The Magnifying Endoscopy Simple Diagnostic Algorithm for Early Gastric Cancer (MESDA-G) has been proposed as a standardized diagnostic system using magnifying endoscopy with narrow-band imaging (M-NBI) for early gastric cancer (EGC), and its usefulness has been reported. However, we sometimes encountered false-negative cases with M-NBI diagnosis (M-NBI diagnostic limitation EGC: MDL-EGC) (Endosc Int Open, 2020). On the other hand, we previously developed automatic analysis software for gastric mucosal blood flow rate and demonstrated that EGCs exhibit significantly slower flow rates compared with reddish non-neoplastic mucosa, supporting its diagnostic utility (J Gastroenterol Hepatol, 2021). Nevertheless, blood flow rate analysis for MDL-EGCs has not yet been investigated. Therefore, the aim of this study was to evaluate blood flow rate in MDL-EGC and clarify the usefulness of the gastric mucosal blood flow rate analysis software.

Among 286 patients (314 lesions) who underwent endoscopic submucosal dissection for EGC at our institution between February 2021 and January 2024, 38 lesions (12.1%) that were diagnosed as “non-cancer” according to MESDA-G were defined as MDL-EGCs and analyzed. Blood flow rate was measured using the automatic analysis software. The diagnostic performance for cancer was assessed using a previously established cutoff value (1.189 mm/s) derived from other test data sets. In addition, lesions were histopathologically classified, and blood flow rates were compared among the groups.

The 38 MDL-EGCs consisted of H. pylori status (current/past/never infection = 1/23/14), location (U/M/L = 15/10/13), macroscopic type (elevated/flat/depressed = 15/10/13), and invasion depth (M/SM = 24/14). Histologically, the lesions were classified into four groups: low-grade differentiated adenocarcinoma (n=12), signet-ring cell carcinoma (sig, n=5), gastric adenocarcinoma of fundic gland mucosa type (GA-FGM, n=5), and gastric adenocarcinoma of fundic gland type (GA-FG, n=16). The mean blood flow rate of all lesions was 1.169 mm/s, and 60.5% of lesions were diagnosed as cancer (<1.189 mm/s) by blood flow analysis (differentiated/sig/GA-FGM/GA-FG =(11/12, 91.7%)/(3/5, 60%)/(3/5, 60%)/(6/16, 37.5%)). Mean blood flow rates by group were 1.06, 1.13, 1.17, and 1.26 mm/s, respectively. GA-FG, in which the surface epithelium is covered by non-neoplastic mucosa, showed the highest blood flow rate, whereas low-grade differentiated adenocarcinoma, in which the tumor is exposed to the surface, showed the lowest blood flow rate. In signet-ring cell carcinoma, 60% (3/5) were correctly diagnosed as cancer by blood flow rate analysis, possibly due to tumor proliferation immediately beneath the non-neoplastic surface epithelium affecting surface blood flow.

This study is the first report to evaluate the usefulness of the analysis system for MDL-EGCs. In MDL-EGCs, blood flow rate analysis was suggested to be useful not only for low-grade differentiated adenocarcinoma and GA-FGM with surface tumor exposure but also for signet-ring cell carcinoma covered by non-neoplastic mucosa. Therefore, this analysis system has the potential to be useful in clinical practice. We hope for further investigation of various types of lesions and the development of a next-generation microvascular blood flow analysis system, which can be applied to clinical endoscopic practice.