Abstract Colorectal cancer (CRC) is a heterogeneous disease, and different subsets of CRC probably have different precursor lesions. Some colorectal polyps are considered precursors to CRC and are broadly classified into: 1) adenomas, including tubular, tubulovillous, and villous adenomas, and 2) serrated polyps, including hyperplasic polyps (HPs), sessile serrated polyps (SSPs), and traditional serrated adenomas (TSAs). Adenomas are established CRC precursors, particularly for advanced lesions ≤ 10 mm, with villous components, or high-grade dysplasia. HPs were previously considered innocuous lesions, but new data suggests that a subset of these may progress to CRC via advanced serrated lesions, such as SSPs and TSAs. To gain insight into the factors associated with initiation and progression for divergent CRC pathways, we tested differences in risk factors for polyps according to histologic type, anatomic site, and severity. Participants, aged 20-79, were sampled from enrollees of Group Health, a large healthcare provider in Washington State, who underwent colonoscopy for any indication between 1998 and 2007. Participants completed a structured interview, and all lesions received a standardized pathology review. We evaluated 1,469 cases, including 628 with adenomas only, 594 with serrated polyps only, and 247 with both types of polyps, as well as 1,037 controls who were free of colorectal pathologies during colonoscopy. We focused on the following risk factors: age, sex, race, education, body mass index (BMI), NSAID use, family history of CRC, hormone use among women, endoscopy screening, smoking status, alcohol use, and physical activity. Multivariable polytomous logistic regression models were used to compare case groups to controls adjusted for all factors. Our study confirmed previously described risk factors for adenomas: age, male sex, obesity, never use of NSAIDs, no previous endoscopy, and physical inactivity. Risk factors for serrated polyps were smoking and never use of estrogen-only therapy. There were statistically significant differences in the association between several risk factors and polyp type. Male sex (P<0.001) and smoking (P<0.001) were stronger risk factors for adenomas than for serrated polyps. With respect to anatomic site, smoking (P=0.05) and never use of estrogen-only therapy (P=0.01) were stronger risk factors for distal/rectal than for proximal adenomas. Smoking (P<0.001) and BMI (P=0.01) were stronger risk factors for distal/rectal than for proximal serrated polyps. With respect to lesion severity, lack of prior endoscopy was a stronger risk factor for advanced adenomas (P<0.001) than for non-advanced adenomas, and higher education level was a stronger risk factor for SSP/TSAs than for HPs (P=0.02). These findings may suggest carcinogenic mechanisms that distinguish the adenoma-carcinoma pathway and the serrated pathway to CRC. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 645. doi:1538-7445.AM2012-645
Read full abstract