| ABSTRACTS I-1 to I-33
ORAL PRESENTATIONS, INVITED SPEAKERS
PATHOGENESIS AND PREVENTION OF FRACTURES AMONG ELDERLY PEOPLE
1Medical School, University of Tampere, Finland
2Accident & Trauma Research Center, UKK Institute, Tampere, Finland
The number and age-specific incidence of low-trauma fractures among the older adults have dramatically increased in many countries during the recent decades and without any population-level intervention the increasing trend is likely to continue thus creating a true public health problem to our aging societies. There is thus a clear need to better understand the etiology, risk factors, pathogenesis, and injury mechanisms of these fractures since improved knowledge on these issues is likely to provide important clues and possibilities for fracture prevention.
Low-trauma fracture (often also called age-related fracture, mild-moderate energy fracture, or osteoporotic fracture) of an older adult is defined as a bone fracture that occurs in an elderly person as a consequence of a mild-moderate trauma only. The most typical trauma of this category is a fall to the floor from standing height or less. Compromised bone strength (osteopenia or osteoporosis) and falling, alone, or more frequently in combination, are the main independent and immediate risk factors of these fractures through which all the other, more distant risk factors (such as aging, inactivity, poor nutrition, smoking, use of alcohol, diseases, medications, impairments, and disabilities) operate.
Of these two, falling, not osteoporosis, is the strongest single risk factor for a fracture, and when a person falls, the type and severity of falling (fall height and energy; fall direction; fall mechanics; anatomical site of the impact; and energy absorption capacity and impact force attenuation of the body-landing surface complex) are crucial in determining whether or not a fracture occurs. In fact, only minority of fractures of older adults are truly osteoporotic (that is, most fractures in the community come from the larger segment of population without osteoporosis), but this does not mean that osteoporosis prevention is useless: we just have to understand that in planning a successful prevention strategy both of these major risk factors have to be taken into account.
Prevention of elderly people's fractures can be divided into three parts: 1. Prevention and treatment of osteoporosis, 2. Prevention of falling, and 3. Prevention of fractures despite osteoporosis and falling (injury-site protection).
Osteoporosis: Osteoporosis is an independent, moderate-level risk factor for fracture and should therefore become prevented and treated. Maximizing peak bone mass and preventing bone loss by exercise, calcium, and vitamin D, and, treatment of established osteoporosis with bone-affecting drugs (HRT, bisphosphonates, SERMs, calcitonin), have a strong scientific basis and have been recommended by many authorities and consensus conferences. A sufficient long-term compliance and adherence to any of these interventions may, however, be a problem. The same naturally concerns programs of fall prevention and injury-site protection.
Falling: Since falling is the strongest single risk factor for a fracture of an older adult, fall prevention is essential. Recent randomized trials have consistently shown that training balance and strength is able to reduce the older person's risk of noninjurious and injurious falls - even cost-effectively. This is to be expected since exercise is able to improve many risk factors of falling, such as muscle strength, flexibility, balance, coordination, proprioception, reaction time, and gait - even in very old and frail people. As yet, only one randomized trial has examined whether fractures can be reduced by exercise, and, this 10-year follow-up study showed that in postmenopausal women regular back-strengthening exercises for 2 years reduced the risk of vertebral fractures by 63%. This finding is in line with many epidemiologic studies, which consistently show that both past and current physical activity is protective against hip fracture, the risk reduction being 20-70%. Many of these studies have even found a dose-response relation between the exercise exposure and the fracture risk. Thus, it seems that we can well recommend regular weight-bearing and other exercises for elderly people - not only to maintain bone health and muscular performance but also to keep the older persons safely on their feet.
In addition to exercise, reducing the number and doses of psychotropic medication has been shown to reduce elderly persons' risk of falling. A similar effect has been seen after cardiac pacing of older adults with cardioinhibitory carotid sinus hypersensitivity. More multifaceted risk factor-modifying interventions have also been successful in preventing falls of the older adults by simultaneously affecting many intrinsic and extrinsic risk factors of falling. The problem with these interventions is that they can not distinguish between the independent role of each modified risk factor, and so it cannot be determined which part of the intervention is effective and which is not. Such multifactorial approaches are also labor intensive and their cost-effectiveness has not been ascertained.
Injury-site protection: Since great majority hip fractures are caused by a sideways fall with direct impact on the greater trochanter of the proximal femur, a fascinating option to prevent the fracture is a specially designed external hip protector, so that at the time of the fall-impact, the impacting force and energy are weakened and diverted away from the greater trochanter by the protector. We conducted a large (n=1801) randomized multicenter trial with the KPH Hip Protector (Respecta, Helsinki) and showed, with intention-to-treat analysis, that the risk of hip fracture was 60% less in the protector group than in the control group (adjusted relative hazard 0.4, 95% CI, 0.2-0.8), and that by protector efficacy or active treatment analysis, the risk reduction was more than 80% if the protectors were actually worn at the time of falling (adjusted relative hazard 0.1, 95% CI, 0.03-0.5). The number needed to treat (NNT) to avoid one hip fracture was 41 persons for one year (95% CI, 25-115), or 8 persons for five years (95% CI, 5-23). These results are rather encouraging to recommend scientifically tested external hip protectors for high-risk frail elderly people, especially those who have fallen before, had fractures, poor balance, and impaired mobility.
Little is known about the injury mechanisms of the age-related fractures other than hip fracture. A recent controlled study from The Accident & Trauma Research Center of the UKK Institute, Tampere, Finland indicates that a great majority of the elderly persons' upper extremity fractures is caused by a direct, fall-induced impact on the fractured site. Although this finding is encouraging concerning possibilities for prevention of these fractures by injury-site protection, no such study has been conducted, and therefore definite recommendations for protecting the older adults' sites other than the hip can not be made at present.
ROLE OF THE WNT/LRP5 PATHWAY IN REGULATION OF BONE MASS AND OSTEOBLAST FUNCTION
Cleveland OH, USA
Abstract not supplied
THE TRANSCRIPTION FACTOR OSTERIX IS REQUIRED FOR OSTEOBLAST DIFFERENTIATION AND BONE FORMATION
B. de Crombrugghe*, X. Zhou, K. Nakashima
The University of Texas M. D. Anderson Cancer Center
The Osterix (Osx) gene, which is specifically expressed in osteoblasts, specifies a polypeptide with typical features of transcription factors. Its DNA binding domain consists of three C2H2-type zinc fingers that have a high degree of identity with similar DNA binding domains in the transcription factors Sp1, Sp3, and Sp4 and its N-terminal segment contains a strong transcription activation domain. Homozygous Osx-mutant mice completely lack bone formation in both endochondral and membranous skeletal elements although they show normal cartilage development. In these mutants mesenchymal precursor cells are arrested in their differentiation into osteoblasts. Indeed osteoblast marker genes such as BSP, osteonectin, osteopontin, and osteocalcin are not expressed and Type I collagen expression is severely reduced to levels commonly seen in other mesenchymal cells. However, Runx2, which encodes a previously identified transcription factor needed for bone formation, is expressed at levels comparable to those found in wild-type osteoblasts. Since no Osx transcripts were detected in skeletal elements of Runx2-null mice, Osx must be downstream of Runx2 in the pathway of osteoblast differentiation.
These properties reveal an essential new step in the pathway of osteoblast differentiation and suggest the following model. Osteoblast progenitors first differentiate into 'preosteoblasts', a process in which Runx2 and its Cbf-beta subunit play an essential role. These preosteoblasts then differentiate into mature, functioning osteoblasts that express characteristic osteoblast marker genes, a process that requires Osx.
Interestingly, Runx2-expressing Osx-null cells arrested in their differentiation into osteoblasts acquire a cell fate characterized by the expression of a series of genes that are typical of chondrocytes. The acquisition of this chondrocyte cell fate by Osx- defective cells implies that Osx is a negative regulator of the chondrocyte phenotype. We, therefore, propose that Osx has a role in the mechanism that segregates the osteoblast and chondrocyte lineages from a common osteochondroprogenitor during endochondral bone formation. By being a negative regulator of Sox9 expression, Osx would commit Runx2-expressing precursor cells to the osteoblast lineage and prevent these cells from adopting a chondrocyte phenotype.
BIOLOGY ON SMART SILICON - SPECULATIONS ON THE NANOTECHNOLOGY OF 2010
The 'laboratory on a chip' is one of today's big ideas. In the form of 'gene chips' (or DNA microarrays, to give them their proper name), such technology is already making a big impact on biomedical research. But where is this technology heading? Taking my cue from the history of microelectronics, I will speculate on the future of 'laboratory on a chip' technology. Just as memory chips have moved from ROM (programmed in the factory) through EPROM (one-off programmable by users) to RAM (fully programmable and reusable), I want to compare current 'gene chips' to ROMs, and suggest how the nanotechnology of the next decade may produce fully programmable laboratories on chips. This should enable, for example, interactive (or 'smart') combinatorial synthesis where steps further down stream can be influenced in a controllable way by real-time 'on-chip' analysis of intermediate products from previous steps. Some speculative ideas, based on a recently published method for
producing surfaces with switchable hydrophobicity, will be briefly discussed. In my account, I will emphasize how nanotechnology is providing many opportunities for fruitful collaborations between physical and life scientists of the 21st century.
PROBING BONE CELL FUNCTION WITH AFM
M. A. Horton*, G. Charras
Bone and Mineral Centre, Dept. of Medicine, University College London, London, UK
AFM is evolving from its traditional use in high-resolution imaging and measurement into a tool for the analysis of biological processes at lower resolution. Driven by the ability of AFM to act as a precise 'nanomanipulator', we have developed a microscope for cell biology applications based upon commercially available equipment. This system has been used to investigate cell adhesion molecules (single molecule receptor-ligand binding, receptor mapping, downstream signaling in living cells) and cellular biomechanics (material properties, strain activated receptor analysis). Our data suggest that AFM can be developed for a wide range of bio- technical applications in biology. An exemplary application is the study of bone cell function (1). The skeleton adapts to mechanical use, though, at the cellular level, the distribution and size of strains generated and their detection are ill understood. The magnitude and nature of strains to which osteoblastic cells respond have been investigated by using the cantilever probe of an atomic force microscope (AFM) as an indenter, with cellular responses analysed via a linked confocal microscope. Two different downstream cell-signaling pathways have been identified in response to mechanical deformation. One is activated upon contact of the AFM tip with the cell surface and depends on activation of stretch-activated ion channels. The second follows stress relaxation after withdrawal of the indenter and requires an intact microtubular cytoskeleton. The cellular responses were modulated by selectively disrupting components of the cytoskeleton or modifying signaling pathways to identify possible transducers of mechanical stimuli. Using these techniques, we have estimated the magnitude of cellular strains required to elicit an intracellular calcium response (a generic 'marker' of activation of cellular signaling), and propose a model that links single cell responses to whole bone adaptation. This AFM-based technique may also aid our understanding of the process of cellular/tissue adaptation to mechanical usage in other organs, such as in skeletal and vascular smooth muscle, systems that are exposed to mechanical forces under normal physiological conditions and in various pathologies.
1: Charras G, Lehenkari P, Horton M. Biotechnological applications of atomic force microscopy. Methods Cell Biol. 2002 68:171
NANOTECHNOLOGY OF COLLAGEN AND BONE: INTERACTION AND STRUCTURE
P. K. Hansma1*, T. Gutsmann1,2, G. Fantner1, J. Kindt1, M. Venturoni1,3
1University of California, Santa Barbara, USA
2Research Center Borstel, Germany
3Bologna University, Italy
The Atomic Force Microscope has been used to study the interactions between collagen molecules both in bone and in pure collagen samples as a function of their ionic environment (James Thompson, Johannes Kindt, Barney Drake, Helen Hansma, Daniel Morse, and Paul Hansma. Bone Indentation recovery time correlates with bond reforming time. Nature 414: 773-775). A great deal is known about bone microstructure and the microcracks that are precursors to its fracture, but little is known about the basic mechanism for dissipating the energy of an impact to keep the bone from fracturing. Bone is a nanocomposite of hydroxyapatite crystals and an organic matrix. Because rigid crystals such as the hydroxyapatite crystals cannot dissipate much energy, the organic matrix, which is mainly collagen, must be involved. A reduction in the number of collagen cross links has been associated with reduced bone strength and collagen is molecularly elongated ('pulled') when bovine tendon is strained.
Using an atomic force microscope, a molecular mechanistic origin for the remarkable toughness of another biocomposite material, abalone nacre, has been found. Here we report that bone, like abalone nacre, contains polymers with 'sacrificial bonds' that both protect the polymer backbone and dissipate energy. The time needed for these sacrificial bonds to reform after pulling correlates with the time needed for bone to recover its toughness as measured by atomic force microscope indentation testing. We suggested that the sacrificial bonds found within or between collagen molecules may be partially responsible for the toughness of bone.
HOW DO WE ASSESS BONE QUALITY?
C. H. Turner
Indiana University, Indianapolis, USA
Bone 'quality' is a poorly defined term used to describe important properties of bone tissue other than bone mineral density (BMD). There are a number of diseases that affect bone quality, including osteogenesis imperfecta, osteopetrosis, and osteomalacia. In addition many pharmacological treatments for osteoporosis change bone quality (in addition to BMD). Important examples include fluoride, which can cause mineralization defects and changes mineral crystal size; antiresorptive drugs that reduce resorption pits in trabeculae; bisphosphonates, which can increase mineralization and microdamage accumulation; and teriparatide that increases bone porosity. Some drug effects on bone quality are advantageous and reduce fracture risk disproportionate to changes in BMD. Other drug effects may be detrimental for bone quality. The most detrimental outcome was with high dose fluoride therapy, which caused mineralization defects and increased fracture rates. These data demonstrate the severe compromise of bone quality that can result if mineralization is compromised. Other effects on bone quality, including increased porosity, mineralization and microdamage accumulation, reduce bone strength and/or toughness in animal models but have not yet been shown to increase fracture risk.
Changes in bone tissue quality are extremely difficult to measure non-invasively. Proper assessment requires direct examination of the tissue. The bone biopsy is the best approach. Many of the tissue-level changes that affect bone quality, including mineralization, porosity, microdamage, and bone turnover, can be measured accurately using standard histomorphometric techniques. Additional information about fracture risk, independent of BMD, can be had with measurements of bone size and shape. Assessment of cross-sectional moment of inertia or section modulus can be done from standard densitometry or CT scans. This information complements BMD. In fact, one can calculate a bone strength index (BSI) as the product of volumetric BMD and the bone section modulus. Measurements such as BSI may determine fracture risk better than BMD alone.
ROLE OF BONE TURNOVER IN BONE QUALITY
University Claude Bernard and Inserm Research Unit 403, Lyon, France
Although initial studies of bone turnover in osteoporosis have used dynamic histomorphometry of the iliac crest, most studies performed in the past 15 years have used biochemical markers of bone turnover. Bone markers can be used to predict the rate of bone loss in postmenopausal women. Three independent studies have shown that high bone turnover is associated with increased bone loss over 4 to 15 years in women 50 to 70 years of age. In addition, we have shown in elderly women that increased bone resorption, i.e. above the premenopausal range, is associated with a twofold increase in the risk of hip fractures and that those with both a low BMD (T score below -2.5) and increased bone resorption have a 4 to 5 fold increase in hip fracture risk. These results have been confirmed by several independent studies in other cohorts. We have recently shown that increased bone turnover predicts the risk of all fragility fractures (including hip, spine and other sites) in a younger cohort of postmenopausal women followed for an average of 5 years. The mechanisms underlying the increased bone turnover in some -but not all- postmenopausal women is unknown. The increase appears to be independent from the residual secretion of 17 b estradiol, (E2), assessed by a highly sensitive radioimmunoassay. Indeed, we found that a low serum E2 predicts the risk of fragility fractures in late postmenopausal women -but not in the elderly- independently of the rate of bone turnover. Some of the post translational modifications of bone collagen may influence bone strength. We have found recently that the degree of isomerization of the C-telopeptide of type I collagen is associated with the subsequent risk of fragility fractures in postmenopausal women, independently of the rate of bone turnover. Experimental data suggest that the ratio of various cross links present in bone matrix is also a significant determinant of bone strength.
In summary, amongst the multiple component of bone quality, bone turnover and the biochemical component of bone matrix appear to play an important role.
IMPORTANCE OF MINERALIZATION FOR BONE STRENGTH AND RESPONSE TO THERAPEUTIC AGENTS
P. J. Meunier*, G. Boivin
INSERM Unité 403, Faculté de Médecine R. Laennec, Lyon, France
Bone strength is determined not only by the volume of bone tissue and the microarchitectural organization of this bone, but also by the degree of mineralization of bone matrix (DMB). The mineralization process consists of a primary deposition of mineral substance on the calcification front, followed by a slow and progressive increase of the mineral deposition named secondary mineralization. In osteoporosis, there is a negative imbalance between bone resorption and bone formation, resulting in bone loss, and microarchitectural deterioration of the trabecular network. Therapeutic agents for osteoporosis could increase bone strength by three separate, but interrelated effects on bone tissue : 1) the prevention of bone loss and thus the preservation of bone microarchitecture, 2) an increase in the volume of bone matrix, and 3) an increase in the DMB to a level similar or higher to that seen in healthy premenopausal women, through a prolongation of the duration of secondary mineralization. Our model is based on the impact of changes in bone remodeling rate on the DMB. Thus, any agent (parathyroid hormone) or event (menopause, ovariectomy) which provoke an augmentation in the 'birthrate' or activation frequency of BMU, induce a decrease of the 'lifespan' of BSU, in other words on the time available for the secondary mineralization. This leads to the fact that new BSU are
resorbed before they have fully completed their secondary mineralization, as proven by the presence of a large amount of uncompletely mineralized BSU and a low mean DMB. Conversely, antiresorptive agents (bisphosphonates, estrogen, SERMs) which cause a marked reduction in the 'birthrate' of BMU, prolong the 'lifespan' of the BSU, allowing a more complete secondary mineralization. This should finally provoke an increase of DMB. This new approach of the determinants of bone strength and the results of our recent studies using antiresorptive (alendronate, raloxifene), forming (teriparatide) and decoupling (strontium ranelate) agents used in the treatment of osteoporosis, emphasize that bone mineral substance is an important factor to take into account in the pathophysiology and the response to therapeutic agents for osteoporosis and other bone conditions.
MODULATION OF ESTROGEN ACTION BY ER ISOFORMS
Abstract not supplied
RISKS AND BENEFITS OF HRT AND SERMS
E. Barrett-Connor, MD, Department of Family and Preventive Medicine, University of California, San Diego, La Jolla, CA 92093-0607, USA
The Women's Health Initiative (WHI) demonstrated for the first time (in a clinical trial) that HRT prevents clinical spine and hip fractures in healthy women unselected for osteoporosis. Raloxifene, the only selective estrogen receptor modulator (SERM) with fracture outcome data in clinical trials, was studied in women selected for low bone density or prior fracture. In these women, raloxifiene was shown to prevent morphometric spine fractures, but not hip fractures. Estrogen is less expensive than raloxifene, but the choice between these regimens will probably be based on extra-skeletal effects.
In the WHI and other clinical trials, HRT increased the risk of heart disease and stroke or had no effect. In the Multiple Outcomes of Raloxifene Evaluation (MORE) there was no overall effect on cardiovascular disease, but in women at high risk of heart disease raloxifene reduced the risk of cardiac events and stroke.
In the WHI, HRT significantly increased the risk of breast cancer in women treated for more than 4 years. In MORE, raloxifene significantly reduced the risk of invasive estrogen-receptor positive breast cancer, with (nonsignificant) differences beginning in the first year.
In WHI, HRT reduced the risk of colon cancer; no effect on colon cancer has been observed with raloxifene.
In separate studies, HRT has not been shown to improve memory in women with or without memory loss/dementia. One raloxifene study showed no change in cognitive function tests overall, but did show improvement in two tests in women aged 70 and older.
In clinical trials, HRT improves quality of life only in women with severe menopause symptoms; HRT is clearly the best medication for hot flushes. Raloxifene does not relieve hot flushes, and makes them worse in 10-20% of women. In clinical trials the hot flushes associated with raloxifene were rarely severe enough to cause women to stop the medication, but women with severe menopause symptoms are not included in long-term trials looking at chronic disease outcomes. Nevertheless, continuation rates in clinical trials are better with raloxifene than with HRT. Main reasons for stopping HRT are bleeding and breast pain.
HRT has serious side effects, the most common of which is deep vein thrombosis/pulmonary emboli. The risk is the same for raloxifene. Raloxifene has no other known serious side effects, while estrogen clearly increases breast cancer risk, and probably increases stroke and heart disease risk.
In terms of chronic disease, the risks of HRT are small, but exceed the benefits, with regard to prevention of chronic disease. The raloxifene risk-benefit ratio in women with osteoporosis appears to be more favorable, with reassuring results presented at this meeting. These results are, however, based on post-hoc analyses, and results from ongoing trials including CORE and RUTH will be needed to better assess the global risk benefit ratio of raloxifene. For the moment, raloxifene is certainly safe with regard to cardiovascular disease and breast cancer, two of the big worries when prescribing HRT to prevent chronic disease in healthy women.
DEVELOPMENT OF NOVEL ANTI-RESORPTIVE THERAPIES TARGETING THE RANKL-RANK SIGNALING PATHWAY: FROM FUNCTIONAL GENOMICS TO BEDSIDE
D. L. Lacey
Amgen, Inc, Thousand Oaks, CA, USA
The discovery of the TNF/TNFR-related osteoprotegerin (OPG)/RANKL/RANK axis and its central role in osteoclast biology has provided the basis for the generation of therapeutics. At Amgen, the OPG/RANKL pathway was fortuitously discovered using functional genomics. Transgenic mice expressing osteoprotegerin hepatically (apoE) had normally-shaped, dense bones histologically consistent with osteopetrosis. While the apoE OPG transgenics had incisors, col1A OPG transgenics exhibit classic rodent osteopetrosis with runting, misshapen long bones, and lack of incisor eruption. The OPG k/o exhibited profound high turnover osteopenia with spontaneous fractures. Recently, Whyte et al have reported that juvenile Paget's disease, an autosomal recessive osteopathy, characterized by rapidly remodeling woven bone, osteopenia, fractures, and progressive skeletal deformity can be attributed to the absence of OPG in humans. The ligand for OPG was identified in a 32D cell library and was identical to RANKL, a ligand for RANK, a TNFR superfamily member found on dendritic cells. RANKL promotes the differentiation, activation, and survival of osteoclasts through RANK. The signaling pathway includes both the JNK and NFkB pathways and also involves members of the TRAF family, notably TRAF6. Knockouts for both RANKL and RANK yielded mice with severe osteopetrosis in addition to uncovering a role for this axis in immune system and mammary gland development. Numerous OPG-based protein constructs were evaluated in mice using the Shenck screening assay. Effective constructs led to the accumulation of metaphyseal bone and cartilage within 5 days of administration and were effective in proresorption challenges (cytokines, hormones) and models of osteoclast-mediated skeletal disease. Overall, RANKL binding proteins have shown beneficial effects in rodent models of postmenopausal osteoporosis, hypercalcemia of malignancy, metastatic bone disease, weightlessness, rheumatoid arthritis, and multiple myeloma. Two forms of OPG have entered early clinical trials and both have shown profound antiresorptive activity following single dose exposures. OPG dose-dependently suppressed bone resorption markers up to 80% rapidly following a single exposure and was well tolerated. While the ultimate success of any therapeutic is difficult to predict in early development, this program is an example of how basic discoveries have provided a strategy to target osteoclasts in diseases characterized by excessive bone resorption.
RELEVANCE OF THE RANKL/RANK/OPG PATHWAY IN CLINICAL PRACTISE
L. C. Hofbauer
Division of Gastroenterology and Endocrinology, Philipps-University, D-35033 Marburg, Germany
The recent characterization of the essential cytokine of osteoclast cell biology, receptor activator of NF-kappaB ligand (RANKL) and its antagonist osteoprotegerin (OPG), have led to a detailed molecular and cellular understanding of bone metabolism in health and disease. RANKL and OPG play an essential role for osteoclast formation and activation, and various metabolic, inflammatory and malignant bone diseases employ this cytokine system to trigger osteoclastic bone resorption. Elucidation of the RANKL/RANK/OPG pathway has important implications for the pathogenesis, diagnosis and treatment of human bone diseases. Most disorders characterized by enhanced osteoclastic bone resorption are due to an imbalance of the RANKL-to-OPG ratio, with RANKL exceeding OPG, including various forms of osteoporosis. Constitutively activating RANK mutations have been reported in familial expansile osteolysis, and inactivating OPG mutations leading to the absence of OPG protein and unopposed effects of RANKL are the cause and mechanism in juvenile Paget's disease. Malignant tumors (multiple myeloma) and inflammatory bone diseases (periodontal disease, rheumatoid arthritis) are characterized by an excessive overproduction of RANKL by tumor and immune cells that overwhelms the endogenous safeguard mechanism of OPG production. OPG (and more recently soluble RANKL) serum levels have been evaluated as novel biochemical markers of bone metabolism. Unexpectedly, OPG serum levels are increased in several forms of osteoporosis and in vascular diseases. This apparent paradox may indicate that OPG serum changes are a consequence rather than a cause of disease. OPG serum levels are decreased in patients with multiple myeloma which is due to degradation of the OPG protein by myeloma cells. Blockade of RANKL by administration of OPG or OPG derivatives, neutralizing RANKL antibodies or inhibitory RANK antibodies has been successfully used to treat various bone diseases including malignant bone diseases, where it has been proven to prevent osteolysis, to mitigate hypercalcemia, to halt tumor progression, to reduce tumor-associated pain, and to prolong survival. Two preliminary controlled studies in humans have demonstrated that administration of an OPG-Fc fusion protein is a safe and efficient therapy that prevented alterations of biochemical markers of bone turnover in postmenopausal osteoporosis and myeloma bone disease.
WHAT TYPE OF EXERCISE IS BENEFICIAL TO THE SKELETON?
UKK Institute for Health Promotion Research, Tampere, Finland
Results from a number of animal experiments have suggested that the rate, magnitude, distribution, frequency and duration of the dynamic strain stimulus are among the important components in bone adaptation. That is, either intensive or unusual exercise programs lead to adaptation in bone architecture. These experiments suggest that some components of total mechanical loading of bones are more effective in influencing bone architecture than others. Application of these basic rules from animal studies to practical exercise regimen include: high-magnitude, repetitive, high- impact and uncustomary loading. In other words a loading pattern that differs from what the bone accustomed to. Physical activity generates loads on bone by means of the magnitude of the physical force and the rate at which this force is applied. Thus, the loads on bone can be determined primarily by the movement conditions (velocity of the segments in motion, number of repetitions, muscular activities) and boundary conditions (anthropometry, fitness level, performance surface, type of shoe). Accordingly, any change in the movement conditions affects the kinetics of the movement and probably also the mechanical stress in different exercise regimens. In terms of skeletal loading, weight training, for instance, creates high-magnitude loading and more specifically, extreme torques in upper extremities whereas lower extremities experience large compressive stresses. The most commonly practised exercise is walking that involves a large number of repetitive loading in lower limbs but is more suited to cardiovascular performance than local bone loading. Exercises that involves speed and power are characterised by fast, forceful, accelatory, impact producing movements often in multiple directions seems to be an appropriate type of training with which to influence osteogenic response, as it would place a variety of forces on bone. Previous studies have reported that athletes engaged in sports producing high strain rates via versatile impact-type movements have much higher bone mass (9-40%), better geometry and stronger structure (the largest differences over 50%) than their sedentary controls. Adaptation to this loading seemed to occur in a site-specific fashion by gross-geometric changes, structural or architectural changes or by their combination. Clinical trials also support the high-impact loading concept. Consequently, the important components of osteogenic exercise stimulus are high strain rates (speed and power) and high peak forces in versatile movements.
NEUROTRANSMITTER FUNCTION IN BONE
Department of Veterinary Basic Sciences, Royal Veterinary College, London, UK
In light of the conservation of genetic information across the animal kingdom, it should not be surprising that signalling systems identified originally in one tissue should have functions in others. Expression by bone cells of bradykinin, VIP, glutamate, dopamine, serotonin and P2 ATP receptors suggests novel mechanisms of intercellular signalling that occur in bone. Since pathophysiology of bone diseases is a function of orchestrated activity of cells that control formation and resorption, novel signalling systems cast new lights on understanding of normal function and disease.
However, involvement of neurotransmitters in intercellular signalling in bone raises a fundamental question: why there should there be similarities between such disparate tissues as the nervous system and bone?
Intercellular communication in the nervous system is rapid, with ability to resolve high frequency repetitions, modulate signals in response to inhibitory or potentiating stimuli and retain persistent record of events as memory. Why should bone, which changes its properties more slowly, require such a mechanism? One reason could be that responses of bone to mechanical loading requires ability to discriminate different rates of application of strains, where important events occur in less than 40msec. In addition, only very short durations of exercise/loading are necessary to induce maximal osteogenic effects. While the idea that strain memory could exist has been considered for some years, data on a possible mechanism for it are sparse. Recently, the potential involvement of neurotransmitter mediated signalling in bone cells has raised the possibility that neuronal-like memory mechanisms may be involved in bone. Preliminary data shows that all the molecules necessary for long term potentiation, the process in the CNS by which memory formation may occur are present in osteoblasts, and that loading in vitro activates the same sequence of events in those cells as neurones.
The implications of understanding of these novel signalling pathways are not clear. However, the density of research effort in neuroscience means that information, reagents and even candidate drugs exist to modify neurotransmitter function. These could impact on bone strength and treatment of osteoporosis.
MECHANICAL LOADING, OSTEOCYTES AND APOPTOSIS
B. S. Noble
Musculoskeletal Research Unit, University of Edinburgh, Edinburgh, UK
Bone is the ultimate 'smart' structural material. It is self-designing, adapting as it does to the prevailing mechanical needs of the organism by adding new bone in regions that require it and undertaking targeted sculpting of each bone internally and externally to remove redundant material. It is also self-repairing, sensing, removing and replacing damaged or mechanically insufficient volumes of bone. The effector cells for this process are the osteoclasts, which share the early part of their lineage with monocyte/macrophages. The mechanism by which cells are precisely targeted to areas requiring resorption is poorly understood.
The osteogenic effects of load engendered strains have been evident for some time, as has the resorption inducing effect of under-loading our skeletons, but the cell types responsible for orchestrating the targeted function of the effector cells is still in question. It is generally considered that the osteocyte is the most likely candidate for this role due to their distribution throughout the bone matrix, their responsiveness to strain and their existence as part of a syncitial network. However, until recently, it has not been possible to identify osteocyte-specific behaviour that is related spatially to the damage / strain environment and is associated with localised remodelling activity.
Some years ago we noted that osteocytes death by apoptosis is over represented in bone tissue which is for a variety of reasons undergoing rapid remodelling. We proposed that the marked apoptosis of osteocytes observed in women and female rats subjected to acute estrogen withdrawal provides a targeting mechanism for inducing the well-documented removal of bone by osteoclasts under osteoporotic conditions. Since then a number of reports have confirmed and extended our findings and a relationship between load engendered, targeted osteoclast activity and osteocyte apoptosis has been noted.
These observations raised the exciting possibility that the targeted removal of bone which is under loaded or contains microcracks might also involve the apoptotic death of the osteocyte. I shall discuss work undertaken which would strengthen this possibility. We will consider the possible mechanisms by which controlled cell death might contribute to the signals for bone removal and repair in the light of work involving cells in other tissue systems.
ROLE OF BIOCHEMICAL MARKERS IN MONITORING ADHERENCE TO TREATMENT
University of Sheffield, Sheffield, UK
Bone turnover markers may be useful in monitoring the response to treatment with antiresorptive therapy. The changes in bone resorption markers are large and occur within the first few weeks of treatment. The size of the change in the individual is related to the reduction in the risk of fracture. The goal of therapy is to decrease bone resorption into the lower half of the reference range for premenopausal women by an amount greater than the least significant change. Not all biochemical markers are the same, for bone resorption markers, the telopeptide assays (NTX, CTX) appear to show the best 'signal to noise' ratio. These markers show large day-to-day variability and this can be minimised by obtaining more than one baseline sample and taking the average. The aims of monitoring treatment with bone turnover markers is to identify patients who do not respond to therapy and to encourage patients to comply with long- term medication. Long-term compliance with medications for osteoporosis is low, and could be improved by knowledge that the treatment is effect after just three months.
ROLE OF BONE DENSITOMETRY AND QUANTITATIVE ULTRASOUND TECHNIQUES IN MONITORING TREATMENT RESPONSE
C. C. Glüer
Medical Physics, Department of Diagnostic Radiology, University Hospital Schleswig-Holstein , Kiel, Germany
The slow response of bone to treatment requires monitoring over several years to establish efficacy of treatment. The small magnitude of changes in bone mineral density (BMD) requires techniques with small long term precision errors. In order to assess the value of radiological techniques such as bone densitometry or Quantitative Ultrasound (QUS) two aspects need to be considered: (1) the technique must offer sufficient longitudinal sensitivity (a function of the ratio of response rate and precision error) to allow decision making after within 1-2 years and (2) the changes measured should reflect changes in bone strength, i.e. probability of fracture. Both of these aspects may depend on the type of treatment selected, e.g. they can be different for bone forming versus anti-resorptive medications and need to be interpreted differently when interpreting bone changes due to exercise.
In view of the increasing variety of different treatment agents it has become apparent that change in bone strength cannot be fully understood if only change in areal density is considered, The association of change in areal BMD with change in fracture rate is modest. Other skeletal aspects including micro-structural changes or alterations in bone tissue quality need to be considered and measured in addition. However, techniques to perform this in vivo are only beginning to emerge. Regarding changes in BMD, a large amount of data on longitudinal sensitivity exists for Dual X- ray Absorptiometry, much less for other densitometry methods such as (peripheral) Quantitative Computed Tomography. Similarly the information on QUS response to treatment is limited and the association of change in QUS variables with change in fracture rate is not known.
We will review existing data for the various techniques and medications. Based on these data and in view of competing techniques such as bone turnover markers the role of bone densitometry and QUS techniques in monitoring treatment response will be outlined.
DO WE NEED TO MONITOR ANTI-OSTEOPOROSIS TREATMENT ANYWAY?
J. E. Compston
Dept of Medicine, University of Cambridge School of Clinical medicine, Cambridge UK
Wider recognition of osteoporosis and development of effective interventions have resulted in a large increase in the number of individuals undergoing treatment. Bone densitometry and/or measurement of biochemical markers of bone turnover are often recommended as means of monitoring the response to treatment, creating a huge industry in the private sector and overburdening osteoporosis services in state-run systems. For monitoring to be justifiable, it must be shown to be cost-effective. This requires randomized controlled trial evidence that monitoring reduces fracture rate by a clinically significant amount and that these benefits outweigh the financial costs involved.
Monitoring is often justified on the basis that it is comforting for the patient and her/his doctor to know that the treatment is working. However, this requires a considerable degree of subterfuge and obfuscation on the part of the physician, since in most cases he/she will not know whether the treatment is working but will be anxious to reassure the patient and encourage their compliance. Far from being reassuring, monitoring may cause anxiety, since to the patient it implies that treatment may fail. Furthermore, although monitoring might improve compliance in a small number of individuals, it may have the reverse effect if an individual is wrongly accused of non-compliance, Finally, since evidence points to a global non-compliance rate of >50% with long-term osteoporosis therapies in clinical practice, the majority of monitoring tests will be performed in patients who are not taking any treatment. When reassured that they are responding, their motivation to take therapy will be further reduced.
A problem central to the issue of monitoring is that there is no agreed definition of a non-responder; indeed, it is uncertain whether true non-responders exist at all. Individuals who fracture whilst on treatment are often described as non-responders but might have sustained more fractures had they not been on treatment. Changes in bone mineral density or biochemical markers only explain a relatively small part of the variance in fracture reduction and cannot be reliably used as indicators of response in individual cases. In the absence of evidence that monitoring significantly improves clinical outcome, there is no justification for its use and the substantial resources dedicated to this practice should be reallocated to more important priorities.
MODELLING HUMAN GENETIC BONE DISEASES USING TRANSGENIC AND OTHER APPROACHES
Charité, Berlin, Germany
Inherited diseases of the skeleton are clinically and genetically heterogeneous. Some affect primarily the overall pattern of the skeleton, others, in contrast, result in growth disturbances or in abnormal bone homeostasis. The identification of specific gene defects in these conditions has unraveled some of the basic principles of bone biology and has opened new avenues for understanding and possibly treating inherited diseases of bone. However, the identification of a gene together with the disease causing mutation is not sufficient to explain the pathobiology behind the condition. Models are needed to characterize the function of a gene within a given biological process and to understand its role in the network of molecular pathways. Mutations have to be modeled in order to understand their consequences. Some mutations result in a loss of function (hypomorph/amorph) whereas other in a gain of function (hypermorph) or even in a new function (neomorph). Examples will be presented how such effects can be studied in animal models. Cleidocranial dysplasia (CCD), for example, is caused by heterozygous loss of function mutations in Runx2, a transcription factor expressed during bone development. Inactivation of one Runx2 allele in the mouse results in a CCD-like phenotype whereas inactivation of both alleles results in a complete loss of osteoblasts and, consequently, bone. This proves that Runx2 is necessary for bone formation. However, overexpression of Runx2 in chick embryos using a retroviral system (RCAS) does not result in extra bone formation indicating that Runx2 is not sufficient to induce bone formation. Rather, Runx2 appears to have an important role in maintaining the early chondrocyte phenotype as indicated by the formation of extra cartilage in infected wings. Furthermore, Runx2 promotes chondrocyte differentiation. The chondrogenic effects of Runx2 can be blocked by expressing a dominant negative form of Runx2 selectively in chondrocytes using a transgenic approach with the collagen type II promotor. Thus, the combination of gene inactivation together with specific or global overexpression in model systems allows the thorough analysis of gene function.
GENETIC DETERMINANTS OF OSTEOPOROSIS
A. G. Uitterlinden
Departments of Internal Medicine, Clinical Chemistry and Epidemiology & Biostatistics Erasmus MC, Rotterdam, The Netherlands
Like other complex diseases and traits, osteoporosis has strong genetic influences. DNA analysis is expected to be a powerful tool in diagnostics of those at risk of disease, in identifying potential drug targets, and in predicting response-to-treatment, and the field of osteoporosis is no exception to those expectations.Therefore intense academic and commercial efforts are ongoing to identify the responsible gene variants.
Linkage analysis by 'genome searches', which should identify chromosomal regions containing disease gene(s), has had limited success sofar to identify complex trait genes. More promising is the analysis of human candidate genes with a known involvement in the biological pathway of interest. Such genes can be identifed by expression analysis of target cells, linkage analysis of monogenic bone diseases, and also by mouse models of the disease. Sofar, candidate genes have come from classic bone pathways, involving estrogen and vitamin D, and from bone matrix molecules such as collagen type I. However, mouse models and analysis of human Mendelian bone disease has also identified novel or hitherto unsuspected 'bone proteins', such as leptin, SOST, and LRP5. Subsequently, genetic-epidemiological analyses of polymorphisms in these candidate genes are performed in large population studies to determine the contribution to phenotypic endpoints of interest, such as risk of fracture.
Candidate gene studies, however, have had their share of scepticism because effects were small, functionality of polymorphisms was not always known, and associations have not always been reproducable across populations. However, while probably hundreds of genes each play a role in bone metabolism, only very few of these have been properly scrutinized for their genetic contribution to osteoporosis including meta-analyses to estimate the true effect size and functional studies to establish mechanism of action.
Importantly, we now also know that a single gene usually carries several polymorphisms which influence its activity and, thus, comprehensive analysis of all genetic variation in a gene (haplotype analysis) is important. In addition, several gene products within a certain biological pathway interact with each other and bio- activities might be modified by environmental factors. Once individual gene effects have been categorized, evaluating these interaction effects will be the major challenge in determinng the role of genetic factors in osteoporosis.
ROLE OF THE PTH/PTHRP RECEPTOR IN BONE GROWTH AND REMODELLING
MGH-Harvard Medical School, Boston MA, USA
The parathyroid hormone/parathyroid hormone-related protein receptor (PTH/PTHrP receptor) mediates both the endocrine actions of parathyroid hormone (PTH) and the auto/paracrine actions of parathyroid hormone-related protein (PTHrP). This places the PTH/PTHRP receptor as a central regulator of both mineral ion homeostasis and bone development. Jansen's metaphyseal chondrodysplasia, a rare autosomal dominant disorder characterized by short-limbed dwarfism and hypercalcemia, is caused by mutant, constitutively active PTH/PTHrP receptors. Consistent with the observation in patients, a transgenic model in which a mutant Jansen PTH/PTHrP receptor was targeted to the growth plate (H223R-Col II) showed delayed mineralization and decelerated chondrocyte maturation in skeletal segments that are formed by endochondral bone development. This finding indicates that the PTH/PTHrP receptor is the main mediator of PTHrP action in the developing endochondral bone. Targeted overexpression of the same mutant Jansen receptor in vivo in cells of the osteoblast lineage (H223R-Col I) demonstrated that stimulation of the PTH/PTHrP receptor in osteoblasts is responsible for both the bone forming and the bone resorbing actions of PTH. Further analysis of H223R-Col I transgenic model also revealed that activation of the PTH/PTH receptor has differential effects on cortical and trabecular bone, and that collagenase activity could be involved in the differential effects of this receptor on discrete bone compartments. Clonogenic marrow stromal cells were isolated from normal and mutant mice, expanded in culture, and transplanted subcutaneously into immunocompromised mice. Normal stromal cells generated a complete ossicle (bone and marrow). In contrast, no marrow was detectable at 4-8 weeks post-transplantation of mutant stromal cells. Interestingly, however, H223R- Col I transgenic had increased hematopoietic stem cell numbers. The hematopoietic stem cell effect was mediated by stroma from these animals, and it could be recapitulated by exogenous application of PTH to wild type stroma containing hematopoietic stem cell cultures. Taken together, these findings strongly suggest that the PTH/PTHrP receptor not only modulates bone growth, but has also a complex and not yet fully elucidated action on the hematopoietic compartment
ROLE OF PTH IN THE PATHOGENESIS AND TREATMENT OF OSTEOPOROSIS
Helen Hayes Hospital and Clinical Medicine, Columbia University NY, USA
1-34hPTH is now approved for treatment of osteoporosis in both men and women at high risk of fracture, in the USA. PTH represents a new paradigm for osteoporosis treatment, since it is the first drug demonstrated to stimulate bone formation and increase the mass of bone tissue in the skeleton. The result is a decreased risk of vertebral and non-vertebral fractures. PTH effects this by two mechanisms. First there is a rapid stimulation of bone formation, within the first few days of initiation of treatment. This is followed by a stimulation of remodeling activity. Both effects are transient and despite continued administration of the drug, bone formation and remodeling activity return to baseline within 2 years. Clinical trials were discontinued when osteosarcomata were found in a rodent toxicology study, although the relevance of this for human use of 1-34hPTH remains unknown. The consequence of these 2 different findings is the recommendation of PTH use for not more than 2 years. One fascinating aspect of PTH action is that bone formation appears to be stimulated on all surfaces of bone, including, at least at some sites, the periosteum. The result is a modest but significant increase in bone size, which may have marked effects on strength. While cellular and molecular mechanisms are still being worked out, one hypothesis, delayed osteoblast apoptosis resembles the finding in mice with the so- called high bone mass gene.
ANTI-TNF THERAPY FOR RHEUMATOID ARTHRITIS: PROSPECTS FOR THE FUTURE
M. Feldmann*, F. M. Brennan, B. M. J. Foxwell, E. Paleolog, R. N. Maini
Faculty of Medicine, Imperial College of Science, Technology and Medicine, UK
Analysis of cytokine disregulation in samples of human synovium provided the major rationale for blocking TNF in rheumatoid arthritis. Clinical trials established this hypothesis using a number of different TNF inhibitory 'biologicals' which are now licensed for human use - first a chimeric antibody, infliximab, then a TNF-receptor fusion protein, etanercept, and most recently a human antibody, adalimumab. All of these induce marked clinical benefit in a majority of patients resistant to other anti- rheumatic therapy. Analysis of the mechanism of action revealed that many processes were affected, with production of many cytokines, chemokines, adhesion molecules down-regulated.
Perhaps the major single mechanism of clinical benefit is the reduction in influx of inflammatory cells into joints. RA is not the only chronic inflammatory disease that benefits from TNF blockade. Crohn's diseases, psoriatic arthritis, ankylosing spondylitis, psoriasis have reported convincing clinical trial data and await approval, many others have successful results. It is likely that the widespread benefit is due to the impact on cell recruitment. In terms of bone biology it is striking that TNF blockade can arrest joint damage in about half the patients, improves others, and in about a third there is improvement in joint images which suggests healing of bones and cartilage damage. That is a very exciting prospect which while needing confirmation in other clinical trials, suggests that some degree of joint repair is possible in humans provided the inflammatory process is sufficiently controlled.
REGULATION OF BONE CELL FUNCTION BY PRO- INFLAMMATORY CYTOKINES
S. R. Goldring
Harvard Medical School, Boston, Massachusetts, USA
Cytokines play a critical regulatory role in the initiation, perpetuation and destructive activities of many inflammatory disorders. Rheumatoid arthritis (RA) represents a paradigm for investigating the role of these pro-inflammatory cytokines on skeletal remodeling. In this condition, proliferation of the synovial lining of diarthrodial joints is accompanied by progressive focal peri-articular bone loss, manifest radiographically by the development of focal joint erosions. This local joint pathology is associated with systemic bone loss and an increased risk of fracture. Histopathological analysis of joint tissues from patients with RA using immunostaining and in situ hybridization techniques indicate that osteoclasts are associated with the development of focal joint erosions. Animal models of inflammatory arthritis, including adjuvant arthritis (Kong et al. Nature 1999; 402:304), serum transfer arthritis (Pettit et al. Am J Pathol 2001;159:1689) and TNF- transgenic mice with spontaneous arthritis (Redlich et al. Arthritis Rheum 2002; 46:785), confirm that osteoclasts are the principal cell type responsible for the pathogenesis of focal bone erosions. Additional studies have helped to identify the cytokines and inflammatory mediators that are involved in the recruitment and activation of bone resorbing cells associated with inflammatory arthritis. Tumor necrosis factor alpha, interleukin-1, receptor activator of NF-kb ligand (RANKL) and a number of other products of activated T cells and synovial fibroblasts are among the factors implicated in the increased focal articular and systemic bone loss. Pro- inflammatory cytokines represent rational therapeutic targets for specifically inhibiting or slowing the progressive bone loss associated with RA and related inflammatory disorders.
MECHANISMS OF BONE LOSS IN ASEPTIC PROSTHETIC LOOSENING
E. M. Schwarz*, D. Campbell, S. Totterman, A. Boyd, R. J. O'Keefe, R. J. Looney
University of Rochester, New York, USA
Although total hip replacement (THR) is amongst the most successful and beneficial medical procedures to date, long-term outcomes continue to suffer from aseptic loosening secondary to periprosthetic osteolysis. Extensive research over the last two decades has elucidated a central mechanism for osteolysis in which wear debris generated from the implant stimulates inflammatory cells to promote osteoclastogenesis and bone resorption. The cytokine tumor necrosis factor alpha (TNF) has been demonstrated to be central to this process and is considered to be a leading target for intervention. Unfortunately, even though FDA approved TNF antagonists are available (etanercept), currently there are no reliable outcome measures that can be used to evaluate the efficacy of a drug to prevent periprosthetic osteolysis. To the end of developing an effective outcome measure, we evaluated the progression of lesion size in 20 patients with established peri-acetabular osteolysis (mean = 29.99 cm3, range = 2.9-92.7 cm3) of an uncemented primary THR over one year, using a novel volumetric computer tomography (3D-CT) technique. We also evaluated polyethylene wear, urine N-telopeptides and functional assessments (WOMAC, SF-36 and Harris Hip Score) for comparison. At the time of entry into the study baseline CT scans were obtained and the patients were randomized to etanercept (25mg s.q., twice/week) and placebo in a double-blinded fashion. CT scans, urine and functional assessments were also obtained at 6 and 12 months. No serious adverse drug related events were reported, but one patient had to have revision surgery before completion of the study due to aseptic loosening. No remarkable differences between the groups were observed. However, the study was not powered to see significant drug effects. 3D-CT data from the 19 patients was used to determine the mean increase in lesion size over 48-weeks, which was 3.19 cm3 (p<0.0013). Analysis of the urine N-telopeptides and functional assessment data failed to identify a significant correlation with wear or osteolysis. In conclusion, volumetric CT was able to measure progression of osteolysis over the course of a year, thus providing a technology that could be used in therapeutic trials. Using the data from this pilot we provide a model power calculation for such a trial.
N. J. Bishop
Academic Unit of Child Health, University of Sheffield, Children's Hospital, Sheffield, S10 2TH, UK
Osteogenesis imperfecta (OI) is the commonest inherited form of bone disease causing osteoporosis. Affected individuals suffer recurrent fractures with resulting pain, deformity and disability. The clinical phenotype varies very widely. Mildly affected individuals have little or no bone deformity or loss of height. Severely affected infants may die at birth or in the first year of life. The Sillence classification defines four forms of OI, but three new types have been described over the last three years. Defining the molecular pathophysiology and its relation to clinical phenotype remains a challenge in all forms of the disease.
The management of OI is multidisciplinary. Recent attention focussed on anti- resorptive therapy using bisphosphonates, with the majority of reported studies being observational in nature. Increases in bone mass, cortical thickness and the height of previously crush-fractured vertebrae, together with the relief of chronic bone pain are consistently reported for children treated with these agents. Multidisciplinary management is required in the majority of moderately to severely affected individuals. Surgery, occupational and physiotherapy remain important aspects of treatment.
This review of OI will provide information on the current system of classification, response to therapy with bisphosphonates and the pathophysiology of the disease at a cellular and molecular level.
AUTOSOMAL DOMINAT OSTEOPETROSIS (ADO). FROM THE PATIENT TO THE MUTATION
Section of Endocrinoloy, Natl. University Hospital. Oslo, Norway
Imbalance between the bone resorptive and formative process can lead to different sclerotic skeletal dysplasias characterised by increased bone mass. Among these, osteopetrosis is caused by defective bone resorption leading to a diffuse, symmetrical osteosclerosis without major modeling defects. ADO is recognised as a rather benign condition, where the diagnosis often is reached by change, however heterogeneity between types and even within families is well known.
Based on a systematic search for the condition, two different types (Type 1 and 2) were described in the County of Funen, Denmark. Both types showed increasing sclerosis with age and both were seen in childhood. Further studies revealed distinct differences between the types at the clinical, biochemical, histomorphometric, and electron microscopic level. ADO 1 is a typical osteoclastopenic disorder, as seen in several murine models characterised by maturation defects in development of the osteoclastic lineage. Interestingly, the disease seems to have increased trabecular bone strength and decreased fracture rate. Linkage analyses mapped this type to chromosome 11q12-13, and further mutation analyses disclosed a missense mutation in exon 4 of the LDL receptor-related protein 5 (LRP5). Mutations in this gene have been associated among others with the High Bone Mass (HBM) phenotype. However, ADO 1 and HBM differs significantly at the clinical and biochemical level and further studies may reveal the exact mechanism by which the LRP5 gene regulates bone mass.
In ADO 2 (Albers-Schönberg disease), numerous huge multinucleated osteoclasts are found at the histological level with a TRAP-positive substance covering adjacent bone trabecular surfaces. This type resembles in many ways the more severe juvenile forms of osteopetrosis, associated with a mutation in the gene encoding for the chloride channel ClCN7. ADO 2 has recently been associated with a missense mutation in the same gene. Chloride channels function as dimers, thus the identified mutation seems to exert a dominant negative effect.
In conclusion, ADO is a heterogeneous group of rather benign osteosclerotic bone dysplasias where the underlying mutations recently have been found. ADO 1 resembles murine mutations with well- known maturation defects in the osteoclastic lineage and is now associated with the LRP5 gene. The defective osteoclast function in ADO 2 is related to the acidification process by mutation in the ClCN 7-gene.
M. P. Whyte
St Louis MO, USA
Skeletal dysplasias are generalized developmental malformations that cause symmetrical bony disturbances. Although individually rare, cumulative prevalence is about 1 per 4,000 - 10,000. Most are heritable and remarkable recent progress is identifying their genetic basis and thereby elucidating their pathogenesis. Consequently, dysplasias are moving increasingly from the domain of dysmorphologists to "metabolic bone disease." Endocrinologists are also consulted for chondrodysplasias because growth plate disturbances can cause metaphyseal irregularity mimicking rachitic disease or vertebral distortion suggesting compression fracture. Although more than 100 seemingly distinct phenotypes have been reported, molecular studies are rapidly improving classification. Indeed, by the end of the 20th Century, a large proportion were known to result from mutations in a small number of genes ("allelism"). Remarkably, 10 genetic loci actually account for the majority of chondrodysplasias. Hence, "consolidation" occurred. For others, separation became possible for clinically and radiographically similar conditions. Graded severity for skeletal dysplasias can involve a single gene (resembling osteogenesis imperfecta where numerous mutations disturb the two genes that encode type I collagen). Aberrations in endochondral bone formation often represent a fundamental pathogenetic defect. The genes are essential and nonredundant for the production of cartilage. Pathogenesis-based classification now places chondrodysplasias, by and large, in two principal groups: (1) gene defects encoding type II collagen (and sometimes other matrix proteins), and (2) craniosynostoses involving FGFR3 mutations. Cartilage matrix protein gene mutations compromise chains of types II, IX, X and IX collagen, matrilin and COMP (typically acting in a dominant negative fashion). Spondyloepiphyseal dysplasias, multiple epiphyseal dysplasias, Stickler dysplasia, Schmid metaphyseal chondrodysplasia, and pseudoachondroplasia represent this category. Mutations that activate receptors (whose downstream signals negatively regulate physeal chondrocyte proliferation and differentiation) include FGF3 defects causing achondroplasia, and PTH/PTHRP receptor mutations causing Jansen syndrome. A third major category disturbs cellular sulfate uptake required for synthesis of cartilage proteoglycans. Nevertheless, diagnosis of dysplasias still is achieved primarily by radiographic "pattern recognition." However, several laboratories [accessed at "GeneTests" on the web (www.genetests.org)] provide either commercial or research laboratory mutation analyses to help with diagnosis.
PATHOGENESIS OF HYPOPHOSPHATAEMIC RICKETS
M. J. Econs
Indianapolis IN, USA
Abstract not supplied
RECENT ADVANCES IN SCLEROSING BONE DYSPLASIAS
W. Van Hul
Dept. Med Genetics, University of Antwerp, Belgium
The group of sclerosing bone dysplasias has longtime been poorly understood at the molecular level. The combination of increased availability of human genome sequence data and newly developed tools and technologies have expedited position cloning efforts aiming at the identification of disease causing genes. This has resulted in the recent identification of a large number of disease genes. The study of the extended group of conditions with an increased bone density has, in some cases, lead to the identification of previously unknown genes involved in bone homeostasis. In other cases, by proving the involvement of previously known genes in sclerosing phenotypes, further insights have been gained on the precise functioning of these genes and the pathways they participate in.
Long time after carbonic anhydrase II-deficiency was proven to lead to an osteopetrotic phenotype, the study of other forms of osteopetrosis has resulted in the characterization of other important partners in the bone resorption process (the Vacuolar H(+)-ATPase and ClCN7). Mutations in coupling factors between bone formation and bone resorption (including TGFb1, OPG and RANK) are also found in conditions with an increased bone density. Finally, the previously unknown SOST gene and the LRP5 gene turned out to be important regulators of the bone formation process by acting on BMP- and Wnt signaling respectively.
In conclusion, positional cloning efforts on monogenic conditions with an abnormal bone density have contributed to the present understanding of bone homeostasis. Furthermore, currently performed association studies with natural variants within the genes identified, will most likely illustrate a role in the pathogenesis of osteoporosis for at least some of these genes.
BASIC RESEARCH STRATEGIES TO IDENTIFY NEW OSTEOPOROSIS TREATMENTS
Yale University School of Medicine, New Haven, CT, USA
ProSkelia Pharmaceuticals, Romainville, France
Advances in the biology of bone cells and of the regulation of bone remodeling have lead to innovative approaches to the treatment of osteoporosis. Osteoporosis drug discovery follows 3 approaches: 1) Pathophysiological, attempting to correct the mechanisms that lead to altered bone remodeling, 2) Anti-resorptives, targeting the osteoclast lineage and 3) Anabolics, targeting the osteoblast lineage. In the pathophysiology of osteoporosis, three major findings have been the discovery of a second receptor for estradiol (ER beta), the finding that the ERs may affect bone cell apoptosis through non-genomic mechanisms and via the AR as well, and the validation of the concept of the SERMs, raising the possibility to selectively target bone and to develop similar compounds for the AR or the PR. In the field of antiresorptives, and beyond bisphosphonates and calcitonin, several recent findings have lead to programs in drug discovery. The main targets are : the vitronectin receptor, cathepsin K, c-Src, the vacuolar proton pump, and the pathway of RANK Ligand and its receptor RANK. In addition, the mevalonate pathway has been demonstrated to be the target for bisphosphonates, leading to a new drug discovery approach. The therapies of the future will also involve bone anabolics. First and foremost, the anabolic potential of PTH given intermittently has been firmly established in human studies, validating the concept of anabolics. An alternative to the use of PTH is the regulation of its secretion by targeting the calcium receptors in the parathyroid gland, antagonising the binding of Ca and thereby increasing PTH levels (calcilytics). Other major findings in this area are the identification of Cbfa-1, the central hypothalamic regulation of bone formation via the adrenergic receptors, and the recent identification of LRP5, a receptor involved in bone mass regulation in humans. Finally, most companies are currently engaged in genetic, genomic and gene expression profiling studies to identify novel genes involved in bone formation that can then be used as novel targets for drug discovery.
NEW TREATMENTS, NEW TRIALS, NEW PARADIGMS FOR PREVENTING FRACTURES
S. R. Cummings
SF Coordinating Center, Pacific Research Institute, San Francisco, CA, USA
Clinical trials have taught us that antiresorptive drugs reduce fracture risk, especially in women with osteoporosis. Additionally, they have taught us that these drugs reduce fracture risk more than can be expected from improvements in BMD. These drugs work quickly but their long-term effects are uncertain. Even long-term treatment with antiresorptives is unlikely to reduce risk of fractures to normal or acceptable levels in patients with sever osteoporosis. One trial of PTH suggests that bone forming agents could change the paradigm of anti-fracture treatment . An effective bone-forming agent could change current paradigm of practice (treatment to
maintain moderately reduced risk) to a new paradigm of treating to achieve goals of nornal bone strength and normal fracture risk. I propose that we need trials of the old vs. new paradigm of prevention.
Despite proving several treatments to effectively reduce fracture risk, only a small proportion of people at high risk of fracture are receiving effective treatment. We need trials of new approaches to delivering care outside of specialist-based osteoporosis practices. We need to find and test new ways to increased the identification of high risk women. We need trials of the effectiveness of inexpensive treatments. And we need to work with companies and governments to make effective treatments affordable world-wide to those who need them most.