In this pilot study, we examined both the longitudinal and circumferential locations of ESD-resected ESCC lesions to determine whether there was circumferential susceptibility. Our main findings were as follows. Among the four circumferential locations, ESCC lesions were most frequently observed in the posterior wall and least frequently in the anterior wall. The median tumor size was the smallest in the anterior wall. A higher proportion of ESCC lesions in the anterior wall had an invasion depth of EP and a size less than 10 mm, and the proportion meeting both criteria was significantly highest in the anterior wall. These findings suggest that the anterior wall has a lower carcinogenic susceptibility for ESCC, whereas the posterior wall exhibited a higher susceptibility.
To date, few studies have evaluated circumferential tumor location [5]. One reason may be that it was previously difficult to identify early-stage ESCC; however, with the recent availability of endoscopes using NBI, this has become much easier [15, 16]. Early ESCC lesions often have a small circumference, making it easy to determine their circumferential location. ESD for ESCC can preserve the esophagus, enabling continued surveillance endoscopy after the resection procedure. This allows for the detection of multiple early ESCC recurrence lesions. In recent studies of ESD for ESCC, both Hazama et al. and Mitsui et al. reported the proportion of lesions in the following order (highest to lowest): posterior, right, left, and anterior [20, 21]. These studies reported that the difficulty of ESD procedures was affected by circumferential location, and thus it has become a focus of attention. Their findings, which indicate that the posterior region is the most affected and the anterior region is the least affected, are consistent with the results of our study.
This study showed that there were fewer lesions on the anterior wall than in other circumferential locations. Notably, the highest proportion of early lesions, with an invasion depth of EP and size less than 10 mm, was found in this location. The International Agency for Research on Cancer (IARC) has identified ethanol in alcoholic beverages, acetaldehyde of alcohol metabolism associated with alcoholic beverage intake, and smoking as obvious esophageal carcinogens [3]. It has also been reported to be involved in carcinogenesis by causing oxidative stress in the gastrointestinal epithelium due to stimulation by ethanol itself [22]. Although the exact mechanism remains unclear, the susceptibility to ESCC development varies depending on the site, potentially due to different exposure to carcinogens. Therefore, we speculated that there is different susceptibility in the circumference location in addition to the middle esophagus.
Although humans often live upright or in a sitting position during the day, most tend to be in a supine position during sleep, and it is thought that saliva tends to stagnate in the esophagus. When considering the esophagus in a horizontal cross-section, it is considered that liquids such as saliva in the esophagus tend to accumulate in the posterior wall due to gravity, and the anterior wall is less frequently exposed to saliva (Fig. 2). Therefore, the fact that the lowest number of ESCC lesions were found in the anterior wall and that the highest proportion of lesions with shallow invasion depth and small size were also in this location, could be due to the low frequency of exposure of the anterior wall to the above-mentioned carcinogenic fluids (alcohol and acetaldehyde). Owing to the general curvature of the vertebrae and the positional relationship of the organs around the esophagus, the middle esophagus is often located on the dorsal side and is the area where fluid is most likely to accumulate in the supine position. Therefore, it is reasonable that ESCC lesions also occur more frequently in the central esophagus from the viewpoint of longitudinal location. Based on the results of this study, care should be taken when examining the anterior wall by endoscopy, as ESCC lesions in this location may be smaller and have a shallow depth of invasion.
This study has several limitations. First, the number of patients was small, and this retrospective study was performed in a single-center setting. Second, longitudinal tumor location was determined using endoscopy alone. This is because early ESCC, for which ESD is indicated, was the subject of the present study. They are usually not identified by esophagography or computed tomography (CT). For lesions that occupied two or more circumferential locations, the circumferential location was defined as the lesion with the longest part in the longitudinal direction. It would be desirable to analyze only lesions localized in the wall in a single direction; however, in the present study, this method would have reduced the sample size further such that sufficient analysis could not be performed. Finally, this study showed that ESCC lesions on the anterior wall were infrequent, small, and often early-stage, indicating a potential carcinogenic site in the posterior wall. However, the circumferential locations of high carcinogenesis could not be directly observed, as we focused on smaller early lesions for this study. This is supported by recent studies by Hazama et al. and Mitsui et al., which showed that the posterior wall was the most common site [20, 21]. Further studies with a greater number of cases and research from multiple institutions would confirm our results.
In conclusion, ESCC lesions in the anterior wall are less frequent, the depth of tumor invasion is shallow, and the tumor size may be small.