Fan et al reported a model for predicting the risk of bone metastases in a patient with bladder cancer. This can help to identify a patient who may benefit from more aggressive management. The authors are from Chengde Medical University and Qingdao University Medical College in China.
Patient selection: bladder cancer
Parameters:
(1) age
(2) lung metastases
(3) liver metastases
(4) brain metastases
(5) N status
(6) T status
(7) histology
(8) grade
(9) primary site
(10) race
Parameter
Finding
Points
age
<= 80 years
21
> 80 years
34.4
lung metastases
no
21
yes
100
liver metastases
no
21
yes
80.2
brain metastases
no
21
yes
84
N status
N0 or N1
21
N2 or N3
69.1
NX
74.7
T status
T1 or T2
21
T3 or T4
24.7
TX
80.9
histology
transitional cell
21
squamous cell
58.6
adenocarcinoma
19.8
other
40.1
grade
1 or 2
21
3 or 4
57.6
primary site
ureteral (C67.67)
1
dome of bladder (C67.1)
5.2
bladder wall (C67.234)
15.4
trigone (C67.0)
21
bladder NOS (C67.9)
22.2
overlapping lesion (C67.8)
33
bladder neck (C67.5)
37
race
black
21
white or other
0
where:
• It is unclear why NX and TX are awarded the highest points for N and T status.
total score =
= SUM(points for all of the parameters)
Interpretation:
• minimum score: 169
• maximum score: 609.9
X =
= (0.02674 * (score)) - 10.75
probability of bone metastases =
= 1 / (1 + EXP((-1) *X))
Performance:
• The area under the ROC curve is 0.81.
To read more or access our algorithms and calculators, please log in or register.
