I'm going to offer you a clinician's perspective on some new insights into managing osteoporosis. I'd like to focus on, first, patient identification and then look at new insights into available treatment options.

First to look at how to identify high-risk patients.

I'll begin with a case of Mrs. Brown, a 70-year-old who presents for a physical. Like the song, she has a lovely daughter, because she's moved in with her daughter and has relocated to your area and is coming to you for a physical. She has a past history of hypertension, asthma, depression, and breast cancer, and her medications right now are amlodipine, montelukast, albuterol, and sertraline. She's appearing pretty healthy. She's petite, 5'3", and 118 pounds. Her pressure is normal, and she appears to be in no acute distress. Routine labs and studies are ordered, including a dual energy x-ray absorptiometry (DXA), appropriate for a 70-year-old.

Her DXA shows a T-score of -1.3 at the lumbar spine and -1.4 at the femoral neck. So the question is, what if anything, would you do for this patient? And what is her risk for fracture? Well, at first blush, you'd say, wow a T-score of -3, -4, not really all that bad and doesn't quite meet standard criteria for treatment.

If you look at her 10-year probability (and this was a study by John Canus looking at Swedish women, so we may need to extrapolate this a little bit to US women), looking at age and femoral neck T-score, her 10-year probability for fracture is actually around 14%, and that really is even higher than say a 50-year-old with a T-score of -2.5 or -3, someone who we would identify much more clearly as being at risk for fracture. The World Health Organization is working on a new algorithm to help us identify patients at higher risk.

Age and bone mineral density (BMD) are independent risks for hip fracture, and right now the T-score, as we know, that defines osteoporosis is a T-score of -2.5. Traditionally, these are the patients, highlighted in blue, who we would target, anybody below a T-score of -2.5 really regardless of age, but if we factor in age and look at a different number, an absolute fracture risk, and look at a 10-year fracture probability, we can really change this paradigm to identifying those with a certain fracture risk and those highlighted here in yellow would really be the ones we would want to target. Our 70-year-old with a hip T-score of -1.4 and other factors, including asthma and perhaps being on steroids periodically and a history of breast cancer in the past, may put this patient really over the top to where we'd want to treat her. Exactly what that absolute fracture risk is going to be is a wavering line right now. Is it going to be 5% 10 year, 10%, 8%? It is going to vary perhapsfrom country to country depending on a country's resources, but this just helps us in changing our look at a high-risk patient and maybe targeting someone whom we might not otherwise have targeted 5 or 6 years ago.

The WHO has proposed an approach to fracture risk assessment and is including other risks, advanced age obviously being at the top of that list, a history of fracture over age 50, family history of hip fracture, use of corticosteroids, alcohol greater than 2 drinks a day, so fortunately it's not just at two, it's greater than two, so most of us should be okay there -- (laughter) -- smoking, and a history of rheumatoid arthritis. Those are under proposal, and hopefully we'll hear more about this shortly.

I am just going to highlight some recent developments with regard to medications. I think you are all pretty familiar with the medications listed here, but I have highlighted in yellow proton pump inhibitors and selective serotonin reuptake inhibitors, which have been recently demonstrated to increase risk for fracture.

With regard to new insights into treatment options, let's again turn to our case.

Mrs. Brown is admitted 2 years later with a hip fracture after tripping over her grandson's toy. Her hip was pinned. She was sent to rehab. She now returns to your office using a walker. Her orthopedist tells her to be sure and check with you concerning whether treatment is needed. Her meds are the same with the addition of calcium and vitamin D that was started a couple of years earlier.

Well, why did Mrs. Brown break her hip? In addition to bad genes, genetic predisposition perhaps, she has weakness, perhaps, or a fall risk due to physical inactivity, due to low vitamin D, which we'll hear more about by Dr. Heaney later. She might have poor vision, she didn't see the toy, low BMD due to lifestyle, or she didn't drink milk growing up. Her medications might be contributing. She might have increased bone turnover, leading to both trabecular and cortical issues, which we will hear more about from Dr. Sebba. So, no matter why she broke her hip, what are we going to do about it?

She's encouraged to do gait and balance exercises and resume calcium with an appropriate dose of vitamin D. Based on recent studies, what could have been done for this patient to have prevented her hip fracture?

Well, if we look at hierarchy of evidence, we do have the ORAG data from 2002 -- Osteoporosis Research Advisory Group -- looking at a meta-analysis of fracture reduction efficacy. And with regard to vertebral fracture reduction, we see that alendronate, risedronate, raloxifene and calcitonin were all shown to reduce spine fractures. With regard to nonvertebral fracture risk reduction, we see that only alendronate and risedronate met the criteria that were significantly reducing those fractures.

They did find in the ORAG analysis that hip fractures were reduced with alendronate, but at the time there was insufficient evidence that other agents reduce hip fractures. More recently, Lieberman evaluated data on nonspine and hip fracture reductions with the available antiresorptive agents.

So he looked at this again, this meta-analysis of randomized, controlled trials and found that, with regard to treatment effects on nonspine fracture risk, with alendronate, whether you break it down to 10 mg or more a day, there was a .51 relative risk of nonspine fracture; with alendronate greater than 5 mg a day, a .66 relative risk for nonspine fracture; with hormone therapy, a .75 relative risk, contributed to in large part by data from the Women's Health Initiative; and with risedronate at 2.5 mg or 5 mg or even at 5 mg a day only, not much difference there, with a .73 or .68 relative risk of nonspine fracture. What about hip fracture risks?

Again, with alendronate at 5 mg or 10 mg a day, looking at hip fracture risk, we have a .5 relative risk. If you look at 5 mg or 10 mg a day and slightly less osteoporotic group, either osteoporotic or a T-score of -2, we have a .55 risk; and with hormone therapy, a .64 risk; and risedronate offers a .74 risk of hip fracture.

This analysis, even though it was more recent, looked at ibandronate data, calcitonin, etidronate, and raloxifene data, and there was insufficient evidence of fracture treatment effects with regard to hip fracture with these agents. A post publication note from this study did describe the RUTH (Rationale and Overview of the Raloxifene Use for the Heart) trial, which had 10,000 women in it looking at raloxifene use and the heart, but also captured fractures, and there was no evidence of nonvertebral fracture benefit with raloxifene. So the conclusion of this paper was that alendronate reduces hip fractures by 45% to 55%, hormone therapy by 25% to 35%, and risedronate by 26% to 27%. These could have been agents that might have helped Mrs. Brown not break her hip.

Well let's do a little case flashback, because you would have started Mrs. Brown on a bisphosphonate at the appropriate time. She never ended up having the hip fracture. So now she is coming in after 5 years of bisphosphonate treatment. Her T-scores are improved.

Now the question is, what evidence do we have with regard to long-term bisphosphonate treatment? Can they be used safely? There are a lot of questions in the press, a lot of questions in our minds, and there is evidence of long-term safety and efficacy? So what is the data?

Well, I can't discuss safety in bisphosphonates without bringing up osteonecrosis of the jaw (ONJ), otherwise referred to by our dental compatriots as bisphosphonate-related osteonecrosis of the jaw or BRON, or B-R-O-N, but I will refer to it fondly as ONJ, which is a poorly-healing oral lesion, usually seen with tooth extraction without head or neck malignancy or radiation resulting in a poorly healing lesion that can be quite annoying for any patient who has it. The known risk factors for ONJ include a diagnosis of cancer. Concomitant therapy such as chemotherapeutic agents, radiation, corticosteroid therapy, and poor oral hygiene, which is really found to be a significant contributor. A recent Australian study showed that this was definitely a factor in their ONJ cases that were reported. Comorbid disorders including significant dental problems, anemia, diabetes, which can affect healing, smoking, and alcohol use.

Well, most cases of ONJ are associated with the IV formulation. Oral bisphosphonates account for approximately 5% of total patients reported worldwide and in the Merck database, the reported incidents with alendronate is .7 per 100,000 person years of exposure. So in all of the clinical trials of alendronate, there have been no reported cases of ONJ, including trials looking at use of alendronate in periodontal disease. Rare cases of ONJ have been seen without risk factors or even without bisphosphonate treatment. In the Horizon study with IV zoledronic acid, there were 3 cases of ONJ noted in that trial, it was the IV trial, but 2 cases were in the placebo group. We clearly need to study this further, but we really don't yet have full definitive answers with regard to ONJ and bisphosphonate. One thing that may be emerging, and is being seen, is that risk may be associated with an increased duration of treatment.

The American Association of Oral and Maxillofacial Surgeons (AAOMS) has issued a position paper on bisphosphonate-related osteonecrosis of the jaw. And this is one of the few times I am referring to hypothesis, rather than evidence-based data, but this is really all we have with regard to ONJ. We don't have any randomized, placebo-controlled, prospective trials evaluating this, except to look at the trials that have already been conducted. It is clearly a very rare event. But their recommendation is, if systemic conditions permit, to delay the start of bisphosphonate until dental health is optimized. So, if you have a patient you've diagnosed with osteoporosis and it looks like their teeth are in great need of repair, you may want to hold off on bisphosphonate therapy, let them get their teeth addressed, then begin treatment 2 to 3 months after healing or after those dental procedures have been carried out. For asymptomatic patients with no clinical risk factors, who are receivingoral bisphosphonates, if they've taken them for less than 3 years, no delay in planned surgery should be done. But if they've been on it for greater than 3 years, these dental surgeons are recommending to suspend bisphosphonate therapy for 3 months, if conditions permit and then restart these agents after healing has occurred. Again, this is based on hypothesis. So keep that in mind that they have issued this, this is why patients are coming into our offices saying, hey doc, my dentist isn't going to do anything to me until you say that it is okay for him to do something. I urge you to go to their Web site, and you can download their paper. You can actually just Google AAOMS and this will pop right up, and you can go right to their site and see this position paper. Well, what about other safety data besides ONJ? Do we have other evidence of long-term safety with bisphosphonates?

And we have with risedronate triple iliac crest biopsies collected at baseline and at 3 to 5 years showing normal mineralization in those biopsies, follow-up with VERT (vertebral efficacy with risedronate) multinational study shows a low incidence of radiographic vertebral fracture after up to 7 years.

What you can see here is the annualized incidents from the VERT multinational carried out through 7 years. You can see with risedronate after 3 years, after 5 years, and after 7 years, that the number of fractures is not increasing, so we have a low incidence of fractures with risedronate carried out through 7 years.

We also have data now 10 years with alendronate from the fracture intervention trial long-term extension or FLEX, the purpose of which was to examine the effects of continuation of alendronate for 10 years, vs 5 years and then placebo for 5 years. Primary endpoint was change in BMD at the total hip, various secondary endpoints, including safety and intolerability and also looking at bone histomorphometry.

Very briefly, the FLEX study design is the (Fracture Intervention Trial) FIT trial and then 2 year, 1 to 2 years of open-label alendronate and then patients who were on alendronate for those first 5 years were randomized to receive placebo, alendronate 5 mg or alendronate 10 mg.

Looking at the primary endpoint total hip BMD, you can see that for 10 years of alendronate treatment, that you're having only a 1% decrease from FLEX baseline carried out through 10 years with the use of alendronate. If you use alendronate for 5 years and then use placebo for 5 years though, of note total hip BMD returns to the FIT baseline and the difference between the 2 groups is 2.4% and the difference from FLEX baseline is 3.4%. You really get a loss of hip BMD with cessation of alendronate therapy after 5 years of therapy.

With regard to time for first nonvertebral fracture, identical lines here with regard to the curves alendronate for 5 years or alendronate for 10 years.

Of interest again, with regard to time for first clinical vertebral fracture, there was a significant difference between the use of alendronate for 10 years, which offered a 55% reduction in clinical vertebral fracture compared with alendronate for 5 years and placebo for 5 years.

Adverse events were really comparable in the 2 groups and comparable to other data on alendronate.

Reassuringly, bone biopsies of the 5-year alendronate, 5-year placebo and 10-year alendronate showed bone of normal mineralization and no quantitative or qualitative differences.

To sum up, the use of the 10-year absolute fracture risk may enhance our ability to identify patients at higher risk for fracture. Recent meta-analyses have demonstrated that alendronate hormone therapy and risedronate reduce hip fractures.

ONJ is fortunately a rare event with oral bisphosphonates, and we can be reassured that bisphosphonate use for up to 10 years maintains the increases in BMD and reduces the relative risks for clinical vertebral fractures, and, most important, there is no evidence of an increase in adverse events or abnormal bone on biopsy. Thank you.

Dr. Bonnick: Thank you, Dr. Burke. I neglected to suggest to all of you that you take a moment to look at the biographies of the speakers in the program book. My introductions from the podium are necessarily short and don't remotely do their careers justice. It is my pleasure to introduce next, Dr. Robert Heaney. Dr. Heaney is the holder of the John A. Creighton University Chair and a Professor of Medicine at Creighton University in Omaha, Nebraska.