Overview, Vol 12, Issue 2

Only doubt is certain and disbelief worth believing.
Without this courage there can be no learning.
Believe nothing.

"The quarterly journal Progress in Osteoporosis began in October 1993 as Advances in Osteoporosis 19 years ago. Its purpose was to provide readers without easy access to the literature with summaries of the most important literature. We now inhabit a virtual world. Information is instantaneously accessible to all at the tap of a keyboard; understanding is not. In the spirit captured by the anonymous author*, the purpose of this publication is to provide critical evaluation of the most important literature and so to provoke discussion. It is our intention to promote dialogue which examines the quality of information published and so its credibility. The opinions expressed are my own and do not necessarily reflect those of the International Osteoporosis Foundation."

We invite readers to comment on and discuss this journal entry at the bottom of the page.

PDF Version

FREE DOWNLOAD of PIO Volume 12-2 Figures

Nonvertebral Fractures: As good as it gets?

We now have several therapies convincingly shown to reduce the risk of vertebral fractures, at least during the first 3 years of treatment (1). Most of these treatments reduce this risk by 50-60% relative to a control group given calcium and vitamin D supplementation. Some, like zoledronic acid and denosumab, reduce it by about 70% which is starting to look respectable (Figure 1 on right). I am not suggesting these treatments are ‘better’ than others. That statement would require evidence based on concurrently conducted comparator trials. None are available or are likely to become available. The sample sizes needed to demonstrate a biologically worthwhile advantage of one treatment over another are prohibitive (2). Nevertheless, treatments should provide a better than 50:50 chance of fracture prevention – is a 50-60% vertebral fracture risk reduction good enough?

This brings us to a real unmet need in this field: nonvertebral antifracture efficacy. While vertebral fractures and trabecular bone loss are ‘flagships’ of osteoporosis, nonvertebral fractures account for 80% of all fractures, 80% of bone is cortical, not trabecular, most bone lost from the appendicular skeleton is cortical, not trabecular and most of this bone loss occurs by intracortical remodeling with the production of intracortical void spaces (porosity) as enlargement and coalescence of Haversian canals progressively cavitates and thins the cortical shell exponentially reducing its resistance to bending (3-5).

The patient consulting you for advice about fracture risk reduction is four times more likely to sustain a nonvertebral than vertebral fracture. Therefore the choice of treatment should be a drug that has been demonstrated to reduce the risk of non-vertebral fractures, not only vertebral fractures. There are not many alternatives available and the expected benefit to the patients is far less than the benefit against vertebral fractures.

A reduction in hip fracture risk is reported for alendronate, risedronate, zoledronic acid, denosumab, and strontium ranelate. In the few studies available examining the benefits of these treatments, the risk reduction is about 40-50%; again, there is only a 50:50 chance that the fracture the patient will have will actually be prevented despite compliance with therapy.

How credible are the data? Well, you judge. There is only limited evidence for anti-hip fracture efficacy in the very group having hip fractures: persons over 70 years of age. Briefly, for alendronate, the reduction in hip fracture risk was based on FIT-1 and this data was based on 22 hip fractures in controls and 11 fractures in the treated group, no patients, or only a few, were older than 80 years of age and most were under 75 years of age. For risedronate, the data were more plentiful; event rates were higher and so the result is more robust. The risk reduction was based on the HIP trial by McClung et al (6) but no reduction in hip fracture risk was detected in persons over 80 years of age selected on risk factors. For zoledronic acid there was a reduction in the HORIZON trial overall but not in those over 75 years of age. For denosumab there was evidence of a reduction in risk in the trial overall, but whether there is evidence in the over 70 year old age group is not known. For strontium ranelate there was a reduction in women over 80 years of age, but this was of borderline statistical significance (7).

If we turn to nonvertebral fractures, the evidence becomes painfully sparse as shown in the above figure: risedronate, zoledronic acid, denosumab and strontium ranelate (1,8). Not only is there little data, the veracity of the data is easily challenged. First, in most, if not all studies, hip fractures are included in these analyses. This is usually unstated, but are the results if hip fractures are excluded from the analyses?

Of the few studies demonstrating any nonvertebral fracture risk reduction, this reduction is about 20%. So, say you have a busy day and you see 100 women over 70 years of age. Of these, 5 will sustain a fracture in that year; quite a high incidence of fractures. The problem is you don't know which of the 100 those 5 will be. Therefore, you have to treatment them all. The problem is, of the 5 sustain nonvertebral fractures, when you treatment all 100, only one will have the fracture averted during treatment, the other four will sustain the fracture despite compliance with therapy. I wonder if informed consent requires giving that sort of information and whether this will be acceptable to patients.

Looking at the quality of the evidence is instructive. There was no evidence for nonvertebral fracture risk reduction with alendronate in the FIT-1 and FIT-2 trials; to achieve statistical significance required a post hoc analysis of the pooled of FIT-1 patients and the patients with osteoporosis in FIT-2 (1). Nonvertebral fracture risk was observed with risedronate in one, but not both multicenter trials. For zoledronic acid, it was reassuring to see a nonvertebral fracture risk reduction in the HORIZON trial and in the post-hip fracture trial, a very difficult trial to execute but a most informative one (1,9). Nonvertebral fracture risk has also been reported in the FREEDOM trial with denosumab.

For the other treatments at the bottom of Figure 1, I am unable to put any data at all. The problems in the design and execution of the trials prevent any real confidence in the data. For example, in the studies of calcium and vitamin D subjects recruited were not deficient in these nutrients, so how can the effect of deficiency on fracture rates be assessed and any potential benefit of intervention be detected? In those studies, almost without exception, compliance was 50% with the intervention. Subanalyses looking at the effects of intervention persons deficient in calcium or vitamin D, or in those who comply with therapy, results in violation of randomization – the single design feature that controls for known and unknown influential covariates. These subanalyses and meta-analyses of subanalyses sometimes manage to squeeze a p<0.05 out of the data, provides enough uncertainty to allow debates at international meetings with a lot of sound and fury signifying nothing.

We are not there yet and the question is why. The answers to this complex question are not available for many reasons but this is a topic for the next issue of Progress in Osteoporosis. Now to a summary of the highlights in therapeutics from the recent IOF–ECCEO12 Congress in Bordeaux, more than a nice place to visit.

Advances in the Therapeutics of Osteoporosis Presented at IOF–ECCEO12

Treatments With a Predominantly Antiresorptive Action


Does denosumab increase bone density during 8 years?

McClung et al reported the 8 years follow-up in postmenopausal women with osteopenia or osteoporosis randomized to placebo, alendronate or denosumab (10). In the extension study, all subjects received open-label denosumab 60 mg Q6M for 4 years. For the 88 subjects who received denosumab for 8 years, BMD at the lumbar spine and total hip increased from baseline by 16.5% and 6.8%, respectively. Reductions in CTX and BSAP were sustained over 8 years of therapy (Figure 2 on right).

This is an important study. The questions are what is the morphological basis underlying the continued increase in BMD and is this beneficial to bone strength or detrimental. The rise in BMD at spine in the second 4 years is ~8%, similar to the rise in the first 4 years. There are three possible explanations for this.

First, this is partly the result of secondary mineralization; the completion of the formation phase of remodeling cycles by secondary mineralization of osteons not removed because remodeling is suppressed. Secondary mineralization is the enlargement of crystals within collagen fibrils with the displacement of the water within them, so the fibril doesn't enlarge but more and more of its volume is occupied by mineral. Secondary mineralization is part of the remodeling ‘transient’; the reversible deficit in matrix and its mineral content that results from the normal delay between completion of excavation of a resorption cavity and its refilling with osteoid which undergoes primary mineralization within a week then slower secondary mineralization (11). Secondary mineralization may take a year but some studies suggest much longer. The duration of completion of secondary mineralization is an area of controversy. Opponents of this explanation hold the view that this cannot be the explanation for the continued rise in BMD because the rise should become asymptotic; as more and more of the bone is fully mineralized there should be flattening of the rise in BMD which should cease to occur after 1-2 years.

This is indeed what appears to be the case with the rise in BMD at the proximal femur. In this graph, the rise in BMD in the second 4 years is ~1%, i.e., approximately asymptotic. This explanation appears most reasonable in my opinion. Why then is there a continued rise in BMD at the spine? An obvious explanation is that this is an artifact resulting from arthritic changes in the intervertebral disc region and facet joints. The third explanation is that there is new bone formation. This is discussed below.

Does denosumab have effects mediated by endogenous PTH?

Denosumab rapidly and markedly reduces bone resorption at the tissue and cellular levels – i.e., the number of remodeling sites initiated upon the internal surfaces of bone decreases because osteoclastogenesis is inhibited and the resorptive activity of osteoclasts present in excavating resoption pits at the time of starting treatment is also prevented. This rapid suppression of resorption is accompanied by a small transitory fall in serum calcium within the normal range and a rise in endogenous PTH secretion (12).

Seeman et al  tested the hypothesis that, in the face of suppressed remodeling, the transitory increase in endogenous PTH will stimulate the activity and lifespan of osteoblasts in existing remodeling cavities, and so will increase the volume of bone deposited in these resorption cavities more than otherwise would be produced, and that the cavity created will be smaller (because osteoclasts have been inhibited from completing resorption) (13). Together, the smaller resorption cavities and the larger volume of bone deposited within them may reduce or abolish the negative bone balance, and perhaps even induce a positive balance resulting in a reduction in porosity or even some bone building effect.

Postmenopausal women with a mean age of 61 years were randomly assigned in a double-blind, double-dummy trial to denosumab 60 mg Q6M (N=83), alendronate 70 mg QW (N=82), or placebo (N=82). PTH was measured and an area under the curve (AUC) for PTH was derived. With placebo and alendronate, porosity increased with increasing PTH. With denosumab, porosity decreased as PTH increased. (Figure 3 on right)

The authors infer that denosumab partially reversed microarchitectural deterioration (i) directly by reducing remodeling intensity and perhaps (ii) indirectly, by a PTH-dependent effect on BMU level bone formation in the setting of full suppression of osteoclast activity. This work is hypothesis generating. The data are consistent with a possible independent effect of endogenous PTH but histomorphometric studies will be needed to determine whether mean wall thickness is increased and mineral appositional rate is increased.

If there is a small anabolic effect of endogenous PTH, could this explain the rise in BMD seen during prolonged therapy? We don’t know. Whatever incremental increase in bone formation there may be, this benefit should be the greatest following the first dose of denosumab when remodeling sites present prior the start of treatment are most plentiful providing a bountiful garden of remodeling sites packed with osteoblasts ready to be stimulated by endogenous PTH. Subsequently, although remodeling begins again in the month prior the need for the next treatment, the numbers of BMUs generated for the next rise in endogenous PTH following the second and subsequent injections is likely to be about half the number of BMUs prior treatment because remodeling does not return to its pretreatment level.

Is there a residual fracture risk reduction after stopping denosumab?

Roux et al reported that fracture risk during 2 years after stopping denosumab remained below that of the controls (14). In 470 placebo and 327 denosumab treated subjects from the FREEDOM trial who discontinued treatment after 2-5 doses, the authors report that fracture rates were below the previously placebo treated subjects. After treatment discontinuation, similar percentages of subjects in both groups sustained a new fracture (9% placebo, 7% denosumab; fracture rate/100 subject-years 13.5 and 9.7, respectively; HR 0.82; 95% CI 0.49, 1.38). (Figure 4 on right)

The question being addressed is important because denosumab has a rapid offset of action. Remodeling markers rise rapidly and may over shoot, so the question is whether this recurrence of remodeling creates stress concentrators – resorption cavities that concentrate stress, like cutting a test tube to make it easier to snap.

The data is difficult to interpret because initiation of therapy, usually bisphosphonates, occurred in both groups after stopping denosumab; in 42% placebo vs. 28% denosumab treated subjects. As there were more placebo treated subjects receiving a new therapy, the nonsignificant lower risk of fracture in the denosumab group may be underestimated; had the control group not been treated, their fracture rate would have been higher. There is no easy solution to this dilemma; analysis excluding patients treated with an alternative agent should be done but this will reduce the sample size, so there will be little power to detect any true residual benefit that remains after stopping denosumab, or, indeed, any increased risk over the control group that might be associated with the overshoot in remodeling after stopping denosumab if one truly exists.

Is long-term denosumab efficacious and safe?

Papapoulos et al reported the results of 6 years of denosumab exposure (15). Women from the FREEDOM placebo group received 3 years of denosumab and women from the denosumab group received 3 more years of denosumab. 4550 (77%) enrolled (N=2207 crossover; N=2343 long-term). In the long-term group, further increases in BMD occurred producing 6-year gains of 15.2% at the lumbar spine and 7.5% at the total hip. In the crossover group, yearly incidences of nonvertebral, new vertebral, clinical vertebral, and clinical fractures were lower than those in the FREEDOM placebo group. Fracture incidence remained low in the long-term group. Incidences of adverse events did not increase. (Figure 5 on right)

The incidence of fractures was low in the latter years, but the question is whether this is attributable to the treatment. There was no control group. It therefore remains possible that there was loss of high risk individuals and the remaining cohort were healthier individuals who would not have had a fracture without treatment. Note, in the first three years of treatment, the incidence in controls decreased year by year, so this potential bias introduced by sampling cannot be ignored. It is unethical to withhold treatment from individuals at high risk for fracture in clinical trials and so continuing a placebo arm cannot be justified. Thus, the challenge of whether antifracture efficacy is maintained after 3 years is not easily addressed.


Does alendronate modify primary mineralization?

Prolonged suppression of bone remodeling may modify the material composition of bone because osteons that normally would be remodeled and replaced with younger bone are not. They undergo more complete secondary mineralization to become homogeneously mineralized and this reduces the resistance to crack propagation (16). Other changes in material properties may occur and to examine these, Bala et al evaluated 150 osteons from iliac cortical bone structural units (BSU) in 6 postmenopausal osteoporotic women treated for ~8 years with alendronate and 5 age-matched controls (17). Cases had a 12% lower elasticity (E) and 6% lower contact hardness (Hc) and higher collagen maturity. Crystallinity index, which is inversely proportional to crystal size/perfection, was higher in alendronate than in controls (25.29±0.76 vs. 24.78±0.70), and inversely correlated with E and Hc (r=‒0.43 and r=‒0.54, respectively). Collagen maturity correlated with E and Hc in the two groups (r ranged from 0.40-0.70, all p<0.001). Treated bone was also less able to plastically resist deformation at constant strain. Alendronate may alter the mineral crystallinity and impair the mechanical behavior at the BSU level. These are small changes, how they affect whole bone strength remains to be determined; but bilateral or unilateral spontaneous proximal femur fractures occurring in association with prolonged antiresorptive therapy is well documented and cannot be ignored as a potentially causal relationship. (Figure 6 on right)

Zoledronic Acid

Does zoledronic acid reduce fracture risk in men?

This is a nicely designed and executed study. It is one of the few, if not the only study convincingly showing a fracture benefit of treatment in men. Boonen et al report a randomized, controlled study in men with osteoporosis. Zoledronic acid reduced vertebral fracture risk by 67% over 24 months (Figure 7 on right). In the 1199-patient double-blind trial, men with osteoporosis aged 50-85 years were randomly assigned to 5 mg (n=588) or placebo (PBO; n=611) infusion at baseline and 12 months (18). Serum testosterone was available in 96% of men. Of these, 146 (26%) zoledronic acid treated and 181 (31%) placebo treated men had total testosterone ≤350 ng/dL. Zoledronic acid reduced the risk of morphometric vertebral fracture by 62% in men with TT >350 ng/dL (p<0.03), and by 72% in those with TT ≤350 ng/dL (p<0.08). The effects on nonvertebral fractures was not reported, presumably this was not significant.

Evidence for sustained fracture risk reduction

Eastell et al report that in the extension of the HORIZON trial, 1233 women who received 3 infusions had 3 additional infusions of zoledronic acid (Z6, n=616) or 3 placebo infusions (Z3P3, n=617) (19). Predictors of new morphometric fracture were a femoral neck and total hip T-score of ≤‒2.5 SD and an incident morphometric vertebral fracture during the core trial which were each variously associated with an OR of 3-5 for subsequent morphometric vertebral fracture. For new nonvertebral fracture occurrence, predictors were incident nonvertebral fracture during Core [HR=2.5(1.2,5.3)] and prevalent vertebral fracture [HR=3.0(1.4,6.3)] (Figures 8, 9 on right). The authors suggest that in women with hip BMD ≥‒2.5 SD, the risk for fracture is low and treatment discontinuation can be considered. In women with hip BMD T-score<‒2.5 SD, continued treatment was interpreted to confer benefit against vertebral fracture.


Cathepsin K Inhibitors

Does odanacatib continue to increase bone density during 5 years therapy?

Odanacatib is a cathepsin K inhibitor that is being assessed in an antifracture efficacy trial at this time and the results are awaited with great anticipation; new treatments are needed in the field as implied in the introduction to this issue of Progress in Osteoporosis. The drug reduces bone resorption by osteoclasts producing more shallow resorption cavities (20). This is an interesting observation for many reasons. If the depth of resorption of each pit is smaller and the resorption pit is refilled with the same or more osteoid, then treatment may reduce the negative bone balance. This is important. The negative bone balance is the necessary and sufficient morphological basis of bone loss in osteoporosis; it is the cause of structural decay and a critical target for its prevention (Figure 10 on right). If the balance is shifted to be positive, i.e., more bone is deposited in a smaller cavity reconstruction of bone may follow – in this case, it is desirable for bone remodeling intensity to continue to be high, as each remodeling event will deposit a small moiety of bone.

Resch et al report that women entering the year 4-5 extension, after 5 years, mean BMD changes for women who received odanacatib 50 mg continuously were: spine 11.9%, femoral neck 9.8%, trochanter 10.9%, total hip 8.5%, and 1/3 radius -1.0% (21). For women who switched from 50 mg to placebo after 2 years, changes were: lumbar spine -0.4%, femoral neck -1.6%, trochanter -1.0%, total hip -1.8%, and 1/3 radius -4.7%. After 5 years, for women continuously receiving 50 mg, mean changes from baseline in remodeling markers were: -67.4% for urine NTX/creatinine and -15.3% for serum BSAP. For women who switched from 50 mg to placebo after 2 years these changes were 6.0% and -11.9%, respectively. (Figure 11 on right)

The question is what is the morphological basis for the continued rise in BMD? Could this be an anabolic effect – i.e., deposition of new bone (not just refilling of the remodeling space transient). Does this treatment result in deposition of osteoid, producing by a positive balance by each BMU? Another possibility is periosteal apposition. Studies in monkeys suggest periosteal apposition occurs with this agent (22). If so, is it sufficient to alter bone morphology – i.e., increase total bone CSA, and so increase resistance to bending?

Yet another possibility that must not be dismissed is that this is a rise in BMD, not an increase in bone mass at all. Noninvasive imaging methods such as DXA and CT scanning do not measure mass, they measure the attenuation of photons produced by mineral. If the same bone mass or bone volume becomes more fully mineralized due to progression of secondary mineralization, the photon attenuation will increase and this is often spoken of as an increase in bone ‘mass’. Neither the mass nor volume of bone has increased – this is a trap for young investigators unaware of the vagaries in language and the abuse of these vagaries that result.

Secondary mineralization is likely to contribute to the rise in BMD because remodeling is suppressed by this agent. A contribution by new bone formation has not been excluded and the studies in monkeys reporting periosteal apposition are of great interest. If remodeling intensity is not reduced as much as it is with the bisphosphonates, then each remodeling event will remove older more mineralized bone, so the extent of secondary mineralization should be less with this class of drug than with bisphosphonates. On the other hand, if remodeling continues and the negative bone balance persists, then each remodeling event will remove bone, even though more slowly than before treatment, and so structural decay will continue but slowly.

If resorption depth is reduced, then the size of the hemiosteon upon a trabecular surface or endocortical surface or the diameter of an osteon within cortical bone will be less; i.e., there will be more interstitial bone between the osteons – unless the continued remodeling at a lower rate maintains osteonal numbers. The osteons will be smaller but there will be more of them if remodeling intensity is not slowed so the net effect is the proportion of the cortical bone that is osteonal remains unchanged. The relevance of this is in microdamage accumulation which occurs more commonly in interstitial bone (bone between osteons). If remodeling is slowed and the osteons are small, then the interstitial bone increases in absolute and relative terms. This may have adverse effects on the material composition of bone.

Does odanacatib modify bone structure?

Odanacatib has been reported to increase periosteal bone formation and cortical thickness in nonhuman primates (22). Brixen et al used QCT to examine the effects of odanacatib on trabecular and cortical bone in a randomized, double-blind, 2-year trial of 214 postmenopausal women with low aBMD who received odanacatib 50 mg or placebo weekly (23). Compared with the placebo-treated women, odanacatib treated women had greater increases in integral and trabecular spine vBMD and compressive strength (estimated using FEA), and integral and hip trabecular vBMD and sideways-fall strength at the hip. Femoral neck cortical thickness increased with odanacatib but declined with placebo. Serum CTX was lower in the odanacatib group than placebo (-43% vs. 3%) but serum P1NP did not differ (ODN -11%, placebo -2%). (Figure 12 on right)

The authors suggest cortical thickness increased, but for this to occur, periosteal apposition and/or endocortical apposition must be documented. There was no evidence provided for either in this study. The way cortical thickness is calculated should be considered. This is a derived value obtained by dividing cortical area by perimeter. There is no such thing as a single cortical ‘thickness’, the thicknesses of the cortex vary at each point around a perimeter of a tubular bone and at each cross-section along its length. The information needed here is what were the periosteal and endocortical circumferences, and cortical area. In addition, what does an increase in cortical ‘density’ mean in morphological terms – did cortical porosity decrease, and/or did tissue mineralization density increase? The increase in cortical density by either mechanism may alter edge detection producing a seeming increase in cortical thickness.

Does ONO-5334 modify bone structure?

This is another well investigated cathepsin K inhibitor that shows promise. Engelke et al randomized postmenopausal women with osteoporosis or osteopenia (with a vertebral fracture) to ONO-5334 (50 mg b.d, 100 or 300 mg qd), placebo or alendronate 70 mg qw double-blind study using QCT (24). Of about 120 women with follow-up scans at 2 years, in the spine, all ONO-5334 doses showed similar changes in trabecular BMD but cortical changes favored 300 mg qd. In the femur, ONO-5334 300 mg qd produced higher BMD increase than other doses, particularly for trabecular BMD. Compared to alendronate, ONO-5334 50 mg bd and 300 mg qd appeared to show equivalent increases in integral and cortical BMD and superior increases in trabecular BMD.


Does oral recombinant calcitonin warrant a revisit?

Binkley et al reported a randomized, double-blind, double-dummy, active- and placebo-controlled, multiple-dose, phase III study to assess the efficacy and safety of oral recombinant calcitonin in 565 postmenopausal women (25). Patients were randomized (4:3:2) to oral recombinant salmon calcitonin (rsCT) (0.2 mg/day), synthetic salmon calcitonin (ssCT) nasal spray (200 IU/day) or placebo for 48 weeks. rsCT increased spine BMD (1.5±3.2%); greater than ssCT nasal spray (0.78±2.9%) or placebo (0.5±3.2%) (Figure 13 on right). Changes in spine BMD in those receiving nasal calcitonin did not differ from placebo. Oral rsCT also resulted in greater improvements in trochanteric and total proximal femur BMD than ssCT nasal spray. Reductions in resorption markers with oral rsCT were greater than those observed in ssCT nasal spray or placebo. Gastrointestinal adverse events were reported by nearly half of women and were the principle reason for premature withdrawals. Oral rsCT was superior to nasal ssCT and placebo for increasing BMD and reducing bone turnover. Oral rsCT was safe and as well tolerated as ssCT nasal spray or placebo. These modest changes are hard to interpret and given the null results of the PROOF trial (except for one of the arms), without evidence of benefits in structure and strength, what is the lesson?

Agents With Other Modes of Action

Strontium Ranelate

Does strontium ranelate increase bone mass in women and in men?

The operative word is ‘mass’. Felsenberg et al report that in 189 women randomized to strontium ranelate (SR) (2 g/day) or alendronate (70 mg/week) during 2 years (26), ultradistal tibia total bone mineral content (BMC) increased by 3.3% and trabecular BMC increased by 2.3% in SR group and by 1.7% and 1.3%, respectively, in ALN group. The moment of inertia (MI) and density-weighted MI increased by 1.2±1.6% and 1.7±2.1%, respectively, in the SR group and by 0.5±1.8% and 0.9±2.4%, respectively, in the ALN group. Mean increases of 0.7±1.8% for the section modulus and 1.3±2.4% for strength strain index (SSI) were found in the SR group, no change was observed in the ALN group. Between-group difference favored SR for each trait. The authors infer greater effects on bone mass and strength parameters at the tibia compared to ALN in women with postmenopausal osteoporosis after 2-year treatment.

These data need to be interpreted cautiously. The word bone ‘mass’ is abused in this field. Osteoid is deposited by osteoblasts and when it is mineralized it is ‘bone’ or ‘bone mass’. When strontium is deposited in bone by substitution for a calcium atom, the attenuation of photons is increased, so there is an increase in what we refer to as the apparent BMD. The same occurs with antiresorptive agents. When these are administered, remodeling intensity decreases and osteons that would have been removed are not. They undergo more complete secondary mineralization – the bone mass or volume do not increase, the mass or volume is more fully mineralized so the apparent density increases. Neither agent makes new bone – there is no evidence that these agents are anabolic. The estimates of bone strength are not direct measurements of bone strength – peak tolerated loads, resistance to bending or torsion, they are mathematically derived estimates that use the apparent density in the formulae.

Does strontium ranelate increase bone mineral density in men?

Yes. Kaufman et al reported the results of a 2-year randomized double-blind placebo-controlled trial (SR 2 g/day/placebo 2:1) in 261 men (27). The ITT population consisted of 243 men, age 72.7±5.7 years with lumbar and femoral neck BMD T-Score of -2.7±1.0 and -2.3±0.7, respectively; 29% of patients had prevalent vertebral fractures. BMD increased in the SR group: lumbar (L2-L4) by 9.8±1.1%; femoral neck by 3.3±0.9% and total hip by 3.7±0.8% (all p<0.001). An improvement in the quality of life was observed (-0.34±0.7 in the SR group vs. -0.07±0.5 in the placebo group (p=0.009). Vertebral fracture incidence was lower in the SR than in the placebo group but not significantly so (5.8% vs. 7.8%). The same was observed for clinical nonvertebral fractures recorded as adverse events (3.5% vs. 4.6%).

Does strontium ranelate protect against fatigue damage?

Strontium ranelate has been demonstrated to reduce the risk of vertebral and non-vertebral fractures in well designed and executed trials. The question is how. Ammann and Rizzoli report SR influences bone microarchitecture and intrinsic bone tissue properties which independently improve estimates of bone strength. The authors suggest that the changes may prevent the formation of microcracks and/or their propagation (28). Vertebrae of intact female rats treated over 8 weeks with SR at 625 mg/kg or with a vehicle were cyclically loaded in axial compression for 100 cycles. The selected peak load corresponded to 5% of the adjacent vertebra maximal load (domain of elastic deformation). The vertebrae were then loaded to failure. Maximal load was 267±19 and 233±20 N in unloaded SR and control groups, respectively. Cyclic loading induced a deterioration of post yield load in control rats (19.81±3.38 vs. 11.80±2.03 N in unloaded vs. fatigue control groups, respectively, p<0.05). This was prevented in SR treated rats (18.42±4.00 vs. 18.78±3.71 N in unloaded vs. fatigue SR groups, respectively). The post yield deflection was unaffected in either group. This suggests less damage accumulation under fatigue loading.

Is strontium ranelate safe?

Clinical trials are designed to assess efficacy, not safety and so post marketing data is needed to evaluate safety. Jakob et al report the results of an observational cohort study to assess safety and treatment persistence with SR during 3 years follow-up in 12,702 postmenopausal women from 7 countries (29). Mean age was 69.0 years with 16.5% of patients being over 80 years. Mean follow-up duration was 32 months and mean treatment duration was 25.2 months (24,956 patient-years of treatment). VTE was reported in 55 patients, an incidence of 2.1/1000 patient-years (95% CI 1.6, 2.8), lower than that observed in patients treated with SR in the phase III studies (7.9/1000 patient-years; 95% CI 6.3, 9.7). No DRESS syndrome or Stevens-Johnson syndrome was reported. Persistence of SR treatment was 80%, 68% and 64% after 12, 24 and 32 months treatment, respectively.

Anabolic Agents

Does antisclerostin antibody reduce fractures in rodent models of osteogenesis imperfecta (OI)?

Yes. Devogelaer et al report that Scl-Ab improved biomechanical properties and reduced fracture rates in OI mice. Seven-week-old OI and control (WT) mice received PBS or Scl-Ab (25 mg/kg twice weekly for 10 weeks (30). Scl-Ab reduced the number of fractures by 56% (2.8±0.6 vs. 6.3±1.5 in controls; p<0.001). In the tibia, ultimate strength increased (midshaft: +30%, proximal: +98% vs. controls), stiffness increased (midshaft: 132%; proximal: 88% vs. controls) and plastic energy increased (midshaft: 125%, proximal: 260% vs. controls). These strength increases were associated with increases in tibia BMD (midshaft: +30%, proximal: +50% vs. PBS) and cortical thickness (midshaft: +40%, proximal: +75% vs. PBS). Scl-Ab therapy also increased BMD and cortical thickness in the humerus and lumbar vertebra so that at the end of therapy, the strength, BMD and cortical thickness of bones of the OI skeleton were similar to WT. Scl-Ab therapy also enhanced the strength, BMD and cortical thickness in tibia, humerus and vertebra of WT normal mice.

Devogelaer et al also assessed the effects of Scl-Ab on fracture rates in axial skeleton of 5-7-week-old OI and WT mice (31). Scl-Ab reduced pelvic fractures by 65% (0.4±0.5 per mouse vs. 1.1±0.8 in PBS; p<0.02) and improved LVB trabecular bone parameters including: BMD (55%), BV/TV (111%), TbTh (40%), TbN (48%) and TbPf (-43%). In WT, Scl-Ab was also associated with improvements in BV/TV (160%), TbTh (33%), TbN (96%) and TbPf (-94%). Scl-Ab therapy decreased SMI in WT mice, but not in OI mice.

Does PTH(1-84) accelerate pelvic fracture healing?

Holzer et al report that in 65 patients with pelvic fractures, 21 received once daily 100 μg PTH(1-84) within two days of admission, 44 patients without PTH treatment served as a control group (32). Functional outcome was assessed using a pain Visual Analogue Scale (VAS) and a Timed Up and Go (TUG) test. In all 21 patients treated with PTH(1-84) pelvic fractures healed at a mean of 7.8 weeks, whereas in patients with no PTH treatment fractures healed after 12.6 weeks. At week 8 all fractures in the treatment group were healed and four fractures in the control group (healing rate 100% vs. 9.1%; (p<0.001). Both the VAS and TUG improved (p<0.001) compared to control. PTH(1-84) accelerates fracture healing in pelvic fractures and improves functional outcome.

Is an analog of human PTHrP anabolic?

Hattersley et al report that BA058, a synthetic analog of hPTHrP(1-34), was assessed in randomized, double-blind, placebo-controlled phase II study of postmenopausal women with osteoporosis randomized to placebo, BA058 20, 40, 80 μg or teriparatide 20 μg for 24 weeks (33). 184 patients completed 6 months treatment. The spine BMD was 1.6% with placebo, 2.9%, 5.2%, and 6.7% with BA058 20, 40 and 80 μg, respectively, and 5.5% with teriparatide. The difference from placebo was significant for BA058 40 and 80 μg and for teriparatide. Further increases in spine BMD were seen during the extension phase (n=55), with a mean percent change at 48 weeks of 0.7% with placebo, 5.1%, 9.8%, 12.9% with BA058 20, 40 and 80 μg, respectively, and 8.6% with teriparatide. A dose dependent increase in total hip BMD was seen at 24 weeks; the mean change was 0.4% with placebo, 1.4%, 2.0%, and 2.6% with BA058 20, 40, and 80 μg, respectively, and 0.5% with teriparatide. At 24 weeks the change in serum and urine markers was significant from baseline for BA058 40 and 80 μg and for teriparatide for P1NP, BSAP, osteocalcin and CTX, and with teriparatide for NTX. BA058 was well tolerated. The proportion of patients with elevated calcium levels was lower with BA058 than with teriparatide. BA058 80 μg resulted in significant BMD gains.


1.  Delmas PD. Treatment of postmenopausal Osteoporosis. Lancet 2002;3589:2018.

2.  Kanis JA, Alexandre JM, Bone HG, et al. Study design in osteoporosis: a European perspective. J Bone Miner Res 2003;18:1133.

3.  Kanis JA, Borgstrom F, Zethraeus N, et al. Intervention thresholds for osteoporosis in the UK. Bone 2005;36:22.

4.  Zebaze RM, Ghasem-Zadeh A, Bohte A, et al. Intracortical remodelling and porosity in the distal radius and post-mortem femurs of women: a cross-sectional study. Lancet 2010;375:1729.

5.  Schaffler MB and Burr DB. Stiffness of compact bone: effects of porosity and density. Biomechanics 1988;21:13.

6.  McClung MR, Geusens P, Miller PD, et al. Hip Intervention Program Study Group. Effect of risedronate on the risk of hip fracture in elderly women. N Engl J Med 2001;344:333.

7.  Seeman E, Vellas B, Benhamou C, et al. Strontium ranelate reduces the risk of vertebral and nonvertebral fractures in women eighty years of age and older. J Bone Miner Res 2006;21:1113.

8.  Chen JS, Sambrook PN. Antiresorptive therapies for osteoporosis: a clinical overview. Nat Rev Endocrinol 2012;8:81.

9.  Lyles KW, Colón-Emeric CS, Magaziner JS, et al. Zoledronic acid and clinical fractures and mortality after hip fracture. N Engl J Med 2007;357:1799.

10.  McClung MR, Lewiecki EM, Bolognese MA, et al. Denosumab continues to affect bone mineral density and biochemical markers of bone turnover for up to 8 years : results of a phase 2 extension trial. Osteoporos Int 2012;23(Suppl 2):S74.

11.  Parfitt AM. Morphologic basis of bone mineral measurements: transient and steady-state effects of treatment in osteoporosis. Miner Electrolyte Metab 1980;4:273.

12.  McClung MR, Lewiecki EM, Cohen SB, et al. Denosumab in postmenopausal women with low bone mineral density. N Engl J Med 2006;354:821.

13.  Seeman S, Libanati C, Austin M, et al. The transitory PTH increase following denosumab administration is associated with reduced intracortical porosity: a distinctive characteristic of denosumab therapy. Osteoporos Int 2012;23(Suppl 2):S76.

14.  Roux C, Brown JP, Beck-Jensen JE, et al. Denosumab discontinuation and associated fracture risk : a FREEDOM trial analysis. Osteoporos Int 2012;23(Suppl 2):S105.

15.  Papapoulos SE, Brown JP, Chapurlat R, et al. Denosumab treatment of postmenopausal women with osteoporosis for 6 years: results from the first 3 yearas of the FREEDOM extension. Osteoporos Int 2012;23(Suppl 2):S76.

16.  Tai K, Dao M, Suresh S, Palazoglu A, Ortiz C. Nanoscale heterogeneity promotes energy dissipation in bone. Nat Mater 2007;6:454.

17.  Bala Y, Depalle B, Douillard T, et al. Bone tissue micromechanical properties are altered independently of mineralization in osteoporotic women long-term treated alendronate. Osteoporos Int 2012;23(Suppl 2):S110.

18.  Boonen S, Reginster JY, Kaufman JM, et al. Efficacy of one-yearly zoledronic acid 5 mg in men with osteoporosis with different levels of serm total testosterone. Osteoporos Int 2012;23(Suppl 2):S79.

19.  Eastell R, Cosman F, Cauley JA, et al. After 3 years of annual zoledronic acid, who should remain on treatment? Results from the HORIZON-PFT extension. Osteoporos Int 2012;23(Suppl 2):S240.

20.  Costa AG, Cusano NE, Silva BC, Cremers S, Bilezikian JP. Cathepsin K: its skeletal actions and role as a therapeutic target in osteoporosis. Nat Rev Rheumatol 2011;7:447.

21.  Resch H, Binkley N, Bone H, et al. Treatment with the cathepsin K inhibitor odanacatib in postmenopausal women with low BMD: 5 year results of a phase 2 trial. Osteoporos Int 2012;23(Suppl 2):S61.

22.  Cusick T, Chen CM, Pennypacker BL, et al. Odanacatib treatment increases hip bone mass and cortical thickness by preserving endocortical bone formation and stimulating periosteal bone formation in the ovariectomized adult rhesus monkey. J Bone Miner Res 2012;27:524.

23.  Brixen K, Chapurlat R, Cheung AM, et al. Imaging and finite element analysis of the spine and hip following 2 years of treatment with odanacatib in postmenopausal women. Osteoporos Int 2012;23(Suppl 2):S62.

24.  Engelke K, Nagasse S, Fuerst T, et al. Effects of the cathepsin K inhibitor, ONO-5334, on BMD as measured by 3D QCT in the hip and the spine after 2 years of treatment. Osteoporos Int 2012;23(Suppl 2):S62.

25.  Binkley N, Bolognese M, Sidrowicz-Bialynicka A, et al. A phase 3 trial of the efficacy and safety of oral recombinant calcitonin: the ORALCAL trial. Osteoporos Int 2012;23(Suppl 2):S78.

26.  Felsenberg D, Beller G, Fiore C, et al. Beneficial effects of strontium ranelate compared to alendronate on bone mass and strength parameters at the tibia in postmenopausal osteoporotic women: a 2-year study. Osteoporos Int 2012;23(Suppl 2):S113.

27.  Kaufman JM, Audran M, Bianchi G, et al. Efficacy and safety of strontium ranelate in the treatment of male osteoporosis. Osteoporos Int 2012;23(Suppl 2):S260.

28.  Ammann P, Rizzoli R. Strontium ranelate fully prevents alteration of bone mechanical properties in response to cyclical loading. Osteoporos Int 2012;23(Suppl 2):S73.

29.  Jakob FJ, Palacios S, Audran M, et al. Final results of a 3 years prospective observational cohort study of patients treated with strontium ranelate. Osteoporos Int 2012;23(Suppl 2):S73.

30.  Devogelaer JP, Behets C, Ammann P, et al. Sclerostin antibody increased bone mass and reduced fractures in osteogenesis imperfecta mice. Osteoporos Int 2012;23(Suppl 2):S74.

31.  Devogelaer JP, Chappard D, Behets C, et al. Sclerostin neutralizing monoclonal antibody decreases significantly the number of pelvic fractures and improves markedly the size and interconnectivity of lumbar trabecular bone in oim/oim mice. Osteoporos Int 2012;23(Suppl 2):S209.

32.  Holzer G, Peichl P, Holzer LA, Maier R. Parathyroid hormone 1-84 accelerates fracture healing in public bones of elderly osteoporotic women. Osteoporos Int 2012;23(Suppl 2):S77.

33.  Hattersley G, Bilezikian JP, Guerriero J, et al. BA0258, a novel analog of human parathyroid hormone-related protein, exerts significant effects on BMD in postmenopausal women with osteoporosis. Osteoporos Int 2012;23(Suppl 2):S77.


dear sir, it article is immense informative to the health professionals.