Skip to main content
Download PDF

HCC prediction models in chronic hepatitis B patients receiving entecavir or tenofovir: a systematic review and meta-analysis

Virology Journal volume 20, Article number: 180 (2023) Cite this article

Abstract

Background

Our study aimed to compare the predictive performance of different hepatocellular carcinoma (HCC) prediction models in chronic hepatitis B patients receiving entecavir or tenofovir, including discrimination, calibration, negative predictive value (NPV) in low-risk, and proportion of low-risk.

Methods

We conducted a systematic literature research in PubMed, EMbase, the Cochrane Library, and Web of Science before January 13, 2022. The predictive performance was assessed by area under receiver operating characteristic curve (AUROC), calibration index, negative predictive value, and the proportion in low-risk. Subgroup and meta-regression analyses of discrimination and calibration were conducted. Sensitivity analysis was conducted to validate the stability of the results.

Results

We identified ten prediction models in 23 studies. The pooled 3-, 5-, and 10-year AUROC varied from 0.72 to 0.84, 0.74 to 0.83, and 0.76 to 0.86, respectively. REAL-B, AASL-HCC, and HCC-RESCUE achieved the best discrimination. HCC-RESCUE, PAGE-B, and mPAGE-B overestimated HCC development, whereas mREACH-B, AASL-HCC, REAL-B, CAMD, CAGE-B, SAGE-B, and aMAP underestimated it. All models were able to identify people with a low risk of HCC accurately. HCC-RESCUE and aMAP recognized over half of the population as low-risk. Subgroup analysis and sensitivity analysis showed similar results.

Conclusion

Considering the predictive performance of all four aspects, we suggest that HCC-RESCUE was the best model to utilize in clinical practice, especially in primary care and low-income areas. To confirm our findings, further validation studies with the above four components were required.

Background

Chronic hepatitis B virus (HBV) infection is associated with life-threatening liver conditions like hepatocellular carcinoma (HCC) [1]. The early detection of HCC and stratified care of distinct risk populations were critical to minimize the harm of liver complications. Patients could be classified into different risk levels of developing HCC over time with HCC risk prediction models. There were models developed in untreated patients like REACH-B (Risk Estimation for HCC in Chronic Hepatitis B)and NGM-HCC (Nomograms for Risk of Hepatocellular Carcinoma) [2, 3], in mixed patients like CU-HCC (Chinese University-HCC), LSM-HCC (Liver Stiffness Measurement based-HCC), GAG-HCC (Guide with Age, Gender, HBV DNA, Core promoter mutations and Cirrhosis), and aMAP (the Age-Male-ALBI-Platelets Score), and models developed in treated patients [4,5,6].

Given that antiviral therapy was commonly employed in the present society, models developed in treated patients may have a greater advantage in the accuracy of predictions. The mREACH-B (modified REACH-B), PAGE-B (Platelet, Age, Gender and HBV), mPAGE-B (modified PAGE-B), HCC-RESCUE (HCC-Risk Estimating Score in CHB patients Under Entecavir), CAMD (the Cirrhosis, Age, Male sex, and Diabetes Mellitus Score), AASL-HCC (Age, Albumin, Sex, Liver Cirrhosis-HCC Scoring System), CAGE-B (Cirrhosis and Age Score), SAGE-B (Stiffness and Age Score), and REAL-B (Real-world Effectiveness from the Asia Pacific Rim Liver Consortium for HBV) were initially developed in patients treated with different antiviral drugs [7,8,9,10,11,12,13,14,15]. It was important to combine information from all derivation and external validation studies for the same model to assess the prediction performance across diverse populations. Furthermore, for some low-risk patients, needless HCC screening and surveillance may result in potential physical, financial, and psychological harms [16]. According to the guideline, HCC surveillance was cost-effective if the annual risk of HCC was ≥ 0.2% per year [17]. As a result, the fraction of low-risk population highlighted by models, as well as the ability to exclude individuals who are unlikely to develop HCC, should be considered.

Presently, entecavir and tenofovir were the first-line medications suggested in the guidelines for antiviral treatment [1, 18, 19]. Thus, our study systematically assessed the prediction performance of the above models in patients treated with entecavir and tenofovir in a meta-analysis.

Methods

The systematic review and meta-analysis was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) and was registered on PROSPERO (ID: CRD42022303167).

Search strategy

We searched literatures published before January 13, 2022 in PubMed, EMbase, the Cochrane Library, and Web of Science. There were no limits on language or publication dates. Keywords of HCC, CHB, prediction models, et al. were used (Content 1 in Additional file 1). We also looked up references in relevant reviews and original publications to see if there were any studies we had overlooked.

Selection criteria

HCC prediction models built in treated CHB patients were selected in our study, including mREACH-B, PAGE-B, mPAGE-B, HCC-RESCUE, CAMD, AASL-HCC, CAGE-B, SAGE-B, and REAL-B. Even though aMAP was created in a mixed population, we included it in our study due to the large sample size in the derivation study. Both derivation studies and validation studies were retrieved. Exclusion criteria were as follows: (1) reviews or meta-analyses, (2) conference abstracts, (3) letters, editorials, and case reports, (4) full-text not in English, (5) update models without external validation, (6) validation cohort including untreated patients or patients treated with other oral antiviral medications, (7) insufficient data for analysis. The study focused on 3-, 5-, and 10-year HCC prediction performance. XX and LJ independently examined titles and abstracts, and studies that met the inclusion criteria were retrieved for full-text evaluation. Two independent investigators were also responsible for data extraction (XX and LJ). Any discrepancies were resolved by a third investigator (YZ).

Data extraction

The following data were extracted from these studies: author, publication year, study type, region, race, setting, recruitment period, sample size, follow-up duration, HCC cases, study interval, type of antiviral treatment received, baseline demographic and medical history (age, sex, proportion of cirrhosis, alcohol abuse, and diabetes), baseline laboratory results (hepatitis B e antigen [HBeAg] status, HBV DNA quantitative, alanine aminotransferase [ALT], platelets, albumin, total bilirubin, alpha-fetoprotein, liver stiffness measurement, and prediction score), area under the receiver operator characteristic curve (AUROC) with 95% confidence interval (CI), observed (O) events and expected (E) events, and negative predictive value (NPV) with 95% CI in the low-risk group. For external validation of different existing models or different cohorts, information was extracted separately.

According to Prediction model Risk of Bias Assessment Tool (PROBAST), which was organized into the following 4 domains: participants, predictors, outcome, and analysis, the quality of the original studies was evaluated.

Statistical analysis

Both derivation and external validation studies were included in meta-analysis. In the meta-analysis, effect measures included AUROC with 95% CI, total O:E ratio with standard error, NPV with 95% CI, and proportion in low-risk group with 95% CI. The predicted occurrences were estimated by multiplying the cumulative HCC incidence by the number of patients in each risk group. Variance of O:E ratio was calculated according to the equation recommended by Debray et al. [20]. When generating 95% CI for average performance, the random-effects model was used. I2 statistics were used to measure between-study heterogeneity. I2 statistic > 50% was regarded as moderate heterogeneity and I2 statistics > 75% was considered as severe heterogeneity. Subgroup analysis was conducted by cirrhotic status (cirrhotic/non-cirrhotic) and race (Caucasian/Asian) to explore the heterogeneity. Meta-regression analysis adjusted by the Hartung-Knapp method was conducted. Besides, sensitivity analysis by omitting anyone research was conducted to study the impact of individual studies on the average performance. Publication bias was analyzed by funnel plots for models that included ten or more studies [21]. All analyses were performed using Review Manager version 5.4 (Cochrane, London, United Kingdom) and StataMP software version 15 (StataCorp LLC, Texas, USA).

Results

We identified 23 publications for the systematic review and meta-analysis after screening 4374 studies from four databases, which included 153,445 CHB patients and 5133 HCC cases (Fig. 1). External validation was not performed in the original research for three model derivation investigations (mREACH-B, PAGE-B, CAGE-B, and SAGE-B). External validation studies or derivation and external validation studies of six models made up the remaining research (HCC-RESCUE, CAMD, mPAGE-B, AASL-HCC, REAL-B, and aMAP). PAGE-B and mPAGE-B were the most commonly externally validated in 19 and 14 studies, respectively, whereas REAL-B and mREACH-B were only validated in one study, respectively. Other models were also frequently validated as follows: CAMD (n = 6), HCC-RESCUE (n = 5), AASL-HCC (n = 4,), CAGE-B and SAGE-B (n = 3), and aMAP (n = 2).

Fig. 1
figure 1

Flow diagram for the systematic analysis and meta-analysis

Full size image

Characteristics of the included studies

The participants were recruited retrospectively using hospital medical records, whereas Hsu et al. and Yip et al. used an insurance database and the Clinical Data Analysis and Reporting System to perform their studies [11, 22]. Different from other studies, Gui et al. compared model performance in cirrhotic patients [23], and Kim et al. studied veterans in United States [24]. Most models were developed in Asian populations, except for PAGE-B, CAGE-B, and SAGE-B, which were derived from Caucasian populations. Except for REAL-B, which was developed in individuals whose treatment regimen included other oral antiviral medicines, most models were developed in patients treated with entecavir or tenofovir. And aMAP was developed in mixed patients with a treatment proportion of 78%. The number of parameters in the models ranged from three to seven. Age and sex were nearly included in all models and other parameters included albumin, total bilirubin, platelets, cirrhosis, liver stiffness measurement, ALT, HBeAg status, diabetes, alcohol abuse, and alpha-fetoprotein (Table 1). The REAL-B and aMAP derivation cohorts were not included in the meta-analysis because their participants did not match the inclusion criteria.

Table 1 Summary of hepatitis B virus-hepatocellular carcinoma prediction models in the derivation studies
Full size table

Risk of bias and applicability assessment

The details of the risk of bias and applicability were depicted in Table S1-2 and Figure S1-2. According to PROBAST, the predictors and outcome had a low risk of bias, but the participants and analysis had a high risk of bias in 17.4% and 52.1% of studies, respectively. In terms of analysis, model calibration was not performed in eight studies (34.8%) and four studies (17.4%) had a small number of HCC cases. Except for 17.4% of research, which had a high risk of participants, most models had a low risk of applicability.

Fig. 2
figure 2

The discrimination (A), calibration (B) performance and negative predictive values in the low-risk group (C) of HCC prediction models in meta-analysis. aHCC events were not reported by Hsu et al. [11], which included 17,984 participants in the study. AUROC, area under the receiver operator characteristic curve; CI, confidence interval; O:E ratio, observed events versus expected events ratio; NPV, negative predictive value; HCC, hepatocellular carcinoma; mREACH-B, Modified Risk Estimation for Hepatocellular Carcinoma in Chronic Hepatitis B; PAGE-B, Platelet, Age, Gender and HBV; mPAGE-B, modified Platelet, Age, Gender and HBV; HCC-RESCUE, HCC-Risk Estimating Score in CHB patients Under Entecavir; CAMD, the Cirrhosis, Age, Male sex, and Diabetes Mellitus Score; AASL-HCC, Age, Albumin, Aex, Liver Cirrhosis-HCC scoring system; aMAP: the Age-Male-ALBI-Platelets Score; CAGE-B, Cirrhosis and Age Score; SAGE-B, Stiffness and Age Score; REAL-B, Real-world Effectiveness from the Asia Pacific Rim Liver Consortium for HBV

Full size image

Meta-analysis

The characteristics of included studies in meta-analysis were shown in Table 2. The pooled 3-, 5-, and 10-year AUROC varied from 0.72 to 0.84 (aMAP: 0.72, 95% CI 0.70–0.75; REAL-B: 0.84, 95% CI 0.78–0.90), 0.74 to 0.83 (mREACH-B: 0.74, 95% CI 0.71–0.76; AASL-HCC: 0.83, 95% CI 0.79–0.86), and 0.76 to 0.86 (SAGE-B: 0.76, 95% CI 0.70–0.83; mPAGE-B: 0.86, 95% CI 0.83–0.89), respectively (Fig. 2, Figure S3, Table S3). When predicting 3-year HCC incidence, REAL-B, AASL-HCC, and HCC-RESCUE models had better discrimination with an AUROC > 0.80, while mREACH-B, PAGE-B and aMAP showed an AUROC < 0.75. When predicting 5-year HCC incidence, AASL-HCC, REAL-B, HCC-RESCUE, and aMAP models performed better with an AUROC ≥ 0.80, followed by mPAGE-B, CAMD, and PAGE-B with an AUROC > 0.75, while mREACH-B showed an AUROC < 0.75. When predicting 10-year HCC incidence, with an AUROC ≥ 0.80, mPAGE-B, HCC-RESCUE, CAMD, and AASL-HCC outperformed CAGE-B, PAGE-B, and SAGE-B (> 0.75).

Table 2 Characteristics of derivation and external validation cohorts included in the meta-analysis
Full size table

The pooled 5- and 10-year total O:E ratio ranged from 0.25 to 1.83 (mPAGE-B: 0.25, 95% CI 0.18–0.31; CAMD: 1.83, 95% CI 1.31–2.35) and 1.99 to 2.10 (SAGE-B: 1.99, 95% CI 0.99–2.99; CAGE-B: 2.10, 95%CI 1.02–3.17), respectively (Fig. 2, Figure S4). The pooled 3-year total O:E ratio of CAMD was 0.77 (95% CI 0.51–1.04). HCC-RESCUE, PAGE-B, and mPAGE-B exhibited an overestimation of HCC development, while AASL-HCC, aMAP, CAMD, CAGE-B, and SAGE-B exhibited an underestimation of HCC development.

The pooled 3-, 5-, and 10-year NPVs ranged from 98.3 to 100% (aMAP: 98.3%, 95% CI 96.3-100.3%; REAL-B: 100%, 95% CI 99.5-100.5%), 99.58–100% (aMAP: 99.6%, 95% CI 99.2–100.0%; AASL-HCC: 100%, 95% CI 99.5-100.5%; HCC-RESCUE: 100%, 95% CI 99.6-100.5%; REAL-B: 100%, 95%CI 99.5-100.5%), and 99.6–100% (PAGE-B: 99.6%, 95% CI 98.4-100.8%; CAGE-B: 100%, 95% CI 99.4-100.7%; SAGE-B: 100%, 95%CI 99.4-100.6%), respectively (Fig. 2, Table S4). All models had a high NPV over 99.5% except for aMAP. The proportion of low-risk population ranged from 14.4 to 53.0% (CAGE-B: 14.4%, 95% CI 12.9–16.0%; aMAP: 53.0%, 95% CI 28.5–77.6%) (Table 3). More than half of the population was identified as low-risk by HCC-RESCUE and aMAP.

Table 3 The proportion of low-risk population classified by the models in meta-analysis
Full size table

Subgroup analysis and meta-regression

The results of subgroup analysis for discrimination and calibration were presented in Table S5. Only three researches compared the performance of PAGE-B, mPAGE-B, and aMAP in cirrhotic and non-cirrhotic individuals. The discrimination performance was generally better in non-cirrhotic patients than cirrhotic patients. PAGE-B, mPAGE-B, HCC-RESCUE, CAMD, and aMAP exhibited greater AUROC in Caucasians, whereas AASL-HCC, CAGE-B, and SAGE-B had comparable discrimination performance in Asians and Caucasians. Several articles reported the model calibration performance, but the difference in cirrhotic and non-cirrhotic population was not reported. The calibration of the subgroup analysis by race was same as that in meta-analysis. And the underestimation of CAMD seems to be more pronounced in Asians than in Caucasians (O:E ratio 2.38 vs. 1.55). We did a meta-regression analysis for model discrimination and calibration performance and found the heterogeneity could not be explained by race (Figure S5).

Publication bias and sensitivity analysis

The funnel plots for the PAGE-B and mPAGE-B model on 5-year discrimination performance were symmetric visually (Fig. 3). Funnel Plots for other models were not analyzed because the number of included studies was small. We mainly discussed the 5- and 10-year predictive performance of model discrimination and calibration, NPV in low-risk, and proportion of low-risk. External validation investigations of REAL-B and mREACH-B were insufficient for sensitivity analysis. After excluding any one research, the pooled 5- or 10-year AUROC of PAGE-B, mPAGE-B, HCC-RESCUE, CAMD, AASL-HCC, CAGE-B, SAGE-B, and aMAP did not change considerably, as shown in Figure S6-7. Sensitivity analysis of calibration was shown in Figure S8-9, and variations in 5-year O:E ratio prediction of CAMD were evident in studies by Hsu and Kim [11, 25]. In Yip et al’s 5-year NPV estimate [22], there was a clear variance in PAGE-B and mPAGE-B (Figure S10). The proportion of low-risk patients detected by AASL-HCC, aMAP, CAMD, PAGE-B, and mPAGE-B did not change significantly in the sensitivity analysis (Figure S11).

Fig. 3
figure 3

Funnel plot with pseudo 95% confidence limits of 5-year AUROC of PAGE-B (A) and mPAGE-B (B)

PAGE-B, Platelet, Age, Gender and HBV; mPAGE-B, modified Platelet, Age, Gender and HBV; AUROC, area under the receiver operator characteristic curve

Full size image

Pair-wise comparison between HCC-RESCUE and other models

We further explored the meta-values of HCC-RESCUE and other models within the same investigations. Only 4 studies have compared the predictive performance of HCC-RESCUE with PAGE-B, mPAGE-B, CAMD, or AASL-HCC. As depicted in Fig. 4, the 5-year AUROC were 0.81 (95% CI 0.77–0.86), 0.80 (95% CI 0.73–0.86), 0.81 (95% CI 0.75–0.87) for HCC-RESCUE, PAGE-B, and mPAGE-B, respectively. The discrimination was also similar between HCC-RESCUE/CAMD (0.81 vs. 0.81) and HCC-RESCUE/AASL-HCC (0.81 vs. 0.83). The proportion of low-risk patients detected by HCC-RESCUE was significantly higher than that by PAGE-B or mPAGE-B (52.4% vs. 23.3% vs. 30%, Table S6).

Fig. 4
figure 4

The pair-wise comparison of 5-year AUROC between HCC-RESCUE and other models within the same investigations. (A) HCC-RESCUE, PAGE-B, and mPAGE-B; (B) HCC-RESCUE and CAMD; (C) HCC-RESCUE and AASL-HCC. AUROC, area under the receiver operator characteristic curve; CI, confidence interval; PAGE-B, Platelet, Age, Gender and HBV; mPAGE-B, modified Platelet, Age, Gender and HBV; HCC-RESCUE, HCC-Risk Estimating Score in CHB patients Under Entecavir; CAMD, the Cirrhosis, Age, Male sex, and Diabetes Mellitus Score; AASL-HCC, Age, Albumin, Aex, Liver Cirrhosis-HCC scoring system

Full size image

Discussion

We conducted a systematic review and meta-analysis of ten HCC prediction models in CHB patients receiving entecavir or tenofovir and compared their predictive performance of discrimination, calibration, NPV in low-risk, and proportion of low-risk. Overall, all models were able to generate satisfactory discrimination with an AUROC > 0.70. In terms of discrimination, calibration, and the capacity to stratify low-risk populations, HCC-RESCUE performed admirably. Different from the previous researches, we also studied the proportions of low-risk in each model and the accuracy of excluding HCC development in low-risk population.

The models included three to seven parameters including age, sex, albumin, total bilirubin, platelets, cirrhosis, liver stiffness measurement, ALT, HBeAg status, diabetes, alcohol abuse, or alpha-fetoprotein. None of the models incorporated viral-related parameters, except for mREACH-B, which included HBeAg status. However, HBV DNA level or HBeAg status were key determinants in models drawn from untreated population or mixed population (CU-HCC, GAG-HCC, LSM-HCC, NGM-HCC) [3,4,5,6]. The difference could be explained by that the antiviral therapy was effective in suppressing virus activity. During the 12-month treatment with entecavir or tenofovir, HBV DNA was undetectable in 80% and 69% of patients, respectively, according to a randomized controlled experiment [26]. HBeAg seroconversion rate was over 40% in patients who received 5-year tenofovir or entecavir treatment [27]. According to Papatheodoridis et al., Caucasian patients receiving long-term entecavir or tenofovir had an 8-year survival rate comparable to the general population if HCC had not developed [28]. Thus, viral-related factors might have little effect in predicting long-term HCC development in antiviral-treated patients. We considered these models developed in treated patients were more suitable to predict HCC incidence in the antiviral era.

According to PROBAST, the participant and analysis were the main sources of the bias. Kim et al. compared performance of different models in veterans, which could lead to a significant risk in participant selection [29]. Gui et al. verified the model performance in patients with CHB-related cirrhosis without considering those who did not develop cirrhosis [23]. While Yip and Güzelbulut et al. included decompensated cirrhosis in their validation cohorts, and we figured that model parameters would be unstable in decompensated cirrhosis, thus increasing the models’ inaccuracy [22, 30]. The bias in the analysis was mostly caused by the limited sample size, which resulted in an unreasonable number of HCC instances, as well as the fact that model performance was evaluated inappropriately due to a lack of calibration evaluation.

Overall, REAL-B, AASL-HCC, and HCC-RESCUE models had the best discrimination performance with an AUROC > 0.8. Interestingly, age, sex, and cirrhosis were all included in the above models. HCC was found to be six times more common in cirrhotic patients than in people without cirrhosis, and males were more likely to acquire HCC than females [31]. Also, the risk of developing HCC increased with age. This may indicate that age, sex, and cirrhosis-based models were more accurate in predicting HCC incidence in treated individuals. Similarly, the CAMD model, which included age, sex, diabetes, and cirrhosis, also performed well. Our findings were consistent with a prior meta-analysis that indicated REAL-B and CAMD had the best discrimination performance, in which HCC-RESCUE was not been investigated and AASL-HCC was been validated in one study [32]. Subgroup analysis showed discrimination of aMAP, PAGE-B, and mPAGE-B was better in non-cirrhotic than cirrhotic patients, which has been reported by Wu et al. [32]. However, discrimination evaluation according to cirrhotic status was not available in other model studies. Besides, there was a tendency that discrimination was better in Caucasians than Asians no matter the model was developed in Asian or Caucasian population. The reason for such a racial disparity was currently unknown. To be mentioned, mREACH-B, aMAP and REAL-B were not widely validated in patients treated with entecavir or tenofovir, further studies on these models were needed to verify our findings. Besides, predictive performance in different subgroups should be considered in further validation.

Regarding model calibration performance, HCC-RESCUE, PAGE-B, and mPAGE-B overestimated HCC development, while AASL-HCC, aMAP, CAMD, CAGE-B, and SAGE-B underestimated HCC development. Calibration was not available in REAL-B and mREACH-B. In the previous study, PAGE-B and mPAGE-B had an overestimation which was the same as our findings, while CAMD has a slight overestimation in 3-year period [32]. To classifying patients with high risk of HCC, we figured that the overestimation was preferable than underestimation. Although overestimation may cause excessive surveillance and financial waste, underestimation would lead to the omission of possible HCC patients, putting patients’ lives in jeopardy. Nevertheless, model calibration was only done in two-thirds of the studies involved. As reported in a previous meta-analysis of HCC prediction models, publication compliance with TRIPOD was 74% [33]. Following model validation studies should pay more attention to the completeness of the article according to transparent reporting of individual prognosis or diagnostic multivariate predictive model (TRIPOD) statement.

All models exhibited a high NPV over 99% in low-risk population, indicating the ability of excluding HCC development was admirable. In the 5- or 10-year study period, almost none of the low-risk patients got HCC. Therefore, intensive supervision was not necessary for these patients, potentially reducing the risk of physical, financial, and psychological harms [16]. To our knowledge, the risk stratification proportions were occasionally reported in separate studies and had not been systematically investigated by meta-analysis. To some extent, the more patients were designated as low-risk, the more medical resources could be saved. According to our findings, HCC-RESCUE and aMAP classified over half of the population as low-risk, following by CAMD and mPAGE-B with 36.8% and 26.6%, respectively, while AASL-HCC, PAGE-B, REAL-B, CAGE-B, and SAGE-B was approximately 20%. Thus, HCC-RESCUE made more patients can be spared from HCC screening and save resources, which was especially useful in primary care and low-income areas.

Our research assessed the predictive ability of ten models at multiple time points, and included subgroup analysis based on race and cirrhosis status. We also ran a sensitivity analysis to verify that the results were reliable. The between-study heterogeneity could be partly explained by race and cirrhotic status. And the results were relatively robust in sensitivity analysis. We also used NPV to assess the accuracy of identifying individuals who would not develop HCC in a given time, and we focused on the low-risk proportion divided by each model, which had never been systematically examined before. We proposed that these two characteristics be investigated in model studies since they could play a key role in allocating HCC screening resources.

However, there were several limitations in our study. First, there were insufficient validation cohorts for mREACH-B, aMAP, and REAL-B, because the mREACH-B derivation study did not illustrate the details of the model, and the latter two were newly developed [7, 14, 15]. Second, model calibration, proportion of each risk group, and NPV in low-risk populations were not depicted in every study, which caused that some models (mREACH and REAL-B) were not analyzed for these performance in the meta-analysis and the number of studies was insufficient for some models (HCC-RESCUE, REAL-B, CAGE-B, and SAGE-B). Besides, over half of the studies had a high risk of bias for participant selection or data analysis, but sensitivity analysis showed that our findings remained stable. Finally, the subgroup analysis of cirrhosis status was incomplete because the difference between cirrhotic and non-cirrhotic patients were not displayed in most cohorts. For the similar reason, the discrimination and calibration results could not be stratified according to treatment received. Further external validation studies with more complete information were needed to confirm our findings.

Conclusion and implications

REAL-B, AASL-HCC, and HCC-RESCUE performed the best discrimination in the meta-analysis of the ten prediction models, although more validation studies of model REAL-B are needed to confirm our findings. AASL-HCC, aMAP, CAMD, CAGE-B, and SAGE-B underestimated HCC development, whereas HCC-RESCUE, PAGE-B, and mPAGE-B overestimated it. Model calibration, proportion of low-risk group, and NPVs were insufficiently reported in many researches, which should be addressed in future model derivation or validation studies. In comparison to other models, HCC-RESCUE identified the most people as low-risk, with a high NPV, indicating that it might be the most appropriate model to be used in primary clinical practice for HCC surveillance.

List of abbreviations: HBV, hepatitis B virus; HCC, hepatocellular carcinoma; CHB, chronic hepatitis B; mREACH-B, modified Risk Estimation for HCC in Chronic Hepatitis B; PAGE-B, Platelet, Age, Gender and HBV; mPAGE-B, modified PAGE-B; HCC-RESCUE, HCC-Risk Estimating Score in CHB patients Under Entecavir; CAMD, the Cirrhosis, Age, Male sex, and Diabetes Mellitus Score; AASL-HCC, Age, Albumin, Sex, Liver Cirrhosis-HCC Scoring System; CAGE-B, Cirrhosis and Age Score; SAGE-B, Stiffness and Age Score; REAL-B, Real-world Effectiveness from the Asia Pacific Rim Liver Consortium for HBV; aMAP, the Age-Male-ALBI-Platelets Score; HBeAg, hepatitis B e antigen; ALT, alanine aminotransferase; AUROC, area under the receiver operator characteristic curve; CI, confidence interval; O:E ratio, observed events versus expected events ratio; NPV, negative predictive value.

Data availability

All data generated or analyzed during this study are included in this published article [and its supplementary information files].

References

  1. Terrault NA, Bzowej NH, Chang KM, Hwang JP, Jonas MM, Murad MH. AASLD guidelines for treatment of chronic hepatitis B. Hepatology. 2016;63(1):261–83.

    Article  PubMed  Google Scholar 

  2. Yang HI, Yuen MF, Chan HL, Han KH, Chen PJ, Kim DY, et al. Risk estimation for hepatocellular carcinoma in chronic hepatitis B (REACH-B): development and validation of a predictive score. Lancet Oncol. 2011;12(6):568–74.

    Article  PubMed  Google Scholar 

  3. Yang HI, Sherman M, Su J, Chen PJ, Liaw YF, Iloeje UH, et al. Nomograms for risk of hepatocellular carcinoma in patients with chronic hepatitis B virus infection. J Clin Oncol. 2010;28(14):2437–44.

    Article  PubMed  Google Scholar 

  4. Wong VW, Chan SL, Mo F, Chan TC, Loong HH, Wong GL, et al. Clinical scoring system to predict hepatocellular carcinoma in chronic hepatitis B carriers. J Clin Oncol. 2010;28(10):1660–5.

    Article  CAS  PubMed  Google Scholar 

  5. Wong GL, Chan HL, Wong CK, Leung C, Chan CY, Ho PP, et al. Liver stiffness-based optimization of hepatocellular carcinoma risk score in patients with chronic hepatitis B. J Hepatol. 2014;60(2):339–45.

    Article  PubMed  Google Scholar 

  6. Yuen MF, Tanaka Y, Fong DY, Fung J, Wong DK, Yuen JC, et al. Independent risk factors and predictive score for the development of hepatocellular carcinoma in chronic hepatitis B. J Hepatol. 2009;50(1):80–8.

    Article  PubMed  Google Scholar 

  7. Lee HW, Yoo EJ, Kim BK, Kim SU, Park JY, Kim DY, et al. Prediction of development of liver-related events by transient elastography in hepatitis B patients with complete virological response on antiviral therapy. Am J Gastroenterol. 2014;109(8):1241–9.

    Article  CAS  PubMed  Google Scholar 

  8. Papatheodoridis G, Dalekos G, Sypsa V, Yurdaydin C, Buti M, Goulis J, et al. PAGE-B predicts the risk of developing hepatocellular carcinoma in Caucasians with chronic hepatitis B on 5-year antiviral therapy. J Hepatol. 2016;64(4):800–6.

    Article  CAS  PubMed  Google Scholar 

  9. Kim JH, Kim YD, Lee M, Jun BG, Kim TS, Suk KT, et al. Modified PAGE-B score predicts the risk of hepatocellular carcinoma in Asians with chronic hepatitis B on antiviral therapy. J Hepatol. 2018;69(5):1066–73.

    Article  PubMed  Google Scholar 

  10. Sohn W, Cho JY, Kim JH, Lee JI, Kim HJ, Woo MA, et al. Risk score model for the development of hepatocellular carcinoma in treatment-naïve patients receiving oral antiviral treatment for chronic hepatitis B. Clin Mol Hepatol. 2017;23(2):170–8.

    Article  PubMed  PubMed Central  Google Scholar 

  11. Hsu YC, Yip TC, Ho HJ, Wong VW, Huang YT, El-Serag HB, et al. Development of a scoring system to predict hepatocellular carcinoma in Asians on antivirals for chronic hepatitis B. J Hepatol. 2018;69(2):278–85.

    Article  PubMed  Google Scholar 

  12. Yu JH, Suh YJ, Jin YJ, Heo NY, Jang JW, You CR, et al. Prediction model for hepatocellular carcinoma risk in treatment-naive chronic hepatitis B patients receiving entecavir/tenofovir. Eur J Gastroenterol Hepatol. 2019;31(7):865–72.

    Article  CAS  PubMed  Google Scholar 

  13. Papatheodoridis GV, Sypsa V, Dalekos GN, Yurdaydin C, Van Boemmel F, Buti M, et al. Hepatocellular carcinoma prediction beyond year 5 of oral therapy in a large cohort of caucasian patients with chronic hepatitis B. J Hepatol. 2020;72(6):1088–96.

    Article  CAS  PubMed  Google Scholar 

  14. Yang HI, Yeh ML, Wong GL, Peng CY, Chen CH, Trinh HN, et al. Real-world effectiveness from the Asia Pacific Rim Liver Consortium for HBV Risk score for the prediction of Hepatocellular Carcinoma in Chronic Hepatitis B Patients treated with oral antiviral therapy. J Infect Dis. 2020;221(3):389–99.

    Article  CAS  PubMed  Google Scholar 

  15. Fan R, Papatheodoridis G, Sun J, Innes H, Toyoda H, Xie Q, et al. aMAP risk score predicts hepatocellular carcinoma development in patients with chronic hepatitis. J Hepatol. 2020;73(6):1368–78.

    Article  CAS  PubMed  Google Scholar 

  16. Kanwal F, Singal AG. Surveillance for Hepatocellular Carcinoma: current best practice and future direction. Gastroenterology. 2019;157(1):54–64.

    Article  PubMed  Google Scholar 

  17. Terrault NA, Lok ASF, McMahon BJ, Chang KM, Hwang JP, Jonas MM, et al. Update on prevention, diagnosis, and treatment of chronic hepatitis B: AASLD 2018 hepatitis B guidance. Hepatology. 2018;67(4):1560–99.

    Article  PubMed  Google Scholar 

  18. Tanaka A. JSH Guidelines for the management of Hepatitis B Virus infection: 2019 update. Hepatol Res. 2020.

  19. European Assoc Study L. EASL 2017 clinical practice guidelines on the management of hepatitis B virus infection. J Hepatol. 2017;67(2):370–98.

    Article  Google Scholar 

  20. Debray TP, Damen JA, Snell KI, Ensor J, Hooft L, Reitsma JB, et al. A guide to systematic review and meta-analysis of prediction model performance. BMJ (Clinical research ed). 2017;356:i6460.

    PubMed  Google Scholar 

  21. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ (Clinical research ed). 1997;315(7109):629–34.

    Article  CAS  PubMed  Google Scholar 

  22. Yip TC, Wong GL, Wong VW, Tse YK, Liang LY, Hui VW, et al. Reassessing the accuracy of PAGE-B-related scores to predict hepatocellular carcinoma development in patients with chronic hepatitis B. J Hepatol. 2020;72(5):847–54.

    Article  CAS  PubMed  Google Scholar 

  23. Gui H, Huang Y, Zhao G, Chen L, Cai W, Wang H, et al. External validation of aMAP hepatocellular carcinoma risk score in patients with chronic Hepatitis B-Related cirrhosis receiving ETV or TDF therapy. Front Med. 2021;8:677920.

    Article  Google Scholar 

  24. Kim HS, Yu X, Kramer J, Thrift AP, Richardson P, Hsu YC, et al. Comparative performance of risk prediction models for hepatitis B-related hepatocellular carcinoma in the United States. J Hepatol. 2022;76(2):294–301.

    Article  CAS  PubMed  Google Scholar 

  25. Kim SU, Seo YS, Lee HA, Kim MN, Kim EH, Kim HY, et al. Validation of the CAMD score in patients with chronic Hepatitis B Virus infection receiving antiviral therapy. Clin Gastroenterol Hepatol. 2020;18(3):693–699e691.

    Article  CAS  PubMed  Google Scholar 

  26. Sriprayoon T, Mahidol C, Ungtrakul T, Chun-On P, Soonklang K, Pongpun W, et al. Efficacy and safety of entecavir versus tenofovir treatment in chronic hepatitis B patients: a randomized controlled trial. Hepatol Res. 2017;47(3):E161–e168.

    Article  CAS  PubMed  Google Scholar 

  27. Xing T, Xu H, Cao L, Ye M. HBeAg Seroconversion in HBeAg-Positive chronic Hepatitis B Patients receiving long-term Nucleos(t)ide Analog Treatment: a systematic review and network Meta-analysis. PLoS ONE. 2017;12(1):e0169444.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Papatheodoridis GV, Sypsa V, Dalekos G, Yurdaydin C, van Boemmel F, Buti M, et al. Eight-year survival in chronic hepatitis B patients under long-term entecavir or tenofovir therapy is similar to the general population. J Hepatol. 2018;68(6):1129–36.

    Article  CAS  PubMed  Google Scholar 

  29. Kim HS, Yu X, Kramer J, Thrift AP, Richardson P, Hsu YC et al. Comparative performance of risk prediction models for hepatitis B-related hepatocellular carcinoma in the United States. J Hepatol. 2021.

  30. Güzelbulut F, Gökçen P, Can G, Adalı G, Değirmenci Saltürk AG, Bahadır Ö, et al. Validation of the HCC-RESCUE score to predict hepatocellular carcinoma risk in caucasian chronic hepatitis B patients under entecavir or tenofovir therapy. J Viral Hepat. 2021;28(5):826–36.

    Article  PubMed  Google Scholar 

  31. Fattovich G, Bortolotti F, Donato F. Natural history of chronic hepatitis B: special emphasis on disease progression and prognostic factors. J Hepatol. 2008;48(2):335–52.

    Article  PubMed  Google Scholar 

  32. Wu S, Zeng N, Sun F, Zhou J, Wu X, Sun Y, et al. Hepatocellular Carcinoma Prediction Models in Chronic Hepatitis B: a systematic review of 14 models and external validation. Clin Gastroenterol Hepatol. 2021;19(12):2499–513.

    Article  CAS  PubMed  Google Scholar 

  33. Yang L, Wang Q, Cui T, Huang J, Jin H. Reporting and Performance of Hepatocellular Carcinoma Risk Prediction Models: Based on TRIPOD Statement and Meta-Analysis. Can J Gastroenterol. 2021, 2021:9996358.

  34. Chen CH, Lee CM, Lai HC, Hu TH, Su WP, Lu SN, et al. Prediction model of hepatocellular carcinoma risk in asian patients with chronic hepatitis B treated with entecavir. Oncotarget. 2017;8(54):92431–41.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Kim MN, Hwang SG, Rim KS, Kim BK, Park JY, Kim DY, et al. Validation of PAGE-B model in asian chronic hepatitis B patients receiving entecavir or tenofovir. Liver Int. 2017;37(12):1788–95.

    Article  CAS  PubMed  Google Scholar 

  36. Kirino S, Tamaki N, Kaneko S, Kurosaki M, Inada K, Yamashita K, et al. Validation of hepatocellular carcinoma risk scores in japanese chronic hepatitis B cohort receiving nucleot(s)ide analog. J Gastroenterol Hepatol. 2020;35(9):1595–601.

    Article  CAS  PubMed  Google Scholar 

  37. Ahn SB, Choi J, Jun DW, Oh H, Yoon EL, Kim HS, et al. Twelve-month post-treatment parameters are superior in predicting hepatocellular carcinoma in patients with chronic hepatitis B. Liver Int. 2021;41(7):1652–61.

    Article  CAS  PubMed  Google Scholar 

  38. Chang JW, Lee JS, Lee HW, Kim BK, Park JY, Kim DY, et al. Validation of risk prediction scores for hepatocellular carcinoma in patients with chronic hepatitis B treated with entecavir or tenofovir. J Viral Hepat. 2021;28(1):95–104.

    Article  CAS  PubMed  Google Scholar 

  39. Chon HY, Lee HA, Suh SJ, Lee JI, Kim BS, Kim IH, et al. Addition of liver stiffness enhances the predictive accuracy of the PAGE-B model for hepatitis B-related hepatocellular carcinoma. Aliment Pharmacol Ther. 2021;53(8):919–27.

    Article  CAS  PubMed  Google Scholar 

  40. Chon HY, Lee JS, Lee HW, Chun HS, Kim BK, Tak WY, et al. Predictive performance of CAGE-B and SAGE-B models in asian treatment-naive patients who started Entecavir for Chronic Hepatitis B. Clin Gastroenterol Hepatol. 2022;20(4):e794–e807.

    Article  CAS  PubMed  Google Scholar 

  41. Lee JS, Lee HW, Lim TS, Shin HJ, Lee HW, Kim SU, et al. Novel liver stiffness-based Nomogram for Predicting Hepatocellular Carcinoma Risk in patients with chronic Hepatitis B Virus infection initiating antiviral therapy. Cancers. 2021;13(23):5892.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Lim J, Chon YE, Kim MN, Lee JH, Hwang SG, Lee HC, et al. Cirrhosis, Age, and liver stiffness-based models predict Hepatocellular Carcinoma in Asian Patients with Chronic Hepatitis B. Cancers. 2021;13(22):5609.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Papatheodoridis GV, Dalekos GN, Idilman R, Sypsa V, Van Boemmel F, Buti M, et al. Predictive performance of newer asian hepatocellular carcinoma risk scores in treated Caucasians with chronic hepatitis B. JHEP Rep. 2021;3(3):100290.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

None.

Funding

This work was supported by the National Major Science and Technology Project of China [Grant number 2017ZX10105001] and National Human Genetic Resources Sharing Service Platform [Grant number 2005DKA21300].

Author information

Author notes

  1. Xiaolan Xu and Lushun Jiang contributed equally to this work.

Authors and Affiliations

  1. State Key Laboratory for Diagnosis and Treatment of Infectious Diseases, National Clinical Research Center for Infectious Diseases, Collaborative Innovation Center for Diagnosis and Treatment of Infectious Diseases, The First Affiliated Hospital, National Medical Center for Infectious Diseases, Zhejiang University School of Medicine, 79 Qingchun Road, Shangcheng District, Hangzhou, 310000, China

    Xiaolan Xu, Lushun Jiang, Yifan Zeng, Liya Pan, Zhuoqi Lou & Bing Ruan

  2. Center for General Practice Medicine, Department of Infectious Diseases, Zhejiang Provincial People’s Hospital (Affiliated People’s Hospital), Hangzhou Medical College, Hangzhou, 310000, China

    Xiaolan Xu

Authors
  1. Xiaolan Xu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Lushun Jiang
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Yifan Zeng
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Liya Pan
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Zhuoqi Lou
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Bing Ruan
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

Conceptualization, BR and XX; methodology, XX; Investigation, XX and LJ; data curation, XX, LJ, and YZ; formal analysis, XX and YZ; writing-original draft, XX; writing—review and editing, ZL and LP; supervision, BR; funding acquisition, BR. All authors have read and agreed to the published version of the manuscript. All authors have read and agreed to the published version of the manuscript.

Corresponding author

Correspondence to Bing Ruan.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Xu, X., Jiang, L., Zeng, Y. et al. HCC prediction models in chronic hepatitis B patients receiving entecavir or tenofovir: a systematic review and meta-analysis. Virol J 20, 180 (2023). https://doi.org/10.1186/s12985-023-02145-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12985-023-02145-5

Keywords

Virology Journal

ISSN: 1743-422X