Switching from tenofovir disoproxil fumarate to tenofovir alafenamide in virologically suppressed patient with chronic hepatitis B.

BACKGROUND AND AIM: Our study evaluated the outcomes of switching from tenofovir disoproxil fumarate (TDF) to tenofovir alafenamide (TAF) in patients with chronic hepatitis B (CHB). We assessed viral and biochemical responses as well as changes in the estimated glomerular filtration rate (eGFR) and bone mineral density (BMD). METHODS: This retrospective multicenter study included CHB patients who achieved virologic response (VR) (HBV DNA < 20 IU/mL) while on TDF and were subsequently switched to TAF between April 2018 and October 2021. RESULTS: This study included 309 patients with a median age of 59 years, and 42.1% were male. The mean duration of TDF and TAF administration were 54.0 and 37.5 months, respectively. All patients maintained VR after switching to TAF. Alanine aminotransferase (ALT) normalization rate significantly increased 6 months after switching (74.8%-83.5%; P = 0.008). Adjusted eGFR significantly improved at 6 months (+5.55 ± 10.52 mL/min/1.73 m2; P < 0.001) and 12 months (+6.02 ± 10.70 mL/min/1.73 m2; P < 0.001) after switching. In the subgroup of patients with renal impairment (eGFR < 60 mL/min/1.73 m2), significant improvement in renal function was observed at 6 months (+0.6 ± 10.5 mL/min/1.73 m2; P < 0.001) and 12 months (+1.0 ± 10.7 mL/min/1.73 m2; P < 0.001) after switching to TAF. In patients with osteoporosis (n = 182), switching to TAF resulted in significant improvement in spine and hip BMD at 12 months, with increases of 9.7% (95% CI: 7.0-12.5) and 9.4% (95% CI: 7.0-11.8), respectively. CONCLUSION: In this real-world study, switching to TAF was effective and safe in patients, with notable improvements in ALT levels, renal function, and BMD.

A Machine Learning Algorithm Facilitates Prognosis Prediction and Treatment Selection for Barcelona Clinic Liver Cancer Stage C Hepatocellular Carcinoma.

BACKGROUND: Given its heterogeneity and diverse clinical outcomes, precise subclassification of BCLC-C hepatocellular carcinoma (HCC) is required for appropriately determining patient prognosis and selecting treatment. METHODS: We recruited 2,626 patients with BCLC-C stage HCC from multiple centers, comprising training/test (n=1,693) and validation cohorts (n=933). The XGBoost was chosen for maximum performance among the machine learning (ML) models. Patients were categorized into low-/intermediate-/high-/very high-risk subgroups which were based on the estimated prognosis, and this subclassification was named the CLAssification via Machine learning of BCLC-C (CLAM-C). RESULTS: The areas under the receiver operating characteristic curve of the CLAM-C for predicting the 6-/12-/24-month survival of patients with BCLC-C were 0.800/0.831/0.715, respectively-significantly higher than those of the conventional models, which was consistent in the validation cohort. The four subgroups had significantly different median overall survivals, and this difference was maintained among various patient subgroups and treatment modalities. Immune-checkpoint inhibitors and transarterial therapies were associated with significantly better survival than tyrosine kinase inhibitors (TKIs) in the low- and intermediate-risk subgroups. In cases with first-line systemic therapy, the CLAM-C identified atezolizumab-bevacizumab as the best therapy particularly in the high-risk group. In cases with later-line systemic therapy, nivolumab had better survival than TKIs in the low-to-intermediate-risk subgroup, whereas TKIs had better survival in the high-to-very high-risk subgroup. CONCLUSIONS: ML modeling effectively subclassified patients with BCLC-C HCC, potentially aiding treatment allocation. Our study underscores the potential utilization of ML modeling in terms of prognostication and treatment allocation in patients with BCLC-C HCC.

Differential liver function at cessation of atezolizumab-bevacizumab versus lenvatinib in HCC: a multicenter, propensity-score matched comparative study.

Background: Atezolizumab+bevacizumab (AB) and lenvatinib have been proposed as first-line treatment options for patients with advanced hepatocellular carcinoma (HCC), but comparative efficacy and associated factors are controversial. Materials and methods: This real-world multicenter study analysed patients with HCC who received AB (n=169) or lenvatinib (n=177). Results: First, 1:1 propensity score matching (PSM) was performed, resulting in 141 patients in both the AB and lenvatinib groups. After PSM, overall survival (OS) was better in the AB group than in the lenvatinib group [hazard ratio (HR)=0.642, P=0.009], but progression-free survival (PFS) did not vary between the two groups (HR=0.817, P=0.132). Objective response rate (ORR) was also similar between AB and lenvatinib (34.8% vs. 30.8%, P=0.581). In a subgroup of patients with objective responses (OR, n=78), OS (HR=0.364, P=0.012) and PFS (HR=0.536, P=0.019) were better in the AB group (n=41) than in the lenvatinib group (n=37). Time-to-progression from time of OR was also better in the AB group (HR=0.465, P=0.012). Importantly, residual liver function was a significant factor related to OS in both treatments. Child-Pugh score following cessation of the respective treatments was better in the AB group (n=105) than in the lenvatinib group (n=126) (median 6 versus 7, P=0.008), and proportion of salvage treatment was also higher in the AB group (52.4% versus 38.9%, P=0.047). When we adjusted for residual liver function or salvage treatment, there was no difference in OS between the two treatments. Conclusion: Our study suggests that residual liver function and subsequent salvage treatments are major determinants of clinical outcomes in patients treated with AB and lenvatinib; these factors should be considered in future comparative studies.

Long-Term HBsAg Titer Kinetics with Entecavir/Tenofovir: Implications for Predicting Functional Cure and Low Levels.

The long-term kinetics of quantitative HBsAg levels in HBV-infected patients treated with entecavir or tenofovir, as well as the role of quantitative HBsAg in predicting functional cure (HBsAg loss) and low HBsAg levels (<2 log IU/mL) remain unclear. Of some 1661 consecutively enrolled patients newly treated with entecavir or tenofovir, we analyzed 852 patients who underwent serial HBsAg level checks every 6-12 months. The primary outcomes included long-term kinetics in HBsAg levels and the rate of functional cure and achieving low HBsAg levels. Over a mean 6.3-year follow-up, the functional cure rate was 2.28% (n = 19), and 12.9% (n = 108) achieved low HBsAg levels. A significant HBsAg level reduction was seen in the first treatment year (p < 0.05), with another stepwise decrease between year 6-7. These trends were pronounced in patients with chronic hepatitis and HBeAg-positivity compared to those with cirrhosis and HBeAg-negativity, respectively. Baseline HBsAg of ≤3 log IU/mL and the first-year HBsAg reduction were key predictors for both functional cure and low HBsAg levels (p < 0.05). In conclusion, our findings elucidate the stepwise reduction in quantitative HBsAg dynamics during high-potency NA therapy (entecavir or tenofovir) along with variations based on different conditions. We also underscore the significance of quantitative HBsAg titer in predicting functional cure and low-HBsAg levels.

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.

A Real-World Comparative Analysis of Atezolizumab Plus Bevacizumab and Transarterial Chemoembolization Plus Radiotherapy in Hepatocellular Carcinoma Patients with Portal Vein Tumor Thrombosis.

This study aimed to compare the treatment outcomes of atezolizumab-plus-bevacizumab (Ate/Bev) therapy with those of transarterial chemoembolization plus radiotherapy (TACE + RT) in hepatocellular carcinoma (HCC) patients with portal vein tumor thrombosis (PVTT) and without metastasis. Between June 2016 and October 2022, we consecutively enrolled 855 HCC patients with PVTT. After excluding 758 patients, 97 patients (n = 37 in the Ate/Bev group; n = 60 in the TACE + RT group) were analyzed. The two groups showed no significant differences in baseline characteristics and had similar objective response and disease control rates. However, the Ate/Bev group showed a significantly higher one-year survival rate (p = 0.041) compared to the TACE + RT group, which was constantly displayed in patients with extensive HCC burden. Meanwhile, the clinical outcomes were comparable between the two groups in patients with unilobar intrahepatic HCC. In Cox-regression analysis, Ate/Bev treatment emerged as a significant factor for better one-year survival (p = 0.049). Finally, in propensity-score matching, the Ate/Bev group demonstrated a better one-year survival (p = 0.02) and PFS (p = 0.01) than the TACE + RT group. In conclusion, Ate/Bev treatment demonstrated superior clinical outcomes compared to TACE + RT treatment in HCC patients with PVTT. Meanwhile, in patients with unilobar intrahepatic HCC, TACE + RT could also be considered as an alternative treatment option alongside Ate/Bev therapy.

An Immunological Perspective on the Mechanism of Drug Induced Liver Injury: Focused on Drugs for Treatment of Hepatocellular Carcinoma and Liver Transplantation.

The liver is frequently exposed to potentially toxic materials, and it is the primary site of clearance of foreign agents, along with many innate and adaptive immune cells. Subsequently, drug induced liver injury (DILI), which is caused by medications, herbs, and dietary supplements, often occurs and has become an important issue in liver diseases. Reactive metabolites or drug-protein complexes induce DILI via the activation of various innate and adaptive immune cells. There has been a revolutionary development of treatment drugs for hepatocellular carcinoma (HCC) and liver transplantation (LT), including immune checkpoint inhibitors (ICIs), that show high efficacy in patients with advanced HCC. Along with the high efficacy of novel drugs, DILI has become a pivotal issue in the use of new drugs, including ICIs. This review demonstrates the immunological mechanism of DILI, including the innate and adaptive immune systems. Moreover, it aims to provide drug treatment targets, describe the mechanisms of DILI, and detail the management of DILI caused by drugs for HCC and LT.

Chemoembolization combined radiofrequency ablation vs. chemoembolization alone for treatment of beyond the Milan criteria viable hepatocellular carcinoma (CERFA): study protocol for a randomized controlled trial.

BACKGROUND: Many previous studies evaluated a combination of transcatheter arterial chemoembolization (TACE) and radiofrequency ablation (RFA) for treating early hepatocellular carcinoma (HCC); however, studies evaluating combination therapy for beyond-the-Milan criteria HCC are scarce. METHODS: A total of 120 patients with beyond-the-Milan criteria HCC who have viable tumour after first TACE will be enrolled in this multi-institutional, parallel, pragmatic, randomized controlled trial. Patients with metastasis, vascular invasion, or a sum of tumour diameter > 8 cm will be excluded. Eligible patients will be randomly assigned to combination TACE and RFA therapy or TACE monotherapy groups. Patients in the combination therapy group will receive a second TACE and subsequent RFA at the viable tumour. Patients in the TACE monotherapy group will receive only second TACE. Patients in both groups will undergo magnetic resonance imaging 4-6 weeks after second TACE. The primary endpoint is 1-month tumour response, and secondary endpoints are progression-free survival, overall response rate, number of treatments until CR, overall survival, and change in liver function. DISCUSSION: Although TACE can be used to treat intermediate-stage HCC, it is difficult to achieve CR by first TACE in most intermediate-stage patients. Recent studies show a survival advantage of combination therapy over monotherapy. However, most studies evaluating combination therapy included patients with a single tumour sized < 5 cm, and no studies included patients with intermediate-stage but more advanced (i.e., beyond-the-Milan criteria) HCC. This study will evaluate the efficacy of combined TACE and RFA therapy for patients with advanced HCC within the intermediate stage. TRIAL REGISTRATION: Clinical Research Information Service (CRiS) KCT0006483.

Safety and effectiveness of direct-acting antivirals in patients with chronic hepatitis C and chronic kidney disease.

BACKGROUND/AIMS: To evaluate the effectiveness and safety of direct acting antivirals (DAAs) available in chronic kidney disease (CKD) patients with hepatitis C virus (HCV) infection in Korea. METHODS: In a retrospective, multicenter cohort study, 362 patients were enrolled from 2015 to 2019. The effectiveness and safety of DAAs including glecaprevir/pibrentasvir, sofosubvir/ribavirin, ledipasvir/sofosbuvir, and daclatasvir/asunaprevir were analyzed for patients according to CKD stage. We evaluated sustained virologic response at week 12 after treatment (SVR12) as primary endpoint. The effectiveness and safety were also evaluated according to CKD stage. RESULTS: Among 362 patients, 307 patients completed DAAs treatment and follow-up period after end of treatment. The subjects comprised 87 patients (62 with CKD stage 3 and 25 with CKD stage (4-5), of whom 22 were undergoing hemodialysis). HCV patients with CKD stage 1 and 2 (estimated glomerular filtration rate [eGFR] ≥ 60 mL/min/1.73 m2) showed SVR12 of 97.2% and 95.4% respectively. SVR12 of CKD stage 3 and 4-5 (eGFR < 60 mL/min/1.73 m2) patients was 91.9% and 91.6% respectively. Patients undergoing hemodialysis achieved SVR12 (90.9%). Treatment failure of DAAs in stage 1, 2, 3, and 4-5 was 2.8%, 2.7%, 1.6%, and 4%. DAAs showed good safety profile and did not affect deterioration of renal function. CONCLUSION: DAAs shows comparable SVR12 and safety in CKD patients (stage 3, 4, and 5) with HCV compared with patients with stage 1 and 2. The effectiveness and safety of DAAs may be related to the treatment duration. Therefore, it is important to select adequate regimens of DAAs and to increase treatment adherence.

A decrease in functional microbiomes represented as Faecalibacterium affects immune homeostasis in long-term stable liver transplant patients.

ABBREVIATIONS: LT, liver transplantation; HCC, hepatocellular carcinoma; IS, immunosuppressants; DC, dendritic cells; Treg, regulatory T; Th17, T helper 17; AST, aspartate transaminase; ALT, alanine transaminase; OUT, operational taxonomic unit; LEfSe, linear discriminant analysis effect size; LDA, linear discriminant analysis; IL, interleukin; TGF, transforming growth factor; GM-CSF, granulocyte-macrophage colony-stimulating factor; IFN, interferon; TNF-α, tumor necrosis factor-α; MIP-1α, macrophage inflammatory protein-1α; IP-10, interferon γ-induced protein; MCP-1, monocyte chemoattractant protein-1; ACR, acute cellular rejection; NF-κB, nuclear factor κB; PT INR, prothrombin time; QC, quality check; PBMC, peripheral blood mononuclear cells; PBS, phosphate-buffered saline; ELISA, enzyme-linked immunosorbent assay.

Long-term Clinical Outcomes and Predictive Factors for Living-donor Liver Transplant Recipients With Biliary Strictures.

BACKGROUND: Biliary strictures frequently occur in living-donor liver transplant (LDLT) recipients. However, long-term clinical outcomes and their associated factors are unclear. METHODS: We analyzed an historical cohort of 228 recipients who underwent LDLT with post-liver transplantation biliary strictures. Endoscopic retrograde cholangiography or percutaneous transhepatic biliary drainage were performed to treat biliary strictures. Patients that experienced persistent jaundice over 3 mo after the initial treatment were defined as a remission-failure group. RESULTS: Median observation period was 8.5 y after the diagnosis of biliary stricture. The 15-y graft survival (GS) rate was 70.6%, and 15-y rate of developing portal hypertension (PH) was 26.1%. Remission failure occurred in 25.0% of study participants. In the multivariate analysis, biopsy-proven acute rejection, and portal vein/hepatic artery abnormalities were risk factors for remission failure. Development of PH, retransplantation, and death were significantly more frequent in the remission-failure group. Remission failure and PH were associated with poor GS. In multivariate analyses, hepatic artery abnormality and biloma were common significant factors that were associated with a poor GS and development of PH. CONCLUSIONS: The insufficient blood supply reflected by hepatic artery abnormality and biloma might be the most important factor that can predict poor long-term survival in LDLT patients with biliary strictures. Future large-scale prospective studies are needed to validate our observations.

Real-life experience of ledipasvir and sofosbuvir for HCV infected Korean patients: a multicenter cohort study.

BACKGROUND/AIMS: To evaluate the efficacy and safety of ledipasvir/sofosbuvir (LDV/SOF) therapy in hepatitis C virus (HCV)-infected Korean patients in a real clinical setting. METHODS: A total of 273 patients who received LDV/SOF therapy between May 2016 and February 2021 were consecutively enrolled and analyzed. A per-protocol analysis was performed to evaluate the virologic response. RESULTS: Seventy-five percent were infected with genotype 1, and 25% were infected with genotype 2. A hundred eightyone (66.3%) patients had chronic hepatitis, 74 (27.1%) had compensated cirrhosis, eight (2.9%) had decompensated cirrhosis, and 10 (3.7%) had undergone liver transplantation. Undetectable HCV RNA at week 4 was achieved in 90.2% (231/256) of patients, 99.2% (250/252) achieved the end of treatment response, and 98.1% (202/206) achieved sustained virologic response at 12 weeks post-treatment (SVR12). According to liver function, the SVR12 rates were 99.3% (135/136) in chronic hepatitis, 96.4% (53/55) in compensated cirrhosis, and 100% (6/6) in decompensated cirrhosis. The SVR12 rates according to the genotype were 98.2% (167/170) for genotype 1 and 97.2% (35/36) for genotype 2. An 8-week LDV/SOF treatment in treatment-naïve chronic hepatitis patients with HCV RNA < 6,000,000 IU/mL at baseline resulted in 100% (23/23) SVR12 rates. Overall, LDV/SOF was tolerated well, with a 0.7% (2/273) discontinuation rate due to adverse events that were unrelated to LDV/SOF. CONCLUSION: LDV/SOF is effective and safe for treating HCV-infected Korean patients with high SVR12 rates.

Patient-Derived Avatar Mouse Model to Predict the Liver Immune Homeostasis of Long-Term Stable Liver Transplant Patients.

Although rejection or tolerance can occur in liver transplantation (LT) patients, there are no reliable non-invasive methods for predicting immune homeostasis. In this study, we developed a humanized mouse model to predict liver immune homeostasis in patients who underwent LT. The patient-derived avatar model was developed by injecting peripheral blood mononuclear cells from healthy controls (HCs) or LT patients with stable, rejection, or tolerance into NOD.Cg-PrkdcscidIL2rgtm1Wjl/SzJ (NSG) mice, followed by injection of human hepatic stellate cells and Carbone tetrachloride (CCl4). After 7 weeks, the patient's T-cell engraftment and liver inflammation in the avatar model were evaluated and compared with the liver histology of LT patients. Changes in liver inflammation following treatment with tacrolimus and/or biguanide derivatives were also examined. The C-X-C Motif Chemokine Receptor 3 (CXCR3)-dependently engrafted patient T cells led to differences in liver inflammation in our model according to the status of LT patients. The livers of avatar models from rejection patients had severe inflammation with more T helper 17 cells and fewer regulatory T cells compared to those of models from tolerance and HCs showing only mild inflammation. Moreover, our model classified stable post-LT patients into severe and mild inflammation groups, which correlated well with liver immunity in these patients. Our models revealed alleviation of inflammation after combination treatment with tacrolimus and biguanide derivatives or monotherapy. Consequently, using our new patient-derived avatar model, we predicted liver immune homeostasis in patients with stable LT without biopsy. Moreover, our avatar model may be useful for preclinical analysis to evaluate treatment responses while reducing risks to patients.

FK506 and Lactobacillus acidophilus ameliorate acute graft-versus-host disease by modulating the T helper 17/regulatory T-cell balance.

BACKGROUND: Graft-versus-host disease (GvHD) is a critical complication after allogeneic hematopoietic stem cell transplantation (HSCT). The immunosuppressants given to patients undergoing allogeneic HSCT disturb the microbiome and the host immune system, potentially leading to dysbiosis and inflammation, and may affect immune function and bone marrow transplantation. The intestinal microbiome is a target for the development of novel therapies for GvHD. Lactobacillus species are widely used supplements to induce production of antimicrobial and anti-inflammatory factors. METHODS: We determined the effect of the combination of Lactobacillus acidophilus and FK506 on GvHD following major histocompatibility complex-mismatched bone marrow transplantation. RESULTS: The combination treatment suppressed IFN-γ and IL-17-producing T cell differentiation, but increased Foxp3+Treg differentiation and IL-10 production. Also, the combination treatment and combination treated-induced Treg cells modulated the proliferation of murine alloreactive T cells in vitro. Additionally, the combination treatment upregulated Treg-related genes-Nt5e, Foxp3, Ikzf2, Nrp1 and Itgb8-in murine CD4+-T cells. The combination treatment also alleviated GvHD clinically and histopathologically by controlling the effector T cell and Treg balance in vivo. Moreover, the combination treatment decreased Th17 differentiation significantly and significantly upregulated Foxp3 and IL-10 expression in peripheral blood mononuclear cells from healthy controls and liver transplantation (LT) patients. CONCLUSIONS: Therefore, the combination of L. acidophilus and FK506 is effective and safe for patients undergoing allogeneic hematopoietic stem cell transplantation.

A Real-World Comparative Study of Microwave and Radiofrequency Ablation in Treatment-Naïve and Recurrent Hepatocellular Carcinoma.

The efficacy and safety of microwave ablation (MWA) compared to radiofrequency ablation (RFA) for patients with treatment-naïve and recurrent hepatocellular carcinoma (HCC) has not been clarified in Korea. There were 150 HCC patients (100 in the RFA group and 50 in the MWA group) enrolled in our study. The primary outcome was one- and two-year disease-free survival (DFS). Secondary outcomes were complete response (CR) rate, two-year survival rate, risk factors for DFS and complication rate. Treatment outcomes were also assessed using propensity-score matching (PSM). The MWA group had better one- and two-year DFS than the RFA group (p = 0.035 and p = 0.032, respectively), whereas the CR rate, two-year survival rate, and complication rate were similar between the two groups with fewer major complications in the MWA group (p = 0.043). Patients with perivascular tumors, high risk of recurrence, and small tumor size (≤3 cm) were more suitable for MWA than RFA. MWA was also an independent factor for favorable one- and two-year DFS. Finally, the MWA group still showed better one- and two-year DFS than the RFA group after PSM. In conclusion, MWA could be an alternative treatment to RFA especially in patients with a high risk of recurrence, perivascular tumors, and small tumor size.

Diagnostic performance of serum exosomal miRNA-720 in hepatocellular carcinoma.

Background/Aim: Hepatocellular carcinoma (HCC) is associated with poor prognosis, largely due to late detection. Highly accurate biomarkers are urgently needed to detect early-stage HCC. Our study aims to explore the diagnostic performance of serum exosomal microRNA (miR)-720 in HCC. Methods: Exosomal miRNA was measured via quantitative real-time PCR. A correlation analysis of exosomal miR-720 and tumor or clinico-demographic data of patients with HCC was performed. The receiver operating characteristic (ROC) curve was used to assess the diagnostic capacity of serum exosomal miR-720 for HCC, in comparison with α-fetoprotein (AFP) and prothrombin induced by vitamin K absence or antagonist-II (PIVKA-II). Results: MiR-720 was chosen as a potential HCC marker via miR microarray based on significant differential expression between tumor and non-tumor samples. Serum exosomal miR-720 was significantly upregulated in patients with HCC (n=114) versus other liver diseases (control, n=30), with a higher area under the ROC curve (AUC, 0.931) than the other markers. Particularly, serum exosomal miR-720 showed superior performance in diagnosing small HCC (<5 cm; AUC, 0.930) compared with AFP (AUC, 0.802) or PIVKA-II (AUC, 0.718). Exosomal miR-720 levels showed marginal correlation with tumor size. The proportion of elevated miR-720 also increased with intrahepatic tumor stage progression. Unlike AFP or PIVKA-II showing a significant correlation with aminotransferase levels, the exosomal miR-720 level was not affected by aminotransferase levels. Conclusions: Serum exosomal miR-720 is an excellent biomarker for the diagnosis of HCC, with better performance than AFP or PIVKA-II. Its diagnostic utility is maintained even in small HCC and is unaffected by aminotransferase levels.