In this talk, we examine design and analysis issues in cure rate models for cancer clinical trials. We demonstrate the use of these models from studies conducted by the Eastern Cooperative Oncology Group and demonstrate their superiority and usefulness over more traditional approaches to design and analysis of clinical trials. Both Bayesian and frequentist methods are discussed and several examples are given to demonstrate these models in a variety of contexts, including modeling univariate survival data, joint modeling of longitudinal and survival data, modeling correlated failure time data, and their use in transformation models for survival data.