JOURNAL OF BONE A N D MINERAL RESEARCH Volume 7, Number 6. 1992 Mary Ann Lkbcrt. he., Publishers

Longitudinal Effect of Tiludronate on Bone Mineral Density, Resonant Frequency, and Strength in Monkeys P. GEUSENS,' J. NIJS,' G. VAN DER PERRE,' R. VAN AUDEKERCKE,' G. LOWET,' S. GOOVAERTS,' A. BARBIER.' F. LACHERETZ,3 B. REMANDET,3 Y. JIANG,' and J. DEQUEKER'

ABSTRACT The effect of Tiludronate on bone was studied in 72 growing monkeys (Papio papio), 36 males and 36 females, aged 4-7 years. They were randomly allocated into four groups (18 animals per group, 9 males and 9 females): group I, controls; group 11, 10 mg/kg/day; group 111, 20 mg/kg/day; and group IV, 40 mg/kg/ day of Tiludronate. A total of 12 animals (6 males and 6 females) in each group were sacrificed at the end of treatment (1 year) and 6 animals (3 males and 3 females) per group 1 year later. Bone mineral density (BMD) was measured by dual-photon absorptiometry. Biomechanics1 properties were evaluated by an impact torsion test and by resonant frequency analysis. Bone mineral measurements indicated that at the end of 1 year of treatment BMD was significantly higher, especially at the distal epiphysis of the radius, than in controls. No significant differences between groups were found in BMD 1 year after stopping treatment. Biomechanical analyses indicated that torsional stiffness increased after treatment. No differences between groups were found 1 year after stopping treatment. Results of resonant frequencies indicated an increased calculated transversal stiffness after treatment and 1 year later and an increased buckling strength 1 year after stopping treatment. In conclusion, the results on the effect of Tiludronate in growing monkeys indicate a profound effect of this drug on bone density and biomechanical properties. The biomechanical results indicate that this drug is safe, with conservation of bone strength despite a change in intrinsic mechanical properties of the bone.

INTRODUCTION ISPHOSPHONATES are

well known for their inhibitory effect on bone resorption.ill The effects of bisphosphonates in clinical applications in Paget's disease"' and hypercalcemia"' are well documented. A new bisphosphonate, Tiludronate (chloro-4-phenylthiomethylene-bisphosphonate, Sanofi) has been assayed in different models of bone resorption in animals'4' and in patients with Paget's disease of bone."' Recently, in a double-blind randomized clinical study, Tiludronate was shown to inhibit significantly postmenopausal bone loss at the lumbar spine."'

B

Bisphosphonates inhibit osteoclastic bone resorption, which can result in an increased bone density,"' especially in growing animals.i8' High doses of bisphophonates can result in impaired mineralization and osteomalacia, as has been shown with EHDP,IP1 or a secondary decrease in bone formation, for example by APD."'] No signs of osteomalacia are found histologically with Tiludronate up to a dose of 28 mg/kg/day in monkeys.'"' No data are available about bone strength during treatment with bisphosphonates, however, except for an increased incidence of spontaneous fractures with high doses of EHDP.'P' Bone mass (BMC)'"' and bone density (BMD)i13'are significantly correlated with bone strength. Bone quality

'Arthritis and Metabolic Bone Disease Research Unit, Catholic University of Leuven, Pellenberg, Belgium. 'Department of Biomedical Engineering, Catholic University of Leuven, Belgium. 'Sanofi Recherche. Montpellier. France.

GEUSENS ET AL.

600 plays an important role in the pathogenesis of fractures,[14) and changes in bone quality can occur before BMC and BMD have changed[14’and/or concomitantly with measurable changes in BMC and BMD,[”’ for example by architectural changes[”’ or changes in the material characteris- STRESS (Nm] tics of the bone, such as ~steomalacia.(~’ In vitro, BMC and BMD can be measured on excised bones[16’and bone quality can be estimated from biomechanical properties at the organ level, such as by rapid torsional loading[”’ or analysis of resonant To evaluate the long-term effect, safety, and reversibility of the effect of Tiludronate on bone mass, density, and strength, growing monkeys were studied by measurements of bone density and biomechanical behavior of the radius, and intercorrelations between bone mass measurements and biomechanical characteristics were evaluated.

b

STIFFNESS

(Nmldegreel

t

MATERIALS AND METHODS A group of 72 baboons (Papiopapio, Senegal) was studied (36 male and 36 female animals). These animals were

caught in the wild between the ages of 4 and 7 years. Growing animals were studied to maximize possible negative effects on bone, such as rickets and fractures. Their initial weight ranged from 2.5 to 8 kg and their final weight from 3.7 to 10 kg. They were housed in individual cages and randomly allocated to four groups, with 18 animals per group (9 males and 9 females): group I, controls; group 11, 10 mg/kg/day; group 111, 20 mg/kg/day, and group IV, 40 mg/kg/day of Tiludronate. This dose range was studied as part of a toxicologic study of the product. The doses used were 4-16 times higher than that used in clinical practice, which is 2.5 mg/kg/day. Tiludronate was given by daily gavage with a solution in distilled water. The product was given orally because it is also studied with oral administration in clinical situations, such as in Paget’s disease,[5’ and in the prevention of postmenopausal bone loss.[6’ A total of 12 animals (6 males and 6 females) from each group were sacrificed at the end of treatment (1 year). A total of 6 animals (3 males and 3 females) per group were sacrificed 1 year later. They were sacrificed using sodium pentobarbital by the intravenous route. A total of 3 animals died during the experiment: 1 male during the 1 year treatment in group I1 and 2 females (1 during 1 year treatment in group I11 and 1 during the 1 year treatment in group IV). Another female from group I11 was sacrificed for ethical reasons during 1 year of treatment. The death of these 4 animals was not related to the treatment. All bones were x-rayed and the length was measured. The wet weight of the bones was measured on a laboratory balance.

Bone mineral measurements BMC and BMD were measured by dual-photon absorptiometry using a Lunar bone mineral analyzer DPX (Lunar Radiation Corporation, Madison, WI). The calibration was performed using the standards supplied by the manu-

FRACTURE

50 degrees

0

FIG. 1. A stress-strain curve of the radius studied biomechanically in torsion.

Spectrum Analyser

Heasti r i ng h p l if ier

Cushion

FIG. 2. Devices for resonant frequency analysis. Excitation was performed with a hammer. Vibrations were recorded by a microphone and fed to a spectrum analyzer.

facturer. Manufacturer supplied software was used as in the clinical situation to calculate BMC (g) and BMD (g/ cm’). Before scanning, the deep-frozen radii were mounted at the bottom on a plexiglass stage in water at a level of 16 cm. The bone was supported on the distal and proximal ends, with the anterior side downward. The stage was placed on the scanner deck so that the diaphysis of the bone was oriented perpendicular to the pass of the photon beam. Measurements were made of the total bone. The results were analyzed in the total radius. They were also ana-

EFFECTS OF TILUDRONATE ON MONKEY BONE

601

The radius can be approximated by a uniform thinlyzed at the proximal and distal epiphyses and at the midshaft. Therefore, a region of 15% of the length of the ra- walled cylindrical tube with elliptical cross section. Tordius was analyzed. For the proximal and distal epiphyses sional strength T, and stiffness S can be calculated(lv.*O] 15% of the length of the radius was analyzed, starting from from the proximal and the distal ends of the radius. At the midshaft, 7.5% of the length was analyzed from both sides from the midpoint of the length of the radius. The reproducibility (one radius, measured five times after reposition- av6 ing) was 2.5, 3.0, 2.6, and 5.1% for the total radius, the distal and proximal epiphyses, and the midshaft, respecS = (2G/p) x M x [(a* + b*)/L(a’ + b’)] tively. where a = semimajor axis of the ellipse, b = semiminor axis of the ellipse, M = mineral mass per unit length Biomechanical tesls (BMC at 50% of the length of the radius), p = mineral After measurements, the bones were tested biomechani- content per cubic centimer cortical bone samples from hucally for the whole bone by a torsion test machine (Biome- man femur,‘”.*1) 7,lr = ultimate shear stress for human chanics Laboratory, Cleveland, OH).‘”) Frozen bone femoral bone.‘”) G = mean shear modulus for human specimens were embedded at the epiphyseal ends in rec- cortical bone at the femur,‘*’’and L = length of the bone. tangular epoxy boxes that fit the grips of the torsion test- Correlations between the measured values for strength and ing machine. Torsional stiffness (N-mldegree), strain stiffness and the values estimated from the model were calenergy absorption before fracture (J). maximum torque culated. If a good correlation between measured and esti(N-m), and maximum angle before fracture (degree) were mated values is found, the geometric model of estimation measured according to methods used for human femora.‘”] is a good predictor, with the prerequisite that the intrinsic These parameters are graphically shown in Fig. 1. material properties (ultimate shear stress and mean shear modulus) are unchanged (since the intrinsic material properties are considered constant). If the correlation is calculated by splitting the calculations according to the treatment groups and different correlations and different slopes RMPL 1 TUOE are found, this indicates a change in material properties. Resonant frequencies were measured at the radius in two planes, one in the sagittal and one in the frontal plane. The bones were placed in free-free conditions on a soft cushion. Vibrations were recorded by a microphone about 1 cm above the bone“8’ (Fig. 2). The excitation was pern/sZ/N formed with a hammer impact at half the length of the radius. The force and microphone signals were fed to a spectrum analyzer, which gave the frequency spectrum, and the resonant frequencies were determined (Fig. 3). The mean of four measurements was calculated for the reme. e E I e e corded signals. Transverse stiffness (E x I / L 3 , where E = e4e FREOUENCY l 5 0 . 0 Hx/OIV.l 1356 Young’s modulus and I = second moment of inertia) was FIG. 3. Display of frequency spectrum on the spectrum calculated as P M , where F = resonant frequency and M = mass (total wet weight). Buckling strength was calcuanalyzer.

TABLEI . BONEMINERAL DENSITY (G/cM*) OF TOTALRADIUS AND 15% OF LENGTHOF PROXIMAL AND DISTAL AND DIAPHYSES OF RADIUSIN MALEAND FEMALE ANIMALS IN DIFFERENT GROUPS AT ENDOF 1 YEAR EPIPHYSES OF TREATMENT (MEANf SEM)a

Dose (mg/kg/day)

n

0 (controls)

12 11 11 11

10 20

40 Kruskal-Wallis Trend

Proximal epiphysis 0.309 0.337 0.362 0.378

f f f f

0.017 0.017 0.016 0.014

< 0.05