Skip to main content

Table 1 Demographics of the PDxBr training and validation cohorts

From: Development and validation of an AI-enabled digital breast cancer assay to predict early-stage breast cancer recurrence within 6 years

 

Train

(N = 1559)

Validation

(N = 516)

Median age, y (range)

60 (24, 90)

60 (28, 90)

Race/ethnicity, N (%)

 Asian

9 (0.6)

6 (1)

 Black

81 (5)

21 (4)

 Latino

22 (1)

8 (1)

 Other

186 (12)

69 (13)

 Unknown

408 (26)

128 (25)

 White

853 (55)

284 (55)

Estrogen receptor, N (%)

 0

204 (13)

68 (13)

 1

1355 (87)

448 (87)

Progesterone receptor, N (%)

 0

292 (19)

100 (19)

 1

1267 (81)

416 (81)

HER2, N (%)

  

 0

1362 (87)

461 (89)

 1

197 (13)

55 (11)

Tumor size (cm)

1.5 ± 1.1 (0.1, 17.0)

1.5 ± 0.9 [0.1, 8.0]

 T1

1168 (75)

398 (77)

 T2

367 (24)

112 (22)

 T3

24 (2)

6 (1)

Stage

 Stage1

1055 (68)

362 (70)

 Stage 2

386 (25)

123 (24)

 Stage IIIA/B

81 (5)

21 (4)

 Stage IIC

36 (2)

10 (2)

 Stage IV

1 (0.1)

0 (0)

Lymph node status

 posLN = 0

1075 (69)

350 (68)

 microLN or isolatedLN and posLN = 0

127 (8)

56 (11)

 1 ≤ posLN ≤ 3

239 (15)

79 (15)

 posLN > 3

118 (8)

31 (6)

Grade

 1

290 (19)

68 (13)

 2

649 (42)

219 (42)

 3

620 (40)

229 (44)

Total events

 0

1339 (86)

447 (87)

 1

220 (14)

69 (13)

Time to event (months)

75.3 [-16.0, 68.0, 200.0]

79.4 [0.0, 69.0, 1173.0]

  1. LN Lymph node