Skip to main content

Table 2 Characteristics of the 16 cancers selected. C-indices are computed with 10 repetitions of 5-fold cross-validation with the ridge regression and all of the genes in the Cox model. Datasets are ordered according to their median C-index (decreasing order)

From: Prognosis of lasso-like penalized Cox models with tumor profiling improves prediction over clinical data alone and benefits from bi-dimensional pre-screening

Cancer

n events

n patients

Censoring rate

Survival - 3 years

C-index

ACC

27

77

0.65

0.75

0.86

KIRP

42

269

0.84

0.87

0.83

UVM

21

77

0.73

0.74

0.79

MESO

73

85

0.14

0.19

0.75

KIRC

175

526

0.67

0.76

0.73

LGG

192

510

0.62

0.56

0.73

CESC

70

288

0.76

0.72

0.7

PRAD

92

490

0.81

0.8

0.69

LIHC

125

358

0.65

0.62

0.67

UCEC

91

541

0.83

0.83

0.67

BLCA

165

371

0.56

0.48

0.65

LAML

92

149

0.38

0.31

0.65

BRCA

145

1079

0.87

0.88

0.64

HNSC

210

491

0.57

0.57

0.64

PAAD

93

175

0.47

0.34

0.63

LUAD

179

488

0.63

0.61

0.62

Back to article page

BMC Cancer

ISSN: 1471-2407

Contact us