Dennis Black, Ph.D.

University of California, San Francisco

Dr. Black reviewed one of the key issues in diagnosis of osteoporosis–use of bone mineral density (BMD) measurements both as a diagnostic tool and as a risk factor for fracture. He began by describing the technology for measuring BMD. Known as a dual energy x-ray absorptiometry or DXA, this test measures BMD at multiple skeletal sites, including the spine and proximal femur. An office-based procedure that involves minimal exposure to radiation, the DXA test costs roughly $125 and is in most instances covered by Medicare.

As illustrated in Figure 3, hip BMD is a good predictor of the risk of fractures in Caucasian women over the age of 65. Those individuals in the highest-quartile with respect to hip BMD (i.e., those with the highest levels of bone mass) are at the least risk of a fracture, while those with low hip BMD have significantly higher risk of fracture. In fact, hip BMD is a better predictor of hip fracture than are standard screening tests for other diseases. For example, BMD levels do a better job of predicting the risk of hip fracture than do cholesterol levels at predicting heart disease. That said, measurement of BMD at other parts of the body (e.g., a wrist, a heel) tends to be less predictive of fracture risk, especially hip fractures. While these peripheral tests have some advantages (they are less costly, emit less radiation, and are portable), they also tend to be less reliable as tools for predicting fracture. Because of these problems, the consensus in the field is that central densitometry should be used for a definitive diagnosis whenever possible.

Figure 3

Femoral Neck (Hip) BMD Predicts Various Types of Fractures. Sources: Black DM, Cummings SR, Genant HK et al. Axial and appendicular bone density predict fractures in older women. J Bone Mineral Res 1992; 7(6):633–638. Cummings SR; Black DM; Nevitt (more...)

Dr. Black noted the development of a classification system that translates BMD scores into diagnosis. This system relates an individual’s bone density to that of a normal population of individuals. This comparison uses a “t-score” to indicate the number of standard deviations that an individual falls below or above a standardized normal BMD (0.89 grams per square centimeter, a level that occurs in most individuals around puberty). The World Health Organization (WHO) created three diagnostic categories based on t-scores. Individuals within one standard deviation of peak BMD are considered normal, those one to 2.5 standard deviations below normal are deemed to be osteopenic or have low bone mass, while those 2.5 or more standard deviations below peak BMD are considered to have osteoporosis. The National Osteoporosis Foundation (NOF) recommends that physicians seriously consider prophylactic therapy for individuals with a t-score that is two or more standard deviations below peak BMD. Individuals with known risk factors should be considered for such treatment if t-scores fall 1.5 or more standard deviations below peak BMD.

BMD, however, is not the only risk factor that is predictive of the potential for fractures. While each one-standard-deviation drop in BMD (equivalent to a 10% to 13% decrease) increases the risk of hip fracture by 2.4 times, other risk factors are also strongly predictive of hip fractures, including age (every five-year increase raises the risk of hip fracture by 50%), weight (every 20% decline in weight after the age of 25 increases the risk of hip fracture by 70%), a history of fracture since the age of 50 (which raises the risk of a hip fracture by 50%), having an existing spine fracture, or having a mother who has fractured her hip. (These latter two factors each double the risk of hip fracture.) As illustrated in Figure 4 below, these clinical risk factors are independently predictive of hip fracture risk. In other words, for a given BMD, individuals with more risk factors have a higher likelihood of suffering a hip fracture.

Figure 4

Clinical Risk Factors Independently Predict Hip Fracture Risk. Source: Cummings SR; Nevitt MC; Browner WS; Stone K; Fox KM; Ensrud KE, et al. Risk factors for hip fracture in white women. N Eng J Med 332(12):767–773, 1995.

Looking to the future in the field of diagnosis, Dr. Black noted that there is a movement to use fracture risk for diagnosis and to decide whether an individual needs treatment. Using BMD and other risk factors, the risk of fracture can be accurately predicted. Ongoing efforts to redefine diagnosis based on risk are currently underway; they have the potential to unify diagnosis across gender, ethnic groups, and countries. But several challenges remain, including how to integrate the risks and consequences of various fracture types, and how to alter treatment recommendations based on fracture risk. At present, treatments such as bisphosphonates have been shown to prevent nonspine and hip fractures only in those with low BMD. Thus, it is not clear what treatment, if any, will be effective for an individual who has normal BMD but has other risk factors. Other controversial issues in diagnosis include how to apply t-score categories to men and non-Caucasian women, determining optimal BMD values for treatment and optimal use of fracture risk for diagnosis, and determining the proper role (if any) for peripheral densitometry. On this latter point, Dr. Black suggested that peripheral BMD measurement could be used as a pre-screening device to raise awareness among the public.

Susan Greenspan, M.D.

Osteoporosis Prevention and Treatment Center

Clifford Rosen, M.D.

St. Joseph Hospital

Dr. Greenspan illustrated the challenge in preventing fractures. As shown in the chart below, the goal is to prevent the young, normal bone from becoming osteoporotic bone. But when osteoporosis does occur, treatments can help to reverse this evolution, making a previously brittle bone stronger. These treatments include antiresorptive therapy that helps prevent bone loss, anabolic therapy that helps build bone, and combinations of therapies, such as two antiresorptive therapies or an antiresorptive therapy combined with an anabolic therapy. One key issue with combination therapy is whether to use the therapies at the same time (concurrently) or one after the other (sequentially).

Figure 5Lumbar Spine

Source: Image courtesy of Ralph Müller, Ph.D., Switzerland, ETH and University Zürich.

Dr. Greenspan and Dr. Rosen reviewed the evidence to date on a variety of antiresorptive and anabolic therapies. Dr. Greenspan cautioned that there is little data that allow for head-to-head comparisons between therapies. She also noted that most studies evaluate improvements in BMD, which are not directly correlated with fracture reduction. That said, evidence suggests that the greater the improvement in BMD, the greater the fracture risk reduction (even though changes in BMD account for only about one-third of overall risk).

Henry G. Bone, M.D.

Michigan Bone and Mineral Clinic

Dr. Bone reviewed how drugs for bone disease are developed and approved. He began by describing the key measures that are used to determine a drug’s efficacy at various stages of development. Early testing focuses on pharmacology measures, including blood and urine levels that help to estimate drug absorption, metabolism, and excretion. In addition, biochemical markers are measured as a means of determining the drug’s impact on bone metabolism. Later in the process, the drug’s impact on bone density is measured at multiple sites. This clinical test is considered very important, since bone mass is a major mechanical determinant of strength and, as noted earlier, is highly predictive of fracture. Fracture rates, however, remain the ultimate clinical outcome. Spine, nonspine, and hip fractures are typically considered separately.

Pre-clinical drug development involves the screening of millions of compounds to find one or a few that have an effect on bone. Once a compound with strong potential is discovered, it is tested on animals to determine the impact on pharmacology and toxicology. Assuming these trials suggest that the compound is safe and potentially effective, human clinical trials begin. Phase I testing involves evaluation of short-term pharmacology and safety issues in healthy individuals. The primary goal is to determine if the drug can be safely tolerated. Phase II may be the most important part of the testing, as it focuses on efficacy, dose-finding, tolerability, and the impact of the drug on biochemical markers and BMD. Roughly 1,000 individuals are involved in this phase for a period of approximately one year. Phase III may involve anywhere from 5,000 to 20,000 individuals. These trials focus on making a definitive determination of the drug’s efficacy and safety, including its three-year impact on BMD and fracture rates at different skeletal sites.

Looking ahead, Dr. Bone urged drug companies and regulators to make sure that drug evaluation is as efficient as possible while still maintaining reliability. He emphasized the importance of rigor during phase II trials designed to determine the appropriate dose, and urged reconsideration of statistical requirements for evaluation of the impact of drugs on secondary fracture sites (if the drug has been proven successful at a primary site). He also called for greater coordination between the U.S. and EU regulatory authorities, and reminded drug developers and regulators not to forget about drug treatments for bone diseases other than osteoporosis.

Eric Orwoll, M.D.

Oregon Health Sciences University

Despite the conventional wisdom that osteoporosis is a “woman’s disease,” many men are affected as well. Just over 13% of men aged 50 will have a fracture sometime in their life. While BMD is a good predictor of fracture risk in men, there are no set criteria for determining when men have osteoporosis and therefore need treatment. Using the same criteria that define osteoporosis in women, roughly one in seven men over the age of 80 has the disease.

Men get osteoporosis for a variety of reasons, although one-third to one-half of the cases are primarily the result of genetics. Environmental factors also play a role, including alcoholism. Men with prostate cancer who are being treated with androgen deprivation therapy are also at greater risk of bone loss.

Treatments such as alendronate and PTH have been shown to markedly increase BMD and reduce the risks of fracture in men (see Figure 6). But men are much less likely to have the disease diagnosed and treated, even after a fracture occurs (see Figure 7).

Figure 6

Therapy in Osteoporotic Men. Sources: Orwoll E; Ettinger M; Weiss S; Miller P; Kendler D; Graham J, et al. Alendronate for the treatment of osteoporosis in men. N Engl J Med 2000 Aug 31; 343(9):604–10. Orwoll E; Belknp JK; Stein RK. Gender specificity (more...)

Figure 7

The Care of Men After Fracture. Sources: Feldstein et al. In press. Kiebzak GM; Beinart GA; Perser K; Ambrose CG; Siff SJ; Heggeness MH. Undertreatment of osteoporosis in men with hip fracture. Arch Intern Med 2002 Oct 28; 162 (19):2217–22.

Jane Cauley, Ph.D.

University of Pittsburgh

Caucasian women have higher rates of osteoporosis than do African-Americans and Asians, but the rate of bone loss in all ethnic groups increases with age. (Differences in body weight and the length of the hip axis appear to be important factors in explaining differences in BMD across ethnic groups.)

African-American women are more likely than Caucasian women to die from a fracture. Risk factors for African-American women are similar to those for Caucasians and include having low body weight in relation to height (a condition known as low body mass index or low BMI–a reading below 22.6 raises the risk of hip fracture by 13.5 times), alcoholism (consuming more than seven drinks a week raises the risk of hip fracture by 4.6 times), a history of stroke (which raises the risk of hip fracture by 3.6 times), and use of ambulatory aids (which raises the risk of stroke by 5.6 times). Each one-standard-deviation decrease in BMD in African-American women raises the risk of fracture by 80%, compared to 40% for Caucasian women. Fortunately, treatments for osteoporosis appear to be effective in African-American women. A study of alendronate, for example, found that the benefits of the drug to African-American women were similar to the benefits for Caucasian women. Dr. Cauley noted that further study is needed on the efficacy of various therapeutic agents across ethnic groups.

Nelson Watts, M.D.

Osteoporosis Center, University of Cincinnati

Bone diseases are often silent conditions that exist for years before any obvious signs manifest (e.g., a fracture). The only way to definitively diagnose the disease is through a bone density test. One of the barriers to testing is coverage, since the cost of the test is relatively high. Since July 1, 1998, Medicare has covered baseline bone density testing for certain individuals that are deemed to be at risk for bone disease (e.g., estrogen-deficient women, patients with abnormalities of the spine, patients receiving long-term glucocorticoids, and patients with primary hyperparathyroidism). Payment is based on diagnoses, which are conveyed by a complex set of codes that are used inconsistently. Along with a lack of insurance coverage, the inability to travel to a testing center may also limit access to diagnostic testing. While the nation has an adequate number of central DXA machines to adequately meet the needs of the 35 million Americans–including 20.5 million women over the age of 65 and 10 million men over the age of 70--who likely need to be screened on a regular basis, the geographic distribution of these machines may make it difficult for individuals in rural regions to get access to a test.

Assuming that individuals are screened, effective pharmacologic therapies for osteoporosis are available. But coverage of these therapies by private insurance is inconsistent, while none of these drugs is covered by Medicare in the outpatient setting. Nonpharmacologic treatments such as calcium and vitamin D, physical therapy, and hip protectors, are also not typically covered.

Dr. Watts noted that the limits to access to screening and treatment for osteoporosis are similar to problems in financing and delivering care for the prevention of other chronic diseases. Looking ahead, he hopes for the development of new diagnostic strategies that are less costly, and for changes in coding and reimbursement to improve access to testing and treatment.

Ethel Siris, M.D.

Columbia University

Access to screening and treatment for osteoporosis begins with awareness of the problem. Dr. Siris shared data from the National Osteoporosis Risk Assessment (NORA) study, which assessed the scope of the problem of osteoporosis and osteopenia in women. This study evaluated 200,000 postmenopausal women without known osteoporosis. All women completed questionnaires on risk factors and had a BMD performed at a single peripheral site (the heel, forearm, or finger) at baseline. The study found that 7.2% of the study population had osteoporosis (defined as a t-score 2.5 or more standard deviations below peak bone mass), while 36.9% had low bone mass (defined as between 1.0 and 2.49 standard deviations below peak bone mass). Bone mass scores declined with age in all ethnic groups, including Caucasians, Asians, African- Americans, Native Americans, and Hispanics. Similarly, fracture rates were significantly higher for those with the lowest t-scores in all ethnic groups (see Figure 8). Perhaps the most interesting finding from the study was that even though the risk of fracture is much higher in individuals with osteoporosis, the greatest absolute number of fractures occurred in individuals with low bone mass (since there are roughly five times more individuals with low bone mass than with osteoporosis). Dr. Siris emphasized the importance of finding ways to diagnose and help these individuals as well as those with osteoporosis.

Figure 8

Osteoporotic Fracture Rates, Population BMD Distribution and Number of Fractures. Source: Siris ES, Miller PD, Abbott TA, Chen Y, Faulkner K, Barrett-Connor E, Berger M, Santora A, Sherwood L. J Bone Miner Res 2001; 16:Suppl 1, S337.

The panel included a presentation from an individual whose personal story highlights the need for better tools for diagnosing and treating osteoporosis.

Annie Lorigan

Ms. Lorigan is a 73-year-old woman with a long history of osteoporosis. She suffered her first fracture 18 years ago, and since that time has had eight additional fractures, each of which caused tremendous pain and required hospital stays and long periods on various medications. Unfortunately, Ms. Lorigan does not tolerate many osteoporosis medications very well. As a result, her treatment consists primarily of estrogen, Vitamin D, and calcium supplements.

Osteoporosis affects every part of her life. She must limit the time she spends with her grandchildren, as well as the types of activities she can enjoy with them. (She has fractured her back three times while playing with her grandchildren.) She finds it impossible to lie down on her back or right side, and finds it difficult to get in and out of bed or a chair. She has had to give up dancing, one of her favorite activities, and feels she has become a “drag” on family members who must slow down to accommodate her limitations.

Ms. Lorigan called for the following actions: a national education campaign focusing on early diagnosis of osteoporosis and other bone diseases (modeled after the diabetes campaign), development of new medications that are easier to tolerate, and greater understanding, patience, and compassion among doctors.