Evaluation of the CAD-EYE System for the Detection of Colorectal Neoplastic Lesions in Patients With Lynch Syndrome

Lynch syndrome (LS) is the most common genetic predisposition syndrome for colorectal cancer (CRC), responsible for around 2-4% of cancers. It is characterized by a pathogenic germline mutation in one of the DNA mismatch repair genes (path_MMR) MLH1, MSH2, MSH6, PMS2 or a deletion in the 3' region of the Epcam gene. Patients followed up for LS are at high risk of developing CRC at an early age, and have a high cumulative CRC risk. In this context, CRC screening by colonoscopy is of major importance, as it is associated with a reduction in both CRC incidence and mortality.

In France, the Institut National du Cancer (INCa) recommends colonoscopy with indigo carmine chromoendoscopy (CE), as it is associated with a significant increase in the adenoma detection rate (ADR) compared with white light. However, EC is not routinely performed in clinical practice, as it is a time-consuming technique requiring a dedicated slot with a trained operator.

Recent years have seen the emergence of artificial intelligence techniques for real-time polyp detection aids or CADe devices. These easy-to-use systems have shown very promising results compared with high-definition (HD) white light. Indeed, data from the first meta-analysis of 5 randomized controlled trials (4354 patients) confirmed a significantly higher ADD in the CADe group than in the HD group (36.6% vs. 25.2%; 95% CI], 1.27-1.62; P < 0.01; I2 Z 42%) 10. The CAD EYE system (Fujifilm) is a CADe device supporting both detection (sensitivity > 95%) and characterization of colonic polyps in real time.

To date, artificial intelligence has never been evaluated for CRC screening in patients followed up for LS. The aim of this work is to evaluate the effectiveness of the CAD EYE system in this specific population. To this end, we intend to conduct a randomized, controlled, non-inferiority trial comparing CAD EYE with CE in patients with LS.