Improving risk stratification and detection of early HCC using ultrasound-based deep learning models

Surveillance for Hepatocellular Carcinoma in Patients With Cirrhosis

Purpose: Hepatocellular carcinoma (HCC) is a leading cause of cancer deaths worldwide, which is mostly diagnosed at advanced and incurable stages. The suboptimal performance of current screening approaches including serum alpha-fetoprotein testing and ultrasound, highlights the urgent need for novel biomarkers to enable risk stratification and early detection of HCC. Based on our previous mechanistic insights with data from human genomic studies available in The Cancer Genome Atlas (TCGA) and from animal models (Cell Rep. 2024.43(9):114676 and Sci Transl Med. 2021.13(624): eabk2267), we have identified functional protein markers associated with HCC such as TGFBR2, MSTN, and PKM2. In this study, we present a model based on the TGF-β and associated immune pathways for the HCC stratification in cirrhotic patients that incorporates biologically relevant proteins to capture ongoing pathophysiological changes that drive hepatocarcinogenesis. Experimental Procedures: A proteomic analysis of 7, 000 proteins (SomaScan) utilizing serum samples (55 uL/each) from 311 individuals with cirrhosis, 18 of whom developed HCC by imaging criteria during the median 5-year follow-up. Using a multivariable logistic regression method, we conducted an unsupervised clustering analysis of TGF-β and the corresponding immune pathway and combined with well-described serum biomarkers to create a risk prediction model. We followed this with ELISA validation of the most promising biomarkers. Results: In our prospective cirrhosis cohort (n=311), 18 HCC cases (5.8%) developed over a median follow-up of 5 years. Using unsupervised hierarchical clustering, we analyzed TGF-β pathway (n=28) and inflammatory/immune response proteins (n=36), grouping patients into two subgroups based on serum protein expression. Among HCC cases, 77.8% showed higher SMAD3 and 66.7% showed higher TGFBR2 expression. Cirrhotic patients with elevated expression of serum HMGB1/IL34/CCL3L1/IL37/SMAD3/TGFBR2 had a six-fold higher HCC risk (10.64% vs. 1.76%). A TGF-β model combining TGFBR2, GPC3, ANGPTL8, and GZMK outperformed the AFP model, with AUCs of 0.88 vs. 0.60 and better sensitivity (0.61 vs. 0.11) at 0.90 specificity. ELISA validation confirmed that MSTN, TGFBR2, and HMGB2 levels were significantly elevated in HCC (p<0.05). Conclusions and Next Steps: This study successfully demonstrates the high performance of our TGF-β and corresponding immune pathway biologically related biomarkers for HCC risk stratification in cirrhotic patients. The next steps will be large-scale validations of this biomarker panel in cirrhotic patients across all racial and ethnic subgroups in Phase II/III studies. Citation Format: Xiyan Xiang, Richard L. Amdur, Kirti Shetty, Herbert Yu, Linda L. Wong, Nyasha Chambwe, Sanjaya K. Satapathy, James M. Crawford, Lopa Mishra. Biomarkers in TGF-β and corresponding Immune pathway stratify HCC-risk in cirrhotic patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 7061.

Predictive models for hepatocellular carcinoma development after sustained virological response in advanced hepatitis C

ObjectiveOur study was designed to explore the utility of 99mTc-HYNIC-PEG4-E[PEG4-c(RGDfK)]2 (99mTc-3PRGD2) for the detection of hepatocellular carcinoma (HCC) and specifically to compare the diagnostic performance of 99mTc-3PRGD2 integrin receptor imaging and 2-18-fluoro-2-deoxy-d-glucose (18F-FDG) metabolic imaging in a nude mouse model.Methods99mTc-3PRGD2 was synthesized using a HYNIC-3PRGD2 lyophilized kit with 99mTcO4 labelling. The nude mouse animal model was established by subcutaneously injecting 5 × 107/ml HepG2 cells into the shoulder flank of each mouse. Biodistribution studies were performed at 0.5, 1, 2 and 4 h after intravenous administration of 0.37 MBq of 99mTc-3PRGD2. Immunohistochemistry was performed to evaluate the expression level of integrin αvβ3 in the HCC tissues. Dynamic imaging was performed using list-mode after the administration of 55.5 MBq of 99mTc-3PRGD2, to reconstruct the multiphase images and acquire the best initial scan time. At 8, 12, 16, 20 and 24 days after inoculation with HepG2 cells, 55.5 MBq of 99mTc-3PRGD2 and 37 MBq of 18F-FDG were injected successively into the nude mouse model, subsequently, simultaneous SPECT/PET imaging was performed to calculate the tumour volume and tumour uptake of 99mTc-3PRGD2 and 18F-FDG.ResultsThe biodistribution study first validated that the tumour uptake of 99mTc-3PRGD2 at the different time points was higher than that of all the other organs tested in the experiment, except for the kidney. Integrin αvβ3 expressed highly in early stage HCC and declined for further necrosis of the tumour tissue. Subcutaneous tumours were visualized clearly with excellent contrast under 99mTc-3PRGD2 SPECT/CT imaging, and the multiphase imaging comparison showed the tumours were prominent at 0.5 h, suggesting that the best initial scan time is 0.5 h post-injection. The comparison of the imaging results of the two methods showed that 99mTc-3PRGD2 integrin receptor imaging was more sensitive than 18F-FDG metabolic imaging for the detection of early stage HCC, meanwhile the tumour uptake of 99mTc-3PRGD2 was consistently higher than that of 18F-FDG. However, as tumour necrosis further increased in HCC tissues, the uptake of 18F-FDG was higher than that of 99mTc-3PRGD2.ConclusionOur study demonstrated that 99mTc-3PRGD2 is a valuable tumour molecular probe for the detection of early stage HCC compared with 18F-FDG, meriting further investigation of 99mTc-3PRGD2 as a novel SPECT tracer for tumour imaging.

Background and aimsDual programmed death 1 (PD-1) and angiogenesis blockade therapy is a frontline treatment for hepatocellular carcinoma (HCC). An accepted model for survival prediction and risk stratification in individual...

Thyroid carcinoma is the only cancer that regards age as an important predictor of thyroid cancer-specific survival (CSS). While the 8th American Joint Committee on Cancer (AJCC) staging system raised the age cutoff from 45 to 55 years for differentiated thyroid carcinoma (DTC) to more accurately predict the prognosis, there is no new information regarding the role of age in the prognosis of anaplastic thyroid carcinoma (ATC). The aim of this study was to determine the optimal age cutoff values for mortality risk stratification in ATC patients. Furthermore, a nomogram to predict ATC CSS was developed in each age group. Patients diagnosed with ATC between 2004 and 2015 were identified in the Surveillance, Epidemiology, and End Results (SEER) database. After applying inclusion and exclusion criteria, a total of 1140 patients were enrolled as cohort 1 to describe the characteristics of ATC, while a total of 556 patients were included as cohort 2 to determine age cutoff values for risk stratification by X-tile program. Training set and testing set were randomly generated to develop and validate a predictive nomogram of CSS in each age group. The 6-month, 1-year, and 2-year survival was 27.6%, 15.1%, and 6.2%, respectively, in cohort 1. X-tile program results showed that the optimal age cutoff values for mortality risk stratification were 65 and 85 years old (p < 0.001). Distant metastasis was independently associated with CSS in patients younger than 85 years old, and these patients benefited more from complete resection of the tumor and radiotherapy/chemotherapy. However, no predictors of CSS were identified in patients over 85 years old, and interventions (surgery, radiotherapy, and chemotherapy) targeting ATC had little role in disease control in these patients. The nomogram was developed and validated based on the independent CSS predictors in each age group. The C-index values of the training set and testing set were 0.735 [95% CI, 0.696-0.774] and 0.733 [95% CI, 0.663-0.804] for CSS in patients of ≤64 years old, while the values were 0.767 [95% CI, 0.730-0.804] and 0.783 [95% CI, 0.718-0.848] in patients of 65-84 years old. All of the C-index values were larger than 0.7, which showed acceptable prediction performance of the nomograms. Age can be used as an auxiliary stratification factor of prognosis in ATC patients. The survival may be improved in patients younger than 85 years old if combination therapy (surgery, radiotherapy, and chemotherapy) was indicated and applicable, while no optimal therapeutic strategy was determined in patients older than 85 years old. The established nomograms can provide good prediction of CSS according to age group.

Liver cancer is the second leading cause of cancer death worldwide, responsible for an estimated 746,000 deaths and 782,000 new cases in 2012 [1]. The countries in the Western Pacific region accounts for 64% and China alone accounts for 51% of the new liver cancer cases and deaths each year. Approximately 80% of hepatocellular carcinoma (HCC), the most common type of liver cancer, is associated with viral hepatitis [1]. In countries with high prevalence of hepatitis B virus (HBV) infection, such as China, up to 80% of HCC is associated with hepatitis B [2]. Liver cancer carries a poor prognosis with a global mortality to incidence ratio of 0.95 [1]. In the USA, the 1-year survival rate remains less than 50% [3]. Asians and Pacific Islanders have the highest incidence of HCC among the different racial/ethnic groups. While every individual with chronic hepatitis B infection (CHB) or HBV carrier is at risk for developing HCC, disease progression leading to cirrhosis is associated with the greatest risk [2]. In an effort to improve survival through early detection of HCC, the American Association for the Study of Liver Diseases recommended HCC screening with abdominal ultrasound (US) at 6–12 month intervals of HBV carriers considered at increased risk for HCC, including those with cirrhosis, family history of liver cancer and Asian male HBV carriers above the age of 40 [4,5]. However, it remains controversial whether population-wide HCC screening of CHB patients results in a reduction in HCC mortality particularly in resource-constraint regions of the world that have the highest burden of HCC. The only two randomized trials available to date are both from China [6,7]. A population-based study in the city of Shanghai, conducted by the Liver Cancer Institute of Fudan University, reported 1-year HCC survival rates improved from 31.2% in the control group to 65.9% in the screened group among 18,816 CHB patients aged 35–59 years with or without cirrhosis, who were randomized into either a group using US and AFP screening every 6 months or a control group [6]. The second study from a rural setting in Qidong, using AFP screening every 6 months without US, did not show any improvement in survival [7]. According to the National Cancer Institute [8], screening for HCC

BackgroundUltrasonography (US) is widely used as a standard surveillance tool for patients who are at a high risk of having hepatocellular carcinoma (HCC); however, conventional B-mode US appears to be insufficient in order to ensure the early detection of HCC. Perfluorobutane allows very stable Kupffer phase imaging for at least 60 min, which is tolerable for examinations of the entire liver. The purpose of our study is to evaluate the added value of contrast-enhanced US using perfluorobutane to that of conventional B-mode US as an HCC surveillance tool for patients with liver cirrhosis.Methods/DesignSCAN (Sonazoid-US for surveillance of hepatoCellulArcarciNoma) is a prospective, multi-institutional, diagnostic trial using an intra-individual comparison design in a single arm of patients. This study was approved by our five institutional review board and informed consent was obtained from all participating. We obtained consent for publication of these data (contrast enhanced US images, CT or MRI images, laboratory findings, age, sex) from all participating patients. All patients will undergo conventional B-mode US immediately followed by contrast-enhanced US. The standardized case report forms will be completed by operating radiologists after B-mode US and contrast-enhanced US, respectively. If any lesion(s) is detected, the likelihood of HCC will be recorded. The primary endpoints are a detection rate of early-stage HCC and a false referral rate of HCC. Intra-individual comparison using Mcnemar’s test will be performed between B-mode US and contrast-enhanced US. The study will include 523 patients under HCC surveillance in five medical institutions in Korea.DiscussionSCAN is the first study to investigate the efficacy of contrast-enhanced US in surveillance using two reciprocal endpoints specialized for the evaluation of a surveillance test. SCAN will provide evidence regarding whether patients can truly benefit from contrast-enhanced US in terms of the detection of early stage HCC while avoiding additional unnecessary examinations. In addition to the study protocol, we elaborate on potentially debatable components of SCAN, including the design of an intra-individual comparison study, study endpoints, composite reference standards, and indefinite imaging criteria regarding the likelihood of HCC.Trial registrationThe date of trial registration (ClincalTrials.gov: NCT02188901) in this study is July 3, 2014. The last patient enrolled in August 30, 2016 and follow up to see the primary end point is still ongoing. All authors have no other relationships/conditions/circumstances that present a potential conflict of interest of relationships. Our study protocol has undergone peer-review by the funding body (GE Healthcare). No other relationships/conditions/circumstances that present a potential conflict of interest. Also, we clearly stated in the 'competing interests' section of my manuscript.

Hepatocellular carcinoma (HCC) remains a significant global health concern, often diagnosed at advanced stages, limiting the efficacy of surgical interventions. Early and accurate diagnosis is critical for improving surgical outcomes and reducing mortality. Traditional biomarkers, such as alpha-fetoprotein (AFP), des-gamma-carboxyprothrombin (DCP), and the lectin-bound fraction of AFP (AFP-L3), show limited sensitivity and specificity. Advanced diagnostic models, including GALAD, TAGALAD, and GAP_TALAD, offer a promising multi-biomarker approach but lack extensive evaluation in surgical contexts. This retrospective study included a cohort of 267 untreated hepatocellular carcinoma patients and 231 control patients (with hepatitis or cirrhosis). We applied the predefined formulas for the TAGALAD, GAP_TALAD, and other models to the cohort data. The diagnostic performance of each model and individual biomarker for detecting HCC was assessed using receiver operating characteristic (ROC) curve analysis to determine the area under the curve (AUC), sensitivity, and specificity at optimal cut-offs. Additionally, key clinical subgroups, including pathologically confirmed HCC, clinically diagnosed HCC, early-stage HCC (TNM I+II), patients with complete data (no imputation), and hepatitis B virus (HBV)-related disease, were also analyzed. TAGALAD and GAP_TALAD demonstrated superior performance compared to the GALAD model and traditional biomarkers across all patient subgroups. Notably, TAGALAD achieved the highest diagnostic accuracy, with an AUC of 0.880, sensitivity of 0.760, and specificity of 0.861, followed closely by GAP_TALAD (AUC = 0.874). Both models demonstrated excellent performance in early-stage HCC detection (TAGALAD AUC = 0.860, GAP_TALAD AUC = 0.867), highlighting their potential in identifying candidates for surgical resection or transplant at an early curative stage. In HBV-related HCC, TAGALAD (AUC = 0.874) and GAP_TALAD (AUC = 0.857) showed superior diagnostic accuracy compared to GALAD (AUC = 0.731) and single biomarkers (AUC = 0.598-0.799). The TAGALAD and GAP_TALAD models offer a robust and reliable framework that supports early diagnosis of HCC. Their superior accuracy indicates a more reliable foundation for identifying candidates for curative surgical interventions, suggesting the potential to refine patient selection. Future research should focus on multi-center validation and the integration of novel biomarkers to further optimize these models for surgical decision-making and personalized treatment strategies.

There are limited data on the effect and evolution of risk factors for hepatocellular carcinoma (HCC) in patients with virologically cured hepatitis C virus (HCV) infection. We conducted a retrospective cohort study of patients with HCV who achieved sustained virological response with direct-acting antivirals from 130 Veterans Administration hospitals during 2014-2018, followed through 2021. Cox proportional hazards models were constructed at 3 landmark times (baseline and 12 and 24 months after sustained virological response) to examine associations between demographic, clinical, and behavioral factors and HCC risk, stratified by cirrhosis status. Among 92,567 patients (32% cirrhosis), 3,247 cases of HCC were diagnosed during a mean follow-up of 2.5 years. In patients with cirrhosis, male sex (hazard ratios [HR]: 1.89, 1.93, and 1.99), cirrhosis duration ≥5 years (HR: 1.71, 1.79, and 1.34), varices (HR: 1.73, 1.60, and 1.56), baseline albumin (HR: 0.48, 0.47, and 0.49), and change in albumin (HR: 0.82 and 0.90) predicted HCC risk at each landmark time. HCV genotype 3, previous treatment, bilirubin, smoking, and race influenced HCC risk at baseline, but their effects attenuated over time. In patients without cirrhosis, diabetes (HR: 1.54, 1.42, and 1.47) and hypertension (HR: 1.59, 1.65, and 1.74) were associated with HCC risk at all landmark times. Changes in fibrosis-4 scores over time were associated with HCC risk both in patients with and without cirrhosis. Risk factors for HCC were different in patients with and without cirrhosis and some also evolved during follow-up. These factors can help with risk stratification and HCC surveillance decisions in patients with cured HCV.

BackgroundThis study aimed to investigate the value of the modified Medical Research Council (mMRC) dyspnea scale in risk stratification and outcome assessment for patients with acute pulmonary embolism (APE).MethodsA retrospective analysis was performed using medical records from a tertiary care center between 2011 and 2023. The study included patients aged 18–80 years who were diagnosed with APE. Participants were categorized into groups based on pulmonary embolism risk stratification, mMRC dyspnea scale, and the presence or absence of adverse outcomes within 1 year, which included in-hospital mortality, all-cause mortality after discharge, and hospital readmission. The associations between the mMRC dyspnea scale and both APE risk stratification and 1-year adverse outcomes were evaluated. The predictive performance of the mMRC dyspnea scale for 1-year adverse prognosis was assessed using receiver operating characteristic (ROC) curve analysis to determine the optimal cut-off threshold.ResultsThe study demonstrated that the mMRC dyspnea scale was significantly positively correlated with risk stratification of APE (P < 0.05). Moreover, the Qanadli score, systolic blood pressure, cardiac troponin, N-terminal pro-brain natriuretic peptide, and the right-to-left ventricular ratio were significantly associated with higher mMRC scores (P < 0.05). ROC curve analysis revealed that an mMRC dyspnea grade of 3 was considered the optimal cut-off value for predicting adverse prognosis within 1 year, with an area under the curve of 0.803 (P < 0.001).ConclusionThe mMRC dyspnea scale demonstrates a significant association with risk stratification in patients with APE. An mMRC dyspnea grade of ≥3 is indicative of a higher risk for adverse outcomes within 1 year and may serve as a valuable prognostic indicator for predicting clinical outcomes in APE patients.

BackgroundDynamic changes in fibrosis markers occur under long-term antiviral treatment (AVT) for chronic hepatitis B. We evaluated prognostic values of on-treatment liver stiffness (LS) compared to ultrasonography findings and determined its optimal cutoff.MethodsThe cumulative probability of hepatocellular carcinoma (HCC) was assessed among 880 patients receiving entecavir or tenofovir for ≥2 years. LS was measured using transient elastography.ResultsAfter ≥2 years’ AVT, the proportion of patients with cirrhosis on ultrasonography decreased from 54.7% to 44.9% and the mean LS decreased from 13.6 to 8.2 kPa (both p<0.001). However, unlike cirrhosis on ultrasonography before AVT (p<0.001), that after ≥2 years’ AVT did not discriminate HCC risk (p=0.792). Using the Contal and O’Quigley’s method, pre-AVT and on-treatment LS of 12.0 and 6.4 kPa, respectively, were chosen as optimal cutoffs to successfully discriminate HCC risk (both p<0.001). However, through stratification using both pre-AVT and on-treatment LS, the prognosis was finally determined according to on-treatment LS of 6.4 kPa, regardless of pre-AVT LS of 12.0 kPa. Using on-treatment LS of 12 kPa suggested by Caucasians with CHB receiving long-term AVT, patients with higher LS were more likely to develop HCC than those with lower LS (p=0.017); however, there was no significant difference between those with on-treatment LS of 6.4–11.9 and ≥ 12.0 kPa (p=0.920).ConclusionFor HCC risk stratification in patients receiving long-term AVT, on-treatment LS cutoff should be lowered to 6.4 kPa, which is more predictive than 12 kPa or cirrhosis on ultrasonography. Further studies are required for validation.

Few genome-wide association studies have examined genetic variants associated with hepatocellular carcinoma (HCC) risk in individuals seronegative for HBsAg and anti-HCV, and the long-term impact of these variants remains uncertain. This multi-stage study analyzed adults >30 years old who were seronegative for HBsAg and anti-HCV. In the genome-wide association study discovery phase, 765 HCC cases and 9,949 controls were analyzed for 308,693 SNPs, with significant SNPs confirmed in community-based (171 HCC cases, 684 controls) and hospital-based (470 HCC cases, 5,460 controls) validation sets. A cohort of 67,909 participants, followed from 2012 to 2021, was used to evaluate the long-term HCC risk associated with these variants. Ten SNPs in PNPLA3/SAMM50 were significantly associated with HCC risk (p <1.62 × 10-7) and were in high linkage disequilibrium. Three SNPs (rs738409, rs2281135, rs2235776) were replicated and demonstrated strong associations with HCC, independent of steatosis. Over 267,238 person-years of follow-up, 32 new HCC cases occurred, with elevated risks observed in individuals homozygous for the risk genotypes: GG (rs738409), AA (rs2281135), and TT (rs2235776). The adjusted hazard ratios (95% CI) were 3.37 (1.32-8.63), 2.80 (1.06-7.37), and 2.64 (0.91-7.66), respectively. In allelic models, carriers of the risk allele had higher HCC risk, with adjusted hazard ratios (95% CI) of 1.88 (1.14-3.10) for rs738409, 1.68 (1.02-2.78) for rs2281135, and 1.61 (0.98-2.67) for rs2235776. This study identified PNPLA3/SAMM50 variants significantly associated with long-term HCC risk in individuals without viral hepatitis. These findings highlight their potential utility as biomarkers for HCC risk stratification and underscore the need for further research into underlying mechanisms. Variants in the PNPLA3/SAMM50 locus were significantly associated with hepatocellular carcinoma (HCC) risk in individuals seronegative for HBsAg and anti-HCV. The stepwise increase in HCC risk with higher numbers of risk alleles suggests their potential utility as biomarkers to identify high-risk individuals without a history of chronic hepatitis B or C virus infection. These findings provide a foundation for genetic risk stratification and precision prevention strategies in populations where non-viral etiologies of HCC are becoming increasingly important.

Despite a recommended multidimensional approach for pulmonary hypertension (PH) risk stratification and guidance of treatment decisions, this may not always be achievable in patients with advanced disease. One issue is the lack of an imaging modality to assess right ventricular (RV) structure and function abnormalities. To explore the risk stratification and prognostic value of cardiac MR feature tracking (MR-FT)-derived RV strain. Retrospective. A total of 80 patients with idiopathic pulmonary artery hypertension (N=52) or chronic thromboembolic PH (N=28). A 1.5 T or 3.0 T, balanced steady-state free precession sequence. All patients underwent laboratory testing, right heart catheterization, and MR imaging (and in 37 cases, a cardiopulmonary exercise test was also performed) within a 1-month period. Cardiac functional parameters and both global longitudinal strain (GLS) and global circumferential strain (GCS) were analyzed. Patients were stratified into low, intermediate, and high-risk groups by guideline suggested stratified values of risk factors. The combined endpoint was death or hospitalization for congestive heart failure assessed during follow-up since the date of MR examination. Kolmogorov-Smirnov's test, independent-sample t-tests, Wilcoxon's rank-sum tests, one-way analysis of variance, χ2 tests or Fisher's exact test, receiver operating characteristic analysis, Kaplan-Meier survival analysis, and Cox regression analysis. A P value < 0.05 was considered statistically significant. The median follow-up duration was 3.4 years. Thirty-five patients presented with combined endpoint including 10 cardiac deaths. RV structural and deformation impairments were significantly associated with combined endpoint (ejection fraction: 31.3% ± 13.2% vs. 38.0% ± 14.8%, hazard ratio [HR: 0.974; GLS: -14.5 [-18.6, -10.9] % vs. -20.4 [-26.0, -13.2] %, HR: 1.071; GCS: -9.8 [-14.5, -7.3] % vs. -12.3 [-19.9, -8.4] %, HR: 1.059). There were significant differences in RVGLS among low, intermediate, and high-risk groups (-19.3% ± 7.2% vs. -17.3% ± 9.4% vs. -11.5% ± 4.4% by cardiac functional class, -21.8% ± 7.3% vs. -19.4% ± 8.2% vs. -12.7 ± 5.3% by NT-proBNP, -19.7% ± 7.7 vs. -15.8% ± 6.5% vs. -12.6% ± 8.2% by cardiac index). RV deformation may aid risk stratification in patients with PH, providing crucial information for RV remodeling, pulmonary hemodynamic condition and exercise capacity. 3 TECHNICAL EFFICACY: Stage 2.

Hepatocellular carcinoma (HCC) incidence has dramatically decreased in patients infected with HCV and HBV due to the widespread use of highly effective antiviral agents. Nevertheless, a substantial proportion of patients with advanced fibrosis or cirrhosis following HCV clearance of in case of HBV control whatever the stage of fibrosis remains at risk of liver cancer development. Cancer predictors in these virus-free patients include routine parameters estimating coexisting comorbidities, persisting liver inflammation or function impairment, and results of non-invasive tests which can be easily combined into HCC risk scoring systems. The latter enables stratification according to various liver cancer incidences and allocation of patients into low, intermediate or high HCC risk probability groups. All international guidelines endorse lifelong surveillance of these patients using semi-annual ultrasound, with known sensibility issues. Refining HCC prediction in this growing population ultimately will trigger personalized management using more effective surveillance tools such as contrast-enhanced imaging techniques or circulating biomarkers while taking into account cost-effectiveness parameters.