Clinical Utility and Diagnostic Accuracy of the GAAD Score in Non-Cirrhotic Patients With Focal Liver Lesions

BACKGROUND & AIMS: Focal liver lesions are frequently encountered, often as incidentalomas, in patients without cirrhosis. There is an urgent unmet need for noninvasive tests to distinguish hepatocellular carcinoma (HCC) from other lesions. We studied the diagnostic accuracy of the GAAD score, which is based on patient age, sex and serum values of α-fetoprotein (AFP) and protein induced by vitamin K absence-II (PIVKA-II; also known as Des-γ-carboxyprothrombin), in non-cirrhotic patients with focal liver lesions. METHODS: In a multi-center prospective study, all consecutive patients presenting with focal liver lesions in a non-cirrhotic liver were enrolled from 2 tertiary referral clinics in the Netherlands. The diagnostic accuracy of the GAAD score was assessed at its predefined cutoff of ≥2.57 and compared with AFP and PIVKA-II alone in the overall study population and after stratification by lesion size and presence of underlying liver disease. RESULTS: A total of 583 patients were included, of whom 73 (12.5%) had HCC. The GAAD score offered superior discrimination over AFP (area under the receiver operating characteristic curve [AUC] 0.96 vs 0.75; P < .01) and PIVKA-II (AUC 0.96 vs 0.91; P = .01). At a GAAD cutoff of ≥2.57, the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were 85%, 96%, 75%, and 98%, with GAAD offering superior specificity over PIVKA-II and superior sensitivity over AFP. The performance of GAAD was consistent regardless of the presence of an underlying chronic liver disease (AUC 0.93 vs 0.97; P = .23), and regardless of the presence of viral hepatitis (AUC 0.99 vs 0.96; P = .05) or steatotic liver disease (AUC 0.92 vs 0.98; P = .19). CONCLUSIONS: In this cohort of non-cirrhotic patients with focal liver lesions, the GAAD score was highly predictive of HCC and offered superior diagnostic accuracy over AFP and PIVKA-II alone. These findings support the use of the GAAD score to guide management decisions in these patients.