Prognostic impact of thoracic duct lymph node metastasis in esophageal squamous cell carcinoma

Abstract Aim We have previously reported the existence of lymph nodes surrounding the thoracic duct ( TDLN) and transthoracic esophagectomy (TTE) with thoracic duct (TD) resection increased the number of lymph nodes (LNs) retrieved. The current study aims to evaluate the prognostic impact of TDLN metastasis in esophageal cancer patients subdivided by its location and comparing the patients’ survival with those with extra‐regional LN metastasis. Methods Patients who underwent TTE with TD resection for esophageal squamous cell carcinoma (ESCC) were reviewed. Patients were classified into those with or without TDLN metastasis, and clinicopathological factors were compared between groups. TDLN was further divided into TDLN‐Ut/Mt/Lt based on the location in the mediastinum. The relapse‐free survival (RFS) and overall survival (OS) were compared between groups. Results Of 232 patients, TDLN metastasis was observed in 17 (7%). RFS and OS were significantly worse in the TDLN metastasis group. TDLN metastasis was shown to be an independent prognostic factor for RFS and OS in the multivariate analysis. The negative prognostic impact of TDLN metastasis was evident in TDLN‐Mt/Lt. The RFS and OS of patients with TDLN metastasis were almost identical to those with positive LN metastasis in extra‐regional LNs. Conclusion TDLN metastasis was proven to be a strong prognostic indicator. Although the TDLN has been included in the classification of regional LN in the current staging systems, it could be independently classified from the current regional LNs. Given that neoadjuvant therapy has been a standard, we might need to introduce adjuvant therapy when TDLN metastasis is observed.


| INTRODUC TI ON
Esophageal cancer that can metastasize widely from an early stage has been regarded as a devastating disease. 1,2 While multidisciplinary treatment advances have improved the survival, 3 esophageal cancer is still the eighth leading cause of cancer-related deaths in 2018. 4 Trans-thoracic esophagectomy (TTE) with lymph node (LN) dissection has been recognized as one of the standard treatments for esophageal cancer, especially in esophageal squamous cell carcinoma (ESCC), where LN metastasis spreads from the cervical to abdominal nodes. 5 When sentinel LNs were evaluated in ESCC at the mid thoracic esophagus, cervical and abdominal LNs were identified as sentinel LNs in more than 10% of patients. 6 Therefore, extensive LN dissection such as three-field lymphadenectomy is recognized as one of the standard treatments for ESCC. 7,8 The fact that the number of LNs retrieved was previously shown to be a prognostic factor 9,10 further supported the concept in which radical lymphadenectomy was a vital procedure for the treatment of patients with esophageal cancer.
The thoracic duct (TD) is the main lymphatic root, which originates from the cistern of chyle and ascends along the thoracic descending aorta, flowing at a left venous angle. The pros and cons of TD resection in esophagectomy have been debatable. We have recently shown that minimally invasive TTE with extensive LN dissection along with TD dissection improved survival especially in cStage I esophageal cancer; that was the first study to suggest the survival benefit of TD resection in TTE. 11 Conversely, two studies reported showing that TD resection could increase the postoperative complication and did not improve the survival. 12,13 When the oncological outcome of TD resection is investigated, attention needs to be given to the LNs surrounding the TD (TDLN). Previously, Udagawa et al 14 showed that the TDLN were the nodes in the adipose tissue surrounding the TD and running between the thoracic esophagus and the descending aorta. They indicated that TD resection with dissection of the TDLN should be performed routinely. Following this report, we also demonstrated the existence of TDLN. In Ivor Lewis esophagectomy, which is conducted for adenocarcinoma, Schurink reported TDLN in a cadaver study. 15 However, there are no studies investigating the prognosis of patients who had metastasis in the TDLN subdivided by its location.
In the present study, we expanded the cohorts of patients with ESCC whose metastatic status of TDLN was pathologically evaluated. The incidence rate of TDLN metastasis was reviewed, and the correlation between TDLN metastasis and survival was analyzed.

| Patients and treatment
As previously reported, since June 2013, the TDLN has been examined independently of paratracheal and aortic LNs in our institution. 16 All patients who underwent TTE with TD resection for esophageal malignancies at the Keio University Hospital between June 2013 and December 2019 were reviewed. We excluded patients with a cT4 tumor, adenocarcinomas, patients that underwent salvage esophagectomy after definitive chemoradiotherapy, and those with R2 resection. As it has been reported that the resection of the TD might influence postoperative fluid retention and liver function, 17,18 the TD was preserved in patients with liver cirrhosis, kidney, or heart diseases.
Clinical staging was performed with esophagoduodenoscopy, esophagography, and computed tomography, and the 8th edition of the TNM classification established by the Union for International Cancer Control (UICC) was used. 19 To define the regional and extra-regional LNs, we referred to the Japanese Classification of Esophageal Cancer 11th edition in which LN was categorized from N1 to N4 based on the location of the primary tumor. 20,21 Then, N 1-3 groups were defined as regional LNs in the current analysis. Extra-regional LNs included upper cervical nodes, common hepatic artery nodes, splenic artery nodes, and infradiaphragmatic nodes.
Based on the Japan Clinical Oncology Group (JCOG) 9907 study, 22 neoadjuvant chemotherapy, using cisplatin and 5-fluorouracil, was the standard treatment since 2007. A regimen consisting of three drugs (cisplatin, 5-fluorouracil, and docetaxel) administered three times every 3 weeks was considered for patients who had either border-line resectable disease or multiple LN metastases at diagnosis. 23 For those who were diagnosed as cT1N0 and underwent upfront surgery, adjuvant chemotherapy was provided when the patients were found to have LN metastasis in the resected specimen.
The present study was approved by the ethics committee of the Keio University School of Medicine.

| Surgical procedure, LN station numbers, and surgical outcomes
As a curative surgery, we performed TTE with right thoracotomy and gastric tube reconstruction in the posterior mediastinal route as a standard surgical procedure at our institution. We performed recon-

| Statistical analysis
We calculated the means and standard deviations, and identified differences using the Student's t-test. We identified differences between categories using the chi-square test or Fisher's exact test. We produced survival curves using the Kaplan-Meier survival method.
We compared two groups using a two-sided log-rank test. We performed all statistical analyses using IBM SPSS Statistics version 28 for Windows (Chicago, IL, USA), and differences were considered significant when P < .05.

| Clinicopathological factors of the patients and surgical outcomes between with and without TDLN metastasis
The characteristics of the patients are shown in Table 1. Of 232 patients, 84% were male and the primary tumor was located at the mid- In terms of the surgical approaches and short-term outcomes (Table S1), there was no significant difference in the amount of intraoperative bleeding, while operative time was significantly shorter in patients with TDLN metastasis. The percentages of thoracoscopic and robotic esophagectomy were significantly higher in the TDLN negative group. The incidence rate of postoperative anastomotic leakage (≥ CD III) and pneumonia (≥ CD II) was 10% and 14%, respectively, and there was no significant difference between patients with and without TDLN metastasis. There was a postoperative mortality in a patient without TDLN metastasis.

| Distribution of TDLN metastasis
The distribution of TDLN metastasis is described in Figure 1 Age (mean ± SD) 65. Following the Japanese Classification of Esophageal Cancer, we compared the survival of patients with TDLN metastasis to that of those who had metastasis in the regional and extra-regional LNs. Extra-regional LNs included upper cervical nodes, common hepatic artery nodes, splenic artery nodes, and infradiaphragmatic nodes.

F I G U R E 1
The distribution of metastasis in the thoracic duct lymph node. All patients who were diagnosed as positive metastasis in the lymph nodes around the thoracic duct were described along with the TNM staging The RFS and OS of patients with TDLN metastasis was almost identical to those with positive LN metastasis in extra-regional LNs ( Figure 5).
When the distribution of initial postoperative recurrence was evaluated (

| D ISCUSS I ON
The current study successfully validated our previous findings that there was the existence of LN metastasis around the TD in ESCC. 16 Subsequently, we further confirmed that TDLN metastasis occurred recurrence. This led us to the conclusion that the prognosis in patients with TDLN metastasis is extremely poor. Through the multivariate analysis, which took into account the tumor stage, we were convinced that the TDLN metastasis was an independent prognostic factor for RFS and OS. Consequently, TDLN metastasis was found to be a strong prognostic indicator in ESCC. Although more sample collection is required, this is the first study which has found a direct association between TDLN metastases for each location and dismal prognosis in esophageal cancer.
In contrast to the value of TDLN metastasis as a prognostic indicator, the efficacy of lymphadenectomy of TDLN with TD resection might not be remarkable. Based on the current study, which was the first study evaluating the prognosis of patients with TDLN metastasis for each location, the incidence rate of TDLN metastasis was relatively low, and its prognosis was proven to be extremely poor. In a previous study, to evaluate the prognostic impact for each LN station, the efficacy index, which takes into account the incidence rate and prognosis of the patients with metastasis in the lesion, was introduced. 25 When efficacy index is calculated for the TDLN, the index of TDLN could be low because of its low incidence rate and high mortality. Furthermore, the survival in patients with TDLN metastasis was almost identical to those with extra-regional LN metastasis that was recognized as distant metastasis ( Figure 5). Although the TDLN has been included in the regional LN classification, it could be independently classified from the current regional LNs and considered as a strong Consequently, those who received adjuvant therapy were compared with those without postoperative treatment. 26 The survival benefit TA B L E 2 Clinicopathological factors, TDLN metastasis, and survival: Uni-and multivariate analysis F I G U R E 5 The survival of thoracic duct lymph node metastasis compared to regional and extra-regional LNs metastasis. To define the regional and extra-regional LNs, we followed the Japanese Classification Esophageal Cancer 11th edition of adjuvant therapy was significantly confirmed by propensity score matching. A randomized control trial comparing perioperative chemotherapy with preoperative therapy demonstrated that perioperative chemotherapy, in which adjuvant chemotherapy was provided after surgery, showed a significantly longer overall survival for ESCC. 27 The efficacy of adding adjuvant chemotherapy, even after neoadjuvant treatment, was also confirmed in a phase-II trial from another cohort. 28  As always, several drawbacks to the resection of the TD need to be considered. TD ligation was reported to induce retroperitoneal fluid retention and lead to intravenous volume loss after surgery. 18,29 In terms of hepatic damage, Guler et al demonstrated that TD ligation had a negative effect on the liver in a canine model of peritonitis, which was induced by the exposure of liver to the endotoxin. 17 Aiko et al reported that TD resection affected the fluid balance and minimized the clinical benefit of enteral feeding after esophagectomy. 29 Furthermore, two papers from Japan reported a negative impact of TD resection in the postsurgical state. Yoshida et al described that TD resection would increase the incidence of pulmonary comorbidity. 12,13 Since the anticipated benefits need to outweigh the potential risks, prospective studies that validate these previous results are warranted.
The present study was limited by its retrospective nature.
However, consecutive patients who underwent TTE were reviewed and all patients with ESCC who underwent TTE between June 2013 and July 2020 were included in this analysis, leading to reduced selection bias. Regarding survival analysis, the follow-up is still immature with respect to the study population. Therefore, the cutoff was set at April 2018 because this study intended to include patients who were followed up for >1 year. As RFS was mainly used in the current analysis, the difference in survival was fairly evaluated.
In conclusion, TDLN metastasis was proven to be a strong prognostic indicator in ESCC. Although the TDLN has been included in the regional LN classification in both the current systems, it could be independently classified from the current regional LNs. Given that neoadjuvant therapy has been a standard, we might need to introduce the adjuvant therapy when TDLN metastasis is observed.
A prospective comparative trial is necessary for conclusions to be arrived at concerning the survival advantage of TD resection in ESCC.