Title: A Bayesian screening approach for hepatocellular carcinoma using two longitudinal biomarkers
Authors: Nabihah Tayob - The University of Texas MD Anderson Cancer Center (United States) [presenting]
Francesco Stingo - University of Florence (Italy)
Kim-Anh Do - The University of Texas MD Anderson Cancer Center (United States)
Ziding Feng - The University of Texas MD Anderson Cancer Center (United States)
Anna Lok - University of Michigan (United States)
Abstract: Advanced hepatocellular carcinoma (HCC) has limited treatment options and poor survival. Early detection of HCC is critical to improve the prognosis of these patients. Current guidelines for high-risk patients include six-month ultrasound screenings but these are not sensitive for early HCC. Alpha-fetoprotein (AFP) is a widely used diagnostic biomarker but has shown limited use in HCC screening with a fixed threshold. Approaches that incorporate longitudinal AFP have shown potentially increased detection of HCC however AFP is not elevated in all HCC cases. We incorporate a second HCC biomarker, des-gamma-carboxy prothrombin (DCP). The Hepatitis C Antiviral Long-term Treatment against Cirrhosis (HALT-C) Trial is a valuable source of data to study biomarker screening. We assume the trajectories of AFP and DCP follow a joint hierarchical mixture model with random change points. Markov chain Monte Carlo methods are used to estimate the posterior distributions used in risk calculations among future patients. The posterior risk of HCC, given longitudinal values of AFP and DCP, is used to determine whether a patient has a positive screen. The screening algorithm was compared to alternatives in the HALT-C Trial (using cross-validation) and in simulations studies under a variety of possible scenarios.