Robot-assisted laparoscopic prostatectomy versus open radical retropubic prostatectomy: early outcomes from a randomised controlled phase 3 study

Articles

Robot-assisted laparoscopic prostatectomy versus open radical retropubic prostatectomy: early outcomes from a randomised controlled phase 3 study John W Yaxley, Geoffrey D Coughlin, Suzanne K Chambers, Stefano Occhipinti, Hema Samaratunga, Leah Zajdlewicz, Nigel Dunglison, Rob Carter, Scott Williams, Diane J Payton, Joanna Perry-Keene, Martin F Lavin, Robert A Gardiner

Summary Background The absence of trial data comparing robot-assisted laparoscopic prostatectomy and open radical retropubic prostatectomy is a crucial knowledge gap in uro-oncology. We aimed to compare these two approaches in terms of functional and oncological outcomes and report the early postoperative outcomes at 12 weeks. Method In this randomised controlled phase 3 study, men who had newly diagnosed clinically localised prostate cancer and who had chosen surgery as their treatment approach, were able to read and speak English, had no previous history of head injury, dementia, or psychiatric illness or no other concurrent cancer, had an estimated life expectancy of 10 years or more, and were aged between 35 years and 70 years were eligible and recruited from the Royal Brisbane and Women’s Hospital (Brisbane, QLD). Participants were randomly assigned (1:1) to receive either robot-assisted laparoscopic prostatectomy or radical retropubic prostatectomy. Randomisation was computer generated and occurred in blocks of ten. This was an open trial; however, study investigators involved in data analysis were masked to each patient’s condition. Further, a masked central pathologist reviewed the biopsy and radical prostatectomy specimens. Primary outcomes were urinary function (urinary domain of EPIC) and sexual function (sexual domain of EPIC and IIEF) at 6 weeks, 12 weeks, and 24 months and oncological outcome (positive surgical margin status and biochemical and imaging evidence of progression at 24 months). The trial was powered to assess health-related and domain-specific quality of life outcomes over 24 months. We report here the early outcomes at 6 weeks and 12 weeks. The per-protocol populations were included in the primary and safety analyses. This trial was registered with the Australian New Zealand Clinical Trials Registry (ANZCTR), number ACTRN12611000661976. Findings Between Aug 23, 2010, and Nov 25, 2014, 326 men were enrolled, of whom 163 were randomly assigned to radical retropubic prostatectomy and 163 to robot-assisted laparoscopic prostatectomy. 18 withdrew (12 assigned to radical retropubic prostatectomy and six assigned to robot-assisted laparoscopic prostatectomy); thus, 151 in the radical retropubic prostatectomy group proceeded to surgery and 157 in the robot-assisted laparoscopic prostatectomy group. 121 assigned to radical retropubic prostatectomy completed the 12 week questionnaire versus 131 assigned to robot-assisted laparoscopic prostatectomy. Urinary function scores did not differ significantly between the radical retropubic prostatectomy group and robot-assisted laparoscopic prostatectomy group at 6 weeks post-surgery (74·50 vs 71·10; p=0·09) or 12 weeks post-surgery (83·80 vs 82·50; p=0·48). Sexual function scores did not differ significantly between the radical retropubic prostatectomy group and robot-assisted laparoscopic prostatectomy group at 6 weeks post-surgery (30·70 vs 32·70; p=0·45) or 12 weeks post-surgery (35·00 vs 38·90; p=0·18). Equivalence testing on the difference between the proportion of positive surgical margins between the two groups (15 [10%] in the radical retropubic prostatectomy group vs 23 [15%] in the robot-assisted laparoscopic prostatectomy group) showed that equality between the two techniques could not be established based on a 90% CI with a Δ of 10%. However, a superiority test showed that the two proportions were not significantly different (p=0·21). 14 patients (9%) in the radical retropubic prostatectomy group versus six (4%) in the robot-assisted laparoscopic prostatectomy group had postoperative complications (p=0·052). 12 (8%) men receiving radical retropubic prostatectomy and three (2%) men receiving robot-assisted laparoscopic prostatectomy experienced intraoperative adverse events. Interpretation These two techniques yield similar functional outcomes at 12 weeks. Longer term follow-up is needed. In the interim, we encourage patients to choose an experienced surgeon they trust and with whom they have rapport, rather than a specific surgical approach. Funding Cancer Council Queensland.

Introduction Prostate cancer is the second most common malignancy in men (excluding keratinocyte cancers) behind lung cancer, with more than 1 million cases estimated to be diagnosed worldwide in 2012.1 Historically, surgery has

been the dominant approach for the treatment of localised disease; however, the high incidence of iatrogenic morbidities associated with open radical retropubic prostatectomy has led to the search for less invasive treatments to improve both oncological and

www.thelancet.com Published online July 26, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30592-X

Published Online July 26, 2016 http://dx.doi.org/10.1016/ S0140-6736(16)30592-X See Online/Comment http://dx.doi.org/10.1016/ S0140-6736(16)30588-8 Department of Urology, Royal Brisbane & Women’s Hospital, Brisbane, QLD, Australia (J W Yaxley FRACS, G D Coughlin FRACS, N Dunglison FRACS, Prof R A Gardiner AM MD); Menzies Health Institute Queensland, Griffith University, Gold Coast, QLD, Australia (Prof S K Chambers PhD, S Occhipinti PhD); The University of Queensland Centre for Clinical Research, Brisbane, QLD, Australia (Prof S K Chambers, H Samaratunga FRCPA, Prof M F Lavin PhD, Prof R A Gardiner); Edith Cowan University, Perth, WA, Australia (Prof S K Chambers, Prof R A Gardiner); Cancer Council Queensland, Brisbane, QLD, Australia (Prof S K Chambers, L Zajdlewicz MOrgPsych); Prostate Cancer Foundation of Australia, Sydney, NSW, Australia (Prof S K Chambers); Aquesta Specialized Uropathology, Brisbane, QLD, Australia (H Samaratunga); Princess Alexandra Hospital, Brisbane, QLD, Australia (H Samaratunga); Deakin University, Melbourne, VIC, Australia (Prof R Carter PhD); Peter Macallum Cancer Centre, Melbourne, VIC, Australia (S Williams MBBS); and Queensland Pathology, Brisbane, QLD, Australia (D J Payton FRCPA, J Perry-Keene FRCPA) Correspondence to: Prof Robert A Gardiner AM, The University of Queensland Centre for Clinical Research, Royal Brisbane & Women’s Hospital, Herston, Brisbane, QLD 4029, Australia [email protected]

1

Articles

Research in context Evidence before this study Robotic-assisted radical prostatectomy was introduced without high level evidence. The published literature so far has consisted of non-randomised longitudinal studies of robotically assisted and open prostatectomies (considered the gold standard in terms of treatment) alone or collated in meta-analyses, although there is randomised controlled trial evidence comparing robotically assisted and laparoscopic prostatectomy showing improved functional results for the robotically assisted procedure.

open prostatectomy and shows no significant difference in outcome for standard oncological and quality of life parameters at 12 weeks. Implications of all the available evidence Over the past 2 decades, patients and clinicians have embraced robotic-assisted prostatectomy in the belief that this approach will result in better patient outcomes. If the short-term findings from this study are maintained with longer follow-up results, this could have implications for patient choice and for health provider decision making.

Added value of this study To the best of our knowledge, this is the first published randomised controlled trial to compare robotic-assisted with

quality of life outcomes. Since the first reported robotassisted laparoscopic prostatectomy by Binder and Kramer2 in 2001, rapid adoption of this surgical technique has been made such that in many countries robot-assisted laparoscopic prostatectomy is becoming the dominant surgical approach for prostatectomy. Advocates of robotassisted laparoscopic prostatectomy claim benefits of the technology lead to improved quality of life and oncological outcomes. Radical retropubic prostatectomy enthusiasts highlight the paucity of high-quality evidence for robotassisted laparoscopic prostatectomy and contend that radical retropubic prostatectomy remains the gold standard technique. This creates confusion for men with localised prostate cancer who are presented with several management options and perspectives from various sources. Perhaps no other topic in the current era of urology has attracted so much attention, controversy, and debate in our community as has robot-assisted laparoscopic prostatectomy versus radical retropubic prostatectomy.3 Building on previous research,4–6 we aimed to assess clinical and quality of life outcomes in radical retropubic prostatectomy compared with robot-assisted laparoscopic prostatectomy.

Methods Study design and participants In this phase 3 randomised controlled trial, men were recruited and managed throughout at the Royal Brisbane & Women’s Hospital (RBWH) with referrals to the public clinics from general practice and the private practices of urologists in SE Queensland in Australia (see Gardiner and colleagues7 for the published protocol). The trial commenced 18 months after the introduction of robotic surgery to our institution, which was the only public centre with a robotic facility in our state. Only one surgeon did the robot-assisted laparoscopic prostatectomy procedures at our institution until completion of the trial. Owing to an administrative oversight trial registration was 2

delayed by 10 months during which time 68 participants (22%) of the sample were consented to the trial. Inclusion criteria were that men must: (1) be newly diagnosed with clinically localised prostate cancer and have chosen surgery as their treatment approach; (2) be able to read and speak English; (3) have no previous history of head injury, dementia, or psychiatric illness; (4) have no other concurrent cancer; (5) have an estimated life expectancy of 10 years or more; and (6) be aged between 35 years and 70 years. Exclusion criteria were: (1) evidence of non-localised prostate cancer clinically; (2) prostate specific antigen (PSA) greater than 20 ng/mL; (3) previous laparoscopic hernia repair; (4) previous pelvic radiotherapy or major pelvic surgery; and (5) another malignancy within the past 5 years with the exception of non-melanoma skin cancer. Ethical approval was obtained from the Royal Brisbane & Women’s Hospital Human Research Ethics Committee.8 Patients provided written informed consent.

Randomisation and masking Participants were randomly assigned (1:1) to receive either radical retropubic prostatectomy or robot-assisted laparoscopic prostatectomy. Randomisation occurred in blocks of ten, with each condition randomly generated five times within each block, to ensure an unpredictable allocation sequence with equal numbers of men in each group at the completion of each block. This sequence was undertaken using a computerised database independent from the study investigators, clinical team, statistician, and research staff. Stratification occurred by age group (40–49 years, 50–59 years, 60-69 years), with the randomisation conducted separately for each stratum; this is, separate blocks of ten were generated for each specific age stratum. After randomisation, patients were telephoned the same day to be notified of their designated surgical procedure and posted an information booklet outlining expectations for their scheduled operation. They were then placed on the RBWH surgical

www.thelancet.com Published online July 26, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30592-X

Articles

waiting list and scheduled for surgery and managed throughout as per standard hospital procedures. This was an open trial; however, study investigators involved in data analysis were blinded to each patient’s condition. Further, a blinded central pathologist reviewed the biopsy and radical prostatectomy specimens.

334 baseline interview 329 completed 5 not completed

326 randomised*

Procedures Management was the same for all patients apart from their form of surgery, which was either robot-assisted laparoscopic prostatectomy or radical retropubic prostatectomy. To reduce surgical heterogeneity, every procedure was done by the same surgeon who had the most expertise in each approach, respectively, consistent with best practice.9 The robot-assisted laparoscopic prostatectomy surgeon had completed a 2-year robotic fellowship, followed by 200 robotic prostatectomies post-fellowship at the commencement of the trial and at trial completion had done more than 1000 robot-assisted laparoscopic prostatectomy procedures. The radical retropubic prostatectomy surgeon had 15 years post-fellowship experience and had done 1500 procedures at the commencement of the trial and more than 2000 radical procedures by trial completion. Cavernous nerve preservation was undertaken on individual surgeon discretion based on clinical staging. A pelvic lymph node dissection was done on protocol for a serum PSA greater than 10 ng/mL, a palpable nodule on digital rectal examination, or primary Gleason grade 4 or higher on biopsy sampling (ie, Gleason 4 + 3, 4 + 4, 4 + 5, or 5 + 4). A standardised template for lymphadenectomy was used by both surgeons, with the upper margin of dissection at the ureteric crossing of the iliac vessels. Epidural or spinal anaesthesia was not used routinely in either group. Patients were assessed at baseline, 24 h (pain), 1 week (pain), 6 weeks, 12 weeks, 6 months, 12 months, and 24 months. All questionnaire measures were treated as continuous and included the Expanded Prostate Cancer Index Composite (EPIC),10 International Index of Erectile Function Questionnaire (IIEF),11 International Prostate Symptom Score (IPSS),12 Short Form 36 Health Survey (SF-36),13 Surgical Pain Scale,14 Revised Impact of Events (RIES),15 Hospital Anxiety and Depression Scale (HADS),16 AQoL-8D,17 and Schover scale.18 The validity and reliability of all quality of life self-report measures have previously been established for this patient population. All clinical data were extracted by medical records review. Intraoperative events (eg, excessive blood loss, arrhythmia) were recorded as per usual practice by the surgical and anaesthetic teams and these data were then collected by the research team upon medical record review. Postoperative complications (eg, wound infection, ileus) were ascertained via medical records and patient interview throughout the trial.

163 assigned to radical retropubic prostatectomy

163 assigned to robot-assisted laparoscopic prostatectomy

163 baseline questionnaire 154 completed 9 not completed

163 baseline questionnaire 156 completed 7 not completed

12 withdrew after randomisation 3 private robotic surgery 3 cardiac issues 1 BMI >40 1 PSA >20 g/mL 1 lymphoma diagnosis 1 changed mind about having surgery 1 anaesthetic issues 1 mesh from hernia repair

6 withdrew after randomisation 1 extensive disease 2 private robot 1 religious beliefs 1 PSA >20 ng/mL 1 extensive head injuries

151 proceeded to surgery

157 proceeded to surgery

6 weeks post-surgery questionnaire

6 weeks post-surgery questionnaire

139 completed 12 not completed

137 completed 20 not completed

1 withdrew because didn’t want to complete SAQs

12 weeks post-surgery questionnaire

12 weeks post-surgery questionnaire

121 completed 29 not completed

131 completed 26 not completed

Figure: Trial profile *An additional eight men were randomised but revoked their consent. Three men revoked their consent after randomisation in the radical retropubic prostatectomy group and the robot-assisted laparoscopic prostatectomy group and a further two men following 12 weeks post-surgery in the robot-assisted laparoscopic prostatectomy group. These men have not been included in this flowchart or in any of the analyses. BMI=body-mass index. PSA=prostate specific antigen. SAQ=Self Administered Questionnaire.

Outcomes Sociodemographic and diagnostic data were assessed at baseline before surgery. Primary outcomes were urinary function (urinary domain of EPIC) and sexual function (sexual domain of EPIC and IIEF) at 6 weeks, 12 weeks, and 24 months. Oncological outcomes (positive surgical margin status and biochemical and imaging evidence of progression at 24 months) were also primary outcomes. Positive margin status was measured as a binary variable. Margins were considered positive when carcinoma was transected by an inked surgical margin; this could be in the setting of organ-confined carcinoma or with extraprostatic extension. We report here the early outcomes at 6 weeks and 12 weeks.

www.thelancet.com Published online July 26, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30592-X

3

Articles

Total (n=326)

Age, years*

Radical retropubic prostatectomy (n=163)

Robot-assisted laparoscopic prostatectomy (n=163)

60·01 (6·24)

60·38 (5·81)

59·64 (6·63) 16 (12%)

Ethnicity† European

31 (12%)

15 (13%)

Indigenous Australian

3 (1%)

2 (2%)

1 (1%)

Asian

8 (2%)

2 (2%)

4 (3%)

Pacific Islander UK/Irish Other

4 (2%)

2 (2%)

2 (1%)

193 (76%)

91 (78%)

102 (74%)

17 (7%)

5 (4%)

12 (9%)

Education† University

30 (12%)

11 (9%)

19 (14%)

Technical

107 (41%)

49 (41%)

58 (42%)

Senior high school

61 (24%)

27 (23%)

34 (24%)

Junior high school

42 (16%)

21 (18%)

21 (15%)

Primary school

13 (5%)

7 (6%)

6 (4%)

6 (2%)

5 (4%)

1 (1%)

144 (58%)

73 (63%)

71 (54%)

7 (3%)

5 (4%)

2 (2%)

54 (22%)

24 (21%)

30 (23%)

Did not complete primary school Employment† Employed full-time Unemployed or looking for work Retired Unwilling to answer

2 (1%)

1 (1%)

1 (1%)

Employed—casual

9 (4%)

2 (2%)

7 (5%)

19 (8%)

5 (4%)

14 (11%)

Employed—part-time Full-time home duties or home carer

1 (1%)

0

11 (4%)

5 (4%)

6 (5%)

<$40 000

89 (34%)

40 (33%)

49 (35%)

$40 001–$60 000

70 (27%)

37 (31%)

33 (24%)

$60 001–$80 000

48 (19%)

22 (18%)

26 (19%)

>$80 001

40 (15%)

17 (14%)

23 (17%)

Unwilling to answer/don’t know

12 (5%)

4 (3%)

Permanently ill/disabled

1 (<1%)

Household income†

8 (6%) (Table 1 continues on next page)

See Online for appendix

Secondary outcomes were pain (assessed with the Surgical Pain Scale), physical and mental functioning (SF-36), fatigue (vitality domain of SF-36), preferencebased utility scores (AQoL-8D), bowel function (EPIC), cancer specific distress (RIES scale), psychological distress (HADS), and time to return to work. The AQoL8D is a multiattribute utility measure that was included for the economic appraisal and will be reported separately. Schover scale data will be reported with the long-term outcome data. Fatigue, IPSS, pad use, and erection data are included in the appendix. We report here the early outcomes up to 12 weeks.

Statistical analysis The trial design was a two-group randomised controlled trial7 with three assessment points (baseline, 6 weeks, and 12 weeks) reported in this paper. Longitudinal analyses will involve comparisons of the rate of change with which men in each surgical treatment group achieves optimal physical and psychological quality of life (urinary, sexual and bowel 4

function, health-related quality of life that includes fatigue, cancer-specific, and psychological distress) applying a twolevel multilevel model. Assuming that at least 70% of the men in each group will remain in the study and assuming further a moderate effect size of d = 0·5 and alpha at 0·05, a total sample size of 400 (ie, 200 per group) would result in over 90% power at 24 months.19 A blinded interim analysis was pre-planned and stipulated in the trial registration when 75% of the target sample size was recruited and was overseen by an independent data monitoring committee (IDMC). This analysis was undertaken in August, 2014 (n=321), and examined differences between the two surgical types on the primary outcomes of urinary and sexual function at 12 weeks post-surgery. Effect sizes were negligible with p values for differences at 12 weeks non-significant (p range 0·26–0·97) and so on the advice of the IDMC recruitment ceased at 326 participants in September, 2014. Group differences were assessed with t tests for all continuous measures and χ² or Fisher’s exact for categorical variables at each timepoint. These included domain-specific and health-related quality of life and pain measures. All tests were two-tailed with an α of 0·05. For variables for which variances were markedly different across groups, log normal t tests were conducted as well. The equivalence of surgical margin status across the groups was examined by comparing the 90% CI of the difference in proportions of positive surgical margins in each group to a prespecified Δ of 0·1. This procedure is equivalent to two one-sided tests procedure.20 All analyses were run on a per-protocol basis. Data were analysed with Stata version 14. The trial was registered with the Australian New Zealand Clinical Trials Registry (ANZCTR), number 12611000661976.

Role of the funding source This study was funded by the Cancer Council Queensland. The funder had no role in the study design, data collection, data analysis, or data interpretation, writing of the report, or the decision to submit the paper for publication.

Results Between Aug 23, 2010, and Nov 25, 2014, 326 men were enrolled, of whom 163 were randomly assigned to radical retropubic prostatectomy and 163 to robot-assisted laparoscopic prostatectomy. 18 withdrew (12 assigned to radical retropubic prostatectomy and six assigned to robotassisted laparoscopic prostatectomy); thus, 151 in the radical retropubic prostatectomy group proceeded to surgery and 157 in the robot-assisted laparoscopic prostatectomy group. The first patient underwent surgery in October, 2010, and the last in March, 2015. One man withdrew 6 weeks post-surgery leaving 150 men in the radical retropubic prostatectomy group and 157 in the

www.thelancet.com Published online July 26, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30592-X

Articles

robot-assisted laparoscopic prostatectomy group remaining in the trial 12 weeks post-surgery (figure 1). Baseline sociodemographic characteristics were well balanced between the two surgery groups (table 1). Urinary function scores did not differ significantly between the radical retropubic prostatectomy group and robot-assisted laparoscopic prostatectomy group at 6 weeks post-surgery (74·50 vs 71·10; p=0·09) or 12 weeks post-surgery (83·80 vs 82·50; p=0·48; table 2). Sexual function scores did not differ significantly between the radical retropubic prostatectomy group and robot-assisted laparoscopic prostatectomy group at 6 weeks post-surgery (30·70 vs 32·70; p=0·45) or 12 weeks post-surgery (35·00 vs 38·90; p=0·18). Descriptive data for the use of pads by men for incontinence and the quality of their erections as well as domain-specific and health-related quality of life outcomes show, in the main, few differences between the trial groups at every timepoint (appendix). Equivalence analyses were done for surgical margins (table 3; present=1, absent=0) between the surgical groups with the two-sided 90% CI of the difference between the proportion of positive margins. The point estimate of the difference between proportions of positive margins was 0·052 with an associated 90% CI of –0·01 to 0·11. The upper bound of this interval was above the specified Δ of 0·10 and so equivalence could not be established. However, when comparing the proportion in the radical retropubic prostatectomy group, which contained fewer positive margins, with the proportion in the robot-assisted laparoscopic prostatectomy group (15 [10%] in the radical retropubic prostatectomy group vs 23 [15%] in the robot-assisted laparoscopic prostatectomy group), analyses showed that these groups were not significantly different (p=0·21). Pain was measured (with the Surgical Pain Scale) postoperatively at 24 h, 1 week, 6 weeks, and 12 weeks after surgery (table 4). For pain at rest, no significant differences were noted between the trial groups at any of the timepoints. For pain during normal activities, participants in the robot-assisted laparoscopic prostatectomy group reported significantly less pain than those in the radical retropubic prostatectomy group in the very early postoperative period (24 h and 1 week timepoints), but not at 6 weeks or 12 weeks after surgery. A similar pattern was noted for the worst reported pain measure; participants in the robot-assisted laparoscopic prostatectomy group reported less pain than those in the radical retropubic prostatectomy group at both 24 h and 1 week post-surgery, but not at later assessments. For physical and mental quality of life as measured by the SF-36, one significant difference between the groups was observed at the 6 week timepoint with patients who received robot-assisted laparoscopic prostatectomy reporting higher physical quality of life than those who received radical retropubic prostatectomy. However, this significant difference was not noted at 12 weeks.

Total (n= 326)

Radical retropubic prostatectomy (n=163)

Robot-assisted laparoscopic prostatectomy (n=163)

(continued from previous page) Comorbid condition‡ Myocardial infarction

21 (7%)

11 (7%)

Heart failure

13 (4%)

6 (4%)

7 (5%)

Angina or abdominal aortic aneurysm

14 (5%)

6 (4%)

8 (5%)

Peptic ulcer disease

14 (5%)

8 (5%)

6 (4%)

7 (2%)

4 (3%)

3 (2%)

Diabetes mellitus

28 (9%)

12 (8%)

16 (10%)

Chronic obstructive pulmonary disease

18 (6%)

9 (6%)

9 (6%)

Cerebrovascular accident

10 (6%)

Dementia

0

0

0

Connective tissue/autoimmune disease

4 (1%)

2 (1%)

2 (1%)

Liver disease

6 (2%)

1 (1%)

5 (3%)

Kidney disease

13 (4%)

8 (5%)

5 (3%)

Bowel disease

10 (3%)

5 (3%)

5 (3%)

Hypertension

117 (38%)

58 (38%)

59 (38%)

Hypercholesterolaemia

103 (33%)

50 (33%)

53 (34%)

Data are mean (SD) or n (%). Numbers differ owing to differential proportions of missing data across sociodemographic variables. *For age, 317 in total, 158 in the radical retropubic prostatectomy group, 159 in the robot-assisted laparoscopic prostatectomy group. †For ethnicity, education, employment, and household income; 253 in total, 118 in the radical retropubic prostatectomy group, and 135 in the robot-assisted laparoscopic prostatectomy group (plus or minus four depending on the amount of missing data). ‡For comorbidity, 307 in total, 152 in the radical retropubic prostatectomy group and 155 in the robot-assisted laparoscopic prostatectomy group.

Table 1: Demographic characteristics by surgery type

No significant differences were noted between the groups for bowel quality of life (table 2). There was a significant difference between groups at 12 weeks for the HADS (3·51 in the radical retropubic prostatectomy group vs 2·30 in the robot-assisted laparoscopic prostatectomy group; p=0·03). Men who received radical retropubic prostatectomy were significantly more distressed at 12 weeks compared with men who received robot-assisted laparoscopic prostatectomy group, however the effect size was small (Cohen’s d=0·27). Return to work measures were examined at 12 weeks post-surgery only for those who had reported being in full-time or part-time work at the time of their surgeries (66 in the radical retropubic prostatectomy group and 63 in the robot-assisted laparoscopic prostatectomy group). No significant difference was noted between the treatment groups in length of time away from work in days (radical retropubic prostatectomy, M=42·71 [95% CI 41·09–53·30]; robot-assisted laparoscopic prostatectomy, M=42·71 [30·98–54·45]; p=0·49). Duration of operation was longer for patients in the radical retropubic prostatectomy group than for those in the robot-assisted laparoscopic prostatectomy group (operating room time, p<0·0001; surgery time, p<0·0001); however, no significant difference was noted between the groups for the amount of time spent in recovery (p=0·95).

www.thelancet.com Published online July 26, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30592-X

5

Articles

Range

Baseline

6 weeks

12 weeks Radical retropubic prostatectomy (n=119)

Robot-assisted laparoscopic prostatectomy (n=129)

p value

0·09

83·80 (81·33–86·17)

82·50 (80·23–84·86)

0·48

32·70 (28·67–36·79)

0·45

35·00 (30·94–39·06)

38·90 (34·83–42·96)

0·18

23·75 (21·03–26·47)

25·63 (22·35–28·92)

0·38

27·56 (24·30–30·81)

30·14 (26·46–30·81)

0·31

0·52

51·96 (50·71–53·20)

53·70 (52·70–54·70)

0·03

59·39 (58·39–60·39)

59·57 (58·51–60·63)

0·81

47·34 (45·61–49·07)

0·17

45·83 (43·72–47·94)

47·99 (46·19–49·79)

0·13

47·45 (45·49–49·41)

49·52 (47·82–51·21)

0·97

93·70 (92·23–95·21)

94·10 (92·74–95·41)

0·72

92·30 (90·82–93·82)

91·40 (89·66–93·14)

0·43

93·70 (91·80–95·54)

94·50 (93·32–95·64)

0·46

0–88

14·92 (12·52-17·33)

12·65 (10·52-14·79)

0·16

8·65 (6·52-10·79)

6·71 (5·19-8·23)

0·15

6·47 (4·65-8·29)

4·30 (2·91-5·69)

0·06

0–42

8·35 (7·25-9·46)

7·82 (6·76-8·87)

0·49

6·85 (5·66-8·04)

6·16 (5·10-7·22)

0·39

7·03 (5·78-8·28)

5·26 (4·16-6·36)

0·04‡

Radical retropubic prostatectomy (n=136)

Robot-assisted laparoscopic prostatectomy (n=131)

p value

0·83

74·50 (72·77–77·17)

71·10 (68·22–73·95)

63·05 (59·32–66·78)

0·25

30·70 (27·18–34·21)

43·96 (40·35–47·58)

46·65 (43·07–50·23)

0·30

0–100

59·29 (58·17–60·41)

59·77 (58·79–60·75)

0–100

45·57 (43·71–47·43)

EPIC—bowel domain

0–100

RIES* HADS†

Radical retropubic prostatectomy (n=152)

Robot-assisted laparoscopic prostatectomy (n=153)

p value

0–100

88·79 (86·74–90·85)

88·50 (86·59–90·40)

EPIC—sexual domain

0–100

59·80 (55·66–63·92)

Domain-specific QoL— IIEF total

0–75

Physical functioning Mental functioning

Primary Urinary function EPIC—urinary domain Sexual function

Secondary Health-related QoL (SF36)

Data are M (95% CI) or n (%). Higher scores indicate better functioning on all scales. Numbers are within plus or minus 3. EPIC=Expanded Prostate Cancer Index Composite. QoL=quality of life. IIEF=International Index of Erectile Function. SF36=Short Form 36. RIES=Revised Impact of Events Scale. HADS=Hospital Anxiety and Depression Scale. *The RIES analyses were repeated after log transforming the data. No significant differences were noted between groups at baseline, 6 weeks, or 12 weeks. †The HADS analyses were repeated after log transforming the data. There was a significant difference between groups at 12 weeks (log transformed means: 3·51 for radical retropubic prostatectomy vs 2·30 for robot-assisted laparoscopic prostatectomy, p=0.03). The effect size was small (Cohen’s d=0.27). There were no significant differences between groups at baseline or 6 weeks.

Table 2: Primary and secondary outcomes across all timepoints and by surgery type

12 (8%) men receiving radical retropubic prostatectomy and three (2%) men receiving robot-assisted laparoscopic prostatectomy experienced intraoperative adverse events. The exact nature of the intraoperative adverse events is not reported due to the low prevalence (p=0·02; table 3). Patients in the radical retropubic prostatectomy group had about three times greater estimated total blood loss (p<0·0001), and a longer average time spent in hospital after surgery (p<0·0001) than with those in the robotassisted laparoscopic prostatectomy group (table 3). All other perioperative measures were comparable (p=0·12–0·95; table 3). The modified Clavien Classification System21 was used to stratify postoperative complications into five grades, increasing in level of intervention (table 3). 24 complications were encountered in 20 (6%) patients; 17 complications were in 14 patients (9%) in the radical retropubic prostatectomy group and seven complications in six patients in the robot-assisted laparoscopic prostatectomy group (4%). The number of patients who experienced postoperative complications did not differ significantly between the two groups (p=0·052). There were two (1%) unplanned Clavien IVa admissions to intensive care units in the open radical retropubic prostatectomy group for cardiac events requiring inotrope medications. No grade 6

VIb or V complications were encountered in either group after surgery to 12 weeks of follow-up. No significant differences were noted between the two surgical groups on pathological indices (table 3; p=0·21–0·94) with the exception of the number of lymph nodes removed; with more lymph nodes recorded for patients in the robot-assisted laparoscopic prostatectomy group (M=6·50) than those in the radical retropubic prostatectomy group (M=3·26).

Discussion A meta-analysis showed improved early functional outcomes for robot-assisted laparoscopic prostatectomy compared with radical retropubic prostatectomy.22,23 Our study does not support this finding, with no significant differences noted at any time up to 12 weeks between the groups with regard to the quality of life parameters of return of early urinary and sexual function. A  recent prospective, controlled, non-randomised trial comparing robot-assisted laparoscopic prostatectomy with radical retropubic prostatectomy noted a modest benefit in preserving erectile function for robot-assisted laparoscopic prostatectomy at 12 months (70% erectile dysfunction with robot-assisted laparoscopic prostatectomy vs 75% with radical retropubic

www.thelancet.com Published online July 26, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30592-X

Articles

prostatectomy).24 The absence of neurovascular bundle tissue on histology was equal in both groups in this study. Although no significant difference between groups in the early ability to penetrate postoperatively was noted, erectile function can continue to improve for up to 3 years after surgery,25 so any difference in outcomes between these surgical techniques might not become apparent for some time. Similar to another recent meta-analysis,26 our study did not show a significant difference in positive surgical margins between the two techniques (15% in the robotassisted laparoscopic prostatectomy group and 10% in the radical retropubic prostatectomy group); these rates were low in comparison with large series data.27 Surgical margin status is an important prognostic oncological factor, with positive surgical margins associated with an increased rate of biochemical relapse and an increased probability of subsequent treatment.28 Positive surgical margin rates for radical prostatectomy vary with surgeon experience, cancer stage, cancer volume, and cancer grade, ranging widely in series from 7% to 32%,29 10% to 44%,30 and in a recent large case series, from 16% to 24%.31 Pathologists’ assessments of positive surgical margins vary depending on experience and expertise. In some cases, interpretation can be difficult and can result in overdiagnosis of positive surgical margin status.32 Negative surgical resection margin rates in organconfined stage pT2 prostate cancer are considered a barometer of surgical technique. By contrast, it is accepted that patients with extracapsular extension (pT3) have an increased risk of positive surgical margins, with positive margin status depending upon several factors, including the extent of extracapsular extension and also the extent that each surgeon excises the neurovascular bundle or tissue around the prostate. In this study, the overall positive surgical margin for pT3 tumours was 10%, with numbers too few for subanalysis. No significant difference in the number of pelvic lymph node dissections done was noted in either group. An unexpected finding was a higher mean number of lymph nodes removed and more positive lymph nodes removed with robot-assisted laparoscopic prostatectomy than with open radical retropubic prostatectomy. Our study was not designed to investigate differences in lymph node retrieval. The number of lymph nodes noted on histology depends on several factors, including the surgical boundaries of dissection, patient and tumour characteristics, the experience of the urologist, size of lymph node packets removed,33,34 and the method of processing and blocking in the laboratory.35 Robot-assisted laparoscopic prostatectomy was shown to provide benefits of a minimally invasive surgical approach. Consistent with the SEER-MEDICARE based results from Hu and colleagues27 and Gandaglia and colleagues,36 there was less blood loss in the robot-assisted laparoscopic prostatectomy group and a shorter length of stay than in the open radical retropubic prostatectomy

Total (n=308)

Radical retropubic prostatectomy (n=151)

Robot-assisted laparoscopic prostatectomy (n=157)

p value

<0·0001

Perioperative outcomes Operative duration Surgery, min

217·97 (47·63)

234·34 (37·07)

202·03 (51·36)

Recovery, min*

107·54 (111·64)

107·12 (146·63)

107·94 (61·18)

0·95

Operating room, min

263·00 (49·79)

280·37 (36·36)

246·08 (55·12)

<0·0001

Intraoperative adverse event Estimated total blood loss, mL

15 (5%) 886·54 (645·62)

12 (8%) 1338·14 (591·47)

3 (2%) 443·74 (294·29)

0·02 <0·0001

Blood transfusions Non-autologous intraoperative

0

0

0

Non-autologous postoperative

7 (2%)

6 (4%)

1 (1%)

Planned

6 (2%)

3 (1%)

3 (2%)

Unplanned

5 (2%)

5 (3%)

0

Readmission

20 (7%)

12 (8%)

Admitted to intensive care unit

·· 0·12 0·18

8 (5%)

·· ·· 0·32

Indwelling catheter, days

8·31 (3·47)

8·42 (3·28)

8·21 (3·64)

0·59

Length of hospital stay, days

2·39 (2·30)

3·27 (1·49)

1·55 (2·61)

<0·0001

Postoperative complications†

20, 24 (6%)

14, 17 (9%)

6, 7(4%)

0·05

Grade I

10, 10 (3%)

6, 6 (4%)

4, 4 (3%)

··

Grade II

5, 6 (2%)

3, 4 (2%)

2, 2 (1%)

··

Grade IIIa

3, 3 (1%)

2, 2 (1%)

1, 1 (1%)

··

Grade IIIb

3, 3 (1%)

3, 3 (2%)

0, 0

··

Grade IVa

2, 2 (<1%)

2, 2 (<1%)

0, 0

··

7·49 (4·08)

7·57 (4·07)

7·41 (4·10)

Pathological characteristics Preoperative PSA, ng/mL Preoperative Gleason score ≤6

0·77 0·51

43 (16%)

20 (15%)

23 (18%)

··

7 (3+4)

126 (48%)

68 (50%)

58 (45%)

··

7 (4+3)

53 (20%)

24 (18%)

29 (22%)

··

8

21 (8%)

9 (7%)

12 (9%)

··

9 (4+5)

19 (7%)

11 (8%)

8 (6%)

··

9 (5+4)

3 (1%)

3 (2%)

0

2·35 (2·25)

2·40 (2·20)

2·31 (2·32)

11 (4%)

5 (3%)

6 (4%)

··

7 (3+4)

144 (47%)

72 (48%)

72 (46%)

··

7 (4+3)

120 (39%)

57 (38%)

63 (40%)

··

2 (1%)

·· ··

Tumour volume, mL Prostatectomy Gleason score ≤6

8

0·61

2 (1%)

0

9 (4+5)

28 (9%)

15 (10%)

13 (8%)

9 (5+4)

3 (1%)

2 (1%)

1 (1%)

Extraprostatic extension None

·· 0·74

·· 0·77

204 (66%)

102 (68%)

102 (65%)

··

Non-focal

87 (28%)

42 (28%)

45 (29%)

··

Focal

17 (6%)

7 (5%)

10 (6%)

Seminal vesicle involvement Not involved Involved

www.thelancet.com Published online July 26, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30592-X

·· 0·24

294 (95%)

142 (94%)

152 (97%)

··

14 (5%)

9 (6%)

5 (3%)

··

(Table 3 continues on next page)

7

Articles

group. No patient required an intraoperative blood transfusion in either group. The mean blood loss in the open radical retropubic prostatectomy group of Total (n=308)

Radical retropubic prostatectomy (n=151)

Robot-assisted laparoscopic prostatectomy (n=157)

p value

(Continued from previous page) Lymphovascular involvement None Present

0·21 234 (76%)

110 (73%)

124 (79%)

74 (24%)

41 (27%)

33 (21%)

Perineural involvement† None Present

·· ·· 0·37

19 (7%)

7 (5%)

12 (8%)

255 (93%)

121 (95%)

134 (92%)

Surgical margins

·· ·· 0·21

Negative

270 (88%)

136 (90%)

134 (85%)

Positive

38 (12%)

15 (10%)

23 (15%)

··

T2

8 (3%)

3 (2%)

5 (3%)

··

≥T3

30 (10%)

12 (8%)

18 (11%)

Neurovascular bundle Absent

··

·· 0·94

212 (69%)

104 (69%)

108 (69%)

··

Present and negative

67 (22%)

32 (21%)

35 (22%)

··

Present and positive

29 (9%)

15 (10%)

14 (9%)

PLND status

·· 0·62

Not performed

196 (64%)

98 (65%)

98 (62%)

··

Performed, –ve

103 (33%)

50 (33%)

53 (34%)

··

Performed, +ve

9 (3%)

3 (2%)

6 (4%)

Lymph node yield

·· 0·0004

Mean (SD)

4·92 (8·48)

3·26 (5·45)

6·50 (10·38)

··

+ve

9 (3%)

2 (1%)

7 (4%)

··

Data are mean (SD), n (%), or number of patients, number of complications (%). PSA=prostate specific antigen. PLND=positive lymph node dissection. *The recovery time analyses were repeated after log transforming the data, which yielded a significant difference between the groups in min in recovery (transformed means: 96·42 for radical retropubic prostatectomy and 112·86 for robot-assisted laparoscopic prostatectomy). Although this difference was significant it was not judged to be of clinical importance. †Numbers of postoperative complication grade are not additive because patients experienced more than one complication: 274 in total had perineural involvement, 128 in the radical retropubic prostatectomy group and 146 in the robot-assisted laparoscopic prostatectomy group; all other numbers are within plus or minus 10 depending on the amount of missing data . Differences are between groups within each timepoint for all p values.

Table 3: Perioperative outcomes, postoperative complications, and pathological characteristics by surgery type

24 h

1 week

12 weeks

Robot-assisted p value laparoscopic prostatectomy (n=152)

Radical retropubic prostatectomy (n=136)

Robot-assisted p value laparoscopic prostatectomy (n=138)

Radical retropubic prostatectomy (n=120)

Robot-assisted laparoscopic prostatectomy (n=130)

p value

0·97

1·98 (1·71–2·24)

1·74 (1·47–2·01)

0·21

0·78 (0·56–0·99)

0·82 (0·55–1·10)

0·79

0·48 (0·29–0·66)

0·39 (0·21–0·58)

0·54

4·60 (4·25–4·95)

<0·0001

3·19 (2·91–3·48)

2·51 (2·19–2·82)

0·002

1·07 (0·84–1·31)

0·97 (0·73–1·21)

0·55

0·61 (0·38–0·83)

0·55 (0·33–0·77)

0·70

5·30 (4·92–5·68)

<0·0001

3·50 (3·12–3·88)

2·37 (2·01–2·73)

<0·0001

0·88 (0·62–1·13)

0·72 (0·49–0·95)

0·37

0·48 (0·25–0·72)

0·49 (0·26–0·72)

0·96

Robot-assisted laparoscopic prostatectomy (n=155)

3·02 (2·69–3·35)

3·01 (2·70–3·32)

Pain during 5·83 activities (5·44–6·23) Worst pain

6·45 (6·08–6·81)

6 weeks

Radical retropubic prostatectomy (n=144)

Radical retropubic prostatectomy (n=148) Pain at rest

1338 cm³ was offset by routine use of the Cell Saver Autologous Blood Recovery System, with six patients requiring postoperative non-autologous blood products, compared with one in the robot-assisted laparoscopic prostatectomy cohort. Patients who received robotassisted laparoscopic prostatectomy experienced fewer intraoperative adverse events. Pain with normal activities, including the worst pain recorded, was significantly less at 24 h and 1 week after surgery for patients in the robotassisted laparoscopic prostatectomy group who also displayed better physical quality of life scores at 6 weeks. However, despite these minimal invasive advantages of robot-assisted laparoscopic prostatectomy, this did not translate into an earlier return to work for men in the robot-assisted laparoscopic prostatectomy group. A limitation of our study was that as a single institution trial in which only two surgeons performed surgeries, our findings might not generalise to other settings. However, this limitation creates other strengths. Our trial commenced after the introduction of robotic technology to our state and after we had optimised integration of its use with our centre—the only public robot facility during most of the trial. This could have been a factor in our ability to successfully recruit, given that international attempts to do a multi-institutional, multisurgeon radical retropubic prostatectomy versus robot-assisted laparoscopic prostatectomy trial have found this challenging. As well, in addition to minimising variability by having standard perioperative procedures with all patients treated in the same specialised urology unit, we controlled for surgeon heterogeneity by having a single surgeon, per study group, who was experienced and expert in the respective surgical approaches, perform all the prostatectomy procedures. We recognise there was a difference in experience of the surgeons and it is likely that the operative technique of the robotic surgeon continued to evolve during the study. However, both surgeons achieved technical outcomes at least equivalent to their peers in international multicentre, multisurgeon publications,24 so it is unlikely a learning curve contributed substantially to the results. Finally, penile

p value

Data are M (95% CI). Higher scores indicate more pain, measured on a scale from 0 “no pain” to 10 “most intense pain”. Differences are between groups within each timepoint for all p values. Numbers are within plus or minus 2 depending on the amount of missing data.

Table 4: Pain ratings across timepoints and by surgery type

8

www.thelancet.com Published online July 26, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30592-X

Articles

rehabilitation was unable to be offered to participants as part of the trial, which could have an effect on long-term erectile functional results affecting both study groups. This randomised phase 3 trial of robot-assisted laparoscopic prostatectomy versus radical retropubic prostatectomy standardised for all relevant parameters did not find a difference in domain-specific quality of life or pathological outcomes at 12 weeks for the two surgical approaches. In brief, both approaches have shown good early results, with minimally invasive benefits seen in the robot-assisted laparoscopic prostatectomy group. Urinary and sexual function are expected to continue to improve with time and, as such, significant differences in functional outcome between these surgical approaches might not become apparent until longer follow-up, which we plan to publish when the final patient has completed his 2 year follow-up assessment. In the interim we encourage patients to choose an experienced surgeon they trust and with whom they have rapport, rather than choose a specific surgical approach. Contributors JWY did the literature search, designed the study, interpreted the data, and wrote the report. GDC did the literature search, interpreted the data, and wrote the report. SKC did the literature search, designed the study, analysed the data, interpreted the data, and wrote the report. SO designed the study, analysed the data, interpreted the data, and wrote the report. HS interpreted the data. LZ collected the data, analysed the data, and wrote the report. ND interpreted the data and wrote the report. RC designed the study. SW designed the study and wrote the report. DJP designed the study. JP-K designed the study. MFL designed the study and wrote the report. RAG did the literature search, designed the study, collected the data, interpreted the data, and wrote the report. Declaration of interests We declare no competing interests. Acknowledgments We thank the Cancer Council Queensland for providing the grant for this research project; the patients for supporting this study; the medical administration at Royal Brisbane & Women’s Hospital, in particular Judy Graves, Keith McNeill, and Barry O’Loughlin, for funding the robotic programme; the Independent Data Monitoring Committee (Joanne Aitken [Cancer Council Queensland; Menzies Health Institute Queensland, Griffith University; School of Public Health and Social Work, QUT], David Smith [Cancer Council New South Wales; Menzies Health Institute Queensland, Griffith University], and Mark Frydenberg AM [Department of Surgery, Monash University; Department of Urology, Monash Health]); and Robyn Medcraft and Nigel Bennett for assistance with data management. References 1 Ferlay J, Soerjomataram I, Ervik M, et al. GLOBOCAN 2012 v1.0, Cancer incidence and mortality worldwide: IARC CancerBase No. 11, 2012 http://globocan.iarc.fr/Pages/fact_sheets_cancer.aspx (accessed July 22, 2016). 2 Binder J, Kramer W. Robotically-assisted laparoscopic radical prostatectomy. BJU Int 2001; 87: 408–10. 3 Sundi D, Han M. Limitations of assessing value in robotic surgery for prostate cancer: what data should patients and physicians use to make the best decision? J Clin Oncol 2014; 32: 1394–95. 4 Wood DP, Schulte R, Dunn RL, et al. Short-term health outcome differences between robotic and conventional radical prostatectomy. Urology 2007; 70: 945–49. 5 Ball AJ, Gambill B, Fabrizio MD, et al. Prospective longitudinal comparative study of early health-related quality-of-life outcomes in patients undergoing surgical treatment for localized prostate cancer: a short-term evaluation of five approaches from a single institution. J Endourol 2006; 20: 723–31.

6 7

8

9

10

11

12

13

14 15 16 17

18

19 20

21

22

23

24

25

26

27

28 29

Smith JA Jr. Radical prostatectomy for low risk carcinoma of the prostate. World J Urol 2008; 26: 443–46. Gardiner RA, Yaxley J, Coughlin G, et al. A randomised trial of robotic and open prostatectomy in men with localised prostate cancer. BMC Cancer 2012; 12: 189. Altman DG, Schulz KF, Moher D, et al. The revised CONSORT statement for reporting randomized trials: explanation and elaboration. Ann Intern Med 2001; 134: 663–94. Atug F, Castle EP, Srivastav SK, Burgess SV, Thomas R, Davis R. Positive surgical margins in robotic-assisted radical prostatectomy: impact of learning curve on oncologic outcomes. Eur Urol 2006; 49: 866–71. Wei JT, Dunn RL, Litwin MS, Sandler HM, Sanda MG. Development and validation of the expanded prostate cancer index composite (EPIC) for comprehensive assessment of health-related quality of life in men with prostate cancer. Urology 2000; 56: 899–905. Rosen RC, Riley A, Wagner G, Osterloh IH, Kirkpatrick J, Mishra A. The international index of erectile function (IIEF): a multidimensional scale for assessment of erectile dysfunction. Urology 1997; 49: 822–30. Barry MJ, Fowler FJ Jr, O’Leary MP, et al. The American Urological Association symptom index for benign prostatic hyperplasia. The Measurement Committee of the American Urological Association. J Urol 1992; 148: 1549–57. Medical Outcomes Trust and Quality Metric Incorporated. SF-36: SF-36v2TM Health Survey; (IQOLA SF36v2 Standard, English (Australia), 7/03). Lincoln, RI: Medical Outcomes Trust and Quality Metric Incorporated, Health Assessment Lab, 1992, 2003. McCarthy M Jr, Chang CH, Pickard AS, et al. Visual analog scales for assessing surgical pain. J Am Coll Surg 2005; 201: 245–52. Horowitz M, Wilner N, Alvarez W. Impact of Event Scale: a measure of subjective stress. Psychosom Med 1979; 41: 209–18. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983; 67: 361–70. Richardson J, Iezzi A, Khan MA, Maxwell A. Validity and reliability of the assessment of quality of life (AQoL)-8D multi-attribute utility instrument. Patient 2014; 7: 85–96. Schover LR, Canada AL, Yuan Y, et al. A randomized trial of internet-based versus traditional sexual counseling for couples after localized prostate cancer treatment. Cancer 2012; 118: 500–09. Jo B. Statistical power in randomized intervention studies with noncompliance. Psychol Methods 2002; 7: 178–93. Schuirmann DJ. A comparison of the two one-sided tests procedure and the power approach for assessing the equivalence of average bioavailability. J Pharmacokinet Biopharm 1987;15: 657–80. Dindo D, Demartines N, 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: 205–13. Ficarra V, Novara G, Ahlering TE, et al. Systematic review and meta-analysis of studies reporting potency rates after robot-assisted radical prostatectomy. Eur Urol 2012; 62: 418–30. Ficarra V, Novara G, Rosen RC, et al. Systematic review and meta-analysis of studies reporting urinary continence recovery after robot-assisted radical prostatectomy. Eur Urol 2012; 62: 405–17. Haglind E, Carlsson S, Stranne J, et al. Urinary incontinence and erectile dysfunction after robotic versus open radical prostatectomy: a prospective, controlled, nonrandomised trial. Eur Urol 2015; 68: 216–25. Isgoren AE, Saitz TR, Serefoglu EC. Erectile function outcomes after robot-assisted radical prostatectomy: is it superior to open retropubic or laparoscopic approach? Sex Med Rev 2014; 2: 10–23. Novara G, Ficarra V, Mocellin S, et al. Systematic review and meta-analysis of studies reporting oncologic outcome after robot-assisted radical prostatectomy. Eur Urol 2012; 62: 382–404. Hu JC, Gandaglia G, Karakiewicz PI, et al. Comparative effectiveness of robot-assisted versus open radical prostatectomy cancer control. Eur Urol 2014; 66: 666–72. Meeks JJ, Eastham JA. Radical prostatectomy: positive surgical margins matter. Urol Oncol 2013; 31: 974–79. Yossepowitch O, Briganti A, Eastham JA, et al. Positive surgical margins after radical prostatectomy: a systematic review and contemporary update. Eur Urol 2014; 65: 303–13.

www.thelancet.com Published online July 26, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30592-X

9

Articles

30

31

32

33

10

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: 2292–95. Suardi N, DellOglio P, Gallina A, et al. Evaluation of positive surgical margins in patients undergoing robot-assisted and open radical prostatectomy according to preoperative risk groups. Urol Oncol 2016; 34: 57. Chuang AY, Epstein JI. Positive surgical margins in areas of capsular incision in otherwise organ-confined disease at radical prostatectomy: histologic features and pitfalls. Am J Surg Pathol 2008; 32: 1201–06. Liss MA, Palazzi K, Stroup SP, Jabaji R, Raheem OA, Kane CJ. Outcomes and complications of pelvic lymph node dissection during robotic-assisted radical prostatectomy. World J Urol 2013; 31: 481–88.

34

35

36

Jung JH, Seo JW, Lim MS, et al. Extended pelvic lymph node dissection including internal iliac packet should be performed during robot-assisted laparoscopic radical prostatectomy for high-risk prostate cancer. J Laparoendosc Adv Surg Tech A 2012; 22: 785–90. Perry-Keene J, Ferguson P, Samaratunga H, Nacey JN, Delahunt B. Total submission of pelvic lymphadenectomy tissues removed during radical prostatectomy for prostate cancer increases lymph node yield and detection of micrometastases. Histopathology 2014; 64: 399–404. Gandaglia G, Sammon JD, Chang SL, et al. Comparative effectiveness of robot-assisted and open radical prostatectomy in the postdissemination era. J Clin Oncol 2014; 32: 1419–26.

www.thelancet.com Published online July 26, 2016 http://dx.doi.org/10.1016/S0140-6736(16)30592-X