Results from prospective studies examining associations between sleep duration and weight gain have been mixed. Melatonin has been hypothesized to mediate the association between sleep duration and weight/body composition. In cross-sectional studies, aMT6s has been shown to be inversely associated with weight/body fat percentage. We examined associations between baseline sleep duration, insomnia status, aMT6s levels with weight/body fat percentage through 6 years, utilizing a subset 690 women who participated in a breast cancer case-control study nested within the WHI-OS. Multi-variable and mixed-effects regression was used to calculate beta-coefficients and 95% confidence intervals. Cross-sectional analyses showed urinary aMT6s levels were inversely associated with BMI and body fat percentage. No associations were observed between sleep patterns and measures of adiposity. The prospective relationship between urinary aMT6 levels and weight/body fat percentage was complex. Age-adjusted mixed models show an association in the interaction term between year and aMT6s with body fat percentage (βinteraction:0.09, pinteraction:0.16, p=0.07), demonstrating the influence of baseline aMT6s and time on changes in outcome. Women with higher baseline aMT6s had a trajectory of increased body fat percentage and weight gain steepest between baseline and year 3, whereas women with lower baseline aMT6 levels had a trajectory of decreased body fat percent and weight between year 3 and year 6. The prospective association between melatonin levels and adiposity measures was unexpected. Future studies with objective measures of sleep and repeated measures of melatonin may shed light of possible explanations for our findings.