The incidence of gastric adenocarcinoma in the East is much higher than in the West; the main etiology is Helicobacter pylori (H. pylori) infection. However, in Japan esophageal adenocarcinoma accounts for only 4.3% of all cases of esophageal cancer, which is a rather low proportion compared to the West.¹ Although the high prevalence of H. pylori infection has contributed to a decrease in the risk of esophageal adenocarcinoma in the East, decreased prevalence of H. pylori infection and increased incidence of obesity have gradually increased the incidence of esophagogastric junction (EGJ) cancer in recent years.² Treatment for EGJ cancer requires special attention to surgical technique, in particular to lymph node dissection. Although surgery is the most effective curative treatment for EGJ cancer, the proportion of R1 or R2 resections is comparatively high. Even after R0 resection, the recurrence rate is high. To improve the R0 resection rate and long-term outcomes, perioperative treatment has been attempted. In this section, we outline the Eastern perspective on the surgical approach and perioperative therapy for EGJ cancer.

The surgical approach for EGJ cancer is mainly based on tumor histology and location. Adenocarcinomas located around the EGJ are usually classified into three categories according to the Siewert system based on their epicenter.³ Mediastinal dissection via a right thoracotomy is usually used to treat Siewert type I adenocarcinomas based on the results of a Dutch randomized controlled trial (RCT).⁴ Lymph node dissection including the upper mediastinum is often performed in Japan because Siewert type I adenocarcinomas can sometimes metastasize to the upper mediastinal lymph nodes.^5,6

A Japanese RCT compared the left transthoracic and transhiatal approaches for mainly Siewert type II and III adenocarcinomas.^7,8 This study demonstrated that there was no survival benefit and higher morbidity associated with the transthoracic approach. The transthoracic approach was also associated with more weight loss, postoperative symptoms, and respiratory dysfunction compared to the transhiatal approach.⁹ Subgroup analysis showed no survival benefit for Siewert type II patients who underwent the transthoracic approach. The transhiatal approach was associated with better survival than the transthoracic approach for Siewert type III patients. Based on the results of this trial, transhiatal resection of the distal esophagus with lymph node dissection of the lower mediastinum is recommended for Siewert type II or III adenocarcinoma. However, we should keep in mind that this trial excluded patients with esophageal invasion over 3 cm because it is difficult to obtain sufficient margins using the transhiatal approach if the tumor has invaded the distal esophagus 3 cm beyond the EGJ.

A Japanese multicenter retrospective study of 315 patients with pT2–T4 Siewert type II adenocarcinoma reported that the overall rate of metastasis or recurrence in the upper and middle mediastinal lymph nodes was 3.8% and 7.0%, respectively.¹⁰ The 5-year overall survival rate in patients with metastasis in the upper or middle mediastinal lymph nodes reached 16.7% after R0 resection. The metastasis rate to the upper and middle mediastinal nodes was significantly higher when the distance from the EGJ to the proximal edge of primary tumor was over 3 cm. Therefore, the right transthoracic approach may be better when there is more than 3 cm of esophageal invasion in terms of both ensuring adequate proximal margins and thorough mediastinal lymph node dissection.

For squamous cell carcinoma (SCC) located around the EGJ, the transthoracic approach with thorough mediastinal lymph node dissection is usually performed in Japan because SCC around the EGJ can easily spread to the upper mediastinum. The incidence of subcarinal lymph node metastasis was reported to be 9.8%.¹¹

Lymphatic flow from the EGJ is mainly directed toward the abdomen rather than the mediastinum. Lymph nodes around the esophageal hiatus (Nos. 19, 20) should be dissected for any type of EGJ cancer. In addition to these nodes, Siewert type I adenocarcinoma frequently metastasizes to the pericardial and lesser curvature nodes (Nos. 1, 2, 3, 7).^12,13 Regarding Siewert type II adenocarcinoma, which represents true adenocarcinoma of the EGJ, some Japanese studies have reported on the incidence of metastasis to the regional lymph nodes of the stomach (Table 93-1).^13–15 For this type of tumor, the pericardial and lesser curvature nodes as well as the suprapancreatic nodes (Nos. 8a, 9, 11p) have a high incidence of metastasis. Thus, these nodes should be dissected in Siewert type II adenocarcinoma. Proximal gastrectomy in lieu of total gastrectomy is acceptable because other perigastric (Nos. 4sa, 4sb, 4d, 5, 6) and splenic hilar nodes (No. 10) have an extremely low incidence of metastasis. Regarding SCC located around the EGJ, the pattern of lymph node metastasis is similar to that for Siewert type II adenocarcinoma.^15,16 Siewert type III adenocarcinomas frequently metastasize to all of the regional lymph nodes of the stomach, similar to gastric adenocarcinoma. Even in the lymph nodes along the greater curvature (Nos. 4sa, 4sb, 4d), the frequency of metastasis was reported to be approximately 10%.¹³ Thus, total gastrectomy would be suitable for Siewert type III adenocarcinoma. Although splenic hilar nodes (No. 10) have a similar metastasis rate, whether splenectomy is indicated remains controversial.

Another route of lymphatic flow is to the paraaortic area (No. 16a2lat) via the left inferior phrenic artery.^17,18 The precise incidence of metastasis to this area has not been revealed in previous studies because the lymph nodes in this area have not been dissected routinely, even in Japan. However, some studies have reported that metastasis to this area was observed in approximately 15% of EGJ cancer patients, representing the most frequent site of nodal recurrence.^13–16 Although a Japanese RCT demonstrated no survival benefit of prophylactic paraaortic lymph node dissection for patients with gastric adenocarcinoma,¹⁹  this trial did not include any patients with EGJ cancer. Mine et al²⁰ reported that prophylactic paraaortic lymph node dissection around the left renal vein was a significant prognostic factor in patients with Siewert type II adenocarcinoma. Considering the high incidence of metastasis to the paraaortic lymph nodes and the result of the retrospective study as mentioned above, dissection of this area may have survival benefit for EGJ cancer. An ongoing nationwide study with a larger cohort will evaluate the possibility of improving survival with prophylactic paraaortic lymph node dissection.