Bisphosphonates: Clinical Implications of Bone Quality

The Impact of Bone Quality on Bone Strength

Mary L. Bouxsein, PhD, Assistant Professor of Orthopedic Surgery at Harvard Medical School, discussed the biomechanics of fractures as well as an emerging paradigm regarding bone quality, including evidence that BMD may be a more limited measure of fracture risk than previously recognized.

A fracture occurs when the load applied to bone exceeds the bone’s strength. Many common activities of daily living place high forces, or loads, on bone. For example, simply rising from a chair without using one’s arms applies a load on the lumbar spine of approximately 170% of body weight. Similarly, bending over and lifting 30 pounds from the floor applies a load on the spine of approximately three times the individual’s body weight. It follows that in an individual with compromised bone strength, these common activities may lead to vertebral fracture. Thus it is essential to note that fracture prevention entails both minimizing the load applied to bone and measures to maintain or improve bone strength.

Until recently, osteoporosis was defined as a skeletal disorder characterized by decreased bone mass and architectural deterioration. In 2001, however, a Consensus Conference on Osteoporosis sponsored by the National Institutes of Health (NIH) proposed this revised definition: Osteoporosis is “a skeletal disorder characterized by compromised bone strength predisposing to an increased risk of fracture.” In this statement, both bone density and bone quality were cited as primary contributors to bone strength. Bone strength is understood to consist of (i) the properties of bone itself (i.e., density or porosity; bone matrix including mineralization, collagen, and the extent of collagen cross-linking; and microdamage) and (ii) bone structure or geometry (i.e., bone size, the distribution of bone mass, and both trabecular and cortical architectural characteristics). These factors determine a bone’s ability to resist fracture, and their deterioration contributes to the pathogenesis of osteoporosis. The objective of osteoporosis therapy is to arrest or prevent this deterioration in order to maintain skeletal integrity and thereby reduce fracture risk.

Density is the primary determinant of the mechanical behavior of bone as a material, explaining 60% to 90% of the various properties. Importantly, however, there is a nonlinear relationship between the strength of bone material and density. Accordingly, small changes in density may have disproportionately greater effect on strength.

Areal BMD by DXA is a standard measurement used clinically to diagnose osteoporosis and to predict risk of fracture. However, BMD measurements are limited in that they cannot measure the three-dimensional geometry of bone, distinguish cortical from cancellous bone, assess trabecular architecture, or evaluate specific properties of the bone matrix.

Clinical observations support the notion that factors other than BMD contribute to fracture risk. For example, increased age is a risk factor for fracture that is independent of BMD such that at a femoral T-score of -2.5, there is a five-fold increase in the risk of hip fracture between the ages of 50 years and 80 years. This difference may be attributable to an increased incidence of falls with advancing age and/or to age-related architectural or structural deterioration that is not detected by BMD. A second observation is that regardless of BMD, previous history of osteoporotic fracture is a risk factor for future fracture. Moreover, the severity of the previous fracture (at least in cases of vertebral fracture) also impacts future fracture risk regardless of BMD. Finally, high bone resorption, as measured by bone turnover markers such as C-terminal cross-linked telopeptide of type 1 collagen or deoxypyridinolone, is a fracture risk that is independent of BMD. Accordingly, a prospective epidemiological study showed that individuals at highest risk for fractures of the femoral neck are those with low femoral BMD and high bone resorption.

Certainly, BMD measurement remains the most reliable tool for diagnosis of osteoporosis, but other clinical factors are important in patient evaluation as well. Compelling clinical evidence indicates that age, positive history of prior fracture, history of a severe prior vertebral fracture and high bone resorption are all important factors for identifying those individuals at highest risk for fracture.

BMD measurement has also been used as a surrogate marker to monitor the efficacy of osteoporosis treatments. Randomized clinical trials have shown, however, that whereas increases in spine BMD with osteoporosis treatments vary markedly, the decrease in the incidence of vertebral fractures is relatively similar across studies. Moreover, it appears that the increase in spine BMD explains a relatively small proportion of the reduction in fracture risk. Furthermore, reductions in fracture risk associated with antiresorptive therapies occur relatively early in treatment, long before maximum gains in BMD are observed. Altogether, these observations imply that factors other than BMD influence age-related bone disease and treatment-related changes in bone strength. Thus they also influence fracture risk.

Among the factors that influence bone’s resistance to fracture are bone geometry, microarchitecture, and properties of the bone matrix itself. Thus, to illustrate how the distribution of mass can influence the mechanical behavior of bone, consider this example. Imagine that a ruler bends easily when loaded to bend along the flat surface but not when it is turned on its side. The total mass of the ruler is unchanged, yet its mechanical behavior changes dramatically just by changing the distribution of this mass relative to the forces applied to it. The implication is that change in bone geometry can have a dramatic effect on strength without much change in mass.

Considering the role of bone microarchitecture, it is well known that there are marked age-related changes in trabecular bone, including a decrease in trabecular bone volume fraction, trabecular thickness, the number of trabeculae and, in particular, a preferential loss of trabecular struts that are oriented in a manner perpendicular to the primary loading direction. A recent study confirmed the contributing role of increased bone resorption to these architectural deteriorations. In this study, iliac crest biopsies were attained at baseline and after one year from early menopausal women receiving either placebo or bisphosphonate (risedronate) therapy. Trabecular microarchitecture was assessed in the biopsies by high-resolution microcomputed tomographic imaging. Notably, untreated women had a 3% decline in spine BMD but a 20% decrease in trabecular bone volume fraction at the iliac crest along with decreases in trabecular number, thickness, and connectivity. However, in women treated with risedronate, this deterioration of trabecular architecture in the iliac crest was prevented.

Deterioration of cortical bone microarchitecture subsequent to in-creased resorption also contributes to increased fracture risk. For instance, patients who suffered hip fracture had preferential thinning of the anterior inferior cortex and increased cortical porosity in the femoral neck. Similar to what has been seen in trabecular bone, it appears that reduction of bone resorption by antiresorptive therapy may prevent cortical architecture deterioration as well. Thus evaluation of iliac crest biopsies from postmenopausal women treated for 2 to 3 years with bisphosphonate therapy (alendronate) revealed that individuals treated with alendronate had 50% lower cortical porosity than subjects receiving placebo. Clinical evidence implies, therefore, that preserving cortical and trabecular bone architecture by antiresorptive treatment may help maintain bone strength.

Finally, changes in the bone matrix itself may influence bone strength and fracture risk. Thus another means of increasing bone mass (and bone strength) is by increasing the degree of mineralization of the bone matrix. When new bone matrix is formed, it is initially universalized. After mineralizing relatively rapidly (primary mineralization) up to approximately 85% of its maximum mineralization, a phase of slower secondary mineralization begins, and full mineralization is achieved after years, or perhaps decades. Note that this second phase of mineralization is influenced markedly by the rate of bone turnover, such that when bone turnover is high, there is not enough time for full completion of secondary mineralization and bone is, therefore, relatively younger and under-mineralized. Evaluation of iliac crest biopsies from individuals treated with antiresorptive therapy has confirmed that reducing bone turnover leads to an increase in the mean degree of
mineralization of the bone matrix. Specifically, compared with baseline, mineralization of the iliac crest bone biopsies increased 5% following 3 years of treatment with calcium plus vitamin D and 7% following the same duration of raloxifene therapy. In comparison, postmenopausal women treated with bisphosphonates, a more potent anti-resorptive therapy, had 7% to 10% higher mineralization values in iliac crest biopsies than women treated with placebo.

Efficacy of Osteoporosis Therapy: What Is the Evidence?

Despite all the properties of bone that may contribute to bone strength, there is no sound means of assessing antifracture efficacy of osteoporosis therapies other than clinical trials that use fracture endpoints. Nelson B. Watts, MD (University of Cincinnati) reviewed the evidence from major clinical trials of FDA-approved drugs that are currently marketed in the United States.

Data published in 2002 from the Women’s Health Initiative indicated that although combination estrogen-based hormone replacement therapy is effective in reducing the risk for hip, vertebral, and other osteoporotic fractures, the overall risk:benefit ratio is such that it is not suitable therapy for the prevention of chronic diseases such as osteoporosis.

Dr. Watts began with data regarding prevention of vertebral fractures. He summarized the data from multiple trials by saying that “alendronate, raloxifene, and risedronate all prevent first vertebral fractures in women, and alendronate, calcitonin, raloxifene, risedronate, and teriparatide will prevent new vertebral fractures in women who have previously had them.”

With regard to the prevention of nonvertebral fracture, Dr. Watts cautioned that the study data are not
easily compared because not all define “nonvertebral fracture” in the same manner. Nonetheless, data from the vertebral fracture arm of the Fracture Intervention Trial (FIT) indicate that alendronate had no effect on nonvertebral fracture except for a significant reduction in risk for hip fracture. However, the Fosamax International Trial (FOSIT) and a post hoc analysis of the FIT data have demonstrated the effectiveness of alendronate for prevention of nonvertebral fractures. With calcitonin and raloxifene, no benefit has been shown for the prevention of nonvertebral fracture including hip fracture. In the Vertebral Efficacy with Risedronate Therapy (VERT) study, risedronate was associated with a significant reduction in nonvertebral fractures. In the only prospective trial in which hip fracture was a primary endpoint, a significant hip fracture benefit was demonstrated with risedronate. In the single pivotal trial of teriparatide, there was a significant reduction in the risk for nonvertebral fracture, but no effect on hip fracture was observed. In summary, “the data suggest that alendronate, risedronate, and teriparatide have demonstrated effectiveness against nonvertebral fractures, but only the two marketed bisphosphonates, alendronate and risedronate, have demonstrated effectiveness against hip fracture.”

Dr. Watts pointed out, however, that an important component of evaluating these agents is the early effect on fracture prevention. In the risedronate VERT studies there was a significant reduction in radiographic vertebral fractures after 12 months of treatment. Post hoc analysis of clinical vertebral fractures found significant reductions after 12 months’ treatment with risedronate, raloxifene, and alendronate. Further analysis with risedronate shows a significant reduction of both vertebral and nonvertebral fracture after 6 months of therapy. With alendronate, a significant reduction in nonvertebral fracture, and in hip fracture in particular, occurs after 18 months. With teriparatide, there appears to be little benefit relative to placebo for 10 months, and the subsequent separation does not become statistically significant for 18 to 24 months.

In addition to the onset of efficacy, it is important to look at the sustained effectiveness of osteoporosis therapies. Extension data from trials indicate that there is a benefit on vertebral fracture associated with the fourth extension year of the raloxifene study and the fourth and fifth years of the risedronate study.

Beyond BMD: Understanding the Link Between Bone Quality and Fracture Risk

Although multiple characteristics of bone contribute to its strength and to the risk for fracture, bone density remains a very important tool in assessing risk for fracture, diagnosing osteoporosis, and monitoring therapy. BMD measurement is important because it is precise, accurate, safe, and readily available. The fundamental concept that has established bone mineral density as a key measurement technology is the relationship between BMD and fracture risk. In general, for a given standard deviation decrease in bone density there is approximately a doubling of fracture risk. In emphasizing this relationship, John P. Bilezikian, MD (College of Physicians and Surgeons, Columbia University) said that “there are very few other relationships between a surrogate endpoint such as bone density and a true endpoint such as fracture risk that are as powerful in clinical medicine.” One would expect on the basis of this observation that an increase in bone density due to treatment would be associated with a reduction in fracture risk.

Although antiresorptive agents increase bone density and reduce fracture, a critical issue is the extent to which this relationship is linked in the therapeutic setting. Several meta-analyses of changes in bone density and reduction in vertebral fracture risk indicate that changes in bone density account only incompletely for the reduction in fracture risk. Fifty percent or more of the reduction in fracture incidence is due to indices that are not part of the bone
density measurement.

Raloxifene, a selective estrogen receptor modulator (SERM), illustrates this point well. The modest change in bone density induced by raloxifene therapy would be expected to be associated with a 10% reduction in vertebral fracture. The actual reduction, however, is considerably higher: 40% to 50%. From these data, Dr. Bilezikian concluded that “just as factors other than reduction in BMD contribute to fracture risk, so do factors other than increased bone density contribute to reduction in fracture risk.” Properties of bone other than its density help to account for the efficacy of antiresorptive agents in preventing fracture.

Maintenance of normal bone remodeling requires a balance between osteoblastic (bone forming) and osteoclastic (bone resorbing) activity. This balance is altered in the aging skeleton in which there is greater bone resorption than bone formation. Bone biopsies taken prior to and at various times after the menopause indicate an increase in the activation frequency, or bone turnover. This increase in number of bone turnover sites creates points of mechanical vulnerability and may lead to microdamage. Changes in bone turnover can be measured by urinary cross-linked N-telopeptides of type 1 collagen (NTX). Treatment with bisphosphonates such as alendronate or risedronate results in a reduction in NTX. This change occurs in parallel with the reduction in vertebral fracture although, for risedronate, there may be a plateau beyond which further reductions in bone turnover may not be associated with additional re-ductions in fracture risk. This plateau may not be the case for alendronate.

Another important factor in accounting for bone strength is microarchitecture. Microarchitecture refers, in part, to the orientation and connectivity of horizontal and vertical trabecular struts. As horizontal struts are lost with aging or via increased bone turnover, the effective vertical length of the trabecular strut becomes greater and, therefore, weaker, according to Euler’s theorem. With the loss of these horizontal cross ties, the vertical struts are weakened in their capacity to withstand stress. Although there is little evidence to support a claim that bisphosphonates reverse microarchitectural deterioration, there is evidence that risedronate therapy may prevent deterioration. Similar data are not yet available for alendronate.

In summary, it can now be said that antiresorptive agents improve bone mineral density, reduce bone turnover, and prevent progression in the microarchitectural deterioration of bone. In order to understand thoroughly how antiresorptives reduce fracture risk, we need to appreciate factors beyond BMD that contribute to bone strength.