Diagnostic accuracy of ultrasound and alpha-fetoprotein measurement for hepatocellular carcinoma surveillance: a retrospective comparative study

Abstract Background Liver cirrhosis and chronic infection with hepatitis B virus (HBV) and hepatitis C virus (HCV) are major risk factors for hepatocellular carcinoma (HCC). International societies guidelines recommend HCC surveillance in patients with cirrhosis and chronic hepatitis. However, the ideal surveillance strategy is still controversial. The aim of this study is to assess and compare the diagnostic accuracy of ultrasound (US) AND alpha-fetoprotein (AFP) measurement in HCC surveillance. Results All patients with cirrhosis and/or infection with HBV and HVC, who received HCC surveillance through US and AFP measurement between January 2013 and January 2016, were retrospectively evaluated. The performance of surveillance using AFP, US, and both in HCC detection was compared. After exclusion, a final cohort of 335 patients were included. Thirty-five patients (10.3%) developed HCCs. US showed no focal lesions in 259 patients, 9 benign lesions (3 cysts, 5 hemangiomas, and 1 focal fat), 16 HCCs and 51 indeterminate findings. All indeterminate findings were worked up by CT and/or MRI. For HCC detection, the area under the receiver operator characteristic (ROC) curve for US surveillance was 0.924 (95% confidence interval: 0.866–0.983, P < 0.001). For surveillance, AFP was 0.829 (95% confidence interval: 0.756– 0.902, P < 0.001) and for combined US and AFP was 0.897 (95% confidence interval: 0.854– 0.941, P < 0.001). When the traditional cutoff value of 20 ng/ml was used, the sensitivity and specificity of AFP were 63% and 88.7%, respectively. ROC curve results for AFP levels demonstrated that using 5.6 ng/ml as a cutoff, AFP will have a 77% sensitivity and 78% specificity for HCC detection. US exhibited a sensitivity and specificity of 45.7% and 100%, respectively. When indeterminate findings were considered as positive surveillance test, US exhibited a sensitivity and specificity of 91.4% and 88.3%, respectively. A combination of US and AFP exhibited a sensitivity and specificity of 97% and 87.3%, respectively. Conclusions US has better accuracy compared to AFP in HCC surveillance. Combined use of US and AFP improves the sensitivity for HCC detection.