Guide your patients through the FDA warning concerning bisphosphonates

A few weeks ago the FDA issued a warning on its website that “there is a possible risk of a rare type of thigh bone (femoral) fracture in people who take drugs known as bisphosphonates to treat osteoporosis.

“FDA says the optimal duration of bisphosphonates treatment for osteoporosis is unknown — an uncertainty the agency is highlighting because these fractures may be related to use of bisphosphonates for longer than 5 years.”

The notice included an Advice for Consumers section that reads:

“If you are currently taking bisphosphonates for osteoporosis, FDA advises that you:

Keep taking your medication unless you are told to stop by your health care professional.
Read the medication guide. It will describe the symptoms of an atypical femur fracture. The guide also advises you to notify your health care professional if you develop symptoms.
Tell your health care professional if you develop new hip or thigh pain (commonly described as dull or aching pain), or have any concerns with your medications.
Report any side effects with your bisphosphonate medication to FDA's MedWatch program.”

This mixed message to “keep taking your medicine” and “tell your health care professional if you develop new hip or thigh pain” is hard to fathom.

What are the data? An upcoming issue of the Journal of Clinical Endocrinology and Metabolism provides data on subtrochanteric and diaphyseal femur fractures from Denmark on 39,567 patients without prior hip fracture who began alendronate (Fosamax, Merck) between 1996 and 2005 compared with 158,268 age- and sex-matched controls. Femur fractures occurred at a rate of 31 per 10,000 patient-years in women and men and 13 per 10,000 patient-years in untreated women and 6 per 10,000 person-years in untreated men. Patients who had received therapy for 9 years had fracture rates similar to those who stopped therapy after only 3 months. In treated women, hip fractures (not subtrochanteric or diaphyseal) occurred at a rate of 164 per 10,000 patient-years vs. 99 per 10,000 patient-years in untreated women. In men, the rates were 133 and 49 per 10,000 patient-years.

This article is available online and I have read it in detail. The data are compelling, but there are seeming “mitigating” circumstances. For example, comorbid conditions leading to an increased risk for fracture were more prevalent in alendronate-treated patients prior to initiating therapy than in the untreated control group. In particular, steroid use was about 10-fold greater in the alendronate-treated group.

Where does that leave us and our patients? FDA has issued warnings for bisphosphonates, estrogen, raloxifene (Evista, Eli Lilly) and teriparatide (Forteo, Eli Lilly), with only two osteoporosis therapies — calcitonin and denosumab (Prolia, Amgen) — escaping these warnings. Should we not use any of the drugs for which the FDA has issued warning notices? Absolutely not! Adequate vitamin D and calcium intake along with regular exercise are a must for minimizing the risk for osteoporosis-related fracture, but there is scant evidence that this approach is sufficient to minimize fracture risk to the same extent as the FDA-approved therapies.

As I have previously noted, my practice is to consider a drug-free holiday for patients in whom consecutive bone density studies 2 years apart demonstrate that bone density is unchanged at the lumbar spine and total femur sites. This is most likely to happen between years 2 and 4 of therapy, but there is a problem: Far too often serial DXA is not performed reliably. It is imperative that you demand the bone density testing site provide you with all of the data concerning the area of the bone that is scanned such that you can compare apples to apples with respect to any changes in bone mineral content. You really cannot make sound judgments if the only data you receive is a T-score and a bone mineral density. This problem is amplified if you are not doing the DXA study in your own practice because the patient’s insurance coverage may send them to different testing sites for serial measurements. Even if the testing sites are using the same make and model DXA system direct comparisons are fraught with error unless the two systems have been calibrated against each other. This rarely happens.

In conjunction with the bone density I monitor a marker of bone resorption (serum CTX or NTX — the decision regarding which marker to use is insurance coverage driven) and bone formation (serum P1NP). The baseline data have limited, if any, influence on the original treatment decision. A repeat marker assay at 3 months should show a significant reduction in the resorption marker in patients treated with antiresorptive drugs and is a reliable guide that the therapy is being taken properly and regularly. With teriparatide, a bone formation stimulation drug, P1NP at 3 months provides equally reassuring data. As long as the patient is doing well and not reporting any issues with therapy there is little need for serial monitoring of these markers.

Markers play a big role in my decision to recommend a drug-free holiday when the BMD has been stable for at least 2 years. (This does not apply to teriparatide therapy since it is only approved for use for 2 years). Most often I find serum CTX or NTX at or even below the lower limit of the reference interval. Continuing to inhibit bone resorption does not seem necessary to me, but I must quickly add that there are very limited data that continuing therapy will do harm, despite the FDA warning.

How long should a drug-free holiday last? To my knowledge, there are no data that provide a solid answer. During the holiday, I monitor the marker every 6 months (an arbitrary decision) with plans to resume therapy as the markers begin to rise or the holiday lasts for 2 years, whichever comes first. During the patient’s drug-free holiday, adequate calcium and vitamin D intake should be maintained, as well as regularly checking that fall prevention is in place.

Is there a better alternative? Sure. If bone resorption has been inhibited to a point where BMD is no longer increasing, it is time to initiate therapy with a formation stimulation drug — teriparatide being the only one available in the United States. However, it may be difficult in getting that covered by the patient’s insurance; try to speak to someone in the company who has at least minimal understanding of metabolic bone disease.

After all of the above what do we tell our clearly and appropriately concerned patients?

Let them know that you have been following this closely and feel comfortable with the FDA-approved osteoporosis therapies. If you have carefully read the literature and don’t feel comfortable with bisphosphonates choose something else.
In the previously untreated patient who is healthy but has a FRAX score that indicates therapy is appropriate, there seems little contraindication to bisphosphonate therapy.
The one exception to that, based on the article I have referred to above, is current and probably also previous exposure to steroid therapy for any extended period of time. Initiating therapy with teriparatide, FDA-approved for treatment of steroid-induced osteoporosis, seems most logical. Again, you may have to fight the insurance company, but be persistent and armed with appropriate scientific data. Patients will balk at a daily injection regimen but once started I have found patient acceptance and adherence to be better than with oral bisphosphonates. Be sure to remind them that they are not weaning male rats and their risk of osteosarcoma is negligible if not nil. Tell them that before they read the black box warning label.
Reassure the patient that you will carefully monitor their response to treatment with blood tests early on and serial DXA over time.
Emphasize calcium, vitamin D, exercise and fall prevention.
Advise your patient not to rely on non-FDA-approved therapies that they see on TV, the internet or read in magazines.

Abrahamson B. J Clin Endocrinol Metab. 2010;doi:10.1210/jc.2010-1571.