Study
Number of patients
Median follow-up (months)
Cohort
UTR prevalence
Median time to UTR (months)
UTR risk factors
UTR location
UTR characteristics
Solsanaa et al. [10]
138
86
NIMBC with CIS
24.6 %
CIS: 21.2 %
No CIS: 2.3 %
38
CIS: 38 (0–56) [mean]
No CIS: 44 (3–122) [mean]
Involvement of prostatic urethra
12 % renal pelvis
88 % distal ureter
32 % bilateral
68 % non-invasive
20 % invasive
12 % unknown
Herr [12]
307
146 (120–216)
NIMBC treated with BCG
25 %
56 (12–181)
Retained bladder
44 % renal pelvis
56 % ureter
23 % non-invasive
77 % invasive
Hurle et al. [13]
591
80–92 (27–143)
All NIMBC
4 %
Low risk: 0.9 %
Intermediate risk: 2.2 %
High risk: 9.8 %
80 (16–230)
Low risk: 80, 91
Intermediate risk: 52 (16–109)
High risk: 49 (16–107)
Grade
Stage
CIS
Multifocality
Recurrent
IVC failure
48 % renal pelvis
36 % distal ureter
16 % multifocal
60 % non-invasive
40 % invasive
Millan-Rodriguez et al. [14]
1,529
All NIMBC
2.6 %
Low risk: 0.6 %
Intermediate risk: 1.8 %
High risk: 4.1 %
Low risk: 122
Intermediate risk: 47 (27–67)
High risk: 39 (10–68)
Grade
Stage
CIS
Multifocality
41 % renal pelvis
41 % ureter
18 % multifocal
47 % non-invasive
33 % invasive
20 % unknown
Canales et al. [15]
375
58 (14–176)
Ta only
3.4 %
Low risk:1.8 %
High risk: 8.9 %
22 months [mean]
Low risk: 43 [mean]
High risk: 24 [mean]
≥2 prior intravesical recurrences
<12 months between intravesical recurrences
77 % renal pelvis
23 % ureter
77 % non-invasive
23 % invasive
Wright et al. [20]
99,388
All bladder cancer (85 % NIMBC)
0.8 %
33
Grade
CIS
Location
Cystectomy
45 % renal pelvis
54 % ureter
55–59 % localized
40–51 % poorly differentiated
Sternberg et al. [16]
935
66
All NIMBC
5.4 %
≤60 months: 57 % recurred
>60 months: 43 % recurred
Stage
CIS
Risk Factors
Patients with T1 NIMBC had nearly double the risk of UTR at 10 years compared to patients with Ta, and patients with at least two prior NIMBC recurrences or an duration of less than 12 months between recurrences had a fivefold higher rate of UTR than those without either (8.9 % vs. 1.8 %) [15, 16]. Other potential risks include vesicoureteral reflux and occupational exposure, although their relative importance is less certain [17–19].
One study stratified patients into low, intermediate and high risk groups according to NIMBC grade, stage, multifocality and presence of CIS, and found UTRs in 0.6, 1.8 and 4.1 % of the patients in each respective risk group [14]. All UTRs occurred in patients who had at least one prior intravesical recurrence, and bladder tumor multifocality tripled the risk. A similar study estimated risk of UTR according to several clinical risk factors [13]. Patients with primary, solitary and low grade NIMBC had a 0.9 % prevalence of UTR, while those with recurrent or multifocal tumors had a 2.2 % prevalence. The highest risk group had a UTR prevalence of 9.8 % and included patients with CIS or high grade tumors and those who failed prior intravesical chemotherapy.
Patients with NIMBC and associated CIS, and those treated intravesical bacillus Calmette Guérin (BCG) are a particurlarly high risk group with up to a 25 % risk of UTR [10, 12]. At 5, 10 and 15 years, patients treated with BCG had a 13, 28 and 38 % cumulative risk of UTR and patients with CIS may have a ten-fold increased risk as compared to patients without CIS. Patients with bladder CIS are also more likely to develop bilateral UTRs, and those with disease in the prostatic urethra have particularly poor outcomes [10].
In an analysis of nearly 100,000 bladder cancer patients from the SEER registry, of whom 85 % had NIMBC, 0.8 % were diagnosed with an UTR [20]. Upper tract recurrences were more common in patients with NIMBC, and risk factors included the presence of bladder CIS, higher grade bladder tumors and bladder tumors located at the ureteral orifice, trigone or bladder neck.
Upper tract recurrences are relatively late events, thus length of follow-up is strongly influential on the risk of developing an UTR. The prevalence of UTR in high risk patients is less than 10 % after 4 years, but may be at least 25 % after 12 years [12–14]. Approximately half of the UTRs occur more than 5 years after NIMBC diagnosis, and some can occur as late as 15 years [12, 16]. Given the length to time needed to develop UTRs and because UTRs typically occur in patients who have not experienced a competing event, such as metastatsis or death, the longer patients are followed in absence of a competing event, the more likely they are to experience an UTR. This explains the seemingly paradoxical observation that patients with invasive bladder tumors and those who have radical cystectomy (RC) have lower rates of UTR [12, 20]. These patients have a higher risk of competing events due to thier aggressive disease, and fewer are at-risk for developing an UTR.
Outcomes
While characteristics of UTRs vary across studies, patients with higher risk bladder tumors tend to develop more aggressive upper tract tumors (Table 10.1). Nearly 80 % of UTRs were muscle invasive in patients with high risk NIMBC, while only 23–32 % were invasive in patients with lower risk tumors [12–15]. From 60 to 96 % of patients with UTRs require aggressive treatment with nephroureterectomy or segmental ureterectomy, although endoscopic treatment can be an option for some [13–15]. Although some have suggested no difference in survival for patients with versus without UTRs [10], 20–32 % will ultimately die from metastastatic disease [12, 13, 15]. Those who die are generally patients with high risk NIMBC and high grade UTRs.
Surveillance After NIMBC Treatment
There is conflicting evidence about whether UTRs are more often diagnosed on surveillance studies or due to the onset of symptoms, however there is no evidence that detecting an asymptomatic UTR translates to better patient outcomes [13, 15, 16]. One series found that approximately two-thirds of UTRs were detected symptomatically and only a quarter were diagnosed asymptomatically on surveillance imaging [16]. CT urography was used most commonly for upper tract surveillance, but proved unnecessary for most patients as the vast majority of UTRs could have been diagnosed with ultrasound or urine cytology. In fact, only 3 out of 51 (6 %) UTRs would have been missed if surveillance CT urography had not been routinely performed on asymptomatic patients.
All professional guidelines recommend a risk stratified approach to upper tract surveillance in asymptomatic patients with a history fo NIMBC (Table 10.2) [5–7, 21]. Because half of all recurrences are detected 5 years after NIMBC diagnosis and this risk does not decrease over time, long-term surveillance in high risk patients is necessary [10–12, 15, 16, 22]. The most common imaging modality is CT urography, which effectively images the renal parenchyma and upper urinary tracts. Other modalities include intravenous pyelography, MR urography, renal ultrasound with retrograde pyelogram and ureteroscopy. Patients with symptoms or positive urinary cytology in the absence of cystoscopic findings should receive upper tract imaging.
Table 10.2
Professional guideline recommendations for upper tract surveillance in asymptomatic patients after NIMBC treatment
NCCN [6] | EAU [5] | ||||
|---|---|---|---|---|---|
Risk group | Upper tract Imaginga | Risk groupb | Upper tract Imaging | Risk groupc | Upper tract Imagingd |
Ta LG | None | Low risk | None | Low risk | None |
Ta HG | Consider every 1–2 years | Intermediate risk | Consider | Intermediate risk | None |
T1 LG | None | High risk | Annually for 2 years, then lengthen interval | High risk | Annually |
T1 HG | Consider every 1–2 years | ||||
Any Tis | Consider every 1–2 years | ||||
10.2.2 Upper Tract Recurrences After Radical Cystectomy
Up to 50 % of patients experience recurrence of urothelial carcinoma following RC, most commonly within 2 years of surgery at metastatic sites within the abdomen and pelvis, chest and bone [23–25]. If patients do not recurr during this early period, there is still an 8 % risk of disease recurrence, most commonly in the upper urinary tracts [24]. Most contemporary series report the risk of UTR following RC between 2 and 6 % with a time to recurrence of 2–3 years (Table 10.3) [10, 25–34]. Over half of UTRs occur in the renal pelvis, up to a quarter are multifocal, and recurrences at the ureteroileal anastomosis are rare [27–31, 33].
Table 10.3
Upper tract recurrence following radical cystectomy
Study | Number of patients | Median follow-up (months) | UTR prevalence | Median time to UTR (months) | UTR risk factors |
|---|---|---|---|---|---|
Kenworthy et al. [28] | 430 | 2.6 % | 40 (9–56) | Distal ureteral involvement | |
Solsana et al. [10] | 225 | CIS: 17.4 % No CIS: 3.9 % | CIS: 18 No CIS: 28 | CIS | |
Sved et al. [30] | 235 | 42 [mean] | 2 % | 40 (16–60) [mean] | Prostatic urethral involvement |
Sanderson et al. [29] | 1,069 | 124 (4–222) | 2.5 % | 40 (5–112) | Superficial urethral involvement |
Tran et al. [32] | 1,329 | 38 (17–70) | 6 % | 25 (1–107) | Distal ureteral involvement |
Furukawa et al. [27] | 583 | 42 (7–77) | 2.1 % | 30 (5–71) | |
Meissner et al. [41] | 322 | 4.7 % | 31 (12–72) | ||
Volkmer et al. [33] | 1,420 | 1.8 % | 39 (4–142) | CIS Recurrent bladder tumor RC for NIMBC Distal ureteral involvement | |
Umbreit et al. [25] | 1,388 | 172 | 4.8 % | 37 (2–174) | Multifocality Positive ureteral margin pT4 Gross hematuria |
Takayanagi et al. [31] | 362 | 48 (0–214) | 3 % | 48 (12–79) | CIS Positive ureteral margin Urethral involvement |
Perlis et al. [34] | 574 | 45 | 4 % | 28 (8–96) |
Risk Factors
Approximately 10 % of RC specimens demonstrate involvement of the distal ureter with carcinoma on final pathology, and this finding is associated with a two to sixfold increased risk of UTR [25, 28, 32, 33]. On one study, 16 % of RC patients with ureteral involvement were diagnosed with UTR a 5 years comapred to 5 % without ureteral involvement [32].
The male and female urethra can be involved with urothelial carcinoma in 15 % and 63 % of RC specimens, respectively [29]. Patients with superficial urethral involvement have a higher 10 year risk of UTR compared to those without it (15–19 % vs. 3–4 %), and some suggest that urethral involvement is a stronger predictor of UTR than ureteral involvement [29, 31]. As expected, prostatic stromal invovement is not consistently associated with an increased risk of UTR, as these patients are more likely to experience cancer-speficic mortality and few are at risk to develop UTR [29].
Bladder tumor characteristics that reflect the presence of aggressive, recurrent or refractory disease increase the risk of UTR. Multifocality and a history of recurrent NIMBC were associated with a two to three times increased risk of UTR [25, 33], and RC patients with NIMBC have a higher risk of UTR than RC patients with invasive disease [33]. Bladder CIS is identified up to 50 % of RC specimens and is associated with a two to sixfold increased risk of UTR [10, 29, 31–33]. Interestingly, bladder CIS triples the risk of ureteral involvement [32].
A recent meta-analysis of over 13,000 patients reported a 0.8–6.4 % prevalence of UTR after RC [35]. On multivariable analysis, lower tumor grade, RC for NIMBC, bladder CIS, positive ureteral margin, positive urethral margin, a history of UTUC and pathologically negative lymph nodes were indepdendently associated with an increased risk of UTR. These results mirror those from individual studies and suggest that bladder tumor characteristics associated with locally aggressive and recurrent disease, but not lethal disease, predict higher rates of UTR.
Similar to UTRs after NIMBC, one of the strongest risk factors for UTR after RC is length of follow-up. As UTR is a relatively late event, patients who develop metastatic disease in the first years following RC usually die and are not at risk to develop UTR. Therefore, RC patients with longer follow-up who do not have a competing event are at an increased risk of UTR. Importantly, this risk of does not decrease over time [22, 32].
Use of Ureteral Frozen Section
Because ureteral involvement is a strong risk factor for UTR, many surgeons routinely send an intraoperative frozen section examination (FSE) of the disetal ureter and perform a stepwise resection in effort to attain a negative margin. Proponents of this practice accept the need to perform a nephroureterectomy if necessary [36]. The utility of routine FSE was addressed in a large RC series where half of the patients had routine intraoperative ureteral FSE [37]. The sensitivty and specificity of FSE to detect ureteral involvement on final pathology was 75 % and 99 %, respectively, and a positive margin on final pathology nearly tripled the risk of UTR. However, most patients with postive FSE did not have enough ureteral length to warrant further step-sectioning, and those who did were infrequently converted to a negative margin on subsequent FSE. Ultimately, conversion from a positve FSE to a negative margin on final pathology did not eliminate the risk of UTR, and ureteral involvement was not associated with overall survival.
Other studies agree that FSE is accurate at detecting ureteral involvement and question its utility given the lack of evidence that ureteral involvement impacts survival, although it remains a topic of debate [36, 38–40]. A different RC series demonstrated that 82.6 % of 178 positive FSEs could be converted to a negative margin on final pathology with step-sectioning [39]. While ureteral involvement was not related to survival, and all patients with a positive FSE had an increased risk of UTR, those with a positive FSE who could be converted to a negative final margin had a slightly lower risk of UTR (HR 4.4, 95 % CI 2.6,7.4) than those who could not be converted to a negative final margin (HR 7.4, 95 % CI 4.3, 16.4).
Because of the pagetoid growth pattern of urothelial carcinoma, a negative FSE cannot rule out proximal upper tract involvement, and patients with a unilateral positive FSM can have a contralateral UTR [39]. With routine transection of the ureters at the common iliac artery, as few as 1.2 % of patients will have a positive ureteral margin on final pathology [38]. While FSE can identify ureteral involvement and may reduce the risk of UTR, it is an added expense that predicts a rare event and does not improve survival. At present, there is no strong evidence to support use of routine ureteral FSE. It is our practice to not send routine FSEs, but instead to perform high transection of the ureters with close monitoring of patients who have ureteral involvement on final pathology.
Outcomes
Upper tract recurrence following RC tends to be aggressive and have poor prognosis. Most patients with UTRs present with advanced stage and lymph node postive tumors, and as many as half present with metastatic disease [26, 27, 29, 31, 33]. Approximately 70 % of patients with UTRs after RC will die from urothelial carcinoma, and median survival is close to 1 year with less than 30 % alive at 5 years [26, 27, 29–31, 33]
Surveillance After RC
Surveillance regimens following RC include frequent office visits, cross-sectional imaging and urine cytology in the first 2–3 years and then at least annually thereafter. Still, some question the value of upper tract surveillance since 40–80 % of UTRs present with a sign or symptom; most commonly gross hematuria, but also flank pain, renal failure, infection and weight loss [26, 28–31, 33]. Importantly, there is no obvious improvement in outcomes for patients who present with an asymptomatic versus symptomatic UTRs.
In one series of 15 UTRs after RC, half were diagnosed due to symptoms and there was no difference in survival between patients with an asymptomatic UTR found on surveillance and a symptomatic UTR [41]. Of the 1,064 surveillance intravenous pyelograms performed, only eight (0.75 %) were abnormal and led to the diagnosis of an asymptomatic UTR. Another study observed that while patients with asymptomatic distant recurrences tended to have improved survival compared to those with symptomatic distant recurrences, this was not seen for patients with urothelial recurrences, including UTRs [42]. Still, patients with UTRs detected asymptomatically on routine surveillance may have better survival compared to those who presented symptomatically (1.6 vs. 3.7 years), although this difference was not significant (p = 0.6) [29].
Urinary biomarkers are also used to monitor for urothelial recurrence. Since the majority of UTRs are high grade, urinary cytology is theoretically a useful test. The sensitivity of urinary cytology for detecting UTRs ranges from 40 % to 100 %, but as many as 90 % of positive cytologies are falsely positive [41, 43]. In a cohort 278 RC patients, only one out of nearly 500 surveillance urine cyologies helped diagnose a UTR that would not have been otherwise detected due to symptoms or abnormal imaging [43]. Fluoroesence in situ hybridization has a 86 % sensitivity and 87 % specificity for detecting UTRs after RC, with a similarly low positive predictive value as cytology [43].
Both urinary cytology and upper tract imaging have low yields for detecting UTR when performed on asymptomatic patients, with 2,000 and 800 patients needed to be screened with cytology and upper tract imaging, respectively, to detect one UTR [35]. While routine cross-sectional imaging may be useful to detect distant recurrences and other postoperative complications, it is currently unknown whether diagnosing an asymptomatic UTR has a more favorible prognosis than one detected symptomatically [44].
A risk-stratified surveillance strategy may improve yield and reduce unnecessary tests. Considering the presence of CIS, recurrent bladder cancer, RC for NIMBC and distal ureteral involvement, patients who had 0, 1–2 and 3–4 of these findings had a 15 year UTR risk of 0.8 %, 8.2 % and 13.1 %, respectively [33]. If all RC patients were imaged annually for 5 years, the chance of diagnosing a UTR in patients with 0, 1–2 and 3–4 risk factors was 1:432, 1:93 and 1:53, respectively. Similarly, when stratifying patients based on the presence of either urethral involvement or bladder CIS, the 5-year prevalence of UTR for patients with either risk present versus neither was 12 % versus 0.9 % [31].
The NCCN currently offers the only guideline recommendations for surveillance following RC. They recommend cross-sectional and upper tract imaging every 3–6 months for 5 years, then as clinically indicated depending on the risk of recurrence [6]. Since UTRs may occur many years after RC and, in the absence of a competing event, the risk does not decrease over time, high risk patients require lifelong surveillance [22, 32]. Going forward, studies are needed to examine whether UTR surveillance strategies that are less reliant on frequent cross-sectional imaging, but instead use urinary cytology, renal ultrasound and physical examination, have comparable safety and effectiveness, both for patients with native bladders as well as after RC.
10.3 Bladder Recurrence Following Treatment for Upper Tract Urothelial Carcinoma
Risk Factors
Identification of risk factors for intravesical recurrence following treatment of UTUC can help guide post-treatment surveillance regimens, and identify patients who would benefit most from adjuvant intravesical treatments. While treatment of intravesical recurrences is usually successful, the required therapy is highly dependent on stage at diagnosis, making early detection critical.
Numerous studies have examined risk factors for intravesical recurrence after nephroureterectomy (NU) for UTUC (Table 10.4). Upper tract tumor risk factors include size [45, 46], presence of upper tract CIS [45, 47], stage [46, 48], grade [49, 50], location [49], and multifocality [2, 46]. Other risk factors are gender [51, 52], incomplete bladder cuff excision [51], a prior history of bladder tumors [53, 54], and surgical approach [46, 47].
Table 10.4
Prevalence and risk factors for bladder recurrence after treatment of UTUC
Study | Number of patients | Median follow-up (months) | Intravesical recurrence rate (%) | Risks for intravesical recurrence |
|---|---|---|---|---|
Sakamoto et al. (1991) [89] | 53 | 36 | Synchronous bladder tumor Multiple UTUC tumors | |
Krough et al. (1991) [59] | 198 | 36 | UTUC location | |
Mukamel et al. (1994) [50] | 69 | 48 | UTUC grade UTUC multifocality | |
Hall et al. (1998) [90] | 252 | 64 | 13 | UTUC stage UTUC treatment modality |
Hisataki et al. (2000) [48] | 69 | 53 | 35 | UTUC stage |
Koga et al. (2001) [51] | 85 | 35 | 34 | Female gender Adjuvant systemic chemotherapy Incomplete ureterectomy |
Kang et al. (2003) [2] | 189 | 91 | 31 | UTUC mutifocality |
Matsui et al. (2005) [46] | 89 | 40 | 42 | UTUC mutifocality UTUC stage UTUC tumor size Surgical approach |
Raman et al. (2005) [53] | 103 | 39 | 50 | Previous bladder cancer |
Zigeuner et al. (2006) [49] | 191 | 33 | 27 | UTUC location UTUC grade |
Novara et al. (2008) [54] | 231
Related posts: Topical Chemotherapy
Surgical Management of Upper Tract Urothelial Carcinoma– Open, Laparoscopic, and Robotic Approaches to Nephroureterectomy and Outcomes
Systemic Chemotherapy for Upper Tract Urothelial Cancer
 Surveillance After Treatment for Upper Tract Transitional Cell Carcinoma
 Beyond Endoscopy-Ultrasound, Optical Coherence Tomography and Confocal Laser Endomicroscopy
 Imaging of Upper Tract Transitional Cell Carcinoma
 Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
Get Clinical Tree app for offline access
|