Incidence Trends
The age-adjusted incidence rate of UCB increased from 8.4 per 100,000 persons in 1975 to 13.1 per 100,000 persons in 2016. The joinpoint regression analysis revealed there were two join points (Fig. 1). The incidence rate showed a rapidly increasing trend from 1975 to 1987, with an APC of 3.3% (95% confidence interval [CI] = 2.8–3.9%, P < 0.0001), while this increase slowed from 1988 to 2001 (APC = 0.8%, 95% CI = 0.4–1.1%, P < 0.0001), and was followed by a decreasing trend from 2002 to 2016 (APC=–0.7%, 95% CI=–0.9% to − 0.5%, P < 0.0001).
Among male UCB patients, the age-adjusted incidence rate increased from 14.9 per 100,000 persons in 1975 to 23 per 100,000 persons in 2016. The joinpoint regression analysis similarly revealed two join points, as for the total population (Fig. 1). The incidence rate in males showed a rapidly increasing trend from 1975 to 1987 (APC = 3.6%, 95% CI = 3.0–4.2%, P < 0.0001), which slowed markedly from 1988 to 2003 (APC = 0.4%, 95% CI = 0.1–0.7%, P < 0.0001), before decreasing from 2004 to 2016 (APC=–0.9%, 95% CI=–1.3% to − 0.6%, P < 0.0001).
Among female UCB patients, the age-adjusted incidence rate increased slightly from 3.8 per 100,000 persons in 1975 to 5.3 per 100,000 persons in 2016. Only one join point was identified (Fig. 1), with the incidence rate showing a slowly increasing trend from 1975 to 1996 (APC = 2.0%, 95% CI = 1.7–2.4%, P < 0.0001), followed by a slowing decreasing trend from 1997 to 2016 (APC=–0.8%, 95% CI=–1.2% to − 0.5%, P < 0.0001).
Characteristic of Patients in Predictive Model
This study included 11,512 postsurgery UCB patients, with 8,058 allocated to the training cohort and 3,454 to the validation cohort. The median follow-up time was 37 months (range 1–119 months). Overall, 2,607 (22.65.0%) patients died of PTCCA during the follow-up. The basic characteristic of the patients are listed in Table 1. The total cohort was aged 70.35 ± 11.71 years, and included 8,980 (78.01%) male patients. There were 10,240 (88.95%) white patients, 656 (5.70%) black patients, and 616 (5.35%) patients of other races. The marital status was divided into 7,209 (62.62%) married patients, 1,471 (12.78%) unmarried or domestic partner or single patients, and 2,832 (24.60%) separated or divorced or widowed patients. There were 11,138 (96.75%) insured and 374 (3.25%) uninsured patients. The primary tumor site in the largest proportion of patients was the anterior, posterior, and lateral walls (n = 4,212, 36.61%), with overlapping lesions/location NOS in 5,539 (48.12%) patients. Most of patients had an undifferentiated grade (n = 7,082, 61.52%) or poorly differentiated grade (n = 2,473, 21.48%), had a localized metastasis stage (n = 9,688, 84.16%), and were at AJCC stage I (n = 7,461, 64.81%) or II (n = 2,312, 20.08%).
Table 1
Baseline characteristic of the patients with urothelial cancer of the bladder
Factors
|
Training cohort (n = 8058)
|
%
|
Validation cohort (n = 3454)
|
%
|
Total (n = 11512)
|
%
|
P
|
Age
|
70.20 ± 11.61
|
70.68 ± 11.92
|
70.35 ± 11.71
|
0.04
|
Sex
|
|
|
|
|
|
|
0.07
|
Male
|
6322
|
78.46
|
2658
|
76.95
|
8980
|
78.01
|
|
Female
|
1736
|
21.54
|
796
|
23.05
|
2532
|
21.99
|
|
Grade
|
|
|
|
|
|
|
0.99
|
Well
|
309
|
3.83
|
134
|
3.88
|
443
|
3.85
|
|
Moderately
|
1059
|
13.14
|
455
|
13.17
|
1514
|
13.15
|
|
Poorly
|
1735
|
21.53
|
738
|
21.37
|
2473
|
21.48
|
|
Undifferentiated
|
4955
|
61.49
|
2127
|
61.58
|
7082
|
61.52
|
|
Race
|
|
|
|
|
|
|
0.97
|
White
|
7171
|
88.99
|
3069
|
88.85
|
10240
|
88.95
|
|
Black
|
458
|
5.68
|
198
|
5.73
|
656
|
5.70
|
|
Other
|
429
|
5.32
|
187
|
5.41
|
616
|
5.35
|
|
Marital status
|
|
|
|
|
|
|
0.06
|
Married
|
5098
|
63.27
|
2111
|
61.12
|
7209
|
62.62
|
|
Single
|
1025
|
12.72
|
446
|
12.91
|
1471
|
12.78
|
|
SDW2
|
1935
|
24.01
|
897
|
25.97
|
2832
|
24.60
|
|
Insurance
|
|
|
|
|
|
|
0.41
|
Yes
|
7789
|
96.66
|
3349
|
96.96
|
11138
|
96.75
|
|
No
|
269
|
3.34
|
105
|
3.04
|
374
|
3.25
|
|
Location
|
|
|
|
|
|
|
0.19
|
Urachus/dome
|
303
|
3.76
|
140
|
4.05
|
443
|
3.85
|
|
Trigone, neck, ureteric orifice
|
951
|
11.80
|
365
|
10.57
|
1316
|
11.43
|
|
Wall
|
2920
|
36.24
|
1294
|
37.46
|
4214
|
36.61
|
|
NOS/overlap
|
3884
|
48.20
|
1655
|
47.92
|
5539
|
48.12
|
|
Stage
|
|
|
|
|
|
|
0.82
|
Localized
|
6782
|
84.16
|
2906
|
84.13
|
9688
|
84.16
|
|
Regional
|
953
|
11.83
|
417
|
12.07
|
1370
|
11.90
|
|
Distant
|
323
|
4.01
|
131
|
3.79
|
454
|
3.94
|
|
AJCC
|
|
|
|
|
|
|
0.41
|
Ⅰ
|
5214
|
64.71
|
2247
|
65.06
|
7461
|
64.81
|
|
Ⅱ
|
1627
|
20.19
|
685
|
19.83
|
2312
|
20.08
|
|
Ⅲ
|
499
|
6.19
|
236
|
6.83
|
735
|
6.38
|
|
Ⅳ
|
718
|
8.91
|
286
|
8.28
|
1004
|
8.72
|
|
Size
|
|
|
|
|
|
|
0.26
|
1-20mm
|
1647
|
20.44
|
746
|
21.60
|
2393
|
20.79
|
|
21-49mm
|
3650
|
45.30
|
1516
|
43.89
|
5166
|
44.87
|
|
≥ 50mm
|
2761
|
34.26
|
1192
|
34.51
|
3953
|
34.34
|
|
Regional lymph nodes removed
|
|
|
|
|
|
|
0.22
|
Yes
|
1574
|
19.53
|
641
|
18.56
|
2215
|
19.24
|
|
No
|
6484
|
80.47
|
2813
|
81.44
|
9297
|
80.76
|
|
Chemotherapy
|
|
|
|
|
|
|
0.45
|
Yes
|
2652
|
32.91
|
1162
|
33.64
|
3814
|
33.13
|
|
No
|
5406
|
67.09
|
2292
|
66.36
|
7698
|
66.87
|
|
Radiation
|
|
|
|
|
|
|
0.42
|
Yes
|
493
|
6.12
|
225
|
6.51
|
718
|
6.24
|
|
No
|
7565
|
93.88
|
3229
|
93.49
|
10794
|
93.76
|
|
SDW, separated or divorced or widowed |
In terms of treatment modalities, the regional lymph nodes were removed in 2,215 (19.24%) patients, 3814 (33.13%) had received chemotherapy, and 718 (6.24%) had received radiation. There were no significant differences between the training and validation cohorts in sex, race, tumor size, marital status, insurance status, differentiate grade, metastasis stage, AJCC stage, tumor location, regional lymph nodes removal status, chemotherapy status, or radiation status (P > 0.05). The patients in the validation cohort were slightly older than those in the training cohort (P = 0.04). The log-rank test showed that the survival time did not differ significantly between the training and validation cohorts (P = 0.3).
Independent Prognostic Factors
Cox regression with the backward stepwise selection method revealed that the statistically significant factors affecting postsurgery UCB survival in the training cohort were the age at diagnosis (hazard ratio [HR] = 1.03, 95% confidence interval [CI] = 1.03–1.04), black race (versus white: HR = 1.22, 95% CI = 1.01–1.47), moderately differentiated grade (versus well-differentiated grade: HR = 1.82, 95% CI = 1.11–2.99), poorly differentiated grade (versus well-differentiated grade: HR = 3.21, 95% CI = 2.00–5.15), undifferentiated grade (versus well-differentiated grade: HR = 3.13, 95% CI = 1.96–5.00), AJCC stage II (versus AJCC stage I: HR = 3.49, 95% CI = 3.08–3.96), AJCC stage III (versus AJCC stage I: HR = 5.92, 95% CI = 3.39–10.33), AJCC stage IV (versus AJCC stage I: HR = 12.97, 95% CI = 7.39–22.75), no regional lymph nodes removed (versus regional lymph nodes removed: HR = 1.82, 95% CI = 1.60–2.07), no chemotherapy (versus received chemotherapy: HR = 1.41, 95% CI = 1.27–1.57), no insurance (versus insured: HR = 1.40, 95% CI = 1.10–1.78), being single or unmarried or domestic partner (versus married, HR = 1.30, 95% CI = 1.13–1.50), being separated or divorced or widowed (versus married: HR = 1.22, 95% CI = 1.10–1.36), tumor size of 21–49 mm (versus 1–20 mm: HR = 1.18, 95% CI = 1.02–1.36), and tumor size of ≥ 50 mm (versus 1–20 mm: HR = 1.48, 95% CI = 1.28–1.71). These independent prognostic factors were used to establish the prognostic risk prediction model of postsurgery UCB.
Construction of the Nomogram
The nine aforementioned variables were used to construct a prognostic nomogram for predicting the 3, 5-, and 8year CSS of postsurgical patients with UCB (Fig. 2). The nomogram shows that the age at diagnosis and the AJCC stage were the strongest factors influencing the prognosis, followed by the differentiation grade, regional lymph nodes removal status, tumor size, race, chemotherapy status, insurance status, and marital status. When using the nomogram, each variable is assigned a score, and the scores for all of the variables are added to obtain the total score. A vertical line is then dropped down from the row showing the total scores to estimate the 3-, 5-, and 8-year survival rates (Fig. 2).
Validation of the Nomogram
The C-index values of the nomogram were 0.820 and 0.823 in the training and verification cohorts, respectively, while those for the AJCC staging system were 0.773 and 0.786. Compared with the AJCC stage, the NRI values for 3, 5, and 8 years of follow-up were 0.34 (95% CI = 0.28–0.41), 0.37 (95% CI = 0.33–0.43), and 0.39 (95% CI = 0.34–0.44), respectively, in the training cohort, and 0.30 (95% CI = 0.20–0.40), 0.33 (95% CI = 0.21–0.41), and 0.32 (95% CI = 0.21–0.42) in the validation cohort; the corresponding IDI values were 0.057, 0.059, and 0.059 in the training cohort, and 0.042, 0.051, and 0.053 in the validation cohort (all P < 0.001). These performance indicators demonstrate that the nomogram showed better discrimination than the AJCC staging system.
The calibration plots showed excellent consistency between the observed and nomogram-predicted probabilities in the training and validation cohorts (Fig. 3). The ROC curve of the predictive model showed good clinical effectiveness in both the training cohort (Fig. 4A), with areas under the ROC curve (AUCs) for 3, 5, and 8 years of follow-up of 0.831, 0.808, and 0.789, respectively, and the validation cohort (Fig. 4B), with corresponding AUCs of 0.811, 0.798, and 0.789.