Effect of optical diagnosis training on recognition and treatment of submucosal invasive colorectal cancer in community hospitals: a prospective multicenter intervention study

Background and study aims: Recognition of T1 colorectal cancer (CRC) is difficult, with sensitivities of 35-60% in Western countries. We evaluated the real-life effects of the implementation of the OPTICAL model, a recently developed structured and validated prediction model, in Dutch community hospitals. Patients and methods: In this prospective multicenter study, 383 endoscopists from 40 hospitals were invited to follow an e-learning on the OPTICAL-model, to increase sensitivity for detecting T1CRC in non-pedunculated polyps. Next, real-life recognition of T1CRC was evaluated in 25 hospitals. Endoscopic and pathologic reports of T1CRCs detected during the next year were collected retrospectively while endoscopists were unaware of this evaluation. Sensitivity for recognition of T1CRC, R0 resection rate, and treatment modality were compared for trained vs. untrained endoscopists and for recognised vs. unrecognised T1CRCs. Results: Within 1 year after the e-learning 251 endoscopists detected 528 non-pedunculated T1CRCs, 118 (47%) of the endoscopist were trained. T1CRCs had a median size of 20mm and were mainly located in the distal colorectum (66%). Trained endoscopists recognised T1CRCs more frequently than untrained endoscopists (sensitivity 74% vs. 62%; mixed model analysis OR 2.90; 95%CI 1.54-5.45. A higher rate of R0 resection was seen for T1CRCs detected by trained endoscopists (69% vs. 56%, OR 1.73; 95%CI 1.03-2.91). Conclusion: Training in optical recognition of T1CRCs in community hospitals was associated with an increase in sensitivity for T1CRCs. Recognition led to a higher rate of en bloc local excision, resulting in higher R0-resection rates. This may be an important step towards more organ-preserving strategies.