Menopausal hormone therapy and risks of colorectal adenomas and cancers in the French E3N prospective cohort: true associations or bias?

Abstract

Menopausal hormone therapy (MHT) use has been quite consistently associated with a decreased colorectal cancer risk; however, data regarding adenomas, types of MHT, or tumour site are limited. We investigated associations between MHT use and colorectal adenoma and cancer risks within the prospective E3N cohort study. In the adenoma study, we analyzed the 13,402 postmenopausal women who underwent a colonoscopy during follow-up (1993-2002), including 1,109 who were diagnosed a first colorectal adenoma. In the cancer study, 525 out of 77,375 postmenopausal women developed a colorectal cancer as the first malignant tumour during follow-up (1992-2008). Ever use of MHT was not significantly associated with colorectal adenoma risk [Hazard Ratio (HR) = 1.13, 95 % Confidence Interval (CI) = 0.99, 1.29], nor with colorectal cancer risk (HR = 0.86, 95 % CI = 0.71, 1.04). However, ever use of estrogens alone was associated with risks of colorectal adenoma and cancer in opposite directions (HR = 1.22, 95 % CI = 1.05, 1.41 and HR = 0.72, 95 % CI = 0.56, 0.94 respectively). Associations were strongest for non-advanced, and for left colon adenomas. Duration, recency of use, type of progestagen or route of estrogen administration were not associated with colorectal tumour risk, nor was the use of estrogen-progestagen MHT. The association between estrogens alone and colorectal cancer risk was only observed in women with previous colonoscopy(ies), but the P value of the interaction test between estrogens alone use and history of colonoscopy was 0.06. Our data suggest a complex relationship between MHT use, colorectal tumour risk and screening strategies, which deserves further investigations in other settings.