Predictive Risk Factors and Scoring Systems Associated with the Development of Hepatocellular Carcinoma in Chronic Hepatitis B.

Chronic hepatitis B (CHB) infection constitutes a leading cause of hepatocellular carcinoma (HCC) development. The identification of HCC risk factors and the development of prognostic risk scores are essential for early diagnosis and prognosis. The aim of this observational, retrospective study was to evaluate baseline risk factors associated with HCC in CHB. Six hundred thirty-two consecutive adults with CHB (n = 632) [median age: 46 (IQR: 24)], attending the outpatients' Hepatology clinics between 01/1993-09/2020 were evaluated. Core promoter mutations and cirrhosis-HCC (GAG-HCC), Chinese University-HCC (CU-HCC), risk estimation for hepatocellular carcinoma in chronic hepatitis B (REACH-B), Fibrosis-4 (FIB-4), and Platelet Age Gender-HBV (PAGE-B) prognostic scores were calculated, and receiver operating curves were used to assess their prognostic performance. HCC was developed in 34 (5.38%) patients. In the multivariable Cox regression analysis, advanced age (HR: 1.086, 95% CI: 1.037-1.137), male sex (HR: 7.696, 95% CI: 1.971-30.046), alcohol abuse (HR: 2.903, 95% CI: 1.222-6.987) and cirrhosis (HR: 21.239, 95% CI: 6.001-75.167) at baseline were independently associated with the development of HCC. GAG-HCC and PAGE-B showed the highest performance with c-statistics of 0.895 (95% CI: 0.829-0.961) and 0.857 (95% CI: 0.791-0.924), respectively. In the subgroup of patients with cirrhosis, the performance of all scores declined. When treated and untreated patients were studied separately, the discriminatory ability of the scores differed. In conclusion, HCC development was independently associated with advanced age, male sex, alcohol abuse, and baseline cirrhosis among a diverse population with CHB. GAG-HCC and PAGE-B showed high discriminatory performance to assess the risk of HCC development in these patients, but these performances declined in the subgroup of patients with cirrhosis. Further research to develop scores more specific to certain CHB subgroups is needed.

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.

Centers for Medicare & Medicaid Services Hierarchical Condition Category score as a predictor of readmission and reoperation following elective inpatient spine surgery.

OBJECTIVE: A universal, objective predictor of postoperative resource utilization following inpatient spine surgery has not been clearly established. The Centers for Medicare & Medicaid Services (CMS) Hierarchical Condition Category (HCC) risk adjustment model, based on a formula using patient demographics and coded diagnoses, is currently used to prospectively estimate financial risk in Medicare Advantage patients; however, the value of this score as a clinical tool is currently unknown. The authors present an analysis evaluating the utility of the CMS HCC score as a universal predictive tool for patients undergoing inpatient spine surgery. METHODS: A total of 1966 consecutive patients (551 with lumbar laminectomy [LL] alone, 592 with lumbar laminectomy and fusion [LF], and 823 with anterior cervical discectomy and fusion [ACDF]) undergoing inpatient spine surgery at a single institution from January 2014 to May 2018 were included in this retrospective outcomes study. Perioperative outcome measures included procedure time, 30-day readmission, reoperation, hospital length of stay (LOS), opioid utilization measured by morphine milligram equivalents (MMEs), and cost of inpatient hospitalization (in US dollars). Published CMS algorithms were incorporated into the electronic health records and used to calculate HCC scores for all patients. Patients were stratified into HCC score quartiles. Linear regression was performed on LOS, procedure time, inpatient opioid consumption, discharge opioid prescriptions, and cost to identify predictors of HCC quartiles when controlling for procedure type. One-way ANOVA and Pearson's chi-square analysis were used to compare perioperative outcomes stratified by HCC score. RESULTS: Across all procedures, the HCC score demonstrated significant association with 30-day readmission (OR 1.45, 95% CI 1.11-1.91, p = 0.007). The average BMI, median American Society of Anesthesiologists score, and 30-day readmission rate were similar across procedures (LL: 30.6 kg/m2, 2, 3.6%; LF: 30.6 kg/m2, 2, 4.6%; ACDF: 30.2 kg/m2, 2, 3.9%; p = 0.265, 0.061, and 0.713, respectively). LOS (p < 0.0001), duration of procedure (p < 0.0001), discharge MME (p = 0.031), total cost (p < 0.001), daily MME (p < 0.001), reoperation (p < 0.001), and 30-day readmission rate (p < 0.001) were significantly different between HCC quartiles. CONCLUSIONS: The HCC score may hold value as an objective, automated predictor of postoperative resource utilization and outcomes, including readmission and reoperation. This may have value as a universal, reproducible tool to target clinical interventions for higher-risk patients.

Validity of eleven prognostic scores with respect to intra- and extrahepatic recurrence of hepatocellular carcinoma after liver transplantation.

INTRODUCTION: Tumor recurrence is the most frequent cause of death after liver transplantation for hepatocellular carcinoma. We selected ten other prognostic classifications to evaluate their potential to predict the risk of recurrence after LT for HCC as compared to the Milan classification. All of the other scores have not been compared with one another in a single cohort. METHODS: Data of 147 consecutive patients transplanted at our department between 1996 and 2014 were analyzed and staged for morphological and functional scores of underlying liver disease. For long-term follow-up, we analyzed intrahepatic (within the liver ± distant metastases) and extrahepatic (distant metastases only) recurrence separately. RESULTS AND CONCLUSIONS: The median survival time for all patients was 106 months. The 5- and 10-year observed survival rates were 61 and 43%, respectively. The observed cumulative 5- and 10-year recurrence rates were 37 and 39%, respectively, 10-year intrahepatic and extrahepatic recurrence rates were 12 and 27%, respectively. Median survival time after diagnosis of first recurrence was 7.5 (0-120) months; 2 and 18 months for all, intra- and extrahepatic recurrence, respectively. UCSF-, up to seven-, Shanghai Fudan- or Duvoux classifications can identify patients with a cumulative 10-year recurrence rate below 20%. The pre-therapeutic AFP level should be considered in addition to the geometry of the intrahepatic lesions.