Skip to main content

Table 3 Univariate and multivariate cox regression analysis of ESCC-PS and clinicopathological characteristics for progression-free survival in 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

Univariate Cox analysis

Multivariate Cox analysis

HR (95%CI)

P value

HR (95%CI)

P value

Age

    

  ≤ 60

1

0.977

  

 > 60

0.990 (0.507–1.935)

   

Gender

    

 Male

1

0.686

  

 Female

0.847 (0.378–1.896)

   

Smoking history

    

 No

1

0.955

  

 Yes

1.021 (0.501–2.081)

   

Drinking history

    

 No

1

0.497

  

 Yes

1.277 (0.631–2.584)

   

T stage

    

 T1-T2

1

0.895

  

 T3-T4

1.067 (0.411–2.766)

   

N stage

    

 N0-N1

1

0.984

  

 N2-N3

0.993 (0.504–1.959)

   

Stage

    

 III

1

0.603

  

 IV

1.219 (0.579–2.566)

   

Lung metastasis

    

 No

1

0.427

  

 Yes

1.412 (0.603–3.311)

   

Bone metastasis

    

 No

1

0.996

  

 Yes

1.003 (0.235–4.288)

   

Liver metastasis

    

 No

1

0.915

  

 Yes

1.059 (0.371–3.025)

   

Radiotherapy

    

 No

1

   

 Yes

0.543 (0.249–1.185)

0.125

  

Chemotherapy

    

 No

1

0.771

  

 Yes

1.121 (0.519–2.420)

   

PD-L1

    

  < 57.5%

1

0.092*

1

0.009*

  ≥ 57.5%

0.486 (0.210–1.124)

 

0.231 (0.077–0.696)

 

ESCC-PS

    

 ESCC-PS 1

1

 < 0.001*

1

 < 0.001*

 ESCC-PS 2

0.393 (0.172–0.897)

 

0.275 (0.108–0.700)

 

 ESCC-PS 3

0.055 (0.018–0.173)

 

0.029 (0.008–0.111)

 
  1. *Statistcal significance