 |
|
|||
Estimating the Cure Fraction among Cancer Patients by using Promotional Time Cure Rate Model with Negative Binomial Distribution
Keywords
Long-term survivors, Negative binomial distribution, Latent variables, Promotional Time Cure Model, Bayesian Approach
Abstract
Modern cancer treatments have substantially improved cure rates and have generated a great interest and need for proper statistical tools to analyze survival data with non-negligible cure fractions. However, the patient's death, which is the event of interest, might happen due to different latent competing causes which can develop cancer. In this paper, a flexible cure rate survival model (Promotional time cure model) has been proposed by assuming that the number of competing causes follows the Negative Binomial distribution and the time to event follows Exponential distribution. Parameter estimation has been done by Bayesian approach to a real dataset of melanoma clinical trial using MCMC technique. Model comparison has been accessed using Deviance information Criterion.