Chapter 13 – Osteoporosis: Prevention and Treatment

Kathryn E. Ackerman, M.D. M.P.H., Fellow, Endocrinology,Diabetes, Hypertension Division
Brigham and Women's Hospital, 221 Longwood Avenue, Boston, MA 02115 KEACKERMAN@partners.org
Meryl S. LeBoff, M.D., Director, Skeletal Health and Osteoporosis Center and Bone Density Unit
Endocrine, Diabetes, Hypertension Division,Brigham and Women's Hospital
221 Longwood Avenue,Boston, MA 02115  MLEBOFF@PARTNERS.ORG

Last Updated November 20, 2008

TO OBTAIN A DOWNLOAD OF THIS CHAPTER IN WORD OR PDF FORMAT, CLICK HERE

INTRODUCTION

Osteoporosis is a major global health problem, resulting in greater than 200 million osteoporotic fractures worldwide each year, including 1.6 million hip fractures.1 In the United States in 2005, there were an estimated two million osteoporosis-related fractures, including approximately 547,000 vertebral fractures, 297,000 hip fractures, 397,000 wrist fractures, 135,000 pelvic fractures, and 675,000 fractures at other sites. The total number of such fractures in the United States is projected to reach over three million by 2025.2

Although only about one-quarter to one-third of vertebral fractures are clinically evident, they can lead to loss of height, kyphosis, restrictive lung disease, abdominal distension, and increased mortality. Hip fractures are the most devastating fractures associated with osteoporosis. Approximately 50% of patients who sustain a hip fracture lose the ability to walk independently. Up to 24% of women and 30% of men die within the first year.3-5

Despite the health consequences of osteoporosis and the availability of effective treatments, it is under-diagnosed and under-treated. For example, although 90% of patients with hip fractures have osteoporosis, in 2007 only 20% of patients with fragility fractures were evaluated and treated.6, 7 In a retrospective study of patients with hip fractures, less than 15% of subjects were diagnosed and less than 13% were treated with medications for osteoporosis, including calcium and vitamin D.8 Yet vitamin D insufficiency is present in up to 96% of hip fracture patients.9 Fracture patients require evaluation of secondary causes and treatment of osteoporosis to help prevent subsequent fractures.

The preceding chapters summarize the pathogenesis and the clinical evaluation of osteoporosis. This chapter will review established therapeutic options and new approaches for the prevention and treatment of osteoporosis. Strategies include both lifestyle and medical approaches to enhance bone strength.

EXERCISE

While pharmacological therapies are much of the focus of this chapter, exercise has been shown to have a significant impact on bone mineral density (BMD). Physical activity helps to maximize BMD during childhood, adolescence and early adulthood, maintain bone mass through mid-life, diminish bone loss with aging, and improve stability and strength to minimize falls and fractures in the elderly.10 However, these benefits come from slow skeletal adaptations to training over time. Because it takes three to four months to complete the bone remodeling cycle of resorption, formation, and mineralization, a minimum of at least six to eight months of an exercise intervention is likely required to achieve a change in bone mass that is quantifiable.10, 11

In a study of premenopausal women, site-specific exercise was performed to determine whether such a program could be used for site-specific BMD improvement, as measured by dual x-ray absorptiometry (DXA). Thirty-five exercising women were randomly assigned to either lower body resistance plus jump exercises or lower and upper body resistance plus jump exercises. These women were compared with 24 age-matched controls. Excluding exercisers with low compliance, at 12 months lumbar spine BMD was significantly greater in the group that incorporated upper extremity exercise than in the other exercise group and controls. Trochanteric BMD was significantly greater in both exercise groups than in the controls.12

In older adults, after peak BMD has been achieved, the goal is to minimize bone loss and fractures. In a 12- month study of 53 postmenopausal women, each was randomly assigned to a strength training program or a power training program. Both groups performed resistance training, some gymnastics, and a home program, but the strength training group lifted weights with slow movements and the power training group lifted weights with fast movements. After 12 months, DXA measurements showed that the power trainers maintained BMD at the spine and total hip, but the strength trainers significantly lost BMD at these sites. Thus, the power training program better minimized bone loss.13

In a 2002 Cochrane Review on exercise for preventing and treating osteoporosis in postmenopausal women (along with its 2004 revision), aerobics, weight bearing and resistance exercises were all shown to have a positive effect on the BMD of the spine in postmenopausal women. Walking also demonstrated improved BMD at the hip.14, 15

It has been established that exercise has a favorable impact on bones. New tools that are undergoing testing in randomized controlled trials, such as vibratory platforms, have even shown benefits in muscle strength and hip BMD.16, 17 However, the optimal exercise, duration, frequency, and percent effort still need to be elucidated. While further research is being conducted on these issues, the best recommendation may be for patients to find a weight-bearing activity they enjoy and will continue to incorporate into their routine three or more days a week.

CALCIUM

Adequate calcium intake is necessary to prevent calcium mobilization from the bone where 99% of calcium is stored. The effects of calcium supplementation on bone depend on age, menopausal status, calcium intake, and vitamin D sufficiency. Increased calcium intake is necessary during acquisition of peak bone mass and with advancing age. Calcium supplementation is ineffective or minimally effective for prevention of bone loss in women within five years of menopause when there may be predominant effects of estrogen deficiency and other hormonal changes.18 A meta-analysis of randomized clinical studies in postmenopausal women of the effects of calcium on bone showed that calcium decreased bone loss by about 2% after two years or more.19 The Institute of Medicine's recommendations for daily calcium intake are shown in Table 1.20 Unless a patient has an underlying disorder of calcium homeostasis, 2,500 mg is considered the upper limit of safety. As maximum absorption of elemental calcium at once is about 500 mg, higher doses are not as effective. Daily intake needs to be divided into multiple doses throughout the day.

While dairy products contain the largest amount of endogenous calcium, many foods including juices, cereals, and cereal bars, may contain added calcium. An 8-ounce glass of calcium-supplemented orange juice or milk contains ~300 mg of elemental calcium. Calcium carbonate contains 40% of elemental calcium and is a commonly used calcium supplement (e.g. Tums™, Oscal™, Caltrate™, and generic preparations). Calcium carbonate should be taken with food because patients with achlorhydria cannot absorb this calcium salt well on an empty stomach.21 Adverse effects of calcium carbonate may include bloating and constipation. Calcium phosphate (e.g. Posture™) may be associated with less constipation and fewer gastrointestinal side effects than calcium carbonate. Calcium citrate (e.g. Citracal™), which contains 24% elemental calcium, is more bioavailable than calcium carbonate and can be taken while fasting. Data from the Women’s Health Initiative (WHI) showed that in 36,282 women who received either calcium (1000 mg total daily intake) plus vitamin D (400 IU daily) or placebo, there was a 17% increased risk of developing renal stones in those assigned calcium plus vitamin D. However, among those compliant with the calcium plus vitamin D regimen, there was a 29% decreased risk of hip fracture over seven years compared with placebo. Thus the risk of hypercalciuria and renal stones with calcium supplementation needs to be balanced with fracture reduction.22

Two online resources helpful for patients to calculate their calcium intake are http://www.nal.usda.gov/fnic/foodcomp/search and http://www.dairycouncilofca.org/Tools/CalciumQuiz. The former allows patients to look up most any food and learn the nutrient content. The latter allows patients to check off the type and quantity of calcium-containing foods they usually consume and then total daily calcium is calculated with suggestions on how to increase calcium intake to recommended levels if needed.

VITAMIN D

Mild vitamin D deficiency is common. Low dietary intake, malabsorption, inadequate sunlight exposure, use of sunblocks, and diminished ability of the skin to synthesize vitamin D can each lead to vitamin D deficiency and secondary hyperparathyroidism. Mild vitamin D insufficiency may not cause symptoms, but can contribute to low bone mass. Severe vitamin D deficiency causes osteomalacia. In addition, vitamin D deficiency has been associated with proximal muscle weakness, increased risk of falls,9, 23-25 and increased rates of various types of cancers (including colorectal, breast, prostate, ovarian, among others).26, 27

Currently, deficient levels of vitamin D are generally defined as a 25-(OH) vitamin D ≤ 20 ng/ml, relative insufficiency as 21 to 29 ng/ml, and sufficient levels of vitamin D to prevent the rise in parathyroid hormone levels as a 25-(OH) vitamin D  30 to 32 ng/ml.28 A study of women hospitalized with hip fractures showed that 57% had 25-(OH) vitamin D levels < 15 ng/ml.29 In a calcium supplementation study, intestinal calcium absorption increased by 65% in women when 25-(OH) vitamin D levels were increased from an average of 20 to an average of 35 ng/ml.30

In a longitudinal, placebo-controlled study in nursing home subjects with low vitamin D levels, calcium (1200 mg daily) plus vitamin D3 (cholecalciferol) (800 IU daily) reduced hip fractures by 43% at 18 months.31 In another study, calcium (500mg daily) and vitamin D3 (700 IU daily) decreased bone loss and led to a 50% reduction in the incidence of non-spine fractures.32 A 2005 meta-analysis combining data from 9829 patients in high-quality trials showed 700-800 IU vitamin D3 daily (with or without calcium supplementation) decreased the relative risk of hip fractures by 26% and non-vertebral fractures by 23% compared with calcium alone or placebo. Fracture reduction was not seen when only 400 IU vitamin D3 was used.33 In a 2007 meta-analysis of calcium or calcium and vitamin D in people  50 years old (patient n = 58,785, 92% women, mean age 67.8 years), there was a 12% risk reduction in all fracture types, 0.54% reduced rate of bone loss at the hip, and 1.19% reduced rate of bone loss in the spine. The fracture risk reduction was much greater in studies with better compliance, and treatment effects were significantly better when calcium doses  1200 mg were used versus <1200 mg and vitamin D doses of  800 IU were used versus < 800 IU.34 These studies support the recommendation that in addition to the calcium recommendations in the preceding section, patients should also have a minimum of 800 IU of vitamin D daily. Other data indicate that 1000 IU or more daily is required.35, 36

Multivitamins typically contain 400 IU of vitamin D3, and many calcium preparations are available with additional vitamin D. Up to 2000 IU tablets of vitamin D3 are available over the counter, and 10,000 IU tablets are available at www.vitD3.com. In the presence of vitamin D deficiency, vitamin D levels should be normalized to a 25-(OH) vitamin D  30- 32 ng/ml to prevent the compensatory rise in parathyroid hormone (PTH) level.37, 38 This may be done in a variety of ways. One approach is shown in table 2. High doses of vitamin D are needed [e.g., 50,000 IU of D2 (ergocalciferol) weekly for 8 weeks or according to the 25-hydroxyvitamin D level], but patients on a high-dose regimen should be monitored closely for the development of hypercalciuria and hypercalcemia.

There is debate about efficacy of vitamin D2 versus D3 in raising 25-(OH) vitamin D levels. Data show that supplemental vitamin D3 (50,000 IU weekly for two weeks) leads to a 1.7-fold greater rise in 25-(OH) vitamin D levels than supplementation with equal molar vitamin D2.39 However, Holick demonstrated that adults ages 18-84 years who received placebo, 1000 IU vitamin D3, 1000 IU vitamin D2, or 500 IU vitamin D2 plus 500 IU vitamin D3 daily for 11 weeks had comparable increases in 25-(OH) vitamin D, suggesting vitamin D2 is as effective as vitamin D3 in maintaining 25-(OH) D status.40 In a recent 6-week randomized study of infants and toddlers with vitamin D deficiency, 2000 IU of vitamin D2 daily, 50,000 IU of vitamin D2 weekly, and 2000 IU vitamin D3 daily were equally efficacious in raising 25-(OH) vitamin D levels.41 More research needs to be done to elucidate the clinically relevant differences in half-lives of ergocalciferol and cholecalciferol and the timing of 25-(OH) D serum measurements after vitamin supplementation.42

HORMONE REPLACEMENT THERAPY

In postmenopausal women, estrogen therapy (ET) and hormone therapy [estrogen plus progesterone (HT)] prevent bone loss and increase BMD through interaction with estrogen receptors on bone cells, activation of tissue-specific genes and proteins, and/or a reduction in cytokines that stimulate osteoclast function.45-48 Combination HT (estrogen + progesterone) was the standard of care for postmenopausal women with low bone mass largely because there was also a presumed benefit, from data generated in observational studies, that HT reduced heart disease by about 50%.49-51 HT had been shown to increase BMD, but for many years, fracture data from randomized controlled trials were limited and did not consistently show fracture reduction.52 For example, in a 2002 analysis of 22 trials of postmenopausal hormone replacement, current users of HT had a decreased relative risk of spine fractures, non-spine fractures, and wrist fractures, but either a decreased or slightly increased risk of hip fractures depending on the study examined.52

The Postmenopausal Estrogen/Progestin Interventions (PEPI) trial, a prospective, placebo-controlled clinical trial, studied the effects of four hormone replacement regimens on bone density in postmenopausal women.53 While estrogen and progestin replacement for three years led to an increase in bone density (3.5-5% at the spine and 1.7% at the hip), a subsequent four year extension in a subset of these women did not show any additional increments in bone mass.54 These data indicated that the greatest gains in bone density occur within the first three years of treatment.

While there are abundant data supporting a bone density benefit, the WHI was the first randomized clinical trial to show that HT decreased hip fractures.55 (See table 3.) The WHI was a large multicenter study of 161,809 postmenopausal women ages 50 to 79 years investigating the effects of hormones (estrogen plus progestin or estrogen alone), calcium and vitamin D, and low-fat diets on cardiovascular disease, fractures, and breast and colorectal malignancies. In July 2002 the estrogen plus progestin arm of this study was stopped after 5.2 years instead of the expected 8.5 years because health risks exceeded benefits.

*HR=Hazard Ratio Kaplan Meier methods; + Nominal Confidence intervals

In the estrogen plus progestin arm of the WHI, 16,608 postmenopausal women were randomized to conjugated estrogen (0.625 mg daily) and medroxyprogesterone (2.5 mg daily) or placebo. During the 5.2 years of treatment, among these women, there were 125 cases of hip fracture (52 in the HT group and 73 in the placebo group), 101 cases of clinical vertebral fractures (41 in the HT group and 60 in the placebo group), and 434 cases of lower arm/wrist fractures (189 in the HT group and 245 in the placebo group). For every 10,000 person-years, the HT group experienced five fewer hip fractures, six fewer clinical vertebral fractures, 18 fewer lower arm/wrist fractures, and 47 fewer total fractures.56

The WHI demonstrated that estrogen plus progestin reduces fractures; however, the harm [increases in breast cancer, coronary heart disease (CHD), pulmonary embolism (PE), and stroke] outweighed the benefits. In this cohort, there were 290 cases of invasive breast cancer (166 and 124 in the HT and placebo groups, respectively), 229 nonfatal myocardial infarctions (133 and 96 in the HT and placebo groups, respectively), 59 CHD deaths (33 and 26 in the HT and placebo groups, respectively), 212 strokes (127 and 85 in the HT and placebo groups, respectively), and 101 cases of PE (70 and 31 in the HT and placebo groups, respectively). These figures represent increases in the risk of breast cancer by 26%, CHD by 29%, stroke by 41%, and PE by 113%. According to absolute risk analyses, for 10,000 person-years of estrogen plus progestin, there were eight more cases of invasive breast cancer, seven more CHD events, eight more strokes, and eight more PEs.55 Of note, the increased risk of stroke and thromboembolic disease was evident in the first one to two years of use. The risk of breast cancer increased after four years of HT and was greater in the older groups.52 In addition, in a smaller analysis of a subset of women older than 65 years, many of whom started taking hormones later in life, there was a twofold increased risk of dementia among estrogen- and progesterone-treated than placebo-treated women.58

The estrogen only arm of the WHI was stopped after 6.8 years because the data did not show prevention of cardiovascular disease (CVD), and the risk of stroke increased. Among the women in the estrogen only arm who had a hysterectomy, about 40% also had an oophorectomy. Compared with placebo, women randomized to estrogen alone showed a 39% reduction in hip fractures and a 39% increase in the risk of stroke. There was no increased risk of breast cancer or an excess risk according to the global index, with a non-significant absolute risk of two events per 10,000 person years.57

In 2008, the WHI researchers published data from an average 2.4 years of follow-up after HT was stopped in the WHI participants. Many patterns of health risks and benefits initially associated with active HT treatment versus placebo observed during the trial were not maintained during the post-intervention phase from July 8, 2002, to March 31, 2005. In three years post-therapy, CVD risks and fracture prevention benefits became insignificant. In addition, cancer risks increased. These data support the theory that HT should be used with caution, as cancer risks may outweigh the benefits, and short-term, early post-menopausal use does not show an ongoing benefit in fracture prevention after cessation.59

The Heart and Estrogen/Progestin Replacement Study (HERS), a large randomized study of 2,763 postmenopausal women with established CHD, also showed that HT use increased the early risk of cardiac events, without long-term benefit when women were followed for up to seven years (HERS II).60, 61 HERS showed no reduction in the risk of fractures. The lack of fracture efficacy may be a result of more use of osteoporosis treatments in the placebo arm and possibly the low percentage of women with osteoporosis (15%) enrolled in the trial.

With data to support that HT increases the risk of breast cancer and CHD, many postmenopausal women stopped this therapy. For women on HRT, discontinuation of therapy is often best achieved by tapering off this treatment over several weeks to prevent or attenuate the development of menopausal symptoms. While earlier studies indicated that rapid bone loss occurred with discontinuation of HT, data have since shown that cessation of HT leads to bone loss comparable to that in women not on HT.54

Because raising serum estradiol in postmenopausal women by a small amount will increase BMD without causing endometrial hyperplasia, the safety and effectiveness of very low-dose transdermal estradiol was studied in a two-year, randomized, placebo-controlled, double-blind trial. 417 postmenopausal women, with intact uterus and BMD Z-scores of -2.0 or higher, received unopposed transdermal estradiol at 0.014 mg daily or placebo. All participants received calcium and vitamin D supplementation. Median plasma estradiol level in the transdermal estradiol group increased from 4.8 pg/mL at baseline to 8.6 pg/mL at two years and was unchanged in the placebo group (p<.001). Lumbar spine BMD increased 2.6% in the estradiol group and 0.6% in the placebo group (p<.001). Total hip BMD increased 0.4% in the estradiol group and decreased 0.8% in the placebo group (p<.001). Bone turnover markers were lower in the estradiol group than the placebo group (p<.001). Endometrial hyperplasia developed in only one of 208 woman in the estradiol group.62 Thus, these data indicate that very low doses of transdermal estrogen for two years result in small increases in bone density.

Tibolone is a synthetic hormone whose metabolites have estrogenic, progestogenic, and androgenic activities. While not FDA approved in the United States, it is used throughout the world to treat menopausal symptoms and osteoporosis. In August 2008, results were published from a randomized study of 4538 postmenopausal women (T-score  –2.5 at the hip or spine or a T-score  –2.0 and radiologic evidence of a vertebral fracture) who received tibolone (1.25 mg daily) or placebo. With a median 34 months of treatment, the tibolone group had a significantly decreased risk of vertebral fracture versus the placebo group (70 cases versus 126 cases per 1000 person-years), and a significantly decreased risk of nonvertebral fracture (122 cases versus 166 cases per 1000 person-years). The tibolone group also had a significant RRR of invasive breast cancer and colon cancer. However, the tibolone group had an increased risk of stroke (relative hazard, 2.19; p = 0.02), for which the study was stopped early. There were no significant differences in the risk of either CHD or venous thromboembolism between the two groups.63 While tibolone is an effective anti-fracture therapy, in the past it has been associated with increased breast cancer risk.64 Conflicting breast cancer data along with stroke risk may limit use for some time.

In summary, for control of moderate to severe vasomotor or other menopausal symptoms, ET or HT (in women with an intact uterus) should be used at the lowest dose for the shortest duration of time and women should be informed about the potential risks. After a careful review of risks and benefits, short-term use of HT (up to five years) at the lowest dose to control menopausal symptoms might be considered. Low doses of estrogen in HT regimens (e.g. 0.3 mg conjugated estrogen or 0.3 mg esterified estrogen) increase bone mass in placebo-controlled trials, although the long-term safety has not been established.65, 66 Transdermal estrogens prevent bone loss and are available in low doses (e.g. 0.014 mg daily). Unlike oral estrogens, in postmenopausal women transdermal estrogens do not adversely affect clotting factors. As an alternative to oral or transdermal estrogen, topical estrogen (e.g. vaginal estrogen rings, pills, or creams) can be used to treat urogenital symptoms. Before estrogen is prescribed, the benefits versus the risks of cardiovascular disease, stroke, and breast cancer should be reviewed. For other perimenopausal symptoms, treatments such as clonidine or venlafaxine (Effexor™) may help reduce hot flashes.67 Other hormonal therapy, such as tibolone may require further risk analysis before being approved in the United States.

SELECTIVE ESTROGEN RECEPTOR MODULATORS

Selective estrogen receptor modulators (SERMs) are a class of drugs that bind to estrogen receptors and can selectively function as agonists or antagonists in different tissues. Raloxifene (Evista™) is a SERM that is Food and Drug Administration (FDA) approved for the prevention and treatment of osteoporosis. The Multiple Outcomes of Raloxifene Evaluation (MORE) study was a randomized clinical trial of the effects of raloxifene versus placebo on bone density and fractures in 7,705 postmenopausal women (mean age of 67 years) with osteoporosis. Compared with placebo, raloxifene treatment for three years increased BMD of the spine by 2.6% and of the femoral neck by 2.1%. Over three years, raloxifene reduced spine fractures by 55% in women without prevalent vertebral fractures and by 30% in women with more than one prevalent vertebral fracture.68 Raloxifene therapy did not lead to a reduction in hip or wrist fractures, but there was not sufficient power to detect a 20% relative risk reduction (RRR) for these fractures. A one-year extension of the MORE study showed vertebral fracture reduction at year four similar to that at year three.69 Data from the MORE study indicate that women with osteoporosis treated with raloxifene had a 76% lower risk of breast cancer than the placebo group.70 In addition, raloxifene decreased LDL-cholesterol by 12%. Recent data, moreover, indicate that raloxifene did not produce an early increase in the risk of CVD and lowered cardiac events by 40% in women with high CVD risk.71

In the Continuing Outcomes Relevant to Evista (CORE) trial, 2725 women from MORE who had received raloxifene 60 or 120 mg daily now received raloxifene 60 mg daily, and 1286 women who had received placebo continued on placebo. After seven years of treatment, compared with the MORE baseline, raloxifene significantly increased lumbar spine BMD (4.3% from baseline, 2.2% from placebo) and femoral neck BMD (1.9% from baseline, 3.0% from placebo). However, the risk of at least one new nonvertebral fracture over the eight years from the beginning of MORE till the end of CORE was similar in the placebo (22.9%) and raloxifene (22.8%) treated groups.72 Thus, raloxifene still has not been shown to decrease the risk of nonvertebral fracture.

The side effects of raloxifene include an increase in deep venous thrombosis similar to use of estrogen, along with a small increase in hot flashes and leg cramps. Raloxifene was approved by the FDA in 2007 for reduction in the risk of invasive breast cancer in post-menopausal women with osteoporosis and postmenopausal women at high risk for invasive breast cancer.

Tamoxifen, a SERM used for the prevention and treatment of estrogen receptor-positive breast cancer, has estrogen-like effects in bone. However, it also stimulates the endometrium and can result in uterine hyperplasia or malignancy.73 Bazedoxifene, lasofoxifene, and arzoxifene are third-generation SERMs, none of which appear to cause endometrial hyperplasia. In fact, arzoxifene has been used in the treatment of endometrial cancer.74, 75 In a study of 7492 postmenopausal women with osteoporosis, women who received bazedoxifene 20 mg daily or 40 mg daily had a significantly lower incidence of new vertebral fracture (2.3% and 2.5% respectively) than those who received placebo (4.1%).76 Bazedoxifene, lasofoxifene and arzoxifene are still in clinical trials and analysis.

CALCITONIN

Calcitonin is a 32-amino acid peptide produced by the parafollicular cells of the thyroid that inhibits bone resorption through direct effects on the osteoclasts. Calcitonin is a highly conserved protein, with human and salmon calcitonin differing by only one amino acid.

Because of its antiresorptive properties, salmon calcitonin has been used in the treatment of metabolic bone disease for over 30 years.77 In osteoporotic subjects, several studies have revealed that parenteral calcitonin (100 IU daily or every other day) prevented bone loss or produced a small increase in the bone density in the forearm, vertebrae, femoral diathesis, or total body. Injectable salmon calcitonin was approved by the FDA in 1984 for the treatment of osteoporosis, although current use is limited because of the availability of calcitonin nasal spray and other more effective medications for the treatment of osteoporosis. An oral formulation of calcitonin is in development.77

Calcitonin nasal spray (Miacalcin™ and Fortical™ 200 IU daily) is a form of calcitonin78 approved by the FDA for the treatment of osteoporosis in women more than five years past menopause. The prospective, five year, placebo-controlled, Prevent Recurrence of Osteoporotic Fractures (PROOF) study examined the effects of Miacalcin™ (100, 200, or 400 IU daily) with calcium (1,000 mg daily) plus vitamin D (400 IU daily) on bone density and fractures. The PROOF study included 1,255 postmenopausal osteoporotic women (average age 69 years) with low bone density (lumbar spine T-score  -2.0).79 Calcitonin nasal spray (200 IU daily) reduced the risk of new vertebral fractures by 33% compared with placebo and by 36% in women with one to five prevalent vertebral fractures. However, there was no effect of calcitonin nasal spray on hip or other non-spine fractures.

Although calcitonin nasal spray decreased spine fractures, there were minimal or no changes in BMD and small decrements in markers of bone turnover. Calcitonin may have alternative effects on microarchitecture and/or bone strength, as was seen when 91 postmenopausal women with osteoporosis received calcitonin nasal spray 200 IU daily or placebo and whose bone microarchitecture was evaluated primarily by high-resolution MRI in the distal radius in a study of the qualitative evaluation of salmon calcitonin therapy (QUEST study). Calcitonin was correlated with preserved bone microarchitecture with significant differences demonstrated for trabecular bone volume, trabecular number, and trabecular spacing at the distal radius versus placebo. These positive effects on trabecular bone microarchitecture were independent of changes in BMD.80

In summary, calcitonin has modest vertebral anti-fracture efficacy through five years, which seems to occur through reduction in bone resorption and preservation of bone microarchitecture rather than primarily by increasing BMD. Anti-fracture efficacy for non-vertebral fractures has not been demonstrated for calcitonin.77 Side effects of calcitonin are minimal and include flushing and pain at the injection site (with injections) and rhinorrhea (with calcitonin nasal spray).

BISPHOSPHONATES

Bisphosphonates are analogs of pyrophosphate, which was used since the 19th century to prevent build-up of calcium crystals in industrial tanks. Pyrophosphate was shown to inhibit both formation and dissolution of calcium crystals in vitro and, when administered subcutaneously, to prevent ectopic calcification in vivo. Oral pyrophosphate cannot be used alone as a regulator of bone turnover because the phosphorous-oxygen-phosphorous structure is rapidly degraded in vivo. Bisphosphonates are resistant to hydrolysis because the phosphorous-carbon-phosphorous structure inhibits bone resorption. They also have a strong affinity for calcium crystals and bind avidly to the surface of bone. Bisphosphonates interrupt osteoclast activity directly through several mechanisms including inhibiting acid production, lysosomal enzymes, and the mevalonate pathway81-83 and indirectly by acting through osteoblasts and macrophages. They also inhibit osteoclast recruitment84 and induce osteoclast apoptosis.85 Thus, through various mechanisms, bisphosphonates reduce the depth of resorption pits (thereby producing positive bone balance at individual bone remodeling units) and decrease the formation of new bone remodeling units.86

Pharmacodynamics

Oral bisphosphonates are poorly absorbed. Less than 3% is absorbed in the fasting state, and absorption is significantly reduced if these drugs are taken with food or beverages other than water. The skeleton rapidly takes up approximately half of the absorbed bisphosphonate, and the remainder is excreted unchanged by the kidney within hours. The drug remains at the bone surface for several weeks before becoming embedded in bone, where it is biologically inert. The embedded drug then remains in bone for many years and is slowly released.

Potency and side effects of the bisphosphonates vary according to the side chains (see Table 4).87, 88

Effective Therapy for Osteoporosis

Alendronate (Fosamax™), risedronate (Actonel™), ibandronate (Boniva™), and zoledronic acid (Reclast™) are all FDA approved for osteoporosis prevention and/or treatment. Their indication and prescribing information are shown in Table 6.

Alendronate

Several longitudinal studies have shown that oral alendronate increases BMD and decreases the risk of osteoporotic fractures. In a meta-analysis of randomized controlled trials published between 1966 and 2007, the efficacy of alendronate in the primary and secondary prevention of osteoporotic fractures in postmenopausal women was evaluated.89 Eleven studies were selected, including three primary prevention studies90-92 and eight secondary prevention studies involving women with low BMD on DXA and/or high prevalence of vertebral fracture.93-100 A total of 12,068 women received at least one year of oral alendronate (6543 women) or placebo (5525 women). Three trials, including the largest secondary prevention trial, Fracture Intervention Trial (FIT), used an initial daily dose of 5 mg and then switched to 10 mg for the remaining study duration. Other studies used 5 mg, 10 mg, or 20 mg of alendronate daily. The length of follow-up ranged from one to four years, and the mean ages were 53 to 78 years. With alendronate 10 mg daily, for secondary prevention there was a significant 45% RRR in vertebral fractures, 23% RRR in non-vertebral fractures, and 53% RRR in hip fractures. For primary prevention, the RRR was only significant for vertebral fractures (45%). No statistically significant differences in adverse events were found in any included study.89

The prevalence of osteoporosis is lower in men than in women. It is estimated that one out of two women and one out of four men over age 50 will develop an osteoporotic fracture.101 Few longitudinal studies have evaluated the efficacy of treatment interventions on bone in osteoporotic men. Orwoll et al. enrolled 241 men with a femoral neck T score of  -2 with a lumbar spine T score  -1 or a history of osteoporotic fracture and a femoral neck T score  -1. Compared with placebo, alendronate significantly increased BMD at each site and decreased markers of bone turnover over two years. From baseline, alendronate increased BMD by 3.1% in the total hip and by 7.1% in the lumbar spine and decreased urinary N-telopeptides by 59% and bone-specific alkaline phosphatase by 38%. The incidence of vertebral fractures was 7.1% in the placebo group versus 0.8% in the alendronate group. There was no significant difference in the incidence of non-vertebral fractures. However, the study did not have statistical power to look at this endpoint.102 Similar results were seen in a smaller study of hypogonadism-induced osteoporosis, indicating no difference in the skeletal response to alendronate in the presence of hypogonadism. In this study of 22 osteoporotic men (mean age 50.2 years) receiving standard testosterone replacement therapy, one half received alendronate 10 mg daily while the other half received placebo. At 12 months, the group receiving alendronate and testosterone had BMD increases at the femoral neck (1.9%) and lumbar spine (8.4%), while the group receiving placebo and testosterone had BMD decreases at the femoral neck (1.4%) and much less of an increase in lumbar spine (3.3%). Both between group differences were significant.103

Alendronate is also effective in the treatment of glucocorticoid-induced osteoporosis. Alendronate increased BMD104, 105 and decreased the incidence of radiographic vertebral fractures at two years (6.8% vs. 0.7%) in glucocorticoid-treated men and women.105

Data show that weekly alendronate (70 mg) is effective and well tolerated, and this dosage has become the standard of care for use of this oral bisphosphonate. Alendronate is suitable for weekly dosing because of its long skeletal retention. In a one year study of 1,258 postmenopausal women with osteoporosis, there were no differences between alendronate 10 mg daily and 70 mg weekly on BMD or markers of bone turnover.106

Long-term treatment with alendronate has beneficial effects on BMD. Bone et al. showed that spine BMD continued to rise in small increments during 10 years of treatment. Femoral neck and trochanter BMD increased during the first three years and then remained stable.107, 108 In an extension of FIT, the FIT Long-term Extension (FLEX) trial, 1099 women who had received alendronate (5 mg daily for two years and 10 mg daily thereafter) were again randomized to receive either 5 or 10 mg alendronate daily or placebo for five more years. With a pooled analysis of the alendronate doses, after five years, the alendronate-treated subjects had significantly better BMD changes at the total hip, femoral neck, lumbar spine, total body, and forearm. These changes included less loss of BMD at the total hip (placebo 3.38% decrease, pooled alendronate 1.02% decrease) and more gain in BMD at the lumbar spine (placebo 1.52% increase, pooled alendronate 5.26% increase). Subjects on placebo had increases in bone turnover markers compared with alendronate users. Alendronate users had lower risk of clinically recognized vertebral fractures, but the cumulative risk of nonvertebral fractures was not significantly different between the alendronate-treated women and those who received placebo. The authors concluded that for many women the discontinuation of alendronate for up to five years did not appear to significantly increase fracture risk, but women at high risk of vertebral fractures may benefit from continued alendronate use.109

This trial has met with some controversy, however. Patients with severe osteoporosis were excluded from enrollment, while those with osteopenia were included. There was an uncontrolled phase between FIT completion and FLEX enrollment. There was also a high dropout rate, limiting statistical power.110 Finally, the alendronate dose of 5 mg is less than the typical regimen used currently. Given the limitations of these analyses, risk of fracture versus benefit of continuing treatment should be individualized.

Risedronate

Risedronate increases BMD and decreases fracture risk among postmenopausal women with osteoporosis. Harris et al. reported data on 2,458 postmenopausal women with established osteoporosis (subjects had either two or more vertebral fractures or one vertebral fracture and lumbar spine T score of -2 or less) and who were randomized to risedronate (5 mg daily) or placebo. Over three years, risedronate increased lumbar spine BMD by 5.4% and femoral neck BMD by 1.6%. Over six months, risedronate decreased the deoxypyridinoline-creatinine ratio by 38% and bone-specific alkaline phosphatase by 35%. Risedronate decreased the risk of new vertebral fractures by 41% and decreased the risk of non-vertebral fractures by 39% at three years.111 Reginster et al. provided further evidence that risedronate protects against fracture among postmenopausal women with established osteoporosis (subjects had two or more vertebral fractures at baseline). Within six months risedronate increased spine and hip BMD and during the first year decreased the risk of new vertebral fracture by 61%. This study demonstrated radiologically that risedronate may have an important early impact on fractures within the first year of therapy. Over three years, the risk of vertebral and non-vertebral fractures was reduced by 49% and 33%, respectively.112

In a large study of 9,331 postmenopausal women, McClung et al. demonstrated that risedronate reduces the risk of hip fractures among postmenopausal women with osteoporosis but not among elderly women selected on the basis of clinical risk factors for hip fracture. Among 5,445 postmenopausal women with osteoporosis, risedronate decreased the risk of hip fractures by 40% over three years. However, among the 3,886 elderly women with clinical risk factors (i.e., difficulty standing from a seated position, poor gait, recent fall-related injury, poor hand-eye coordination, recent smoking, maternal history of hip fracture), risedronate had no effect on the incidence of hip fractures.113 These results highlight the importance of a BMD measurement and/or the presence of prevalent fractures instead of only clinical risk factors in identifying patients likely to benefit from bisphosphonate therapy. (Please see FRAX™ CALCULATOR below).

Risedronate therapy also reduces fracture risk in men. In an open-label, randomized, controlled trial, 316 men with osteoporosis received either oral risedronate 5 mg daily or alfacalcidol 1 mcg daily. At 12 months, lumbar spine BMD increased by 4.7% in the risedronate group versus 1.0% in the alfaclcidol group. Significant increases in BMD at the total hip and femoral neck were also seen in the risedronate group. The risedronate users also had a 60% lower risk of vertebral fracture than controls.114

Risedronate is effective in the prevention and treatment of glucocorticoid-induced osteoporosis in men and women. Risedronate (5 mg daily) prevented glucocorticoid-induced bone loss115 and reduced the risk of radiographic vertebral fractures by 70% after one year of treatment.116

Weekly risedronate (35 mg) is effective and well tolerated.117-119 Brown et al. randomized 1,468 women to daily or weekly risedronate. The increase in lumbar spine BMD at one year was similar between groups. Weekly risedronate was well tolerated, and the occurrence of adverse events was similar in daily and weekly treatment groups.117 Long-term treatment with risedronate has beneficial effects on bone. Mellstrom et al. showed that lumbar spine BMD continued to increase during seven years of treatment and fracture data demonstrate no loss of anti-fracture efficacy during the six to seven-year time period.120

Recently, monthly dosing of risedronate became available (75 mg two consecutive days a month or now 150 mg once a month). Both monthly dosing regimens were shown to be non-inferior in efficacy and safety to the 5 mg daily regimen at one year.121, 122 Thus, monthly risedronate provides a new alternative regimen for the prevention and treatment of osteoporosis.

Alendronate versus Risedronate

In an extension of the double-blind Fosamax Actonel Comparison Trial international study (FACTS-International), postmemopausal women with osteoporosis were randomly assigned to alendronate 70 mg per week or risedronate 35 mg per week for 24 months. BMD increased significantly from baseline in both groups, but with significantly larger increases with alendronate at each site. At the lumbar spine alendronate increased BMD 6.0% versus risedronate’s 4.2%, at the total hip alendronate increased BMD 3.7% versus risedronate’s 2.4%, at the greater trochanter alendronate increased BMD 5.2% versus risedronate’s 3.7%, and at the femoral neck alendronate increased BMD 3.2% versus risedronate’s 2.3%. Significant decreases from baseline were seen for all four bone turnover markers in both groups, but with significantly larger decreases with alendronate. There were no significant differences in UGI symptoms or other adverse events. No fracture results were compared, however, because of limited power.123 These results support previous findings.124-126

The RisedronatE and ALendronate (REAL) cohort study was a retrospective analysis of patients receiving weekly alendronate or risedronate for one year according to healthcare utilization records in the United States. The studied outcome was new nonvertebral fractures as a group (hip, wrist, humerus, clavicle, pelvis, leg) and hip fractures during the 12 months of therapy. In the study population of 33,830, there were 507 nonvertebral fractures and 109 hip fractures. The incidence of nonvertebral fractures in the risedronate patients (2.0%) was 18% lower than in the alendronate patients (2.3%). The incidence of hip fractures in the risedronate patients (0.4%) was 43% lower (95% CI 13% – 63%) than in the alendronate patients (0.6%).127

There are still no head-to-head prospective randomized control trials for the comparison of bisphosphonates’ effects on fracture prevention. Thus, studies like REAL, with their possible flaws in baseline characteristics and fracture and compliance reporting from medical records, along with meta-analyses, with variations in study populations and study designs, are still our only tools. To date, from the studies available, alendronate appears to increase BMD more than risedronate, and risedronate appears to have an advantage in early fracture prevention at one year. Both alendronate and risedronate are effective treatment options for osteoporosis. It is well-established that there is poor compliance with oral bisphosphonates, with up to 50% of patients discontinuing treatment.128-130 Therefore, less frequent dosing regimens including monthly risedronate or a yearly IV bisphosphonate (see zoledronic acid below) may prove more effective for some individuals.

Zoledronic Acid

Zoledronic acid, an intravenous bisphosphonate, has been FDA approved for years for the treatment of hypercalcemia of malignancy, multiple myeloma, and bone metastases from solid tumors. In August 2007, zoledronic acid (Reclast®) became the second intravenous bisphosphonate after ibandronate (Boniva®) to be FDA approved for treatment of postmenopausal osteoporosis. It is considerably more potent than other available bisphosphonates. Thus small doses and longer dosing intervals may be used.131 Reid et al. showed that zoledronic acid increases BMD and decreases markers of bone turnover in postmenopausal women. A single infusion of 4 mg of zoledronic acid increased BMD at the lumbar spine by about 4.5% and suppressed markers of bone turnover (serum C-telopeptide and the ratio of urinary N-telopeptide to creatinine) by approximately 50% to 65% at 12 months.132 The observed effects on bone were similar to those achieved with daily oral bisphosphonates.

In the Health Outcomes and Reduced Incidence with Zoledronic Acid Once Yearly (HORIZON) study, a double-blind, placebo-controlled trial of 7765 post-menopausal women with osteoporosis were randomly assigned to receive a single 15-minute infusion of 5 mg of zoledronic acid or placebo at baseline, at 12 months, and at 24 months. The patients were followed over 36 months. In addition to positive BMD and bone turnover marker findings, treatment with zoledronic acid was associated with 70% RRR in morphogenic vertebral fractures and 41% RRR in hip fractures compared with placebo. Nonvertebral fractures, clinical fractures, and clinical vertebral fractures were reduced by 25%, 33%, and 77%, respectively. While adverse events, including change in renal function, were similar in both study groups, serious atrial fibrillation (AF) occurred slightly more frequently in the zoledronic acid group.133 Further analysis of the trial data and possible risk factors for AF are presented below under Atrial Fibrillation.134

In a HORIZON sub-study of 2,127 patients with a recent hip fracture (the Recurrent Fracture Trial), zoledronic acid was associated with a significant reduction in new clinical fractures and the risk of death. There was a 35% reduction in any new clinical fracture and a 28% reduction in deaths from any cause in the zoledronic acid-treated group compared with the placebo-treated group.135 Most subjects in the multinational study received 50,000 to 125,000 IU vitamin D at least two weeks prior to the zoledronic acid infusion. Once yearly infusion of zoledronic acid after vitamin D repletion, therefore, produces a large decrease in spine and hip fractures, and in some populations may have a survival benefit.

Other Bisphosphonates

Ibandronate (oral and IV) is FDA-approved for the prevention and treatment of postmenopausal osteoporosis. It has also been shown to increase bone density and decrease vertebral fractures with both an oral daily regimen (2.5 mg daily) and an intermittent regimen (20 mg every other day for 12 doses every three months, 150 mg monthly).136-138 However, to date, it has not been shown to decrease hip fractures. Thus, because there are other bisphosphonates with more efficacy data available for reduction of hip and non-spine fractures, ibandronate should currently not be used for first-line treatment.

Pamidronate does not have FDA approval for use in osteoporosis; however, it is occasionally used “off-label” for patients with a variety of conditions, including esophageal abnormalities (i.e., stricture or achalasia) and organ transplants. Usually 30 to 60 mg is infused over two to four hours every three months. Pamidronate has been shown to increase BMD, but no fracture data are available.139-143

Adverse Effects

GI Effects

In general, the bisphosphonates are safe medications. Oral bisphosphonates are associated with some GI symptoms, and rare cases of severe esophagitis have been reported with alendronate and oral pamidronate. However, Lanza et al. carried out a placebo-controlled endoscopic study in 277 subjects and found that the incidence of upper GI symptoms and endoscopic lesions was similar in the placebo and weekly alendronate groups.144 While in controlled trials the incidence of GI adverse effects did not differ in alendronate versus placebo groups, in clinical practice some patients discontinue bisphosphonates because of adverse GI experiences. Weekly alendronate appears to be better tolerated than daily alendronate,144 and weekly risedronate is well tolerated among patients who discontinued alendronate because of adverse upper GI experiences.119

In a two-week head-to-head endoscopic trial, risedronate was associated with fewer gastric ulcers than was alendronate. No significant between-group difference was noted in esophageal or duodenal endoscopic scores. However, the study was not powered to detect a difference in these endpoints.145 In a 12-month head-to-head comparison trial between alendronate and risedronate there was no difference in the incidence of upper GI side effects between the alendronate and risedronate groups.125 In a pooled analysis of nine studies involving 10,068 men and women who received placebo (n = 5048) or 5 mg of risedronate sodium (n = 5020) for up to three years of daily treatment, risedronate was not associated with an increased frequency of adverse GI tract effects, even among patients at high risk for such events.146

Because of the risk of esophagitis, alendronate is contraindicated for patients with esophageal abnormalities such as stricture or achalasia, and both alendronate and risedronate are contraindicated for patients who are unable to stand or sit upright for at least 30 minutes after drug administration because of increased risk of adverse esophageal effects. .

Osteonecrosis of the Jaws

Bisphosphonate-associated osteonecrosis of the jaws (ONJ) has drawn attention in recent years. ONJ may remain asymptomatic for weeks or months, but then is noticed when there is exposed bone in the oral cavity, pain, swelling, loose teeth, and/or drainage. It is more common after dental procedures such as tooth extraction. It has been hypothesized that ONJ is the result of bone remodeling suppression combined with additional factors such as dental intervention or infection.147 In 2005 the FDA requested that all oral and IV bisphosphonates include a class “precaution” labeling for ONJ. There have been no cases reported in randomized, placebo-controlled trials of alendronate, risedronate, or ibandronate. However, in a 2006 Medline review, 368 published cases were found, 94% of which involved patients receiving intravenous bisphosphonates, 85% of which involved patients with multiple myeloma or metastatic cancer. Only 4% of patients had osteoporosis and data suggests a time- and dose-dependent effect. 60% of reported cases of ONJ occurred after dentoalveolar surgery for infections (tooth extractions), and the remaining 40% were likely related to infection, denture trauma, or other oral trauma.148 Based on both published and unpublished data, the risk of ONJ associated with oral bisphosphonate treatment for osteoporosis seems low, estimated between one in 10,000 and less than one in 100,000 patient-treatment years.149 Some experts suggest stopping bisphosphonates during a time before and after-invasive dental procedures. There is no evidence that this improves outcomes, but it may be a cautious strategy along with maintaining routine dental hygiene.

Atrial Fibrillation

In the HORIZON trial, serious AF was seen more frequently in patients who received IV zoledronic acid (50 subjects, 1.5%) than in those who received placebo (20 subjects 0.5%).133 As mentioned above, significant risk factors were active tacharrythmia, congestive heart failure, previous bisphosphonate use, and advanced age.134 In a review of the results from FIT, there were more serious AF cases in the alendronate group (47 subjects, 1.5%) than in the placebo group (31 subjects, 1.0%), but the difference was not statistically significant.150 These findings raised concern, as this was a new issue with bisphosphonate use. In a case-control study published in 2008, researchers identified 719 women with confirmed AF between October 1, 2001, and December 31, 2004, and 966 control women without AF, matched for age, presence or absence of treated hypertension, and calendar year. More AF subjects than controls had ever used alendronate (n = 47, 6.5% versus n = 40, 4.1%).151 A review of data from multiple trials did not find an association between risedronate use and AF.152 It is unclear how bisphosphonates may increase the risk of AF. Hypotheses include the release of inflammatory cytokines when IV bisphosphonates are administered, calcium shifts that can occur with IV and potent oral bisphosphonates, and relative binding affinity of the various bisphosphonates to bone. Both cytokines and calcium shifts may increase the risk of AF, and thus far, strongly binding bisphosphonates (zoledronic acid and alendronate) have shown more of an AF association than a weakly binding bisphosphonate (risedronate). Further investigations will need to better examine the possible bisphosphonate/AF relationship.

Recently the FDA released a review of spontaneous post-marketing reports of AF associated with oral and IV bisphosphonates and did not identify a risk of AF.153, 154 The FDA continues to monitor such reports. Given the efficacy of bisphosphonates for the prevention and treatment of osteoporosis, and the minimal risk of AF, currently it seems that the benefits of bisphosphonate treatment in patients with osteoporosis outweigh the risk of AF.

Other Precautions

Acute-phase reactions (e.g. fever, malaise, myalgia) may occur and are seen in up to 10% to 20% of subjects following intravenous administration of pamidronate or zoledronic acid.132, 133 Patients often are premedicated with acetaminophen, and symptoms are usually mild and transient. Hypocalcemia may occur, but this is usually mild and asymptomatic. To avert marked hypocalcemia it is important to ensure that the patient is vitamin D sufficient, which according to the authors’ practices, can best be achieved by checking a 25-hydroxy vitamin D level prior to each infusion. In addition, calcium and creatinine levels should be tested before treatment as well.

Bisphosphonates are excreted by the kidneys and because of lack of experience should not be used for patients with severe renal insufficiency (creatinine clearance < 35ml/min). Studies in cancer patients, in whom cumulative doses are several-fold higher than in osteoporosis patients, show that age, concomitant non-steroidal anti-inflammatory drug use, prior pamidronate use, history of hypercalcemia, renal disease, hypertension, and smoking are risk factors for renal failure.155, 156 Further studies using lower doses of bisphosphonates for patients with renal insufficiency are warranted.

Drug Administration

Oral bisphosphonates should be taken in the morning with water on an empty stomach. Because oral bisphosphonates are poorly absorbed, patients should wait at least 30 minutes before ingesting other beverages, food, or medications. To help patients avoid esophageal irritation, they are instructed to swallow oral bisphosphonates with six to eight ounces of water and to remain upright for at least 30 minutes and until they have had their first meal of the day.157

Intravenous preparations must be infused slowly to avoid renal toxicity.

When choosing a bisphosphonate, the best available efficacy data support that alendronate and risedronate should be first-line. Both drugs have been on the market for more than 10 years and have favorable safety profiles when used in the indicated populations. Alendronate currently has the lowest cost, as it is available in generic form. Generic risedronate has been FDA approved, but will not be available until 2011. While oral ibandronate is popular for its monthly dosing schedule, it has only been shown to reduce the incidence of vertebral fractures, and now there is a monthly alternative with risedronate. In addition, ibandronate’s IV dosing is more expensive and more frequent than zoledronic acid. Thus, it has a limited role in osteoporosis treatment. In patients who are unable to comply with the administration requirements of the oral agents, and in those who experience intolerable GI effects, intravenous zoledronic acid is an appropriate therapy. Like alendronate and risedronate, it reduces the incidence of vertebral and nonvertebral fractures. Zoledronic acid (5 mg infusion once a year) should also be considered in patients with a recent hip fracture (after two weeks to 90 days, and optimally after vitamin D repletion) because of its demonstrated secondary prevention of other fractures and evidence to support a survival benefit.

PARATHYROID HORMONE

Anabolic Action on Bone

In 2002, the FDA approved teriparatide (Forteo™), injectable recombinant human PTH (1-34), for the treatment of men and postmenopausal women with osteoporosis who are at high risk for fracture. The biologically active fragment PTH (1-34) has properties similar to the full-length molecule PTH (1-84), which is being actively investigated and is approved for use in Europe.158 Antiresorptive agents, such as estrogen, raloxifene, and the bisphosphonates, increase BMD by up to 8%. However, many patients with osteoporosis have lost as much as 30% of their peak bone mass. Thus, agents that trigger more dramatic recovery of BMD are desirable.159 PTH directly stimulates bone formation and can have robust effects on BMD.

It may seem paradoxical that PTH increases BMD, given that primary hyperparathyroidism is associated with low bone mass. Severe longstanding primary hyperparathyroidism can cause high bone turnover and resorption, fractures, and bone cysts. However, mild primary hyperparathyroidism is often asymptomatic and has variable effects on bone. While cortical BMD (e.g., in the distal third of the radius) tends to be decreased in mild primary hyperparathyroidism, trabecular BMD (e.g. in the vertebral bodies) is often well preserved even in postmenopausal women.160, 161 Thus, the effects of PTH on bone metabolism are complex.

Animal studies show that PTH is capable of both anabolic and catabolic actions on bone. PTH stimulates both bone formation and bone resorption; the net effect on BMD depends on the balance between these two processes.162 A continuous infusion of PTH increases both formation and resorption and leads to bone breakdown.162, 163 However, intermittent exposure preferentially increases formation, thereby producing an anabolic effect on bone.162, 164, 165 Therefore, PTH can increase or decrease BMD depending on the pattern of exposure. Dosing PTH in a manner leading to stimulation of bone formation before causing bone resorption has become known as maximizing the “anabolic window” of PTH.166

Cellular Mechanisms

PTH acts directly on osteoblasts and cells of the osteoblast lineage. PTH promotes differentiation of pre-osteoblasts to osteoblasts163 and inhibits osteoblast apoptosis, thereby increasing the number of active osteoblasts167. Furthermore, PTH triggers the production of several growth factors in bone cells, including insulin-like growth factor I (IGF-I).163, 168

Clinical Studies

In 1976, Reeve et al. published the first clinical trial of PTH for osteoporosis in humans. The investigators treated four postmenopausal osteoporotic women with 100 mg PTH (1-34) daily for six months and noted marked increases in bone turnover (as measured by isotopic tracer and histological methods) with greater increases in formation than resorption.169 In a subsequent study of osteoporotic men and women, PTH (1-34) increased iliac trabecular bone mass by approximately 70%.170 Slovik et al. treated eight osteoporotic men with PTH (1-34) plus 1,25-(OH) vitamin D. After one year of treatment, BMD of the lumbar spine (measured by quantitative computed tomography) increased almost twofold and BMD of the radius (measured by single photon absorptiometry) remained stable.171

Subsequent trials using DXA as a measure of BMD have confirmed these earlier studies. A beneficial effect of PTH (1-34) on BMD has been documented in a variety of patient populations including men with idiopathic osteoporosis,172, 173 young women with estrogen deficiency caused by gonadotropin-releasing hormone (GnRH) agonists,174, 175 patients with glucocorticoid-induced osteoporosis,176, 177 and postmenopausal women with osteoporosis.178-180 In men, daily doses of 400 IU PTH (1-34) increased lumbar spine BMD by 13.5% and femoral neck BMD by 2.9% over 18 months. BMD in the distal third of the radius did not change.172 Finkelstein et al. studied the effects of PTH in young women with acute estrogen deficiency caused by a GnRH agonist (nafarelin acetate). At 12 months, PTH (1-34) (40 µg subcutaneously daily) plus nafarelin acetate (200 µg intranasally twice daily) maintained BMD in the hip and total body and increased BMD in the lateral lumbar spine by 7.5% [the control group of nafarelin acetate (200 µg intranasally twice daily) without PTH significantly lost BMD at most measured sites including 4.9% from the lateral lumbar spine].175 PTH has been shown to increase markers of bone turnover by as much as 200%,172, 176 with earlier effects on markers of bone formation than on markers of bone resorption.176 Paired iliac crest biopsies reveal that, in addition to increasing bone mass, PTH also improves trabecular bone microarchitecture.181

Fracture data

In addition to increasing BMD, PTH has been shown to decrease fracture risk. In a three-year trial, investigators randomized 52 postmenopausal women on HT (T score -2.5 or less or baseline vertebral fracture) to PTH (1-34) (25 µg daily) or placebo. The PTH + HT group (average age 57.7 years) showed progressive increases in BMD over three years, whereas the HT alone group (average age 62.9 years) had no significant change in BMD. The PTH + HT group showed an increase of 13.4% in the spine, 4.4% in the total hip, and 3.7% in the total body at three years. PTH + HT therapy was associated with a 75-100% decrease in the rate of radiographic vertebral fractures compared with HT alone. (This small study did not have sufficient power to assess the effect of PTH on non-vertebral fractures.) Of note, subjects were followed for one year after discontinuing PTH (HT was continued in all subjects), and their BMD remained stable,178, 179suggesting that PTH may have sustained benefit on bone with sustained HT.

Combination Therapy

The effects of concurrent or sequential therapy with PTH and antiresorptive agents have been studied. Black et al. compared the effects of PTH (1-84), alendronate, or both in combination in postmenopausal women.182 At one year, spine DXA had increased in all three groups. There was no difference in spine DXA between the PTH group and the combination group. However, the PTH group had a significantly greater increase in volumetric BMD of the spine on quantitative CT than the alendronate and combination groups. Finkelstein et al. also carried out a study in men.183 PTH (1-34) was started at month six, and all three groups were followed for 30 months. Spine BMD as measured by both DXA and quantitative CT increased to a greater degree in the PTH group than in the alendronate and combination groups. Thus, these studies show no evidence of synergy between PTH and alendronate. Furthermore, alendronate may impair the anabolic activity of PTH. It is hypothesized that PTH is less effective when bone turnover is suppressed. In a non-randomized analysis, Ettinger et al. found that prior treatment with alendronate prevented PTH-induced increases in BMD, while the group previously treated with raloxifene experienced the expected increase in BMD with PTH treatment. Baseline markers of bone turnover were lower in the alendronate group.184 This is an important area for further research, as many patients being considered for PTH therapy have been previously treated with antiresorptive agents.

While concurrent treatment with PTH and alendronate does not appear to be additive, bisphosphonate therapy initiated immediately upon completion of PTH course is beneficial. Rittmaster et al. demonstrated that PTH followed by alendronate produces progressive increases in BMD. In this study, 66 postmenopausal women were randomized to either 50 µg of recombinant human PTH (1-84) daily or placebo for the first year, and then all subjects were treated with alendronate on an open label extension for the second year. During the first year, the PTH group gained 4.3% BMD at the lumbar spine while the placebo group gained 1.3%. During the second year, the PTH group gained 6.3% BMD at the lumbar spine while the placebo group gained 5.7%. Thus, subjects previously treated with PTH continued to gain BMD with subsequent alendronate therapy.159 Black et al. extended their trial mentioned above.182 Post-menopausal women who had received PTH (1-84) in year one were randomly assigned to an additional year of placebo (n = 60) or alendronate (n = 59). Over two years, alendronate after PTH (1-84) led to significant increases in BMD compared to placebo after PTH (1-84), most notable at trabecular bone areas of the spine as assessed by quantitative CT [31% increase in alendronate after PTH (1-84) group versus14% increase in placebo after alendronate group]. Significant BMD loss was seen in year two in the placebo after PTH (1-84) group185. Kurland et al. reported similar findings in men.186 Twenty-one men were followed for up to two years after discontinuing PTH (1-34). Those who decided to go on bisphosphonate therapy immediately upon completion of the PTH course gained an additional 8.9% BMD at the lumbar spine at two years, while the men who did not go on bisphosphonate therapy lost 3.7% BMD at the lumbar spine at one year. These studies support the immediate use of bisphosphonates upon completion of the recommended 18- to 24-month course of PTH therapy to consolidate the increases in bone density.

Adverse Effects

In general, teriparatide, recombinant human PTH (1-34), injections are well tolerated. PTH is cleared from the circulation within four hours of subcutaneous administration.119 A daily injection is necessary and transient redness at the injection site has been noted. Headache and nausea occur in less than 10% of subjects receiving a daily dose of 20 µg. Mild, early, transient hypercalcemia can occur, but severe hypercalcemia is rare. Increases in urinary calcium (by 30 µg per day) and serum uric acid concentrations (by 13%) are seen but do not appear to have clinical consequences.

Fisher 344 rats treated with nearly life-long daily teriparatide have an increased risk of osteosarcoma. To date there is no reported increase in prevalence of osteosarcoma in humans treated with teriparatide, and no association has been found between primary hyperparathyroidism and osteosarcoma. On the basis of the above data, teriparatide should not be used for patients at increased risk for bone tumors. The FDA has a black box warning about osteosarcoma in rodents treated with teriparatide and the manufacturer warns against using teriparatide in the following settings: Paget's disease or unexplained elevations of alkaline phosphatase, open epiphyses in children or young adults, bone metastases, prior radiation therapy involving the skeleton, metabolic bone disease other than osteoporosis, and hypercalcemia.

Drug Administration

Teriparatide is supplied in a disposable pen device for subcutaneous injection into the thigh or abdomen. The recommended dosage is 20 µg once a day for no more than two years. The safety and efficacy of the teriparatide preparation have not been evaluated beyond two years of treatment.

As mentioned above, PTH (1-84) is available as osteoporosis therapy in Europe, but not yet approved in the US. Daily nasal spray of PTH (1-34) and a transdermal preparation are under investigation.187, 188

OTHER THERAPEUTIC CONSIDERATIONS

STRONTIUM

Strontium ranelate is made of two atoms of the divalent cation strontium along with ranelic acid. It is distributed in bone, more in cancellous bone than cortical bone, and more in new bone than old bone.189 Strontium inhibits bone resorption and stimulates bone formation. (Of note, when assessing BMD with DXA after a patient has been treated with strontium ranelate, the BMD needs to be corrected for bone strontium content).190

While receiving calcium and vitamin D, 1442 postmenopausal women with osteoporosis and at least one vertebral fracture received 2 g of oral strontium ranelate or placebo for three years in the Spinal Osteoporosis Therapeutic Intervention (SOTI) trial. There was a risk reduction in new fractures in the strontium group of 49% percent in the first year of treatment and 41% at the end of the three-year study. BMD in the strontium-treated group increased significantly from baseline by 12.7% at the lumbar spine, 7.2% at the femoral neck, and 8.6% at the total hip, with differences between the strontium- and placebo-treated groups of 14.4%, 8.3%, and 9.8%, respectively. At three years, the BMD at the lumbar spine, adjusted for the strontium content, increased 6.8% from baseline in the strontium-treated group and decreased 1.3% in the placebo group (p < 0.001).191

In the Treatment of Peripheral Osteoporosis (TROPOS) study, published in 2005, 2 g daily of strontium ranelate or placebo was randomly allocated to 5091 postmenopausal women with osteoporosis. Fracture rate and BMD were followed over three years. There was a 16% risk reduction in all nonvertebral fractures and a 19% risk reduction in major fragility fractures (hip, wrist, pelvis, sacrum, ribs, sternum, clavicle, humerus) in the strontium ranelate-treated patients compared with the placebo group. Among women at high risk of hip fracture, the RRR for hip fracture was 36% and for morphogenic vertebral fractures was 39%. At three years, strontium ranelate increased BMD 8.2% at the femoral neck and 9.8% at the total hip.192 In February, 2008, the full five-year data of the TROPOS study were published, showing a 15% RRR in nonvertebral fractures, a 43% RRR in hip fractures, and a 24% RRR in morphogenic vertebral fractures in the strontium versus placebo groups. In addition, BMD continued to increase from the three-year to five-year benchmark of strontium treatment at all sites.193

In a small osteoporosis prevention study, 40 early postmenopausal women received strontium ranelate 1 g daily for two years while 40 received placebo. After two years, the annual increase for strontium-adjusted lumbar spine BMD values was +0.66% compared with -0.5%. Femoral neck and total hip BMD were also significantly increased in the treatment group compared with placebo, thus suggesting a possible role for strontium ranelate not only in the treatment of osteoporosis, but also in prevention.194

While there have not been head-to-head trials of the anti-fracture efficacy of strontium ranelate versus other conventional anti-osteoporotic treatments, based on results from many individual therapies’ pivotal studies, strontium ranelate is effective in decreasing vertebral, nonvertebral, and hip fractures.195, 196

In general, strontium is well-tolerated. Diarrhea may occur, but in published trials of strontium ranelate treatment versus placebo, there were no significant differences in withdrawal rate based on side effects. Additional research will help elucidate possible increases in rates of venous thromboembolism, PE, headache, seizure, memory loss, and disturbance of consciousness that were suggested in earlier trials.197

Strontium ranelate has been approved as an osteoporosis treatment in Europe since 2004, and is scheduled for future review by the FDA in 2009 in the United States.

DENOSUMAB

The receptor activator of nuclear factor kappa B ligand (RANKL) is an important regulator of bone remodeling. It is secreted by osteoblasts and binds to its receptor, RANK, located on osteoclasts. The binding of RANKL to RANK positively affects osteoclast proliferation, differentiation, activation, and survival. Osteoprotegrin is an endogenous cytokine and decoy receptor that binds RANKL and inhibits osteoclast activation.198

Denosumab is a structurally human monoclonal antibody to RANKL and mimics the actions of osteoprotegrin. McClung et al. reported the effects of denosumab versus placebo in 412 post-menopausal women with low BMD (T-score –1.8 to - 4.0 at the lumbar spine or –1.8 to -3.5 at the proximal femur). Subjects who received denosumab were given subcutaneous injections every three months (6, 14, or 30 mg) or every six months (14, 60, 100, or 200 mg). At one year, variable dosing schedules of denosumab increased lumbar spine BMD 3.0 to 6.7% (compared with a decrease of 0.8% in the placebo-treated group), increased total hip BMD 1.9 to 3.6% (compared with a decrease of 0.6% in the placebo-treated group), and increased distal radius BMD 0.4 to 1.3% (compared with a decrease of 2% in the placebo-treated group). Denosumab also suppressed bone turnover markers.199 These results were confirmed in a 12-month extension of the trial published by Lewiecki et al. in November 2007. The next 12 months of treatment demonstrated continued increases in BMD, with the most striking results at the lumbar spine, where the two-year increase in BMD was 4.13 to 8.89%.200.

Bone et al. conducted a two- year, randomized double-blind placebo-controlled study of 332 postmenopausal women with lumbar spine BMD T-scores between –1.0 and –2.5. Subjects received 60 mg of subcutaneous denosumab every six months or placebo. After two years, denosumab significantly increased lumbar spine BMD compared with placebo (6.5% versus -0.6%; p < 0.0001). Denosumab also significantly increased BMD at the total hip, one-third radius, and total body (p < 0.0001 versus placebo). Increased distal radius volumetric BMD (p < 0.01), improved hip structural analysis parameters, and significantly suppressed bone turnover markers were also noted in the denosumab group.201

The overall incidence of adverse events was similar between both treatment and placebo groups in the above studies. The most common adverse events in denosumab-treated subjects in both the Lewiecki et al. and Bone et al. studies were common infections.200, 201 In the Lewiecki et al. study, incidences of hypertension and urinary tract infection were significantly higher in the denosumab-treated women than those who received placebo (10.5% vs. 0% for both)200.

In a recently published phase III, multi-center, double-blind study, 1189 postmenopausal women (T-score  2.0 at the lumbar spine or total hip) received subcutaneous denosumab injections (60 mg every six months) plus oral placebo weekly (n = 594) or oral alendronate weekly (70 mg) plus subcutaneous placebo injections (n = 595). Denosumab significantly increased BMD compared with alendronate at one year at all measured skeletal sites (12-month treatment difference: 0.9% total hip; 0.6% femoral neck; 1.0% trochanter; 1.1% lumbar spine; 0.6% 1/3 radius; p  0.0002 all sites). Denosumab led to significantly greater reductions of bone turnover markers compared with alendronate. There were no significant differences in adverse events between the treatment groups.202

Denosumab may have advantages over current osteoporosis therapies: infrequent dosing (every six months), easy subcutaneous self-administration, and rapid, effective, but reversible antiresorptive activity. The above Phase III trial data suggest better BMD and bone turnover improvement with denosumb compared with alendronate. Further trials with fracture outcomes, comparisons of denosumab to other antiresorptive therapy, and safety and efficacy studies of patients switching from other osteoporosis therapies to denosumab will likely follow. Denosumab is not yet FDA approved.

OTHER EMERGING THERAPIES

Other new therapies for osteoporosis under investigation include calcium-sensing-receptor antagonists, sclerostin inhibitors, integrin antagonists, and cathepsin-K inhibitors.203 Administration of calcium-sensing-receptor antagonists leads to a transient rise in endogenous PTH, thus mimicking the effects of intermittent exogenous administration of PTH (1-84) and PTH (1-34).204 Sclerostin, produced by osteocytes, inhibits osteoblast differentiation and ectopic bone formation. Animal studies have demonstrated that treatment with sclerostin antibodies blocked this effect and was associated with increased bone mass.205, 206 Integrins mediate cell-cell and cell-matrix interaction. Thus, integrin inhibitors disrupt osteoclast interaction with the extracellular matrix, preventing bone resorption. In a study of postmenopausal women, an v3 integrin antagonist, L-000845704, increased lumbar spine, total hip, and femoral neck BMD compared with placebo along with decreasing markers of bone turnover.207 Cathepsin-K is expressed in osteoclasts and is a cysteine protease involved in bone resorption. In randomized, placebo-controlled trials of postmenopausal women, cathepsin K inhibitors, ondanacatib and balicatib, both were associated with increases in BMD and decreases in bone resorption markers.208-210 All of the above bone modulation approaches are continuing to be explored with further trials underway.

FRAX™ CALCULATOR

The most important complication of osteoporosis is fracture, whose risk is influenced by multiple variables. Thus BMD is not the only information that should be used to assess fracture risk. FRAX™ is a fracture risk assessment tool developed by the World Health Organization to estimate the probability of fracture in postmenopausal women and older men using a combination of clinical risk factors with or without BMD measurements. The FRAX™ calculator can be found at http://www.shef.ac.uk/FRAX/index.htm.

A patient’s 10-year probability of fracture is calculated by entering the information in table 6. BMD does not need to be part of the calculator. However, if BMD is used, the T-score and Z-score values provided by commercial DXA machines do not precisely correspond to those used for the FRAX™ calculations. A "FRAX™ Patch" was developed to calculate the T-score to use with FRAX™ from various DXA BMD results, accounting for different machines, some of which use normative data that are gender and ethnic group specific. The "FRAX™ Patch" is found at http://www.nof.org/frax_patch_full.htm.

To use the "FRAX™ Patch", the brand of DXA machine on which the patient was measured is selected and then the femoral neck BMD (or total hip BMD if femoral neck not used) values are entered in units of g/cm². The T-score to use for the FRAX™ calculator is then provided by the program.

Table 5. FRAX™ calculator information
Age	years
Sex	male or female
Weight	kgs
Height	cms
Previous fracture	yes or no
Parent with fractured hip	yes or no
Current smoker	yes or no
Glucocorticoid use	yes or no
Rheumatoid arthritis	yes or no
Secondary osteoporosis	yes or no
3 or more alcohol units a day	yes or no
Femoral neck or total hip BMD if available	T-score

Bone density manufacturers now have FDA approval to include the FRAX™ calculator in their reports. With the help of the FRAX™ calculator and the “FRAX™ patch,” the National Osteoporosis Foundation (NOF) updated its clinical guidelines in 2008 for indications for osteoporosis therapy in postmenopausal women and men over age 50.

They recommend treatment:

1) after a hip or spine fracture

2) with a BMD spine or proximal femur T-score ≤ -2.5

3) with a BMD between -1 and -2.5 and one of the following:

a) history of fragility fracture since age 50

b) 10-year risk of major fracture of 20% or more (calculated by FRAX™)

c) 10-year risk of hip fracture of 3% or more (calculated by FRAX™).211

As with many predictive algorithms, there are limitations to FRAX™. The FRAX™ calculator is useful in identifying fracture risk over 10 years, particularly in patients with osteopenia. However, there are limitations to the tool, including a failure to account for risk factors appropriately. For example, if a BMD and a secondary cause for osteoporosis (e.g. aromatase inhibitor in a woman or a GnRH agonist in a male) are entered in FRAX™, the secondary cause for osteoporosis is no longer considered in the 10-year absolute fracture risk assessment. If the current BMD is osteopenic, FRAX™ will underestimate future fracture risk in such a patient. Thus, it is critically important to use clinical judgment and consider clinical risk factors for a given patient.212

CONCLUSION

At present, a number of safe and very effective therapies for osteoporosis are available. Antiresorptive agents, such as the bisphosphonates, estrogen, and raloxifene, produce modest improvements in bone mass and reduce fractures, in some instances within one year of use. Teriparatide, recombinant human PTH (1-34), is the only anabolic agent currently available in the United States, and is most useful when the recommended 18- to 24-month course is followed by bisphosphonate therapy. Therapies such as strontium and denosumab may be available within the next few years, while other novel therapies continue in Phase III trials.

Table 6. lists the currently available osteoporosis drugs approved by the FDA, their dosage, indication, and general efficacy.

* PMO= postmenopausal osteoporosis

† GIO= glucocorticoid-induced osteoporosis

Acknowlegment: The authors would like to thank Katharine H. Mikulec, M.D. for her contributions to the previous Endotext chapter last updated in 2005.

Conflicts of Interest:

Kathryn E. Ackerman, M.D. M.P.H has no conflicts of interest.

Meryl S. LeBoff, M.D. Stock ownership of Amgen and General Electric Company; no other conflicts of interest.