Skip to main content

Table 2 Univariable and multivariable Fine-Gray competing risks models predicting PRCC recurrence, while accounting for the competing risk of death without previous recurrence

From: The VENUSS prognostic model to predict disease recurrence following surgery for non-metastatic papillary renal cell carcinoma: development and evaluation using the ASSURE prospective clinical trial cohort

 

Univariable

Multivariable

Variable

Category

Coeff

SE

SHR

95% CI

p

Coeff

SE

SHR

95% CI

p

Tumour size

0.1–4.0 cm

Reference

Reference

 

4.1–10.0 cm

1.64

0.34

5.15

2.60–10.2

< 0.001

1.00

0.38

2.73

1.31–5.68

0.007

> 10 cm

2.52

0.40

12.38

5.63–27.2

< 0.001

1.09

0.49

2.97

1.14–7.74

0.026

T stage

pT1

Reference

Reference

 

pT2

1.47

0.41

4.36

1.94–9.8

< 0.001

0.69

0.44

1.99

0.84–4.69

0.12

pT3

2.09

0.30

7.99

4.48–14.3

< 0.001

0.91

0.36

2.48

1.21–5.05

0.013

pT4

4.22

0.28

68.3

39.4–118.4

< 0.001

1.15

0.63

3.15

0.92–10.8

0.068

N stage

pNx/pN0

Reference

Reference

 

pN1

2.55

0.36

12.8

6.37–25.9

< 0.001

1.49

0.46

4.42

1.78–11.0

0.001

Nuclear grade

G1 or G2

Reference

Reference

 

G3 or G4

1.55

0.26

4.73

2.85–7.87

< 0.001

0.93

0.30

2.53

1.42–4.53

0.002

Venous tumour thrombus

Absent

Reference

Reference

 

Renal vein

2.15

0.48

8.56

3.32–22.1

< 0.001

1.07

0.49

2.93

1.12–7.67

0.029

IVC

2.44

0.35

11.5

5.77–23.0

< 0.001

0.95

0.45

2.57

1.08–6.16

0.034

Papillary type

Type 1

Reference

 
 

Type 2

1.07

0.31

2.91

1.58–5.34

< 0.001

     

Fat invasion

Present

0.93

0.28

2.53

1.47–4.33

< 0.001

     

Tumour necrosis

Present

1.01

0.27

2.74

1.61–4.67

< 0.001

     

Sarcomatoid features

Present

1.33

0.57

3.77

1.24–11.5

0.019

     

Surgical margin

Positive

0.45

0.46

1.56

0.63–3.86

0.34

     
  1. Data on papillary type 1 or 2 were available in 493 patients. On multivariable analysis, papillary type, tumour necrosis, fat invasion and sarcomatoid features were removed from the model during the backward selection process