Background: Hepatocellular carcinoma (HCC) is a cancer with poor prognosis. Indonesia is a country with high prevalence of chronic hepatitis B infection. The performance of alpha fetoprotein (AFP) as a tumor marker in HCC surveillance is primarily influenced by the etiology of the underlying liver disease. We aimed to determine the best cut-off value of AFP biomarker examination for HCC surveillance in patients with chronic hepatitis B infection.
Methods: The study collected medical record data of the Hepatobiliary Division of Dr. Cipto Mangunkusumo Hospital from the period of 2017 to 2023. A total of 506 subjects with chronic hepatitis B of all spectrums (hepatitis B without cirrhosis, liver cirrhosis, and early-stage HCC, BCLC 0 and A) were included by total sampling that was performed from 26 July 2023 to 31 August 2023. Determination of the AFP cut-off value was carried out using the receiver operating characteristics (ROC) method. Results: For HCC surveillance caused by hepatitis B virus, ROC curve analysis resulted in an area under the curve (AUC) of 0.792 (95% CI, 0.719-0.866), and the cut-off value with the highest Youden index was 8.7 ng/ml, with 58% sensitivity, 94% specificity, positive predictive value (PPV) 56.14%, negative predictive value (NPV) 94.43%, positive likelihood ratio (LR+) 10.08, and negative likelihood ratio (LR-) 0.46.
Conclusion: The cut-off value of AFP in HCC surveillance on hepatitis B specific etiology is lower than the cut-off value of AFP in previous HCC surveillance which was not etiologically specific. The cut-off value of 8.7 ng/ml produces the best sensitivity and specificity for the cut-off value for HCC surveillance with hepatitis B etiology.
Keywords: AFP; chronic hepatitis B; cut-off value; hepatocellular carcinoma; surveillance.