After Cystectomy, Is It Justified to Perform a Bladder Replacement for Patients with Lymph Node Positive Bladder Cancer?

European Urology

European Urology 42 (2002) 344±349

After Cystectomy, Is It Justified to Perform a Bladder Replacement for Patients with Lymph Node Positive Bladder Cancer? Thierry Lebret, Jean-Marie Herve, Laurent Yonneau, Vincent Molinie, Philippe Barre, Pierre-Marie Lugagne, Martine Butreau, Laurent Mignot, Henry Botto* Department of Urology, HoÃpital Foch, 40 rue Worth, 92151 Suresnes, France Accepted 29 May 2002

Abstract Purpose: After cystectomy for bladder cancer, when pelvic lymph nodes are positive, bladder replacement remains controversial. The aim of this study was to evaluate the outcome of patients who underwent neobladder replacement despite bladder cancer metastasis to the regional lymph nodes. Materials and Methods: From 1981 to 1997, a total of 504 consecutive cystectomies for bladder cancer were performed at our institution. For 150 patients, pelvic lymphadenectomy were positive, nevertheless 71 patients underwent a neobladder replacement (50 N1 and 21 N2). The distribution of patients by clinical stage, according to the TNM 97 classi®cation, was 4 T1, 14 T2, 32 T3 and 21 T4. No patient showed signs of metastasis on diagnosis. Results: Five-year disease speci®c survival rate of the entire group (71 patients) was 46%. With a mean follow-up of 8.3 years (3.2±20 years), 25 patients (35%) were alive and free of disease (72% with day continence), ®ve patients were alive with recurrence (three bone metastasis, one chest metastasis and one with local recurrence), 41 patients died, (three non-cystectomy related). Of the 46 patients who recurred, a total of eight patients had local recurrence. For ®ve patients, a severe dysfunction of the plasty appeared: two needed de®nitive bladder drainage until they died, one patient became totally incontinent, one patient needed a conversion of the plasty to Bricker ileal conduit. For the remaining patient the tumor involvement provoked recto-plasty-cutaneous ®stula. All these ®ve patients died in the 6 months after the plasty dysfunction appeared. Conclusions: Although prognosis in bladder cancer metastasis to the regional lymph nodes has been reported to be poor, this study demonstrates that after cystectomy, it is justi®ed to propose a neobladder replacement to well selected patients. Local recurrence only occurred in 11% of patients and there was no damage to enteroplasty function for nearly half of the patients, and considering bene®t to the quality of life, orthotopic bladder substitution should be considered as the preferential diversion in this patient population. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Bladder neoplasms; Radical cystectomy; Lymph nodes; Lymphadenectomy; Enteroplasty

1. Introduction Orthotopic bladder reconstruction has become the standard procedure, as well as the preferred method of urinary diversion, after cystectomy for localized *

Corresponding author. Tel. ‡33-1-46-25-24-64; Fax: ‡33-1-46-25-20-24. E-mail address: [email protected] (H. Botto).

bladder carcinoma. Signi®cant progress has permitted this procedure to be proposed to a majority of patients of both sexes in order to improve quality of life after bladder removal. Although radical cystectomy and pelvic lymphadenectomy are routinely performed, the use of bladder replacement remains a subject of debate, when lymph nodes are positive. Skinner, and then Vieweg et al. for Memorial SloanKettering Cancer Center, were the ®rst to emphasize

0302-2838/02/$ ± see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 2 - 2 8 3 8 ( 0 2 ) 0 0 3 2 0 - 2

T. Lebret et al. / European Urology 42 (2002) 344±349

the curability of some positive cases by lymphadenectomy performed at cystectomy [1,2]. Nevertheless, the local recurrence risk has been reported to be strongly associated with the extent of nodal cancer involvement [3]. In fact, the 5- and 10-year recurrence-free survival for these patients has been reported to be respectively 30±35% and 25±30% [4,5]. As recurrence may compromise the bene®t of bladder reconstruction, it is of major interest to study the outcome of positive lymph node patients when orthotopic neobladder has been performed after cystectomy. The real question is: is it justi®ed to avoid enteroplasty for those patients, what should be proposed for enteroplasty candidates when identi®ed as having node positive bladder cancer during cystectomy revealed by frozen samples? Since 1985, at our institution, in an attempt to provide patients with a more acceptable means for storing and voiding urine, despite positive lymph nodes, we proposed orthotopic bladder replacement at cystectomy. This study retrospectively evaluated the outcome of patients in an attempt to address these questions. 2. Materials and methods From 1981 to 1997, a total of 504 consecutive patients (41 females and 463 males, mean age 62 years, range 38±90) underwent a cystectomy with lymphadenectomy for transitional cell carcinoma. Pathological stage was assigned based on the radical cystectomy and pelvic lymph node dissection specimen and classi®ed according to the TNM 97 staging system (Table 1). For 150 of these patients (29.8%), after a meticulous pelvic iliac lymphadenectomy, positive nodes were found on examination of frozen sections and then con®rmed by ®nal analysis. For four patients, microscopic metastasis was not revealed by frozen section but at the time of ®nal pathologic review (false negative frozen section analysis). Patient selection criteria for orthotopic bladder reconstruction included patients and cancer factors. Patient contraindications were age (over 80-year old) or signi®cant co-morbid health problem factors or psychological inability to assume enteroplasty. Table 1 Pathological stage of the 504 cystectomies performed in our institution from 1981 to 1997a Negative lymph nodes

Positive lymph nodes

Total

pT1 N0 M0 ˆ 37 pT2a N0 M0 ˆ 158 pT2b N0 M0 ˆ 55 pT3 N0 M0 ˆ 61 pT4 N0 M0 ˆ 43

pT1 N1 M0 ˆ 0 pT2a N1 M0 ˆ 18 pT2b N1±2 M0 ˆ 12 pT3 N1±2 M0 ˆ 49 pT4 N1±2 M0 ˆ 58 pT2±4 N1±3 M1 ˆ13

pT1 ˆ 7.3% pT2a ˆ 34.9% pT2b ˆ 13.3% pT3 ˆ 21.8% pT4 ˆ 20.0% M1 ˆ 2.6%

N0 ˆ 70.2%

N‡ ˆ 29.8%

Cancer factors were advanced pelvic involvement of the tumor: N3 lymph node metastasis (or bulky nodal disease) or inability to keep the native urethra. In fact, the urethra was removed only if there was intraoperative positive urethral frozen-section analysis as previously described [6]. The patients with unresectable tumors were not included in this study. Based on this exclusion criteria, 71 patients with positive pelvic nodes (50 N1 and 21 N2) had a bladder reconstruction, 67 were men and 4 women and this population formed the basis of this analysis. Mean patient age was 61.2 years (38±78 years). Pathological staging of the tumor and lymph nodes was performed according to 1997 TNM classi®cation of American Joint Committee on Cancer. The incidence of lymph node involvement increased with tumor in®ltration: 4 patients with T1, 14 with T2, 32 with T3 and 21 with T4. No patients, at the time of diagnosis demonstrated any presence of metastasis. During lymphadenectomy, limits of dissection included the inferior circum¯ex iliac vein, lateral pelvic wall, medial bladder wall and common iliac vein. Node frozen samples were analyzed by two experienced uro-pathologists. The technique used to perform cystectomy remained unchanged during the entire period. In men the urinary bladder, prostate and seminal vesicles were removed, in women the bladder resection included the uterus for two patients. All the ileal neobladders were constructed by removing 45± 50 cm long ileal segment 15 cm proximal to the ileocecal valve. The speci®c type of enteroplasty was determined primarily by the period (see Table 2): from 1981 to 1987, Camey I tubularized procedure [7] was applied (24 patients) from 1988 to 1992, Camey II procedure [8] (22 patients) and from 1992 onwards ``Z shape'' procedure (25 patients). For this last reservoir construction technique, the detubularized ileum was oriented into a Z shape with 3 cm  15 cm long limbs. The pouch was constructed with running and locking mucosa and seromuscular absorbed suture. The two extremities were ®xed to the psoas muscle. Direct implantation of the ureter was performed using an end-to-side anastomosis with an indwelling catheter for 14 days. Urinary continuity was then restored by an anastomosis of the neobladder to the membranous urethra over a 22 CH catheter. For the entire procedure (Camey I, II or Z), catheter drainage of the reservoir was maintained for 2 weeks. All the patients received post-operative adjuvant chemotherapy according to the MVAC protocol (52 patients) or other platinum based poly-chemotherapy protocol (19 patients). To avoid neobladder resorption during chemotherapy, patients had indwelling catheter bladder drainage. Table 2 Urinary diversions after cystectomy Bricker ileal conduit

Enteroplasty

Others diversions (Coffey, cutaneous ureterostomy, Kock pouch) 41

N

71

242

N‡

68

71

Total

139

313

a

According to 1997 TNM classificationÐN1: one positive lymph node <2 cm, N2: positive lymph node(s) >2 and <5 cm, N3: positive lymph node(s) >5 cm.

345

95 Camey I 74 Camey II 73 Z shape 24 Camey I 22 Camey II 25 Z shape

11

52

346

T. Lebret et al. / European Urology 42 (2002) 344±349

All the patients were followed post-operatively at 6 months intervals the ®rst 2 years and annually thereafter. Radiographic evaluation of the enteroplasty and upper urinary tracts was performed with an intravenous pyelography or abdominal/pelvic computerized tomography scans. Bladder cancer recurrence were classi®ed as local (pelvic), distant (metastasis) or both. Local recurrence occurred within the soft tissue ®eld of exenteration. Distant recurrences were de®ned as those that occurred outside the pelvis.

3. Results There were no intraoperative deaths in this series (71 patients) but 11 patients had complications (15%) requiring prolonged hospitalization, such as nonfatal pulmonary embolism, sepsis or small bowel obstruction. With a mean follow-up of 8.3 years (range 3.2±20 years, median 6.7 years), 25 patients were alive and free from disease. Continence and voiding patterns were evaluated with a standard questionnaire during follow-up. Continence was de®ned as good: no protection or a light safety pad but otherwise completely dry, satisfactory: only one pad during day and poor: more than one pad during the day or evening. Of these patients, 18 (72%) reported good continence, in contrast two (8%) reported satisfactory continence and ®ve (20%) poor continence (Table 3). Three patients required intermittent catheterization for chronic retention. Five patients are alive with recurrence (three bone metastasis, one chest metastasis and one with local recurrence), four of these patients maintain good or satisfactory continence. The remaining patient (with local recurrence) developed a retention and maintained continence by catheterizing before going to bed and upon awaking in the morning. Forty-one patients died, three of which were noncystectomy related (central nervous system event and

two myocardial failures). Of the 38 patients who died of bladder cancer, 18 were in the ®rst 2 years after cystectomy, eleven in the third year and nine after 3 years. At the time of death 13 had bone metastasis, 10 liver, 5 chest, 3 peritoneal carcinosis, 4 retroperitoneal nodes, 2 mediastinal nodes. Only one patient had a con®rmed localized recurrence. Of the 46 patients who recurred, a total of eight patients had local recurrence (one T1 primitive tumor, one T2, three T3 and three T4). For three patients ureteronephrosis required urethral catheterization with double J stent.  Four patients developed a bulky frozen pelvic recurrence and two of them required de®nitive bladder drainage until they died (2 and 6 months after the catheter was left in place). For three of these patients the plasty was invaded by tumor and neobladders held less than 100 cc; for the remaining patient, recurrence occurred at the back of the plasty with sphincter involvement, inducing a total incontinence, the patient died 6 weeks after.  For two patients, ®stula between rectum and plasty occurred. For one patient a temporary 24 Ch French catheter and a de®nitive left colostomy restored satisfactory continence. For the other patient a conversion of the plasty to Bricker ileal conduit was performed to retain a comfortable 6-months end of life.  One patient developed a rectal tumoral stenosis treated by dilatation then colostomy. This permitted good continence to be maintained.  For the remaining patient (51-year-old woman) an anterior recurrence occurred, 5 months after cystectomy. First a cutaneous ®stula appeared, and was treated with a simple appliance then tumor involvement produced a recto-plasty-cutaneous ®stula (Fig. 1). The patients died 3 weeks later.

Table 3 Continence of the 71 patientsa Continence

Camey I Patients alive free of desease

Camey II or Z Patients with tumoral recurrence

Total

Patients alive free of desease

Patients with tumoral recurrence

Good Satisfactory Poor

4 2 3

6 4 5

14 0 2

20 3 8

44 9 18

Total

9

15

16

31

71

a

Continence was defined asÐgood: no protection or a light safety pad but otherwise completely dry; satisfactory: only one pad during day; poor: more than one pad during the day or evening. For the 30 patients alive, last evaluations are reported while for the 41 patients who died, evaluation 6 months before the death are reported. Concerning continence, there is no statistical difference between the two populations.

T. Lebret et al. / European Urology 42 (2002) 344±349

347

Fig. 1. CT scan: 12 months after cystectomy and enteroplasty, a 51-year-old woman developed local recurrence with a recto-enteroplasty-cutaneous ®stula.

4. Discussion This present study attempted to clarify the issue of whether or not bladder replacement is an acceptable diversion for N‡ patient, in other words we attempted to answer a very practical and useful question which remains the subject of debate, i.e. is it justi®ed to avoid bladder reconstruction in patients with node positive bladder cancer? This can be argued because risk of recurrence is high and local tumor involvement may have altered the neobladder voiding function. The concept that positive bladder cancer patients are nor suitable for enteroplasty is a common fallacy. In fact, many urologist are reluctant to perform an orthotopic neobladder and prefer a cutaneous ureterostomy or an ileal conduit. The enigma of the appropriate management of these patients becomes more perplexing in view of the bene®t to the quality of life when enteroplasty is performed. Even if patients are informed of the possibility of creating a stoma, most patients are candidates for orthotopic neobladder. During the procedure we are often not aware of the exact in®ltration of the tumor and therefore tumor staging lacks greater precision. Therefore, the only parameter at our disposal is pelvic node status. Prior to surgery, CT scan has

limited accuracy mainly because of its inability to detect small lymph node metastasis [9]. The ®nal decision of whether or not to perform bladder reconstruction must be made during the procedure. As previously reported [10], frozen sample examination at cystectomy is reliable. However, we must point out the high speci®city (no false positive results) of routine frozen pathological examination during cystectomy, in this series, as all the positive nodes were ®nally con®rmed following pathologic analysis. Until recently, positive lymphadenectomy was considered incurable and after palliative cystectomy, ureterostomy was considered as the diversion of choice. Recent series have reported improved outcome in some patients with node positive disease [1,5,11,12]. It is probable that extended pelvic iliac lymph node dissection and adjuvant platinum based poly-chemotherapy may contribute to the long term survival of this population. Recently Ennis et al. demonstrated that M-VAC chemotherapy has a profound impact on pelvic but not on metastatic failure, [13], and despite lymph node metastasis, a 5-year survival rate, whatever the pathology staging, is generally admitted to be approximately 30% [5]. In this present series, our results were slightly better. However, in this patient cohort, we excluded N3

348

T. Lebret et al. / European Urology 42 (2002) 344±349

patients and patients with major invasive tumor involvement. As previously reported in the literature, no apparent difference was observed in the early complication rate when evaluated by the type of urinary diversion, however, this does not appear to in¯uence morbidity and mortality [5,14], as most early complications can be managed without surgery [15]. The main reason to support orthotopic bladder substitution is the improvement in quality of life [16]. Patients without an abdominal stoma have an improved body image, sexuality, sociability and global sense of well-being [17]. For positive nodes patients without recurrence similar functional results have been achieved. In fact, continence for non recurrent patients was comparable to generally accepted results [16,18,19], i.e. 80±90% of daytime continence. Our series also revealed that 72% of patients report complete daytime continence while more than 60% were continent at night. These poor results are partially explained by the limited number of patients who underwent detubularized neobladder, at the initial stages of our experience. Although this present study is retrospective and nonrandomized, its main interest was to assess the outcome of patient with recurrence. A total of 46 patients (65%) developed a bladder cancer recurrence. Most of these patients died of distance metastasis probably unrecognized at surgery, in fact 47% of deaths occurred within 2 years and 76% within 3 years. A total of 46% compare to 29% [20]. Most patients developed a distant recurrence (83%) while only 11% developed a local recurrence. In fact, it should kept in mind that even a solitary positive lymph node represents a systemic disease [10]. For patients with poor life expectancy, the bene®ts of enteroplasty, by improving quality of life, must be considered. In this series, 90.2% of the patients died with functional enteroplasty bladder substitution. These data trend to suggest that recurrence does not often affect the void-

ing of the orthotopic bladder substitution, even for patients with locally advanced disease. During chemotherapy treatment, neobladder drainage is required to avoid re-absorption by digestive mucosa of the bladder substitution. If we focus on the eight patients with local recurrence it is important to note that for three of these patients the recurrence did not alter the enteroplasty voiding function. For the remaining patients alteration (total incontinence, ®stula or retention) occurred in the last 6 months life. However, these patients had previously retained satisfactory continence. These results con®rm data previously reported by Hautmann and Simon [21]. In fact, on their series the local recurrence risk after orthotopic urinary reconstruction was 12%, and most patients were able to anticipate normal neobladder function despite recurrence. In conclusion, previous studies indicate a pelvic failure rate of about 10% [22,23], and these results are in agreement with this present study. In 71 patients of our population 11% had local recurrence, while slightly more than half of these patients developed a voiding dysfunction of the plasty during the last 6 months of their life. Moreover, these four patients had a mean 17 months life with a continent plasty. Based on these results the authors suggest that, after cystectomy, it could be justi®ed to propose an ECP to a well selected patients. In fact, patients who have small volume nodal disease and are treated with adjuvant chemotherapy have a low enough pelvic recurrence rate and a better quality of life to justify the use of a neobladder. Acknowledgements Richard Medeiros provided valuable editorial assistance.

References [1] Skinner DG. Management of invasive bladder cancer: a meticulous pelvic node dissection can make a difference. J Urol 1982;128:34±6. [2] Vieweg J, Whitmore Jr WF, Herr HW, Sogani PC, Russo P, Sheinfeld J, Fair WR. The role of pelvic lymphadenectomy and radical cystectomy for lymph node positive bladder cancer. Cancer 1994;73: 3020±8. [3] Smith JA, Whitmore Jr WA. Regional lymph node metastasis from bladder cancer. J Urol 1981;126:591±3. [4] Lebret T, Herve JM, Yonneau L, Barre P, Lugagne PM, Butreau M, et al. Study of survival after cystectomy for bladder cancer. Report of 504 cases. Prog Urol 2000;10:553±60. [5] Stein JP, Lieskovsky G, Cote R, Groshen S, Feng AC, Boyd S, et al. Radical cystectomy in the treatment of invasive bladder cancer: longterm results in 1054 patients. J Clin Oncol 2001;19:666±75.

[6] Lebret T, Herve JM, Barre P, Gaudez F, Lugagne PM, Barbagelatta M, Botto H. Urethral recurrence in transitional cell carcinoma of the bladder, predictive value of preoperative latero-montanal biopsies and urethral frozen section during prostato-cystectomy. Eur Urol 1998;33:170±4. [7] Lilien OM, Camey M. 25-year experience with replacement of the human bladder (Camey procedure). J Urol 1984;132:886±91. [8] Camey M, Richard F, Botto H. Bladder replacement by ileocystoplasty. In: King LR, Stone AR, Webster GD, editors. Bladder reconstruction and continent urinary diversion. Chicago: Year Book Medical Publisher, 1987 336±59. [9] Paik ML, Scolieri MJ, Brown SL, Spirnak JP, Resnick M. Limitations of computerized tomography in staging invasive bladder cancer before radical cystectomy. J Urol 2000;163:1693±6.

T. Lebret et al. / European Urology 42 (2002) 344±349 [10] Pollack A, Zagars GK, Cole CJ, Dinney CPN, Swanson DA, Grossman HB. The relationship of local control to distant metastasis in muscle invasive bladder cancer. J Urol 1995;154:2059±64. [11] Vieweg J, Gschwend JE, Herr HW, Fair WR. Pelvic lymph node dissection can be curative in patients with node positive bladder cancer. J Urol 1999;161:449±54. [12] Leissner J, Hohenfellner R, ThuÈroff JW, Wolf HK. Lymphadenectomy in patients with transitional cell carcinoma of the urinary bladder; signi®cance for staging and prognosis. BJU Int 2000;85: 817±23. [13] Ennis RD, Petrylak DP, Singh P, et al. The effect of cystectomy, and perioperative methotrexate, vinblastine, doxorubicin and cisplatin chemotherapy on the risk and pattern of relapse in patients with muscle invasive bladder cancer. J Urol 2000;163:1413±8. [14] Amling CL, Thrasher JB, Frazier HA, et al. Radical cystectomy for stage Ta, Tis and T1 transitional cell carcinoma of the bladder. J Urol 1994;151:31±5. [15] Figueroa AJ, Stein JP, Dickinson M, et al. Radical cystectomy in the elderly patient: an updated experience in 404 patients. Cancer 1998;83:141±7. [16] Hollowell CMP, Christiano AP, Steinberg GD. Technique of Hautmann ileal neobladder with chimney modi®cation: interim results in 50 patients. J Urol 2000;163:47±51.

Editorial Comment

R.E. Hautmann, Ulm, Germany One of the initial deterrents of orthotopic diversion was the risk for a local recurrence of cancer. The question, whether a meticulous lymph node dissection is an option or a must in bladder cancer surgery has meanwhile been answered. The experience of recent decades shows increased survival with suf®cient removal of lymph nodes in bladder cancer. A meticulous lymph node dissection makes a difference and indeed the survival rate as result of this procedure in node positive bladder cancer was 35%.

349

[17] Weijerman PC, Schurmans JR, Hop WC, et al. Morbidity and quality of life in patients with orthotopic and heterotopic continent urinary diversion. Urology 1998;51:51±6. [18] Studer UE, Danuser H, Merz V, et al. Experience in 100 patients with an ileal low pressure bladder substitute combined an afferent tubular isoperistaltic segment. J Urol 1995;154:49±56. [19] Blute ML, Gburek BM. Continent orthotopic urinary diversion in female patients: early Mayo Clinic experience. Mayo Clin Proc 1998;73:501±7. [20] Mills RD, Turner WH, Fleischmann A, Markwalder R, Thalman GN, Studer UE. Pelvic lymph nodes metastases from bladder cancer: outcome in 83 patients after radical cystectomy and pelvic lymphadenectomy. J Urol 2001;166:19±23. [21] Hautmann RE, Simon J. Ileal neobladder and local recurrence of bladder cancer: patterns of failure and impact on function in men. J Urol 1999;162:1963±6. [22] Frazier HA, Robertson JE, Dodge RK, et al. The value of pathological factors in predicting cancer-speci®c survival among patients treated with radical cystectomy for transitional cell carcinoma of the bladder and prostate. Cancer 1993;71:3993±4001. [23] Schoenberg MP, Walsh PC, Breazeale DR. Local recurrence and survival following nerve sparing radical cystoprostatectomy for bladder cancer: 10-year follow-up. J Urol 1996;155:490±4.

The local recurrence rate following orthotopic urinary reconstruction is 10%. Survival following diagnosis of local recurrence is limited despite multimodality therapy. However, most patients can anticipate normal neobladder function even in the presence of recurrent disease or until the time of death. Thus, the use of orthotopic diversion after cystectomy in patients with locally advanced bladder cancer, including macroscopically positive lymph nodes, is safe. The authors con®rm the important experience other reported in selected cases.