Table 3 Single-factor and multifactor logistic analyses of model predictive performance.

From: AI-based prediction of androgen receptor expression and its prognostic significance in prostate cancer

  

Univariate analysis

Multivariate analysis

 

OR(95%CI)

P value

OR(95%CI)

P value

Age

 

0.992(0.967–1.018)

0.545

  

BMI

 

1.038(0.975–1.104)

0.242

  

TPSA

< 4

0.411 (0.177–0.955)

0.039

0.573 (0.214–1.536)

0.268

4–10

0.471 (0.293–0.757)

0.002

0.627 (0.357–1.102)

0.105

10–20

0.506 (0.315–0.811)

0.005

0.547 (0.319–0.937)

0.028

> 20

    

Volume

 

0.997(0.989–1.004)

0.390

  

Margin

Y/N

0.531(0.361–0.782)

0.001

0.600(0.379–0.950)

0.029

Seminal vesicle invasion

Y/N

2.503(1.612–3.887)

< 0.001

1.932(1.091–3.421)

0.024

Extracapsular invasion

Y/N

1.617(1.087–2.405)

0.018

0.938(0.565–1.556)

0.803

T

< T4/T4

2.558(0.491–13.330)

0.265

  

N

N1/NO

1.177(0.809–1.713)

0.394

  

M

M1/MO

1.009(0.141–7.229)

0.993

  

GS

 

1.295(1.099–1.527)

0.002

1.086(0.885–1.334)

0.428

AR

0

    

1

1.686(0.994–2.861)

0.053

1.875(1.044–3.368)

0.035

2

1.813(1.090–3.016)

0.022

2.178(1.225–3.872)

0.008

3

6.400(0.687–59.616)

0.100

  

Her2

0

    

1

1.020(0.493–2.113)

0.957

  

2

0.989(0.666–1.470)

0.958

  

Ki67

 

1.019(0.996–1.042)

0.103

  

P53

Wild/Mutant

1.640(1.106–2.431)

0.014

1.606(1.049–2.461)

0.029

ERG

-/+

1.298(0.846–1.992)

0.233