Abstract:4577 Background: Hepatocellular carcinoma (HCC) is a leading cause of cancer-related deaths worldwide. Though biannual ultrasound surveillance with or without alpha-fetoprotein (AFP) testing is recommended for at-risk patients, its sensitivity for early-stage HCC detection is suboptimal. We therefore evaluated performance of a biomarker panel incorporating methylated DNA markers (MDMs) and proteins for early HCC detection in at-risk patients with chronic liver disease. Methods: In an international, multicenter, case-control study, blood specimens were collected from patients with HCC per AASLD criteria and controls matched for age and liver disease etiology. All patients had underlying cirrhosis or chronic HBV infection. Whole blood was collected in cell-free DNA stabilizing and serum-separation tubes and shipped to a central laboratory for processing. The levels of 5 MDMs, AFP, and AFP-L3 were assessed along with age and sex. We used 537 samples in a 5-fold validation for developing a LASSO regression algorithm to classify samples as HCC positive or negative. Model robustness was tested by perturbing the data in silico and analyzing results with the predictive algorithm. Algorithm performance was compared to AFP alone and the GALAD score (Gender, Age, AFP-L3, AFP, and DCP). Results: The study included 136 HCC cases (81 early-stage—BCLC stage 0/A) and 401 controls. With specificity set at 89%, we developed a model using sex, AFP, and 3 MDMs (HOXA1, TSPYL5, B3GALT6) with higher sensitivity (70%) for early-stage HCC compared to GALAD (54%) or AFP (31% at 20 ng/mL or 52% at ≥7.7 ng/mL) (Table). The AUC for the HCC marker panel was 0.91 (95% CI 0.89 – 0.94) compared to GALAD (0.88; 95% CI 0.85 – 0.91) or AFP (0.84; 95% CI 0.81 – 0.87). The panel performed similarly in viral (AUC = 0.94) and non-viral (AUC = 0.89) etiologies. Conclusions: The robust algorithm based on novel blood-based biomarkers presented here provides higher sensitivity for early-stage HCC compared to other available blood-based biomarkers and, therefore, could significantly impact HCC clinical management and patient outcomes. Further clinical studies to validate the algorithm are ongoing. Clinical trial information: NCT03628651 . Performance of HCC marker panel. Sensitivity % (95% CI) Specificity % (95% CI) All-Stage AUC (95% CI) Early-Stage All-Stage Current HCC marker panel 70 (60 – 80) 81 (74 – 87) 89 (86 – 92) 0.91 (0.89 – 0.94) GALAD 54 (43 – 65) 68 (59 – 75) 89 (86 – 92) 0.88 (0.85 – 0.91) AFP (≥7.7 ng/mL) 52 (41 – 63) 64 (56 – 72) 89 (86 – 92) 0.84 (0.81 – 0.87) AFP (20 ng/mL) 31 (21 – 42) 46 (38 – 55) 98 (96 – 99)

Comparison between models for detecting hepatocellular carcinoma in patients with chronic liver diseases of various etiologies: ASAP score versus GALAD score

Diagnostic Performance Comparisons of Two Commonly Used Multi-Biomarker-based Scores for Detection of Hepatocellular Carcinoma in Non-Alcoholic Fatty Liver Disease

ASAP Score versus GALAD Score for detection of hepatitis C-related hepatocellular carcinoma: A multicenter case-control analysis

Validation of the GALAD Model and Establishment of GAAP Model for Diagnosis of Hepatocellular Carcinoma in Chinese Patients

Validation of the GALAD model for early diagnosis and monitoring of hepatocellular carcinoma in Chinese multicenter study

Comparison of diagnostic performance of AFP, DCP and two diagnostic models in hepatocellular carcinoma: a retrospective study

The Performance of GALAD Score for Diagnosing Hepatocellular Carcinoma in Patients with Chronic Liver Diseases: A Systematic Review and Meta-Analysis.

The GALAD scoring algorithm based on AFP, AFP-L3, and DCP significantly improves detection of BCLC early stage hepatocellular carcinoma

Meta-analysis of the GALAD model for diagnosing primary hepatocellular carcinoma

Algorithm for blood-based panel of methylated DNA and protein markers to detect early-stage hepatocellular carcinoma with high specificity.

Performance of GALAD score and serum biomarkers for detecting NAFLD-related HCC: a systematic review and network meta-analysis

Role of GALAD serological model in the clinical diagnosis of primary hepatocellular carcinoma

O-10 VALIDATION OF SERUM BIOMARKER PANELS FOR EARLY HCC DETECTION: RESULTS FROM A LARGE PROSPECTIVE LATIN AMERICAN MULTICENTER STUDY

Performance Comparisons of Two Biomarker-Based Panels for Early Diagnosis of Hepatocellular Carcinoma in Nonalcoholic Fatty Liver Disease: A Multi-Institutional Case-Control Analysis

Diagnostic and prognostic performances of GALAD score in staging and 1-year mortality of hepatocellular carcinoma: A prospective study

Diagnostic Performance of AFP, AFP-L3, or PIVKA-II for Hepatitis C Virus-Associated Hepatocellular Carcinoma: A Multicenter Analysis

The Combination of Serum Oligosaccharide Chain (G-Test), Alpha-Fetoprotein, and Aspartate Aminotransferase to Alanine Aminotransferase Ratio Provides the Optimal Diagnostic Value for Early Detection of Hepatocellular Carcinoma

Multimarker Panels for Detection of Early Stage Hepatocellular Carcinoma: A Prospective, Multicenter, Case‐Control Study

Validation of models in patients with hepatocellular carcinoma: comparison of Hong Kong Liver Cancer with Barcelona Clinic Liver Cancer staging system in a Chinese cohort.

A Phase 3 Biomarker Validation of GALAD for the Detection of Hepatocellular Carcinoma in Cirrhosis