Renal Tolerability Profile of Novel, Potent Bisphosphonates in Two Short-Term Rat Models

Transcription

1 0 Pharmacology & Toxicology 1997,80, Printed in Denmark All rights reserved Copyright 0 ISSN Renal Tolerability Profile of Novel, Potent Bisphosphonates in Two Short-Term Rat Models Jonathan R. Green, Yves Seltenmeyer, Knut A. Jaeggi and Leo Widler Ciba-Geigy Ltd., Pharmaceuticals Division, Basel, Switzerland (Received October 1 I, 1996, Accepted December 11, 1996) Abstract: Bisphosphonates are used clinically to inhibit bone resorption but they may also cause renal damage. For the profiling of new, potent bisphosphonates, their adverse renal effects were investigated in 2 rat models. In the first model, bisphosphonate was repeatedly injected (1 mg/kg, subcutaneously) over 2 weeks and the urinary excretion of malate dehydrogenase was monitored to assess nephrotoxic potential. Of the 6 new compounds tested, 3 markedly elevated malate dehydrogenase whereas 3 others caused only minor changes similar to those observed with 6 reference bisphosphonates that are already used clinically. On the basis of a therapeutic index (inhibition of bone resorption versus renal effects) fold greater than that of other analogues, the compound CGP was further profiled. In the second model, CGP or pamidronate was infused ( mg/kg, intravenously) into anaesthetized rats and the serum urea concentration was monitored as an indicator of renal dysfunction. Both compounds elevated serum urea in a time- and dosedependent manner, but the Dloo value for CGP was 3.8-fold higher than that of pamidronate. It is concluded that CGP (zoledronate) has a low nephrotoxic potential and can be further developed as a new, potent inhibitor of bone resorption. Bisphosphonates have found widespread clinical use as inhibitors of osteoclastic activity in a variety of diseases characterized by elevated bone resorption, notably malignant bone disease, Paget s disease and more recently, osteoporosis (Fleisch 1995). The bisphosphonate group is a powerful calcium chelator, a property which leads to the formation of polymeric aggregates with cations (Grabenstetter & Cilley 197 1; Wiers 1971) and also explains the high affinity that these compounds exhibit for bone (Bisaz et al. 1978). For most bisphosphonates, 50-80% of the drug reaching the systemic circulation is rapidly bound to bone and the remainder is excreted unchanged via the kidneys (Fleisch 1995). In rats the clearance of bisphosphonates exceeds the glomerular filtration rate, indicating active tubular secretion (Troehler et al. 1975). This rapid excretion exposes the kidneys to high concentrations of bisphosphonate or bisphosphonate aggregates which may cause renal damage, especially when large doses are given intravenously to patients with malignant disease in whom renal function may already be impaired. In clinical studies with bisphosphonates renal failure, transient proteinuria and alterations in creatinine clearance have been observed following intravenous infusion of large doses of etidronate, clodronate and tiludronate (Bournameaux et al. 1983; Yates et al. 1985; Kanis et al. 1983; Dumon et al. 1991). Similarly in rats, high doses of clodronate cause proteinuria prior to the onset of histopathological changes characterized by proximal tubular necrosis (Alden et al. 1989). These adverse effects can usually Author for correspondence: Jonathan R. Green, Ciba-Geigy Ltd., K , CH-4002 Basel, Switzerland (fax ). be avoided by slowly administering small doses of bisphosphonate in large volumes of fluid (Kanis et al. 1987; Mian et al. 1991). Therefore, in the profiling of new, potent bisphosphonates there was clearly a need to investigate renal tolerability to ascertain whether antiresorptive potency could be dissociated from renal intolerability, and to ensure that only those compounds with an improved therapeutic ratio (i.e. bone versus kidney effects) entered further development. To this end, 2 in vivo rat models were used to compare the renal tolerability of new bisphosphonates (subsequently referred to as preclinical compounds ) to that of established bisphosphonates which are already marketed or in advanced clinical trials ( clinical compounds ). These investigations were part of the early preclinical profiling of research compounds, they were not intended to replace the more extensive mandatory toxicological studies. Materials and Methods Animals. All the experiments were performed with male rats, g body weight, strain Tif:RAIf(SPF), supplied by the Ciba- Geigy animal farm, Sisseln, Switzerland. Test compounds. The 6 clinical (generic names) and 6 preclinical (CGP numbers) compounds tested in this study were: alendronate =4-amino-l-hydroxybutylidene-l, I-bisphosphonic acid etidronate = I-hydroxyethylidene-I, I-bisphosphonic acid clodronate =dichloromethylenebisphosphonic acid ibandronate = 1 -hydroxy-3-(methylpentylamino)propylidene- 1, I-bisphosphonic acid pamidronate = 3-amino- 1 -hydroxypropylidene- I, 1 -bisphosphonic acid risedronate =2-(3-pyridinyl)-l-hydroxyethylidene- 1,l-bisphosphonic acid

2 226 JONATHAN R. GRN T AL. CGP =( l-hydroxy-3-{methyl-(3-phenylpropyl) -amino}-propy1idene)-1, 1 -bisphosphonic acid CGP 42166= (5-methyl-2-thiazolyl) -aminomethylbisphosphonic acid CGP 42446=2-(imidazol-l-yl)- 1-hydroxyethylidene -1,l-bisphosphonic acid CGP 46360=( 1 -hydroxy-3-{ methyl-(2-phenylethyl) -amino}-propylidene)- 1,l-bisphosphonic acid CGP 47072= 3-(N-(phenoxypropyl)-N-methylamino) -1-hydroxypropylidene-1,l-bisphosphonic acid CGP 47875=(5-ethyl-2-thiazolyl) -aminomethylbisphosphonic acid They were all synthesized by the Chemistry Department, Pharmaceuticals Division, Ciba-Geigy Ltd., Basel. To improve solubility, the free acid was converted to its sodium salt by addition of a slight 2-molar excess of 1N NaOH. Ten-fold concentrated stock solutions of the bisphosphonates were then prepared in phosphate-buffered saline, ph 7.4 and diluted for use in 0.9% (w/v) saline. Thyroparathyroidectomized rat assay. The antiresorptive activity of the bisphosphonates was tested in vivo by their ability to inhibit the hypercalcemia induced by 1,25-dihydroxyvitamin D3 (1,25-diOH Dj) in thyroparathyroidectomized rats. The assay is a modification (Green et al. 1994) of a previously published method (Russell et al. 1970). Under the conditions of the assay, a reduction in the serum calcium concentration in the thyroparathyroidectomized rat in vivo is interpreted as indirect evidence for an inhibition of bone resorption by the test compound. Test compounds were administered subcutaneously at various doses once daily for 4 days. The potency of a test compound was calculated by graphical interpolation from a semi-logarithmic plot (percent hypercalcaemia versus log dose) as the dose required to reduce the 1,25-diOH D3-induced hypercalcaemia by 50% (D5,, value). Renal tolerability afer repeated subcutaneous injections. The nephrotoxic potential of repeated injections of bisphosphonate in rats was assessed in a modification of published models which had demonstrated that urinary excretion of the cytosolic enzyme malate dehydrogenase is a sensitive early marker of renal toxicity (Alden et al. 1990; Fent et al. 1988; Zbinden et al. 1988). Groups of 3 rats were first accustomed to housing in single cages for 3 days and then transferred to individual metabolic cages for 5 days (days 4-8) and two 24 hr control urine samples were collected from noon to noon on days 4-5 and 67. After a 2 day recovery period in normal single cages, the rats were returned to individual metabolic cages for 5 days (days 11-15), bisphosphonate (1 mgkg subcutaneously) or saline was injected daily for 5 days and four 24 hr urine samples were collected between days The rats were again allowed a 2 day recovery period in single cages and then returned to metabolic cages for a second treatment cycle. Further bisphosphonate (1 mg/kg) or saline injections were given once daily on days 18-21, and four 24 hr urines were collected between days Finally the animals were killed by carbon dioxide inhalation at the end of day 22. Urinary malate dehydrogenase enzymatic activity was measured in duplicate using a colorimetric assay based on the oxidation of nicotinamide adenine dinucleotide (NADH to NAD+) coupled to the reduction of oxaloacetate to malate by malate dehydrogenase. The assay contained the following components: 2.8 ml phosphate buffer, O.lM, ph 7.5; 0 p1 oxaloacetic acid, 15 mm; 50 pl NADH, 12 mm; 50 pi urine. After addition of the urine sample to the reaction mixture, the decrease in optical absorbance at 340 nm was recorded for 5 min. in a spectrophotometer. nzymatic activity was calculated using a molar absorption coefficient for NADH of 6.22 M-'. cm-' and results were expressed as total units of enzymatic activity/24 hr urine sample. One unit of enzymatic activity (U) hydrolyzed 1 pmole of substrate / min. at ". The compound CGP was used a positive control since it had already been shown to be nephrotoxic to rats in a day toxic- ity trial, causing elevated urea, creatinine and renal enzymatic activity in serum, and renal tubular necrosis (data on file at Ciba-Geigy Ltd., Basel). The clinical compounds etidronate, clodronate and pamidronate were regarded as negative controls since they have been used extensively in patients and are well tolerated if administered correctly. In addition the 3 newer clinical compounds alendronate, risedronate and ibandronate were also tested, as well as the 5 other preclinical compounds CGP 41389, CGP 42446, CGP 46360, CGP and CGP Renal function after a single intravenous infusion. In order to assess further the acute renal tolerability of bisphosphonates selected from the first screen, a second rat model was developed in which the serum urea concentration was the readout. Rats were anaesthetized with Inactin@ (0 mg/kg intraperitoneally) and their body temperature was automatically maintained at 37" with an infrared lamp controlled by a rectal thermocouple. The trachea was cannulated to maintain a patent airway. The jugular vein was cannulated for the intravenous administration of the test substance and for the collection of blood samples. Solutions of pamidronate and CGP in 0.9% (w/v) saline were infused intravenously for 1 hr at a rate of 0.8 mhr. Both compounds were tested at 4 doses (1.5, 5.0, 15 and 50 mg/kg) on groups of 6 animals. An additional control group was infused for 1 hr with saline alone. Venous blood samples (-0.2 ml) were collected at 0, 1, 2, 3 and 4 hr and immediately centrifuged for 3 min. at 1,500Xg. The resulting sera were diluted ten times with 0.9% saline and their urea content was determined in duplicate with a commercial assay kit (Urea nitrogen kit, catalogue no. 640, Sigma Diagnostics Co.). At the end of the experiment the animals were killed with an overdose of Inactin.@ Data analysis. Data were analyzed with the Instat statistics program, GraphPad Software Inc., San Diego, U.S.A. All the results are expressed as medians with maximum and minimum values. Statistical analysis was performed using the Kruskal-Wallis non-parametric analysis of variance test and by Dunn's multiple comparisons test against the relevant control. Spearman's non-parametric rank correlation test was used to determine the degree of association between 2 variables. Results Antiresorptive potency of bisphosphonates in the thyroparathyroidectomized rat. As inhibitors of bone resorption in the thyroparathyroidectomized rat model, the 12 bisphosphonates tested varied greatly in antiresorptive potency (table 1, column 2). The Ds0 values of the clinical compounds etidronate, clodronate, pamidronate, alendronate, risedronate and ibandronate (>,000, 1,0, 85, 8.0, 2.4, 1.1 pglkg, respectively) reflect the successive improvement in potency as each new compound was developed. All the new preclinical compounds in the CGP series were highly potent the thyroparathyroidectomized rat, with DSo values in the range pglkg. To our knowledge, the most potent bisphosphonate synthesized to date is CGP (D pg/kg), which is more than 5 orders of magnitude more potent than etidronate, the first bisphosphonate to be developed for clinical use. Renal tolerability after repeated subcutaneous injections. The data on the effects of repeated injections of bisphosphonates (1 mgkg, subcutaneously) on urinary malate dehydrogenase excretion are presented graphically in fig. 1 and 2.

3 0 1 2o alendronate s RNAL TOLRABILITY OF BISPHOSPHONATS IN RAT MODLS I I clodronate... ]:&:.&/... Y '5 etidronate ibandronate : 0 t 1 0 Q I I I i pamidronate risedronate I I " I I I 1 1 ' ' 1 I I I ' 1 1 ' days Fig. I. ffect of 2 weeks treatment with 6 clinical bisphosphonates on the daily urinary excretion of malate dehydrogenase in the rat. Test compound was injected subcutaneously at a dose of 1 mg/kg on days and (a total of 9 injections). Complete 24 hr urine samples were collected between days 415, 6/7, 11/15 and 18/ 22. Solid circles with bars represent the median, maximum and minimum malate dehydrogenase (MDH) values from 3 animals treated with each bisphosphonate. The dotted lines indicate the median, maximum and minimum MDH values throughout each experiment for 3 saline-treated control animals. The day 7 samples for pamidronate were lost due to operator error. The total malate dehydrogenase activity excreted in the eight 24 hr urine samples collected after the first injection of the test compound is given in table 1, column 3. In the control animals injected with saline, the median malate dehydrogenase excretion for each experiment ranged from 0.62 to 2.8 U/day, with individual minimum and maximum values of 0.14 and 8.1, respectively (dotted lines). The daily median malate dehydrogenase excretion increased from 0.8 U/day at day 5 to 1.9 U/day at day 22, but this trend was not statistically significant (Spearman's r coefficient = 0.3 5, P = 0.33). Treatment with the 6 clinical bisphosphonates produced a small increase in malate dehydrogenase excretion after each cycle of injections (fig. 1). Apart from 2 data points which slightly exceeded the maximum control values (ibandronate, 5.0 units on day 22; risedronate, 5.4 units on day 13), the daily malate dehydrogenase excretion always fell within the range of their respective controls. The total malate dehydrogenase excretion ranged from units, compared to a control range of units (table I). In the series of new bisphosphonates, the 3 compounds CGP 41389, CGP and CGP showed a similar minor trend towards elevated malate dehydrogenase excretion with maximal values of 4.2, 5.4 and 6.3, respectively (fig. 2). With only 3 animals per group a statistical analysis was not possible but it is unlikely that these effects are significant. By contrast, the known nephrotoxic compound CGP produced a marked 60-fold increase in malate dehydrogenase excretion over the control value at day 15, followed by a smaller peak at day 22. CGP and CGP produced a similar, but less pronounced (1 1 and 1 6-fold, respectively) elevation of malate dehydrogenase excretion at day 22, preceded by a smaller peak (fig. 2). The total malate dehydrogenase excretion showed an identical pattern: 3 compounds (CGP 42166, CGP 46360, CGP 47875) markedly increased malate dehydrogenase excretion, whereas 3 others (CGP 41389, CGP 47072, CGP 42446) produced an effect similar to the clinical reference compounds (table 1). In order to rank compounds, a therapeutic ratio (table 1, column 4) was calculated by dividing the reciprocal of the bone antiresorptive Ds0 value from the thyroparathyroidectomized assay in mg/kg by the total units of malate dehydrogenase activity excreted. It is clear from table 1 that, CGP CGP CGP T I CGP [I l0i-,4jr... 0 : I L-rrrtrrrrtrrrr i1+'l>' 'lk'lb' '212 days Fig. 2. ffect of 2 weeks treatment with 6 preclinical bisphosphonates on the daily urinary excretion of malate dehydrogenase in the rat. The range for the off-scale malate dehydrogenase (MDH) value of CGP at day 15 was 57-75, and for CGP at day 22, Further details are given in the legend to fig. 1. 0

4 228 JONATHAN R. GRN T AL. Table 1. Bone antiresorptive potency, short-term renal tolerability, and therapeutic ratio of 12 bisphosphonates in the rat. Compound DSO (P& s.c.)~ Total MDH (U)b Therapeutic ratioc Alendronate 8.0 (1.8, 19, 13) Clodronate 1,0 (500, 20, 7) tidronate >,000 (4) 18.5 <0.003 Ibandronate 1.1 (0.36, 3.0, ) Pamidronate 85 (18, 180, 27) Risedronate 2.4 (0.85, 6.5, 11) CGP (0., 1.9, 9) CGP (0.25,, 6) CGP (26, 190, 14) CGP (0.22, 3.5, 11) CGP (0., 0.75, ) CGP (0.17, 9.0, 11) a bone antiresorptive potency in the thyroparathyroidectomized rat with 1,25-diOH D3-induced hypercalcaemia, expressed as median with the minimum and maximum values from n experiments in parenthesis total units of malate dehydrogenase excreted in the eight 24 hr urine samples collected after initiating treatment with the test compound. For the control animals injected with saline, the range was units therapeutic ratio comparing nephrotoxic potential versus bone antiresorptive potency, calculated by dividing the reciprocal of the DSo value in the thyroparathyroidectomized assay (mglkg) by the total urinary malate dehydrogenase excretion on the basis of therapeutic ratio, CGP was far superior to the other compounds by a wide margin, so it was.p tm- CI 2 CI C 8 e! a Q) CI) c ta c CI Q) 0- I.-I CGP pamidronate I I I I I I I I hours Fig. 3. Acute effect of an intravenous infusion of CGP and pamidronate on the median serum urea level in the rat. Bisphosphonate at doses of 1.5 (m), 5.0 (O), 15 (V) and 50 (A) mg/kg, or saline (0), was administered intravenously to groups of 6 rats for 1 hr (horizontal bar). The asterisk * denotes a P-value<O.OS in Dunn s multiple comparison test against the corresponding control at the same time point. * selected for additional studies. Statistical analysis of the data for all 12 compounds confirmed that there was no significant correlation between renal toxicity and antiresorptive potency (Spearman s r coefficient= 0.05, P=O.9). Renal function after a single intravenous infusion. In order to profile further CGP 42446, its effect was directly compared to that of pamidronate in the second renal tolerability model (fig. 3). During the 4 hr experimental period there was no significant change in the serum urea concentration in the control animals. Both bisphosphonates produced a dose-dependent and time-dependent increase in serum urea concentration. In comparison with the corresponding saline control values at the same time point, the serum urea concentration was significantly different (P<0.05) after 2 hr with pamidronate at doses of 15 and 50 mg/kg, and with CGP only at a dose of 50 mg/kg. The effect of CGP was less marked than that of pamidronate at all time points for all doses except 1.5 mg/ kg, where the results for both compounds were not significantly different from the saline controls. From the doseresponse curves it could be calculated that the dose of compound required to elevate the serum urea concentration by 0% at 4 hr (Dloo) was mg/kg for pamidronate and 38 mg/kg for CGP Discussion In order to profile bisphosphonates for their renal tolerability, a method was adopted from the literature (Fent el al. 1988). It demonstrated that urinary excretion of the enzyme malate dehydrogenase was one of the most sensitive and reliable markers of renal damage, even before histological changes were evident. Although this method had been validated with 13 nephrotoxic reference compounds, none was structurally related to a bisphosphonate so the known

5 RNAL TOLRABILITY OF BISPHOSPHONATS IN RAT MODLS 229 nephrotoxic bisphosphonate CGP was tested as a positive control. The marked elevation of urinary malate dehydrogenase excretion observed after only 4 injections of CGP confirmed the model s suitability for detecting nephrotoxic bisphosphonates. In a similar acute screening model for nephrotoxicity in rats, large doses of clodronate caused an early elevation of another urinary enzyme, N- acetyl-13-d-glucosaminidase (Alden et al. 1990). For the further validation of the assay the 6 clinical bisphosphonates were tested as negative controls since they must have successfully gone through toxicology trials before being administered to humans. Despite the large differences in structure and antiresorptive potency, none of these compounds produced a marked elevation of urinary malate dehydrogenase excretion in the rat under the conditions of the assay. This is consistent with the clinical data which indicate that these compounds are generally well-tolerated when given at the recommended doses required to inhibit bone resorption (Fleisch 1995). Of the preclinical compounds apart from CGP 42166, both CGP and CGP also produced a large increase in malate dehydrogenase excretion so they were excluded from further profiling. Of the remaining compounds, CGP was clearly superior to CGP and CGP in terms therapeutic ratio. In comparison to pamidronate, CGP has a 900-fold higher therapeutic ratio (0.88 versus 790, respectively) due to its greater potency as an inhibitor of bone resorption. While these therapeutic ratios obtained in the rat may not directly reflect tolerability in man, especially for the earlier compounds that are already marketed or in clinical trials, they are valuable as used here to differentiate between compounds within a structural class. In clinical practice the measurement of blood urea levels offers a simple method to assess renal function, with persistent elevation above the normal range of mmol/l indicating impaired renal circulation or glomerular malfunction. By analogy, the acute renal screening model was developed in the rat using the serum urea concentration as a measure of renal function. Although this method is not as sensitive as radionucleotide clearance studies it was considered adequate for early profiling. Pamidronate has been used extensively in man intravenously at doses of approximately mg/kg without any untoward renal effects (Fitton & McTavish 1991). Likewise, in the rat model, a 1 hr intravenous infusion of pamidronate at a dose of 1.5 mg/kg produced no significant effect on the serum urea concentration over the 4 hr observation period. Higher doses ( mg/kg) of both pamidronate and CGP produced a dose-dependent increase in serum urea, indicative of a progressive impairment of renal function. From the Dloo values of 38 and mg/kg, respectively, it can be concluded that a 1 hr infusion of CGP is 3.8-fold less potent than pamidronate as an inhibitor of renal function in rats. It must be emphasized that these doses of CGP ( mg/kg) are 2-3 orders of magnitude greater than the minimum effective clinical dose of approximately 5 pg/kg required to inhibit bone resorption in patients with Paget s disease (Schaffer et al. 1996). In conclusion, the results of these in vivo rat studies comparing bone anti-resorptive potency and nephrotoxic potential for 12 compounds identified CGP as an outstanding new bisphosphonate with a therapeutic ratio almost 3 orders of magnitude higher than that of pamidronate. Moreover, the data indicate that increased potency as an inhibitor of bone resorption is not directly associated with a parallel increase in renal toxicity. In the acute model of renal function, intravenous CGP was again superior to pamidronate, in this case by a factor of 3.8. Since pamidronate has an excellent renal tolerability record based on many years of widespread clinical use, it is unlikely that renal tolerability will be an issue in the further development of CGP References Alden, C. L., R. D. Parker & D. F. astman: Development of an acute model for the study of chloromethanediphosphonate nephrotoxicity. Toxicol. Pathol. 1989, 17, Alden, C. L., J. L. Burns, R. D. Parker, J. L. nglehart & V. W Dennis: Characterization of the early ultrastructural and biochemical events occurring in dichloromethane diphosphonate nephrotoxicity. Toxicol. Pathol. 1990, 18, Bisaz, S., A. Jung & H. Fleisch: Uptake by bone of pyrophosphate, diphosphonates and their technetium derivatives. Clin. Sci. Mol. Med. 1978, 54, Bounameaux, H. M., J. Schifferli, J. P Montani, A. Jung & Chatelanat: Renal failure associated with intravenous diphosphonates. Lancer 1983, I, 471. Dumon, J. C., A. Magritte & J. J. Body: fficacy and safety of the bisphosphonate tiludronate for the treatment of tumor-associated hypercalcemia. Bone Miner. 1991, 15, Fent, K.,. Mayer & G. Zbinden: Nephrotoxicity screening in rats: a validation study. Arch. Toxicol. 1988, 61, Fitton, A. & D. McTavish: Pamidronate: a review of its pharmacological and therapeutic efficacy in resorptive bone disease. Drugs 1991, 41, Fleisch, H. Bisphosphonates in bone disease:,from the laboraiory the patient. Parthenon Publishing Group Ltd., New York & London, Grabenstetter, R. J. & W. A. Cilley: Polynuclear complex formation in solutions of calcium ion and ethane-1-hydroxy-1, I-diphosphonic acid. I. Complexometric and ph titrations. J. Phys. Chem. 1971,15, Green, J. R., K. Miiller & K. A. Jaeggi: Preclinical pharmacology of CGP , a new, potent, heterocyclic bisphosphonate compound. J. Bone Miner. Res. 1994, 9, Kanis, J. A,, C. J. Preston, A. J. l? Yates, R. C. Percival, K. I. Mundy & R. G. G. Russell: ffects of intravenous diphosphonates on renal function. Lancet 1983, I, Kanis, J. A., G. H. Urwin, R.. S. Gray, M. N. C. Beneton,. V. McCloskey, N. A. T. Hamdy & S. A. Murray: ffects of intravenous etidronate disodium on skeletal and calcium metabolism. Amer. J. Med. 1987, 82, suppl. 2A, Mian, M., F. Beghe, A. Caprio, R. Aloj & A. Bertelli: Tolerability and safety of clodronate therapy in bone diseases. Int. J. Clin. Pharm. Res. 1991,11, Russell, R. G. G., R. C. Muhlbauer, S. Bisaz, D. A. Williams & H. Fleisch: The influence of pyrophosphate, condensed phosphates, phosphonates and other phosphate compounds on the dissolution of hydroxyapatite in viiro and on bone resorption induced by parathyroid hormone in tissue culture and in thyroparathyroidectomised rats. Calcif Tissue Res. 1970, 6,

6 230 JONATHAN R. GRN T AL. Schaffer, A. V., H. Buckler, W. Ryan & P. Richardson: A double blind placebo-controlled trial using intravenous zoledronate (CGP 42446) in patients with Paget s disease of bone. J. Bone Miner. Res. 1996, 11, suppl. 1, S373. Troehler, U., J.? Bonjour & H. Fleisch: Renal secretion of diphosphonates in rats. Kidney Int. 1975, 8, Wiers, B. H.: Polynuclear complex formation in solutions of calcium ion and ethane-l-hydroxy-l,i-diphosphonic acid. 11. Light scattering, sedimentation, mobility, and dialysis measurements. J. Phys. Chem. 1971,75, Yates, A. J. P., R.. S. Gray, G. H. Urwin, C. J. Preston, R. G. G. Russell, R. C. Rrcival, R. M. Atkins, N. A. T. Hamdy, M. N. C. Beneton & J. A. Kanis: Intravenous clodronate in the treatment and retreatment of Paget s disease of bone. Lancet 1985, I, Zbinden, G., K. Fent & M. H. Thouin: Nephrotoxicity screening in rats; general approach and establishment of test criteria. Arch. Toxicol. 1988, 61,