Laparoscopic surgery for rectal carcinoma — slash back?
Thomas Rösch, Hamburg and Karl-Hermann Fuchs, Frankfurt
NEJM 2015; 372: 1324-32
|A Randomized Trial of Laparoscopic versus Open Surgery for Rectal Cancer|
|H. Jaap Bonjer, M.D., Ph.D., Charlotte L. Deijen, M.D., Gabor A. Abis, M.D., Miguel A. Cuesta, M.D., Ph.D., Martijn H.G.M. van der Pas, M.D., Elly S.M. de Lange-de Klerk, M.D., Ph.D., Antonio M. Lacy, M.D., Ph.D., Willem A. Bemelman, M.D., Ph.D., John Andersson, M.D., Eva Angenete, M.D., Ph.D., Jacob Rosenberg, M.D., Ph.D., Alois Fuerst, M.D., Ph.D., and Eva Haglind, M.D., Ph.D., for the COLOR II Study Group*, Amsterdam, Barcelona, Gothenburg, Kopenhagen, Regensburg|
Laparoscopic resection of colorectal cancer is widely used. However, robust evidence to conclude that laparoscopic surgery and open surgery have similar outcomes in rectal cancer is lacking. A trial was designed to compare 3-year rates of cancer recurrence in the pelvic or perineal area (locoregional recurrence) and survival after laparoscopic and open resection of rectal cancer.
In this international trial conducted in 30 hospitals, we randomly assigned patients with a solitary adenocarcinoma of the rectum within 15 cm of the anal verge, not invading adjacent tissues, and without distant metastases to undergo either laparoscopic or open surgery in a 2:1 ratio. The primary end point was locoregional recurrence 3 years after the index surgery. Secondary end points included disease-free and overall survival.
A total of 1044 patients were included (699 in the laparoscopic-surgery group and 345 in the open-surgery group). At 3 years, the locoregional recurrence rate was 5.0% in the two groups (difference, 0 percentage points; 90% confidence interval [CI], −2.6 to 2.6). Disease-free survival rates were 74.8% in the laparoscopic-surgery group and 70.8% in the open-surgery group (difference, 4.0 percentage points; 95% CI, −1.9 to 9.9). Overall survival rates were 86.7% in the laparoscopic-surgery group and 83.6% in the open-surgery group (difference, 3.1 percentage points; 95% CI, −1.6 to 7.8).
Laparoscopic surgery in patients with rectal cancer was associated with rates of locoregional recurrence and disease-free and overall survival similar to those for open surgery. (Funded by Ethicon Endo-Surgery Europe and others; COLOR II (ClinicalTrials.gov number, NCT00297791.
|Effect of Laparoscopic-Assisted Resection vs Open Resection of Stage II or III Rectal Cancer on Pathologic Outcomes
The ACOSOG Z6051 Randomized Clinical Trial
|James Fleshman, MD; Megan Branda, MS; Daniel J. Sargent, PhD; Anne Marie Boller, MD; Virgilio George, MD; Maher Abbas, MD;Walter R. Peters Jr, MD; Dipen Maun, MD; George Chang, MD; Alan Herline, MD; Alessandro Fichera, MD; Matthew Mutch, MD; StevenWexner, MD; Mark Whiteford, MD; John Marks, MD; Elisa Birnbaum, MD; David Margolin, MD; David Larson, MD; Peter Marcello, MD; Mitchell Posner, MD; Thomas Read, MD; John Monson, MD; SherryM.Wren, MD; PeterW. T. Pisters, MD; Heidi Nelson, MD|
|Dallas, Rochester, Chicago, Indianapolis, Abu Dhabi, Columbia, Houston,Nashville, St. Louis, Weston, Portland, Wynnewood, New Orleans, Burlington, Paolo Alto|
Evidence about the efficacy of laparoscopic resection of rectal cancer is incomplete, particularly for patients with more advanced-stage disease.
To determine whether laparoscopic resection is noninferior to open resection, as determined by gross pathologic and histologic evaluation of the resected proctectomy specimen.
Design, Setting, and Participants
A multicenter, balanced, noninferiority, randomized trial enrolled patients between October 2008 and September 2013. The trial was conducted by credentialed surgeons from 35 institutions in the United States and Canada. A total of 486 patients with clinical stage II or III rectal cancer within 12 cm of the anal verge were randomized after completion of neoadjuvant therapy to laparoscopic or open resection.
Standard laparoscopic and open approaches were performed by the credentialed surgeons.
Main Outcomes and Measures
The primary outcome assessing efficacy was a composite of circumferential radial margin greater than 1 mm, distal margin without tumor, and completeness of total mesorectal excision. A 6%noninferiority margin was chosen according to clinical relevance estimation.
Two hundred forty patients with laparoscopic resection and 222 with open resection were evaluable for analysis of the 486 enrolled. Successful resection occurred in 81.7%of laparoscopic resection cases (95%CI, 76.8%-86.6%) and 86.9%of open resection cases (95%CI, 82.5%-91.4%) and did not support noninferiority (difference, −5.3%; 1-sided 95%CI, −10.8%to _; P for noninferiority = .41). Patients underwent low anterior resection (76.7%) or abdominoperineal resection (23.3%). Conversion to open resection occurred in 11.3%of patients. Operative time was significantly longer for laparoscopic resection (mean, 266.2 vs 220.6 minutes; mean difference, 45.5 minutes; 95%CI, 27.7-63.4; P < .001). Length of stay (7.3 vs 7.0 days; mean difference, 0.3 days; 95%CI, −0.6 to 1.1), readmission within 30 days (3.3%vs 4.1%; difference, −0.7%; 95%CI, −4.2%to 2.7%), and severe complications (22.5%vs 22.1%; difference, 0.4%; 95%CI, −4.2%to 2.7%) did not differ significantly. Quality of the total mesorectal excision specimen in 462
Conclusions and Relevance
Among patients with stage II or III rectal cancer, the use of laparoscopic resection compared with open resection failed to meet the criterion for noninferiority for pathologic outcomes. Pending clinical oncologic outcomes, the findings do not support the use of laparoscopic resection in these patients.
JAMA 2015; 314: 1356-1363
|Effect of Laparoscopic-Assisted Resection vs Open Resection on Pathological Outcomes in Rectal Cancer
The ALaCaRT Randomized Clinical Trial
|Andrew R. L. Stevenson, MB BS, FRACS; Michael J. Solomon, MB BCh, MSc, FRCSI, FRACS; JohnW. Lumley, MBBS, FRACS; Peter Hewett, MB BS, FRACS; Andrew D. Clouston, MB BS, PhD, FRCPA; Val J. Gebski, MStat; Lucy Davies, MSc; Kate Wilson, BA, MPH;Wendy Hague,MB BS, PhD, MBA; John Simes, BSc (Med), MB BS, SM, FRACP, MD; for the ALaCaRT Investigators|
|Brisbane, Sydney, Adelaide u. a.|
Laparoscopic procedures are generally thought to have better outcomes than open procedures. Because of anatomical constraints, laparoscopic rectal resection may not be better because of limitations in performing an adequate cancer resection.
To determine whether laparoscopic resection is noninferior to open rectal cancer resection for adequacy of cancer clearance.
Design, Setting, and Participants
Randomized, noninferiority, phase 3 trial (Australasian Laparoscopic Cancer of the Rectum; ALaCaRT) conducted between March 2010 and November 2014. Twenty-six accredited surgeons from 24 sites in Australia and New Zealand randomized 475 patients with T1-T3 rectal adenocarcinoma less than 15 cm from the anal verge.
Open laparotomy and rectal resection (n = 237) or laparoscopic rectal resection (n = 238).
Main Outcomes and Measures
The primary end point was a composite of oncological factors indicating an adequate surgical resection, with a noninferiority boundary of Δ = −8%. Successful resection was defined as meeting all the following criteria: (1) complete total mesorectal excision, (2) a clear circumferential margin (_1 mm), and (3) a clear distal resection margin (_1 mm). Pathologists used standardized reporting and were blinded to the method of surgery.
A successful resection was achieved in 194 patients (82%) in the laparoscopic surgery group and 208 patients (89%) in the open surgery group (risk difference of −7.0% [95%CI, −12.4%to _]; P = .38 for noninferiority). The circumferential resection margin was clear in 222 patients (93%) in the laparoscopic surgery group and in 228 patients (97%) in the open surgery group (risk difference of −3.7%[95%CI, −7.6%to 0.1%]; P = .06), the distal margin was clear in 236 patients (99%) in the laparoscopic surgery group and in 234 patients (99%) in the open surgery group (risk difference of −0.4%[95%CI, −1.8%to 1.0%]; P = .67), and total mesorectal excision was complete in 206 patients (87%) in the laparoscopic surgery group and 216 patients (92%) in the open surgery group (risk difference of −5.4%[95%CI, −10.9%to 0.2%]; P = .06). The conversion rate from laparoscopic to open surgery was 9%.
Conclusions and Relevance
Among patients with T1-T3 rectal tumors, noninferiority of laparoscopic surgery compared with open surgery for successful resection was not established. Although the overall quality of surgery was high, these findings do not provide sufficient evidence for the routine use of laparoscopic surgery. Longer follow-up of recurrence and survival is currently being acquired.
What you need to know
Since the early 2000s, meta-analyses based on smaller studies have shown that open and laparoscopic surgery for rectal carcinoma produce similar results. The most recently published meta-analysis to appear in a major journal — as so often, by a research group in China (1) — included six randomized studies with 1033 patients, including one dating from 2003 with 28 patients (sic!) from a local Brazilian journal published by the University of São Paolo (Rev Hosp Clin Fac Med Univ São Paolo). It is to be hoped that the authors of the meta-analysis read the study as a full publication in the original language. The other studies had been published in English-language journals, but were in some cases so small (with totals of 99–204 cases) that it would need a brilliant statistician to identify any differences that were not gigantic. The only reasonable study, the CLASSIC trial (n = 381) (2) found no difference in the circumferential resection margin, but a slightly lower (HR 0.72) 3-year survival rate, although with an identical disease-free 3-year survival. Overall, the results of the “meta-analysis” for the two procedures are equivalent so far as the statistical analysis is concerned, but the hazard ratio figures suggest that the case numbers are nevertheless not sufficient (2). Another meta-analysis by a Chinese group was published in 2014, this time with 16 studies, most of which could already have been included in the 2011 study (3). So caution needs to be observed with meta-analyses.
A large European study including 1044 patients (the COLOR II Study) was published in April this year, showing equivalent oncological results with the two procedures at the 3-year follow-up in relation to locoregional recurrences, disease-free survival, and overall survival. The case numbers (from 30 hospitals over a 4-year period) are thus similar to those in the six studies included in the meta-analysis described above. However, very advanced tumors (including T4) were excluded, and the boundary used was 15 cm ab ano. Randomization was 2 : 1 for surgery; just under 60% and around one-third of the patients received radiotherapy or chemotherapy preoperatively. Depending on the parameters, follow-up data after 3 years were available for 74–89% of the patients. The study was designed to establish equivalence or noninferiority (4). After 3 years, the results were as follows:
|Disease-free survival||74.8%||70.8%||4.0 n.s.|
|Overall survival||86.7%||83.6%||3.1 n.s.|
In both groups, the circumferential resection margin was not tumor-free (defined as a distance of 2 mm) in 10% of cases, and the other data on complete resection were also nearly identical.
Subsequently, two smaller randomized studies from the USA/Canada (n = 489) and from Australia/New Zealand (n = 475) comparing the two procedures have just been published in JAMA. They both report poorer results in the oncological parameters for the laparoscopic variant (no follow-up is available). The North American study (5) from 35 hospitals presented the results for 486 patients who were included, with 462 evaluable patients with rectal carcinomas (up to 12 cm ab ano, no T4) over 5 years — i.e., an average of just under three patients per center and year. This does not mean that inexperienced physicians were involved, but rather that as usual in everyday clinical work, most patients were not included for reasons of time or for other organizational reasons, or did not provide consent (the relevant figures are not given). In addition, the numbers of patients included varied at different centers, and it is also not clear whether all of the hospitals started the study at the same time point. This patient selection, which has to be accepted with almost all multicenter studies, can however in principle mean that the representativeness of the data for all of the relevant patients is at least debatable. The study in Australia and New Zealand (6) used a similar approach, with 475 patients being included at 24 centers over four and a half years (473 were evaluated). T4 tumors were excluded here as well, and as in Europe the rectum was 15 cm long. Neither study offers any follow-up as yet, so that the main outcome parameters were histopathological, consisting of the same composite score based on the distal and circumferential resection margin (> 1 mm), and the quality of the entire mesorectal excision with blinded pathology. Complete resection was noted when all of the end points were reached, either completely or almost completely in North America (i.e., R0 with a margin of less than 1 mm) or completely in Australia. The results were then as follows:
|USA / Canada||92.1%||95.1%|
|Australia / New Zealand||82%||89%|
However, due to the different definitions used, it is worthwhile — at least for a surgeon — to look more closely at the three individual parameters mentioned above. If the 1-mm boundary in both studies is strictly observed, the complete resections achieved were as follows:
|Distal resection margin||Laparoscopic||Open|
It can be seen from this that the distal resection margin was not a problem, but the circumferential one certainly was, particularly in the USA (although in both groups). Comparison with the European study, in which only the circumferential resection margin was reported, reveals the differences most clearly.
The contradictory results are not very easy to explain. The NEJM study is larger and uses oncological outcomes instead of surrogate parameters, but follow-up results are lacking. Could it be due to the quality of the surgery? But that was quite well established with minimal numbers and a reviewed video of a procedure before the start of the study; despite that, the European groups perhaps had better expertise. However, the statistics are difficult for laypersons to read; the means for the main outcome parameters (composite score, see above) are within the assumed range of noninferiority in both studies. In the methodology, the studies assumed a range of 6% (North America) to 8% (Australia) for noninferiority. In the composite score, North America reached differences of 3% and Australia 7% — both thus apparently in the normal area, and the P values are also below 0.05. Despite that, the overall range has to be used. In this case, the lower boundary for the actual range of results (lower boundary, 95% confidence interval) in North America was –10.8% (10% was assumed) and in Australia –12.4% (8% was assumed), so that the target was just missed. But when you have to dig deeper into the statistical deliberations in order to understand the results, slight concerns arise. Perhaps both studies were simply not sufficiently powered? In the end, the criticism of the study’s main parameter, which is a surrogate parameter, stands. As the authors themselves mention, the follow-up may show different conclusions — hopefully both studies are sufficiently powered for it. More patience in resisting pressures to publish would therefore be helpful sometimes … From our point of view, the NEJM study will until then continue to be the superior one for evidence, and we would not want to leave unchallenged the warning that laparoscopic surgery should be downgraded. The aim should certainly be to carry out these procedures in specialized centers, so that experience can be better concentrated.
- Huang MJ, Liang JL, Wang H, Kang L, Deng YH, Wang JP Laparoscopic-assisted versus open surgery for rectal cancer: a meta-analysis of randomized controlled trials on oncologic adequacy of resection and long-term oncologic outcomes. Int J Colorectal Dis. 2011 Apr;26(4):415-21
- Jayne DG, Guillou PJ, Thorpe H, Quirke P, Copeland J, Smith AM, Heath RM, Brown JM; 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 Jul 20;25(21):3061-8.
- Zhang FW, Zhou ZY, Wang HL, Zhang JX, Di BS, Huang WH, Yang KH. Laparoscopic versus open surgery for rectal cancer: a systematic review and meta-analysis of randomized controlled trials. Asian Pac J Cancer Prev. 2014;15(22):9985-96.
- Bonjer HJ, Deijen CL, Abis GA, Cuesta MA, van der Pas MH, de Lange-de Klerk ES, Lacy AM, Bemelman WA, Andersson J, Angenete E, Rosenberg J, Fuerst A, Haglind E; COLOR II Study Group. A randomized trial of laparoscopic versus open surgery for rectal cancer. N Engl J Med. 2015 Apr 2;372(14):1324-32
- Fleshman J, Branda M, Sargent DJ, Boller AM, George V, Abbas M, Peters WR Jr, Maun D, Chang G, Herline A, Fichera A, Mutch M, Wexner S, Whiteford M, Marks J, Birnbaum E, Margolin D, Larson D, Marcello P, Posner M, Read T, Monson J, Wren SM, Pisters PW, Nelson H. Effect of Laparoscopic-Assisted Resection vs Open Resection of Stage II or III Rectal Cancer on Pathologic Outcomes: The ACOSOG Z6051 Randomized Clinical Trial. JAMA. 2015 Oct 6;314(13):1346-55.
- Stevenson AR, Solomon MJ, Lumley JW, Hewett P, Clouston AD, Gebski VJ, Davies L, Wilson K, Hague W, Simes J; ALaCaRT Investigators. Effect of Laparoscopic-Assisted Resection vs Open Resection on Pathological Outcomes in Rectal Cancer: The ALaCaRT Randomized Clinical Trial. JAMA. 2015 Oct 6;314(13):1356-63.