Table 1 Characteristics of the study patients.

From: Prediction of pathologic complete response to neoadjuvant systemic therapy in triple negative breast cancer using deep learning on multiparametric MRI

 

Development group

Testing group (R)

p1

Testing group (P)

p2

No. of patients

130

32

 

48

 

Mean age ± SD (y) at diagnosis

48 ± 10

51 ± 13

0.21

50 ± 11

0.31

Overall clinical stage

I

13

4

0.91

10

0.15

II

99

24

33

III

18

4

5

Tumor stage

T1

21

5

–

12

–

T2

92

25

32

T3

14

2

4

T4

3

0

0

Nodal stage

N0

88

23

0.50

34

–

N1

29

4

11

N2

6

3

0

N3

7

2

3

Median stromal tumor infiltrating lymphocytes (IQR)

10 (26)

10 (16)

0.48

10 (23)

0.96

Median Ki-67 (IQR)

75 (25)

75 (40)

0.47

75 (40)

0.81

Histology type

Invasive ductal

119

28

0.51

45

–

Metaplastic

8

2

3

Other

3

2

0

pCR status

pCR

65

14

0.53

22

0.62

Non-pCR

65

18

26

  1. Testing group (R): retrospective independent testing group. Testing group (P): prospective blinded testing group. p1: p-value of the comparison between the development and the retrospective independent testing groups. p2: p-value of the comparison between the development and the prospective blinded testing groups. The groups’ ages were compared using a t-test; stromal tumor infiltrating lymphocytes and Ki-67 were compared using the nonparametric Mann–Whitney U test; and other non-continuous variables were compared using the chi-squared test.
  2. SD standard deviation, IQR interquartile range, pCR pathologic complete response.
Back to article page