Skip to main content
Figure 2 | Breast Cancer Research

Figure 2

From: Benefits of biomarker selection and clinico-pathological covariate inclusion in breast cancer prognostic models

Figure 2

Performance, model size distribution and variable stability of reduced models as described in Figure 1 for lower NPI and higher NPI risk groups. Left column: Patients with an Nottingham Prognostic Index (NPI) of more than 4.4. Right column: NPI of 4.4 or less. The final reduced model (RM) for the lower NPI group consists of 7 variables, whereas the final reduced model for the higher NPI group consists of 11 partially overlapping variables. For example, CCNB1 is one of the most robust variables in the higher NPI group, but is the least robust variable in the lower NPI group. ATD-AUCROC, average time-dependent area under the receiver operator characteristic curve; ER, estrogen receptor; FM, full models; HER, human epidermal growth factor receptor; PR, progesterone receptor.

Back to article page