Elective management of transitional cell carcinoma of the distal ureter: can kidney-sparing surgery be advised?

Authors


Urs E. Studer, Department of Urology, University of Bern, Anna-Seiler-Haus, Inselspital, CH-3010 Bern, Switzerland.
e-mail: urology.berne@insel.ch

Abstract

OBJECTIVE

To determine the long-term oncological outcome of patients with primary transitional cell carcinoma (TCC) of the distal ureter electively treated with either kidney-sparing surgery (KSS) or radical nephroureterectomy (RNU) in a retrospective, non-randomized, single-centre study.

PATIENTS AND METHODS

Of 43 consecutive patients with a primary solitary distal ureter TCC, 19 had KSS, consisting of distal ureter resection with bladder cuff excision and ureter reimplantation, and 24 had RNU with bladder cuff excision.

RESULTS

The median (range) age at surgery was 69 (31–86) years for the KSS group and 73 (59–87) years for the RNU group, patients in the latter having worse hydronephrotic kidneys. The median (range) follow-up was 58 (3–260) months. A recurrent bladder tumour was diagnosed after a median of 15 months in five of the 19 patients treated by KSS and after a median of 5.5 months in eight of the 24 treated by RNU. Five of the 19 patients treated by KSS and six of the 24 treated by RNU died from metastatic disease despite chemotherapy. Recurrence-free, cancer-specific and overall survival were comparable in the two groups. In two patients (11%) treated by KSS an ipsilateral upper urinary tract TCC recurred after 42 and 105 months, respectively.

CONCLUSION

Treatment by distal ureteric resection is feasible in patients with primary TCC of the distal ureter. The long-term oncological outcome seems to be comparable with that of patients treated by RNU. Furthermore, kidney preservation is advantageous if adjuvant or salvage chemotherapy is required.

Abbreviations
RNU

radical nephroureterectomy

KSS

kidney-sparing surgery

UUT

upper urinary tract.

INTRODUCTION

Radical nephroureterectomy (RNU) with bladder cuff excision is considered the standard treatment for patients with TCC of the upper urinary tract (UUT), because the reported recurrence rate in the ipsilateral urothelium is high (up to 58%), tumours are frequently multifocal (up to 44%) and the incidence of contralateral synchronous and metachronous involvement is low (2–5%) [1,2]. By eliminating the risk of ipsilateral recurrence, radical surgery requires less frequent follow-up examinations, thereby reducing costs [3].

In patients with solitary kidneys, bilateral synchronous tumours or severely impaired renal function, kidney-sparing surgery (KSS) has been used for UUT TCC [4,5]. Due to the encouraging oncological results in these patients, a kidney-sparing approach has also been proposed for patients with no imperative indications [6,7]. Kidney preservation might also be important if platinum-based chemotherapy becomes necessary for future treatment.

Whether KSS represents a valid alternative to standard RNU for elective indications remains a matter of debate, as long-term survival data from the few reported series are limited [8–15]. There are no randomized clinical trials, and they would be difficult to conduct because TCC of the distal ureter is rare, patients might not be willing to participate and those with hydronephrosis tend to gave RNU. Therefore, the aim of the present retrospective, non-randomized, single-centre study was to compare the long-term oncological outcome in patients with primary TCC of the distal ureter electively treated with either KSS or RNU.

PATIENTS AND METHODS

A consecutive series of 61 patients with TCC of the distal ureter treated at our institution between December 1974 and December 2004 was analysed retrospectively. Patients with primary, solitary, unilateral TCC of the distal (intrapelvic) ureter and a normal-functioning contralateral kidney were included. Patients with previous UUT TCC, previous or concomitant radical cystoprostatectomy for TCC, ureteric TCC treated by endourological procedures, distant metastasis at diagnosis, other malignancies or incomplete follow-up data were excluded. Forty-three patients (70%) fulfilled these criteria and were eligible for analysis.

The preoperative evaluation included: (i) serum creatinine, creatinine clearance and, if necessary, diuretic renal scintigraphy; (ii) cytological examination of voided urine after fluid intake, 40 mg furosemide and ambulation (forced diuresis); (iii) cysto-urethroscopy with biopsy and/or transurethral resection of bladder tumour whenever present, and cytological examination of bladder barbotage and urine from selective UUT catheterization; (iv) excretory and/or retrograde urography, CT of the abdomen/pelvis, chest X-ray and bone scan. Routine ureteroscopy with tumour biopsy was not used, to avoid tumour cell spillage into the UUT.

Based on the urography findings, preoperative hydronephrosis was graded as absent, mild (ureter only dilated), moderate (ureter and renal pelvis dilated) or severe (ureter, renal pelvis and calyces dilated). Tumour size was measured by urography or CT.

Patients with severe hydronephrosis and loss of renal parenchyma all had RNU, assuming that functional recovery of the kidney after KSS would be minimal. In the others, the decision to use either KSS or RNU was made by the operating surgeon. Nineteen patients had KSS, consisting of en bloc resection of the distal ureter with a bladder cuff. The ureteric resection margin was 2 cm proximal to the ureter tumour. Intraoperative frozen-sections of the ureter margins were not analysed routinely. The ureter was then reimplanted into the bladder using the Politano-Leadbetter technique with a psoas hitch. RNU was used in 24 patients and consisted of en bloc removal of the entire kidney and ureter with a bladder cuff. Regional lymph nodes were dissected if an invasive TCC was suspected during surgery, or enlarged (>1 cm in diameter) nodes were detected.

Ureteric TCC was confirmed histopathologically in all cases. Tumour stage and grade were designated according to the 2002 International Union Against Cancer TNM classification [16] and the 1973 WHO grading system [17].

Adjuvant chemotherapy was offered to patients with lymph node metastases. Patients were followed according to a prospective standardized institutional protocol. Cytology from forced diuresis urine and bladder barbotage after cysto-urethroscopy was done every 6 months for the first year and at yearly intervals thereafter. Excretory urography was used in all patients after a year and if considered necessary thereafter. For muscle-invasive (stage ≥ pT2) tumours, CT of the abdomen/pelvis and a chest X-ray were taken at yearly intervals.

The study end points were bladder recurrence-free survival, UUT recurrence rate, cancer-specific survival and overall survival. Kaplan-Meier estimates were compared by the log-rank test. Patients were censored at the time of the last known follow-up visit. Differences in patient characteristics between the groups were analysed using the Mann–Whitney U-test for quantitative variables and the chi-square test for categorical variables, with P < 0.05 considered to indicate significance. The results were analysed in collaboration with the Department of Mathematical Statistics, University of Bern, Switzerland.

RESULTS

The median (range) age of the patients was 72 (31–86) years and median (range) follow-up 58 (3–260) months; the patient characteristics before surgery were similar in the two groups, except for severe hydronephrosis, which was absent in the KSS group (Table 1). The ratio of carcinoma-type to papillary-type tumour on selective UUT catheterization cytology was higher, albeit not significantly, in the RNU than in KSS group (56% vs 38%; data not shown).

Table 1.  Preoperative patient and tumour characteristics of the two treatment groups
CharacteristicKSS (19)RNU (24)P
  1. NS, not significant.

Median (range):
 age, years69 (31–86)73 (59–87)NS
 age-adjusted creatinine clearance, mL/min51 (20–198)56 (20–149)NS
Presenting symptom, n (%)  NS
 macroscopic haematuria 11 (58)14 (58) 
 flank pain 5 (26) 6 (25) 
 microscopic haematuria 1 (5) 1 (4) 
 diagnosis during follow-up of bladder tumour 2 (11) 3 (13) 
With hydronephrosis, n (%)  <0.001
 absent or mild 9 (47) 4 (17) 
 moderate10 (53) 6 (25) 
 severe 0 (0) 14 (58) 
Median (range) tumour size, mm20 (5–25)20 (10–25)NS
Positive cytology results with, n (%)  NS
 forced diuresis 8 (42)10 (42) 
 bladder barbotage 6 (32) 8 (33) 
 selective UUT catheterization13 (68)16 (67) 
With previous bladder TCC, n (%) 3 (16) 3 (13)NS
With concomitant bladder TCC, n (%) 5 (26) 9 (38)NS
Median (range) follow-up, months50 (3–191)49 (3–261)NS
Tumour characteristics, n (%)
Stage
 pTa13 (68) 11 (46) 
 pT1 0 (0) 6 (25) 
 pT2 4 (21) 1 (4) 
 pT3 2 (11) 6 (25) 
Grade
 G1 1 (5) 1 (4) 
 G213 (68) 11 (46) 
 G3 5 (26)12 (50) 
Growth pattern
 papillary14 (74)13 (54) 
 solid 2 (11) 6 (25) 
 mixed 3 (16) 5 (21) 
Associated Tis in distal ureter 2 (11) 1 (4) 
Surgical margins
 negative19 (100)24 (100) 
 positive 0 0 
Lymph node status
 pNX 9 (47) 6 (25) 
 pN0 8 (42)15 (63) 
 pN+ 2 (11) 3 (13) 

Before surgery, six of the 43 patients (14%) had a history of superficial bladder cancer and the distal ureteric TCC was diagnosed a median (range) of 48.5 (5–156) months later. A concomitant bladder cancer was found in 14 of the 43 patients (33%); nine of these had a TaG1/G2, one a carcinoma in situ (Tis) and four a T1G3 bladder cancer. The bladder cancer was multifocal in six of these 14 patients. The side of the bladder cancer corresponded to the side of the distal ureter tumour in 13 of the 14 patients.

Patients treated by KSS had pTa/pT1 distal ureter TCC in 68% and pT2/pT3 in 32%, and patients treated by RNU in 71% and 29%, respectively (Table 1). Grade 3 tumour was diagnosed in 26% of KSS and in 50% of RNU patients. Lymph nodes were dissected in 10 KSS (53%) and in 18 RNU patients (75%). Positive nodes were found in two KSS (11%) and in three RNU patients (13%), all associated with muscle-invasive high-grade TCC of the ureter.

A second, synchronous TCC of the UUT was found in the RNU specimen of three patients, involving the renal calyces in two (pTaG2 and pT3G3) and the mid-ureter (pT1G3) in one. Preoperative CT and excretory urography were reviewed and considered negative for concomitant ipsilateral UUT tumours. All three patients had a history of recurrent and multifocal superficial bladder cancer. The two patients with invasive disease died from metastatic disease, despite RNU, at 6 and 13 months, respectively. The patient with pTaG2 tumour is alive with no evidence of disease. He had had ureteroscopy with tumour biopsy before RNU.

Adjuvant chemotherapy was given to all patients with positive lymph nodes, two and three in the KSS and RNU group, respectively.

Recurrent bladder TCC was diagnosed in five KSS (26%) and in eight RNU patients (33%) after a median of 15 and 6 months, respectively. The bladder tumour was TaG1/2 in three, T1G3 in seven and T2G3 in three patients. Except for one patient, tumour recurrences were all multifocal and occurred in the peri-ostial area of the ureter tumour side. The median (95% CI) bladder recurrence-free survival after 5 and 10 years was 82  (63–95)% and 39 (21–67)% in KSS and 52 (39–71)% and 22 (11–59)% in RNU patients, respectively (log-rank test, P = 0.117).

A recurrent ipsilateral UUT tumour was diagnosed in two patients (11%) after KSS. One was initially treated for a pTaG2 and the other for a pT2G3 ureteric tumour. The former was treated with an endoscopic retrograde tumour resection for a TaG2 TCC in the distal reimplanted ureter after 42 months, and by radical cystectomy for BCG-refractory bladder Tis together with a distal ureter resection for a second TaG2 recurrence after 85 months. The latter had a distal ureterectomy in conjunction with a radical cystectomy for multifocal T1G3 bladder cancer after 105 months.

A recurrent contralateral TCC of the renal pelvis occurred in one patient in the KSS group with a pTaG2 distal ureter tumour, after 3 months. The patient refused further treatment and died from metastatic disease 2 months later. CT before surgery and excretory urography of this patient were reviewed and found to be negative for concomitant UUT tumours.

Of the 43 patients, 20 (47%) have died, 11 of them from progressive disease despite chemotherapy. The remaining 23 patients (53%) are alive with no evidence of disease (Fig. 1). The median (95% CI) cancer-specific survival was 64 (43–96)% after 5 and 10 years in KSS and 66 (46–93)% after 5 and 10 years in RNU patients, respectively (log-rank test, P = 0.896; Fig. 2A). Overall survival after 5 and 10 years was 52 (32–82)% and 43 (23–77)% in KSS, and 56 (36–85)% and 37 (15–92)% in RNU patients, respectively (log-rank test, P = 0.693; Fig. 2B). When survival was stratified by tumour stage (pTa/T1 vs pT2/T3), no patients with pTa/pT1 tumours died from disease, regardless of the type of surgery, if four (two per group), who also had a metachronous and invasive TCC either in the renal pelvis or the bladder, were excluded from the analysis. Conversely, three of six KSS patients and four of seven RNU patients with pT2/pT3 tumours died from systemic disease.

Figure 1.

The long-term oncological results for all 43 patients: (A) bladder recurrence-free survival; median (95% CI), 65 (55–89)% at 5 years and 39 (23–63)% at 10 years: (B) cancer-specific survival: 65 (51–86)% at 5 and 10 years: (C) overall survival: 54 (39–74)% at 5 years and 42 (28–63)% at 10 years.

Figure 2.

The long-term oncological results for the two treatment groups: KSS (19) and RNU (24): (A) the cancer-specific survival at 5 and 10 years was not statistically significantly different (log rank test, P = 0.896); (B) the overall survival at 5 and 10 years was not statistically significantly different (log rank test, P = 0.693).

DISCUSSION

The treatment of choice for TCC of the distal ureter remains controversial. According to the recently updated European Association of Urology guidelines [3], it is still considered to be RNU. For distal ureteric TCC the option of KSS is mentioned, but no clear recommendations are made. The National Comprehensive Cancer Network guidelines [18] conclude that KSS might be the preferable option for distal ureteric TCC if clinically feasible, but still advocate radical surgery in patients with high-grade tumours.

The concept of KSS might be supported by the fact that in patients with UUT TCC, tumour stage, grade and location within the UUT are the most important prognostic factors, irrespective of the treatment. Tumours of the distal ureter are more common than those of the mid and proximal ureter, and are more frequently solitary, of smaller volume and of lower stage and grade than their renal pelvic or upper ureteric counterparts [2,13]. Furthermore, they are less often associated with recurrent disease within the UUT. Recurrences, if any, occur almost exclusively distal to the primary tumour site and are frequently also of lower stage and grade [8,15,19]. Taking these data together, tumours of the distal ureter should be the most amenable in the UUT for elective KSS [2,19].

Mazeman [8] was the first to show that recurrences after KSS for distal ureter TCC usually do not occur proximal to the site of the primary tumour, which corroborates the theory of intraluminal tumour-cell seeding, the mechanism most frequently involved in TCC recurrence [20]. Interestingly, two of the present patients had a recurrent tumour proximal to the primary one after KSS; both had had ureteroscopy with tumour biopsy before surgery. The question then arises as to whether this proximal tumour recurrence might be the result of cell seeding during this procedure. It would therefore be wise to avoid any endoscopic manipulation of the ureteric tumour before KSS. Both patients finally required a cystectomy, and it cannot ultimately be excluded that they might have been spared cystectomy had they been operated radically early on. However, cystectomy was necessary 7 and 9 years after KSS and might rather be attributable to the biological malignant potential of the tumour than to the surgery.

However, three patients in the RNU group had concomitant TCC located more proximally, which had not been diagnosed before RNU. Two of them had invasive tumours and died from metastatic disease despite standard radical treatment. With KSS the outcome would probably not have been different. The third patient had a superficial lesion which would have been detected in due time during the standard follow-up, and treated without compromising survival.

In the present series there was no significant difference in bladder recurrence-free survival between the groups, in accordance with others [11,15]. Similar to the values reported by Leitenberger et al.[15] and Kang et al.[21], 92% of the bladder tumour recurrences in the present patients were in the peri-ostial area on the side of the ureteric tumour.

There was also no difference in cancer-specific and overall survival between the treatment groups; there was a trend to better cancer-specific survival in patients with other than muscle-invasive (pTa/pT1) tumours than in those with muscle-invasive (pT2/pT3) tumours, in both groups. No statistical subgroup analysis was used, as there were too few patients.

Comparisons between the present and other series should be interpreted cautiously. In the present series only patients with distal ureteric tumours were included and treated either by KSS or RNU. The other series were more heterogeneous and included tumours within the entire UUT, and which were treated by a variety of surgical techniques [8–15]. In most of them no information was available on the standardized follow-up protocols. Despite this, the finding that none of the present patients with low-grade, non-muscle-invasive tumour died from disease is in accordance with the 100% 5-year overall survival rates reported by others [9,11,12] for this group of patients. As in the present patients, the 5-year cancer-specific survival for muscle-invasive disease was also poor, with rates as low as 25%[12]. Furthermore, and again similar to the present findings, no survival difference was detected in relation to the surgical approach (KSS or RNU) adopted in their series. This finding was also corroborated by Heney et al.[10] and Anderström et al.[13], suggesting that the biology of the tumour is ultimately decisive. Conversely, Zungri et al.[14] found a significant difference in the 5-year overall survival rate in a (small) group of 12 patients with high-grade, muscle-invasive tumours, with 57% survival in the RNU group and 33% in the KSS group.

There are evident limitations to the present study, the main one being that the two groups were not comparable for preoperative hydronephrosis and pathological tumour grade. All patients treated by KSS had either mild or moderate hydronephrosis, whereas two-thirds of those treated by RNU had severe hydronephrosis. High-grade (G3) tumours were found in a quarter of the KSS and half of the RNU group. Thus, it is possible that patients in the RNU group had longer lasting tumours, as indicated by the hydronephrosis, and more advanced or aggressive disease, and that these differences might have provided the KSS group with a more favourable prognosis. The surgeon’s choice might also have been influenced by the results of the selective UUT cytology, as there were more patients with carcinoma-type cytology in the RNU group. However, there were too few patients to allow a subgroup analysis, and over-interpretation should be avoided. It might also be argued that over the long accrual period of three decades, disease presentation, treatment and outcome might have changed. However, a recent analysis evaluating the temporal trends of UUT TCC in terms of incidence and survival in the USA over the last 20 years detected only a slight increase in the incidence of ureteric tumours and low-stage lesions, but no significant difference in 5-year cancer-specific survival stratified by tumour stage [22]. It is likely that these findings also apply to the European population.

Based on the results of the present retrospective, non-randomized analysis of a relatively homogeneous series of patients with primary TCC of the distal ureter, and published data, KSS appears to be a feasible treatment option in patients with low-grade, non-muscle-invasive disease, and might thus avoid potential over-treatment. It does not seem to jeopardize the long-term oncological outcome, provided that the follow-up is meticulous.

For patients with muscle-invasive disease, whose prognosis appears to be poor irrespective of the treatment used, radical surgery might be more advisable, as suggested by the European Association of Urology and National Comprehensive Cancer Network guidelines. KSS in these patients might be considered on an individual basis, depending on age, comorbidities, and the requirement for good renal function for eventual adjuvant or salvage chemotherapy.

CONFLICT OF INTEREST

None declared.

Ancillary