Background: Bisphosphonates are antiresorptive drugs commonly used to treat osteoporosis. Once treatment is initiated, it is usually continued. Small studies have shown either bone marrow density (BMD) increases or maintenance with continued treatment compared with slight BMD decreases or maintenance with discontinuation. Long-term studies of these medications, as well as fracture risk data, are lacking. The Fracture Intervention Trial (FIT) studied the long-term effects of alendronate (Fosamax) on BMD and fracture risk in postmenopausal women.

The Study: This randomized, blinded, placebo-controlled trial followed women 55 to 81 years of age for an average of 3.8 years and then afterward for optional open-label treatment. Participants were randomized to placebo or to lower or higher doses of alendronate, depending on whether vertebral deformity was present. This study was in the FIT Long-term Extension (FLEX) part of the trial, which randomized patients to 5- or 10-mg dosing for a 10-year total or to discontinuation after five years. FLEX participants had completed at least three years of active alendronate treatment and had a hip T-score greater than −3.5 at FLEX baseline. All women in the FLEX trial received alendronate 10 mg per day, 5 mg per day, or placebo for five years. Patients also were offered calcium and vitamin D supplementation.

Patients received telephone calls every three months to encourage adherence and to discuss problems, and physical examinations were performed annually. Baseline dual energy x-ray absorptiometry was performed at various bony sites. Biochemical markers were obtained and analyzed at the end of the study. Spine radiography was performed to confirm suspected fractures. All occurring fractures were documented.

Results: Of the 1,099 women enrolled in FLEX, 437 were randomized to receive placebo, 329 to receive alendronate 5 mg per day, and 333 to receive alendronate 10 mg per day. More than one third of patients had vertebral fractures, and 60 percent had a history of postmenopausal clinical fracture. At five years, women taking either dose of alendronate maintained a better total hip BMD (1.02 percent BMD decline) compared with placebo (3.38 percent BMD decline). Similar patterns were found at other bony sites. Women taking alendronate for 10 years had greater BMD gains than women taking placebo following five years of alendronate use.

Markers of bone turnover rose gradually in patients discontinuing alendronate, whereas those continuing treatment had stable markers. No statistically significant difference was noted among groups in the overall occurrence of clinical and nonvertebral fractures. Nineteen percent of the placebo group sustained nonvertebral fractures, compared with 18.9 percent of the alendronate group. There was a statistically significant decrease in risk of vertebral fractures in the alendronate users (2.4 percent) compared with the placebo group (5.3 percent).

Conclusion: Women who continued alendronate treatment after the initial five years had higher hip and spine BMD, lower bone remodeling, and a lower rate of clinical vertebral fractures; there was no difference in nonvertebral fractures. The authors conclude that, unless patients are at high risk of clinical vertebral fracture, those who have taken alendronate for five years can discontinue the drug for up to five years without a significant increase in fracture risk.