Screening based on the risk of cancer calculation from Bayesian hierarchical changepoint and mixture models of longitudinal markers

The standard approach to early detection of disease with a quantitative mar ker is to set a population-based fixed reference level for making further i ndividual screening or referral decisions. For many types of disease, inclu ding prostate and ovarian cancer, additional information is contained in th e subject-specific temporal behavior of the marker, which exhibits a charac teristic alteration early in the course of the disease. In this article we derive a Bayesian approach to screening based on calculation of the posteri or probability of disease given longitudinal marker levels. The method is m otivated by a randomized ovarian cancer screening trial in the United Kingd om comprising 22,000 women screened over 4 years with an additional 5 years of follow-up on average. Levels of the antigen CA125 were recorded annuall y in the screened arm. CA125 profiles of cases and controls from the U.K. t rial are modeled using hierarchical changepoint and mixture models, posteri or distributions are calculated using Markov chain Monte Carlo methods, and the model is used to calculate the Bayesian posterior risk of having ovari an cancer given a new subject's single or multiple longitudinal CA125 level s. A screening strategy based on the risk calculation is then evaluated usi ng data from an independent screening trial of 5,550 women performed in Swe den. A longitudinal CA125 screening strategy based on calculation of the ri sk of ovarian cancer is proposed. Simulations of a prospective trial using a strategy based on the risk calculated from longitudinal CA125 values indi cate potentially large increases in sensitivity for a given specificity com pared to the standard approach based on a fixed CA125 reference level for a ll subjects.