Artificial Intelligence in Colonoscopy Surveillance for Lynch Syndrome: Emerging Evidence, Lessons Learned From...

Area of Science:

• Gastroenterology
• Oncology
• Medical Imaging

Background:

• Lynch syndrome (LS) is the most common hereditary colorectal cancer (CRC) syndrome, featuring accelerated carcinogenesis and high adenoma miss rates.
• Artificial intelligence (AI) and computer-aided detection (CADe) have improved adenoma detection in average-risk populations.
• The efficacy of AI in LS surveillance, where cancer development and standards differ, remains unclear.

Purpose of the Study:

• To review current evidence on AI-assisted colonoscopy in Lynch syndrome surveillance.
• To contextualize findings within the broader research on AI in average-risk CRC screening.
• To evaluate the safety and efficacy of AI integration in LS surveillance protocols.

Main Methods:

• Narrative review of existing literature on AI-assisted colonoscopy in LS.
• Inclusion of findings from randomized controlled trials, specifically the CADLY and TIMELY trials.
• Comparison of AI performance in LS versus average-risk CRC screening populations.

Main Results:

• AI can be safely integrated into high-quality surveillance for Lynch syndrome patients.
• Current data do not demonstrate that AI aids expert colonoscopists in improving overall adenoma or advanced neoplasia detection rates.
• Adequate baseline procedural quality is a prerequisite for evaluating AI's impact.

Conclusions:

• AI integration into Lynch syndrome surveillance appears safe.
• AI has not yet proven superior to expert colonoscopists for increasing adenoma detection in LS.
• Further investigation is required to establish AI's definitive role in enhancing LS surveillance outcomes.