Table 9 Independent cohorts.

From: Accurate prediction of breast cancer survival through coherent voting networks with gene expression profiling

GEO

45255 (ch)

45255 (ho)

45255 (chho)

37181

7390

2034

End point

os

os

os

dfs

os

rfs

Therapy

No-yes-no

No-no-yes

No-yes-yes

No-no-no

No-no-no

Yes-no-no

n.p.

8

16

13

119

181

264

\(>5y\)

3

6

4

59

24

95

\(<5y\)

5

10

9

60

157

169

n.a.

8

16

13

106

179

258

Kappa

1.0

0.58

1.0

0.35

0.36

0.12

Sen.

1.0

0.8

1.0

0.66

0.7

0.90

Spe.

1.0

0.81

1.0

0.70

0.89

0.67

OR

15.0

18.0

36.0

4.59

20.86

20.12

CI-Lo

0.66

1.24

1.77

2.01

4.93

2.53

CI-Hi

339

260

731

10

88.2

159

OR p-val

0.19

0.03

0.01

3.7E−4

3.0E−5

1.8E−4

AUC

1.0

0.89

1.0

0.72

0.70

0.63

AUC p-val

0.01

0.005

0.003

2.6E−5

7.5E−4

1.0E−4

  1. Results of leave-one-out evaluation with optimal multigene fingerprints derived from Metabric data sets. Therapy class: (RAD, CHE, HOR). Endpoint (e.p.) is os overall survival, dfs disease-free survival, rfs relapse-free survival. Confidence interval for odds ratio at 95% confidence. n.a. number of answers.
Back to article page