Disease-free survival: (non-)parametric estimation
| Authors | |
|---|---|
| Year of publication | 2011 |
| Type | Conference abstract |
| MU Faculty or unit | |
| Citation | |
| Description | Treatment efficacy in patients with a disease uses to beexpressed by the concept of disease-free survival, i.e. the probability of staying in a remission after its achievement or after a therapeutic intervention. However, this concept does not allow to evaluate the proportion of disease-free patients in subsequent remission after further possible relapses. The method proposed by Klein et al. enables to estimate the probability of being in first and second remissions. The aim of the presentation is to present two new methods of estimation the probability of being in any of remissions. The first one extends the non-parametric estimation proposed by Klein et al. that is based on Kaplan-Meier estimators of survival functions. The second one utilizes a multistate model and it adopts the Wood method for matrix model parameters identification based on quadratic programming to estimate probabilities of remissions and relapses of any rank. The methods are illustrated on data of CML patients. |