AbstractBackground:Exposure to higher levels of melatonin may be associated with lower breast cancer risk, but epidemiologic evidence has been limited. We examined the relationship in a case–control study nested within the Diagnostisch Onderzoek Mammacarcinoom (DOM) study and conducted a meta-analysis of prospective studies.Methods:Concentrations of 6-sulfatoxymelatonin (aMT6s) in prediagnostic first morning urine voids were measured in 274 postmenopausal women diagnosed with breast cancer and 274 matched controls from the DOM study. Conditional logistic regression models were used to estimate multivariable adjusted ORs of breast cancer for thirds of aMT6s. Meta-analysis of this and previous prospective studies of urinary melatonin with breast cancer risk estimated the inverse-variance weighted averages of study-specific log RRs of breast cancer for the highest versus lowest levels of aMT6s.Results:In the DOM study, the ORs of breast cancer for the middle and highest versus lowest thirds of aMT6s were 0.70 [95% confidence interval (CI), 0.45–1.09] and 0.72 (95% CI, 0.44–1.19), respectively. In the meta-analysis of the DOM study with six previous studies (2,296 cases), RR of breast cancer for the highest versus lowest levels of aMT6s was 0.87 (95% CI, 0.76–1.01).Conclusions:Results from the DOM study, together with the published prospective data, do not support a strong association of melatonin with breast cancer risk.Impact:This study adds to the relatively scarce prospective data on melatonin in relation to breast cancer risk. The totality of the prospective evidence does not clearly show an association between melatonin and breast cancer risk, but further data are needed to be able to exclude a modest association.