Use of Gender, Age, Alfa-fetoprotein (AFP), and Des-gamma-carboxyprothrombin (PIVKA-II) or GAAD Score in Addition to Ultrasound for Surveillance of People At-risk for Developing Hepatocellular Carcinoma in Asia in Order to Detect Early Liver Cancer

Overview

HCC surveillance is currently limited by underutilization and the suboptimal performance of AFP. This prospective, single-arm study investigates whether the GAAD score (Gender, Age, AFP, and PIVKA-II) enhances HCC detection when added to standard-of-care surveillance. High-risk patients will undergo US plus GAAD score testing every six months for two years. The primary analysis compares the relative true positive rate (rTPR) and relative false positive rate (rFPR) of surveillance modalities (US, AFP, GAAD) against combined strategies (US+AFP; US+GAAD), utilizing a 2.57 GAAD cut-off. Secondary endpoints include longitudinal biomarker kinetics, early-stage HCC detection rates, and the impact on downstream imaging (CT/MRI) volume. Ultimately, this study seeks to define the role of GAAD as a surveillance adjunct and inform future clinical guidelines for biomarker-enhanced HCC screening.

This is a prospective, non-randomized, open-label, interventional diagnostic utility study Study procedure Upon enrollment, patients will be followed for 2 years. Surveillance visits will take place at every 6 months intervals according to national guidelines and usual clinical practice. Biomarkers will be assessed using routine blood draw taken at each surveillance visit. Hence, no additional blood draw will be required as per study protocol. Blood collection will be conducted according to local's standard procedures. The following parameters will be assessed at baseline and each follow-up visit. * Characteristics (age, sex, height, weight) * Blood test results (platelets, liver and kidney function, PT, viral infection, liver fibrosis markers) * Fib4 * BMI * Child-Pugh Score * AFP, PIVKA, GAAD * Ultrasound (US): liver evaluation, including existing liver steatosis and its degree if available (according to local clinical practice and used guidelines) * Diabetes type (ICD-10) * Dyslipidemia status * Diagnostic of Hypertension * Central obesity status (defined as waist more than 80 cm - women and 90 cm - male) The Elecsys® AFP, and Elecsys® PIVKA-II will be measured using the Elecsys analyzer, and the GAAD score for each case will be calculated using automatic calculation flow navifyTM Algo Suite. * Elecsys® AFP and Elecsys® PIVKA will be tested and interpreted independent of GAAD as per investigators discretion and/or according to local clinical practice. * GAAD - calculate using navify AlgoSuite In case of positive biomarker/score or imaging, result patients will be subjected further diagnostic workup (recall procedure) with multiphasic contrast-enhanced CT or MRI as described below. Due to its better performance, MRI is the preferred modality but CT can be performed alternatively. Recall procedure For MRI/CT, the following protocols must be applied in order to provide optimal diagnostic performance in accordance with local current clinical practice and Guidelines recommendation (AASLD1, EASL2 and APASL3). The recall procedure with multiphasic contrast-enhanced CT or MRI will be triggered for each patient fulfilling one of the following criteria: * Patients with any suspicious lesion ≥ 1 cm on ultrasound should undergo diagnostic evaluation with multiphasic contrast-enhanced CT or MRI * AFP ≥ 20 ng/ml or rising AFP (on two consecutive tests or doubling of AFP levels) * Patients with limited ultrasound visualization may undergo surveillance contrast-enhanced MRI or multiphase CT based on medical evaluation of treating physician in line with local current clinical practice. For patients with new lesion or lesion \<1 cm lesion on abdominal ultrasound will undergo repeat short-interval evaluation (ultrasound, AFP, PIVKA measurements and GAAD score calculation) in approximately 3-4 months . As part of the study, PIVKA II and GAAD score will be performed and patients with GAAD score ≥ 2.57 will be followed up with multiphasic contrast-enhanced CT or MRI as recall procedure. Clinical decisions will be based on local radiologic assessment. Based on multiphasic contrast-enhanced CT or MRI - the diagnosis of HCC will be made in line with current clinical practice: * in case of negative results the normal follow up procedure will continue; * for positive results / HCC confirmation - adequate diagnostic, clinical and treatment decisions will be made in line with current clinical practice (outside of the study) For patients with positive findings in any surveillance modality (refers to US, AFP or GAAD; if PIVKA-II alone is above MDP cut-off, this will not be considered a positive case), the following additional parameters will be documented upon diagnostic workup: * imaging modality and imaging protocol * imaging result * pathology result (if applicable) * LiRADS Score. Right to withdraw Subjects have the right to withdraw from the study at any time for any reason. If lost to follow-up, the assigned study staff will try to contact the subject by telephone followed by registered mail to establish and document as completely as possible the reason for the withdrawal. Subjects will be allowed to miss one scheduled visit. Subjects with \>1 missed visits will be excluded from further follow-up. Subjects will be informed of circumstances under which their participation may be terminated by the responsible investigator without the subject's consent. The investigator may withdraw subjects from the study in the event of intercurrent illness, adverse events, lack of compliance with the study and/or study procedures (e.g., study visits), or any reason where it is felt that it is in the best interest of the subject to be terminated from the study. Any administrative or other reasons for withdrawal will be documented and explained to the subject. If the reason for removal of a subject from the study is an Adverse Event, the principal specific event will be recorded in the medical record. 5\. Statistics Primary and secondary endpoint(s) Primary endpoint Exploratory Clinical performance of Elecsys® AFP, Elecsys® PIVKA-II, GAAD -score for the aid in diagnosis of early and all stages stage Hepatocellular Carcinoma (HCC). Exploratory endpoint * To compare the clinical performance of GAAD algorithm versus US, AFP, PIVKA, as standalone and US +AFP (by the stage and etiology). * To assess the biomarker kinetics of Elecsys AFP, Elecsys PIVKA-II, GAAD leading up to HCC diagnosis in HCC confirmed cases. * Number of additional early-stage HCCs identified through GAAD * HCC prevalence * Missed surveillance visits/drop-out rate due to missed visit * Distribution of GAAD Score among different Child-Pugh subgroups, etiologies and HCC cases and controls * Change in GAAD score over time, in relation to development of HCC and potential confounding factors * Number of additional CT/MRI imaging triggered positive GAAD, AFP or US and calculation of the relative specificity (rFPR) with the corresponding 95% confidence intervals (using Cheng and Macaluso 1997). Statistical methods Primary endpoints: ● Exploratory Clinical performance of Elecsys® AFP, Elecsys® PIVKA-II, GAAD -score for the aid in diagnosis of early and all stages stage Hepatocellular Carcinoma (HCC). - descriptive (PPV, NPV, Sens, Spec AUC/ROC) Exploratory endpoints, * Contingency tables comparing AFP, PIVKA-II, US, and GAAD will be used to investigate all possible combinations (e.g. US vs. GAAD), both for the cases and for the controls. Besides PPV and NPV values for AFP, PIVKA-II, US, and GAAD will be determined as well as ROC curves and AUC values. Time dependent changes in AFP, PIVKA-II, US, and GAAD shall be investigated as an experimental endpoint as well as possible changes in the clinical outcome over the study duration. * Incidence of HCC: Cox-regression analysis, log-rank test, receiver operating characteristic curve analysis. * Association of Elecsys AFP, Elecsys AFP-L3, Elecsys PIVKA-II, GAAD -score with HCC detection: Logistic regression analysis, Spearman correlation * Besides, the prevalence for HCC and the drop-out rate shall be estimated together with the 95% confidence intervals using binomial distributions. Sample size, level of significance, and power Sample size estimation: As HCC has a low annual prevalence (consider 1.5% \& 2% for now), the precision of the sensitivity estimate (i.e., width of the 95% confidence interval) is the main prerequisite for the sample size, as sensitivity is calculated from HCC cases. Assuming a decent target of 20% sensitivity 95% confidence interval width, a true sensitivity of 86.5% (MCE study), and a prevalence of 1.5% \& 2% over two years follow-up, a sample size of 2100 samples would be required to achieve a 95% confidence interval width of 19% (1.5% annual prevalence) and 16% (2% annual prevalence) respectively, which is considered sufficient.