- Email: [email protected]
Impact of surgical waiting time on survival in patients with upper tract urothelial carcinoma: A national cancer database study
Urologic Oncology: Seminars and Original Investigations ] (2017) ∎∎∎–∎∎∎
Original article
Impact of surgical waiting time on survival in patients with upper tract urothelial carcinoma: A national cancer database study Leilei Xia, M.D., Benjamin L. Taylor, M.D., Jose E. Pulido, M.D., Thomas J. Guzzo, M.D., M.P.H.⁎ Division of Urology, Department of Surgery, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA Received 8 May 2017; received in revised form 20 June 2017; accepted 14 September 2017
Abstract Purpose: To evaluate the impact of surgical waiting time (SWT) on the survival outcome in patients with upper tract urothelial carcinoma (UTUC). Materials and methods: We identified patients with nonmetastatic UTUC who underwent radical nephroureterectomy (RNU) between 2004 and 2013 in the National Cancer Database. The association between SWT and overall survival (OS) was evaluated using Cox proportional hazards regression. SWT was categorized into 6 groups: SWT ≤ 7 days, SWT 8 to 30 days, SWT 31 to 60 days, SWT 61 to 90 days, SWT 91 to 120 days, and SWT 121 to 180 days. Multivariable analyses were adjusted for patient, tumor, and facility-related factors. Results: A total of 3,581 patients were included in the final overall cohort and 2,397 (66.9%) patients had the higher-risk disease (highgrade or ≥pT2). Multivariable Cox regressions showed that patients in the groups of SWT 31 to 60 days, SWT 61 to 90 days, and SWT 91 to 120 days had similar OS compared with patients who had SWT of 8 to 30 days in the overall cohort and higher-risk cohort. Patients with SWT 121 to 180 days had worse OS (HR ¼ 1.61, 95% CI: 1.19–2.19, P ¼ 0.002 in the overall cohort; HR ¼ 1.56, 95% CI: 1.11–2.20, P ¼ 0.010 in the higher-risk cohort). Conclusions: Increased SWT from diagnosis to RNU appears to be not associated with worse OS within 120 days after the diagnosis of UTUC but SWT 4 120 days may be associated with worsened survival. These findings might have important implications for trial design in the evaluation of neoadjuvant chemotherapy for UTUC and future clinical practice. r 2017 Published by Elsevier Inc.
Keywords: Upper tract urothelial carcinoma; Ureter; Renal pelvis; Time factors; Survival; Outcome assessment
1. Introduction Upper tract urothelial carcinoma (UTUC) is a rare malignancy accounting for 5% to 7% of all renal cancers and 5% to 10 % of all urothelial carcinomas [1]. Although selected low-risk disease can be managed with kidneysparing surgery, radical nephroureterectomy (RNU) (with bladder cuff removal) remains the standard of care for the majority of nonmetastatic UTUCs [2–4]. Despite radical surgery, high-risk UTUCs usually have a poor prognosis
⁎
Correspondence to: Perelman Center for Advanced Medicine, West Pavilion, 3rd Floor, 3400 Civic Center Boulevard, Philadelphia, PA 19104. Tel.: þ1-215-662-2891; fax: þ1-215-662-3955. E-mail address: [email protected] (T.J. Guzzo). http://dx.doi.org/10.1016/j.urolonc.2017.09.013 1078-1439/r 2017 Published by Elsevier Inc.
and the long-term survival for UTUC has not improved very much for the past 20 to 30 years [2,5–7]. There is an imperative need to further investigate the role of perioperative therapies, particularly chemotherapy, in the management of UTUC. Although there is a lack of level 1 evidence for neoadjuvant chemotherapy (NAC) in patients with UTUC, NAC appears to be a promising approach for selected patients with high-grade or muscle-invasive disease [8–10]. Also, with advances in imaging and diagnostic ureteroscopy technology, our ability to clinically stage UTUC has improved. However, the lack of level 1 evidence for NAC in UTUC and the potential for delaying definitive surgical therapy in patients who may not respond continues to be a significant concern.
2
L. Xia et al. / Urologic Oncology: Seminars and Original Investigations ] (2017) ∎∎∎–∎∎∎
It is intuitive that excessive SWT for cancer surgery may cause disease progression and result in unfavorable prognosis. It is also likely that different cancers have different growth rates and biological behavior. Therefore, it is currently unknown as to whether there is a potential “safe window” of SWT for specific cancers [11,12]. Some would argue that definitive surgery for aggressive cancers such as UTUC should be performed with the shortest SWT possible and it is a moot point to investigate the so-called “safe window” [3]. However, in the modern health care system, SWT may be inevitable and the fact that UTUC is potentially aggressive raise the need to evaluate the impact of SWT on survival outcomes. The results may provide guidance regarding the preoperative assessment timeline. Also, if the “safe window” is long enough, it may partially alleviate the concerns regarding delaying RNU for legitimate reasons such as NAC in practice and clinical trials. It can also add valuable information for institutions and health care policy makers in setting parameters for wait times in various clinical conditions. There is limited data regarding the impact of SWT on the survival outcomes in patients with UTUC [13–15]. Due to small sample sizes, previous studies categorize patients into only 2 groups (early RNU vs. delayed RNU), which can potentially bias the results. Given these drawbacks, there remains no consensus about the “safe window” or cutoff for removal of tumors after diagnosis [2–4]. Given the fact that UTUC is a relatively rare malignancy, the current question is unlikely to be answered in a prospective trial or singleinstitution retrospective study. We hypothesized that there may be a “safe window” in terms of the effect of SWT on UTUC outcome and the question should be answered with the nonbinary categorization of SWT. We, therefore, investigated the association of SWT and outcome using the National Cancer Database (NCDB). NCDB is a wellvalidated database for cancer outcomes research including the effect of SWT on survival [16–18]. In this study, we took the advantages of prospectively collected data and large sample size of NCDB analyzing the impact of SWT on survival after UTUC.
long-term survival [18]. This study used de-identified data and institutional review board approval was not required. 2.2. Study population and variables
2. Materials and methods
All patients with UTUC in the NCDB from 2004 to 2013 were identified. Urothelial histology was determined via International Classification of Diseases for Oncology, Third Edition, codes 8120 to 8139. The study population was defined as patients who underwent RNU (Code 40 in the NCDB). We excluded patients with metastatic disease at the time of diagnosis, those with other cancer diagnoses, those who died within 90 days of RNU, those who had NAC, those who had other treatments other than chemotherapy, and those with missing data in relevant co-variables or outcomes of interest. SWT was defined as the number of days between the date of initial diagnosis and the date on which the RNU was performed. We also excluded cases with RNU occurred on the same day as the diagnosis and patients who had SWT more than 180 days considering the rare occurrence in the database. Co-variables included age, sex, race, reporting facility type, income level, education level, comorbidity status measured by the Charlson/Deyo score (CDS, categorized as 0, 1, and ≥ 2), tumor location (renal pelvis vs. ureter), tumor grade, pT stage, pN stage, lymph node dissection (LND), and adjuvant chemotherapy (AC). We used the general cancer grading system (grade 1—well differentiated, grade 2—moderately differentiated, grade 3—poorly differentiated, and grade 4—undifferentiated) in our study. Although both general grading system and 2004 World Health Organization/International Society of Urological Pathology (WHO/ISUP) system (low-grade vs. high-grade) were included in the NCDB recently, WHO/ISUP grading system had significantly more missing values. As described and confirmed by previous studies, grades 1 to 2 can be considered as WHO/ISUP low-grade tumors and grades 3 to 4 as WHO/ISUP high-grade tumors [19–21]. The primary outcome of interest was overall survival (OS). Follow-up time was defined as the number of months between the date of diagnosis and the date on which the patient was last contacted or died.
2.1. Data source
2.3. Statistical analysis
Data of current study was derived from the NCDB. The NCDB is a joint project of the American Cancer Society and the Commission on Cancer (CoC) of the American College of Surgeons. The NCDB, established in 1989, is a nationwide, hospital-based, comprehensive clinical surveillance resource oncology data set that currently captures 70% of all newly diagnosed malignancies in the United States annually. NCDB contains comprehensive data about reporting facility, patient demographics, cancer identification, disease stage, treatments, short-term outcomes, and
The primary objective of the study was to assess the association between SWT and OS. Time intervals between the date of initial diagnosis and date of RNU were set at 30day increments, with the last 2 intervals combined owing to smaller numbers in each. Also, the first interval was separated into 2 groups with the consideration that small SWT may indicate nonelective surgery and may have worse outcomes [22]. Our a priori analysis also showed less favorable OS in patients who had SWT ≤ 7 days vs. patients who had SWT of 8 to 30 days. Therefore, SWT
L. Xia et al. / Urologic Oncology: Seminars and Original Investigations ] (2017) ∎∎∎–∎∎∎
was ultimately categorized into 6 groups: SWT ≤ 7 days, SWT 8 to 30 days, SWT 31 to 60 days, SWT 61 to 90 days, SWT 91 to 120 days, and SWT 121 to 180 days. Baseline demographics and tumor characteristics were described. Unadjusted OS was estimated and assessed by the Kaplan–Meier curves and univariable Cox regression analysis. Data were reported as medians (interquartile ranges [IQRs]) and frequencies (percentages) for continuous and categorical variables, respectively. Multivariable Cox regression analysis was performed to assess the associations between all the variables and OS. Hazard ratios (HRs) with 95% confidence intervals (CIs) were used to assess the strength of the associations. All tests were 2-sided and a P o 0.05 was considered statistically significant. All analyses were performed using STATA 14 (StataCorp LP, College Station, TX).
2.4. Primary analysis We performed the primary multivariable Cox regression analyses in the overall included cohort and a subgroup of patients with higher-risk disease (grades 3– 4 or ≥pT2) [4]. Given the difficulty of clinically staging UTUC before RNU and the fact that tumor grade in the NCDB is based on final pathologic diagnosis, we chose to use final pathologic T stage and grade to define the higher-risk subgroup. In the primary analyses, SWT was considered as a categorical variable as described earlier and SWT 8 to 30 days was used as the reference. A sensitivity analysis excluding patients with SWT ≤ 7 days was also performed.
2.5. Secondary analysis Based on the results of primary analyses, we then performed the secondary Cox regressions (univariable and multivariable) with SWT considered as a continuous variable (per day increase) in the overall cohort and higher-risk cohort with the SWT in the range of 1 to 120 days and 8 to 120 days.
2.6. Post-hoc analysis Based on the results of primary analyses and secondary analyses, we did a post-hoc analysis comparing patient characteristics between SWT of 1 to 120 days and SWT of 121 to 180 days. The objective was to explore if there are any differences in any variables between the 2 groups. Continuous variables were compared using Mann–Whitney U test and categorical variables were compared using Fisher’s exact test.
3
3. Results 3.1. Study cohorts A total of 3,581 patients were included in the final overall cohort. Characteristics of the overall cohort are presented in Table 1. Median age was 71 years old (interquartile range [IQR] 63–78 years); 1,543 (43.1%) patients were female and 3,347 patients (93.5%) were white. In the overall cohort, 230 (6.4%) patients had SWT ≤ 7 days, 1,628 (45.5%) patients had SWT ≤ 30 days, 2,878 (80.4%) patients had SWT ≤ 60 days, 3,350 (93.5%) patients had SWT ≤ 90 days, and 3,493 (97.5%) patients had SWT ≤ 120 days. The median follow-up was 40.4 months (IQR: 23.3–65.2 months). Three-year OS and 5-year OS for the overall cohort were 74.0% (95% CI: 72.4%–75.5%) and 61.9% (95% CI: 60.0%–63.7%), respectively. A total 2,397 (66.9%) patients had higher-risk disease based on our predetermined criteria (grades 3–4 or ≥ pT2). Three-year OS and 5-year OS for the higher-risk cohort were 66.6% (95% CI: 64.5%–68.5%) and 54.3% (95% CI: 51.9%–56.6%), respectively. 3.2. Unadjusted OS Kaplan–Meier survival estimates and univariable Cox regression analyses of the association between SWT and OS are shown in Fig. 1. In the overall cohort (n ¼ 3,581), compared with patients having SWT of 8 to 30 days, patients having SWT of 1 to 7 days had worse OS (HR ¼ 1.29, 95% CI: 1.03–1.62, P ¼ 0.029); patients having SWT of 121 to 180 days also had worse OS (HR ¼ 2.05, 95% CI: 1.52–2.77, P o 0.001). In the higher-risk cohort (n ¼ 2,397), compared with patients having SWT of 8 to 30 days, only patients having SWT of 121 to 180 days had worse OS (HR ¼ 2.11, 95% CI: 1.51–2.95, P o 0.001). 3.3. Primary analysis Results of primary multivariable Cox regression analysis adjusting for all the co-variables are shown in Table 2. In the overall cohort (n ¼ 3,581), compared with patients having SWT of 8 to 30 days, patients having SWT of 1 to 7 days had worse OS (HR ¼ 1.32, 95% CI: 1.06–1.67, P ¼ 0.016); patients having SWT of 121 to 180 days also had worse OS (HR ¼ 1.61, 95% CI: 1.19–2.19, P ¼ 0.002). In the higher-risk cohort (n ¼ 2,397), compared with patients having SWT of 8 to 30 days, only patients having SWT of 121 to 180 days had worse OS (HR ¼ 1.56, 95% CI: 1.11–2.20, P ¼ 0.010). Results of sensitivity analyses in the overall cohort and higher-risk cohort excluding SWT of 1 to 7 days are shown in Table 3. The sensitivity analyses confirmed the results of primary analyses and showed that patients having SWT of 121 to 180
L. Xia et al. / Urologic Oncology: Seminars and Original Investigations ] (2017) ∎∎∎–∎∎∎
4
Table 1 Baseline characteristics of the overall included cohort, stratified by SWT Variable
All patients (n ¼ 3,581) SWT (days) 1–7 (n ¼ 230) 8–30 (n ¼ 1,398) 31–60 (n ¼ 1,250) 61–90 (n ¼ 472) 91–120 (n ¼ 143) 121–180 (n ¼ 88)
Age, years 71 Sex Male 2,038 Female 1,543 Race White 3,347 Black 137 Other 97 Facility type Community 1,929 Academic 1,652 Median incomea o$48,000 1,395 ≥$48,000 2,186 Education levelb ≥13% 1,359 o13% 2,222 CDS 0 2,402 1 893 ≥2 286 Tumor location Renal pelvis 2,428 Ureter 1,153 Tumor grade 1–2 1,273 3–4 2,308 pT stage opT2 1,865 ≥pT2 1,492 pTx 224 pN stage pN0 565 pNþ 147 pNx 2,869 LND No 2,845 Yes 736 AC No 3,224 Yes 357 a
(63–78)
71 (63–79)
70 (63–78)
70 (62–78)
72 (64–80)
73 (64–80)
73 (67–80)
(56.9%) (43.1%)
131 (57.0%) 99 (43.0%)
758 (54.2%) 640 (45.8%)
747 (59.8%) 503 (40.2%)
255 (54.0%) 217 (46.0%)
89 (62.2%) 54 (37.8%)
58 (65.9%) 30 (34.1%)
(93.5%) (3.8%) (2.7%)
208 (90.4%) 9 (3.9%) 13 (5.7%)
1,326 (94.9%) 39 (2.8%) 33 (2.4%)
1,162 (93.0%) 53 (4.2%) 35 (2.8%)
441 (93.4%) 21 (4.5%) 10 (2.1%)
129 (90.2%) 10 (7.0%) 4 (2.8%)
81 (92.1%) 5 (5.7%) 2 (2.3%)
(53.9%) (43.1%)
119 (51.7%) 111 (48.3%)
798 (57.1%) 600 (42.9%)
680 (54.4%) 570 (45.6%)
232 (49.2%) 240 (50.9%)
63 (44.1%) 80 (55.9%)
37 (42.1%) 51 (57.9%)
(39.0%) (61.0%)
90 (39.1%) 140 (60.9%)
551 (39.4%) 847 (60.6%)
486 (38.9%) 764 (61.1%)
182 (38.6%) 290 (61.4%)
47 (32.9%) 96 (67.1%)
39 (44.3%) 49 (55.7%)
(38.0%) (62.0%)
83 (36.1%) 147 (63.9%)
508 (36.3%) 890 (63.7%)
482 (38.6%) 768 (61.4%)
184 (39.0%) 288 (61.0%)
61 (42.7%) 82 (57.3%)
41 (46.6%) 47 (53.4%)
(67.1%) (24.9%) (8.0%)
151 (65.7%) 57 (24.8%) 22 (9.6%)
977 (69.9%) 325 (23.3%) 96 (6.9%)
838 (67.0%) 309 (24.7%) 103 (8.2%)
305 (64.6%) 125 (26.5%) 42 (8.9%)
80 (55.9%) 48 (33.6%) 15 (10.5%)
51 (58.0%) 29 (33.0%) 8 (9.1%)
(67.8%) (32.2%)
170 (73.9%) 60 (26.1%)
946 (67.7%) 452 (32.3%)
824 (65.9%) 426 (34.1%)
317 (67.2%) 155 (32.8%)
104 (72.7%) 39 (27.3%)
67 (76.1%) 21 (23.9%)
(35.6%) (64.4%)
66 (28.7%) 164 (71.3%)
482 (34.5%) 916 (65.5%)
467 (37.4%) 783 (62.6%)
173 (36.7%) 299 (63.4%)
61 (42.7%) 82 (57.3%)
24 (27.3%) 64 (72.7%)
(52.1%) (41.7%) (6.3%)
116 (50.4%) 102 (44.4%) 12 (5.2%)
682 (48.8%) 636 (45.5%) 80 (5.7%)
680 (54.4%) 489 (39.1%) 81 (6.5%)
273 (57.8%) 169 (35.8%) 30 (6.4%)
80 (55.9%) 52 (36.4%) 11 (7.7%)
34 (38.6%) 44 (50.0%) 10 (11.4%)
(15.8%) (4.1%) (80.1%)
41 (17.8%) 12 (5.2%) 177 (77.0%)
212 (15.2%) 66 (4.7%) 1,120 (80.1%)
195 (15.6%) 41 (3.3%) 1,014 (81.1%)
81 (17.2%) 18 (3.8%) 373 (79.0%)
23 (16.1%) 5 (3.5%) 115 (80.4%)
13 (14.8%) 5 (5.7%) 70 (79.6%)
(79.5%) (20.5%)
175 (76.1%) 55 (23.9%)
1,111 (79.5%) 287 (20.5%)
1,005 (80.4%) 245 (19.6%)
373 (79.0%) 99 (21.0%)
114 (79.7%) 29 (20.3%)
67 (76.1%) 21 (23.9%)
(90.0%) (10.0%)
206 (89.6%) 24 (10.4%)
1,231 (88.1%) 167 (12.0%)
1,131 (90.5%) 119 (9.5%)
439 (93.0%) 33 (7.0%)
138 (96.5%) 5 (3.5%)
79 (89.8%) 9 (10.2%)
Household income for each patient’s area of residence. The number of adults in each patient’s area of residence who did not graduate from high school.
b
days had worse OS compared with patients having SWT of 8 to 30 days (overall cohort [n ¼ 3,351]: HR ¼ 1.65, 95% CI: 1.20–2.20, P ¼ 0.002; higher-risk cohort [n ¼ 2,232]: HR ¼ 1.58, 95% CI: 1.12–2.23, P ¼ 0.008). 3.4. Secondary analysis Based on the results of primary analyses, we performed the secondary analyses with SWT considered as a continuous variable in patients with SWT of 1 to 120 days and SWT of 8 to 120 days. All the analyses were separately performed in the overall cohort and higher-risk cohort. The results of secondary univariable and multivariable Cox
regression analyses are shown in Supplementary Table 1 (SWT range: 1–120 days) and Supplementary Table 2 (SWT range: 8–120 days). Increased SWT (per day) was not shown to be associated with worse OS in the overall/ higher-risk cohorts with SWT of 1 to 120 days or SWT of 8 to 120 days. 3.5. Post-hoc analysis Based on our primary and secondary analyses showing that SWT within 120 days was the potential cutoff in terms of OS difference, we did the post-hos analysis comparing patient characteristics between SWT of 1 to 120 days
L. Xia et al. / Urologic Oncology: Seminars and Original Investigations ] (2017) ∎∎∎–∎∎∎ Overall cohort (n=3,581)
Higher-risk cohort (n=2,397) 1.00 Overall survival
Overall survival
1.00 0.75 0.50 0.25 0.00
0.75 0.50 0.25 0.00
0
12
24
36
48
60
72
84
96
108
120
0
12
24
36
Follow-up (months) Number at risk SWT 1-7 d SWT 8-30 d SWT 31-60 d SWT 61-90 d SWT 91-120 d SWT 121-180 d
5
230 1398 1250 472 143 88
186 1256 1144 435 130 77
147 1018 951 344 104 54
116 799 701 247 79 33
92 620 524 188 54 22
74 428 371 136 42 17
40 277 247 92 29 12
48
60
72
84
96
108
120
15 113 65 31 6 3
9 48 24 15 2 2
1 16 5 2 0 0
0 0 0 0 0 0
Follow-up (months)
21 180 151 60 11 8
13 78 58 24 4 3
2 24 14 4 1 0
0 0 0 0 0 0
Number at risk SWT 1-7 d SWT 8-30 d SWT 31-60 d SWT 61-90 d SWT 91-120 d SWT 121-180 d
165 963 811 308 86 64
124 838 721 275 75 55
93 645 574 211 56 33
75 486 396 145 38 17
57 378 282 104 26 10
46 258 187 70 19 8
24 167 120 45 14 4
SWT 1-7 d
SWT 8-30 d
SWT 1-7 d
SWT 8-30 d
SWT 31-60 d
SWT 61-90 d
SWT 31-60 d
SWT 61-90 d
SWT 91-120 d
SWT 121-180 d
SWT 91-120 d
SWT 121-180 d
SWT 8-30 d 1-7 d 31-60 d 61-90 d 91-120 d 121-180 d
HR (95%CI) Reference 1.29 (1.03-1.62) 1.04 (0.91-1.18) 1.13 (0.95-1.35) 1.06 (0.78-1.42) 2.05 (1.52-2.77)
P value 0.029 0.584 0.174 0.724 < 0.001
SWT 8-30 d 1-7 d 31-60 d 61-90 d 91-120 d 121-180 d
HR (95%CI) Reference 1.17 (0.90-1.51) 1.08 (0.93-1.25) 1.17 (0.96-1.43) 1.15 (0.81-1.62) 2.11 (1.51-2.95)
P value 0.251 0.339 0.121 0.442 < 0.001
Fig. 1. Kaplan–Meier curves and univariable Cox regression of the association between SWT and OS in the overall cohort and higher-risk cohort.
(n ¼ 3,493) and SWT of 121 to 180 days (n ¼ 88). The results showed that compared with SWT 1 to 120 days, patients in the SWT 121 to 180 days group were significantly older (73 y vs. 71 y, P ¼ 0.005), more likely to get treated in the academic centers (57.9% vs. 45.8%, P ¼ 0.030), and more likely to have ≥pT2 disease (50.0% vs. 41.5%) (Supplementary Table 3). There were no significant differences in Charlson/Deyo score or tumor grade.
4. Discussion To the best of our knowledge, this is the first study to evaluate the impact of SWT on the survival of UTUC patients in a large national facility-based database. Our results showed that SWT of RNU appeared to be not associated with OS within 120 days of diagnosis. The absence of an association in the 120-day window was consistent in both the overall cohort and a subgroup of patients with higher-risk disease (grades 3–4 or ≥pT2). However, SWT 4 120 days appeared to be associated with worsened OS. There are controversies in previously published reports about the prognostic significance of an increasing SWT in UTUC [13–15]. To be noted, previous studies were singleor multicenter retrospective studies with small sample sizes. Patients were generally categorized into 2 groups (early RNU vs. delayed RNU) by a single-interval cutoff value. Lee et al. [13,23] included 138 patients with UTUC and used the SWT cutoff of 1 month (30.5 days). The results showed that delayed RNU was associated with worse cancer-specific survival (CSS) and recurrence-free survival (RFS) in the ureteral tumor cohort (n = 80) but not in the overall UTUC cohort [13,23]. A study from MD Anderson
Cancer Center (n = 240) used the 3-month cutoff and demonstrated no significant differences in RFS, CSS, and OS between the early and delayed RNU [14]. Another multicenter study also showed no statistical differences in CSS and RFS between patients who had RNU at o3 months vs. ≥3 months after diagnosis in both the overall cohort (n ¼ 187) and a subgroup of patients (n ¼ 90, 48.1%) with muscle-invasive disease (≥pT2) [15]. The authors also analyzed the SWT as a continuous variable (per month) and the results suggested that a delay in the interval from diagnosis to RNU is associated with more advanced disease stage but not RFS or CSS [15]. Compared with previous studies in the literature, our study has 2 features in terms of study design. First, a significantly larger sample size not only provides more statistical power but also allows for analysis of multiple delay groups of varying lengths. We chose the 30-day increment and considered 1 to 7 days as a separate group based on previous national database studies [16,22,24]. The 30-day interval was also partially in line with SWT cutoffs of previously described reports [13–15]. Second, analyses are more extensive in our study. We repeated all the analyses in the higher-risk cohort to make the results more informative. Sensitivity analyses and secondary analyses were also performed to ensure the results are more reliable. Overall, the current study has advantages and contributes to the existing literature on this topic with some novel findings. As it is unlikely that a randomized clinical trial investigating the impact of SWT on cancer outcomes will ever occur, our analyses of the largest cancer database from the United States may provide the most demonstrative evidence on this topic. Although there is the possibility that higher-risk/sicker patients had shorter SWT than lower-risk/healthier patients,
6
L. Xia et al. / Urologic Oncology: Seminars and Original Investigations ] (2017) ∎∎∎–∎∎∎
Table 2 Multivariable Cox regression in the overall cohort and higher-risk cohort. Variable
SWT 8–30 d 1–7 d 31–60 d 61–90 d 91–120 d 121–180 d Age, per 1 y Sex Male Female Race White Black Other Facility type Community Academic Median income o$48,000 ≥$48,000 Education level ≥13% o13% CDS 0 1 ≥2 Tumor location Renal pelvis Ureter Tumor grade 1–2 3–4 pT stage opT2 ≥pT2 pTx pN stage pN0 pNþ pNx LND No Yes AC No Yes
Overall cohort (n ¼ 3,581)
Higher-risk cohort (n ¼ 2,397)
HR (95% CI)
P value
HR (95% CI)
P value
Reference 1.32 (1.06–1.67) 1.11 (0.97–1.27) 1.09 (0.91–1.30) 1.00 (0.74–1.35) 1.61 (1.19–2.19) 1.049 (1.043–1.056)
0.016 0.126 0.360 0.976 0.002 o0.001
Reference 1.24 (0.95–1.61) 1.10 (0.94–1.27) 1.07 (0.88–1.31) 0.94 (0.66–1.34) 1.56 (1.11–2.20) 1.043 (1.036–1.051)
0.114 0.231 0.510 0.744 0.010 o0.001
Reference 0.87 (0.78–0.98)
0.017
Reference 0.88 (0.77–1.00)
0.046
Reference 1.05 (0.78–1.42) 0.94 (0.67–1.32)
0.737 0.735
Reference 1.00 (0.71–1.40) 0.92 (0.65–1.32)
0.987 0.667
Reference 1.04 (0.92–1.16)
0.546
Reference 1.02 (0.89–1.16)
0.775
Reference 0.95 (0.84–1.09)
0.465
Reference 1.01 (0.87–1.18)
0.853
Reference 0.91 (0.80–1.04)
0.153
Reference 0.85 (0.73–0.98)
0.030
Reference 1.21 (1.06–1.38) 1.70 (1.42–2.04)
0.004 o0.001
Reference 1.27 (1.10–1.47) 1.64 (1.32–2.04)
0.001 o0.001
Reference 1.05 (0.93–1.18)
0.440
Reference 1.07 (0.93–1.23)
0.332
Reference 1.38 (1.20–1.59)
o0.001
Reference 1.76 (1.21–2.54)
0.003
Reference 2.46 (2.14–2.82) 1.71 (1.36–2.14)
o0.001 o0.001
Reference 2.76 (2.33–3.27) 2.25 (1.70–2.98)
o0.001 o0.001
Reference 2.74 (2.10–3.57) 1.59 (0.85–2.97)
o0.001 0.150
Reference 2.44 (1.85–3.21) 1.45 (0.72–2.92)
o0.001 0.296
Reference 1.29 (0.70–2.38)
0.407
Reference 1.23 (0.63–2.43)
0.545
Reference 1.15 (0.96–1.37)
0.140
Reference 1.10 (0.91–1.33)
0.306
we adjusted for patient-related, tumor-related, and facilityrelated factors in the multivariable regression models and the results were consistent. Given the fact that there was not a substantial difference in outcome with SWT in the 120day window, this argues in favor of the use of NAC as it is unlikely to result in worse surgical outcomes in patients who may not respond to systemic therapy [8–10]. Compared with AC, one obvious advantage of NAC is that
significantly more patients would be eligible for cisplatinbased multiagent chemotherapy. Studies have consistently shown that estimated glomerular filtration rate (eGFR) is significantly diminished after RNU, which may eliminate cisplatin-based chemotherapy as a therapeutic option in about half of the high-risk patients if it is deferred to the adjuvant setting [25,26]. Other potential advantages of NAC could include the early eradication of micrometastases
L. Xia et al. / Urologic Oncology: Seminars and Original Investigations ] (2017) ∎∎∎–∎∎∎ Table 3 Sensitivity analysis–multivariable Cox regression in the overall cohort and higher-risk cohort excluding SWT 1 to 7 days SWT
8–30 d 31–60 d 61–90 d 91–120 d 121–180 d
Overall cohort (n ¼ 3,351)
Higher-risk cohort (n ¼ 2,232)
HR (95% CI)
P value
HR (95% CI)
P value
Reference 1.11 (0.97–1.27) 1.09 (0.92–1.30) 1.02 (0.75–1.38) 1.65 (1.20–2.20)
0.126 0.322 0.911 0.002
Reference 1.10 (0.94–1.28) 1.08 (0.88–1.32) 0.97 (0.68–1.38) 1.58 (1.12–2.23)
0.221 0.461 0.862 0.008
Adjusted for the same co-variables in Table 2.
and tumor down-staging [8–10]. The level 1 evidence in favor of NAC for muscle-invasive bladder cancer can be considered as a further justification for giving chemotherapy in the neoadjuvant setting [27]. From the future research standpoint, multicenter collaborative prospective trials to investigate NAC for UTUC patients may be the optimal setting given the rarity of UTUC. Similarly, our results may also further support the safety of meticulous risk-stratification for UTUC, most notably by ureteroscopic evaluation. Previous studies have shown that ureteroscopy with biopsy and/or tumor ablation before RNU did not adversely affect the postoperative disease status [28,29]. There are several limitations that should be considered when interpreting our study. First, heterogeneity of the initial diagnosis date in the NCDB might bias the results. The date of initial diagnosis in the NCDB is defined as the date of first clinical or histologic confirmation or if in retrospect, the patient had cancer at an earlier date [30]. We strictly used the time between the date of initial diagnosis and the date of RNU as SWT considering that it might be the optimal method to define SWT in the NCDB [16,17,30]. NCDB is a well-validated administrative database and initial diagnosis date in the real-world practice may suggest it is the date when RNU was indicated or at least considered [18]. Second, there is no information regarding causes of delay in surgery [17]. Causes of delay may relate to various factors and those factors may also be the confounding factors of survival. Third, other important endpoint outcomes such as disease progression, DR, and CSS are not available in the NCDB, which further weakens our study. Finally, NCDB is a hospital-based clinical cancer registry but not population-based. Cancer patients are captured into the NCDB on the basis of the hospital where they get the diagnosis and/or treatment. The inherent biases of NCDB may limit the interpretation of the results in the general population.
5. Conclusions Our results suggest that increased SWT from diagnosis to RNU appears to be not associated with worse OS within
7
120 days after the diagnosis of UTUC but SWT 4 120 days may be associated with worsened survival. These findings might have important implications for trial design in the evaluation of NAC for UTUC and future clinical practice. Nonetheless, further prospective studies assessing cancer-specific outcomes are required to validate our findings. Acknowledgments The data used in this study are derived from a deidentified NCDB file. The American College of Surgeons and the Commission on Cancer have not verified and are not responsible for the analytic or statistical methodology employed, or the conclusions drawn from these data by the investigators. Appendix A. Supplementary material Supplementary data are available in the online version of this article at http://dx.doi.org/10.1016/j.urolonc.2017.09. 013.
References [1] Soria F, Shariat SF, Lerner SP, et al. Epidemiology, diagnosis, preoperative evaluation and prognostic assessment of upper-tract urothelial carcinoma (UTUC). World J Urol 2017;35:379–87. [2] Roupret M, Babjuk M, Comperat E, et al. European Association of Urology guidelines on upper urinary tract urothelial cell carcinoma: 2015 update. Eur Urol 2015;68:868–79. [3] Roupret M, Babjuk M, Comperat E, et al. European guidelines on upper tract urothelial carcinomas: 2013 update. Eur Urol 2013;63: 1059–71. [4] Hu M, Jacobs BL, Montgomery JS, et al. Sharpening the focus on causes and timing of readmission after radical cystectomy for bladder cancer. Cancer 2014;120:1409–16. [5] Leow JJ, Orsola A, Chang SL, Bellmunt J. A contemporary review of management and prognostic factors of upper tract urothelial carcinoma. Cancer Treat Rev 2015;41:310–9. [6] Eylert MF, Hounsome L, Verne J, Bahl A, Jefferies ER, Persad RA. Prognosis is deteriorating for upper tract urothelial cancer: data for England 1985–2010. BJU Int 2013;112:E107–13. [7] Brown GA, Busby JE, Wood CG, et al. Nephroureterectomy for treating upper urinary tract transitional cell carcinoma: time to change the treatment paradigm? BJU Int 2006;98:1176–80. [8] Leow JJ, Martin-Doyle W, Fay AP, Choueiri TK, Chang SL, Bellmunt J. A systematic review and meta-analysis of adjuvant and neoadjuvant chemotherapy for upper tract urothelial carcinoma. Eur Urol 2014;66:529–41. [9] Matin SF, Margulis V, Kamat A, et al. Incidence of downstaging and complete remission after neoadjuvant chemotherapy for high-risk upper tract transitional cell carcinoma. Cancer 2010;116:3127–34. [10] Porten S, Siefker-Radtke AO, Xiao L, et al. Neoadjuvant chemotherapy improves survival of patients with upper tract urothelial carcinoma. Cancer 2014;120:1794–9. [11] Bourgade V, Drouin SJ, Yates DR, et al. Impact of the length of time between diagnosis and surgical removal of urologic neoplasms on survival. World J Urol 2014;32:475–9.
8
L. Xia et al. / Urologic Oncology: Seminars and Original Investigations ] (2017) ∎∎∎–∎∎∎
[12] Neal RD, Tharmanathan P, France B, et al. Is increased time to diagnosis and treatment in symptomatic cancer associated with poorer outcomes? Systematic review. Br J Cancer 2015;112(Suppl. 1): S92–S107. [13] Lee JN, Kwon SY, Choi GS, et al. Impact of surgical wait time on oncologic outcomes in upper urinary tract urothelial carcinoma. J Surg Oncol 2014;110:468–75. [14] Sundi D, Svatek RS, Margulis V, et al. Upper tract urothelial carcinoma: impact of time to surgery. Urol Oncol 2012;30:266–72. [15] Waldert M, Karakiewicz PI, Raman JD, et al. A delay in radical nephroureterectomy can lead to upstaging. BJU Int 2010;105:812–7. [16] Bleicher RJ, Ruth K, Sigurdson ER, et al. Time to surgery and breast cancer survival in the United States. JAMA Oncol 2016;2:330–9. [17] Strohl AE, Feinglass JM, Shahabi S, Simon MA. Surgical wait time: a new health indicator in women with endometrial cancer. Gynecol Oncol 2016;141:511–5. [18] Boffa DJ, Rosen JE, Mallin K, et al. Using the National Cancer Database for outcomes research: a review. JAMA Oncol 2017. http://dx. doi.org/10.1001/jamaoncol.2016.6905:[Epub ahead of print]. [19] Matulewicz RS, Frainey BT, Oberlin DT, Meeks JJ. High-risk of adverse pathologic features in patients with clinical T1 high-grade bladder cancer undergoing radical cystectomy. J Natl Compr Canc Netw 2016;14:1403–11. [20] Chappidi MR, Kates M, Johnson MH, Hahn NM, Bivalacqua TJ, Pierorazio PM. Lymph node yield and tumor location in patients with upper tract urothelial carcinoma undergoing nephroureterectomy affects survival: a U.S. population-based analysis (2004–2012). Urol Oncol 2016;34:531 e15–531, e24. [21] Charlton ME, Adamo MP, Sun L, Deorah S. Bladder cancer collaborative stage variables and their data quality, usage, and clinical
[22] [23] [24]
[25]
[26]
[27]
[28]
[29]
[30]
implications: a review of SEER data, 2004-2010. Cancer 2014;120 (Suppl. 23):3815–25. Elit LM, O’Leary EM, Pond GR, Seow HY. Impact of wait times on survival for women with uterine cancer. J Clin Oncol 2014;32:27–33. Roupret M. Surgery: surgical wait times for UTUC—a race against the clock? Nat Rev Urol 2014;11:665–6. Kulkarni GS, Urbach DR, Austin PC, Fleshner NE, Laupacis A. Longer wait times increase overall mortality in patients with bladder cancer. J Urol 2009;182:1318–24. Kaag MG, O'Malley RL, O'Malley P, et al. Changes in renal function following nephroureterectomy may affect the use of perioperative chemotherapy. Eur Urol 2010;58:581–7. Lane BR, Smith AK, Larson BT, et al. Chronic kidney disease after nephroureterectomy for upper tract urothelial carcinoma and implications for the administration of perioperative chemotherapy. Cancer 2010;116:2967–73. Advanced Bladder Cancer Meta-analysis Collaboration: Neoadjuvant chemotherapy in invasive bladder cancer. a systematic review and meta-analysis. Lancet 2003;361:1927–34. Boorjian S, Ng C, Munver R, et al. Impact of delay to nephroureterectomy for patients undergoing ureteroscopic biopsy and laser tumor ablation of upper tract transitional cell carcinoma. Urology 2005;66:283–7. Nison L, Roupret M, Bozzini G, et al. The oncologic impact of a delay between diagnosis and radical nephroureterectomy due to diagnostic ureteroscopy in upper urinary tract urothelial carcinomas: results from a large collaborative database. World J Urol 2013;31:69–76. Bilimoria KY, Ko CY, Tomlinson JS, et al. Wait times for cancer surgery in the United States: trends and predictors of delays. Ann Surg 2011;253:779–85.
Original article
Impact of surgical waiting time on survival in patients with upper tract urothelial carcinoma: A national cancer database study Leilei Xia, M.D., Benjamin L. Taylor, M.D., Jose E. Pulido, M.D., Thomas J. Guzzo, M.D., M.P.H.⁎ Division of Urology, Department of Surgery, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA Received 8 May 2017; received in revised form 20 June 2017; accepted 14 September 2017
Abstract Purpose: To evaluate the impact of surgical waiting time (SWT) on the survival outcome in patients with upper tract urothelial carcinoma (UTUC). Materials and methods: We identified patients with nonmetastatic UTUC who underwent radical nephroureterectomy (RNU) between 2004 and 2013 in the National Cancer Database. The association between SWT and overall survival (OS) was evaluated using Cox proportional hazards regression. SWT was categorized into 6 groups: SWT ≤ 7 days, SWT 8 to 30 days, SWT 31 to 60 days, SWT 61 to 90 days, SWT 91 to 120 days, and SWT 121 to 180 days. Multivariable analyses were adjusted for patient, tumor, and facility-related factors. Results: A total of 3,581 patients were included in the final overall cohort and 2,397 (66.9%) patients had the higher-risk disease (highgrade or ≥pT2). Multivariable Cox regressions showed that patients in the groups of SWT 31 to 60 days, SWT 61 to 90 days, and SWT 91 to 120 days had similar OS compared with patients who had SWT of 8 to 30 days in the overall cohort and higher-risk cohort. Patients with SWT 121 to 180 days had worse OS (HR ¼ 1.61, 95% CI: 1.19–2.19, P ¼ 0.002 in the overall cohort; HR ¼ 1.56, 95% CI: 1.11–2.20, P ¼ 0.010 in the higher-risk cohort). Conclusions: Increased SWT from diagnosis to RNU appears to be not associated with worse OS within 120 days after the diagnosis of UTUC but SWT 4 120 days may be associated with worsened survival. These findings might have important implications for trial design in the evaluation of neoadjuvant chemotherapy for UTUC and future clinical practice. r 2017 Published by Elsevier Inc.
Keywords: Upper tract urothelial carcinoma; Ureter; Renal pelvis; Time factors; Survival; Outcome assessment
1. Introduction Upper tract urothelial carcinoma (UTUC) is a rare malignancy accounting for 5% to 7% of all renal cancers and 5% to 10 % of all urothelial carcinomas [1]. Although selected low-risk disease can be managed with kidneysparing surgery, radical nephroureterectomy (RNU) (with bladder cuff removal) remains the standard of care for the majority of nonmetastatic UTUCs [2–4]. Despite radical surgery, high-risk UTUCs usually have a poor prognosis
⁎
Correspondence to: Perelman Center for Advanced Medicine, West Pavilion, 3rd Floor, 3400 Civic Center Boulevard, Philadelphia, PA 19104. Tel.: þ1-215-662-2891; fax: þ1-215-662-3955. E-mail address: [email protected] (T.J. Guzzo). http://dx.doi.org/10.1016/j.urolonc.2017.09.013 1078-1439/r 2017 Published by Elsevier Inc.
and the long-term survival for UTUC has not improved very much for the past 20 to 30 years [2,5–7]. There is an imperative need to further investigate the role of perioperative therapies, particularly chemotherapy, in the management of UTUC. Although there is a lack of level 1 evidence for neoadjuvant chemotherapy (NAC) in patients with UTUC, NAC appears to be a promising approach for selected patients with high-grade or muscle-invasive disease [8–10]. Also, with advances in imaging and diagnostic ureteroscopy technology, our ability to clinically stage UTUC has improved. However, the lack of level 1 evidence for NAC in UTUC and the potential for delaying definitive surgical therapy in patients who may not respond continues to be a significant concern.
2
L. Xia et al. / Urologic Oncology: Seminars and Original Investigations ] (2017) ∎∎∎–∎∎∎
It is intuitive that excessive SWT for cancer surgery may cause disease progression and result in unfavorable prognosis. It is also likely that different cancers have different growth rates and biological behavior. Therefore, it is currently unknown as to whether there is a potential “safe window” of SWT for specific cancers [11,12]. Some would argue that definitive surgery for aggressive cancers such as UTUC should be performed with the shortest SWT possible and it is a moot point to investigate the so-called “safe window” [3]. However, in the modern health care system, SWT may be inevitable and the fact that UTUC is potentially aggressive raise the need to evaluate the impact of SWT on survival outcomes. The results may provide guidance regarding the preoperative assessment timeline. Also, if the “safe window” is long enough, it may partially alleviate the concerns regarding delaying RNU for legitimate reasons such as NAC in practice and clinical trials. It can also add valuable information for institutions and health care policy makers in setting parameters for wait times in various clinical conditions. There is limited data regarding the impact of SWT on the survival outcomes in patients with UTUC [13–15]. Due to small sample sizes, previous studies categorize patients into only 2 groups (early RNU vs. delayed RNU), which can potentially bias the results. Given these drawbacks, there remains no consensus about the “safe window” or cutoff for removal of tumors after diagnosis [2–4]. Given the fact that UTUC is a relatively rare malignancy, the current question is unlikely to be answered in a prospective trial or singleinstitution retrospective study. We hypothesized that there may be a “safe window” in terms of the effect of SWT on UTUC outcome and the question should be answered with the nonbinary categorization of SWT. We, therefore, investigated the association of SWT and outcome using the National Cancer Database (NCDB). NCDB is a wellvalidated database for cancer outcomes research including the effect of SWT on survival [16–18]. In this study, we took the advantages of prospectively collected data and large sample size of NCDB analyzing the impact of SWT on survival after UTUC.
long-term survival [18]. This study used de-identified data and institutional review board approval was not required. 2.2. Study population and variables
2. Materials and methods
All patients with UTUC in the NCDB from 2004 to 2013 were identified. Urothelial histology was determined via International Classification of Diseases for Oncology, Third Edition, codes 8120 to 8139. The study population was defined as patients who underwent RNU (Code 40 in the NCDB). We excluded patients with metastatic disease at the time of diagnosis, those with other cancer diagnoses, those who died within 90 days of RNU, those who had NAC, those who had other treatments other than chemotherapy, and those with missing data in relevant co-variables or outcomes of interest. SWT was defined as the number of days between the date of initial diagnosis and the date on which the RNU was performed. We also excluded cases with RNU occurred on the same day as the diagnosis and patients who had SWT more than 180 days considering the rare occurrence in the database. Co-variables included age, sex, race, reporting facility type, income level, education level, comorbidity status measured by the Charlson/Deyo score (CDS, categorized as 0, 1, and ≥ 2), tumor location (renal pelvis vs. ureter), tumor grade, pT stage, pN stage, lymph node dissection (LND), and adjuvant chemotherapy (AC). We used the general cancer grading system (grade 1—well differentiated, grade 2—moderately differentiated, grade 3—poorly differentiated, and grade 4—undifferentiated) in our study. Although both general grading system and 2004 World Health Organization/International Society of Urological Pathology (WHO/ISUP) system (low-grade vs. high-grade) were included in the NCDB recently, WHO/ISUP grading system had significantly more missing values. As described and confirmed by previous studies, grades 1 to 2 can be considered as WHO/ISUP low-grade tumors and grades 3 to 4 as WHO/ISUP high-grade tumors [19–21]. The primary outcome of interest was overall survival (OS). Follow-up time was defined as the number of months between the date of diagnosis and the date on which the patient was last contacted or died.
2.1. Data source
2.3. Statistical analysis
Data of current study was derived from the NCDB. The NCDB is a joint project of the American Cancer Society and the Commission on Cancer (CoC) of the American College of Surgeons. The NCDB, established in 1989, is a nationwide, hospital-based, comprehensive clinical surveillance resource oncology data set that currently captures 70% of all newly diagnosed malignancies in the United States annually. NCDB contains comprehensive data about reporting facility, patient demographics, cancer identification, disease stage, treatments, short-term outcomes, and
The primary objective of the study was to assess the association between SWT and OS. Time intervals between the date of initial diagnosis and date of RNU were set at 30day increments, with the last 2 intervals combined owing to smaller numbers in each. Also, the first interval was separated into 2 groups with the consideration that small SWT may indicate nonelective surgery and may have worse outcomes [22]. Our a priori analysis also showed less favorable OS in patients who had SWT ≤ 7 days vs. patients who had SWT of 8 to 30 days. Therefore, SWT
L. Xia et al. / Urologic Oncology: Seminars and Original Investigations ] (2017) ∎∎∎–∎∎∎
was ultimately categorized into 6 groups: SWT ≤ 7 days, SWT 8 to 30 days, SWT 31 to 60 days, SWT 61 to 90 days, SWT 91 to 120 days, and SWT 121 to 180 days. Baseline demographics and tumor characteristics were described. Unadjusted OS was estimated and assessed by the Kaplan–Meier curves and univariable Cox regression analysis. Data were reported as medians (interquartile ranges [IQRs]) and frequencies (percentages) for continuous and categorical variables, respectively. Multivariable Cox regression analysis was performed to assess the associations between all the variables and OS. Hazard ratios (HRs) with 95% confidence intervals (CIs) were used to assess the strength of the associations. All tests were 2-sided and a P o 0.05 was considered statistically significant. All analyses were performed using STATA 14 (StataCorp LP, College Station, TX).
2.4. Primary analysis We performed the primary multivariable Cox regression analyses in the overall included cohort and a subgroup of patients with higher-risk disease (grades 3– 4 or ≥pT2) [4]. Given the difficulty of clinically staging UTUC before RNU and the fact that tumor grade in the NCDB is based on final pathologic diagnosis, we chose to use final pathologic T stage and grade to define the higher-risk subgroup. In the primary analyses, SWT was considered as a categorical variable as described earlier and SWT 8 to 30 days was used as the reference. A sensitivity analysis excluding patients with SWT ≤ 7 days was also performed.
2.5. Secondary analysis Based on the results of primary analyses, we then performed the secondary Cox regressions (univariable and multivariable) with SWT considered as a continuous variable (per day increase) in the overall cohort and higher-risk cohort with the SWT in the range of 1 to 120 days and 8 to 120 days.
2.6. Post-hoc analysis Based on the results of primary analyses and secondary analyses, we did a post-hoc analysis comparing patient characteristics between SWT of 1 to 120 days and SWT of 121 to 180 days. The objective was to explore if there are any differences in any variables between the 2 groups. Continuous variables were compared using Mann–Whitney U test and categorical variables were compared using Fisher’s exact test.
3
3. Results 3.1. Study cohorts A total of 3,581 patients were included in the final overall cohort. Characteristics of the overall cohort are presented in Table 1. Median age was 71 years old (interquartile range [IQR] 63–78 years); 1,543 (43.1%) patients were female and 3,347 patients (93.5%) were white. In the overall cohort, 230 (6.4%) patients had SWT ≤ 7 days, 1,628 (45.5%) patients had SWT ≤ 30 days, 2,878 (80.4%) patients had SWT ≤ 60 days, 3,350 (93.5%) patients had SWT ≤ 90 days, and 3,493 (97.5%) patients had SWT ≤ 120 days. The median follow-up was 40.4 months (IQR: 23.3–65.2 months). Three-year OS and 5-year OS for the overall cohort were 74.0% (95% CI: 72.4%–75.5%) and 61.9% (95% CI: 60.0%–63.7%), respectively. A total 2,397 (66.9%) patients had higher-risk disease based on our predetermined criteria (grades 3–4 or ≥ pT2). Three-year OS and 5-year OS for the higher-risk cohort were 66.6% (95% CI: 64.5%–68.5%) and 54.3% (95% CI: 51.9%–56.6%), respectively. 3.2. Unadjusted OS Kaplan–Meier survival estimates and univariable Cox regression analyses of the association between SWT and OS are shown in Fig. 1. In the overall cohort (n ¼ 3,581), compared with patients having SWT of 8 to 30 days, patients having SWT of 1 to 7 days had worse OS (HR ¼ 1.29, 95% CI: 1.03–1.62, P ¼ 0.029); patients having SWT of 121 to 180 days also had worse OS (HR ¼ 2.05, 95% CI: 1.52–2.77, P o 0.001). In the higher-risk cohort (n ¼ 2,397), compared with patients having SWT of 8 to 30 days, only patients having SWT of 121 to 180 days had worse OS (HR ¼ 2.11, 95% CI: 1.51–2.95, P o 0.001). 3.3. Primary analysis Results of primary multivariable Cox regression analysis adjusting for all the co-variables are shown in Table 2. In the overall cohort (n ¼ 3,581), compared with patients having SWT of 8 to 30 days, patients having SWT of 1 to 7 days had worse OS (HR ¼ 1.32, 95% CI: 1.06–1.67, P ¼ 0.016); patients having SWT of 121 to 180 days also had worse OS (HR ¼ 1.61, 95% CI: 1.19–2.19, P ¼ 0.002). In the higher-risk cohort (n ¼ 2,397), compared with patients having SWT of 8 to 30 days, only patients having SWT of 121 to 180 days had worse OS (HR ¼ 1.56, 95% CI: 1.11–2.20, P ¼ 0.010). Results of sensitivity analyses in the overall cohort and higher-risk cohort excluding SWT of 1 to 7 days are shown in Table 3. The sensitivity analyses confirmed the results of primary analyses and showed that patients having SWT of 121 to 180
L. Xia et al. / Urologic Oncology: Seminars and Original Investigations ] (2017) ∎∎∎–∎∎∎
4
Table 1 Baseline characteristics of the overall included cohort, stratified by SWT Variable
All patients (n ¼ 3,581) SWT (days) 1–7 (n ¼ 230) 8–30 (n ¼ 1,398) 31–60 (n ¼ 1,250) 61–90 (n ¼ 472) 91–120 (n ¼ 143) 121–180 (n ¼ 88)
Age, years 71 Sex Male 2,038 Female 1,543 Race White 3,347 Black 137 Other 97 Facility type Community 1,929 Academic 1,652 Median incomea o$48,000 1,395 ≥$48,000 2,186 Education levelb ≥13% 1,359 o13% 2,222 CDS 0 2,402 1 893 ≥2 286 Tumor location Renal pelvis 2,428 Ureter 1,153 Tumor grade 1–2 1,273 3–4 2,308 pT stage opT2 1,865 ≥pT2 1,492 pTx 224 pN stage pN0 565 pNþ 147 pNx 2,869 LND No 2,845 Yes 736 AC No 3,224 Yes 357 a
(63–78)
71 (63–79)
70 (63–78)
70 (62–78)
72 (64–80)
73 (64–80)
73 (67–80)
(56.9%) (43.1%)
131 (57.0%) 99 (43.0%)
758 (54.2%) 640 (45.8%)
747 (59.8%) 503 (40.2%)
255 (54.0%) 217 (46.0%)
89 (62.2%) 54 (37.8%)
58 (65.9%) 30 (34.1%)
(93.5%) (3.8%) (2.7%)
208 (90.4%) 9 (3.9%) 13 (5.7%)
1,326 (94.9%) 39 (2.8%) 33 (2.4%)
1,162 (93.0%) 53 (4.2%) 35 (2.8%)
441 (93.4%) 21 (4.5%) 10 (2.1%)
129 (90.2%) 10 (7.0%) 4 (2.8%)
81 (92.1%) 5 (5.7%) 2 (2.3%)
(53.9%) (43.1%)
119 (51.7%) 111 (48.3%)
798 (57.1%) 600 (42.9%)
680 (54.4%) 570 (45.6%)
232 (49.2%) 240 (50.9%)
63 (44.1%) 80 (55.9%)
37 (42.1%) 51 (57.9%)
(39.0%) (61.0%)
90 (39.1%) 140 (60.9%)
551 (39.4%) 847 (60.6%)
486 (38.9%) 764 (61.1%)
182 (38.6%) 290 (61.4%)
47 (32.9%) 96 (67.1%)
39 (44.3%) 49 (55.7%)
(38.0%) (62.0%)
83 (36.1%) 147 (63.9%)
508 (36.3%) 890 (63.7%)
482 (38.6%) 768 (61.4%)
184 (39.0%) 288 (61.0%)
61 (42.7%) 82 (57.3%)
41 (46.6%) 47 (53.4%)
(67.1%) (24.9%) (8.0%)
151 (65.7%) 57 (24.8%) 22 (9.6%)
977 (69.9%) 325 (23.3%) 96 (6.9%)
838 (67.0%) 309 (24.7%) 103 (8.2%)
305 (64.6%) 125 (26.5%) 42 (8.9%)
80 (55.9%) 48 (33.6%) 15 (10.5%)
51 (58.0%) 29 (33.0%) 8 (9.1%)
(67.8%) (32.2%)
170 (73.9%) 60 (26.1%)
946 (67.7%) 452 (32.3%)
824 (65.9%) 426 (34.1%)
317 (67.2%) 155 (32.8%)
104 (72.7%) 39 (27.3%)
67 (76.1%) 21 (23.9%)
(35.6%) (64.4%)
66 (28.7%) 164 (71.3%)
482 (34.5%) 916 (65.5%)
467 (37.4%) 783 (62.6%)
173 (36.7%) 299 (63.4%)
61 (42.7%) 82 (57.3%)
24 (27.3%) 64 (72.7%)
(52.1%) (41.7%) (6.3%)
116 (50.4%) 102 (44.4%) 12 (5.2%)
682 (48.8%) 636 (45.5%) 80 (5.7%)
680 (54.4%) 489 (39.1%) 81 (6.5%)
273 (57.8%) 169 (35.8%) 30 (6.4%)
80 (55.9%) 52 (36.4%) 11 (7.7%)
34 (38.6%) 44 (50.0%) 10 (11.4%)
(15.8%) (4.1%) (80.1%)
41 (17.8%) 12 (5.2%) 177 (77.0%)
212 (15.2%) 66 (4.7%) 1,120 (80.1%)
195 (15.6%) 41 (3.3%) 1,014 (81.1%)
81 (17.2%) 18 (3.8%) 373 (79.0%)
23 (16.1%) 5 (3.5%) 115 (80.4%)
13 (14.8%) 5 (5.7%) 70 (79.6%)
(79.5%) (20.5%)
175 (76.1%) 55 (23.9%)
1,111 (79.5%) 287 (20.5%)
1,005 (80.4%) 245 (19.6%)
373 (79.0%) 99 (21.0%)
114 (79.7%) 29 (20.3%)
67 (76.1%) 21 (23.9%)
(90.0%) (10.0%)
206 (89.6%) 24 (10.4%)
1,231 (88.1%) 167 (12.0%)
1,131 (90.5%) 119 (9.5%)
439 (93.0%) 33 (7.0%)
138 (96.5%) 5 (3.5%)
79 (89.8%) 9 (10.2%)
Household income for each patient’s area of residence. The number of adults in each patient’s area of residence who did not graduate from high school.
b
days had worse OS compared with patients having SWT of 8 to 30 days (overall cohort [n ¼ 3,351]: HR ¼ 1.65, 95% CI: 1.20–2.20, P ¼ 0.002; higher-risk cohort [n ¼ 2,232]: HR ¼ 1.58, 95% CI: 1.12–2.23, P ¼ 0.008). 3.4. Secondary analysis Based on the results of primary analyses, we performed the secondary analyses with SWT considered as a continuous variable in patients with SWT of 1 to 120 days and SWT of 8 to 120 days. All the analyses were separately performed in the overall cohort and higher-risk cohort. The results of secondary univariable and multivariable Cox
regression analyses are shown in Supplementary Table 1 (SWT range: 1–120 days) and Supplementary Table 2 (SWT range: 8–120 days). Increased SWT (per day) was not shown to be associated with worse OS in the overall/ higher-risk cohorts with SWT of 1 to 120 days or SWT of 8 to 120 days. 3.5. Post-hoc analysis Based on our primary and secondary analyses showing that SWT within 120 days was the potential cutoff in terms of OS difference, we did the post-hos analysis comparing patient characteristics between SWT of 1 to 120 days
L. Xia et al. / Urologic Oncology: Seminars and Original Investigations ] (2017) ∎∎∎–∎∎∎ Overall cohort (n=3,581)
Higher-risk cohort (n=2,397) 1.00 Overall survival
Overall survival
1.00 0.75 0.50 0.25 0.00
0.75 0.50 0.25 0.00
0
12
24
36
48
60
72
84
96
108
120
0
12
24
36
Follow-up (months) Number at risk SWT 1-7 d SWT 8-30 d SWT 31-60 d SWT 61-90 d SWT 91-120 d SWT 121-180 d
5
230 1398 1250 472 143 88
186 1256 1144 435 130 77
147 1018 951 344 104 54
116 799 701 247 79 33
92 620 524 188 54 22
74 428 371 136 42 17
40 277 247 92 29 12
48
60
72
84
96
108
120
15 113 65 31 6 3
9 48 24 15 2 2
1 16 5 2 0 0
0 0 0 0 0 0
Follow-up (months)
21 180 151 60 11 8
13 78 58 24 4 3
2 24 14 4 1 0
0 0 0 0 0 0
Number at risk SWT 1-7 d SWT 8-30 d SWT 31-60 d SWT 61-90 d SWT 91-120 d SWT 121-180 d
165 963 811 308 86 64
124 838 721 275 75 55
93 645 574 211 56 33
75 486 396 145 38 17
57 378 282 104 26 10
46 258 187 70 19 8
24 167 120 45 14 4
SWT 1-7 d
SWT 8-30 d
SWT 1-7 d
SWT 8-30 d
SWT 31-60 d
SWT 61-90 d
SWT 31-60 d
SWT 61-90 d
SWT 91-120 d
SWT 121-180 d
SWT 91-120 d
SWT 121-180 d
SWT 8-30 d 1-7 d 31-60 d 61-90 d 91-120 d 121-180 d
HR (95%CI) Reference 1.29 (1.03-1.62) 1.04 (0.91-1.18) 1.13 (0.95-1.35) 1.06 (0.78-1.42) 2.05 (1.52-2.77)
P value 0.029 0.584 0.174 0.724 < 0.001
SWT 8-30 d 1-7 d 31-60 d 61-90 d 91-120 d 121-180 d
HR (95%CI) Reference 1.17 (0.90-1.51) 1.08 (0.93-1.25) 1.17 (0.96-1.43) 1.15 (0.81-1.62) 2.11 (1.51-2.95)
P value 0.251 0.339 0.121 0.442 < 0.001
Fig. 1. Kaplan–Meier curves and univariable Cox regression of the association between SWT and OS in the overall cohort and higher-risk cohort.
(n ¼ 3,493) and SWT of 121 to 180 days (n ¼ 88). The results showed that compared with SWT 1 to 120 days, patients in the SWT 121 to 180 days group were significantly older (73 y vs. 71 y, P ¼ 0.005), more likely to get treated in the academic centers (57.9% vs. 45.8%, P ¼ 0.030), and more likely to have ≥pT2 disease (50.0% vs. 41.5%) (Supplementary Table 3). There were no significant differences in Charlson/Deyo score or tumor grade.
4. Discussion To the best of our knowledge, this is the first study to evaluate the impact of SWT on the survival of UTUC patients in a large national facility-based database. Our results showed that SWT of RNU appeared to be not associated with OS within 120 days of diagnosis. The absence of an association in the 120-day window was consistent in both the overall cohort and a subgroup of patients with higher-risk disease (grades 3–4 or ≥pT2). However, SWT 4 120 days appeared to be associated with worsened OS. There are controversies in previously published reports about the prognostic significance of an increasing SWT in UTUC [13–15]. To be noted, previous studies were singleor multicenter retrospective studies with small sample sizes. Patients were generally categorized into 2 groups (early RNU vs. delayed RNU) by a single-interval cutoff value. Lee et al. [13,23] included 138 patients with UTUC and used the SWT cutoff of 1 month (30.5 days). The results showed that delayed RNU was associated with worse cancer-specific survival (CSS) and recurrence-free survival (RFS) in the ureteral tumor cohort (n = 80) but not in the overall UTUC cohort [13,23]. A study from MD Anderson
Cancer Center (n = 240) used the 3-month cutoff and demonstrated no significant differences in RFS, CSS, and OS between the early and delayed RNU [14]. Another multicenter study also showed no statistical differences in CSS and RFS between patients who had RNU at o3 months vs. ≥3 months after diagnosis in both the overall cohort (n ¼ 187) and a subgroup of patients (n ¼ 90, 48.1%) with muscle-invasive disease (≥pT2) [15]. The authors also analyzed the SWT as a continuous variable (per month) and the results suggested that a delay in the interval from diagnosis to RNU is associated with more advanced disease stage but not RFS or CSS [15]. Compared with previous studies in the literature, our study has 2 features in terms of study design. First, a significantly larger sample size not only provides more statistical power but also allows for analysis of multiple delay groups of varying lengths. We chose the 30-day increment and considered 1 to 7 days as a separate group based on previous national database studies [16,22,24]. The 30-day interval was also partially in line with SWT cutoffs of previously described reports [13–15]. Second, analyses are more extensive in our study. We repeated all the analyses in the higher-risk cohort to make the results more informative. Sensitivity analyses and secondary analyses were also performed to ensure the results are more reliable. Overall, the current study has advantages and contributes to the existing literature on this topic with some novel findings. As it is unlikely that a randomized clinical trial investigating the impact of SWT on cancer outcomes will ever occur, our analyses of the largest cancer database from the United States may provide the most demonstrative evidence on this topic. Although there is the possibility that higher-risk/sicker patients had shorter SWT than lower-risk/healthier patients,
6
L. Xia et al. / Urologic Oncology: Seminars and Original Investigations ] (2017) ∎∎∎–∎∎∎
Table 2 Multivariable Cox regression in the overall cohort and higher-risk cohort. Variable
SWT 8–30 d 1–7 d 31–60 d 61–90 d 91–120 d 121–180 d Age, per 1 y Sex Male Female Race White Black Other Facility type Community Academic Median income o$48,000 ≥$48,000 Education level ≥13% o13% CDS 0 1 ≥2 Tumor location Renal pelvis Ureter Tumor grade 1–2 3–4 pT stage opT2 ≥pT2 pTx pN stage pN0 pNþ pNx LND No Yes AC No Yes
Overall cohort (n ¼ 3,581)
Higher-risk cohort (n ¼ 2,397)
HR (95% CI)
P value
HR (95% CI)
P value
Reference 1.32 (1.06–1.67) 1.11 (0.97–1.27) 1.09 (0.91–1.30) 1.00 (0.74–1.35) 1.61 (1.19–2.19) 1.049 (1.043–1.056)
0.016 0.126 0.360 0.976 0.002 o0.001
Reference 1.24 (0.95–1.61) 1.10 (0.94–1.27) 1.07 (0.88–1.31) 0.94 (0.66–1.34) 1.56 (1.11–2.20) 1.043 (1.036–1.051)
0.114 0.231 0.510 0.744 0.010 o0.001
Reference 0.87 (0.78–0.98)
0.017
Reference 0.88 (0.77–1.00)
0.046
Reference 1.05 (0.78–1.42) 0.94 (0.67–1.32)
0.737 0.735
Reference 1.00 (0.71–1.40) 0.92 (0.65–1.32)
0.987 0.667
Reference 1.04 (0.92–1.16)
0.546
Reference 1.02 (0.89–1.16)
0.775
Reference 0.95 (0.84–1.09)
0.465
Reference 1.01 (0.87–1.18)
0.853
Reference 0.91 (0.80–1.04)
0.153
Reference 0.85 (0.73–0.98)
0.030
Reference 1.21 (1.06–1.38) 1.70 (1.42–2.04)
0.004 o0.001
Reference 1.27 (1.10–1.47) 1.64 (1.32–2.04)
0.001 o0.001
Reference 1.05 (0.93–1.18)
0.440
Reference 1.07 (0.93–1.23)
0.332
Reference 1.38 (1.20–1.59)
o0.001
Reference 1.76 (1.21–2.54)
0.003
Reference 2.46 (2.14–2.82) 1.71 (1.36–2.14)
o0.001 o0.001
Reference 2.76 (2.33–3.27) 2.25 (1.70–2.98)
o0.001 o0.001
Reference 2.74 (2.10–3.57) 1.59 (0.85–2.97)
o0.001 0.150
Reference 2.44 (1.85–3.21) 1.45 (0.72–2.92)
o0.001 0.296
Reference 1.29 (0.70–2.38)
0.407
Reference 1.23 (0.63–2.43)
0.545
Reference 1.15 (0.96–1.37)
0.140
Reference 1.10 (0.91–1.33)
0.306
we adjusted for patient-related, tumor-related, and facilityrelated factors in the multivariable regression models and the results were consistent. Given the fact that there was not a substantial difference in outcome with SWT in the 120day window, this argues in favor of the use of NAC as it is unlikely to result in worse surgical outcomes in patients who may not respond to systemic therapy [8–10]. Compared with AC, one obvious advantage of NAC is that
significantly more patients would be eligible for cisplatinbased multiagent chemotherapy. Studies have consistently shown that estimated glomerular filtration rate (eGFR) is significantly diminished after RNU, which may eliminate cisplatin-based chemotherapy as a therapeutic option in about half of the high-risk patients if it is deferred to the adjuvant setting [25,26]. Other potential advantages of NAC could include the early eradication of micrometastases
L. Xia et al. / Urologic Oncology: Seminars and Original Investigations ] (2017) ∎∎∎–∎∎∎ Table 3 Sensitivity analysis–multivariable Cox regression in the overall cohort and higher-risk cohort excluding SWT 1 to 7 days SWT
8–30 d 31–60 d 61–90 d 91–120 d 121–180 d
Overall cohort (n ¼ 3,351)
Higher-risk cohort (n ¼ 2,232)
HR (95% CI)
P value
HR (95% CI)
P value
Reference 1.11 (0.97–1.27) 1.09 (0.92–1.30) 1.02 (0.75–1.38) 1.65 (1.20–2.20)
0.126 0.322 0.911 0.002
Reference 1.10 (0.94–1.28) 1.08 (0.88–1.32) 0.97 (0.68–1.38) 1.58 (1.12–2.23)
0.221 0.461 0.862 0.008
Adjusted for the same co-variables in Table 2.
and tumor down-staging [8–10]. The level 1 evidence in favor of NAC for muscle-invasive bladder cancer can be considered as a further justification for giving chemotherapy in the neoadjuvant setting [27]. From the future research standpoint, multicenter collaborative prospective trials to investigate NAC for UTUC patients may be the optimal setting given the rarity of UTUC. Similarly, our results may also further support the safety of meticulous risk-stratification for UTUC, most notably by ureteroscopic evaluation. Previous studies have shown that ureteroscopy with biopsy and/or tumor ablation before RNU did not adversely affect the postoperative disease status [28,29]. There are several limitations that should be considered when interpreting our study. First, heterogeneity of the initial diagnosis date in the NCDB might bias the results. The date of initial diagnosis in the NCDB is defined as the date of first clinical or histologic confirmation or if in retrospect, the patient had cancer at an earlier date [30]. We strictly used the time between the date of initial diagnosis and the date of RNU as SWT considering that it might be the optimal method to define SWT in the NCDB [16,17,30]. NCDB is a well-validated administrative database and initial diagnosis date in the real-world practice may suggest it is the date when RNU was indicated or at least considered [18]. Second, there is no information regarding causes of delay in surgery [17]. Causes of delay may relate to various factors and those factors may also be the confounding factors of survival. Third, other important endpoint outcomes such as disease progression, DR, and CSS are not available in the NCDB, which further weakens our study. Finally, NCDB is a hospital-based clinical cancer registry but not population-based. Cancer patients are captured into the NCDB on the basis of the hospital where they get the diagnosis and/or treatment. The inherent biases of NCDB may limit the interpretation of the results in the general population.
5. Conclusions Our results suggest that increased SWT from diagnosis to RNU appears to be not associated with worse OS within
7
120 days after the diagnosis of UTUC but SWT 4 120 days may be associated with worsened survival. These findings might have important implications for trial design in the evaluation of NAC for UTUC and future clinical practice. Nonetheless, further prospective studies assessing cancer-specific outcomes are required to validate our findings. Acknowledgments The data used in this study are derived from a deidentified NCDB file. The American College of Surgeons and the Commission on Cancer have not verified and are not responsible for the analytic or statistical methodology employed, or the conclusions drawn from these data by the investigators. Appendix A. Supplementary material Supplementary data are available in the online version of this article at http://dx.doi.org/10.1016/j.urolonc.2017.09. 013.
References [1] Soria F, Shariat SF, Lerner SP, et al. Epidemiology, diagnosis, preoperative evaluation and prognostic assessment of upper-tract urothelial carcinoma (UTUC). World J Urol 2017;35:379–87. [2] Roupret M, Babjuk M, Comperat E, et al. European Association of Urology guidelines on upper urinary tract urothelial cell carcinoma: 2015 update. Eur Urol 2015;68:868–79. [3] Roupret M, Babjuk M, Comperat E, et al. European guidelines on upper tract urothelial carcinomas: 2013 update. Eur Urol 2013;63: 1059–71. [4] Hu M, Jacobs BL, Montgomery JS, et al. Sharpening the focus on causes and timing of readmission after radical cystectomy for bladder cancer. Cancer 2014;120:1409–16. [5] Leow JJ, Orsola A, Chang SL, Bellmunt J. A contemporary review of management and prognostic factors of upper tract urothelial carcinoma. Cancer Treat Rev 2015;41:310–9. [6] Eylert MF, Hounsome L, Verne J, Bahl A, Jefferies ER, Persad RA. Prognosis is deteriorating for upper tract urothelial cancer: data for England 1985–2010. BJU Int 2013;112:E107–13. [7] Brown GA, Busby JE, Wood CG, et al. Nephroureterectomy for treating upper urinary tract transitional cell carcinoma: time to change the treatment paradigm? BJU Int 2006;98:1176–80. [8] Leow JJ, Martin-Doyle W, Fay AP, Choueiri TK, Chang SL, Bellmunt J. A systematic review and meta-analysis of adjuvant and neoadjuvant chemotherapy for upper tract urothelial carcinoma. Eur Urol 2014;66:529–41. [9] Matin SF, Margulis V, Kamat A, et al. Incidence of downstaging and complete remission after neoadjuvant chemotherapy for high-risk upper tract transitional cell carcinoma. Cancer 2010;116:3127–34. [10] Porten S, Siefker-Radtke AO, Xiao L, et al. Neoadjuvant chemotherapy improves survival of patients with upper tract urothelial carcinoma. Cancer 2014;120:1794–9. [11] Bourgade V, Drouin SJ, Yates DR, et al. Impact of the length of time between diagnosis and surgical removal of urologic neoplasms on survival. World J Urol 2014;32:475–9.
8
L. Xia et al. / Urologic Oncology: Seminars and Original Investigations ] (2017) ∎∎∎–∎∎∎
[12] Neal RD, Tharmanathan P, France B, et al. Is increased time to diagnosis and treatment in symptomatic cancer associated with poorer outcomes? Systematic review. Br J Cancer 2015;112(Suppl. 1): S92–S107. [13] Lee JN, Kwon SY, Choi GS, et al. Impact of surgical wait time on oncologic outcomes in upper urinary tract urothelial carcinoma. J Surg Oncol 2014;110:468–75. [14] Sundi D, Svatek RS, Margulis V, et al. Upper tract urothelial carcinoma: impact of time to surgery. Urol Oncol 2012;30:266–72. [15] Waldert M, Karakiewicz PI, Raman JD, et al. A delay in radical nephroureterectomy can lead to upstaging. BJU Int 2010;105:812–7. [16] Bleicher RJ, Ruth K, Sigurdson ER, et al. Time to surgery and breast cancer survival in the United States. JAMA Oncol 2016;2:330–9. [17] Strohl AE, Feinglass JM, Shahabi S, Simon MA. Surgical wait time: a new health indicator in women with endometrial cancer. Gynecol Oncol 2016;141:511–5. [18] Boffa DJ, Rosen JE, Mallin K, et al. Using the National Cancer Database for outcomes research: a review. JAMA Oncol 2017. http://dx. doi.org/10.1001/jamaoncol.2016.6905:[Epub ahead of print]. [19] Matulewicz RS, Frainey BT, Oberlin DT, Meeks JJ. High-risk of adverse pathologic features in patients with clinical T1 high-grade bladder cancer undergoing radical cystectomy. J Natl Compr Canc Netw 2016;14:1403–11. [20] Chappidi MR, Kates M, Johnson MH, Hahn NM, Bivalacqua TJ, Pierorazio PM. Lymph node yield and tumor location in patients with upper tract urothelial carcinoma undergoing nephroureterectomy affects survival: a U.S. population-based analysis (2004–2012). Urol Oncol 2016;34:531 e15–531, e24. [21] Charlton ME, Adamo MP, Sun L, Deorah S. Bladder cancer collaborative stage variables and their data quality, usage, and clinical
[22] [23] [24]
[25]
[26]
[27]
[28]
[29]
[30]
implications: a review of SEER data, 2004-2010. Cancer 2014;120 (Suppl. 23):3815–25. Elit LM, O’Leary EM, Pond GR, Seow HY. Impact of wait times on survival for women with uterine cancer. J Clin Oncol 2014;32:27–33. Roupret M. Surgery: surgical wait times for UTUC—a race against the clock? Nat Rev Urol 2014;11:665–6. Kulkarni GS, Urbach DR, Austin PC, Fleshner NE, Laupacis A. Longer wait times increase overall mortality in patients with bladder cancer. J Urol 2009;182:1318–24. Kaag MG, O'Malley RL, O'Malley P, et al. Changes in renal function following nephroureterectomy may affect the use of perioperative chemotherapy. Eur Urol 2010;58:581–7. Lane BR, Smith AK, Larson BT, et al. Chronic kidney disease after nephroureterectomy for upper tract urothelial carcinoma and implications for the administration of perioperative chemotherapy. Cancer 2010;116:2967–73. Advanced Bladder Cancer Meta-analysis Collaboration: Neoadjuvant chemotherapy in invasive bladder cancer. a systematic review and meta-analysis. Lancet 2003;361:1927–34. Boorjian S, Ng C, Munver R, et al. Impact of delay to nephroureterectomy for patients undergoing ureteroscopic biopsy and laser tumor ablation of upper tract transitional cell carcinoma. Urology 2005;66:283–7. Nison L, Roupret M, Bozzini G, et al. The oncologic impact of a delay between diagnosis and radical nephroureterectomy due to diagnostic ureteroscopy in upper urinary tract urothelial carcinomas: results from a large collaborative database. World J Urol 2013;31:69–76. Bilimoria KY, Ko CY, Tomlinson JS, et al. Wait times for cancer surgery in the United States: trends and predictors of delays. Ann Surg 2011;253:779–85.