Positive Surgical Margins in Robotic-Assisted Radical Prostatectomy: Impact of Learning Curve on Oncologic Outcomes

Transcription

1 european urology 49 (2006) available at journal homepage: Laparoscopy Positive Surgical Margins in Robotic-Assisted Radical Prostatectomy: Impact of Learning Curve on Oncologic Outcomes Fatih Atug a, *, Erik P. Castle a, Sudesh K. Srivastav b, Scott V. Burgess a, Raju Thomas a, Rodney Davis a a Center for Minimally Invasive Urologic Surgery, Department of Urology, Tulane University Health Sciences Center, New Orleans, Louisiana, USA b Department of Biostatistics, School of Public Health and Tropical Medicine, Tulane University, New Orleans, Louisiana, USA Article info Article history: Accepted February 21, 2006 Published online ahead of print on March 10, 2006 Keywords: Learning curve Robotic prostatectomy Surgical margins Abstract Objective: The presence of positive surgical margins following radical prostatectomy is a known risk factor for disease recurrence and may lead to adjuvant treatment. Our goal was to assess the incidence of positive surgical margins in our series of robotic-assisted radical prostatectomy (RARP) and its relationship to our learning curve. Methods: Between February 2003 and August 2005, 140 patients underwent RARP by the same surgical team at our institution. The records of our first 100 consecutive RARPs were retrospectively reviewed. The patients were divided into three groups based on the time of surgery: group I included the first 33 cases; group II included the second 33 cases; and group III comprised the last 34 cases. We compared the incidence and location of positive surgical margins among the groups. Additional variables evaluated included the patient s prostate-specific antigen (PSA) level, preoperative/postoperative Gleason score, clinical/pathologic stage, and pathologic tumour volume. Results: The positive margin rates were 45.4%, 21.2%, and 11.7% for groups I, II, and III, respectively. The difference in positive margin rates in the three groups was statistically significant ( p = ). Positive margin rates declined specifically at the apex and bladder neck when comparing the first 33 patients to the last 34 patients. Patient demographics and preoperative staging variables were comparable among all three groups, with no statistically significant differences among them. Conclusions: This study illustrates that experience gained with time led to a decrease in the incidence of positive surgical margins. We do not feel that a selection bias affected our results because clinical and pathologic staging was evenly distributed within the three study groups. A steady reduction in positive surgical margin rates demonstrates a learning curve, of approximately 30 * Corresponding author. Department of Urology, Tulane University Health Sciences Center, 1430 Tulane Avenue, SL-42, New Orleans, LA 70112, USA. Tel: ; Fax: address: fatug@tulane.edu (F. Atug) /$ see back matter # 2006 Elsevier B.V. All rights reserved. doi: /j.eururo

2 european urology 49 (2006) Introduction Positive surgical margins are associated with higher rates of biochemical, local, and systemic progression [1,2]. The primary goal of any urologist is to minimise the risk of positive surgical margins and hence reduce the risk of recurrence. New surgical techniques for prostatectomy should be held to this same standard of minimising positive surgical margins and risk of disease recurrence. Robotic urologic surgery, a new and exciting emerging frontier in the field of urology, has tremendous potential to progress as a treatment option for prostate cancer in the future. It is important that urologists keep abreast of these new technologies, their limitations, and the possibility of incorporating them into urologic practice. Robotic-assisted laparoscopic radical prostatectomy (RARP) is gaining popularity for the treatment of clinically localised prostate cancer. As with any new surgical technique, an associated learning curve is necessary to attain proficiency. One may assume that the initial histopathology reports of the prostate following RARP will have higher rates of positive surgical margins, as surgeons are still new to the robotic experience. This has been found in reports of laparoscopic radical prostatectomies [3]. The impact of a learning curve on surgical margin positivity in RARP has yet to be fully assessed. In the current study, we analysed the incidence of positive surgical margins and any association this may have with the robotic learning curve. 2. Methods 2.1. Patients and procedures One hundred forty patients underwent RARP by the same surgical team at our institution. Three faculty members performed the RARP procedures in random rotation, and all three were considered experienced in advanced laparoscopic techniques. After obtaining Institutional Review Board approval, the records of the first 100 consecutive patients were reviewed. Standard preoperative assessment included digital rectal examination (DRE), prostate-specific antigen (PSA), and transrectal ultrasound-guided biopsies. We used an operative technique similar to what has been described previously [4]. Nerve sparing was performed when clinically indicated, even early in our experience. Patients referred or presented to our institution were offered this procedure in all cases when indicated and not selected based on any specific preoperative variable. Considerations for nerve sparing were also based on clinical stage, age, and presence of high-risk parameters, such as a Gleason score, PSA levels, and the number and the percentage of tumour involvement in these cores [5]. In addition, because all but 2 of the 100 patients had preoperative Gleason scores of 7 (Table 1) and were all clinically localised, each patient who was sexually active preoperatively was offered a nerve-sparing option. At this point in our learning curve, we were comfortable in offering this desirable option. For analysis, the patients were classified into three groups according to the chronological order based on procedure date. Group I consisted of the first 33 patients in our series, group II consisted of the second 33, and group III consisted of the last 34 patients. We compared the incidence of positive surgical margins among the three groups. Additional variables evaluated were the preoperative PSA levels, location of positive margins, preoperative and postoperative Gleason pattern and score, preoperative clinical stage, final pathologic stage, and prostate and pathologic tumour volume (Tables 1 and 2). Patients were also stratified based on high- or low-risk features and compared among the three groups (Table 3). High-risk and low-risk clinical profiles, for having positive surgical margins, were assessed according to the risk profile described by Wieder and Soloway [5]. Specimens were processed according to the technique described by True [6]. The apex and bladder-neck cones were amputated and sectioned in the sagittal plane. The remaining specimen was sectioned transversely at intervals of 2 3 mm. All prostate blocks were labeled according to location, which Table 1 Preoperative data of patients Group I Group II Group III p No. of patients Mean age, yr Mean PSA, ng/ml Preoperative Gleason scores Mean total Gleason score Clinical stage T T T

3 868 european urology 49 (2006) Table 2 Postoperative data of patients Group I Group II Group III p No. of patients with positive surgical margins 15 (45.4%) 7 (21.2%) 4 (11.7%) No. of patients with negative surgical margins 18 (54.6%) 26 (78.8%) 30 (88.3%) Positive margins for pathologic stages pt2 10/26 (38.4%) 4/29 (13.7%) 1/28 (3.6%) PT3 3/5 (60%) 2/3 (66%) 2/5 (40%) PT4 2/2 (100%) 1/1 (100%) 1/1 (100%) Positive margin location Bladder neck a Apex a Other a Postoperative Gleason scores Total Gleason score, mean Pathologic stage pt2a pt2b pt3a pt3b pt Final pathologic tumour volumes, mean, % Pathologic prostate volumes, mean, g a There may have been more than one location for each patient with a positive margin. allowed for whole-mount reconstruction if necessary. Positive surgical margins were defined as the presence of tumour tissue on the inked surface of the specimen. Descriptive results were reported for all studied parameters. The x 2 test at the 5% significance level was used to compare mean proportion of surgical margins values (positivity/ negativity) and risk profiles for surgical margin positivity among the three groups. To test the homogeneity for Gleason score, clinical stages, pathologic stages, and nerve sparing for the three groups, the x 2 test was performed. Analysis of variance method was used to compare mean age, PSA, and final pathologic tumour volume values for the three groups. All statistical analysis tests were performed with the Statistical Analysis Software 9.1 (SAS Institute, Cary, NC) and graphs were plotted using R-software (The R Foundation for Statistical Computing, Vienna, Austria) Statistical analysis 3. Results Table 1 lists our results. The mean patient age was 58 yr (range, yr). After stratifying based on number, the margin positive rates were 45.4%, 21.2%, and 11.7% for groups I, II and III, respectively. Patient demographics and preoperative staging variables are displayed in Table 1. The mean PSA level of all patients was 7.06 ng/ml. The mean preoperative PSA values were 7.25 ng/ml, 6.25 ng/ ml, and 6.67 ng/ml for the three groups, respectively Table 3 Risk profile of patients for surgical margin positivity Group 1 (cases 1 33) Group 2 (cases 33 66) Group 3 (cases ) p Low-risk profile, n a 24 (72.7%) 14 (42.4%) 14 (41.79%) High-risk profile, n b (n) (%) 9 (27.3%) 19 (57.6%) 20 (58.21%) Nerve sparing, n Bilateral 27 (81.8%) 30 (90.8%) 30 (90.8%) Unilateral 3 (9.09%) 2 (6.06%) 1 (2.9%) None 3 (9.09%) 1 (3.03%) 3 (8.8%) a PSA 10 ng/ml, Gleason score 7 or2 cores involved. b PSA 10 ng/ml, Gleason score >7 or3 cores involved.

4 european urology 49 (2006) ( p = ). The mean preoperative and postoperative Gleason scores were 5.91, 6.21, 6.3 and 6.3, 6.5, 6.7 for the first, second, and third groups, respectively. There was a statistically significant difference among the three groups for preoperative Gleason scores ( p = 0.037). In particular, group I and group III were found significant in the post-hoc analysis, using the Tukey test. The factors that could affect the surgical margin status, such as PSA level, Gleason pattern, and final pathologic stage, were evenly distributed among the three groups (Tables 1 and 2). The majority of the positive surgical margins were at the apex and the bladder neck in group I and decreased significantly with experience in group III. Surprisingly, our margin positive rate decreased with experience, despite having greater numbers of high-risk patients in groups II and III. In addition, nerve sparing was performed in equal numbers of men in all three groups and was not a statistically significant factor ( p = 0.683). However, stratifying patients based on high- and low-risk features proved to be statistically significant ( p = 0.010), with more low-risk patients in group I and more high-risk patients in the last group (Table 3). 4. Discussion The incidence of positive surgical margins in previously reported open prostatectomy series varies from 16% to 46% [7]. The incidence in pure laparoscopic series is 16 26% [8 11]. According to the reported series in RARP, the overall surgical margin positivity rate varies between 6% and 35.5% [12,13]. In our RARP patients, the rate of positive margins decreased dramatically, from 45.4% to 11.7%. Although various clinical and pathologic parameters have been associated with the risk of a positive surgical margin, we feel that the learning curve may play a significant role in outcomes of RARP. Scardino et al. established that although the clinical and pathologic features of cancer are associated with the risk of a positive surgical margin in radical prostatectomy specimens, the technique used by the surgeon is also a risk factor [14]. Lower rates of positive surgical margins for high-volume surgeons suggest that experience and careful attention to surgical details, adjusted for the characteristics of the cancer being treated, can decrease positive surgical margin rates and improve cancer control with radical prostatectomy. Furthermore, increasing surgical experience has also been shown to improve outcomes following procedures, including esophagectomy, pancreatectomy, and primary surgery for colon and breast cancer [14]. Clearly, the surgeon s experience, especially during the learning curve, can have a significant effect on oncologic outcome in any cancer surgery. In a recent study, Rassweiler and colleagues analyzed the data of 5824 patients who underwent laparoscopic radical prostatectomy (LRP) in 18 centres in Europe. They found the surgical margin rates were comparable to open surgery and were dependent on surgical experience. Additionally, the results of the study confirmed the efficacy of the training program with safe transfer of the learning curve for LRP. They demonstrated that the learning curve flattened in a reasonable time and they anticipated adequate transfer of existing open and laparoscopic expertise for further optimisation of oncologic and functional outcomes [15]. In general, the rate of positive surgical margins in radical prostatectomies has declined during the last two decades. Reasons for this decline include the stage migration that has been witnessed following the introduction of PSA screening. However, surgical skill and technical modifications may also play a role in the decrease. Improvements in surgical technique, such as more meticulous dissection of apex and bladder neck, have been reported to contribute to the decline in rates of positive margins [14]. In a recent study, Graefen et al demonstrated that surgical experience and refinements in technique, particularly for a nervesparing procedure, allows for excellent cancer control in combination with good and reproducible functional results [16]. In this series, our technique and location for trocar placement did not vary. With experience, our use of cold scissors and bipolar cautery increased. As with any foray into new technology and surgical procedures, we believe that such minor technique variations are inevitable with experience, which may further explain the role of surgical experience in ultimately improving outcomes, such as decreasing positive surgical margins. When comparing outcomes of intervention, selection bias is always a concern. In this study, preoperative clinical stages and final pathologic stages were evenly distributed among the three groups, suggesting that there was not a selection bias. Furthermore, there were higher preoperative and postoperative numbers of Gleason and Gleason in the last group when compared to the first group. Additionally, there were high numbers of preoperative and postoperative Gleason scores 7 in groups II and III when compared to the first group.

5 870 european urology 49 (2006) In fact, group III had the lowest margin positive rate, despite having 58.2% of patients with higher risk features and 90.8% of patients undergoing nerve-sparing prostatectomy. Thus, selection bias did not seem to play a role in our series. We analysed the impact of our surgical technique on the incidence and location of positive margins. We feel that our technique evolved primarily with added focus on performing meticulous apical and bladder-neck dissections. Positive margins decreased from 5 to 0 at the bladder neck for groups I and III, respectively. Over time, our comfort levels with the bladder-neck dissection increased and the transition through the posterior bladder neck into the prerectal space became easier. The apical dissection also improved early in our series, as we became more comfortable with the increased magnification and ability to see and suture ligate the dorsal venous complex and surrounding attachments. This decreased blood loss and improved vision, which directly contributed to improved apical dissection ability. Positive margin rates declined from 8 to 3 at the apex for groups I and III, respectively. Although these are subjective assessments, our margin positive rates decreased dramatically in both these areas. Ultimately, long-term follow-up of these patients is essential to assess any effect the positive surgical margin rates and robotic learning curve may have on biochemical recurrence and overall survival. Based on this study, we believe that for a surgeon well versed in open radical prostatectomy and laparoscopic surgery, approximately 30 roboticassisted prostatectomy procedures are needed to gain proficiency. Though other authors have alluded to having gained proficiency in RARP without advanced laparoscopic experience [13], one has to be realistic and make such a recommendation to the average practicing urologist, who may not have the luxury of having skilled assistants (i.e., fellows and urologic residents). Surgical skills and learning curves differ from surgeon to surgeon. When considering the various steps, from the initial task of gaining pneumoperitoneum and adequate trocar placement to successfully completing the RARP to safe exit, we have estimated that approximately 30 patients would constitute a safe window of experience to overcome the learning curve. 5. Conclusion Our results suggest that the experience gained with RARP leads to a decrease in the incidence of positive surgical margins. Once the learning curve is overcome, improved dissection of the apex and bladder neck likely contributes to the decline in margin positive rates. This study demonstrates that surgeon experience and learning curve of a new technique may affect and predict the oncologic outcomes as much as standard preoperative variables. Based on the rationale presented, approximately 30 RARP cases are needed to gain proficiency for those experienced in open radical surgery and laparoscopy. References [1] Cheng L, Darson ME, Bergstralh EJ, Slezak J, Myers RP, Bostwick DG. Correlation of margin status and extraprostatic extension with progression of prostate carcinoma. Cancer 1999;86:1775. [2] Catalona WJ, Smith DS. 5-year tumor recurrence rates after anatomical radical retropubic prostatectomy for prostate cancer. J Urol 1994;152:1837. [3] Baumert H, Fromont G, Adorno Rosa J, Cahill D, Cathelineau X, Vallancien G. Impact of learning curve in laparoscopic radical prostatectomy on margin status: prospective study of first 100 procedures performed by one surgeon. J Endourol 2004;18: [4] Menon M, Tewari A, Peabody J, the VIP Team. Vattikuti Institute prostatectomy: technique. J Urol 2003;169: [5] Wieder JA, Soloway MS. Incidence, etiology, location, prevention and treatment of positive surgical margins after radical prostatectomy for prostate cancer. J Urol 1998;160: [6] True LD. Surgical pathology examination of the prostate gland: practice survey by American Society of Clinical Pathologists. Am J Clin Pathol 1994;102:572. [7] Sofer M, Hamilton-Nelson KL, Schlesselman JJ, Soloway MS. Risk of positive margins and biochemical recurrence in relation to nerve-sparing radical prostatectomy. J Clin Oncol 2002;20:1853. [8] Bollens R, Vanden Bossche M, Roumeguere T, et al. Extraperitoneal laparoscopic radical prostatectomy. Results after 50 cases. Eur Urol 2001;40:65. [9] Rassweiler J, Sentker L, Seemann O, Hatzinger M, Rumpelt HJ. Laparoscopic radical prostatectomy with the Heilbronn technique: an analysis of the first 180 cases. J Urol 2001;166:2101. [10] Gill IS, Zippe CD. Laparoscopic radical prostatectomy: technique. Urol Clin North Am 2001;28:423. [11] Katz R, Salomon L, Hoznek A, de la Taille A, Antiphon P, Abbou CC. Positive surgical margins in laparoscopic radical prostatectomy: the impact of apical dissection, bladder neck remodeling and nerve preservation. J Urol 2003; 169: [12] Menon M, Tewari A, Vattikuti Institute Prostatectomy Team. Robotic radical prostatectomy and the Vattikuti Urology Institute technique: an interim analysis of results and technical points. Urology 2003;61:15 20.

6 european urology 49 (2006) [13] Ahlering TE, Skarecky D, Lee D, Clayman RV. Successful transfer of open surgical skills to a laparoscopic environment using a robotic interface: initial experience with laparoscopic radical prostatectomy. J Urol 2003;170: [14] Eastham JA, Kattan MW, Riedel E, et al. Variations among individual surgeons in the rate of positive surgical margins in radical prostatectomy specimens. J Urol 2003;170: [15] Rassweiler J, Stolzenburg J, Sulser T, et al. Laparoscopic radical prostatectomy the experience of the German Laparoscopic Working Group. Eur Urol 2006;49: [16] Graefen M, Walz J, Huland H. Open retropubic nervesparing radical prostatectomy. Eur Urol 2006;49: Editorial Comment Herve Baumert baumertherve@yahoo.fr Since the first reports of laparoscopic radical prostatectomies were published at the end of the 1990s, the approach has been readily adopted worldwide. The advantages of a minimally invasive surgery seduced patients and surgeons alike. However, this technique is difficult to learn because it requires skill to perform a precise dissection and a watertight anastomosis. The concept of the radical prostatectomy is contradictory. One has to dissect the prostate as close as possible to limit the risks of functional side effects in this radical procedure, as opposed to a supposedly wide excision. Ideally, all patients should be margin free and fully recover continence and potency fully. Does the robot help to achieve this challenging aim? It seems clear that this machine allows surgeons, without any laparoscopic experience, to make their learning curve shorter [1,2]. The threedimensional visions added to the articulation of the instruments led to this shorter learning period. However, no randomised study allows us to conclude, today, that robotic-assisted surgeons do a better job than skilled laparoscopists. In this present study, the positive margin rate decreased from 45.4% to 11.7%, in comparable patients. This leads me to two comments. First, after limited procedures (around 60), these surgeons had a very low positive margin rate (3.6% for pt2) comparable to or better than skilled laparoscopists who have performed hundreds of laparoscopic prostatectomies [3,4]. This can be due to the instrument articulation that makes the dissection easier, especially at the apex, a key point of the operation regarding positive margins. These articulated instruments allow as well, a softer and safer manipulation and dissection of the prostate, probably reducing the risk of false-positive margins (capsular incision) that can occur with laparoscopic instruments during surgical exposure. Second, the positive margin rate of the first group of patients (45.4%) is difficult to accept in this day and age. All teams new to robotic or laparoscopic surgery should initiate their programs with mentors to avoid sacrificing the first patients. References [1] Tewari A, El-Hakim A, Leung RA. Robotic prostatectomy: a pooled analysis of published literature. Expert Rev Anticancer Ther 2006;6: [2] Ahlering TE, Skarecky D, Lee D, Clayman RV. Successful transfer of open surgical skills to a laparoscopic environment using a robotic interface: initial experience with laparoscopic radical prostatectomy. J Urol 2003;170: [3] Touijer K, Kuroiwa K, Saranchuk JW, et al. Quality improvement in laparoscopic radical prostatectomy for pt2 prostate cancer: impact of video documentation review on positive surgical margin. J Urol 2005;173: [4] Rassweiler J, Stolzenburg J, Sulser T, et al. Laparoscopic radical prostatectomy: the experience of the German Laparoscopic Working Group. Eur Urol 2006;49: Editorial Comment Jens Rassweiler jens.rassweiler@slk-kliniken.de The introduction of the da Vinci system still has a significant impact on the diffusion of laparoscopic radical prostatectomy (LRP), mainly in the United States. Even centres that reported excellent results with open retropubic or perineal radical prostatectomy switched to the robotassisted laparoscopic radical prostatectomy (RARP) technique. One reason for this represents the patient s demand to be operated by the robot. The authors of this article are such an example; Raju Thomas and colleagues were able to realise a radical perineal prostatectomy within a 24-h hospital stay [1]; nevertheless they had to offer RARP.

7 Based on their experience and with the help of the device, they were able to establish a successful program of RARP including three different surgeons. The authors can be congratulated on the clear presentation of their data, which differ significantly from others. Both Menon (their reference [4]) and Ahlering (their reference [13]) reported that the learning curve with the da Vinci system requires only 12 to 20 cases compared to 60 to 80 of LRP. The present article clearly demonstrates that there is an ongoing learning curve with continuous improvement of the outcome as in LRP and that the learning curve to reach the standards of open radical prostatectomy included 66 cases. Of course as with LRP (their reference [15]), the next generations will definitively not need to pass the same learning curve, if trained step-by-step at an experienced center [2]. In conclusion, the excellent results of LRP can be reproduced by use of the da Vinci device and may require a shorter learning curve for novices. However, some questions and concerns remain: (1) Do these 20 or fewer cases really justify the investment? (2) Is there any evidence, that RARP produces results superior to LRP? (3) What is the future of LRP in the United States? There is no doubt that there is an extreme RARP boom in the United States, for example, >30 devices in the State of New York. Mainly because of different reimbursement systems, this will not likely occur in Europe. Moreover, actually there exists no comparative study proving superior results of RARP versus LRP. In my opinion, RARP cannot provide any significant advantage with respect to the operative technique of nerve-sparing radical prostatectomy, but there are still some deficiencies (ie, lack of instruments, no haptic sense). The future will show whether the next generation of robots will be able to overcome these drawbacks or the next generation of surgeons will be able to master the cost-effective LRP. References [1] Ruiz-Deja G, Davis R, Srivastav SK, M Wise A, Thomas R. Outpatient radical prostatectomy: impact of standard perineal approach on patient outcome. J Urol 2001;166: [2] Frede T, Erdogru T, Zukosky D, Gulkesen H, Teber D, Rassweiler J. Comparison of training modalities for performing laparoscopic radical prostatectomy: experience with 1,000 patients. J Urol 2005;174:673 8.