Figure 5

Boxplots of C-index values for the 1000 test data sets. Predictions were made using Cox regression models fitted to each training set. Models were derived from either gene expression (GE) data only, clinical data only, or results from both gene expression and clinical data. The green line shown on each panel represents the C-index corresponding to the 10 year Adjuvant! Online risk scores calculated for both disease mortality and disease recurrence, respectively. Clinical data included for both outcomes were patient stage of disease at diagnosis (1, 2, and 3 or 4), ER status (+/-), and PR status (+/-). Genes included in both the OS and DFS models were PGR, GABRP, TBC1D9, SLC39A6 and LRBA, while NAT1 was also included in the model for DFS.