Seminário UFSCar/USP - 08/04/2011

LOCAL: Sala de seminários ? DEs-UFSCar

Título: A New Cure Rate Survival Model Under A Hybrid Latent Activation Scheme

Patrick Borges (Doutorado)

Resumo:

In lifetimes studies, the occurrence of an event (such as tumor detection or death) might be caused by one, among many, competing causes. Furthemore, both the number of causes and the time-to-event associated with each cause may not be observed. The number of causes can be zero, corresponding a cure fraction. In this paper, we propose a method of estimating the numerical characteristics of unobservable stages (such as initiation, promotion and progression) of carcinogenesis from data on tumor size at detection in the presence of latent competing causes. To this end, a general survival model for spontaneous carcinogeneses under a hybrid latent activation scheme has been developed to allow for a simple pattern of the dynamics of tumor growth. It is assumed that a tumor becomes detectable when its size attains some threshold level (proliferation of malignant cells), which is treated as a random variable. We assume the number of initiated cells (competing causes of initiation stage) and the number of malignant cells (competing causes of progression stage) to follow weighted Poisson distributions. The advantage of this model is that it incorporates into the analysis characteristics of the stage of tumor progression as well as the proportion of initiated cells that die before tumor induction. Parameter estimation of the proposed model is discussed by means of maximum likelihood estimation method. Finally, we illustrate the usefulness of the proposed model by applying it to a real data on melanoma.

Seminário UFSCar/USP - 08/04/2011 - 14h00