Telomere length differences between colorectal polyp subtypes: a colonoscopy-based case-control study

Sheetal Hardikar, Andrea N. Burnett-Hartman, Amanda I. Phipps, Melissa P. Upton, Lee-Ching Zhu, Polly A. Newcomb

Short telomeres have been associated with increased risk of many cancers, particularly cancers of the gastrointestinal tract including esophagus and stomach. However, the association between telomere length (TL) and colorectal cancer and its precursors, colorectal polyps, is not clear. We investigated the relationship between TL and risk of colorectal polyp subtypes in a colonoscopy-based study in western Washington. Participants were 35-79 year-old enrollees at an integrated health care system, who underwent a colonoscopy between 1998 and 2007 (n = 190), completed a self-administered questionnaire, provided blood samples, and were distinguished as having adenomas, serrated polyps, or as polyp-free controls through a standardized pathology review. Telomere length (T) relative to a single copy gene (S) was measured in circulating leukocytes from stored buffy coat samples using quantitative polymerase chain reaction. Multivariable polytomous logistic regression was used to compare case groups with polyp-free controls and other case groups; adjusted odds ratios (OR) and 95% confidence intervals (CI) were estimated. TL in the shortest tertile (T/S ratio < 0.58) was associated with increased risk of adenomas and serrated polyps [OR (95%CI) were 1.77(0.81-3.88) and 2.98(1.15-7.77), respectively). When evaluated by lesion severity within each pathway, short TL was more strongly associated with advanced adenomas and sessile serrated polyps [OR (95% CI) = 1.90(0.76-4.73) and 3.82(0.86-16.86), respectively], although the associations were not statistically significant. Our results suggest that short TL may be associated with an increased risk of colorectal polyps in both the adenoma-carcinoma and serrated pathways. The risk was particularly notable for sessile serrated polyps, although the association was not statistically significant and sample size was limited.