J. vet. Pharmacol. Therap. 34, 388396. doi: 10.1111/j.1365-2885.2010.01244.x.

Studies on the pharmacokinetics and pharmacodynamics of mirtazapine in healthy young cats

J. M. QUIMBY* D. L. GUSTAFSON* B. J. SAMBER & K. F. LUNN*
*Department of Clinical Sciences, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, CO, USA; Merial, Analytical R&D, North Brunswick, NJ, USA

Quimby, J. M., Gustafson, D. L., Samber, B. J., Lunn, K. F. Studies on the pharmacokinetics and pharmacodynamics of mirtazapine in healthy young cats. J. vet. Pharmacol. Therap. 34, 388396. Mirtazapine pharmacokinetics was studied in 10 healthy cats. Blood was collected before, and at intervals up to 72 h after, oral dose of 3.75 mg (high dose: HD) or 1.88 mg (low dose: LD) of mirtazapine. Liquid chromatography coupled to tandem mass spectrometry was used to measure mirtazapine, 8-hydroxymirtazapine and glucuronide metabolite concentrations. Noncompartmental pharmacokinetic modeling was performed. Median half-life was 15.9 h (HD) and 9.2 h (LD). Using MannWhitney analysis, a statistically signicant difference between the elimination half-life, clearance, area under the curve (AUC) per dose, and AUC dose of the groups was found. Mirtazapine does not appear to display linear pharmacokinetics in cats. There was no signicant difference in glucuronidated metabolite concentration between groups. Pharmacodynamics was studied in 14 healthy cats administered placebo, LD and HD mirtazapine orally once in a crossover, blinded trial. In comparison with placebo, cats ingested signicantly more food when mirtazapine was administered. No difference in food ingestion was seen between HD and LD, but signicantly more behavior changes were seen with the HD. Limited serum sampling during the pharmacodynamic study revealed drug exposure comparable with the pharmacokinetic study, but no correlation between exposure and food consumed. Mirtazapine (LD) was administered daily for 6 days with no drug accumulation detected. (Paper received 23 June 2010; accepted for publication 13 September 2010) Jessica M Quimby, Department of Clinical Sciences, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, 300 West Drake Road, Fort Collins, CO 80523, USA. E-mail: jquimby@colostate.edu

INTRODUCTION Anorexia is a common clinical sign in cats and is associated with many systemic diseases. An appropriate diagnostic investigation is necessary to determine the underlying cause, but in many cases, anorexia may continue for an extended period of time during patient evaluation and initial therapy. Nutritional support is a key component of successful patient management. While the effects of inadequate nutrition in the diseased state may be clinically subtle, they involve many organ systems and are associated with increased morbidity and mortality (Sanderson, 2000). In the cat specically, the onset of hepatic lipidosis following anorexia or hyporexia is well documented (Center, 2005). In the acutely ill patient, management of anorexia may result in more rapid recovery. Options for nutritional support include the use of appetite stimulants, placement of feeding tubes and administration of total parenteral nutrition. In the chronically ill
388

patient, management of anorexia likely results in improved quality of life as well as improvements in survival (Kopple, 1994). For example, appetite stimulants may be helpful in the long-term management of chronic renal failure, particularly as ingestion of specially formulated renal diets has been shown to improve outcome (Elliott et al., 2000; Plantinga et al., 2005; Ross et al., 2006). Mirtazapine is a presynaptic a2-adrenergic receptor antagonist that increases noradrenergic and serotonergic neurotransmission by blocking presynaptic inhibitory receptors, resulting in increased norepinephrine release into the synaptic cleft and therefore increased postsynaptic availability (de Boer, 1996). The serotonergic effects occur through 5-HT1 receptor-like activity as well as enhancement of serotonergic transmission by norepinephrine. In contrast to these effects, mirtazapine is also a potent postsynaptic 5-hydroxytryptamine (HT)3 and 5-HT2 antagonist. It is thought that the receptor-specic anti-serotonergic effects help ameliorate side effects of the
2010 Blackwell Publishing Ltd

Mirtazapine pharmacokinetics and pharmacodynamics in cats 389

drug (de Boer, 1996). Although the original mechanism of action studies was performed in rats and no information is available for cats, mirtazapine likely exerts its purported effect through analogous mechanisms in the species of interest. Mirtazapine was originally introduced to human medicine as an antidepressant; however, it has recently attracted interest in veterinary medicine owing to several desirable effects, namely its signicant anti-nausea, anti-emetic and appetite-stimulating properties. These effects appear to be a result of antagonism of the 5-HT3 receptor, which is an important receptor in the physiology of emesis. Noradrenergic effects appear to be responsible for the antidepressant properties (Timmer et al., 2000). An initial uncontrolled clinical trial (Cahill, 2006) and numerous anecdotal reports have encouraged continued clinical use of mirtazapine in veterinary medicine. Doses for cats and dogs have been extrapolated from human doses. A dose of 1.88 or 3.75 mg (1 8 or 1 4 of a 15 mg tablet) every 3 days has been recommended for cats (Cahill, 2006); however, no pharmacological studies have been reported to date in support of these extrapolations. Anecdotal observations by the authors noted that in most cases, the effect of the drug appears to have dissipated by the second day after administration, implying a shorter dosing interval may be more appropriate. Side effects that have been noted clinically in cats include tremors, muscle twitching and hyperactivity (Cahill, 2006). Human pharmacokinetic data have demonstrated that a number of factors affect the metabolism of mirtazapine. Mirtazapine displays linear pharmacokinetics over a wide dose range in humans, with an elimination half-life (t1 2k) range of 2040 h. Time to maximum serum concentration (Tmax) ranges from 1 to 2 h, and maximum serum concentration (C max) is reported to range from 56.1 to 74.5 lg L in healthy men with a concomitant area under the curve to innity (AUC) range of 397950 lg Lh (Timmer et al., 2000). No concentrationeffect relationship has been found in humans treated with mirtazapine, and when used at therapeutic doses, the serum concentration has been reported to range from 5 to 100 lg L (Timmer et al., 2000). Differences in metabolism were found between sexes and ages, and for patients suffering from hepatic or renal impairment (Timmer et al., 2000). To provide accurate dosing recommendations, it would therefore be prudent to investigate the pharmacokinetics of mirtazapine in veterinary patients. This is particularly important in feline patients as the drug may be metabolized more slowly owing to the decreased availability of some metabolic pathways in this species. For example, cats lack glucuronyl transferase, which is necessary for glucuronide conjugation (Lainesse et al., 2006) As mirtazapine is metabolized initially by demethylation and oxidation, followed by conjugation to glucuronic acid, some differences in drug metabolism and elimination for felines would be expected (Timmer et al., 2000). Therefore, this project was designed to assess the pharmacokinetics and pharmacodynamics of the commonly prescribed doses of mirtazapine in healthy cats to provide information to guide the use of mirtazapine as an appetite stimulant in cats.
2010 Blackwell Publishing Ltd

MATERIALS AND METHODS Drug preparation Commercially available generic 15 mg mirtazapine (Aurobindo Pharma USA, Inc., Dayton, NJ, USA) tablets were compounded by the pharmacy at the Colorado State University Veterinary Medical Center according to Professional Compounding Centers of America protocol. The method used is guaranteed to produce accurate compounding within 10% of the target dose. Analysis of random compounded capsules for mirtazapine content using liquid chromatography coupled to tandem mass spectrometry (LC MS MS) showed accuracies of 94.5 4.6% to the intended content and a stability of at least 6 months as formulated. The capsules were compounded within 1 month of utilization and stored at room temperature. For the pharmacodynamic study, groups of capsules (lactose placebo, 1.88 and 3.75 mg mirtazapine) were manufactured and randomly coded by the pharmacy staff who kept the code concealed until completion of the study. Pharmacokinetic sampling Seven purpose-bred and three privately owned young adult healthy cats were enrolled in the pharmacokinetic study after undergoing a physical examination, complete blood count, serum biochemistry prole and urinalysis to rule out systemic disease. Breeds included nine domestic shorthairs and one domestic longhair. Six cats were spayed females, and four cats were castrated males. All portions of the project were approved by the Institutional Animal Care and Use Committee at Colorado State University, and all owners signed consent forms prior to participation. Ten cats were randomly assigned to either the high-dose (HD) group that received 3.75 mg or to the low-dose (LD) group that received 1.88 mg of mirtazapine. Capsules were administered orally, followed by 3 mL of water given with a syringe. The cats were fasted for 12 h before beginning the study. A jugular catheter was placed under ketamine (20 mg per cat IV) and butorphanol (0.1 mg kg IV) sedation 3 h prior to mirtazapine administration to allow for ease of sample collection. Blood samples (1.5 mL) were collected prior to, and at 0.25, 0.5, 1, 4, 8, 24, 48 and 72 h after mirtazapine administration. Samples were centrifuged within 10 min of collection, and serum was harvested and stored at )80 C until analysis. Mirtazapine analysis Mirtazapine was measured using liquid chromatography coupled to tandem mass spectrometry (LC MS MS) using a method modied from one previously published (Doherty et al., 2007). The analysis was carried out in the Pharmacology Core at the Colorado State University Veterinary Medical Center. An LC MS MS-based assay was developed and validated for the analysis of mirtazapine in cat serum. For sample preparation, a 200 lL aliquot of serum was transferred to a 1.5-mL polypropylene microcentrifuge tube, and 10 ng of trazodone (10 lL of 1 lg mL) was added as an internal standard. Proteins were then

390 J. M. Quimby et al.

precipitated by the addition of 500 lL of acetonitrile and vortexing for 10 min. Samples were centrifuged at 17 900 g for 10 min, and the resulting supernatant was collected. Standards and quality assurance quality control (QA QC) samples were constructed by spiking 200 lL of blank cat serum with known amounts of mirtazapine and processing as described above. Positive-ion electrospray ionization mass spectra were determined with a 3200 Q-TRAP triple quadrupole mass spectrometer (Applied Biosystems, Foster City, CA, USA) with a turbo ionspray source interfaced to a 1200 Series Binary Pump SL liquid chromatography system (Agilent Technologies Inc, Santa Clara, CA, USA) and HTC-PAL autosampler (Leap Technologies, Carrboro, NC, USA). Chromatography was performed on a SunFire C18, 2.5-lm, 4.6 50 mm column (Waters Corporation, Milford, MA, USA) with a liquid chromatography gradient employing 10 mM ammonium acetate (mobile phase A) and acetonitrile (mobile phase B). Initial chromatographic conditions were 60:40 (mobile phase A mobile phase B) with increasing B from 40 to 95% from 0 to 2 min, holding at 95% B from 2 to 2.75 min and re-establishing initial conditions from 2.75 to 4 min. The ow rate was 1 mL min, injection volume was 420 lL, and the total analysis time was 4 min. Assay performance for each batch was assessed utilizing at least 10% QA QC samples dispersed among unknown samples at low (1 ng mL), mid (10 ng mL) and high (100 ng mL) ranges of the standard curve (0.5500 ng mL) with batches failing if >25% of the QA QC samples were outside of the accepted level of 85% accuracy. Accuracy of QA QC samples among the batches analyzed for this study ranged from 94.5 4.6 to 92.2 6.8%. The lower limit of quantitation (LLOQ) for this assay was based on the level of detection with >85% accuracy and a CV (%) <15% and was determined to be 0.5 ng mL. Assay performance was linear to >500 ng mL, but 500 ng mL was used as the upper limit to the assay as utilized because of no unknown sample measurements exceeding this value. The mass spectrometer settings were as follows: turbo ion-spray temperature, 550 C; ion-spray voltage, 1500 V; declustering potential, 48 V; entrance potential, 4.5 V; collision energy, 37 V; collision cell exit potential, 2 V; collision cell entrance potential, 11 V; collision gas, N2, medium; curtain gas, N2, 35 units; nebulizer gas, N2, 40 units; and auxiliary gas, N2, 20 units. Samples were quantied by the internal standard reference method in the multiple reaction monitoring mode by monitoring the transition m z 266 195 for the analyte mirtazapine and m z 372 176 for the internal standard trazodone. 8-Hydroxymirtazapine and 8-Hydroxymirtazapine glucuronide analysis For the analysis of 8-hydroxymirtazapine and 8-hydroxymirtazapine glucuronide, aliquots of cat serum samples were treated with either b-glucuronidase or saccharic acid. The addition of saccharic acid to samples was for the inhibition of endogenous glucuronidase activity. To each 100 lL aliquot of cat serum was added 1 ng of trazodone (10 lL of 100 ng mL) followed by either 2500 units (50 lL of 50 units lL) of b-glucuronidase or 50 lL of

300 lM saccharic acid. Samples were vortexed briey to mix and incubated at 37 C for 1 h. Samples were then prepared by acetonitrile protein precipitation as described for mirtazapine. Chromatographic conditions used were identical to those used for mirtazapine, and the mass spectrometer settings used were as follows: turbo ion-spray temperature, 350 C; ion-spray voltage, 1500 V; declustering potential, 48 V; entrance potential, 7 V; collision energy, 37 V; collision cell exit potential, 2 V; collision cell entrance potential, 11 V; collision gas, N2, low; curtain gas, N2, 35 units; nebulizer gas, N2, 40 units; and auxiliary gas, N2, 20 units. 8-Hydroxymirtazapine was monitored as a transition m z 282 211. The relative amount of 8-hydroxymirtazapine and 8-hydroxymirtazapine glucuronide in samples was calculated based on a standard curve utilizing mirtazapine owing to the lack of availability of 8-hydroxymirtazapine for use as a standard. 8-Hydroxymirtazapine values were determined from the saccharic acidtreated samples, and the amount of 8-hydroxymirtazapine glucuronide was calculated as the amount in b-glucuronidase minus saccharic acidtreated samples. Supplies and reagents b-Glucuronidase (G4511), mirtazapine (M0443), saccharic acid (S0375) and trazodone (T6154) were purchased from SigmaAldrich Co. (St Louis, MO, USA). All other reagents were of reagent grade or better. Pharmacodynamic study This was a randomized, double-blind, placebo crossover study. Fifteen privately owned young adult healthy cats were enrolled in the pharmacodynamic study after undergoing a physical examination, complete blood count, serum biochemistry prole and urinalysis to rule out systemic disease. One cat was excluded owing to the development of an upper respiratory infection. Breeds included seven domestic shorthairs, four domestic longhairs, two siamese mixes and one Burmese. Eight cats were spayed females, and six cats were neutered males. Cats were fasted for 12 h before beginning the study. Fourteen cats each received placebo, 1.88 and 3.75 mg of mirtazapine on separate occasions, 1 week apart in a predetermined random sequence. Capsules were administered orally, followed by 3 mL of water administered by syringe. Before capsule administration, a baseline behavioral assessment was carried out (see Appendix 1; the behavioral scores from each observation over the 8-h period were tallied to give the total behavior score for that particular treatment). One hour after capsule administration, the cats received a behavioral assessment and one cup of their usual dry diet, which was weighed on a gram scale (ZL200630016361.1, China), the accuracy of which was assessed by measuring National Institute of Standards and Technology traceable calibration weights. Blood samples were collected for mirtazapine serum concentration at 2 and 8 h after capsule administration. At 1-h intervals for an 8-h period, the cats remaining food was weighed and a behavioral assessment was carried out by a single blinded observer. The cats were released to their owners 9 h after
2010 Blackwell Publishing Ltd

Mirtazapine pharmacokinetics and pharmacodynamics in cats 391

mirtazapine placebo administration. Owners were asked to complete a home assessment sheet (see Appendix 2) including perceived changes in appetite, vocalization and activity the evening of, and the morning after, the administration of mirtazapine placebo. Estimation of AUC in pharmacodynamic study cats utilizing a limited sampling approach Blood samples were collected 2 and 8 h after capsule administration for the analysis of mirtazapine serum concentration; samples were handled as previously described. Estimation of the AUC of mirtazapine for cats in the pharmacodynamic study was determined (Panetta et al., 2003). A multiple linear regression analysis showed that the 8-h time point alone gave the best correlation to the AUC in the 10 animals from the full pharmacokinetic study with an r2 = 0.909. Utilizing this model to predict the AUC024 in the training set gave an accuracy and precision (CV%) of 79.4 21.5 and a median accuracy and range of 84.8 and 49.698.1, respectively. Assessment of drug accumulation The accumulation factor at steady-state following multiple doses was estimated using the following equation: Accumulation Factor = 1 1-e-kel*T. The terminal elimination rate was used for estimating the accumulation factor as the applicable kel and the dosing interval (T) were set at 24 h. In addition to calculation of accumulation based on the pharmacokinetic data, LD mirtazapine was administered daily to determine in vivo whether signicant drug accumulation would occur. Four cats (two female and two male) from the seven purpose-bred cats that participated in the pharmacokinetic study were utilized for the daily dosing study. The cats received LD mirtazapine orally once daily for 6 days. On days 3 and 6, blood samples were collected immediately before, and 2 h after, capsule administration and processed as previously described. Subjective behavior assessment was performed during this period (objective assessment was performed in the pharmacodynamic portion of the project). Pharmacokinetic analysis Data were analyzed, and pharmacokinetic parameters were calculated utilizing noncompartmental methods. Data from the 72-h samples were not included as some cats did not have detectable levels of drug at this time. Pharmacokinetic parameters that were determined included AUC0t, AUC, t1 2k, Tmax and Cmax. Because mirtazapine was administered via an extravascular route, absorbed dose is equal to D (dose) bioavailability (F), and thus, the estimation of the parameters dependent on D [clearance (CL) and volume of distribution (Vd)] is represented as CL F and V F (Wagner, 1993). The data from the initial pharmacokinetic time course study were utilized to develop a limited sampling approach for pharmacokinetic analysis during the pharmacodynamic study. Analysis of the AUC data vs. specic time point values from the 5
2010 Blackwell Publishing Ltd

HD and 5 LD animals was conducted by multiple linear regression (MINITAB v15.1.1.0; Minitab Inc., State College, PA, USA) to determine whether one to three samples could accurately predict the total AUC value. (Panetta et al., 2003) The AUC vs. the 8-h time point resulted in the best prediction with an r2 = 0.909. Therefore, the AUC for the pharmacodynamic portion of the studies was estimated based on the value of the 8-h time point utilizing the equation AUC (ng mL h) = )21.4 + 28.6 [8 h time point (ng mL)].

Statistical analysis Comparison of the pharmacokinetic parameters between HD and LD groups was performed with MannWhitney statistical analysis. Comparison of percentage of food ingested and the total behavioral scores for each day during the pharmacodynamic study was performed with repeated measures ANOVA and Bonferroni post hoc analysis. Spearman rank correlation was used to explore the relationship between drug exposure and pharmacodynamic effect. PRISM software (GraphPad Software, Inc., La Jolla, CA, USA) was used for the analysis. Comparison of at-home assessment parameters was performed using a mixed-effects logistic regression model (Littell, 2006) with SAS software (SAS, version 9.2; SAS Institute Inc., Cary, NC, USA). Cat identity was nested within treatment sequence (the primary exposure of interest) to account for random and repeated effects. Treatment sequence was included as a random effect in each model, regardless of whether a signicant association could be identied when treatment sequence was analyzed as a xed effect. Even though incomplete washout was not expected, this method conservatively accounts for possible incomplete washout (Littell, 2006). For all analyses, a P-value of <0.05 was considered to be statistically signicant.

RESULTS Pharmacokinetic study Ten young adult healthy cats with normal physical examination, serum biochemistry prole, complete blood count and urinalysis were entered in the study. Demographics for these cats are illustrated in Table 1. Comparison of the mg kg dose of mirtazapine between the HD and LD groups, using a Mann Whitney test, demonstrated that the doses were signicantly different (P = 0.015). No cats experienced adverse effects from drug administration although increased activity, social interaction and vocalization were noted subjectively in nine cats (objective assessment was performed in the pharmacodynamic portion of the project). Pharmacokinetic parameters were calculated using a noncompartmental model and are shown in Table 2. Overall, there was a statistically signicant difference between the elimination half-life, CL, AUC dose and AUC dose of the two groups when they were compared with MannWhitney nonparametric statistics (P = 0.03 for all comparisons). Drug concentration curves for cats in the HD and LD groups are shown in Fig. 1.

392 J. M. Quimby et al. Age (years) Study type PK single dose (n = 10 cats) Daily multiple dosing (n = 4 cats) PD single dose (n = 14 cats) Median 2.3 1.4 3.3 Range 1.44.1 1.41.4 1.04.5 Weight (kg) Median 5.3 5.1 4.2 Range 3.97.8 3.876.45 2.96.8 Dose (mg kg) Dose group Median LD HD LD HD LD HD 0.44 0.62 0.37 NA 0.45 0.89 Range 0.350.48 0.480.73 0.290.49 NA 0.270.64 0.561.27 Table 1. Summary of demographic data by study

HD, high dose; LD, low dose.

Table 2. Pharmacokinetic parameters of orally dosed mirtazapine in healthy cats High dose (HD) (3.75 mg) Pharmacokinetic parameter Cmax (ng mL) Cmax Dose (ng mL) (mg kg) Tmax (h) Half-life (h)* AUC (ng mL h) AUC Dose (ng mL h) (mg kg)* AUC (ng mL h) AUC Dose (ng mL h) (mg kg)* CL F (L h kg)* Vdz F (L)

Low dose (1.88 mg) Mean SD Median 45.3 126.2 1 9.2 364.9 830 376.1 891 1.12 16 Range 34.1136.9 77.9281.9 14 8.714.3 289.4562.0 7711157 310.9570 8161173 0.851.27 11.323.2 Mean SD 73.1 45.5 170 96 1.6 1.3 10.3 2.3 397 105.6 926 180 407.4 102.1 953 173 1.08 0.19 16.2 4.9

Median 183.9 250.6 1 15.9 942.9 1622 1047.9 1670.2 0.59 13.7

Range 40.1272.8 82.9399.4 0.54 11.021.8 443.81458.6 9172135 501.81498.4 1037.12297.3 0.430.96 6.922.4

156.5 92.4 243 132 1.5 1.4 15.3 4.7 990.2 372.4 1567 483 1088.6 400.1 1725 527 0.63 0.22 14.1 6.2

Values represent the median, range and mean SD of pharmacokinetic parameters determined in ve separate animals. *Signicantly (P < 0.05) different between the high- and the low-dose group as determined by a MannWhitney test. CL, clearance; Vdz, volume of distribution; AUC, area under the curve.

Metabolite analysis revealed that measurable quantities of 8-hydroxymirtazapine, the rst product produced by Phase 1, were detected (Fig. 2). Median AUC was 4.16 ng mL h with a range of 3.534.59 ng mL (HD) and 2.63 ng mL h with a range of 1.943.31 ng mL (LD). Thus, there was a trend toward higher values in the HD group, but this did not reach statistical signicance (P = 0.08). The amount of metabolite undergoing glucuronidation was also measurable, and the median AUC was 11.63 ng mL h with a range of 9.2411.72 ng mL (HD) and 14.32 ng mL h with a range of 12.8924.69 ng mL(LD), suggesting the same or slightly lower concentration in the HD group compared to the LD group (Fig. 3). This difference was not statistically different (P = 0.19).

Pharmacodynamic study Fourteen young adult healthy cats with normal physical examination, serum biochemistry prole, complete blood count

and urinalysis were entered in the study. Demographics for these cats are illustrated in Table 1. The effect of mirtazapine administration on food consumption is illustrated in Fig. 4. There was a statistically signicant difference (P = < 0.0001) between the amount of food eaten after administration of either LD or HD mirtazapine in comparison to placebo. Cats ate an average of 53.2% of their food when receiving the LD and 56.7% of their food when receiving the HD in comparison to eating 13% of their food when receiving the placebo. There was no signicant difference in the amount of food consumed between the two doses of mirtazapine. Nine of 14 cats refused food entirely when administered placebo, and 8 9 of these cats ate after mirtazapine was administered at either dose. One cat refused food entirely regardless of treatment administered. The effect of mirtazapine administration on interaction, activity level and vocalization scores is illustrated in Figs 57. There was a statistically signicant difference between administration of either LD or HD mirtazapine in comparison with
2010 Blackwell Publishing Ltd

Mirtazapine pharmacokinetics and pharmacodynamics in cats 393

Fig. 3. Glucuronidated metabolite concentrations. The glucuronidated metabolite is present in essentially the same quantities in the lowdose and high-dose groups (P = 0.19).

Fig. 1. Drug concentration curves for cats in the high-dose and low-dose groups. Note that the lines deviate, which represents a statistically signicant difference in half-life (P = 0.03) between doses.

Fig. 4. Effect of administration of low-dose mirtazapine, high-dose (HD) mirtazapine or placebo on food consumption. There is a statistically signicant difference in the amount of food ingested between placebo and either low-dose or HD mirtazapine (P < 0.0001). There is no signicant difference in the amount of food ingested between mirtazapine doses.

Fig. 2. 8-Hydroxymirtazapine metabolite concentrations. 8-Hydroxymirtazapine appears to be present in higher concentrations in the highdose group, although the difference was not statistically signicant (P = 0.08).

placebo for all three parameters (interaction P < 0.0001; activity level P < 0.0001; vocalization P = 0.0028). Bonferronis post hoc analysis determined that there was a statistically signicant
2010 Blackwell Publishing Ltd

difference (P < 0.05) between the LD and HD for interaction and vocalization scores, but not for activity level (95% condence interval: )5.7 to )0.15 for interaction and )5 to )0.7 for vocalization). Cats were also assessed for the presence of tremors or twitching, and no events were recorded during the 8-h observation period. At-home owner assessment of the effect of mirtazapine on appetite, activity and vocalization is illustrated in Table 3. Using mixed-effects logistic regression, it was determined that cats receiving mirtazapine still had a signicantly increased appetite approximately 12 h after drug administration when compared to cats that received placebo (P = 0.03); however, there was no signicant difference in appetite 24 h after drug administration (P = 0.18). No statistically signicant

394 J. M. Quimby et al.

difference was seen for vocalization at either time point (12 h: P = 0.05, 24 h: P = 0.3) or activity at either time point (12 h: P = 0.15, 24 h: P = 0.33). Estimation of AUC in pharmacodynamic study cats utilizing a limited sampling approach A limited sampling approach was utilized to estimate the AUC of mirtazapine in the cats for the pharmacodynamic studies (Panetta et al., 2003). The AUC024 estimated in the pharmacodynamic cats using this limited sampling model was 440.2 137.5 and 1073.3 414.4 ng mL h for the LD and HD, respectively. These values were not signicantly different from those calculated in the full-course pharmacokinetic study at these same doses and show essentially the same ratio in exposure between the HD and LD groups. No correlation between dose and effect was found when drug exposure, as represented by AUC, was compared to amount of food eaten (r = 0.53), vocalization (r = 0.50), activity (r = 0.64) and interaction (r = 0.53). Assessment of drug accumulation Calculation of drug accumulation from the pharmacokinetic data indicated that minimal accumulation (AF = 1.2) was expected with daily administration of the LD. Demographics of the four cats that received LD mirtazapine orally daily for 6 days are presented in Table 1. A median day 3 trough concentration of 1.4 ng mL (range: 0.223.36 ng mL) and a median 2 h postadministration drug concentration of 38.6 0.2 ng mL (range: 38.438.8 ng mL) was observed. On day 6, the median trough concentration was 1.53 ng mL (range: 03.27 ng mL), and the median 2 h postadministration drug concentration was 29.5 ng mL (range: 24.754.8 ng mL). These concentrations were similar to the range measured in the pharmacokinetic curve at the corresponding time point (range 25.280.2 ng mL; median 36.8 ng mL). No evidence of drug accumulation was detected. Subjectively, no cats experienced changes in activity, social interaction and vocalization during the 6-day time period (objective assessment was performed in the pharmacodynamic portion of the project).

Fig. 5. Effect of administration of low-dose (LD) mirtazapine, high-dose (HD)mirtazapine or placebo on interaction scores. There is a statistically signicant difference in interaction scores between placebo and either LD or HD mirtazapine (P < 0.0001). There is also a statistically signicant difference in interaction scores between administration of LD or HD (P < 0.05).

Fig. 6. Effect of administration of low-dose (LD) mirtazapine, high-dose (HD) mirtazapine or placebo on activity level scores. There is a statistically signicant difference in activity level scores between placebo and either LD or HD mirtazapine (P < 0.0001). There is no signicant difference in activity level scores between mirtazapine doses.

DISCUSSION The efcacy of mirtazapine as an appetite stimulant in cats is supported by the results of this study, which demonstrated a signicant increase in food consumption after administration of the drug. Eighty-eight percent of healthy cats that refused to eat their normal diet while in the hospital subsequently ate when mirtazapine was administered. No signicant difference in the amount of food consumed was seen between the two doses; however, a signicant increase in behavior such as vocalization and interaction was noted in the HD group. Limited sampling during the pharmacodynamic study allowed an estimation of drug exposure to be calculated. No correlation between drug exposure and clinical effect of food consumption could be
2010 Blackwell Publishing Ltd

Fig. 7. Effect of administration of low-dose (LD) mirtazapine, high-dose (HD) mirtazapine or placebo on vocalization scores. There is a statistically signicant difference in vocalization scores between placebo and either LD or HD mirtazapine (P < 0.0001). There is also a statistically signicant difference in vocalization scores between administration of LD or HD (P < 0.05).

Mirtazapine pharmacokinetics and pharmacodynamics in cats 395 Table 3. At-home assessment of appetite, activity and vocalization after administration of mirtazapine. For each treatment type, the number of cats out of 14 that experienced and increased vs. decreased normal appetite, vocalization or activity is noted Appetite Placebo 1014 h after administration 2024 h after administration Increased Normal or decreased Increased Normal or decreased 1 14 13 14 0 14 14 14 HD 6 14 8 14 2 14 12 14 LD 9 14 5 14 4 14 10 14 Placebo 1 14 13 14 0 14 14 14 Vocalization HD 7 14 7 14 3 14 11 14 LD 5 14 9 14 2 14 12 14 Placebo 1 14 13 14 1 14 13 14 Activity HD 6 14 8 14 3 14 11 14 LD 4 14 10 14 4 14 10 14

HD, high dose; LD, low dose.

appreciated, which is consistent with human data (Timmer et al., 2000). The implication of these ndings is that similar efcacy with less potential for side effects is experienced with the lower dose, and this may be the most appropriate initial dose for many cats. The results of this study also suggest that the dosing interval in cats is signicantly different from that previously postulated by Cahill: 1 8 to 1 4 of a 15 mg tablet orally every 3 days (Cahill, 2006). The half-life in humans ranges from 20 to 40 h, and the drug is administered daily without substantial accumulation (Timmer et al., 2000). In contrast, the half-life of mirtazapine in cats at the lowest effective dose (1.88 mg) is approximately 9 h. Based on this half-life, the drug could theoretically be administered daily in normal cats without substantial drug accumulation. This is supported by the almost negligible trough serum concentrations that were measured during the daily dosing study. Although only a small number of cats were used and therefore the data were only subjectively described, 24-h trough concentrations were close to the limits of detection and no accumulation was seen after daily dosing for 6 days. The authors recognize that serum levels may not necessarily relate to drug effect in cats and that individual variation between cats could potentially result in greater drug accumulation than that measured in this study. In an effort to evaluate the duration of effect of the drug, assessments were carried out in the home environment during the pharmacodynamic study. Observations recorded by owners in a blinded fashion indicated that the majority of cats did experience clinical effects of the drug the night after the medication was administered, and a statistically signicant difference in appetite and behavior was noted for both HD and LD when compared to placebo. Although a few cats still experienced effects 24-h postadministration, the majority of cats had no apparent clinical effect at that time point and no statistically signicant difference in appetite or behavior was noted for either dose in comparison with placebo. These results conceivably may have been affected by unknown bias on the part of the owners or unknown factors affecting the cats in the 24 h after their participation in the pharmacodynamic study. Dose proportionality is exhibited with mirtazapine in human pharmacokinetic studies (Timmer et al., 2000), and regardless of the dose administered, the elimination half-life remains the same. This does not appear to be true in cats, based on the data collected in this study, as there was a signicant difference
2010 Blackwell Publishing Ltd

between the elimination half-lives of the HD and LD groups. The pharmacokinetic study was not a crossover study in design, and this may have potentially affected the data and interpretation. However, a signicant difference in drug exposure and CL does appear to exist between the dose groups, while Vd remains constant. There does not appear to be a signicant difference in absorption between the two groups, so this is unlikely to play a role. Obesity is one factor that could affect CL, while Vd remains the same (Dunn et al., 1991). It was noted that one cat in the HD group was obese; however, the other cats involved in the study were not considered to be. This could potentially have affected the study results. Alternatively, the results of the glucuronide metabolite data may provide some explanation for the apparent lack of linear pharmacokinetics in this species as there is some evidence of a trend indicating glucuronide saturation. The accumulation of higher concentrations of the glucuronide metabolite might have been predicted in the HD group, as there were slightly higher concentrations of 8-hydroxymirtazapine detected in that group. However, there was essentially the same amount, if not less, of glucuronide metabolite present in the HD group. Possible explanations for this result include saturation of the glucuronide enzyme system, which is known to be decient in cats (Lainesse et al., 2006), or inhibition of metabolism by the 8-hydroxymirtazapine metabolite. This would be concerning not only because of a potential for more drug side effects but also because depletion of glucuronidation ability could affect metabolism and serum concentrations of other drugs and supplements. These ndings suggest that it may be more appropriate to administer the 1.88 mg dose of mirtazapine to avoid these potential effects on drug metabolism. In conclusion, administration of mirtazapine to healthy young cats resulted in a signicant increase in appetite. No difference in the amount of food ingested was seen between the HD and LD, but signicantly more behavior changes were seen with the HD. A signicant difference in half-life was also seen between the HD and LD, and thus, this drug does not appear to display linear pharmacokinetics in the cat. The half-life of mirtazapine administered at the lowest effective dose (1.88 mg) appears to be shorter than that previously postulated and is potentially consistent with daily dosing in young healthy cats. Data collected on clinical effect after a single dose and serum concentration after repeated dosing support this conclusion. As it is known that age, renal and hepatic impairment affect the half-life of mirtazapine in humans, further studies are needed to

396 J. M. Quimby et al.

clarify the effects of these factors on the half-life of mirtazapine in cats. Clinical effects could potentially be different in elderly or ill animals, and daily dosing may not be appropriate in these patients.

APPENDIX 1: BEHAVIORAL ASSESSMENT Interaction a) Hiding sits at back of cage b) Gets up to greet you when cage is opened c) Soliciting attention at front of cage Vocalization (meows per 15 s) a) 010 b) 1015 c) 15+ Activity a) Sedentary hiding b) Walking around normally c) Frantic hyperactivity Tremors a) None b) Mild twitch bobble (subtle) c) Noticeable head bobble or twitching

ACKNOWLEDGMENTS This study was funded by a grant from the Winn Feline Foundation. The study sponsors had no involvement in the planning or execution of the study. Paul S. Morley graciously provided statistical assistance.

REFERENCES
de Boer, T. (1996) The pharmacologic prole of mirtazapine. Journal of Clinical Psychiatry, 57 (Suppl. 4), 1925. Cahill, C. (2006) Mirtazapine as an antiemetic. Veterinary Forum, 23, 3436. Center, S.A. (2005) Feline hepatic lipidosis. Veterinary Clinics of North America. Small Animal Practice, 35, 225269. Doherty, B., Rodriguez, V., Leslie, J.C., McClean, S. & Smyth, W.F. (2007) An electrospray ionisation tandem mass spectrometric investigation of selected psychoactive pharmaceuticals and its application in drug and metabolite proling by liquid chromatography electrospray ionisation tandem mass spectrometry. Rapid Communications in Mass Spectrometry, 21, 20312038. Dunn, T.E., Ludwig, E.A., Slaughter, R.L., Camara, D.S. & Jusko, W.J. (1991) Pharmacokinetics and pharmacodynamics of methylprednisolone in obesity. Clinical Pharmacology and Therapeutics, 49, 536549. Elliott, J., Rawlings, J.M., Markwell, P.J. & Barber, P.J. (2000) Survival of cats with naturally occurring chronic renal failure: effect of dietary management. Journal of Small Animal Practice, 41, 235242. Kopple, J.D. (1994) Effect of nutrition on morbidity and mortality in maintenance dialysis patients. American Journal of Kidney Diseases, 24, 10021009. Lainesse, C., Frank, D., Meucci, V., Intorre, L., Soldani, G. & Doucet, M. (2006) Pharmacokinetics of clomipramine and desmethylclomipramine after single dose intravenous and oral administrations in cats. Journal of Veterinary Pharmacology and Therapeutics, 29, 271278. Littell, R. (2006) SAS for Mixed Models, 2nd edn, SAS Institute Inc., Cary, NC. Panetta, J.C., Iacono, L.C., Adamson, P.C. & Stewart, C.F. (2003) The importance of pharmacokinetic limited sampling models for childhood cancer drug development. Clinical Cancer Research, 9, 50685077. Plantinga, E.A., Everts, H., Kastelein, A.M. & Beynen, A.C. (2005) Retrospective study of the survival of cats with acquired chronic renal insufciency offered different commercial diets. Veterinary Record, 157, 185187. Ross, S.J., Osborne, C.A., Kirk, C.A., Lowry, S.R., Koehler, L.A. & Polzin, D.J. (2006) Clinical evaluation of dietary modication for treatment of spontaneous chronic kidney disease in cats. Journal of the American Veterinary Medical Association, 229, 949957. Sanderson, B.J. (2000) Management of anorexia. In Current Veterinary Therapy XIII. Ed. Bonagura, J., pp. 6974. WB Saunders Co, Philadelphia. Timmer, C.J., Sitsen, J.M. & Delbressine, L.P. (2000) Clinical pharmacokinetics of mirtazapine. Clinical Pharmacokinetics, 38, 461474. Wagner, J. (1993) Pharmacokinetics for the Pharmaceutical Scientist. Technomic Publishing Company Inc., Lancaster, PA.

APPENDIX 2: AT-HOME ASSESSMENT Please observe your cat tonight and tomorrow morning, ll out this sheet and return to Dr. Quimby. Cat: __________________________________ Date: ____________________________ The night you bring your cat home Was your cats appetite: Normal Increased Decreased Was your cats vocalization: Normal Increased Decreased Was your cats activity level: Normal Increased Decreased The morning after Was your cats appetite: Normal Increased Decreased Was your cats vocalization: Normal Increased Decreased Was your cats activity level: Normal Increased Decreased Additional comments:

2010 Blackwell Publishing Ltd