Pain 100 (2002) 219–229
`
`Research papers
`
`PAIN
`
`www.elsevier.com/locate/pain
`
`Disease modifying and anti-nociceptive effects of the bisphosphonate,
`zoledronic acid in a model of bone cancer pain
`
`Katharine Walkera, Stephen J. Medhurstb, Bruce L. Kiddc, Markus Glattd,
`Mick Bowesb, Sadhana Patelb, Kara McNairb, Adam Kesinglandb, Jonathan Greend, Otto Chanc,
`Alyson J. Foxb, Laszlo A. Urbanb,*
`
`aPurdue Biopharma LP, 201 College Road East, Princeton, NJ 08540, USA
`bNovartis Institute for Medical Sciences, 5 Gower Place, London WC1E 6BN, UK
`cW. Harvey Research Institute, St Bartholomew’s and Royal London School of Medicine, Charterhouse Square, London, UK
`dNovartis Pharma AG, Basel, Switzerland
`
`Received 11 July 2001; received in revised form 22 October 2001; accepted 29 January 2002
`
`Abstract
`
`Inoculation of syngeneic MRMT-1 mammary tumour cells into one tibia of female rats produced tumour growth within the bone associated
`with a reduction in bone mineral density (BMD) and bone mineral content (BMC), severe radiological signs of bone destruction, together
`with the development of behavioural mechanical allodynia and hyperalgesia. Histological and radiological examination showed that chronic
`treatment with the bisphosphonate, zoledronic acid (30 mg/kg, s.c.), for 19 days significantly inhibited tumour proliferation and preserved the
`cortical and trabecular bone structure. In addition, BMD and BMC were preserved and a dramatic reduction of tartrate resistant acid
`phosphatase-positive polykaryocytes (osteoclasts) was observed. In behavioural tests, chronic treatment with zoledronic acid but not the
`significantly less effective bisphosphonate, pamidronate, or the selective COX-2 inhibitor, celebrex, attenuated mechanical allodynia and
`hyperalgesia in the affected hind paw.
`Zoledronic acid also attenuated mechanical hyperalgesia associated with chronic peripheral neuropathy and inflammation in the rat. In
`contrast, pamidronate or clodronate did not have any anti-hyperalgesic effect on mechanical hyperalgesia in the neuropathic and inflam-
`matory pain models.
`We conclude that zoledronic acid, in addition to, or independent from, its anti-metastatic and bone preserving therapeutic effects, is an
`anti-nociceptive agent in a rat model of metastatic cancer pain. This unique property of zoledronic acid amongst the bisphosphonate class of
`compounds could make this drug a preferred choice for the treatment of painful bone metastases in the clinic.
`q 2002 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved.
`
`Keywords: Bone; Cancer; Pain; Bisphosphonates; Zoledronic acid
`
`1. Introduction
`
`Metastatic bone disease is common in patients with
`breast, prostate and lung cancer and multiple myeloma,
`and is a frequent cause of morbidity in advanced cancer
`patients (Rubens, 1998). Metastatic bone tumour produces
`an imbalance between bone resorption and bone formation,
`with severe remodelling of the bone. This leads to hyper-
`calcaemia, structural weakening and consequent pathologi-
`cal fractures and compression in 8–30% of patients with
`bone metastasis.
`
`* Corresponding author. Tel.: 144-20-7333-2126; fax: 144-7387-4116
`E-mail address: laszlo.urban@pharma.novartis.com (L.A. Urban).
`
`Also, most patients with bone metastasis suffer from
`pain resulting from structural damage, periosteal irritation
`and nerve entrapment. Pain induced by bone metastases is
`frequent and difficult to treat because of its intermittent and
`progressive nature. Bone cancer often produces dull spon-
`taneous pain, which is aggravated by movement. Opioids
`are commonly used for malignant bone pain, however,
`their side effects prevent appropriate control (Mercadante,
`1997).
`The mechanisms underlying bone cancer pain are unclear
`but may involve changes in osteoclastic activity leading to
`bone resorption, and inflammatory activity provoked by
`cytokine and prostaglandin production by the cancer cells
`(Fulfaro et al., 1998).
`Bisphosphonates are a relatively new class of drug,
`
`0304-3959/02/$20.00 q 2002 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved.
`PII: S 0 3 0 4 - 3 9 5 9 ( 0 2 ) 0 0 0 4 0 - 4
`
`

`

`220
`
`K. Walker et al. / Pain 100 (2002) 219–229
`
`which are potent inhibitors of osteoclastic activity (Fleisch,
`1991), cancer cell proliferation, production of cytokines
`such as IL-6 and the metalloproteinase MMP-1 (Derenne
`et al., 1999; Senarante et al., 2000). Bisphosphonates have
`been shown to be efficacious in clinical conditions such as
`tumour
`induced hypercalcaemia and metastatic bone
`disease (Berenson et al., 2000; Body, 2000). Clodronate,
`pamidronate and zoledronic acid are first, second and third-
`generation bisphosphonates,
`respectively, which differ
`from one another by the substitution of active side chains
`(Fulfaro et al., 1998). They are all stable pyrophosphate
`analogues, with a characteristic P–C–P bond, an essential
`structure that prevents bisphosphonates from hydrolysis by
`phosphatases. They strongly bind to hydroxy apatite in the
`bone, which explains their curious pharmacokinetic beha-
`viour characterised by rapid clearance from the plasma and
`long-term accumulation in the bone. Anecdotal clinical
`evidence suggests that in addition to their disease modify-
`ing activity, these compounds are effective against pain
`associated with bone cancer (see Fulfaro et al., 1998).
`This is supported by recent preclinical studies showing
`that bisphosphonates have analgesic activity in non-
`cancer related nociceptive tests in mice (Bonabello et al.,
`2001).
`Study of the effectiveness of novel agents against meta-
`static bone cancer pain is hindered by the lack of animal
`models. A recent report described a murine model of bone
`cancer pain (Schwei et al., 1999) induced by sarcoma cells
`injected into the proximal epiphysis of the femur. Tumour
`growth was associated with the development of mechanical
`allodynia and pain-related behaviour. In this model osteo-
`protegerin, a TNF receptor family protein molecule, was
`reported to be effective in attenuating pain-related beha-
`viour such as mechanical allodynia (Honore et al., 2000;
`Luger et al., 2001). Although therapeutic effects of osteo-
`protegerin have been recently reported in postmenopausal
`women (Bekker et al., 2001), it is not known whether the
`compound works
`in metastatic pain.
`In comparison,
`bisphosphonates are used for the clinical
`indication of
`bone metastasis (Berenson et al., 2000; Body et al.,
`1999) and their effect on metastatic tumour-related pain
`is recorded (Berenson et al., 2001; Hortobagyi et al.,
`1996). To this date, there is no preclinical evidence for
`the anti-nociceptive effects of bisphosphonates in animal
`models of permanent pain.
`The aim of the present study was to examine the effects of
`bisphosphonates in bone cancer induced pain in comparison
`to other chronic pain models, namely neuropathic and
`inflammatory pain. Spontaneous pain, which is the main
`symptom of metastatic bone tumour, cannot be measured
`in animal models, therefore we investigated relevant pain-
`related behaviour such as mechanical allodynia and hyper-
`algesia. For this study we used a novel rat model of bone
`cancer pain, which develops 15–20 days after inoculation
`of the syngeneic mammary tumour cell line MRMT-1 into
`the tibial bone.
`
`2. Methods
`
`2.1. Preparation of cells
`
`Syngeneic MRMT-1 rat mammary gland carcinoma cells
`were obtained from Novartis Pharma (Basel). Cells were
`cultured in medium containing RPMI 1640 (Gibco), 10%
`foetal bovine serum (FBS, heat-inactivated), 1% l-gluta-
`mine and 2% penicillin/streptomycin. Cells were released
`from the plastic by brief exposure to 0.1% w/v trypsin, and
`then prepared for injection. Ten millilitres of medium was
`centrifuged for 3 min at 1200 rpm, and the resulting pellet
`was washed twice with 10 ml of Hank’s balanced salt solu-
`tion (HBSS [Hank’s]; Gibco) without Ca21, Mg21 or phenol
`red and then centrifuged for 3 min at 1200 rpm. The final
`pellet was suspended in 1 ml of Hank’s solution and cells
`were counted using a haemocytometer. Cells were diluted to
`achieve final concentrations for injection and kept on ice
`until injection. For the sham group, MRMT-1 cells prepared
`in the same way and diluted to achieve final concentrations
`for injection, and then inactivated by boiling for 20 min.
`
`2.2. Induction of bone cancer
`
`All animal procedures were carried out in accordance
`with the UK Animals (Scientific Procedures) Act, 1986
`and associated guidelines. Female Sprague–Dawley rats
`(150–180 g; Charles River) were maintained in a controlled
`lighting environment, six to a cage with water and food
`supply ad libitum.
`Animals were anaesthetised using Enflurane (Abbott) and
`the left leg was shaved and the skin disinfected with 70% (v/
`v) ethanol. A 1 cm rostro-caudal incision was made in the
`skin over the proximal half of the tibia to expose the bone
`with minimal damage to the surrounding muscle or blood
`vessels. Using a 23 gauge needle the bone was pierced 5 mm
`below the knee joint distal to the epiphysial growth plate
`and the needle was pushed down the intramedullary canal of
`the bone to create a cavity. Using a 5 ml Hamilton syringe,
`3 ml of culture medium containing approximately 3 £ 103
`cells was injected into one cavity. Control animals received
`the same volume of heat-killed cells (HKC) or medium
`only. The injection site was closed using bone wax and
`the wound closed with metal clips and dusted with Aureo-
`mycin antibiotic powder. Animals were placed in thermo-
`regulated cages until they have regained consciousness and
`then returned to the home cage.
`
`2.3. Drug administration
`
`Zoledronic acid, pamidronate or clodronate (supplied at
`the disodium salts by Novartis Pharma, Basel) were
`dissolved in sterile 0.01 M phosphate buffered saline
`(PBS) for subcutaneous (s.c.) administration (injection
`volume: 1 ml/kg). Celebrex was dissolved in 20% crema-
`phor/PBS (0.01 M) for daily s.c. administration. The dosing
`regimen for the bisphosphonates was as follows: once daily
`
`

`

`K. Walker et al. / Pain 100 (2002) 219–229
`
`221
`
`on the day of, and day 2, 5, 7, 9, 12, 14, 16, 19 after MRMT-
`1 cell inoculation. Two groups (Boissier et al., 1997, 2000)
`of cancer cell-treated animals were set up for permanent
`pamidronate treatment (30 and 100 mg/kg) with the same
`dosing regimen than that for zoledronic acid. Also, a control
`non-treated group of animals (Bonabello et al., 2001) was
`added.
`
`2.4. Radiology
`
`Radiological analysis was carried out on ipsilateral and
`contralateral hind limbs 20 days following cell injection.
`Limbs were placed on Industrex X-ray film (Kodak) and
`exposed to an X-ray source (Faxitron)
`for 1 min at
`30 KVP. Radiological scores for the tibia were given
`based on blind analysis of projected radiographs by two
`independent radiologists, as follows: (Bekker et al., 2001)
`MRMT-1 (3 £ 103) treated, (Berenson et al., 2000) heat-
`killed MRMT-1 treated, (Berenson et al., 2001) MRMT-1
`cell inoculation followed by treatment with 10 mg/kg, zole-
`dronic acid (s.c.) (Body, 1997), MRMT-1 cell inoculation
`followed by treatment with 30 mg/kg, zoledronic acid (s.c.)
`(Body, 2000), MRMT-1 cell inoculation followed by treat-
`ment with vehicle, zoledronic acid (s.c.) and (Body et al.,
`1999) naı¨ve animals (n ¼ 8 for each group). We used a
`similar scoring system to that of Honore et al. (2000): 0 ¼
`normal bone structure without any sign of deterioration;
`1 ¼ small radiolucent lesions in the proximal epiphysis,
`close to the site of the injection; 2 ¼ increased number of
`lesions; 3 ¼ loss of medullary bone, plus
`radiolucent
`erosion of the cortical bone; 4 ¼ full thickness unicortical
`bone loss; 5 ¼ full
`thickness bicortical bone loss and
`displaced fractures.
`
`2.5. Bone mass density and content measurement
`
`A Hologic QDR 1000 Plus instrument (Hologic Inc.,
`USA) was used with the Regional High Resolution Acces-
`sory for small animal applications. Bone mineral density
`(BMD) and bone mineral content (BMC) were determined
`for the whole tibiae and the proximal metaphysis, 19 days
`after intratibial injection of Hank’s solution, cancer cell
`inoculation, heat-killed cancer cell inoculation, cancer cell
`inoculation together with zoledronic acid (10 or 30 mg/kg)
`repeated treatment (see protocol for treatment under drug
`administration) and cancer cell inoculation together with a
`single dose of zoledronic acid (100 mg/kg, s.c.) on day 19.
`Specimens were positioned in a plastic cuvette and covered
`with 70% ethanol for measurement. For high-resolution
`recordings the X-ray beam was collimated to a diameter
`of 9 mm, line spacing and point resolution were set at
`0.254 and 0.127 mm, respectively. Data were computed as
`BMD (g/cm2) and BMC (g). Statistical analysis: analysis of
`variance (ANOVA), followed by Dunnett’s test.
`
`2.6. Bone histology
`
`For histology, rats were terminally anaesthetised by an
`i.p. injection of pentobarbitone and perfused with phosphate
`buffer containing 4% paraformaldehyde. The tibia was
`dissected and embedded in metacrylate. Serial sections of
`6 mm thickness were cut without demineralisation on a
`Reichert-Jung microtome. Sections were stained with
`haematoxylin and eosin to visualise the structure of the
`tumour and the bone.
`To identify polykaryocytes and/or osteoclasts, a tartrate
`resistant acid phosphatase (TRAP) stain was used (Hayman
`et al., 2000). Goldner’s Trichrom staining was used to visua-
`lise/differentiate bone and osteoid structures, osteoclasts
`and cytoarchitectonics of the tissue (Romeis, 1989).
`All methods were performed 20 days after the initial
`treatment on the following experimental groups: 3 £ 103
`MRMT-1 cell injection; 3 £ 103 heat-killed MRMT-1 cell
`injection;
`injection of vehicle only (Hank’s solution);
`3 £ 103 MRMT-1 cell injection and treatment with zoledro-
`nic acid, (30 mg/kg once daily, for 19 days, see dosing regi-
`men).
`
`2.7. Bone cancer pain model
`
`Mechanical allodynia was measured as the hind paw
`withdrawal response to stimulation with von Frey filaments
`with logarithmically incremental stiffness that deliver force
`to the hind paw measured in milliNewton [range: 0.2–20.9;
`Kim and Chung (1992)]. The test environment consisted of
`a perspex box situated on a wire mesh platform. The rat was
`placed in the test box and allowed to settle for 5–10 min.
`Each von Frey filament was applied to the plantar surface of
`the hind paw, in ascending order beginning with the 1 g
`filament. A single trial consisted of six to eight applications
`of the filament within a 2–3 s period. At least 2 min were
`allowed between each trial, and five trials were administered
`for each von Frey filament. A trial was suspended immedi-
`ately when a hind paw withdrawal response occurred. An
`indication of the amount of paw withdrawals in the five
`trials was expressed as percent response frequency: (number
`trials) £ 100 ¼ % response
`of
`paw withdrawals/five
`frequency).
`Mechanical hyperalgesia was assessed as the difference
`in weight born by the ipsilateral compared to the contral-
`ateral limb, measured using a Dual Channel Weight Aver-
`ager, (Churchill Electronic Services Ltd). Rats were placed
`in a perspex chamber designed so that each hind paw was
`resting on a separate transducer pad, which recorded over
`3 s the distribution of the animals’ body weight on each
`paw. For each rat two readings from each paw were taken
`and then averaged. Results are presented as weight bearing
`difference (WBD; contralateral reading – ipsilateral read-
`ing). Raw data were analysed using one-way ANOVA
`followed by Dunnett’s post hoc test ð*P , 0:05Þ, repeated
`
`

`

`222
`
`K. Walker et al. / Pain 100 (2002) 219–229
`
`measures of ANOVA, and post hoc tests including Student’s
`t and Tukey’s HSD tests.
`
`2.8. Inflammatory hyperalgesia
`
`Mechanical hyperalgesia was examined in a model of
`permanent inflammatory pain. Paw withdrawal thresholds
`to an increasing pressure stimulus were measured by the
`Randal-Sellito technique using an analgesymeter (Ugo
`Basile, Milan), in naı¨ve animals prior to an intraplantar
`injection of Freund’s complete adjuvant (FCA) into the
`left hind paw. Twenty-four hours later paw withdrawal
`thresholds were measured again, prior to (predose) and
`then from 10 min to 6 h following drug or vehicle adminis-
`tration. Data are presented as absolute values (g).
`
`2.9. Neuropathic hyperalgesia
`
`Mechanical hyperalgesia was examined in a model of
`neuropathic pain induced by partial ligation of the left scia-
`tic nerve (Seltzer et al., 1990). Approximately 14 days
`following surgery mechanical withdrawal
`thresholds of
`both the ligated (ipsilateral) and non-ligated (contralateral)
`paw were measured prior to (predose) and then from 10 min
`to 6 h following drug or vehicle administration. Data are
`presented as absolute values (g).
`All experiments were carried out using groups of six
`animals. Stock concentrations of drugs were dissolved in
`distilled water and subsequent dilutions were made in
`0.9% saline for s.c. administration in a volume of 4 ml/kg.
`All drugs were made up in plastic vials and kept in the dark.
`Statistical analysis was carried out on withdrawal thresh-
`old readings (g) using ANOVA with repeated measures
`followed by Tukey’s HSD test. Efficacy refers to the maxi-
`mal reversal of hyperalgesia observed at the doses used.
`
`3. Results
`
`3.1. Radiological analysis of tumour development in the
`tibia
`No radiological change (score ¼ 0) was found in non-
`treated animals, in animals treated with HKCs or with
`Hank’s solution (not shown) 20 days after initiation of the
`experiment. In contrast, 20 days after MRMT-1 cell injec-
`tion, full thickness bicortical bone loss and in one case
`displaced fractures were observed in the tibiae ðn ¼ 16Þ.
`Radiological scores ranged from 3.5 to 3.75 when assayed
`independently by two radiologists (Table 1). No extra-
`osseal tumour growth/dissemination or contralateral radi-
`ological signs of bone erosion were observed.
`Animals treated with 10 mg/kg zoledronic acid showed a
`tendency towards a decrease in the radiological scores,
`although this was not significant (Table 1). However,
`repeated treatment with 30 mg/kg zoledronic acid, produced
`a highly significant prevention of bone destruction with
`
`radiological scores of 0.88 compared to 3.5–3.75 observed
`in animals with bone cancer without zoledronic acid treat-
`ment (n ¼ 8; P , 0:001; ANOVA, Tukey’s HSD test;
`Table 1). Mainly small radiolucent lesions were observed
`in animals treated with zoledronic acid, with occasional,
`minimal
`loss of the medullary bone, compared to the
`dramatic loss seen in vehicle treated animals.
`Repeated treatment with pamidronate at a similar dose
`range (30 or 100 mg/kg) did not affect the radiological
`scores (Table 1). After repeated treatment with 100 mg/kg
`pamidronate signs of full thickness unicortical bone loss and
`medullary erosions were regularly observed.
`
`3.2. BMD and BMC
`
`Both BMD and BMC were significantly diminished in the
`MRMT-1 treated tibiae compared to bones treated with
`culture medium (Hank’s) or HKCs. This reduction was
`apparent when measurements were taken from either the
`proximal tibia or the whole bone (Fig. 1). Chronic treatment
`with zoledronic acid (10 or 30 mg/kg)
`significantly
`increased both BMD and BMC of the bone in comparison
`to the cancerous, non-treated bones. In contrast, a single
`dose of 100 mg/kg zoledronic acid on day 19 after
`MRMT-1 cell inoculation did not affect the reduction in
`BMD or BMC caused by cancer cell injection.
`
`3.3. Histological evaluation of tumour growth and
`osteoclast activity
`
`Bones inoculated with MRMT-1 cells showed infiltration
`of bone marrow space by malignant tumour (haematoxylin
`and eosin), typically composed of sheets of neoplastic cells
`with moderately pleomorphic hyperchromatic nuclei, in
`keeping with a primary sarcoma, 20 days after tumour cell
`inoculation (Fig. 2C). Compact tumour growth alternated
`with small areas of degenerating and necrotising tumour
`
`Table 1
`Effects of zoledronic acid and pamidronate on radiological scores in bone
`cancera
`
`Treatment
`
`Investigator #1
`
`Investigator #2
`
`Cancer/vehicle
`Cancer/zoledronic acid 10 mg/kg
`Cancer/zoledronic 30 mg/kg
`Cancer/vehicle
`Cancer/pamidronate 30 mg/kg
`Cancer/pamidronate 100 mg/kg
`
`3.5 ^ 0.45
`2.63 ^ 0.2
`0.88 ^ 0.24*
`3.5 ^ 0.29
`3.75 ^ 0.34
`3.25 ^ 0.34
`
`3.75 ^ 0.34
`2.75 ^ 0.18
`0.88 ^ 0.32*
`3.63 ^ 0.28
`4.13 ^ 0.24
`3.38 ^ 0.35
`
`a All animals were treated with MRMT-1 cells. Two groups received
`chronic treatment with 10 and 30 mg/kg zoledronic acid (s.c.), respectively.
`Pamidronate was administered chronically at 30 and 100 mg/kg s.c. daily
`doses, respectively. For the dosing regimen, see Section 2). Radiological
`scores were obtained 19 day after MRMT-1 cell inoculation. Each radio-
`gram was scored blindly by two independent investigators (#1 and #2). Data
`are expressed as mean ^ SEM; *P , 0:001; ANOVA,
`followed by
`Tukey’s HSD test.
`
`

`

`K. Walker et al. / Pain 100 (2002) 219–229
`
`223
`
`A
`
`0.4
`
`B
`
`0.15
`
`~.10
`0
`::& m
`0.05
`
`0.00
`
`Treatment
`
`** **
`
`**
`
`**
`
`**
`
`Treatment
`
`** **
`
`** **
`
`C
`
`0.3
`.-.
`"'
`~0.2
`~
`C
`io.1
`
`0.0
`
`D
`
`0.4
`
`__ o.a
`s
`0 :e0,2
`m
`0.1
`
`0.0
`
`Treatment
`
`Treatment
`
`Fig. 1. The effects of zoledronic acid on the bone mineral density (BMD; A,C) and bone mineral content (BMC; B,D) in the tibia. Columns represent averaged
`data from six animals 20 days after injection of either Hank’s solution (Hank’s), heat-killed MRMT-1 cells (HKC) or live 3 £ 103 MRMT-1 cells (cancer) and
`three treatment groups (chronic treatment with 10 and 30 mg/kg zoledronic acid; 100 mg/kg acute treatment with zoledronic acid on day 19). Statistical
`analysis: **P , 0:01; one way ANOVA, followed by Dunnet’s test, comparisons against the HKC group. Note, that after chronic treatment with zoledronic
`acid, BMD and BMC also showed increase in comparison to the untreated control groups (Hank’s and HKC). For detailed treatment protocol, see Section 2.
`(A,B) Data obtained from the proximal tibia (1/3 of the whole length); (C,D) BMD and BMC for the whole bone.
`
`tissue (Fig. 2E, F) indicating a dynamic progression of the
`tumour.
`The extent of bone marrow replacement by tumour
`ranged from 35 to 80% ðn ¼ 5Þ. No specimen treated with
`killed MRMT-1 cells contained tumour (Fig. 2B). Although
`the tumour was retained within the bone, in some cases the
`destruction of the cortical bone was complete with only the
`periosteum left (Fig. 2C, E). The osteoblast lining of the
`compact bone largely disappeared. Chronic treatment with
`30 mg/kg zoledronic acid largely prevented the destruction
`of the bone, with islands of limited tumour growth and
`proliferation of connective tissue cells (Fig. 2D, G, H).
`The thickness of the cortical bone was preserved and most
`of the trabecular arrangement remained intact (Fig. 2D).
`There was little sign of the normally present osteoblast
`lining of the calcified bone.
`A set of sections were stained to detect TRAP containing
`osteoclasts (Fig. 2I, J). TRAP positive profiles, which are
`associated with the bone and rare in control (Hank’s or HKC
`treated) bone, appeared scattered in the tumour mass and
`
`close to the remnants of the trabeculae (Fig. 2I). In addition
`to the dramatic increase of the number of TRAP positive
`profiles within the boundary of the tumour, the intensity of
`the staining was enhanced (compare Fig. 2I and J). In
`contrast,
`these profiles were almost completely absent
`from sections obtained from zoledronic acid treated bones
`(Fig. 2J).
`Goldner’s staining showed diminished calcification of the
`cancerous bone, which was restored by zoledronic acid
`treatment (not shown).
`
`3.4. Effect of chronic administration of zoledronic acid on
`mechanical allodynia and mechanical hyperalgesia
`
`Allodynia develops gradually in rats treated with MRMT-
`1 cells starting about 12 days after inoculation (Fig. 3A).
`After a 19 day treatment regimen (see Section 2) of zole-
`dronic acid (10 or 30 mg/kg; once daily) mechanical allo-
`dynia was attenuated. An overall rightward shift was seen
`with both doses, however, only 30 mg/kg reached the level
`
`

`

`224
`
`K. Walker et al. / Pain 100 (2002) 219–229
`
`Fig. 2. The effects of zoledronic acid on the structure of the bone and the development of the tumour and the proliferation of polykaryocytes. Haematoxylin and
`eosin; 6 mm thick sections. Longitudinal sections of the proximal tibia obtained from a control animal (A), from an animal treated with heat-killed MRMT-1
`cells (B), from MRMT-1 cell inoculated animal without zoledronic acid treatment (C) and from an animal treated with 30 mg/kg zoledronic acid for 2 weeks
`after MRMT-1 cell inoculation (D) 20 days after initiation of the tumour. Inoculation of live MRMT-1 cells produced a large, expansive tumour with a
`complete destruction of the cortical bone and loss of the bone marrow (C). Chronic treatment with zoledronic acid preserved the structure of the bone (D). (E,F)
`Different areas from the solid tumour developing in the bone. Dotted line in E marks the direct boundary between tumour cells and the periosteum after total
`destruction of the cortical bone. (F) Most of the tumour contains densely packed cells with occasional small regions of necrotic areas (arrow). (G,H) Illustration
`of the effects of chronic zoledronic acid treatment. (G) Large areas contain tissue derbis with signs of karyolysis and complete lack of tumour cells. (H) The
`trabecular system is well maintained and the bone marrow space is alternately filled with cells of fusiform appearance (open arrow) and with cells of the bone
`marrow (filled arrow). (Note that the colour was affected in sections taken from zoledronic acid-treated animals.) Chronic treatment with zoledronic acid
`prevents the appearance of polykaryocytes and osteclasts in the tibia after injection of 3 £ 103 MRMT-1 cells (I,J). TRAP staining (6 mm thick sections) of the
`bone from a rat inoculated with MRMT-1 cells (I), and from MRMT-1 cell inoculated and zoledronic acid (30 mg/kg) treated animals (J), 20 days after
`inoculation of MRMT-1 cells. TRAP is present in large polykaryocytes with multiple nuclei. Note the large amount of scattered, strongly TRAP positive cells
`in the cancerous bone. Most of these profiles reside within the tumour and are not associated with any bone structure (I). In contrast, bone after chronic
`treatment with zoledronic acid (30 mg/kg) contains only few, faintly stained TRAP positive profiles close to the bone (J; arrows).
`
`of significance in the range of 4–10 g (Fig. 3A). In contrast,
`pamidronate given by the same dosing regimen at 30 or
`100 mg/kg, did not alter the mechanical allodynia (Fig. 3B).
`The selective COX-2 enzyme inhibitor, Celebrex, admi-
`nistered once daily for 19 days (30 mg/kg) did not affect
`mechanical allodynia (Fig. 3C).
`Mechanical hyperalgesia measured as a difference in
`weight bearing between ipsi- and contralateral hind limbs
`was observed only in animals treated with MRMT-1 cells,
`reaching a peak at 12 days postinjection and persisting until
`at least 20 days. Chronic treatment with 30 mg/kg zoledro-
`nic acid attenuated this mechanical hyperalgesia with a
`significant reduction in the difference between ipsi- and
`contralateral weight bearing (P , 0:05; Fig. 3D). The effect
`
`of 10 mg/kg zoledronic acid was not significant. Pamidro-
`nate (30 or 100 mg/kg) did not produce any significant
`change in the WBD.
`
`3.5. Inflammatory hyperalgesia
`
`A single s.c. administration of zoledronic acid (0.003–
`0.1 mg/kg) produced a pronounced dose-dependent reversal
`of FCA-induced mechanical hyperalgesia (Fig. 4). Its effect
`was rapid in onset with a maximal reversal of 93% observed
`30 min following administration, and of short duration with
`no significant reversal 3 h following administration. At the
`highest dose tested there was an increase in contralateral
`paw withdrawal threshold.
`
`

`

`K. Walker et al. / Pain 100 (2002) 219–229
`
`225
`
`---vehicle (PBS)
`
`:::= 10µ!)"kg zoledromc acid
`
`30µ !)"kg zoledronic acid
`_,.... Hank's intra-tibially (n=12)
`
`Von frey filament (g)
`
`10
`
`-A.- Vehicle (PBS)
`_,,_ 30 mg/kg celebrex
`
`A
`
`100
`
`80
`
`C
`
`0
`
`0.1
`
`100
`
`90
`
`BO
`
`70
`
`60
`
`50
`
`40
`
`30
`
`20
`
`10
`
`0
`
`-10+-o-....-r-;;i:-----........ .....,..,""""'l""--................. ....-~(cid:173)
`o.1
`
`10
`
`Von Frey ftlament (g)
`
`B
`
`100
`
`90
`
`BO
`
`70
`
`ill
`
`40
`
`l
`~ 60
`! 50
`I 30
`c!
`
`20
`
`10
`
`0
`
`-10
`0.1
`
`Q)
`
`D 100
`90
`§ 80
`~ 70
`!!:
`,::, 60
`g>
`aj 50
`~
`E 40
`.!2>
`Q) 30
`~
`
`20
`
`10
`
`0
`
`-•-Vehicle (PBS)
`:::aOµg/kg pamld1011ate
`l 0(lllg,l<g pamldronate
`
`Von Irey filament (g)
`
`10
`
`Vehicle
`
`30µgkg 1 00µgkg
`Pamidronate
`
`Fig. 3. Effects of chronic treatment with zoledronic acid on the nociceptive behaviour in the bone cancer model. (A) The effects of chronic treatment with
`zoledronic acid (10 and 30 mg/kg; n ¼ 14=group); (B) pamidronate (30 and 100 mg/kg, n ¼ 8=group); and (C) celebrex (30 mg/kg; n ¼ 8=group) on mechanical
`allodynia, measured on day 19. (A) Shows lack of mechanical allodynia in Hank’s inoculated animals (P ¼ no cancer cell inoculation). (*P , 0:05; ANOVA,
`followed by Tukey’s HSD test, compared to vehicle treated control). (D) The effects of zoledronic acid and pamidronate on the weight bearing difference
`between the hindlimbs 19 days after MRMT-1 cell inoculation. (*P , 0:05; ANOVA, Tukey’s HSD test, compared to vehicle treated control; nðvehicleÞ ¼ 22;
`nðzoledronic acidÞ ¼ 12; nðpamidronateÞ ¼ 8Þ.
`
`Pamidronate (0.03–1.0 mg/kg) and clodronate (0.3–
`10 mg/kg) produced no reversal of mechanical hyperalgesia
`following a
`single
`s.c.
`administration. Pamidronate
`produced a small but significant decrease in ipsilateral and
`contralateral paw withdrawal thresholds at the highest dose
`tested of 1 mg/kg which was apparent at the later time points
`(Fig. 4). Clodronate similarly produced slight reductions in
`contralateral withdrawal thresholds (Fig. 4).
`
`3.6. Neuropathic hyperalgesia
`
`All bisphosphonates were used in the neuropathic model
`within the same dose range as in the experiments addressing
`inflammatory hyperalgesia. Zoledronic acid produced a
`moderate reversal of mechanical hyperalgesia with maximal
`efficacy of 40% observed 30 min following s.c. administra-
`tion (Fig. 5). However, it is of interest that the immediate
`effect of the highest dose (0.1 mg/kg) of zoledronic acid
`
`caused a slight decrease in ipsilateral paw withdrawal thresh-
`olds at 10 min following administration prior to the observed
`increase in thresholds and reversal of hyperalgesia. The
`reduction in thresholds was pronounced on the contralateral
`paw at 10 and 30 min following administration (Fig. 5).
`Pamidronate did not significantly affect the ipsilateral paw
`withdrawal thresholds, which resulted in approximately a
`20% reversal of hyperalgesia 1 h following administration.
`However, there was again a transient fall in contralateral
`withdrawal thresholds 10 min postadministration. A similar
`reduction was observed with clodronate although it did not
`produce any subsequent reversal of hyperalgesia (Fig. 5).
`
`4. Discussion
`
`The main finding of this study is that chronic treatment
`with zoledronic acid, a third-generation bisphosphonate,
`
`

`

`226
`
`K. Walker et al. / Pain 100 (2002) 219–229
`
`significantly prevented the development of bone tumour in a
`novel rat model of bone cancer, and in addition it produced
`anti-allodynic
`and anti-hyperalgesic
`effects. Another
`bisphosphonate, pamidronate, in a similar dose range did
`not alter the radiological scores or nociception in the same
`
`model. When tested in conventional models of persistent
`inflammatory and neuropathic pain, zoledronic acid alone
`showed anti-hyperalgesic activity.
`For this study we developed a novel model of bone
`cancer, in which tumour growth was induced by injection
`
`lpsilateral
`
`I
`
`O vehicle
`• 0.003mgkg-1
`& 0.01 mgkg-1
`Y 0.03 mgkg-1
`• 0.1 mgkg-1
`
`Zoledronic acid
`
`Contralaterlli
`
`o vehicle
`• 0.003 mgkg-1
`• 0.01 mgkg-1
`Y 0.o3mgkg-I
`• 0.1 mgkg-1
`
`naive predose 10 min 30 min 1 h
`
`3 h
`
`6 h
`
`Time post-dose
`
`naive predose 10 min 30 min I h
`Time post-dose
`
`3 h
`
`6 h
`
`Parnidronate
`
`•
`
`Ipsilateral
`
`O vehicle
`• 0.o3mgkg-I
`• 0.lmgkg-1
`" 0.3 mgkg-1
`•
`1 mgkg-1
`
`50
`
`40
`
`120
`:§ 110
`'O
`'.g 100
`"' ~ 90
`-;
`80
`~
`j 70
`-~
`60
`~
`
`50
`
`40
`
`Contralateral
`
`I
`
`I ~ ~
`
`•
`
`a vehicle
`• 0.03 mgkg-1
`• 0.1 mgkg-1
`,, 0.3mgkg-1
`• 1 mgkg-1
`
`Naive predose 10 min 30 min 1 h
`
`3 h
`
`6 h
`
`Naive predose 10 min 30 min 1 h
`
`3h
`
`6h
`
`Time post-dose
`
`Time post-dose
`
`Ipsilateral
`
`I
`
`D vehicle
`
`• 0.3mgkg-I
`... 1 mgkg-1
`" 3 mgkg-1
`• l0mgkg-1
`
`Clodronate
`
`Con tralateral
`
`I
`
`a vehicle
`• 0.3mgkg-1
`... 1 mgkg-1
`" 3 mgkg- 1
`• IOmgkg-1
`
`120
`
`3110
`~ 100
`I ,s 90
`
`-;
`80
`~
`j 70
`-~
`60
`~
`ll. 50
`
`40
`
`120
`:§ 110
`:9
`
`..8 100 I 90
`-;
`~ 80
`3
`70
`;I'
`~ 60
`ll.
`50
`
`'O
`
`40
`
`Naivepred