Outcomes of metachronous para-aortic lymphadenectomy in colorectal cancer: a systematic review of the literature

Introduction and aim Para-aortic lymph node metastasis associated with colorectal cancer is a very rare occurrence, but at the same time an important predictor of survival. Despite its importance, there is still no gold standard management strategy, particularly for lymph nodes detected during follow-up, after resection of the primary tumour. Therefore, this review was undertaken to examine the evidence available on the surgical and non-surgical management of metachronous para-aortic lymph node metastasis (m-PALNM) in colorectal cancer treatment. Methods This is a systematic review using the patient, intervention, comparison, outcome and study strategy. The literature search was undertaken using Cochrane, MEDLINE, EMBASE and PubMed databases with the following MeSH terms: lymph node excision, para-aortic lymph node, metastasis, metachronous and colorectal neoplasms. Results Five original papers met the study criteria including 188 patients in total (55.3% male, 44.7% female). Surgical resection of the m-PALND was the management of choice in 64% of patients. Reporting styles on survival outcomes were heterogeneous. However, patients undergoing surgical management for m-PALNM had longer disease-free survival and overall survival rates. Conclusion There is significant paucity in the evidence available on the management of m-PALNM. However, the evidence reported by this review suggests that surgical management should be considered whenever possible, with the aim of prolonging survival. Future randomised trials are needed in order to provide further high-level evidence on m-PALNM management.


Introduction
Para-aortic lymph node metastasis (PALNM) associated with colorectal cancer (CRC), unlike liver and lung metastasis, is a rare occurrence with incidence rates reported as < 2% [1].Despite this uncommon occurrence, PALNM is amongst the paramount prognostic factors for CRC [2].
There is still no clearly defined gold standard management strategy for PALNM.A significant contributing factor to this dilemma is the variation in definition of PALNM.According to the Japanese Society for Cancer of the Colon and Rectum PALNM is classified as regional or stage III disease [1,3].On the contrary, the American Joint Committee on Cancer defined PALNM as disseminated stage IV disease and the TNM staging system as M1 disease [1,4].This discrepancy unavoidably causes inconsistency in reporting and measuring management outcomes.
PALNM may occur synchronously or metachronously, with synchronous PALNM (s-PALNM) resulting in more advanced disease at diagnosis, and poorer patient survival outcomes, when compared with metachronous PALNM (m-PALNM) [5].The lack of unambiguously clear distinctions between the two PALNM types in published reports [2,6,7], is also another contributing factor to the ambiguity of an optimal management strategy for PALNM in CRC.
Resection of synchronous and metachronous CRC-associated hepatic or pulmonary metastases is now widely practised after significant evidence of improved survival and morbidity outcomes has been presented [8,9].However, there is relative paucity in the literature available for the surgical and non-surgical management of PALNM in CRC, with a distinct lack of published randomised controlled trials, and the predominant evidence originating from retrospective studies of which a significant amount are case reports.Several studies report the outcomes of simultaneous resection of s-PALNM and the primary CRC [10][11][12][13][14][15][16].Although the evidence supporting this approach still remains weak, the most difficult dilemma refers to the management of m-PALNM, as resection of metachronous disease requires a second laparotomy (or laparoscopy), which could increase patients' morbidity and mortality for a questionable benefit.The aim of this study is to review and summarise the outcomes of surgical and nonsurgical management of metachronous PALNM in CRC and define the pathways for further research in this field.

Eligibility criteria
Eligibility criteria were determined using the population, intervention, comparison, outcome and study design (PICOS) strategy [17].The relevant population included patients who had a diagnosis of isolated metachronous para-aortic lymph node metastasis (m-PALNM) on imaging (computed tomography and/or magnetic resonance imaging and/or positron emission tomography).Surgical management or resection was the intervention of interest.The comparison population were patients who underwent non-surgical management.The outcomes included, but were not limited to, disease-free survival, overall survival, histological characteristics and times to recurrence.Articles which included data on both m-and s-PALNM or other metastatic sites were only included if the m-PALNMrelated data could be isolated from all others.Study designs included were randomised control trials, non-randomised trials, case-control studies, cohort studies (Levels I-III evidence).Levels of evidence were delineated using the Centre for Evidence-Based Medicine's criteria for levels of evidence for therapeutic studies.("Appendix 1") [18,19].
Papers containing non-published series, papers not providing original data, papers not written in English, case series, case reports and expert opinions (levels IV and V evidence) were excluded.Other exclusion criteria were: editorials, letters, non-human studies, cadaveric studies and articles in which PALNM occurred concurrently with distant metastases.

Information sources
A literature search using Cochrane, MEDLINE, EMBASE and PubMed databases was conducted independently by one reviewer (OO) using MeSH terms: lymph node excision, para-aortic lymph node, metastasis, metachronous and colorectal neoplasms ("Appendix 2").Studies were only included if published from the year 2000 and onwards.Bibliographies of relevant studies were searched for additional papers which met the inclusion criteria.The search strategy is summarised in the PRISMA flowchart ("Appendix 2").

Definitions
Para-aortic lymph nodes (PALNs) were defined as lymph nodes surrounding the abdominal aorta and inferior vena cava, which were located in the area from the uppermost part of the origin of the celiac trunk to the lower margin of the aortic bifurcation [3].

Data extraction
Data were extracted on the following categories: study type, participant demographics, diagnostic modalities utilised, resection margins, primary tumour locations, tumour histology, additional therapeutic modalities utilised, surgery-related morbidity and mortality, overall survival, disease or recurrencefree survival/ interval, recurrence rates and recurrence sites.

Assessment of risk of bias
The risk of bias assessment was performed using the National Heart, Lung and Blood Institute's (NIH) Quality Assessment Tool [20].The tool assesses quality based on 14 questions and provides an overall quality assessment of good, fair or poor.

Synthesis of results
Data were extracted to populate a pre-defined series of tables addressing the clinical characteristics described above.

Ethics review
The protocol was reviewed by the research and development department of Central Manchester University Hospitals NHS Foundation Trust who stated that as there is no direct clinical contact, and therefore further ethics committee review was not required.

Search results
The search strategy provided a total of 1897 results, which were critically reviewed for eligibility of inclusion ("Appendix 1").A final total of 5 papers met the study criteria, with two studies originating from South Korea, and one each from Canada, France and the USA (Table 1).There were a total of 188 participants (55.3% male, 44.7% female).Surgical management for metachronous paraaortic lymph node metastasis occurred in 64% of patients (121/188), compared with non-surgical management in 36% (67/188).The location of the primary tumour was the colon in 70% of patients, and the rectum in 30%.(Table 2).

Surgery and short-term outcomes
Of the 121 patients who underwent surgical resection of m-PALNM, 100% underwent an open surgical approach (Table 2).Resection margins were R0 in 88% of patients and R1/R2 in 12% of patients.Five cases were abandoned intraoperatively (Table 3).There was a lack of consistent reporting on the amount of resected and/or positive lymph nodes.The complication rate was 40% within 30 days of surgery, including all grades of complications according to the Clavien-Dindo classification system [21].Overall reported mortality was 0.8% within 30 days of surgery (Table 2).

Neoadjuvant and adjuvant treatment
Neoadjuvant and adjuvant therapeutic interventions involved chemotherapy (CT), radiotherapy (RT) or a combination of both.CT and RT use was variable amongst the studies (Table 4).CT regimens were also variable, which reflects the long study period of this systematic review.

Survival
Reporting styles on survival characteristics were heterogeneous.Patients undergoing surgical management for m-PALNM had longer disease-free survival (DFS) and overall survival (OS) rates.Three of the studies compared survival between surgical and non-surgical cohorts.(Table 5) Min et al. reported DFS and OS between resected and non-resected groups of 22 vs 18 months (mean, p = 0.049) and 42.5 vs 18.7 months (mean, p = 0.034) respectively [22].Kim et al. reported DFS and OS rates between resected and non-resected groups of 24.4 months vs 21.6 months (mean, p = 0.603) and 77 vs 62 months (median, p = 0.055) respectively.[23]Shibata et al. reported 2-year and 5-year OS rates for resected vs nonresected groups of 60% vs 20% and 15% vs 0% respectively [24].Time to m-PALNM is displayed in Table 6.

Risk of bias assessment
The NIH's quality assessment tool for observational cohort and cross-sectional studies was utilised to assess the internal and external validity of the included studies.The overall quality was good for two studies and fair for the remaining three (Table 7).All studies lacked in the justification of sample size; however, this is reasonable as PALNM is a rare pathology, and all studies were retrospective in nature.Additionally, three studies did not adequately defined survival outcome measures [22,24,25], whilst one did not report the definition for PALNM [24].

Discussion
Managing metachronous para-aortic lymph node metastases (m-PALNM) requires understanding of their prognostic significance, weighing up the risk of repeat resection and complications of surgery with overall survival, and comparing surgical  [26].Less is known about the survival benefit of resection of metachronous PALNM.The aim of this systematic review was to summarise the outcomes of surgical and non-surgical management of metachronous PALNM in colorectal cancer (CRC) to allow patients and clinicians to make informed decisions.At the time of submission of this article, there were two very recently published review articles.Chen et al. reviewed both synchronous PALNM (s-PALNM) and m-PALNM articles [29].Their m-PALNM articles included four of the studies also included in this article, except the study by Razik et al. [30] This article did not include a meta-analysis or risk-of-bias assessment and focused on examining survival outcomes and post-op complications.The authors were unable to make definitive recommendations and concluded that a prospective, multicentre study would be required to fully delineate the oncological outcomes of PALND.Wang et al. similarly reviewed both s-PALNM and m-PALNM articles with a focus on overall survival (OS) and post-operative complications [31].The authors performed a meta-analysis for survival and reported that there was a 5-year OS benefit for s-PALNM and m-PALNM resection compared to nonoperative treatment.This article concluded that although PALND (synchronous or metachronous) conferred survival benefits, there is significant paucity in the available literature on PALND and large randomised trials are warranted to confirm the benefits of lymphadenectomy.This present review is not free from limitations.The number of studies extracted from the literature search is rather small, and the overall study period ranges between 1988 and 2014.Treatment strategies have changed for both operative and non-operative management of metastatic colorectal cancer within this period.Surgical strategies for managing metastases from CRC have been more aggressive during the past two decades, and chemotherapy guidelines have changed not only in terms of timing and duration of treatment but also in terms of the recommended regimens, especially after the broad utilisation of biological agents [32].Secondly, the included studies represent small cohorts of patients, and subsequently the evidence derived from their results is rather weak.Pooling and meta-analysing the available data in order to provide a more robust conclusion from this systematic review was not feasible due to the heterogeneity in reporting of the survival outcomes.We acknowledge that the exclusion of studies not published in the English language will have contributed to this small sample size.Also, there was significant variability in the studies with regard to reports on resection margins and the use of neoadjuvant/ adjuvant chemotherapy and radiation, which makes the results difficult to generalise.
Despite the above limitations it is worth highlighting that resection of m-PALMN is reported to be a safe operation despite the fact that patients with recurrent disease undergo a second surgical procedure in a relatively short period after their colectomy.According to older reports, surgical resection for m-PALNM was thought to carry a high risk of complications [15], but this systematic review shows a 40% risk of complications of all grades and a low 30-day mortality rate of 0.8%.Given the fact that resection of m-PALMN has been reported to be open surgery in 100% of the cases, these morbidity and mortality rates should be considered reasonable.Most importantly, this systematic review has demonstrated a significantly improved overall survival in patients undergoing surgical resection for m-PALNM compared to non-operative strategies.This is supported by three out of the five included studies, which compared survival outcomes between surgical and non-surgical management.Clear resection margins were reported in 88% of the cases, which is in keeping with the overall survival benefit observed after surgical management of m-PALNM.
In conclusion, the available evidence, as reported by this systematic review, suggests that surgical management of m-PALNM should be considered a safe procedure aiming to prolong survival.Based on the so far reported outcomes, a future randomised trial would be a safe and ethical step forward to provide more evidence in this field, ideally in a multicentre setting due to the low incidence of m-PALNM.In the era of multimodality treatment for metastatic colorectal cancer, surgical management of isolated metachronous lymph node metastases definitely represents a considerable option in carefully selected patients.

Appendix 1
Centre for Evidence-Based Medicine's criteria for levels of evidence for therapeutic studies.

4 . 6 . 7 .Y 8 .
Were all the subjects selected or recruited from the same or similar populations (including the same time period)?Were inclusion and exclusion criteria for being in the study prespecified and applied uniformly to all participants?For the analyses in this paper, were the exposure(s) of interest measured prior to the outcome(s) being measured?Was the timeframe sufficient so that one could reasonably expect to see an association between exposure and outcome if it existed?Y (for short-term survival) Y (for short-term survival) Y NR For exposures that can vary in amount or level, did the study examine different levels of the exposure as related to the outcome (e.g.categories of exposure, or exposure measured as continuous variable)?definition of isolated RPR) 10.Was the exposure(s) assessed more than once over time?NA (exposure = treatment for PALNM) NA (exposure = PALNM and its treatment) NA (exposure = PALNM and its treatment) NA (exposure = PALNM and its treatment) NA (exposure = PALNM and its treatment)

Table 2
Diagnosis, disease profile and surgery

Table 3
Ichikawa et al. describedan increased survival rate amongst patients having radical paraaortic lymph node resection at the time of primary resection compared to those having targeted lymphadenectomy

Table 6
Time to recurrence