The Value of Paratracheal Lymphadenectomy in Esophagectomy for Adenocarcinoma of the Esophagus or Gastroesophageal Junction: A Systematic Review of the Literature

Background The role of upper mediastinal lymphadenectomy for distal esophageal or gastroesophageal junction (GEJ) adenocarcinomas remains a matter of debate. This systematic review aims to provide a comprehensive overview of evidence on the incidence of nodal metastases in the upper mediastinum following transthoracic esophagectomy for distal esophageal or GEJ adenocarcinoma. Methods A literature search was performed using Medline, Embase and Cochrane databases up to November 2020 to include studies on patients who underwent transthoracic esophagectomy with upper mediastinal lymphadenectomy for distal esophageal and/or GEJ adenocarcinoma. The primary endpoint was the incidence of metastatic nodes in the upper mediastinum based on pathological examination. Secondary endpoints were the definition of upper mediastinal lymphadenectomy, recurrent laryngeal nerve (RLN) palsy rate and survival. Results A total of 17 studies were included and the sample sizes ranged from 10-634 patients. Overall, the median incidence of upper mediastinal lymph node metastases was 10.0% (IQR 4.7-16.7). The incidences of upper mediastinal lymph node metastases were 8.3% in the 7 studies that included patients undergoing primary resection (IQR 2.0-16.6), 4,4% in the 1 study that provided neoadjuvant therapy to the full cohort, and 10.6% in the 9 studies that included patients undergoing esophagectomy either with or without neoadjuvant therapy (IQR 8.9-15.8%). Data on survival and RLN palsy rates were scarce and inconclusive. Conclusions The incidence of upper mediastinal lymph node metastases in distal esophageal adenocarcinoma is up to 10%. Morbidity should be weighed against potential impact on survival.

Esophagectomy in the context of bi-or trimodality treatment achieves a 5-year survival rate of 40-50% in patients with locally advanced esophageal cancer. 1,2 An adequate lymphadenectomy is an essential part of esophagectomy and most surgeons routinely remove at least the abdominal and mediastinal lymph node stations (i.e. two-field lymphadenectomy). 3 Recent studies emphasized the need for a proper lymph node dissection in esophagectomy, as the lymph node yield is directly related to overall survival. 4,5 Nonetheless, there is an ongoing debate regarding the mediastinal lymph node stations that should be dissected routinely. Particular disagreement exists on the role of an upper mediastinal lymphadenectomy involving the paratracheal lymph nodes (stations 2 and 4 or 105 and 106 according to the American and Japanese classification systems, respectively). 6,7 While removal of these stations might improve locoregional control of disease, the paratracheal lymph nodes are located in a complex anatomical area containing the recurrent laryngeal nerves and several major vascular structures. [8][9][10] A paratracheal lymph node dissection, particularly on the left side, is surgically challenging, especially in the open approach. Hence, the potential oncological merits of paratracheal lymphadenectomy need to be weighed against its possible risks in terms of surgical morbidity.
While upper mediastinal lymph node metastases most often occur with squamous cell carcinoma, several studies have shown that paratracheal lymph node metastases are also found in patients with distal adenocarcinoma, even after neoadjuvant therapy. [11][12][13] In this perspective, the entire mediastinal peri-esophageal lymph node network should theoretically be removed to optimize the chances of curation and survival following esophagectomy for both adeno-and squamous cell carcinoma. However, available evidence on this topic is conflicting and there is substantial variation regarding the definition of an upper mediastinal lymphadenectomy. 14 Moreover, there is controversy regarding the survival benefit of extensive mediastinal lymphadenectomy in patients who received neoadjuvant chemoradiation (nCRT).
To provide an objective overview of current evidence on the clinical value of paratracheal lymphadenectomy for adenocarcinoma of the distal esophagus or gastroesophageal junction (GEJ), a systematic review of the literature was performed.

Study Selection, Eligibility Criteria
Titles and abstracts were scrutinized by the same two independent researchers (AG and FK) to determine their suitability for inclusion. The full text of potentially relevant articles was retrieved and assessed for inclusion by the same two authors. All studies reporting on the primary endpoint, i.e. the incidence of pathologically confirmed upper mediastinal lymph node metastases in patients undergoing transthoracic esophagectomy for distal esophageal or GEJ adenocarcinoma, were included. Secondary endpoints included the exact definition of performed upper mediastinal lymphadenectomy, recurrent laryngeal nerve injury rate, and survival. Upper mediastinal lymphadenectomy was defined as part of a total mediastinal lymphadenectomy. 13 Case reports, studies with fewer than 10 patients, reviews, poster abstracts, animal studies, and non-English-language articles were excluded. If authors from the same institution had published a primary paper and then an updated analysis with a larger patient cohort, the most recent publication was included. The reference list of articles obtained was searched to identify additional articles. Any discordances between the two authors regarding study inclusion were resolved between all coauthors. The quality of all selected articles was scored according to the Centre for Evidence-Based Medicine (CEBM) Levels of Evidence, 2011 version.

Data Abstraction
For eligible studies, data were extracted from the original articles, including publication year, country of origin, sample size, age, sex, histology, tumor location, neoadjuvant treatment, pathological T and N stage, definition of performed upper mediastinal lymphadenectomy, number of patients with pathologically positive nodes in the upper mediastinum, recurrent laryngeal nerve palsy rate, and survival. If data from any of the above items were not reported in the study, items were indicated as 'NR' (not reported). The extracted data were presented per study in tables. As the number of studies was limited and the variability in study design was considerable, no metaanalyses were performed. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed for the conduction and reporting of this systematic review.

Inclusion
The primary literature search through the PubMed, Embase, and Cochrane libraries identified 543 studies. The search results and selection process are summarized in Fig. 1. The first screening discarded 496 papers, based on title and/or abstract, leaving 47 studies for full-text assessment. Thirty studies were excluded due to rates of upper mediastinal lymph node metastases not being reported (n = 19), mediastinal nodal status reported altogether without stratifying upper mediastinal lymph nodes (n = 5), reporting on fewer than 10 patients (n = 4), using the same case series as a more recent paper from the same authors (n = 1), and results not being reported in absolute numbers (n = 1). The remaining 17 studies 15-31 were included in this review.

Overall and Disease-Free Survival
Survival data for patients with pathologically positive nodes in the upper mediastinum after esophagectomy were reported in six studies, as shown in Table 5. Two studies reported on 5-year overall survival that ranged between 17 and 44.4%. 19,26 One study 15 reported a median disease-free survival of 15.4 months, while the remaining three studies reported on median survival (8 months), 21 5-year diseasefree survival (0%), 24 and 3-year overall survival of 53%. 30

Recurrent Laryngeal Nerve Palsy Rate
Only one study reported on the recurrent laryngeal nerve (RLN) palsy rate, which occurred in 8 of 129 patients in that study (6.2%). 20 Note that this rate was from all patients submitted to transthoracic esophagectomy in that case series, and was not specific for patients who received an upper mediastinal lymphadenectomy (67 patients in that group).

DISCUSSION
Based on this systematic review, the mean incidence of upper mediastinal lymph node metastases was up to 10% in patients undergoing transthoracic esophagectomy for distal esophageal or GEJ adenocarcinoma. Substantial variation was found in the reported incidences, which is probably at least partially explained by different definitions being used for an upper mediastinal lymphadenectomy. Furthermore, in most studies (9/17), selected patients underwent mediastinal lymph node dissection.
Transthoracic esophagectomy is currently considered to achieve the best oncological outcomes, as it allows a thorough mediastinal lymphadenectomy to maximize the chances of removing all affected lymph nodes. Data regarding the optimal extent of lymphadenectomy for esophageal adenocarcinoma in the era of neoadjuvant treatment are contradictory. On the one hand, there is evidence in favor of extensive lymphadenectomy conferring survival benefit; a meta-analysis including 26 studies supported the benefit of an increased lymph node yield from esophagectomy for overall and disease-free survival, even after neoadjuvant treatment. 4 A study in Asian subjects evaluated the impact of dissecting specific lymph node stations on survival and concluded that paratracheal lymphadenectomy has therapeutic value in patients with esophageal carcinoma; 12 however, this study could not be included in the current systematic review since it mainly included patients with squamous cell carcinoma and did not separately report the outcomes of patients with adenocarcinoma. A significant overall survival benefit was not found in selected patients when comparing transhiatal resection with a limited lymphadenectomy versus transthoracic esophagectomy with an extended lymphadenectomy. 32 A trend towards improved long-term survival at 5 years using the extended transthoracic approach compared with the limited transhiatal approach was found in another study. 33 Nevertheless, data on dissecting specific lymph node stations are lacking and most surgeons do not routinely dissect the upper mediastinal lymph nodes for distal esophageal and GEJ tumors, as the balance between potential oncological merits and risk of recurrent laryngeal nerve injury is not clear. This systematic review found that the incidence of upper mediastinal lymph node metastasis is as high as 10% in these patients, suggesting that involvement of the upper mediastinal nodes may be present, even in distal adenocarcinomas. Ideally, preoperative imaging should identify these lymph nodes, allowing surgeons to select patients who are most eligible for upper mediastinal lymphadenectomy. However, the reliability of clinical lymph node staging remains poor as the sensitivity of positron emission tomography/computed tomography (PET/CT) scanning for the detection of esophageal lymph node metastases is only 34-74%. 34 Although the sensitivity of endoscopic ultrasound (EUS) is somewhat higher for lymph nodes adjacent to the esophagus (75-84%), this diagnostic modality has a considerably lower specificity (65-75%) and is less accurate in identifying abnormal lymph nodes located more distantly. 35,36 Although a combination of PET/CT and EUS CT chemotherapy, RT radiotherapy, CRT chemoradiotherapy, NR not reported is used in some centers for preoperative lymph node staging, diagnostic accuracy is generally considered to be insufficient to perform a targeted lymph node dissection. This could be an argument for esophagectomy with routine removal of all mediastinal lymph nodes at risk, including those located in the upper mediastinum. On the other hand, increased understanding of metastatic patterns based on tumor characteristics might carry potential to guide decision making. In a nationwide Japanese study that was included in this systematic review, it was shown that lymph node metastases in JES stations 105, 106L, and 106R (which correspond to AJCC stations 2 and 4) occurred in 0%, 0%, and 3%, respectively, of patients with GEJ adenocarcinoma or squamous cell carcinoma that extended \4 cm in the esophagus. 20 Such insights, which are expected to be amplified within the next years by initiatives such as the TIGER trial (NCT03222895), 37 might shed light on indications for performing a less aggressive lymph node dissection in particular subgroups in the future. Although upper mediastinal lymph node metastases are observed in up to 10% of patients with distal esophageal adenocarcinoma, the impact of upper mediastinal lymphadenectomy on survival remains unclear based on this systematic review. Hence, high-quality evidence on this topic is lacking and more research is needed to evaluate the clinical value of upper mediastinal lymphadenectomy. Even if future studies are able to identify an association between upper mediastinal lymphadenectomy and survival after esophagectomy for adenocarcinoma, investigators should be aware of the phenomenon referred to as stage migration, which may be the result of harvesting more lymph nodes. 38 Future studies should thus strive to improve our understanding of the role of upper mediastinal lymphadenectomy on staging and survival in patients undergoing esophagectomy for adenocarcinoma.
With potential survival benefits on one side of the scale, the risk of complications should be weighed up against the other. As paratracheal stations 2 and 4 are adjacent to the RLNs, dissection of this region might increase the risk of iatrogenic RLN injury, with reported incidences in experienced centers ranging from 5 to 26%. [39][40][41] RLN injury can result in hoarseness and difficulty swallowing, which reduces quality of life and increases the risk of aspiration   44 The strength of this study is its clinical relevance, as it represents the first systematic review that aims to provide insight into the value of upper mediastinal lymphadenectomy for patients undergoing esophagectomy for adenocarcinoma with and without neoadjuvant therapy. One of the limitations of the study is the heterogeneity in neoadjuvant treatment and patient selection. Another limitation seems to be the wide variation in definitions that were used for an upper mediastinal lymphadenectomy in the existing literature, which in fact highlights the low level of evidence on the topic at this moment. In the absence of studies that are suitable for pooling of data in order to generate a reasonable level of evidence, esophageal surgeons are currently operating mainly based on their personal convictions or those of their former teachers. To increase the transparency and comparability of studies on this topic, future studies should report the exact lymph node stations that are part of the lymphadenectomy. Furthermore, the authors suggest the dissection of stations 2 and 4 in both sides as the only definition of a full upper mediastinal lymphadenectomy as part of a total mediastinal lymphadenectomy. By increasing the number of comparable studies, it will be possible to establish evidence-based recommendations on this topic.

CONCLUSION
This study demonstrates that the incidence of upper mediastinal lymph node metastases is up to 10% in patients undergoing transthoracic esophagectomy for distal esophageal or GEJ adenocarcinoma. The diagnostic accuracy of current imaging techniques is insufficient for the detection of upper mediastinal lymph node metastases, and the effect of neoadjuvant treatment on node positivity in this region is unclear. Although the impact on overall survival is not clear, if morbidity could be reduced, surgeons could consider standard upper mediastinal lymphadenectomy for distal esophageal adenocarcinoma to potentially improve the oncological outcome. 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/.