Assessment of the Diagnostic Performance of Endoscopic Ultrasonography After Conventional Endoscopy for the Evaluation of Esophageal Squamous Cell Carcinoma Invasion Depth.

Importance: Distinguishing between mucosal and submucosal cancers is important for selecting the optimal treatment for patients with esophageal squamous cell carcinoma (ESCC); however, standard procedures for diagnosing cancer invasion depth have not yet been determined. Objective: To evaluate the diagnostic performance of endoscopic ultrasonography (EUS) after conventional endoscopy for the evaluation of ESCC invasion depth. Design, Setting, and Participants: This prospective single-arm confirmatory diagnostic study comprising 372 patients with T1 esophageal cancer was conducted at 41 secondary or tertiary hospitals in Japan. Enrollment began on July 20, 2017; patients were enrolled in 2 steps, with the first registration occurring from August 4, 2017, to December 11, 2019, and the second from August 9, 2017, to December 11, 2019. After the completion of all first and second registration examinations, patients received treatment and were followed up for 30 days, with follow-up ending on February 14, 2020. Patients were eligible for inclusion if they had pathologically or endoscopically diagnosed esophageal cancer with T1 clinical depth of invasion. Interventions: In the first registration, nonmagnifying endoscopy (non-ME) and magnifying endoscopy (ME) were used to diagnose cancer invasion depth. In the second registration, patients from the first registration who had cancers invading the muscularis mucosa or submucosa were enrolled and received EUS. After completion of the protocol examinations, patients received treatment with endoscopic resection or esophagectomy. The pathological results of the resected specimens were used as the reference standard for evaluating cancer invasion depth. Main Outcomes and Measures: The primary end point was the proportion of overdiagnosis of submucosal cancer (defined as invasion depth >200 µm) after receipt of non-ME and ME, with or without the addition of EUS. The secondary end points were underdiagnosis, sensitivity, and specificity. Results: Among 372 patients enrolled in the first registration, 371 received non-ME and ME. Of those, 300 patients were enrolled in the second registration, and 293 patients received EUS. A total of 269 patients (217 men [80.7%]; median age, 69 years; interquartile range, 62-75 years) were included in the final analysis. The addition of EUS was associated with a 6.6% increase in the proportion of overdiagnosis (from 16 of 74 patients [21.6%; 95% CI, 12.9%-32.7%] after non-ME and ME to 29 of 103 patients [28.2%; 95% CI, 19.7%-37.9%] after the addition of EUS; 1-sided P = .93). All subgroup analyses found similar increases in overdiagnosis of submucosal cancer. The addition of EUS was associated with a 4.5% reduction in the proportion of underdiagnosis (from 57 of 195 patients [29.2%; 95% CI, 23.0%-36.2%] after non-ME and ME to 41 of 166 patients [24.7%; 95% CI, 18.3%-32.0%] after the addition of EUS). After non-ME and ME, diagnostic sensitivity was 50.4% (95% CI, 41.0%-59.9%), specificity was 89.6% (95% CI, 83.7%-93.9%), and accuracy was 72.9% (95% CI, 67.1%-78.1%). After the addition of EUS, diagnostic sensitivity was 64.3% (95% CI, 54.9%-73.1%), specificity was 81.2% (95% CI, 74.1%-87.0%), and accuracy was 74.0% (95% CI, 68.3%-79.1%). Conclusions and Relevance: This study found that the addition of EUS was not associated with improvements in the diagnostic accuracy of cancer invasion depth. These findings do not support the routine use of EUS after conventional endoscopy for evaluating the invasion depth among patients with T1 ESCC.

Real-time assessment of video images for esophageal squamous cell carcinoma invasion depth using artificial intelligence.

BACKGROUND: Although optimal treatment of superficial esophageal squamous cell carcinoma (SCC) requires accurate evaluation of cancer invasion depth, the current process is rather subjective and may vary by observer. We, therefore, aimed to develop an AI system to calculate cancer invasion depth. METHODS: We gathered and selected 23,977 images (6857 WLI and 17,120 NBI/BLI images) of pathologically proven superficial esophageal SCC from endoscopic videos and still images of superficial esophageal SCC taken in our facility, to use as a learning dataset. We annotated the images with information [such as magnified endoscopy (ME) or non-ME, pEP-LPM, pMM, pSM1, and pSM2-3 cancers] based on pathologic diagnosis of the resected specimens. We created a model using a convolutional neural network. Performance of the AI system was compared with that of invited experts who used the same validation video set, independent of the learning dataset. RESULTS: Accuracy, sensitivity, and specificity with non-magnified endoscopy (ME) were 87%, 50%, and 99% for the AI system and 85%, 45%, 97% for the experts. Accuracy, sensitivity, and specificity with ME were 89%, 71%, and 95% for the AI system and 84%, 42%, 97% for the experts. CONCLUSIONS: Most diagnostic parameters were higher when done by the AI system than by the experts. These results suggest that our AI system could potentially provide useful support during endoscopies.

Prospective small bowel mucosal assessment immediately after chemoradiotherapy of unresectable locally advanced pancreatic cancer using capsule endoscopy: a case series.

In this case series, three consecutive patients with unresectable locally advanced pancreatic cancer (ULAPC) underwent capsule endoscopy (CE) before and after chemoradiotherapy (CRT) to evaluate duodenal and jejunal mucosa, and to examine the relationship between CE findings and dose distribution. CE after CRT showed duodenitis and proximal jejunitis in all three patients. The most inflamed region was the third part of the duodenum, and in dose distribution, this was the closest region to the center of irradiation. This case series shows that CE can safely diagnose acute duodenitis and proximal jejunitis caused by CRT for ULAPC, and that dose distribution is possible to predict the degree of duodenal and jejunal mucosal injuries.