Accurate preoperative staging of adenocarcinoma of esophagogastric junction (AEG) is difficult, especially in lymph node staging. The postoperative treatment plan can be determined according to accurate pathological diagnosis in patients with high recurrence risk. Postoperative adjuvant chemotherapy aims to control potential micrometastasis after surgery and reduce the risk of local recurrence and distant metastasis.
We found that the main sites of local recurrence of AEG were anastomosis and mediastinum in the thoracic cavity (8.1% vs 7.6%; 5.6% vs 4.9%, in the surgery and surgery plus chemotherapy groups, respectively). The recurrence rate in lymph nodes around the celiac axis was 9.7% and 9.2% in the surgery and surgery plus chemotherapy groups, respectively, suggesting that Siewert II AEG may metastasize both downward and upward, showing bidirectional features. Multicenter studies have shown that proximal gastrectomy, lower esophageal resection and local lymph node resection are the minimum requirements for surgical treatment of AEG. Expanding esophagectomy and enlarging the scope of lymph node dissection does not improve local recurrence . Other studies  have shown that the overall survival is not significantly related to the dissected mediastinal lymph nodes. In our study, there was no difference in the recurrence rate between patients treated with transthoracic or abdominal surgery. However, the local recurrence rate in the surgery group was higher than that in the surgery plus chemotherapy group (23.5% vs 21.8%). Although the Kaplan-Meier curve showed no significant difference in local recurrence between the two groups, the survival curve of the surgery plus chemotherapy group was above that of the surgery group, suggesting that the surgery plus chemotherapy had a tendency to reduce local recurrence,and the effect was not significant, which might be related to the surgical injury and blockage of blood vessels and lymph nodes in the surgical area, which to some extent affect inhibition of potential micrometastasis in the surgical area by chemotherapeutic drugs. Currently, neoadjuvant chemotherapy and neoadjuvant radiochemotherapy are the hotspots in the field of cancer research. There is also evidence [16,17,18] to indicate that neoadjuvant chemotherapy or neoadjuvant radiochemotherapy can reduce local recurrence of AEG and improve prognosis, but there is still controversy about this treatment . Due to the fear of cancer  and concern about complications associated with neoadjuvant therapy, neoadjuvant chemotherapy in our medical center is not smoothly administered. Only 2.4% of our patients received neoadjuvant chemotherapy; therefore, adjuvant treatment after surgery is still valuable in patients.
Patients with AEG have a high risk of hematogenous dissemination after surgery. Liver metastasis can occur as early as 32 days postoperatively. Distant metastasis sites vary; mainly the liver and lungs, with multiple subcutaneous, ovarian, and small intestinal metastases being less common. Compared with surgery alone, surgery plus chemotherapy reduced hematogenous dissemination (51.2% vs 42.0%). Distant pleural and peritoneal metastases occurred in some patients via implantation or hematogenous dissemination of cancer cells. But whether implantation of cancer cells occurred during surgery or was caused by long-term hematogenous dissemination is unclear. In univariate Cox analysis, it was found that the ratio of pleural and peritoneal metastases in the surgery plus chemotherapy group was lower than that in the surgery group (3.8% vs 4.0%; 11.4% vs 14.6%) but these differences were not significant (P= 0.563, 0.099).
In our study, within 730 days after surgery, most patients had recurrence (62.0%, 190/306); 72.3% (89/123) in the surgery group and 55.1% (101/183) in the surgery plus chemotherapy group, indicating that advanced AEG has high potential of early recurrence. Previous studies have shown that survival rates of patients with tumor recurrence are low, especially in those with earlier recurrence [21,22]. Multivariate Cox analysis showed that patients with higher pN stage were more prone to recurrence. Surgery plus chemotherapy may be beneficial to RFS, but the recurrence rate was still high. Therefore, for AEG patients, we should focus on pN staging, modify the follow-up protocol within 2 years after surgery, and increase the frequency of follow-up appropriately, so as to find timely evidence of recurrence, and formulate subsequent treatment strategies. At the same time, we found an interesting phenomenon in that the age of patients in the surgery plus chemotherapy group was lower than that in the surgery group (median age 62 years vs 65 years, P<0.001). Univariate Cox analysis showed that older age was associated with recurrence after surgery. This indicated that the younger patients had higher compliance with subsequent treatment, thus improving the prognosis.
This retrospective study had some limitations: Although there were inclusion criteria, there may have been selection bias; for example, patients who were lost to follow-up may have had no recurrence. Unfortunately, we were unable to obtain sufficient data on follow-up treatment after relapse. This study was conducted in a single center, which helped us to understand the recurrence mode of advanced AEG, and improve subsequent treatment strategies. Strengthening health education and improving patient compliance, and increasing the surgical area (including lower mediastinum and upper abdomen) for radiotherapy may also help to reduce local recurrence. However, the ARTIST (Adjuvant Chemoradiation Therapy in Stomach Cancer) trial  found that additional postoperative radiotherapy had the same benefit as chemotherapy alone in preventing recurrence of gastric cancer. Therefore, while focusing on randomized clinical trials[24,25], we also need to develop and optimize treatment strategies that accord with the medical conditions of the local area.We found that stage III AEG mainly metastasizes to distant sites, predominantly through hematogenous transmission. We recommend increasing the number of follow-up visits within 2 years after surgery, especially in the second year. Surgery plus chemotherapy can improve RFS and reduce distant metastasis, but they do not have a beneficial role in controlling local recurrence.