Best Practice & Research Clinical Rheumatology Vol. 15, No. 3, pp. 483±496, 2001
doi:10.1053/berh.2001.0162, available online at http://www.idealibrary.com on
10 Emerging therapies in osteoporosis Brian R. MacDonald*
Director Musculoskeletal Clinical Research, North American Medical Aairs, GlaxoSmithKline Pharmaceuticals, 1250 S. Collegeville Road, PO Box 5089, Collegeville, PA 19426-0989, USA
Vice President Musculoskeletal Diseases, GlaxoSmithKline Pharmaceuticals Drug Discovery, 709 Swedeland Road, PO Box 1539, King of Prussia, PA 19406-0939, USA
The approval of alendronate in 1994 marked a watershed in the treatment of osteoporosis. Before that time there was no therapy for which unequivocal proof of ecacy existed. Since then several more agents, all from the anti-resorptive class, have also been approved for use in the treatment of this disease and the range of indications for alendronate has been extended to include the prevention of osteoporosis in women with lesser degrees of bone loss, the treatment of glucocorticoid-induced osteoporosis and, most recently, the treatment of male osteoporosis. Despite this there are still several areas of unmet medical need in this disease, including the availability of well tolerated and convenient therapies and treatments that will go beyond the levels of ecacy oered by current therapies. An intense eort is now being directed towards meeting these unmet needs with the improvement of existing therapies and the development of novel agents that will provide superior long term bene®t. Important and exciting drug targets are yielding novel compounds with anti-resorptive activity or anabolic eects to complement current anti-resorptives. Despite this eort considerable obstacles to the successful development of these compounds remain, not least the stringent safety requirements needed to provide an acceptable risk-to-bene®t pro®le and the increasing diculties of conducting placebo controlled studies in patients at high risk of fracture. Key words: osteoporosis treatment; anti-resorptive; anabolic; PTH.
INTRODUCTION Osteoporosis is an increasingly important public health issue that often has catastrophic results at the level of the individual. It is estimated that up to 40% of women will experience an osteoporotic fracture during their lifetime and many of these will be hip fractures.1 The consequences of hip fractures are obvious and well documented. Less well recognized are the debilitating consequences of multiple vertebral fractures including chronic pain, diminished respiratory capacity and *All correspondence to: Brian R. MacDonald. Tel: 610-917-5855; Fax: 610-917-4100; E-mail: [email protected]
c 2001 Harcourt Publishers Ltd. *
484 B. R. MacDonald and M. Gowen
decreased mobility. It is clear that a range of eective, well tolerated therapies for this disease will have important health, economic and clinical bene®ts.2 As humans age, bone loss occurs progressively from a peak level reached during the third decade. This bone loss results in a progressive decrease in bone strength and thus an increasing risk of fracture. Eventually the loss of strength becomes so profound that fractures may result from minimal external force, the classic presentation of an osteoporotic fracture. Epidemiological data indicate that the risk of osteoporotic fracture in women begins to increase at a signi®cantly earlier age than in men although the latter are at considerable risk in more elderly populations.3 The primary reasons for this gender dierence are the lower peak bone mass achieved by women and the accelerated bone loss that occurs in the post-menopausal period. Bone loss is caused by imbalances in the cellular activities that control bone remodelling. This fundamental concept is central to the development of both established and novel therapies for eective treatment of this disease. Bone remodelling, the process by which bone structure is maintained and adapted to the mechanical environment, is the product of a well ordered sequence of cellular events, as described in Figure 1. The initial event of the sequence is the activation of osteoclast formation and activity which, after a reversal period, is followed by a phase of osteoblast recruitment and bone formation. When the amount of bone resorbed is not matched by the amount of subsequent bone formation (as is the case from the time of peak bone mass onwards) the result is a net bone loss at each remodelling site. Over a period of years successive remodelling cycles in the same anatomical site leads to trabecular thinning and eventually to perforation and decreased cortical thickness. These structural eects translate into a reduction in bone strength. An increased activation frequency of remodelling cycles accelerates this process even though the dierence between the rates of formation and resorption is not markedly aected (i.e. both are raised). Resorption
Formation Lining cells
Figure 1. Schematic diagram of the bone remodelling cycle.
Emerging therapies in osteoporosis 485
The scienti®c basis for the development of most current anti-osteoporotic therapies is the modulation of the activities of cells within the bone remodelling cycle in order to correct the imbalance described earlier and thus eliminate or reverse the gradual loss of bone. The most important class of compounds that aect bone remodelling are anti-resorptives. This diverse group acts by reducing the formation and/or the activity of osteoclasts, which will reduce the amount of bone resorbed and also the frequency with which one remodelling cycle follows its predecessor in the same anatomical location. However, other cellular therapeutic targets may be identi®ed in the future. These could include bone lining cells on areas of quiescent bone, which may be stimulated to an osteoblast phenotype and osteocyte function which appears to play a role in transducing changes in the mechanical environment into biological activity. THE RECENT HISTORY OF OSTEOPOROSIS TREATMENT The introduction of eective therapies The approval of alendronate for the treatment of osteoporosis was a key event in the medical management of this disease. This was the ®rst agent to be approved on the basis of a clinical development plan that was in accordance with the USA Food and Drug Administration (FDA) guidelines, which called for a positive eect on fracture incidence over 3 years of therapy. This is an important benchmark and the alendronate database has since been extended by the landmark Fracture Intervention Study to provide high quality data on fracture outcomes.4,5 Since then several other anti-resorptive agents have met similar standards including raloxifene6 and, most recently, risedronate.7 These compounds all display the features expected of an anti-resorptive drug: reduction in the rate of bone remodelling at a biochemical and histological level accompanied by an increase in bone mineral density (BMD) as well as a reduction in fracture rates. Interestingly the magnitude of change in bone remodelling and BMD is not directly re¯ected in the degree of reduction in vertebral fracture rates since raloxifene and alendronate both reduce fracture rates by approximately 50%, despite a greater increase in BMD with alendronate than with raloxifene. However, a meta-analysis of published studies indicates that, in general, larger changes in BMD are associated with larger decreases in fracture incidence.8 The real bene®t of more powerful anti-resorptive therapy may be observed in the eect on vertebral fractures in women at higher risk of fracture (e.g. those with prevalent fractures) and in the eect on the incidence of nonvertebral fractures. In both of these populations alendronate appears to be superior to raloxifene. These data, coupled with the potential bene®t of raloxifene on the incidence of breast cancer, have tended to lead to the use of alendronate in those who are at high risk of fracture (treatment of osteoporosis or secondary prevention of fractures) whereas raloxifene is used in a younger population with lower fracture risk (prevention of ®rst fracture) who may also be concerned about reducing the risk of other diseases such as breast cancer or cardiovascular disease. Unmet medical need It is worth noting that despite these advances in the ®eld there are still signi®cant obstacles to the successful treatment of osteoporosis. Even the positive eect of bisphosphonates on BMD, which reaches a maximum of approximately 10% after several years of therapy, is modest when considered against the fact that many osteoporotic women may have lost up to a third of their bone mass before the ®rst fracture and a
486 B. R. MacDonald and M. Gowen
signi®cant amount of the trabecular microarchitecture. The clinical correlate of this is that with long term (3 years) therapy the maximum reduction in fracture rates with anti-resorptive therapy remains at approximately 50%. Thus additional increments in ecacy may be obtained by therapies that are more eective in improving bone mass and repairing trabecular microarchitecture. This is unlikely to be accomplished by antiresorptive therapy even if the ability to reduce bone resorption is maximized because, by de®nition, anti-resorptive therapy can only operate within the bone remodelling envelope which represents only a small proportion of the bone surface at any given time. For the same reason the onset of bene®t of anti-resorptive therapy may be delayed. To date, the earliest available positive data are after 1 year of therapy.7 A key component of the maintenance of long-term ecacy is the willingness of the patient to continue to take daily therapy. A critical component of persistence of therapy is the overall tolerability and convenience of the medication. Many of the marketed antiresorptive therapies face challenges in this area. Bisphosphonates are not absorbed in the presence of calcium and therefore must be taken in the absence of food or calcium containing liquids. There are also important safety concerns about oesophageal erosion and other upper gastrointestinal (GI) damage, particularly with alendronate. This has caused further complications in the dosing regimen such that the drug must be taken in the early morning with water only and the patient must then remain standing to prevent the tablet from lodging in the oesophagus. Despite these precautions there is still evidence that alendronate is associated with adverse upper GI symptomatology, increased consumption of antacids and other GI protective drugs and low compliance.9 Although raloxifene does not have problems associated with dosing it does cause unwanted pharmacological eects that include a worsening of vasomotor symptoms in approximately 30% of patients and an appreciable incidence of leg cramps.10 More serious side eects of venous thromboembolism have been observed with raloxifene and are thought to be associated with oestrogenic activity. Oestrogen therapy (whether unopposed or in combination with progestin therapy) causes breast tenderness and unscheduled vaginal bleeding in the early months of therapy in a high proportion of subjects, and may cause an increased risk of breast cancer with long term use. Nasal calcitonin is also burdened by tolerability problems related to the delivery system and to a perception of marginal ecacy. Thus, there appears to be a compelling need to provide patients with additional choices to be able to maintain long term therapy, if not with a single agent then with a series of dierent drugs to maintain a continuous period of anti-resorptive therapy for sucient time to have a meaningful clinical eect. It is against this backdrop of modest ecacy and poor long term tolerability that new agents for the treatment of osteoporosis will emerge. However, the fact that there are now approved therapies for this disease complicates the ability to approve subsequent compounds. Primarily this is caused by the ethical and trial recruitment problems of performing placebo-controlled studies, particularly in high risk populations, coupled with the extremely large study populations that would be needed to perform fracture outcome studies with active comparator drugs. IMPROVEMENTS IN EXISTING THERAPIES Bisphosphonates In an eort to overcome some of the problems associated with the current therapies several approaches are under consideration. The bioavailability and dosing problems associated with bisphosphonates may be countered by attempts to use parenteral
Emerging therapies in osteoporosis 487
formulations or weekly oral dosing. The latter does not eliminate the need for a strict adherence to the dosing regimen but allows this to happen less frequently.11 Whether this will improve persistence with the therapy remains to be determined. An interesting approach to the use of bisphosphonates is to use intermittent injections (for example the use of ibandronate every 3 months). With such infrequent dosing the size of the dose is of critical importance in order to prevent breakthrough of bone resorption in the weeks preceding the next injection, which can undo the bene®t of the previous months of anti-resorptive cover. This was demonstrated by the failure of the intravenous (I.V.) ibandronate development programme in osteoporosis due to lack of ecacy. Hormone based therapies The use of oestrogen replacement therapy to prevent post-menopausal bone loss and reduce fracture risk is one of the mainstays of anti-osteoporosis therapy. Many dierent preparations of oestrogen have been used including conjugated oestrogens, containing a variety of dierent compounds of varying oestrogenicity, and synthetic oestradiol, containing a single de®ned compound. The safety and ecacy pro®le of all of these products is similar in terms of the clinical pro®le of oestrogenic activity in dierent target tissues. Even though oestrogenic compounds are eective anti-resorptive therapies in terms of changes in BMD and biochemical markers of bone metabolism, long term therapy is still the exception rather than the rule. This is primarily because of the short term tolerability problems of breast tenderness and unscheduled vaginal bleeding and the more long term concerns of an increased risk of breast cancer. The latter issue has not been de®ned with clarity, largely because of the problems of making de®nitive conclusions on the basis of retrospective epidemiological studies. Even if the incidence is increased, the absolute risk of developing breast cancer remains low and enhanced screening may reduce morbidity and mortality. Nevertheless this issue creates a powerful negative image in the minds of patients and prescribers alike. The concept of selective oestrogen receptor modulation (SERM) has been recently advanced to describe the actions of compounds that can have either oestrogenic or antioestrogenic eects depending on the tissue or cell type under investigation. From a clinical perspective the ideal SERM would provide oestrogen agonist eects in bone, central nervous system, cardiovascular system, skin and urogenital epithelium but oestrogen antagonist activity in breast and uterine tissue. To date no such compound is available. Raloxifene, the only marketed product in this class that is indicated for the prevention and treatment of osteoporosis, provides modest anti-resorptive activity in bone that does not reproduce the eects of oestradiol at a molecular level12 and may not provide the required level of ecacy for women with more severe osteoporosis and very low bone mass. It is an oestrogen antagonist in breast and uterine tissue that provides a reduction in the incidence of oestrogen receptor positive breast cancer without any apparent change in endometrial histology. However, the increase in vasomotor symptoms reported in clinical studies is also indicative of oestrogen antagonist eects and may represent an unacceptable barrier to long term therapy for many women. Several other compounds are in clinical development in the hope that a better SERM pro®le may be found, which will address some of the de®ciencies of the clinical pro®le of raloxifene. In terms of complexity of development SERMs represent a major clinical research challenge. Many of the potential bene®ts of SERMs are related to the prevention of disease events such as breast cancer or fractures. De®nitive proof of ecacy, therefore, requires large scale outcome studies addressing changes in the incidence of multiple diseases.
488 B. R. MacDonald and M. Gowen
NOVEL ANTI-RESORPTIVE THERAPIES The osteoclast is uniquely capable of degrading and removing calci®ed matrix and has evolved a number of specialized features to enhance this process. It is a multinucleated cell, re¯ecting high metabolic and synthetic requirements, that is formed from mononuclear precursors in the activation phase of the bone remodelling cycle. The process of osteoclast formation is a control point in the cycle, being modulated by many dierent hormones and cytokines. Once the mature osteoclast is formed it attaches to the bone surface. This binding is mediated by the interaction of the avb3 integrin (vitronectin receptor) with bone matrix proteins such as osteopontin and forms a tight sealing zone around an area underneath the osteoclast, which may be considered to be an extracellular lysosome. Into this space are extruded protons, to maintain a very low pH for the dissolution of minerals, and proteases for the degradation of matrix proteins. Within the cell complex signalling and transport pathways allow for cell motility, control of cellular activity and removal of degraded bone matrix and mineral. This complex cell therefore oers numerous targets that may be exploited to discover and develop novel anti-resorptive therapies. These targets may be considered to be either unique to the osteoclast or more widely expressed but serving a unique function in the osteoclast. a vb 3 integrin (vitronectin receptor) antagonists This integrin appears to be the key mediator of osteoclast binding to bone matrix and thus the formation of a sealing zone around the acidic, protease-rich resorption compartment.13 Abrogation of the binding of the integrin to the matrix should therefore result in inhibition of osteoclast function by allowing the cell to lose contact with the matrix with the subsequent dissipation of the biochemical machinery of bone resorption. This hypothesis was proven by the observation that monoclonal antibodies against this integrin were capable of inhibiting bone resorption in vivo.14 Since then several groups have developed synthetic small molecules that are capable of reproducing this eect both in vitro and in vivo in preclinical studies.15,16 There are a large number of potential integrin heterodimer couplets with dierent cellular expression and function despite closely related protein sequences. Consequently one of the diculties in developing these molecules is maintaining speci®city compared to other integrins, which might lead to unwanted eects in other cell types (e.g. excessive bleeding caused by abrogation of platelet function secondary to antagonism of the GpIIaIIIb). This has actually been exploited by inhibition of the avb5 integrin as well as the avb3 integrin. avb5 is expressed on osteclast precursor cells and is thought to mediate, in part, the fusion of osteoclast precursors to form multinucleated cells thus targeting the activation of osteoclast formation as well as the activity of the mature osteoclast.17 Although the avb3 integrin performs a unique and essential function for the osteoclast its expression is not limited to this cell type.18 It is also expressed in developing new blood vessels in physiological states such as wound healing as well as in disease states such as tumour growth19, in¯ammation20 and retinal neovascularization. This activity may suggest additional therapeutic activities in the treatment of in¯ammatory joint disease, malignancy and proliferative retinopathy/macular degeneration. Another challenge in the development of these compounds has been optimizing oral bioavailability and half-life to be consistent with the dosing regimen required for long term persistence of therapy.18 Despite these obstacles encouraging data have been produced from several companies demonstrating that in the ovariectomized rat the
Emerging therapies in osteoporosis 489
use of these antagonists caused a prevention of the accelerated bone loss that follows ovariectomy.21 Compounds from this class are now moving into the early stages of human studies with exciting potential for treating a range of human diseases in addition to osteoporosis. c-src inhibitors Although not related to integrins per se the c-src protein is involved in the intracellular signalling following the binding of avb3 to matrix ligands. Speci®c inhibitors of c-src have been developed and shown to be eective anti-resorptive agents in preclinical animal studies.22 Interestingly, in addition to reductions in osteoclast activity these agents also appear to increase the rate of bone formation, re¯ecting a role for c-src in osteoblast function. This combination of anti-resorptive and anabolic eects may represent a pro®le that leads to superior ecacy compared to pure anti-resorptives. Cathepsin K inhibitors One of the most exciting developments in osteoclast cell biology in recent years has been the discovery that much of the matrix degrading capacity of the osteoclast is mediated by a single protease designated as cathepsin K.23 Since its discovery it has been shown to be the most abundant cathepsin produced by the osteoclast and to mediate the cleavage of collagen at sites that correspond to naturally observed sites of cleavage in vivo.24 Deletion of the cathepsin K gene in mice results in osteopetrosis in which cortical and trabecular bone volume is markedly increased whereas osteoclast activity is markedly decreased.25 Consistent with these observations in mouse knockout models, deletion mutations of this gene in humans results in pycnodysostosis, a form of osteopetrosis.26 Of great importance to the utility of cathepsin K as a drug target has been the fact that its expression appears to be highly restricted to the osteoclast lineage. Consequently, provided compounds that are selective for cathepsin K without inhibiting other closely related proteases can be synthesized, the potential for unwanted pharmacological eects may be limited. Not surprisingly this is a very attractive area for drug discovery and many companies are active in this area. There are no compounds reported to be in the clinic to date although published data indicates that selective compounds that have in vivo ecacy have been developed.27 It is highly likely that within the next few years clinical trials of cathepsin K inhibitors will be initiated. Vacuolar HATPase inhibitors Even though cathepsin K is essential for the digestion of bone matrix it requires an acidic environment for optimal function and for facilitating the dissolution of mineral from the bone surface to expose the matrix. This acidic extracellular microenvironment between the osteoclast and the bone matrix is maintained by the vacuolar H ATPase, which extrudes protons across the rued border of the cell. Carbonic anhydrase activity provides the source of protons and deletion of the Type II form of this enzyme produces a mild form of osteopetrosis in humans. In order to maintain intracellular pH within normal limits, K ATPases on the opposite surface of the cell are involved in the extrusion of bicarbonate. Proton pumps are not unique to the osteoclast. Perhaps the most obvious role is in gastric acid secretion but renal acid base balance also utilizes this enzyme machinery. Interestingly inhibitors of the gastric proton pump such as omeprazole do not have
490 B. R. MacDonald and M. Gowen
signi®cant inhibitor eects on the vacuolar enzyme present in the osteoclast. Potent and speci®c inhibitors of the osteoclast vacuolar H ATPase have been developed and found to be very potent anti-resorptive therapies in the ovariectomized rat.28 The recent isolation of the 116 kDa osteoclast-speci®c subunit of the vacuolar H ATPase has oered the prospect of deployment of ecient drug discovery screening techniques against this target, a vital requirement in order to be able to provide the chemical diversity needed for the discovery of a range of novel compounds. Inhibitors of cytokine activity Osteoclast formation and activity is largely controlled by locally acting cytokines. Over the past 20 years a large number of dierent cytokines have been shown to stimulate bone resorption in vitro and in vivo. Overproduction of some of these cytokines such as interleukin-1 and tumour necrosis factor a has been implicated in the excessive bone resorption associated with rheumatoid arthritis or post-menopausal osteoporosis.29 Inhibition of these cytokines therefore represents a potential approach for inhibiting bone resorption. Several potential approaches may be considered including the use of soluble receptors or receptor antagonists or small molecule inhibitors of key enzymes that control cytokine synthesis such as p38 mitogen activated protein (MAP) kinase.30 Several such compounds are in clinical development for the treatment of rheumatoid arthritis and other p38 inhibitors have been shown to prevent bone loss in the ovariectomized rat.31 The recent discovery that the receptor activator of nuclear factor kB ligand (RANKL) cytokine is an extremely potent bone resorbing factor32 and that its soluble ligand osteoprotogerin is a correspondingly potent inhibitor of bone resorption33 oers another potential approach to the treatment of osteoporosis. Preliminary clinical studies have indicated that parenteral delivery of osteoprotogerin has profound anti-resorptive activity, considerably more powerful than currently marketed bisphosphonates in terms of reduction in biochemical markers of bone resorption.34 Whether this can be translated into superior eects on bone strength remains uncertain because profound inhibition of bone resorption may lead to the accumulation of microcrack damage that would normally be repaired by bone remodelling. Long term use of such a compound could therefore lead to a weakening of the material properties of the bone matrix.
ANABOLIC THERAPIES In order to provide ecacy beyond that produced by anti-resorptive therapy, agents that stimulate bone formation are required. Such agents may act on the cells of the osteoblastic lineage independently of the bone remodelling cycle or may lead to increases in bone mass by reversing the imbalance between resorption and formation in the bone remodelling cycle. A potential concern with anabolic therapies that aect bone formation independently of the bone remodelling cycle (perhaps best described as aecting bone modelling) is that the new bone that is formed may not be mechanically useful either because it is formed in areas that are not primarily concerned with bone strength or because it is composed of woven bone that does not have the superior material properties of `physiological' lamellar bone. Under such circumstances the relationship between change in BMD and change in fracture rate that appears to hold true for anti-resorptive therapy may be absent. An example of
Emerging therapies in osteoporosis 491
this is ¯uoride which, when tested at relatively high doses caused large increases in BMD but no reduction and possibly a small increase in fracture rates.35 Another important consideration with anabolic therapies is the long term maintenance of increased bone strength. The `mechanostat' hypothesis36 of the regulation of bone strength suggests that as muscle mass decreases with ageing the bones detect the decreased force applied by weaker muscles as relative disuse, with consequent bone loss to a new, lower steady state. Continued loss of muscle mass therefore provokes a progressive loss of bone. Under these circumstances a large positive increment in bone mass caused by an anabolic agent, if not accompanied by a commensurate increase in strength, may be quickly lost once therapy is discontinued, as a result of bone remodelling. Consequently the use of anabolic therapies may require concomitant or subsequent anti-resorptive therapy. Parathyroid hormone and analogues Parathyroid hormone (PTH) has important eects on both bone formation and resorption. It stimulates osteoclast activation and increases the rate of bone resorption. It also causes the activation and dierentiation of bone lining cells into mature bone forming osteoblasts. In situations of continuous elevation of PTH levels, such as primary hyperparathyroidism, the bone resorptive eect predominates, leading to a decline in BMD. Clinical studies have demonstrated that when exogenous PTH is administered daily the intermittent increase in PTH levels allow the bone formation eect to predominate, thus producing large increases in BMD.37 The eect of injected PTH is most marked when given with background anti-resorptive therapy and both 13438 and full length 1-84 peptides are eective.39 The predominant eect of PTH is on trabecular bone. Thus the earliest and largest changes are seen in the lumbar spine, particularly with quantitative computed tomography (QCT) measurements that speci®cally assess changes in trabecular bone.40 The eect on cortical bone is also detectable in the presence of background antiresorptive therapy but the onset is delayed by approximately 6 months.38 In the absence of anti-resorptive therapy no change in cortical BMD is apparent but this does not necessarily imply that there is no impact of this therapy on cortical bone. In fact the thickness of the cortical bone is increased due to bone formation on the endocortical surface but this is not re¯ected in increased bone mass due to relatively poor mineralization. The net eect on bone strength in bones with a large cortical component such as the proximal femur is therefore likely to be positive. In a recent large scale clinical trial PTH 1-34, given as stand alone therapy, was shown to cause a 65% reduction in vertebral fracture rates after a mean of 21 months of therapy in a population of women with low BMD and pre-existing fractures.41 Reductions in non-vertebral fractures were also observed. This landmark study indicates that the bone formed in response to PTH is biomechanically relevant and heralds a new era of combination therapy for the treatment of osteoporosis that promises to provide greater ecacy than has been seen with conventional antiresorptive therapy. An important additional bene®t of combination therapy is that it is much easier to design and recruit patients for studies in which the control group is `best available therapy`. Initially it is likely that such therapeutic regimens will be focused on those at highest risk of fracture. However, with greater clinical experience it is likely that patients with lower risk will eventually be candidates for therapy to enable restoration of `normal' BMD before fracture risk rises to very high levels. In this clinical paradigm it will be important to prevent rapid loss of bone after the anabolic
492 B. R. MacDonald and M. Gowen
therapy has terminated. However, it has been shown that a sequential regimen of PTH 1-84 followed by anti-resorptive therapy is capable of maintaining and possibly enhancing the BMD increases seen during the initial period of PTH therapy.39 Although PTH therapy holds promise as a means of improving on the ecacy of bisphosphonates some drawbacks of this approach remain unresolved. PTH has other important roles in calcium regulation, particularly in the control of renal calcium excretion. It is thought that the temporary in¯uence of raised PTH levels will not have a signi®cant impact on the control of calcium excretion. As the product becomes more widely used following regulatory approval the clinical impact of transient renal eects, if any, is likely to become clearer. A major concern of regulators has been the observation of osteosarcoma formation in animals treated with PTH. This has provoked a cautious approach to the testing of other PTH preparations, with restriction to high risk study populations, even though human osteosarcoma has not been observed in association with hyperparathyroidism. This observation may possibly represent an artefact of performing studies in rodents, because of the presence of open epiphyses throughout life, but even the possibility of a life threatening complication of therapy may act as a limitation on the duration of PTH therapy until much greater experience with the drug is available. The most obvious drawback with PTH therapy in the treatment of osteoporosis is the need for daily subcutaneous injection. This is likely to lead to signi®cant reductions in persistence of therapy because, unlike de®ciency states such as diabetes, there are no immediate life threatening consequences to not continuing with injections. Furthermore, the disease itself may be considered to be asymptomatic until the late stages when fracture has occurred unlike conditions such as rheumatoid arthritis where the patient has a compelling reason to continue injected therapy that modi®es symptoms. In practice this approach to therapy is likely to remain as a niche indication for those with very severe disease who have failed or are unwilling to contemplate other therapeutic approaches. In response to this problem several approaches are in development to either deliver PTH by a more acceptable route or to generate the biological eect of PTH on bone formation. It may be possible to formulate PTH for delivery by the intranasal, inhaled or oral route, although a major hurdle to be overcome will be the probable low bioavailability and consequent high variability in drug levels. In addition, the potential for local adverse eects of PTH in the nasal passages or in the lung may preclude these routes of administration. Calcium receptor antagonists A novel approach to producing a PTH anabolic eect has been to rely on the parathyroid gland to provide the transient PTH increase. Release of PTH from the parathyroid cell is controlled by the level of extracellular calcium, a homeostatic signal that is transduced by the cell surface calcium sensing receptor.42 Thus a decrease in extracellular calcium provokes release of PTH due to decreased binding of calcium to the calcium sensing receptor. It has been hypothesized that compounds that can mimic the eect of calcium would be capable of reducing PTH secretion in situations of excess such as primary hyperparathyroidism and, conversely, that antagonists of the eect of calcium on the receptor would increase the release of PTH. The use of calcimimetic drugs in humans has since been demonstrated to reduce PTH levels in primary hyperparathyroidism.43 Studies in animal models have also demonstrated that calcium receptor antagonists can increase PTH production44 and provoke an anabolic eect in bone45, thus providing the
Emerging therapies in osteoporosis 493
promise of novel therapies that can be used to deliver endogenously produced PTH to the bone forming cells in patients with osteoporosis. Although conceptually elegant there are important hurdles to be overcome with this approach. In particular it is important to ensure that the release of PTH from the parathyroid gland is a transient phenomenon and that the amount of PTH released is sucient to provide a clinically relevant anabolic eect. In addition the calcium receptor may have other roles in other cells that could lead to unwanted pharmacological eects. Instead of delivering PTH, either exogenously administered or endogenously stimulated, it may be possible to mimic the eect of PTH on the PTH receptor by the use of PTH receptor agonists. Peptide analogues of PTH have been shown to produce large increases in BMD in animal models but suer from the same drawbacks as PTH in terms of drug delivery. To date no small molecule PTH receptor agonists have been disclosed. HMG CoA reductase inhibition Considerable interest has recently been generated in the possibility that HMG CoA reductase inhibitors (statins) may produce an anabolic eect in bone by increasing the production of bone morphogenetic protein 2, which stimulates bone formation.46 An additional potential eect of statins on bone metabolism is to inhibit bone resorption, which could be a valuable synergistic eect.47 Evidence from animal models indicates that local or systemic administration of statins produces large increases in bone mass. Retrospective epidemiological studies have suggested that statin usage is associated with decreased fracture incidence.48,49 However, the only prospective study performed to date did not demonstrate the large changes in bone mass that would be expected from an anabolic therapy.50 One possible explanation for this dierence is that the clinically approved doses that provide lipid lowering bene®t are too low to provide a meaningful anabolic eect in bone. It will be interesting to see if the newer, more potent agents that are now available will also provide more robust eects in bone. If the promise of statin therapy on bone can be con®rmed the dual eect on bone and lipids will represent an attractive therapeutic option for elderly individuals who may be at increased risk of both osteoporosis and cardiovascular disease.
SUMMARY This review has illustrated the potential of the novel emerging therapies for the treatment and prevention of osteoporosis with a few speci®c examples of both antiresorptive and anabolic therapies. It is certain that within the ranks of the academic and industrial research groups in this area many more potential areas are under investigation. We have also attempted to describe some of the hurdles that confront successful drug development in this ®eld, not least the large scale and long term studies that are required for regulatory approval and the stringent safety and tolerability pro®les that must be met for successful treatment of osteoporotic patients. Experience tells us that of the novel compounds emerging into clinical research studies in this area most will fail to become marketed therapies. However the bene®ts and de®ciencies of the currently marketed drugs give us a clear perspective on what clinical pro®le is needed for the osteoporotic drugs of the future, whatever their molecular or cellular target. Given the diculties of drug development in this disease
494 B. R. MacDonald and M. Gowen
Practice points . osteoporosis is a widely prevalent disease that may have devastating consequences for aected individuals . eective therapies are available and should be widely used, particularly in patients at high risk of fracture . the emergence of novel, well tolerated and safe therapies will enhance treatment options and increase diagnostic awareness
Research agenda . the unmet needs in the treatment of osteoporosis are for an increased choice of well tolerated, safe drugs and for drugs with enhanced ecacy . alternative clinical research paradigms are needed to increase the feasibility and commercial viability of the development of novel osteoporosis therapies . dialogue with regulators is essential to enable such changes to become accepted as valid approaches to the assessment of the risk to bene®t pro®le of novel osteoporosis therapies
it is incumbent on all researchers in the ®eld to focus resources on those potential products that will have a high chance of meeting this unmet need.
REFERENCES 1. Melton LJ III. Who has osteoporosis? A con¯ict between clinical and public health perspectives. Journal of Bone and Mineral Research 2000; 15: 2309±2314. 2. Ross P, Wasnich R, MacLean C et al. A model for estimating the potential costs and savings of osteoporosis prevention strategies. Bone 1988; 9: 337±347. 3. Orwoll ES. Osteoporosis in men. New Dimensions in Osteoporosis 1999; 1: 2±8, 12. 4. Black D, Cummings S, Karpf D et al. Randomised trial of eect of alendronate on risk of fracture in women with existing vertebral fractures. Lancet 1996; 348: 1535±1541. 5. Cummings S, Black D, Thompson D et al. Eect of alendronate on risk of fracture in women with low bone density but without vertebral fractures. Journal of the American Medical Association 1998; 280: 2077±2082. 6. Ettinger B, Black D, Mitlak B et al. Reduction of vertebral fracture risk in postmenopausal women with osteoporosis treated with raloxifene. Journal of the American Medical Association 1999; 282: 637±645. 7. Harris ST, Watts NB, Genant HK et al. Eects of risedronate treatment on vertebral and nonvertebral fractures in women with postmenopausal osteoporosis. Journal of the American Medical Association 1999; 282: 1344±1352. 8. Wasnich R & Miller P. Antifracture ecacy of antiresorptive agents are related to changes in bone density. Journal of Clinical Endocrinology and Metabolism 2000; 85: 231±236. 9. Colby CJ, Levin TR & Boyko WL. Costs of acid related disorders associated with alendronate use for osteoporosis in a large managed care organization. Journal of Bone and Mineral Research 1999; 14: S207. 10. Evista1, Medical Economics Company, Montrale, NY. Raloxifene hydrochloride labeling information. Physicians Desk Reference 54th Edition 2000; 1583±1585. 11. Schnitzer T, Bone HG, Crepaldi G et al. Therapeutic equivalence of alendronate 70 mg once-weekly and alendronate 10 mg daily in the treatment of osteoporosis. Aging Clinical and Experimental Research. 2000; 12: 1±12.
Emerging therapies in osteoporosis 495 12. Nuttall ME, Fisher PW, Suva LJ & Gowen M. The selective estrogen receptor modulators idoxifene and raloxifene have fundamentally dierent cell-speci®c estrogen-response element (ERE)-dependent/ independent mechanisms in vitro. European Journal of Cancer 2000; 36: 63±64. 13. Nakamura I, Pilkington MF, Lakkakorpi PT et al. Role of avb3 integrin in osteoclast migration and formation of the sealing zone. Journal of Cell Science. 1999; 112: 3985±3993. 14. Crippes BA, Engleman VW, Settle SL et al. Antibody to Beta(3) integrin inhibits osteoclast-mediated bone resorption in the thyroparathyroidectomized rat. Endocrinology 1996; 137: 918±924. 15. Engleman VW, Nickols GA, Ross FP et al. A peptidomimetic antagonist of the avb3 integrin inhibits bone resorption in vitro and prevents osteoporosis in vivo. Journal of Clinical Investigation 1997; 99: 2284±2292. 16. Lark MW, Strap GB, Hwang SM et al. Design and characterization of orally active Arg-Gly-Asp peptidomimetic vitronectin receptor antagonist, SB 265123, for the prevention of bone loss in osteoporosis. Journal of Pharmacology and Experimental Therapeutics 1999; 291: 612±617. 17. Sago K, Teitelbaum SL, Venstrom K et al. The integrin alpha(v)beta(5) is expressed on avian osteoclast precursors and regulated by retinoic acid. Journal of Bone and Mineral Research 1999; 14: 32±38. 18. Miller WH, Keenan RM, Willette RN & Lark MW. Identi®cation and in vivo ecacy of small-molecule antagonists of integrin av b3 (the vitronectin receptor). Drug Discovery Today 2000; 5: 397±408. 19. Eliceiri BP & Cheresh DA. The role of av integrins during angiogenesis: insights into potential mechanisms of action and clinical development. Journal of Clinical Investigation 1999; 103: 1227±1230. 20. Storgard C & Stupack DG, Jonczyk A et al. Decreased angiogenesis and arthritic disease in rabbits treated with an avb3 antagonist. Journal of Clinical Investigation. 1999; 103: 47±54. 21. Lark M, Stromp G, Dodds R et al. Antagonism of the osteoclast vitronectin receptor with an orally active nonpeptide inhibitor prevents cancellous bone loss in the ovariectomized rat. Journal of Bone and Mineral Research 2001; 16: 319±327. 22. Missbach M, Jeschke M, Feyen J et al. A novel inhibitor of the tyrosine kinase Src suppresses phosphorylation of its major cellular substrates and reduces bone resorption in vitro and in rodent models in vivo. Bone 1999; 24: 437±449. 23. Drake FH, Dodds RA, James IE et al. Cathepsin K, but not cathepsins B, L, or S, is abundantly expressed in human osteoclasts. Journal of Biological Chemistry 1996; 271: 12511±12516. 24. Garnero P, Borel O, Byrjalsen I et al. The collagenolytic activity of cathepsin K is unique among mammalian proteinases. Journal of Biological Chemistry 1998; 273: 32347±32352. 25. Gelb BD, Shi GP, Chapman HA et al. Pycnodysostosis, a lysosomal disease caused by cathepsin K de®ciency. Science 1996; 273: 1236±1238. 26. Gowen M, Lazner F, Dodds R et al. Cathepsin K knockout mice develop osteopetrosis due to a de®cit in matrix degradation but not demineralization. Journal of Bone and Mineral Research 1999; 14: 1654±1663. 27. Thompson SK, Halbert SM, Bossard MJ et al. Design of potent and selective human cathepsin K inhibitors that span the active site. Proceedings of the National Academy of Sciences of the USA 1997; 94: 14249±14254. 28. Visentin L, Valente M, Dodds RA et al. SB-242784, a selective inhibitor of the osteoclast v-H -ATPase, prevents bone loss in ovariectomized rats. Bone 1998; 23: S181. 29. Paci®ci R, Vannice JL, Rifas L et al. Monocytic secretion of interleukin-1 receptor antagonist in normal and osteoporotic women: eects of menopause and estrogen/progesterone therapy. Journal of Clinical Endocrinology and Metabolism 1993; 77: 1135±1141. 30. Lee JC, Laydon JT, McDonnell PC et al. A protein kinase involved in the regulation of in¯ammatory cytokine biosynthesis. Nature 1994; 372: 739±746. 31. Bradbeer JN, Stroup GB, Homan SJ et al. An orally active inhibitor of cytokine synthesis prevents bone loss in the ovariectomized rat. Journal of Bone and Mineral Research 1996; 11: S125. 32. Lacey DL, Timms E, Tan HL et al. Osteoprotegerin ligand is a cytokine that regulates osteoclast dierentiation and activation. Cell 1998; 93: 165±176. 33. Tsuda E, Goto M, Mochizuki S et al. Isolation of a novel cytokine from human ®broblasts that speci®cally inhibits osteoclastogenesis. Biochemical and Biophysical Research Communications 1997; 234: 137±142. 34. Bekker PJ, Holloway D, Nakanishi A et al. The eect of a single dose of osteoprotogerin in postmenopausal women. Journal of Bone and Mineral Research 2001; 16: 348±360. 35. Riggs BL, Hodgson SF, O'Fallon WM et al. Eect of ¯uoride treatment on the fracture rate in postmenopausal women with osteoporosis. New England Journal of Medicine 1990; 322: 802±809. 36. Frost HM. Bone `mass' and the `mechanostat': a proposal. Anatomical Record 1987; 219: 1±9. 37. Lindsay R, Nieves J, Formica C et al. Randomised controlled study of eect of parathyroid hormone on vertebral-bone mass and fracture incidence among postmenopausal women on oestrogen with osteoporosis. Lancet 1997; 350: 550±555. 38. Roe EB, Sanchez SD, del Puerto GA et al. Parathyroid hormone 1-34 (hPTH 1-34) and estrogen produce dramatic bone density increases in postmenopausal osteoporosis ± results from a placebo-controlled randomized trial. Journal of Bone and Mineral Research 1999; 14: S137.
496 B. R. MacDonald and M. Gowen 39. Rittmaster RS, Bolognese M, Ettinger MP et al. Enhancement of bone mass in osteoporotic women with parathyroid hormone followed by alendronate. Journal of Clinical Endocrinology and Metabolism 2000; 85: 2129±2134. 40. Cann CE, Roe EB, Sanchez SD et al. PTH eects in the femur: envelope-speci®c responses by 3DQCT in postmenopausal women. Journal of Bone and Mineral Research 1999; 14: S137. 41. Neer RM, Arnaud C, Zanchetta JR et al. Recombinant human PTH [rhPTH 1-34] reduces the risk of spine and non spine fractures in postmenopausal osteoporosis. Endocrine Society 82nd Annual Meeting, Clinical Trials Symposium Session 2000; 42. 42. Nemeth EF. Calcium receptors as novel drug targets. In Bilezekian JP, Raisz LG & Rodan G (eds) Principles of Bone Biology, pp 1019±1035. San Diego: Academic Press, 1996. 43. Silverberg SJ, Bone HG III, Marriott TB et al. Short-term inhibition of parathyroid hormone secretion by a calcium-receptor agonist in patients with primary hyperparathyroidism. New England Journal of Medicine 1997; 337: 1506±1510. 44. Nemeth EF, Fox J, Delmar EG et al. Stimulation of parathyroid hormone secretion by a small molecule antagonist of the calcium receptor. Bone 1998; 23: S156. 45. Gowen M, Stroup GB, Dodds RA et al. Antagonizing the parathyroid calcium receptor stimulates parathyroid hormone secretion and bone formation in osteopenic rats. Journal of Clinical Investigation 2000; 105: 1595±1604. 46. Mundy G, Garrett R, Harris S et al. Stimulation of bone formation in vitro and in rodents by statins. Science 1999; 286: 1946±1949. 47. Sato T, Morita I & Murota S. Involvement of cholesterol in osteoclast-like cell formation via cellular fusion. Bone 1998; 23: 135±140. 48. Meier C, Schlienger R, Kraenzlin ME et al. HMG-CoA reductase inhibitors and the risk of fractures. Journal of the American Medical Association 2000; 283: 3205±3210. 49. Wang P, Solomon D, Mogun H & Avom J. HMG-CoA reductase inhibitors and the risk of hip fractures in elderly patients. Journal of the American Medical Association 2000; 283: 3211±3216. 50. Watanabe S, Fukumoto S, Takeuchi Y et al. Eects of one year treatment with statins on bone mass and metabolism. Journal of Bone and Mineral Research 2000; 15: S194.