Abstract
Background/Aim: Clinical staging in the eighth edition of the Union for International Cancer Control TNM classification (TNM8) is reported to predict the prognosis of patients with gastric cancer. However, there have been no reports on using the TNM8 for prognostic stratification of patients with adenocarcinoma of the esophagogastric junction (AEG). This study aimed to investigate whether it was possible to stratify the prognosis of patients who underwent curative surgery for Siewert type II/III AEG according to the TNM8 clinical stage (cStage). Patients and Methods: This study included patients with Siewert type II/III AEG who underwent curative surgery between 2000 and 2019 at Kanagawa Cancer Center. Those who received neoadjuvant chemotherapy were excluded. We investigated the survival of patients with AEG of each TNM8 cStage. Results: This study included 138 patients, among whom 102 (74%) had Siewert type II and 36 (26%) had Siewert type III AEG. A total of 50, 38, 43, and seven patients were classified with cStage I, II, III, and IV, respectively. The median duration of follow-up of the survivors was 54.7 months. The 5-year overall survival rate of the entire cohort was 65.8%, whereas for patients with cStage I, II, III and IV was 81.6%, 69.0%, 54.3% and 14.3%, respectively. The hazard ratio with reference to cStage I was 1.83, 3.07, and 8.13 for cStage I, III, and IV, respectively, increasing in a stepwise manner. Conclusion: TNM8 Clinical staging is able to stratify the prognosis of patients with Siewert type II/III AEG.
Adenocarcinoma of the esophagogastric junction (AEG) has become a major topic among surgeons and oncologists because the incidence of AEG has gradually increased in Western countries and in Eastern Asia, including Japan (1-11). Compared with common gastric cancer (GC), most cases of AEG are found at an advanced stage and the prognosis is generally poor (12, 13), However, there are also reports that planned perioperative chemotherapy (14, 15) and extensive lymphadenectomy (16) for AEG improve outcomes. Therefore, accurately evaluating the preoperative clinical stage (cStage) and the associated prognosis is crucial to making appropriate treatment decisions, especially for planning surgery, which remains the foundation for treatment.
In 2017, the clinical staging system of AEG was first defined in the eighth edition of the Union for International Cancer Control TNM classification (TNM8), a global classification system (17). In Japan, the 15th edition of the Japanese Classification of Gastric Carcinoma (18), in which the clinical and pathological classifications were unified with those of the TNM8, was published in the same year.
In the TNM8, it was reported that the cStage is able to predict the prognosis of patients with GC (19), however, patients with AEG were excluded from that study. Thus, whether the TNM8 cStage can stratify the prognosis of patients with AEG is unclear. This study aimed to investigate whether it is possible to stratify the prognosis of patients with AEG according to the TNM8 cStage.
Patients and Methods
Patients. This study included 200 patients who underwent surgery for AEG at Kanagawa Cancer Center from 2000 to 2019 and was approved by the Institutional Review Board of the Kanagawa Cancer Center (2022 Epidemiological Study-23). The inclusion criteria were patients with histologically proven adenocarcinoma, those with AEG type II or III according to the Siewert classification, and those who underwent curative surgery with D1 or greater radical lymphadenectomy. The exclusion criteria were patients with a history of cancer, including GC or AEG, in the last 5 years before surgery, those with Siewert type I AEG, and those who received preoperative therapy with neoadjuvant chemotherapy and radiation. The Siewert classification was strictly determined from the findings of preoperative endoscopy, contrast radiography, macroscopic images of the resected specimens, and pathological and endoscopic findings. The surgical approach and extent of lymph node dissection were carefully determined considering the tumor location and preoperative staging, including tumor depth of invasion, nodal metastasis, distant metastasis, and comorbidities. Finally, 138 patients were assessed in this study (Figure 1). Table I shows the clinicopathological characteristics of the patients.
Selection of the patient cohort of the present study. Among 200 patients who underwent surgery for AEG at Kanagawa Cancer Center from 2000 to 2019, 19 with Siewert type I adenocarcinoma of the esophagogastric junction and 43 patients who received neoadjuvant chemotherapy were excluded. The remaining 138 patients were assessed in this study.
Clinicopathological characteristics of study patients.
cStage of AEG. The AEG cStage was determined according to the TNM8. Clinical and pathological stage groupings are shown in Table II and Table III. cT, cN, and cM were determined using gastro-esophageal endoscopy, contrast-enhanced multirow detector computed tomography (CT), and fluorodeoxyglucose positron-emission tomography–CT by a team comprised of endoscopists, oncologists, and surgeons. We performed a prospective validation study to diagnose cT and cN staging based on multirow detector CT findings from 2006 to 2008, which was previously reported by our Institution (20), and we determined cT and cN staging based on that study to diagnose all patients. The evaluation criteria for cT staging were as follows: Clinical cT1 tumors: those that could not be found on imaging; T2/3 tumors: those with focal or diffuse thickening of the gastric wall and a smooth outer border of the wall; cT4a tumors: transmural tumors with obvious blurring of at least one-third of the tumor extent; and cT4b tumors: those with obliteration of the fat plane between the gastric tumor and the adjacent organ. For N staging, lymph nodes with short-axis diameters larger than 8 mm were considered positive for metastasis.
Clinical stage according to the eighth edition of the Union for International Cancer Control TNM classification (17).
Pathological stage according to the eighth edition of the Union for International Cancer Control TNM classification (17).
Surgery. Surgery was performed via the trans-hiatal esophageal approach or right transthoracic approach. In principle, we selected the trans-hiatal abdominal approach for patients with adenocarcinoma with an esophageal involvement of <3.0 cm and employed the right transthoracic approach in patients with an esophageal involvement of ≥3.0 cm. In patients who underwent surgery via the right transthoracic approach, we dissected lymph node numbers 1, 2, 3a, 7, 8a, 9, 11p, 11d, 16a2lat, 19, 20, 105, 106recR, 106recL, 107, 108, 109R, 109L, 110 111 and 112. In patients with cStage II/III AEG who underwent surgery via the transhiatal abdominal approach, we performed lower esophagectomy and total gastrectomy with dissection of lymph node numbers 1, 2, 3, 7, 8a, 9, 11p, 11d, 16a2lat, 19, 20, 110, 111 and 112. In patients with cStage I AEG, we performed a lower esophagectomy and proximal gastrectomy with dissection of lymph node numbers 1, 2, 3a, 7, 8a, 9, 19, 20, 110, 111 and 112 via the abdominal esophageal hiatal approach.
The endpoint of this study was overall survival of patients with AEG according to the TNM8 cStage.
Statistical analyses. Data are presented as the median and range unless otherwise stated. All statistical analyses were performed using EZR version 1.37 (Jichi University, Tochigi, Japan), and statistical significance in all tests was set at p<0.05 and a 95% confidence interval. The chi-squared test was used to compare patient populations. The survival curves were calculated using the Kaplan-Meier method and compared using the log-rank test. The hazard ratio (HR) of each cStage referenced to stage I was calculated using the Cox proportional hazards regression model.
Results
Almost 75% of the patients had Siewert type II tumors, and 25% had type III tumors. Most patients had advanced tumors; >70% of the patients underwent lower esophagectomy and total gastrectomy, while <10% underwent total esophagectomy via the right transthoracic approach. Almost 50% of the patients received postoperative adjuvant chemotherapy after surgery.
Distribution of TNM8 stages. Table IV shows the clinical T, N, M, and stage grouping based on the TNM8. Approximately 70% of patients had advanced primary tumors, and 40% had nodal metastases according to the TNM8 clinical evaluation criteria. Table V shows the grouping of pathological stages. Although all patients were diagnosed as cM0 before surgery, 11 were diagnosed with pathological stage IV disease after surgery. Six patients had positive peritoneal lavage cytology, two had small levels of peritoneal dissemination near the primary tumor, two had distant lymph node metastasis, and one had liver metastasis.
Distribution of patients with adenocarcinoma of the esophagogastric junction by clinical stage according to the eighth edition of the Union for International Cancer Control TNM classification (17).
Distribution of patients with adenocarcinoma of the esophagogastric junction by pathological stage according to the eighth edition of the Union for International Cancer Control TNM classification (17).
Survival outcomes. Figure 2A shows the survival curves of patients stratified by cStage. The median duration of follow-up of survivors was 54.7 months. The 5-year overall survival (OS) rate was 65.8%. In clinical staging, the 5-year OS rates were 81.6% for cStage I, 69.0% for cStage II, 54.3% for cStage III and 14.3% for cStage IV (p<0.01). In the pathological (p) staging, 81.9%, 81.6%, 52.7% and 0% were stage I, II, III and IV, respectively. Figure 2B shows the survival curves of patients stratified by cT. The 5-year OS rate was approximately 80% in cT1, cT2 and cT3, and was not statistically different between cT1 and cT2 (p=0.48), cT2 and cT3 (p=0.25), and cT3 and cT4a, (p=0.29). The 5-year OS rate was 51.0% in cT4a and 14.3% in cT4b, which was a significant difference (p=0.032). Figure 2C shows the survival curves of patients stratified by cN. The 5-year OS rate was 74.3% for cN0 and 55.5% for cN1-3, which was a significant difference (p=0.016).
Overall survival (OS) curves of patients with adenocarcinoma of the esophagogastric junction. A: OS according to clinical stage (cStage). The median follow-up of the survivors was 58.3 months. The 5-year OS rates were 81.6%, 68.0%, 54.3%, and 14.3% for cStages I, II, III, and IV, respectively. B: OS stratified by cT. The 5-year OS rates were 83.3%, 83.0%, 78.8%, 51.0% and 14.3% in cT1, cT2, cT3, cT4a and cT4b, respectively, and were not significantly different between cT1 and cT2 (p=0.48), cT2 and cT3 (p=0.25), and cT3 and cT4a (p=0.29). The 5-year OS rate was significantly lower for the cT4b group than the cT4a group (p=0.032). C: The 5-year OS rates stratified by cN were 74.3% in cN0 and 55.5% in cN1-3, which were significantly different (p=0.016).
Risk analysis. We calculated the HR for each cStage (Table VI). The HRs for cStages II, III and IV were 1.83, 3.07, and 8.13, respectively; the HR increased in a stepwise manner with cStage.
Hazard ratio (HR) for overall survival by clinical stage (cStage) according to the eighth edition of the Union for International Cancer Control TNM classification (TNM8) (17) with reference to cStage I.
Concordance rates between cT and pT, and cN and pN. Table VII and Table VIII show the diagnostic concordance for cT and cN. The concordance rates were 83.3% for cT1, 36.8% for cT2, 46.7% for cT3, 58.2% for cT4a, and 85.7% for cT4b (Table VII). The rates for cN were 64.1% for cN0, 24.2% for cN1, 30.0% for cN2, and 85.7% for cN3 (Table VIII).
Concordance rate between clinical (cT) and pathological (pT) T stage by the eighth edition of the Union for International Cancer Control TNM classification (17).
Concordance rate between clinical (cN) and pathological (pN) N stage by the eighth edition of the Union for International Cancer Control TNM classification (17).
Discussion
In this study, we examined the prognosis of patients with Siewert types II and III AEG and found that the TNM8 clinical staging system appropriately stratified patients according to their outcomes. The OS rate according to the clinical staging was similar to that of the historical control in common GC.
In this study, serosal invasion was evaluated on the basis of the procedure and criteria described by Hasegawa et al. (20). Although the distributions of cT and pT were similar, the concordance rates were not excellent. In this examination, overdiagnosis was observed rather than underdiagnosis. Bando et al. reported an excellent prognostic impact of cStage proposed in the TNM8 (19); however, the concordance rates were 25.1% for cT2, 36.1% for cT3, and 57.6% for cT4, which is similar to our study, in which the concordance rates between the cT and pT stages were low.
The distributions of cN and pN negativity and positivity were similar (Table VIII). However, the accuracy, sensitivity, and specificity were 64.1%, 86.2%, and 65.0%, respectively. In this examination, underdiagnosis was more common than correct diagnosis. This is probably due to the insufficient accuracy in counting the number of metastatic lymph nodes on CT. In our study, the concordance rates between cN and pN were not excellent (Table VIII). Improving the accuracy of cN may be required. A cut-off value of 8 mm for the short-axis diameter is commonly used for the clinical diagnosis of metastasis. However, lymph node assessment by size alone has its limitations. The low sensitivity and specificity in this study were similar to those in previous studies (21, 22).
The TNM8 cStage appropriately stratified patients with Siewert types II and III cancer in our cohort in terms of their OS. A stepwise increase in HR was observed in this study. Moreover, the survival rate in pStage III was higher than that in cStage III, which may be because those underestimated cStage III were correctly assigned to pStage IV. More patients had pStage IV disease than had cStage IV disease, which may have been affected by the positive peritoneal cytology, which was difficult to diagnose preoperatively when ascites was not observed on the CT scan.
The OS for each cStage in this study was similar to that of historical controls. As TNM8 cStage and pStage categories are different, a direct comparison of stage-concordance rates is impossible. However, in this study, the OS by cStage was similar to that by pStage.
This study had several limitations. Firstly, it was a retrospective study conducted at a single institution. Secondly, the number of patients was relatively small. Thus, the present results need to be confirmed in a multicenter prospective study. Finally, the concordance rates for the clinical and pathological T and N factors were not high. Therefore, improving the accuracy of clinical T and N staging is necessary.
In conclusion, clinical staging by TNM8 appropriately stratified patients with Siewert type II/III AEG in terms of their prognosis. Clinical staging by TNM8 may be useful in deciding the treatment strategy for AEG.
Acknowledgements
This work was supported by JSPS KAKENHI Grant Number JP20K08997. The Authors would like to thank the patients, their families, and the staff at Kanagawa Cancer Center for their participation in this study.
Footnotes
Authors’ Contributions
TY and SH designed the study. The data collection and literature search were performed by HW, SH, and TO. The data analysis and interpretation were performed by HW and SH. The data interpretation was performed by all Authors. The article and figures were drafted by HW, SH, and TO. Finally, the article was revised and approved by all Authors. Thus, all Authors actively participated in this study.
Conflicts of Interest
The Authors have no actual or potential conflicts of interest to declare in relation to this study.
- Received March 17, 2023.
- Revision received April 9, 2023.
- Accepted April 13, 2023.
- Copyright © 2023 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.