Use of age and BMD to predict fracture risk in men on androgen deprivation therapy

Abstract

9517 Background: Androgen deprivation therapy (ADT) decreases bone mineral density by reducing estrogens to castrate levels and, as a result, increases fracture risk. We recently completed a two-year trial in 1382 men in which we examined the ability of toremifene to reduce fracture risk in men on ADT. Herein we describe analyses of the placebo group to assess the baseline characteristics associated with new fractures. Methods: We conducted a randomized double blind placebo controlled trial in 1382 men with histologically confirmed prostate cancer on ADT. Entry criteria included age ≥ 50 years, continous ADT for 6 months or longer or intermittent ADT for 12 months or longer. Subjects on intermittent ADT at enrollment had to remain on continuous ADT for their duration on study. Subjects were randomized to receive either 80mg toremifene citrate daily or matching placebo. The primary end point was the incidence of new morphometric vertebral fractures. Secondary endpoints included bone mineral density, clinical fragility fractures, bone turnover markers, lipid profile, hot flashes, and gynecomastia. Results: The modified intent to treat (MITT) population included subjects who took study medication and had an on-study radiograph. There were 467 subjects in the placebo MITT population. To identify factors associated with fracture risk in men on ADT we compared placebo subjects who suffered a fracture or during the first year on study suffered 7% or greater bone loss with those placebo subjects who did not. Baseline characteristics included: BMD (spine, hip, femoral neck), Age, history of fracture, ADT duration, bone turnover markers, and race. Logistic regression models of the probability of fracture/bone loss as a function of country showed that each of BMD at all sites, age, race, CTX, and history of previous fracture independently predicted fracture/7% bone loss. When all characteristics were analyzed in a multivariable model lower spine BMD (p=0.006) and older age (p=0.018) were significantly associated with incident fractures. Conclusions: In prostate cancer patients on ADT older age and lower baseline spine BMD were associated with a greater risk of fracture in untreated patients. [Table: see text]