PHARMACOKINETICS OF SNX-111, A SELECTIVE N-TYPE CALCIUM CHANNEL BLOCKER, IN RATS AND CYNOMOLGUS MONKEYS

Transcription

1 /97/ $02.00/0 DRUG METABOLISM AND DISPOSITION Vol. 25, No. 3 Copyright 1997 by The American Society for Pharmacology and Experimental Therapeutics Printed in U.S.A. PHARMACOKINETICS OF, A SELECTIVE N-TYPE CALCIUM CHANNEL BLOCKER, IN RATS AND CYNOMOLGUS MONKEYS SCOTT BOWERSOX, JAAP MANDEMA, KATALIN TARCZY-HORNOCH, GEORGE MILJANICH, AND ROBERT R. LUTHER Department of Pharmacology (S.B., K.T.-H., G.M., R.R.L.), Neurex Corporation; and Department of Anesthesiology (J.M.), Stanford University School of Medicine (Received September 18, 1996; accepted December 6, 1996) ABSTRACT:, a selective N-type voltage-sensitive calcium channel blocker, is in clinical trials for the treatment of ischemia-induced brain injury and chronic pain. Pharmacokinetic studies were conducted in rats and cynomologus monkeys to determine the disposition of this compound when it is administered for 24 hr by continuous, constant-rate intravenous infusion. Venous blood samples for determination of plasma levels were collected at regular intervals immediately before, during, and after dosing. Plasma concentrations of equivalents were measured by radioimmunoassay. Pharmacokinetic parameters were derived from plasma concentration-time data using a two-compartment pharmacokinetic model. Results showed close correspondences between pharmacokinetic parameters determined for both is the synthetic form of -conopeptide MVIIA, a selective N-type neuronal VSCC 1 blocker found in the venom of the piscivorous marine snail, Conus magus (1). It is a 25 amino acid, polycationic peptide containing six cysteine residues linked by three disulfide bridges (1). has been shown in animal studies to bind to N-type VSCCs with high affinity (K d 10 pm) and block transmitter release from nerve endings by preventing depolarizationinduced calcium influx (2, 3). It has both analgesic and neuroprotective actions when it is administered to laboratory animals (4 8) and is currently under evaluation in human clinical trials for the relief of pain and for preventing encephalopathies associated with head trauma and CABG surgery. The current investigation determined the pharmacokinetics of when the compound is administered to rats and cynomologus monkeys by continuous, constant-rate intravenous infusion for 24 hr. A summary of the principal findings of this study has been previously reported (9). 1 Abbreviations used are: VSCC, voltage-sensitive calcium channel; CABG, coronary artery bypass graft; RIA, radioimmunoassay; CL, clearance; V SS, volume of distribution at steady-state; t 1 2,, plasma half-life of ; t 1 2,, plasma half-life of ; C SS, steady-state concentration; AUC, area under the curve; ED 50, median effective dose; 4-VO, 4-vessel occlusion. Send reprint requests to: Dr. Scott Bowersox, Neurex Corporation, 3760 Haven Avenue, Menlo Park, CA species. There were no consistent gender- or dose-related differences in calculated kinetic parameters. In all cases, apparent steady-state plasma concentrations were achieved within 2 4 hr of initiating infusion. Steady-state volume of distribution values were 40% of body weight, indicating extravascular dissemination of to both extracellular and intracellular fluids. Elimination curves contained two exponential components. The fast component (rat t 1 2, hr; monkey t 1 2, hr) accounted for 97% of the unit impulse disposition function. The apparent terminal half-life ranged from 4.61 hr (rat) to 6.48 hr (monkey). Current findings constitute the first description of the pharmacokinetics of a member of the -conopeptide family of neuronal calcium channel blockers. Materials and Methods Test Article. acetate was prepared as a lyophilized powder containing 10% acetic acid in salt form and 10% water. The compound was synthesized by a solid-phase method using conventional t-boc chemistry with modifications (10), after procedures previously described for -conopeptide MVIIC (11). Purity was determined to be 98% by reversed-phase HPLC using two different solvent systems. The compound was diluted to appropriate dosing concentrations using 0.9% sodium chloride for injection U.S.P. All solutions were sterile-filtered through a m filter before dosing. Subjects and Study Design. Rats. Sixty Sprague-Dawley rats of either sex, weighing between 169 and 275 g, were used. Animals were assigned to six treatment groups, each consisting of 5 males and 5 females. All rats were surgically implanted with indwelling femoral and jugular vein catheters for intravenous infusions and blood sampling, respectively, the day before test article administration. Animals were given continuous, constant-rate intravenous infusions of either mg/kg/hr (10 mg/kg) or 1.67 mg/kg/hr (40 mg/kg) of for 24 hr at an infusion rate of 2 ml/kg/hr. To avoid drawing an excessive volume of blood from any one animal, three separate groups of rats were bled for each dose level to characterize overlapping segments of the plasma concentration-time curve. Blood samples ( 0.5 ml) were collected before the start of the infusion at 0.25, 1, 2, 6, 12, and 24 hr during the infusion (group 1 rats), and at 0, 0.08, 0.25, 0.5, 1, 2, and 3 hr (group 2 rats), and at 6, 8, 10, 12, and 18 hr (group 3 rats) postinfusion. Monkeys. Two groups of cynomologus monkeys, weighing between 3.2 and 4.4 kg (2 males and 2 females per group), were administered either mg/kg/hr (11.2 mg/kg) or 1.04 mg/kg/hr (25 mg/kg) of for 24 hr at a rate of 1 ml/kg/hr. Dose levels were chosen to provide sufficiently high plasma concentrations for accurate measurement of without causing toxicity. Animals were randomly assigned to the high- and low-dose groups based on body weight, using a method that ensured equal weight distribution among groups. Each monkey was implanted with an indwelling femoral vein catheter for test article infusions 7 days before treatment commenced. A blood sample ( 1 ml) taken from the femoral, brachial, or saphenous vein was collected from each animal before the start of infusion and at 0.25, 2, 4, 8, 12, and 24 hr during infusion. Additional samples ( 1 ml) were taken at 5, 15, 30, and 60 min and at 2, 3, 4, 6, 8, 10, 12, and 24 hr after the end of the infusion period. RIA. quantitation was performed by RIA after methods described by Barksdale et al. (12), with slight modifications. -specific antibody was raised in rabbits using conjugated to bovine serum albumin and

2 380 BOWERSOX ET AL. FIG. 1.Plasma concentration of as a function of time for all rats receiving a 24-hr infusion of either mg/kg/hr (A) or 1.67 mg/kg/hr (B). Each point represents the mean of multiple determinations from a single plasma sample. Infusion of started at time 0 and was terminated at time 24 hr. was used with a radioiodinated derivative prepared and purified according to the method of Kristipati et al. (13). The assay was validated in rat and monkey plasma using 100 l of sample; linear logit log calibration curves were obtained over an concentration range of 4 pg/100 l to333 pg/100 l. It has been determined that an oxidized form of the compound, [sulfoxymethionine (12)], is present in preparations of at concentrations of 3% and is 50% cross-reactive in the RIA. No metabolites of have yet been identified; however, preliminary studies using HPLC separation of plasma samples from human subjects dosed with have found that 97% of the measured immunoreactivity is measured in the -containing fractions. Pharmacokinetic Analysis. Within 30 min of collection, blood samples were centrifuged and plasma was separated and stored frozen at approximately 20 C until analyzed for content. quantitation was performed by validated RIA yielding a lower limit of quantitation of 40.0 pg/ml plasma. Intraassay and interassay reproducibility were within limits of 10% and 20%, respectively. concentrations were measured in samples at each time point. Values derived from points off the linear portion of the RIA standard curve were discarded as were obvious outliers in any set of triplicate measures (i.e. values 2 times the mean of the remaining values). Because the actual times of plasma sample collection generally deviated from the nominal times by no more than a few seconds, nominal times of plasma sample collection were used for the determination of pharmacokinetic parameters. The pharmacokinetic modeling assumed that the concentrations measured by RIA were those of parent compound and not immunoreactive metabolites. Plasma concentration-time data were fitted to a two-compartment pharmacokinetic model using the nonlinear regression program, NONMEM (14), according to the following equation: 2 Ai Cp R i i 1 e i t tr e it, (1) where Cp is the predicted plasma concentration, R is the infusion rate, A i and i are the coefficients and exponents of the unit impulse disposition function, and t r is the infusion duration. t t r is set to 0 if t t r. A proportional error model was used to characterize the variability of the measured and predicted values. The residual error, i, was modeled as follows: Yi Cp i 1 i, (2) where Yi is the ith concentration measurement, and Cp i is the concentration predicted by the pharmacokinetic model. The i s are assumed to be normally distributed with mean 0 and variance 2. Gender and dose were evaluated as covariables on all parameters. The log likelihood parameters test was used to test the significance of each fit (p 0.01). The difference in 2 times the log of the likelihood between a full and reduced model is asymptotically 2 -distributed with the degrees of freedom equal to the difference in the number of parameters between the two models. Results Rats. Individual plasma concentrations at each time point for all data sets were used for the pharmacokinetic analysis. Although coefficients of variation for replicate measures at each time point were consistently 10%, values at some time points were not used because they were obvious outliers. A total of 332 plasma concentrations were available for analysis. Plasma concentration vs. time data for all rats receiving and 1.67 mg/kg/hr are shown in fig. 1. For direct comparison of concentration vs. time data for each sex, the concentration values for rats receiving 1.67 mg/kg/hr were normalized to those receiving mg/kg/hr by multiplying by 0.25 (fig. 2). Visual inspection of the plots indicated: 1) plasma concentration approximated steady-state within 2 hr of initiating infusion; 2) disposition after cessation of infusion was composed of at least two exponential components; 3) the time course of drug disposition was similar regardless of dose; and 4) the time course of drug disposition was comparable for males and females. Quantitative analyses confirmed these qualitative observations. concentration-time data were fitted to a two-compartment pharmacokinetic model using CL, V SS, and t 1 2, and t 1 2, as parameters. Because there were no statistically significant differences in kinetic parameters calculated independently for either sex or SNX-

3 NEURONAL CALCIUM CHANNEL BLOCKER PHARMACOKINETICS 381 FIG. 2.Concentration-time curves for male (A) and female (B) rats receiving a 24-hr, intravenous infusion of or 1.67 mg/kg/hr. Each point is the mean of multiple determinations from a single plasma sample taken from each rat. For the purpose of direct comparison, concentration values for the 1.67 mg/kg/hr group were normalized to those of the mg/kg/hr group. infusion was initiated at time 0, and infusion was terminated at time 24 hr. 111 dosage level, data for all animals were combined and analyzed with the two-compartment model. Pharmacokinetic parameters are given in table 1. From the CL and rate of infusion, C SS s were calculated to be 1.1 and 4.2 mg/liter for the low- and high-dose groups, respectively. During the infusion, steady-state was rapidly approached. The times to reach 50%, 90%, and 95% of the C SS s were 0.4, 1.4, and 2.1 hr, respectively. With respect to disposition, most of the AUC was associated with the fast phase of the disposition function. From the fit to all data, the fast component contributed 96% of the AUC, and the slow component contributed 4%. More than 99% of the compound was cleared from plasma by 18 hr after cessation of infusion. Approximately 93% of animals receiving 1.67 mg/kg/hr exhibited slight tremors during the infusion period, compared with 17% of those receiving the mg/kg/hr dose. Tremors were observed within 1 hr of infusion onset, before plasma reached steady-state levels. Because there were no changes in the incidence of tremor after plasma levels reached steady-state, ED 50 concentrations for tremor were estimated to lie between 1.1 and 4.2 mg/liter. Monkeys. A total of 428 plasma concentrations measured in 152 plasma samples were available for analysis. Comparison of individual plasma concentrations of at each time point during infusion of and after infusion of disclosed no consistent differences between values for male and female animals. Data for both sexes were therefore combined for pharmacokinetic analyses. With the exception of one animal in the high-dose group, coefficients of variation for replicate measures of plasma at each time point were consistently 10%. Plasma concentrations for each of the four monkeys receiving either the or 1.04 mg/kg/hr dose are plotted as a function of time after initiation of infusion in fig. 3. The fitted curves for the two dose groups are shown as solid lines. Apparent C SS s were achieved within 4 hr of initiating infusion, and the decline in concentrations after cessation of infusion contained at least two exponential components. The two-compartment pharmacokinetic model (eq. 1) was parameterized in CL, V SS, t 1 2, and t 1 2,. Results of this analysis are presented in table 2. Measured plasma concentrations of closely approximated those predicted, confirming the appropriateness of the model used. The time to reach 50%, 90%, and 95% of the steady-state plasma concentration was 0.75, 2.7, and 3.7 hr, respectively. Steady-state plasma concentrations were estimated to be 1.4 mg/liter (low-dose group) and 3.2 mg/liter (high-dose group). With respect to disposition, most of the AUC was associated with the fast phase of the disposition function. For the fit to all data (lowand high-dose data combined), the fast phase contributed 97% of the AUC, and the slow phase contributed 3%. All animals of the high-dose (1.04 mg/kg/hr) group and half (2 of 4) of the animals of the low-dose (0.467 mg/kg/hr) group exhibited slight tremors during the infusion period. Tremors were observed within 30 min of initiating infusion and before plasma concentrations reached 50% of steady-state levels. Because all animals of both dose groups exhibited tremors after 24 hr of infusion, the EC 100 plasma concentration was estimated to be no more than the steady-state value for the low-dose group (1.4 mg/liter). Discussion Current findings constitute the first description of the pharmacokinetics of a member of the -conopeptide family of neuronal calcium channel blockers. The aim of this study was to determine the pharmacokinetics of the selective, N-type VSCC blocker,, in rats and cynomologus monkeys when the compound is administered by continuous, intravenous infusion for 24 hr. Time-concentration data for both species were best fit to a two-compartment pharmacokinetic model and yielded similar estimates of kinetic parameters. No dosedependency in pharmacokinetics was observed, nor did the estimated parameters vary by gender.

4 382 BOWERSOX ET AL. TABLE 1 Pharmacokinetic parameters for administered to Sprague-Dawley rats by continuous, constant-rate intravenous infusion for 24 hr Parameters (values in right column) are derived from pooled data for all groups. Values given are the means SEM. Male Female Parameter mg/kg/hr 1.67 mg/kg/hr mg/kg/hr 1.67 mg/kg/hr All Data CL (liters/hr/kg) V SS (liters/kg) t 1 2, (hr) t 1 2, (hr) FIG. 3.Plasma concentrations of as a function of time for four monkeys receiving 24-hr infusion of either mg/kg/hr (A) or 1.04 mg/kg/hr (B). Each point represents the mean of triplicate determinations from a single plasma sample taken from each monkey at the indicated time point. Time 0is the time of infusion onset, and infusion was terminated at time 24 hr. TABLE 2 Mean ( SD) pharmacokinetic parameters for administered to cynomologus monkeys by continuous, constant-rate intravenous infusion for 24 hr Model parameterized in CL, V SS, t 2,, and t 1 1 2, (method 2). Values given are the means SEM. Parameter mg/kg/hr 1.04 mg/kg/hr All Data CL (liters/hr/kg) V SS (liters/kg) t 1 2, (hr) t 1 2, (hr) Distribution volumes at apparent steady-state exceeded plasma volumes. Steady-state volume of distribution values were 40% of body weight, indicating that was distributed to both extracellular and intracellular fluid compartments. This finding is consistent with previous findings (unpublished) that showed that binding of to plasma proteins is relatively low (rat: 56%, monkey: 73%), nonspecific, nonsaturable, and of low affinity; thus, unlikely to lead to significant drug retention in the vascular space. Based on current findings, it would seem that continuous, intravenous infusion is the most appropriate mode of drug delivery for achieving and maintaining therapeutic blood levels of. Although the apparent terminal half-life of is several hours, most of the area under the unit impulse disposition curve is associated with a more rapid phase of elimination (t 1 2, 44 min in the monkey). Thus, a repetitive, intravenous bolus dosing paradigm would require an unrealistically short dosing interval to achieve desired average plasma concentrations of without producing large fluctuations. Therefore, a continuous, intravenous infusion paradigm is being used in clinical studies of (15, 16). The results of this study have important clinical implications. Intravenously administered produces significant neuroprotection in animal models of global and focal cerebral ischemia, suggesting that N-type VSCC blockers may be useful for the prevention of human brain injury in such settings as head trauma and stroke. is neuroprotective in the rat 4-VO model of transient, global forebrain ischemia when a single dose is administered intravenously as late as 24 hr after the ischemic insult (7). Neuroprotective potency is increased when is administered for 24 hr by continuous, constant-rate intravenous infusion rather than bolus injection (ED 50 : intravenous bolus injection 12.5 mg/kg vs. intravenous infusion 0.4 mg/kg/24 hr) (17). Neuroprotective doses of adminis-

5 NEURONAL CALCIUM CHANNEL BLOCKER PHARMACOKINETICS 383 tered by continuous intravenous infusion 24 hr (ED 50 ) achieve steady-state plasma concentrations of 15 ng/ml (unpublished observation). The neuroprotective effects of are clearly mediated by blockade of central N-type neuronal VSCCs, as retains its neuroprotective efficacy in the rat 4-VO model when it is administered intracerebroventricularly in doses as low as 0.1 g (18). Because neuroprotective potency is enhanced by continuous intravenous infusion relative to bolus injection, the continuous intravenous administration regimen has been chosen for use in phase I and II clinical trials. The low-dose continuous intravenous infusion regimen reduces the likelihood of encountering potential adverse effects of that may be the result of higher peak plasma concentrations associated with bolus dose administration. The continuous intravenous infusion regimen also allows for the treatment of both primary and secondary ischemic events that are encountered commonly following severe head trauma. Phase I human safety and tolerability studies have been conducted with administered by continuous intravenous infusion in doses ranging from ng/kg/hr (0.3 g/kg/24 hr) to 42 g/kg/hr (1 mg/kg/24 hr) for 24 hr. Pharmacological activity (sympathetic nervous system inhibition secondary to N-type VSCC blockade) has been demonstrated in humans (15). Doses as high as 42 g/kg/hr (1 mg/kg/24 hr) were safe and well tolerated (16), and preliminary unpublished human pharmacokinetic data suggest that doses substantially 42 g/kg/hr achieve target blood levels of 15 ng/ml (ED 50 rat 4-VO). Phase II human safety, efficacy, and feasibility studies are underway in patients with severe closed head trauma and, as previously indicated, in patients undergoing elective CABG with cardiopulmonary bypass to treat the postpump encephalopathy syndrome. References 1. B. M. Olivera, L. J. Cruz, V. de Santos, G. W. Le Cheminant, D. Griffin, R. Zeikus, et al.: Neuronal calcium channel subtypes using -conotoxin from Conus magus venom. Biochemistry 26, (1987). 2. S. Gaur, R. Newcomb, B. Rivnay, J. R. Bell, J. Ramachandran, and G. P. Miljanich: Calcium channel antagonist peptides define several components of transmitter release in the hippocampus. Neuropharmacology 33, (1994). 3. R. Kristipati, L. Nadasdi, K. Tarzcy-Hornoch, G. P. Miljanich, J. Ramachandran, and J. Bell: Characterization of the binding of -conopeptides to different classes of non-l-type neuronal calcium channels. Mol. Cell. Neurosci. 5, (1994). 4. A. Malmberg and T. Yaksh: Voltage-sensitive calcium channels in spinal nociceptive processing: blockade of N- and P-type channels inhibits formalin-induced nociception. J. Neurosci. 14, (1994). 5. S. R. Chaplan, J. W. Pogrel, and T. L. Yaksh: Role of voltage-dependent calcium channel subtypes in experimental tactile allodynia. J. Pharmacol. Exp. Ther. 269, (1994). 6. K. Valentino, R. Newcomb, T. Gadbois, T. Singh, S. S. Bowersox, S. Bitner, et al.: A selective N-type calcium channel antagonist protects against neuronal loss after global ischemia. Proc. Natl. Acad. Sci. U.S.A. 90, (1993). 7. A. M. Buchan, S. Z. Gertler, H. Li, D. Xue, Z.-G. Huang, K. E. Chaundy, et al.: A selective N-type Ca 2 channel blocker prevents CA1 injury 24 hours following severe forebrain ischemia and reduces infarction following focal ischemia. J. Cereb. Blood Flow Metab. 14, (1994). 8. S. Takizawa, K. Matsushima, H. Fujita, K. Nanri, S. Ogawa, and Y. Shinohara: A selective N-type calcium channel antagonist reduces extracellular glutamate release and infarct volume in focal cerebral ischemia. J. Cereb. Blood Flow Metab. 15, (1995). 9. S. S. Bowersox and R. R. Luther: N-type voltage-sensitive calcium channel antagonist. Drugs Fut. 19, (1994). 10. D. Yamashiro and C. H. Li: New segment synthesis of alpha-inhibin-92 by the acyl disulfide method. Int. J. Peptide Protein Res. 31, (1988). 11. D. R. Hillyard, V. D. Monje, I. M. Mintz, B. P. Bean, L. Nadasdi, J. Ramachandran, et al.: A new Conus peptide ligand for mammalian presynaptic calcium channels. Neuron 9, (1992). 12. C. M. Barksdale, G. D. Nordblom, G. M. Miljanich, K. Tarczy-Hornoch, R. Kristipati, and A. R. Kugler: Development of a radioimmunoassay for the novel N-type calcium channel blocker, (CI-1009). J. Clin. Ligand Assay in press. 13. R. Kristipati, L. Nadasdi, K. Tarczy-Hornoch, K. Lau, G. P. Miljanich, J. Ramachandran, and J. R. Bell: Characterization of the binding of -conopeptides to different classes of non-l-type neuronal calcium channels. Mol. Cell. Neurosci. 5, (1994). 14. S. L. Beale and L. B. Sheiner: NONMEM User s Guide. NONMEM Project Group, University of California, San Francisco, R. R. Luther, W. D. Sperzel, and K. L. Valentino: Pharmacological activity of, the first N-type, neuron specific, calcium channel blocker, is demonstrated in man. J. Clin. Pharmacol. 34, 1014 (1994). 16. R. R. Luther, N. Tich, and W. D. Sperzel: A double-blind, placebocontrolled safety and tolerability study of. J. Clin. Pharmacol. 35, 993 (1995). 17. T. Singh, T. Gadbois, K. Valentino, R. R. Luther, G. P. Miljanich, and J. Ramachandran: 24 hour infusion of protects against neuronal death after cerebral ischemia with minimal adverse effects. Soc. Neurosci. Abstr. 19, 850 (1993). 18. K. Valentino, T. Gadbois, T. Singh, S. S. Bowersox, S. Bitner, and A. Justice, et al.: -Conopeptides prevent damage to CA1 neurons caused by global ischemia. Soc. Neurosci. Abstr. 17, 1079 (1991).