External validation of hepatocellular carcinoma risk scores in patients with chronic hepatitis B virus infection in China.

Several scores have been proposed in untreated or treated patients with chronic hepatitis B (CHB) to predict risks of hepatocellular carcinoma (HCC) occurrence. However, it is still unclear which score suits all chronic hepatitis B virus (HBV)-infected patients well, regardless of whether they are chronic carriers or CHB patients. In this study, we validated and compared the predictability of CU-HCC, REACH-B, PAGE-B and mPAGE-B in patients with chronic HBV infection in China. 1,786 patients with no history of HCC were recruited, with 978 carriers and 808 CHB patients on antiviral therapy. Patients were classified into low- and high-risk groups according to the predefined cut-off values of 5, 8, 10 and 9 for CU-HCC, REACH-B, PAGE-B and mPAGE-B. The median follow-up period was 43.7months, during which 18 (1.0%) patients developed HCC. The areas under the receiver operating characteristic curves (AUROCs) of CU-HCC, REACH-B, PAGE-B and mPAGE-B scores to predict HCC risk at 36 months were 0.815, 0.703, 0.794 and 0.825, respectively (all p < 0.05). No significant difference among AUROCs of these scores was observed except those of mPAGE-B and REACH-B at 36 months. The cumulative incidence of HCC in low- and high- risk groups based on CU-HCC, REACH-B, PAGE-B and mPAGE-B were 0.4% vs. 3.2%, 0.7% vs. 1.5%, 0.2% vs. 1.3%, and 0.2% vs. 1.7% at 36 months, respectively (all p < 0.05, except PAGE-B, log-rant test). Both CU-HCC and mPAGE-B scores accurately predict HCC risk in Chinese chronic HBV-infected patients. Patients with CU-HCC <5 or mPAGE-B <9 could be exempt from HCC surveillance within 36 months.

Incidence and predictors of hepatocellular carcinoma in Caucasian chronic hepatitis B patients receiving entecavir or tenofovir.

BACKGROUND & AIMS: The risk of hepatocellular carcinoma (HCC) in Caucasian patients with chronic hepatitis B (CHB), treated with entecavir (ETV) or tenofovir (TDF), is unclear. We evaluated the incidence and predictors of HCC and the accuracy of existing HCC risk scores in Caucasian CHB patients receiving ETV/TDF. METHODS: This large, multicentre, retrospective cohort study included 1666 adult Caucasian CHB patients under ETV/TDF for 39 months. CHB without cirrhosis, compensated and decompensated cirrhosis were present in 67%, 39%, and 3% of patients, respectively. The predictability of baseline parameters and three risk scores (GAG-HCC, CU-HCC, and REACH-B), developed in Asian patients, was assessed. RESULTS: The cumulative probability of HCC was 1.3%, 3.4%, and 8.7% at year-1, year-3, and year-5 after ETV/TDF onset. Older age and lower platelets were strong independent HCC predictors in the total population and in the subgroups of cirrhotic and non-cirrhotic patients, while liver disease severity was an independent HCC predictor in the total population and in the cirrhotics. GAG-HCC, CU-HCC, and REACH-B risk scores were associated with HCC development only in the univariable but not in the multivariable analyses and offered poor to modest predictability. CONCLUSIONS: HCC can still develop in Caucasian CHB patients treated with ETV/TDF. Besides the well-known predictors of HCC, such as older age, male gender and more advanced liver disease, lower platelets represent an independent factor of higher HCC risk. The applicability and predictability of HCC risk scores developed in Asian patients are poor or modest in Caucasian CHB patients, for whom different risk scores are required.

Deep learning model for prediction of hepatocellular carcinoma in patients with HBV-related cirrhosis on antiviral therapy.

BACKGROUND & AIMS: Personalised risk prediction of the development of hepatocellular carcinoma (HCC) among patients with liver cirrhosis on potent antiviral therapy is important for targeted screening and individualised intervention. This study aimed to develop and validate a new model for risk prediction of HCC development based on deep learning, and to compare it with previously reported risk models. METHODS: A novel deep-learning-based model was developed from a cohort of 424 patients with HBV-related cirrhosis on entecavir therapy with 2 residual blocks, including 7 layers of a neural network, and it was validated using an independent external cohort (n = 316). The deep-learning-based model was compared to 6 previously reported models (platelet, age, and gender-hepatitis B score [PAGE-B], Chinese University HCC score [CU-HCC], HCC-Risk Estimating Score in CHB patients Under Entecavir [HCC-RESCUE], age, diabetes, race, etiology of cirrhosis, sex, and severity HCC score [ADRESS-HCC], modified PAGE-B score [mPAGE], and Toronto HCC risk index [THRI]) using Harrell's concordance (c)-index. RESULTS: During a median 5.2 yr of follow-up (inter-quartile range 2.8-6.9 yr), 86 patients (20.3%) developed HCC. The deep-learning-based model had a Harrell's c-index of 0.719 in the derivation cohort and 0.782 in the validation cohort. Goodness of fit was confirmed by the Hosmer-Lemeshow test (p >0.05). Moreover, this model in the validation cohort had the highest c-index among the 6 previously reported models: PAGE-B (0.570), CU-HCC (0.548), HCC-RESCUE (0.577), ADRESS-HCC (0.551), mPAGE (0.598), and THRI (0.587) (all p <0.001). The misclassification rate of this model was 23.7% (model accuracy: 76.3%) in the validation group. CONCLUSIONS: The deep-learning-based model had better performance than the previous models for predicting the HCC risk in patients with HBV-related cirrhosis on potent antivirals. LAY SUMMARY: For early detection of hepatocellular carcinoma, it is important to maintain regular surveillance. However, there is currently no standard prediction model for risk stratification that can be used to establish a personalised surveillance strategy. We develop and validate a deep-learning-based model that showed better performance than previous models.

INASL Position Statements on Prevention, Diagnosis and Management of Hepatitis B Virus Infection in India: The Andaman Statements.

Hepatitis B Virus (HBV) infection is one of the major causes of morbidity, mortality and healthcare expenditure in India. There are no Indian consensus guidelines on prevention, diagnosis and management of HBV infection. The Indian National Association for Study of the Liver (INASL) set up a taskforce on HBV in 2016, with a mandate to develop consensus guidelines for diagnosis and management of HBV infection, relevant to disease patterns and clinical practices in India. The taskforce first identified contentious issues on various aspects of HBV management, which were allotted to individual members of the taskforce who reviewed them in detail. A 2-day round table discussion was held on 11th and 12th February 2017 at Port Blair, Andaman & Nicobar Islands, to discuss, debate, and finalize the consensus statements. The members of the taskforce reviewed and discussed the existing literature threadbare at this meeting and formulated the 'INASL position statements' on each of the issues. The evidence and recommendations in these guidelines have been graded according to the Grading of Recommendations Assessment Development and Evaluation (GRADE) system with minor modifications. The strength of recommendations (strong: 1, weak: 2) thus reflects the quality (grade) of underlying evidence (A, B, C, D). We present here the INASL position statements on prevention, diagnosis and management of HBV in India.