Institutional Outcomes Should Be a Determinant in Decision to Perform Laparoscopic Proctectomies for Rectal Cancer

Purpose Minimally invasive rectal cancer (RC) resection has become common, despite recent high-profile trials failing to show non-inferiority to open proctectomy. We hypothesized that at a high-volume center, laparoscopic resection may have superior outcomes compared to those seen in ALaCaRT and ACOSOG Z6051.


Introduction
The goal of surgical resection for rectal cancer is to optimize cancer-free survival and lower the risk of both local and distant recurrence. Due to the lack of longitudinal data, much of the current surgical research has focused on early outcomes including intactness of the mesorectum, and negative circumferential and distal margins [1][2][3][4][5][6][7][8][9]. There is an ongoing debate regarding the optimal operative approach for management of resectable rectal cancer. It is uncertain whether minimally invasive surgery (MIS) provides equivalent short-term and/or long-term outcomes for these patients in comparison to traditional open resection.
The CLASSIC trial found that patients undergoing laparoscopic proctectomy had a higher rate of positive circumferential margins compared to open surgery [3]. However, follow-up data failed to demonstrate an oncologic difference in long-term outcomes [10]. ACOSOG Z6051 and ALaCaRT trials failed to demonstrate non-inferiority of short-term outcomes for MIS proctectomy compared to open resection [6,7]. On recently published two-year follow-up results, both studies did not demonstrate significant differences in rates of disease-free survival or recurrence between laparoscopic and open approaches, though estimates of treatment effect in ALaCaRT favored open over laparoscopic approach [11,12]. These results have amplified concerns regarding oncologic safety of MIS approaches for rectal cancer.
High hospital volume, defined as more than 25 laparoscopic proctectomies/year, has been associated with better outcomes in rectal cancer [13]. It is therefore reasonable to expect that oncologic safety following laparoscopic proctectomy may vary based on institutional experience. While the feasibility of laparoscopic approaches for surgical resection of rectal cancer has been demonstrated in the setting of randomized trials, institutional level studies evaluating the feasibility, safety and efficacy of laparoscopic proctectomies are limited.
Our aim was to compare our experience with laparoscopic proctectomies to assess outcomes in comparison with data from ACOSOG and ALaCaRT trials. We hypothesized that a high-volume minimally invasive center, where surgeons preferentially select MIS techniques, would have better outcomes following a laparoscopic proctectomy than currently reported in the literature. This would justify a preference for laparoscopic approach in carefully selected patients based on individual center outcomes.

Materials And Methods
Following Institutional Review Board (IRB) approval, a retrospective review of patients undergoing laparoscopic resection for rectal cancer at University Hospitals Cleveland Medical Center from 2007 to 2015 was performed. Patients who underwent surgery for rectal cancer were identified from the medical record using International Classification of Disease 9th Revision diagnosis code 154.1. Inclusion and exclusion criteria were adopted from the ACOSOG and ALaCaRT trials. Patients were included if they were over 18 years of age, had a primary rectal adenocarcinoma (defined as <15 cm from the anal verge by endoscopic or radiologic measurement) diagnosed by histology and were planned to undergo laparoscopic resection. Patients were excluded if they had recurrent disease, were scheduled for a palliative resection, multivisceral resection or intraoperative radiation therapy, if they had clinical stage T4 tumors, stage IV disease or unknown preoperative clinical staging information.
Patient information collected included demographics, comorbidities, use of neoadjuvant and adjuvant chemoradiotherapy, clinical and pathological staging, operative details, complete total mesorectal excision (TME), specimen grade (when available), hospital length of stay and postoperative complications. Patients underwent clinical evaluation including proctoscopy, computed tomography, ultrasound and magnetic resonance imaging as deemed clinically appropriate. The decision for laparoscopic surgery was at surgeon discretion and preference. Study data were collected and managed using a REDCap database hosted at University Hospitals Cleveland Medical Center [14].

Primary and secondary outcomes
The primary outcome was the composite endpoint of negative distal and circumferential resection margins and TME grade, when available. This was modeled on the ALaCaRT and ACOSOG primary study endpoints. Secondary endpoints included conversion rate and significant intraoperative and postoperative morbidity (Clavien-Dindo class 3-5 complications). This data was then compared to published outcomes for rectal cancer to determine if they matched or exceeded acceptable standards for minimally invasive surgery.

Statistical analysis
Continuous variables are shown as the mean with standard deviation (SD) or median with interquartile range (IQR) and were compared using two-way Student t-test or Wilcoxon rank sum test where appropriate. Categorical variables are displayed as frequencies with percentages and were compared using Chi-square or Fischer's exact where appropriate. A p-value less than 0.05 was considered significant. Statistical analysis was performed using Stata/SE 14.2 (StataCorp, College Station, TX, USA).

Results
There were 386 patients who underwent minimally invasive proctectomy between 2007 and 2015 at our institution. In 148 (38.3%) cases, the proctectomy was performed for malignancy. Of these, 89 (60%) patients met the inclusion criteria and were included in the analysis. Patients with multi-visceral resection (n = 19), recurrent disease (n = 20), stage 4 disease (n = 14) and intraoperative radiation (n = 6) were excluded. Mean age at the time of surgery was 66 years and 54 (60.7%) patients were male. Nearly 80% of patients received neoadjuvant chemotherapy and radiation.

Technical and oncologic success
A clear circumferential margin was obtained in 87 (97.8%) cases. A negative distal margin was achieved in 87 (97.8%) cases. Completeness of TME based on pathologic assessment was not standardized until 2012; therefore, TME completeness was only available for 33 cases. The 33patient cohort was statistically compared to our larger 89-patient group and was found to be clinically similar in the relevant demographic, patient and tumor characteristic, operative detail and outcome parameters. TME was graded as complete/near complete in 31 (93.9%) of the cases (29 complete, four nearly complete). The mean number of lymph nodes examined was 18.3 (SD 5). Overall, 87.9% of cases were considered "pathologically successful" based on negative circumferential and distal margins and complete/near complete TME. The majority of tumors (69.7%) were well or moderately differentiated. These results are summarized in Table  3.

Outcomes compared to published literature
We compared our data to historical data provided from the ACOSOG and ALaCaRT trials. Demographic data between our study and the ACOSOG and ALaCaRT studies was generally similar, with a few noticeable differences ( Table 4). In our experience, the location of the tumor was slightly higher in the rectum, with only 24.4% being low compared to 51.2% for the ACOSOG and 35% for the ALaCaRT. We also had fewer patients who received preoperative chemoradiation (79.8%) compared to the 95% of ACOSOG, but more than the 50% of ALaCaRT participants who received preoperative radiation.  and our study had a lower proportion of hand-assisted or robotic assisted procedures (11.4% vs. 31.3% p < 0.001) than the ACOSOG trial. Compared to ACOSOG, conversion rate (24.7% vs. 11%, p = 0.001) and 30-day readmission rates (12.4% vs. 3.3%, p = 0.002) were higher in our experience ( Table 5).   Overall, our pathologic outcomes were favorable when compared to ACOSOG results. Our CRM margin was negative in 97.8% compared to the 87.9% of ACOSOG (p < .001). The overall "success rate" was higher in our study compared to in ACOSOG (87.9% vs. 81.7%, p = 0.47), although this value did not reach statistical significance. Lymph node harvests were similar between studies.

Current Study ALaCaRT P-value
Pathologic outcomes between our study and ALaCaRT were similar for most measures, including negative distal margin, completeness of TME, and percent successful resection. Our radial margin was negative in 98.9% of patients, compared to 93% (p = 0.047) in the ALaCaRT data. The tumors in the two studies had similar sizes and histologic grades, although a greater proportion of patients in our study had node negative disease (N0 79.8% vs. 62%, p < 0.01).
This study demonstrates that at a high-volume minimally invasive center, MIS proctectomy for surgical resection of rectal cancer can be performed safely with excellent short-term oncologic outcomes. While much of the current debate regarding the use of MIS for resection of rectal cancer is focused on the feasibility of technique, we believe that the preference for minimally invasive approaches should be based on institutional and surgeon data.
To this point, we found that the selection of a minimally invasive approach was justified in our experience as we were able to achieve a high overall negative CRM margin in 98% of the cases and an overall success rate of 88% in the 33-patient cohort with reported TME grade. While this success rate surpasses 81.7% in ACOSOG and 82% in ALaCART, it is comparable to the open surgical success seen in ACOSOG (86.9%) and ALaCART (89%). We have demonstrated that individual centers can achieve results that compare favorably to the outcomes of MIS approaches and are similar to results of open approaches reported in ACOSOG and ALaCaRT clinical trials. These results emphasize the importance of institutional level evaluation of outcomes to guide decision making regarding the appropriate surgical approach. We believe that our good outcomes can be attributable to careful patient selection of best candidates for surgery and high volume of experience of each surgeon practicing in our institution at the time.
The findings from our study are consistent with the current literature and adds to the growing evidence in favor of the use of laparoscopic approaches. Arezzo et al. performed a systematic review and meta-analysis, evaluating the outcomes of laparoscopic approaches compared to open resection [9]. In their meta-analysis they included results from clinical trials and retrospective studies and found that the incidence of CRM margin involvement was 7.9% after laparoscopic resection compared to 6 There are several limitations to this study. Standardization of surgeon and pathologist oncologic evaluation over time presents complications in normalizing the data. We were only able to report completeness of TME in under half of our procedures. Therefore, we were only able to report "successful resection" in a proportion of our patients. It may be inaccurate to extrapolate high rates of complete TME to our earlier patients. The lack of uniformity in oncologic outcomes between ACOSOG and ALaCaRT complicates direct comparison. Examples of this include the use of means versus medians, and distal resection margin of 1 mm versus greater than 1 mm. Additionally, our study cohort is limited by its sample size, which restricts our ability to demonstrate statistical significance in some cases. Lastly, there were differences in tumor demographics between our study and ACOSOG and ALaCaRT. A greater proportion of our tumors were higher in the rectum, which may affect rates of "successful" surgery. Additionally, our conversion and APR rates were higher than the trial populations. Surgeon specific and institutional specific data should not be extrapolated to other institutions, and surgeons need to know their own data to make decisions about the best possible approach in each surgeon's hands.
Decisions regarding choice of surgical technique should be based on data obtained from the institution and surgeon, rather than based on technique. Similar to baseball, one would not give a bat a batting average, the batting average would be attributed to the batter, using their bat of choice. Comparably, we hope that the debate regarding the use of minimally invasive surgery will focus on surgeon and institutional outcomes, rather than global evaluation of the techniques.

Conclusions
Minimally invasive proctectomy for the surgical resection of rectal cancer can be performed safely, but the decision for the selection of minimally invasive approach should be based on the institutional evaluation of outcomes for optimal results.

Additional Information Disclosures
Human subjects: Consent was obtained by all participants in this study. University Hospitals Cleveland Medical Center Institutional Review Board issued approval 11-15-21C. The IRB reviewed and approved this submission. Per Federal regulation, changes MAY NOT be made to any element of the current research without prior IRB approval, except to eliminate an immediate and apparent hazard to subjects enrolled in the study. Per Federal regulation, the research may not continue beyond the Approval End date. You must submit a continuing review form 6-8 weeks before this Approval End date in order to maintain IRB approval. Failure to maintain IRB approval is human subjects non-compliance. Please note that even if your study falls into a category that does not require an Approval End date, the institution may require a yearly "checkin" to confirm the status of the study. . Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work.