Abstract
BACKGROUND AND AIMS: Lynch syndrome (LS) is caused by pathogenic mutations in mismatch repair (MMR) genes. There are limited data on differences in colorectal cancer (CRC) surveillance by MMR genes, and an international consensus on surveillance based on genes is not established. We aimed to evaluate colonoscopy and esophagogastroduodenoscopy (EGD) surveillance outcomes and compare CRC surveillance findings by the mutated gene. METHODS: One hundred one patients with LS were included and colonoscopy results were compared by MMR mutation. Primary outcomes included the development and recurrence of adenoma, CRC, high-grade dysplasia, advanced adenoma, and sessile serrated lesions. Logistic regressions evaluated the relationship between genes and the development or recurrence of primary outcomes. Survival analysis evaluated primary outcomes in patients with ≥ 2 colonoscopies. EGD results were summarized. RESULTS: Three hundred twenty seven colonoscopies were reviewed. Compared to PMS2, MLH1 was associated with a higher risk of advanced adenoma/high-grade dysplasia/CRC development (odds ratio [OR] 9.85, 95% confidence interval [CI]: 1.97-77.24) and MSH2 was associated with a higher risk of adenoma development (OR 4.17, 95% CI: 1.11-17.61). Among those with > 2 colonoscopies, MLH1 (hazard ratio 18.98, 95% CI: 1.31-274.51) and MSH6 (hazard ratio 15.03, 95% CI: 1.16-194.65) had a higher risk of sessile serrated lesions compared to MSH2. Among patients who had adenoma detected once, MLH1 had a higher risk of adenoma recurrence compared to MSH6 (OR 14.59, 95% CI: 1.53-244.30) and PMS2 (OR 47.15, 95% CI: 4.26-984.28). MSH2 had a higher risk of adenoma recurrence compared to PMS2 (OR 11.89, 95% CI: 1.38-164.78). Of 170 EGDs, an actionable finding was identified in 16% of patients during their first 3 EGDs. CONCLUSION: Surveillance colonoscopy outcomes differed in patients with LS and suggest the need to guide surveillance based on MMR gene mutation.