Meta-analysis of predictive models to assess the clinical validity and utility for patient-centered medical decision making: application to the CAncer of the Prostate Risk Assessment (CAPRA)

Background

The Cancer of the Prostate Risk Assessment (CAPRA) was designed and validated several times to predict the biochemical recurrence-free survival after a radical prostatectomy. Our objective was to study the usefulness of the CAPRA score for stratified medicine.

Methods

We proposed a meta-analysis based a literature search using MEDLINE. Pooled survival curves per CAPRA-based strata were estimated by the distribution-free approach with random effects proposed by Combescure, Foucher and Jackson. The discrimination capacities at 5-years post-RP were evaluated by using time-dependent summary ROC (SROCt) curves as proposed by Combescure, Daurès and Foucher.

Data extraction

The previous estimations were based on the following data. The survival probabilities were extracted from a digitalized picture by using the R packages ReadImages and digitize. Many papers do not provide the number of at-risk patients over the time, so we applied the method proposed by Parmar et al. in order to obtain estimations. The obtained data van be downloaded here