Skip to main content

Table 1 Clinical and pathologic characteristics of patients with HNSCC in the training, Internal validation, and external validation cohorts

From: Development and validation of MRI-derived deep learning score for non-invasive prediction of PD-L1 expression and prognostic stratification in head and neck squamous cell carcinoma

Clinical Characteristics

 

Training Cohort (n = 267)

Internal validation cohort (n = 115)

External validation cohort (n = 134)

 

P value

Age (mean ± SD, years)

 

 61.70 ± 8.50

 59.85 ± 8.16

 58.76 ± 8.63

 

0.003

Histological differentiation

     

0.505

 

Well

31 (11.61%)

11 (9.57%)

14 (10.45%)

  
 

Moderately

115 (43.07%)

48 (41.74%)

47 (35.07%)

  
 

Poorly

121(45.32%)

56 (48.69%)

73 (54.48%)

  

T stage

     

0.043

 

2

89 (33.33%)

30 (26.09%)

46 (34.33%)

  
 

3

120 (44.94%)

46 (40.00%)

46 (34.33%)

  
 

4

58 (21.73%)

39 (33.91%)

42 (31.34%)

  

N stage

     

0.321

 

0

85 (31.84%)

34 (29.57%)

33 (24.63%)

  
 

1

83 (31.09%)

31 (26.96%)

40 (29.85%)

  
 

2

92 (34.46%)

42 (36.52%)

53 (39.55%)

  
 

3

7 (2.61%)

8 (6.95%)

8 (5.97%)

  

Smoking

     

0.519

 

No

72 (27.00%)

29 (25.22%)

42 (31.34%)

  
 

Yes

195 (73.00%)

86 (74.78%)

92 (68.66%)

  

Drink

     

0.758

 

No

185(85.5%)

79 (68.70%)

88 (65.67%)

  
 

Yes

82 (14.5%)

36 (31.30%)

46 (34.33%)

  

Sex

     

0.196

 

Male

238 (89.14%)

109 (76.20%)

123 (91.79%)

  
 

Female

29(10.86%)

6 (23.80%)

11 (8.21%)

  
  1. P value < 0.05 is considered as a significant difference.SD, standard deviation