Myeloma bone disease: pathophysiology and management

Transcription

1 Review Annals of Oncology 16: , 2005 doi: /annonc/mdi235 Published online 31 May 2005 Myeloma bone disease: pathophysiology and management E. Terpos 1,2 * & M.-A. Dimopoulos 3 1 Department of Hematology, 251 General Airforce Hospital, Athens, Greece; 2 Department of Hematology, Faculty of Medicine Imperial College London, Hammersmith Hospital, London, UK; 3 Department of Clinical Therapeutics and Internal Medicine, University of Athens School of Medicine, Athens, Greece Received 5 December 2004; accepted 4 February 2005 Bone disease is a major feature of multiple myeloma. Myeloma-induced bone destruction is the result of an increased activity of osteoclasts, which is not accompanied by a comparable increase of osteoblast function. Recent studies have revealed that new molecules such as the receptor activator of nuclear factor-kappa B (RANK), its ligand (RANKL), osteoprotegerin (OPG), and macrophage inflammatory protein-1a are implicated in osteoclast activation and differentiation, while proteins such as dickkopf-1 inhibit osteoblastic bone formation. These new molecules seem to interfere not only with the biology of myeloma bone destruction but also with tumour growth and survival, creating novel targets for the development of new antimyeloma treatment. Currently, bisphosphonates play a major role in the management of myeloma bone disease. Clodronate, pamidronate and zoledronic acid are the most effective bisphosphonates in symptomatic myeloma patients. Biochemical markers of bone remodeling have been used in an attempt to identify patients more likely to benefit from early treatment with bisphosphonates. Furthermore, using microarray techniques, myeloma patients may be subdivided into molecular subgroups with certain clinical characteristics, such as propensity for lytic lesions that may need early prophylactic treatment. Recent phase I studies with recombinant OPG and monoclonal antibodies to RANKL appear promising. Key words: bone disease, bisphosphonates, macrophage inflammatory protein-1 alpha (MIP-1a), multiple myeloma, osteoprotegerin, receptor activator of nuclear factor-kappa B ligand (RANKL) Introduction Multiple myeloma (MM), which accounts for approximately 1% of all cancer-related deaths in Western countries, is characterized by the accumulation of malignant plasma cells in the bone marrow leading to impaired hematopoiesis and bone disease, which includes mainly lytic lesions, pathological fractures, hypercalcemia and osteoporosis. Myeloma bone disease is the result of an increased activity of osteoclasts, which is not accompanied by a comparable increase of osteoblast function, thus leading to enhanced bone resorption. The interaction of plasma cells with bone marrow microenvironment is crucial for the activation of osteoclasts. Several recent studies have provided new insights into the pathogenesis of MM bone disease. Apart from cytokines, such as interleukin-6 (IL-6), macrophage colony-stimulating factor (M-CSF), interleukin-1 beta (IL-1b), tumor necrosis factors (TNFs) and interleukin-11 (IL-11), which are known to have osteoclast activating function (OAF), the characterization of newer molecules such as the receptor activator of nuclear factor-kappa B (RANK), its ligand (RANKL), osteoprotegerin *Correspondence to: Dr E. Terpos, 5 Marathonomahon street, Drossia Attikis, 14572, Greece. Tel: ; Fax: ; [email protected] (OPG) and macrophage inflammatory protein-1 alpha (MIP- 1a), has provided further insight into the pathogenesis of MM bone disease and has formed the basis for development of new therapeutic approaches. Biology of MM bone disease Increased osteoclast activity is evident in multiple myeloma The interactions between myeloma cells and bone marrow microenvironment leads to osteoclast activation and proliferation, and subsequently to increased bone resorption through the production of different cytokines with OAF activity (Figure 1). Interleukin-6. IL-6 is produced mainly by marrow stromal cells and is a growth factor for both osteoclasts and myeloma cells, stimulating their proliferation and preventing their apoptosis. The primary effect of IL-6 on osteoclast formation is to increase the pool of the early osteoclast precursors that in turn differentiate into mature osteoclasts. Serum levels of IL-6 and its receptor (IL-6R) are increased in MM and correlate with MM stage, disease activity and disease-free survival [1]. However, neither anti-il-6 nor anti-il-1b blocking was able to inhibit the osteoclastogenic effect of OAF factors secreted by q 2005 European Society for Medical Oncology

2 1224 BONE MARROW OPG Myeloma Cells RANKL(?) (-) CD138 DKK1 IL-6 VCAM-1 VCAM-1 α4β1 integrin RANKL RANKL BMSCs OPG (-) RANKL RANK IL-6 IL-11, IL-1b, bfgf TNFα, M-CSF MIP-1α, IL-3, HGF Osteoclast precursor TRACP-5b Activated osteoclasts Collagen type-1 degradation produxts: NTX, ICTP, CTX Bone resorption Bone matrix Figure 1. Myeloma cells adhere to bone marrow stromal cells (BMSCs) through binding of VLA-4 [a4b1 intrergrin; present on the surface of multiple myeloma (MM) cells] to vascular cell adhesion molecule-1 (VCAM-1), which is expressed on stromal cells. The adherence of MM cells to BMSCs/ osteoblasts enhances the production of the receptor activator of nuclear factor-kappa B ligand (RANKL), macrophage colony stimulating factor (M-CSF), and other cytokines with osteoclast activating function (OAF) activity (IL-6, IL-11, IL-1b, TNFs, bfgf), while it suppresses the production of osteoprotegerin (OPG, the decoy receptor of RANKL). The above cytokines also modify the bone marrow microenvironment, up-regulating the RANKL expression and secretion by both BMSCs and osteoblasts. Furthermore, myeloma cells produce MIP-1a, HGF and VEGF, which enhance the proliferation and differentiation of osteoclast precursors. MIP-1a can also activate intergrins to further induce cell adhesion, taking part in a paracrine pathway of inducing adherence of MM cells to stromal cells through VLA-4/VCAM-1 interactions, thus stimulating osteoclast activation. Myeloma cells may express RANKL, while OPG binds both surface and soluble RANKL inhibiting osteoclast development and bone resorption. Syndecan 1 (CD138) expressed on the surface of, and secreted from, the myeloma cells can bind soluble OPG, thus preventing its inhibitory effect on RANKL function. Therefore the ratio of RANKL/OPG is increased, leading to osteoclast differentiation, proliferation and activation, and to increased bone resorption, as is reflected by the increased levels of bone resorption markers (TRACP-5b, NTX, ICTP, CTX). IL-6 might also play its role in myeloma through the activation of MIP-1a. All these phenomena emphasise the multiple complex interactions between myeloma cells and BMSCs. myeloma cell lines, while the administration of anti-il-6 antibodies in MM had no antimyeloma effect [2]. Interleukin-1b. IL-1b has potent OAF activity; it enhances the expression of adhesion molecules and induces paracrine IL-6 production, resulting in osteolytic disease. Increased levels of IL-1b were detected in the supernatant of cultures of freshly isolated myeloma cells and in in vivo models of human MM [3]. Elevated IL-1b mrna levels were also detected in MM patients, while anti-il-1b antibodies failed to completely abolish OAF activity of myeloma bone marrow [4]. Interleukin-3. IL-3 mrna levels were found to be increased in myeloma cells and IL-3 protein levels were increased in bone marrow plasma from MM patients. Furthermore, IL-3 in combination with MIP-1a or RANKL significantly enhanced human osteoclast formation and bone resorption compared with MIP-1a or RANKL alone. IL-3 also stimulated the growth of myeloma cells independently of the presence of IL- 6. These data suggest that increased IL-3 levels are present in the marrow microenvironment of myeloma patients, increasing bone destruction and tumor cell growth [5]. TNF-a. TNF-a is found at high levels in the supernatant of plasma cell cultures from MM patients. The effects of TNF-a are mediated by stimulation of the proteolytic breakdown of I- kappa B [the inhibitor of nuclear factor-kappa B (NF-kB)], leading to NF-kB activation and enhancement of gene transcription, including IL-6 and adhesion molecules, which are involved in promoting bone resorption [6]. Hepatocyte growth factor (HGF). HGF is involved in angiogenesis, epithelial cell proliferation and osteoclast activation. HGF and its receptor (c-met) are expressed on myeloma cells, which have the ability to convert HGF into its active form [7]. HGF up-regulates the expression of IL-11 from human osteoclast-like cells, while transforming growth factor-beta 1 (TGFb1) and IL-1 potentate the effect of HGF on IL-11 secretion. Thus, HGF is an indirect factor involved in myeloma bone disease. Furthermore, when serum HGF levels were elevated

3 1225 in MM patients they predicted for poor survival and lack of response to chemotherapy [8]. Vascular endothelial growth factor (VEGF). VEGF is a multifunctional cytokine that has a major role in tumor neovascularization and has been recently implicated in osteoclastogenesis in MM. VEGF is expressed by myeloma cells and it binds to the VEGFR-1 receptor that is predominantly expressed on osteoclasts. VEGF directly enhances osteoclastic bone resorption and survival of mature osteoclasts. It can substitute for M-CSF in the induction of osteoclast recruitment in mice with M-CSF gene deficiency. Moreover, VEGF enhances the production of IL-6 from stromal cells, while IL-6 stimulates VEGF expression and secretion by myeloma cells, suggesting the existence of paracrine interactions between myeloma and marrow stromal cells triggered by VEGF and IL-6 [9]. Osteopontin. Osteopontin is a non-collagenous matrix protein produced by various cells including osteoblasts, osteoclasts and myeloma cells. It is involved in a number of physiologic and pathologic events including adhesion, angiogenesis, apoptosis and tumour metastasis. Marrow cells from myeloma patients with advanced disease produced increased levels of osteopontin compared with asymptomatic MM or monoclonal gammopathy of undetermined significance (MGUS) patients. Furthermore, plasma osteopontin levels of MM patients were significantly higher than those of MGUS and controls, and correlated with both disease progression and bone destruction. These observations suggest that myeloma cells actively produce osteopontin, which contributes to osteoclastic bone resorption [10]. The role of MIP-a pathway in osteoclast activation in MM MIP-1a is a member of the CC chemokine family and is primarily associated with cell adhesion and migration. MIP-1a is chemotactic for monocytes and monocyte-like cells, including osteoclast precursors. It is produced by myeloma cells and directly stimulates osteoclast formation and differentiation in a dose dependent way, through the receptors CCR1 and CCR5, which are expressed by osteoclasts. Moreover, the addition of a neutralizing antibody against MIP-1a to human marrow cultures treated with freshly isolated marrow plasma from patients with MM blocks MIP-1a-induced osteoclast formation [11]. MIP-1a mrna has been detected in myeloma cells, while MIP-1a protein levels were elevated in the bone marrow plasma of MM patients and correlated with disease stage and activity. MIP-1a was also elevated in the blood of myeloma patients with severe bone disease, but not in MGUS patients with increased bone resorption [12 14]. MIP-1a has also been found to stimulate proliferation, migration and survival of plasma cells in both in vitro and in vivo studies [11]. Mice, which were inoculated with myeloma cells and treated with a monoclonal rat anti-mouse MIP- 1a antibody, showed a reduction of both paraprotein and lytic lesions. In addition, MIP-1a enhanced adhesive interactions between myeloma and marrow stromal cells, increasing the expression of RANKL and IL-6, which further increased bone destruction and tumor burden [15]. These observations are in accordance with the recent finding that myeloma patients with high MIP-1a serum levels had poor prognosis [13]. The effect of the RANKL/RANK/OPG system on osteoclasts in MM New insights into the pathophysiology of osteoclastogenesis have emerged recently with the characterization of three new molecules that belong to the TNF superfamily, namely RANKL, RANK and OPG. RANKL is encoded by a single gene at human chromosome 13q14. Alternative splicing of RANKL mrna allows expression of a type II transmembrane glycoprotein or a soluble ligand. Soluble RANKL (srankl) can be also released from its membrane-bound state by metalloproteinases. RANKL is expressed by activated T cells, marrow stromal cells and osteoblasts and binds to its receptor, RANK, which is expressed by osteoclast precursors, chondrocytes and mature osteoclasts. The binding of RANKL on RANK promotes osteoclast maturation and activation [16]. OPG is encoded by a single gene on chromosome 8q24 and is mainly secreted by marrow stromal cells. It is the decoy receptor for RANKL that blocks the RANKL RANK interaction and thus inhibits osteoclast differentiation and function [17]. Therefore, it is the balance between the expression of RANKL and OPG that determines the extent of osteoclast activity and subsequent bone resorption. Osteoclastogenesis requires contact between osteoclast precursors and stromal cells/osteoblasts. These accessory cells express M-CSF and RANKL that are essential to promote osteoclastogenesis. M-CSF expands the pool of osteoclast precursors and RANKL in turn stimulates it to commit to osteoclast phenotype. Thus, stromal cells and osteoblasts are the target cells of most osteoclastogenic factors that exert their effect by enhancing RANKL expression, such as parathyroid hormone (PTH) and vitamin D3. The expression of RANKL is enhanced by glucocorticoids, IL-1b, TNF-a, IL-11, PTH, prostaglandin-e2 and vitamin D3, and is decreased by TGF-b. The expression of OPG is increased by IL-1b, TNF-a, TGF-b and 17b-estradiol, while glucocorticoids, vitamin D3 and PTH reduce OPG production [18]. Furthermore, both IL-6 and IL- 11 support human osteoclast formation by a RANKL-dependent mechanism, and the presence of the RANKFc that blocks the RANK RANKL interaction inhibits stromal cell-induced secretion of IL-6 and IL-11 [19]. The importance of RANKL and OPG as regulators of osteoclastogenesis has become evident from experiments with transgenic mice. Mice that lack either RANKL or RANK or that over-express OPG develop osteopetrosis because of decreased osteoclast activity [20]. Conversely, OPG knockout mice have numerous osteoclasts and develop osteoporosis and multiple fractures, since OPG cannot inhibit RANKL activity [21]. Myeloma cells have the ability to up-regulate the expression of RANKL and down-regulate the expression of OPG at both mrna and protein level in pre-osteoblastic or stromal cell

4 1226 co-cultures [22]. Therefore, RANKL expression has been found to be increased in bone marrow biopsies from patients with MM, while RANKL is over-produced by stromal cells, osteoblasts and activated T-cells in areas infiltrated by MM [23]. An interesting and controversial question has arisen recently about the direct expression or production of RANKL by human myeloma cells. Some researchers have found that myeloma cells did not express RANKL and did not produce srankl [22]. Furthermore, microarray technology studies showed that RANKL gene expression has not been detected in myeloma cells of MM patients [24]. However, other groups have detected RANKL expression in myeloma cells [25]. Despite this controversy, the available data suggest that the RANKL/OPG system is mainly involved in the activation of osteoclasts by myeloma cells indirectly through the bone marrow environment. OPG expression is reduced in bone marrow specimens from myeloma patients. The adhesive interactions of myeloma cells with bone marrow stromal cells inhibit OPG production both at the mrna and protein level [22]. Furthermore, myeloma cells decrease OPG availability by internalizing it through syndecan-1 and degrading it within their lysosomal compartment [26]. Thus, in MM, the regulation of OPG at both transcriptional and post-translational level reduces the availability of OPG in the marrow microenvironment, leading to reduced inhibition of RANKL and increased osteoclast activation. Indeed, when serum OPG levels were evaluated in MM patients, they were found to be decreased, while serum levels of srankl and srankl/opg ratio were increased. Moreover, the srankl/opg ratio correlated with the extent of bone disease and makers of bone resorption, confirming the importance of the RANKL/OPG pathway in the pathogenesis of MM bone disease in humans [27]. Is increased osteoclast function crucial for myeloma cell growth? Osteoclasts seem to enhance growth and survival of myeloma cells more potently than stromal cells, while they protect them from apoptosis. The adherence of myeloma cells to osteoclasts resulted in increased IL-6 and osteopontin production from osteoclasts. Subsequently, IL-6 and osteopontin in combination enhanced MM cell growth and survival, which were only partially suppressed by a simultaneous addition of anti- IL-6 and anti-osteopontin antibodies, and were completely abrogated by inhibition of cellular contact between myeloma cells and osteoclasts. These observations demonstrate that interactions of myeloma cells with osteoclasts augment MM growth and survival, and thereby create a vicious cycle leading to extensive bone destruction and MM cell expansion [28]. Osteoblast function is also impaired in MM Histomorphometric evaluation of osteoblast activity in bone biopsies from MM patients has revealed osteoblast inhibition, since no evidence of bone regeneration was detectable either within the skeletal lesions or in their vicinity. Functional exhaustion of osteoblasts has been also postulated by the inverse relation between biochemical indicators of osteoid production, namely serum osteocalcin (OC) and bone-specific alkaline phosphatase (balp), and the presence of osteolytic lesions [3]. Silvestris et al. [29] showed that osteoblasts undergo apoptosis promptly in the presence of myeloma cells from patients with severe bone disease. It has been suggested that, in the myeloma bone microenvironment, both high cytokine levels and physical interaction between malignant plasma cells with osteoblasts lead to accelerated apoptosis of osteoblasts and defective new bone formation [29]. A recent study by Tian et al. [30], reported that myeloma cells produce dickkopf 1 (DKK1) protein, an inhibitor of the Wnt signaling pathway, which is crucial for osteoblast differentiation. Marrow plasma from patients with MM that contained >12 ng/ml of DKK1 inhibited osteoblast differentiation. Furthermore, gene expression levels of DKK1 correlated with the extent of bone disease [30]. The presence of a soluble factor produced by myeloma cells that suppresses osteoblast differentiation is a very important finding, which, however, does not entirely explain why myeloma bone lesions do not heal even in patients in complete remission. There may possibly be a longlasting change in the marrow microenvironment that results in an inability of osteoblast precursors to differentiate, despite the absence of myeloma cells. Figure 1 summarizes the currently available data on pathogenesis of myeloma-induced bone disease. Molecular subtypes of myeloma and lytic bone disease A relation between the presence of bone lytic lesions on magnetic resonance imaging (MRI) and molecular characteristics of myeloma patients has been recently described, and is depicted in Table 1 [30, 31]. There are myeloma patients who belong in certain molecular subtypes (i.e. in subtypes 1, 2) and have greater incidence of presence of lytic lesions on MRI and higher DKK1 expression. The confirmation of these data will lead to a molecular classification of myeloma patients with respect to the different biological and clinical features and subsequently to earlier use of agents with antiresorbing activity in patients at higher risk of developing lytic disease. Management of myeloma bone disease The management of myeloma bone disease includes mainly the use of bisphosphonates, radiotherapy, adequate analgesia for bone pain and, rarely, surgical procedures. Bisphosphonates Bisphosphonates inhibit osteoclast recruitment and maturation, prevent the development of monocytes into osteoclasts, induce osteoclast apoptosis and interrupt their attachment to the bone. Furthermore, anti-myeloma activity of pamidronate and zoledronic acid has been suggested [32, 33]. Possible mechanisms

5 1227 Table 1. Relation of myeloma molecular subtypes and lytic lesions on MRI Molecular subtype IgH gene translocation or cyclin D expression One or more lesions on MRI DKK1 signal > 1000 a on MRI positive versus MRI negative patients (%) 1 11q13 or 6p21; cyclin D1 or 94% 66 versus 0 cyclin D3 2 Frequent hyperdiploidy 89% 81 versus 13 but no 11q13; cyclin D1 3 Other translocation; cyclin D2 71% 45 versus p16; MMSET and FGFR3 57% 50 versus q23 or 20q11; c-maf or mafb 55% 17 versus 0 Adapted from Robbiani et al. [31]. a Values are the percentages of patients within each subgroup with a microarray signal (Affymetrix Hu95Av2) for DKK1 of >1000. MRI, magnetic resonance imaging; IgH, immunoglobulin heavy chain; DKK1, dickkopf 1; MMSET, multiple myeloma SET domain; FGFR3, fibroblast growth factor receptor 3; c-maf and mafb are v-maf avian musculoaponeurotic fibrosarcoma oncogene homologues. include the reduction of IL-6 secretion by bone marrow stromal cells or the expansion of gamma/delta T cells with possible anti-mm activity. Several studies have evaluated the role of bisphosphonates in patients with MM [34 46] (Table 2). Etidronate. Etidronate was found to be ineffective in two placebo-controlled studies in myeloma patients [34, 35]. Clodronate. Two major, placebo-controlled, randomized trials have been performed to date in MM. Lahtinen et al. [36] reported the reduction of the development of new osteolytic lesions by 50% in myeloma patients who received oral clodronate for 2 years. The benefits of clodronate were independent of the presence of lytic lesions at baseline. In the other study [37], although there was no difference in overall survival between the two groups, clodronate patients who did not have vertebral fractures at baseline seem to have a survival advantage (59 versus 37 months). After 1 year of follow-up, both vertebral and non-vertebral fractures, as well as the time to first non-vertebral fracture and severe hypercalcemia, were reduced in the clodronate group. At 2 years, the patients who received clodronate had better performance status and less myeloma-related pain than patients treated with placebo [38, 39]. Pamidronate. Two, double-blind, placebo-controlled trials have been performed to date in patients with MM using pamidronate. Brincker et al. [40] carried out a trial in which patients were randomized to receive either oral pamidronate or placebo, in addition to conventional treatment. The authors found no reduction in skeletal-related events (SREs). However, patients treated with oral pamidronate experienced fewer episodes of severe pain. The overall negative result of this study was attributed to the low absorption of orally administered bisphosphonates. In the second trial, patients with advanced disease and at least one lytic lesion were randomized to placebo or Table 2. Major double-blind, placebo-controlled, trials on bisphosphonates in multiple myeloma (MM) Author(s) [ref.] Type of bisphosphonate Dosage No. of MM patients Reduction of pain Reduction of SREs Belch et al. [34] Etidronate 5 mg/kg/day, p.o. 173 No No No Daragon et al. [35] Etidronate 10 mg/kg/day, p.o., for 4 months 94 No No No Lahtinen et al. [36] Clodronate 2.4 g/day, p.o., for 2 years 350 Yes Yes NE & Laakso et al. [37] McCloskey et al. [38, 39] Clodronate 1.6 g/day, p.o. 530 Yes Yes +/ b Brincker et al. [40] Pamidronate 300 mg/day, p.o. 300 Yes No No Berenson et al. [41, 42] Pamidronate 90 mg, i.v., every 4 weeks for 21 cycles 392 Yes Yes +/ c Menssen et al. [46] Ibandronate 2 mg, i.v., monthly 198 No No No Berenson et al. [43] a Zoledronic acid 2 or 4 mg, i.v., monthly 108 Yes Yes NE Rosen et al. [44, 45] a Zoledronic acid 4 or 8 mg, i.v., monthly 513 Yes Yes Yes Survival benefit a Pamidronate-controlled trial. b In a post-hoc analysis, patients without vertebral fracture at entry survived significantly longer on clodronate (median survival was 23 months longer than in similar patients receiving placebo). c Survival in the patients with more advanced disease was significantly increased in the pamidronate group (median survival 21 versus 14 months; P = adjusted for baseline serum b2-microglobulin and Eastern Cooperative Oncology Group performance status). SREs, skeletal related events (new lytic lesions, vertebral and non-vertebral fractures, need for radiation or surgery to the bone); NE, not evaluated.

6 1228 intravenous pamidronate [41, 42]. The mean number of SREs per year and the median time to the first skeletal event were reduced in the pamidronate group. Pain scores and quality of life were also significantly improved in the pamidronate group. Although there was no difference in terms of survival between the two treatment groups, this study identified a subgroup of patients, who had received more than one previous anti-myeloma regimen, in which pamidronate was associated with prolonged survival [42]. The Cochrane Myeloma Review Group has reported a meta-analysis based on 11 trials and involving 2183 assessable patients. This review concluded that both pamidronate and clodronate reduce the incidence of hypercalcemia, the pain index, and the number of vertebral fractures in myeloma patients [47]. Zoledronic acid. Berenson et al. [43] compared the effects of zoledronic acid and pamidronate in a phase II randomized trial. Zoledronic acid at doses of 2.0 and 4.0 mg and pamidronate at a dose of 90 mg each significantly reduced SREs in contrast to 0.4 mg zoledronic acid. This phase II trial failed to show any superiority of zoledronic acid compared to pamidronate in terms of SREs, in contrast to a large phase III study showing a superior effect of zoledronic acid at 4.0 or 8.0 mg over pamidronate for the treatment of hypercalcemia of malignancy. Therefore, a large phase III, randomized, double-blind, study was performed to compare the effects of zoledronic acid and pamidronate [44]. There was no difference in terms of time to the first SRE between treatment groups. The skeletal morbidity rate was slightly lower in patients treated with zoledronic acid (4.0 mg). N-telopeptide of collagen type-i (NTX), a marker of bone resorption, showed better normalisation in patients treated with 4.0 mg of zoledronic acid compared to pamidronate, but that was the only reported difference between the treatment groups. The long-term follow-up analysis confirmed that zoledronic acid was of similar efficacy and safety with pamidronate in MM patients [45]. Ibandronate. Ibandronate has been used effectively in the treatment of hypercalcemia of malignancy. However, a randomized, double-blind, placebo-controlled trial failed to show any effect of 2.0 mg of ibandronate on reducing bone morbidity or on prolonging survival in MM [46]. Another study has shown that pamidronate produced a greater reduction of biochemical markers of bone resorption, IL-6 and b 2 -microglobulin than ibandronate (4.0 mg), while there was no difference between the two bisphosphonates in terms of SREs during the 10-month period of follow up [48]. What is the optimal duration of bisphosphonates in myeloma patients? This question has not been answered to date because the issue has never been the subject of any clinical trial. However, due to the benefit of bisphosphonates on performance status, quality of life and possibly on survival in a subset of patients, the clinician has to decide on the optimal duration, taking into account the potential palliative benefits of bisphosphonates and the adverse events that may be manifested. We believe that symptomatic myeloma patients should continue bisphosphonate administration for life with adequate follow-up of renal and liver functions. The time of initiating bisphosphonate treatment is also controversial. The American Society of Clinical Oncology has suggested that myeloma patients with lytic lesions or osteopenia should be treated with bisphosphonates, but there is no such recommendation for patients with solitary plasmacytoma or smoldering/indolent myeloma without documented lytic bone disease [49]. In accordance with this recommendation, two recent studies have shown that prophylactic administration of pamidronate does not improve overall progression-free survival but may decrease the development of skeletal events in stage I MM [50, 51]. Monitoring bisphosphonate treatment in MM Imaging modalities and bone densitometry are of limited value in assessing improvement or deterioration of myeloma bone disease. Therefore, biochemical markers of bone turnover have been used in MM to identify subsets of patients who are most at risk of bone complications or will benefit the most from bisphosphonate treatment, and also for predicting disease progression. A variety of markers of bone resorption [NTX, C-telopeptide of collagen type-i (ICTP/CTX), tartrateresistant acid phosphatase isoform-5b (TRACP-5b), pyridinoline and deoxypyridinoline)] and bone formation (balp, OC and procollagen type-i N- or C-propeptide) have been studied. Both ICTP and NTX have shown a significant decrease after pamidronate or zoledronic acid administration [43, 52, 53]. High levels of ICTP and NTX correlated with bone disease progression during conventional anti-myeloma treatment [54]. Furthermore, TRACP-5b, which is produced only by activated osteoclasts, was increased in MM patients, correlated with the extent of bone disease, reduced during pamidronate administration and had a possible predictive value [55]. Bone resorption markers and srankl/opg ratio have also been found to have prognostic value in MM [27]. In addition, these markers may become normal after high dose chemotherapy with autologous stem cell support [56]. However, further trials are needed to establish the predictive value of these markers before introducing them into routine use. Radiotherapy Radiotherapy is mainly used for the management of solitary plasmacytoma when there is evidence of symptomatic spinal cord compression, and for extensive and symptomatic lytic lesions. For solitary plasmacytoma, treatment with 4500 cgy (4000 cgy for vertebral lesions) provides excellent local control [57]. For painful bone lesions, pain relief is usually obtained with doses of 3000 cgy in fractions. Patients with generalized pain due to multiple-site involvement may be treated with single-dose hemi-body irradiation, to doses of 600 cgy to the upper and 800 cgy to the lower hemi-body. Experience with double hemi-body irradiation over a 6-year period showed a 95% reduction of bone pain in myeloma

7 1229 patients with relapsed/refractory disease, and 20% of them were able to discontinue opiate analgesia [58]. Vertebroplasty Percutaneous vertebroplasty which consists of percutaneous injection of bone cement into the vertebral body under fluoroscopy guidance, has been introduced in the management of spinal fractures. Early results in patients with metastases, myeloma or osteoporotic compression fractures are very promising as 80% of patients with pain unresponsive to medical treatment experience pain relief [59]. Kyphoplasty represents a modification of vertebroplasty that, in addition to stabilizing the vertebra and relieving pain, aims to restore the vertebral body back towards its original height [60]. These new techniques require further evaluation. Surgery Surgery has a role in the management of selected MM patients. Fractures of the femora or humeri require prompt fixation with an intramedullary rod, followed by radiotherapy. Decompression laminectomy is rarely necessary in patients with known myeloma, although radioresistant myeloma or retropulsed bone fragment may require surgical intervention [61]. RANKL/OPG system as a target for novel anti-myeloma treatment Intravenous administration of either RANK-Fc, a fusion protein of the murine RANK with the human IgG region, or recombinant OPG markedly reduced not only bone resorption and skeletal destruction, but also tumor burden in myeloma animal models [62, 63]. Body et al. [64] attempted to disrupt the RANK/RANKL/OPG interaction in 28 myeloma patients who were randomized to receive a single dose of either recombinant OPG or pamidronate. OPG caused a rapid and sustained dose-dependent decrease in NTX, comparable to that observed with pamidronate, without having severe sideeffects; however, the development of anti-opg antibodies seems to eliminate the role of recombinant OPG in myeloma treatment [64]. Another recent study in 49 post-menopausal women with osteoporosis, confirmed the safety and bone antiresorptive effect of a single subcutaneous dose of a human monoclonal antibody to RANKL [65]. These results warrant further clinical trials targeting the RANKL/OPG pathway. Conclusions Bone disease remains a major problem in the management of patients with MM. Oral clodronate, intravenous pamidronate or zoledronic acid should be used in myeloma patients with osteolytic bone disease. However, many important issues, such as the most effective bisphosphonate, the time of initiation, the duration of treatment, and the use of markers to select high-risk patients, have not yet been clarified. Additional studies focusing on these issues are therefore required. Furthermore, the emergence of new molecules, which are involved in the pathogenesis of MM bone disease (RANKL, OPG, MIP-1a), may allow the development of new agents with antimyeloma activity. References 1. Kyrtsonis MC, Dedoussis G, Baxevanis C et al. Serum interleukin-6 (IL-6) and interleukin-4 (IL-4) in patients with multiple myeloma (MM). Br J Haematol 1996; 92: van Zaanen HC, Lokhorst HM, Aarden LA et al. Chimaeric anti-interleukin 6 monoclonal antibodies in the treatment of advanced multiple myeloma: a phase I dose-escalating study. Br J Haematol 1998; 102: Alsina M, Boyce B, Devlin RD et al. Development of an in vivo model of human multiple myeloma bone disease. Blood 1996; 87: Donovan KA, Lacy MQ, Gertz MA, Lust JA. IL-1beta expression in IgM monoclonal gammopathy and its relationship to multiple myeloma. Leukemia 2002; 16: Lee JW, Chung HY, Ehrlich LA et al. IL-3 expression by myeloma cells increases both osteoclast formation and growth of myeloma cells. Blood 2004; 103: Lam J, Takeshita S, Barker JE et al. TNF-alpha induces osteoclastogenesis by direct stimulation of macrophages exposed to permissive levels of RANK ligand. J Clin Invest 2000; 106: Tjin EP, Derksen PW, Kataoka H et al. Multiple myeloma cells catalyze hepatocyte growth factor (HGF) activation by secreting the serine protease HGF-activator. Blood 2004; 104: Seidel C, Lenhoff S, Brabrand S et al. Hepatocyte growth factor in myeloma patients treated with high-dose chemotherapy. Br J Haematol 2002; 119: Dankbar B, Padro T, Leo R et al. Vascular endothelial growth factor and interleukin-6 in paracrine tumor-stromal cell interactions in multiple myeloma. Blood 2000; 95: Standal T, Hjorth-Hansen H, Rasmussen T et al. Osteopontin is an adhesive factor for myeloma cells and is found in increased levels in plasma from patients with multiple myeloma. Haematologica 2004; 89: Choi SJ, Oba Y, Gazitt Y et al. Antisense inhibition of macrophage inflammatory protein 1-alpha blocks bone destruction in a model of myeloma bone disease. J Clin Invest 2001; 108: Abe M, Hiura K, Wilde J et al. Role for macrophage inflammatory protein (MIP)-1alpha and MIP-1beta in the development of osteolytic lesions in multiple myeloma. Blood 2002; 100: Terpos E, Politou M, Szydlo R et al. Serum levels of macrophage inflammatory protein-1 alpha (MIP-1a) correlate with the extent of bone disease and survival in patients with Multiple Myeloma. Br J Haematol 2003; 123: Politou M, Terpos E, Anagnostopoulos A et al. Role of receptor activator of nuclear factor-kappa B ligand (RANKL), osteoprotegerin and macrophage protein 1-alpha (MIP-1a) in monoclonal gammopathy of undetermined significance (MGUS). Br J Haematol 2004; 126: Oyajobi BO, Franchin G, Williams PJ et al. Dual effects of macrophage inflammatory protein-1alpha on osteolysis and tumor burden in the murine 5TGM1 model of myeloma bone disease. Blood 2003; 102: Hsu H, Lacey DL, Dunstan CR et al. Tumor necrosis factor receptor family member RANK mediates osteoclast differentiation

8 1230 and activation induced by osteoprotegerin ligand. Proc Natl Acad Sci USA 1999; 96: Simonet WS, Lacey DL, Dunstan CR et al. Osteoprotegerin: a novel secreted protein involved in the regulation of bone density. Cell 1997; 89: Hofbauer LC, Neubauer A, Heufelder AE. Receptor activator of nuclear factor-kappab ligand and osteoprotegerin: potential implications for the pathogenesis and treatment of malignant bone diseases. Cancer 2001; 92: Giuliani N, Colla S, Morandi F, Rizzoli V. The RANK/RANK ligand system is involved in intereleukin-6 and interleukin-11 up-regulation by human myeloma cells in the bone marrow microenvironment. Haematologica 2004; 89: Kong YY, Yoshida H, Sarosi I et al. OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis. Nature 1999; 397: Bucay N, Sarosi I, Dunstan CR et al. Osteoprotegerin-deficient mice develop early onset osteoporosis and arterial calcification. Genes Dev 1998; 12: Giuliani N, Bataille R, Mancini C et al. Myeloma cells induce imbalance in the osteoprotegerin/osteoprotegerin ligand system in the human bone marrow environment. Blood 2001; 98: Giuliani N, Colla S, Sala R et al. Human myeloma cells stimulate the receptor activator of nuclear factor-kb ligand (RANKL) in T lymphocytes: a potential role in multiple myeloma bone disease. Blood 2002; 100: Shaughnessy JD, Barlogie B. Interpreting the molecular biology and clinical behavior of multiple myeloma in the context of global gene expression profiling. Immunol Rev 2003; 194: Heider U, Langelotz C, Jakob C et al. Expression of receptor activator of nuclear factor kappab ligand on bone marrow plasma cells correlates with osteolytic bone disease in patients with multiple myeloma. Clin Cancer Res 2003; 9: Standal T, Seidel C, Hjertner O et al. Osteoprotegerin is bound, internalized, and degraded by multiple myeloma cells. Blood 2002; 100: Terpos E, Szydlo R, Apperley JF et al. Soluble receptor activator of nuclear factor {kappa}b ligand (RANKL)/osteoprotegerin (OPG) ratio predicts survival in multiple myeloma. Proposal for a novel prognostic index. Blood 2003; 102: Abe M, Hiura K, Wilde J et al. Osteoclasts enhance myeloma cell growth and survival via cell-cell contact: a vicious cycle between bone destruction and myeloma expansion. Blood 2004; 104: Silvestris F, Cafforio P, Calvani N, Dammacco F. Impaired osteoblastogenesis in myeloma bone disease: role of upregulated apoptosis by cytokines and malignant plasma cells. Br J Haematol 2004; 126: Tian E, Zhan F, Walker R et al. The role of the Wnt-signaling antagonist DKK1 in the development of osteolytic lesions in multiple myeloma. N Engl J Med 2003; 349: Robbiani DF, Chesi M, Bergsagel PL. Bone lesions in molecular subtypes of multiple myeloma. N Engl J Med 2004; 351: [letter]. 32. Croucher PI, De Hendrik R, Perry MJ et al. Zoledronic acid treatment of 5T2MM-bearing mice inhibits the development of myeloma bone disease: evidence for decreased osteolysis, tumor burden and angiogenesis, and increased survival. J Bone Miner Res 2003; 18: Gordon S, Helfrich MH, Sati HI et al. Pamidronate causes apoptosis of plasma cells in vivo in patients with multiple myeloma. Br J Haematol 2002; 119: Belch AR, Bergsagel DE, Wilson K et al. Effect of daily etidronate on the osteolysis of multiple myeloma. J Clin Oncol 1991; 9: Daragon A, Humez C, Michot C et al. Treatment of multiple myeloma with etidronate: results of a multicentre double-blind study. Eur J Med 1993; 2: Lahtinen R, Laakso M, Palva I et al. Randomised, placebo-controlled multicentre trial of clodronate in multiple myeloma. Lancet 1992; 340: Laakso M, Lahtinen R, Virkkunen P, Elomaa I. Subgroup and costbenefit analysis of the Finnish multicentre trial of clodronate in multiple myeloma. Br J Haematol 1994; 87: McCloskey EV, MacLennan IC, Drayson MT et al. A randomized trial of the effect of clodronate on skeletal morbidity in multiple myeloma. MRC Working Party on Leukaemia in Adults. Br J Haematol 1998; 100: McCloskey EV, Dunn JA, Kanis JA et al. Long-term follow-up of a prospective, double-blind, placebo-controlled randomized trial of clodronate in multiple myeloma. Br J Haematol 2001; 113: Brincker H, Westin J, Abildgaard N et al. Failure of oral pamidronate to reduce skeletal morbidity in multiple myeloma: a double-blind placebo-controlled trial. Danish-Swedish co-operative study group. Br J Haematol 1998; 101: Berenson JR, Lichtenstein A, Porter L et al. Efficacy of pamidronate in reducing skeletal events in patients with advanced multiple myeloma. Myeloma Aredia Study Group. N Engl J Med 1996; 334: Berenson JR, Lichtenstein A, Porter L et al. Long-term pamidronate treatment of advanced multiple myeloma patients reduces skeletal events. Myeloma Aredia Study Group. J Clin Oncol 1998; 16: Berenson JR, Rosen LS, Howell A et al. Zoledronic acid reduces skeletal-related events in patients with osteolytic metastases. Cancer 2001; 91: Rosen LS, Gordon D, Kaminski M et al. Zoledronic acid versus pamidronate in the treatment of skeletal metastases in patients with breast cancer or osteolytic lesions of multiple myeloma: a phase III, doubleblind, comparative trial. Cancer J 2001; 7: Rosen LS, Gordon D, Kaminski M et al. Long-term efficacy and safety of zoledronic acid compared with pamidronate disodium in the treatment of skeletal complications in patients with advanced multiple myeloma or breast carcinoma: a randomized, double-blind, multicenter, comparative trial. Cancer 2003; 98: Menssen HD, Sakalova A, Fontana A et al. Effects of long-term intravenous ibandronate therapy on skeletal-related events, survival, and bone resorption markers in patients with advanced multiple myeloma. J Clin Oncol 2002; 20: Djulbegovic B, Wheatley K, Ross J et al. Bisphosphonates in multiple myeloma. Cochrane Database Syst Rev 2002; 3: CD Terpos E, Viniou N, de la Fuente J et al. Pamidronate is superior to ibandronate in decreasing bone resorption, interleukin-6 and beta 2-microglobulin in multiple myeloma. Eur J Haematol 2003; 70: Berenson JR, Hillner BE, Kyle RA et al. American Society of Clinical Oncology clinical practice guidelines: the role of bisphosphonates in multiple myeloma. J Clin Oncol 2002; 20: Musto P, Falcone A, Sanpaolo G et al. Pamidronate reduces skeletal events but does not improve progression-free survival in early-stage untreated myeloma: results of a randomized trial. Leuk Lymphoma 2003; 44: Caparrotti G, Catalano L, Feo C et al. Perspective study on pamidronate in stage I multiple myeloma. Hematol J 2003; 4: [letter].

9 Terpos E, Palermos J, Tsionos K et al. Effect of pamidronate administration on markers of bone turnover and disease activity in multiple myeloma. Eur J Haematol 2000; 65: Terpos E, Palermos J, Viniou N et al. Pamidronate increases markers of bone formation in patients with multiple myeloma in plateau phase under interferon-alpha treatment. Calcif Tissue Int 2001; 68: Abildgaard N, Brixen K, Eriksen EF et al. Sequential analysis of biochemical markers of bone resorption and bone densitometry in multiple myeloma. Haematologica 2004; 89: Terpos E, de la Fuente J, Szydlo R et al. Tartrate-resistant acid phosphatase isoform 5b: a novel serum marker for monitoring bone disease in multiple myeloma. Int J Cancer 2003; 106: Terpos E, Politou M, Szydlo R et al. Autologous stem cell transplantation normalizes abnormal bone remodeling and srankl/osteoprotegerin ratio in patients with multiple myeloma. Leukemia 2004; 18: Dimopoulos MA, Moulopoulos LA, Maniatis A, Alexanian R. Solitary plasmacytoma of bone and asymptomatic multiple myeloma. Blood 2000; 96: McSweeney EN, Tobias JS, Blackman G et al. Double hemibody irradiation (DHBI) in the management of relapsed and primary chemoresistant multiple myeloma. Clin Oncol 1993; 5: Peh WC, Gilula LA. Percutaneous vertebroplasty: indications, contraindications, and technique. Br J Radiol 2003; 76: Dudeney S, Lieberman IH, Reinhardt MK, Hussein M. Kyphoplasty in the treatment of osteolytic vertebral compression fractures as a result of multiple myeloma. J Clin Oncol 2002; 20: Wedin R. Surgical treatment for pathologic fracture. Acta Orthop Scand Suppl 2001; 72: Pearse RN, Sordillo EM, Yaccoby S et al. Multiple myeloma disrupts the TRANCE/ osteoprotegerin cytokine axis to trigger bone destruction and promote tumor progression. Proc Natl Acad Sci USA 2001; 98: Vanderkerken K, De Leenheer E, Shipman C et al. Recombinant osteoprotegerin decreases tumor burden and increases survival in a murine model of multiple myeloma. Cancer Res 2003; 63: Body JJ, Greipp P, Coleman RE et al. A phase I study of AMGN- 0007, a recombinant osteoprotegerin construct, in patients with multiple myeloma or breast carcinoma related bone metastases. Cancer 2003; 97 (3 Suppl): Bekker PJ, Holloway DL, Rasmussen AS et al. A single-dose placebo-controlled study of AMG 162, a fully human monoclonal antibody to RANKL, in postmenopausal women. J Bone Miner Res 2004; 19: