Upper urinary tract urothelial carcinoma (UTUC)

JOURNAL OF ENDOUROLOGY Volume 29, Number 8, August 2015 ª Mary Ann Liebert, Inc. Pp. --- --- DOI: 10.1089/end.2015.0044 Original Research Immediate Nephroureterectomy or After Attempting Conservative Treatment, on Elective Indications, for Upper Urinary Tract Urothelial Carcinoma: Comparison of the Pathology Reports on a Retrospective Monocentric Study Mattieu Haddad, MD, Jonathan Cloutier, MD, Jean Nicolas Cornu, MD, PhD, Luca Villa, MD, Jean-Baptiste Terrasa, MD, Sabrina Benbouzid, MD, Marie Audouin, MD, Olivier Cussenot, MD, PhD, and Olivier Traxer, MD, PhD Abstract Objective: Conservative treatment (CT) with flexible ureteroscopy and laser ablation is an alternative to radical nephroureterectomy (RNU) for the treatment of the upper urinary tract urothelial carcinoma (UTUC). The purpose of this study was to compare the pathology results obtained after immediate RNU or after attempt of CT for elective indication. Patients and Methods: A retrospective study was conducted in a single tertiary center. All patients who had an RNU for urothelial carcinoma between 2007 and 2012 have been included. The patients were classified into two groups: group 1 is immediate RNU, and group 2 is RNU after CT (only elective indications). Preoperative data collected were as follows: age, sex, chronic kidney failure, radiological classification for cancer staging (TNM), tumor size, localization, and multifocal indication of CT. The pathological RNU data collected were tumor stage and grade. The T stage was divided into two groups (primary endpoint): pta-t1-t2 and pt3-t4. The v 2 test and Mann Whitney was performed to compare the independent qualitative and quantitative variables, Results: A total of 51 patients were included (40 patients in the immediate RNU group and 11 patients in the delayed RNU group after CT). Patients in both groups had comparable characteristics regarding age, sex, location, T stage, and preoperative tumor grade. On final pathology, 23 tumors were classified as pta-t1-t2 in the immediate RNU group compared with 6 in the delayed RNU group. Seventeen and five tumors were classified as T3 in group 1 and group 2, These results were not significantly different between both groups ( p = 0.866). The pathological RNU grade was not significantly different between the groups. Conclusion: Within the limits of this retrospective study, the pathological RNU data showed no significant difference when RNU was done immediately or after CT for UTUC. Introduction Upper urinary tract urothelial carcinoma (UTUC) represents 5% 10% of urothelial tumors and less than 1% of the genitourinary tract tumors. 1 They can affect any place covered by the urothelium, but they are more often located in the pyelocaliceal system (in two-thirds of cases). 2 If the computed-tomography scanner urography (CT-Scan) has a diagnostic sensitivity of 40% 96% for the UTUC depending on the size, it can have a false negative for small tumors, namely those of less than 3 mm. 3 According to the actual recommendations, 1 the therapeutic options for the treatment of UTUC are as follows: radical nephroureterectomy (RNU) with bladder cuff excision, segmental ureterectomy (in very selected cases), or conservative treatment (CT) with flexible ureteroscopy (URS) and Holmium:YAG laser tumor vaporization. Nowadays, the RNU is still the treatment of reference for UTUC. 1 The CT is an alternative to the radical treatment and may be proposed in face of an imperative indication (solitary kidney, chronic kidney failure, or bilateral tumors) or with an elective indication: unifocal tumor of less than 1 cm, low-grade, without Department of Urology, Tenon Hospital, Paris, France. 1

2 HADDAD ET AL. suspicion of local infiltration, regional or systemic extension. The patient needs accepting close monitoring with CT scan and frequent URS with general anesthesia. One of the main issues of CT with URS is the relatively high risk of ipsilateral tumor recurrences. If specific recommendations propose the realization of an RNU in case of recurrence, 1 there is no specific recommendation in the circumstance of recurrence for patients under CT for imperative indication. For some authors, iterative URS with treatment can be done as often as local conditions allowed. In case of patients under CT with elective indications, the exact timing or indication to stop the CT and perform the RNU is not of mutual consent between urologists. According to the literature, there are 13% 19% of patients treated conservatively who will ultimately undergo an RNU. 4 7 The objective of this study was to compare the pathological RNU data for UTUC made immediately after the diagnosis or after CT for elective indications, to determine whether the disease progressed to the higher stage and grade with CT. Patients and Methods A retrospective, monocentric study was conducted in a tertiary reference center. The institutional review board approved the study. The records of patients who had an RNU for UTUC between January 2007 and December 2012 were reviewed. The patients were classified into two groups: group 1 is immediate RNU, as a primary treatment option, following the diagnosis and group 2 is delayed RNU following at least one CT by URS with curative intent. Some patients had an URS just for biopsy and were classified as group 1. Patients with imperative indications were excluded due to the risk of bias. Main indications for diagnostic URS were refinement of the diagnosis or uncertainty regarding the diagnosis. Recommendations of the European Association of Urology have been used to perform a conservative treatment for UTUC: unifocal tumor, less than 1 cm, low grade, no sign of infiltrative lesion on CT urography, and understanding of close follow-up. Preoperative data collected were as follows: age at diagnosis, sex, presence or absence of kidney failure, radiological classification for cancer staging (TNM), tumor size, location, number, and indication of CT if necessary. For patients with a CT, the following additional data were collected: date of the first and last URS, number of therapeutic URS, biopsy T stage, and grade. All patients were systematically treated with a digital flexible ureteroscope (Storz Flex-XCÔ or Olympus URF-VÔ) with a narrow band imaging (NBI), while available, during the initial CT as well as during the follow-up to optimize tumor detection. A single expert operator performed all the URS (O.T.). Intraoperative data collected for patients at the RNU were as follows: date of surgery, type of incision, tumor side, location, and size. The data were tumor classification for cancer staging after surgery on a pathology report (ptnm). The ptnm classification was divided in two groups: pta-t1-t2 and pt3-t4. The tumor location categorization was as follows: caliceal, renal pelvis, ureteral, and multifocal. The primary outcome was to compare the UTUC ptnm stage after the RNU between the immediate RNU group and the RNU group after intent of CT. The XlStat2013 software (Addin Soft, Paris, France) was used for statistical analysis. The v 2 and Mann Whitney test were used to compare the independent qualitative and quantitative variables, Results Population Atotalof51patientswereincluded(40patientsinthe immediate RNU group and 11 in the delayed RNU group). Among the 40 patients in group 1, 12 (30%) underwent a diagnostic URS, with biopsies and without an attempt of CT. The other 28 patients (70%) of group 1 had immediate RNU without biopsy and with correlation of clinical, biological, and radiological symptoms. Three of 11 patients Table 1. Patients Characteristics According to Their Group and Postoperative Staging RNU (n = 40) CT (n = 11) p Age mean (years) 68.8 12.4 (40 90) 69.1 11.2 (50 84) 0.792 Sex Male 27 (67%) 6 (54%) 0.426 Female 13 (33%) 5 (46%) Mean radiological size (mm) a 47.6 20.3 (11 90) 22.7 9.6 (12 40) 0.008 Tumor location Caliceal 15 (37%) 3 (27%) 0.845 Renal pelvis 12 (30%) 3 (27%) Ureteral 7 (18%) 3 (27%) Multifocal 6 (15%) 2 (19%) T stage Ta-T1-T2 23 (61.5%) 6 (54.5%) 0.866 T3-T4 17 (38.5%) 5 (45.5%) Lymph node involvement (N + ) 10 (25%) 0 (0%) Histopathological grade Low 6 (15%) 4 (36%) 0.114 High 34 (85%) 7 (64%) a Twenty three RNU patients and five CT patients did not have the radiological size of their tumor. CT = conservative treatment; RNU = radical nephroureterectomy; T = tumor stage.

IMMEDIATE RNU OR CT FOR UTUC: PATHOLOGY REPORTS 3 (27%) in group 2 have failed the intent of CT at the first control URS, mostly due to intraoperative complications such as difficulty to control bleeding. The median time to RNU in group 2 was 143 days. Twenty three of 51 patients (45%) had at least one URS. Patients in both groups had comparable characteristics regarding age, sex, tumor location, stage, and grade (Table 1). Tumor characteristics At the diagnosis, the majority of the tumors were localized into a calyx (Table 1). The location was not significantly different between the two groups ( p = 0.845). The only characteristic that differed between the two groups was the radiological diameter of the tumor: The average size was 47.6 and 22.7 mm for groups 1 and 2, respectively ( p = 0. 008). Pathologic data Grade. Preoperative biopsies in the CT group revealed seven tumors of high grade, two of low grade, and two that were not contributive. In the immediate RNU group, of the 12 patients who had a diagnostic URS, 6 biopsies showed highgrade tumors (50%), 1 showed low-grade tumors (8%), 2 showed tumors that were not contributive (17%), and 3 showed missing data (25%) (Table 2). Final pathology after RNU, including immediate RNU (group 1) and after attempt of conservative treatment (group 2), found 10 low-grade tumors (20%): 6 in group 1 compared with 4 in group 2. Forty one tumors were high-grade ones (80%): 34 in group 1 and 7 in group 2. These values were not significantly different between both groups ( p = 0.114). From the 13 preoperative biopsies that showed high-grade tumors, 2 were finally low-grade ones (17%) and all of them came from the conservative treatment group. From the four preoperative low-grade tumors, two were found to be highgrade tumors (67%), one came from the CT group, and the other came from the immediate RNU group. Table 2. Preoperative Biopsy Analysis After CT RNU (n = 40) (%) CT (n = 11) (%) p T stage a Ta-T1-T2 4 (10) 7 (63.6) T3-T4 0 (0) 0 (0) Histological grade b Low 1 (2.5) 2 (18.2) High 6 (15) 7 (63.6) a Eight and 4 biopsies during ureteroscopy were not contributive for the stage T in the immediate RNU group and CT group (8 of 12), b Two biopsies were not contributive, and there were three missing data on the RNU group. Two biopsies were not contributive to the CT group. Tumor stage. Preoperative biopsies in CT group showed seven tumors classified Ta-T1-T2 (63.6%) and four as not contributive (36.4%). In the immediate RNU group, four tumors were classified Ta-T1-T2 (10%), and eight biopsies were not contributive. Biopsies were considered as not contributive when they were too crushed for the pathologist to see the invasion of the layers. Final pathology after RNU, including immediate RNU (group 1) and after attempt of conservative treatment (group 2), reported 23 tumors classified as pta-t1-t2 in the immediate RNU group (61.5%) and 6 in the delayed RNU group (54.5%). Seventeen (38.5%) and five (45.5%) tumors were classified as pt3-t4 for the immediate RNU and delayed RNU group, There was no significant difference between the two groups ( p = 0.866). Lymph node involvement. There were 10 patients with positive lymph nodes in the immediate RNU group (25%) compared with 0 in the CT group. Immediate RNU and CT groups had one and two missing data, Biopsy interpretation. Eight of 12 biopsies performed in the immediate RNU were noncontributive for the stage as well as 4 of 11 patients in the delayed RNU group. In total, 12 of 23 biopsies (52.2%) did not obtain a tumor stage. For histological grade, 2 of 12 and 2 of 11 biopsies were noncontributive for group 1 and group 2, In total, 4 of 23 biopsies (17.4%) did not obtain the tumor grade with the biopsy. Discussion The data analyzed did not find a significant difference between the two groups on the primary endpoint of the study (number of tumors pta-t1-t2 and pt3-t4). Patients with delayed RNU after intent of CT had no significantly worse pathological results than patients with an immediate RNU at the diagnosis. Several studies have compared the data after CT with those obtained after nephroureterectomy. 6,7 However, no randomized control study comparing the two approaches has been published. Data are, therefore, coming mainly from case series and retrospective cohorts. Gadzinski et al. compared the CT with the immediate RNU. 7 This retrospective study evaluated the monitoring of patients by analyzing postoperative complications, 5-year cancer-specific survival, overall survival, occurrence of metastases at 5 years, and recurrence of urothelial tumors. According to this study, when CT is technically feasible, it is the best practice for low-grade UTUC with a rate of postoperative complications that are threefold lower (9.3% in the CT group compared with 29% in the RNU group) and overall survival, specific 5-year cancer survival, cancer recurrence, and occurrence of metastasis at 5 years that were not significantly different for low-grade tumors. However, this is at the price of higher re-intervention rate. The same group compared the immediate RNU with delayed RNU after CT. 6 They found that the failure of CT did not affect the overall survival after RNU. The 5-year survival of patients with a delayed RNU after CT was 64% compared with 59% for patients who have had immediate RNU. Cancer-specific survival was 91% compared with 80% for patients with delayed and immediate RNU, The 5-year metastasis-free survival was 77% in patients who had delayed RNU against 73% of patients who had immediate RNU. None of these data showed significant differences between the two groups ( p > 0.05).

4 HADDAD ET AL. Other authors make the comment that the results of such studies must be interpreted cautiously, because cohorts are small and the groups are probably not comparable as the CT group is assumed to contain more low-grade tumors as compared with high-grade tumors in the immediate RNU group. 4 This is a selection bias, as tumor characteristics are more favorable in the CT group than in the immediate RNU. Our study was potentially biased by the fact that patients were often addressed in our reference center for complex cases, hence the importance of initial radiological tumor size in the CT group. Moreover, 25.5% of patients had biopsies of a high grade with 45.1% of patients presenting UTUC of more than 1 cm. Despite these characteristics, no difference in the two groups was found (Table 1). Even given the methodological limitations of the study, these data raise the question of re-evaluating the criteria of CT for UTUC. Furthermore, patients with imperative indication for CT were excluded from the analysis, as the limits of the CT in these patients are often pushed over the guidelines to avoid terminal kidney failure after treatment of the UTUC. The main problem of the CT is that it is difficult to obtain an initial pathology to follow exactly the appropriate management. At present, there is no effective biopsy forceps that could pass in a working channel of the flexible ureteroscopes. Biopsies taken during the URS are often uninterpretable, because the sample is too small or crushed. 8 The development of the ureteroscopes advance toward decreasing the diameter of the endoscopes to obtain an easier upper urinary tract access while trying to maintain or even enlarge the working channel to facilitate irrigation and instrument manipulation. An expansion of the operating channel could allow the development of an effective and easy-to-use biopsy forceps that would contribute to improving the accuracy of grade and maybe the T stage of the tumor with a single biopsy specimen, thus helping in the management of UTUC. Wang et al. conducted a study on the association between endoscopic biopsies and pathologic results after RNU. 9 Of 184 patients who underwent an endoscopic exploration before the RNU, 15% of biopsies were noncontributive and 11% had no histological grade. Of patients with low-grade tumors, 96% had a more aggressive histological grade after the RNU. Grade 2 on tumor biopsy was proved to be 40% of time a grade 3 on the final pathology report. Smith et al. also analyzed the correlation between the histopathology of CT biopsies and the pathology RNU data. 10 They had 43% of patients classified as low grade and/or T1 who have been reclassified as high grade and/or invasive. The URS alone has a little role in the classification of urothelial tumors. 1 The literature agrees that biopsies are interpretable in a third of cases, and in 40% 96% of cases, biopsies underestimate the pathology of the tumor. In addition, URS with biopsies are operator dependent, and it is now necessary to obtain better tools to produce simple and reproducible biopsies of quality to be sure of CT improvement. 11 Moreover, the flexible ureteroscopes do not have the same sensitivity to diagnose an UTUC. 12 The fiber optic ureteroscopes have often insufficient resolution to recognize all the small tumors embedded in the cavities. Among digital ureteroscopes, the model from Olympus provides the NBI vision, which allows distinguishing between inflammation and tumors in patients with an inflammatory urothelium that is often present due to their previous UTUC treatments and Double-J ureteral stent. 13 Since this year, Storz offers the Storz Professionnal Image Enhancement System; however, up to now, no study has evaluated this technology for the UTUC. Furthermore, stopping a CT in favor of an RNU is sometimes based on subjective factors without always a clear indication. This problem is accentuated by this operatordependent procedure and the heterogeneity of the flexible ureteroscopes available in different centers that decide to do CT for a UTUC event with only fiber optic ureteroscopes. Moreover, there is a problem of radiological classification. The CT scan performed in these patients has insufficient sensitivity for an appropriate TNM classification of small tumors. 3 Radiological characteristics are usually urothelium wall thickness or a filling defect in the excretory phase. These radiological signs are often missing for flat tumors such as carcinoma in situ, and the radiological TNM classification is poorly evaluated. This study has some limitations. First, it is a retrospective study design and selection bias is inevitable between the groups. There is also a small number of patients, and no longterm follow-up is available. However, the aim of the study was principally to compare whether there is a difference between the ptnm stage of delayed RNU due to CT and immediate RNU. Another limitation was the decision to cease the CT is dependent on the surgeon s perspective, but here, only one expert surgeon has taken all the commitment to abstain from the CT and opt for a nephroureterectomy. Conclusion Within the limits of this retrospective study, the definitive pathologic examination showed no significant difference in the case of immediate RNU or after attempt of a conservative ureteroscopic management. Author Disclosure Statement O.T. is a consultant for Coloplast, Rocamed, Olympus, and AMS. All other authors have nothing to disclose. References 1. Roupret M, Babjuk M, Comperat E, Zigeuner R, Sylvester R, Burger M, et al. European guidelines on upper tract urothelial carcinomas: 2013 update. Eur Urol 2013;63:1059 1071. 2. Roupret M. [Conservative management of urothelial carcinomas of the upper urinary tract]. Prog Urol 2012; 22:861 866. 3. Puech P, Rouprêt M, Renard-Penna R, Lemaître L, Colin P. Imaging of urothelial carcinomas of the upper tract: State of the art for the yearly scientific report of the National French Association of Urology. Prog Urol 2014;24:987 999. 4. Cutress ML, Stewart GD, Tudor EC, Egong EA, Wells- Cole S, Phipps S, et al. Endoscopic versus laparoscopic management of noninvasive upper tract urothelial carcinoma: 20-year single center experience. J Urol 2013;189: 2054 2060. 5. Cutress ML, Stewart GD, Zakikhani P, Phipps S, Thomas BG, Tolley DA. Ureteroscopic and percutaneous management of upper tract urothelial carcinoma (UTUC): Systematic review. BJU Int 2012;110:614 628.

IMMEDIATE RNU OR CT FOR UTUC: PATHOLOGY REPORTS 5 6. Gadzinski AJ, Roberts WW, Faerber GJ, Wolf JS, Jr. Longterm outcomes of immediate versus delayed nephroureterectomy for upper tract urothelial carcinoma. J Endourol 2012;26:566 573. 7. Gadzinski AJ, Roberts WW, Faerber GJ, Wolf JS, Jr. Longterm outcomes of nephroureterectomy versus endoscopic management for upper tract urothelial carcinoma. J Urol 2010;183:2148 2153. 8. Al-Qahtani SM, Legraverend D, Gil-Diez de Medina S, Sibony M, Traxer O. Can we improve the biopsy quality of upper urinary tract urothelial tumors? Single-center preliminary results of a new biopsy forceps. Urol Int 2014;93:34 37. 9. Wang JK, Tollefson MK, Krambeck AE, Trost LW, Thompson RH. High rate of pathologic upgrading at nephroureterectomy for upper tract urothelial carcinoma. Urology 2012;79:615 619. 10. Smith AK, Stephenson AJ, Lane BR, Larson BT, Thomas AA, Gong MC, et al. Inadequacy of biopsy for diagnosis of upper tract urothelial carcinoma: Implications for conservative management. Urology 2011;78:82 86. 11. Taneja SS. Re: Inadequacy of biopsy for diagnosis of upper tract urothelial carcinoma: Implications for conservative management. J Urol 2012;187:1583 1584. 12. Audenet F, Traxer O, Yates DR, Cussenot O, Rouprêt M. Potential role of photodynamic techniques combined with new generation flexible ureterorenoscopes and molecular markers for the management of urothelial carcinoma of the upper urinary tract. BJU Int 2012;109:608 613; discussion 613 614. 13. Bus MT, de Bruin DM, Faber DJ, et al. Optical diagnostics for upper urinary tract urothelial cancer: Technology, thresholds, and clinical applications. J Endourol 2015;29: 113 123. Address correspondence to: Olivier Traxer, MD, PhD Department of Urology Tenon Hospital 4 rue de la CHINE Paris 75020 France E-mail: olivier.traxer@tnn.aphp.fr Abbreviations Used CT ¼ conservative treatment CT-Scan ¼ computed-tomography scanner urography NBI ¼ narrow band imaging RNU ¼ radical nephroureterectomy URS ¼ ureteroscopy UTUC ¼ upper urinary tract urothelial carcinoma