Open versus laparoscopic surgery for mid or low rectal cancer after neoadjuvant chemoradiotherapy (COREAN trial): short-term outcomes of an open-label randomised controlled trial

Articles

Open versus laparoscopic surgery for mid or low rectal cancer after neoadjuvant chemoradiotherapy (COREAN trial): short-term outcomes of an open-label randomised controlled trial Sung-Bum Kang, Ji Won Park, Seung-Yong Jeong, Byung Ho Nam, Hyo Seong Choi, Duck-Woo Kim, Seok-Byung Lim, Taek-Gu Lee, Dae Yong Kim, Jae-Sung Kim, Hee Jin Chang, Hye-Seung Lee, Sun Young Kim, Kyung Hae Jung, Yong Sang Hong, Jee Hyun Kim, Dae Kyung Sohn, Dae-Hyun Kim, Jae Hwan Oh

Summary Background The safety and short-term efficacy of laparoscopic surgery for rectal cancer after preoperative chemoradiotherapy has not been demonstrated. The aim of the randomised Comparison of Open versus laparoscopic surgery for mid and low REctal cancer After Neoadjuvant chemoradiotherapy (COREAN) trial was to compare open surgery with laparoscopic surgery for mid or low rectal cancer after neoadjuvant chemoradiotherapy.

Lancet Oncol 2010; 11: 637–45

Methods Between April 4, 2006, and Aug 26, 2009, patients with cT3N0–2 mid or low rectal cancer without distant metastasis after preoperative chemoradiotherapy were enrolled at three tertiary-referral hospitals. Patients were randomised 1:1 to receive either open surgery (n=170) or laparoscopic surgery (n=170), stratified according to sex and preoperative chemotherapy regimen. Short-term outcomes assessed were involvement of the circumferential resection margin, macroscopic quality of the total mesorectal excision specimen, number of harvested lymph nodes, recovery of bowel function, perioperative morbidity, postoperative pain, and quality of life. Analyses were based on the intention-to-treat population. Patients continue to be followed up for the primary outcome (3-year disease-free survival). This study is registered with ClinicalTrials.gov, number NCT00470951.

See Reflection and Reaction page 606

Findings Two patients (1·2%) in the laparoscopic group were converted to open surgery, but were included in the laparoscopic group for analyses. Estimated blood loss was less in the laparoscopic group than in the open group (median 217·5 mL [150·0–400·0] in the open group vs 200·0 mL [100·0–300·0] in the laparoscopic group, p=0·006), although surgery time was longer in the laparoscopic group (mean 244·9 min [SD 75·4] vs 197·0 min [62·9], p<0·0001). Involvement of the circumferential resection margin, macroscopic quality of the total mesorectal excision specimen, number of harvested lymph nodes, and perioperative morbidity did not differ between the two groups. The laparoscopic surgery group showed earlier recovery of bowel function than the open surgery group (time to pass first flatus, median 38·5 h [23·0–53·0] vs 60·0 h [43·0–73·0], p<0·0001; time to resume a normal diet, 85·0 h [66·0–95·0] vs 93·0 h [86·0–121·0], p<0·0001; time to first defecation, 96·5 h [70·0–125·0] vs 123 h [94·0–156·0], p<0·0001). The total amount of morphine used was less in the laparoscopic group than in the open group (median 107·2 mg [80·0–150·0] vs 156·9 mg [117·0–185·2], p<0·0001). 3 months after proctectomy or ileostomy takedown, the laparoscopic group showed better physical functioning score than the open group (0·501 [n=122] vs –4·970 [n=128], p=0·0073), less fatigue (–5·659 [n=122] vs 0·098 [n=129], p=0·0206), and fewer micturition (–2·583 [n=122] vs 4·725 [n=129], p=0·0002), gastrointestinal (–0·400 [n=122] vs 4·331 [n=129], p=0·0102), and defecation problems (0·535 [n=103] vs 5·327 [n=99], p=0·0184) in repeated measures analysis of covariance, adjusted for baseline values. Interpretation Laparoscopic surgery after preoperative chemoradiotherapy for mid or low rectal cancer is safe and has short-term benefits compared with open surgery; the quality of oncological resection was equivalent. Funding The National Cancer Center, South Korea.

Introduction Laparoscopic resection for rectal cancer does not have level 1 evidence in surgical practice, although laparoscopic surgery for colon cancer has been growing in popularity based on oncological evidence.1–5 Subset analysis of the UK Medical Research Council (MRC) Conventional versus Laparoscopic-Assisted Surgery in Colorectal Cancer (CLASICC) trial reported a 34% conversion rate with 59% morbidity for 30 days after www.thelancet.com/oncology Vol 11 July 2010

laparoscopic surgery for rectal cancer.6 Although a randomised trial of sigmoid colon cancer, including upper rectal cancer, showed that laparoscopic surgery had short-term benefits with similar oncological outcomes to open surgery,3 the results were difficult to interpret because of the heterogeneity of tumour locations.3,4 Laparoscopic procedures for rectal cancer are regarded as technically demanding7 because total mesorectal excision (TME) and autonomic nerve

Published Online June 17, 2010 DOI:10.1016/S14702045(10)70131-5

Department of Surgery (S-B Kang MD, D-W Kim MD, T-G Lee MD), Department of Radiation Oncology (Prof J-S Kim MD), Department of Pathology (H-S Lee MD), and Department of Internal Medicine (J H Kim MD), Seoul National University College of Medicine, Seoul National University Bundang Hospital, Seongnam, South Korea; Center for Colorectal Cancer (J W Park MD, H S Choi MD, D Y Kim MD, H J Chang MD, S Y Kim MD, D K Sohn MD, D-H Kim MD, J H Oh MD) and Center for Clinical Trials (B H Nam PhD), Research Institute and Hospital, National Cancer Center, Goyang, South Korea; Division of Colorectal Surgery, Department of Surgery, Seoul National University College of Medicine, Seoul National University Hospital, Seoul, South Korea (S-Y Jeong MD); and Department of Colon and Rectal Surgery (S-B Lim MD) and Department of Oncology (K H Jung MD, Y S Hong MD), University of Ulsan College of Medicine, Asan Medical Center, Seoul, South Korea Correspondence to: Dr Jae Hwan Oh, Center for Colorectal Cancer, Research Institute and Hospital, National Cancer Center, 323 Ilsan-ro, Ilsandong-gu, Goyangsi, Gyeonggi-do 410-769, South Korea [email protected]

637

Articles

For the COREAN trial protocol see http://ncc.re.kr/common/ downloadByFileURL.jsp?path=/ downloadFiles/Protocol179.pdf

preservation are prerequisites for functional and oncological safety. Several studies have shown that laparoscopic surgery has technical benefits, such as a magnified view, over open surgery.8–10 A few randomised trials involving patients with mid or low rectal cancer have shown that laparoscopic surgery does not compromise oncological outcomes compared with open types of surgery,8,10,11 but these trials were based on small samples and did not control for preoperative chemoradiotherapy. There have been no randomised trials demonstrating the safety of laparoscopic surgery after preoperative chemoradiotherapy for mid and low rectal cancer. The German Rectal Cancer Study Group trial12 showed that preoperative chemoradiotherapy improves 5-year locoregional recurrence and sphincter preservation compared with postoperative chemoradiotherapy in patients with stage T3, T4, or node-positive disease. The National Comprehensive Cancer Network recommends that resectable cT3N0 or any cTN1–2 lesions should be initially treated with preoperative chemoradiation.13 Only a single case-matched study has compared short-term outcomes of laparoscopy versus open surgery after preoperative chemoradiotherapy in mid or low rectal cancer.14 The COREAN (Comparison of Open versus laparoscopic surgery for mid and low REctal cancer After Neoadjuvant chemoradiotherapy) randomised controlled trial was designed to assess the safety and efficacy of laparoscopic surgery compared with that of open surgery for mid or low rectal cancer after preoperative chemoradiotherapy. Here, we report the short-term outcomes of this trial.

379 patients with cT3N0-2 mid or low rectal cancers were eligible for preoperative chemoradiotherapy of 1408 total rectal cancers

Methods Patients This randomised, controlled, open-label, parallel group trial compared open versus laparoscopic surgery after preoperative chemoradiotherapy in patients with mid or low rectal cancer. Patients were recruited at three tertiaryreferral hospitals in South Korea: the National Cancer Center, the Seoul National University Hospital, and the Seoul National University Bundang Hospital. The primary endpoint is 3-year disease-free survival (DFS); patients continue to be followed up for this endpoint. Patients with mid or low rectal cancer without distant metastasis after preoperative chemoradiotherapy were eligible for inclusion in the study. All patients had rectal adenocarcinomas that were 9 cm or less from the anal verge, which were clinically diagnosed as cT3N0–2 lesions based on pelvic CT, transanal ultrasonography, and MRI. Patients received a fluoropyrimidine-based chemoradiotherapy regimen preoperatively. A dose of 50·4 Gy was given, which included 45 Gy in 25 fractions to the pelvis and a 5·4 Gy boost in three fractions to the primary tumour over 5·5 weeks. The most common chemotherapy regimens were: two cycles of an intravenous bolus of fluorouracil (400 mg/m² per day) and racemic D,L-leucovorin (20 mg/m² per day) for 3 days in the first and fifth weeks of radiotherapy; or continuous oral administration of capecitabine (825 mg/m² twice daily) during radiotherapy. All patients were recommended to receive adjuvant chemotherapy for 4 months. Exclusion criteria included a synchronous distant metastasis, another malignancy, severe cardiac or pulmonary disease, pregnancy, severe medical disease, and intestinal obstruction or perforation. The study was approved and overseen by the institutional review board of each participating hospital, and all patients gave written informed consent.

Randomisation and masking 39 excluded 19 refused to participate 6 distant metastasis 14 refused surgery

340 patients randomised

170 allocated to open surgery 170 received open surgery

170 allocated to laparoscopic surgery 168 received laparoscopic surgery 2 converted to open

3-month follow-up 0 death 0 missing data except for questionnaires (response rate: 77%)

3-month follow-up 0 death 0 missing data except for questionnaires (response rate: 75%)

170 included in analysis

170 included in analysis

After preoperative chemoradiotherapy, patients were randomly assigned 1:1 to receive either laparoscopic or open surgery using a block permutation approach. Random numbers were computer generated and patients were stratified according to sex and preoperative chemotherapy regimen. Allocation was communicated by telephone by the trial coordinator at the central office at the National Cancer Center.

Procedures

Figure 1: Trial profile

638

Surgery was done 6–8 weeks after completion of preoperative chemoradiotherapy, by seven surgeons from three hospitals who had experience with a median number of 75 (range 28–150) cases of laparoscopic colorectal resection before the trial. These surgeons were qualified through live demonstrations held by the Korean Laparoscopic Colorectal Surgery Study Group. Each surgeon submitted a videotape of their laparoscopic rectal surgery, which was reviewed by a trial steering committee www.thelancet.com/oncology Vol 11 July 2010

Articles

to assess the surgeon’s oncological technique, including the level of inferior mesenteric vessels, TME quality, and preservation of autonomic nerves. All patients underwent TME and autonomic-nerve preservation, and the extent of resection was the same for the open and laparoscopic methods. Access in laparoscopic surgery was obtained using five trocars. The inferior mesenteric artery was ligated close to its origin with clips, using the medial approach. For tension-free anastomosis, full splenic flexure mobilisation was done in case of a lack of redundancy of the sigmoid colon as an intraoperative finding.15 The surgeon performed fine dissection with monopolar cautery or harmonic scalpel into the presacral space, while keeping the proper plane of dissection between the fascia propria of the rectum and the presacral fascia. Mobilisation of the mesorectum was done very cautiously, to avoid any damage to the underlying hypogastric nerve plexus. Once the rectum was completely mobilised, and after an adequate distal resection margin was guaranteed by digital rectal examination, the distal lumen of the tumour was clamped and rectal washout was done with a 5% povidone iodine solution. One or two endoscopic linear stapling devices were introduced through the right lower port and the rectum was transected. Surgical specimens were subsequently removed via a 4–6-cm gridiron incision in the left lower quadrant, under a wound protector sleeve. Bowel anastomoses were performed intracorporally by double staple technique or by transanal suture. Conversion to an open procedure was defined as an abdominal incision larger than necessary for specimen retrieval.3 A protective loop ileostomy was recommended for all patients, and intestinal continuity was re-established after completion of postoperative adjuvant therapy. Diet was resumed as soon as the first flatus had been passed. Patient-controlled analgesia was given on demand and calculated as the amount of morphine used. Patients were discharged if they considered themselves sufficiently recovered, with tolerable diet for 24 h, analgesic-free, safe ambulation, and afebrile status without major complications. Three pathologists, masked to allocation, examined the resected specimen according to an established technique, which focuses on the extent of circumferential resection margin (CRM), macroscopic quality of the TME specimen, and number of harvested lymph nodes. The CRM was considered positive when the distance from the tumour to the mesorectal fascia was 1 mm or less. The macroscopic quality of the specimen was graded as described by Nagtegaal and colleagues.16 After surgery, tumour stage was assessed according to the TNM classification system.17 Preoperative chemoradiotherapy response was classified using the tumour regression grade (TRG) scale proposed by Dworak and co-workers.18 Postoperative pain was measured using the present pain intensity (PPI) index and a visual analogue scale (VAS).19 Quality of life (QoL) was assessed preoperatively and 3 months after proctectomy or ileostomy takedown in patients who www.thelancet.com/oncology Vol 11 July 2010

received ileostomy, using the validated Korean version of the European Organisation for Research and Treatment of Cancer (EORTC) QLQ-C30 questionnaire (version 3.0)20 and the colorectal cancer module QLQ-CR38.21 Open (n=170) Age (years)

Laparoscopic (n=170)

59·1 (9·9)

57·8 (11·1)

110 (64·7%)

110 (64·7%)

60 (35·3%)

60 (35·3%)

24·1 (3·2)

24·1 (3·2)

I

65 (38·2%)

69 (40·6%)

II

98 (57·6%)

96 (56·4%)

III

7 (4·1%)

5 (2·9%)

Sex Male Female Body-mass index (kg/m2) ASA grade

Preoperative CEA (ng/mL)

2·91 (0·27)

2·75 (0·24)

Tumour distance from AV (cm)

5·3 (2·5)

5·6 (2·3)

0

105 (61·8%)

93 (54·7%)

1

44 (25·9%)

51 (30·0%)

≥2

21 (12·4%)

26 (15·3%)

Previous abdominal surgery

Clinical N classification cN (–)

52 (30·6%)

59 (34·7%)

cN (+)

118 (69·4%)

111 (65·3%)

116 (68·2%)

116 (68·2%)

54 (31·8%)

54 (31·8%)

C

23 (13·5%)

19 (11·2%)

IC

1 (0·6%)

3 (1·8%)

CIC

13 (7·7%)

11 (6·5%)

UFT

17 (10·0%)

21 (12·4%)

Preoperative chemotherapy Bolus FU/LV Oral fluoropyrimidines±others

Data are n (%) or mean (SD). ASA=American Society of Anesthesiologists. CEA=carcinoembryonic antigen serum level. AV=anal verge. FU=fluorouracil. LV=leucovorin. C=capecitabine. IC=capecitabine and irinotecan. CIC=cetuximab, irinotecan, and capecitabine. UFT=oral uracil and tegafur.

Table 1: Baseline characteristics

Open (n=170)

Laparoscopic (n=170)

p value <0·0001*

Surgery time (min), mean (SD)

197·0 (62·9)

244·9 (75·4)

EBL (mL)

217·5 (150·0–400·0)

200·0 (100·0–300·0)

0·006†

Blood transfusion Yes

1 (0·6%)

No

169 (99·4%)

Length of incision (cm)

20·0 (18·0–23·0)

0 170 (100%) 5·0 (4·5–6·0)

1·000‡ ·· <0·0001†

Procedures Abdominoperineal resection

24 (14·1%)

19 (11·2%)

Low anterior resection with CAA

33 (19·4%)

33 (19·4%)

··

113 (66·5%)

118 (69·4%)

··

Yes

129 (88·4%)

138 (91·4%)

No

17 (11·6%)

13 (8·6%)

Low anterior resection with DS

0·708§

Diverting ileostomy¶ 0·386§ ··

Data are n (%) or median (IQR) unless otherwise stated. EBL=estimated blood loss. CAA=coloanal anastomosis. DS=double stapling technique. *Student’s t test. †Wilcoxon rank-sum test. ‡Fisher’s exact test. §χ2 test. ¶146 patients underwent sphincter preservation procedures in open group, 151 patients in laparoscopic group.

Table 2: Surgery data

639

Articles

Open (n=170) Tumour size (cm) Number of harvested lymph nodes

Laparoscopic (n=170)

p value

2·4 (1·8–3·2)

2·2 (1·3–3·2)

0·252*

18 (13·0–24·0)

17 (12·0–22·0)

0·085* 0·228†

Tumour differentiation Well differentiated

17 (10·0%)

30 (17·7%)

146 (85·9%)

134 (78·8%)

··

Poorly differentiated

4 (2·4%)

2 (1·2%)

··

SRC/mucinous

2 (1·2%)

3 (1·8%)

··

Unknown

1 (0·6%)

1 (0·6%)

··

Moderately differentiated

Tumour Regression Grade scale 1

35 (20·6%)

25 (14·7%)

2

89 (52·4%)

74 (43·5%)

0·027‡ ··

3

24 (14·1%)

31 (18·2%)

··

4

22 (12·9%)

40 (23·5%)

··

ypT0

24 (14·1%)

40 (23·5%)

ypTis

1 (0·6%)

5 (2·9%)

··

ypT1

6 (3·5%)

9 (5·3%)

··

ypT2

40 (23·5%)

41 (24·1%)

··

ypT3

96 (56·5%)

73 (42·9%)

··

ypT4

3 (1·8%)

2 (1·2%)

··

ypT classification 0·054†

ypN classification ypN0

113 (66·5%)

135 (79·4%)

ypN1

43 (25·3%)

18 (10·6%)

··

ypN2

14 (8·2%)

17 (10·0%)

··

Proximal resection margin (cm)

0·002‡

13·0 (10·0–18·1)

13·0 (9·3–17·0)

0·442*

Distal resection margin (cm)

2·0 (1·0–3·5)

2·0 (1·0–3·5)

0·543*

Radial resection margin (cm)

0·8 (0·4–1·2)

0·9 (0·5–1·3)

0·307*

Positive (≤1 mm)

7 (4·1%)

5 (2·9%)

0·770†

Negative (>1 mm)

163 (95·9%)

165 (97·1%)

CRM ··

Macroscopic quality of TME specimen 127 (74·7%)

123 (72·4%)

Nearly complete

Complete

23 (13·5%)

33 (19·4%)

0·414† ··

Incomplete

11 (6·5%)

8 (4·7%)

··

Unknown

9 (5·3%)

6 (3·5%)

··

Data are n (%) or median (IQR). SRC=signet ring cell. yp=pathological stage classified after pretreatment. CRM=circumferential resection margin. TME=total mesorectal excision. *Wilcoxon rank-sum test. †Fisher’s exact test. ‡χ2 test.

Table 3: Pathological characteristics of tumours

Statistical analysis The sample size for this study was based on a noninferiority design. The expected 3-year DFS for the open surgery group was 75%. Allowing a difference of 15% as the non-inferiority margin, 2·5% type 1 error, 85% power, and a follow-up loss of 10%, a total of 340 patients were required. All comparisons between groups were based on the intention-to-treat population (all randomised patients were analysed according to their assigned treatment group). Clinical and pathological variables were analysed with the χ² test (Fisher’s exact test) and Student’s t test or the Wilcoxon rank–sum test, depending on the distribution of the variables. Sensitivity analyses were done to assess 640

whether any systematic differences existed between patients who responded and those who did not, with respect to clinical outcomes. Because these sensitivity analyses did not show any significant differences, the outcome analyses are presented in unadjusted form. Pain scores and QoL questionnaires were analysed using the analysis of covariance (ANCOVA) method with repeated measures. For the EORTC QLQ-C30 and QLQ-CR38, the global health status, functioning, and symptom subscale scores were reported on scales from 0–100. On the function scales, higher scores indicated better function, and on the symptom scales, higher scores indicated more severe symptoms. Statistical analysis of QoL outcomes evaluated differences between the two groups with respect to changes from preoperative scores (baseline) to those at 3-month follow-up. We adjusted for baseline values to allow for possible differences at baseline. Statistical analyses were done with STATA version 10.0. All statistical tests were two-sided. A p value of less than 0·05 was considered significant. This study is registered with ClinicalTrials.gov, number NCT00470951.

Role of the funding source The study sponsor had no involvement in trial design, collection, analysis, or interpretation of data, or the writing of the report. The corresponding author had full access to all data and had final responsibility for the decision to submit the report for publication.

Results Between April 4, 2006, and Aug 26, 2009, 340 patients were randomly assigned to receive laparoscopic or open surgery (figure 1). None of the patients had a metastatic lesion detected before or during surgery. Two patients in the laparoscopic group were converted to open surgery because of a difficult dissection in a narrow pelvis and intraoperative bleeding; for the purposes of the analyses presented here, these patients were included in the laparoscopic surgery group. There were no deaths during the first 3 months after surgery. The open and laparoscopic groups were balanced in terms of their baseline characteristics, clinical staging before preoperative chemoradiotherapy, and chemotherapy regimens (table 1). Surgery time was longer in the laparoscopic group than in the open group (table 2). Blood transfusion requirements were similar between the two groups, although the estimated blood loss was slightly less in the laparoscopic group (table 2). The incision length used for specimen retrieval was smaller in the laparoscopic group than in the open group (table 2). Similar operative procedures were done in both groups (table 2). Tumour size, the number of harvested lymph nodes, and tumour differentiation were similar in both groups (table 3). TRG and the pathological nodal stage were different, although the depth of invasion did not differ between the two groups (table 3). The proximal, distal, and radial resection margins were similar www.thelancet.com/oncology Vol 11 July 2010

Articles

between the two groups, as was CRM positivity (table 3). Among patients who underwent sphincter preservation procedures, CRM positivity was similar between groups (3·4% [five of 146] in the open group vs 2·7% [ four of 151] in the laparoscopic group, p=0·750). Among patients who underwent abdominoperineal resection, CRM positivity was similar between groups (8·3% [two of 24] in the open group vs 5·3% [one of 19] in the laparoscopic group, p=1·000). Macroscopic quality of the TME specimen was similar between the two groups (table 3). Time required to pass the first flatus, to resume a normal diet, and to the first defecation were shorter after laparoscopic surgery than open surgery (table 4). The length of hospital stay was longer in the open group than in the laparoscopic group, but the difference was not significant (table 4). The two groups had similar overall perioperative complication rates (table 4). Wound discharge, including seroma and superficial surgical site infection, were more common in the open group than in the laparoscopic group (table 4). Acute voiding difficulties requiring urinary catheter were more frequent after laparoscopic surgery than open surgery (table 4), but all patients recovered within 1 month. Six patients underwent reoperation, for ileostomy-related complications in two patients and ileus in four patients. Pain scores according to PPI and VAS for the 3 postoperative days were lower after laparoscopic surgery than after open surgery, whereas preoperative pain scores did not differ between the two groups (figure 2). The total amount of morphine used via a patient-controlled analgesia pump was less in the laparoscopic group than in the open group (median 107·2 [80·0–150·0] mg vs 156·9 [117·0–185·2] mg, p<0·0001). Compliance rates for the EORTC QLQ-C30 and EORTC QLQ-CR38 questionnaires were similar between groups (168 of 170 [98·8%] patients preoperatively and 131 of 170 [77·1%] at 3 months postoperatively in the open group; 164 of 170 [96·5%] preoperatively and 127 of 170 [74·7%] at 3 months postoperatively in the laparoscopic group). In the open group, 99 of 118 (83·9%) patients responded to questions regarding defecation problems in the QLQCR38 questionnaire, as did 103 of 122 (84·4%) patients in the laparoscopic group, after 57 patients who did not have an ileostomy takedown (28 in the open group and 29 in the laparoscopic group) and 43 patients who underwent abdominoperineal resection (24 patients in the open group and 19 patients in the laparoscopic group) were excluded. Response rates for sexual enjoyment or problems were low (12·4% [21 of 170] in the open group vs 10·6% [18 of 170] in the laparoscopic group, p=1·000) because of physical status, old age, or low sexual activity: 76 patients (44·7%) in the open group and 81 patients (47·6%) in the laparoscopic group were currently sexually active. Preoperative QLQ-C30 scores were similar between the two groups for most scales, except for sleep disturbance (mean 23·0 [SD 32·2] open group vs 13·7 [25·3] laparoscopic group, p=0·004; figure 3). In the repeated measures ANCOVA adjusted for baseline values, the www.thelancet.com/oncology Vol 11 July 2010

Open (n=170)

Laparoscopic (n=170)

p value

Time to pass first flatus (h)

60·0 (43·0–73·0)

38·5 (23·0–53·0)

Time to resume liquid diet (h)

68 (50·0–89·0)

48 (42·0–69·0)

<0·0001*

Time to resume normal diet (h)

93·0 (86·0–121·0)

85·0 (66·0–95·0)

<0·0001*

Time to first defecation (h)

123 (94·0–156·0)

96·5 (70·0–125·0)

<0·0001*

Postoperative hospital stay (days) Perioperative complications

9 (8·0–12·0) 40 (23·5%)

8 (7·0–12·0)

0·603† 0·499‡

0

2 (1·2%)

Pelvic abscess

1 (0·6%)

0

22 (12·9%)

Wound discharge

11 (6·5%)

0·056*

36 (21·2%)

Anastomotic leakage Ileus§

<0·0001*

1·000‡

17 (10·0%)

0·395†

2 (1·2%)

0·020‡

Chylous ascites

1 (0·6%)

0

Acute voiding difficulty

7 (4·1%)

17 (10·0%)

0·034†

Bleeding

3 (1·8%)

1 (0·6%)

0·623‡

Stoma-related complication¶

0

1 (0·6%)

1·000‡

Bacteraemia

1 (0·6%)

1 (0·6%)

1·000‡

Nerve injury||

1 (0·6%)

1 (0·6%)

1·000‡

Yes

3 (1·8%)

3 (1·8%)

··

No

167 (98·2%)

167 (98·2%)

··

Reoperation

1·000‡

1·000‡

Data are n (%) or median (IQR). *Wilcoxon rank-sum test. †χ2 test. ‡Fisher’s exact test.§Requiring a nasogastric drainage before discharge. ¶Mucocutaneous separation. ||Brachial plexus injury.

Table 4: Postoperative recovery and complications

laparoscopic group showed better physical functioning (0·501 in 122 patients vs – 4·970 in 128 patients, p=0·0073) and less fatigue (–5·659 in 122 patients vs 0·098 in 129 patients, p=0·0206) than the open group 3 months after surgery. No differences were seen preoperatively between the two groups on the QLQ-CR38 scale scores, except for male sexual problems (23·3 [SD 29·9] in the open group vs 11·7 [21·5] in the laparoscopic group, p=0·010; figure 4). In both groups, sexual function was better 3 months after surgery than at baseline (open group 92·5 [16·2] vs 83·6 [24·5], p<0·0001; laparoscopic group 90·9 [16·3] vs 81·2 [24·0], p<0·0001). Male sexual problems were higher 3 months after surgery than at baseline in both groups (open group 45·2 [38·8] vs 23·3 [29·9], p=0·0188; laparoscopic group 46·6 [40·2] vs 11·7 [21·5], p<0·0001); there was no difference between the two groups (p=0·2855). In the repeated measures ANCOVA adjusted for baseline values, the laparoscopic group showed fewer micturition problems (–2·583 in 122 patients vs 4·725 in 129 patients, p=0·0002), gastrointestinal problems (–0·400 in 122 patients vs 4·331 in 129 patients, p=0·0102), and defecation problems (0·535 in 103 patients vs 5·327 in 99 patients, p=0·0184) than the open group 3 months after surgery. Sensitivity analyses confirmed that missing data had no effect on the interpretation of results (data not shown).

Discussion To our knowledge, this is the first randomised trial comparing open and laparoscopic surgery after preoperative chemoradiotherapy for mid or low rectal cancer. The study 641

Articles

A

VAS †

6

† †

Score

4

2

Open surgery Laparoscopic surgery 0

B

PPI

4 * 3

Score

† 2

1

0 Preop

POD 1

POD 2

POD 3

Figure 2: Actual pain scores with visual analogue scale (VAS; A) and present pain intensity (PPI; B) Higher scores indicate severe pain. Pairs of dots show scores for open surgery (blue) and laparoscopic surgery (red). Dots are mean values and whiskers indicate 95% CI. Preop=preoperative. POD=postoperative days. *p<0·05 in repeated measures ANCOVA adjusted for baseline values. †p <0·01 in ANCOVA.

shows that laparoscopic surgery for rectal cancer after preoperative chemoradiotherapy does not jeopardise shortterm surgical outcomes such as CRM positivity, macroscopic quality of the TME specimen, or the number of harvested lymph nodes, which are associated with longterm oncological outcomes, and does not increase morbidity or mortality compared with open surgery. This trial shows the safety of laparoscopic surgery when done by skilled surgeons, despite the laparoscopic procedure being technically more difficult for rectal cancer than for other cancers.22 In the present study, laparoscopic surgery showed short-term benefits over open surgery in terms of earlier bowel movement, less pain, and less blood loss, although not in terms of the length of hospital stay. QoL was better in the laparoscopy group than in the open group 3 months after surgery, although an improvement in QoL for only 1 month has been reported in previous studies.23,24 The short-term advantages of laparoscopy are important to justify the time and expense of the procedure, provided the long-term oncological outcomes are the same or better. This trial found a clinically significant difference in the physical functioning component of the EORTC QLQ-C30 3 months after 642

surgery in favour of a laparoscopic approach. This better physical functioning might have resulted from less fatigue and fewer gastrointestinal and defecation problems after laparoscopic surgery, as shown in a previous study,25 and is consistent with the short-term benefits, such as earlier bowel movement, less pain, and a smaller incision. The current study shows fewer micturition problems 3 months after laparoscopic surgery, whereas analysis of patients in the MRC CLASICC trial noted similar outcomes in bladder function with laparoscopic and open resection.26 We suspect that fewer micturition problems might be a result of better preservation of the autonomic nerve and less traumatic surgery, attributable to the magnified view of laparoscopic surgery.8–10 The COST (Clinical Outcomes of Surgical Therapy) and CLASICC trials showed only slight short-term improvement in QoL when laparoscopic surgery was done by a surgeon with experience of at least 20 laparoscopic-assisted resections,2,6,23,24 which has been criticised as probably insufficient experience for participation in a randomised trial.3,6 Our trial was done by seven surgeons at three tertiary-referral hospitals, in which more than 200 rectal cancers are treated annually. The experience levels of these surgeons were in the top third of their peers, from a professional survey of surgeries undertaken (unpublished data). However, acute voiding difficulties requiring catheter insertion were more common in the laparoscopic group. Although the reasons for this difficulty are unknown, the transient neuropraxia can be explained by the broader retractors allowed by open surgery, which produce less traction injury to nerves than the small, grasping instruments used for retraction in laparoscopy. 3 months after surgery, sexual function had improved from baseline values in both groups, as was shown in the COLOR (COlon cancer Laparoscopic or Open Resection) trial,27 where both groups improved in some aspects of QoL at 3 months compared with preoperative scores.23 This improvement might be associated with a responseshift tendency—a previous study has shown that patients scored higher on QoL than did the general population.25 High levels of self-reported QoL at different points in time can have different meanings.28 In the present trial, pathological analysis showed low CRM positivity and good macroscopic quality of TME specimens, which are the most important prognostic factors in rectal cancer surgery. Additionally, the conversion rate was low compared with previous studies.2,3,6 There are a number of possible reasons for these favourable oncological short-term outcomes. First, recent trials have shown significant improvements in surgical techniques,5 surgical instruments,27 perioperative patient management, and oncological safety compared with the pioneering clinical trials.1–6 Second, this study might have reduced the oncological risk caused by local invasion because we excluded cT4 lesions after evaluation by MRI and other imaging methods. The laparoscopic approach might increase oncological risk during rectal www.thelancet.com/oncology Vol 11 July 2010

Articles

A

Score

90

80

80

60

70

70

50

60

60

0

0

0

E

F

Cognitive functioning

100

100

80

90

90

70

80

80

60

70

70

50

60

60

0

0

0

H

Fatigue

I

Nausea and vomiting

40

40

30

30

20

20

20

10

10

10

0

0

0

30 * Score

90

90

G

J

Score

†

70

Emotional functioning

K

Dyspnoea

L

Constipation

40

40

40

30

30

30

20

20

20

10

10

10

0

0

0

M

N

Sleep disturbance

O

Appetite loss

40

40

40

30

30

30

20

20

20

10

10

10

0

0 Preop

3 months

Role functioning

100

90

40

Score

C

Physical functioning

100

Open surgery Laparoscopic surgery

80

D

Score

B

Global quality of life

Social functioning

Pain

Diarrhoea

Financial difficulties

0 Preop

3 months

Preop

3 months

Figure 3: Actual EORTC QLQ-C30 scores On the global quality of life (A) and function scales (B–F), higher scores indicate better function. On the symptom scales (G–O), higher scores indicate more severe symptoms. Pairs of dots show scores for open surgery (blue) and laparoscopic surgery (red). Dots show mean values and whiskers indicate 95% CI. EORTC=European Organisation for Research and Treatment of Cancer. Preop=preoperative. *p<0·05 in repeated measures ANCOVA adjusted for baseline values. †p <0·01 in ANCOVA.

cancer surgery,6 whereas there is no evidence that laparoscopic surgery is better than open surgery for oncological outcomes. Therefore, we suggest that for www.thelancet.com/oncology Vol 11 July 2010

oncological safety, cT4 lesions should not be indicated for laparoscopic surgery. Third, body-mass index was low in this series compared with previous trials;6,27 the same 643

Articles

A

Score

Body image

B

100

90

90

90

80

80

80

70

70

70

0

0

0

D

Open surgery Laparoscopic surgery

Future perspective

E

100

Score

80

0 Symptoms in GI tract

†

20

10

10

0

0

I

Female sexual problems

40

40

40

30

30

30

20

20

10

10

10

0

0

0

20

*

Chemotherapy side-effects

30

20

H

Sexual enjoyment

40

30

60

G

F

Micturition problems

40

40

Score

C

Sexual function

100

100

Male sexual problems

Preop

J

Defecation problems

K

Weight loss

40

40 *

30 Score

3 months

30

20

20

10

10 0

0 Preop

3 months

Preop

3 months

Figure 4: Actual EORTC QLQ-CR38 scores On the function scales (A–D), higher scores indicated better function. On the symptom scales (E–K), higher scores indicated more severe symptoms. Pairs of dots show scores for open surgery (blue) and laparoscopic surgery (red). Dots show mean values and whiskers indicate 95% CI. EORTC=European Organisation for Research and Treatment of Cancer. Preop=preoperative. *p<0·05 in repeated measures ANCOVA adjusted for baseline values. †p<0·01 in ANCOVA.

results might not be possible in larger patients. In this study, laparoscopic surgery did not increase anastomotic leakage compared with open surgery. We believe that a diverting ileostomy can reduce anastomotic leakage, as noted in a recent meta-analysis of the role of defunctioning stoma in surgery for low rectal cancer.29 The abdominoperineal resection rates for mid and low rectal cancer were low in our study compared with other studies,30 although whether abdominoperineal resection rate is a marker of surgical quality is controversial.30 The 644

long-term anorectal function and QoL of patients with sphincter preservation will be assessed in a future study. Our study had some limitations. First, we did not collect data on the immediate QoL outcomes at 1 week or 1 month after surgery, when QoL scores are lower and differences between the two groups might have been more pronounced. Representative trials have shown that most QoL scales show improvement by 2–3 months after surgery.2,6,23 Second, there were differences in the distribution of TRG and after-pretreatment pathological N classifications www.thelancet.com/oncology Vol 11 July 2010

Articles

between the two groups, which should be adjusted for long-term oncological analysis. Third, the 15% noninferiority margin for sample size is large, but has been used in previous studies1 and was based on practical constraints because of the number of patients with cT3N0-2 mid or low rectal cancer that can accrue in three hospitals. The COREAN trial shows that laparoscopic surgery is feasible and does not increase short-term oncological risks, which are predicted by CRM positivity and macroscopic quality of TME specimens, after preoperative chemoradiotherapy for mid or low rectal cancer. The results of this trial also suggest that laparoscopic surgery results in a better quality of life for up to 3 months after surgery. The outcomes in this trial were achieved by seven highly skilled laparoscopic specialists, showing that excellent results can be achieved in expert hands. These results do not, however, prove that laparoscopic rectal resection is the preferred surgical method for routine use by all surgeons. Contributors S-YJ, BHN, S-BL, DYK, KHJ, HSC, and S-BK were responsible for the conception and design of this study. S-BK, JWP, HSC, D-WK, T-GL, DYK, J-SK, HJC, H-SL, SYK, YSH, JHK, DKS, and D-HK collected and assembled data. S-BK, JWP, BHN, JHO, and S-YJ analysed and interpreted the data. S-BK, BHN, and JHO wrote the report, which was approved by all authors. Conflicts of interest The authors declared no conflicts of interest. Acknowledgments The COREAN trial was supported by a grant from the National Cancer Center (grant 0910200). References 1 Lacy AM, García-Valdecasas JC, Delgado S, et al. Laparoscopy-assisted colectomy versus open colectomy for treatment of non-metastatic colon cancer: a randomised trial. Lancet 2002; 359: 2224–29. 2 Clinical Outcomes of Surgical Therapy (COST) Study Group. A comparison of laparoscopically assisted and open colectomy for colon cancer. N Engl J Med 2004; 350: 2050–59. 3 Leung KL, Kwok SP, Lam SC, et al. Laparoscopic resection of rectosigmoid carcinoma: prospective randomised trial. Lancet 2004; 363: 1187–92. 4 Jayne DG, Guillou PJ, Thorpe H, et al. UK MRC CLASICC Trial Group. Randomized trial of laparoscopic-assisted resection of colorectal carcinoma: 3-year results of the UK MRC CLASICC Trial Group. J Clin Oncol 2007; 25: 3061–68. 5 The COlon Cancer Laparoscopic or Open Resection Study Group. Survival after laparoscopic surgery versus open surgery for colon cancer: long-term outcome of a randomised clinical trial. Lancet Oncol 2009; 10: 44–52. 6 Guillou PJ, Quirke P, Thorpe H, et al. Short-term endpoints of conventional versus laparoscopic-assisted surgery in patients with colorectal cancer (MRC CLASICC trial): multicentre, randomised controlled trial. Lancet 2005; 365: 1718–26. 7 D’Anniblae A, Morpurgo E, Fiscon V, et al. Robotic and laparoscopic surgery for treatment of colorectal disease. Dis Colon Rectum 2004; 47: 2162–68. 8 Zhou ZG, Hu M, Li Y, et al. Laparoscopic versus open total mesorectal excision with anal sphincter preservation for low rectal cancer. Surg Endosc 2004; 18: 1211–15. 9 Gouvas N, Tsiaoussis J, Pechlivanides G, et al. Quality of surgery for rectal carcinoma: comparison between open and laparoscopic approaches. Am J Surg 2009; 198: 702–08. 10 Lujan J, Valero G, Hernandez Q, Sanchez A, Frutos MD, Parrilla P. Randomized clinical trial comparing laparoscopic and open surgery in patients with rectal cancer. Br J Surg 2009; 96: 982–89.

www.thelancet.com/oncology Vol 11 July 2010

11

12

13

14

15

16

17 18

19

20 21

22

23

24

25

26

27

28

29

30

Ng SS, Leung KL, Lee JF, et al. Laparoscopic-assisted versus open abdominoperineal resection for low rectal cancer: a prospective randomized trial. Ann Surg Oncol 2008; 15: 2418–25. Sauer R, Becker H, Hohenberger W, et al. Preoperative versus postoperative chemoradiotherapy for rectal cancer. N Engl J Med 2004; 351: 1731–40. Engstrom PF, Arnoletti JP, Benson AB, et al. NCCN Clinical Practice Guidelines in Oncology: rectal cancer. J Natl Compr Canc Netw 2009; 7: 838–81. Denoya P, Wang H, Sands D, Nogueras J, Weiss E, Wexner SD. Short-term outcomes of laparoscopic total mesorectal excision following neoadjuvant chemoradiotherapy. Surg Endosc 2010; 24: 933–38. Park JS, Kang SB, Kim DW, Lee KH, Kim YH. Laparoscopic versus open resection without splenic flexure mobilization for the treatment of rectum and sigmoid cancer: a study from a single institution that selectively used splenic flexure mobilization. Surg Laparosc Endosc Percutan Tech 2009; 19: 62–68. Nagtegaal ID, van de Velde CJ, van der Worp E, Kapiteijn E, Quirke P, van Krieken JH. Macroscopic evaluation of rectal cancer resection specimen: clinical significance of the pathologist in quality control. J Clin Oncol 2002; 20: 1729–34. Greene FL, Rage DL, Fleming ID, et al. AJCC Cancer Staging Manual, 6th edn. New York: Springer-Verlag, 2002. Dworak O, Keilholz L, Hoffmann A. Pathological features of rectal cancer after preoperative radiochemotherapy. Int J Colorectal Dis 1997; 12: 19–23. Huskisson EC. Visual analogue scale. In: Melzack R, ed. Pain measurement and assessment, 1st edn. New York; Raven Press, 1983: 33–37. Yun YH, Park YS, Lee ES, et al. Validation of the Korean version of the EORTC QLQ-C30. Qual Life Res 2004; 13: 863–68. Sprangers MA, te Velde A, Aaronson NK. The construction and testing of the EORTC colorectal cancer-specific quality of life questionnaire module (QLQ-CR38). European Organization for Research and Treatment of Cancer Study Group on Quality of Life. Eur J Cancer 1999; 35: 238–47. Targarona EM, Balague C, Pernas JC, et al. Can we predict immediate outcome after laparoscopic rectal surgery? Multivariate analysis of clinical, anatomic, and pathologic features after 3-dimensional reconstruction of the pelvic anatomy. Ann Surg 2008; 247: 642–49. Janson M, Lindholm E, Anderberg B, Haglind E. Randomized trial of health-related quality of life after open and laparoscopic surgery for colon cancer. Surg Endosc 2007; 21: 747–53. Weeks JC, Nelson H, Gelber S, Sargent D, Schroeder G, for the Clinical Outcomes of Surgical Therapy (COST) Study Group. Shortterm quality-of-life outcomes following laparoscopic-assisted colectomy vs open colectomy for colon cancer: a randomized trial. JAMA 2002; 287: 321–28. Pucciarelli S, Del Bianco P, Efficace F, et al. Health-related quality of life, faecal continence and bowel function in rectal cancer patients after chemoradiotherapy followed by radical surgery. Support Care Cancer 2010; 18: 601–08. Jayne DG, Brown JM, Thorpe H, Walker J, Quirke P, Guillou PJ. Bladder and sexual function following resection for rectal cancer in a randomized clinical trial of laparoscopic versus open technique. Br J Surg 2005; 92: 1124–32. The COlon cancer Laparoscopic or Open Resection Study Group. Laparoscopic surgery versus open surgery for colon cancer: shortterm outcomes of a randomised trial. Lancet Oncol 2005; 6: 477–84. Hagedoorn M, Sneeuw KC, Aaronson NK. Changes in physical functioning and quality of life in patients with cancer: response shift and relative evaluation of one’s condition. J Clin Epidemiol 2002; 55: 176–83. Hüser N, Michalski CW, Erkan M, et al. Systematic review and meta-analysis of the role of defunctioning stoma in low rectal cancer surgery. Ann Surg 2008; 248: 52–60. Tilney HS, Heriot AG, Purkayastha S, et al. A national perspective on the decline of abdominoperineal resection for rectal cancer. Ann Surg 2008; 247: 77–84.

645