Malignancies of the renal pelvis and ureter

INTRODUCTION — The mucosal surfaces of the renal collecting tubules, calyces, and pelvis, as well as the ureter, bladder, and urethra, all have the same embryologic origin, and the term "urothelium" is used to delineate the lining surface epithelium [1]. This common embryologic origin has important implications for the management of patients with these tumors.

The clinical manifestations, diagnosis, and treatment of malignancies arising in the urothelium of the upper urinary tract (renal pelvis or ureter) are reviewed here. Urothelial tumors originating in the bladder and urethra are discussed separately. (See "Overview of the initial approach and management of urothelial bladder cancer" and "Urethral cancer".)

PATHOLOGY

Primary tumors — Urothelial (formerly known as transitional cell) carcinomas of the upper urinary tract (UTUC), bladder, and urethra tend to be multifocal. This phenomenon is termed "field cancerization" and is thought to be caused by exposure of the urothelium to potential carcinogens that are either excreted in the urine or activated by hydrolyzing enzymes in the urine. (See "Pathology of bladder neoplasms" and "Epidemiology and risk factors of urothelial (transitional cell) carcinoma of the bladder", section on 'Risk factors'.)

Primary tumors arising in the renal pelvis include urothelial carcinomas, squamous cell carcinomas, and adenocarcinomas.

●Over 90 percent of tumors of the renal pelvis and ureter are of urothelial origin and are histologically identical to those originating in the bladder.

●Squamous cell carcinomas account for approximately 8 percent of tumors of the renal pelvis [2]. These tumors are associated with a poorer prognosis than urothelial tumors because they tend to be sessile and deeply invasive at presentation. Squamous cell carcinomas have been associated with antecedent calculi or chronic infection [3].

●Primary adenocarcinoma and small cell carcinoma arising in the renal pelvis is exceedingly rare.

Drop metastases — Urothelial tumors can spread to urothelial structures that are either distal or proximal to the primary tumor (so called "drop metastases"). Upper urinary tract tumors frequently are multiple or occur synchronously with bladder tumors. This may reflect the propensity of neoplastic cells to flow down from the renal pelvis or reflux from the bladder to the ureter, forming an invasive implant.

Drop metastases may be difficult to differentiate from separate primary lesions. Recurrent bladder tumors often cluster around the ureteral orifice of an affected ureter, thus providing evidence for their origin as metastases from the original primary tumor [4]. (See "Epidemiology and risk factors of urothelial (transitional cell) carcinoma of the bladder", section on 'Risk factors'.)

Metastatic tumors — Tumors from outside the genitourinary tract may spread to the renal pelvis or ureter by hematogenous or lymphatic channels, or may involve these structures by direct extension [5-7]. Signs and symptoms of metastatic disease to the urinary tract usually appear late in the course of disease and may only be recognized preterminally or at autopsy.

Direct invasion of the ureter by a non-genitourinary malignancy most commonly occurs with carcinoma of the cervix, carcinoma of the colon, or retroperitoneal lymphoma. Most compress rather than invade the ureter. Treatment is directed toward the primary tumor and relief of bilateral ureteral obstruction.

CLINICAL MANIFESTATIONS

Epidemiology — Tumors of the upper urinary tract are almost twice as common in men compared with women, with a mean age at diagnosis of 73 years [8]. The age-adjusted annual incidence rates of ureteral and renal pelvis cancer in the Surveillance, Epidemiology, and End Results (SEER) database was 0.91 and 1.15/100,000 person-years during the period from 1997 to 2005. The incidence of upper urinary tract tumors has slowly increased, and there has been an associated shift toward earlier-stage disease.

Multifocality is common since the entire urothelial surface is affected by the same carcinogenic influences (the "field cancerization" effect), although bilateral upper tract tumors occur in less than 2 percent of cases. Areas of invasive or in situ carcinoma may present either synchronously or metachronously.

Relationship to urothelial bladder cancer — Approximately 17 percent of patients will have a concurrent bladder cancer at presentation [9]. The incidence of upper tract tumors in patients with primary bladder tumors is increased compared with the general population, and 40 to 50 percent will have urothelial cancer of the bladder either at presentation or subsequently [10]. In those with a history of carcinoma in situ, the cumulative incidence of upper tract tumors is 20 to 25 percent at ten years [11]. (See "Epidemiology and risk factors of urothelial (transitional cell) carcinoma of the bladder", section on 'Upper urinary tract cancer'.)

Environmental causes — A number of other environmental factors have been associated with the preferential development of urothelial cancer of the upper urinary tract:

●Chemical carcinogenesis – Chemical carcinogenesis due to cigarette smoke or occupational exposures is associated with urothelial carcinoma of the bladder and is also an important factor in upper tract urothelial carcinoma [10]. (See "Epidemiology and risk factors of urothelial (transitional cell) carcinoma of the bladder", section on 'Chemical carcinogenesis'.)

●Balkan endemic nephropathy – In the Balkan region (Bulgaria, Greece, Romania, former Yugoslavia), an indolent inflammatory process of the renal interstitium (so called "Balkan endemic nephropathy") is associated with the development of urothelial tumors of the renal pelvis and ureter [10]. In that geographic region, these urothelial tumors account for almost 50 percent of all renal cancers. (See "Balkan endemic nephropathy".)

●Lynch syndrome – Patients with Lynch syndrome have abnormalities in DNA mismatch repair. These abnormalities are classically associated with colorectal and endometrial tumors. Studies have shown that these abnormalities are also associated with urothelial neoplasms, particularly of the ureter and renal pelvis [12-14]. The lifetime risk for urologic cancer has been estimated to be approximately 8 percent by age 70 years [14]. (See "Lynch syndrome (hereditary nonpolyposis colorectal cancer): Clinical manifestations and diagnosis".)

●Arsenic exposure – Exposure to arsenic has been associated with the development of upper urinary tract cancer in Taiwan [15-17]. (See "Arsenic exposure and chronic poisoning", section on 'Cancer'.)

●Phenacetin use – Habitual use of analgesics that contain phenacetin has been associated with a significant increase in the incidence of urothelial tumors of the renal pelvis and ureter [18-21]. (See "Urinary tract malignancy and atherosclerotic disease in patients with chronic analgesic abuse", section on 'Urinary tract malignancy'.)

●Chinese herb nephropathy – Chinese herb nephropathy can cause a progressive renal fibrosis that is frequently associated with urothelial malignancy of the renal pelvis [22,23]. (See "Nephropathy induced by aristolochic acid (AA) containing herbs".)

Clinical presentation — Hematuria is present at diagnosis in 70 to 80 percent of patients. Obstruction of the ureter or ureteropelvic junction due to a tumor mass causes flank pain in 20 to 40 percent of cases. Other urinary tract symptoms, such as those associated with bladder irritation, and constitutional symptoms occur in less than 10 percent of cases [10].

The physical examination is usually unrevealing. In rare cases, a flank mass, caused either by the tumor or associated hydronephrosis, may be palpated.

DIAGNOSIS — Radiologic imaging procedures and ureteroscopy are the initial diagnostic tools for patients with a suspected tumor of the ureter or renal pelvis.

Patients with urothelial carcinoma of the upper urinary tract (UTUC) must be carefully evaluated at presentation for multiple synchronous tumors. Because of the high incidence of multiple tumors, the entire urinary tract, including the urethra and bladder mucosa, should be visualized either radiographically or endoscopically.

Radiologic imaging — The diagnosis of a tumor of the renal pelvis or ureter is usually suspected based upon an abnormal computed tomography (CT) or retrograde pyelography (image 1) [1,24,25]. CT urography has virtually replaced intravenous pyelogram (IVP), offering better visualization of the urinary collecting system as well as evaluating for direct extension or nodal involvement [26,27].

Gadolinium-enhanced magnetic resonance urography is an alternative for assessment of the upper urinary tract, but gadolinium should not be used in patients with severe kidney impairment due to the risk of nephrogenic systemic fibrosis [10].

Typical findings on radiographic imaging are a mass in the renal pelvis (image 2). CT or magnetic resonance imaging (MRI) of the abdomen and pelvis may also detect extension of the tumor outside the collecting system, the presence of adjacent organ involvement, and/or the presence of distant metastases.

The most common finding on retrograde pyelography is a solitary filling defect that can be observed in 50 to 75 percent of patients. Tumors of the renal pelvis may also present with multiple filling defects, ureteropelvic junction (UPJ) obstruction with hydronephrosis (image 3), infundibular stenosis, splaying of the calyces (suggesting a kidney mass), or nonvisualization of the kidney collecting system. Non-contrast spiral CT can identify renal pelvic or ureteral calculi, which may appear as a filling defect on the contrast study and therefore be confused with a tumor of the collecting system.

Ureteropyeloscopy with cystoscopy — Flexible fiberoptic ureteroscopes allow direct visualization of the entire collecting system. This procedure is generally safe and well tolerated [28,29]. In a series of 113 patients with upper tract urothelial tumors undergoing ureteroscopy, the diagnosis was confirmed in 74 percent of upper tract tumors [30].

Ureteropyeloscopy allows biopsy to confirm the diagnosis and may permit endourologic treatment in carefully selected patients. However, such biopsy samples obtained during ureteropyeloscopy are small and can potentially underestimate the risk of more severe disease. As an example, in a series of 56 patients initially thought to have low-grade, non-invasive tumor, 24 (43 percent) were subsequently found to have either high-grade or invasive disease in a repeat biopsy or surgical resection specimen obtained at median of 1.5 months later. (See 'Endourologic surgery' below.)

Brush biopsy — Retrograde brushing of the upper urinary tract was initially described in 1973 [31]. An angiographic catheter is passed into the ureter transurethrally and positioned adjacent to the suspicious lesion. A small nylon or steel bristle brush is passed inside the lumen, and material entrapped by the bristles is sent for histologic and cytologic examination. This technique allows accurate diagnosis of moderately differentiated urothelial cancer, but it may produce false-negative results in patients with low-grade lesions [32].

Although ureteroscopy and direct biopsy have largely replaced retrograde brush biopsy, brushing and cytology may be useful in evaluating patients with positive upper tract cytology who do not have visible lesions on radiography or ureteroscopy.

Urine cytology — Low-grade urothelial cancers of the upper tract generally are not associated with positive cytology, but higher-grade lesions may generate a positive cytology. In patients with an unexplained positive cytology and negative cystoscopy/biopsy, evaluation of the upper tracts and prostatic urethra is mandatory.

Cytologic examination of urine is less reliable for upper urinary tract neoplasms than for bladder cancers. Voided specimens for urinary cytology have a poor yield for low-grade tumors, and positive findings may reflect a synchronous bladder tumor. Retrograde collections at the time of ureteroscopy that are aided by furosemide diuresis may yield better results.

A pathologist with particular expertise in cytology is critical for the interpretation of such specimens. Cytologically, nuclear characteristics generally indicate the tumor grade, but a direct correlation of the cytologic grade with histologic tumor grade and invasiveness is lacking. As an example, in one series there was agreement between cytology and histologic assessment in 70 percent of patients with upper tract urothelial carcinoma [33].

Urine biomarkers — Evaluation of a wide range of urinary biomarkers is an area of active research for patients with known or suspected urothelial neoplasms. None of these tests have sufficient diagnostic reliability to replace endoscopy or urine cytology.

The status of these urinary tumor markers is discussed separately. (See "Urine biomarkers for the detection of urothelial (transitional cell) carcinoma of the bladder".)

STAGING — The eighth edition (2017) of the tumor, node, metastasis (TNM) system is widely used to stage tumors of both the renal pelvis and ureters (table 1) [34]. Any nodal involvement or distant metastasis constitutes stage IV disease in the TNM system. The TNM stage is correlated with outcome following definitive treatment. (See 'Prognosis' below.)

Urothelial cancer of the upper urinary tract may spread by direct extension, via the lymphatics, or hematogenously. Regional lymph nodes are commonly the initial site of metastasis [35].

Routine staging procedures should include cystoscopy to exclude associated bladder cancer, a chest radiograph or CT, radionuclide bone scan if there are symptoms suggesting bone involvement or an elevated bone-derived alkaline phosphatase, and an evaluation of hepatic, kidney, and hematologic function. CT or MRI of the abdomen may be performed to assess for the presence of retroperitoneal lymphadenopathy, as well as other sites of disease extension.

TREATMENT OF LOCALIZED DISEASE — Surgery is the only potentially curative treatment modality for urothelial carcinoma of the renal pelvis or ureter. Surgery may be carried out via open, laparoscopic, endoscopic, or percutaneous approaches.

Several factors should be considered when determining the surgical approach:

●There is very little difference between kidney-conserving and kidney-sacrificing procedures in terms of operative risk. In fact, nephroureterectomy may be performed more quickly and more easily than some less extensive procedures.

●The likelihood of local recurrence, especially with tumors of higher grade, may be higher for kidney-sparing as opposed to kidney-sacrificing procedures.

●Removal of the entire ureter and kidney obviates the need for close follow-up for metachronous ipsilateral tumors or local recurrence. This is the predominant site of recurrence and is the major focus of follow-up after kidney-sparing procedures.

A systematic review conducted by the European Association of Urology (EAU) incorporated data from 22 nonrandomized studies that compared kidney-sparing surgical approaches with radical nephroureterectomy. This analysis suggested that patients with low-grade noninvasive carcinoma had a similar survival outcome when managed with kidney-sparing approaches compared with those managed with radical nephroureterectomy.

Nephroureterectomy — Nephroureterectomy, using either a laparoscopic (with or without robotic assistance) or an open approach, with excision of a cuff of normal bladder and bladder mucosa is the gold standard for most patients with urothelial tumors of the renal pelvis or ureter [36]. Less extensive procedures (endourologic surgery, segmental ureterectomy) are appropriate for patients with localized low-grade lesions [37-39]. (See 'Endourologic surgery' below and 'Segmental ureterectomy' below.)

For patients with tumors arising in the renal pelvis, removal of the entire ipsilateral ureter is recommended because approximately 20 percent of patients with residual ureteral stumps will develop tumors within the stump. Similarly, total ureterectomy and nephrectomy are recommended for patients with ureteral cancers because of the high incidence of multiple ipsilateral lesions [40].

Laparoscopic versus open nephroureterectomy — Laparoscopic and open nephroureterectomy have been compared in a meta-analysis that included a total of 4328 patients in 21 studies who underwent surgery [41]. The main controversy with the minimally invasive approach is the optimal management of the distal ureter and cuff of bladder. In general, minimally invasive approaches offer reduced morbidity with equal efficacy. However, the most important factor in choosing a surgical technique is the expertise and experience of the surgeon.

●There were no statistically significant differences in the rates of intraoperative or postoperative complications or in the rate of perioperative mortality.

●There was a lower rate of bladder recurrence with a laparoscopic approach, although the metastasis rate was not significantly different. There were no statistically significant differences in overall survival at two and five years. The interpretation of these data is limited, since patients managed with a laparoscopic approach were more likely to have Ta/Tis or T1 disease and less likely to have T3 or T4 lesions (table 1).

Port site recurrence is a rare complication; in a retrospective series of 116 laparoscopic procedures, only one port metastasis was observed (0.9 percent) [42].

The use of robotic laparoscopic nephroureterectomy is increasing, but supporting evidence for this technique is limited and mainly low quality [43].

Lymph node dissection — All high-grade and/or high-stage lesions should have regional lymph node dissection. The importance of lymph node status as a prognostic parameter was studied in a multicenter series of 1130 consecutive patients who underwent nephroureterectomy for upper tract urothelial cancer [44]. The five-year cancer-specific survival rates were significantly better in those with negative lymph nodes compared with those not undergoing a lymph node dissection or with positive lymph nodes (77 versus 69 versus 35 percent, respectively).

However, it is not clear whether lymph node dissection improves survival in patients with urothelial cancer of the upper urinary tract or whether the identification of patients with positive lymph nodes simply identifies a poor-prognosis subset and also may identify patients who may benefit from adjuvant chemotherapy [44,45].

Kidney-sparing surgical approaches — Kidney-sparing surgical approaches definitely have a role for patients with low-risk disease and normally functioning kidneys, as well as for those with a solitary kidney and/or impaired kidney function.

Patient selection — Kidney-sparing approaches may be particularly useful in several settings:

●For patients with a solitary kidney, compromised kidney function, bilateral lesions, or Balkan endemic nephropathy, kidney-sparing surgery is essential whenever possible. The preferred treatment is local excision of a renal pelvic lesion, with or without partial nephrectomy [46]. (See "Definitive surgical management of renal cell carcinoma", section on 'Partial nephrectomy'.)

●Selected lesions involving the renal pelvis and calyceal system may be considered for conservative treatment, provided they are not high-grade urothelial cancers. Ureteroscopy may be helpful to evaluate the degree of local extension prior to surgery. Small biopsy specimens obtained at ureteroscopy should be interpreted with caution, since limitations in sampling may miss areas of high-grade and/or invasive disease [47]. (See 'Ureteropyeloscopy with cystoscopy' above.)

●In patients with a ureteral tumor, ureteral excision or ureterectomy with replacement by ileum may be indicated in carefully selected cases. Local excision with adequate margins and reanastomosis should be limited to patients with a localized polypoid filling defect and a low-grade, low-stage tumor.

For patients with a solitary low-grade tumor in the lower third of the ureter, distal ureterectomy with regional lymphadenectomy and ureter reimplantation with a psoas hitch represents the preferred procedure. In patients with higher-grade lesions, the likelihood of synchronous upper tract lesions favors nephroureterectomy. Tumors in the middle or upper third of the ureter are generally best treated by nephroureterectomy, unless cytologic and brush biopsy specimens demonstrate a low-grade and presumably low-stage tumor. In such patients, excisional biopsy with ureteroureterostomy can be considered.

Patients with carcinoma in situ may benefit from instillation of Bacillus Calmette-Guerin (BCG). However, retrospective data suggest that this endourologic procedure is not as effective in preventing recurrence in patients with invasive lesions as in those with Tis lesions [48]. (See 'In situ disease' below.)

For all patients who undergo a kidney-sparing procedure, frequent surveillance is required because of the high risk of recurrent disease [39,49]. As an example, in a single-institution series, 83 patients with a normal contralateral upper tract were managed with an endoscopic approach [39]. At a median follow-up of 4.6 years, 46 patients developed 76 upper tract recurrences, and bladder recurrences were identified in 37 patients. Overall, 27 patients required nephroureterectomy, and nine patients died of urothelial cancer.

Endourologic surgery — The development of equipment and techniques to directly visualize the upper urinary tract has evolved into kidney-sparing approaches that are appropriate for carefully selected patients. These approaches include retrograde ureteropyeloscopic resection, either alone or in combination with antegrade percutaneous techniques.

Retrograde ureteropyeloscopic resection — Endoscopic resection of renal pelvis and ureteral tumors is a newer option for patients with low-grade, limited-stage tumors [28,50,51]. The preferred approach is to access and resect the tumor through a retrograde ureteropyeloscopy.

Results with this approach are illustrated by a series of 90 patients with an antecedent history of urothelial carcinoma of the bladder, who were subsequently treated endoscopically for urothelial cancers of the renal pelvis or ureter [50]. At a median follow-up of over four years, 105 upper urinary tract recurrences were documented in 55 patients (61 percent). In addition, 38 patients experienced a total of 91 recurrences in the bladder. At last follow-up, 17 patients had died of urothelial cancer at a median of 3.4 years after the original diagnosis.

However, whether endoscopic approaches are equivalent to surgical management is unclear. A Surveillance, Epidemiology, and End Results (SEER) database study compared 151 patients with low-grade upper urinary tract cancer with 302 propensity-matched patients managed with surgery. Although there was no difference at two years, those treated with surgery subsequently had better overall and cancer-specific survival. In the 53 percent of patients initially treated endoscopically who subsequently underwent surgery, cancer specific-survival remained lower than for those initially treated with surgery.

Percutaneous techniques — Percutaneous management can be used in conjunction with retrograde ureteropyeloscopy for patients with low-grade, early-stage disease if a complete resection is not feasible from below. In this approach, a catheter is placed percutaneously into the renal pelvis, and then passed in an antegrade fashion to allow the resection to be completed.

In one series, percutaneous resection salvaged 9 of 15 kidneys in such patients after a median follow-up of 63 months [52]. Less well-selected patients may have a worse prognosis, with ipsilateral recurrence rates as high as 41 percent in one report of 34 patients (eight multifocal, five high-grade) followed for an average of 51 months after percutaneous management [53].

This technique is not considered standard of care because of the risk of tumor spillage into the perirenal space and should only be used in highly selected patients for whom nephroureterectomy is not possible. Rare cases of port site recurrence have been reported [54].

Segmental ureterectomy — Distal ureterectomy with reimplantation of the ureter is the preferred procedure for low-grade distal tumors. Segmental ureterectomy may also be appropriate for carefully selected patients with compromised kidney function.

This approach is supported by a SEER database analysis that included 569 patients who underwent segmental ureterectomy and 1222 treated with nephroureterectomy [55]. At a median follow-up of 30 months, the five-year cancer-specific mortality-free rates were not significantly different for segmental ureterectomy compared with nephroureterectomy with or without bladder cuff removal (87 versus 82 versus 81 percent, respectively).

Kidney sparing non-surgical approaches

Low-grade noninvasive disease (Jelmyto) — For patients with treatment-naïve or recurrent low-grade upper tract urothelial carcinoma (UTUC) confined to the renal pelvis in whom complete tumor ablation is not feasible, we offer intracavitary administration of pyelocalyceal mitomycin (Jelmyto) as an alternative to nephroureterectomy. These patients commonly experience recurrent disease within the upper urinary tract but have limited nonsurgical options. In a prospective single arm phase III trial, pyelocalyceal mitomycin demonstrated an initial complete response rate of 59 percent [56,57]. While this agent is a reasonable alternative to surgical options such as nephroureterectomy, clinicians should discuss the risks and benefits of both approaches with their patients.

Pyelocalyceal mitomycin is a formulation of mitomycin admixed with a reverse thermogelation hydrogel; at cold temperatures, the solution becomes a viscous liquid that can be injected via a ureteral catheter or nephrostomy tube. At body temperature, a semi-solid, water soluble ureteral gel forms that dissolves over a four- to six-hour period. This allows the mitomycin to be retained in the renal pelvis, overcoming one of the primary limitations of intracavitary therapy. Pyelocalyceal mitomycin is administered at 4 mg/mL weekly for six weeks via a nephrostomy tube or a ureteral catheter [56,57]. The total instillation volume is based on volumetric measurement on pyelography and should not exceed 15 mL (60 mg) total per dose. Patients with a clinical complete response to therapy may continue monthly instillations for up to 11 months. Pyelocalyceal mitomycin should be avoided in patients with impaired kidney function (ie, glomerular filtration rate [GFR] <30 mL/min).

Of note, the formulation of pyelocalyceal mitomycin is different from intravesical mitomycin, which is used to treat non-muscle invasive urothelial bladder tumors. Intravesical mitomycin is diluted in a different solution (typically normal saline) and administered via a urinary catheter. Further details on the administration and efficacy of intravesicular mitomycin is discussed separately. (See "Treatment of primary non-muscle invasive urothelial bladder cancer".)

An open-label, single arm trial (OLYMPUS) evaluated the efficacy of pyelocalyceal mitomycin in 71 patients with treatment-naïve or recurrent low-grade upper tract urothelial carcinoma [56,57]. Tumors ranged between 5 and 15 mm in size and were located above the ureteropelvic junction. Tumor debulking was allowed in some patients prior to treatment as long as there was at least 5 mm of residual tumor. Primary disease response was assessed at three months using urine cytology, ureteroscopy, and biopsy as clinically indicated. Patients with a complete response at three months were offered continuing treatment with monthly pyelocalyceal mitomycin instillations for up to 11 months. At median follow-up of 12 months, complete responses (CR) at three months were seen in 41 patients (58 percent); within this cohort of complete responders, 23 patients (56 percent) had continued CR at 12 months. The estimated durability of response at 12 months after primary disease evaluation was 82 percent and median time to recurrence was not estimable.

Common treatment-related adverse events included ureteric stenosis (44 percent), urinary tract infection (32 percent), hematuria (32 percent), flank pain (30 percent), nausea (25 percent), and impaired kidney function (20 percent). Cytopenias were uncommon, such as anemia (14 percent), thrombocytopenia (4 percent), and leukopenia (3 percent). Approximately one-half of patients with an adverse event related to the kidney/urinary system were treated with a temporary ureteral stent.

Based on the results of this trial, pyelocalyceal mitomycin was approved by the US Food and Drug Administration (FDA) for the treatment of patients with low-grade UTUC [58].

In situ disease — Some patients with isolated in situ UTUC (ie, no invasive disease and negative random biopsies of the bladder and prostatic urethra) can be managed with instillation of BCG [10,48,59]. However, this approach should be limited to patients in whom definitive therapy is not feasible. (See "Treatment of primary non-muscle invasive urothelial bladder cancer".)

In one report of 55 consecutive patients treated for lesions in 64 kidneys, recurrence occurred in 2 of 42 kidneys (5 percent) treated for Tis and 9 of 22 (41 percent) of those with Ta or T1 lesions at a median follow-up of 42 months [48]. Caution should be utilized with instillation of BCG in the upper urinary tract to prevent intravasation and BCG sepsis.

Adjuvant systemic therapy for high-risk disease

Adjuvant platinum-based chemotherapy — For patients with high-risk disease (ie, those with muscle invasive [pT2 or greater] and/or any node positive disease) who have undergone nephroureterectomy, we suggest adjuvant chemotherapy with gemcitabine plus platinum-based chemotherapy rather than surveillance. This approach improved disease-free survival (DFS) in patients with muscle invasive or node positive disease in a randomized phase III trial (Peri-Operative chemotherapy versus sUrveillance in upper Tract urothelial cancer [POUT]) [60] and is associated with improved DFS and overall survival in observational studies [61-63].

Based on POUT, we suggest administering four cycles of adjuvant gemcitabine and cisplatin to fit, high-risk patients who are eligible for cisplatin-based chemotherapy. For those who are ineligible for cisplatin-based chemotherapy, we administer four cycles of adjuvant gemcitabine plus carboplatin. Eligibility criteria for cisplatin-based chemotherapy are discussed separately. (See "Treatment of metastatic urothelial carcinoma of the bladder and urinary tract", section on 'Defining eligibility for systemic therapy'.)

Patients with high-risk disease after nephroureterectomy have high rates of relapse, leading to increased risk of disease progression and death. Data supporting improved outcomes with the addition of adjuvant platinum-based chemotherapy are as follows:

●The POUT trial, conducted in the United Kingdom, was an open-label phase III trial that enrolled 261 patients with either muscle invasive (ie, pT2-4, N any stage) or node positive disease (ie, T any stage, N1-3) who had undergone nephroureterectomy [60]. Patients were randomly assigned to either four cycles of adjuvant gemcitabine plus platinum-based chemotherapy (either cisplatin or carboplatin) initiated within 90 days of surgery, or surveillance. At a median follow-up of 31 months, adjuvant chemotherapy improved DFS compared with surveillance (three-year DFS 71 versus 46 percent, hazard ratio [HR] 0.45, 95% CI 0.30-0.68).

A DFS benefit was consistently seen across all prespecified subgroups, including those treated with gemcitabine plus cisplatin (HR 0.35) versus gemcitabine plus carboplatin (HR 0.66). Although gemcitabine plus cisplatin is the preferred adjuvant regimen for high-risk disease, based on these data, patients who are ineligible for cisplatin-based chemotherapy may still benefit from adjuvant gemcitabine plus carboplatin. There are no randomized trials that directly compare gemcitabine plus cisplatin with gemcitabine plus carboplatin in the adjuvant setting for UTUC, so the relative survival benefit for each specific regimen remains unclear.

Grade ≥3 toxicities were seen in 44 percent of patients treated with adjuvant chemotherapy, including neutropenia (36 percent); thrombocytopenia (10 percent); and nausea, vomiting, or febrile neutropenia (6 percent each). Metastasis-free and overall survival are secondary endpoints and have not been reported.

●Observational studies and meta-analyses suggest an overall survival benefit with adjuvant chemotherapy [61,63]. As an example, in a systematic review and meta-analysis of peri-operative chemotherapy in UTUC, adjuvant chemotherapy conferred a benefit for overall survival (pooled HR 0.77; 95% CI 0.64-0.92, across 14 studies and 7983 patients), cancer specific survival (pooled HR 0.79; 95% CI 0.69-0.91, across 18 studies and 5659 patients) and disease-free survival (pooled HR 0.52; 95% CI 0.38-0.70 across four studies and 602 patients). Most studies were retrospective but, there were two prospective randomized trials that provided high-quality evidence for benefit with adjuvant chemotherapy [63].

The role of adjuvant chemotherapy in patients with resected urothelial bladder cancer is discussed separately. (See "Adjuvant therapy for muscle-invasive urothelial carcinoma of the bladder", section on 'Patients eligible for cisplatin-based chemotherapy'.)

Adjuvant nivolumab — One year of adjuvant nivolumab is an alternative option for patients with resected UTUC at high-risk for disease recurrence (ie, pT3-T4a disease and/or nodal involvement) and are ineligible for (or decline) adjuvant cisplatin-based chemotherapy.

A placebo-controlled phase III trial (Checkmate 274) established a DFS benefit for adjuvant nivolumab in patients with resected high-risk urothelial carcinoma, including those with tumors of the urinary bladder (79 percent) and those with UTUC of the renal pelvis or ureters (21 percent) [64]. A subgroup analysis suggests a larger effect size for adjuvant nivolumab for urothelial carcinoma of the bladder than UTUC, although the study was not powered to evaluate this specific patient subgroup. Results of this trial for the entire study population are discussed separately. (See "Adjuvant therapy for muscle-invasive urothelial carcinoma of the bladder", section on 'Adjuvant nivolumab' and "Adjuvant therapy for muscle-invasive urothelial carcinoma of the bladder", section on 'Patients ineligible for cisplatin-based chemotherapy'.)

Postoperative intravesical chemotherapy — The role of postoperative intravesical chemotherapy (eg, mitomycin, pirarubicin) to prevent secondary bladder tumors is supported by randomized trials and is consistent with guidelines from the European Association of Urology (EAU) and the National Comprehensive Cancer Network (NCCN) [10,65,66]. Some experts offer single-dose intravesical chemotherapy to patients after nephroureterectomy independent of stage and grade. Randomized trials suggest reduced bladder cancer recurrence rates with this approach compared with surgery alone [10]. We do not suggest the use of maintenance intravesical chemotherapy, which did not improve recurrence free-survival when compared with single-dose intravesical chemotherapy in a randomized trial [67].

The following studies investigated postoperative intravesical chemotherapy:

●Single dose of intravesical chemotherapy – In a randomized European multicenter trial of 140 patients with upper tract tumors, the addition of a single dose of mitomycin following nephroureterectomy decreased the incidence of bladder recurrence at one year (16 versus 27 percent) [65]. Similarly, another randomized phase II trial reported reduced two-year bladder cancer recurrence rates using a single dose of intravesical pirarubicin compared with observation (17 versus 42 percent) [68].

●Maintenance versus single-dose intravesical chemotherapy – In a randomized trial of 74 patients treated with nephroureterectomy, one-year bladder cancer recurrence-free survival was similar for maintenance intravesical chemotherapy with epirubicin (administered over one year) compared with a single dose of chemotherapy (83 versus 87 percent) [67].

The efficacy of adjuvant chemotherapy in these patients is discussed separately. (See 'Adjuvant platinum-based chemotherapy' above.)

No role for postoperative radiation therapy — Postoperative (adjuvant) radiation therapy (RT) does not have a role in patients who have undergone complete resection of an upper urinary tract urothelial carcinoma. The role of adjuvant RT in muscle invasive urothelial carcinoma of the bladder is discussed separately. (See "Adjuvant therapy for muscle-invasive urothelial carcinoma of the bladder", section on 'Decisions regarding adjuvant radiation'.)

IS THERE A ROLE FOR NEOADJUVANT THERAPY? — The role for neoadjuvant chemotherapy in urothelial carcinoma of the upper urinary tract (UTUC) is not established. There is a theoretical advantage for using platinum-based chemotherapy prior to surgery, after which overall kidney function will be reduced. In patients with UTUC treated with neoadjuvant platinum-based chemotherapy, observational studies and meta-analyses suggest improved disease-free and overall survival compared with surgery alone [63,69,70] and initial clinical trials have evaluated the efficacy of various regimens [71-73]. However, widespread acceptance of this approach has been hampered by limited high-quality evidence or indications for selection of appropriate candidates for neoadjuvant chemotherapy.

The use of neoadjuvant therapy for muscle-invasive urothelial bladder cancer is discussed separately. (See "Neoadjuvant treatment options for muscle-invasive urothelial bladder cancer".)

POSTTREATMENT SURVEILLANCE — Because of the risk of metachronous urothelial tumors, posttreatment cancer surveillance of the bladder and contralateral ureter is also required.

For patients who have undergone a nephroureterectomy, we monitor with cystoscopy every six months for several years and then annually thereafter, based upon our approach in those who have been treated for a non-muscle invasive bladder cancer. For patients who have undergone a kidney-sparing procedure, surveillance should be extended to include the upper tract because of the risk of recurrence in the renal pelvis or ureter. This may include upper tract imaging (CT or magnetic resonance urography) and/or ureteroscopy [66].

Patients with urothelial carcinoma of the upper urinary tract (UTUC) are at high risk for the subsequent development of urothelial tumors of the bladder [74-79]. This was illustrated by a series of 82 patients who had complete resection of a urothelial cancer of the renal pelvis or ureter, in which urothelial carcinoma was subsequently diagnosed in 36 (44 percent) at a median interval of 14 months [78]. The bladder tumors were frequently multifocal (mean, 2.1 per patient).

TREATMENT OF METASTATIC DISEASE — The treatment of metastatic urothelial carcinoma of the upper urinary tract (UTUC) is the same as that used for metastatic urothelial bladder cancer (algorithm 1). Treatment is mostly extrapolated from studies of metastatic urothelial bladder cancer, which often include patients with UTUC. Of note, chronic kidney disease is common in patients with UTUC, which can impact selection of therapy [80]. Available treatment options are discussed in detail separately. (See "Treatment of metastatic urothelial carcinoma of the bladder and urinary tract".)

PROGNOSIS — The prognosis following definitive treatment for urothelial tumors of the upper urinary tract is related to the stage of the primary tumor and the presence or absence of regional lymph node involvement, as well as tumor grade [81-83]. (See 'Staging' above.)

The importance of stage was illustrated by a series of 1363 patients with urothelial cancer of the upper urinary tract treated with curative intent at 12 centers between 1992 and 2006 [81]. Nephroureterectomy was performed with open and laparoscopic approaches in 77 and 23 percent of cases, and lymphadenectomy was performed in 43 percent. Median follow-up was 37 months.

The following findings were noted:

●For the entire cohort, the five-year cancer-specific survival was 73 percent.

●The prognosis worsened with increasing stage of the primary tumor. The five-year cancer-specific survival rates for patients with pT0/Ta/Tis, pT1, pT2, pT3, and pT4 disease were 94, 91, 75, 54, and 12 percent, respectively (table 1).

●Patients with low-grade tumors had a significantly better prognosis than those with high-grade tumors (five-year cancer-specific survival 89 versus 63 percent).

●Patients with negative nodes or who did not undergo lymphadenectomy had a better prognosis than those with positive lymph nodes (five-year cancer-specific survival 77 versus 35 percent).

There is no major difference in prognosis between urothelial carcinomas arising the ureter compared with those originating in the renal pelvis [84,85].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Cancer of the kidney and ureters".)

SUMMARY AND RECOMMENDATIONS

●Treatment of localized UTUC – For patients with urothelial carcinoma (formerly known as transitional cell carcinoma) of the upper urinary tract (UTUC; which includes the renal pelvis and ureter) and without evidence of metastatic or unresectable locally advanced disease, we recommend surgical resection (Grade 1B). (See 'Treatment of localized disease' above.)

•Surgical approaches – For most patients, nephroureterectomy with excision of a cuff of normal bladder and bladder mucosa is the preferred procedure. For patients with a low-grade lesion of the distal ureter without evidence of invasion, a segmental resection with reimplantation of the ureter is preferred. (See 'Treatment of localized disease' above.)

-The frequent occurrence of synchronous or metachronous multifocal tumors makes a less extensive resection contraindicated. (See 'Nephroureterectomy' above.)

-Open and laparoscopic surgical approaches appear to be equally safe and effective, assuming adequate expertise is available and a complete resection of the intramural ureter with bladder cuff is performed. (See 'Laparoscopic versus open nephroureterectomy' above.)

•Indications for kidney-sparing surgery – For carefully selected patients, we suggest endourologic, kidney-sparing surgery rather than nephroureterectomy (Grade 2B). (See 'Endourologic surgery' above.)

-Patients in whom this approach is particularly appropriate include those with a solitary kidney, compromised kidney function, bilateral lesions, or Balkan endemic nephropathy. Those with low-grade, low-stage tumors of the ureter or renal pelvis may also be candidates for this approach, and this should be used in patients with high-grade tumor only in highly selected cases. (See 'Patient selection' above.)

-Treatment options in these settings include retrograde ureteropyeloscopy alone or in conjunction with an antegrade percutaneous resection. (See 'Endourologic surgery' above.)

•Pyelocalyceal mitomycin – For patients with treatment-naïve or recurrent low-grade disease in whom complete tumor ablation is not feasible, we offer pyelocalyceal mitomycin as an alternative to nephroureterectomy. (See 'Low-grade noninvasive disease (Jelmyto)' above.)

•Adjuvant systemic therapy – For fit, high-risk patients who have undergone nephroureterectomy, we suggest adjuvant platinum-based chemotherapy rather than surveillance (Grade 2B). We define high risk as muscle invasive (pT2 or greater) and/or node positive disease.

-For those who are cisplatin eligible, our preferred regimen is gemcitabine plus cisplatin. For those who are ineligible for cisplatin, we offer gemcitabine plus carboplatin. (See 'Adjuvant platinum-based chemotherapy' above.)

-One year of adjuvant nivolumab is an alternative option for patients with resected UTUC at high risk for disease recurrence (ie, pT3-T4a disease and/or nodal involvement) and are ineligible for (or decline) adjuvant cisplatin-based chemotherapy. (See 'Adjuvant nivolumab' above.)

●Treatment of metastatic UTUC – For patients with metastatic UTUC, the approach to systemic therapy is the same as that used for metastatic urothelial bladder cancer patients. (See "Treatment of metastatic urothelial carcinoma of the bladder and urinary tract".)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Philip W Kantoff, MD, who contributed to earlier versions of this topic review.