Machine learning-based clinical decision support system for treatment recommendation and overall survival prediction of hepatocellular carcinoma: a multi-center study.

The treatment decisions for patients with hepatocellular carcinoma are determined by a wide range of factors, and there is a significant difference between the recommendations of widely used staging systems and the actual initial treatment choices. Herein, we propose a machine learning-based clinical decision support system suitable for use in multi-center settings. We collected data from nine institutions in South Korea for training and validation datasets. The internal and external datasets included 935 and 1750 patients, respectively. We developed a model with 20 clinical variables consisting of two stages: the first stage which recommends initial treatment using an ensemble voting machine, and the second stage, which predicts post-treatment survival using a random survival forest algorithm. We derived the first and second treatment options from the results with the highest and the second-highest probabilities given by the ensemble model and predicted their post-treatment survival. When only the first treatment option was accepted, the mean accuracy of treatment recommendation in the internal and external datasets was 67.27% and 55.34%, respectively. The accuracy increased to 87.27% and 86.06%, respectively, when the second option was included as the correct answer. Harrell's C index, integrated time-dependent AUC curve, and integrated Brier score of survival prediction in the internal and external datasets were 0.8381 and 0.7767, 91.89 and 86.48, 0.12, and 0.14, respectively. The proposed system can assist physicians by providing data-driven predictions for reference from other larger institutions or other physicians within the same institution when making treatment decisions.

Hepatic hemangioma: proportion and predictor of surgical treatment with emphasis on its growth rate.

BACKGROUND/AIMS: There are limited studies on the management of hepatic hemangiomas (HHs). We investigated the proportion and predictors of surgical resection and analyzed HH growth rates in addition to associated factors. METHODS: A retrospective case-control study of patients treated in 2 centers was conducted. Thirty-six patients who underwent surgical resection were assigned to the case group. Patients who did not undergo surgical treatment were randomly sigselected at a 1:10 ratio and assigned to the control group (n = 360). Baseline characteristics, clinical course and surgical outcomes were analyzed. RESULTS: The proportion of surgically treated HH patients was 0.3% (36 per 11,049). The longest diameter at diagnosis (mean ± standard deviation) was 7.7 ± 5.2 cm in the case group and 2.4 ± 1.8 cm in the control group (p < 0.001). In the multivariate analysis, the presence of more than 2 HHs (odds ratio [OR] 7.64, 95% confidence interval [CI] 1.40-41.72) and a growth rate of more than 4.8%/year (OR 30.73, 95% CI 4.86-194.51) were independently associated with surgical treatment. Symptom development during follow-up was related to HH size > 10 cm (OR 10.50, 95% CI 1.06-103.77, p = 0.04). The subgroup analysis showed substantial growth in 41.3% with an overall mean annual growth rate of 0.14 cm. CONCLUSION: Approximately one in 300 patients with an HH underwent surgical treatment. Multiple HHs and a growth rate of more than 4.8%/year were indications for surgical treatment. Nearly half of the HHs showed growing pattern in our study.

Risk factors associated with late hepatocellular carcinoma detection in patients undergoing regular surveillance.

Hepatocellular carcinoma (HCC) has a very poor prognosis with a 5-year survival rate of < 20%; hence, early diagnosis is crucial. Despite regular checkups for high-risk groups of HCC, there are a few cases in which it is discovered as a late-stage HCC. Therefore, this study aimed to investigate the characteristics of patients with delayed HCC detection during regular surveillance. Between January 2010 and December 2020, we analyzed patients with newly diagnosed HCCs who underwent HCC surveillance by ultrasound or computed tomography scan at least twice and were followed up for more than 1 year for hepatitis B, hepatitis C, and chronic liver disease. The mean age of 223 HCC patients was 70 years, of which 152 were male, accounting for 68.1%. Among them, 196 patients (87%) were diagnosed with Barcelona clinic liver cancer stage 0 or A, while 27 (13%) were diagnosed with Barcelona clinic liver cancer stages B and C. When classified according to the TNM criteria, 154 patients (69%) were in stage I, and 69 (31%) were in stage II or higher. Multivariate analysis was performed to identify the risk factors for patients diagnosed with late-stage HCC. The Child-Turcotte-Pugh (CTP) score was identified as a highly significant factor (P = .002, HR 1.547, 95% CI 1.177-2.032), whereas the presence of cirrhosis, body mass index, and sex had no significant effect. We found that in patients with chronic liver disease who were screened regularly, those with higher CTP scores were more likely to be diagnosed with HCC in the late-stages. Therefore, although the presence of cirrhosis is also important for HCC surveillance, careful attention is needed in patients with high CTP scores.

Metformin administration is associated with enhanced response to transarterial chemoembolization for hepatocellular carcinoma in type 2 diabetes patients.

Transarterial chemoembolization (TACE) is often used as a locoregional therapy for early hepatocellular carcinoma (HCC) when local ablation or resection are not feasible, but incomplete response and recurrence are commonly observed. In this study, we sought to determine the association between metformin administration and TACE outcomes for single nodular HCC in patients with type 2 diabetes mellitus (T2DM). The retrospective cohort analysis included 164 T2DM patients with single nodular HCC who underwent TACE as an initial treatment, and 91 were exposed to metformin before and after TACE. Propensity score (PS) matching was used to balance covariates. Logistic regression analysis was used to determine the predictors of tumor response after TACE, and Cox regression analysis assessed independent predictors of local tumor recurrence (LTR) in patients with complete response after TACE. Metformin use was associated with significantly higher objective response rate (ORR) in the overall and PS-matched cohort (79.1% vs. 60.3 and 78.7% vs. 57.5%; p = 0.008 and p = 0.029, respectively). Logistic regression analysis showed that metformin use was an independent predictor of ORR in all and PS-matched patients (odds ratio = 2.65 and 3.06; p = 0.016 and 0.034, respectively). Cox regression analysis showed metformin administration was an independent predictor for lower LTR in all and PS-matched patients (hazard ratio = 0.28 and 0.27; p = 0.001 and 0.007, respectively). Metformin administration is associated with better initial response and lower local recurrence after TACE for single nodular HCC in T2DM.

Incidence and Pattern of Aminotransferase Elevation During Anti-Hypertensive Therapy With Angiotensin-II Receptor Blockers.

BACKGROUND: Angiotensin type II receptor blockers (ARBs) are the most widely used anti-hypertensive drugs. This study aimed to elucidate the likelihood and pattern of ARB-induced liver injury in a hospital-based cohort. METHODS: Data of patients receiving fimasartan (n = 5,543), candesartan (n = 6,406), valsartan (n = 6,040), and losartan (n = 9,126) were retrieved from the clinical data warehouse of two tertiary hospitals. Patients with alanine aminotransferase (ALT) levels > 5 times the upper normal limit were assessed according to the Roussel Uclaf Causality Assessment Method (RUCAM). RESULTS: A total of 27,115 patients were enrolled, including 14,630 (54.0%) men, with a mean age of 64.6 years (standard deviation, 13.6). During 31,717 person-years of ARB therapy, serum ALT levels > 120 IU/L were found in 558 (2.1%) person-years, and levels > 200 IU/L were found in 155 (0.6%) person-years. The incidence of ALT elevation > 120 IU/L per 106 cumulative defined daily doses was 6.6, 3.6, 3.9, and 4.0 in the fimasartan, candesartan, valsartan, and losartan groups, respectively (P = 0.002). An ALT level > 200 IU/L with RUCAM score ≥ 6 was found in 20 patients, suggesting probable drug-induced liver injury for 11 (0.2%) patients receiving fimasartan, five (0.1%) receiving candesartan, four (0.1%) receiving valsartan, and none receiving losartan (P < 0.001). CONCLUSION: Approximately 2% of patients receiving ARB therapy had significant ALT elevation (4.24/106 cumulative defined daily doses [cDDDs]), which was associated with probable ARB-related liver injury in 0.07% of patients (0.15/106 cDDDs). Elevation of ALT was more commonly associated with fimasartan than the other ARBs. Clinicians should be aware of the possibility of ARB-related ALT elevation in patients with unexplained chronic abnormal ALT.

Elevated alpha-fetoprotein in asymptomatic adults: Clinical features, outcome, and association with body composition.

BACKGROUND AND AIM: Apparently healthy individuals with elevated serum alpha-fetoprotein (AFP) levels (>7 ng/mL) for unknown causes visit clinics. We investigated their clinical characteristics, outcomes, and relationship with body fat deposition and muscle mass. METHODS: The case group included asymptomatic 137 individuals with "elevated AFP level" (R772) diagnostic code from 2009 to 2018 in a tertiary hospital. The control group enrolled 274 age- and sex-matched patients with <5 cm hepatic hemangiomas. Hepatic, visceral, and psoas muscle adiposity and psoas muscle index (PMI) were measured in the subgroups of 45 cases and 90 controls with pre-contrast computed tomography (CT) images. RESULTS: The case group (mean age 47.5 years, male 35.8%) showed higher AFP levels (10.3 vs 2.5 ng/mL, p<0.001) and total bilirubin (0.8 vs 0.7 mg/dL, p<0.001), but a lower body mass index (22.2 vs 23.3 kg/m2, p = 0.011) and alanine aminotransferase levels (17.0 vs 19.0 IU/L, p = 0.047) than the controls. During 13 months of median follow-up, there was no cancer or liver disease development. The AFP levels were stable. In the subgroups with CT images, cases showed a lower proportion of hepatic steatosis (4.4% vs 18.9%, p = 0.023), higher psoas muscle attenuation (48.2 vs 43.8 Hounsfield units, p<0.001) and higher PMI (5.7 vs 4.2 cm2/m2, p<0.001) than the controls. CONCLUSION: Elevated AFP levels in asymptomatic individuals may play a role in expressing a protective phenotype against hepatic steatosis, myosteatosis, and sarcopenia. AFP levels in patients with elevated AFP were stable during follow-up without liver injury or cancer development. Interaction between AFP expression and steatosis warrants further study.

Hepatocellular carcinoma with Budd-Chiari syndrome due to membranous obstruction of the inferior vena cava with long-term follow-up: a case report.

Membranous obstruction of the inferior vena cava (MOVC) is a rare subset of Budd-Chiari syndrome (BCS) with a subacute onset that is often complicated by cirrhosis and hepatocellular carcinoma (HCC). Here we report a case of recurrent HCC in a patient with cirrhosis and BCS that was treated with several episodes of transarterial chemoembolization followed by surgical tumorectomy, whereas the MOVC was successfully treated with balloon angioplasty followed by endovascular stenting. The patient was followed up for 9.9 years without anticoagulation and experienced no stent thrombosis. After the tumorectomy, the patient was HCC-free for 4.4 years of follow-up.