On the Prognostic Value of Gene Expression Signatures for Censored Data

As part of the validation of any statistical model, it is good statistical practice to quantify the amount of prognostic information represented by the model; this includes gene expression signatures derived from high-dimensional microarray data. Several approaches exist for right-censored survival data that measure the gain in prognostic information compared to established clinical parameters or biomarkers in terms of explained variation or explained randomness. They are either model-based or use estimates of the prediction accuracy.