Nephroureterectomy for upper tract urothelial carcinoma recurrence in bladder cancer patients treated with radical cystectomy: a multicentric propensity score matched analysis on predictors, practice patterns and survival outcomes Authors: Valerio Santarelli, Vincenzo Asero, Roberta Corvino, Ettore De Berardinis, Alessandro Sciarra, Giuseppe Basile, Yasmin Abu-Ghanem, Rajesh Nair, Muhamad Shamin Khan, Ramesh Thurairaja, Wojciech Krajewski, Jan Laszkiewicz, Simone Albisinni, Keiichiro Mori, Felix Guerrero-Ramos, David D’Andrea, Andrea Gallioli, Ekaterina Laukhtina, Elisabeth Grobet-jeandin, Maxence Reynard, Andrea Mari, Laura S Mertens, Francesco Soria, Jeremy Teoh, Francesco Claps, Antonio Amodeo, Aleksander Ślusarczyk, Roberto Contieri, Sisto Perdonà, José Daniel Subiela, Karl H Tully, Renate Pichler, Alessandro Antonelli, Riccardo Bertolo, Francesco Ditonno, Rodolfo Hurle, Felice Crocetto, Roman Mayr, Angelo Porreca, Filippo Marino Published: 2026 ABSTRACT AbstractAim(I) To determine clinicopathological determinants of metachronous Upper Tract Urothelial Carcinoma (UTUC) requiring Radical Nephroureterectomy (RNU) after Radical Cystectomy (RC). (II) To evaluate long-term survival of patients who underwent RC + RNU compared with matched RC only controls.MethodsPatients undergoing RNU for metachronous UTUC were extracted from a multi-institutional RC database. A 1:2 Propensity Score Match (PSM) was performed based on age, gender, BMI, CCI, Smoking Status, and cT stage between RC only and RC + RNU patients. Simon-Makuch plots, landmark analyses and Multivariable Cox regressions were adopted to compare survival outcomes.ResultsOf 1804 RC patients, 85 (4.7%) underwent subsequent RNU. At multivariate regression, younger age, smoking history, BCG exposure, NMIBC, CIS and positive ureteric margins were identified as positive predictors of upper tract recurrence. After PSM, 81 RC + RNU patients were matched to 157 RC only controls. Median time from RC to RNU was 29 (18–47) months, with the majority of UTUC diagnosed at a muscle-invasive stage (70.9%). Simon-Makuch curves demonstrated worse Cancer Specific Survival (CSS) of the RC + RNU cohort (HR: 6.41, 95% CI 3.16–13.04). At landmark analyses, RNU was consistently associated with an increased mortality risk from the 24th month onward. Multivariable Cox regression identified RNU and pN + as the only significant predictors of worse CSS (respectively, HR: 6.55, 95% CI 2.93–14.64 and HR: 8.45, 95% CI 3.24–21.99).ConclusionsThe worse survival outcomes and high rates of locally advanced disease found at RNU underscore the need for standardized, risk-stratified, long-term follow-up of the remnant urinary tract following RC.Supplementary InformationThe online version contains supplementary material available at 10.1007/s00345-026-06520-z.Keywords: Radical cystectomy, Radical Nephroureterectomy, Upper tract urothelial carcinoma, Urothelial recurrence, Bladder cancer Introduction Upper Urinary Tract (UUT) recurrence in the remnant urothelium represents a persistent, life-long risk following Radical Cystectomy (RC) [1]. Accordingly, latest European Association of Urology (EAU) guidelines suggest stringent UUT follow-up schedules for BC patients, with 4–10% of RC patients developing an Upper Urinary Tract Urothelial Carcinoma (UTUC). Radical nephroureterectomy (RNU) is the standard of care for high-risk UTUC [2, 3]. However, RNU following RC represents an additional burden in the already demanding treatment pathway for urothelial disease. Notably, only few risk factors for UUT recurrence have been previously identified [1, 4]. Among these, bladder carcinoma in situ (CIS), multifocal disease, non-muscle invasive bladder cancer (NMIBC), and involvement of the distal ureter or prostatic urethra are recognized as the most prominent predictors for secondary UTUC [4]. Beyond these pathology-based features, limited evidence exists regarding the influence of surgical approach during RC, type of urinary diversion, and patient-related risk factors. Despite the well-accepted recurrence risk, an effective standardized and risk-stratified follow-up regimen after RC for BC is still lacking [5]. Particularly, no consensus was reached regarding the timing of urinary cytology or the role of cross-sectional imaging in detecting late recurrences (i.e., beyond 5 years), which frequently involve the UUT [6, 7]. Besides its direct impact on cancer-specific survival outcomes, UTUC recurrence and its subsequent treatment further compromise overall health status primarily by worsening renal function and ultimately increasing cardiovascular events and all-cause deaths [2, 5, 6]. To date, real-world evidence on non-pathological determinants of UTUC recurrence after RC remains scarce. Moreover, limited data are available regarding the impact of RNU on survival outcomes of RC patients. To address this gap, we specifically selected patients who underwent RC for BC and subsequently RNU for an UUT recurrence to evaluate clinical and pathological determinants of UUT recurrence. After Propensity Score Matching on selected preoperative clinical characteristics, long-term survival outcomes were finally evaluated. Materials and methods Study population Data from 23 tertiary referral centers in Europe, North America and Asia on patients who underwent RC with curative intent from 2005 to 2024 was collected from a multi-institutional, prospectively maintained database of the Young Academic Urologists (YAU)—Urothelial Carcinoma Working Group. Subsequently, the same dataset was retrospectively reviewed to extract patients who developed a UUT recurrence requiring RNU. Before study initiation, data-transfer agreement and Institutional Review Boards (IRB) approval from each of the included centers were obtained. All patient-related information was anonymized prior to data sharing. Exclusion criteria were previous Radiation Therapy or Chemotherapy performed for other purposes other than a Neoadjuvant/Adjuvant setting, RC performed for other indications rather than curative intent for a cN0M0 BC, estimated life expectancy < 5 years and concomitant or previous RNU. Statistical analysis Statistical analyses were conducted using the R software environment for statistical computing and graphics (R version 4.2.2, R Foundation for Statistical Computing, Vienna, Austria). Baseline pertinent study cohort characteristics were summarized with descriptive statistics. Multivariable binary regression model was constructed to determine clinical and pathological determinants of UUT recurrence following RC. To balance baseline characteristics between the RC alone and the RC + RNU cohorts, a 1:2 Propensity Score Match (PSM) analysis using the nearest neighbour method with a calliper size of 0.2 was performed based on age, gender, BMI, CCI, Smoking Status, and Clinical T Stage. Relevant study information and outcomes of the matched cohorts were subsequently reported stratified according to RNU status (no RNU vs. RNU). To account for the time-dependent impact of performing RNU on long-term survival status, univariate survival analysis was performed using the Simon-Makuch variant of Kaplan–Meier plots and landmark analyses. Survival for all time-to-event analyses was defined as the time between radical cystectomy and death or last patient contact. Cox proportional hazards regression was used to estimate Hazard Ratio (HR) with 95% Confidence Interval (CI). Finally, multivariable Cox regression models with RNU-status as a time-dependent covariate were constructed for both Cancer Specific Survival (CSS) (defined as any death from urothelial cancer) and Overall Survival (OS). Results Descriptive statistics of the study population In total, 1804 patients were identified (supplementary Table 1). Of these, 85 developed UTUC requiring RNU. Patients who underwent RNU after RC for a metachronous UTUC were more likely to undergo RC for a NMIBC diagnosis (67.1% vs. 31.5%, p < 0.001) and with a concomitant Carcinoma in Situ (CIS) (43.5% vs. 17.9%, p < 0.001). The majority of RC + RNU patients were previously exposed to intravesical BCG (n = 44, 51.8%), compared to only n = 317, 18.4% in the RC only cohort (p < 0.001). A pN + status at RC was more frequent in the RC alone group (n = 331, 18.1% vs. n = 2, 2.4%). After 1:2 PSM, 81 RC + RNU patients were successfully matched with 157 patients in the RC alone group, meaning that n = 4 RC + RNU patients did not find a suitable match, and n = 5 were matched with a single RC alone patient. Characteristics of the matched cohort are reported in Table 1. Despite the similar NMIBC rates after matching on preoperative cT stage, patients who developed an UUT recurrence treated by RNU remained more frequently exposed to BCG (53.8% vs. 28.7%, p < 0.001). At final pathological analysis, RC alone patients were more likely to achieve a pT0 stage (no evidence of residual tumour, 26.1% vs. 6.2%), despite the similar NAC rates (23.6% vs. 17.3%, p = 0.321). However, MIBC rates at final RC specimens were similar between the two groups (46.5% vs. 43.2%, p = 0.713). As expected, patients who developed a subsequent RNU were more likely to present a positive ureteric margin at the final RC histopathological report (0.6% vs. 8.6%, p = 0.004). After a significantly longer follow-up (61.0 months [IQR 31.0–89.0] vs. 26.0 [6.0–60.0], p < 0.001), only 50%, n = 40 of RC + RNU were still alive, vs. 79.6%, n = 125 RC only patients, p < 0.001. Table 1General characteristics of the study population after propensity score matchingVariableOverall N = 238RC Only (N = 157)RC + RNU (N = 81)p-valueAge, median (IQR)68.0 (60.0–73.0)67.0 (58.0–74.0)68.8 (60.8–72.9)0.642Gender, n (%)0.543Male180 (75.6%)116 (73.9%)64 (79.0%)Female58 (24.4%)41 (26.1%)17 (21.0%)BMI, median (IQR)26.3 (23.5–29.0)26.3 (23.3–28.4)26.4 (24.0–30.0)0.246Charlson Comorbidity Index, median (IQR)5.0 (3.0–6.0)5.0 (3.0–6.0)5.0 (3.0–6.0)0.398Smoking Status, n (%)0.901No59 (24.8%)40 (25.5%)19 (23.5%)Active/Former179 (75.2%)117 (74.5%)62 (76.5%)BCG exposure, n (%) < 0.001No150 (62.9%)112 (71.3%)38 (46.3%)Yes88 (37.1%)45 (28.7%)43 (53.8%)MIBC at TURBT, n (%)0.212No138 (58.1%)86 (54.8%)52 (64.6%)Yes100 (41.9%)71 (45.2%)29 (35.4%)Clinical T stage (cT), n (%)0.893cT00 (0.0%)0 (0.0%)0 (0.0%)cT198 (41.2%)64 (40.8%)34 (42.0%)cT295 (39.9%)64 (40.8%)31 (38.3%)cT36 (2.5%)4 (2.5%)2 (2.5%)cT47 (2.9%)3 (1.9%)4 (4.9%)cTis16 (6.7%)11 (7.0%)5 (6.2%)cTa16 (6.7%)11 (7.0%)5 (6.2%)CIS at TURBT, n (%) < 0.001No173 (72.8%)128 (81.5%)45 (55.1%)Yes65 (27.2%)29 (18.5%)36 (44.9%)Clinical N stage (cN), n (%)0.936cN0234 (98.3%)154 (98.1%)80 (98.7%)cN + 4 (1.7%)3 (1.9%)1 (1.3%)Neoadjuvant Chemotherapy, n (%)0.321No187 (78.6%)120 (76.4%)67 (82.7%)Yes51 (21.4%)37 (23.6%)14 (17.3%)Urinary diversion, n (%)0.064Ileal conduit131 (55.0%)81 (51.6%)50 (61.7%)Neobladder70 (29.4%)49 (31.2%)21 (25.9%)Cutaneous ureterostomy35 (14.7%)27 (17.2%)8 (9.9%)Ileal Pouch2 (0.8%)0 (0.0%)2 (2.5%)Pathologic T stage, n (%) < 0.001pT046 (19.3%)41 (26.1%)5 (6.2%)pT140 (16.8%)18 (11.5%)22 (27.2%)pT242 (17.6%)21 (13.4%)21 (25.9%)pT349 (20.6%)41 (26.1%)8 (9.9%)pT417 (7.1%)11 (7.0%)6 (7.4%)pTis34 (14.3%)19 (12.1%)15 (18.5%)pTa10 (4.2%)6 (3.8%)4 (4.9%)MIBC at RC, n (%)0.713NMIBC130 (54.6%)84 (53.5%)46 (56.8%)MIBC108 (45.4%)73 (46.5%)35 (43.2%)pN stage at RC, n (%)0.010pN0214 (89.9%)135 (86.0%)79 (97.5%)pN + 24 (10.1%)22 (14.0%)2 (2.5%)CIS at RC0.385No147 (61.8%)101 (64.3%)46 (56.8%)Yes91 (38.2%)56 (35.7%)35 (43.2%)Variant Histology at RC, n (%)0.767No204 (85.6%)134 (85.4%)70 (86.1%)Yes34 (14.4%)23 (14.6%)11 (13.9%)Ureteral Margin, n (%)0.004Negative230 (96.6%)156 (99.4%)74 (91.4%)Positive8 (3.4%)1 (0.6%)7 (8.6%)Time to RNU, median (IQR)29.0 (18.0–47.0)NA (NA-NA)29.0 (18.0–47.0)Neoadjuvant CHT before RNU, n (%)No76 (93.8%)0 (NA%)76 (93.8%)Yes5 (6.2%)0 (NA%)5 (6.2%)Clinical T stage at RNU, n (%)cT00 (0.0%)0 (NA%)0 (0.0%)cT116 (20.8%)0 (NA%)16 (20.8%)cT232 (41.6%)0 (NA%)32 (41.6%)cT316 (20.8%)0 (NA%)16 (20.8%)cT44 (5.2%)0 (NA%)4 (5.2%)cTis2 (2.6%)0 (NA%)2 (2.6%)cTa3 (3.9%)0 (NA%)3 (3.9%)cTx4 (5.2%)0 (NA%)4 (5.2%)Clinical N stage RNU, n (%)cN069 (86.3%)0 (NA%)69 (86.3%)cN + 11 (13.8%)0 (NA%)11 (13.8%)Tumor location RNU, n (%)Ureter28 (34.6%)0 (NA%)28 (34.6%)Pelvicalyceal system37 (45.7%)0 (NA%)37 (45.7%)Both16 (19.8%)0 (NA%)16 (19.8%)Pathologic T stage RNU, n (%)pT05 (6.3%)0 (NA%)5 (6.3%)pT118 (22.8%)0 (NA%)18 (22.8%)pT221 (26.6%)0 (NA%)21 (26.6%)pT328 (35.4%)0 (NA%)28 (35.4%)pT47 (8.9%)0 (NA%)7 (8.9%)Pathologic N stage RNU, n (%)pN032 (40.0%)0 (NA%)32 (40.0%)pN14 (5.0%)0 (NA%)4 (5.0%)pN210 (12.5%)0 (NA%)10 (12.5%)pN32 (2.5%)0 (NA%)2 (2.5%)pNx32 (40.0%)0 (NA%)32 (40.0%)VH RNU, n (%)No72 (88.9%)0 (NA%)72 (88.9%)Yes9 (11.1%)0 (NA%)9 (11.1%)LVI RNU, n (%)No55 (67.9%)0 (NA%)55 (67.9%)Yes26 (32.1%)0 (NA%)26 (32.1%)CIS RNU, n (%)No54 (66.7%)0 (NA%)54 (66.7%)Yes27 (33.3%)0 (NA%)27 (33.3%)Adjuvant CHT RNU, n (%)No69 (85.2%)0 (NA%)69 (85.2%)Yes12 (14.8%)0 (NA%)12 (14.8%)Last FU, months, median (IQR)34.5 (12.0–63.0)26.0 (6.0–60.0)61.0 (31.0–89.0) < 0.001Follow-Up Status, n (%) < 0.001Alive165 (69.6%)125 (79.6%)40 (50.0%)Dead of disease48 (20.3%)23 (14.6%)25 (31.3%)Dead of other causes24 (10.1%)9 (5.7%)15 (18.8%)RC radical cystectomy, RNU radical nephroureterectomy, IQR interquartile range, BMI body mass index, BCG bacillus calmette-guérin, MIBC muscle-invasive bladder cancer, TURBT transurethral resection of bladder tumor, CIS carcinoma in situ, NMIBC non-muscle-invasive bladder cancer, CHT chemotherapy, VH variant histology, LVI lymphovascular invasion, FU follow-up, NA not applicable General characteristics of the study population after propensity score matching RC radical cystectomy, RNU radical nephroureterectomy, IQR interquartile range, BMI body mass index, BCG bacillus calmette-guérin, MIBC muscle-invasive bladder cancer, TURBT transurethral resection of bladder tumor, CIS carcinoma in situ, NMIBC non-muscle-invasive bladder cancer, CHT chemotherapy, VH variant histology, LVI lymphovascular invasion, FU follow-up, NA not applicable Risk factors for UUT recurrence requiring RNU in the unmatched cohort Results of multivariable binary logistic regression analysis exploring predictors of UUT recurrence requiring RNU among patients submitted to RC for BC are reported in Table 2. Older age reduced the risk of a subsequent RNU (Adjusted Odds Ratio (aOR): 0.940, 95% CI 0.921–0.959, p < 0.001). History of smoking (either former or active) and BCG exposure were significantly associated with an increasing risk of RNU (aOR: 1.81, 95% CI 1.012–3.277, p = 0.046 for smoking status and aOR: 2.12, 95% CI 1.21–5.77, p = 0.021 for BCG exposure). Regarding histological determinants, NMIBC, concomitant CIS, and positive ureteric margin were all significant independent predictors of UUT recurrence (respectively aOR: 4.67, 95% CI 2.68–8.13, p < 0.001; aOR: 3.49, 95% CI 2.11–5.79, p < 0.001 and aOR 10.71, 95% CI 2.97–38.60, p < 0.001). Finally, the Urinary Diversion of choice, gender, preoperative NAC and RC approach had no significant impact on the risk of UTUC development in RC patients. Table 2Multivariable binary regression exploring risk factors for UTUC development in the unmatched cohortVariableaOR95% CIp valueAge0.940.92–0.96 < 0.001Former/Active Smoker1.821.01–3.230.046Gender (female)1.010.59–2.030.782NAC0.870.45–1.690.679BCG exposure2.121.21–5.770.021RC approach1.020.41–2.140.863NMIBC4.672.68–8.13 < 0.001Variant Histology1.280.61–2.670.506CIS3.492.11–5.79 < 0.001Ureteral Margin10.712.97–38.60 < 0.001Urinary Diversion (UCS vs. IC)0.930.44–1.960.859Urinary Diversion (NB vs. IC)0.430.161–1.120.084aOR adjusted odds ratio, CI confidence interval, NAC neoadjuvant chemotherapy, BCG bacillus calmette-guérin, RC radical cystectomy, NMIBC non-muscle-invasive bladder cancer, CIS carcinoma in situ, UCS cutaneous ureterostomy, IC ileal conduit, NB neobladder Multivariable binary regression exploring risk factors for UTUC development in the unmatched cohort aOR adjusted odds ratio, CI confidence interval, NAC neoadjuvant chemotherapy, BCG bacillus calmette-guérin, RC radical cystectomy, NMIBC non-muscle-invasive bladder cancer, CIS carcinoma in situ, UCS cutaneous ureterostomy, IC ileal conduit, NB neobladder Survival outcomes by treatment group Overall survival Simon-Makuch survival analyses were carried out to compare OS distributions between patients who underwent RC alone vs. those who subsequently required RNU both in the matched and unmatched cohorts (supplementary Fig. 1a and Fig. 1a). RNU was treated as a time-dependent covariate, allowing patients to contribute to the no-RNU group until the time of RNU and subsequently to the RNU group. In the matched cohort, Restricted Mean Survival Time (RMST) was 133.9 months (SE = 5.79) for the RC alone group and 53.3 months (SE = 8.44) for the RC + RNU group. Results of Cox proportional hazards regression with RNU as a time-dependent covariate showed a significantly higher risk of all-cause death in patients requiring RNU (HR = 6.17, 95% CI 3.48–10.94, p < 0.001). A Simon-Makuch survival analysis comparing overall survival (OS) distributions between the RC alone group and the RC + RNU group in the matched cohort; B Landmark analysis for OS according to RNU status (matched cohort) Landmark analyses were performed at 6, 12, 24, 36, 48 and 60 months both in the unmatched (supplementary Fig. 1b) and matched (Fig. 1b) cohorts, to assess the association between RNU and OS. Classification of the patients varied at each landmark based on RNU status (no RNU vs. RNU). In both cohorts, RNU was not significantly associated with worse OS at early landmarks (6 and 12 months). However, from 24 months onwards, RNU was consistently associated with a significantly increased overall mortality risk, reaching an HR of 3.99 (95% CI 1.49–10.71, p = 0.006) at 60 months. Cancer specific survival Supplementary Fig. 2a and Fig. 2a depict results of Simon-Makuch survival analysis comparing CSS distributions in the unmatched and matched cohorts, respectively. After 1:2 PSM, RMST was 143.4 months (SE = 5.23) for the RC alone group and 75.9 months (SE = 11.03) for the RC + RNU group after PSM. Results of Cox proportional hazards regression with RNU as a time-dependent covariate showed a significantly higher risk of cancer-related death in patients requiring RNU (HR = 6.41, 95% CI 3.16–13.04, p < 0.001). A Simon-Makuch survival analysis comparing Cancer Specific Survival (CSS) distributions between the RC alone group and the RC + RNU group in the matched cohort; B Landmark analysis for CSS according to RNU status (matched cohort) The temporal effect of RNU on CSS was assessed with Landmark analyses carried out at 6, 12, 24, 36, 48 and 60 months before and after PSM (Supplementary Fig. 2b and Fig. 2b). Similarly to OS, RNU was not significantly associated with worse CSS at early landmarks (6 and 12 months). Subsequently, the impact of RNU on CSS increased, reaching statistical significance at 24 months (HR = 2.62, 95% CI 1.09–6.30, p = 0.031), 36 months (HR: 2.54, 95% CI 0.91–7.06, p = 0.075), 48 months (HR = 3.40, 95% CI 1.16–9.99, p = 0.026), and 60 months (HR = 4.38, 95% CI 1.12–17.12, p = 0.034). Multivariable predictors of survival outcomes Cox proportional hazard regression analyses with RNU as a time-dependent covariate were performed to identify predictors of OS and CSS in the matched cohort (Table 3). Undergoing RNU after RC was independently associated with a significantly increased risk of both all-cause and cancer-related death (HR 5.44, 95% CI 2.89–10.24, p < 0.001 for OS and HR: 6.55, 95% CI 2.93–14.64, p < 0.001 for CSS). LN involvement at RC was also a strong independent predictor of mortality in both models (HR: 6.25, 95% CI 2.78–14.04, p < 0.001 for OS and HR: 8.45, 95% CI 3.24–21.99, p < 0.001 for CSS). Older age was significantly associated with worst OS (HR 1.05 per year, 95% CI 1.02–1.08, p < 0.001) but not with CSS (HR 1.03, 95% CI 0.99–1.06, p = 0.143). Gender, smoking status, MIBC at TURBT, CIS in the bladder, histology at TURBT, pathological T stage at RC, and type of urinary diversion were not significantly associated with either OS or CSS (p > 0.05). Table 3Multivariable cox regression analysis with RNU as a time-dependent covariate exploring independent predictors of overall survival (OS) and cancer specific survival (CSS)OSCSSVariableHR95% CIp-valueHR95% CIp-valueRNU5.442.89–10.24 < 0.0016.552.93–14.64 < 0.001Age1.051.02–1.08 < 0.0011.030.99–1.060.143Gender (female)0.730.38–1.410.3500.440.18–1.070.069Smoking Status0.870.46–1.640.6570.680.32–1.430.306MIBC at TURBT1.000.57–1.750.9991.140.58–2.230.709CIS in the bladder0.940.54–1.640.8340.870.43–1.780.709Histology at RC1.360.70–2.650.3631.090.47–2.530.839pT RC1.260.71–2.230.4251.430.70–2.890.325pN RC (positive)6.252.78–14.04 < 0.0018.453.24–21.99 < 0.001Urinary diversion1.200.87–1.640.2620.970.65–1.450.894OS overall survival, CSS cancer-specific survival, HR hazard ratio, CI confidence interval, RNU radical nephroureterectomy, MIBC muscle-invasive bladder cancer, TURBT transurethral resection of bladder tumor, CIS carcinoma in situ, RC radical cystectomy Multivariable cox regression analysis with RNU as a time-dependent covariate exploring independent predictors of overall survival (OS) and cancer specific survival (CSS) OS overall survival, CSS cancer-specific survival, HR hazard ratio, CI confidence interval, RNU radical nephroureterectomy, MIBC muscle-invasive bladder cancer, TURBT transurethral resection of bladder tumor, CIS carcinoma in situ, RC radical cystectomy Finally, a subgroup analysis was performed to identify UTUC-related predictors of OS and CSS among patients who underwent RNU after RC (Supplementary Table 2). Positive lymph nodes and Variant histology were the only independent predictors of survival outcomes (respectively, HR: 2.19, 95% CI 1.00–6.72, p = 0.049 and HR: 4.12, 95% CI 1.56–10.89, p = 0.004 for OS; HR: 3.96, 95% CI 1.54–10.14, p = 0.004 and HR: 4.30, 95% CI 1.29–14.33, p = 0.017 for CSS). Discussion The majority of systemic BC recurrences typically develop within the first two years following RC, while urothelial recurrence within the remnant UUT has been reported to develop throughout a patient’s lifetime [8–12]. In our multicenter cohort of 1804 RC patients, 4.71% developed an UUT recurrence requiring RNU. Considering that RNU is the standard of care for high-risk UTUC, and a rate of 61% of UUT recurrences after RC managed with RNU reported by a contemporary review of post-RC UTUC management, our results are in line with the available literature[13–15]. In light of the prognostic advantage of early diagnosis for UTUC, current EAU guidelines suggest CT scan every six months until the third year, followed by annual imaging thereafter [13]. In the present study, we reported a median time from RC to RNU of 29 months, with an IQR of 18 to 47 months. These findings could suggest the need to extend strict follow-up beyond the 2-year landmark, particularly for those patients with known risk factors [9, 15]. Our results confirm the positive association between CIS, NMIBC, positive ureteral margin and metachronous UTUC, while other clinical determinants emerged, such as younger age at RC and a positive smoking history. Compared to the other well-accepted predictors of UUT recurrence, the association between a positive history of BCG exposure and development of metachronous UTUC is reported in fewer cohorts [7, 16–18]. The higher rates of BCG exposures in the RC + RNU cohort suggests a greater proportion of BCG-unresponsive cases, potentially reflecting more aggressive tumor biology [19, 20]. A pN + status at RC was more frequent in the RC alone group. This pattern is consistent with the poorer prognosis of pN + disease, which may reduce the time window to observe metachronous UUT recurrence. Most UTUCs were found to have adverse pathologic features at the time of RNU, with a predominance of muscle-invasive disease and a substantial proportion of lymph node-positive disease. The high prevalence of adverse pathological characteristics can be partially explained by the fact that only metachronous UTUC managed with RNU were included in the present study. Nonetheless, the predominance of locally advanced disease may reflect inadequate standardization of surveillance protocols, reliance on non-contrast studies in patients with compromised renal function yielding lower diagnostic sensitivity, limited accuracy of cytological examination, and the tendency for UTUC to manifest later in the post-cystectomy period when monitoring intensity has typically been reduced. Both in the matched and unmatched cohorts, prognosis following RNU after RC was notably poor. These findings underscore the aggressive biological behavior of metachronous UTUC following RC for bladder cancer. To date, only few retrospective studies reported poor survival after RNU in patients previously treated with RC [21, 22]. UTUC in post-cystectomy patients is typically diagnosed following symptomatic presentation–namely hematuria, flank pain, or pyelonephritis–which often reflects advanced overlooked disease [14]. Nonetheless, the detection of asymptomatic UTUC is associated with a 30% reduction of mortality compared to symptomatic presentation [23]. Early detection of an UUT recurrence could not only minimize cancer-related mortality, but also reduce all-cause deaths and improve quality of life, by avoiding radical management in favour of kidney-sparing strategies [24]. Taken together, our results suggest that a one-size-fits-all follow-up protocol may be insufficient after RC. Patients carrying multiple risk factors—namely concomitant CIS, history of recurrent NMIBC, positive ureteral margins at RC, younger age, and BCG-failure—represent a high-risk subgroup warranting intensified and prolonged UUT surveillance, potentially with semi-annual CT urography beyond the standard three-year window. Moreover, our results corroborate the reported worse survival outcomes of variant histology (VH) in UTUC patients at RNU [25]. Nonetheless, whether these prognostic implications should translate into prompt adjuvant therapy or intensified follow-up remains unclear, particularly for the already vulnerable population of RC patients. Several limitations of our study must be considered. First, the retrospective nature and relatively small sample size of the RC + RNU cohort could have influenced the results or masked significant associations. Second, although PSM allowed us to obtain similar and comparable study groups, other relevant confounders might not have been evaluated. Third, since the timing of RNU was not uniform and follow-up duration significantly differed between the two groups, survival outcomes might have been affected by time-dependent biases such as lead-time bias and immortal time bias. Lastly, relevant information, such as clinical presentation, time from UTUC diagnosis to surgical treatment and adopted surveillance schedules, were not available. Conclusions In the present multicentric, propensity score-matched analysis, approximately 5% of patients undergoing RC with curative intent for BC were subsequently submitted to RNU for a metachronous UTUC. Moreover, most of the UUT recurrences were diagnosed more than two years after RC. Established risk factors, such as concomitant CIS, NMIBC, and positive ureteral margins were confirmed as significant predictors of UUT recurrence, while younger age, BCG-failure and smoking history emerged as possible additional clinical determinants. Even after adequate matching for baseline characteristics, patients who developed a metachronous UTUC treated with RNU after RC exhibited markedly worse survival outcomes compared to the RC-alone cohort. The high prevalence of locally-advanced disease at RNU, together with the impaired survival outcomes, suggest that current surveillance protocol might be insufficient to detect early stage UTUC, particularly for high-risk cases carrying multiple risk factors. Our findings corroborate the need for standardized, risk-stratified, stringent and long-term follow-up of the remnant urinary tract following RC. In the future, large prospective studies are required to further explore the optimal surveillance strategy of BC survivors. Supplementary Information Below is the link to the electronic supplementary material. Supplementary file1 (DOCX 110 KB) Supplementary file2 (DOCX 105 KB) Supplementary file3 (DOCX 20 KB) Supplementary file4 (DOCX 15 KB) Author contributions V.S. conceived and designed the study, performed statistical analysis, and wrote the main manuscript text. V.A., R.C., and G.B. contributed to manuscript preparation and data acquisition. E.D.B. and A.S. contributed to data acquisition and manuscript preparation. Y.A-G., R.N., M.S.K., R.T., W.K., J.L., S.A., E.L., E.G-J., M.R., A.M., L.S.M., A.G., R.Co., S.P., J.D.S., K.H.T., R.P., A.An., R.B., F.D., R.H., F.Cr., R.Ma., A.P., F.M., G.S., G.T., B.M.C.R., C.S., L.V., P.Z., M.D.A., P.S., L.A., R.D.G., and G.D.N. contributed to data acquisition. K.M., F.G-R., D.D., F.S., J.T., F.C., A.A., A.Ś., J.D.S., M.Mo., and L.A. also contributed to manuscript preparation. A.Mo., N.P., B.I.C., G.Ma., B.P., and F.D.G. provided supervision. F.D.G. contributed to study conceptualization and design. All authors reviewed and approved the final manuscript. Funding Open access funding provided by Università degli Studi di Roma La Sapienza within the CRUI-CARE Agreement. None. Data availability Additional data is partially available upon request to the corresponding author. Declarations Conflict of interest The authors declare no competing interests. Human ethics and consent to participate This study is a retrospective observational study conducted in compliance with the European General Data Protection Regulation (EU GDPR 2016/679) and applicable local regulations. The study involved only the analysis of fully anonymized data and did not affect patient care. All procedures were conducted in accordance with the ethical principles of the 1964 Helsinki Declaration and its later amendments or comparable ethical standards, and followed local institutional guidelines at each participating center. Before study initiation, data-transfer agreement and Institutional Review Boards (IRB) approval from each of the included centers were obtained. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary file1 (DOCX 110 KB) Supplementary file2 (DOCX 105 KB) Supplementary file3 (DOCX 20 KB) Supplementary file4 (DOCX 15 KB) Data Availability Statement Additional data is partially available upon request to the corresponding author.