Oncological Evaluation of Prostate Sparing Cystectomy: The Montsouris Long-Term Results

Oncological Evaluation of Prostate Sparing Cystectomy: The Montsouris Long-Term Results François Rozet,* Guillaume Lesur, Xavier Cathelineau, Eric Barret, Gordon Smyth, Shawn Soon and Guy Vallancien From the Department of Urology, Institut Mutualiste Montsouris and Université Paris Descartes, Paris, France

Purpose: Prostate sparing cystectomy provides an alternative therapeutic option in highly selected patients with invasive bladder cancer who wish to avoid the significant functional side effects of traditional surgery. Concern exists regarding the oncological safety of this technique especially with regard to the presence of prostatic urothelial cancer and incidental prostate adenocarcinoma. We present the long-term oncological outcomes of a large series of patients treated at a single institution. Materials and Methods: Between October 1992 and June 2004 a total of 117 patients were selected for prostate sparing cystectomy after meeting the inclusion criteria. Results: Nine patients underwent radical cystoprostatectomy after prostate urothelial carcinoma was diagnosed intraoperatively. Long-term oncological results (2 years minimum followup [mean 55 months]) are presented here for the remaining patients. The rate of local and distant recurrence was 4.7% and 34%, respectively, at 20 months. Overall survival at 5 years in our study was 67%, and 5-year disease-free survival rates were 77% for pT2 N0 or less, 44% for pT3 N0 or greater and 22% for pN⫹ disease. Of 6 patients found to have prostate adenocarcinoma in transurethral prostate resection specimens 1 was treated with high intensity focused ultrasound and 5 were followed with active surveillance, 2 of whom later died of bladder cancer. Conclusions: We report oncological data on the largest prospective cohort of patients, with the longest followup, treated by prostate sparing cystectomy to date. Outcomes are comparable with the largest published series of cystoprosatectomies. With appropriate screening the risk of a clinically significant prostate cancer appears to be low. For certain selected patients this technique represents a valuable additional option for treatment. Key Words: urinary bladder neoplasms, cystectomy, treatment outcome, prostatic neoplasms, prostatectomy

adical cystoprostatectomy remains the definitive surgical management of muscle invasive bladder cancer. However, this treatment carries important functional consequences. The rates of daytime and nighttime incontinence with orthotopic bladder substitution are 5% to 18% and 13% to 27%, respectively, and erectile dysfunction occurs in 75% to 80% of patients.1 Preservation of the neurovascular bundles, as demonstrated by Walsh and Donker during radical prostatectomy, reduces these side effects.2 In the early 1990s Schilling and Friesen proposed transprostatic cystectomy to preserve the neurovascular bundles and the external sphincter.3 In an attempt to improve functional outcomes in terms of urinary and sexual function without compromising oncological principles, we have preserved the prostatic capsule in selected patients with a low risk of prostate cancer or invasion of the prostate by bladder tumor.4 The long-term oncological efficacy of this technique remains controversial.5 We present the oncological outcomes of patients treated with prostate sparing cystectomy for transitional cell carcinoma

R

Submitted for publication October 5, 2007. * Correspondence: Department of Urology, Institut Mutualiste Montsouris, 42 Boulevard Jourdan, 75014, Paris (telephone: 00 33 1 56 61 66 20; FAX: 00 33 1 56 61 66 38; e-mail: [email protected]).

0022-5347/08/1796-2170/0 THE JOURNAL OF UROLOGY® Copyright © 2008 by AMERICAN UROLOGICAL ASSOCIATION

of the bladder at a single institution with a minimum of 2 years of followup. MATERIALS AND METHODS From October 1992 to June 2004, 131 patients fitting the inclusion criteria for prostate sparing cystectomy were treated at the Institut Mutualiste Montsouris. To exclude bias 10 patients who received neoadjuvant therapy were not included in analysis. An additional 4 patients with nonurothelial tumors were also excluded from study. The inclusion criteria are listed in the Appendix. The majority of patients (67%) underwent TURP at the time of cystectomy while 16% had the TURP several days before surgery. Toward the end of the study TURP was replaced by simple adenomectomy, initially open (Millins prostatectomy) (3%) then laparoscopic (14%). A full description of our surgical technique of prostate sparing cystectomy has been previously published.4 The TNM 2002 classification was used to evaluate the pathological stage of the bladder tumor and lymph nodes. Followup analysis was performed using 3 pathological groups of localized tumor (pT2 or less, pN0), locally advanced disease (greater than pT2, pN0) and lymph node positive disease (pN⫹). In the case of no residual tumor (pT0, 11 of 108 patients), the tumor was classified according to the highest transurethral bladder tumor resection stage.

2170

Vol. 179, 2170-2175, June 2008 Printed in U.S.A. DOI:10.1016/j.juro.2008.01.112

ONCOLOGICAL EVALUATION OF PROSTATE SPARING CYSTECTOMY The patients were followed every 3 months for the first year, then every 6 months for the next 2 years and yearly thereafter. The clinical evaluation included history, physical examination, routine blood work including PSA, urinalysis and radiological evaluation including excretory urogram, computerized tomography of the thorax, abdomen and pelvis and isotope bone scan if indicated. All data were collected prospectively in a computerized database (Siemens-Clinicom®). Bladder cancer recurrences were classified as local or distant. Local recurrence included the surgical bed or pelvic lymph nodes, or disease relapse in the prostate/urethra. Distant relapse involved the lymph nodes above the iliac bifurcation, visceral organs or bones.6 Patients who had local and distant recurrences were allocated to the latter group.7 Statistical analysis was performed using the SPSS® program. The primary end points were overall survival, cancer specific survival, disease-free survival and metastasis-free survival which were described using Kaplan-Meier curves. Survival differences were compared according to pT stage by the log-rank test. Statistical significance was set at p ⱕ0.05. Multivariate analysis was done using Cox regression analysis. Tests were 2-sided. RESULTS A total of 117 patients with urothelial carcinoma were treated with prostate sparing cystectomy without neoadjuvant therapy. Perioperative frozen sections demonstrated prostatic urothelial carcinoma in 9 patients, all of whom were treated with radical cystoprostatectomy. Of these cases the final pathology revealed no evidence of prostatic urothelial carcinoma in 1 patient. Four patients had CIS in the prostatic urethra and 2 had invasion of the prostatic stroma by urothelial carcinoma. Two patients had prostate adenocarcinoma. Therefore, 108 patients were treated with prostate sparing cystectomy. Two-year minimum followup is presented in this study. Two patients were ultimately lost to followup, but only after 24.8 and 65.4 months without recurrence, respectively. The patient demographics and pathological stages are presented in table 1. Mean postoperative followup was 55 ⫾ 3.6 months (range 3 to 146.2). One patient died of sepsis 3 months after surgery. Overall 44 patients died. Of these 37 died of cancer and 7 of other causes. There were 41 patients (38.7%) who presented with local or distant recurrence (table 2). Five patients (4.7%) presented with local recurrence at a mean of 44 months from

2171

TABLE 2. Local and metastatic recurrence Local Recurrence No./total No. stage (%): pTis-a-1N0 pT2N0 pT3–4N0 pTN⫹

1/21 2/41 2/29 0/15

Totals Mean mos to recurrence (range)

(5) (5) (7)

2/21 9/41 14/29 11/15

5/106 (4.7) 44 (5–78)

No. TNM stage (%): Overall pTa/pTis pT1 pT2 pT3 pT4a No. grade (%): 1 2 3 Mean pt age ⫾ SD Mean mos followup ⫾ SD

surgery. One patient with initial pT4 pathology presented with a pelvic recurrence at 30 months postoperatively and was treated with salvage chemotherapy. Three patients (pT2 ⫻ 2, pT4 ⫻ 1) presented with pelvic lymph node metastases at 36 months and were also treated with salvage chemotherapy. One patient (pTa) had local recurrence at the prostatic urethra at 40 months postoperatively and was treated with transurethral resection. This patient has had no further recurrence 10 years after surgery. Distant metastases developed in 36 patients (34%) with a mean delay of 20 months. The location of the metastases were lymph nodes (above the iliac bifurcation) (26%), liver (20%), lung (16%), peritoneal carcinomatosis (11%), bone (9%), brain (7%), cutaneous (4%), mediastinal (3%), cavernous (3%) and splenic (1%). The overall 5-year survival rate for all patients was 67%. Cancer specific survival was 71%. Disease-free survival was 77% for patients with pT2 N0 or less disease (79% for pTisa-1 N0 and 76% for pT2 N0), 44% for those with pT3 N0 or greater disease and 22% for those with pN⫹ disease (see figure). The only statistically significant predictor of survival at 5 years was lymph node status (80 ⫾ 5 months for pN negative disease vs 19 ⫾ 11 for pN positive disease, p ⬍0.00001). There was no significant difference in survival due to the presence of CIS (68 ⫾ 9 vs 66 ⫾ 6 months, p ⫽ 0.7). On multivariate analysis the only independent

1,0

0,8

CUMULATIVE

0,6

≤pTa-is-1-2 N0

0,4

Overall

pN⫹

108 (100) 14 (13) 8 (7) 42 (39) 41 (38) 3 (3)

15 (14) 0 0 1 (1) 13 (12) 1 (1)

10 (9) 21 (20) 77 (71) 62 ⫾ 9 55 ⫾ 3.6

(10) (22) (48) (73)

36/106 (34) 20 (2–67)

SURVIVAL

TABLE 1. Patient characteristics

Metastatic Recurrence

≥pT3-4 N0 0,2

N+ p=0.00001

0,0 0

30

60

90

120

MONTHS AFTER CYSTECTOMY

Disease-free survival by stage

150

2172

ONCOLOGICAL EVALUATION OF PROSTATE SPARING CYSTECTOMY

predictor of recurrence was the presence of multifocal tumors (Cox regression analysis p ⫽ 0.01). Mean preoperative PSA was 2.2 ⫾ 2 ng/ml. Prostate adenocarcinoma was diagnosed in 6 patients (5%) during the final pathological examination of the TURP specimens which was not seen on the initial, intraoperative frozen section examination. The Gleason score was 6 (3 ⫹ 3) for this entire cohort. One patient was treated with HIFU and PSA was less than 0.2 ng/ml after 59 months of followup. The other 5 patients were followed with active surveillance and of these 2 died of bladder cancer at 33 and 47 months, respectively, postoperatively. The remaining 3 patients are alive with a median followup of 34 months. A further 3 patients were newly diagnosed with prostate cancer during followup at 39, 63 and 76 months, respectively. In all 3 cases the prostate cancer was Gleason grade 6 (3 ⫹ 3). One patient underwent treatment with HIFU, the second was treated with brachytherapy and the last with intermittent hormonal therapy. In the remainder of the cohort the mean PSA was 1.62 ng/ml at 5 years of followup and 1.53 ng/ml at 10 years.

(tables 2 and 3).7,12,13 In our series the rate of local recurrence was 4.7%. The rate of distant recurrence after radical cystoprostatectomy varies between 13% and 35%.7,12,13 Hautmann et al reported an 18% distant recurrence rate with a mean survival of 12 months.12 In all, 46% did not have local recurrence in conjunction with their metastasis. Maderbascher et al found the rate of distant metastasis to be 32%.7 In patients with pT2 N0 disease the rate of distant recurrence was 13% as reported by Stein et al and 25% in the Maderbascher et al series.7,13 In our series the rate of distant metastasis was 34% at 20 months and the rate of 22% for patients with pT2 N0 is comparable to other cystoprostatectomy series. In the series of Botto et al, which included 34 prostate sparing cystectomies with a followup of 26 months, 7 patients (21%) presented with recurrence.10 Of 7 patients 6 (pT2 N0 or less ⫻ 4, pT3 N0 or greater ⫻ 1, pN⫹ ⫻ 1) presented with metastasis without local recurrence.

Prostate sparing cystectomy is a therapeutic option in highly selected patients with muscle invasive bladder cancer. The main advantage of this procedure when compared to the standard of care, radical cystoprostatectomy, is the improvement in functional outcomes.4,8 –11 The cancer control efficacy of this technique remains controversial and is derived from a few small series with limited followup. The importance of the current series is derived from the large number of patients (106), with a minimum of 2 years followup. The mean followup is 55 months and this in combination with the limited number of patients lost to followup (2) allows for meaningful analysis. The heterogeneous nature of the published long-term results of cystoprostatectomy makes direct comparison to prostate sparing surgery difficult. Between 10% and 20% of cystectomies were in females and 16% to 48% of patients received neoadjuvant therapy. Variable subgroup divisions between 1 study and another also make comparison difficult.7,12,13

Survival The overall survival reported in large series of cystoprostatectomies varies between 58% and 66% after 5 years (table 3).7,12,13 This is comparable to the overall survival of 67% reported in our series of prostate sparing cystectomy. In 2001 Stein et al reported a series of 1,054 patients, of which 20% were women and 15% received neoadjuvant chemotherapy.13 The 5-year disease-free survival was 85% for stages pT2 N0 or less, 58% for stages pT3 N0 or greater and 35% for pN⫹. Two large series of cystoprostatectomies without neoadjuvant treatment have recently been reported. Madersbacher et al reported on a series of 507 patients (21% women) with a mean followup of 45 months.7 The 5-year disease-free survival was 73% for stages pT2 N0 or less, 56% for stages pT3 N0 or greater and 33% for pN⫹. Hautmann et al reported on a series of 788 patients (17% women) with a mean followup of 54 months.12 The 5-year disease-free survival was 91%, 76%, 58%, 40% and 21% for stages pTais-1, pT2, pT3, pT4 and pN⫹, respectively. These results compare favorably with our 5-year survival without recurrence of 77% for stages pT2 N0 or less (79% for pTa-is-1, 76% for pT2), 44% for stages pT3 N0 or greater and 22% for stage pN⫹.

Local and Distant Recurrence Local recurrences appear mostly during the first 2 years after radical cystoprostatectomy. The rate of local recurrence overall in the published literature is less than 10%

Prostate Urothelial Cancer Risk Pettus et al reported an incidence of prostatic urothelial carcinoma of 33% in a series of 235 patients who underwent cystoprostatectomy for invasive bladder cancer.14 Of these

DISCUSSION

TABLE 3. Oncological results

No. pts Mean mos followup No. locoregional recurrence (%) Mean mos to recurrence No. metastatic recurrence (%) Mean mos to recurrence Total recurrent disease (%) % 5-Yr global survival % 5-Yr disease-free survival % pT2N0 or less: pTis-a-1 pT2 % pT3N0 or greater % pN⫹

Present Series

Stein et al13

Madersbacher et al7

Hautmann et al12

106 55 5 (4.7) 44 36 (34) 20 41 (38.7) 67 60 77 79 76 44 22

1,054 122 77 (7) 18 234 (22) 12 311 (30) 66 68 85

507 45 40 (8) Not significant 179 (35) Not significant 219 (43) 59 62 73

788 54 73 (9) 7 140 (18) 12 239 (30) 58 68

58 35

56 33

91 76 58 (pT3), 40 (pT4) 21

ONCOLOGICAL EVALUATION OF PROSTATE SPARING CYSTECTOMY

2173

77 patients 64% had stromal invasion while the remaining 36% had CIS of the urethra and/or the periurethral/peripheral prostatic ducts only. On multivariate analysis carcinoma in situ of the bladder and trigone/bladder neck involvement were significantly associated with prostatic urothelial carcinoma in the final specimen. In our series the entire prostatic urethra as well as the prostatic transitional zone were completely removed by TURP or simple prostatectomy at surgery and sent for frozen section to exclude any transitional cell carcinoma/carcinoma in situ at these sites. The frozen section examination result excluded 9 patients in the current series and highlights the crucial role of the pathologist in interpreting these frozen section preparations. Frozen section examination was also performed on the prostatic base of the resected cystectomy specimen to confirm negative surgical margins. In their series of 118 cystoprostatectomies for transitional cell carcinoma of the bladder, Lebret et al demonstrated that when urethral margins were negative on frozen section no recurrence was observed after a 10-year minimum followup.15

comes. It is also important to note that extended lymph node dissections were not performed in this patient cohort in keeping with the practice of our unit at the time. All patients underwent standard pelvic lymph node dissection only. Our final pathological results show 3 pT4a tumors which perhaps points to potential limitations of intraoperative frozen section examination.

Prostate Adenocarcinoma Risk The risk of synchronous prostate adenocarcinoma varies from 11% to 48% in the reported series.14,16 –20 However, in these series prostate cancer screening was not performed preoperatively. Furthermore, the only significant variable associated with the risk of prostate cancer was patient age, with older patients in the cystoprostatectomy series than in the prostate sparing cystectomy series.14 Not all prostate cancer detected in the cystoprostatectomy series are clinically significant. In 1989 Kabalin et al reported an incidence of occult prostate cancer of 38% in men with bladder cancer, but only 13% of these cancers had a tumor volume greater than 0.1 cc.17 Revelo et al found occult prostate cancer in 41% (50 of 121) of unscreened cystoprostatectomies, of which less than half (48%) were clinically insignificant.20 In our series 9 patients were diagnosed with prostate adenocarcinoma. In 6 cases the final pathological examination found a Gleason grade 6 adenocarcinoma of the prostate, missed during the frozen section analysis, and in 3 cases the tumor was diagnosed during followup. The false-negative cases during frozen section may perhaps be explained by the fact that the primary focus at frozen section was to exclude urothelial carcinoma and, therefore, that small foci of prostate cancer may have been overlooked. The risk of development of a significant prostate cancer in our series of prostate sparing cystectomy appears to be relatively low. This is undoubtedly due to the preoperative screening of the patients with total and free PSA, DRE and TRUS examination. Systematic biopsies were not done in this series but they remain an option in the future, for example in patients with a PSA greater than 2.5 ng/ml, to optimize the selection process. When detected during followup the prostate cancers were all Gleason grade 6 and could all be treated effectively. Furthermore, for patients with bladder cancer life expectancy is reduced compared to the general population (overall survival of 67% after 5 years in this series) and is certainly linked more to the bladder cancer than the prostate cancer. We recognize that the present study has some limitations. Our patient cohort included 10 patients (9%) who had grade 1 disease and 21 patients (20%) who had grade 2 disease, potentially causing bias toward more favorable out-

APPENDIX

CONCLUSIONS The technique of prostate sparing cystectomy is controversial due to the paucity of information available concerning its oncological efficacy. To our knowledge this study reports the largest series published to date with the longest followup. It demonstrates oncological results comparable to those of the largest series of cystoprostatectomies. While the gold standard in treating muscle invasive bladder cancer remains cystoprostatectomy, we demonstrate that prostate sparing cystectomy is an additional option for treating highly selected patients who want to be offered curative therapy with minimal side effects.

Inclusion Criteria for Prostate Sparing Cystectomy Preoperative criteria

Perioperative criteria

Full informed consent Normal DRE Normal TRUS PSA less than 4 ng/dl Free-to-total PSA ratio greater than 15% or TRUS biopsy negative No prostatic urothelial carcinoma (on frozen section of TURP or simple prostatectomy specimen) Intraoperative frozen section of prostate base, distal ureters and trigone

Abbreviations and Acronyms CIS DRE HIFU PSA TRUS TURP

⫽ ⫽ ⫽ ⫽ ⫽ ⫽

carcinoma in situ digital rectal examination high intensity focused ultrasound prostate specific antigen transrectal ultrasound transurethral prostate resection

REFERENCES 1.

Turner WH, Bitten A and Studer UE: Reconstruction of the urinary tract after radical cystectomy: the case for continent urinary diversion. Urology 1997; 49: 663. 2. Walsh PC and Donker PJ: Impotence following radical prostatectomy: insight into etiology and prevention. J Urol 1982; 128: 492. 3. Schilling A and Friesen A: Transprostatic selective cystectomy with an ileal bladder. Eur Urol 1990; 18: 253. 4. Vallancien G, Abou El Fettouh H, Cathelineau X, Baumert H, Fromont G and Guillonneau B: Cystectomy with prostate sparing for bladder cancer in 100 patients: 10-year experience. J Urol 2002; 168: 2413. 5. Hautmann RE and Stein JP: Neobladder with prostatic capsule and seminal-sparing cystectomy for bladder cancer: a step in the wrong direction. Urol Clin North Am 2005; 32: 177. 6. Yossepowitch O, Dalbagni G, Golijanin D, Donat SM, Bochner BH, Herr H et al: Orthotopic urinary diversion after cys-

2174

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

ONCOLOGICAL EVALUATION OF PROSTATE SPARING CYSTECTOMY

tectomy for bladder cancer: implications for cancer control and patterns of disease recurrence. J Urol 2003; 169: 177. Madersbacher S, Hochreiter W, Burkhard F, Thalmann GN, Danuser H and Markwalder R: Radical cystectomy for bladder cancer today–a homogeneous series without neoadjuvant therapy. J Clin Oncol 2003; 21: 690. Muto G, Bardari F, D’Urso L and Giona C: Seminal sparing cystectomy and ileocapsuloplasty: long-term followup results. J Urol 2004; 172: 76. Colombo R, Bertini R, Salonia A, Naspro R, Ghezzi M and Mazzoccoli B: Overall clinical outcomes after nerve and seminal sparing radical cystectomy for the treatment of organ confined bladder cancer. J Urol 2004; 171: 1819. Botto H, Sebe P, Molinie V, Herve JM, Yonneau C and Lebret T: Prostatic capsule- and seminal-sparing cystectomy for bladder carcinoma: initial results for selected patients. BJU Int 2004; 94: 1021. Horenblas S, Meinhardt W, Ijzerman W and Moonen LF: Sexuality preserving cystectomy and neobladder: initial results. J Urol 2001; 166: 837. Hautmann RE, Gschwend JE, de Petriconi RC, Kron M and Volkmer BG: Cystectomy for transitional cell carcinoma of the bladder: results of a surgery only series in the neobladder era. J Urol 2006; 176: 486. Stein JP, Lieskovsky G, Cote R, Groshen S, Feng AC and Boyd S: Radical cystectomy in the treatment of invasive bladder cancer: long-term results in 1,054 patients. J Clin Oncol 2001; 19: 666. Pettus JA, Al-Ahmadie H, Barocas DA, Koppie TM, Herr H, Donat SM et al: Risk assessment of prostatic pathology in patients undergoing radical cystoprostatectomy. Eur Urol 2008; 53: 370. Lebret T, Herve JM, Barre P, Gaudez F, Lugagne PM, 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 1998; 33: 170. Pritchett TR, Moreno J, Warner NE, Lieskovsky G, Nichols PW and Cook BA: Unsuspected prostatic adenocarcinoma in patients who have undergone radical cystoprostatectomy for transitional cell carcinoma of the bladder. J Urol 1988; 139: 1214. Kabalin JN, McNeal JE, Price HM, Freiha FS and Stamey TA: Unsuspected adenocarcinoma of the prostate in patients undergoing cystoprostatectomy for other causes: incidence, histology and morphometric observations. J Urol 1989; 141: 1091. Montie JE, Wood DP Jr, Pontes JE, Boyett JM and Levin HS: Adenocarcinoma of the prostate in cystoprostatectomy specimens removed for bladder cancer. Cancer 1989; 63: 381. Zerbib M and Bouchot O: Prostate cancer incidence on specimen of cystoprostatectomy for infiltrative bladder cancer. Prog Urol 2005; 15: 1262. Revelo MP, Cookson MS, Chang SS, Shook MF, Smith JA Jr and Shappell SB: Incidence and location of prostate and urothelial carcinoma in prostates from cystoprostatectomies: implications for possible apical sparing surgery. J Urol 2004; 171: 646.

EDITORIAL COMMENT This article describes the long-term oncological results of a novel approach to radical cystectomy whereby the prostate is largely spared. The surgical approach was crafted to optimize functional recovery (of potency and continence), and maintain cancer recurrence and mortality rates equivalent to a more conventional approach to radical cystectomy with

complete removal of the prostate. This innovative approach was initiated in 1992 and 133 patients were accrued during a 12-year period. The authors report 67% probability of 5-year overall survival, and local and distant recurrence rates of 4.7% and 34%, respectively. The authors conclude that these outcomes are comparable to the largest radical cystectomy series reported to date. There are several limitations and cautions that merit discussion regarding the hypothesis testing, surgical technique and data analysis. The rationale for performing a prostate and seminal vesicle sparing cystectomy is to preserve potency and optimize continence following orthotopic neobladder reconstruction. Although not the subject of the current article, the authors previously assessed these outcomes using a mailed questionnaire and reported these results in their first 100 patients in 2002 (reference 4 in article). No details were provided regarding response rates (were all of the data derived from the results of the questionnaire or completed by other means?), validation of the instrument or the instruments used. They reported that of the 61 patients who were potent preoperatively 82% maintained potency defined as unassisted intercourse. Continence was defined as dry with no pads, and was reported as 97% during the day and 98% at night with nocturia 1 to 3 times at 1-year followup based on 88 of the 100 patients included in the series. These phase II data are encouraging yet raise some questions. Why would one offer prostate sparing surgery to a patient with preexisting erectile dysfunction? Does prostate sparing optimize continence? Does it increase the probability of urinary retention requiring intermittent catheterization? A randomized clinical trial comparing prostate sparing to a conventional nerve sparing approach with complete removal of the prostate or perhaps an alternative approach such as apical prostate sparing or socalled prostate capsule sparing can answer these questions. The Montsouris technique involves dissection along the posterior bladder wall anterior to the seminal vesicles then transecting the prostate distal to the bladder neck, maintaining at least a 5 mm margin from the tumor which is usually enough for a good safety margin (reference 4 in article). This raises concern about using this approach in a patient with a posterior based cancer and the potential for involvement of the perivesical lymph nodes. The local only recurrence rate was low (4.7%) and comparable to radical cystoprostatectomy. However, this may understate the true incidence of local recurrence as patients with local recurrence and distant recurrence were lumped into the distant recurrence group but only the sites of distant recurrence were reported. This technique necessarily involves opening of the bladder and the resected prostatic fossa which may result in contamination of the pelvis with urine containing exfoliated malignant cells. Could the same functional results be obtained without the need for TURP and meticulous dissection of the neurovascular bundle within the lateral prostatic fascia? The incidence of prostatic transitional cell carcinoma and prostatic adenocarcinoma is 29% to 43% and 41%, respectively, when evaluated using a whole-mount step section technique (reference 20 in article).1 Therein lies the major concern with leaving the majority of the prostate intact. TURP with frozen section will identify some but not all of the patients with transitional cell carcinoma of the prostatic urothelium. We described several patterns of involvement of

ONCOLOGICAL EVALUATION OF PROSTATE SPARING CYSTECTOMY prostatic transitional cell carcinoma and found that of the 32% of patients identified with prostatic transitional cell carcinoma the majority (57%) had stromal invasion, and 39% of these patients had either direct penetration from the bladder tumor through the bladder neck or posterior penetration with periprostatic and seminal vesicles.2 These patients are not always identified by TURP and this prostate sparing technique would pose a significant risk of dissecting across the tumor or at worst a positive surgical margin. This prostate sparing technique has been reported by others for nonmuscle invasive bladder cancer, nonurothelial malignancy and benign conditions, eg interstitial cystitis.3,4 In patients with benign conditions or those with refractory nonmuscle invasive cancer desiring to maintain fertility the Montsouris technique can accomplish oncological control while preserving excellent functional recovery. For a broader application to muscle invasive transitional cell carcinoma, the data from the present study are valuable from a technical and functional outcome standpoint but require more rigorous testing with phase III trials. Seth P. Lerner Scott Department of Urology Baylor College of Medicine Houston, Texas 1.

Shen SS and Lerner SP: Prostatic transitional cell carcinoma: pathologic features and clinical management. Expert Rev Anticancer Ther 2007; 7: 1155. 2. Shen SS, Lerner SP, Muezzinoglu B, Truong LD, Amiel G and Wheeler TM: Prostatic involvement by transitional cell carcinoma in patients with bladder cancer and its prognostic significance. Hum Pathol 2006; 37: 726. 3. Colombo R, Bertini R, Salonia A, Da Pozzo LF, Montorsi F, Brausi M et al: Nerve and seminal sparing radical cystectomy with orthotopic urinary diversion for select patients with superficial bladder cancer: an innovative surgical approach. J Urol 2001; 165: 51. 4. Spitz A, Stein JP, Lieskovsky G and Skinner DG: Orthotopic urinary diversion with preservation of erectile and ejaculatory function in men requiring radical cystectomy for nonurothelial malignancy: a new technique. J Urol 1999; 161: 1761.

2175

REPLY BY AUTHORS The aim of our study was to provide the long-term oncological results of our series of highly selected patients treated with prostate sparing cystectomy for bladder cancer. We deliberately did not include functional data as these have been previously published (reference 4 in article). With regard to defining local and distant recurrences, we have strictly adhered to the nomenclature used in the reporting of major radical cystoprostatectomy series. We define local recurrence as including the surgical bed or pelvic lymph nodes, or disease relapse originating in the urethra or prostate, according to the definition of Yossepowitch et al (reference 6 in article). We also used their definition of distant/metastatic recurrence as involving the lymph nodes above the iliac bifurcation, visceral organs or bones. In common with Stein et al, “bladder cancer recurrences were classified as either local or distant. Patients who had both local and distant recurrences were allocated to the latter group” (reference 13 in article). We devoted a significant amount of our discussion to issues of prostatic urothelial carcinoma and prostate adenocarcinoma. Again we contend that the reason for the lower incidence of prostate adenocarcinoma in our patient cohort is that it represents a highly selected and screened population, distinct from those described in standard reports of radical cystoprostatectomy. The previous article from our institution on this technique states that of the 165 males who were candidates for cystectomy “57 (39%) were not candidates for prostate sparing due to associated prostate cancer, transitional cell carcinoma of the prostatic urethra or bladder tumor invading the prostate” (reference 4 in article). We acknowledge the concerns raised by this controversial and novel technique, which we advocate for only the most highly selected patients. We agree that a randomized clinical trail would answer such questions but suggest that, in terms of the numbers required for significance, this might only be feasible within the context of a major multi-center study. We hope that the publication of these results, which to our knowledge represent the longest oncological follow-up to date for this type of surgery, will add to the debate on the management of bladder cancer.