Endoscopic imaging modalities for diagnosing the invasion depth of superficial esophageal squamous cell carcinoma: a systematic review

Evaluating the usefulness of considering the size and morphological type of type B2 vessel area based on Japan Esophageal Society classification in estimating tumor invasion depth in superficial esophageal squamous cell carcinomas: study protocol for a prospective observational study (Japan BEES study)

BACKGROUND

Accurate evaluation of tumor invasion depth is essential to determine the appropriate treatment strategy for patients with superficial esophageal cancer. The pretreatment tumor depth diagnosis currently relies on the magnifying endoscopic classification established by the Japan Esophageal Society (JES). However, the diagnostic accuracy of tumors involving the muscularis mucosa (MM) or those invading the upper third of the submucosal layer (SM1), which correspond to Type B2 vessels in the JES classification, remains insufficient. Previous retrospective studies have reported improved accuracy by considering additional findings, such as the size and macroscopic type of the Type B2 vessel area, in evaluating tumor invasion depth. Therefore, this study aimed to investigate whether incorporating the size and/or macroscopic type of the Type B2 vessel area improves the diagnostic accuracy of preoperative tumor invasion depth prediction based on the JES classification.

METHODS

This multicenter prospective observational study will include patients diagnosed with MM/SM1 esophageal squamous cell carcinoma based on the Type B2 vessels of the JES classification. The tumor invasion depth will be evaluated using both the standard JES classification (standard-depth evaluation) and the JES classification with additional findings (hypothetical-depth evaluation) for the same set of patients. Data from both endoscopic depth evaluations will be electronically collected and stored in a cloud-based database before endoscopic resection or esophagectomy. This study's primary endpoint is accuracy, defined as the proportion of cases in which the preoperative depth diagnosis matched the histological depth diagnosis after resection. Outcomes of standard- and hypothetical-depth evaluation will be compared.

DISCUSSION

Collecting reliable prospective data on the JES classification, explicitly concerning the B2 vessel category, has the potential to provide valuable insights. Incorporating additional findings into the in-depth evaluation process may guide clinical decision-making and promote evidence-based medicine practices in managing superficial esophageal cancer.

TRIAL REGISTRATION

This trial was registered in the Clinical Trials Registry of the University Hospital Medical Information Network (UMIN-CTR) under the identifier UMIN000051145, registered on 23/5/2023.

Endoscopic Diagnosis of Superficial Gastrointestinal Cancer

White light image (WLI) findings are important for detection and characterization in the GI tract. However, magnified endoscopic examination with image enhanced endoscopy (IEE-NE) is becoming increasingly important for qualitative diagnosis of GI neoplastic lesions. IEE-ME is extremely useful for diagnosis of invasion depth in esophageal squamous cell cancer (ESCC) and colorectal cancer, whereas macroscopic findings of WLI are still useful in Barrett's adenocarcinoma (BAC) and gastric cancer. IEE-ME is also useful for diagnosis of tumor extent in BAC and gastric cancer, whereas chromoendoscopy with indigo carmine is useful in colorectal cancer and iodine staining is indispensable in ESCC.

Human-Like Artificial Intelligent System for Predicting Invasion Depth of Esophageal Squamous Cell Carcinoma Using Magnifying Narrow-Band Imaging Endoscopy: A Retrospective Multicenter Study

INTRODUCTION

Endoscopic evaluation is crucial for predicting the invasion depth of esophagus squamous cell carcinoma (ESCC) and selecting appropriate treatment strategies. Our study aimed to develop and validate an interpretable artificial intelligence-based invasion depth prediction system (AI-IDPS) for ESCC.

METHODS

We reviewed the PubMed for eligible studies and collected potential visual feature indices associated with invasion depth. Multicenter data comprising 5,119 narrow-band imaging magnifying endoscopy images from 581 patients with ESCC were collected from 4 hospitals between April 2016 and November 2021. Thirteen models for feature extraction and 1 model for feature fitting were developed for AI-IDPS. The efficiency of AI-IDPS was evaluated on 196 images and 33 consecutively collected videos and compared with a pure deep learning model and performance of endoscopists. A crossover study and a questionnaire survey were conducted to investigate the system's impact on endoscopists' understanding of the AI predictions.

RESULTS

AI-IDPS demonstrated the sensitivity, specificity, and accuracy of 85.7%, 86.3%, and 86.2% in image validation and 87.5%, 84%, and 84.9% in consecutively collected videos, respectively, for differentiating SM2-3 lesions. The pure deep learning model showed significantly lower sensitivity, specificity, and accuracy (83.7%, 52.1% and 60.0%, respectively). The endoscopists had significantly improved accuracy (from 79.7% to 84.9% on average, P = 0.03) and comparable sensitivity (from 37.5% to 55.4% on average, P = 0.27) and specificity (from 93.1% to 94.3% on average, P = 0.75) after AI-IDPS assistance.

DISCUSSION

Based on domain knowledge, we developed an interpretable system for predicting ESCC invasion depth. The anthropopathic approach demonstrates the potential to outperform deep learning architecture in practice.