Rise and fall of total mesorectal excision with lateral pelvic lymphadenectomy for rectal cancer: an updated systematic review and meta-analysis of 11,366 patients

Abstract The role of lateral lymph node dissection (LLND) during total mesorectal excision (TME) for rectal cancer is still controversial. Many reviews were published on prophylactic LLND in rectal cancer surgery, some biased by heterogeneity of overall associated treatments. The aim of this systematic review and meta-analysis is to perform a timeline analysis of different treatments associated to prophylactic LLND vs no-LLND during TME for rectal cancer. Methods A literature search was performed in PubMed, SCOPUS and WOS for publications up to 1 September 2020. We considered RCTs and CCTs comparing oncologic and functional outcomes of TME with or without LLND in patients with rectal cancer. Results Thirty-four included articles and 29 studies enrolled 11,606 patients. No difference in 5-year local recurrence (in every subgroup analysis including preoperative neoadjuvant chemoradiotherapy), 5-year distant and overall recurrence, 5-year overall survival and 5-year disease-free survival was found between LLND group and non LLND group. The analysis of post-operative functional outcomes reported hindered quality of life (urinary, evacuatory and sexual dysfunction) in LLND patients when compared to non LLND. Conclusion Our publication does not demonstrate that TME with LLND has any oncological advantage when compared to TME alone, showing that with the advent of neoadjuvant therapy, the advantage of LLND is lost. In this review, the most important bias is the heterogeneous characteristics of patients, cancer staging, different neoadjuvant therapy, different radiotherapy techniques and fractionation used in different studies. Higher rate of functional post-operative complications does not support routinely use of LLND. Supplementary Information The online version contains supplementary material available at 10.1007/s00384-021-03946-2.


Introduction
Although commonly performed in urologic (1) and gynaecologic (2) surgery, the role of lateral lymph node dissection (LLND) is still a very controversial topic in rectal cancer treatment (3). This procedure, reported in Japan in the 1970s (4,5), was standardized by Moriya at the end of the 1980s: "On the basis of the extent of lateral node spread, two types of lateral node dissection were performed, consisting of preservation of internal iliac vessels (conventional) and en-bloc excision of these vessels (extended)" (6).
Currently, total mesorectal excision (TME) remains the gold standard for surgical treatment of mid and low rectal cancer. In contrast, the place of LLND remains a matter of controversy between Eastern and Western surgical guidelines (7)(8)(9)(10)(11)(12). The main conceptual difference is the fact that the lateral pelvic lymph nodes are considered as localized disease in Japanese clinical practice, whereas the West treats them as systemic disease (13)(14)(15). For this reason, in Japan, prophylactic LLND is always performed in patients with stage II/III lower rectal cancer, whereas in the West, chemoradiotherapy (CRT) is routinely performed, thus generally avoiding a more invasive surgical approach (16).
To date, seven systematic reviews and meta-analyses have provided the highest levels of evidence to support the role of LLND for rectal cancer (17)(18)(19)(20)(21)(22)(23). This new systematic review and meta-analysis aims to perform an updated analysis of the different types of treatments associated with prophylactic LLND vs. no-LLND (NLLND) in rectal cancer surgery.

Methods
We performed a systematic review adhering to AMSTAR 2 principles (24). A literature search was performed from two authors (R.C., F.B.) in PubMed, SCOPUS and WOS for publications up to 1 September 2020. The protocol for this study was registered on PROSPERO, a prospective international database for reviews under the registration number 42020186525.
Inclusion criteria We considered RCTs (randomized control trial) and CCTs (clinical control trials) comparing patients with rectal cancer who underwent rectal resection and TME with versus without LLND.
Exclusion criteria Patients having surgery without TME.
The Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines were followed (25)(ESM6). The keywords used for PubMed database research were: "extended lymphadenectomy," "pelvic lymphadenectomy," "lateral lymph-node dissection," "total mesorectal excision," "rectal resection," "rectal cancer," and their combinations. The search strategy performed on PubMed was the We also manually searched the references of identified articles and relevant reviews and searched conference proceedings, theses and published abstracts on Google scholar. No language restriction was applied.
Outcomes The primary outcomes were the incidence of local recurrence and distant recurrence at 5 years. The secondary outcomes were the 5-year overall and disease-free survival and the incidence of urinary dysfunction (retention), urinary incontinence, evacuatory dysfunction and sexual dysfunction.
The assessment of methodological quality was performed independently by two authors (RC, CR). The risk of bias of randomized control trials (RCTs) was assessed using methods described in the Cochrane Handbook for Systematic Reviews of Interventions (26) and the ROBINS-I tool (27) for observational studies. In the ROBINS-I tool, risk of bias is assessed within specified domains, including bias due to confounding, bias in selection of participants into the study, bias in classification of interventions, bias due to deviations from intended interventions bias due to missing data, bias in measurement of outcomes, bias in selection of the reported result and overall bias. Bias assessments were tabulated with explanation. Disagreements were resolved via discussion between the investigators. Graphic representation of the results was produced using the Robvis online tool (28) (ESM4-5).

Statistical analysis
This meta-analysis was conducted using the Review Manager (RevMan version 5.3.5) computer program (Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014).
The dichotomous outcomes were pooled with a randomeffects model with the Mantel-Haenszel method to estimate risk ratios (RRd) and their 95% confidence intervals (29). Clinical heterogeneity was tested using τ2, Cochrane's Q and I 2 statistics. We considered an I 2 value exceeding 50% to be indicative of heterogeneity (30).
We used a random-effect analysis model for the high clinical heterogeneity and statistically significant higher chi 2 value and I 2 (31). In all remaining circumstances, we used the random-effects model.
The following subgroup analyses were performed to reduce the heterogeneity:

Characteristics of the studies
The 28 included studies were published between 1994 and 2020; patients were enrolled between 1985 and 2016 (Table 1). In all studies, the cancer was located at the rectum,  Risk of bias Seven domains for the potential risk of bias of included RCTs using methods described in the Cochrane Handbook for Systematic Reviews of Interventions were analysed (26). All studies were rated as unclear risk of random Fig. 1 Forest plot, 5-year local recurrence sequence generation (selection bias) and five studies for allocation concealment (selection bias). Blinding of participants and personnel and incomplete outcome data were rated as high risk in all included studies. Five studies were rated as low risk of selection bias for selective reporting (reporting bias) and other bias. The ROBINS-I tool was used to evaluate the quality of the comparative studies.

Discussion
Rectal cancer represents the third leading cause of death worldwide with a steadily increasing incidence (70,71). The concept of TME, introduced by Heald, has revolutionized the treatment by reducing the local recurrence rates from up to 40% to 4-8% (15,71). TME does not include the removal of the lateral pelvic lymph nodes but only those found within the mesorectal fascia and along the course of the mesenteric vessels. The efficacy of the excision of the pelvic lateral lymph nodes is still a controversial topic (72,73).
The lymphatic drainage of the rectum, especially for the most distal/lower rectum, through the submucosal plexus, drains in three trunks: the upper branch, which flows into the lymphatic channels of the lower mesenteric vein; the middle branch, draining to the lymph nodes surrounding the internal, external and common iliac vessels; the lower branch draining to the inguinal lymph nodes (15) (Table 2; Fig. 8). These lateral regional lymphatic areas outside of the mesorectum are classified into six regions near the following arteries: the internal  pudendal (outside of the pelvic plexus), the internal iliac (proximal to the superior vesical artery), the common iliac, the external iliac, the obturator and the presacral regions. Among these regions, the internal iliac artery and obturator regions have the highest rate of nodal involvement (22-61%) and are called the 'vulnerable field' (13,14).
The different approaches to the LLND between the East and West stems from the concept that pelvic lateral lymph nodes are considered regional according to Japanese authors and staging systems. The Western world, with the latest AJCC guidelines (AJCC 8 th edition), confirms pelvic lateral lymph node stations as remote stations. This is mainly debated in the case of stage II and III low rectal cancer. The involvement of lymph node stations in the iliac and obturator regions varies from 10.6 to 25.5% (15) stage II and III rectal cancer below the peritoneal reflection. More specifically, pelvic extra-regional lymph node involvement is reported in 5.4% of T1 cases, 8.2% for T2, 16.5% for T3 and 37.2% for T4 (58). For this reason, Japanese surgeons suggest performing TME with bilateral pelvic lymphadenectomy without neoadjuvant treatment, as they expect that the risk of intrapelvic recurrence decreases by 50%, and 5-year survival improves by 8 to 9% (7,8).
On the contrary, surgeons of the Western world generally treat rectal cancer with a classical TME and often considering neoadjuvant CRT(74), preserving LLND for patients with clinically suspected lateral pelvic lymph node metastasis (9)(10)(11).
The comparison between LLND versus CRT for lateral pelvic lymph nodes mainly concerns the rate of local pelvic recurrence. The only RCT comparing these two surgical techniques is JCOGO212 (41), which compared TME vs. TME and lateral pelvic lymphadenectomy in patients who had no lateral pelvic lymphadenopathy before surgery. The rate of local recurrence decreased from 12.6% in cases of TME alone to 7.4% when TME was associated with lateral lymphadenectomy. A limitation of this study was the choice of not performing preoperative CRT before TME, even when it would have been indicated according to Western guidelines (12,74). Long-term followup of JCOGO212 confirms the non-inferiority of TME alone compared to TME with pelvic lymphadenectomy in patients without clinically identifiable pelvic lymph node involvement. The study concludes that pelvic lateral lymphadenectomy should only be performed in patients with radiological evidence of lymph node involvement.
Other studies (54) confirm that the risk of pelvic recurrence rises to 19.5% in patients with lateral pelvic lymph nodes of a size more than 7 mm after neoadjuvant therapy. On the other hand, there is little evidence on the true efficacy of bilateral pelvic lymphadenectomy for low rectum carcinomas without clinical evidence of bilateral pelvic lymphadenopathy (41). Although TME alone should not be considered inferior to TME with lateral lymphadenectomy, surgery extended to lateral pelvic lymph nodes reduces the risk of pelvic recurrence, especially in radiologically positive cases.
The main point of the discussion remains the risk of lateral pelvic lymph node metastases even after neoadjuvant CRT. The literature (72) reports a high percentage (up to 30-40%) of pelvic lymph node involvement even after neoadjuvant CRT.
The results from the present analysis confirm that the more radical and invasive surgical approach does not appear to be the safest and optimal way to treat these patients. The comparison between LLND and NLLND groups showed no Regarding overall survival, the cumulative analysis also revealed a lack of any advantage of LLND, but the subgroup analysis did show improved overall survival in the group with LLND plus neoadjuvant CRT.
The main concern for the more invasive surgical approach of LLND is additional complications. It is recognized that higher occurrence of urinary, defecatory and sexual dysfunctions is found after LLND (3,75), despite the introduction of nerve-sparing techniques. In the present analysis, the incidence of urinary retention and incontinence and sexual dysfunctions was directly compared in patients with and without LLND. The only statistically significant difference was the higher incidence of urinary retention in patients undergoing LLND. Another possible confounding factor is the fact that the comparison in most cases was carried out on patients without CRT, which is a procedure also burdened with similar and potentially additional, functional complications. More targeted studies are needed to assess the safety and quality of life following LLND surgery.
An important limitation of the present analysis is the possible bias introduced by the high heterogeneity of the clinical and oncological status of the included patients. Furthermore, our analysis could be expanded and completed by examining other data, such as the number of harvested lymph nodes, additional lymph node metastases detected in the LLND group and differences in functional outcomes between minimally invasive surgeries versus open resections.

Conclusion
Our results suggest that TME with LLND does not offer an oncological advantage over TME without LLND. The advantage of LLND in the pre-neoadjuvant CRT era is lost after the implementation of neoadjuvant CRT. The addition of adjuvant CRT to LLND appears to contribute towards better survival and diminishes the rate of local recurrences. Whilst incurring in heterogeneity of data analysing currently available literature, the evidence would suggest that there is no place for routine LLND in the management of rectal cancer.
These findings reiterate the importance of careful selection of patients for LLND through an improved definition of pathological lymph nodes. Improved imaging techniques to accurately define a reliable cut-off size and describe radiological abnormalities that accurately predict involvement of pelvic lymph nodes are needed. Further studies, preferably prospective, that focus on survival and its association with surgical technique are needed to establish an evidence-based cut-off, which would aid in identifying precise indications for LLND.
Funding Open access funding provided by Università degli Studi di Ferrara within the CRUI-CARE Agreement.

Conflict of interest The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.