Epidemiology

Urothelial cell carcinoma (UCC) is the fourth most common cancer worldwide, after prostate (or breast), lung, and colorectal cancer. Among UCC, upper tract urothelial carcinoma (UTUC) is rare and accounts for 5% to 10% of all urothelial carcinomas with an annual incidence of 2 cases per 100,000 inhabitants in Western countries. UTUC is more common in men than in women with a male-to-female ratio of 2:1 and the mean age at the diagnosis of 65 years old. It is located twice more often in the renal pelvis than in the ureter and in about 20% of the cases a concomitant urothelial carcinoma of the bladder (UCB) is present. UTUC tends to recur in the bladder or progress through the lymphatic and vascular systems to distant organs. To date, much of the decision-making in UTUC comes from knowledge acquired in UCB because of its relative rarity. Over the last 5 years, a large body of evidence has proven that UTUC and UCB have common features but differ as well in a significant degree.

Risk Factors

Tobacco and occupational exposure are the main UTUC exogenous risk factors in most countries. The relative risk of developing UTUC in the case of tobacco exposure has been estimated to be 2.5 to 7, depending on the number of years of exposure and the number of cigarettes smoked per day. Occupational exposure to certain aromatic amines has an estimated risk of 8.3. However, exposure risk decreased since the 1960s because certain chemical substances, such as benzidine and β-naphthalene, have been banned from industrial production. The most exciting discovery in the last 10 years in UTUC has been the unraveling of aristolochic acid as an iatrogenic global risk factor for UTUC development.

Balkan endemic nephropathy and Chinese herbs nephropathy, which are the same disease, are specifically related to UTUC. They are characterized by a mutation of the p53 gene as a consequence of the exposure to aristolochic acid, a potent carcinogen derived from Aristolochia plants, which is used as an herbal ingredient or remedy. A significant use of this plant has also been registered in Taiwan, where the incidence of UTUC is estimated to be approximately 20% to 25% of all urothelial cancers, the highest worldwide. Unfortunately, clinical trials targeting these selected populations are lacking and further efforts are required to lower the exposure to this deadly substance, which is integral in many traditional Chinese and Indian pharmacopoeia.

Rare cases of UTUC linked to hereditary nonpolyposis colorectal carcinoma (HNPCC, Lynch syndrome) have also been reported. UTUC patients younger than 60 years, with a personal history of an HNPCC-associated cancer, a first-degree relative younger than 50 years of age with HNPCC-associated cancer, or 2 first-degree relatives with HNPCC-associated cancer should be screened for hereditary cancers.

Symptoms and Screening

UTUC are mainly silent in the first stages, diminishing the chance of an early diagnosis. Principal symptoms of UTUC are macrohematuria or microhematuria, followed by flank pain and lumbar mass. Systemic symptoms, such as fatigue, loss of weight, and fever, may suggest the presence of a metastatic disease when associated with UTUC.

Diagnosis

The diagnosis and evaluation of a patient with a suspected UTUC requires an accurate investigation, combining imaging and endoscopy, to better stage the patient and to carefully select for individualized therapy.

Computed tomography (CT) urography is nowadays the first-line imaging examination used, because of a higher ability to identify small lesions with good anatomic characterization, compared with intravenous excretory urography and ultrasonography. The sensitivity and specificity of CT urography to detect UTUC are continuously improving and are estimated to be in the range of 67% to 100% and 93% to 99%, respectively. Radiation exposure, allergic reactions, or nephropathy after contrast medium employment need to be considered before performing this examination. Magnetic resonance urography remains a valid alternative for some patients presenting with a contraindication to iodinated contrast material, with an estimated 74% rate of detection for lesions less than 2 cm. Retrograde ureterorenopyelography through a ureteral catheter or during ureteroscopy (URS) remains an option to be considered to investigate UTUC, but it does not allow the prediction of tumor grade or stage.

Positive urinary cytology may be the first sign of UTUC in the case of normal cystoscopy and after excluding carcinoma in situ (CIS) of the bladder and prostatic urethra. Because cytology of UTUC is less sensitive compared with that of UCB cytology, it is recommended to perform it in situ before the contrast injection in case of retrograde ureteropyelography.

Flexible URS associated with biopsy is an important diagnostic tool to evaluate patients with suspected UTUC. A reduction of misdiagnoses from 15.5% to 2.1% has been estimated when routine URS was performed ( P = .021). In addition to diagnosis, URS allows visual and pathologic evaluation of the tumor, thereby helping in clinical decision-making regarding treatment. It should be especially applied preoperatively in uncertain cases when radical surgery could be avoided or in the case of solitary kidney. Basket device biopsies seem to have a higher rate of successful diagnosis with a more precise definition of grade compared with other biopsy devices. Complications associated with URS, such as urinary infection or iatrogenic lesion of the urinary upper tract, however, need to be taken into consideration. Finally, despite the delay between diagnosis and radical surgery, diagnostic URS does not seem to significantly affect oncologic outcomes. URS allows not only diagnosis, collection of information that imposes staging, and prognostication but also therapy in those patients benefitting from conservative therapy.

Epidemiology

Urothelial cell carcinoma (UCC) is the fourth most common cancer worldwide, after prostate (or breast), lung, and colorectal cancer. Among UCC, upper tract urothelial carcinoma (UTUC) is rare and accounts for 5% to 10% of all urothelial carcinomas with an annual incidence of 2 cases per 100,000 inhabitants in Western countries. UTUC is more common in men than in women with a male-to-female ratio of 2:1 and the mean age at the diagnosis of 65 years old. It is located twice more often in the renal pelvis than in the ureter and in about 20% of the cases a concomitant urothelial carcinoma of the bladder (UCB) is present. UTUC tends to recur in the bladder or progress through the lymphatic and vascular systems to distant organs. To date, much of the decision-making in UTUC comes from knowledge acquired in UCB because of its relative rarity. Over the last 5 years, a large body of evidence has proven that UTUC and UCB have common features but differ as well in a significant degree.

Risk Factors

Tobacco and occupational exposure are the main UTUC exogenous risk factors in most countries. The relative risk of developing UTUC in the case of tobacco exposure has been estimated to be 2.5 to 7, depending on the number of years of exposure and the number of cigarettes smoked per day. Occupational exposure to certain aromatic amines has an estimated risk of 8.3. However, exposure risk decreased since the 1960s because certain chemical substances, such as benzidine and β-naphthalene, have been banned from industrial production. The most exciting discovery in the last 10 years in UTUC has been the unraveling of aristolochic acid as an iatrogenic global risk factor for UTUC development.

Balkan endemic nephropathy and Chinese herbs nephropathy, which are the same disease, are specifically related to UTUC. They are characterized by a mutation of the p53 gene as a consequence of the exposure to aristolochic acid, a potent carcinogen derived from Aristolochia plants, which is used as an herbal ingredient or remedy. A significant use of this plant has also been registered in Taiwan, where the incidence of UTUC is estimated to be approximately 20% to 25% of all urothelial cancers, the highest worldwide. Unfortunately, clinical trials targeting these selected populations are lacking and further efforts are required to lower the exposure to this deadly substance, which is integral in many traditional Chinese and Indian pharmacopoeia.

Rare cases of UTUC linked to hereditary nonpolyposis colorectal carcinoma (HNPCC, Lynch syndrome) have also been reported. UTUC patients younger than 60 years, with a personal history of an HNPCC-associated cancer, a first-degree relative younger than 50 years of age with HNPCC-associated cancer, or 2 first-degree relatives with HNPCC-associated cancer should be screened for hereditary cancers.

Symptoms and Screening

UTUC are mainly silent in the first stages, diminishing the chance of an early diagnosis. Principal symptoms of UTUC are macrohematuria or microhematuria, followed by flank pain and lumbar mass. Systemic symptoms, such as fatigue, loss of weight, and fever, may suggest the presence of a metastatic disease when associated with UTUC.

Diagnosis

The diagnosis and evaluation of a patient with a suspected UTUC requires an accurate investigation, combining imaging and endoscopy, to better stage the patient and to carefully select for individualized therapy.

Computed tomography (CT) urography is nowadays the first-line imaging examination used, because of a higher ability to identify small lesions with good anatomic characterization, compared with intravenous excretory urography and ultrasonography. The sensitivity and specificity of CT urography to detect UTUC are continuously improving and are estimated to be in the range of 67% to 100% and 93% to 99%, respectively. Radiation exposure, allergic reactions, or nephropathy after contrast medium employment need to be considered before performing this examination. Magnetic resonance urography remains a valid alternative for some patients presenting with a contraindication to iodinated contrast material, with an estimated 74% rate of detection for lesions less than 2 cm. Retrograde ureterorenopyelography through a ureteral catheter or during ureteroscopy (URS) remains an option to be considered to investigate UTUC, but it does not allow the prediction of tumor grade or stage.

Positive urinary cytology may be the first sign of UTUC in the case of normal cystoscopy and after excluding carcinoma in situ (CIS) of the bladder and prostatic urethra. Because cytology of UTUC is less sensitive compared with that of UCB cytology, it is recommended to perform it in situ before the contrast injection in case of retrograde ureteropyelography.

Flexible URS associated with biopsy is an important diagnostic tool to evaluate patients with suspected UTUC. A reduction of misdiagnoses from 15.5% to 2.1% has been estimated when routine URS was performed ( P = .021). In addition to diagnosis, URS allows visual and pathologic evaluation of the tumor, thereby helping in clinical decision-making regarding treatment. It should be especially applied preoperatively in uncertain cases when radical surgery could be avoided or in the case of solitary kidney. Basket device biopsies seem to have a higher rate of successful diagnosis with a more precise definition of grade compared with other biopsy devices. Complications associated with URS, such as urinary infection or iatrogenic lesion of the urinary upper tract, however, need to be taken into consideration. Finally, despite the delay between diagnosis and radical surgery, diagnostic URS does not seem to significantly affect oncologic outcomes. URS allows not only diagnosis, collection of information that imposes staging, and prognostication but also therapy in those patients benefitting from conservative therapy.

Patient Characteristics

Female patients tend to have more advanced pathologic T stage and higher-grade tumor than male patients. However, gender alone is no longer considered to be an independent prognostic factor of CSS. Also, ethnicity does not directly influence UTUC survival, even though some data show different tumor characteristics between Caucasians and Asians. Older patients diagnosed with UTUC tend to have worse survival but many of them are still cured with radical surgical treatment, suggesting that age alone is not sufficient to influence clinical decision.

Tumor Characteristics

Tumor stage and grade are the most powerful prognostic factors influencing UTUC outcome. The 5-year cancer-related survival in UTUC patients is higher than 90% for pTa/pT1/CIS tumors, less than 50% for pT2/pT3 tumors, and less than 10% for pT4 tumors. The impact of positive lymph nodes, number of positive nodes, and extranodal extension are also well-established prognostic factors. It has been estimated that the incidence of positive lymph nodes is approximately 0%, 5%, 24%, and 84% in patients with pTa/pT1/CIS, pT2, pT3, and pT4 UTUC, respectively.

Tumor multifocality should be considered an independent prognostic factor of RFS and CSS in patients with organ-confined UTUC treated with radical nephroureterectomy (RNU). In the case of a single lesion, ureteral location tends to have a worse oncologic outcome compared with renal pelvis or calice location. As already shown in other types of cancers, lymphovascular invasion (LVI) is a strong independent prognostic factor for CSS and metastasis-free survival in UTUC patients, in particular when a lymph node dissection (LND) is not performed or in those without lymph node metastasis. It has been estimated that LVI is present in approximately 20% of RNU samples, and its inclusion in the pathologic report is strongly recommended. Finally, extensive tumor necrosis, tumor size 3 cm or greater, sessile tumor architecture, and hydronephrosis also negatively influence UTUC prognosis.

Genetic and Biological Markers

Genetic and epigenetic biomarkers, based on tissue, urine, and blood, have achieved good sensitivities and specificities for UTUC diagnosis in pilot studies, but none has so far been validated. Their efficacy on large patient cohorts or for follow-up of UTUC patients to detect early recurrences still needs to be proven.

Interestingly, some studies have determined some specific genetic and epigenetic differences between UTUC and UCB. The micro-RNA expression pattern seems to play a role in progressing and nonprogressing UTUC and a combination of new micro-RNAs may help to distinguish UTUC with a high probability of recurrence, improving patients’ follow-up and CSS. Among the oncofetal proteins, insulinlike growth factor messenger RNA binding proteins 3 and 5 have shown interesting results as independent factors associated with RFS, CCS, and all-cause mortality. Few data are available on DNA methylation and UTUC detection and prognosis. GDF15, TMEFF2, VIM promoter methylation, GATA binding protein 3, KI-67, Bcl-xL, calreticulin, and annexin A2 and A3 are other valuable and promising diagnostic and prognostic biomarkers, but robust clinical data are still lacking.

Increasing evidence suggests that host inflammatory response and procoagulant and fibrinolytic factors are stage-independent and grade-independent predictors of survival in patients with various solid tumors, including UTUC. Elevated preoperative neutrophil-lymphocyte ratio and erythrocyte sedimentation rate were independent risk factors of disease recurrence and cancer-specific mortality in UTUC patients treated with RNU. Hemoglobin seems to be a promising marker as predictor of RFS and CSS in UTUC patients, being associated with aggressive tumor features, especially if a perioperative blood transfusion is required. Finally, C-reactive protein, sodium, and albumin also may play a role for prognostication of UTUC patients treated with RNU.

Taken together, these findings advocate testing of biomarkers in correlative clinical trials to validate their utility in UTUC management.

Conservative Surgery

Although no randomized trials comparing UTUC conservative treatment with RNU are available nowadays, the nephron-sparing approach may be a valid alternative, especially in the case of renal insufficiency, solitary functional kidney, bilateral disease, or low-risk tumors. Given the frequent bladder and ipsilateral upper tract recurrences, strict follow-up is required and, therefore, patient selection is crucial. Despite a better protection of kidney function, endoscopic treatment of UTUC has a rate of local and bladder recurrence of approximately 25% and 15%, respectively. Moreover, a change in grade and/or stage from the diagnostic biopsy to the endoscopic resection may occur in more than one-third of UTUC patients managed conservatively, with a considerable risk of undergrading and understaging. The percutaneous approach in UTUC management could be considered for low-grade tumors in the lower caliceal system, even though only small case series have been reported and technical improvements in new flexible ureteroscopes are replacing this procedure.

UTUC segmental resection has the main advantage of preserving renal function and conserving a good oncologic control, but should be reserved only for selected cases. Complete distal ureterectomy and neocystostomy are a valid and oncologically safe alternative for large-volume noninvasive low-grade UTUC located in the distal ureter, which are not possible to manage endoscopically. Ureteroureterostomy is a nonstandard option for proximal ureter or mid-ureter UTUC not completely resectable by the endoscopic approach in patients in whom kidney function preservation is imperative. Segmental resection of the iliac and lumbar ureter and open resection of the renal pelvis tumor are hardly performed anymore because of the high risk of tumor spillages, complications, and recurrences, and the improved endoscopic tools.

Adjuvant Topical Agents

Despite a large experience on intravesical therapies in noninvasive UCB, only a few reports on the efficacy of instillation with chemotherapy or immunotherapy for UTUC treated with conservative intent have been published so far. Instillation of topical agents is technically feasible anterogradely by means of a percutaneous nephrostomy or retrogradely through a single J or double JJ ureteric stent. To avoid complications, intracavitary pressure should be maintained low and absence of perforation or obstruction of the system needs to be proven. Although administration of bacillus Calmette-Guérin (BCG) or mitomycin C seems to be safe, its efficacy is still debated.

Patients with CIS UTUC treated with a curative intent by antegrade perfusions of BCG tend to have a better RFS, progression-free survival, and RNU-free survival compared with those treated with adjuvant intention after endoscopic resection of pTa/T1 tumors. It is now becoming widely accepted that BCG is mainly indicated in CIS patients. Finally, it is difficult to obtain significant evidence on mitomycin C given that there are only a few studies available describing its use. Another type of adjuvant therapy is single-dose intravesical mitomycin C after RNU, and now, without evidence-based proof yet, after conservative endoscopic management of UTUC.

Follow-up of Bladder After Conservative Therapies

UTUC patients treated with conservative therapies need vigilant surveillance of the bladder and the ipsilateral upper tract. Currently, a combination of CT urography, ureteroscopy, cystoscopy, urinary cytology, and cytology in situ seems to be the most effective method. No consensus exists regarding the exact follow-up schedule, but at least routinely controls during the first 5 years are advisable.

Radical Nephroureterectomy

The gold-standard treatment of localized high-risk UTUC is an RNU, which consists of en block removal of the kidney, renal pelvis, extramural and intramural ureter, ureteral orifice, and a cuff of bladder around the ureteral orifice. In patients with a neobladder or urinary diversion (eg, from bladder cancer progression), resection of the ureteral anastomosis and cuff of bowel should be performed instead. From an oncologic standpoint, it is imperative to (1) avoid entry into the urinary tract to prevent tumor spillage or seeding and (2) completely excise the ureter distal to the tumor and bladder cuff because of a high risk of recurrence arising in the remnant ureteral tissue translating into worse overall survival.

Various surgical techniques for managing the distal ureter exist. The conventional approaches of bladder cuff removal via open surgery through the transvesical or extravesical approach appear similar in terms of oncologic outcomes. The “pluck” technique was introduced by McDonald and colleagues in 1952 but was not popularized until 1995 when this endoscopic approach to the distal ureter was found to have equivalent oncologic outcomes and reduced operative time. These 3 competing methods of bladder cuff excision (transvesical, extravesical, and endoscopic) were recently evaluated by the UTUC Collaboration in a large study of 2681 patients, and no differences in terms of RFS, CSS, and overall survival were found. However, patients who underwent the endoscopic approach were at significantly higher risk of intravesical recurrence compared with those who underwent the transvesical ( P = .02) or extravesical approaches ( P = .02), with no differences between the latter 2 groups. Regardless of technique chosen to manage the distal cuff, the underlying principle and consensus remains to avoid tumor spillage in the peritoneal cavity.

Overall, oncologic outcomes after RNU remain relatively poor. The UTUC Collaboration reported results from 1363 patients treated with RNU at 12 academic centers and found that 5-year RFS and CSS probabilities were 69% and 73%, respectively. Furthermore, a study from MD Anderson Cancer Center showed that across the study period from 1986 to 2004, disease-specific survival rates were unchanged, highlighting that current treatment paradigms may need to be augmented with multimodal therapy to improve outcomes.

Lymphadenectomy

Regional lymphadenectomy or LND allows for optimal staging of disease. Numerous retrospective studies have found a survival benefit in patients with greater or equal to a pT2 disease. Because preoperative clinical staging is challenging, frequently inaccurate, some advocate performing LND for all patients undergoing RNU. The template for LND depends on tumor location. Based on a comprehensive mapping study, Kondo and colleagues have proposed the following: for right-sided renal pelvic and midureter and upper-ureteral tumors, LND should include the hilar nodes, the paracaval, inter-aortocaval, and retrocaval lymph nodes, from the level of the renal hilus to the aortic bifurcation. For left-sided renal pelvic tumors and mid ureters and upper ureters, dissection should include para-aortic and hilar nodes from the renal hilus to the aortic bifurcation. The ipsilateral pelvic lymph nodes along the common, external, internal iliac, and obturator vessels should also be retrieved. A large multicenter retrospective study of 552 patients found that in the subgroup of patients with pN0 disease (n = 412), removal of at least 8 lymph nodes was a strong independent predictor of cancer-specific mortality (hazard ratio [HR]: 0.42, P = .004). The benefits of LND lie in improved staging and local control as well as reduced risk of locoregional nodal relapse. Therefore, although randomized trials are not available to clarify its role on overall survival, it is recommended when technically feasible.

Type of Surgical Approach

Most RNUs are performed via an open approach. A population-based study using the Nationwide Inpatient Sample between 1998 and 2009 found that 90.8% of patients with nonmetastatic UTUC underwent open RNU, whereas the rest underwent a minimally invasive approach. As the urologic community gains experience with minimally invasive abdominal and pelvic surgery, the utilization of laparoscopic and robotic RNU is expected to increase correspondingly. Although peritoneal dissemination and port-site metastases are possible and have been reported, these represent rare occurrences and have not deterred the adoption of the minimally invasive approach for RNU.

The available data suggest comparable oncologic outcomes following either open or minimally invasive RNU. The only prospective randomized study examining this by Simone and colleagues included 80 patients who underwent either open or laparoscopic RNU. The authors failed to identify a significant difference in 5-year CSS (open: 89.9% vs laparoscopic: 79.8%, P = .2). A recent systematic review and meta-analysis of 21 nonrandomized studies with 1235 cases and 3093 controls examining oncologic outcomes of open versus laparoscopic RNU concluded that short-term to mid-term oncologic outcomes appear similar between open and laparoscopic RNU. At this time, long-term results are sparse, especially for high-risk disease (eg, pT2, N+), with oncologic equivalence between open and laparoscopic RNU reported in the longest median follow-up of 13.7 years.

Laparoscopic RNU is associated with improved perioperative outcomes. The aforementioned randomized trial found shorter hospital length of stay and decreased blood loss among those receiving laparoscopic compared with open RNU. A recent population-based study similarly reported that laparoscopic RNU was associated with fewer adverse intraoperative and postoperative outcomes compared with open RNU. Robot-assisted laparoscopic RNU remains in its infancy, with only 10 studies reporting their initial experience, demonstrating feasibility and safety but lacking data at this time to confirm equivalent perioperative and long-term oncologic outcomes.

Neoadjuvant and Adjuvant Chemotherapy

Compared with UCB, there has been less robust evidence in terms of randomized clinical trials to clarify the role of perioperative chemotherapy in UTUC. Neoadjuvant chemotherapy is a promising and emerging approach for UTUC patients, particularly those with good renal function and histologic evidence of high-grade muscle-invasive disease. Pooled results from 2 retrospective studies found a 59% benefit in disease-specific survival (HR 0.41, P = .005). Two prospective single-center phase 2 trials accrued patients with locally advanced urothelial cancers, including UTUC, and reported substantial pathologic downstaging (to inferior or equal to pT1N0) in up to 75% of patients. Two ongoing trials evaluating the efficacy of neoadjuvant gemcitabine and cisplatin ( NCT01663285 , NCT01261728 ) are highly anticipated because positive results may shift the paradigm in the management of this disease.

There is a growing body of evidence supporting the use of adjuvant chemotherapy. Adjuvant chemotherapy has strengths apparent to most: (1) accurate pathologic staging is available postoperatively; (2) any subclinical metastases can be eradicated to maximize the patient’s overall and disease-specific survival. However, the downside is that a considerable proportion of patients will develop postoperative renal insufficiency potentially preventing the use of systemic therapies. Although cisplatin-based chemotherapy seems to be the most efficacious regimen to be administered from available evidence, it is also rather nephrotoxic, thereby limiting its utility. The latest meta-analysis investigating the role of adjuvant chemotherapy found that a pooled overall survival benefit (HR = 0.43, 95% CI: 0.21–0.89, P = .023), across 3 cisplatin-based studies, represents a 57% benefit in OS among those treated with adjuvant chemotherapy compared with those who received surgery alone.

Radiotherapy

The aim of radiotherapy is to optimize local control of disease. This treatment targets the renal fossa, course of the ureter to the entire bladder, and the paracaval and para-aortic lymph nodes at risk of harboring micrometastatic or metastatic disease. Radiotherapy may be beneficial in delaying bladder tumor relapse and improving overall survival in both adjuvant and salvage settings. There may be prolonged overall and disease-free survival when administered concurrently with cisplatin-based chemotherapy and a prescribed dose of 50 Gy or greater. However, the systemic nature of UTUC is highlighted by the fact that 2 studies examining adjuvant radiotherapy have not found an OS benefit.