Prediction of development of hepatocellular carcinoma using a new scoring system involving virtual touch...

This study aimed to establish a new scoring system that combined several risk factors, including virtual touch quantification (VTQ) values and fasting plasma glucose (FPG) levels, for predicting the development of hepatocellular carcinoma (HCC) in patients with chronic liver disease.

A total of 1808 chronic liver disease patients who underwent VTQ measurement were analyzed. Risk factors for developing HCC were selected by multivariate Cox proportional hazards models.

VTQ (>1.33 m/s), FPG (≥110 mg/dl), sex (male), age (≥55 years), and α-fetoprotein (AFP) level (≥5 ng/ml) were independently selected as risk factors for HCC development by multivariate analysis. Using these parameters, we established a new scoring system (0 to 5 points), based on VTQ, FPG, sex, age, and AFP level, named VFMAP. As compared with the low VFMAP score group (0 or 1 point), the hazard ratio for the incidence of HCC was 17.37 [95 % confidence interval (CI), 2.35-128.40] in the intermediate-score group (2 or 3 points) and 66.82 (95 % CI, 9.01-495.80) in the high-score group (4 or 5 points). The area under the receiver operating characteristic curve of the VFMAP score for predicting HCC development within 5 years was 0.82 (95 % CI, 0.76-0.87), indicating a moderate diagnostic value. A VFMAP cutoff value of 3 excluded HCC within 5 years with a high negative predictive value (98.2 %).

The VFMAP score accurately predicted HCC in patients with chronic liver disease.