Randomized Trial of Studer Pouch versus T-Pouch Orthotopic Ileal Neobladder in Patients with Bladder Cancer Eila C. Skinner,* Adrian S. Fairey, Susan Groshen, Siamak Daneshmand, Jie Cai, Gus Miranda and Donald G. Skinner From the Department of Urology, Stanford University, Stanford (ECS), and USC Institute of Urology, Keck Medical Center of USC, Los Angeles (SG, SD, JC, GM, DGS), California, and Department of Urology, University of Alberta, Edmonton, Canada (ASF)

Purpose: The need to prevent reflux in the construction of an orthotopic ileal neobladder is controversial. We designed the USC-STAR trial to determine whether the T-pouch neobladder that included an antireflux mechanism was superior to the Studer pouch in patients with bladder cancer undergoing radical cystectomy. Materials and Methods: This single center, randomized, controlled trial recruited patients with clinically nonmetastatic bladder cancer scheduled to undergo radical cystectomy with neobladder. Eligible patients were randomly assigned to undergo T-pouch or Studer ileal orthotopic neobladder. Treatment assignment was not masked. The primary end point was change in renal function from baseline to 3 years. The CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) equation was used to calculate the estimated glomerular filtration rate. Results: Between February 2002 and November 2009, 237 patients were randomly assigned to T-pouch ileal orthotopic neobladder and 247 to Studer ileal orthotopic neobladder. Baseline characteristics did not differ between the groups. Between baseline and 3 years the estimated glomerular filtration rate decreased by 6.4 ml/minute/1.73 m2 in the Studer group and 6.6 ml/minute/1.73 m2 in the T-pouch group (p¼0.35). Multivariable analysis showed that type of ileal orthotopic neobladder was not independently associated with 3-year renal function (p¼0.63). However, baseline estimated glomerular filtration rate, age and urinary tract obstruction were independently associated with 3-year decline in renal function. Cumulative risk of urinary tract infection and overall late complications were not different between the groups, but the T-pouch was associated with an increased risk of secondary diversion related surgeries. Conclusions: T-pouch ileal orthotopic neobladder with an antireflux mechanism did not prevent a moderate reduction in renal function observed at 3 years compared to the Studer pouch, but did result in an increase in diversion related secondary surgical procedures.

Abbreviations and Acronyms BC ¼ bladder cancer eGFR ¼ estimated glomerular filtration rate GFR ¼ glomerular filtration rate ONB ¼ ileal orthotopic neobladder SRI ¼ surgical re-intervention UDSRI ¼ urinary diversion related surgical re-intervention USC ¼ University of Southern California UTI ¼ urinary tract infection Accepted for publication March 6, 2015. Presented at annual meeting of American Urological Association, San Diego, California, May 4-8, 2013. Study received institutional review board approval. * Correspondence: Department of Urology, Stanford University, 300 Pasteur Drive, Suite S287, Stanford, California 94305-5118 (telephone: 650-724-3332; FAX: 650-723-4055; e-mail: [email protected]).

Key Words: urinary diversion, cystectomy

RADICAL cystectomy with pelvic lymph node dissection and urinary diversion is a standard treatment for muscle invasive and high risk nonmuscle invasive bladder cancer.1 Ileal orthotopic

neobladder has become an accepted and in some institutions is the preferred form of urinary diversion after radical cystectomy. However, there is controversy about the need to provide

0022-5347/15/1942-0433/0 THE JOURNAL OF UROLOGY® Ó 2015 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC.

http://dx.doi.org/10.1016/j.juro.2015.03.101 Vol. 194, 433-440, August 2015 Printed in U.S.A.

www.jurology.com

j

433

434

STUDER POUCH VERSUS T-POUCH ORTHOTOPIC ILEAL NEOBLADDER

an antireflux mechanism as part of the construction of the neobladder. Long-term kidney damage observed with ileal conduit diversion has been presumed to be in part due to reflux, and early studies suggested continent diversion with antireflux techniques was associated with less upper tract scarring and bacteruria.2,3 However, Song4 and Hautmann5 et al found no differences in renal function or hydronephrosis in refluxing or nonrefluxing ONB in nonrandomized retrospective studies. In addition, past efforts to prevent reflux have been associated with the development of upper tract obstruction as confirmed in a within-patient randomized study of 80 patients by Shabaan et al.5e7 We designed the USC-STAR trial to determine whether the T-pouch with an antireflux mechanism would result in less renal function decline at 3 years compared to the Studer pouch in patients with BC undergoing radical cystectomy.

METHODS Study Design and Participants USC-STAR was a prospective, single center, randomized, phase 3 trial comparing T-pouch with Studer ONB in patients with BC. Patients were recruited between February 2002 and November 2009. Eligible patients were scheduled to undergo radical cystectomy for clinical nonmetastatic BC, and were deemed candidates for orthotopic urinary diversion by the treating surgeon using accepted criteria. Prior pelvic radiation therapy was allowed. Patients were excluded from study if they were undergoing cystectomy for benign disease or any malignancy other than BC, and if they had a history of other malignancy within 5 years except nonmelanoma skin cancer, low risk prostate cancer or treated in situ cervical cancer. Approximately 10% of patients were unable to have a neobladder because of a positive frozen section urethral margin and they were not included in the study. The institutional review board at the University of Southern California approved the study. All participants provided written informed consent. An interim analysis for safety was performed after the initial 100 patients were accrued.

Randomization and Masking Patients were randomly allocated (1:1) to T-pouch or Studer ONB. Randomization was done with permuted blocks generated by a congruence algorithm. Randomization was not stratified. There was no attempt to mask the assignment of neobladder type from the patient or staff.

Procedures Surgical technique. All patients underwent open radical cystectomy, extended bilateral pelvic lymph node dissection and urinary diversion by 1 of 5 experienced surgeons at one of the primary USC hospitals as previously described.8,9 All surgeons participating in the trial were experienced with both types of neobladder construction.

Perioperative management followed the same clinical pathway in both arms. Both types of neobladder were constructed as previously described.10,11 The reservoir in both techniques was formed from a 44 cm segment of distal ileum that was completely detubularized and double-folded using the Kock technique.12 In the Studer pouch an additional 12 to 15 cm proximal ileum was left intact as a tubularized afferent limb into which the ureters were implanted (fig. 1, A). In the T-pouch the afferent limb was separated from the reservoir portion and incorporated into an extraserosal tunnel that functions as a flap valve to prevent reflux (fig. 1, B). The urethra in either pouch was anastomosed to the end of the suture line (fig. 1, A) or to a separate opening in the pouch according to surgeon preference (in equal percentages in both groups), and the pouch closed with running absorbable suture. In both procedures the ureters were implanted separately using the Leadbetter/Bricker technique. Adjuvant therapy. Patients received neoadjuvant or adjuvant chemotherapy at the discretion of the treating urologist in consultation with a medical oncologist. No patient received postoperative pelvic radiation. Followup. Scheduled study visits occurred at 3 weeks, 4 to 6 months, 1 year, 2 years and 3 years after surgery. At each visit history, physical examination, blood work, urine culture, cystogram and radiologic evaluation (renal ultrasound, computerized tomography of the abdomen and pelvis with contrast, magnetic resonance imaging with gadolinium, or excretory urogram) were performed. Laboratory work, radiologic studies and postoperative complications were recorded prospectively, and efforts were made to capture all events including those that were managed at other hospitals or physician offices. End points. The primary end point was change in renal function from baseline to 3 years. Secondary end points included changes in renal function from baseline to 6, 12 and 24 months, reduction in eGFR 10 ml/minute/1.73 m2 or greater, early (less than 90 days) and late (90 days to 3 years) complications, urinary tract infection that was symptomatic or was treated, any surgical re-intervention or urinary diversion related surgical re-intervention defined as an open or endoscopic procedure under anesthesia, cancer recurrence and overall survival. Renal function was assessed using the CKD-EPI equation.13 Complications were analyzed and graded by a blinded adjudicator using the modified Clavien complication grading system.14

Statistical Analysis Sample size calculation was based on the primary end point. A pre-study power calculation assumed a 70% survival rate and 20% lost to followup rate at 3 years. Overall 210 evaluable patients per group allowed detection of a 15% difference between groups with a power of 85% and a 1-sided significance level of 0.05. The analysis used the intent to treat principle, including all eligible randomized patients who were alive at 3 years, regardless of what treatment they actually received. Missing data were not imputed. The Fisher’s exact test and Pearson’s

STUDER POUCH VERSUS T-POUCH ORTHOTOPIC ILEAL NEOBLADDER

435

Figure 1. Modified Studer pouch (A) and T-pouch (B). Reprinted with permission from Campbell-Walsh Urology, 10th ed. Edited by AJ Wein, LR Kavoussi, AC Novick et al. Philadelphia: Elsevier Saunders 2012.

chi-square test were used to evaluate baseline categorical variables. The Kruskal-Wallis test was used to evaluate baseline continuous variables. Primary end point. The Wilcoxon 2-sample test was used to compare between-group changes in renal function from baseline to 3 years. Multivariable linear regression analysis was used to examine predictors of 3-year renal function. Variables included in the regression model were type of ONB, baseline eGFR, age, urinary tract obstruction, gender, pathological TNM stage, perioperative chemotherapy use and ASAÒ score. Diabetes and hypertension were not included individually because they were closely associated with baseline renal function. Secondary end points. The Wilcoxon 2-sample test was used to compare between-group changes in renal function from baseline to 6, 12 and 24 months. Pearson’s chi-square test was used to compare early complications. The cumulative incidence method was used to assess reduction in eGFR 10 ml/minute/1.73 m2 or greater, late complications, UTI, SRI, UDSRI and recurrence to account for competing risks using Gray’s test. Fine and Gray competing risk regression analysis was used to examine predictors of reduction in eGFR 10 ml/minute/1.73 m2 or greater, late complications, UTI, SRI, UDSRI and recurrence. Variables included in the regression models included type of ONB, age, gender, pathological TNM stage, perioperative chemotherapy use and baseline estimated GFR. The Kaplan-Meier method and multivariable Cox proportional hazards regression analysis were used to analyze overall mortality. Differences in overall mortality were assessed with the log rank statistic. Cox’s regression proportionality hazards assumptions were tested and no violations of the proportionality were

found. Data were analyzed using SASÒ version 9.2. A 2-sided p 0.05 indicated statistical significance.

Role of the Funding Source The USC Department of Urology funded the study. The corresponding author had full access to all the data in the study and takes responsibility for the integrity of the data and accuracy of data analysis.

RESULTS Flow of Participants through the Trial Figure 2 shows the trial profile. A total of 484 patients were randomized to Studer pouch (247) or T-pouch (237). Complete baseline data were available in 423 patients and these are included in this study. By 3 years 124 patients had died (29%) and an additional 39 were lost to followup or missing renal function data (13%), leaving 260 patients. Demographic, clinical and pathological characteristics did not differ between the groups (table 1). There was no difference in the incidence of preoperative diabetes, hypertension or the use of perioperative chemotherapy in the 2 groups. Renal Function Primary renal function outcome. Table 2 shows the

primary renal function outcome. Mean baseline eGFR was 85.8 ml/minute/1.73 m2 in the Studer group and 83.9 ml/minute/1.73 m2 in the T-pouch group (p¼0.39). Between baseline and 3-year eGFR decreased by a mean of 6.4 ml/minute/1.73 m2 in

436

STUDER POUCH VERSUS T-POUCH ORTHOTOPIC ILEAL NEOBLADDER

Table 1. Baseline characteristics Studer Pouch No. pts Median age (range) No. gender (%): M F

Figure 2. Flow of participants through trial

Demographic 213 67 (25e86) 180 33 Clinical

(85) (15)

No. ml/min/1.73 m2 eGFR (%): Less than 40 0 (0) 40e59 6 (3) 60e89 126 (59) 90 or Greater 81 (38) 27 (17e48) Median kg/m2 body mass index (range) No. TNM stage (%):* T2N0M0 or less 186 (89) T3N0M0 or greater 12 (6) TanyN1e3M0 10 (5) No. periop chemotherapy (%): Neoadjuvant 16 (8) Adjuvant 44 (21) No. ASA score (%): 1e2 54 (25) 3e4 159 (75) Pathological No. TNM stage (%):† T2N0M0 or less 125 (59) T3N0M0 or greater 43 (20) TanyN1e3M0 45 (21) No. soft tissue surgical margin (%): Neg 211 (99) Pos 2 (1) No. histological type (%): Urothelial carcinoma 185 (87) Other 28 (13)

T-Pouch

p Value

210 66 (36e89)

0.93

183 27

(87) (13)

0.49

2 (1) 15 (7) 124 (59) 69 (33) 27 (18e53)

0.08

179 13 11

(88) (6) (6)

0.16

22 43

(5) (21)

0.31 1.00

67 143

(32) (68)

0.16

123 31 56

(58) (15) (27)

0.21

206 4

(98) (2)

0.45

184 26

(88) (12)

0.88

0.36

* Data missing in 12. † 1997 American Joint Committee on Cancer TNM classification system.

the Studer group and 6.6 ml/minute/1.73 m2 in the T-pouch group (p¼0.35). Multivariable linear regression analysis showed that type of ONB was not independently associated with 3-year renal function (p¼0.63). However, baseline eGFR (Beta þ0.54, p

Randomized Trial of Studer Pouch versus T-Pouch Orthotopic Ileal Neobladder in Patients with Bladder Cancer.

The need to prevent reflux in the construction of an orthotopic ileal neobladder is controversial. We designed the USC-STAR trial to determine whether...
1MB Sizes 4 Downloads 16 Views