Skip to main content

Table 1 Baseline characteristics of patients in training cohort and validation cohort

From: From pixels to patient care: deep learning-enabled pathomics signature offers precise outcome predictions for immunotherapy in esophageal squamous cell cancer

Patient characteristics

Training Cohort (N = 120)

Validation Cohort (N = 43)

P value

Age (Median)

61.21

60.35

0.582

Gender

   

 Male

108 (90.0%)

34 (79.1%)

0.066

 Female

12 (10.0%)

9 (20.9%)

 

Smoking history

   

 No

57 (47.5%)

25 (58.1%)

0.231

 Yes

63 (52.5%)

18 (41.9%)

 

Drinking history

   

 No

72 (60.0%)

23 (53.5%)

0.457

 Yes

48 (40.0%)

20 (46.5%)

 

 T stage

   

 T1-T2

29 (24.2%)

5 (11.6%)

0.083

 T3-T4

91 (75.8%)

38 (88.4%)

 

N stage

   

 N0-N1

59 (49.2%)

17 (39.5%)

0.277

 N2-N3

61 (50.8%)

26 (60.5%)

 

Stage

   

 III

31 (25.8%)

13 (30.2%)

0.577

 IV

89 (74.2%)

30 (69.8%)

 

Lung metastasis

   

 No

101 (84.2%)

34 (79.1%)

0.447

 Yes

19 (15.8%)

9 (20.9%)

 

Brain metastasis

   

 No

118 (98.3%)

43 (100.0%)

 > 0.999

 Yes

2 (1.7%)

0 (0%)

 

Bone metastasis

   

 No

110 (91.7%)

41 (95.3%)

0.734

 Yes

10 (8.3%)

2 (4.7%)

 

Liver metastasis

   

 No

104 (86.7%)

38 (88.4%)

0.775

 Yes

16 (13.3%)

5 (11.6%)

 

Radiotherapy

   

 No

32 (26.7%)

12 (27.9%)

0.875

 Yes

88 (73.3%)

31 (72.1%)

 

Chemotherapy

   

 No

23 (19.2%)

12 (27.9%)

0.231

 Yes

97 (80.8%)

31 (72.1%)

 

PD-L1

   

  < 57.5%

26 (63.4%)

34 (79.1%)

0.112

  ≥ 57.5%

15 (36.6%)

9 (20.9%)

 

ESCC-PS

   

 ESCC-PS 1

32 (26.7%)

19 (44.2%)

0.071

 ESCC-PS 2

57 (47.5%)

13 (30.2%)

 

 ESCC-PS 3

31 (25.8%)

11 (25.6%)