In 2016, the Japan Gastroenterological Endoscopy Society and the Japanese Gastric Cancer Association (JGCA) jointly issued the guidelines for ESD and EMR of EGC, which divided indications for endoscopic resection into absolute and expanded indications[18]. The former has good long-term prognostic evidence, while the latter lacks reliable long-term prognostic results.

Tumor lesions with a risk of LNM less than 1% for which endoscopic resection is considered to have the same effect as radical surgery are classified as absolute indications for ESD therapy[6]. Based on the research results of 5265 patients with EGC by Gotoda et al[19], Japanese gastric cancer treatment guidelines 2014 (version 4) clearly stated non-ulcerated and differentiated intramucosal carcinoma with a tumor diameter ≤ 2 cm as the absolute indication for endoscopic treatment of EGC[20] (Table 1). This absolute indication range is the same as that mentioned in guidelines for ESD and EMR for EGC[18]. Based on the results of the JCOG0607 trial[21], Japanese gastric cancer treatment guidelines 2018 (5th edition) adjusted tumor diameter > 2 cm, differentiated intramucosal carcinoma without ulcer lesions and tumor diameter ≤ 3 cm, and differentiated intramucosal carcinoma with ulcer lesions as the absolute indications for ESD. On the other hand, undifferentiated intramucosal carcinoma without ulcer and with a diameter ≤ 2 cm was considered an expanded indication for ESD[22] (Table 1). This version of guidelines is the same as the Korean Practice Guideline for Gastric Cancer 2018 for the absolute and expanded indications for ESD in EGC[23]. In addition, the ongoing study of JCOG1009/1010, which explores the efficacy and safety of ESD in the treatment of undifferentiated cT1a gastric cancer, has completed a 5-year follow-up. The results of the study confirmed that the patients with undifferentiated intramucosal carcinoma with tumor diameter < 2 cm, without ulcers had a satisfactory prognosis after endoscopic treatment, and no LNM occurred[24,25]. It is believed that this indication is expected to be included in the absolute indication range of endoscopic resection, which will lay a theoretical foundation for further elaboration of ESD indications.

Due to the low incidence of EGC in European and American countries, endoscopic treatment still lacks relevant evidence-based medicine. Currently, research results of endoscopic therapy for EGC are mainly based on the data from related studies in Japan and South Korea. The European Society of Gastrointestinal Endoscopy guidelines and the National Comprehensive Cancer Network (NCCN) guidelines fully adopt the absolute indications for endoscopic treatment recommended by the JGCA for endoscopic resection[26-28]. However, whether the above ESD treatment indications are suitable for the Chinese population is yet to be validated in high-quality clinical trials. Therefore, some research centers in China are carrying out exploratory research on ESD indications for EGC, hoping to establish reasonable ESD indications that meet the characteristics of the Chinese population[29]. Meanwhile, a staging diagnosis scheme for EGC suitable for China’s national conditions was proposed to further achieve the purpose of precision treatment and improve the life quality and prognosis of patients.

The absolute indications for ESD for the treatment of EGC have been unanimously approved, but the application of expanded indications is still controversial. Among them, assessment of the risk of LNM is the key to determining the optimal therapy for patients with expanded indications. A study from Korea found that 17.6% of the patients who were assessed to meet the expanded indications for ESD before surgery were proved to be non-compliant with the ESD indications after surgery, while only 6.7% of patients who were assessed as absolute indications before surgery did not meet the indications[30]. It indicates that improving the accuracy of preoperative diagnosis of expanded indications is a prerequisite for the rational application of ESD in the treatment of EGC. Therefore, although expanded indications can benefit some patients with EGC, its exact efficacy is still being explored. A meta-analysis showed that en bloc resection rates (93.6% vs 97%, P < 0.0001) and radical resection rates (82.4% vs 94%, P < 0.0001) were significantly lower in patients eligible for the expanded indications than in those with absolute indications, but there was no statistically significant difference in long-term survival (P = 0.37)[31]. Another retrospective study from South Korea used propensity score matching to analyze 522 patients who were eligible for expanded indications and underwent surgery or endoscopic treatment[32]. The study found that the overall and tumor-specific survival rates were not statistically different between the two groups, but the 5-year relapse-free survival rate in the surgery group was better than that in the endoscopic group (96.7% vs 92.7%, P < 0.001)[32]. Further comparison of patient recurrence patterns showed that there was no significant statistical difference in LNM rate and distant metastasis rate between the two groups, but the metachronous metastasis rate in the endoscopic group was higher than that in the surgical group[32]. This result indicates that the recurrence pattern of patients with expanded indications for endoscopic treatment is mainly local recurrence. Therefore, endoscopic treatment may be a good option for patients with expanded indications under the condition of ensuring sufficient surgical margins and regular postoperative re-examination.

The JCOG0607 trial, which is conducted in Japan, suggested that ESD treatment is safe and effective for EGC patients with expanded indications. A total of 470 patients eligible for expanded indications for ESD were included in the study. The results showed that the en bloc rate of ESD was 99.1%, the curative resection rate was 67%, and the delayed bleeding and perforation rates were 8.5% and 2.6%, respectively. Of these, 86.8% of patients with NCR received surgical treatment, and the 5-year survival rate of all patients was 97%[21]. Based on the results of this study, it was confirmed that ESD was reasonable and safe in the treatment of patients with EGC who partially met the expanded indications. On this basis, the endoscopic treatment indications of the Japanese gastric cancer treatment guidelines 2018 (5^th edition) have been revised to make the guidelines more in line with the needs of clinical treatment. Therefore, the current expanded indications for endoscopic treatment mainly include undifferentiated intramucosal carcinoma with tumor diameter ≤ 2 cm and without ulcer lesions and differentiated submucosal carcinoma with tumor diameter < 3 cm and invasion depth <500 µm[22]. Recently, with the publication of JCOG1009/1010 results, the application scope of existing ESD indications will be further expanded, benefiting more patients with EGC[24] (Table 1).

Although the above research results support the application of ESD expanded indications, some skeptical studies pointed out that compared with patients with absolute indications, patients with expanded indications had a higher rate of LNM, especially for undifferentiated intramucosal carcinoma with a diameter ≤ 2 cm [25/972 (2.6%), reference range = 6.79, P = 0.004] and differentiated submucosal carcinoma with a diameter < 3 cm [8/315 (2.5%), reference range = 6.30, P = 0.004][33]. Therefore, the current debate on expanded indications mainly focuses on undifferentiated carcinoma and submucosal infiltrating carcinoma. These two types of lesions seem to have a higher risk of LNM, which may not only increase the rate of NCR but also pose a challenge to preoperative evaluation of lesions. Therefore, we suggest that endoscopic therapy for patients with expanded indications should be selectively carried out by experienced centers in the context of clinical trials.

Curative resection refers to the complete resection of the lesions with negative margins and no vascular and lymphatic infiltration which meet the absolute and expanded indications. Complete resection is an important condition for curative resection and complete reconstruction after segmental resection of the lesions can also be considered as meeting the criteria for curative resection. According to the Japanese gastric cancer treatment guidelines 2014 (version 4)[20], complete resection, tumor diameter ≤ 2 cm, differentiated intramucosal carcinoma without ulceration, negative horizontal and vertical margins, and no lymph node or vascular infiltration are required for absolute indications. For expanded indications, one of the following four requirements is required: (1) Tumor diameter > 2 cm, differentiated intramucosal carcinoma without ulcer; (2) Tumor diameter ≤ 3 cm, differentiated intramucosal carcinoma with ulcer; (3) Tumor diameter ≤ 2 cm, undifferentiated intramucosal carcinoma without ulcer; and (4) Tumor diameter ≤ 3 cm, differentiated submucosal carcinoma with invasion depth < 500 µm. In addition, the horizontal and vertical resection margins should be negative without lymphatic and vascular infiltration (Table 2). Curative resection of EGC is the ultimate goal of endoscopic therapy and the key is that clinicians need to have a full grasp of the indications for endoscopic therapy. However, postoperative pathology confirmed that part of EGC did not reach the standard of curative resection after endoscopic treatment. Cho et al[34] analyzed the literature on the efficacy of ESD in the treatment of EGC in Eastern and Western countries in recent years and found that the en bloc rate was 92%-97% and curative resection rate was 73.6%, which indicated that NCR still had a certain proportion in postoperative pathological evaluation. Therefore, the Japanese gastric cancer treatment guidelines 2018 (5^th edition) updated the expression of ‘curative/non-curative resection’ in the evaluation of ESD radical resection to ‘endoscopic curability (eCura)’[22]. In these guidelines, curative resection, expanded curative resection, and NCR were changed to eCura A, eCura B, and eCura C, respectively (Table 2). Studies have found that patients with curative resection still have a potential recurrence risk after surgery with a local recurrence rate of 0.13%-1.3%[35], an incidence of simultaneous carcinoma and metachronous carcinoma of 4.0%-12.9% and 2.5%-5.1%, respectively[35-37], and 5-year and 10-year cumulative risk rates as high as 9.5% and 22.7%, respectively[35]. Therefore, for patients with eCura A and eCura B, the Japanese guidelines recommend close follow-up observation to monitor the occurrence of metachronous gastric cancer and LNM[38], and regular high-quality endoscopy follow-up can detect more than 95% of metachronous carcinomas and regular abdominal computed tomography can monitor the presence or absence of LNM and distant organ metastasis.

NCR refers to the situation that does not meet the criteria for curative resection or expanded curative resection after endoscopic resection, and its incidence is approximately 14.3%-21.4%[39-42]. NCR includes eCura C1 and eCura C2, among which eCura C1 refers to non-en bloc resection and positive horizontal margins, while other situations belong to eCura C2[22].

For patients with eCura C1, the guidelines recommend additional ESD remedial resection, surgical treatment, and close follow-up. Follow-up is a feasible strategy for patients having only positive horizontal margins with a low rate of LNM. A study found that in 77 patients with positive horizontal margins after ESD, only 11.9% had local recurrence without distant metastasis after 60 mo of follow-up, and the 5-year overall survival rate was 94.2%[43]. Other studies have found a higher risk of recurrence in patients whose tumors were partitioned, but no tumor-related deaths during the 10-year follow-up period were observed[44]. However, due to the lack of evidence from randomized controlled studies, there is still no accepted standard treatment for NCR of eCura C1.

For eCura C2 patients with high-risk factors for LNM, the guidelines recommend additional surgical treatment. Suzuki et al[45] divided 1969 EGC patients with NCR into the additional surgery group and the observation group and found that the 5-year overall survival rates of the two groups were 91% and 75.5% (P < 0.001), and the disease-specific survival rates was 99.0% and 96.8% (P = 0.013), respectively. Therefore, although the current treatment of eCura C2 is still controversial, most evidence shows that additional surgery can benefit patients’ survival[39,46-48]. However, salvage surgery also increases the risk of surgical complications and reduces the patient’s postoperative life quality, and it is possible for these patients to obtain postoperative pathological specimens without residual cancer.

Therefore, for patients diagnosed with NCR after ESD, two factors need to be considered in the formulation of remedial strategies: (1) Positive margin or local recurrence; and (2) LNM. In the absence of LNM, complete excision can be achieved by ESD again, regardless of positive margin or local recurrence. However, how to predict the risk of LNM after NCR of ESD is the key to guiding treatment after NCR[49].

To assess the risk factors for LNM in patients with NCR, Hatta et al[40] proposed the eCura scoring system to make treatment decisions. Five factors including tumor size (1 point), invasion depth (1 point), lymphatic invasion (3 points), venous invasion (1 point), and vertical margin positive (1 point) were included in the eCura scoring system. Patients with a total score of 0-1, 2-4, and 5-7 were classified as low-risk, medium-risk, and high-risk groups, with LNM rates of 2.5%, 6.7%, and 22.7%, respectively(Table 3). The eCura scoring system was used to conduct internal verification on 905 patients with EGC without additional surgical treatment. The results showed that the 5-year tumor-specific survival rates of the low-, medium-, and high-risk groups were 99.6%, 96%, and 90.2%, respectively (P < 0.01)[40]. In a follow-up study, compared to the patients with additional surgery, Hatta et al[41] demonstrated that a higher risk of tumor recurrence (HR = 3.13, P = 0.024) and no significant difference in specific tumor-related mortality (reference range = 2.66, P = 0.063) in high-risk patients as per the eCura scoring system. This indicated that the additional radical surgery after ESD is of great significance to improve the prognosis of the high-risk group, while close follow-up is also a feasible option for low-risk group. In addition, Niwa et al[50] retrospectively analyzed 47 patients with EGC and found that the eCura scoring system was also applicable to the selection of additional surgery after the NCR of ESD. However, in clinical practice, patients with undifferentiated EGC often choose radical gastrectomy, and there is a selection bias. Therefore, the study did not recommend the use of eCura scoring system to evaluate risk level and formulate therapy in patients with undifferentiated EGC. Consequently, how to put forward a more accurate model to predict LNM and tumor recurrence to reduce unnecessary surgical trauma is still a research hotspot in the future.

Thus, there are both correlations and differences between eCura and the eCura scoring system[22,40]. eCura is mainly used for the curative evaluation of EGC patients undergoing endoscopic resection. As per the guidelines, undifferentiated carcinoma or carcinoma with a tumor diameter > 2 cm that invades the submucosa are classified as eCura C2. However, the eCura scoring system is mainly for eCura C2 patients with EGC to predict the risk of LNM, so the histological type is not included as an evaluation index (Table 4).

For additional surgery after NCR, the range of gastric resection is not clearly specified in the guidelines, but the resection range of EGC can be referred to. According to the European Society for Medical Oncology guidelines, a distal gastrectomy should be performed if the proximal margin of resection is more than 5 cm from the tumor, otherwise a total gastrectomy should be considered[55]. According to the NCCN guidelines, adequate gastrectomy for T1b-T3 stage tumors is recommended to achieve a negative pathologic margin. Distal gastrectomy is preferred for distal gastric tumors, while both proximal and total gastrectomy are available for proximal gastric tumors[27]. Yamasaki et al[56] conducted a prospective multicenter controlled study on early upper stomach cancer, which confirmed that compared with total gastrectomy, patients with proximal gastrectomy had good safety and short-term and long-term efficacy. Therefore, the Japanese guidelines recommend that a safe margin of 2 cm should be ensured for T1 patients and preoperative endoscopic positioning should be performed for tumors with unclear boundaries. Distal gastrectomy should be performed for lower stomach cancer and pylorus-preserving gastrectomy (PPG) and proximal gastrectomy should be considered for tumors in the middle of the stomach (more than 4 cm from the pylorus) and upper stomach, respectively[22].

Function-preserving gastrectomy is performed to maximize the postoperative life quality of patients by ensuring the radical resection of the tumor. With the gradual improvement in people’s requirements for quality of life, the treatment of EGC has gradually shifted from radical gastrectomy to function-preserving gastrectomy which includes PPG, laparoscopic-endoscopic combined partial gastrectomy, proximal gastrectomy, segmental gastrectomy, and local gastrectomy. Although segmental gastrectomy and local gastrectomy can theoretically achieve the effects of radical oncology, there is still a lack of high-quality research evidence, so it is not often used in clinical practice. The indications for PPG, which have been studied extensively in recent years, are mainly for EGC patients with cT1N0, tumor lesions located at the greater curvature of the gastric body, and the distance from the pylorus of more than 4 cm. The advantages of surgery are mainly reflected in the reduced incidence of dumping syndrome and bile reflux due to pyloric resection, as well as better food storage[57,58] (Figure 2). Studies have confirmed that the probability of suprapyloric LNM in T1 stage EGC in the middle of the stomach is only 0.2%. Therefore, lymph node dissection in the suprapyloric region can be omitted or only partially dissected to preserve the hepatic branch, the celiac branch, and the pyloric branch of the vagus nerve and the right gastric vessel, so as to seek a balance between the radical resection of the tumor and the function preservation as far as possible, and at the same time reduce the incidence of postoperative gastric emptying and improve the postoperative life quality of patients[59,60]. We also think that anastomotic methods might be associated with gastric emptying disorder occurrence after PPG, that is, manual suture might be better than Stapler. However, we need to perform large-sample clinical trials to verify this in the future. In addition, another study conducted short- and long-term follow-ups of 2898 Japanese patients with EGC in the middle of the stomach who underwent either PPG or distal gastrectomy[61]. It was found that there were no statistically significant differences in mortality, incidence of postoperative complications, and 3-year and 5-year survival rates between the two groups[61]. Meanwhile, Tsujiura et al[62] evaluated the nutritional status of 465 patients undergoing PPG surgery and found that the serum total protein, albumin, and hemoglobin could be maintained at a good level, and the bodyweight ratio could be restored to 93.24% ± 7.29% one year after the surgery. Therefore, PPG can achieve the same therapeutic effects as distal gastrectomy for patients with T1N0 EGC in the middle of the stomach, but the grasp of indications, especially the accuracy of preoperative diagnosis, is an important factor affecting the therapeutic effect.

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Indications for pylorus-preserving gastrectomy. For early gastric cancer patients with cT1N0, when the tumor lesion is located at the greater curvature of the gastric body and the distance from the pylorus is more than 4 cm, they are suitable for pylorus-preserving gastrectomy (PPG) surgery. The advantages of PPG include reduced incidence of dumping syndrome and bile reflux due to pyloric resection, as well as better food storage.