Perioperative Outcomes and Oncologic Efficacy from a Pilot Prospective Randomized Clinical Trial of Open Versus Robotic Assisted Radical Cystectomy

Perioperative Outcomes and Oncologic Efficacy from a Pilot Prospective Randomized Clinical Trial of Open Versus Robotic Assisted Radical Cystectomy Dipen J. Parekh,* Jamie Messer, John Fitzgerald, Barbara Ercole and Robert Svatek From the Department of Urology, University of Texas Health Sciences Center at San Antonio, San Antonio, Texas

Abbreviations and Acronyms BMI ⫽ body mass index EBL ⫽ estimated blood loss LN ⫽ lymph node LOS ⫽ length of stay ORC ⫽ open radical cystectomy RARC ⫽ robotic assisted laparoscopic radical cystectomy Accepted for publication September 13, 2012. Study received institutional review board approval. Clinical Trial Registration NCT01157676 (www. clinicaltrials.gov). * Correspondence: Department of Urology, University of Miami Miller School of Medicine, Dominion Tower, 5th floor, 1400 NW 10th Ave., Miami, Florida 33136 (telephone: 305-243-6591; e-mail: [email protected]).

Purpose: Robotic assisted laparoscopic radical cystectomy for bladder cancer has been reported with potential for improvement in perioperative morbidity compared to the open approach. However, most studies are retrospective with significant selection bias. Materials and Methods: A pilot prospective randomized trial evaluating perioperative outcomes and oncologic efficacy of open vs robotic assisted laparoscopic radical cystectomy for consecutive patients was performed from July 2009 to June 2011. Results: To date 47 patients have been randomized with data available on 40 patients for analysis. Each group was similar with regard to age, gender, race, body mass index and comorbidities, as well as previous surgeries, operative time, postoperative complications and final pathological stage. We observed no significant differences between oncologic outcomes of positive margins (5% each, p ⫽ 0.50) or number of lymph nodes removed for open radical cystectomy (23, IQR 15–28) vs robotic assisted laparoscopic radical cystectomy (11, IQR 8.75–21.5) groups (p ⫽ 0.135). The robotic assisted laparoscopic radical cystectomy group (400 ml, IQR 300 –762.5) was noted to have decreased estimated blood loss compared to the open radical cystectomy group (800 ml, IQR 400 –1,100) and trended toward a decreased rate of excessive length of stay (greater than 5 days) (65% vs 90%, p ⫽ 0.11) compared to the open radical cystectomy group. The robotic group also trended toward fewer transfusions (40% vs 50%, p ⫽ 0.26). Conclusions: Our study validates the concept of randomizing patients with bladder cancer undergoing radical cystectomy to an open or robotic approach. Our results suggest no significant differences in surrogates of oncologic efficacy. Robotic assisted laparoscopic radical cystectomy demonstrates potential benefits of decreased estimated blood loss and decreased hospital stay compared to open radical cystectomy. Our results need to be validated in a larger multicenter prospective randomized clinical trial. Key Words: urinary bladder neoplasms, cystectomy, robotics, randomized controlled trial, laparoscopy

474

www.jurology.com

AN estimated 73,510 new cases of bladder cancer with approximately 14,880 deaths were expected in 2012.1 Although the majority of patients present with noninvasive bladder cancer, where the focus of therapy is

prevention of progression and recurrence, 25% to 30% will present with muscle invasive bladder cancer.2 Radical cystectomy is the treatment of choice for patients with muscle invasive bladder cancer and for select pa-

0022-5347/13/1892-0474/0 THE JOURNAL OF UROLOGY® © 2013 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION

http://dx.doi.org/10.1016/j.juro.2012.09.077 Vol. 189, 474-479, February 2013 RESEARCH, INC. Printed in U.S.A.

AND

OPEN VERSUS ROBOTIC ASSISTED RADICAL CYSTECTOMY

tients with high grade nonmuscle invasive bladder cancer.3 Radical cystectomy has proven efficacy with regard to local control and long-term disease-free survival.4 However, the procedure can be associated with high perioperative morbidity. In the last decade minimally invasive approaches including robotic assisted approaches have emerged as viable surgical options for many urological malignancies with the promise of decreased morbidity with shorter hospital stays, faster recovery and fewer narcotic analgesic requirements.5–7 In recent years several studies have documented the feasibility of minimally invasive approaches to muscle invasive bladder cancer.5,8,9 These studies have demonstrated that robotic assisted radical cystectomy has potential benefits of lower estimated blood loss, fewer transfusions, improved return of bowel function and decreased hospital stay.10 –14 Unfortunately many of these studies are retrospective single institution studies. To date only 1 prospective randomized study has been performed reporting potential benefits in perioperative morbidity and equivalent operative pathological parameters.15 The treatment of bladder cancer can be associated with high expense, and currently robotics have been implemented based on the data from retrospective studies and the perception that minimally invasive approaches may help decrease the morbidity associated with the treatment of muscle invasive bladder cancer.16 –18 However, unfortunately this has not been demonstrated in a prospective randomized fashion. The primary goal for this analysis is to provide preliminary data from a single institution prospective randomized trial comparing oncologic efficacy and perioperative morbidity in patients with invasive bladder cancer treated with an open vs robotic approach. We unequivocally submit that our study is a pilot trial performed mainly to establish the feasibility and safety of the robotic approach, in addition to generating important preliminary data to determine statistical analyses and outcome measures for a phase 3 multi-institutional clinical trial.

MATERIALS AND METHODS Patients After obtaining institutional review board approval, a prospective randomized trial of open vs robotic assisted radical cystectomy was performed at our institution. The clinical trial was registered at clinicaltrials.gov under the unique identifier NCT01157676. Patients were eligible for study enrollment if they demonstrated biopsy proven bladder cancer of clinical stage T1–T3, N0, M0 and were considered candidates for an open or robotic approach at the discretion of the treating surgeon. Exclusion criteria were 1) inability to give informed consent, 2) multiple prior abdominal and pelvic open surgical procedures that

475

would preclude a safe robotic approach, 3) morbid obesity that would preclude the robotic approach, 4) clinical T4 bladder cancer, 5) clinical lymph node positive bladder cancer with grossly enlarged pelvic or retroperitoneal lymph nodes, 6) any preexisting condition that precludes safe initiation or maintenance of pneumoperitoneum for a prolonged period, 7) age younger than 30 or older than 90 years and 8) pregnancy.

Surgical Procedure Open radical cystectomy was performed in a standard manner with pelvic lymph node dissection performed in all patients. Urinary diversion was performed at the discretion of the surgeon. Lymph node dissection for both groups was performed superiorly to the level of the ureteral crossing of the common iliac vessels, and included the internal iliac artery and obturator fossa, and laterally to the genitofemoral nerve including the tissue overlying the psoas muscle. The extent of lymph node dissection was similar in the ORC and RARC groups. RARC was performed with 3, 8 mm robotic trocars; 2, 12 mm assistant trocars and 1, 12 mm camera trocar. After each procedure patients were placed on a standardized post-cystectomy care plan.

Outcomes Measured Primary outcomes were to establish the feasibility of randomizing patients, obtaining preliminary data on oncologic efficacy and measuring perioperative outcomes. Oncologic efficacy was assessed with surgical margin status and total LN count. Perioperative outcomes measured were EBL, operative time (defined as incision to closure), transfusion requirement, time to return of bowel function and length of stay. Complications were measured using the Clavien classification system to allow a standardized approach to perioperative complications.19

Randomization and Statistical Analysis Patients who met the inclusion criteria were randomized to open or robotic radical cystectomy at their preoperative clinic visit using a computerized randomization program (www.randomization.com) (see figure). This program generated a list of surgical slots numbered 1 through 60, and randomly assigned open or robotic assisted cystectomy to each slot (30 slots for each procedure). Each assignment was placed in a sealed envelope with the corresponding slot number written on the outside. At the time of consent the lowest numbered envelope remaining was opened and the patient was assigned to the surgical procedure listed on the piece of paper inside the envelope. The surgical team and the patient were then made aware of the type of surgery. Fisher’s exact test and the chi-square test were used to evaluate the association between categorical variables. Differences in variables with a continuous distribution across dichotomous categories were assessed using the Mann-Whitney U test. All reported p values were 2-sided and statistical significance was set at p ⬍0.05. All statistical tests were performed with Stata® 10.1.

RESULTS From July 2009 to June 2011 a total of 46 patients met the study inclusion criteria, of which 39 were

476

OPEN VERSUS ROBOTIC ASSISTED RADICAL CYSTECTOMY

47 patients with muscle invasive bladder cancer who met inclusion criteria Excluded: -

-

-

2 patients excluded prior to cystectomy due to clinical T4 disease 3 patients excluded due to desire for robotic surgery 2 patients declined participation

40 patients randomized

20 patients randomized to ORC

20 Patients randomized to RARC

Primary End Points •

Oncologic Efficacy

Soft Tissue Margins Lymph Node counts Pathologic Stage •

Perioperative Outcomes

Secondary Endpoints •

Quality of Life Outcomes - Vanderbilt Cystectomy Index at 3,6,9 and 12 months

•

Functional Recovery - Activities of Daily Living (ADL), Instrumental ADL, Timed Get up and Go/Grip Strength at 3,6 months

Randomization schema

randomized with data available for analysis. Each group was similar with regard to age, gender, race, BMI, comorbidities (American Society of Anesthesiologists), previous abdominal procedures and final pathological stage (table 1). Of the 20 patients in the robotic arm none required conversion to an open procedure. We found no difference in median operative times, median units transfused, transfusion percentage, median LOS or complications. The robotic group had a significantly lower EBL with a trend

toward faster return of bowel function and fewer prolonged hospitalizations which were not significant (table 2). Complication rates of the 2 groups were not significantly different with regard to Clavien grade 2 or greater complications. The complications noted in the robotic group were cardiac arrhythmia, cerebrovascular stroke, pneumonia, renal failure, evisceration and ileus. For the open group the complications noted were cardiac arrhythmia, wound infection (2), fistula formation (1) and ileus. With regard to perioperative pathological measures

OPEN VERSUS ROBOTIC ASSISTED RADICAL CYSTECTOMY

Table 3. Perioperative pathological outcomes

Table 1. Patient characteristics RARC No. pts Median age (IQR) No. gender: M F Median kg/m2 BMI (IQR) Median ASA No. pathological stage (%): pT0 pTis pT1 pT2 pT3 pT4

ORC

p Value

20 20 69.5 (62.3–74) 64.5 (59.8–72.3)

0.595 0.429

18 16 2 4 27.6 (24.2–29.9) 28.3 (26.1–32.3) 0.072 3 3 Not significant 0.108 2 (10) 3 (15) 2 (10) 6 (30) 3 (15) 3 (15) 3 (15) 1 (5) 3 (15) 2 (10) 7 (35) 5 (25)

of oncologic outcomes we found no significant difference in positive margins, the use of perioperative chemotherapy or LN counts. There was 1 positive margin in each group with both patients having pT4 disease (table 3).

DISCUSSION In the last several years the minimally invasive robotic approach has come to the forefront of attention for many urological malignancies including RARC for invasive bladder cancer. The surgical robot has been aggressively marketed during the last decade with the promise of reducing perioperative morbidity, and improving oncologic and functional outcomes in many organ sites.17,18 However, prospective randomized data are lacking. Several retrospective series have demonstrated the feasibility of RARC.9,11,12,15,20 The potential benefits reported include decreased EBL, fewer transfusions, more rapid return of bowel function, shorter LOS and possible fewer postoperative complications. This comes with a significant learning curve to the robotic approach and a possible increase in operative time as noted in a prior prospective study.15 Table 2. Perioperative outcomes RARC

477

ORC

p Value

No. pts 20 19 Median operative mins (IQR) 300 (240–366) 285.5 (240–321.3) 0.329 Median ml EBL (IQR) 400 (300–762.5) 800 (400–1,125) 0.003 Median units blood 0 (0–4) 2 (0–4) 0.410 transfused (IQR) No./total No. transfusions (%) 8/20 (40) 10/20 (50) 0.268 Median days LOS (IQR) 6 (5–9.5) 6 (6–9.3) 0.288 No./total No. LOS 5 days or 7/20 (35) 2/20 (10) 0.030 less (%) Median days to diet (IQR) 4 (3–6.8) 5.5 (3–7) 0.500 No./total No. Clavien 2 or 5/20 (25) 5/20 (25) 0.500 greater complications (%)

RARC No. pts No./total No. periop Adjuvant or neoadjuvant chemotherapy (%) No./total No. pos surgical margins (%) Median LNs (IQR) No./total No. LN pos (%)

ORC

p Value

20 8/20

(40)

19 8/20

(40)

0.500

1/20

(5)

1/20

(5)

0.500

23 (15–28) 4/20 (20)

0.135 0.500

11 (8.8–21.5) 4/20 (20)

The early reports on RARC were single institution retrospective reviews with comparison of the robotic cohort to historical controls or retrospective comparative studies with significant selection bias toward the robotic approach.12,13,20 There has been one prior prospective randomized analysis comparing RARC to ORC with important differences compared to our study.15 While the open and robotic cohorts were similar with regard to age, gender distribution, BMI and American Society of Anesthesiologists classification, our study had a larger number of patients with locally advanced disease (pT3 and pT4). Our randomization scheme was significantly different in that in the earlier study, patients were randomized sequentially 5 at a time to open or robotic cystectomy, whereas in our analysis each consecutive patient was randomized. We decided to perform this trial after the primary faculty involved in the study, with independent experience of more than 100 ORCs, had performed approximately 50 RARCs, with the intent of eliminating bias induced by the learning curve for the robotic approach. We noted no significant difference in median operative time between the RARC and ORC groups. Prior studies demonstrated that there is a significant learning curve for the robotic approach, and while after the first 20 cases a gradual reduction in operative times can be perceived, our study attempted to eliminate this confounder by using faculty who were past this learning curve.8,21 Several prior studies demonstrated a significant increase in operative times associated with the robotic approach, and the single previous prospective randomized series showed a difference of 4.2 vs 3.5 hours for the robotic vs the open group, respectively.13,15,22 In our series we showed that the robotic approach can have operative times similar to those of the open approach, addressing a previous major limitation of the robotic approach. We noted a significantly lower EBL in the robotic group, similar to several previous studies.10,13,15,23,24 We also noted a trend toward decreased transfusions in the perioperative period in the robotic group that was not statistically significant. This may be a product of the small sample size in our study as well as

478

OPEN VERSUS ROBOTIC ASSISTED RADICAL CYSTECTOMY

significant selection bias in favor of the robotic approach in previously reported studies. While the mean EBL for RARC was only 400 cc, we acknowledge that the perioperative transfusion rate may seem high. However, as seen in several cystectomy cases, most of these transfusions were not given in the intraoperative setting or for intraoperative blood loss, but were in response to a gradual decrease in hemoglobin levels found in elderly patients with significant comorbidities and a history of cardiac disease in the postoperative period. In our series we noted that there were no differences with regard to LOS. However, we did note that the RARC cohort had a trend toward fewer prolonged hospitalizations (LOS greater than 5 days), although this was not significant. We also noted that the RARC group had a trend toward a faster return to a regular diet compared to the ORC group. Nix et al demonstrated that RARC was associated with a shorter median time to return of flatus (2.3 vs 3.2 days, p ⫽ 0.0013) but a similar LOS (5.1 vs 6.0 days, p ⫽ 0.239).15 Several retrospective analyses have demonstrated a significantly lower LOS compared to the open group that again may be reflective of the selection bias from a retrospective study.5,9,25 The difference in return of bowel function has not translated to shorter LOS in our analysis or the previous prospective study.15 We noted that the complication rate was similar between the cohorts. Previous retrospective studies have reported lower complication rates with RARC compared to ORC that may be a reflection of significant selection bias.15 Oncologic outcomes after radical cystectomy can be measured by several parameters, but LN yield and positive surgical margin status have previously been shown to serve as surrogates for oncologic outcomes.3,26 In our current study we noted no significant difference between LN yields for either group. Although there is an appearance of fewer LNs in RARC, this finding was not significant. A possible reason for this apparent discrepancy could be the different sampling methods of LN collection between the operations. For the robotic groups, at the completion of lymphadenectomy for each side, nodes are submitted as right and left pelvic lymph nodes, whereas in the open group LNs are handed off as discrete anatomical packets. The variability in specimen collection and sampling leading to differences in LN count between RARC and ORC has been similarly reported recently.27 Several previous series have demonstrated that the robotic approach yields LN counts similar to those of the open approach.10,15 Lymph node metastasis can be found in 25% of patients at radical cystectomy, which is similar to our study as well as the previously reported randomized study.15,28

We noted no difference in positive surgical margins between the 2 surgical approaches. The 2 patients with positive surgical margins had pT4 disease with gross extravesical extension. Previous series have noted similar outcomes when comparing surgical margin status after robotic cystectomy.15,25,29 It should be noted that in our current analysis 50% of the patients had pT3 or greater disease in the robotic cohort, demonstrating that RARC is feasible in locally advanced disease. Almost all of the patients in our study underwent surgery at a single institution by a single surgeon proficient in both approaches. This significantly reduced surgeon induced bias in our study. Smith et al previously reported on a multi-surgeon, multi-institutional cohort of 227 patients from 4 centers all undergoing robotic assisted radical cystectomy with mean operating time of 5.5 hours and a mean EBL of 256 ml, with 70% of patients with a LOS of 5 days or less.30 This is similar to our current study, except our operative time was shorter, and it should also be noted this study did not have any comparison to the open cohort at these institutions. We acknowledge the limitations of our current study. The primary aim of the study was to demonstrate the feasibility of randomizing and enrolling patients in a pilot setting to obtain enough outcome measures in several variables to determine the optimal sample size to conduct future multi-institutional studies. Our study was limited by the sample size and the lack of power to detect differences in oncologic efficacy as well as perioperative outcomes. We also limited measures of oncologic efficacy in this study to measured perioperative pathological parameters. We plan on collecting intermediate and long-term future survival data on these patients to determine the true oncologic outcomes of progression-free and overall survival. We are also currently assessing data on quality of life, activities of daily living, handgrip strength and mobility data for all patients who have completed the analysis. The strength of our study is the prospective randomized nature that eliminates selection biases that may have been present in prior retrospective analyses. We also believe that our study along with the recently reported similar study demonstrates that a prospective randomized trial comparing traditional open and robotic approaches in bladder cancer is feasible.

CONCLUSIONS The results from our preliminary pilot study demonstrate unequivocal feasibility in performing a pro-

OPEN VERSUS ROBOTIC ASSISTED RADICAL CYSTECTOMY

spective randomized clinical trial comparing open and robotic approaches for radical cystectomy in patients with invasive bladder cancer. While our study suggests potential benefits of decreased EBL, lower transfusion requirements and fewer prolonged lengths of stay with RARC compared to ORC while maintaining oncologic efficacy, these results

479

still need to be validated in a larger multicenter setting. We have joined with several institutions nationally to build on the present study, and have started a phase 3 multi-institutional prospective randomized clinical trial comparing RARC and ORC in patients with invasive bladder cancer that is currently actively accruing.

REFERENCES 1. Siegel R, Naishadham D and Jemal A: Cancer statistics, 2012. CA Cancer J Clin 2012; 62: 10. 2. Lopez-Beltran A: Bladder cancer: clinical and pathological profile. Scand J Urol Nephrol Suppl 2008; 218: 95. 3. Huang GJ and Stein JP: Open radical cystectomy with lymphadenectomy remains the treatment of choice for invasive bladder cancer. Curr Opin Urol 2007; 17: 369. 4. Shabsigh A, Korets R, Vora KC et al: Defining early morbidity of radical cystectomy for patients with bladder cancer using a standardized reporting methodology. Eur Urol 2009; 55: 164. 5. Haber GP, Crouzet S and Gill IS: Laparoscopic and robotic assisted radical cystectomy for bladder cancer: a critical analysis. Eur Urol 2008; 54: 54. 6. Haber GP, Campbell SC, Colombo JR Jr et al: Perioperative outcomes with laparoscopic radical cystectomy: “pure laparoscopic” and “open-assisted laparoscopic” approaches. Urology 2007; 70: 910. 7. Taylor GD, Duchene DA and Koeneman KS: Hand assisted laparoscopic cystectomy with minilaparotomy ileal conduit: series report and comparison with open cystectomy. J Urol 2004; 172: 1291. 8. Menon M, Hemal AK, Tewari A et al: Nervesparing robot-assisted radical cystoprostatectomy and urinary diversion. BJU Int 2003; 92: 232. 9. Fergany AF and Gill IS: Laparoscopic radical cystectomy. Urol Clin North Am 2008; 35: 455. 10. Guru KA, Kim HL, Piacente PM et al: Robotassisted radical cystectomy and pelvic lymph node dissection: initial experience at Roswell Park Cancer Institute. Urology 2007; 69: 469.

11. Hemal AK: Robotic and laparoscopic radical cystectomy in the management of bladder cancer. Curr Urol Rep 2009; 10: 45.

21. Pruthi RS, Smith A and Wallen EM: Evaluating the learning curve for robot-assisted laparoscopic radical cystectomy. J Endourol 2008; 22: 2469.

12. Wang GJ, Barocas DA, Raman JD et al: Robotic vs open radical cystectomy: prospective comparison of perioperative outcomes and pathological measures of early oncological efficacy. BJU Int 2008; 101: 89.

22. Rhee JJ, Lebeau S, Smolkin M et al: Radical cystectomy with ileal conduit diversion: early prospective evaluation of the impact of robotic assistance. BJU Int 2006; 98: 1059.

13. Pruthi RS and Wallen EM: Robotic assisted laparoscopic radical cystoprostatectomy: operative and pathological outcomes. J Urol 2007; 178: 814. 14. Khan MS, Shah SS, Hemel A et al: Roboticassisted radical cystectomy. Int J Med Robot 2008; 4: 197. 15. Nix J, Smith A, Kurpad R et al: Prospective randomized controlled trial of robotic versus open radical cystectomy for bladder cancer: perioperative and pathologic results. Eur Urol 2010; 57: 196. 16. Konety BR, Joyce GF and Wise M: Bladder and upper tract urothelial cancer. J Urol 2007; 177: 1636. 17. Yates DR, Vaessen C and Roupret M: From Leonardo to da Vinci: the history of robot-assisted surgery in urology. BJU Int 2011; 108: 1708. 18. Richards KA, Kader K and Hemal AK: Robotic radical cystectomy: where are we today, where will we be tomorrow? ScientificWorldJournal 2010; 10: 2215. 19. Dindo D, Demartines N and Clavien PA: Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004; 240: 205. 20. Novara G and Ficarra V: Robotic-assisted laparoscopic radical cystectomy: where do we stand? Int J Clin Pract 2009; 63: 185.

23. Kauffman EC, Ng CK, Lee MM et al: Critical analysis of complications after robotic-assisted radical cystectomy with identification of preoperative and operative risk factors. BJU Int 2010; 105: 520. 24. Murphy DG, Challacombe BJ, Elhage O et al: Robotic-assisted laparoscopic radical cystectomy with extracorporeal urinary diversion: initial experience. Eur Urol 2008; 54: 570. 25. Ng CK, Kauffman EC, Lee MM et al: A comparison of postoperative complications in open versus robotic cystectomy. Eur Urol 2010; 57: 274. 26. Buscarini M, Josephson DY and Stein JP: Lymphadenectomy in bladder cancer: a review. Urol Int 2007; 79: 191. 27. Davis JW, Gaston K, Anderson R et al: Robot assisted extended pelvic lymphadenectomy at radical cystectomy: lymph node yield compared with second look open dissection. J Urol 2011; 185: 79. 28. Leissner J, Hohenfellner R, Thuroff JW et al: Lymphadenectomy in patients with transitional cell carcinoma of the urinary bladder; significance for staging and prognosis. BJU Int 2000; 85: 817. 29. Yuh B, Padalino J, Butt ZM et al: Impact of tumour volume on surgical and pathological outcomes after robot-assisted radical cystectomy. BJU Int 2008; 102: 840. 30. Smith AB, Raynor M, Amling CL et al: Multiinstitutional analysis of robotic radical cystectomy for bladder cancer: perioperative outcomes and complications in 227 patients. J Laparoendosc Adv Surg Tech A 2012; 22: 17.