Metachronous Multiple Gastric Cancer Discovered as Endoscopic Curability C2 during Regular Follow-Up after Gastric Endoscopic Submucosal Dissection

Introduction: The objective of this study was to clarify characteristics of metachronous endoscopic curability C2 (eCura C2) cancer during post-endoscopic submucosal dissection (ESD) follow-up. Methods: Of 4,355 gastric lesions treated by ESD at our hospital during 2005–2021, 657 were metachronous. After excluding lesions found ≥2 years since the prior examination or in the gastric remnant, the remaining 515 were analyzed. Study 1: We compared 35 eCura C2 cancers and 480 eCura A-C1 cancers. Study 2: Endoscopic findings of the 35 lesions were examined to determine why they had been missed. Results: Mean tumor size was larger (34.0 mm vs. 12.1 mm, p < 0.01) and the proportions of mixed-type and poorly differentiated cancers were higher (highly:mixed:poorly, 34.3:57.1:8.6 vs. 94.2:5.0:0.8, p < 0.01) in the eCura C2 group. Study 2: At the prior examination, 4 lesions were noticed but considered benign, 2 lacked sufficient imaging, 19 were detectable on imaging but missed, and 10 were not detectable on imaging. Over half the lesions that were detectable but missed at the prior examination were in the lesser curvature, many being type IIa–IIb lesions with color similar to the background mucosa. All lesions not detectable on imaging at the prior examination were mixed-type or poorly differentiated type. Discussion: Metachronous cancer detected as eCura C2 cancers was significantly larger, and a significantly higher proportion was mixed-type or poorly differentiated cancers, compared with eCura A-C1 cancers. Possible reasons why these lesions were missed include rapid progression of mixed-type and poorly differentiated cancers, and poor recognition that lesions showing only slight color changes may be present at the lesser curvature.


Introduction
Endoscopic submucosal dissection (ESD) is widely accepted today as an effective minimally invasive local treatment for early gastric cancer with negligible risk of lymph node metastasis and has favorable long-term outcomes [1][2][3][4][5][6]. However, the possibility of metachronous gastric karger@karger.com www.karger.com/ddi cancer development remains after ESD because the gastric mucosa is preserved after this local treatment [7][8][9][10][11]. In principle, endoscopic follow-up examination once or twice a year is recommended [12], but despite such regular follow-ups, we sometimes encounter lesions that do not meet the criteria of curative resection. In the latest Japanese gastric cancer treatment guidelines (5th edition), noncurative resection of primary early gastric carcinoma is subdivided into endoscopic curability (eCura) C1 and eCura C2 based on the completeness of resection and the risk of lymph node metastasis [12]. Considering that eCura C1 is a technical problem during endoscopic treatment, we investigated the characteristics of metachronous cancer detected as eCura C2 cancers during regular follow-up after ESD.

Indications and Patients
A total of 4,355 gastric lesions were treated by ESD at our hospital from January 2005 to December 2021, and 657 were metachronous lesions found during follow-up. Lesions found ≥2 years since the prior examination, those in the gastric remnant, and those followed up at other hospitals were excluded, leaving 515 lesions for comparative analysis (501 treated by ESD, 13 by surgery, and 1 by chemotherapy). We defined "eCura C2" cancers as those requiring surgery or chemotherapy without ESD, although this does not fit the strict definition of eCura C2. Study 1: We compared 35 eCura C2 cancers and 480 eCura A-C1 cancers. Study 2: Endoscopic findings of these 35 eCura C2 cancers were examined to find reasons why they were missed. Followup esophagogastroduodenoscopy (EGD) was started within 2 months after the initial ESD. Surveillance EGD was performed every 6-12 months in principle. This study was approved by the Ethics Committee of Toranomon Hospital (No. 1828).

Evaluations
We retrospectively analyzed patient background factors, endoscopic findings, histopathology, Helicobacter pylori status, atrophy of background gastric mucosa, examination interval, endoscopic findings of the initially treated lesion, and histopathology. All endoscopic findings were reviewed for data related to the size, location, and gross findings of the tumor.
Longitudinal tumor locations were classified as being in the upper third (cardia, fundus, and upper body), middle third (midbody and lower body), and lower third (angle, antrum, and prepylorus) of the stomach [13]. Circumferential tumor locations were defined as the anterior wall, posterior wall, lesser curvature, and greater curvature. The largest diameter of the gastric lesion was measured as the tumor size. The endoscopic gross morphology of the tumors was classified based on the Paris Classification [14]. The presence of atrophy was assessed endoscopically by an expert endoscopist. Gastric atrophy was defined based on endoscopic evaluation and categorized using the Kimura-Takemoto classification [15]. For eCura C2 cancers, whether images of the same site were taken at the prior examination, whether biopsy of the lesion site had been performed, and whether the lesion was detectable on imaging taken at the prior examination were assessed.

Definitions of eCura
Curability of endoscopic resection was defined based on the Japanese Guidelines for Gastric Cancer Practice 2018, 5th edition [12]. eCura A or B was defined as cases where the tumor had a negative vertical margin, had no lymphovascular invasion, and met one of the following conditions: (1) dominantly differentiated-type tumor, pT1a, and no ulcerative findings; (2) dominantly differentiated-type tumor, pT1a, ulcerative findings, and tumor size ≤3 cm; (3) dominantly differentiated-type tumor, SM1 (tumor invaded submucosa less than 500 μm from the muscularis mucosa), and tumor size ≤3 cm; or (4) dominantly undifferentiated-type tumor, pT1a, and tumor size ≤2 cm. eCura C1 was defined as cases where the tumor was dominantly differentiated-type and corresponded to eCura A or B but either had not been resected en bloc or had positive horizontal margins. eCura C2 was defined as cases where the resection did not correspond to eCura A, B, and C1.

Stratifications of Reasons Why Lesions Were Missed
Reasons why the eCura C2 cancers were missed at the prior examination were classified into the following 4 patterns based on endoscopy reports and images ( Fig. 1): "qualitative errors in diagnosis (lesions noticed)," when lesions had been noticed but judged as inflammation or adenomas, or followed after diagnosis of inflammation by biopsy; "inadequate endoscopic observation (observation not conducted)," when endoscopic images of the lesion site were not taken during the prior examination; "errors in detection (lesions not noticed)," when lesions were missed even though they were detectable during the prior examination (Fig. 2); and "difficulties in detection (lesions difficult to notice)," when lesions were undetectable even after reviewing images of the lesion site (Fig. 3).
Five board-certified fellows of the Japan Gastroenterological Endoscopy Society (hereinafter referred to as board-certified fellows) assessed whether images of the lesion site were taken at the prior examination and, if so, whether lesions were retrospectively detectable in the images. A majority vote was taken when opinions differed.

Definition of Metachronous Gastric Cancer
In general, synchronous gastric cancers are often defined as secondary cancers found within 1 year after endoscopic treatment, but in this study, we included those cancers in metachronous gastric cancers to examine cases where lesions were missed by endoscopy.

Statistical Analysis
Data are presented as the mean ± standard deviation. Statistical analysis was performed using the χ 2 test and the Mann-Whitney U test. All statistical analyses were performed using Stata version 14 (StataCorp, College Station, TX, USA). A p value of <0.05 was considered significant.

Study 1
Patient characteristics are shown in Table 1. Mean age was around 75 years in both groups, and the sex ratio was not significantly different between the groups. There were no significant differences in longitudinal (upper third, middle third, lower third) and circumferential locations (anterior wall, greater curvature, posterior wall, lesser curvature) of lesions between the groups. The mean tumor size was significantly larger in the eCura C2 group than in the eCura A-C1 group (34.0 mm vs. 2.1 mm, p < 0.01). Tumor histopathology showed the proportions (%) of mixed-type and poorly differentiated cancers were significantly higher in the eCura C2 group than in the eCura A-C1 group (highly differentiated:mixed:poorly differentiated, 34.3:57.1:8.6 vs. 94.2:5.0:0.8, p < 0.01).
The SM invasion rate was significantly higher in the eCura C2 group than in the eCura A-C1 group (M:SM or deeper, 9:24 vs. 470:10, p < 0.01). There was no significant difference between the groups in examination interval (eCura C2 group, 8.8 ± 3.6 months; eCura A-C1 group, 8.2 ± 4.0 months), longitudinal and circumferential locations of initially treated lesions, histopathological findings, and the number of lesions. Also, there was no significant difference in time since the prior examination, H. pylori status, atrophy of background mucosa, or map-like redness. The percentage of lesions examined A typical image of lesions missed due to "errors in detection." a1 A large lesion covering the cardia, upper body, and fornix, and a depression on the posterior wall of the cardia were found. ESD was performed, and pathological findings for the lesion (120 × 65 mm) were type 0-IIa+IIb, tub2>1, SM2 (2,100 μm), ly0, v1, ul(−), HM0, VM1. a2 An image of the same location as in a1, taken during endoscopic examination 12 months earlier. A regional type IIb lesion with color similar to that of the background mucosa was detectable. b1 A large lesion covering from the incisura angularis of the lesser curvature to the middle part of the body was confirmed (NBI image). Surgery was performed, and pathological findings for the lesion (70 × 55 mm) were tub2>por, sig>tub1, M, ul(+), ly0, V0. b2 An image of the same location as in b1, taken during endoscopic examination 10 months earlier. A regional yellowish type IIb lesion in the incisura angularis of the lesser curvature was detectable." NBI, narrow band imaging.  Table 2 shows reasons why eCura C2 cancers were missed at the prior examination. "Qualitative errors in diagnosis" was the reason why 4 lesions were missed: 2 after diagnosis of inflammation by biopsy and 1 each judged to be adenoma or benign erosion by endoscopy. H. pylori eradication therapy had been performed in all cases. "Inadequate endoscopic observation" was the reason why 2 lesions were missed: both did not have images taken at the prior examination and were not examined by a specialist. One was located at the lesser curvature in the middle third, and the other at the anterior wall of the lower third. "Errors in detection" was the reason why 19 lesions were missed: the lesions were large (mean tumor size, 42.2 ± 29.2 mm) and more than half of them were located at the lesser curvature. Unlike the majority of lesions that are red and depressed, a relatively large number (more than half of the total) of these lesions were type IIa-IIb with a color similar to or slightly more yellow than the background mucosa, and these slight differences in color against the background failed to be recognized. Depressed or protruding lesions had newly developed when they were detected (Fig. 2). Longitudinal and circumferential locations of metachronous lesions were different from those of the initially treated lesions in 73.7% and 89.5% of cases, respectively, suggesting that areas close to the postoperative scar were the focus of the observation and locations far from the scar might not have been closely examined during observation. "Difficulties in detection" was the reason for missing 10 lesions. These were not detectable at the prior examination, and all of them were mixed-type or poorly differentiated cancers. All of these lesions changed drastically in a short period of time, indicating their rapid growth. Further, a relatively large proportion (40%) was located at the greater curvature.

Discussion
In the current era, where ESD is performed instead of endoscopic mucosal resection, the rate of complete en bloc resection has improved markedly, enabling local recurrence to be avoided. ESD is an excellent treatment that preserves gastric function, thereby contributing to the maintenance of QOL. However, the gastric mucosa from which early gastric cancer arose is preserved intact    because ESD is a local treatment. Thus, even though lesions are dissected endoscopically, careful follow-up is necessary for synchronous lesions that were missed at the initial examination and for metachronous cancer that may occur in the future [7][8][9][10][11]. The occurrence rate of synchronous and metachronous cancer in patients who underwent endoscopic treatment for early gastric cancer is relatively high (5-15%), albeit with some variation [7,10,[16][17][18][19][20][21]. Reported risk factors for metachronous gastric cancer are older age, persistent H. pylori infection, and severe atrophy [18,[22][23][24][25]. The performance of endoscopic equipment has improved markedly in recent years. However, the ability of humans to detect lesions is limited; thus, it is impossible to completely avoid missing lesions, regardless of how remarkably the equipment performs. The rate of missed gastric cancer by endoscopy is reported to be 10-25% [26,27] or 9.4% (95% confidence interval, 5.7-13.1%) by a meta-analysis [28]. Areas where lesions are often missed include the lesser curvature at the cardiac region, the posterior wall of the body, between the folds on the greater curvature of the body, and the antrum during peristalsis. If missed lesions are defined as those that were missed, although they were detectable on endoscopic images taken at the prior examination, 20-30% of the lesions in the eCura A-C1 group in this study would be categorized as missed lesions. Nevertheless, many of them were detected during follow-up examination and curatively resected by ESD. Curative resection is likely achievable in cases of early gastric cancer even with a delay in detection of approximately 1-2 years. Given that, this study did not examine missed lesions and instead examined lesions that had been missed at the prior examination and were detected as eCura C2 cancers.
This study showed that the mean tumor size was significantly larger and the proportions of mixed-type and poorly differentiated cancers were significantly higher in the eCura C2 group than in the eCura A-C1 group. Also, the percentage of lesions examined by nonspecialists was significantly higher in the eCura C2 group than in the eCura A-C1 group. There were no differences between the groups in age, H. pylori status, and degree of atrophy, which are known high-risk factors for the development of metachronous cancer. Similarly, there were no significant differences in the longitudinal and circumferential locations of metachronous lesions, examination intervals, the presence/absence of map-like redness, longitudinal and circumferential locations of initially treated lesions, and histopathology.
This study examined reasons why metachronous lesions were missed and found that approximately 2% of them were not found on images taken at the prior examination, indicating the limit of endoscopic examination. Those lesions in this study were either mixed-type or poorly differentiated cancers that grew rapidly in a short period of time, indicating that tumor growth rate was a considerable factor. Even though regular follow-up  Coloration of lesion at last endoscopy Reddish 52.6 (10) Same color to pale yellow 47.4 (9) Morphology of lesion at last endoscopy 0-IIa 21. is performed, we sometimes encounter such cases and therefore need to be fully aware of them. Even with regular follow-up, it is impossible to avoid surgery in all cases, but the rate of surgery can be reduced. In this study, additional therapy was indicated for 6.8% of metachronous lesions, approximately 70% of which could have been detected and treated earlier (lesions missed because of "inadequate endoscopic observation," "qualitative errors in diagnosis," and "errors in detection"). If surgery becomes necessary because lesions were missed at the prior examination, it will affect patients' QOL considerably. What we can do first is to perform thorough gastric lavage to improve detection of lesions and minimize the chances of missed observation by taking images of the whole stomach in accordance with the predetermined protocol. Then, endoscopists need to keep in mind that there may be a large lesion that can be easily missed (type IIa-IIb lesions in the lesser curvature, the color of which is similar to or slightly more yellow than the background mucosa); they then need to perform observation to detect the slightest changes in color under multiple conditions (e.g., observing the same area with different degrees of air insufflation and from different angles). Given that many lesions are likely to be missed if inadequate time is spent for EGD [29], a certain length of time will be needed to capture images over a wider area.
To improve the quality of diagnosis, endoscopy using indigo carmine, second-generation narrow band imaging (EVIS LUCERA ELITE), third-generation narrow band imaging (EVIS X1), and linked color imaging [30] will be useful. A diagnostic system combining image-enhanced endoscopy and magnifying endoscopy (VS classification system) has been established and demonstrated extremely high rates of correct diagnosis [31][32][33]. Furthermore, if malignancies are suspected by endoscopy, reexamination should be performed even if biopsy results are negative.
This study has several limitations. First, it is a retrospective single-institution study with a limited number of patients. Second, the follow-up period was relatively short. A further long-term prospective study is awaited. Nevertheless, to our knowledge, no previous study has compared eCura C2 metachronous cancers with eCura A-C1 metachronous cancers, and thus, this study has significance.

Conclusions
Metachronous cancer, found as eCura C2 cancers through regular follow-up after ESD, was significantly large and contained a significantly higher proportion of mixed-type and poorly differentiated cancers compared with those found as eCura A-C1 cancers. Reasons why these lesions were missed included rapid progression of mixed-type and poorly differentiated tumors (in 2% of total cases) and poor awareness of lesions with slight changes in color and negligible depression or protrusion. Endoscopists must be vigilant to perform careful observation, with full awareness of the risk of missing lesions.

Statement of Ethics
Because of the retrospective nature of this study, written informed consent was not obtained from the patients, and the need for informed consent was waived by the Ethics Committee of Toranomon Hospital. The study was conducted according to the ethical principles stated in the 1964 Helsinki Declaration. In our study, research and publication ethics were followed, and the relevant rules were followed as well. Ethical approval of the study was obtained from the Ethics Committee of Toranomon Hospital (approval number: 1828).