In contrast to the available data for colon cancer, there is limited quality data for the application of laparoscopy for rectal cancer. There are several single-institution reports supporting the feasibility and equivalent long-term outcomes of laparoscopy. Nevertheless, the CLASICC study provides the highest level of evidence to date. Current on-going studies, including the American College of Surgeons Oncology Group (ACSOG) Z6051 and the Robotic versus Laparoscopic Resection for Rectal Cancer (ROLARR), aim to provide a rigorous evaluation of these approaches to the treatment of rectal cancer.
Classic: rectal cancer
The CLASICC study found that the laparoscopic arm for rectal cancer was more likely to have undergone a total mesorectal excision. This supports the ability to maintain standard of care using a laparoscopic approach. Short-term results, however, found a trend towards positive circumferential resection margins in those undergoing laparoscopic LAR (lap: 12%, open: 6%, (p = 0.19). This raised concerns of increased risk for local recurrence [
9].
Three and 5-year data, however, did not identify an increased risk for local recurrence among patients having laparoscopic operation for rectal cancer. Specifically, the 5-year local recurrence rate for LAR was 9.4% for laparoscopy and 7.6% for open (p = 0.740). Five-year data was not provided for the local recurrence rates for patients in the APR group, but there was no significant difference identified. Overall, the distant recurrence rate was 20.9% found in 111 cases at 5 years. There was no significant difference between either technique (lap: 21%, open: 20.6%; p = 0.820) [
10,
11].
Overall survival was also equivalent for laparoscopic and open resection of rectal cancer (lap: 60.3%, open: 52.9%; p = 0.132). Similar results were found regardless of whether patients underwent LAR or APR (LAR: lap: 62.8%, open: 56.7%;p = 0.247, APR: lap: 53.2%, open: 41.8%; p = 0.310). In fact, data at three years suggested a trend towards improved survival with laparoscopy in Dukes’ A patients, though this did not persist at 5 years (p = 0.491). This suggests that there is no difference in overall survival at 5 years between treatment arms for any stage of rectal cancer [
10,
11]. Cancer-free survival at 5 years was also not significantly different for patients with rectal cancer (lap: 53.2%, open: 52.1%; p = 0.953). The 5 year cancer-free survival was not significantly different for LAR (lap: 57.7%, open: 57.6%; p = 0.832) or for APR (lap: 41.4%, open: 36.2%; p = 0.618).
Patients requiring conversion from laparoscopic to open operation during resection for rectal cancer fare worse with significantly decreased 5 year overall survival (lap: 62.4%, open: 58.5%, conversion 49.6%; p = 0.005). Based on sensitivity analysis, this worse overall survival outcome was maintained even for surgeons with lower than average conversion rate, suggesting surgeon-related factors were unlikely related.
The Comparison of Open versus laparoscopic surgery for mid and low Rectal cancer After Neoadjuvant chemoradiotherapy (COREAN) trial is another randomized trial recently published showing equivalent short-term oncologic outcomes between open and laparoscopic surgery for rectal cancer. The trial included 340 patients with cT3N0-2 mid or low rectal cancer without distant metastasis. Conversion to an open procedure occurred in 2 patients (1.2%). Along with equivalence of oncologic outcomes, other short-term outcomes such as recovery of gastrointestinal function, analgesia requirement, and return to physical function showed improvement in the laparoscopic group [
17].
COLOR II is an ongoing non-inferiority RCT that has 27 participating international sites with 739 patients, focusing on the outcomes from laparoscopic versus open rectal cancer resection. [
18,
19] Several individual studies have shown that laparoscopic total mesorectal excision is feasible and safe but CLASSICC is the only RCT to reliably support laparoscopic resection as an adequate and equivalent treatment for rectal cancer. Inclusion criteria for COLOR II include a single rectal mass within 15 cm of the anal verge on rigid proctoscopy or distal to the conjugate line on CT/MRI. Preoperative imaging is required to exclude distant metastasis. Tumors amenable to local excision and those with radiographic features suggestive of local invasion are excluded. The primary endpoint for this study will be loco-regional recurrence at 3 years.
Three other smaller, single-center RCT studies were included along with the CLASICC study in the Cochrane meta-analysis of operative approaches to rectal cancer. Those smaller trials will be discussed here. The first by Araujo et al. included 28 patients and focused on laparoscopic versus open APR after neoadjuvant therapy. No conversions were performed. Mean follow up was 47 months and only postoperative complications and local recurrence were noted. They found significantly fewer lymph nodes harvested in the laparoscopic group (5.5 nodes) compared to open (11.9) (p = 0.04). They attributed this difference to the small number of patients. At 47 months of follow up, two local recurrences were noted in the open group and none in the laparoscopic group [
20].
The second trial included in the Cochrane analysis by Zhou et al. included 171 patients, comparing open and laparoscopic low and ultralow anterior resection. Included patients had the distal margin of tumor distal to the peritoneal reflection and 1.5 cm above the dentate line. Total mesenteric excision with anal sphincter preservation was accomplished in all patients. The average operative time was not significantly different (lap: 120 min, open: 106 min; p > 0.05) although blood loss was significantly less with laparoscopy (lap: 20 mL, open: 92 mL; p < 0.05). The outcomes of this study included postoperative recovery and local recurrence. There was no significant difference in days until start of fluid intake (p = 0.713) or in days of analgesia (p = 0.225). Days to first bowel movement was significantly shorter in the laparoscopic group (lap: 1.5 days, open: 2.7 days; p = 0.009). The number of hospital days was also significantly shorter in the laparoscopy group (lap: 8.1 days, open: 13.3 days; p = 0.001). Postoperative complications such as urinary retention, infection, obstruction and anastomotic leakage were significantly decreased with laparoscopy (lap: 6.1%, open: 12.4%; p = 0.016). Two port site recurrences were noted in the laparoscopic group and 3 pelvic local recurrences in the open group. Statistical significance was not reported. No mortalities were noted in the 1 to 16 month follow-up. The authors conclude that adequate resection of low rectal cancers can be performed laparoscopically but is a technically challenging approach. Long-term results regarding survival were not reported in this study [
21].
The third study in the Cochrane analysis by Braga et al. followed the 5-year outcomes in 391 patients with colorectal cancer. Of this group, 134 patients had rectal cancer with 68 in the laparoscopic group and 68 in the open group. The outcomes in this study were not classified between colon and rectal cancer patients. No difference was found with respect to number of lymph nodes recovered. Distal and radial margins were negative in all patients. The rate of conversion was 4.2%. Outcomes measured included postoperative morbidity and disease-free survival. The rate of anastomotic leak was not significantly different (lap: 4.7%, open: 6.9%; p = 0.46). Reoperation was required in 6.3% of laparoscopic cases due to anastomotic leak in seven, adhesive disease in three, and bleeding in two cases compared to 9.4% of open cases, with anastomotic leak in 11, adhesive disease in six, and bleeding in two cases (p = 0.34). The length of stay was significantly shorter for the laparoscopic group (lap: 9.4 days, open: 12.7 days; p = 0.0001). Long-term complications were also noted to be significantly lower in the laparoscopic group overall (lap: 6.8%, open: 14.9%; p = 0.02). Follow-up was between 15 to 60 months. Five-year overall and disease-free survival were not significantly different between open or laparoscopic groups although p-values were not provided. Of note, the study describes that the local recurrence rate in rectal cancer was 7.3% in the laparoscopic group and 8.8% in the open group, with no p-value provided [
22].
The ACSOG Z6051 study published data of their phase II pilot study in 2011 supporting that laparoscopic-assisted resection had both acceptable oncologic and perioperative clinical outcomes when compared to open resection. The pilot study was created with the intention to provide baseline parameters for a planned randomized control trial. This included the data of 54 patients with stage I to III rectal cancer obtained from 2001 to 2005. Exclusion criteria included stage IV cancer, pregnancy, and patients with ASA IV and V. Three surgeons performed the laparoscopic procedures, all with extensive laparoscopic experience (>300 procedures for colorectal disease and >20 laparoscopic rectal dissections). Pouch reconstruction and hand-assisted approaches were based on surgeon preference. Follow-up data was collected up to 5 years post-operatively.
Perioperative results described significantly earlier return of bowel function in the laparoscopic group (p = 0.03). There was no significant difference in the complication rate (lap: 22.2%, open: 32.4%; p = 0.178). Local recurrence was similar (lap: 2%, open: 4.2%, p = 0.417) as was 5-year overall survival (lap: 90.8%, open: 88.5%; p = 0.261) and disease-free survival (lap: 80.8%, open: 75.8%; p = 0.390). All cases of local recurrence occurred within 2 years. Conversion to open resection occurred in 6 cases (11.1%) and reasons included difficulty obtaining sufficient length and difficulty in completing the anastomosis. A positive circumferential rectal margin was found in one laparoscopic case, compared to 7 in the open group (p > 0.05). This was noted to be the opposite of CLASICC findings. Nevertheless the 3-year local recurrence and survival were not different between either laparoscopic or open approaches, which is also seen in the ASCOG Z6051 study.
Authors of the Z6051 pilot trial identify the study size as a major limitation, despite well-matched groups. Nevertheless, the current data suggests that the laparoscopic approach provides the potential for acceptable outcomes [
23]. The Phase III randomized controlled trial arm of the study is currently recruiting patients.
As the aforementioned studies provide encouraging data supporting the efficacy of laparoscopic approaches to rectal cancer, other centers seek to evaluate the outcomes of robotic-assisted laparoscopy in this setting. Given the technical difficulties of applying laparoscopy to the confined pelvic space, robotic assistance has the advantage of providing manipulation of instruments with 7-degrees of freedom of movement, as well as enhancement of dexterity and field of view. To date, there are a number of case series as well as one published randomized trial with 18 patients showing no difference in the outcomes, conversion rates or operative time comparing robotic-assistance to standard laparoscopy [
24].
The Robotic versus Laproscopic Resection for Rectal Cancer (ROLARR) trial is a prospective, randomized, controlled, multi-center, unblinded superiority which began recruiting in 2010 seeking to primarily investigate rate of conversion, circumferential margin positivity, 3-year local recurrence and overall outcomes of the robotic-assisted approach. Other outcomes of interest include cost-effectiveness and quality of life The authors aim for about 400 patients to maintain 80% power. Participating surgeons are required to have performed at least 10 robotic-assisted resections. Along with measuring overall oncologic outcomes, ROLARR seeks to evaluate the clinical benefits of robotics, including preservation of normal bladder and sexual function. Given the costs of robotic systems, these measures are essential to justify the use of robotics [
25].
In conclusion, the larger CLASICC study and the three smaller single-center studies support that the laparoscopic approach to rectal cancer appears to provide an adequate oncologic excision with similar long-term outcomes. Current studies such as the Z6051, ROLARR and COLOR II studies can provide further insight on the outcomes of these minimally invasive approaches.