Table 2 Comparison of clinicopathological features between the training set and the validation set.

From: A risk stratification model to predict chemotherapy benefit in medullary carcinoma of the breast: a population-based SEER database

Characteristics

Training set

Validation set

p value

No. patients

n = 545

n = 136

 

Age at diagnosis (%)

0.537

 < 40

87 (16.0)

19 (14.0)

 

 40–65

349 (64.0)

94 (69.1)

 

 > 65

109 (20.0)

23 (16.9)

 

Race (%)

0.493

 Black

134 (24.6)

40 (29.4)

 

 White

364 (66.8)

84 (61.8)

 

 Other

47 (8.6)

12 (8.8)

 

Marital status (%)

0.338

 Married

288 (52.8)

65 (47.8)

 

 Single

257 (47.2)

71 (52.2)

 

Grade (%)

1

 I–II

33 (6.1)

8 (5.9)

 

 III–IV

512 (93.9)

128 (94.1)

 

T stage (%)

0.391

 T1

238 (43.7)

53 (39.0)

 

 T2

270 (49.5)

76 (55.9)

 

 T3–T4

37 (6.8)

7 (5.1)

 

N stage (%)

0.771

 N0

424 (77.8)

108 (79.4)

 

 N1–N3

121 (22.2)

28 (20.6)

 

ER status (%)

0.134

 Negative

349 (64.0)

97 (71.3)

 

 Positive

196 (36.0)

39 (28.7)

 

PR status (%)

0.481

 Negative

440 (80.7)

114 (83.8)

 

 Positive

105 (19.3)

22 (16.2)

 

HER2 status (%)

0.694

 Negative

484 (88.8)

123 (90.4)

 

 Positive

61 (11.2)

13 (9.6)

 

Subtype (%)

0.156

 HR−/HER2−

292 (53.6)

86 (63.2)

 

 HR−/HER2+ 

38 (7.0)

6 (4.4)

 

 HR+/HER2−

192 (35.2)

37 (27.2)

 

 HR+/HER2+ 

23 (4.2)

7 (5.1)

 

Surgery (%)

1

 No/unknown

11 (2.0)

3 (2.2)

 

 Yes

534 (98.0)

133 (97.8)

 

Radiation (%)

0.443

 No

266 (48.8)

72 (52.9)

 

 Yes

279 (51.2)

64 (47.1)

 

Chemotherapy (%)

0.784

 No/unknown

141 (25.9)

33 (24.3)

 

 Yes

404 (74.1)

103 (75.7)

 
  1. ER estrogen receptor, PR progesterone receptor, HER2 human epidermal growth factor receptor 2, TNBC triple-negative breast cancer.