Studies Probe Possible Link Between Bisphosphonates and Femoral Fractures

Conflicting data on whether long-term bisphosphonate therapy may cause a rare type of low-energy femoral fracture are creating challenges for physicians and patients in weighing the potential risks of such medications against their benefits in reducing hip fractures. But the US Food and Drug Administration (FDA) and other groups say the data so far do not support changes in prescribing practices.

Case reports describing low-energy atypical subtrochanteric femur fractures in women who have taken bisphosphonates for several years, as well as preliminary data suggesting reductions in bone quality with prolonged use, have raised concerns about the safety of the drugs. So far, however, an ongoing review of data from the drugs' manufacturers and other studies by the FDA has not found an increased risk of such fractures associated with the drugs, according to a statement from the agency on March 10.

Additionally, an analysis of 3 placebo-controlled trials published online on March 24 found such atypical femoral fractures to be rare in women taking bisphosphonates (with a combined rate of 2.3 per 10 000 patient-years in the trials, based on 12 fractures in 10 women), and found no significant increase in risk of such fractures in women taking the drugs (Black DM et al. N Engl J Med. 10.1056/nejmoa1001086). To better understand these events, the FDA said it will continue its review and will consult with an international, multidisciplinary task force created by the American Society for Bone and Mineral Research (ASBMR) to assess the data. A report from the task force is expected later this year.

In the meantime, physicians and patients should understand that the risk of atypical subtrochanteric femur fractures is small compared with the risk of more common types of hip fracture associated with osteoporosis, which bisphosphonates have been demonstrated to reduce. Elizabeth Shane, MD, cochair of the ASBMR task force and professor of medicine at Columbia University in New York City, explained that of 100 hip fractures, approximately 2 to 4 would occur in the subtrochanteric femur region, and that patients taking bisphosphonates account for only one-third of those.

“It's very important that physicians and patients don't rush out and stop bisphosphonates because of the concern about these fractures, and perhaps increase their risk of the more common types of hip fracture,” said Shane.

The agency began a safety review of the drugs in 2008, spurred by the emergence of several case reports of women experiencing atypical subtrochanteric femur fractures after several years of taking bisphosphonates. The FDA's assessment included data from an analysis of a cross-sectional study of nearly 12 000 women with proximal femur fractures, as well as data from a matched cohort study of 5187 individuals who had used alendronate and had any nonhip fracture (including femoral fractures) and 10 374 controls who had any nonhip fractures (including femoral fractures) but who were never exposed to bisphosphonates (Abrahamsen B et al. J Bone Miner Res. 2009;24[6]:1095-1102).

In the cross-sectional study, the scientists found that 7% of the patients with atypical femur fracture and 7% of the patients with hip fracture had taken alendronate. The cohort study found a significantly increased risk of subtrochanteric fracture (hazard ratio [HR], 1.64; 95% confidence interval [CI], 1.05-2.58) in patients taking alendronate compared with controls and increased risk of hip fracture in patients taking alendronate compared with controls (HR, 1.50; 95% CI, 1.26-1.79). They also found that patients who were highly adherent to alendronate use had a lower risk of either type of fracture. Because of the similar rates of subtrochanteric and hip fracture in patients taking alendronate, the authors concluded that atypical femoral fractures are likely related to osteoporosis itself and not the drug.

However, results from a pair of studies presented at the American Academy of Orthopaedic Surgeons’ annual meeting in New Orleans in March have raised new questions. These studies suggested that long-term use of bisphosphonates may reduce bone quality in ways that might lead to atypical femoral fractures in a subset of women. Brian Gladnick, a medical student at Weill Cornell Medical College in New York City, presented results from a prospective pilot study of bone samples taken from 21 postmenopausal women who presented to the emergency department with femoral fractures; 12 had been exposed to bisphosphonates and 9 had not been exposed. Gladnick's team found reduced heterogeneity of bone tissue in the treated group, which may lead to more brittle bone.

The second study, presented by Anthony Ding, a medical student at the Columbia University College of Physicians and Surgeons, compared the bone of 111 postmenopausal women with osteoporosis, including 61 treated with bisphosphonates for at least 4 years, and 50 controls taking calcium and vitamin D. As expected, the researchers found that the structural integrity of bone in the control group deteriorated over time. In the treated group, bone structure improved during the first 4 years of treatment, but later began to deteriorate.

The secondary analysis, funded by Merck and Novartis, examined data from 3 pharmaceutical company–funded placebo-controlled trials involving 14 195 women. In this analysis, Dennis M. Black, PhD, of the University of California, San Francisco, and colleagues failed to find a significantly increased risk of these atypical femoral fractures, though the authors note it may have been inadequately powered to do so. The authors found that the relative hazard for alendronate use vs placebo was 1.03 (95% CI, 0.06-16.46) in the Fracture Intervention Trial (FIT), which followed up 6549 women for 3 to 4.5 years; a 1.33 relative hazard (95% CI, 0.12-14.67) for alendronate use in the Fit Long-Term Extension Trial (FLEX), which randomly assigned 1099 women from the alendronate group of FIT to 1 of 2 doses alendronate or placebo and followed them up for an additional 5 years; and a 1.50 relative hazard (95% CI, 0.25-9.00) for zoledronic acid in the Health Outcomes and Reduced Incidence with Zolendronic Acid Once Yearly (HORIZON) Pivotal Fracture Trial (PFT), which followed up 7736 women for 3 years. Only 12 such fractures were identified in 10 women from the 3 studies.

The authors concluded that such adverse events are rare, and do not outweigh the substantial benefits of preventing hip fractures. For example, the authors calculated that treating 1000 women with bisphosphonates for 3 years would prevent 100 fractures, while the hypothetical risk of atypical femur fractures would be about 0.3 per 1000 women treated for 3 years. However, the authors note the limitations of the study, such as wide 95% CIs and the exclusion of women taking certain types of concurrent medications. Moreover, only 1 of the 3 studies followed up a total of 1009 women for longer than 4.5 years.

Until these concerns are resolved, the FDA recommends that physicians be aware of the potential risk of a femoral fracture and discuss with their patients the benefits of the drugs as well as any known or potential risks. Physicians should also continue to follow the directions on the drugs' labels. Patients are urged by the FDA to not stop taking their medications unless their physician instructs them to discontinue and to discuss any concerns they have about their medication with their physician. Patients should also report any emergent hip or thigh pain to their physician. Both physicians and patients are advised to report adverse events, including fractures, to the agency's MedWatch program (https://www.accessdata.fda.gov/scripts/medwatch/medwatch-online.htm).

Shane said the FDA's recommendations are reasonable, but that it would also be prudent for physicians to reevaluate patients who have been on long-term bisphosphonate therapy to determine why they began the therapy, how their condition is progressing, and whether markers of bone turnover are overly suppressed. She explained that recommendations for when to initiate bisphosphonate therapy have changed, so that some patients who started taking the drugs years ago would not be recommended as candidates for such therapy today. For example, younger individuals who have osteopenia but are at low risk of fracture most likely would not be treated with bisphosphonates now.

More information about the FDA's review and recommendations are available online at http://www.fda.gov/Drugs/DrugSafety/PostmarketDrugSafetyInformationforPatientsandProviders/ucm203891.htm.