Lectures and seminars PhD student seminar: Fabrizio Di Mari

Abstract

When information on the cause of death is unavailable or unreliable, such as in many population-based studies, the Relative Survival (RS) framework is the preferred approach for estimating net survival, which represents survival in a hypothetical scenario where the disease of interest is the only possible cause of death. In the context of cancer, net survival often reaches a plateau, indicating that a portion of diagnosed patients is cured, as they face the same risk of dying as a comparable group of healthy individuals with similar demographic characteristics. Most RS mixture cure models estimate the fraction of cured patients using logistic regression. However, this functional form is somewhat arbitrary, and misspecification can severely distort the resulting cure estimates.

To address this, we propose using flexible functions of the covariates within the framework of Generalized Additive Models and Neural Networks. We design an EM algorithm for these RS cure models and conduct a simulation study to compare our approach with the classical RS Weibull mixture cure model. Finally, we apply our methodology to a real-world dataset from a historical Italian cancer registry, providing insights into the survival outcomes of Italian colon cancer patients.

Joint work with:

Roberto Rocci, Sapienza University of Rome, Rome, Italy

Roberta De Angelis, National Institute of Health, Rome, Italy

About the speaker

Fabrizio Di Mari is visiting MEB from Rome, where he is a PhD student in Methodological Statistics at Sapienza University of Rome, and his research focuses on developing new methodologies for cure and prevalence estimation in population-based cancer studies.