UT tumors are identified most frequently as an obstruction or a filling defect detected by a contrasted exam such as intravenous urography (excretory urogram), retrograde pyelography or, more commonly, CT urography (30,31,32). CT and MRI studies are equally valuable for resolving the differential diagnosis that includes calculi, blood clots, external compression, and inflammatory lesions (ureteritis cystica) (33). Cross sectional imaging with CT or MRI is important for evaluating the extent of the disease and possible invasion of adjacent organs and structures, and for detecting regional lymphadenopathy or distant metastasis (34). CT urography is more accurate than excretory urography and should be considered the method of choice for radiographic evaluation of patients with hematuria (35,36). Moreover, threedimensional (3D) reconstruction provides an anatomic full view of the upper and lower urinary tract that may also be useful for surgical planning. Transurethral retrograde and antegrade percutaneous approaches, when the lower urinary tract is not accessible, provide anatomic detail, allow collection of urine samples for biomarker assessment and access for cold cup and/or brush biopsy in order to establish a diagnosis. Relief of obstruction, with placement of a stent or nephrostomy tube, may provide symptomatic relief and optimize renal function, which will be critical if neoadjuvant chemotherapy is considered. Furthermore, since there is no absorption of the contrast medium through the urothelium, the retrograde pyelography is a safe alternative to CT urography or intravenous pyelography in patients with a contrast allergy (34) or chronic renal disease in whom intravenous gadolinium is contraindicated because of the risk of nephrogenic systemic fibrosis (37).

Urine cytology is a very useful tool for diagnosis and surveillance of UC in the UT, but sensitivity is poor for low-grade tumors. Voided urine samples are helpful to give an initial impression but do not distinguish upper and lower urinary tract origin. Ureteral catheterization with a washing technique to collect selective cells from the renal pelvis may identify up to 80% of high-grade lesions (38). The use of brushing techniques and postbrushing wash may improve sensitivity to 90% (39,40,41). The selective collection of urine samples should occur prior to retrograde pyelography as the high-osmolar contrast agents may alter the cytologic features. Similar to bladder cancer, a number of molecular biomarkers have been studied in the upper urinary tract. These include telomerase (42,43), loss of heterozygosity at 9p21 and other microsatellites regions (44,45,46,47), tumor ploidy (48), p27 (49), FISH (50,51,52), and the nuclear matrix protein-based marker (NMP22) (53,54,55). All seem to improve overall sensitivity but are not as specific as cytology. p53 immunohistochemistry of cytologic specimens has also been shown to correlate well with the presence of UC tumors in the UT (56,57,58). A comprehensive review of urinary markers is beyond the scope of this chapter. However, several new molecular markers may provide greater sensitivity than cytology, particularly for low-grade lesions, but they are less specific with high rates of false positives (59,60,61). The past decade witnessed an explosion of new markers, and further studies are warranted to define their role in the diagnosis, surveillance, and treatment of UT tumors.

Visualization of the entire collecting system combined with the capabilities of performing direct-vision biopsy through a flexible ureteroscope has become the method of choice for diagnosis. Without the limitations of the rigid ureteroscopes, the flexible instruments provide increased accuracy for diagnosis by allowing more effective exploration of more proximal areas and hard-to-reach angulated calyces (62,63). The current generation of flexible ureteroscopes has high-definition (HD) and narrow-band imaging (NBI) capabilities, which provide high-quality images and the potential to detect occult dysplastic and papillary lesions. However, because of the small size of the scopes and thin ureteral wall, the staging capability of ureteroscopic methods is limited by the size of current instrumentation and risk of perforation. Nevertheless, the concordance between the ureteroscopic biopsy specimen and the final nephroureterectomy specimen grade is high, ranging from 90% to 92% (64,65). Because of the limited size of the biopsy obtained through the ureteroscope, CT or MRI is critical for final staging and planning of treatment.

The most commonly used guidelines for staging are those set out by the American Joint Committee on Cancer (AJCC), using the TNM (tumor, node, metastasis) system shown in Table 21A.1 (66).

Grade has been shown to correlate with the aggressiveness of the tumors, and the system traditionally used for bladder lesions can also be applied to UT tumors. Recently, the threetier (grades 1, 2, and 3, or low-, intermediate-, and high-grade) system has been replaced by a two-tier system (low- and highgrade), but because of subtleties in the grade 2 category, it is actually recommended that, for clarification, both systems be used simultaneously (67,68).

UCs of the UT are a heterogeneous group of lesions. As in the bladder, these include low-grade noninvasive tumors that are associated with a high probability of recurrence and a low risk of progression, carcinoma in situ, and invasive tumors that are usually high grade and behave more aggressively, and frequently metastasize to lymph nodes and distant sites.

Due to the high frequency of synchronous multiple tumors in the ipsilateral renal pelvis and/or ureter and an increased risk of recurrence, the mainstay of the treatment with curative intent is radical excision of the ipsilateral kidney and ureter, including a bladder cuff and the intramural ureter. The major downside of this approach is the reduction of renal parenchyma and total renal function. Thus, the greatest challenge lies in finding the balance between conservative nephron sparing endoscopic resection or nephroureterectomy. Small low-grade tumors when accessible can be identified and treated conservatively by minimally invasive methods and require meticulous follow-up due to the high risk of recurrence. High-grade tumors require definitive treatment with nephroureterectomy. Conservative endoscopic management may be appropriate for patients not medically fit for major surgery or when nephroureterectomy would result in a requirement for hemodialysis, However, nonextirpative therapy exposes the patient to a high risk of progression and a missed opportunity for cure (69).

Because invasion of the renal sinus fat is a common finding in invasive high-grade tumors of the renal pelvis, removal of the kidney, including all the perirenal fat and Gerota fascia, is mandatory. UT tumors usually metastasize initially to regional retroperitoneal and pelvic lymph nodes. Therefore, a thorough ipsilateral lymph node dissection is also recommended. The extent of the dissection is determined by the anatomic lymphatic drainage of the upper urinary tract. For renal pelvic tumors, this is similar to that described for renal cell carcinoma and includes, on the left, suprahilar and para-aortic nodes, and on the right, suprahilar, paracaval, and interaortocaval nodes. For ureteral tumors, the proximal ureter drains to the ipsilateral retroperitoneal nodes similar to testis cancer, and drainage
of the mid- and distal ureter includes common iliac and pelvic nodes. The use of perioperative chemotherapy is guided by inference from UC of the bladder as there are no randomized clinical trial data to guide us. Some centers are investigating the use of neoadjuvant chemotherapy for patients with high-grade tumors or clinical evidence of regional node metastasis (70,71). The use of adjuvant chemotherapy may be limited by reduction in renal function limiting the use of cisplatin-based chemotherapy (71).

This procedure, which is the gold standard treatment for patients with UTUC and a normal contralateral kidney, involves the complete resection of ipsilateral kidney and ureter, including its intramural portion, and requires the resection of a bladder cuff with at least 1 cm of clear margin in the bladder mucosa. The ipsilateral adrenal gland should be spared if possible, unless there is direct involvement from an upper calyx tumor or presence of a nodule suspicious for metastasis. If a UT tumor is identified in the surgical specimen of a radical nephrectomy executed for a suspected renal cell carcinoma, a second procedure must be performed to completely excise the remaining ureteral stump, since the risk for recurrence in the stump has been reported to range from 20% to 84% (81,83,90,91,92,93). The same principle of complete ureteral resection applies to patients who have previously undergone cystectomy and urinary diversion. The cuff around the ureteroenteric anastamosis of the reservoir/conduit should also be removed to ensure complete resection.

The procedure can be performed through open access or laparoscopically. For open surgery, a single thoracoabdominal extrapleural extraperitoneal incision provides optimum access from the diaphragm to the pelvis, optimizing exposure to the retroperitoneum for both the nephroureterectomy and the lymphadenectomy. Some surgeons prefer a two-incision approach with a flank incision for the nephrectomy portion and a distal incision either Gibson, Pfannenstiel, or infraumbilical midline for the distal ureter. Although a long, single midline incision can be used, proximal exposure to the suprahilar region may be challenging depending on the patient’s body habitus.

There is controversy about the best method of management of the distal ureter and the bladder cuff excision. Several surgical techniques have been described, but basically the bladder cuff can be removed through (a) a cystostomy (open or laparoscopic) with the optional closing by a two-layer suture; (b) a transurethral dissection or disarticulation using a Collins knife with “pluck” technique when removing en bloc with the ureter, or with the distal part removed through the urethra; and (c) stapling the ureterovesical transition with or without cystoscopic intramural ureter unroofing and fulguration (94,95,96,97,98,99,100,101,102,103,104). Although reasonable results have been reported for all these techniques, it seems that leaving an open cystostomy is not advisable in patients with high-grade lesions, distal ureteric location, previous bladder tumors, or contralateral UT tumors. The stapler technique has the advantage of close excision of the uretero-vesical transition, but it risks leaving part of the cuff and intramural ureter unresected, or an exposed staple line in the bladder, predisposing to stone formation. In fact, increased positive margins and recurrence rates have been reported with this technique (99). The largest series reporting on the outcomes of patients from the SEER registry undergoing radical nephroureterectomy, with or without excision of the bladder cuff, demonstrated significantly improved cancer-specific survival in patients with T3-4 and or N+ disease, who had undergone bladder cuff excision. Because the same benefit was not observed in patients with lower stage disease and urothelial recurrence was not evaluated, the safety of avoiding the bladder cuff excision in patients with T1-2 tumors is unclear (97). Removal of the bladder cuff should remain in 2010 the standard of care accompanying any radical nephroureterectomy or distal ureterectomy, regardless of the technique. The authors’ preference is to perform the modified pluck technique with ligation of the distal ureter (to avoid spillage) associated with open or laparoscopic approach, followed by cystostomy closure using absorbable two-layer suture.

Usually indicated for invasive or high-grade lesions situated in the mid- or distal portions of the ureter, this approach is supported by the fact that ipsilateral tumor recurrences tend to be located distal to the previous site, with proximal recurrences being a rare event (77,105). The margins of resection should be clear at least 1 cm from the resection line, and the ends of the affected segment resected must be closed in order to avoid spillage. The principles of complete excision of the intramural ureter with a cuff of bladder are identical to nephroureterectomy. In patients with invasive or high-grade lesions, lymph node dissection involving the main drainage area associated with tumor location is recommended, as is the bladder cuff excision in all cases involving distal ureterectomy.

In a recent study, analyzing 2,044 patients with ureteral UC extracted from the SEER database, the authors compared the results of cancer-specific survival between segmental resection and nephroureterectomy, with and without bladder cuff excision. These patients had a median follow-up of 30 months and were characterized by stages pT1, pT2, pT3, and pT4 in 37.2%, 30.5%, 25.2%, and 7%, respectively; and grade 1, 2, 3, and 4 lesions in 6.1%, 32.6%, 37.6%, and 23.7%, respectively. They observed that there was no statistically significant difference between the three approaches, even when stratified by stage, with 5-year cancer-specific survival rates of 86.6%, 82.2%, and 80.5%, respectively. These findings support the use of segmental resections, whenever it is technically feasible, with the additional benefit of preserving renal function in these patients (106).

Discussion of reconstruction after segmental resections is not within the scope of this chapter, but it is important to keep in mind the options for reconstruction and their impact on future surveillance of the remaining urothelial tract when planning a segmental resection of UT. Usually, the ureteroneocystostomy is made directly in a nonrefluxing fashion, assisted by a psoas hitch technique or a Boari flap. The risk for recurrence in the ipsilateral UT is reported to be up to 70%, and it varies according to the stage and grade of the tumor (107). Therefore, long-term surveillance is needed in these patients.

Kondo et al. (108) explored the patterns of lymph node metastasis according to the location of the tumor in the UT. They and others have suggested that the proximal limit for lesions located in the renal pelvis and upper ureter should start at the level of the renal hilum (108,109,110,111,112,113,114). There is a risk, however, of leaving behind potential lymph node-bearing tissue in the suprahilar area, and we prefer to include those nodes in the lymphadenectomy specimen as pointed out recently by Dalbagni (115).

Following principles established for radical cystectomy and pelvic lymphadenectomy, recent retrospective multicenter studies suggest that lymph node dissection provides more accurate staging and more accurate risk assessment of subsequent progression. There is also the potential for improving cancer-specific and disease-free survival in selected patients with limited and localized lymph node only metastatic disease (109,110,111,112,113,114,116). The largest series is a multicenter study performed in 13 centers worldwide, including 1,130 consecutive patients with pT1-4 primary UTUC treated with radical nephroureterectomy. They observed that lymph node status (pN+ vs. pNx vs. pN0) was significantly associated with disease-free and cancer-specific survival, and also that the extent of lymph node dissection affected the outcomes. The hazard ratios between pN+ versus pN0 were 2.185 and 2.120 (both p < 0.001), and between pNx versus pN0 were 1.398 and 1.424 (p = 0.018 and p = 0.016), for disease-free survival and cancer-specific survival, respectively (112). Although there is no prospective trial supporting a therapeutic benefit, it makes sense to include a regional anatomic node dissection in patients with high-grade disease in order to optimize pathologic staging. In patients with clinical or pathologic evidence of node metastasis, the node dissection is more of an imperative in order to optimize locoregional cancer control. If gross evidence of node metastasis is found at the time of initial exploration, consideration should be given to backing out in order to give chemotherapy followed by salvage nephroureterectomy and lymphadenectomy (112,113,114,115,116,117,118,119).

As for radical cystectomy (116,117,118,121), the number of nodes removed is associated with cancer-specific survival in patients with pN0 disease, suggesting a potential curative benefit (113). Roscigno et al. (113) determined that a minimum of eight lymph nodes should be removed by lymph node dissection during a radical nephroureterectomy in order to maximize recurrence-free and cancer-specific survival. The number of nodes identified in the pathologic specimen may be a surrogate for the adequacy of the lymphadenectomy. It is likely however that the anatomic extent of the node dissection determines the potential benefit of the lymphadenectomy (115).

Therefore, for tumors located in the renal pelvis or proximal ureter, the ipsilateral hilar, preaortic and paraaortic for tumors on the left or paracaval nodes for tumors on right should be included. For right-side tumors, interaortocaval nodes should be also included in the template, as these are the primary sites of metastasis (112,113,114,117). An anatomic lymph node dissection for UC of the renal pelvis and/or upper ureter should extend from the crus of the diaphragm to the bifurcation of the aorta. It should include all nodes described above anterior, posterior, and lateral to the ipsilateral great vessel (aorta on the left side and vena cava on the right side), with interaortocaval nodes added for right-side tumors. For lesions located in the midureter, it is reasonable to dissect from the renal hilum to the common iliac nodes including the nodes associated with the ipsilateral great vessel, and for lesions located in the distal (pelvic) ureter, the inclusion of pelvic (common, external, and internal iliacs, and obturator) nodes (114).

Newer technology and improved optical systems have catapulted percutaneous and ureteroscopic techniques into the routine armamentarium of the contemporary urologist. This approach, also called nephron sparing, was initially limited to patients with solitary kidney, bilateral disease, renal insufficiency, and those who were not medically fit for a radical nephroureterectomy. With safety demonstrated by early adopters, select patients with unilateral disease and normal contralateral kidney function also began to be treated endoscopically (39,65,76,118).

A plethora of technology is available including cold cup biopsy forceps that fit through flexible ureteroscopes, ureteroresectoscopes, electrocoagulation/fulguration, and ablation using holmium:YAG and neodymium:YAG lasers. Regardless of the technique, the principles of treatment should remain the same. Biopsy should be attempted when technically feasible, and then, depending on the size, stage and grade of the lesion, ablation of the tumor (74). Antegrade percutaneous access may also be used alone or in combination with the retrograde approach. The main advantages of the percutaneous technique are better visibility, easy access to the calyces and proximal ureter, capacity for larger instrumentation, and access to the UT in patients with urinary diversions or UT reconstructions that pose difficult retrograde instrumentation (76). However, there is always a potential risk of seeding in the percutaneous tract and occurrence of pyelovenous lymphatic backflow due to increased irrigation pressure that may expose the patient to a risk of inadvertent systemic spread of disease. The incidence of these phenomena is considered rare, with only a few case reports present in the literature (119,120,121,122,123). Although larger contemporary series of percutaneous and ureteroscopic management of UTUC have not reported any recurrence associated with these mechanisms even with ureteral perforation, the principle of maintaining a closed urinary system to avoid spillage should be respected, whenever possible (124,125,126). The retrograde approach has the advantage of maintaining a closed urinary system and lower morbidity. The size limitations imposed by smaller diameter ureteroscopes have been progressively overcome with new scopes incorporating HD optics and NBI technology, which provide improved clarity of visualization for the endoscopic procedures. NBI may provide improved sensitivity for assessing the margins adjacent to the area of resection, and identifying mucosal lesions that may appear normal under regular white-light standard optics. With improvements in technology, we expect a potential expansion in the use of endoscopic management. However, in order to avoid unnecessary manipulation of large aggressive lesions that may not be amenable to endoscopic resection, or that are invasive, a careful preoperative evaluation should always take place before considering endoscopic ablation. In these cases, the chances of complete endoscopic resection are reduced and the theoretical risk of dissemination of the tumor associated with perforation should not be ignored.