The modern management of upper tract urothelial cancer: surgical treatment
Francis X. Keeley Jr, Consultant Urologist, Bristol Urological Institute, Southmead Hospital, Bristol BS10 5NB, UK.
upper urinary tract.
Primary urothelial carcinoma of the upper tract is a rare urological disease, accounting for ≈ 5% of all urothelial tumours and 7–10% of all kidney tumours [1,2]. It has a propensity for multifocality, local recurrence and development of metastatic disease. Reports show a slight increase of incidence during the last 30 years, as well as a slight improvement in survival . However, the current management recommendations are based mainly on retrospective series and expert opinion, as there are no controlled randomized trials .
Almost 100 years have passed since Albarran described radical nephroureterectomy (RNU) with en bloc excision of the peri-ureteric bladder cuff . Since then, and through the decades, it became the standard treatment for most patients with TCC involving the upper urinary tract (UUT). This approach meets all conditions required for oncological surgery, including excision of the entire tumour with an adequate surgical margin, control of local recurrence and assessing the anatomical spread of the tumour . This last issue is most important as several studies show that survival of patients with UUT-TCC depends mainly on stage and grade of the tumour.
The general concept of oncological surgery has changed significantly in the last 15 years. Advances in diagnosis, staging and treatment of patients with cancer have resulted in improved survival and an overall change in the prognosis of several malignant diseases. This has lead to modifications in surgical techniques, giving rise to minimally invasive and organ-sparing procedures. Much of this change has come about because of advances in diagnosis and adjuvant therapy.
In recent years several investigators reported the results of endoscopic treatment of UUT tumours, initially based on efforts to preserve renal function in patients with single kidneys or bilateral tumours. In this article we review current concepts and oncological outcomes of the surgical management of UUT-TCC considering recent technical developments. Is the standard RNU still necessary? Can organ-sparing techniques be used safely?
RNU: CLINICAL ASPECTS
Several issues, particularly related to prognostic factors, must be considered when evaluating the therapeutic efficacy of radical surgical treatment for UUT tumours. These are based mainly on large series analysed retrospectively and published in the last three decades [1,2,7–15].
Urothelial carcinoma is commonly multicentric. Adequate identification of all UUT lesions is challenging with the available diagnostic methods. Unilateral multifocal UUT tumours account for 6.2–27%[2,7,10,11] and their frequency apparently correlates with differentiation . RNU provides the best local treatment of all lesions within the ipsilateral tract, but it might result in overtreatment in some cases.
Bilateral synchronous UUT tumours have a low incidence, occurring in 0.01–4% of cases [1,2,8]. Bilateral metachronous ureteric and renal pelvic carcinomas are also rare. In a recent review of 786 patients, 3.1% developed a contralateral tumour after a median follow-up of 46 months, with an estimated incidence of 6.5% at 15 years . In that study, bladder cancer was identified as the most important risk factor for developing a bilateral metachronous tumour. However, the low risk of both synchronous and metachronous tumour incidence makes radical surgery a safe and rational therapy in patients with a healthy contralateral kidney, especially in moderate- to high-grade disease.
Hall et al. identified the surgical procedure as an independent predictor of overall disease recurrence in a multivariate analysis in a retrospective study of 252 cases, being significantly higher in patients who had parenchymal-sparing surgery, including distal or segmental ureterectomy, excision of the tumour and endoscopic ablation. However, recurrence in the remaining UUT in patients undergoing simple nephrectomy or partial NU is also high. Huben et al. reported a 40% rate in the ureteric stump in patients with simple nephrectomy, while further reports found a 19–23% recurrence rate in the preserved urinary tract after these conservative procedures [9,11].
The low incidence of recurrence after radical treatment makes disease surveillance less complicated , in that there is no need to include ureteroscopy in the follow-up schedule. IVU should be recommended only in patients with a history of bladder cancer, as subsequent contralateral tumour development is higher in these cases, at up to 7.4%. The risk of developing subsequent tumours in the bladder is 13.4–42%[1,8,9,11,13,14]. This makes a periodic endoscopic examination of the bladder imperative; the frequency is set depending on the risk factors previously identified.
Disease-specific survival and risk of distant metastases depend mainly on tumour stage [1–3,8,16] and grade [8,9,16]. A high proportion of UUT tumours are high-stage and high-grade, and ureteric tumours are associated with a worse outcome than those within the kidney . A review of the Surveillance, Epidemiology and End Results public database for TCC of the UUT from 1985 to 1996 (5052 patients) showed an incidence of 45.1% for locally advanced and metastatic tumours . In a series of 139 patients over a period of 20 years in Mannheim, 43.8% had at least pT2 tumours ; other surgical series reported similar proportions (34–59%) [1,2,9,11]. Considering the better results of radical surgery than for more conservative approaches in terms of recurrence and survival, it follows from the above that RNU is the safest treatment for most TCC of the UUT.
RNU: TECHNICAL ASPECTS
Radical surgery for UUT-TCC includes the perinephric fat and Gerota’s fascia. Johansson et al. compared in an unrandomized way patients treated with intrafascial nephrectomy and perifascial nephrectomy. The difference in prognosis was significant in favour of the second group, and could not be explained by differences in extent of tumour infiltration, tumour grade or tumour size. However, patients in the first group received extirpation of a varying length of the ureter. Further studies analysing this issue are not available; however, it is widely accepted that Gerota’s fascia and perinephric fat should be excised during RNU.
Routine ipsilateral adrenalectomy is also commonly performed in the setting of RNU. By contrast to patients with RCC there are no available data on the rate of adrenal metastasis or for direct invasion by renal pelvic tumours. The role of adrenalectomy is therefore not clearly defined.
The role of lymphadenectomy is still under debate. Lymph nodes are the most common metastatic site in UUT-TCC; the prognosis of patients with lymph node metastases was reported to be generally poor [7,14,15]. As the presence of lymph node metastases seems to be a sign of systemic disease, it is unlikely that there would be any therapeutic benefit in routine lymphadenectomy.
Skinner  proposed a routine lymphadenectomy in all patients; its extent would be determined by the location of the primary tumour. However, a therapeutic lymphadenectomy would only be effective in cases where the tumour-bearing lymph nodes were near the primary lesion. Contralateral lymph node metastases are rare. The appropriate extent of lymphadenectomy has not been established.
Few clinical analyses have assessed the value of lymphadenectomy. Komatsu et al. suggested a potential diagnostic role of lymphadenectomy for identifying candidates for adjuvant chemotherapy, considering the strong prognostic information of nodal status. In an unrandomized retrospective study, Miyake et al. found a significant benefit to lymphadenectomy in patients with no lymphatic vessel invasion in the specimen. The authors suggested a prophylactic eradication of minimal metastatic disease as the reason for this prognostic advantage. The recently reported prolonged survival of patients with extended lymphadenectomy during cystectomy for urothelial bladder cancer  provides the basis for prospective studies in patients with UUT-TCC.
The laparoscopic approach has gained wide acceptance in renal surgery because of the lower postoperative morbidity. The procedure maintains the established surgical principles of open surgery, with complete urothelial excision and adequate surgical margins. A comparative report on the long-term follow-up  showed improved 5-year survival rates for the laparoscopic cases, probably because of patient selection. A more recent study showed no significant difference in 7-year disease-specific survival rates of patients undergoing open or laparoscopic NU .
There are still concerns about some specific risks. The incidence of port-site metastases was reported as 1.6% in a review . Seeding is also a potential cause of local recurrence after transurethral detachment of the orifice. However, the technique of ureterectomy has not been standardized yet. The growing surgical experience and technical expertise in this field are expected to reduce these specific technical concerns.
The ultimate goal of endoscopic management is cancer control while simultaneously preserving renal function and integrity of the urinary tract. Initially reserved for patients with solitary kidneys or bilateral disease, it has gained acceptance in the management of small, low-stage and low-grade tumours. Recent reports have shown that in these situations local control of disease and long-term renal preservation are achievable. These procedures include percutaneous and retrograde ureteroscopic approaches.
Many reports of minimally invasive surgery for TCC of the UUT have emerged in recent years, but only a few studies have had reasonably many patients. Most series are small and include all types of indications, ranging from elective to palliative. Tumour characteristics (location, size, grade, multifocality) vary widely among the series. The heterogeneous nature of the patient population and the inconsistent reporting of results makes objective interpretation of these reports challenging.
Percutaneous treatment of UUT-TCC involves gaining access into an appropriate calyx in a similar manner to percutaneous nephrolithotomy. The tumour is biopsied, then either resected or cauterized, much like transurethral treatment of bladder TCC. Adjuvant topical therapy can be given after surgery via a nephrostomy tube. Some authors recommend treating the nephrostomy tract with localized radiotherapy, using an iridium wire .
Several relatively large retrospective case series have reported recurrence rates for low-grade tumours (grades 1–2) of 25–30% and disease-specific survival rates of 96–100%. However, most series with a mixture of low- and high-grade tumours report higher recurrence rates. Palou et al. recently reported a 44% local recurrence rate and a 73% renal preservation rate in 34 patients after a mean follow-up of 51 months, with percutaneous treatment; 5.9% of patients died from disease.
Compared to percutaneous treatment, ureteroscopy has the advantage of preserving the integrity of the urinary tract. The advent of small-diameter, flexible ureteroscopes combined with laser fibres has allowed effective treatment throughout the UUT, even in the proximal ureter and intrarenal collecting system. The procedure involves the following steps, which were described previously : diagnostic ureteroscopy, including rigid and flexible ureterorenoscopy to assess the entire collecting system; obtaining urine and biopsy tissue for cytological assessment; and a combination of Nd:YAG laser to coagulate and Ho:YAG laser to ablate any visible tumours.
A significant disadvantage of the ureteroscopic approach is the small size of the equipment, which limits the volume of tumour that can be treated in one procedure. This might explain the finding that the size of the tumour is reported as an independent risk factor for the failure of ureteroscopic treatment, although it is likely that tumour bulk might simply correlate with higher grade and stage . The size limitation dictates that ureteroscopic treatment requires a ‘second look’ procedure to ensure complete eradication of the tumour.
Retrospective case series show recurrence rates of 31–65% and disease-free rates of 35–86%. Low-grade tumours treated electively, with a normal contralateral kidney, have disease-free rates as high as 91%. By contrast, Daneshmand et al. recently reported a series of 30 cases of ureteroscopic treatment and surveillance in a heterogeneous patient group, many of whom had a solitary kidney. An average of 3.4 recurrences occurred after a mean follow-up of 38 months. The average time to first recurrence was 7 months and 20% of tumours progressed. Overall, 20 of the 30 patients had their kidney preserved.
Both the percutaneous and ureteroscopic approaches are associated with a high risk of ipsilateral recurrences, making follow-up mandatory. Much like the follow-up for bladder cancer, endoscopic follow-up has been shown to be more sensitive for recurrences than a radiological survey . Surveillance ureteroscopy is usually carried out at 3 and 6 months, then every 6 months for a year, and annually thereafter. Provided patients are carefully selected, i.e. with small-volume, low-grade and low-stage disease, the kidney can be preserved in most cases with a strict follow-up regimen.
Concern about the limited radiographic and endoscopic staging of UUT tumours is critical. Grading is possible but has pitfalls and limitations . Furthermore, UUT-TCC appears to have, on average, a worse prognosis than TCC of the bladder . Given the above points, there is a high risk of poor selection of cases due to undergrading or understaging, which might result in inappropriate treatment. Endoscopic treatment failures might ultimately lead to metastases and death despite delayed NU. This risk must be balanced against the loss of renal function after NU and the small risk of bilateral tumours.
The essential message from previous reports on the endoscopic treatment of UUT-TCC is that patients must be selected carefully for this approach. Patient selection is based on tumour size, grade, multifocality, and presence of frankly malignant cells on cytology. Single, low-grade tumours of <15 mm in the absence of positive urine cytology predicts a good outcome with endoscopic treatment, provided patients have regular ureteroscopic surveillance. The great majority of patients with UUT-TCC do not fall into this category; therefore, endoscopic treatment should be reserved for a small subset of patients. Given that stage is often not obtainable from biopsy specimens, it behoves the clinician to be very wary of undergrading and understaging the disease.
ENDOSCOPIC TREATMENT VS RNU
The lack of prospective comparative studies, let alone randomized controlled trials, makes this comparison especially difficult. Despite the limitations listed above, endoscopic treatment has been shown to have a similar overall survival as in patients undergoing RNU, in a recent case series . However, by selecting out patients with a good prognosis, endoscopic management for elective indications might be expected to have a higher survival rate than radical surgery. For patients with more risk factors, endoscopic treatment runs the risk of undertreating the tumour but has the clear benefit of preserving renal function for some patients.
Recurrence rates of bladder TCC also appear to be similar after endoscopic or extirpative surgical treatment . Moreover, ureteroscopy before NU does not appear to confer a disadvantage in terms of survival compared to immediate NU based on imaging alone; furthermore, a delay of >6 months before NU in patients undergoing attempted endoscopic treatment does not appear to affect survival . Thus, there appears to be little if any harm caused by ureteroscopic biopsy and attempted treatment of suitable cases, provided patients have an adequate follow-up.
Lastly, in comparison to RNU, endoscopic treatment requires a strict and long-term surveillance protocol, similar to the follow-up of TCC of the bladder, first to ensure complete eradication of the tumour bulk and second to detect recurrences in a timely manner.
Open RNU, including the complete urothelial excision and adequate surgical margins, still represents the reference standard for treating TCC of the UUT. There is no clarity about the role of lymphadenectomy or the necessity of adrenalectomy. New techniques of evaluation, diagnosis and surgical treatment are not yet established, despite some encouraging results in small series. However, until we have non- or minimally invasive tools to adequately predict tumour stage and aggressiveness, there will always be a significant risk of understaging and undertreatment in conservatively managed tumours. Considering that most UUT tumours are high-grade and high-stage, it follows that endoscopic treatment will be inappropriate for most cases.
The laparoscopic approach with adherence to standard radical surgical principles seems to be comparable to that of open surgery for cancer outcomes. Moreover, it offers the established advantages of this procedure in terms of morbidity and convalescence. Finally, as the survival of patients is determined by tumour stage and grade, further strategies of adjuvant systemic therapy in addition to diagnostic surveillance methods must be developed and validated to improve the outcome of high-risk tumours.
CONFLICT OF INTEREST