Bisphosphonates: Preclinical ReviewJONATHAN R. GREEN

Novartis Pharma AG, Basel, Switzerland

Key Words. Bisphosphonates · Zoledronic acid · Bone resorption · Apoptosis · Antitumor effects · Mevalonate pathway

ABSTRACT

Bisphosphonates effectively inhibit osteoclast-medi-ated bone resorption and are integral in the treatment ofbenign and malignant bone diseases. The evolution of bis-phosphonates over the past 30 years has led to the devel-opment of nitrogen-containing bisphosphonates (N-BPs),which have a mechanism of action different from that ofthe nonnitrogen-containing bisphosphonates. Studiesconducted over the past decade have elucidated themechanism of action and pharmacologic properties of theN-BPs. N-BPs exert their effects on osteoclasts and tumorcells by inhibiting a key enzyme in the mevalonate path-way, farnesyl diphosphate synthase, thus preventing pro-tein prenylation and activation of intracellular signalingproteins such as Ras. Recent evidence suggests that N-BPs also induce production of a unique adenosinetriphosphate analogue (Apppi) that can directly induceapoptosis. Our increased understanding of the pharma-cologic effects of bisphosphonates is shedding light on the

mechanisms by which they exert antitumor effects. As aresult of their biochemical effects on protein prenylation,N-BPs induce caspase-dependent apoptosis, inhibitmatrix metalloproteinase activity, and downregulateααvββ3 and ααvββ5 integrins. In addition, zoledronic acid(Zometa®; Novartis Pharmaceuticals Corp.; EastHanover, NJ and Basel, Switzerland) exerts synergisticantitumor activity when combined with other anticanceragents. Zoledronic acid also inhibits tumor cell adhesionto the extracellular matrix and invasion throughMatrigel™ and has antiangiogenic activity. A growingbody of evidence from animal models demonstrates thatzoledronic acid and other bisphosphonates can reduceskeletal tumor burden and prevent metastasis to bone.Further studies are needed to fully elucidate these bio-chemical mechanisms and to determine if the antitumorpotential of bisphosphonates translates to the clinical set-ting. The Oncologist 2004;9(suppl 4):3-13

The Oncologist 2004;9(suppl 4):3-13 www.TheOncologist.com

Correspondence: Jonathan R. Green, Ph.D., Novartis Pharma AG, Klybeckstrasse 141, WKL-125.901, CH-4002 Basel,Switzerland. Telephone: 41-61-696-4415; Fax: 41-61-696-3849; e-mail: jonathan.green@pharma.novartis.com ReceivedJuly 19, 2004; accepted for publication August 3, 2004. ©AlphaMed Press 1083-7159/2004/$12.00/0

INTRODUCTION

Bisphosphonates are ideally suited for the treatment ofmetabolic bone disease because they bind avidly to the bonemineral at sites of active bone metabolism, where they

achieve therapeutic concentrations. The pioneering work ofFleisch and colleagues showed that bisphosphonates not onlyinhibit dissolution of hydroxyapatite crystals, but also affectosteoclast metabolism and function [1-3]. Bisphosphonates

TheOncologist®

LEARNING OBJECTIVES

After completing this course, the reader will be able to:

1. Describe the mechanism of action of first-generation and nitrogen-containing bisphosphonates.

2. Explain how the mechanism of action of the bisphosphonates might directly affect tumor growth.

3. Discuss how the bisphosphonates might be incorporated into both the prevention and treatment of cancer.

Access and take the CME test online and receive 1 hour of AMA PRA category 1 credit at CME.TheOncologist.comCMECME

are released during bone resorption and are internalized byosteoclasts, leading to inhibition of bone resorption andinduction of osteoclast apoptosis [4-6].

There is now extensive preclinical evidence that bis-phosphonates also have antitumor activity, as evidenced byreduced proliferation and viability of tumor cell lines invitro and reduced skeletal tumor burden and slower pro-gression of bone lesions in animal models. Several mecha-nisms have been proposed to explain these observations.Bisphosphonates may render the bone a less favorablemicroenvironment for tumor cell growth by reducingtumor-induced osteolysis and local release of growth fac-tors, and bisphosphonates may also have direct antitumoreffects. Bisphosphonates inhibit proliferation and induceapoptosis of a variety of human tumor cell lines in vitro [7-17]. Bisphosphonates also inhibit tumor cell adhesion to theextracellular bone matrix, inhibit invasion of tumor cellsthrough Matrigel™, and have antiangiogenic and immuno-modulatory activities. Consistent with these findings,bisphosphonates have been shown to inhibit the formationor progression of bone metastases and/or reduce skeletaltumor burden in a variety of animal models. The mecha-nisms responsible for the observed antitumor effects of bisphosphonates are beginning to be elucidated.

MECHANISM OF ACTION

Bisphosphonates accumulate in the mineralized bonematrix and are released during bone resorption. First-gener-ation, nonnitrogen-containing bisphosphonates are metabo-lized by osteoclasts to nonhydrolyzable cytotoxic ATPanalogues [18-21]. For example, clodronate is metabolizedto AppCC12p, which, at high concentrations, inhibits mito-chondrial ATP/adenosine diphosphate (ADP) translocase,thereby causing loss of the mitochondrial membrane poten-tial and direct induction of apoptosis [22-24]. The highaffinity of bisphosphonates for bone mineral and subse-quent uptake by activated osteoclasts during bone resorp-tion ensures that cytotoxic concentrations of thesemetabolites only accumulate within osteoclasts. However,they may also accumulate within tumor cells growing in thebone because tumor cells stimulate osteolysis.

In contrast, nitrogen-containing bisphosphonates (N-BPs),which include pamidronate (Aredia®; Novartis Pharma-ceuticals Corp.; East Hanover, NJ), alendronate (Fosamax®;Merck and Company, Inc.; West Point, PA), ibandronate(Bondronat®; Hoffmann-La Roche Inc.; Nutley, NJ), rise-dronate (Actonel®; Proctor and Gamble Pharmaceuticals, Inc.;Cincinnati, OH), and zoledronic acid (Zometa®; NovartisPharmaceuticals Corp.), affect osteoclast activity and survivalthrough a different mechanism. After internalization, thesecompounds inhibit a key enzyme, farnesyl diphosphonate

(FPP) synthase, in the biosynthetic mevalonate pathway (Fig.1) [5, 25-28]. As a result, N-BPs interfere with a variety ofcellular functions essential for the bone-resorbing activityand survival of osteoclasts [28, 29]. Several intermediates inthis pathway, including farnesyl pyrophosphate and geranyl-geranyl pyrophosphate, are required for the posttranslationalmodification (i.e., prenylation) of guanosine triphosphate-binding proteins such as Ras, Rho, and Rac [30]. These sig-naling molecules are involved in the regulation of cellproliferation, cell survival, and cytoskeletal organization [26,28, 31, 32]. In particular, inhibition of protein prenylationand Ras signaling within osteoclasts leads to defects in intra-cellular vesicle transport [33]. As a result, osteoclasts cannotform a tight-sealing zone or ruffled borders, which arerequired for bone resorption.

Recent studies have shown that N-BPs may have yetanother mechanism of action. As shown in Figure 1 [25, 34],N-BPs also induce production of an intracellular ATP ana-logue known as Apppi (triphosphoric acid 1-adenosin-5′-ylester 3-[3-methylbut-3-enyl] ester), which may directlyinduce apoptosis similar to AppCC12p (i.e., a metabolite ofclodronate) [34]. Inhibition of FPP synthase results in accu-mulation of isopentenyl diphosphonate, which can bemetabolized to Apppi. The ability of various bisphospho-nates (at 0.1-µM concentrations) to inhibit recombinanthuman FPP synthase activity (Fig. 2) [35] correlates wellwith their ability to induce production of Apppi in J774

4 Bisphosphonates: Preclinical Review

Statins HMG CoAreductase

HMG CoA

Mevalonate

Mevalonatepyrophosphate

Dimethylallylpyrophosphate

Isopentenylpyrophosphate+ AMP Apppi

Geranylpyrophosphate

FPP-synthaseN-BPs

Farnesylpyrophosphate Farnesol

Dolichol

Ubiquinone

Proteinfarnesylation

Squalene

Cholesterol

Geranyl-Geranylpyrophosphate

Protein geranyl-geranylation

Figure 1. Schematic representation of the mevalonate pathway andthe effects of nitrogen-containing bisphosphonates. Abbreviation:HMG CoA = 3-hydroxy-3-methylglutaryl coenzymeA. Adapted withpermission from Gober et al. [25].

cells. In this assay, clodronate serves as a negative controlwith little or no effect on FPP synthase activity, and itinduces no Apppi production. Of the N-BPs tested, zole-dronic acid demonstrated the highest potency in terms ofboth inhibition of FPP synthase activity (Fig. 2) [35] andproduction of Apppi [34]. The implication of this work isthat N-BPs, by inhibiting FPP synthase, can potentiallyinduce apoptosis of osteoclasts and tumor cells by at leasttwo distinct pathways.

It has been demonstrated that the potency of various N-BPs with respect to inhibition of FPP synthase activityalso correlates well with their observed potency in terms ofinhibiting bone resorption in vitro (Table 1) [35, 36]. Amongthe N-BPs tested, zoledronic acid was the most potentinhibitor of FPP synthase activity, followed by risedronateand ibandronate. The relative potencies of these three N-BPs

in the FPP synthase assay correlate closely with their relativepotencies in terms of inhibiting vitamin D3-induced calciumrelease from mouse calvaria cultures [36]. These data suggestthat inhibition of FPP synthase is a central mechanism bywhich N-BPs inhibit osteoclast-mediated bone resorption.Moreover, osteoclasts isolated from animals treated with N-BPs show a profound suppression of FPP synthase activity[37]. Therefore, FPP synthase appears to be one of the keymolecular targets of N-BPs. Inhibition of protein prenylationin cancer cells is also thought to be responsible for many ofthe observed antitumor effects of N-BPs.

EVIDENCE OF ANTITUMOR EFFECTS

In Vitro StudiesBisphosphonates have demonstrated direct antitumor

activity against a variety of tumor cell lines at concentrationsranging from 5-2,000 µM [38]. Bisphosphonates cause dose-and time-dependent inhibition of proliferation and induceapoptosis of myeloma, breast cancer, prostate cancer, pancre-atic cancer, lung cancer, and osteosarcoma cell lines in vitro[7-17, 39]. For example, in studies with the human MDA-MB-231 breast cancer cell line, clodronate, pamidronate, andzoledronic acid demonstrated dose-dependent effects on cellviability, with 50% inhibitory concentration (IC50) values of700, 40, and 15 µM, respectively [12]. This effect was notcaused by calcium chelation and appeared to be specific totumor cells. Similar results have been observed with otherbreast cancer cell lines. For instance, dose-dependentincreases in the proportion of apoptotic cells have beenobserved when Hs 578T cells were incubated with zoledronicacid [12]. Studies with MCF-7 cells have further shown thatthe effects of zoledronic acid on tumor cell viability and apop-tosis could be reversed by incubation with geranylgeraniol,indicating that inhibition of protein prenylation can inducetumor cell apoptosis [39]. In vitro studies with several differ-ent prostate cancer cell lines, including PC-3 and LNCaP23.1, have shown that zoledronic acid inhibits proliferation,induces apoptosis, and causes cell-cycle arrest in a dose-dependent manner [40, 41]. These are just a few examples ofthe antitumor activity of bisphosphonates against a wide vari-ety of human and murine tumor cell lines, and in everyinstance where several bisphosphonates have been tested,zoledronic acid has demonstrated the most potent activity.

In vitro studies have also shown that combining N-BPswith a variety of standard anticancer agents results in addi-tive or synergistic antitumor effects against a range oftumor cell lines [39, 42-49]. Jagdev et al. [39] were the firstto show that the combination of zoledronic acid (10 µM)plus paclitaxel (Taxol®; Bristol-Myers Squibb; Princeton,NJ; 2 nM) enhanced apoptosis of MCF-7 breast cancer cells

Green 5

FP

P s

ynth

ase

(% c

ontr

ol)

100

75

50

25

0Clodronate Ibandronate Risedronate Zoledronic acid

*

**

Figure 2. The effects of equimolar concentrations (0.1 µm) ofbisphosphonates on recombinant human FPP synthase activity.*p < 0.001. Adapted with permission from Dunford et al. [35].

Table 1. Relative potencies of bisphosphonates with respect to inhibition of FPP synthase and bone resorption activity

FPP synthase Bone resorptionBisphosphonate IC50 (µM)a IC50 (µM)b

Zoledronic acid 0.02 0.002

Risedronate 0.10 0.01

Ibandronate 0.31 0.02

Alendronate 0.50 0.05

Pamidronate 0.85 0.2

aMean values calculated from dose-response plots of inhibition ofFPP synthase in J774 cell homogenates based on three experiments.Data from Dunford et al. [35].bInhibition of 1,25-dihydroxyvitamin D3-induced calcium releasefrom mouse calvaria in vitro. Data represent means of severalexperiments. Data from Green et al. [36].

fourfold compared with either agent alone. Similar resultswere reported when ibandronate or zoledronic acid werecombined with epirubicin (Ellence®; Pharmacia andUpjohn; Portage, MI) plus docetaxel (Taxotere®; AventisPharmaceuticals Inc.; Bridgewater, NJ) [50]. In cultures ofprimary breast cancer cells, the concentrations of epirubicinplus docetaxel were gradually reduced to suboptimal levelsby serial dilution until there was little or no inhibition oftumor cell growth with these two drugs alone. However, theaddition of 15 µM ibandronate or zoledronic acid resulted in35% and 70% inhibitions of tumor cell growth, respectively.Combinations of zoledronic acid with taxanes also havedemonstrated synergistic antitumor activity against prostatecancer cell lines. The combination of zoledronic acid (12.5 or 25 µM) with suboptimal concentrations of doc-etaxel (≤0.1 ng/ml) demonstrated additive and dose-depen-dent cytotoxic effects on PC-3 prostate cancer cells at 72hours [45]. In addition, recent studies with DU-145 prostatecancer and MCF-18 breast cancer cell lines have demon-strated that the combination of zoledronic acid with thecyclooxygenase-2 inhibitor SC236 had additive inhibitoryeffects on tumor cell growth [48, 49]. These and many otherstudies have shown that N-BPs can enhance the cytotoxicand cytostatic effects of standard anticancer agents used totreat a variety of solid tumors. Recently, zoledronic acidalso has been shown to possess antileukemic activityagainst a Philadelphia chromosome-positive cell line, andthe combination of zoledronic acid with imatinib mesylate(Gleevec®; Novartis Pharmaceuticals Corp.) demonstratedsynergistic antileukemic activity [51]. These findings areintriguing given that Bcr-Abl stimulates the Ras signalingpathway.

Two recent studies with breast cancer cell lines havebegun to shed light on the possible mechanisms underlyingthe observed synergy between N-BPs and anticancer agents.In the first of these studies, MCF-7 breast cancer cells wereincubated with zoledronic acid (25 µM) and doxorubicin(Adriamycin®; Pharmacia and Upjohn; Peapack, NJ; 0.05µM), and the effects of different sequences and incubationperiods were tested [43]. The only combination that resultedin synergistic enhancement of apoptosis was doxorubicin for24 hours followed by zoledronic acid for 1 hour, whereaszoledronic acid administered either before doxorubicin ortogether with doxorubicin did not increase apoptosis (Fig. 3)[43]. This suggests that exposure to doxorubicin at a concen-tration that was not cytotoxic sensitized tumor cells to thecytotoxic effects of zoledronic acid. A similar but oppositeeffect of sequencing was also reported by the same groupwhen zoledronic acid (25 µM) was combined with tumornecrosis factor (TNF)-related apoptosis inducing ligand(TRAIL) at a concentration of 10 ng/ml. Among five combi-

nations tested, the only combination that produced synergis-tic apoptotic effects was zoledronic acid for 48 hours fol-lowed by TRAIL for 24 hours. This combination yielded14.7% apoptotic cells compared with 0.7% for zoledronicacid alone and 2.7% for TRAIL alone (p < 0.001). Theintriguing results of these in vitro studies suggest a novelapproach to enhance the synergy between N-BPs andchemotherapeutic agents in the clinical setting.

Animal ModelsThese in vitro findings are supported by data from many

animal models showing that the newer N-BPs can significantlyreduce the number and size of osteolytic lesions in tumor-bear-ing mice, reduce skeletal tumor burden, induce tumor cellapoptosis in bone lesions, reduce serum levels of tumor mark-ers, and prevent formation of bone metastases (Table 2) [9, 40,52-60]. Some models have even shown effects on visceraltumors and an improvement in survival of tumor-bearingmice, but these findings have been less consistently observed.Animal studies have focused on models of multiple myeloma,breast cancer, and prostate cancer. Using radiographic, histo-logic, and histomorphometric techniques, these studies haveshown that N-BPs can inhibit the formation or progression ofbone metastases and/or reduce skeletal tumor burden.Bisphosphonates have been administered either at the time oftumor cell inoculation (i.e., prevention setting) or after bonemetastases are established (i.e., treatment setting).

Several models of multiple myeloma have been estab-lished and have consistently shown reductions in tumor-induced osteolysis and skeletal tumor burden by N-BPs [52,53, 59]. For example, in the 5T2 model of murine myeloma,treatment with zoledronic acid (120 µg/kg s.c. twice weekly)after osteolytic lesions were established significantlyreduced development of new osteolytic lesions, reduced thebone surface occupied by osteoclasts and inhibited tumor-induced osteolysis, and reduced skeletal tumor burden asevidenced by serum paraprotein levels [53]. Moreover,treatment with zoledronic acid at the first sign of circulating

6 Bisphosphonates: Preclinical Review

15

10

5

0

Apo

ptos

is (

%)

Control DOX alone50 nM, 24 hours

ZOL alone25 µM, 1 hour

DOX +ZOL

DOX thenZOL

1.5% 1.7%

13.6%

ZOL thenDOX

Figure 3. Synergistic apoptotic effect of zoledronic acid (ZOL) incombination with doxorubicin (DOX) on MCF-7 breast cancer cellsin vitro. Adapted with permission from Neville-Webbe et al. [43].

paraprotein significantly prolonged disease-free survival inthose mice.

Numerous studies in breast cancer models have alsobeen reported. Zoledronic acid was shown to inhibit pro-gression of established bone metastases and development ofnew bone metastases in two models of breast cancer [54,55]. In nude mice injected with human MDA-MB-231breast cancer cells and allowed to develop osteolytic lesions,mice treated with zoledronic acid (0.2, 1.0, or 5.0 µg/days.c.) had significantly less radiographic bone lesion area, by>80%, than controls [54]. Ibandronate (1.0 µg/day) andalendronate (10 µg/day) resulted in nonsignificant reduc-tions in bone lesion area (65% and 55% reductions, respec-tively). These findings were confirmed in anotherindependent study using highly sensitive in vivo imaging ofMDA-MB-231 cells genetically engineered to express greenfluorescent protein [55]. Similar studies have been con-ducted with ibandronate in nude mice bearing MDA-MB-231 breast cancer cells, which typically form both adrenaland bone metastases. Treatment with ibandronate (4 µg/days.c.) inhibited the radiographic progression of establishedosteolytic lesions and decreased skeletal tumor burden com-pared with controls [9]. However, administration of iban-dronate at the time of tumor cell inoculation resulted in atwofold increase in adrenal tumor load [61]. In a murinebreast cancer model, treatment with zoledronic acid (5 µg/day) for 7 days after injection of 4T1 murine mam-mary tumor cells (i.e., prevention setting) markedlydecreased the formation of new bone metastases at day 28[54]. The observed decrease in radiographic bone lesionarea was accompanied by an increase in the number of apop-totic osteoclasts and apoptotic tumor cells in bone lesions.

Notably, the 4T1 mammary tumor model has providedthe first in vivo evidence of synergy between zoledronic acidand chemotherapy. In animals with established orthotopictumors, the combination of zoledronic acid (250 µg/kg) with20 mg/kg/day of oral UFT (a combination of uracil and tega-fur [4:1 molar ratio]) reduced skeletal tumor burden moreeffectively than either agent alone [56]. Similarly, the com-bination of ibandronate with doxorubicin (150 µg/day) hasbeen investigated in the MDA-MB-231 model and wasshown to have additive antitumor effects in bone [61].

Studies in a prostate cancer model have also recently beenreported. In those studies PC-3 and LuCaP 23.1 cells wereinjected directly into the tibia of mice [40]; PC-3 cells formosteolytic lesions, and LuCaP 23.1 cells form osteoblasticlesions. The treatment group received zoledronic acid (5 µgs.c. twice weekly) either at the time of tumor cell injection orafter tibial tumors were established (7 days for PC-3 tumorsand 33 days for LuCaP 23.1 tumors). Treatment with zole-dronic acid significantly inhibited growth of both osteolyticand osteoblastic metastases by radiographic analysis (Fig. 4)[40] and also reduced skeletal tumor burden as evidenced bya significant decrease in serum levels of prostate-specificantigen in animals bearing LuCaP 23.1 tumors [40]. Theobserved reduction in serum prostate-specific antigen levelsprovides compelling direct evidence of the antitumor activityof zoledronic acid in this animal model.

The potential of zoledronic acid to prevent bone metas-tasis was also demonstrated in an animal model of prostatecancer [62]. In that model, mice were injected with PC-3cells, and the incidence of bone metastases was studied innormal mice and in mice that were rendered androgen defi-cient by surgical castration. Androgen-deficient mice devel-

Green 7

Table 2. Animal models demonstrating antitumor effects of zoledronic acid

Study Tumor cells Bisphosphonate Dose and schedule

Yaccoby et al. [52] Primary human myeloma cells Zoledronic acid 100 µg/kg/week s.c.

Pamidronate 1.3 mg/kg every 2 weeks s.c.

Croucher et al. [53] 5T2 murine myeloma Zoledronic acid 120 µg/kg twice weekly s.c.

Green et al. [54] MDA-MB-231 human breast cancer Zoledronic acid 0.2-5.0 µg/day s.c.

Ibandronate 1.0 µg/day s.c.

Alendronate 10 µg/day s.c.

Peyruchaud et al. [55] MDA-MB-231 human breast cancer Zoledronic acid 3 µg/day × 12 days s.c.

Hiraga et al. [56] 4T1 murine mammary carcinoma Zoledronic acid + 250 µg/kg

UFT 20 mg/kg/day

Hiraga et al. [9] MDA-MB-231 human breast cancer Ibandronate 4 µg/day s.c.

Nobuyuki et al. [57] 4T1 murine mammary carcinoma Zoledronic acid 0.5 or 5 µg every 4 days i.v.

Corey et al. [40] PC-3 and LuCaP 23.1 prostate cancer Zoledronic acid 5 µg/mouse twice weekly s.c.

Adapted with permission from Green [58].

oped significantly more bone metastases than did intact con-trol mice. This strongly suggests that excess bone resorptioncaused by androgen ablation can stimulate metastasis of PC-3 cells to the bone. This is consistent with current hypothesesof tumor cell metastasis to bone. Moreover, daily treatmentwith zoledronic acid significantly reduced the incidence ofbone metastases in both normal and castrated mice.

Although the majority of animal models suggest that theprimary antitumor effect of bisphosphonates is manifested inthe bone, where they reach the highest concentrations, pre-liminary data from the 4T1 mammary tumor model havedemonstrated that zoledronic acid can also inhibit visceralmetastases [57]. 4T1 cells expressing luciferase were used toquantitate tumor burden in mice. After tumor cell inocula-tion, mice were treated with zoledronic acid (5.0 µg every 4 days), which resulted in lower tumor burdens than in con-trols not only in bone but also in the liver and lungs [57, 63].Treatment with zoledronic acid also prolonged survival oftumor-bearing mice. In the same model, ibandronate (4 µg/day s.c.) had no effect on lung or liver metastases or onsurvival [61, 63]. As mentioned above, treatment of micebearing 5T2 myeloma cells with zoledronic acid (120 µg/kgs.c. twice weekly) also resulted in a significantly longer dis-ease-free survival time (median, 47 days versus 35 days foruntreated controls; p < 0.01) [53].

These animal models provide convincing evidence of thepotential of bisphosphonates, particularly the more potent N-BPs, to reduce tumor burden in bone and inhibit formationand progression of bone metastases in a variety of tumor mod-els. It is less clear whether bisphosphonates have antitumoractivity beyond the bone, but it appears that zoledronic acidmay be sufficiently potent to inhibit extraskeletal tumor cellgrowth and to prolong survival in some animal models.

MECHANISMS OF ANTITUMOR EFFECTS

The precise mechanisms responsible for the antitumoreffects of bisphosphonates are beginning to be elucidated.Bisphosphonates appear to make the bone a less favorablesite for tumor cell growth via the inhibition of osteoclast-mediated bone resorption and osteoclastogenesis, therebyreducing the release of growth factors that stimulate tumorgrowth in bone. In addition, bisphosphonates directlyinhibit tumor cell growth and survival and the ability oftumor cells to colonize the bone. Either or both of thesemechanisms may be at work.

ApoptosisOne of the primary mechanisms responsible for the

direct antitumor activity of bisphosphonates is inductionof tumor cell apoptosis. Both N-BPs and non-N-BPsappear to induce apoptosis of osteoclasts and tumor cellsby activation of caspases [9, 12-14, 16, 17, 64, 65]. Onemechanism by which bisphosphonates induce apoptosisis through production of ATP analogues (either as directmetabolites or as a result of inhibition of the mevalonatepathway), which can disrupt mitochondrial ATP/ADPtranslocase. A recent study investigating the mechanismby which zoledronic acid induced apoptosis in humanbreast cancer cell lines (MDA-MB-231 and MCF-7)indicated that this response was associated withcytochrome c release from the mitochondria and subse-quent caspase-3 activation [66]. It appears that N-BPsmay induce cytochrome c release by modulating expres-sion of Bcl-2, a key antiapoptotic regulatory protein[66]. These events may be precipitated by inhibition ofRas activation, which requires protein prenylation(specifically farnesylation) [66].

8 Bisphosphonates: Preclinical Review

Figure 4. The effects ofzoledronic acid (5 µg s.c.twice weekly) on tumorvolume (assessed radi-ographically) in mice withtibial tumors from PC-3and LuCaP 23.1 prostatecancer cell lines. Zole-dronic acid was adminis-tered either at the time oftumor cell injection (i.e.,prevention) or after tibialtumors were established(i.e., treatment), whichwas 7 days postinjectionfor PC-3 and 33 dayspostinjection for LuCaP.*p < 0.03; **p < 0.001.Adapted with permissionfrom Corey et al. [40].

80

60

40

20

0

Tum

or v

olum

e (%

tiss

ue v

olum

e)

Vehicle Prevent Treat Vehicle Prevent Treat

Zoledronic acid: 5 µg/mouse (twice a week)

* ***

**

PC-3 (lytic) LuCaP 23.1 (blastic)

Inhibition of Tumor Cell Adhesion and Invasion of theExtracellular Bone Matrix

Bisphosphonates have also been shown to inhibit adhe-sion of tumor cells to extracellular matrix (ECM) proteins andto inhibit the process of tumor cell invasion and metastasis[67-70]. Using an in vitro Matrigel™-based invasion assay,Boissier et al. have shown that bisphosphonates inhibit theability of human breast and prostate cancer cells to invade theECM [67]. In this assay, zoledronic acid and ibandronatecaused dose-dependent inhibition of cell invasion through theMatrigel™ at extremely low concentrations (≤10–10 M) that didnot inhibit tumor cell motility or induce significant apoptosis.Clodronate was approximately six orders of magnitude lesspotent in the same assay. Furthermore, the combination ofibandronate with taxanes enhanced the inhibitory effect onMDA-MB-231 cell invasion [69].

One contributory mechanism may be the inhibition ofmatrix metalloproteinase (MMP) activity, which is necessaryfor tumor cell invasion of the ECM [67, 71, 72]. Bisphos-phonates have been shown to inhibit the activity of MMPsproduced by tumor cell lines, and this seems to correlate withreduced invasiveness in the Matrigel™ assay [71, 72]. Forexample, zoledronic acid was shown to inhibit the productionof MMP-2 and MMP-9 by PC-3 cells [40]. These data sug-gest a potential mechanism by which N-BPs may inhibit tumorcell invasion of the bone but cannot explain the apparentdependence of this effect on protein prenylation.

Recent studies suggest that inhibition of tumor celladhesion to ECM proteins and invasion through Matrigel™

is dependent on inhibition of protein prenylation. In onestudy, zoledronic acid was shown to dose-dependentlyinhibit adhesion of MCF-7 and MDA-MB-231 cells to avariety of matrix proteins (Fig. 5) [73], and this inhibitoryeffect was overcome by addition of either farnesol or ger-anylgeraniol or by the addition of a broad spectrum caspaseinhibitor [73]. Similar findings have been reported for alen-dronate. The inhibitory effect of alendronate on tumor cellinvasion through Matrigel™ was reversed by the addition of geranylgeraniol andtrans-trans-farnesol

[74]. Therefore, inhibition of the mevalonate pathway andinduction of caspase activity are important for theinhibitory effects of N-BPs on tumor cell adhesion to theECM and on invasiveness. Further, it has been shown thatan activating Ras mutation enhances adhesion of a normalbreast epithelial cell line to ECM proteins, suggesting thatincreased Ras activation in response to growth factor recep-tor signaling may increase the metastatic potential of breastcancer cells [73]. Thus, by inhibiting protein prenylationand Ras signaling, zoledronic acid should reduce themetastatic potential of tumor cells.

Antiangiogenic EffectsIn vitro and in vivo studies have further demonstrated that

zoledronic acid has antiangiogenic effects. In vitro assayswith human umbilical vein endothelial cells (HUVECs) haveshown that zoledronic acid dose-dependently inhibited theproliferation of HUVECs induced by fetal calf serum andbasic fibroblast growth factor (bFGF), and these findings havebeen confirmed in vivo [75]. Systemic administration of zole-dronic acid to mice resulted in potent inhibition of angiogen-esis induced by s.c. implants impregnated with bFGF, witha dose of 3 µg/kg producing a 50% efficacy (ED50) [75]. Ithas also been reported that zoledronic acid can reduce bone-tumor-associated angiogenesis in the murine 5T2 myelomamodel [53]. In another series of experiments, zoledronicacid, as well as ibandronate, risedronate, and clodronate,inhibited formation of capillary-like tubules by HUVECs invitro [76]. In vivo, zoledronic acid and ibandronate, but notclodronate, decreased revascularization (as measured byvessel area) of the ventral prostate gland in castrated ratstreated with testosterone [76].

The inhibitory effect of zoledronic acid on endothelialcell adhesion and migration appears to be mediated, at leastin part, by modulation of integrins (e.g., αvβ3 and αvβ5) thatare involved in angiogenesis [77, 78]. Interestingly, αvβ3

integrin is also required for osteoclasts to adhere tightly tothe bone and form resorption lacunae during active bone

Green 9

Figure 5. Percent adherentcells versus controls whenMDA-MB-231 humanbreast cancer cells wereincubated with zoledronicacid for 24 hours (0.1 or100 µM) before culture on plates coated with vari-ous extracellular matrixproteins. Adapted with permission from Pickeringet al. [73].

Control

0.1 µM Zoledronic acid

100 µM Zoledronic acid24-hour treatment

Collagen IVitronectinFibronectinLaminin

Adh

eren

t cel

ls (

% c

ontr

ol)

100

50

0

resorption, and αvβ3 expression confers on tumor cells agreater propensity to metastasize to bone [79]. In fact, asmall molecule inhibitor of αvβ3 was recently shown toeffectively prevent metastasis of MDA-MB-435 breast can-cer cells to bone [80]. Therefore, effects on αvβ3 could havepleiotropic effects on bone resorption and tumor metastasis.In addition, it has recently been reported that zoledronicacid decreases the survival of HUVECs by sensitizing themto TNF-induced programmed cell death [78]. Zoledronicacid also appears to modulate serum levels of proangio-genic growth factors such as vascular endothelial growthfactor and bFGF in cancer patients [81]. These studies sug-gest a variety of potential mechanisms to account for theobserved antiangiogenic effects of bisphosphonates.

CONCLUSIONS AND FUTURE DIRECTIONS

There is now extensive in vitro and in vivo preclinicalevidence that bisphosphonates, particularly the morepotent N-BPs, have antitumor activity and can reduce

skeletal tumor burden. The evidence that they have antitu-mor activity outside the bone is more tenuous. A varietyof potential mechanisms to explain these observed antitu-mor effects have been proposed, including indirect effectson tumor cell growth in bone via inhibition of boneresorption and osteoclastogenesis. In addition, bisphos-phonates clearly have the potential to directly induceapoptosis of tumor cells, inhibit tumor cell adhesion to theECM, reduce the metastatic potential of tumor cells, andinhibit angiogenesis. Further research is ongoing to fullyelucidate the molecular mechanisms involved and todetermine the most effective dose and schedule of bispho-sphonates to maximize their antitumor potential in the clin-ical setting, either alone or in combination with standardantineoplastic agents.

ACKNOWLEDGMENT

Jonathan Green is a full time employee of NovartisPharma AG and holds stock in the company.

10 Bisphosphonates: Preclinical Review

REFERENCES

1 Fleisch H. Development of bisphosphonates. Breast CancerRes 2002;4:30-34.

2 Russell RG, Croucher PI, Rogers MJ. Bisphosphonates: phar-macology, mechanisms of action and clinical uses. OsteoporosInt 1999;9(suppl 2):S66-S80.

3 Stronski SA, Bettschen-Camin L, Wetterwald A et al. Bisphos-phonates inhibit 1,25-dihydroxyvitamin D3-induced increase ofosteocalcin in plasma of rats in vivo and in culture medium of ratcalvaria in vitro. Calcif Tissue Int 1988;42:248-254.

4 Flanagan AM, Chambers TJ. Inhibition of bone resorption bybisphosphonates: interactions between bisphosphonates,osteoclasts, and bone. Calcif Tissue Int 1991;49:407-415.

5 Rogers MJ, Gordon S, Benford HL et al. Cellular and molec-ular mechanisms of action of bisphosphonates. Cancer2000;88(suppl 12):2961-2978.

6 Sato M, Grasser W, Endo N et al. Bisphosphonate action.Alendronate localization in rat bone and effects on osteoclastultrastructure. J Clin Invest 1991;88:2095-2105.

7 Aparicio A, Gardner A, Tu Y et al. In vitro cytoreductiveeffects on multiple myeloma cells induced by bisphosphonates.Leukemia 1998;12:220-229.

8 Derenne S, Amiot M, Barille S et al. Zoledronate is a potentinhibitor of myeloma cell growth and secretion of IL-6 andMMP-1 by the tumoral environment. J Bone Miner Res1999;14:2048-2056.

9 Hiraga T, Williams PJ, Mundy GR et al. The bisphosphonateibandronate promotes apoptosis in MDA-MB-231 human breastcancer cells in bone metastases. Cancer Res 2001;61:4418-4424.

10 Lee MV, Fong EM, Singer FR et al. Bisphosphonate treat-ment inhibits the growth of prostate cancer cells. Cancer Res2001;61:2602-2608.

11 Mackie PS, Fisher JL, Zhou H et al. Bisphosphonates regulatecell growth and gene expression in the UMR 106-01 clonal ratosteosarcoma cell line. Br J Cancer 2001;84:951-958.

12 Senaratne SG, Pirianov G, Mansi JL et al. Bisphosphonatesinduce apoptosis in human breast cancer cell lines. Br J Cancer2000;82:1459-1468.

13 Shipman CM, Rogers MJ, Apperley JF et al. Bisphosphonatesinduce apoptosis in human myeloma cell lines: a novel anti-tumour activity. Br J Haematol 1997;98:665-672.

14 Shipman CM, Croucher PI, Russell RG et al. The bisphos-phonate incadronate (YM175) causes apoptosis of humanmyeloma cells in vitro by inhibiting the mevalonate pathway.Cancer Res 1998;58:5294-5297.

15 Sonnemann J, Eckervogt V, Truckenbrod B et al. The bisphos-phonate pamidronate is a potent inhibitor of human osteosar-coma cell growth in vitro. Anticancer Drugs 2001;12:459-465.

16 Takahashi R, Shimazaki C, Inaba T et al. A newly developedbisphosphonate, YM529, is a potent apoptosis inducer ofhuman myeloma cells. Leuk Res 2001;25:77-83.

17 Tassone P, Tagliaferri P, Viscomi C et al. Zoledronic acidinduces antiproliferative and apoptotic effects in human pan-creatic cancer cells in vitro. Br J Cancer 2003;88:1971-1978.

18 Rogers MJ, Russell RG, Blackburn GM et al. Metabolism ofhalogenated bisphosphonates by the cellular slime mouldDictyostelium discoideum. Biochem Biophys Res Commun1992;189:414-423.

19 Rogers MJ, Ji X, Russell RG et al. Incorporation of bisphospho-nates into adenine nucleotides by amoebae of the cellular slimemould Dictyostelium discoideum. Biochem J 1994;303:303-311.

20 Pelorgeas S, Martin JB, Satre M. Cytotoxicity of dichloro-methane diphosphonate and of 1-hydroxyethane-1,1-diphos-phonate in the amoebae of the slime mould Dictyosteliumdiscoideum. A 31P NMR study. Biochem Pharmacol1992;44:2157-2163.

21 Frith JC, Mönkkönen J, Blackburn GM et al. Clodronate andliposome-encapsulated clodronate are metabolized to a toxicATP analog, adenosine 5′-(beta, gamma-dichloromethylene)triphosphate, by mammalian cells in vitro. J Bone Miner Res1997;12:1358-1367.

22 Lehenkari PP, Kellinsalmi M, Napankangas JP et al. Furtherinsight into mechanism of action of clodronate: inhibition ofmitochondrial ADP/ATP translocase by a nonhydrolyzable, ade-nine-containing metabolite. Mol Pharmacol 2002;61:1255-1262.

23 Rogers MJ, Watts DJ, Russell RGG. Overview of bisphos-phonates. Cancer 1997;80(suppl 8):1652-1660.

24 Selander KS, Monkkonen J, Karhukorpi EK et al. Character-istics of clodronate-induced apoptosis in osteoclasts andmacrophages. Mol Pharmacol 1996;50:1127-1138.

25 Gober HJ, Kistowska M, Angman L et al. Human T cellreceptor gammadelta cells recognize endogenous mevalonatemetabolites in tumor cells. J Exp Med 2003;197:163-168.

26 Luckman SP, Hughes DE, Coxon FP et al. Nitrogen-contain-ing bisphosphonates inhibit the mevalonate pathway and pre-vent post-translational prenylation of GTP-binding proteins,including Ras. J Bone Miner Res 1998;13:581-589.

27 Rogers MJ, Frith JC, Luckman SP et al. Molecular mecha-nisms of action of bisphosphonates. Bone 1999;24(suppl5):73S-79S.

28 Russell RG, Rogers MJ, Frith JC et al. The pharmacology ofbisphosphonates and new insights into their mechanisms ofaction. J Bone Miner Res 1999;14(suppl 2):53-65.

29 Hughes DE, Wright KR, Uy HL et al. Bisphosphonates promoteapoptosis in murine osteoclasts in vitro and in vivo. J BoneMiner Res 1995;10:1478-1487.

30 Luckman SP, Coxon FP, Ebetino FH et al. Heterocycle-con-taining bisphosphonates cause apoptosis and inhibit boneresorption by preventing protein prenylation: evidence fromstructure-activity relationships in J774 macrophages. J BoneMiner Res 1998;13:1668-1678.

31 Oliff A. Farnesyltransferase inhibitors: targeting the molecularbasis of cancer. Biochim Biophys Acta 1999;1423:C19-C30.

32 Zhang FL, Casey PJ. Protein prenylation: molecular mecha-nisms and functional consequences. Annu Rev Biochem1996;65:241-269.

33 Alakangas A, Selander K, Mulari M et al. Alendronate dis-turbs vesicular trafficking in osteoclasts. Calcif Tissue Int2002;70:40-47.

34 Mönkkönen H, Lehenkari PP, Kellinsalmi M et al. A newmechanism of action for bisphosphonates: apppi dedicatedcytotoxicity of N-BPs. Bone 2004;34:S66-S67.

35 Dunford JE, Thompson K, Coxon FP et al. Structure-activity rela-tionships for inhibition of farnesyl diphosphate synthase in vitro

and inhibition of bone resorption in vivo by nitrogen-containingbisphosphonates. J Pharmacol Exp Ther 2001;296:235-242.

36 Green JR, Müller K, Jaeggi KA. Preclinical pharmacology ofCGP 42'446, a new, potent, heterocyclic bisphosphonatecompound. J Bone Miner Res 1994;9:745-751.

37 Fisher JE, Rodan GA, Reszka AA. In vivo effects of bisphos-phonates on the osteoclast mevalonate pathway. Endocrinology2000;141:4793-4796.

38 Clezardin P. The antitumor potential of bisphosphonates.Semin Oncol 2002;29(suppl 21):33-42.

39 Jagdev SP, Coleman RE, Shipman CM et al. The bisphos-phonate, zoledronic acid, induces apoptosis of breast cancercells: evidence for synergy with paclitaxel. Br J Cancer2001;84:1126-1134.

40 Corey E, Brown LG, Quinn JE et al. Zoledronic acid exhibitsinhibitory effects on osteoblastic and osteolytic metastases ofprostate cancer. Clin Cancer Res 2003;9:295-306. Erratum in:Clin Cancer Res 2003;9:1574-1575.

41 Oades GM, Senaratne SG, Clarke IA et al. Nitrogen contain-ing bisphosphonates induce apoptosis and inhibit the meval-onate pathway, impairing Ras membrane localization inprostate cancer cells. J Urol 2003;170:246-252.

42 Matsumoto S, Kimura S, Segawa H et al. Efficacy of com-bining the third generation bisphosphonate, zoledronate withimatinib mesylate in suppressing small cell lung cancer cellline proliferation. Proc Am Soc Clin Oncol 2003;22:684.

43 Neville-Webbe HL, Evans CA, Coleman RE et al. Zoledronicacid (ZOL) in combination with doxorubicin (DOX) has syn-ergistic effects on apoptosis of breast cancer cells (BCCs) invitro [abstract]. Presented at the IVth International Conferenceon Cancer-Induced Bone Diseases, December 7-9, 2003, SanAntonio, TX.

44 Tassone P, Forciniti S, Galea E et al. Growth inhibition and syn-ergistic induction of apoptosis by zoledronate and dexametha-sone in human myeloma cell lines. Leukemia 2000;14:841-844.

45 Ullén A, Lennartsson L, Hjelm-Eriksson M et al. Additive/syn-ergistic anti-tumoral effects on prostate cancer cells in vitro fol-lowing treatment with a combination of docetaxel andzoledronate [poster]. Presented at the European WinterOncology Conference (EWOC-8), January 19-24, 2003, Flims,Switzerland.

46 Ullén A, Lennartsson L, Hjelm-Eriksson M et al. Additive/syn-ergistic anti-tumoral effect on prostate cancer cells in vitro fol-lowing treatment with a combination of gemcitabine andzoledronic acid. Proc Am Soc Clin Oncol 2003;22:432.

47 Vogt U, Wassmann K, Bosse U et al. Enhancement of breasttumor growth inhibition (BTGI) in vitro through combination ofCMF, epirubicin/cyclophosphamide (EC), epirubicin/paclitaxel(ET) and epirubicin/docetaxel (EDoc) with ibandronate (IB) orzoledronic acid (ZOL). Proc Am Soc Clin Oncol 2003;22:55.

48 Witters L, Crispino J, Javeed M et al. Inhibition of growth ofa human prostate cancer cell line with the combination ofzoledronic acid and a COX-2 inhibitor. Proc Am Soc ClinOncol 2002;21:5b.

Green 11

49 Witters LM, Crispino J, Fraterrigo T et al. Effect of the com-bination of docetaxel, zoledronic acid, and a COX-2 inhibitoron the growth of human breast cancer cell lines. Am J ClinOncol 2003;26:S92-S97.

50 Schlotter CM, Vogt U, Bosse U et al. Enhancement of breasttumor growth inhibition (BTGI) in vitro through combinationof CMF, epirubicin/cyclophosphamide (EC), and epiru-bicin/paclitaxel (ET) with ibandronate (IB) or zoledronic acid(ZOL). J Bone Miner Res 2001;16(suppl 1):S191.

51 Kuroda J, Kimura S, Segawa H et al. The third-generationbisphosphonate zoledronate synergistically augments theanti-Ph+ leukemia activity of imatinib mesylate. Blood2003;102:2229-2235.

52 Yaccoby S, Pearse RN, Johnson CL et al. Myeloma interactswith the bone marrow microenvironment to induce osteoclasto-genesis and is dependent on osteoclast activity. Br J Haematol2002;116:278-290.

53 Croucher PI, De Raeve H, Perry MJ et al. Zoledronic acid treat-ment of 5T2MM-bearing mice inhibits the development ofmyeloma bone disease: evidence for decreased osteolysis,tumor burden and angiogenesis, and increased survival. J BoneMiner Res 2003;18:482-492.

54 Green J, Gschaidmeier H, Yoneda T et al. Zoledronic acidpotently inhibits tumour-induced osteolysis in two models ofbreast cancer metastasis to bone. Ann Oncol 2000;11(suppl 4):14.

55 Peyruchaud O, Winding B, Pecheur I et al. Early detection ofbone metastases in a murine model using fluorescent humanbreast cancer cells: application to the use of the bisphosphonatezoledronic acid in the treatment of osteolytic lesions. J BoneMiner Res 2001;16:2027-2034.

56 Hiraga T, Ueda A, Tamura D et al. Effects of oral UFT com-bined with or without zoledronic acid on bone metastasis in the4T1/luc mouse breast cancer. Int J Cancer 2003;106:973-979.

57 Nobuyuki H, Hiraga T, Williams PJ et al. The bisphospho-nate zoledronic acid inhibits metastases to bone and liver withsuppression of osteopontin production in mouse mammarytumor. J Bone Miner Res 2001;16(suppl 1):S191.

58 Green JR. Bisphosphonates in cancer therapy. Curr Opin Oncol2002;14:609-615.

59 Cruz JC, Alsina M, Craig F et al. Ibandronate decreases bonedisease development and osteoclast stimulatory activity in an invivo model of human myeloma. Exp Hematol 2001;29:441-447.

60 Alvarez E, Westmore M, Galvin RJ et al. Properties of bisphos-phonates in the 13762 rat mammary carcinoma model oftumor-induced bone resorption. Clin Cancer Res 2003;9:5705-5713.

61 Michigami T, Hiraga T, Williams PJ et al. The effect of thebisphosphonate ibandronate on breast cancer metastasis tovisceral organs. Breast Cancer Res Treat 2002;75:249-258.

62 Padalecki SS, Carreon M, Grubbs B et al. Androgen deprivationenhances bone loss and prostate cancer metastases to bone: prevention by zoledronic acid. Oncology 2003;17(suppl):32.

63 Yoneda T, Hashimoto N, Hiraga T. Bisphosphonate actionson cancer. Calcif Tissue Int 2003;73:315-318.

64 Benford HL, McGowan NWA, Helfrich MH et al.Visualization of bisphosphonate-induced caspase-3 activityin apoptotic osteoclasts in vitro. Bone 2001;28:465-473.

65 Fromigue O, Lagneaux L, Body JJ. Bisphosphonates inducebreast cancer cell death in vitro. J Bone Miner Res2000;15:2211-2221.

66 Senaratne SG, Mansi JL, Colston KW. The bisphosphonatezoledronic acid impairs Ras membrane [correction of impairsmembrane] localisation and induces cytochrome c release inbreast cancer cells. Br J Cancer 2002;86:1479-1486. Erratumin: Br J Cancer 2002;87:1340.

67 Boissier S, Ferreras M, Peyruchaud O et al. Bisphosphonatesinhibit breast and prostate carcinoma cell invasion, an earlyevent in the formation of bone metastases. Cancer Res2000;60:2949-2954.

68 Boissier S, Magnetto S, Frappart L et al. Bisphosphonatesinhibit prostate and breast carcinoma cell adhesion to unmin-eralized and mineralized bone extracellular matrices. CancerRes 1997;57:3890-3894.

69 Magnetto S, Boissier S, Delmas PD et al. Additive antitumoractivities of taxoids in combination with the bisphosphonateibandronate against invasion and adhesion of human breastcarcinoma cells to bone. Int J Cancer 1999;83:263-269.

70 van der Pluijm G, Vloedgraven H, van Beek E et al.Bisphosphonates inhibit the adhesion of breast cancer cells tobone matrices in vitro. J Clin Invest 1996;98:698-705.

71 Teronen O, Heikkila P, Konttinen YT et al. MMP inhibitionand downregulation by bisphosphonates. Ann NY Acad Sci1999;878:453-465.

72 Heikkila P, Teronen O, Moilanen M et al. Bisphosphonatesinhibit stromelysin-1 (MMP-3), matrix metalloelastase(MMP-12), collagenase-3 (MMP-13) and enamelysin (MMP-20), but not urokinase-type plasminogen activator, and dimin-ish invasion and migration of human malignant andendothelial cell lines. Anticancer Drugs 2002;13:245-254.

73 Pickering LM, Mansi JL, Colston KW. Adhesion of breastcancer cells to extracellular matrices is inhibited by zole-dronic acid and enhanced by aberrant Ras signalling. ProcAm Soc Clin Oncol 2003;22:863.

74 Virtanen SS, Väänänen HK, Härkönen PL et al. Alendronateinhibits invasion of PC-3 prostate cancer cells by affectingthe mevalonate pathway. Cancer Res 2002;62:2708-2714.

75 Wood J, Bonjean K, Ruetz S et al. Novel antiangiogeniceffects of the bisphosphonate compound zoledronic acid. J Pharmacol Exp Ther 2002;302:1055-1061.

76 Fournier P, Boissier S, Filleur S et al. Bisphosphonatesinhibit angiogenesis in vitro and testosterone-stimulated vas-cular regrowth in the ventral prostate in castrated rats. CancerRes 2002;62:6538-6544.

77 Bonjean K, Bellahcene A, Locigno R et al. Zoledronate modu-lates endothelial cell surface receptors involved in angiogenesis.Proc Am Assoc Cancer Res 2001;42:106.

78 Bezzi M, Hasmim M, Bieler G et al. Zoledronate sensitizesendothelial cells to tumor necrosis factor-induced programmed

12 Bisphosphonates: Preclinical Review

cell death. Evidence for the suppression of sustained activa-tion of focal adhesion kinase and protein kinase B/Akt. J BiolChem 2003;278:43603-43614.

79 Pécheur I, Peyruchaud O, Serre CM et al. Integrinalpha(v)beta3 expression confers on tumor cells a greaterpropensity to metastasize to bone. FASEB J 2002;16:1266-1268.

80 Welch DR, Harms JF, Samant RS et al. The small moleculeαvβ3 antagonist (S247) inhibits MDA-MB-435 breast cancermetastasis to bone. Oncology 2003;17(suppl):18.

81 Santini D, Vincenzi B, Dicuonzo G et al. Zoledronic acidinduces significant and long-lasting modifications of circulat-ing angiogenic factors in cancer patients. Clin Cancer Res2003;9:2893-2897.

Green 13