Objective: Using a nested case–referent design we evaluated the relationship between plasma levels of six carotenoids, alpha-tocopherol, and retinol, sampled before diagnosis, and later breast cancer risk.
Methods: In total, 201 cases and 290 referents were selected from three population-based cohorts in northern Sweden, where all subjects donated blood samples at enrolment. All blood samples were stored at −80 °C. Cases and referents were matched for age, age of blood sample, and sampling centre. Breast cancer cases were identified through the regional and national cancer registries.
Results: Plasma concentrations of carotenoids were positively intercorrelated. In analysis of three cohorts as a group none of the carotenoids was found to be significantly related to the risk of developing breast cancer. Similarly, no significant associations between breast cancer risk and plasma levels of α-tocopherol or retinol were found. However, in postmenopausal women from a mammography cohort with a high number of prevalent cases, lycopene was significantly associated with a decreased risk of breast cancer. A significant trend of an inverse association between lutein and breast cancer risk was seen in premenopausal women from two combined population-based cohorts with only incident cases. A non-significant reduced risk with higher plasma α-carotene was apparent throughout all the sub-analyses.
Conclusion: In conclusion, no significant associations were found between plasma levels of carotenoids, α-tocopherol or retinol and breast cancer risk in analysis of three combined cohorts. However, results from stratified analysis by cohort membership and menopausal status suggest that lycopene and other plasma-carotenoids may reduce the risk of developing breast cancer and that menopausal status has an impact on the mechanisms involved.