Multimodal kidney‐preserving approach in localised and locally advanced high‐risk upper tract urothelial carcinoma

Abstract Objectives Multimodal kidney‐preserving (MKP) strategies may be an option for patients with localised or locally advanced high‐risk upper tract urothelial carcinoma (UTUC) who have a relative contraindication for nephroureterectomy (NU). Materials and methods We studied patients with UTUC who were managed with MKP strategies, consisting of systemic anticancer therapy, with or without local/topical strategies after endoscopic control of intraluminal tumours. Primary end points were overall survival (OS) and progression‐free survival (PFS). Results Fourteen patients received MKP treatment between August 2013 and April 2020. Median baseline estimated glomerular filtration rate was 43 mL/min/1.73m2. MKP was mainly pursued to avoid dialysis (10/14, 71%), followed by low performance status and/or comorbidities (2/14, 14%). All patients had received systemic therapy: chemotherapy (64%) and immunotherapy (36%). Endoscopic control and/or laser ablation was feasible in 7 (50%) patients. Calculated overall risk of non‐organ confined disease was 35%. Predicted 2‐year and 5‐year relapse‐free probability (RFP) was 74% (24–92%) and 62% (10–85%), respectively. Median follow‐up was 31 months (95% CI: 22.6, NE), median OS was 48.1 months (95% CI: 48.1, NE) and 2‐year OS probability was 0.89 (95% CI: 0.71, 1). Median metastases‐free survival was 48.1 months (95% CI: 26.8, NE), median PFS was 22.4 months (95% CI: 15.6, NE) and 2‐year PFS probability was 0.48 (0.26, 0.89). Conclusion Management of high‐risk localised or locally advanced UTUC with MKP strategies was associated with good tolerance, preservation of renal function, and comparable PFS and OS to predicted in vulnerable patients. Prospective studies with more patients are needed to evaluate these possible benefits relative to current standards.


K E Y W O R D S
Bacillus Calmette-Guerin, chemotherapy, endoscopic resection, immunotherapy, mitomycin, upper tract urothelial carcinoma 1 | INTRODUCTION Nephroureterectomy (NU) is part of the standard-of-care treatment for high-grade or clinically infiltrating upper tract urothelial carcinoma (UTUC) and includes the removal of the entire kidney, ureter and ipsilateral bladder cuff. 1,2 An estimated 60% of UTUC have a muscle invasive tumour (i.e., ≥T2) at time of diagnosis as compared with 15-25% of bladder tumours, and understaging of UTUC on imaging is frequent. 3,4 Invasive tumours significantly impact outcomes in UTUC with a 5-year cancer-specific-mortality free rate of 50-60%, as compared with 80-90% in non-invasive tumours. 5,6 Recent data have highlighted the benefit of perioperative systemic therapy for high-risk UTUC, and surgery has long remained an important consolidative strategy in this disease. [7][8][9][10] However, there are indications for multimodality kidney-preserving (MKP) approaches, which may be alternatives to NU in select patients where surgery may carry higher morbidity, such as patients with bilateral involvement, tumours in a solitary kidney, chronic kidney disease (CKD) or poor performance status. This is a relatively common scenario that can arise for patients with underlying CKD, which is strongly associated with UTUC, 11 and who have multiple synchronous or metasynchronous recurrences of UTUC. Current MKP modalities are indicated largely for low-grade tumours and may be delivered surgically or endoscopically, through either antegrade or retrograde access, and are well described in the literature. 1,2 Endoscopic management of UTUC is associated with high recurrence rates, which can be reduced by application of topical therapy such as mitomycin. 12,13 However, there is minimal data for outcomes in UTUC patients who have a relative contraindication for NU and higher risk disease associated with higher rates of progression. In these cases, a 'non-traditional' approach using endoscopic local control in addition to systemic management to manage an invasive component may provide an alternative to NU and dialysis or severe CKD.
The objective of the present study is to evaluate the clinical characteristics and treatment outcomes of patients with localised or locally advanced high-risk UTUC with vulnerable kidney function or who refused standard of care options, focusing on the ability of this strategy to prevent NU or dialysis and progression to metastasis.

| Patients and follow-up protocol
Baseline characteristics and clinical outcomes were retrospectively collected for patients with localised or locally advanced high-risk UTUC who were treated at the University of Texas MD Anderson Cancer Center (MDACC), Houston, Texas. All patients had baseline and surveillance chest, abdomen and pelvis imaging with contrast enhanced computed tomography or magnetic resonance imaging with or without gadolinium depending on contraindications, as well as cystoscopy with ureteroscopy. Patients were followed with serial imaging every 3 months in the first 1-2 years then every 6 months for at least 2 additional years; urine cytology was not performed a as part of routine surveillance but may have been obtained selectively. Ureteroscopy was done for biopsy, disease assessment and, if feasible, attempted local control using a Holmium or diode laser. Patients did not undergo serial ureteroscopy. In cases of apparent radiographic complete response (CR) or near CR, a 'confirmatory look' ureteroscopy with laser ablation of any residual disease was performed.
Patients with locally advanced or node positive (N2+) disease were excluded. This study was approved by the Institutional Review Board (IRB) of MDACC protocol RCR05-0521 with waiver of informed consent. In order to provide a reference, we used the preoperative nomogram reported by Petros et al. to predict risk of non-organ confined (pT3-4 or pN+) disease 14 and the nomogram reported by Freifeld et al. to predict disease recurrence following radical NU. 15

| End points
End points of interest included overall survival (OS), 5-year OS probability, progression-free survival (PFS) and 5-year dialysis-free probability.
Given that many patients had history of concordant or discordant tumours elsewhere in urothelial tract, date of diagnosis was defined as that for the UTUC managed with 'non-standard'/MKP approach. OS was calculated as the duration from diagnosis to date of death or to the date of last follow-up for patients alive. Two-year and 5-year OS probability was calculated as the percentage of patients alive at 2 and 5 years from diagnosis, respectively. PFS was calculated as the duration from diagnosis to date of progression (local and distant) or date of death, whichever occurred first, or to the date of last follow-up for patients alive and without progression. Five-year dialysis-free probability was calculated as the percentage of patients free from dialysis at 5 years from diagnosis.

| Statistical analysis
Continuous variables were summarised using descriptive statistics, and categorical variables were tabulated with frequency and percentage. The Kaplan-Meier method was used to estimate the time to event outcomes, and the log rank test was used to compare these outcomes between subgroups of patients. SAS software v9.4 (SAS Institute Inc., Cary, NC) and Splus software v8.2 (TIBCO software Inc., Palo Alto, CA) were used for statistical analysis.

| Baseline characteristics of patients
Between August 2013 and April 2020, of 353 patients with UTUC, 21 were recommended treatment using MKP strategies, and 7 did not return to our centre for treatment, leaving 14 who are the subject of this study. Table 1 summarises the baseline clinical characteristics of the patients. Median age was 74 (range: 57-89), 50% (7/14) of patients were male and 43% had a history of smoking. Median estimated glomerular filtration rate (eGFR) at baseline was 43 mL/ min/1.73m 2 (range: 22-87). Avoiding dialysis was the most common reason (10/14, 71%) to elect MKP strategy over NU; 43% (6/14) of patients had prior contralateral nephrectomy. Low Eastern Cooperative Oncology Group (ECOG) performance status (PS) of 2-3 was the second most common reason to avoid surgery, occurring in 14% (2/14). Two-thirds of the patients had a history of a prior UTUC, and half had a prior bladder cancer; the prior management of these 10 patients with previous urothelial cancer diagnoses is included in Table 1.

| DISCUSSION
We report on our experience using MKP management strategies to treat challenging and under-recognised scenarios of localised or locally advanced high-risk UTUC with relative contraindications for NU. Contraindications to NU were either due to poor baseline renal function and/or poor PS with surgery-prohibitive comorbidities. The 5-year OS probability rate was 38% (95% CI 9-100%) with MKP management strategies, which might initially appear lower than historic 5-year OS rate of 80.2% 16  Cutress data showed a 5-year disease-specific survival of 60-79% but F I G U R E 1 Overall and progressionfree survival pointed out that this is likely an overestimate due to biassed reporting and raw figures uncensored to overall vital status; in fact, studies that appropriately censored patients showed 5-year disease specific survival of 32-38% for grade 3 disease. 17 The risk of progression may be as high as 88% by 2-3 years in high-grade cases. 17 In our cohort, we had a predicted 2-year and 5-year RFP of 74% (24, 92) and 63% (15,85), respectively, if patients had undergone NU. The actual esti-  Limitations include the small sample size, the inherent selection bias associated with retrospective studies and lack of a comparator arm; use of the preoperative nomograms served as a way to provide a reference for expected disease outcomes if NU had been selected.
Small sample size resulted in the imprecise estimates of 5-year OS probability as compared with the 2-year OS probability. The different endoscopic, topical and systemic therapies that patients received are also a limitation; however, they show that management of these patients is a case-by-case decision. The common features in our 14 patients were their need to avoid NU because of poor renal function and the predicted poor tolerability to dialysis and surgery. These patients appeared to have better survival compared with historical data with endoscopic-only management and comparable outcomes to nomogram prediction, with the added benefit of renal preservation in a highly vulnerable situation. Given the lack of data with kidney preservation in high-grade UTUC, prospective data with a larger cohort would be important to corroborate our observations and to inform clinical practice.

| CONCLUSION
In patients with node-negative localised or locally advanced high-risk UTUC with relative contraindication to NU, management with MKP strategies was associated with good local and systemic control, was well tolerated and showed promising data on overall, recurrence, progression and dialysis-free survival. These kidney-preserving strategies require a multidisciplinary team approach and should be individualised for each patient after discussion of the benefits and the possibility of a non-curative approach, as well as the need for close surveillance to avoid suboptimal oncologic outcomes.