a decade of bisphosphonate bone complications: what it has ... · oral & craniofacial tissue...

309Oral & Craniofacial Tissue Engineering

A Decade of Bisphosphonate Bone Complications: What It Has Taught Us About Bone PhysiologyRobert E. Marx, DDS1

While the AIDS epidemic of the 1980s taught the medical and dental professions much about immune cells and the immune system’s cellular relationships, the bisphosphonate-induced osteonecrosis epidemic of the past decade has taught these same professions much about bone turnover, bone cell cross talk, the response and functional relationship of bone cells to loading, and drug effects on cellular dynamic relationships. The present article explores the literature as well as both evidence- and experience-based data to discuss known bone pathologies and physiologic mechanisms as well as uncover new findings: (1) bone remodeling is the mechanism by which bone adapts to loading stresses, termed either bone modeling or Wolff’s law, and it is also the mechanism for bone renewal; (2) osteoclastic bone resorption triggers bone renewal at a rate of about 0.7%/day by its release of growth factors; (3) bisphosphonates prevent the renewal of old and injured bone, thus making it brittle and more likely to fracture over time; (4) bisphosphonates have a half-life in bone of 11 years because of their irreversible binding to bone via their central carbon atom; (5) when administered intravenously, bisphosphonate loads bone and accumulates in bone 142.8 times faster than when administered orally; (6) osteoclastic resorption of bisphosphonate-loaded bone results in osteoclast death in which the cell bursts, releasing the bisphosphonate molecules to reenter the local bone or bone marrow in a re-dosing effect; (7) endosteal osteoblasts are dependent on the osteoclastic resorption/growth factor release/new bone formation mechanism of bone renewal, whereas periosteal osteoblasts are not; and (8) it is likely that endosteal osteoblasts and periosteal osteoblasts have different cell membrane receptors and arise from separate embryologic niches. Oral CraniOfaC Tissue eng 2012;2:309–320. doi: 10.11607/octe.0058

Key words: bisphosphonates, bisphosphonate-induced osteonecrosis of the jaw, bone modeling, bone remodeling, bone resorption, osteoblasts, osteoclasts, osteoporosis

The emergence of the AIDS epidemic in 1980 caught the fields of medical science, immunol-ogy, and medical and dental education unaware

and unprepared.1 At that time, the immune system was simply understood as a two-class system consisting of phagocytes (eg, neutrophils, macrophages, etc) and secretory cells (eg, lymphocytes and plasma cells). The secretory cells were further categorized as T cells from thymus dependency–producing lymphokines and

B cells from Peyer’s patches, lymph nodes, and other unknown sites related to the bursa of Fabricius in birds (hence the name B cells). It was taught that B cells produce antibodies, and they were further identified as producing various classes of antibodies (eg, im-munoglobulins M, G, A, E). However, the identifica-tion of a virus with a specific predilection to a certain component of the immune system stimulated a frenetic search and study of immune cells and how they work, leading to the recognition of T-helper cells, the specific CD4 cell, chemokine receptor 5 and CXC chemokine receptor 4, natural killer T cells, antigen-presenting cells, etc.2 Knowledge of immune cells and the im-mune system grew rapidly within a short time and led to focused treatments for human immuno deficiency vi-rus infection/AIDS, as well as an improved knowledge and treatment of other immune-based pathologies.3,4

1 Professor of Surgery and Chief, Division of Oral and Maxillofacial Surgery, University of Miami Miller School of Medicine, Miami, Florida, USA.

Correspondence to: Dr Robert E. Marx, 9380 SW 150th Street, Suite 190, Miami, FL 33176, USA. Email: [email protected]

©2012 by Quintessence Publishing Co Inc.

© 2013 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER. © 2013 BY QUINTESSENCE PUBLISHING CO, INC. PRINTING OF THIS DOCUMENT IS RESTRICTED TO PERSONAL USE ONLY. NO PART MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM WITHOUT WRITTEN PERMISSION FROM THE PUBLISHER.

Marx

310 Volume 2, Number 4, 2012

Now, in a similar sense, bisphosphonates have caused osteonecrosis of the jaws (ONJ), spawning a renewed interest and intense research in the last 10 years into how bone remodels and renews itself as well as how it ages, becomes injured, and fractures.

Although cases of ONJ occurred in drug company–sponsored trials in the 1990s, they were dismissed and not reported.5–7 Therefore, the initial identification of ONJ caused by bisphosphonates to the scientific community occurred in a 2002 textbook.8 This was followed by four publications9–12 in the latter half of 2003 and first half of 2004 and has since been fol-lowed by more than 1,300 publications, which have reported more than 15,000 cases, confirming the epidemic predicted in 2003.9 However, the majority of these publications have focused on prevalence, the clinical entity of exposed bone, and prevention and treatment recommendations.13–17 Few have focused on the actual bone cells involved and what their newly discovered expanded role is in ONJ or even in normal skeletal homeostasis.

SPECIFIC BONE SCIENCE LESSONS LEARNED FROM BISPHOSPHONATES

Bone Remodeling Rates and Cellular Life Spans The fundamental effect of bisphosphonates is the impairment and apoptosis (cellular death) of not only adult osteoclasts but osteoclast precursors in the bone marrow (Figs 1a and 1b).18,19 Osteoclasts arise

from hematopoietic stem cells in bone marrow and mature as mononuclear cells through the influence of transcription factors c-Fos and PU.1 (Fig 2).20 They are secreted into the general circulation, still as a mononuclear cell, which then fuses into a multinu-cleated cell through the effects of interleukin-1 and interleukin-6 as well as receptor activator of nuclear factor kappa-B ligand (RANKL). However, this mature osteoclast is a terminal cell with a life span of only 14 days.21 This is in contrast to osteoblasts, which be-come osteocytes, at which point they have a life span of about 180 days. Osteocytes are not mere quiescent cells but are tasked with the responsibility of acting as mechanoreceptors through their pseudopodia within their canaliculi as they traverse the mineral matrix. Osteocytes sense tension and pressure to regulate the resorption of their own mineral matrix through the se-cretion of osteoprotegerin (OP).22 That is, osteocytes sense tension and pressure, which in the jaws is high-est in the lamina dura about the teeth, the edentulous alveolar ridge in denture wearers, and the posterior lingual cortex in the region of the mandibular molars—the areas of highest incidence of ONJ. This is a result of the balance between the proresorption effects of RANKL and the inhibition of remodeling provided by OP (Fig 3). In areas of greater stress, the osteoblasts/osteocytes reduce their secretion of OP and increase their secretion of RANKL so that the area can resorb and re-form to accommodate this greater stress, a classic adaptive response.23,24 If the now-stimulated osteoclast attempts to resorb this stressed bone that

Hematopoieticstem cell

Committed osteoclastprecursor (PU.1+)

Macrophage colonystimulating factor RANK ligand

RANK ligandIL-1

Fusion of multiplemononuclear osteoclasts

via RANK ligandIL-1 and IL-6

Quiescentmultinucleated

osteoclast

Activeosteoclast

Differentiatedmononucleated

osteoclast cFos+and RANK+

Fig 1a Bisphosphonate-poisoned osteo-clast with ballooned cytoplasm and more nuclei than usual, which are small and dis-rupted.

Fig 1b Two osteoclasts. One shows dispersed chromatin and ballooned cy-toplasm, the other is a “ghost cell” of shrunken cytoplasm; both were caused by bisphosphonate poisoning.

Fig 2 Maturation process of an osteoclast from a mononuclear cell bone marrow precursor to a mature cell involved in bone remodeling.

a b


Marx


contains a bisphosphonate, it begins to do so but dies off before completing the resorption, thus disrupting the resorption-renewal cycle. The osteocytes then succumb to the stress and die before their 180-day life span, or they live out their life span but are not replaced. The bone may first become overmineralized (sclerotic) (Fig 4), but then it often becomes necrotic and exposed (Fig 5).

How do osteoclasts know when to resorb bone during normal skeletal remodeling? Osteoclasts self- secrete RANKL as an autocrine function and will resorb any bone that is vulnerable to be resorbed.23–25 This vulnerability is determined by the age and health of the bone itself. That is, young undamaged osteocytes se-crete sufficient amounts of OP to inhibit osteo clastic

bone resorption. As osteocytes age, become fewer in number, or are injured, however, their OP secretion drops, allowing osteoclasts to overcome this inhibi-tion and resorb the mineral matrix.25 Therefore, osteo-clasts are able to recognize not just necrotic bone and sequester it, but they recognize aged and injured bone as well, which keeps the skeleton in a dynamic renewal of young elastic bone capable of responding to the stresses placed upon it (Fig 6).

The Coupling of Bone Resorption and Bone Remodeling (Renewal)The effects of bisphosphonates on the jaws help to clarify the too often interchangeable and incorrectly used terms bone modeling, bone remodeling, and

RANK ligandosteoclast activation

RANK receptors

OPG inactivation of RANK receptor

BMP receptorsILG receptors

Preosteoblast/osteoblast

Differentiationand bone formation

Release of BMP,ILG1, ILG2

Osteoclast

Fig 3 The competitive inhibition and stimulation of osteoclasts by the paracrine effects from osteoblasts and by their own autocrine effects.

Fig 4 Sclerotic (overmineralized) bone caused by bisphospho-nate therapy.

Fig 5 Exposed necrotic bone caused by bisphosphonate therapy.


Marx


bone renewal. Bone modeling is the morphologic change in bone (eg, thickening or thinning) of a bone in response to adaptive stress. The lamina dura is a good example of this. Here, within a marrow space, is an unexpected rim of dense cortical bone induced by Sharpey fibers from the periodontal ligament arising from root cementum and inserting into it. This trans-mits occlusal forces to the bone, which responds by laying down denser bone that is capable of withstand-ing this stress, which becomes the lamina dura. When a tooth is extracted, the lamina dura is seen to fade away radiographically (Fig 7). Before tooth extraction, there is resorption and new bone apposition, with the net result of a gain in bone density because of dental occlusal loading. After tooth extraction, the resorp-tion followed by new apposition of bone results in a net loss of bone density to a degree equal to that of the surrounding trabecular bone in the marrow space because of the loss of occlusal loading stress. There-fore, the lamina dura remodels to the density of mar-row space trabecular bone and becomes indistinct on radiographs. This action of osteoclast-mediated bone resorption followed by new bone apposition is referred to as bone remodeling. Therefore, bone re-modeling is the mechanism by which bone modeling occurs. In a similar sense, old and dying bone that has lived out its average life span of 180 days undergoes renewal by bone resorption coupled with new bone apposition by this very same bone remodeling activity. Therefore, bone remodeling is also the mechanism by which bone renewal is achieved.

The example of the lamina dura represents a steady state of bone modeling. However, it is well known that both hyperocclusion and bisphosphonates inde-pendently cause a thickening of the lamina dura and together cause an even more prominent thickening of the lamina dura (Fig 8). Hyperocclusion transmits increased forces to the lamina dura through the peri-odontal ligament. The lamina dura thickens in response

up to a point beyond which it may break down, resulting in periodontal bone loss. Bisphosphonates prevent the resorption phase of lamina dura remodeling, resulting in hypermineralization of the lamina dura with old unre-modeled bone. If the bisphosphonate dose accumula-tion over time in the lamina dura becomes too great, the bone dies and becomes exposed. This is why the observed bone exposure of bisphosphonate-induced osteonecrosis always begins in the alveolar bone or over the remodeling surface of a torus. This is the case even when it is first seen as a spontaneous exposure of the lingual cortex in the mandibular molar region; this occurs because axial loading of the mandibular molars is actually exerted on the lingual cortex rather than the inferior border (Fig 9). It also explains why the exposed bone in this area is always above the mylohyoid ridge, which anatomically is alveolar bone (Fig 10).

The Biochemistry of BoneBone is type 1 collagen wrapped as a left-handed triple helix in which crystals of hydroxyapatite are in-terspersed. However, upon formation of the mineral matrix, osteoblasts insert trace quantities of the growth factors bone morphogenetic protein (BMP) and insu-linlike growth factors 1 and 2 (ILG-1 and ILG-2).25 When an osteoclast begins normal resorption of old or injured bone, it seals itself to the bone surface and establishes a closed compartment that will become the Howship lacuna. It then secretes protons (acid pH < 1) to dissolve the hydroxyapatite crystals and col-lagenase to break down the bone collagen. However, in doing so it liberates the acid-insoluble BMP, ILG-1, and ILG-2, which in turn causes a proliferation and differentiation of local resident stem cells and osteo-progenitor cells to re-form the bone with younger and more elastic tissue (Fig 11).26

This coupling of bone resorption followed by new bone formation (bone remodeling) is not only the ba-sis of bone renewal but the fundamental physiology of

Fig 6 Osteoclasts seen here are resorb-ing viable bone because of its age and micro damage.

Fig 7 A faded and less distinct lamina dura than the origi-nal is present months after extraction. This change is caused by remodeling; a less dense bone has been laid down in re-sponse to the absence of functional loading.


Marx


osteo penia and osteoporosis for which the oral bisphos-phonates alendronate, residronate, and ibandronate are prescribed (and more recently the intravenous [IV] form of zolendronate known as Reclast and the RANKL in-hibitor denosumab subcutaneously), ie, bone renewal throughout the entire skeleton as well as the focal areas of interest for osteopenia and osteoporosis. The lumbar spine and femoral head remodel with a net gain in bone mass until an individual’s early 20s.27 After the early to mid-20s, a slow but steady decline in bone mass takes place in both men and women. This is because less than 100% of the bone that is resorbed about every 180 days is replaced by newly forming bone. Over the years, this slight shortfall in bone renewal adds up to a significant loss of bone mass. It is heightened in wom-en after menopause because estrogen is one of the

requirements for osteoblast differentiation from bone marrow osteoprogenitor cells.28 Therefore, postmeno-pausal women undergo a more rapid and significant de-cline in bone mass (crudely measured as bone mineral density by the dual energy x-ray absorptiometry scan).29 Thus, the logical strategy but also the fallacy of treating osteopenia/osteoporosis with bisphosphonates comes to light.30–33

The Strategy and Fallacy of Bisphosphonates for Osteopenia/OsteoporosisBisphosphonates inhibit osteoclast-mediated bone resorption and renewal. Therefore, similar to their effect on the lamina dura, they retain existing bone; under normal function, this increases mineralization, resulting in an increase in bone mineral density and

Fig 11 Normal bone remodeling cycle. Bone is renewed by replacing old bone with new, more elastic bone.

Fig 8 Prominent hypermineralized lamina dura from bisphosphonate therapy, indicating the accu-mulation of older unremodeled bone.

Fig 9 Cone beam computed tomographic scan shows that the axial loading of mandibular mo-lars is on the lingual cortex.

Fig 10 Exposed bone of the mandibular lin gual cortex representing bisphosphonate- induced osteonecrosis.

Circulating mononuclear cells

Osteoclast precursors

Osteoblast precursors

BMPILG1ILG2

BMPILG1ILG2

Osteoblasts

Quiescent osteoclasts

Active osteoclasts

Osteoid OsteoidMatureregenerated bone

1 month 3 months 6 monthsTime


Marx


theoretically an increased resistance to osteoporosis-related fractures.28,31 However, preventing osteoclast-mediated bone resorption/renewal results only in the retention of old bone, which lives out its natural life span and becomes brittle and thus even more prone to fracture.32,34 This is why the more recent literature is reporting an increasing number of femur fractures actu-ally caused by alendronate rather than prevented by it (Fig 12)35–38 and why independent investigators have not been able to duplicate the prevention of osteo porosis fractures first published by the drug companies.39

Simply stated, osteoclasts are a normal cell in skeletal homeostasis. Removing them and their use-ful function by toxic drugs can be expected to result in, and has resulted in, the same complications seen in the genetic loss of osteoclast function in the dis-ease known as osteopetrosis,40 ie, exposed bone in the maxilla and/or mandible and leg fractures. With bisphosphonates, it is not a matter of genetic expres-sion but one of dose and accumulation in bone over time. This is why, in September 2011, the U.S. Food and Drug Administration (FDA) encouraged physicians to reassess osteopenia/osteoporosis patients treated with oral bisphosphonates after 3 years of treatment and discouraged treatment beyond 5 years.41

The Biochemistry of Bisphosphonates Related to Bone Bisphosphonates are a synthetic of pyrophosphates, which are diphosphosphonates,42 as exemplified by the common bone scan technetium-99 methylene di-phosphonate. Such scans are used routinely to show areas of increased activity in bone; the technetium is

water soluble and therefore rapidly eliminated through the kidneys. Any view of a total-body technetium bone scan will identify a generally greater uptake in the jaws, indicative of the greater turnover rate of jawbone (Fig 13). That, together with occlusal loading and den-ture wearing, is why the jaws are more vulnerable to bisphosphonates and why exposed-bone ONJ only occurs in the jaws.

In the synthesis of bisphosphonates, the central-backbone oxygen atom of the pyrophosphates is re-placed by a carbon atom, which imparts irreversible binding to the hydroxyapatite in bone (Fig 14).43,44 The result is a documented half-life in bone of 11 years and an accumulation to toxic effects with repeated doses that can result in ONJ or femur fractures.45 This explains the greater numbers and greater severity of ONJ cases with longer-term use and the fact that the vast majority of femur fractures occur only after 6 or more years of alendronate use (Fig 12).

While the central-backbone carbon atom imparts irreversible binding to bone, the R-group attached to the central carbon atom confers potency (Fig 14). Ni-trogen within this R-group side chain increases po-tency sufficiently such that non–nitrogen-containing bisphosphonates, eg, etidronate and tiludronate, are not known to cause ONJ. Variations in the configu-ration of the nitrogen-containing side chains confer different degrees of potency, with zolendronate be-ing the most potent of the IV bisphosphonates and therefore causing the majority of IV ONJ cases46 and alendronate being the most potent of the oral bisphos-phonates and therefore causing the majority of oral ONJ cases.47–49

Fig 12 Subtrochanter midshaft fracture of the femur caused by long-term alendro-nate (Fosamax) use.

Fig 13 Technetium-99 methylene diphosphonate bone scans show a generalized in-creased uptake in the jaws, which is indicative of a higher turnover rate.


Marx


The Pharmacodynamics of Bisphosphonates and BoneIt has been widely recognized that ONJ cases caused by IV bisphosphonates occur after fewer doses, are more severe, require surgical intervention more often and of a greater magnitude, and are less responsive to drug holidays.47,48 This is a result of the immediate uptake and binding of bisphosphonates to bone via the IV route, as compared to the oral route, where less than 0.7% is absorbed through the gastrointestinal mucosa before it circulates and binds to bone.7,47,50,51

Therefore, for each milligram of the equally potent IV bisphosphonate pamidronate (Aredia) vs the oral bisphosphonate alendronate, there is a 1/0.007 bone absorption ratio, meaning that bone absorption and accumulation is 142.8 times higher for an IV than an oral bisphosphonate. This explains not only the differ-ence in severity and the degree of exposed bone but why as few as four doses of an IV bisphosphonate can place patients at a significant risk of ONJ, while it may take 100 or more weekly doses (2 years) of an oral bisphosphonate to create the same risk level.

Because all bisphosphonates affect the osteoclast precursors in bone marrow as well as those osteo-clasts remodeling mineralized bone in the general skeleton,19,52 IV doses affect the bone marrow pre-cursors quickly and more severely. This reduces the ability of the bone marrow to replace lost osteoclasts that attempt to resorb bisphosphonate-loaded bone in the jaws at other skeletal units. By contrast, the poor gastrointestinal absorption of oral bisphosphonates represents a trickle effect into the bone marrow and affects the precursor cells more weakly, allowing them

to keep pace with lost osteoclasts during resorption of bisphosphonate-loaded bone in the jaws and else-where. This retained ability to replace a population of dying osteoclasts explains why drug holidays of 9 months or longer are effective in reducing the risk of ONJ in an oral bisphosphonate individual but usually are not in a patient that has been treated with an IV bisphosphonate.

Drug Holidays and BisphosphonatesA drug holiday is a temporary discontinuation of a bisphosphonate to allow bone marrow recovery and the formation of new osteoclasts to repopulate the population that has been depleted by the bisphospho-nate. As stated, because IV bisphosphonates confer 142.8 times the bone loading dose of oral bisphos-phonates, the depletion of osteoclast precursors and the accumulation of the bisphosphonate in bone are too great to be predictably reversed by a drug holiday. However, because an oral bisphosphonate’s systemic absorption is so low, its effects on the osteoclast bone marrow precursors are less pronounced, allowing the bone marrow to keep pace with the loss. Therefore, a drug holiday becomes effective because a residual population of marrow stem cells and osteoclast pro-genitor cells repopulates the osteoclast pool to re-turn bone remodeling/renewal to levels that remain depressed but are still consistent with bone healing (Figs 15 and 16).

Arbitrary drug holidays that have been recom-mended by various associations are not based on evidence.53–55 This is because the length of a drug hol-iday depends on the duration and dose of the various

Fig 14 The basic bisphosphonate molecule.

OH group is the site that binds to hydroxyapatite.

P–C–P is necessary for irreversible binding to hydroxyapatite.

The P–C–P backbone makes the bisphosphonate unable to be hydrolyzed and is essential for biologic activity.

R2 chain determines potency.Nitrogen-containing side chains are most potent.

O = P C P = O

HO

HO

HO

OH

OH

R2


Marx


bisphosphonates used. For instance, alendronate is recommended and marketed for 70 mg per week, as compared to residronate, at 35 mg per week, and iban-dronate, at 150 mg per month (35 mg/week). Perhaps this is why alendronate has accounted for more than 96% of ONJ cases caused by an oral bisphosphonate and why it is the only oral bisphosphonate known to cause fractures of the femur.38,48

The length of a drug holiday from an oral bisphos-phonate or even the very need for a drug holiday may be guided by a serum morning fasting c-terminal telo-peptide (CTX) test. Although the clinical utility of this blood test remains controversial, with several studies showing data to attest to its clinical correlation48,56–59 and others denying its usefulness,53,60 it is frequently used as the primary biochemical marker of bone turn-over. In fact, it has been used in nearly every study of the therapeutic effectiveness of oral bisphosphonates

and those of many of the IV bisphosphonates that were presented to the FDA.5,6,30,33,61

The CTX measures an eight–amino-acid sequence of collagen released by osteoclasts during bone re-sorption.56 However, since most cancers break down collagen during their invasion of normal tissue, the CTX results are unreliably higher in cancer patients, making it a useless study in this group. Therefore, the clini-cal utility of the CTX is limited to noncancer patients on oral bisphosphonates. In addition, in patients who have taken or are taking methotrexate or prednisone, the CTX is unreliably low and is also not recommend-ed. These patients are usually those with autoimmune diseases such as rheumatoid arthritis or systemic lupus erythematosus. Because methotrexate reduces the bone marrow stem cell population, fewer osteo-clasts are produced, resulting in lower bone turnover and therefore a lower CTX level. Because prednisone

Fig 15 Persistent exposed bone despite a long-term (3 years) drug holiday in a patient who received Zometa for breast cancer metastasis for 2.5 years before the drug holiday.

Fig 16a Exposed bone in a patient who received Fosamax for 4 years.

Fig 16b After a 9-month drug holiday, the exposed bone is par-tially sequestered, allowing its removal.

Fig 16c The site has healed as a result of the repopulation of osteoclasts and the return of bone remodeling following the drug holiday.

a

cb


Marx


inhibits collagen synthesis, including type 1 collagen in bone, the release of collagen fragments during bone turnover is reduced, resulting in a lower CTX level.56,59

Normal CTX level tests in bisphosphonate-naïve adults range between 350 and 500 pg/mL.48,50,56 CTX levels in patients taking an oral bisphosphonate may range from 20 to 55 pg/mL, depending on the spe-cific drug, the dose, and the number of doses taken. Studies have shown that a CTX value of 150 pg/mL is sufficient for healing in alveolar bone after surgery and represents a goal for a drug holiday.48,56 Upon begin-ning a drug holiday in an uncomplicated osteopenia/ osteoporosis patient, a low CTX is usually seen to gradually rise to a level of 150 pg/mL within 9 months. However, one notable exception highlights an important lesson to be learned about osteoclasts and bisphos-phonates. That is, in patients who have taken alendro-nate steadily for 7 years or more, the CTX is usually very low (30 to 60 pg/mL). Upon beginning a drug hol-iday, the CTX is seen to rise, as one would expect from bone marrow recovery. However, once the CTX rises to levels around 100 pg/mL, CTX levels paradoxically fall again to an intermediate level of 60 to 90 pg/mL. This seems to be the result of the drug holiday allow-ing bone marrow recovery and the regeneration of new osteoclasts; these new osteoclasts begin resorbing heavily bisphosphonate-loaded bone. In doing so, they die and burst (apoptosis), releasing the metabolically stable bisphosphonate into the local environment,

where some of it becomes bound once again to ad-jacent bone; most is taken up in the circulation, ef-fectively re-dosing the body with the bisphosphonate, thereby suppressing the bone marrow osteoclast pre-cursors once again and lowering the CTX value.

Endosteum and Periosteum The mandible and maxilla, like most bones, have an outer rim of cortical bone covered by a thin perios-teum on their outer cortical surface. On the inner corti-cal surface is a layer of endosteum, which unlike the periosteum is not a defined membrane but a series of connected cells that line the surface of a network of interconnected trabecular bone. This trabecular bone network extends from the buccal cortex to the lingual cortex, which is lined by endosteal cells throughout. Between these bone trabeculae are various marrow spaces, which in the adult contain a dwindling number of true hematopoietic and mesenchymal stem cells as well as age-related fibrofatty marrow.

In resected specimens of both IV and oral bisphos-phonate–induced ONJ, there is a noted formation of red viable periosteal new bone that can be separat-ed from the underlying necrotic bone in many cases (Figs 17a to 17c). Additionally, new periosteal bone can be seen radiographically to form months and years after resection (Fig 18). This indicates that the network of trabecular bone in the marrow space and the inner cortical surface are dependent on the

Fig 17a Resection specimen from a case of IV bisphosphonate–induced osteonecrosis shows a red surface of periosteal newly formed bone.

Fig 17b Necrotic central marrow core and cortex, indicating that periosteal bone formation is independent of osteoclast function, whereas the endosteum is not.

Fig 17c Lingual surface of specimen seen in Fig 17a, showing prominent periosteal new bone formation on all surfaces.

cb

a


Marx


osteoclast-mediated coupling of bone renewal and homeostasis that is induced by bone resorption/ release of growth factors/new bone formation, where-as the outer cortical surface and periosteal cells do not depend on this coupling.62–64 This further implies a significant genetic difference between periosteal and endosteal cells and likely a different embryology as well. Moreover, it has also been noted that when recombinant human BMP is used in contact with both endosteum and periosteum, the regenerated bone arises solely from the endosteum, implying that peri-osteal cells lack the receptors for BMP (Fig 19).65 This has already been translated to the clinical arena by sheer experience, as surgeons do not rely on the periosteum to regenerate bone when using recombi-nant human BMP-2 in an absorbable collagen sponge as a graft but instead increase its contact with a great-er area of endosteal bone surface.65

CONCLUSIONS

1. The toxic effects of bisphosphonates on bone and the resulting cases of osteonecrosis of the jaw, similar to human immunodeficiency virus/AIDS, have reinforced the age-old axiom that “disease teaches us much about normal.”

2. Bone remodeling is the mechanism by which bone renews itself and by which it adapts to functional stress, often termed Wolff’s law or bone modeling.

3. Osteoclastic bone resorption is coupled to bone remodeling and therefore bone modeling and bone renewal by the growth factors bone morphogenet-ic protein and insulinlike growth factors 1 and 2.

4. The inhibition of bone remodeling by bisphospho-nates results in the retention of old hypermineral-ized brittle bone, which may appear as a thickened

lamina dura or an area of generalized sclerosis in the jaws, may die off to result in osteonecrosis of the jaw, or may result in subtrochanteric fractures of the femur.

5. Bisphosphonates are irreversibly bound to bone and can be released from bone only by osteoclas-tic acid–based resorption. This irreversible bind-ing is caused by the backbone carbon atom in this relatively small molecule. Potency is determined by the nitrogen-containing side chain on the back-bone carbon atom.

6. Osteonecrosis caused by intravenous bisphos-phonates is more prevalent, more severe, less responsive to drug holidays, and less responsive to debridement surgeries than osteonecrosis in-duced by oral bisphosphonates, because intra-venous bisphosphonates load and accumulate in bone at a rate 142.8 times higher than oral bisphosphonates.

7. Drug holidays are useful only in cases of oral bisphosphonate use uncomplicated by cancer, methotrexate, or prednisone because the 0.7% gastrointestinal absorption of oral bisphospho-nates induces only a gradual toxicity to the bone marrow, allowing the marrow to keep pace, recov-er in response to a drug holiday, and repopulate the lost osteoclasts.

8. Osteoclasts that resorb bone during a drug holi-day after several years of bisphosphonate use re-lease the bisphosphonate, which is then taken up in the circulation to act in a re-dosing effect on the bone marrow as well as bone formation. This is reflected by lower c-terminal telopeptide values.

9. While the endosteum is dependent on the cas-cade of bone resorption/release of growth factors/ new bone formation, periosteum is not and is therefore resistant to bisphosphonate toxicity.

Fig 19 Regenerated bone from recombinant human BMP-2 is seen to arise from existing bone surfaces rather than from the periosteum.

Fig 18 Four-year follow-up of a resection defect from a patient with IV bisphosphonate–induced osteonecrosis shows radiographically demon-strable periosteal new bone formation.


Marx


10. Periosteal and endosteal cells are likely genetically and embryologically different, as they respond in opposite fashions to bisphosphonates and bone morphogenetic protein.

ACkNOWLEDgMENTS

The authors reported no conflicts of interest related to this study.Several of the images in this article are reprinted from Marx,50 with permission from Quintessence Publishing Co Inc.

REFERENCES

1. Ho DD, Neumann AU, Perelson AS, Chen W, Leonard JM, Markowitz M. Rapid turnover of plasma virions and CD4 lymphocytes in HIV infection. Nature 1995;373:123–126.

2. Griesser H, Mak TW. Developmental and functional biology of B and T lymphocytes. In: Canellos GP, Lister TA, Sklar JL (eds). The Lymphomas. Philadelphia: WB Saunders, 1998:19–42.

3. Cotton DJ, Powderly WG. Antiretroviral therapy. In: Powderly WG (ed). Manual oOf HIV Therapeutics. Philadelphia: Lippincott-Raven, 1997:42–49.

4. Hanauer SB, Sandburn WJ, Rutzerts P, Fedorak RN, Lukas M, et al. Human anti-necrosis factor monoclonal antibody (adalimumab) in China’s disease: The classic trial. Gastroen-terology 2006;130:323–333.

5. Novartis AG. Aredia. Pamidronate disodium for injection, for intravenous infusion. Product Information Sheet. Basel, Switzerland: Novartis AG, 2004:1.

6. Novartis AG. Zometa: Zoledronic acid injection. Product Information Sheet. 2011. http://www.pharma.us.novartis.com/product/pi/pdf/reclast.pdf. Accessed 18 November 2012.

7. Merck & Co. Inc. Fosamax (alendronate sodium) tablets and oral solution. Product Information Sheet. 2012. http://www.merck.com/product/usa/pi_circulars/f/fosamax/fosamax_pi.pdf. Accessed 18 November 2012.

8. Marx RE, Stern DS (eds). Oral and Maxillofacial Pathology: A Rationale for Diagnosis and Treatment. Carol Stream, IL: Quintessence, 2002.

9. Marx RE. Pamidronate (Aredia) and zolendronate (Zometa) induced avascular necrosis of the jaws: A growing epidemic. J Oral Maxillofac Surg 2003;61:1115–1157.

10. Ruggiero SL, Mekrota B, Rosenberg TJ, Engroff SL. Osteo-necrosis of the jaws associated with the use of bisphospho-nates. A review of 63 cases. J Oral Maxillofac Surg 2004;62: 527–534.

11. Migliorati CA. Bisphosphonates and oral cavity avascular bone necrosis. J Clin Oncol 2003;21:4253–4254.

12. Wang J, Goodger NM, Pogrel MA. Osteonecrosis of the jaws associated with cancer chemotherapy. J Oral Maxillofac Surg 2003;61:1104–1107.

13. Etminan M, Amirzadek K, Matthew IR, Brophy JM. Use of oral bisphosphonates and the risk of aseptic osteonecrosis: A nested case-control study. J Rheumatol 2008 Apr;35(4): 691–695. Epub 2008 Jan 15.

14. Bonacina R, Mariani U, Villa F, Villa A. Preventive strategies and clinical implications for bisphosphonate-related osteonecrosis of the jaw: A review of 282 patients. J Can Dent Assoc 2011;77:b147.

15. Hess LM, Jeter JM, Benham-Hutchins M, Albert MD. Factors associated with osteonecrosis of the jaw among bisphospho-nate users. Am J Med 2008;121:475–483.

16. Dimopoulos M, Kastridis E, Anagnostopbilos A, et al. Osteonecrosis of the jaw in patients with multiple myeloma treated with bisphosphonates: Evidence of increased risk after treatment with zoledronic acid. Hematologica 2006;91: 968–997.

17. Badros A, Weckel D, Goloubeva O, Rapopirt A, Fentin R, et al. Osteonecrosis of the jaw in multiple myeloma patients: Clinical features and risk factors. J Clin Oncol 2006;24: 945–952.

18. Glowacki J. The deceiving appearance of osteoclasts. N Engl J Med 2009;36:80–82.

19. Eslami B, Zhou S, Van Ekeren I, Le Boff MS, Glowacki J. Reduced osteoclastogenesis and RANKL expression in marrow from women taking alendronate. Calcif Tissue Int 2011 Apr;88:272–280. Epub 2011 Feb 15.

20. Jameson JL, Kasper DL, Faui AS, Braunwald E, Lingo DL, et al. Harrison’s Endocrinology. New York: McGraw Hill Medical, 2006:411–414.

21. Boyle W, Simonet W, Lacey P. Osteoclast differentiation and activation. Nature 2003;423:337–341.

22. Hofbauer K, Khosla S, Dunstan CR, et al. The roles of osteopro-tegerin and osteoprotegerin ligand in the paracrine regulation of bone resorption. J Bone Miner Res 2000;15:2–12.

23. Hofbauer L, Heufeldes A. Role of receptor activator of nuclear factor-kappa B ligand and osteoprotegerin in bone cell biology. J Mol Med 2001;79:243–253.

24. Hofbauer L, Kuhne C, Viereck V. The OPG/RANKL/RANK system in metabolic bone disease. J Musculoskelet Neuron Interact 2004;4:268–275.

25. Mohan S, Baylink DJ. Bone growth factors. Clin Orthop 1991; 263:30–48.

26. Wergedal JE, Mohan S, Lundy M, Baylink DJ. Skeletal growth factors and other factors known to be present in bone matrix stimulate proliferation and protein synthesis in human bone cells. J Bone Miner Res 1990;5:179–186.

27. Epstein S. The roles of bone mineral density, bone turnover, and other factors in reducing fracture risk during antiresorp-tive therapy. Mayo Clin Proc 2005;80:379–388.

28. Riggs L. The mechanisms of estrogen regulation of bone resorption. J Clin Invest 2000;106:1203–1204.

29. Bonnick S. Monitoring osteoporosis therapy with bone densitometry: A vital tool or regression toward mediocrity? J Clin Endocrinol Metab 2000;10:343–345.

30. Bone HG, Hosking D, Devogalaer JP, et al. Ten years’ experience with alendronate for osteoporosis in postmeno-pausal women. N Engl J Med 2004;350:1189–1199.

31. Alonso-Cuello P, Garcia-Franco AL, Guyatt G, Moynihan R. Drugs for preosteoporosis: Prevention or disease mongering? BMJ 2008;336:126–129.

32. Edwards MH, McCrae FC, YoungMia SA. Alendronate-relat-ed femoral diaphysis fracture. What should be done to predict and prevent subsequent fracture of the contralateral side. Osteoporos Int 2010;21:701–703.

33. Black DM, Thompson DE, Vauer DC, et al. Fracture risk reduction with alendronate in women with osteoporosis: The fracture intervention trial. J Clin Endocrinol Metab 2000;85:4118–4124 [erratum 2001 Feb;86:938].

34. Marini JC. Do bisphosphonates make children’s bones better or brittle? N Engl J Med 2003;349:423–426.

35. Nevraser AS, Lane JM, Lenart BA, Edobor-Osula F, Lorich DG. Low energy femoral shaft fractures associated with alendronate use. J Orthop Traum 2008;22:346–350.


Marx


36. Park-Wyllie LY, Mamdani MM, Juirlink DN, et al. Bisphospho-nate use and atypical fractures of the femoral shaft. N Engl J Med 2011;36:1728–1737.

37. Schilecher J, Michaelsion K, Aspenberg P. Bisphosphonate use and atypical fractures of the femoral shaft. N Engl J Med 2011;36:1728–1737.

38. Atik OS, Suluvova F, Gormeli G, Yildririm A, Ali AK. Insuffi-ciency femoral fractures in patients undergoing prolonged alendronate therapy. Eklem Hastalok Carrahisi 2010;21: 56–59.

39. Adami S, Isaia G, Luisetto G, et al. Osteoporosis treatment and fracture incidence. The ICARO longitudinal study. Osteoporosis Int 2008;19:1219–1223.

40. Gorlin RJ, Cohen MM Jr, Hennekam RCM. Osteopetrosis. In: Gorlin RJ, Cohen MM Jr, Hennekam RCM (eds). Syndromes of the Head and Neck, ed 4. New York: Oxford University Press, 2001:285–289.

41. Food and Drug Administration. Background document for meeting of advisory committee for reproductive health drugs and drug safety and risk management advisory committee. FDA, Sept. 9, 2011. http://www.fda.gov/downloads/advisory committees/committees meeting materials/drugs/drug safety and risk management advisory committee/UCM 270958.pdf. Accessed 19 November 2012.

42. Glowacki J. Bisphosphonates and bone. Ortho J Harv Med School 2005;7:64–65.

43. Russell RGG, Croucher PI, Rogers MJ. Bisphosphonates: Pharmacology, mechanisms of action and clinical uses. Osteoporos Int 1999;(suppl 2):566–580.

44. Rogers MJ, Gordon S, Benford HL, et al. Cellular and molecular mechanisms of action of bisphosphonates. Cancer 2000;88(suppl):2961–2979.

45. Llasseter KC, Porras AG, Denker A, et al. Pharmocokinetic considerations in determining the terminal half-lives of bisphosphonates. Clin Drug Invest 2005;25:107–114.

46. Marx RE, Sawatari Y, Fortin M, Broumand V. Bisphosphonate-induced exposed bone (osteonecrosis/osteopetrosis) of the jaws. Risk factors, recognition, prevention, and treatment. J Oral Maxillofac Surg 2005;63:1567–1575.

47. Marx RE. Oral and Intravenous Bisphosphonate-Induced Osteonecrosis of the Jaws. History, Etiology, Prevention, and Treatment. Hanover Park, IL: Quintessence, 2007.

48. Marx RE, Cillo JE, Ulloa JJ. Oral bisphosphonate induced osteonecrosis: Risk factors, prediction of risk using serum CTX testing, prevention and treatment. J Oral Maxillofac Surg 2007;65:2397–2404.

49. Migliorai CA, Casiglia J, Epstein J, et al. Managing the care of patients with bisphosphonate-associated osteonecrosis. An American Academy of Oral Medicine Position Paper. J Am Dent Assoc 2005;136:1658–1668.

50. Marx RE. Oral and Intravenous Bisphosphonate-Induced Osteonecrosis of the Jaws, ed 2. Hanover Park, IL: Quintes-sence, 2011.

51. Porras AG, Holland SD, Gerz BJ. Pharmacokinetics of alendronate. Clin Pharmacokinet 1999;6:316–328.

52. Van Beek ER, Lowck CWGM, Papapoulos SE. Bisphospho-nates suppress bone resorption by a direct effect of early osteoclasts precursors without affecting the osteoclastogenic capacity of osteogenic cells. Bone 2002;30:60–74.

53. Ruggiero SL, Dodson TB, Assael LA, et al. American Association of Oral and Maxillofacial Surgeons position paper on bisphosphonate-related osteonecrosis of the jaws—2009 update. J Oral Maxillofac Surg 2009;67(5 suppl):2–12.

54. Khosla S, Burr D, Cauley J, et al. Bisphosphonate associated osteonecrosis of the jaw: Report of a task force of the American Society for Bone and Mineral Research. J Bone Miner Res 2007;22:1479–1489.

55. Advisory Task Force on Bisphosphonate-Related Osteo-necrosis of the Jaws. American Association of Oral and Maxillofacial Surgeons position paper on bisphosphonate-related osteonecrosis of the jaws. J Oral Maxillofac Surg 2007;65:369–376.

56. Rosen HN, Moses AC, Garber J, et al. Serum CTX. A new marker of bone resorption that shows treatment effect more often than other markers because of low coefficient of variability and large changes with bisphosphonate therapy. Calcif Tissue Int 2000;66:100–103.

57. Kunchur R, Need A, Hughes T, Goss A. Clinical investigation of C-terminal cross linking telopeptide test in prevention and management and management of bisphosphonate-associated osteonecrosis of the jaws. J Oral Maxillofac Surg 2009;67: 1167–1173.

58. Kwon YD, Oh JY, Kim DY, Chung DJ, Park YD. Retrospective study of two biochemical markers for the risk assessment of oral bisphosphonate-related osteonecrosis of the jaws. Can they be utilized as risk markers? Clin Oral Implants Res 2011 Jan;22:100–105. Epub 2010 Oct 13.

59. Greenspan SL, Parker RA, Fergusin L, Rosen HN, Maitland-Ramsey L, Karpf DB. Early changes in biochemical markers of bone turnover predict the long term response to alendronate therapy in representative elderly women; A randomized clinical trial. J Bone Miner Res 1998;13:1431–1438.

60. Dodson TB. CTX and its role in managing patients exposed to oral bisphosphonates letter to editor. J Oral Maxillofac Surg 2010;68:487–494.

61. Black DM, Schwartz AV, Ensrud KE, et al; FLEX Research Group. Effects of continuing or stopping alendronate after 5 years of treatment. The Fracture Intervention Trial Long-Term Extension (FLEX): A randomized trial. JAMA 2006;294: 2927–2938.

62. Feher R, Koivunemi A, Koivunemi M, et al. Bisphosphonates do not inhibit periosteal bone formation in estrogen deficient animals and allow enhanced bone modeling in response to mechanical loading. Bone 2010 Jan;46:203–207. Epub 2009 Oct 24.

63. Allen MR, Iwata K, Phipps R, Burr DB. Alterations in canine vertebral bone turnover, micro damage accumulation, and biochemical properties following 1 year treatment with clinical treatment doses of risedronate or alendronate. Bone 2006; 39:872–879.

64. Gourion-Arsiguaud SG, Allen MR, Burr DB, Vashishth D, Tang SY, Boskey AL. Bisphosphonate treatment modifies canine bone mineral and matrix properties and their heteroge-nicity. Bone 2010 Mar; 46:666–672. Epub 2009 Nov 17.

65. Triplett RG, Nevins M, Marx RE, et al. Pivotal, randomized, parallel evaluation of recombinant human bone morphoge-netic protein-2/absorbable collagen sponge and autogenous bone graft for maxillary sinus floor augmentation. J Oral Maxillofac Surg 2009;67:1947–1960.


a decade of bisphosphonate bone complications: what it has ... · oral & craniofacial tissue...

Documents