Estimating Progression-Free Survival When the Progression Status of Some Patients Is Unknown

• Activity Number: 336
• Type: Topic Contributed
• Date/Time: Tuesday, August 3, 2010 : 10:30 AM to 12:20 PM
• Sponsor: ENAR
• Author(s): Ying Yuan*+ and Peter F. Thall and Johannes Wolff
• Companies: MD Anderson Cancer Center and MD Anderson Cancer Center and MD Anderson Cancer Center
• Keywords: Imputation ; Latent variables ; Missing values ; Missing at random ; Survival analysis

Abstract

In oncology and many other disease areas progression-free survival time, defined as the minimum of the times to disease progression and death, frequently is used to characterize the effects treatments and covariates. For diseases where progression always precedes death, the problem of estimating the progression time distribution becomes more complicated if, for some subjects, either the survival time is known but the progression time is not known, or survival time is right censored and it is not known whether the subject's disease progressed prior to the censoring time. For data having these missingship structures, we formulate a family of parametric likelihood functions and present methods for estimating the progression time distribution. We illustrate the proposed methodology using a pediatric brain tumor dataset.