Recent cohort studies demonstrated reduced breast cancer risks among women with a history of fractures or low bone mineral density (BMD). In the Study of Osteoporotic Fractures, each standard deviation increase in distal radius BMD was associated with a 50% increased risk over three years of follow-up, while in the Framingham study, women in the highest quartile of metacarpal bone mass had a 3.5-fold higher risk than women in the lowest quartile. The impact of the severity and timing of bone loss on risk has not yet been investigated, and the extent to which other risk factors (family history, anthropometric factors, physical activity, and exogenous hormones) modify the relationship with BMD is unknown.

To elaborate on these research questions, we are conducting a follow-up study of 22,695 postmenopausal women who volunteered for the Fracture Intervention Trial (FIT), a trial of the bone-enhancing drug alendronate. This large cohort includes extensive baseline information on major breast cancer risk factors, and thus is ideal for evaluating potential interactions with BMD and the effects of BMD on other cancer sites. Endometrial cancer has been reported to occur more frequently among women with a history of fracture, but no previous studies have specifically investigated its relationship to BMD.

We are investigating whether BMD of the proximal femur predicts breast cancer risk; whether breast cancer risk factors among postmenopausal women modify the relationship with BMD; whether BMD predicts endometrial or other cancers; and whether measurable biomarkers offer further etiologic clues about BMD and cancer risk.

We have contacted the surviving members of FIT to ascertain incident cancers. Risk factors and fracture history are being updated through a self-administered questionnaire. To supplement the serum samples collected at baseline, we are using a nested case-control study approach to collect buccal cell specimens, which may be useful for measuring a variety of biomarkers, including endogenous hormones and genetic polymorphisms involved in either bone growth (e.g., vitamin D receptor) or hormone metabolism (e.g., CYP17, COMT). Retrieval of operative and pathology reports is being used to validate self-reported cancers. The social security numbers and contacts names provided by FIT participants when they completed the baseline questionnaire are facilitating comprehensive follow-up and a National Death Index search for those who cannot be located. The baseline data, the established cooperation of this study population, and the collection of additional biospecimens should enable this study to answer important questions about BMD in breast and endometrial cancers.

Recent cohort studies demonstrated reduced breast cancer risks among women with a history of fractures or low bone mineral density (BMD). In the Study of Osteoporotic Fractures, each standard deviation increase in distal radius BMD was associated with a 50% increased risk over three years of follow-up, while in the Framingham study, women in the highest quartile of metacarpal bone mass had a 3.5-fold higher risk than women in the lowest quartile. The impact of the severity and timing of bone loss on risk has not been investigated, and the extent to which other risk factors (family history, lifestyle, and exogenous hormones) modify the relationship with BMD is unknown.

To elaborate on these research questions, we conducted a follow-up study of postmenopausal women who volunteered for the Fracture Intervention Trial (FIT), a trial of the bone-enhancing drug alendronate. The BFIT follow-up study includes 15,595 of the 22,695 FIT volunteers. Surviving members of FIT were contacted to ascertain incident cancers and to provide updated risk factor and fracture history through a self-administered questionnaire. To supplement baseline serum samples, we used a nested case-control approach to collect buccal cell specimens for biomarker measurement, including endogenous hormones and genetic polymorphisms involved in either bone growth (e.g., vitamin D receptor) or hormone metabolism (e.g., CYPI7, COMT). Operative and pathology reports were used to validate self-reported cancers. The social security numbers and contact names provided by FIT participants at baseline facilitated comprehensive follow-up and a National Death Index search for those who could not be located.

This large cohort includes extensive baseline information on major breast cancer risk factors, and thus is ideal for evaluating potential interactions with BMD and the effects of BMD on other cancer sites. Endometrial cancer has been reported to occur more frequently among women with a history of fracture, but no previous studies have investigated its relationship to BMD. We are investigating whether proximal femur BMD predicts breast cancer risk; whether breast cancer risk factors among postmenopausal women modify the relationship with BMD; whether BMD predicts cancer risk; and whether biomarkers offer etiologic clues about BMD and cancer risk. Currently, we are examining: 1) the relationship of serum adipocytokines to endometrial cancer risk, and 2) the relationships of serum estrogens and metabolites to postmenopausal breast cancer risk. The baseline and follow-up data and the collection of additional biospecimens should enable us to answer important questions about BMD and other cancers.

• INCLUSION CRITERIA:

Women previously enrolled in FIT and provided informed consent.

The B FIT study included 15,595 of the 22,695 who were screened for participation in the Fracture Intervention Trial (FIT), a randomized clinical trial originally designed to test whether alendronate, a bispohosphonate, would reduce the rate of fractures in women with low bone mineral density (Black et al. 1993). The FIT clinical trial (1992-1998) enrolled approximately 6000 volunteers who had low BMD. The B FIT follow-up study includes all women (regardless of baseline BMD) who volunteered for FIT. In 1992-1993, postmenopausal women (ages 55-80) completed an extensive questionnaire, donated a baseline blood sample, underwent a bone mineral density scan, and provided clinical examination data. As part of the B FIT study, these women were followed (median 10.3 years) to ascertain incident cancer outcomes and incident fractures through the period of 2001-2004.