Building a Prognostic Model for Breast Cancer Survival
John Newell (National University of Ireland, Galway)
Friday 14th March, 2014 15:00-16:00 Maths 204
In this talk I will present results and ideas arising from a project where the primary aim was to build a prognostic model for modelling time to disease recurrence and death for invasive breast cancer patients. Follow up time, patient characteristics and genetic markers were collected retrospectively from 647 women in the West of Ireland.
Initially I will present extensions to plots of the Kaplan Meier estimator and will then introduce a novel new semi-parametric estimator for the under used summary in Survival Analysis namely the Mean Residual Life (MRL) function. The MRL, which at time t measures the expected remaining lifetime of an individual having survived up to time t, provides a clear and simple summary of the effect of a treatment or a risk factor in units of time, avoiding hazard ratios or probability scales which require careful interpretation.
An additional complexity exists when building prognostic models due to the presence of missing data; in the example presented a complete case analysis with both clinical and pathological biomarkers reduces the number of cases to 103 patients. The results of a simulation study, used to compare the performance of variable selection techniques used on imputed data compared to the benefits of using penalisations such as the LASSO or Ridge regression will be given.
The proposed model will be presented as a dynamic nomogram using the Shiny package in R.