Colonoscopy surveillance in Lynch syndrome: what it prevents and what it does not.

Abstract

Lynch syndrome (LS), synonymous with hereditary non-polyposis colorectal cancer (HNPCC), is caused by germline pathogenic variants in MLH1, MSH2, MSH6 or PMS2, which confer an elevated lifetime risk of colorectal cancer (CRC). Since the early 2000s, colonoscopic surveillance has been recommended to reduce CRC incidence via polypectomy and mortality via early detection, with intervals now being tailored by gene. Early non-randomised studies suggested that surveillance imparted reductions in both CRC incidence and mortality. However, more recent prospective registry data, including the Prospective Lynch Syndrome Database (PLSD), consistently report high CRC incidence, despite regular colonoscopy. Nevertheless, mortality from CRC is low in individuals under surveillance, indicative of a beneficial effect of early cancer detection, likely through detection at a less advanced tumour stage. The apparent discrepancies between historical and recent studies regarding incidence reduction may reflect methodological issues (selection, confounding, overdiagnosis, immortal-time bias). However, LS tumourigenesis often follows an accelerated or alternative pathway (DNA mismatch repair (MMR)-deficient adenomas and MMR-deficient crypt foci) that may bypass conventional, easily detected precursor lesions. This may limit the impact of surveillance on incidence but preserve the mortality benefit through earlier-stage detection. Adherence, procedure quality and surveillance interval further influence outcomes. When weighting the evidence, large prospective consortium studies do not provide evidence in support of a substantial reduction in CRC incidence with 1-2 yearly surveillance for carriers of pathogenic MLH1/MSH2 variants over longer intervals. However, smaller retrospective studies have suggested potential benefits. Less intensive schedules appear appropriate for carriers of pathogenic PMS2 variants due to their low penetrance and more conventional adenoma-carcinoma pathways. In short, the evidence is strong that colonoscopic surveillance in LS reliably reduces CRC mortality but incompletely prevents incidence, especially in carriers with pathogenic variants whose cancers arise through an accelerated pathway and without a visible precursor. Improving adherence and endoscopic quality, considering adjunctive techniques (eg, faecal immunochemical testing between surveillance intervals, chromoendoscopy/AI (artificial intelligence) assistance), and exploring complementary strategies (eg, aspirin chemoprevention, biomarker-guided risk) are priorities. This narrative review synthesises current evidence, highlighting the need for robust future studies to optimise patient surveillance.