-
3
-
Korean J Hepatol. 2006;12(3):404-411. Published online January 1, 2000.
- Optimal Cut-off Value of PIVKA-II for Diagnosis of Hepatocellular Carcinoma - Using ROC Curve -
-
-
- Abstract
- Background/Aims
Protein induced by vitamin K absence or antagonist-II (PIVKA-II), also known as des-carboxyprothrombin (DCP), can be used as an alternative tool to alpha-fetoprotein (AFP) for surveillance of hepatocellular carcinoma (HCC). The aims of the present study were to compare PIVKA- II levels between the patients with HCC and patients with non-HCC chronic liver disease, to evaluate the correlation of PIVKA-II and AFP in HCC patients, and finally to estimate the optimal cut-off value for PIVKA-II for the diagnosis of HCC with using the receiver operating characteristic (ROC) curve. Methods: A total of 227 consecutive patients with HCC (n=42) or chronic liver disease (n=185) were enrolled in this study. HCC was diagnosed histologically or by imaging such as computed tomography, magnetic resonance imaging or angiography. The serum PIVKA-II and AFP levels were measured by electrochemiluminoimmunoassay with using the Haicatch PIVKA-II kit and by immunoradiometric assay, respectively. Results: The PIVKA-II level in the HCC patients was significantly higher than the non-HCC chronic liver disease patients (903.0±1156.7 vs. 111.7±211.0 mAU/ mL, respectively, P<0.01). PIVKA-II and AFP showed a statistical correlation in HCC patients (r=0.46, P<0.01). The sensitivity and specificity of PIVKA-II for the diagnosis of HCC were 66.7% and 74.1%, respectively, and when tasted together with AFP, the sensitivity was increased by 85.7%. For the ROC curve of PIVKA-II in HCC patients, the specificity of a 250 mAU/mL level of PIVKA-II was 95%. Conclusions: PIVKA-II was as useful surveillance tool for differentiating HCC from chronic liver disease, and a PIVKA-II value of 250 mAU/ mL was proposed as a significant cut-off value for diagnosis of hepatocellular carcinoma.
Keywords :Hepatocellular carcinoma; PIVKA-II; alpha-Fetoproteins; ROC curve; Diagnosis
