0022-5347/04/1724-1342/0 THE JOURNAL OF UROLOGY® Copyright © 2004 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 172, 1342–1347, October 2004 Printed in U.S.A.
DOI: 10.1097/01.ju.0000138208.07426.19
THE MANAGEMENT OF URETHRAL TRANSITIONAL CELL CARCINOMA AFTER RADICAL CYSTECTOMY FOR INVASIVE BLADDER CANCER PETER E. CLARK,* JOHN P. STEIN, SUSAN G. GROSHEN,† GUS MIRANDA, JIE CAI, GARY LIESKOVSKY AND DONALD G. SKINNER From the Departments of Urology (JPS, GM, GL, DGS), and Preventive Medicine (SGG, JC), University of Southern California/Norris Cancer Center, Los Angeles, California and Department of Urology, Comprehensive Cancer Center of Wake Forest University School of Medicine (PEC), Winston-Salem, North Carolina
ABSTRACT
Purpose: Previous reports have identified risk factors for urethral recurrence following radical cystectomy for transitional cell carcinoma (TCC). However, reports of the clinical presentation, treatment and outcome in these patients are lacking. We report our experience with the diagnosis, management and outcome of urethral TCC after radical cystectomy for bladder cancer. Materials and Methods: A database of 1,054 patients who underwent radical cystectomy and urinary diversion for TCC from 1971 to 1997 was retrospectively reviewed. All patients with urethral TCC after surgery were identified. Results: Urethral TCC was diagnosed in 47 men a median of 18.5 months (range 2 to 116) after cystectomy with 20 (42%) diagnosed within 1 year. Symptomatic recurrence developed in 24 of 42 evaluable patients (57%), 21 had bloody urethral discharge and 7 had pain or a palpable mass. A total of 13 patients (31%) were asymptomatic with abnormal cytology. The remaining 5 patients underwent prophylactic urethrectomy based on cystectomy pathology. Overall 41 patients underwent urethrectomy, which was total in 36 and distal with perineal urethrostomy in 5, including later conversion to total urethrectomy in 2. Overall at a median followup of 26 months (range 3 to 275) since diagnosis 36 of 47 patients (76%) were dead, including 25 of metastatic disease. Only 10 patients (21%) remained disease-free. Median overall survival in patients with urethral TCC after radical cystectomy was only 28 months after the diagnosis of urethral TCC. Urethral stage (superficial vs invasive disease) at diagnosis was the most import predictor of overall survival in this cohort of patients. Conclusions: Most patients with urethral recurrence present with symptoms. However, screening cytology alone still detects a significant proportion. The median survival of patients with urethral TCC after radical cystectomy is only 28 months after diagnosis. Urethral stage (superficial vs invasive disease) at diagnosis is the most import predictor of overall survival in this cohort of patients. KEY WORDS: urethra; carcinoma, transitional cell; cystectomy; neoplasm recurrence, local; survival
The incidence of urethral recurrence after radical cystectomy for invasive transitional cell carcinoma (TCC) has been well defined in the literature and in large series it is between 0.7% and 18%.1 In a review of 2,062 patients across 18 studies the overall risk of urethral recurrence after radical cystectomy for TCC was 10.1%.1 A recent large series from the University of Southern California of 436 patients showed a urethral recurrence rate of 7.9%.2 With the development of the orthotopic neobladder the issue of preserving and subsequently managing the retained urethra has taken on new importance. There is now a large body of literature defining the risk factors for urethral recurrence, allowing surgeons to select carefully the proper candidate for orthotopic bladder substitution.1– 4 Risk factors for urethral recurrence in the male patient are papillary tumors, multifocality, bladder neck involvement, associated carcinoma in situ (CIS), upper tract TCC, and prostatic involvement with superficial TCC and invasion of the stroma.1
Despite meticulous patient selection urethral recurrences can still develop. Surprisingly little study has been done of the diagnosis and treatment in these patients. We examined the diagnosis and management of urethral TCC after radical cystectomy at our institution. In addition, we evaluated the clinical outcomes in this group of patients and the prognostic variables for overall survival.
MATERIALS AND METHODS
Between 1971 and 1997 1,054 patients (843 men and 211 women) underwent radical cystectomy for bladder TCC at University of Southern California. Median age was 66 years (range 22 to 93) and median followup was 10.1 years. The clinical outcome in these patients has been reported previously.5 Of the 1,054 patients 47 (4.4%) were found to have urethral involvement with TCC at some point after radical cystectomy, including 5 who underwent staged prophylactic urethrectomy based on final cystectomy pathology and who, therefore, technically speaking did not have urethral recurrence. All 48 patients who underwent en bloc urethrectomy at cystectomy were excluded. Also excluded were patients who underwent prophylactic urethrectomy after radical cystectomy in whom the final pathology was negative for TCC. All 47 patients with urethral TCC diagnosed after radical
Accepted for publication May 21, 2004. * Correspondence: Department of Urology, Wake Forest University Health Sciences, Medical Center Blvd., Winston-Salem, North Carolina 27157 (telephone: 336-716-5702; FAX: 336-716-5711; e-mail:
[email protected]). † Financial interest and/or other relationship with Epimmune Imaging. 1342
URETHRAL TRANSITIONAL CELL CANCER AFTER CYSTECTOMY
cystectomy were male (5.6%). There were no cases of subsequent urethral involvement in the 44 women with orthotopic urinary diversion included in the initial analysis. Mean age at cystectomy was 68.4 years (range 48 to 93). A detailed description of the standard preoperative assessment, operative technique and routine postoperative care has been reported previously.5 Patients underwent standard radical cystectomy with en bloc bilateral pelvic lymphadenectomy. Table 1 lists final pathological findings in the 47 patients with urethral TCC following cystectomy. The type of urinary diversion at the time of cystectomy was an ileal conduit in 21 patients, a continent cutaneous reservoir in 10, an orthotopic neobladder in 14 and ureterosigmoidostomy in 2. After cystectomy patients were followed at 4-month intervals for year 1, 6-month intervals for year 2 and annually thereafter. Evaluation included a history and physical examination, blood chemistries, chest x-ray, excretory urogram and pouchogram as indicated. Urethral wash cytology or voided cytology was generally performed on an annual basis, although this was modified in patients deemed at higher risk for urethral recurrence. At recurrence cystourethroscopy was reserved for patients with asymptomatic positive cytology, in which the diagnosis was in doubt, or those with an orthotopic neobladder, in which it was critical to determine the extent of urethral involvement. Five patients with an orthotopic neobladder and urethral recurrence distal to the urethral sphincter were offered distal urethrectomy with perineal urethrostomy. Patients with an orthotopic neobladder and more extensive involvement, including those involving the urethrointestinal anastomosis, underwent total urethrectomy with resection of the anastomosis and conversion to a continent cutaneous reservoir. Patients with a continent cutaneous reservoir or ileal conduit at urethral recurrence underwent standard total urethrectomy.6 Three select patients with only CIS of the urethra and an orthotopic neobladder were offered a trial of intraurethral instillations of 5-fluorouracil (5-FU) and close monitoring. A dose of 10 cc 5% 5-FU cream was mixed with an equal amount of Lubrifax (Mid-Town Petroleum, Inc., Bridgeview, Illinois) or Efudex (Valeant Pharmaceuticals International, Costa Mesa, California) and injected intraurethrally twice weekly for 1 week, weekly for 6 weeks and then monthly for at least 6 months. The patient underwent careful monitoring with voided cytologies and repeat cystourethroscopy at 12 weeks. Viscous xylocaine jelly could not be used because it causes 5-FU to precipitate. Survival following urethral recurrence (subsequent survival) was the primary clinical outcome in this study. Survival was calculated from the date of diagnosis of urethral recurrence to the date of death regardless of cause. Patients who
TABLE 1. Pathological characteristics at radical cystectomy and urethral TCC diagnosis in all 47 patients No. Pts Bladder pathological subgroup: Organ confined (CIS, pTa–pT2) Extravesical disease (pT3/4) Lymph node pos Bladder grade:21 Low/moderate High (grade 3) Urethral stage: Superficial (CIS or pTa) Invasive Unknown Urethral grade:* Low/moderate High Unknown * Determined by standard histopathology/hematoxylin and eosin
26 14 7 2 45 26 19 2 2 41 4 staining.
1343
were still alive at the last followup were censored at that time. Kaplan-Meier plots were used to estimate the probabilities (proportion) of survival every year since urethral recurrence. The log rank test (overall and stratified) was used to compare differences in survival in subgroups. The Kruskal-Wallis test was used to examine the difference in time to the diagnosis of urethral recurrence by screening cytology alone or by clinical symptoms. All p values reported in the analyses were 2-sided. RESULTS
Urethral recurrence developed 2 to 116 months (median 18.5) after radical cystectomy. A total of 20 patients (43%) had recurrence within 1 year, whereas 5 (11%) had recurrence more than 5 years after cystectomy, including 1 at 9.5 years. Of the 47 patients in the study 42 had data available with regard to symptoms at recurrence (table 2). Of the 42 patients 24 (57%) had a symptomatic urethral recurrence, 13 (31%) were detected based on screening cytology alone and 5 (12%) underwent planned prophylactic urethrectomy due to diffuse bladder and prostatic urethral CIS on final pathology after cystectomy, including 1 with a positive urethral margin with lymphovascular invasion. In patients with symptomatic recurrence the median time to recurrence was 30.9 months (range 5 to 116). In the 13 patients diagnosed by screening cytology alone median time to recurrence was 11.6 months (range 2 to 55, p ⫽ 0.026). Of the patients 33 had cytological data available at urethral TCC diagnosis. In this group only 2 patients had negative cytology. In the remaining 31 patients (94%) cytology was positive. A total of 14 patients underwent urethroscopic examination at urethral recurrence. In 11 cases gross tumor on endoscopic examination was confirmed on biopsy. In the remaining 3 cases random biopsy or urethral washings were positive for malignancy. A total of 41 patients underwent urethrectomy at the initial urethral diagnosis of TCC, including 36 total urethrectomies. Five of the 14 patients (36%) with an orthotopic urinary diversion had urethral recurrence clinically confined to an area in the mid/distal urethra and underwent distal urethrectomy and perineal urethrostomy. Two of these patients subsequently underwent total urethrectomy and conversion to a continent cutaneous urinary reservoir due to a positive margin on final pathology. At urethrectomy 28 other operative procedures were performed simultaneously, including conversion to a continent cutaneous reservoir (9), insertion of a penile prosthesis (7), partial penectomy (1), removal of an artificial urinary sphincter and/or penile prosthesis (3) and total penectomy (3). Table 1 shows the stage of urethral TCC at diagnosis. Three patients with an orthotopic urinary diversion and low stage disease at urethral recurrence or CIS were treated with intraurethral instillations of 5-FU with or without endoscopic tumor resection. All 3 patients had an initial response to therapy. One patient with a small Ta tumor and adjacent CIS never had recurrence and remained free of disease after 7 years of followup. The remaining 2 patients had local recurrence within 10 to 12 months, ultimately
TABLE 2. Presentation mode of urethral TCC after radical cystectomy Presentation Mode Symptom
No. Pts
Symptomatic:* Hematuria/bloody urethral discharge Pain/palpable mass Voiding habit change Abnormal cytology, asymptomatic Prophylactic urethrectomy† * More than 1 symptom in some patients. † Based on radical cystectomy pathology.
24 21 7 4 13 5
1344
URETHRAL TRANSITIONAL CELL CANCER AFTER CYSTECTOMY
requiring urethrectomy. These 2 patients eventually died of metastatic TCC. Three patients were treated with systemic chemotherapy alone at urethral recurrence. Two patients had metastatic disease at recurrence, while 1 had infiltrating TCC at the urethroenteric anastomosis of an orthotopic urinary diversion. The latter patient underwent external beam radiation therapy in addition to chemotherapy and was the only one to remain disease-free at followup 26 months after the initial recurrence. The remaining 2 patients died of metastatic TCC 15 and 17 months, respectively, from the time of urethral recurrence. Overall at a median followup of 26 months (range 3 to 275) since the time of urethral TCC diagnosis 36 of 47 patients (76%) died, 25 including (53%) of metastatic TCC. Ten patients (21%) had no evidence of disease and 1 remained alive 148 months after recurrence but disease status was unknown. Median overall survival after the urethral TCC diagnosis was 28 months with a 5-year actuarial survival of 35.2% (fig. 1). Patients with superficial TCC of the urethra at urethral diagnosis (CIS alone or pTa) had significantly better overall survival than patients with invasive disease (median survival 58.5 vs 17.1 months, p ⫽ 0.017, fig. 2). This was the single best predictor of overall survival from the time of diagnosis of urethral disease. When stratified by pathological stage at cystectomy, there was no apparent relationship between bladder stage and overall survival (p ⫽ 0.90, fig. 3). Overall survival from the time of urethral diagnosis was not significantly influenced by whether patients were symptomatic at diagnosis or whether they had been diagnosed based on cytology alone (p ⫽ 0.59, fig. 4). Similarly when patients were stratified by the time from radical cystectomy to the diagnosis of urethral TCC, there was no difference in overall survival (p ⫽ 0.26, fig. 5). Overall survival was also not influenced by the date of cystectomy when divided into the decade of surgery (1970s vs 1980s vs 1990s, p ⫽ 0.53). DISCUSSION
In the current series more than 40% of patients who were found to have urethral TCC were diagnosed within 1 year
after cystectomy and the median time to diagnosis was 18 months. This is in accord with most series in the literature, in which the median time was 1.6 years to 2.2 years.1 Nevertheless, there are several reported cases of late urethral recurrences, including those in this series, up to 20 years after radical cystectomy.2, 7 While relatively rare, this highlights the importance of lifelong followup in these patients. Most symptomatic patients presented with bloody urethral discharge or gross hematuria, as noted in the earlier report of Schellhammer and Whitmore.7 However, 4 patients with an orthotopic neobladder presented with a change in voiding habits, which in 1 was the only patient complaint. It is important to consider urethral recurrence in the differential diagnosis when a patient with an orthotopic neobladder presents with any change in voiding habits. Patients who have urethral recurrence after radical cystectomy have relatively poor median overall and disease specific survival.8, 9 In our study the median overall survival was 28 months from the time of urethral diagnosis. The most important predictor of overall survival in the current series was urethral TCC stage at diagnosis. Bladder stage at cystectomy and year of cystectomy did not influence survival. Unfortunately there were not enough patients with low/moderate grade disease to assess adequately any influence on survival. Prior evidence in the literature for the relative influence of bladder vs urethral pathology on overall and disease specific survival is sparse and contradictory. Most studies show no direct comparison between the relative influence of urethral vs bladder pathology or did not use historical controls. Nevertheless, several studies suggest that patients with invasive or overt urethral recurrences have a poor outcome, as demonstrated in the current analysis.10 –13 A recent study of Lin et al showed that in 24 male patients who underwent urethrectomy after radical cystectomy only cystectomy pathological stage and not urethral stage predicted overall survival.14 The reason for this contrast is likely due to differences in the patient population between the 2 studies. In the current study patients who had pT0 disease at urethrectomy were excluded and 42% of our patients had invasive disease. In the study of Lin et al 37% of the
FIG. 1. Actuarial overall survival in 47 patients with urethral TCC after radical cystectomy for bladder cancer
URETHRAL TRANSITIONAL CELL CANCER AFTER CYSTECTOMY
1345
FIG. 2. Actuarial overall survival according to stage at diagnosis of urethral TCC from time of diagnosis, stratified by whether urethral TCC was superficial (Ta and/or CIS) or invasive (T1 or greater).
FIG. 3. Actuarial overall survival stratified by stage at cystectomy and time of urethral TCC diagnosis, stratified by whether bladder TCC at radical cystectomy was organ confined (OC) (pTa to pT2 or CIS), extravesical (EV) (pT3 or pT4) or lymph node positive (LN⫹).
patients had pT0 disease and only 12.5% had invasive disease. Relatively few groups have attempted to determine whether the method of diagnosis influences the outcome in patients with urethral recurrence. In agreement with our findings, the study of Lin et al also showed that the method of diagnosis did not impact survival.14 Several series have documented poor survival in patients with symptomatic urethral disease recurrence.7, 15–17 Other studies have assessed patients who had asymptomatic urethral recurrences diagnosed by cytology alone.10, 18 –20 In general they showed a favorable outcome in this group of patients. Except for Lin et al14 none of these groups compared outcomes between symp-
tomatic and asymptomatic men. Although this is the largest series of urethral recurrences reported in the literature to address prognostic issues there were still only 47 patients in this series. It remains possible that small nonsignificant differences in the current analysis would be significant if more patients were included. As with any retrospective review, there are some important potential weaknesses with this report. There is the potential for stage migration as well as changes in the treatment of patients in the 25-year time frame of this database. In an effort to address this we analyzed the effect of date of cystectomy and/or urethrectomy on prognosis and found no influence on outcome. Another poten-
1346
URETHRAL TRANSITIONAL CELL CANCER AFTER CYSTECTOMY
FIG. 4. Actuarial overall survival from time of urethral TCC diagnosis stratified by whether patient was symptomatic at urethral TCC diagnosis or whether diagnosis was based on cytology alone.
FIG. 5. Actuarial overall survival stratified by time from radical cystectomy for bladder TCC to urethral TCC diagnosis, stratified by whether diagnosis was made within 11⁄2 years after radical cystectomy.
tial weakness is that some patient records are old and at times incomplete. Despite exhaustive efforts to supplement the database via patient contact and records from elsewhere with appropriate permission there remained a small proportion of patients in whom not all relevant data were available for analysis. Furthermore, we recognize that this is a retrospective review and, therefore, subject to potential selection bias that may not be readily measurable. Nevertheless, this is the largest reported series of urethral recurrences in the modern era and the median followup across the entire database is 10 years. Despite the poor overall prognosis in patients with ure-
thral recurrence after radical cystectomy the percent with recurrence (5.9%) is quite small. If all risk factors proposed to date for urethral recurrence were to serve as contraindications to orthotopic diversion, approximately 70% of patients would be ineligible.2 When en bloc prophylactic urethrectomies are performed, 57% to 100% the time there is no urethral TCC identified.1 Therefore, the data presented in the current study do not mean that patients cannot undergo orthotopic diversion. Rather, properly selected patients with a negative urethral margin on frozen section at radical cystectomy can safely undergo orthotopic diversion. However, it is imperative that the surgeon and
URETHRAL TRANSITIONAL CELL CANCER AFTER CYSTECTOMY
patient should be committed to lifelong followup of the native urethra. CONCLUSIONS
Most patients with urethral recurrence present with symptoms. However, screening cytology alone still detects a significant proportion. Median survival in patients with urethral TCC after radical cystectomy is only 28 months from the time of diagnosis. Urethral stage (superficial vs invasive disease) at diagnosis is the most import predictor of overall survival in this cohort of patients. REFERENCES
1. Freeman, J. A., Esrig, D., Stein, J. P. and Skinner, D. G.: Management of the patient with bladder cancer. Urethral recurrence. Urol Clin North Am, 21: 645, 1994 2. Freeman, J. A., Tarter, T. A., Esrig, D., Stein, J. P., Elmajian, D. A., Chen, S.-C. et al: Urethral recurrence in patients with orthotopic ileal neobladders. J Urol, 156: 1615, 1996 3. Lebret, T., Herve, J. M., Barre, P., Gaudez, F., Lugagne, P. M., Barbagelatta, M. et al: Urethral recurrence of transitional cell carcinoma of the bladder. Predictive value of preoperative latero-montanal biopsies and urethral frozen sections during prostatocystectomy. Eur Urol, 33: 170, 1998 4. Bell, C. R., Gujral, S., Collins, C. M., Sibley, G. N. and Persad, R. A.: The fate of the urethra after definitive treatment of invasive transitional cell carcinoma of the urinary bladder. BJU Int, 83: 607, 1999 5. Stein, J. P., Lieskovsky, G., Cote, R., Groshen, S., Feng, A. C., Boyd, S. et al: Radical cystectomy in the treatment of invasive bladder cancer: long-term results in 1,054 patients. J Clin Oncol, 19: 666, 2001 6. Ahlering, T. and Lieskovsky, G.: Surgical treatment of urethral cancer in the male patient. In: Diagnosis and Management of Genitourinary Cancer. Edited by D. G. Skinner and G. Lieskovsky. Philadelphia: W. B. Saunders Co., chapt. 43, p. 627, 1988 7. Schellhammer, P. F. and Whitmore, W. F., Jr.: Transitional cell carcinoma of the urethra in men having cystectomy for bladder
1347
cancer. J Urol, 115: 56, 1976 8. Zabbo, A. and Montie, J. E.: Management of the urethra in men undergoing radical cystectomy for bladder cancer. J Urol, 131: 267, 1984 9. Poole-Wilson, D. S. and Barnard, R. J.: Total cystectomy for bladder tumours. Br J Urol, 43: 16, 1971 10. Ahlering, T. E., Lieskovsky, G. and Skinner, D. G.: Indications for urethrectomy in men undergoing single stage radical cystectomy for bladder cancer. J Urol, 131: 657, 1984 11. Hendry, W. F., Gowing, N. F. and Wallace, D. M.: Surgical treatment of urethral tumours associated with bladder cancer. Proc R Soc Med, 67: 304, 1974 12. Raz, S., McLorie, G., Johnson, S. and Skinner, D. G.: Management of the urethra in patients undergoing radical cystectomy for bladder carcinoma. J Urol, 120: 298, 1978 13. Nurmi, M., Puntala, P. and Ekfors, T.: Urethral recurrence after cystoprostatectomy for bladder carcinoma. Scand J Urol Nephrol, 23: 1, 1989 14. Lin, D. W., Herr, H. W. and Dalbagni, G.: Value of urethral wash cytology in the retained male urethra after radical cystoprostatectomy. J Urol, 169: 961, 2003 15. Cordonnier, J. J. and Spjut, H. J.: Urethral occurrence of bladder carcinoma following cystectomy. J Urol, 87: 398, 1962 16. Baron, J. C., Gibod, L. B. and Steg, A.: Management of the urethra in patients undergoing radical cystectomy for bladder carcinoma. Eur Urol, 16: 283, 1989 17. Clark, P. B.: Urethral carcinoma after cystectomy: the case for routine urethrectomy. J Urol, 90: 173, 1984 18. Hickey, D. P., Soloway, M. S. and Murphy, W. M.: Selective urethrectomy following cystoprostatectomy for bladder cancer. J Urol, 136: 828, 1986 19. Faysal, M. H.: Urethrectomy in men with transitional cell carcinoma of bladder. Urology, 16: 23, 1980 20. Bardot, S. F., Montie, J. E. and Levin, H. S.: Urethral recurrence after cystoprostatectomy. J Urol, suppl., 145: 338A, abstract 501, 1991 21. Bergkvist, A., Ljungqvist, A. and Moberger, G.: Classification of bladder tumours based on the cellular pattern. Preliminary report of a clinical-pathological study of 300 cases with a minimum follow-up of eight years. Acta Chir Scand, 130: 371, 1965