Comparison of Oncologic Outcomes for Open and Laparoscopic Nephroureterectomy: A Multi-Institutional Analysis of 1249 Cases

EUROPEAN UROLOGY 56 (2009) 1 9 available at www.sciencedirect.com journal homepage: www.europeanurology.com Platinum Priority Urothelial Cancer Editorial by Evangelos Liatsikos and Panagiotis Kallidonis on pp. 10 12 of this issue Comparison of Oncologic Outcomes for Open and Laparoscopic Nephroureterectomy: A Multi-Institutional Analysis of 1249 Cases Umberto Capitanio a,b, Shahrokh F. Shariat c, Hendrik Isbarn a, Alon Weizer d, Mesut Remzi e, Marco Roscigno b, Eiji Kikuchi f, Jay D. Raman g, Christian Bolenz h, Karim Bensalah i, Theresa M. Koppie j, Wassim Kassouf k, Mario I. Fernández l, Philipp Ströbel h, Jeffrey Wheat d, Richard Zigeuner m, Cord Langner m, Matthias Waldert e, Mototsugu Oya f, Charles C. Guo n, Casey Ng g, Francesco Montorsi b, Christopher G. Wood n, Vitaly Margulis n, Pierre I. Karakiewicz a, * a Cancer Prognostics and Health Outcomes Unit, University of Montreal Health Center, Montreal, Quebec, Canada b Department of Urology, Vita-Salute San Raffaele, Milan, Italy c Department of Urology, Memorial Sloan-Kettering Cancer Center, New York, NY, USA d University of Michigan, Ann Arbor, MI, USA e Medical University of Vienna, Vienna, Austria f Keio University School of Medicine, Tokyo, Japan g Cornell University, New York, NY, USA h Mannheim Medical Center, University of Heidelberg, Mannheim, Germany i University of Rennes, Rennes, France j University of California Davis, Sacramento, CA, USA k McGill University, Montreal, Quebec, Canada l Clinica Alemana de Santiago, Santiago, Chile m Medical University Graz, Graz, Austria n University of Texas M.D. Anderson Cancer Center, Houston, TX, USA Article info Article history: Accepted March 11, 2009 Published online ahead of print on April 3, 2009 Keywords: Laparoscopy Nephroureterectomy Recurrence Survival Upper tract urothelial carcinoma Abstract Background: Data regarding the oncologic efficacy of laparoscopic nephroureterectomy (LNU) compared to open nephroureterectomy (ONU) are scarce. Objective: We compared recurrence and cause-specific mortality rates of ONU and LNU. Design, setting, and participants: Thirteen centers from three continents contributed data on 1249 patients with nonmetastatic upper tract urothelial carcinoma (UTUC). Measurements: Univariable and multivariable survival models tested the effect of procedure type (ONU [n = 979] vs LNU [n = 270]) on cancer recurrence and cancerspecific mortality. Covariables consisted of institution, age, Eastern Cooperative Oncology Group (ECOG) performance status score, pt stage, pn stage, tumor grade, lymphovascular invasion, tumor location, concomitant carcinoma in situ, ureteral * Corresponding author. Cancer Prognostics and Health Outcomes Unit, University of Montreal Health Center (CHUM), 1058, rue St-Denis, Montréal, Québec, Canada, H2X 3J4. Tel. +1 514 890 8000 35336; Fax: +1 514 412 7363. E-mail address: pierre.karakiewicz@umontreal.ca (P.I. Karakiewicz). 0302-2838/$ see back matter # 2009 European Association of Urology. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.eururo.2009.03.072

2 EUROPEAN UROLOGY 56 (2009) 1 9 cuff management, previous urothelial bladder cancer, and previous endoscopic treatment. Results and limitations: Median follow-up for censored cases was 49 mo (mean: 62). Relative to ONU, LNU patients had more favorable pathologic stages (pt0/ta/ Tis: 38.1% vs 20.8%, p < 0.001) and less lymphovascular invasion (14.8% vs 21.3%, p = 0.02) and less frequently had tumors located in the ureter (64.5 vs 71.1%, p = 0.04). In univariable recurrence and cancer-specific mortality models, ONU was associated with higher cancer recurrence and mortality rates compared to LNU (hazard ratio [HR]: 2.1 [p < 0.001] and 2.0 [p = 0.008], respectively). After adjustment for all covariates, ONU and LNU had no residual effect on cancer recurrence and mortality ( p = 0.1 for both). Conclusions: Short-term oncologic data on LNU are comparable to ONU. Since LNU was selectively performed in favorable-risk patients, we cannot state with certainty that ONU and LNU have the same oncologic efficacy in poor-risk patients. Longterm follow-up data and morbidity data are necessary before LNU can be considered as the standard of care in patients with muscle-invasive or high-grade UTUC. # 2009 European Association of Urology. Published by Elsevier B.V. All rights reserved. 1. Introduction Upper tract urothelial carcinoma (UTUC) is a rare disease, accounting for 5 10% of all renal tumors and 5 6% of all urothelial tumors [1,2]. Radical nephroureterectomy (RNU) represents the standard of care for UTUC, especially for muscle-invasive and/or high-grade disease. Open nephroureterectomy (ONU), however, may be associated with significant morbidity [1]. The role of laparoscopy has been examined in several urologic malignancies [3 6]. In some pathologies, laparoscopic procedures have been shown to be equally effective as open surgery with respect to cancer control outcomes. Moreover, laparoscopic surgery may be associated with less morbidity compared to open surgery. Laparoscopic radical nephrectomy, for example, has become the standard of care at institutions with available laparoscopic expertise [5]. For other procedures such as the radical prostatectomy, the debate about the benefits and potential disadvantages of laparoscopic versus open surgery is ongoing [3,4,6]. In some areas of urologic oncology, the available data regarding the merits of laparoscopy are currently too immature to allow valid comparisons [7,8]. RNU for UTUC represents one such area [9 12]. Within the past decade, only four single-center reports compared cancer control outcomes of laparoscopic nephroureterectomy (LNU) with those of ONU (Table 1). The sample size of LNU cases ranged between 20 and 66, and the cumulative number of subjects was 187 [9 12]. Because of the paucity of data and the single-center nature of these studies, there is a need for large, multicenter analysis of the oncologic efficacy of LNU compared to ONU. Toward this aim, we retrospectively analyzed the data from 1249 patients treated with either LNU or ONU at 13 centers from nine countries on three continents. 2. Materials and methods 2.1. Patient selection In this institutional review board approved study, all participating sites provided the necessary institutional data-sharing agreements prior to Table 1 Results of four single-center studies comparing cancer control outcomes of patients with upper tract urothelial carcinoma who were treated with either open or laparoscopic nephroureterectomy Author Study design No. of patients Year of surgery Mean follow-up, mo Results Waldert et al [9] Retrospective n = 102 1999 2006 ONU: 41 No statistically significant differences were ONU: n = 59 LNU: 41 recorded between the two groups with respect to LNU: n =43 disease recurrence and cancer-specific mortality Roupret et al [10] Retrospective n = 46 1994 2004 ONU: 78 No statistically significant differences were ONU: n = 26 LNU: 68.5 recorded between the two groups with respect to LNU: n =20 disease recurrence and cancer-specific mortality Hsueh et al [11] Retrospective n = 143 1998 2005 ONU: 53.6 No statistically significant differences were ONU: n = 77 LNU: 37.6 LNU: n =66 recorded between the two groups with respect to disease recurrence and cancer-specific mortality Manabe et al [12] Retrospective n = 224 2000 2004 ONU: 28 No statistically significant differences were recorded ONU: n = 166 LNU: 13.6 LNU: n =58 ONU = open nephroureterectomy; LNU = laparoscopic nephroureterectomy. between the two groups with respect to disease recurrence and cancer-specific mortality

EUROPEAN UROLOGY 56 (2009) 1 9 3 initiation of the study. A total of 13 academic centers worldwide provided data. A computerized databank was generated for data transfer. After combining the data sets, reports were generated for each variable to identify data inconsistencies and other data integrity problems. Through regular communication with all sites, resolution of all identified anomalies was achieved before analysis. Prior to final analysis, the database was frozen and the final data set was produced for the current analysis. This study comprised 1453 patients who underwent RNU between 1987 and 2007. Regional lymphadenectomy was generally performed in patients with suspicious lymph nodes on preoperative axial imaging or with adenopathies detected during intraoperative examination. Extended lymphadenectomy was not routinely performed. There were three management options for the distal ureter. First, no bladder cuff was excised. Second, a bladder cuff was excised during open or laparoscopic surgery. Third, a bladder cuff was removed via an endourologic approach. Patients who received neoadjuvant or adjuvant chemotherapy regimens were excluded (n = 41 [2.8%] and n = 163 [11.2%], respectively), as these treatments may confound the effect of nephroureterectomy on recurrence and cancer-specific mortality. This left 1249 assessable patients. 2.2. Pathologic evaluation All surgical specimens were processed according to standard pathologic procedures, and all slides were re-reviewed by genitourinary pathologists at each participating institution according to prospectively defined criteria. All pathologists were blinded to clinical outcomes. All specimens were histologically confirmed to be urothelial carcinomas. UTUC was defined as urothelial carcinomas located in the renal pelvis or calices as well as tumors located within the ureter. Tumors were staged according to the 2002 American Joint Committee on Cancer/Union Internationale Contre le Cancer (AJCC/UICC) Tumor-Node-Metastasis (TNM) classification [13]. Tumor grading was assessed according to the 1998 World Health Organization/International Society of Urologic Pathology (WHO/ISUP) consensus classification [14]. Lymphovascular invasion was defined as the presence of tumor cells within an endothelium-lined space without underlying muscular walls. The presence of concomitant carcinoma in situ (CIS) and the tumor architecture were also assessed in every representative section. In case of concomitant renal pelvis and ureteral tumors, the dominating tumor was used to determine the primary location. 2.3. Follow-up Patients were generally followed every 3 4 mo for the first year following RNU, every 6 mo from the second through the fifth year, and annually thereafter. Follow-up consisted of a history, physical examination, routine blood work and serum chemistry studies, urinary cytology, chest radiography, cystoscopic evaluation of the urinary bladder, and radiographic evaluation of the contralateral upper urinary tract. Elective bone scans, chest computed tomography (CT), or magnetic resonance imaging were performed when clinically indicated. Disease recurrence was defined as local failure in the operative site, regional lymph nodes, or distant metastases. Cause of death was determined by the treating physicians and was defined after chart review and corroboration by death certificate or by death certificate alone. All patients who were coded as dead of UTUC also had evidence of upper tract cancer recurrence. In cancer-specific mortality analyses, perioperative mortalities (any death within 30 d of surgery or before discharge) were censored. 2.4. Statistical analysis The t test and the chi-square test were used to evaluate the association between variables between the two procedure types (LNU vs ONU). Recurrence and survival rates after nephroureterectomy were estimated using the Kaplan-Meier method. Univariable and multivariable Cox regression models addressed recurrence and cancer-specific mortality after nephroureterectomy. Covariables consisted of institution, age, Eastern Cooperative Oncology Group (ECOG) performance status (PS) score, T stage, tumor grade, tumor location, concomitant CIS, lymphovascular invasion, N stage, ureteral management, previous bladder cancer, and previous endoscopic surgery. All reported p-values are 2-sided, and statistical significance was set at 0.05. Statistical analyses were performed with SPSS 13.0 (SPSS Inc, Chicago, IL, USA) or S-PLUS Professional software (MathSoft Inc, Seattle, WA, USA). 3. Results Table 2 shows the distribution of patient characteristics. Data were stratified according to procedure type (LNU vs Table 2 Descriptive characteristics of 1249 patients treated with open or laparoscopic nephroureterectomy for urothelial carcinoma (UC) of the upper urinary tract between 1987 and 2007 Variable Open nephroureterectomy (n = 979) Laparoscopic nephroureterectomy (n = 270) p-value Patient characteristics Age, yr 0.01 Mean (median) 68.3 (69) 70.2 (70) Range 27 97 36 94 Gender, n (%) 0.8 Male 665 (67.9) 181 (67.0) Female 314 (32.1) 89 (33.0) Race, n (%) 0.5 Caucasian 809 (82.6) 228 (84.4) Other 170 (17.4) 42 (15.6) ECOG PS, n (%) 0.01 0 697 (71.2) 168 (62.2) 1 3 282 (28.8) 102 (37.8) Previous bladder cancer, n (%) 0.2 Yes 253 (25.8) 60 (22.2) No 726 (74.2) 210 (77.8)

4 EUROPEAN UROLOGY 56 (2009) 1 9 Table 2 (Continued ) Variable Open nephroureterectomy (n = 979) Laparoscopic nephroureterectomy (n = 270) p-value Tumor characteristics pt stage, n (%) <0.001 pt0/ta/tis 204 (20.8) 103 (38.1) pt1 229 (23.4) 69 (25.6) pt2 202 (20.6) 35 (13.0) pt3 306 (31.3) 59 (21.9) pt4 38 (3.9) 4 (1.5) pn stage, n (%) <0.001 pnx 564 (57.6) 204 (75.6) pn0 345 (35.2) 60 (22.2) pn1 70 (7.2) 6 (2.2) Grade, n (%) 0.5 Low 395 (40.3) 103 (38.1) High 584 (59.7) 167 (61.9) Tumor location, n (%) 0.043 Renal pelvis 631 (64.5) 192 (71.1) Ureter 348 (35.5) 78 (28.9) Lymphovascular invasion, n (%) 0.02 Yes 209 (21.3) 40 (14.8) No 770 (78.7) 230 (85.2) Concomitant CIS, n (%) 0.5 Yes 254 (25.9) 65 (24.1) No 725 (74.1) 205 (75.9) Surgical characteristics Number of cases performed by each participating institution, n (%) <0.001 UT MD Anderson Cancer Center, Houston, TX, USA 119 (12.2) 60 (2.2) Weill Medical College of Cornell University, NY, USA 41 (4.2) 27 (10.0) Clinica Alemana de Santiago, Santiago, Chile 2 (0.2) 4 (1.5) University of Vienna, Vienna, Austria 54 (5.5) 24 (8.9) McGill University, Montreal, Quebec, Canada 81 (8.3) 17 (6.3) CHU Pontchaillou, Rennes, France 20 (2.0) 1 (0.4) UT Southwestern Medical Center, Dallas, TX, USA 18 (1.8) 23 (8.5) Keio University School of Medicine, Tokyo, Japan 141 (14.4) 26 (9.6) UC Davis, Sacramento, CA, USA 21 (2.1) 16 (5.9) University of Graz, Graz, Austria 222 (22.7) 1 (0.4) University Vita-Salute, Milan, Italy 109 (11.1) 0 (0) University Hospital Mannheim, Mannheim, Germany 118 (12.1) 9 (3.3) University of Michigan Hospital, Ann Arbor, MI, USA 33 (3.4) 62 (23.0) Perioperative mortality, n (%) 6 (0.6) 1 (0.4) 1.0 Management of the ureter, n (%) <0.001 Without excision of a bladder cuff 576 (58.8) 135 (50.0) Open or laparoscopic cuff excision 398 (40.7) 79 (29.3) Cuff excision via endoscopy 5 (0.5) 56 (20.7) Lymphadenectomy performed, n (%) <0.001 Yes 415 (42.4) 66 (24.4) No 564 (57.6) 204 (75.6) 0.2 Number of removed lymph nodes (when lymphadenectomy was performed) Mean (median) 5.5 (4) 4.7 (3) Range 1 34 1 31 Outcome Recurrence analysis Follow-up of censored patients, mo Mean (median) 73.0 (63.0) 31.0 (25.9) Range 0.1 250.0 0.1 148.0 Number of events (%) 250 (25.5) 27 (10.0) 5-yr recurrence free survival, % 76.2 86.8 Cancer-specific mortality analysis Follow-up of censored patients Mean (median) 70.7 (59.0) 30.6 (25.6) Range 0.1 250.0 0.1 148.0 Number of events (%) 217 (22.8) 21 (7.8) 5-yr cancer-specific mortality-free survival, % 73.1 85.8 ECOG PS = Eastern Cooperative Oncology Group performance status; CIS = carcinoma in situ.

EUROPEAN UROLOGY 56 (2009) 1 9 5 Table 3 Univariable and multivariable Cox regression models predicting recurrence and cause-specific mortality of upper tract urothelial carcinoma after surgery with curative intent Recurrence Cancer-specific mortality UVA MVA UVA MVA Variable HR; p-value HR; p-value HR; p-value HR; p-value Procedure Open vs laparoscopic 2.1; p < 0.001 1.44; p = 0.1 2.00; p = 0.008 1.54; p = 0.1 Age (continuously coded) 1.02; p = 0.002 1.01; p = 0.4 1.03; p < 0.001 1.02; p = 0.04 T stage NA; p < 0.001 NA; p < 0.001 NA; p < 0.001 NA; p < 0.001 T1 vs Ta/Tis 2.51; p = 0.006 1.78; p = 0.09 2.29; p = 0.02 1.74; p = 0.1 T2 vs Ta/Tis 5.49; p < 0.001 3.39; p < 0.001 5.87; p < 0.001 3.73; p < 0.001 T3 vs Ta/Tis 13.94; p < 0.001 8.00; p < 0.001 14.68; p < 0.001 8.79; p < 0.001 T4 vs Ta/Tis 51.87; p < 0.001 21.90; p < 0.001 49.03; p < 0.001 22.05; p < 0.001 N stage N+ vs N0 5.00; p < 0.001 1.73; p = 0.003 5.21; p < 0.001 1.71; p = 0.007 Grade High vs low 4.90; p < 0.001 2.65; p < 0.001 4.90; p < 0.001 2.45; p < 0.001 Lymphovascular invasion Yes vs no 3.07; p < 0.001 1.32; p = 0.06 3.44; p < 0.001 1.48; p = 0.02 ECOG PS 1 vs 0 1.84; p < 0.001 1.61; p = 0.001 1.70; p < 0.001 1.46; p = 0.02 Tumor location Ureter vs renal pelvis 1.0; p = 0.7 1.14; p = 0.3 1.03; p = 0.8 1.25; p = 0.1 Ureteral management NA; p = 0.2 NA; p = 0.3 NA; p = 0.1 NA; p = 0.2 Open cuff vs no cuff 0.94; p = 0.6 1.61; p = 0.1 1.19; p = 0.2 1.87; p = 0.07 Endoscopic cuff vs no cuff 0.42; p = 0.06 1.08; p = 0.9 0.47; p = 0.1 1.02; p = 1.0 Previous bladder cancer Yes vs no 0.95; p = 0.7 1.40; p = 0.04 1.11; p = 0.5 1.58; p = 0.008 Previous endoscopy Yes vs no 0.59; p = 0.02 0.78; p = 0.3 0.67; p = 0.08 0.78; p = 0.4 Concomitant CIS Yes vs no 1.23; p = 0.1 0.94; p = 0.7 1.20; p = 0.2 0.89; p = 0.4 Institution NA; p < 0.001 NA; p = 0.009 NA; p = 0.002 NA; p = 0.08 UVA = univariable; MVA = multivariable; HR = hazard ratio; ECOG PS = Eastern Cooperative Oncology Group performance score; CIS = carcinoma in situ. ONU). Relative to ONU, LNU patients were older ( p =0.01) and had worse PS scores ( p = 0.01). Conversely, LNU patients had more favorable pathologic stages ( p < 0.001), more frequent papillary architecture ( p = 0.02), less lymphovascular invasion ( p = 0.02), and less frequent primary tumor location in the ureter versus the renal pelvis ( p = 0.043). There was no statistically significant difference in the rates of perioperative mortality between the two groups ( p = 1.0). The 5-yr recurrence-free survival estimates and the 5-yr cancer-specific mortality-free survival estimates were 86.8% and 85.8% for patients treated with LNU versus 76.2% and 73.1% for patients treated with ONU, respectively (Table 2). In univariable unadjusted analyses, ONU was associated with statistically significantly higher recurrence (hazard ratio [HR]: 2.1; p < 0.001) and cancer-specific mortality (HR: 2.0; p < 0.001) rates than LNU (Table 3). After adjustment for tumor stage, no statistically significant differences with respect to recurrence and cancer-specific mortality were recorded for ONU versus LNU (Figs. 1 and 2). Moreover, in the multivariable models that were adjusted for the effects of all other tested covariates that may confound the relation between procedure type and recurrence or cancer-specific mortality, no difference between ONU and LNU with regard to disease recurrence ( p = 0.1) and cancer-specific mortality ( p = 0.1) was recorded (Table 3). Lack of residual statistical significance for the variable defining procedure type can, therefore, be interpreted as the result of controlling for various selection biases that were present when treatment assignment was performed. 4. Discussion The current standard of care for UTUC consists of ONU [1]. LNU, however, is becoming more commonly applied and is becoming established as an alternative to ONU at centers with adequate laparoscopic expertise [15,16]. The first reports comparing LNU to ONU were published in 1993 [17,18]. Since then, only four other reports compared longterm cancer control of LNU and ONU patients [9 12]. Their individual sample sizes were limited (20 66 LNU patients). Consequently, the evidence supporting the efficacy of LNU versus ONU remains weak [9 12]. As such, the use of LNU cannot be safely recommended as an alternative to ONU in formal guidelines.

6 EUROPEAN UROLOGY 56 (2009) 1 9 Fig. 1 Recurrence-free survival rates of patients treated with either laparoscopic or open nephroureterectomy, according to pt stage. Given the relatively low incidence of UTUC, no single institution has sufficient numbers of ONU and LNU cases to validly address the efficacy of procedure type (ONU vs LNU) on recurrence and cause-specific mortality rates. Moreover, single-institution series may be fraught with selection biases related to subject assignment to ONU or LNU procedures. Therefore, to ensure adequate statistical power and generalizability, we combined the 1249 patients treated with nephroureterectomy at 13 centers from nine countries on three continents. The comparison between the characteristics of LNU and ONU cases revealed very important selection biases that distinguished LNU from ONU cases. Specifically, LNU cases had more favorable pathologic characteristics than ONU cases (Table 2). Selection of clinically and pathologically more favorable cases is not a rarity when a novel surgical technique is being evaluated. Such practice of selective referral of favorable-risk patients toward LNU resulted in substantially better recurrence and survival patterns for LNU cases in univariable analyses (Table 3). To control for these important selection biases, we relied on multivariable modeling techniques. After adjusting for the effects of tumor stage, grade, patient PS score, and other important confounding variables, no difference in disease recurrence or cause-specific mortality rates were observed between the two procedures (ONU vs LNU) (Table 3). Taken together, our results indicate that LNU is offered to favorable-risk patients and that its results may appear to be better than those of ONU. ONU is offered to poorer risk patients and is, expectedly, associated with higher recurrence and mortality rates; however, when we adjusted for the effects of all tumor and patient characteristics, there was no difference in the oncologic efficacy of ONU versus LNU. Our findings could be interpreted as direct evidence of equivalence between LNU and ONU when recurrence and mortality are considered; however, the majority of the patient population treated with LNU had favorable disease characteristics. Therefore, our findings of equivalence between ONU and LNU are mainly limited to patients with biologically and clinically less aggressive disease such as those with non muscle-invasive stage disease, absence of nodal metastasis, and absence of lymphovascular invasion (Table 2). Although not all of these features can be identified preoperatively, many pathologic features are closely related to T and N stages that are known before surgery. Since LNU was selectively performed in favorable-risk patients, we cannot state with certainty that ONU and LNU have the same oncologic efficacy in poor-risk patients [19,20]. This question can only be addressed by a formal randomized clinical trial. Given the low incidence of UTUC, it is unlikely that such a trial will ever be performed and completed. After more than a decade of debate about the merits and the disadvantages of laparoscopic radical

EUROPEAN UROLOGY 56 (2009) 1 9 7 Fig. 2 Cancer-specific mortality-free survival rates of patients treated with either laparoscopic or open nephroureterectomy, according to pt stage. prostatectomy versus open surgery, no randomized trial has been performed to assess the oncologic and functional differences between these treatment modalities [3,4]. Nonetheless, a multi-institutional randomized trial should be strongly encouraged. Although the pathologic features of LNU cases were more favorable than those of ONU cases, LNU was more frequently performed in older patients and in individuals with worse PS scores (Table 2). Age and PS score represent known selection criteria for laparoscopic surgery due to lower morbidity relative to open surgery [21,22]. Some limitations apply to our findings. The selection of patients into one or the other treatment type represents the most important bias. Given the more limited number of LNUs, the expertise with this procedure is invariably lower than the expertise with ONU. We do not have data addressing individual surgical volume or institution volume for purposes of adjustment [23]. Moreover, we could not include data on body mass index or comorbidity. These and other variables defining patient characteristics should also be considered in further analyses. Exclusion of patients exposed to chemotherapy represents a potential limitation of the current study, as the included patients with lymph node metastasis did not receive adjuvant chemotherapy. This exclusion may be interpreted as undertreatment. No convincing data, however, demonstrate a survival benefit of adjuvant or salvage chemotherapy for patients with lymph node metastasis after nephroureterectomy. Therefore, we restricted our analysis to patients unexposed to any form of chemotherapy to avoid an unanticipated effect on survival. In patients treated with radical cystectomy, for example, the use of adjuvant chemotherapy was associated with worse survival [24,25]. Lack of routine and standardized use of CT imaging may have affected recurrence rates. Recurrence, for example, may be detected at a later time if CT scans are not used at standard intervals. Lack of routine and standard imaging, however, has not affected the reported mortality rates, which represents the ultimate measure of efficacy. As in other disease models, the accuracy of death certificates is not perfect. A higher degree of accuracy could be expected from prospective studies that are performed in meticulous fashion. Lack of data on local recurrence also represents an important limitation of this analysis. The heterogeneity of approaches to the management of the distal ureter adds another bias. An analysis restricted to patients who underwent a bladder cuff resection would be ideal. Unfortunately, such a measure would seriously compromise the study size. Restriction of the study population to patients treated with bladder cuff excision could also introduce another bias, since a bladder cuff was not removed in a substantial proportion of LNU (50.0%) and ONU (58.8%) patients. Therefore, nephroureterectomies are commonly performed without bladder cuff removal. Moreover, a high number of patients in the current series did not undergo a lymphadenectomy, and the median number of removed lymph nodes was relatively low (four and three,

8 EUROPEAN UROLOGY 56 (2009) 1 9 respectively), which may have compromised exact staging. It is conceivable that a more extensive lymph node dissection could improve cancer control outcomes [26]. No convincing data, however, indicate the prognostic benefit of an extended lymphadenectomy relative to a standard lymphadenectomy or to no lymphadenectomy. Consequently, there is no established standard of care regarding the performance and/or the extent of lymphadenectomy at the time of nephroureterectomy. Finally, the retrospective nature of this study represents another important bias that cannot be controlled for with statistical tools, and the multicenter nature of the study may further confound this limitation. Consequently, the validity of our findings regarding the equivalence of LNU versus ONU cancer control outcomes requires confirmation in prospective studies. Despite those limitations, our findings are derived from the largest available series and are more generalizable than any of the previous reports [9 12]. Although our findings require corroboration, they do provide strong data about the oncologic equivalence of both approaches (LNU vs ONU). 5. Conclusions Our data can be used as evidence for equivalent cancer control outcomes (recurrence and mortality) between LNU and ONU in patients with predominantly favorable clinical and pathologic UTUC features. Further analyses ideally, a randomized trial are needed to generalize these conclusions to patients with more unfavorable disease characteristics. Author contributions: Pierre Karakiewicz had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Capitanio, Isbarn, Shariat, Karakiewicz. Acquisition of data: Shariat, Weizer, Remzi, Roscigno, Kikuchi, Raman, Bolenz, Bensalah, Koppie, Kassouf, Fernández, Ströbel, Wheat, Zigeuner, Langner, Waldert, Oya, Guo, Ng, Montorsi, Wood, Margulis, Karakiewicz. Analysis and interpretation of data: Capitanio, Shariat, Karakiewicz, Isbarn. Drafting of the manuscript: Capitanio, Shariat, Karakiewicz, Isbarn. Critical revision of the manuscript for important intellectual content: Capitanio, Shariat, Weizer, Remzi, Roscigno, Kikuchi, Raman, Bolenz, Bensalah, Koppie, Kassouf, Fernández, Ströbel, Wheat, Zigeuner, Langner, Waldert, Oya, Guo, Ng, Montorsi, Wood, Margulis, Karakiewicz, Isbarn. Statistical analysis: Capitanio, Shariat, Karakiewicz, Isbarn. Obtaining funding: Karakiewicz. Administrative, technical, or material support: None. Supervision: Capitanio, Shariat, Weizer, Isbarn, Remzi, Roscigno, Kikuchi, Raman, Bolenz, Bensalah, Koppie, Kassouf, Fernández, Ströbel, Wheat, Zigeuner, Langner, Waldert, Oya, Guo, Ng, Montorsi, Wood, Margulis, Karakiewicz. Other (specify): None. Financial disclosures: I certify that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None. Funding/Support and role of the sponsor: Pierre Karakiewicz is partially supported by the University of Montreal Health Center Urology Associates, Fonds de la Recherche en Santé du Quebec, the University of Montreal Department of Surgery, and the University of Montreal Health Center (CHUM) Foundation, which supported this analysis. References [1] Oosterlinck W, Solsona E, van der Meijden APM, et al. EAU guidelines on diagnosis and treatment of upper urinary tract transitional cell carcinoma. Eur Urol 2004;46:147 54. [2] Munoz JJ, Ellison LM. Upper tract urothelial neoplasms: incidence and survival during the last 2 decades. J Urol 2000;164:1523 5. [3] Touijer K, Eastham JA, Secin FP, et al. Comprehensive prospective comparative analysis of outcomes between open and laparoscopic radical prostatectomy conducted in 2003 to 2005. J Urol 2008; 179:1811 7, discussion 1817. [4] Parsons JK, Bennett JL. Outcomes of retropubic, laparoscopic, and robotic-assisted prostatectomy. Urology 2008;72:412 6. [5] Eskicorapci SY, Teber D, Schulze M, Ates M, Stock C, Rassweiler JJ. Laparoscopic radical nephrectomy: the new gold standard surgical treatment for localized renal cell carcinoma. ScientificWorldJournal 2007;7:825 36. [6] Rassweiler J, Hruza M, Teber D, Su L-M. Laparoscopic and robotic assisted radical prostatectomy critical analysis of the results. Eur Urol 2006;49:612 24. [7] Haber G-P, Crouzet S, Gill IS. Laparoscopic and robotic assisted radical cystectomy for bladder cancer: a critical analysis. Eur Urol 2008;54:54 64. [8] Huang GJ, Stein JP. Open radical cystectomy with lymphadenectomy remains the treatment of choice for invasive bladder cancer. Curr Opin Urol 2007;17:369 75. [9] Waldert M, Remzi M, Klingler HC, Mueller L, Marberger M. The oncological results of laparoscopic nephroureterectomy for upper urinary tract transitional cell cancer are equal to those of open nephroureterectomy. BJU Int 2009;103:66 70. [10] Roupret M, Hupertan V, Sanderson KM, et al. Oncologic control after open or laparoscopic nephroureterectomy for upper urinary tract transitional cell carcinoma: a single center experience. Urology 2007;69:656 61. [11] Hsueh TY, Huang YH, Chiu AW, Huan SK, Lee YH. Survival analysis in patients with upper urinary tract transitional cell carcinoma: a comparison between open and hand-assisted laparoscopic nephroureterectomy. BJU Int 2007;99:632 6. [12] Manabe D, Saika T, Ebara S, et al. Comparative study of oncologic outcome of laparoscopic nephroureterectomy and standard nephroureterectomy for upper urinary tract transitional cell carcinoma. Urology 2007;69:457 61. [13] Greene FL, Balch CM, Fleming ID, et al. editors. AJCC Cancer Staging Handbook: TNM Classification of Malignant Tumors. 6th ed. Chicago, IL: American Joint Committee on Cancer; 2002. [14] Epstein JI, Amin MB, Reuter VR, Mostofi FK. The World Health Organization/International Society of Urological Pathology consensus classification of urothelial (transitional cell) neoplasms of the urinary bladder. Bladder Consensus Conference Committee. Am J Surg Pathol 1998;22:1435 48. [15] Muntener M, Nielsen ME, Romero FR, et al. Long-term oncologic outcome after laparoscopic radical nephroureterectomy for upper tract transitional cell carcinoma. Eur Urol 2007;51:1639 44. [16] Tsujihata M, Nonomura N, Tsujimura A, Yoshimura K, Miyagawa Y, Okuyama A. Laparoscopic nephroureterectomy for upper tract transitional cell carcinoma: comparison of laparoscopic and open surgery. Eur Urol 2006;49:332 6. [17] Janetschek G, Reissigl A, Peschel R, Bartsch G. Laparoscopic nephroureterectomy. Br J Urol 1993;72:987 8. [18] Rassweiler JJ, Henkel TO, Potempa DM, Coptcoat L, Alken P. The technique of transperitoneal laparoscopic nephrectomy, adrenalectomy and nephroureterectomy. Eur Urol 1993;23:425 30.

EUROPEAN UROLOGY 56 (2009) 1 9 9 [19] Li C-C, Chang T-H, Wu W-J, et al. Significant predictive factors for prognosis of primary upper urinary tract cancer after radical nephroureterectomy in Taiwanese patients. Eur Urol 2008;54: 1127 35. [20] Novara G, De Marco V, Gottardo F, et al. Independent predictors of cancer-specific survival in transitional cell carcinoma of the upper urinary tract: multi-institutional dataset from 3 European centers. Cancer 2007;110:1715 22. [21] Kercher KW, Joels CS, Matthews BD, Lincourt AE, Smith TI, Heniford BT. Hand-assisted surgery improves outcomes for laparoscopic nephrectomy. Am Surg 2003;69:1061 6. [22] Matin SF, Madsen LT, Wood CG. Laparoscopic cytoreductive nephrectomy: the M. D. Anderson Cancer Center experience. Urology 2006;68:528 32. [23] Roscigno M, Cozzarini C, Bertini R, et al. Prognostic value of lymph node dissection in patients with muscle-invasive transitional cell carcinoma of the upper urinary tract. Eur Urol 2008;53:794 802. [24] Shariat SF, Karakiewicz PI, Palapattu GS, et al. Nomograms provide improved accuracy for predicting survival after radical cystectomy. Clin Cancer Res 2006;12:6663 76. [25] Karakiewicz PI, Shariat SF, Palapattu GS, et al. Nomogram for predicting disease recurrence after radical cystectomy for transitional cell carcinoma of the bladder. J Urol 2006;176:1354 61, discussion 1361 2. [26] Brausi MA, Gavioli M, De Luca G, et al. Retroperitoneal lymph node dissection (RPLD) in conjunction with nephroureterectomy in the treatment of infiltrative transitional cell carcinoma (TCC) of the upper urinary tract: impact on survival. Eur Urol 2007;52:1414 20.