SEARCH

SEARCH BY CITATION

Keywords:

  • bladder cancer mortality;
  • cancer-specific mortality;
  • competing-risks;
  • nephroureterectomy;
  • upper-tract urothelial carcinoma

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Conflict of interest
  9. References
  10. Supporting Information

Objectives

To examine the rates of cancer-specific mortality, other-cause and bladder cancer mortality in patients with upper-tract urothelial carcinoma undergoing radical nephroureterectomy.

Methods

Relying on the Surveillance, Epidemiology, and End Results database, 9899 patients treated with radical nephroureterectomy were identified. A 20-strata graphical aid was constructed using age (<60, 60–69, 70–79, >79 years) and American Joint Committee on Cancer/TNM stage (pT1N0/x, pT2N0/x, pT3N0/x, pT4N0/x, pTanypN1–3) as stratifying variables. The 5-year cancer-specific mortality, other-cause and bladder cancer mortality rates were generated through competing-risks Poisson regression methodologies. Multivariable competing-risks regression models were used to test the effect of age and stage on three different end-points: cancer-specific mortality, other-cause and bladder cancer mortality.

Results

Overall, 1797 (18.1%), 891 (9.1%) and 3090 (31.2%) patients died of cancer-specific mortality, other-cause and bladder cancer mortality, respectively. Following stratification according to age and stage, the proportion of patients who succumbed to cancer-specific mortality (11.7–21.9%) and other-cause mortality (8.9–30.4%) increased with age. In contrast, with increasing stage, the proportion of patients who died of cancer-specific mortality increased (7.2–37.5%), whereas the proportion of other-cause mortality remained stable (18.9–22.0%). The rate of bladder cancer mortality increased with advancing stage. At multivariable competing-risk regression model, besides age and stage, women, type of surgery, grade and location were associated with higher cancer-specific mortality. Furthermore, ureteral location, stage and grade were associated with bladder cancer mortality.

Conclusions

The developed graphical aid for prediction of cancer-specific mortality, other-cause, and bladder cancer mortality according to age and stage in patients with upper-tract urothelial carcinoma undergoing radical nephroureterectomy can be useful for physicians and patients during clinical counseling.


Abbreviations & Acronyms
AJCC/TNM

American Joint Committee on Cancer/TNM

CSM

cancer-specific mortality

HR

hazard ratio

RNU

radical nephroureterectomy

SEER

Surveillance, Epidemiology, and End Results

UTUC

upper-tract urothelial carcinoma

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Conflict of interest
  9. References
  10. Supporting Information

The standard of care for patients with invasive, non-metastatic UTUC remains to be RNU.[1] Albeit poor cancer control outcomes amongst patients with advanced tumor stage and/or those who harbor lymph node metastases (5-year CSM-free rates: 28.3–35%),[2-4] chronological age and age-related conditions represent non-negligible considerations in selecting surgical candidates.[3, 5] Specifically, elderly patients might be less healthy, and less capable of tolerating cancer manifestations and related therapies compared with their younger and healthier counterparts.[6] In contrast, advanced age at surgery has also been associated with a higher risk of dying from UTUC after surgery.[3, 5, 7]

Recently, several prognostic models have been developed to predict oncological outcomes in patients with UTUC.[8-10] These tools might be helpful to guide preoperative patient counseling, postoperative follow up and administration of additional cancer therapies. In this context, although several models graphically illustrate the risk of CSM and other-cause mortality in patients diagnosed with other malignancies,[11-13] no such tool exists for patients diagnosed with UTUC treated with RNU. Consequently, the primary objective of the current study was to identify individuals who might benefit the most from RNU, and those who might only gain minimal benefit from surgical intervention. In addition, given the high recurrence rates in the bladder (5-year recurrence rate: 35%) of UTUC patients as previously reported,[14, 15] we sought to assess the rates of bladder cancer mortality. We relied on competing-risks regression models to graphically illustrate the 5-year CSM, other-cause and bladder cancer mortality rates in patients with UTUC treated with RNU.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Conflict of interest
  9. References
  10. Supporting Information

Population source

The SEER program database (1988–2009) was used to extract the study population. The SEER collects patient demographics, and publishes cancer incidence and survival data from population-based cancer registries, currently covering approximately 28% of the USA population.[16] Data from 1988 to 2009 from 18 SEER registries were abstracted. The characteristics of the SEER population are comparable with that of the general population of the USA.[17]

Study cohort

Only individuals with a diagnosis of UTUC were included (International Classification of Disease for Oncology [ICD-O] codes: C65.9, C66.9). Those with histological subtypes other than that of urothelial carcinoma (ICD-O-3 codes 8120 and 8130) were excluded from the current analysis. Patients who underwent a RNU with or without bladder cuff were identified. Exclusions consisted of those less than 18 years-of-age (n = 1), metastatic disease (n = 530), unknown tumor extent (n = 441), unknown nodal status (n = 52) and unknown metastatic status (n = 205). This resulted in 9899 assessable individuals.

Stratified variables

For the purpose of analysis, stratification was carried out according to patient age at diagnosis categories (<60, 60–69, 70–79 and >79 years) and disease stage according to the most recent AJCC/TNM classification (pT1N0/x, pT2N0/x, pT3N0/x, pT4N0/x, pTanypN1–3). This stratification was chosen for its simplicity and clinical pertinence, as previously done in other tumor malignancies.[13] The stratification of patients according to age and disease staging groups resulted in 20 different combinations.

Outcomes

The cause of death was defined using the SEER cause of death code (29020 and 29030). Deaths from UTUC were coded as CSM. Deaths from urothelial carcinoma of the urinary bladder were coded as bladder cancer mortality. All other events were considered as other-cause mortality.

Statistical analyses

Means, medians and interquartile ranges were reported for continuous variables. Frequencies and proportions were reported for categorical variables. The independent t-test and χ2-test were used to compare the statistical significance of differences in means and proportions, respectively. These tests were carried out using spss version 20 (SPSS, Chicago, IL, USA).

The statistical analyses consisted of different steps. First, a graphical illustration of the 5-year CSM, other-cause and bladder cancer mortality rates was generated using competing-risks Poisson regression analyses, as previously described.[18-20] For each age and stage strata, survival estimates were calculated. As RNU with bladder cuff currently represents the standard of care in the treatment of UTUC,[21] we carried out separate analyses exclusively in a cohort of patients receiving this surgical approach (n = 6283).

Second, separate multivariable competing-risks regression models were used to test the effect of AJCC/TNM stage and age groups on CSM. Covariates consisted of sex, race, type of surgery, primary tumor location, year of diagnosis and tumor grade. The competing-risks regression methodology allowed us to account for the effects of other-cause and bladder cancer mortality, providing the most unbiased estimates of CSM. Indeed, this statistical approach avoids overestimation of CSM, as censoring as a result of other-cause and bladder cancer mortality could artificially reduce the cohort of patients at risk of dying from UTUC.[18-20]

Third, we repeated our analyses exclusively in a subcohort of patients with only one primary tumor, thereby excluding patients with a concomitant disease, such as bladder cancer, at UTUC diagnosis. These analyses were carried out using the R statistical package (v2.15.2). All tests were two-sided, with a significance level set at P < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Conflict of interest
  9. References
  10. Supporting Information

Baseline descriptives

The study population consisted of 9899 patients who underwent radical RNU for UTUC between 1988 and 2009. Baseline patient characteristics are described in Table 1. Average age was 71 years (median 73). The majority of patients were male (n = 5823, 59%) and Caucasian (n = 8820, 89%). Respectively, 1485 (15%), 2415 (24.4), 3681 (37.2) and 2318 (23.4%) of patients were aged <60, 60–69, 70–79 and >79 years. Overall, the AJCC/TNM staging distribution was 34.7, 16.2, 33.7, 7.4 and 8.0% for pT1N0/x, pT2N0/x, pT3N0/x, pT4N0/x and pTanyN1/3, respectively. When patients were stratified according to age at surgery (namely, <60, 60–69, 70–79 and >79 years), statistically significant differences were recorded with respect to sex, surgery type, the rate of lymph node dissection and tumor grade (all P ≤ 0.002). Mean and median follow up was 104 and 98 months. Overall, 1797 (18.1%), 891 (9.1%), and 3090 (31.2%) deaths as a result of CSM, other-cause and bladder cancer mortality occurred, respectively.

Table 1. Characteristics of patients diagnosed with UTUC and who underwent RNU between 1988 and 2009 within 17 SEER registries
 Overall populationAge <60 yearsAge 60–69 yearsAge 70–79 yearsAge >79 yearsP-value
No. patients (%)98991485 (15)2415 (24.4)3681 (37.2)2318 (23.4)
Age (years)     
Mean (median)71.2 (73)52.4 (54)65.1 (66)74.7 (75)84.1 (83.5)
Range22–9922–5960–6970–7980–99
Sex     <0.001
Men5823 (58.8%)996 (65%)1546 (64%)2139 (58.1%)1142 (49.2%)
Women4076 (41.2%)489 (35%)869 (36%)1542 (41.9%)1176 (50.8%)
Race     <0.001
White8820 (89.1%)1300 (87.5%)2111 (87.4%)3281 (89.1%)2128 (91.8%)
Other1079 (10.9%)185 (12.5%)304 (12.6%)400 (10.9%)190 (8.2%)
Primary tumor location     <0.001
Renal pelvis6819 (68.8%)1145 (77.1%)1620 (67.1%)2428 (65.9%)1626 (70.1%)
Ureter3080 (31.2%)340 (22.9%)795 (32.9%)1253 (34.1%)692 (29.9%)
Year of surgery     <0.001
1988–19962529 (25.6%)403 (27.1%)676 (28%)965 (26.2%)485 (20.9%)
1997–20022622 (26.5%)391 (26.3%)603 (25%)1058 (28.7%)570 (24.7%)
2003–20062698 (27.2%)399 (26.9%)632 (26.1%)978 (26.6%)689 (29.7%)
2007–20092050 (20.7%)292 (19.7%)504 (20.9%)680 (18.5%)574 (24.7%)
Type of surgery     0.002
RNU with bladder cuff6283 (64.5%)909 (61.2%)1590 (65.8%)2362 (64.2%)1422 (61.3%)
RNU without bladder cuff3616 (36.5%)576 (38.8%)825 (34.2%)1319 (35.8%)896 (38.7%)
pT stage     <0.001
pT13493 (35.2%)577 (39.3%)911 (37.7%)1299 (35.3%)706 (30.4%)
pT21674 (16.9%)236 (15.7%)398 (16.4%)656 (17.8%)384 (16.6%)
pT33788 (38.2%)538 (36.1%)897 (37.2%)1370 (37.3%)983 (42.5%)
pT4944 (9.7%)134 (9%)209 (8.7%)356 (9.6%)245 (10.5%)
pN stage     <0.001
pNx7561 (76.3%)1057 (71.2%)1799 (74.5%)2852 (77.5%)1853 (79.9%)
pN01545 (15.6%)296 (19.9%)416 (17.2%)526 (14.2%)307 (13.3%)
pN1–3793 (8.1%)132 (8.9%)200 (8.3%)303 (8.3%)158 (6.8%)
AJCC/TNM staging     <0.001
pT1N0/x3438 (34.7%)567 (38.2%)900 (37.3%)1273 (34.6%)698 (30.1%)
pT2N0/x1605 (16.2%)222 (14.9%)385 (15.9%)628 (17.1%)370 (16%)
pT3N0/x3334 (33.7)469 (31.6%)772 (32%)1206 (32.8%)887 (38.3%)
pT4N0/x729 (7.4%)95 (6.4%)158 (6.5%)271 (7.4%)205 (8.8%)
pTanypN1/3793 (8%)132 (8.9%)200 (8.3%)303 (8.2%)158 (6.8%)
Tumor grade     <0.001
Grade I489 (4.9%)100 (6.7%)142 (5.9%)157 (4.2%)90 (3.9%)
Grade II2590 (26.1%)472 (31.7%)676 (28%)971 (26.4%)471 (20.3%)
Grade III3786 (38.3%)488 (32.9%)902 (37.4%)1450 (39.5%)946 (40.8%)
Grade IV2550 (25.8%)352 (23.8%)579 (23.9%)918 (24.9%)701 (30.2%)
NA484 (4.9%)73 (4.9%)116 (4.8%)185 (5%)110 (4.8%)
Follow up (months) of censored patients     <0.001
Mean (median)104 (88)105 (90)104.7 (88)107.4 (95)93.6 (73)
Range0–2630–2630–2620–2600–241
Vital status at last contact     <0.001
Alive4121 (41.6%)948 (63.8%)1160 (48.1%)1326 (36%)687 (29.6%)
Deceased due to UTUC1797 (18.1%)204 (13.7%)390 (16.1%)679 (18.4%)524 (22.6%)
Deceased due to bladder cancer891 (9.1%)119 (8.1%)233 (9.6%)345 (9.4%)194 (8.4%)
Deceased due to other causes3090 (31.2%)214 (14.4%)632 (26.2%)1331 (36.2%)913 (39.4%)

Poisson regression analyses

Figure > 1a shows the smoothed cumulative mortality estimates at 5 years of follow up after surgery. The lowest CSM rate was recorded in patients aged <60 years with pT1pNx/0 disease (3.2%). Conversely, the highest rate of CSM was recorded in patients aged >79 years with lymph node metastases (53.7%). The rate of bladder cancer mortality increased with advancing stage, as well as with advancing age: 15.8% in septuagenarians with pT4N0/x UTUC. These results were confirmed when evaluating only patients undergoing RNU with bladder cuff (Fig. 1b; n = 6283).

figure

Figure 1. Smoothed, model-derived, 5-year cumulative mortality estimates of (a) patients affected by UTUC undergoing RNU overall, and (b) in patients receiving RNU with bladder cuff (b). image, Other-cause mortality; image, CSM; image, bladder cancer mortality; image, survival.

Download figure to PowerPoint

Multivariate competing-risk regression models

When focusing on CSM, the present results showed that women (HR 1.16), increasing age (>79 years, HR 1.72), advancing stage (pTanyN1/3, HR 5.09), high-grade (HR 1.71), RNU without bladder cuff (HR 1.29) and renal pelvic tumors (HR 1.12; all P ≤ 0.03) were associated with a higher risk of the outcome after adjusting for all other covariates, including that of other-cause mortality (Table 2). When focusing on other-cause mortality, the present results showed that only men (HR 1.12) and increasing age (HR 4.25, all P ≤ 0.002) emerged as statistically significantly associated with a higher risk of the outcome (Table 2). With respect to bladder cancer mortality, the present results showed that men (HR 1.4), ureteral tumors (HR 1.49), increasing AJCC/TNM stage (HR 2.28) and high-grade UTUC (HR 1.63; all P ≤ 0.005) were associated with a higher risk of the outcome (Table 2).

Table 2. Multivariable competing-risks regression models addressing CSM, other-cause and bladder cancer mortality in patients diagnosed with UTUC undergoing RNU
 CSMOther-cause mortalityBladder cancer mortality
HR (95% CI)P-valueHR (95% CI)P-valueHR (95% CI)P-value
Age at surgery (years)      
60–69 vs <601.20 (1.01–1.42)0.022.04 (1.75–2.38)<0.0011.17 (0.94–1.46)0.15
70–79 vs <601.32 (1.13–1.55)<0.0013.13 (2.71–3.61)<0.0011.10 (0.89–1.36)0.35
>80 vs <601.72 (1.46–2.02)<0.0014.25 (3.65–4.94)<0.0011.03 (0.82–1.30)0.74
Pathological T and N stages      
pT2N0/x vs pT1N0/x1.56 (1.30–1.87)<0.0010.93 (0.84–1.03)0.181.36 (1.09–1.68)0.005
pT3N0/x vs pT1N0/x2.61 (2.27–3.01)<0.0010.90 (0.83–0.98)0.021.67 (1.40–2.00)<0.001
pT4N0/x vs pT1N0/x4.18 (3.51–4.98)<0.0010.80 (0.68–0.93)0.0052.14 (1.67–2.75)<0.001
pTanyN1/3 vs pT1N0/x5.09 (4.29–6.04)<0.0010.72 (0.61–0.86)<0.0012.28 (1.78–2.91)<0.001
Sex      
Female vs male1.16 (1.05–1.27)0.0010.89 (0.83–0.96)0.0020.71 (0.62–0.82)<0.001
Race      
Non-white vs white0.99 (0.85–1.16)0.90.98 (0.87–1.1)0.71.04 (0.85–1.29)0.66
Year of diagnosis0.96 (0.95–0.97)<0.0010.97 (0.96–0.98)<0.0010.99 (0.98–1.00)0.71
Type of surgery      
RNU without bladder cuff vs RNU with bladder cuff1.29 (1.17–1.42)<0.0010.95 (0.88–1.02)0.211.05 (0.91–1.21)0.46
Primary tumor location      
Ureter vs renal pelvis0.89 (0.80–0.99)0.031.04 (0.97–1.12)0.221.49 (1.29–1.71)<0.001
Tumor grade      
Grade II vs Grade I0.89 (0.71–1.11)0.320.89 (0.80–1.00)0.051.17 (0.87–1.58)0.29
Grade III vs Grade I1.64 (1.33–2.01)<0.0010.83 (0.74–0.93)0.0011.74 (1.30–2.31)<0.001
Grade IV vs Grade I1.71 (1.38–2.12)<0.0010.85 (0.74–0.97)0.011.63 (1.20–2.21)<0.001

Subanalysis

In analyses focusing exclusively on patients with only one primary tumor diagnosis of UTUC (n = 4216; see Fig. S1a), CSM rate was lowest amongst patients in the younger age group and lower disease stage (4.2%), and highest amongst patients in the oldest age group with lymph node metastases (60.0%). With respect to bladder cancer mortality, the 5-year rate ranged from 1.4% in patients aged <60 years to 13.5% in those aged 60–69 years with pT4N0/x disease. After adjusting for all other covariates, competing-risks regression analyses showed comparable results to that of the entire cohort (data not shown). Similar results were obtained when evaluating a cohort of patients with only one primary tumor diagnosis of UTUC undergoing RNU with bladder cuff (n = 2604; see Fig. S1b).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Conflict of interest
  9. References
  10. Supporting Information

The standard treatment of care for patients diagnosed with UTUC is RNU.[1] For such individuals, cancer control outcomes have been shown to vary according to disease stage (5-year CSM-free rates: 12.4–93.5%)[4] and age (5-year CSM-free rates: 29–84%).[5, 22] Whilst CSM might increase with advancing stage, older patients are prone to other-cause mortality.[13] Consequently, an adequate selection of RNU candidates, considering both age and disease stage, remains primordial. For several other urological malignancies, the development of a visual aid that simultaneously shows CSM and other-cause mortality rates has been explored.[11-13] However, no such graphical tool currently exists for UTUC. As such, our primary objective was to assess the rates of CSM and other-cause mortality in patients diagnosed with UTUC using a large contemporary population-based cohort. Through the reliance of stringent statistical methods, CSM and other-cause mortality rates at 5 years after RNU were generated for each age and stage strata. In addition, as several important studies have shown that a recurrence to the bladder occurs in approximately three out of 10 individuals treated with RNU,[14, 15] we further complimented our analyses by also focusing on bladder cancer mortality rates.

Primarily, our analyses show that CSM was clearly determined by pathological tumor and nodal extents, as measured using the AJCC/TNM stages, irrespective of age. For example, it ranged from 7.2% for patients with pT1N0/x to 37.5% for patients with node-positive UTUC. That said, advancing age also portended to higher rates of CSM at 5 years after RNU, regardless of disease stage. For example, it ranged from 11.7% for those aged <60 years to 21.9% for patients >79 years, regardless of AJCC/TNM stage. In multivariable analyses, sex, year of diagnosis, type of surgery, primary tumor location and tumor grade represented additional predictors of CSM. Interestingly, our observations show for the first time that women are significantly more likely to die from UTUC relative to men. In this context, when focusing on the SEER database, Lughezzani et al. reported that females are more likely to have advanced pathological stage and grade at RNU compared with males.[23] However, they failed to show an effect of sex on the risk of CSM, after accounting for other confounders. The inclusion of individuals with less aggressive disease (pT1–3N0/x disease), shorter follow up (53 months) and a smaller number of patients evaluated (n = 4850) might be responsible for the lack of significance.[23]

Importantly, when considering other-cause mortality in addition to CSM, our analyses show that although age has a direct association with increasing other-cause mortality, locally-advanced pathological stages (e.g. pT3–4N0/x) or patients with lymph node metastases at RNU (pTanyN1–3) represent the main drivers of mortality. For example, the rate of CSM was as high as 43% amongst individuals with pT3–4N0/x disease aged >79 years. For the same disease stage, other-cause mortality rates varied minimally. In comparison, the effect of other-cause mortality was more detrimental amongst individuals with localized disease (pT1–2N0/x), and increased with advancing age.

As an additional clinically pertinent end-point, we also assessed the rates of bladder cancer mortality in patients diagnosed with UTUC after RNU. In this regard, our analyses show that the risk of succumbing to bladder cancer mortality was non-negligible in the entire cohort, thereby suggesting that recurrence to the bladder might be particularly worrisome. Specifically, the overall bladder cancer mortality rates at 5 years after RNU ranged from 3.2 to 15.8%, after adjusting for both CSM and other-cause mortality. Of interest, our visual aid shows that advancing disease stage resulted in higher risk of bladder cancer mortality. Furthermore, in multivariable analyses, our results showed that high-grade and/or node-positive metastatic disease, as well as ureteral tumor location, emerged as significant predictors of higher bladder cancer mortality after RNU, as previously observed.[24, 25] To exclude the possibility of patients who had a concomitant bladder cancer diagnosis at RNU, we confirmed our findings by restricting our patient cohort to those with only one primary UTUC diagnosis.

Several explanations might be advanced for the observed phenomenon: (i) UTUC and bladder cancer could be two different manifestations of the same panurothelial disease, which increases the risk of a bladder recurrence after UTUC, and vice versa;[14, 15, 26, 27] and (ii) cancer-cell spillage to the bladder[25] is enhanced given a UTUC ureteral location, namely the intraluminal seeding hypothesis,[28] thus increasing the risk of bladder cancer growth. Because of the nature of the SEER database, it was not possible to further explore such hypotheses. That said, the results merit further investigation from future reports with detailed information on the aforementioned points.

From a clinical standpoint, the current report shows that disease stage is strongly indicative of CSM, even in older individuals. In consequence, it could be suggested that the consideration of RNU should not be neglected amongst older individuals with an aggressive disease, especially if their general health permits it. Indeed, according to the society of geriatric oncology,[29] old age alone cannot be considered a criterion for surgery omission. Specifically, a comprehensive functional assessment should be endorsed to evaluate the patients' ability to withstand surgery.[29] At the same time, our findings might raise caution against the benefit of surgical intervention in older patients with localized UTUC. In this regard, the risk of succumbing to CSM varied minimally, whereas the risk of other-cause mortality increased with advancing age. Indeed, it is well established that old age is unequivocally associated with a weaker defence mechanism, and a lower ability to withhold treatment-induced complications and toxicities.[6] As such, the role of surgery should be cautiously apprehended in this subcohort.

In summary, the current results support the role of RNU in the context of aggressive disease stage. In older individuals, the consideration of surgical intervention should be prudently evaluated, especially given a localized UTUC stage, where limited cancer control benefit might be expected. Overall, the developed graphical tool may be used by physicians and patients during clinical counseling, and for follow-up planning after surgery. The current report might also serve to encourage future studies that pertain to the assessment of bladder cancer recurrence after a UTUC diagnosis, and the exploration of more concrete hypotheses, using a more detailed database.

Despite several strengths, the present study was not devoid of limitations. First, the lack of additional prognostic determinants of mortality prevented us from adjusting for other important characteristics, such as smoking status, obesity or the presence of other comorbid conditions at UTUC diagnosis, as well as the administration of adjuvant treatment. The lack thereof could have under- or overestimated the effect of stage and age on multivariable analyses. Furthermore, although our analyses were supplemented with a multivariable model that sought to identify determinants of CSM, because of the nature of the SEER database, we recognize that other characteristics (e.g. biomarkers), which we lacked, might be associated with the end-points examined. Second, whereas our analyses sought to explore bladder cancer mortality as a secondary end-point of interest, because of the purported high recurrence rates to the bladder after UTUC diagnosis, it was not possible to actually examine recurrence rates, given that this end-point is not readily available in the database. In consequence, although we attempted to adjust for the presence of a concomitant bladder cancer diagnosis at UTUC diagnosis by restricting the cohort to patients with only one primary diagnosis at RNU, caution is warranted in the interpretation of such results. Third, because of the population-based and retrospective nature of the database, the lack of central pathology represents an additional limitation to the study. In this context, we were not able to adjust our analyses for important pathological predictors of recurrence and CSM, such as lymphovascular invasion, concomitant carcinoma in situ and surgical margins.[30] Additionally, we were not able to discriminate patients treated with an open or laparoscopic RNU. However, we should underline that several studies showed equivalence in terms of short-term oncological outcomes between the two surgical approaches, even in selected patients with locally advanced disease.[30, 31] Finally, cause of death attribution, tumor and nodal extent assignment are variable in the SEER database. Whilst previous reports have reported on the validity and reliability of such codings, the large number of patients and areas covered by the SEER might nonetheless result in minor fluctuations in the coding of such variables. Other studies relying on the SEER database are also limited by such factors.[3, 13, 22] That said, the SEER remains a large contemporary cohort that is considered highly generalizable of the USA population.

Concluding, the present study provides a valuable graphical aid for the estimation of CSM, bladder cancer and other-cause mortality in UTUC patients treated with RNU. This tool serves to better stratify the risk-benefit ratio of surgery. Of note, the present findings support the role of RNU in individuals with advanced UTUC. However, an adequate selection of surgical candidates remains of utmost importance in the setting of older individuals with localized UTUC.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Conflict of interest
  9. References
  10. Supporting Information

Pierre I Karakiewicz is partially supported by the University of Montreal Health Centre Urology Specialists, Fonds de la Recherche en Santé du Québec, the University of Montreal Department of Surgery and the University of Montreal Health Centre (CHUM) Foundation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Conflict of interest
  9. References
  10. Supporting Information
  • 1
    Roupret M, Zigeuner R, Palou J et al. European guidelines for the diagnosis and management of upper urinary tract urothelial cell carcinomas: 2011 update. Eur. Urol. 2011; 59: 584594.
  • 2
    Roscigno M, Shariat SF, Margulis V et al. Impact of lymph node dissection on cancer specific survival in patients with upper tract urothelial carcinoma treated with radical nephroureterectomy. J. Urol. 2009; 181: 24822489.
  • 3
    Lughezzani G, Jeldres C, Isbarn H et al. Nephroureterectomy and segmental ureterectomy in the treatment of invasive upper tract urothelial carcinoma: a population-based study of 2299 patients. Eur. J. Cancer 2009; 45: 32913297.
  • 4
    Margulis V, Shariat SF, Matin SF et al. Outcomes of radical nephroureterectomy: a series from the Upper Tract Urothelial Carcinoma Collaboration. Cancer 2009; 115: 12241233.
  • 5
    Shariat SF, Godoy G, Lotan Y et al. Advanced patient age is associated with inferior cancer-specific survival after radical nephroureterectomy. BJU Int. 2010; 105: 16721677.
  • 6
    Nielsen ME, Shariat SF, Karakiewicz PI et al. Advanced age is associated with poorer bladder cancer-specific survival in patients treated with radical cystectomy. Eur. Urol. 2007; 51: 699706; discussion 706–8.
  • 7
    Chromecki TF, Ehdaie B, Novara G et al. Chronological age is not an independent predictor of clinical outcomes after radical nephroureterectomy. World J. Urol. 2011; 29: 473480.
  • 8
    Roupret M, Hupertan V, Seisen T et al. Prediction of cancer specific survival after radical nephroureterectomy for upper tract urothelial carcinoma: development of an optimized postoperative nomogram using decision curve analysis. J. Urol. 2013; 189: 16621669.
  • 9
    Xylinas E, Kluth L, Mangal S, Karakiewicz PI, Shariat SF. Predictive tools for clinical decision-making and counseling of patients with upper tract urothelial carcinoma. World J. Urol. 2013; 31: 3136.
  • 10
    Cha EK, Shariat SF, Kormaksson M et al. Predicting clinical outcomes after radical nephroureterectomy for upper tract urothelial carcinoma. Eur. Urol. 2012; 61: 818825.
  • 11
    Hollingsworth JM, Miller DC, Daignault S, Hollenback BK. Five-year survival after surgical treatment for kidney cancer: a population-based competing risk analysis. Cancer 2007; 109: 17631768.
  • 12
    Albertsen PC, Hanley JA, Gleason DF, Barry MJ. Competing risk analysis of men aged 55 to 74 years at diagnosis managed conservatively for clinically localized prostate cancer. JAMA 1998; 280: 975980.
  • 13
    Lughezzani G, Sun M, Shariat SF et al. A population-based competing-risks analysis of the survival of patients treated with radical cystectomy for bladder cancer. Cancer 2011; 117: 103109.
  • 14
    Hisataki T, Miyao N, Masumori N et al. Risk factors for the development of bladder cancer after upper tract urothelial cancer. Urology 2000; 55: 663667.
  • 15
    Kang CH, Yu TJ, Hsieh HH et al. The development of bladder tumors and contralateral upper urinary tract tumors after primary transitional cell carcinoma of the upper urinary tract. Cancer 2003; 98: 16201626.
  • 16
    Howlader N, Noone AM, Krapcho M et al. SEER Cancer Statistics Review, 1975–2009 (Vintage 2009 Populations). National Cancer Institute, Bethesda, MD, 2011. [Cited 1 Jul 2013.] Available from URL: http://seer.cancer.gov/csr/1975_2009_pops09/, based on November 2011 SEER data submission, posted to the SEER web site, April 2012.
  • 17
    Ries LAG, Melbert D, Krapcho M et al. SEER Cancer Statistics Review, 1975–2004. National Cancer Institute Web site, Bethesda, MD, 2004. [Cited 1 Jul 2013.] Available from URL: http://seer.cancer.gov/csr/1975_2004/
  • 18
    Gray JR. A class of K-sample tests for comparing the cumulative incidence of a competing risk. Ann. Statist. 1988; 16: 11411154.
  • 19
    Jason PF, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J. Am. Stat. Assoc. 1999; 94: 496509.
  • 20
    Muller H-G, Wang J-L. Hazard rate estimation under random censoring with varying kernels and bandwidths. Biometrics 1994; 50: 6176.
  • 21
    Lughezzani G, Sun M, Perrotte P et al. Should bladder cuff excision remain the standard of care at nephroureterectomy in patients with urothelial carcinoma of the renal pelvis? A population-based study. Eur. Urol. 2010; 57: 956962.
  • 22
    Yap SA, Schupp CW, Chamie K et al. Effect of age on transitional cell carcinoma of the upper urinary tract: presentation, treatment, and outcomes. Urology 2011; 78: 8792.
  • 23
    Lughezzani G, Sun M, Perrotte P et al. Gender-related differences in patients with stage I to III upper tract urothelial carcinoma: results from the Surveillance, Epidemiology, and End Results database. Urology 2010; 75: 321327.
  • 24
    Kates M, Badalato GM, Gupta M et al. Secondary bladder cancer after upper tract urothelial carcinoma in the US population. BJU Int. 2012; 110: 13251329.
  • 25
    Zigeuner RE, Hutterer G, Chromecki T et al. Bladder tumour development after urothelial carcinoma of the upper urinary tract is related to primary tumour location. BJU Int. 2006; 98: 11811186.
  • 26
    Mullerad M, Russo P, Golijanin D et al. Bladder cancer as a prognostic factor for upper tract transitional cell carcinoma. J. Urol. 2004; 172 (6 Pt 1): 21772181.
  • 27
    Miyake H, Hara I, Kamidono S et al. Multifocal transitional cell carcinoma of the bladder and upper urinary tract: molecular screening of clonal origin by characterizing CD44 alternative splicing patterns. J. Urol. 2004; 172: 11271129.
  • 28
    Hafner C, Knuechel R, Zanardo L et al. Evidence for oligoclonality and tumor spread by intraluminal seeding in multifocal urothelial carcinomas of the upper and lower urinary tract. Oncogene 2001; 20: 49104915.
  • 29
    Audisio RA, Bozzetti F, Gennari R et al. The surgical management of elderly cancer patients; recommendations of the SIOG surgical task force. Eur. J. Cancer 2004; 40: 926938.
  • 30
    Lughezzani G, Burger M, Margulis V et al. Prognostic factors in upper urinary tract urothelial carcinomas: a comprehensive review of the current literature. Eur. Urol. 2012; 62: 100114.
  • 31
    Ariane MM, Colin P, Ouzzane A et al. Assessment of oncologic control obtained after open versus laparoscopic nephroureterectomy for upper urinary tract urothelial carcinomas (UUT-UCs): results from a large French multicenter collaborative study. Ann. Surg. Oncol. 2012; 19: 301308.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgment
  8. Conflict of interest
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
iju12267-sup-0001-si.zip1690K

Fig. S1 Smoothed, model-derived, 5-year cumulative mortality estimates in (a) patients with only one primary tumor diagnosis of UTUC undergoing RNU overall, and in (b) patients receiving RNU with bladder cuff. image, Other-cause mortality; image, CSM; image, bladder-cancer mortality; □, survival.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.