Predictors of Incident Serrated Polyps: Results from a Large Multicenter Clinical Trial.

Abstract

Serrated polyps (SP) are important colorectal cancer precursors, yet their epidemiology is incompletely understood. We measured risk factors for incident sessile-serrated lesions (SSL) and microvesicular (MVHP) and goblet-cell rich (GCHP) hyperplastic polyp subtypes. We conducted a cohort study of patients undergoing colonoscopic surveillance nested within a chemoprevention trial. Outcomes of interest were ≥1 SPs, including SSLs, MVHPs, and GCHPs specifically. Multivariable generalized estimating equation models were used to estimate adjusted risk ratios (RR) and 95% confidence intervals (CI) for different polyp types. Among 2,102 participants, a total of 1,615 SPs (including 212 SSLs) were found among 758 participants during follow-up. Prior history of SPs was strongly associated with subsequent occurrence of SPs. There was no apparent association between age, sex, or education and risk of SPs. Black participants were at lower risk of SSLs and MVHPs, but higher risk of GCHPs compared with white participants [RR, 0.40; 95% CI, 0.16-0.99); RR, 0.63 (95% CI, 0.42-0.96); and RR, 1.83 (95% CI, 1.23-2.72) respectively]. Alcohol and smoking exposure were also associated with SPs, including hyperplastic polyp subtypes in particular. In this prospective study, the risk of SP subtypes differed by race, alcohol, and smoking status, and prior history of SPs. Risk factor associations for SPs differ from risk factors for conventional adenomas, supporting the concept of etiologic heterogeneity of colorectal cancer. These findings allow for better risk stratification of patients undergoing colorectal cancer screening and could inform screening test selection.