Abstract
Purpose
Population studies of patients undergoing radical cystectomy (RC) for bladder cancer (BC) suggest that a more extended lymph node dissection (LND) increases survival. However, information regarding LNDs of patients undergoing RC with a history of radiation therapy for BC is largely unknown. This study aims to define the lymph node yield (LNY) in patients undergoing RC for BC following radiation of the bladder using the surveillance epidemiology and end results (SEER) database.
Methods
Data were collected using SEER 18 registries from 1988 to 2013 to identify patients undergoing RC for BC. Data on extent and yield of LND were obtained. Logistic regression and multivariate Cox proportional hazard regression were done to identify predictors of adequate LND and all-cause mortality, respectively.
Results
In total, 27,451 patients were identified, of which, 27,362 (99.7%) were radiation naïve and 89 (0.3%) had prior radiation therapy for BC. The average LNY in radiation naïve patients (15, SD [13.5]) was slightly higher than the LNY in patients with prior radiation (12.3 SD [9.2], p = 0.157). Prior radiation was not an independent predictor of overall mortality (HR = 1.3, 95% CI [0.98–1.7]; p = 0.076).
Conclusions
A lower proportion of patients with a history of radiation underwent a LND. The LNYs of radiation naïve patients, and those with a history of radiation, were not statistically different; however, the proportion of irradiated patients was small. Further investigation will be required to elucidate the patient and provider characteristics that contribute to the similar LNYs.
Keywords: Lymphadenectomy, Lymph node dissection, Cystectomy, Radiation therapy
Introduction
Population-based studies of patients undergoing surgical treatment for bladder cancer (BC) suggest that increasing the number of lymph nodes removed increases survival [1–3]. This survival benefit is traditionally attributed to factors including the removal of undetectable micrometastatic disease in patients with both nodal and non-nodal disease [4] and better staging with more stage-appropriate care [5]. High volume centers show that node count, as an indirect measurement of lymphadenectomy extent, demonstrates a correlation with survival, particularly when a minimum of 10 nodes are removed, though debate surrounds this number [1, 5–9].
Lymph node counts after cystectomy are typically dependent on patient characteristics [10]. BC patients are increasingly being considered for bladder-sparing approaches using chemoradiotherapy regimens [11]. Radical cystectomy (RC), following failure of bladder-sparing approaches, may be met with a more prolonged and difficult dissection [12]. Prior radiation therapy creates a desmoplastic reaction in the targeted and surrounding tissues, resulting in difficulty defining tissue planes. This may adversely impact the surgery and potentially increase complication rates. Historically, morbidity and mortality rates in patients undergoing RC following radiation therapy have been high, ranging from 44 to 66%, and 15 to 33%, respectively [13, 14].
While several series have demonstrated the short-term morbidity and mortality of RC after radiation therapy, population-based information on lymph node dissection (LND) patterns and lymph node yield (LNY) in these cases is lacking. Such an investigation will help standardize lymph node obtainment patterns and illustrate the potential benefit of LND in this patient population. Thus, our main objective was to compare the LNYs of patients who undergo RC after radiation therapy to the LNYs of radiation naïve patients undergoing primary RC. Our secondary objective was to investigate the survival effect of an LND in these patients.
Methods
Data were collected using the surveillance epidemiology and end results (SEER) 18 registries from 1988 to 2010. Malignancies of the bladder were selected for using ICD-O-3 site code C670–C679. Radiation treatment information for each case was obtained, and cases with external beam radiation for previously diagnosed BC were selected. BC-specific surgical treatment information was obtained using the Site Specific Surgery variable (codes 40, 50, 60 and 70 were defined as RC) and the RX Summary Variable (codes 50, 60–64, 70–74, and 80 were defined as RC). Only cases undergoing RC at some point for BC were included, and grouped based on radiation history. Patients undergoing RC without any history of radiation therapy before or after cystectomy were defined as “radiation naïve” patients. Patients undergoing RC with a history of prior external beam radiation therapy for BC were defined as “irradiated” patients.
LND and LNY information was obtained using the variable, Regional Nodes Examined (00 defined as a case with no LND, 01–90 defined as a LND with known LNY, and 96–98 equal to a LND with an unknown LNY). For analytic purposes, we considered a LNY ≥ 10 nodes as an adequate dissection.
All histology types of BC were included with ICD-O-3 codes 8020–8031 defined as transitional cell carcinoma (TCC), 8070–8078 defined as squamous cell carcinoma (SCC), and the remaining codes defined as “Other.” Cancer grade was obtained using SEER extent of disease 10 and SEER American Joint Committee on Cancer variables. Demographic information obtained included age at BC diagnosis, gender, race, and SEER county-specific rural–urban classification with population information.
The primary objectives of this study were to compare the LNYs of those undergoing RC with and without prior bladder radiation. This also includes the proportion of patients who had an adequate LND. The secondary objectives were to determine predictors of undergoing an adequate LND, and the subsequent ramifications on survival.
Nominal variables and proportion comparisons were evaluated with Chi-square analysis. Univariate and multivariate logistic regression were used to determine variables associated with a LNY ≥ 10 nodes. Kaplan–Meier curves with log-rank tests were obtained to determine overall survival based on LNY and radiation history. For multivariate survival analysis, Cox proportional hazard ratios were used to compare the relative risk of overall mortality. All statistical analyses and data management were conducted using SPSS Version 21.0 (Chicago, IL) and R Version 3.3.1 (Vienna, Austria).
Results
Patient characteristics
A total of 89 patients undergoing RC following radiation therapy were identified, and compared to 27,362 radiation naïve patients undergoing RC. The median follow-up time was 28 months [IQR 11–75]. Table 1 shows the baseline patient and tumor characteristics in the aforementioned two groups. No significant difference was seen in age (p = 0.811), gender (p = 0.191), race (p = 0.824), or in registered county population characteristics (p = 0.684). Compared to radiation naïve patients, irradiated patients presented with significantly more stage T3 (27.3 vs. 17.5%) and stage T4 disease (9.1 vs. 3.8%; p = 0.012). Nodal disease was not significantly different between irradiated and radiation naïve patients (13.6 vs. 17.6%; p = 0.4). No difference in tumor grade was observed; however, irradiated patients did demonstrate a higher rate of metastatic disease (9 vs. 4.7%; p = 0.001).
Table 1.
Patient characteristics
| Covariate | Radiation naive | Irradiated | p value |
|---|---|---|---|
| N (%) | 27,362 | 89 | – |
| Age, years | |||
| < 69 | 14,719 (53.8) | 49 (55.1) | 0.811 |
| ≥ 70 | 12,643 (46.2) | 40 (44.9) | |
| Gender | |||
| Male | 20,124 (73.5) | 60 (67.4) | 0.191 |
| Race | |||
| White | 24,382 (89.1) | 82 (92.1) | 0.824 |
| County classification | |||
| Metro county | 2431 (8.9) | 9 (10.1) | 0.684 |
| Non-metro county | 24,931 (91.1) | 80 (89.8) | |
| Histology | |||
| TCC | 24,229 (88.5) | 79 (87.6) | 0.931 |
| SCC | 1258 (4.6) | 4 (4.5) | |
| Other | 1875 (6.9) | 7 (7.9) | |
| T stage (N0 M0) | |||
| Ta | 461 (1.7) | 0 (0) | 0.012 |
| Tis | 225 (0.8) | 2 (2.3) | |
| T1 | 3008 (11) | 19 (8) | |
| T2 | 9447 (34.7) | 28 (31.8) | |
| T3 | 4771 (17.5) | 24 (27.3) | |
| T4 | 1036 (3.8) | 8 (9.1) | |
| Nodal disease (M0) | |||
| N + | 4798 (17.6) | 12 (13.6) | 0.4 |
| Metastatic disease | |||
| M + | 1276 (4.7) | 8 (9) | 0.001 |
| Grade | |||
| I | 332 (1.3) | 1 (1.3) | 0.864 |
| II | 1977 (7.6) | 6 (7.7) | |
| III | 11,401 (44.1) | 38 (48.7) | |
| IV | 12,145 (47) | 33 (42.3) |
All the bold values are statistically significant
Lymph node yield
LNY among patients who underwent LND during RC was calculated. Those with prior radiation therapy and radiation naïve cases had no statistical difference in LNYs (12.3 [SD 9.2] vs. 15.0 [SD 13.5]; p = 0.157). A low proportion of irradiated patients, compared to radiation naïve patients, received an adequate LND, defined by LNY of greater than 10 (32.1 vs. 45.3%, p = 0.016). Using logistic regression, we determined the predictors of undergoing an adequate LND (Table 2). A prior radiation history was not an independent predictor of adequate dissection.
Table 2.
Logistic regression to identify variables associated with an adequate lymph node dissection, using the cutoff of at least 10 nodes retrieved
| Covariate | Univariate |
Multivariate |
||
|---|---|---|---|---|
| Odds ratio [95% CI] | p value | Odds ratio [95% CI] | p value | |
| Prior radiation | 0.57 [0.36–0.91] | 0.017 | 0.66 [0.40–1.09] | 0.1 |
| Age ≥ 70 | 0.70 [0.67–0.74] | < 0.001 | 0.70 [0.66–0.74] | < 0.001 |
| Male | 1.09 [1.03–1.15] | 0.003 | 1.04 [0.98–1.10] | 0.262 |
| Race | ||||
| Black | 0.94 [0.85–1.04] | 0.225 | 0.94 [0.84–1.05] | 0.257 |
| Other | 1.22 [1.08–1.37] | 0.001 | 1.24 [1.09–1.40] | 0.001 |
| SCC histology | 0.85 [0.75–0.95] | 0.006 | 1.13 [0.99–1.29 | |
| Grade | ||||
| 1 | Ref | Ref | ||
| 2 | 1.39 [1.07–1.80] | 0.013 | 1.26 [0.97–1.65] | 0.09 |
| 3 | 1.65 [1.29–2.11] | < 0.001 | 1.41 [1.09–1.82] | < 0.001 |
| 4 | 2.64 [2.06–3.37] | < 0.001 | 2.19 [1.70–2.83] | < 0.001 |
| T stage | ||||
| T0, Ta, Tis, Tx | Ref | Ref | ||
| T1/T2 | 2.86 [2.52–3.24] | < 0.001 | 2.32 [1.99–2.71] | < 0.001 |
| T3/T4 | 2.95 [2.60–3.34] | < 0.001 | 2.11 [1.81–2.46] | < 0.001 |
| N1 | 2.26 [2.11–2.41] | < 0.001 | 2.08 [1.94–2.23] | < 0.001 |
| M1 | 1.00 [0.89–1.13] | 0.933 | 1.09 [0.95–1.25] | 0.238 |
Patient survival
The median survival time for irradiated patients was 22 months (95% confidence interval (95% CI) [14.3–29.7]), while the median survival time for radiation naïve patients was 43 months (95% CI [41.4–44.6], p = 0.0136). Figure 1 displays the Kaplan–Meier curve for these two groups of patients.
Fig. 1.
a Kaplan–Meier curve demonstrates that the survival of patients with prior radiation therapy for bladder cancer was lower compared to the survival of radiation naïve patients. However, controlling for all patient characteristics, a Cox regression analysis revealed that radiation history was not an independent predictor of mortality. An adequate lymph node dissection, obtaining 10 or more nodes, improved overall survival in b radiation naïve patients, but did not affect survival in c patients with prior radiation for bladder cancer
In radiation naïve patients, an adequate LND conferred a higher survival time compared to the survival of patients with a LNY < 10 (60 months (95% CI [56.4–63.6] vs. 35 months (95% CI [33.4–36.6]), p < 0.001). On the contrary, in irradiated patients, an adequate LND yielded no survival benefit (15 months (95% CI [9.0–21.0] vs. 25 months (95% CI [17.1–32.9]), p = 0.131). Figure 1 compares the Kaplan–Meier curves of radiation naïve patients to patients with prior radiation, determining the impact of an adequate LND during RC on overall survival. Patients with prior radiation who received an adequate LND tended to have more stage T3/T4 cancer, and more nodal disease compared to patients who received an inadequate LND (Table 3).
Table 3.
Cox proportional hazard regression of predictors for overall mortality
| Covariate | Univariate |
Multivariate |
||
|---|---|---|---|---|
| Hazard ratio [95% CI] | p value | Hazard ratio [95% CI] | p value | |
| Age ≥ 70 | 1.67 [1.62–1.72] | < 0.001 | 1.64 [1.60–1.70] | < 0.001 |
| Race | ||||
| White | 1 | Ref | 1 | ref |
| Black | 1.22 [1.15–1.30] | < 0.001 | 1.30 [1.21–1.39] | < 0.001 |
| Other | 0.86 [0.80–0.93] | < 0.001 | 0.86 [0.80–0.93] | < 0.001 |
| Sex | ||||
| Female | 1 | Ref | 1 | ref |
| Male | 0.89 [0.86–0.92] | < 0.001 | 0.97 [0.93–1.00] | 0.068 |
| Histology | ||||
| UCC | 1 | Ref | 1 | ref |
| SCC | 1.35 [1.26–1.44] | < 0.001 | 1.33 [1.23–1.44] | < 0.001 |
| Grade | ||||
| 1 | 1 | Ref | 1 | ref |
| 2 | 1.21 [1.04–1.41] | 0.016 | 1.32 [1.12–1.55] | 0.001 |
| 3 | 1.38 [1.19–1.60 | < 0.001 | 1.61 [1.38–1.88] | < 0.001 |
| 4 | 1.28 [1.10–1.48] | 0.001 | 1.52 [1.31–1.78] | < 0.001 |
| Metastasis | 3.20 [3.00–3.40] | < 0.001 | 3.25 [3.04–3.46] | < 0.001 |
| LNY ≥ 10 | 0.75 [0.73–0.78] | < 0.001 | 0.77 [0.74–0.80] | < 0.001 |
| Prior radiation | 1.35 [1.06–1.71] | 0.015 | 1.27 [0.98–1.66] | 0.076 |
Using Cox proportional regression models, an adequate LND was independently associated with a lower all-cause mortality, HR 0.77 (95% CI [0.74–0.80]). A history of prior radiation was not an independent predictor of all-cause mortality, HR 1.27 (95% CI [0.98–1.66]). Predictors associated with higher all-cause mortality include age ≥ 70 years old, African-American race, a histology of squamous cell carcinoma, disease grade > 1, and metastatic disease (Table 4).
Table 4.
Describes the characteristics of patients with prior radiation who underwent a lymph node dissection
| Covariate, N (%) | LNY ≥ 10 n = 27 |
LNY < 10 n = 48 |
p value |
|---|---|---|---|
| Age ≥ 70 | 11 (40.7) | 27 (47.4) | 0.569 |
| Male | 19 (70.4) | 37 (64.9) | 0.620 |
| Race | |||
| Black | 3 (11.1) | 1 (1.8) | 0.169 |
| Other | 1 (3.7) | 2 (3.5) | |
| County classification | 0.211 | ||
| Metro country | 1 (3.7) | 7 (12.3) | |
| Non-metro county | 26 (96.3) | 50 (87.7) | |
| SCC histology | 0 (0) | 4 (7.0) | 0.24 |
| Grade | 0.261 | ||
| 1 | 1 (3.8) | 0 (0) | |
| 2 | 1 (3.8) | 5 (10.4) | |
| 3 | 10 (38.5) | 24 (50.0) | |
| 4 | 14 (53.8) | 19 (39.6) | |
| T stage | 0.021 | ||
| T0, Ta, Tis, Tx | 3 (11.1) | 5 (8.9) | |
| T1/T2 | 3 (11.1) | 23 (41.1) | |
| T3/T4 | 21 (77.8) | 28 (50.0) | |
| Node positive Dx | 7 (25.9) | 5 (8.9) | 0.039 |
| Metastatic Dx | 2 (7.4) | 6 (10.5) | 0.649 |
Discussion
In this retrospective study of the SEER database, we describe the proportion of patients undergoing a LND, and the LNY of patients undergoing RC following radiation therapy for bladder malignancy. The decision to perform a LND, irrespective of the extent, is typically dependent on surgeon characteristics, rather than patient characteristics. A prior SEER study determined that up to 30% of RCs were performed without a lymphadenectomy [10]. Though a thorough LND confers diagnostic and therapeutic benefits, these facets have yet to be fully explored in patients who have sought prior bladder-sparing treatments, such as radiation, for BC. These situations are rare, but present with unique challenges. As it is, urologists agree that the LND and urinary diversion are the most challenging aspects of a robotically assisted RC, and the challenge increases with the desmoplastic tissue changes following radiation therapy.
Here, we observed a 13% higher rate of adequate LND in radiation naïve patients. Surgeon variables are generally the driving force behind the decision to perform a LND, however. For example, urologists that practice at a teaching hospital or have an association with a major cancer center are more likely to perform a LND [10]. Exploration of temporal trends has unearthed a rising proportion of patients who undergo LND during RC. A SEER study determined that in 1988, 63% of patients underwent a LND, with this proportion increasing until 2004, when 84% of patients underwent a LND [15]. Surgical factors, such as margin status, and LND extent, heavily influence oncologic outcomes and thus explain the increasing proportion of patients receiving a LND during RC [16, 17].
We demonstrate that a lower proportion of irradiated patients receive an adequate LND. One possible solution may be to encourage adherence to guidelines for template lymphadenectomy, frequently cited as a superior metric of surgical quality [18–21]. The paucity of data regarding lymphadenectomy in salvage RC makes these guidelines difficult to apply in this patient population. Furthermore, observance of guidelines can be variable, and a template for a LND does not enforce the quality of dissection. In a post hoc review of the SWOG 4B951 trial, a study determining the efficacy of adjuvant chemotherapy in p53-positive bladder tumors, 31% of patients met all the LNY and anatomic extent of LND benchmarks for protocol eligibility, highlighting the variability of adherence to surgical quality metrics [22]. Enforcing surgical benchmarks remains challenging in radiation naïve patients as well as our results show that 45% of patients with no prior bladder radiation received an adequate LND.
We show that irradiated and radiation naïve patients who underwent a LND had similar mean LNYs. Once a LND is performed, multiple factors, including surgeon factors, disease severity, the method of obtaining lymph nodes, and pathology practices, may influence the lymph node count, resulting in large variability [23]. As an example, the same surgical technique performed at two institutions resulted in largely different median LNYs (72 vs. 40, p ≤ 0.001), yet the rate of metastatic disease was similar (23 vs. 24%) [19]. Despite the potential factors introducing variability in LNYs, our results highlight that once a LND is performed in a patient with prior radiation, their lymph node count is similar to that of radiation naïve patients.
Our results echo prior studies and show that a LNY ≥ 10 was independently associated with decreased mortality. We also show that prior radiation was not associated with mortality. In a similar manner, Ramani et al. [11] determined that the overall mortality of patients receiving primary RC is no different from that of patients receiving a salvage cystectomy after primary radiotherapy. The Kaplan–Meier curve of LND in patients with prior radiation (Fig. 1) appears to show a paradoxical survival benefit in patients who received an inadequate LND. However, this benefit was not statistically significant and may have been driven by the higher proportion of patients with T3/T4 disease and positive nodal status in patients who received an adequate LND. Yet the low number of irradiated patients may also play a role in this outcome. Additionally, we cannot account for the varying LND templates between centers. Still, given the independent survival benefit, an adequate LND in patients with a radiation history may then provide similar benefits as they would to radiation naïve patients; however, this will require prospective studies to fully elucidate.
Limitations of this study should also be considered. First, this is a retrospective SEER database study and may be impacted by selection biases, as unaccounted patient characteristics may confound the LNY. Additionally, the LNY is affected by factors other than patient and surgeon dependent aspects, such as variable practices in pathologic processing or node counting of the sample. However, this limitation may be unavoidable, particularly in studies that determine LNYs from multiple institutions. Finally, the sample size of patients with prior bladder radiation therapy (n = 89) therapy was low, and larger, prospective studies may prove beneficial when exploring the role of a LND in these patients. This small sample size may be the reason that prior radiation did not have an independent effect on mortality.
Conclusions
When compared to radiation naïve patients, a lower proportion of irradiated patients underwent an adequate LND. The LNY of radiation naïve patients was lower than those with prior bladder radiation; however, this difference was not statistically significant. A history of prior radiation was not independently associated with mortality. As other studies have shown, a LNY ≥ 10 nodes was independently associated with overall survival. Further investigation will be required to elucidate the patient and provider characteristics that contribute to the similar LNYs.
Acknowledgements
This research was supported by the Intramural Research Program of the NIH, National Cancer Institute, Center for Cancer Research.
Footnotes
Compliance with ethical standards
Conflict of interest None.
References
- 1.Konety BR, Joslyn SA, O’Donnell MA (2003) Extent of pelvic lymphadenectomy and its impact on outcome in patients diagnosed with bladder cancer: analysis of data from the Surveillance, Epidemiology and End Results Program data base. J Urol 169(3):946–950 [DOI] [PubMed] [Google Scholar]
- 2.Siemens DR et al. (2015) Lymph node counts are valid indicators of the quality of surgical care in bladder cancer: a population-based study. Urol Oncol 33(10):425.e15–425.e23 [DOI] [PubMed] [Google Scholar]
- 3.Hollenbeck BK et al. (2008) Hospital lymph node counts and survival after radical cystectomy. Cancer 112(4):806–812 [DOI] [PubMed] [Google Scholar]
- 4.Herr HW et al. (2002) Impact of the number of lymph nodes retrieved on outcome in patients with muscle invasive bladder cancer. J Urol 167(3):1295–1298 [PubMed] [Google Scholar]
- 5.Herr H et al. (2004) Standardization of radical cystectomy and pelvic lymph node dissection for bladder cancer: a collaborative group report. J Urol 171(5):1823–1828 (discussion 1827–1828) [DOI] [PubMed] [Google Scholar]
- 6.Leissner J et al. (2000) Lymphadenectomy in patients with transitional cell carcinoma of the urinary bladder; significance for staging and prognosis. BJU Int 85(7):817–823 [DOI] [PubMed] [Google Scholar]
- 7.Wright JL, Lin DW, Porter MP (2008) The association between extent of lymphadenectomy and survival among patients with lymph node metastases undergoing radical cystectomy. Cancer 112(11):2401–2408 [DOI] [PubMed] [Google Scholar]
- 8.Abdollah F et al. (2012) Stage-specific impact of pelvic lymph node dissection on survival in patients with non-metastatic bladder cancer treated with radical cystectomy. BJU Int 109(8):1147–1154 [DOI] [PubMed] [Google Scholar]
- 9.Cole AP et al. (2015) Temporal trends in receipt of adequate lymphadenectomy in bladder cancer 1988 to 2010. Urol Oncol 33(12):504.e9–504.e17 [DOI] [PubMed] [Google Scholar]
- 10.Hedgepeth RC et al. (2011) Variation in use of lymph node dissection during radical cystectomy for bladder cancer. Urology 77(2):385–390 [DOI] [PubMed] [Google Scholar]
- 11.Ramani VA et al. (2010) Differential complication rates following radical cystectomy in the irradiated and nonirradiated pelvis. Eur Urol 57(6):1058–1063 [DOI] [PubMed] [Google Scholar]
- 12.Efstathiou JA et al. (2012) Long-term outcomes of selective bladder preservation by combined-modality therapy for invasive bladder cancer: the MGH experience. Eur Urol 61(4):705–711 [DOI] [PubMed] [Google Scholar]
- 13.Nieuwenhuijzen JA et al. (2004) Salvage cystectomy after failure of interstitial radiotherapy and external beam radiotherapy for bladder cancer. BJU Int 94(6):793–797 [DOI] [PubMed] [Google Scholar]
- 14.Eisenberg MS et al. (2010) Early complications of cystectomy after high dose pelvic radiation. J Urol 184(6):2264–2269 [DOI] [PubMed] [Google Scholar]
- 15.Hellenthal NJ et al. (2009) Trends in pelvic lymphadenectomy at the time of radical cystectomy: 1988 to 2004. J Urol 181(6):2490–2495 [DOI] [PubMed] [Google Scholar]
- 16.Herr HW et al. (2004) Surgical factors influence bladder cancer outcomes: a cooperative group report. J Clin Oncol 22(14):2781–2789 [DOI] [PubMed] [Google Scholar]
- 17.Skinner EC, Stein JP, Skinner DG (2007) Surgical benchmarks for the treatment of invasive bladder cancer. Urol Oncol 25(1):66–71 [DOI] [PubMed] [Google Scholar]
- 18.National Comprehensive Cancer Network. Bladder cancer (version 1.2017). 2017 January 5, 2017. https://www.nccn.org/professionals/physician_gls/PDF/bladder.pdf
- 19.Dorin RP et al. (2011) Lymph node dissection technique is more important than lymph node count in identifying nodal metastases in radical cystectomy patients: a comparative mapping study. Eur Urol 60(5):946–952 [DOI] [PubMed] [Google Scholar]
- 20.van de Putte EEF et al. (2015) Lymph node count at radical cystectomy does not influence long-term survival if surgeons adhere to a standardized template. Urol Oncol 33(12):504.e19–504.e24 [DOI] [PubMed] [Google Scholar]
- 21.Ghersi D, Stewart LA, Palmar MKB et al. (1999) Neoadjuvant cisplatin, methotrexate, and vinblastine chemotherapy for muscle-invasive bladder cancer: a randomised controlled trial. International collaboration of trialists. Lancet 354(9178):533–540 [PubMed] [Google Scholar]
- 22.Mata DA et al. (2015) Variability in surgical quality in a phase III clinical trial of radical cystectomy in patients with organ-confined, node-negative urothelial carcinoma of the bladder. J Surg Oncol 111(7):923–928 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Stein JP et al. (2007) Radical cystectomy with extended lymphadenectomy: evaluating separate package versus en bloc submission for node positive bladder cancer. J Urol 177(3):876–881 (discussion 881–882) [DOI] [PubMed] [Google Scholar]