Clinical Determinants of Extraurinary Tract Recurrence and Survival after Radical Surgery for pT2 Upper Tract Urothelial Carcinoma

Simple Summary Although upper tract urothelial carcinoma (UTUC) is a relatively rare malignancy in Western countries, recurrence and distant metastasis are common even after definitive surgery. Many prognostic factors have been identified from previous studies, allowing clinicians to better stratify risk to select patients for perioperative systemic therapy; however, the applicability of adjuvant chemotherapy for patients with stage II UTUC after radical surgery remains unclear. In this study, we found that patients with primary tumor location at ureter or renal pelvis plus synchronous ureter had more frequent disease relapse and worse long-term oncological outcomes than other patients. Male sex, older age, history of previous bladder cancer, and positive surgical margins remain important unfavorable prognostic factors for recurrence and survival. Additional treatment and closer surveillance in patients with these negative prognostic factors are warranted despite complete pathological removal of the tumor. Abstract Background: Oncologic outcomes for pT2N0M0 upper tract urothelial carcinoma (UTUC) after nephroureterectomy are not well defined, with most previous studies focused on a heterogeneous population. Therefore, we aimed to investigate the clinical determinants of extraurinary tract recurrence and survival after radical surgery in patients with localized UTUC. Methods: We retrospectively identified 476 patients with pT2N0M0 UTUC who underwent radical nephroureterectomy or ureterectomy between October 2002 and March 2022. To evaluate the prognostic impact, patients were divided into renal pelvic, ureteral, and both-region (renal pelvis plus synchronous ureter) groups based on tumor location. The outcomes included recurrence-free survival (RFS), cancer-specific survival (CSS), and overall survival (OS). Associations were evaluated using multivariable Cox regression analyses for prognostic factors and Kaplan–Meier analyses for survival curves. Results: The renal pelvic, ureteral, and both-region groups consisted of 151 (31.7%), 314 (66.0%), and 11 (2.3%) patients, respectively. Kaplan–Meier analyses comparing the three tumor types showed significant differences in 5-year RFS (83.6% vs. 73.6% vs. 52.5%, p = 0.013), CSS (88.6% vs. 80.7% vs. 51.0%, p = 0.011), and OS (83.4% vs. 70.1% vs. 45.6%, p = 0.002). Multivariable analyses showed that age >60 years, previous bladder cancer history, ureteral involvement (ureteral and both-regional groups), and positive surgical margins were significant negative prognostic factors for the studied outcomes. Conclusions: Patients with pT2 UTUC and presence of ureteral involvement had more frequent disease relapse. Subsequent adjuvant therapy regimens and close follow-up in patients with negative prognostic factors are warranted despite complete pathological removal of the tumor.


Introduction
Urothelial carcinoma is characterized by neoplastic growth of the entire urothelium, including the upper (renal pelvis and ureter) and the lower (bladder and urethra) urinary tract. Although upper tract urothelial carcinoma (UTUC) is a relatively uncommon malignancy in Western countries, making up only 5-10% of all urothelial carcinomas [1,2], it has a more advanced stage and worse differentiation than bladder cancer, as 62% of UTUCs are muscle-invasive at diagnosis compared to 35% of bladder cancers [3].
Many significant prognostic factors have been proposed based on previously published data [4], allowing clinicians to better stratify risk to select patients for subsequent adjuvant management; however, the use of adjuvant chemotherapy for patients with pT2 UTUC after radical surgery remains controversial. The 2022 National Comprehensive Cancer Network guidelines state that "adjuvant chemotherapy should be considered for patients with no platinum-based neoadjuvant treatment administered and pT3-4 or pN+ disease after surgery" [6]. Contrastingly, a phase 3, open-label, randomized controlled trial that enrolled 260 patients with pT2-4 or pN+ UTUC, of whom 74 (28%) had stage pT2 disease, concluded that adjuvant platinum-based chemotherapy after nephroureterectomy significantly improved disease-free survival [7].
Due to the lack of data on the utility of adjuvant therapy and population heterogeneity in previously published studies, prognostic predictors to identify patients with pT2 UTUC who are more likely to have extraurinary tract recurrence and should receive adjuvant chemotherapy and/or radiotherapy remain insufficient [4,8]. Therefore, this retrospective study aimed to evaluate the association between clinical characteristics and RFS, cancer-specific survival (CSS), and overall survival (OS) of the pT2 UTUC population and provide information to guide the postoperative management and prognostication of patients with pT2 UTUC after radical surgery. We hypothesized that the known risk factors for disease recurrence and survival after radical surgery would apply to the localized UTUC population.

Patient Selection
This study was performed with the approval and oversight of the Institutional Review Board (IRB No. 202100779B0). We retrospectively reviewed the medical charts of 476 consecutive patients with pT2 UTUC who were treated with radical nephroureterectomy or ureterectomy between October 2002 and March 2022 at three main branches of Chang Gung Memorial Hospital (Figure 1), which span northern to southern regions of Taiwan with high overall disease coverage [9]. Radical nephroureterectomy with bladder cuff excision is our institution's standard treatment for patients with pT2 UTUC, with segmental ureteral resection performed in patients with distal ureteral tumors, serious renal insufficiency, or a solitary kidney. Patients with neoadjuvant chemotherapy, radiographic metastases, or retroperitoneal lymph node size > 1 cm were excluded. All patients underwent cystoscopy, chest radiography, and computed tomography (CT) urography or magnetic resonance urography (e.g., if any contraindications to CT urography were present) for preoperative risk stratification. In selected patients, diagnostic ureteroscopy, chest CT, and bone scan were used.
with high overall disease coverage [9]. Radical nephroureterectomy with bladder c cision is our institution's standard treatment for patients with pT2 UTUC, with seg ureteral resection performed in patients with distal ureteral tumors, serious renal i ciency, or a solitary kidney. Patients with neoadjuvant chemotherapy, radiograph tastases, or retroperitoneal lymph node size > 1 cm were excluded. All patients und cystoscopy, chest radiography, and computed tomography (CT) urography or ma resonance urography (e.g., if any contraindications to CT urography were prese preoperative risk stratification. In selected patients, diagnostic ureteroscopy, che and bone scan were used.

Pathological Evaluation
All the surgical specimens were examined by urologic pathologists at our insti Tumors were staged according to the 2017 TNM classification by the American Join mittee on Cancer for UTUC. Tumor grading was assessed according to the 2016 Health Organization/International Society of Urological Pathology consensus cla tion. The pathological characteristics collected for predicting prognosis included location, tumor grade, multifocal disease, carcinoma in situ (CIS), lymphovascula sion (LVI), and surgical margin.

Outcome Measures
After surgery, patients were generally seen every three months for the first two every six months from the third through fifth year, and annually thereafter. Foll generally consisted of medical history, physical examination, blood laboratory tes nary cytology, renal ultrasound, and cystoscopic evaluation. Diagnostic imaging o chest radiography and CT urography were used at least annually to detect locore recurrence and distant metastasis. Chest CT and bone scans were performed when cally indicated.

Pathological Evaluation
All the surgical specimens were examined by urologic pathologists at our institution. Tumors were staged according to the 2017 TNM classification by the American Joint Committee on Cancer for UTUC. Tumor grading was assessed according to the 2016 World Health Organization/International Society of Urological Pathology consensus classification. The pathological characteristics collected for predicting prognosis included tumor location, tumor grade, multifocal disease, carcinoma in situ (CIS), lymphovascular invasion (LVI), and surgical margin.

Outcome Measures
After surgery, patients were generally seen every three months for the first two years, every six months from the third through fifth year, and annually thereafter. Follow-up generally consisted of medical history, physical examination, blood laboratory tests, urinary cytology, renal ultrasound, and cystoscopic evaluation. Diagnostic imaging of both chest radiography and CT urography were used at least annually to detect locoregional recurrence and distant metastasis. Chest CT and bone scans were performed when clinically indicated.
The following clinical characteristics that may be associated with the outcomes were collected: sex, age at surgery, contralateral UTUC history (previous/synchronous/metachronous), bladder cancer history (previous/synchronous/metachronous), hydronephrosis grade, American Society of Anesthesiologists (ASA) score, diagnostic ureteroscopy (with/without biopsy), surgical approach (open/laparoscopic/robotic), surgical procedure (nephroureterectomy/ureterectomy), and estimated glomerular filtration rate (eGFR). Hydronephrosis grade was assessed by preoperative imaging, including CT, excretory urography, and renal ultrasonography. Hydronephrosis was reported as grade 0, no caliceal or pelvic dilatation; 1, pelvic dilatation only; 2, mild caliceal dilatation; 3, severe caliceal dilatation; and 4, caliceal dilatation accompanied by renal parenchymal atrophy [10]. eGFR was calculated using the 2021 Chronic Kidney Disease Epidemiology creatinine-based equation [11], which was the most widely used equation and recommended by the National Kidney Foundation and the American Society of Nephrology [12].
To evaluate the impact of clinical features on recurrence and survival, patients were divided into renal pelvic, ureteral, and both-region (renal pelvis plus synchronous ureter) groups based on the location of the muscle-invasive tumor at radical surgery (pT2).
Disease recurrence was defined as locoregional failure or distant metastases. Metachronous UC in the remnant genitourinary tract was not considered in the analysis of recurrence [13,14]. RFS interval was defined as the time between radical surgery and the first extraurinary tract recurrence, CSS interval was defined as the time between radical surgery and death from UTUC, and OS interval was defined as the time between radical surgery and death from any cause. Additionally, patients who died within 30 days of radical surgery or during hospital admission were censored at the time of death for the CSS and OS analysis [15].

Statistical Analysis
Continuous and categorical variables are presented as median values with interquartile ranges (IQR) and proportions, respectively. One-way ANOVA followed by the Tukey-Kramer test for post hoc comparisons [16] and chi-square test were used to compare continuous and categorical variables in the three groups, respectively. Survival curves were analyzed using the Kaplan-Meier method, and differences were determined using the log-rank test. The prognostic factors for RFS, CSS, and OS were estimated using the Cox proportional hazards regression model in the univariate and multivariate analyses. Only those factors with p < 0.05 in univariable analysis were further evaluated in multivariable analysis. All reported p values were two-sided, and statistical significance was set at p < 0.05. All statistical analyses were performed using SPSS version 20 (IBM Corp, Armonk, NY, USA) or Prism version 9 (GraphPad Software, San Diego, CA, USA).

Baseline Characteristics
Patients' clinical and pathological features stratified by tumor location are presented in Table 1. In the studied cohort, the renal pelvic, ureteral, and both-region groups comprised 151 (31.7%), 314 (66.0%), and 11 (2.3%) patients, respectively. The median age was 70.7 years (IQR, 62.4-77.0 years), and the proportion of female patients was 52.9% in the study population.
The proportion of patients with synchronous bladder cancer, metachronous bladder cancer, and extraurinary tract recurrence was significantly higher in the ureteral and bothregion groups than in the renal pelvic group (p = 0.003, 0.004, and 0.025, respectively). Hydronephrosis, ureteroscopic biopsy, and ureterectomy were more commonly performed in the ureteral group than in the renal pelvic and both-region groups (p < 0.001, 0.001, and <0.001, respectively). Patients in the both-region group had a significantly higher proportion of multifocal disease than those in the renal pelvic and ureteral groups (p < 0.001). LVI was more common in the renal pelvic and both-region groups than in the ureteral group (p = 0.013). No statistically significant differences were observed in sex, age, contralateral UTUC history, previous bladder cancer history, ASA score, surgical approach, tumor grade, CIS, positive surgical margin, and eGFR between the groups (all, p > 0.05).

Discussion
Although many possible risk factors associated with recurrence and survival have been proposed [4], prognostic factors for patients with pT2 UTUC after radical surgery remain unclear. This may be due to the rare incidence of the disease and highly heterogeneous population enrolled in previous studies. Therefore, we only included patients with pT2N0M0 UTUC in this series. Although there are no data on the efficacy of postoperative chemotherapy or radiotherapy in terms of recurrence and mortality in patients with pT2N0M0 UTUC after radical surgery, our findings confirm the prognostic significance of several variables associated with disease recurrence and survival. Subsequent adjuvant therapy regimens and close follow-up in patients with poor prognostic factors are warranted despite complete pathological removal of the disease.
Previous studies have reported controversial results regarding the impact of primary tumor location on the outcome of UTUC treatment. Some studies failed to identify a difference in cancer-specific mortality between renal pelvic and ureteral tumors [17,18]. On the contrary, a retrospective study from multiple institutions, similar to our findings, showed a worse CSS for ureteral and both-region tumors than for renal pelvic tumors, even when adjusted for stage [19]. There are several possible explanations for the conflicting results between the present and previously published studies. In an international collaborative study from 13 centers worldwide, Raman et al. enrolled 1249 patients with UTUC managed by radical nephroureterectomy and assigned them into renal pelvic and ureteral groups [17]. After adjustment for pathologic tumor classification, grade, and lymph node status, tumor location did not independently predict cancer-specific mortality. Potential bias in this study may lie in the fact that tumors involving both the renal pelvis and ureteral regions were classified based on the dominant tumor location (in accordance with stage or size) under either the renal pelvic or ureteral group. In cases where renal pelvic and ureteral tumors are of the same stage, the tumor size is used to identify the tumor location. We believe this methodology can result in misclassification and bias, especially in an international retrospective study. Although the both-region group contained only 11 cases in this study, we postulate that tumors involving both the renal pelvis and ureter should be analyzed as distinct entities to avoid misclassification. In further analysis, tumor location remained a significant prognostic factor for RFS, CSS, and OS after excluding the both-region group. We also reported several important patient-related factors (e.g., history of previous bladder cancer, history of previous UTUC, and ASA score) and tumor-related factors (e.g., hydronephrosis, multifocality, CIS, lymphovascular invasion, and surgical margins) that were not assessed by Raman et al. [17]. In a similar report using administrative data from nine registries of the SEER database, Isbarn et al. identified 2824 patients treated with nephroureterectomy for UTUC and divided them into dichotomies according to primary tumor location [18]. Although the main findings were not different in terms of oncologic outcomes between patients with renal pelvic and ureteral tumors, data were collected by medical files review at participating institutions, thus introducing discrepancies in the interpretation of study variables. Overcoming these limitations, the study variables in the present study were reviewed by two independent urologists (YCH and HLL). Furthermore, this study was not a multi-institutional collaborative study, and practice patterns, including patients' access to care, disease management, surgical techniques, and follow-up after surgery, were relatively uniform at our institute.
The poor prognosis of ureteral involvement can be explained in several ways. Compared to renal pelvic tumor with the natural barriers of renal parenchyma, perirenal fat, and Gerota's fascia, ureteral involvement has a thin wall containing an extensive plexus of blood and lymphatic vessels, enabling easier invasion and spread of tumor cells [20]. Higher prevalence of hydronephrosis in ureteral involvement is also associated with more pronounced eGFR deterioration after radical nephroureterectomy [21], thereby restricting the use of postoperative cisplatin-based chemotherapy. Interestingly, a previous history of bladder cancer increased the risk of cancer-specific death by a 2.54-fold factor relative to no previous history of bladder cancer. Indeed, the proportion of multifocality and CIS was significantly higher in the group with a history of previous bladder cancer, which is a well-known predictor of poor outcomes after UTUC, than in the group with no previous history of bladder cancer, and this result is consistent with previous findings [22,23].
However, the effect of sex on the prognosis of UTUC after radical surgery remains unclear. Sikic et al. reported a 2.92-fold higher risk of cancer-specific death in female patients aged 59 years and older than in male patients [24]. Milojevic et al. found no significant difference in the CSS between female and male patients treated with radical nephroureterectomy [25]. In contrast, Wu et al. showed that male patients with UTUC were associated with more metachronous bladder cancer and higher cancer-specific mortality compared to female patients with UTUC [26]. Our observations that male patients with UTUC have worse cancer-specific mortality compared to female patients based on multivariate analysis are consistent with the previous findings. Multiple factors, including genetic background, environmental exposure, tumor biology, hormonal variation, and anatomical factors, may play a role in the reported sex differences. However, this finding is not in line with recently published data [27], and further epidemiologic and molecular research is required to address the impact of sex on the incidence, progression, and metastasis of patients with UTUC.
Unlike surgical margins, the effects of age on clinical outcomes in patients with UTUC have rarely been discussed. A large population-based study using the SEER database showed that older age is directly associated with a decrease in CSS after adjustment for stage, grade, and treatment type [28]. Reasons for this may be include changes in the biological potential of tumor cells, decreased host immunity with advancing age, or even different choices of treatment in elderly patients compared with younger patients [29]. In the current study, 3.2% of patients treated with segmental ureterectomy were aged < 60 years, and 13.0% were aged > 80 years, given the higher risk of residual disease [30]. Furthermore, elderly patients are less likely to undergo salvage chemotherapy for disease relapse [31], which has been shown to be associated with improved survival. These results indicate that treatment choice, at least in part, may account for the worse outcomes in older patients.
The current study had several limitations. First, due to the retrospective design of this study, biases are inevitable, as segmental ureteral resection in patients with distal ureteral tumors, with a serious renal insufficiency, or having a solitary kidney was chosen depending on patient preference after discussion with the treating urologist. Second, a centralized pathological review is lacking. The specimens being evaluated by various genitourinary pathologists over a long period could have led to discrepancies in the interpretation of the pathologic specimens. Third, the number of patients was too small to draw definite conclusions, particularly in the both-region group. Fourth, this study lacked data on adjuvant treatment. We could not confirm the effect of perioperative chemotherapy, immunotherapy, or radiotherapy on survival outcomes. Despite these limitations, our study was a relatively large cohort study focusing on the outcomes of pT2 UTUC after radical surgery. Our results indicate that patients with tumors located in the ureter or renal pelvis plus synchronous ureter could be candidates for additional treatment and closer follow-up after radical surgery. Prospective assessments to obtain a definitive role of adjuvant therapy in patients with pT2 UTUC are warranted.

Conclusions
Patients with pT2 UTUC and presence of ureteral involvement (ureteral and bothregional groups) had more frequent disease relapse and worse long-term oncological outcomes than other patients. Male sex, older age, history of previous bladder cancer, and positive surgical margins remain important unfavorable prognostic factors for recurrence and survival. Our findings support the need for more stringent follow-up strategies and subsequent adjuvant treatment in patients with those poor prognostic factors despite complete pathological removal of the disease. Prospectively large-scale studies investigating the role of tumor location in patients with pT2 UTUC are needed to obtain a definitive statement regarding this matter.