Development of a prediction model for 10-year risk of hepatocellular carcinoma in middle-aged Japanese: the Japan Public Health Center-based Prospective Study Cohort II

Abstract

Objective: The purpose of the present study was to develop a risk estimation model for the 10-year risk of hepatocellular carcinoma (HCC) that could be easily used in a general population to aid in the prevention of HCC.

Methods: Our prediction model was derived from data obtained on 17,654 Japanese aged 40 to 69 years who participated in health checkups (follow-up: 1993-2006). Cox proportional hazards regression was applied to obtain coefficients for each predictor.

Results: During follow-up, a total of 104 cases of HCC were newly diagnosed. After checking the model fit, we incorporated age, sex, alcohol consumption, body mass index, diabetes, coffee consumption, and hepatitis B and C virus infection into the prediction model. The model showed satisfactory discrimination (Harrell's c-index=0.94) and was well calibrated (the overall observed/expected ratio=1.03, 95% confidence interval=0.83-1.29). We also developed a simple risk scoring system. Those subjects with total scores of 17 or more under this system (score range: -1 to 19) had an estimated 10-year HCC risk of over 90%; those with 4 points or less had an estimated risk of less than 0.1%.

Conclusion: We developed a simple 10-year risk prediction model for HCC in the Japanese general population as a public education tool.