbid LC and HCC (45.92%).The Hepatitis E group had the highest levels of AST, ALT, GGT, ALP, Fn, and most of the variations in expression compared with the other groups were statistically significant (P < 0.05).

Table 2 .
Characteristics of the study population.

Table 3 .
The performance of ApoB/ApoA1 ratio and AFP for diagnosing HCC from LC.

Table 4 .
The performance of ApoB/ApoA1 ratio and AFP for diagnosing HCC from CHB.

Table 5 .
The performance of ApoB/ApoA1 ratio and AFP for diagnosing HCC and AFP negative HCC from HC.

Table 6 .
Correlation of prevalence of HCC with serum ApoB/ApoA1 ratios in LC patients.Logistic regression was used to examine the correlations between serum levels of ApoB/ApoA1 ratio and HCC in LC patients.

Table 7 .
Correlation of prevalence of LC with serum ApoB/ApoA1 ratios in CHB patients.Logistic regression was used to examine the correlations between serum levels of ApoB/ApoA1 and LC in CHB patients.