SUMMARY: In this prospective study in 2,137 perimenopausal and early postmenopausal women who were followed over a 13.1-year period of time, we observed no association between bone mineral density measured at the beginning of menopause and the subsequent risk of breast cancer. INTRODUCTION: This study aimed to investigate the relationship between BMD and the risk of breast cancer (BC) in young postmenopausal women. METHODS: As part of a clinical research program, 2,137 women who were perimenopausal or within their 5 first postmenopausal years were scanned between 1988-1990 and reviewed on average 13.1 years after their initial examination. Ninety-eight incident BC cases were recorded throughout the follow-up. RESULTS: Women with incident BC significantly differed from those who had never had BC with regard to age at menarche, age of birth of 1st child, familial history of BC and postmenopausal hormone therapy (PHT) use. There was no significant difference between the two groups for baseline DXA of the spine. There was a trend for BC cases for having lower femoral neck BMD compared to women without BC. However, women with low BMD were more likely to have taken PHT by the end of the study. In Cox multivariate analyses the relationship between BC risk and femoral neck BMD no longer existed. CONCLUSIONS: There was no relationship between BMD measured within the first postmenopausal years and the risk of BC, which makes unlikely the possibility of using BMD as a predictor factor for BC in early postmenopausal women.