HCC Risk Scores: Useful or Not?

Abstract

The advent and efficacy of surveillance for hepatocellular carcinoma (HCC) has necessitated the refinement of assessing who is at risk for this cancer. Initially, risk was assessed for all individuals with hepatitis B and all those with cirrhosis. However, the majority of these individuals do not develop HCC so that providing surveillance for all is a waste of resources. There are now many different scores that have been developed that allow better identification of who is at risk and who is not. Specific models have been developed for hepatitis B before and on treatment, for hepatitis C before and after treatment, and for cirrhosis in general. There are also models for assessing risk in the general population. Some models can only be applied to patients coming from the population in which the score was developed (e.g., hepatitis B). Others are more generalizable. Many lack external validation. With some exceptions, the models do not attempt to assess the score at which surveillance should start. Overall, the models provide some useful guidance as to who does not need to undergo surveillance, but the long-term performance and how changes in risk score correlate with changes in HCC risk has not been completely assessed.