BIOPHARMACEUTICS & DRUG DISPOSITION, VOL. 13, 255-262 (1992)

PHARMACOKINETICS OF KETOTIFEN AFTER ORAL ADMINISTRATION TO HEALTHY

MALE SUBJECTS

A. GRAHNEN, A. LONNEBO, o BECK,* s-K ECKERNAS, B. DAHLSTROM AND B. LINDSTROM

Pharmaco Medical Consultants ( P M C ) A B, Uppsala and *Department of Clinical Pharmacology, Karolinska Institute, Stockholm, Sweden

ABSTRACT

The pharmacokinetics of 2 mg ketotifen from four different oral dosage forms were examined in two randomized, balanced cross-over studies. Forty healthy male subjects participated. Each of 20 subjects received two capsule formulations and each of the other 20 subjects received two syrup formulations. Ketotifen concentrations in plasma were determined by amodified GC-MS method. The limit of quantitation was40 pgml-I. Inter-day precision and accuracy calculated from quality control samples were 16.3 per cent (-1.9 per cent), 19.8 per cent (+4.5 per cent) and 23.6 per cent ( + 5 9 per cent) at plasma concentration levels of 86 (n = 18), 215 (n = 19) and 343 (n = 18) pg ml-I, respectively. Ketotifen was rapidly absorbed from all dosage forms reaching C,,, in the order of 400 pg ml-l after the syrup formulations and 300 pg ml-l after the capsule formulations within 2 to 4 h. The syrup formulations showed a significantly more rapid rate of absorption as assessed by T,,,. No significant differences in extent of absorption between dosage forms were observed. The terminal elimination half-life of ketotifen varied between subjects from 7 to 27 hours with a mean of about 12 h. The minor pharmacokinetic difference between dosage forms observed in this study is unlikely to be of clinical significance.

KEY WORDS Ketotifen Pharmacokinetics GC-MS

INTRODUCTION

Ketotifen is a cycloheptathiophenone derivative, shown to have anti-allergic and anti-anaphylactic properties.' The effects of ketotifen are similar to those of disodium cromoglycate. A number of controlled clinical studies have demon- strated ketotifen to be an effective and safe treatment for asthma.*"

Currently available data on the pharmacokinetic properties of ketotifen are extremely sparse in the literature. A recent review article claims that the bioavai-

Correspondence to: Dr Anders Grahnin, PMC Drug Study Unit AB, Glunten, S-751 83 Uppsala, Sweden.

0 142-2782/92/04025 5-08$05.00 0 1992 by John Wiley & Sons, Ltd.

Received 28 February 1991 Accepted 29 October 1991

256 A . GRAHNEN ET A L .

lability of ketotifen is about 50 per cent following oral administration. The disposition is stated to be biphasic with a terminal elimination half-life of 22 h.l Julien-Larose er al. have described a GC-MS method for the determination of ketotifen and its primary metabolites in clinical plasma sample^.^ Ketotifen levels after a 2 mg single oral dose were low in the order of 500-600 pg ml-I. In plasma, most of the drug material was found as ketotifen-glucuronide (inac- tive), norketotifen (active), and 10-OH ketotifen. In plasma, the glucuronide metabolites were present in amounts 5 to 20 times higher than ketotifen. The active norketotifen metabolite was present in lower amounts as compared to ketotifen. Utilizing 3H-ketotifen, Kennedy demonstrated a terminal elimination half-life for total radioactivity of about 22 h.* In a multiple dose study, Schmidt- Redemann et al. found ketotifen levels in plasma to be about 2 to 3 ng ml-* following a 1 mg b.i.d. dosage regime.y These data were obtained from a radioimmunoassay (RIA) where assay specifications (sensitivity and selectivity) were not reported. The GC-MS method as described by Julien-Larose7 was, in our hands, neither easily reproducible for clinical plasma samples, nor sensi- tive enough for a detailed pharmacokinetic analysis of ketotifen.

The aim of this study was to further develop the previously published GC-MS method for ketotifen and to determine some basic pharmacokinetic parameters of ketotifen after single oral dose administration of different dosage forms available in Europe. Ketotifen, developed by Sandoz, is marketed in a number of European Community countries as Zaditen@ through Sandoz-Wander. Generic formulations of ketotifen are marketed by Hexal@-Pharma in Germany. For the treatment of asthma, capsule formulations are available for adults whereas syrup formulations are available for treatment of childhood asthma.

METHODS

Subjects and study design

The study, approved by the Ethics Committee of the University of Uppsala and by the Department of Drugs of the Swedish National Board of Health and Welfare, was conducted in accordance with good clinical practice (GCP) standards and the Declaration of Helsinki. Forty normal, healthy, male subjects ranging in age from 21 to 34 years with a mean of 25 k 3 years and ranging in weight from 59 to 85 kg were selected. They were all within * l o per cent of their ideal body weight for their height and body frame according to the Metropolitan Insurance Company Bulletin, 1983.

The subjects took part in two separate experiments. Twenty subjects received two different capsule formulations of ketotifen (Zaditen@ capsules-Wander, Germany and Ketotifen capsules-HexaP-Pharma, Germany). The other half of the group (20) received two different syrup formulations of ketotifen (Ketoti- fen syrupHexal@-Pharma, Germany, and Zaditen@ s y r u p w a n d e r , Ger- many). Each subject received 2 mg ketotifen.

KETOTIFEN KINETICS 257

Subjects were required to refrain from all drugs from 7 days prior to drug administration and during the whole study. Alcohol was not allowed during the study. A minimum of a 1-day washout period was observed between doses. After an 8-10 h fast, a 2 mg single oral dose of ketotifen was administered to each subject in a two-treatment, two-period, randomized, balanced cross- over design. No food was allowed until 4 h after drug administration.

Blood samples (10 ml) were collected in evacuated glass heparinized tubes prior to and at 0.5, 1, 1.5, 2, 2.5, 3, 4, 5, 6, 8, 10, 12, 24, and 36 h after drug administration. The samples were left for 30 min and were then centrifuged at 3000 rev min-l for 10 min. The separated plasma was then transferred to a test tube and placed in a freezer (-20") until analysis of ketotifen. The samples were analysed within 12 months. Quality control samples (stored together with clinical samples) showed stability during the time span of the analysis.

Ketotifen assay

Ketotifen concentrations in plasma samples were determined by a GC-MS method. The assay method adopted was a modification of the method published by Julien-Larose.' Ketotifenhydrogenfumarate was obtained from Hexal@'- Pharma & Co., KG, GFR. Pizotifenmaleate (internal standard: is.) was obtained from The National Corporation of Swedish Pharmacies. All other chemicals were of analytical or reagent grade.

A Finnegan model Incos 50 mass spectrometer equipped with a HP 5890 gas chromatograph fitted with a moving needle injector was used for plasma determinations. A 25 m HP Ultra 2 fused silica column (film thickness 0-53 pm) was used and operated with helium as carrier gas. The column was directly inserted into the ion source. The pressure at the injector inlet was 14 psi. The temperature in the injector and transfer line was 250". A temperature program starting at 200" (0.1 min) to 300" at 20" min-' was employed. The mass spectr- ometer was operated in the electron impact mode with an ion energy of 70eV and a source temperature of 170". The molecular ions monitored were m/z 295 (ketotifen) and m/z 295 (pizotifen, is.). The peak area ratio was used to construct the calibration graph.

To 1.0 ml plasma was added 50 pl i.s. solution (41 ng ml-I) and 0.5 ml of NaOH. The mixture was extracted with 3 ml of toluene and whirl-mixed for 1 min. After centrifugation for 5 min at 1000 g the organic phase was transferred to a conical tube and evaporated under nitrogen gas at 40". The residue was dissolved in 20 p1 of ethanol and a 2 pl aliquot was transferred into the injector.

Calibration graphs were constructed by spiking plasma samples with ketoti- fen in the range 80 to 804 pg ml-l and analysing the samples according to the method. Quality control samples were prepared in the concentrations 85.8, 214.5, and 342.2 pg ml-l plasma. One sample of each concentration was ana- lysed together with each batch of unknown (analytical) samples and calibration

258 A. GRAHNEN ET AL.

graph. The quality control samples were stored together with the analytical samples at -20".

Data analysis

Areas under the ketotifen plasma concentration-time curves were calculated for each subject and treatment. The relative bioavailability of the different dosage forms of ketotifen was evaluated from the area under the curve (AUC), the time for maximal plasma concentration (T,,,) and the maximum plasma level (C,,,,,). AUCo-, was calculated by the trapezoidal rule. AUCO-, was calcu- lated as AUCo-, + ClaSt/A. The elimination rate constant ( A ) was estimated by regression analysis of the terminal part of the plasma concentration-time curve. Due to the extremely low levels of ketotifen in plasma after 4 h it was not possible to accurately assess AUCo-, for all subjects.

Analysis of variance (ANOVA) was used to assess differences in the pharma- cokinetic parameters associated with treatment by each formulation. An unpaired student's t-test was used to test for differences between capsule and syrup treatments. Wilcoxon's signed rank test was used as a non-parametric test for differences in the other parameter not normally distributed ( Tmax). ANOVA was performed using the computer program BMDP (UCLA). Sources of variation were: mean, sequence, treatment, period and error. Confidence intervals (90 per cent) according to the two one-sided test procedurelo were used to test for equivalence within dosage forms (capsule vs capsule, syrup vs syrup).

RESULTS

Ketotifen assays

The calibration graphs were linear in the concentration range used with all coefficients of correlation > 0.9762 (mean 0.9899). The limit of quantitation of the method allowed the determination of ketotifen down to 40 pg ml-I. Inter-day precision (expressed as CV) and accuracy (within brackets) calculated from the quality control samples were 16.3 per cent (- 1.9 per cent), 19.8 per cent (+4.5 per cent) and 23-6 per cent (+5.9 per cent) at the plasma concen- trations levels of 85.8 (n = 18), 214.5 (n = 19) and 343.2 (n = 18) pgml-I, respecti- vely. Figure 1 shows examples of ion chromatograms for internal standard, plasma background, and a representative plasma sample.

Pharmacokinetic parameters

The mean ketotifen plasma levels and pharmacokinetic data for each dosage form are summarized in Figure 2 and Table 1. As indicated in Table 1 , AUCo-,was not estimated for all subjects. This was due to the extremely low

KETOTIFEN KINETICS 259

mlz 295 PZ

a

i I !

Figure 1. Monitoring of representative ions for the determination of ketotifen in plasma: (a) m/z 295 ion chromatogram of pizotifen (PZ) used as internal standard; (b) plasma background when monitoring the m/z 309 ion representative for ketotifen (KT); (c) m/z 309 ion chromatogram from a plasma sample containing 280 pg ml-’ of ketotifen

levels of ketotifen in plasma observed 4 h (and onwards) after the observed peak concentration. When the estimated residual area contributed to more than 20 per cent of the total area, the estimates were excluded from mean calculations. Maximum plasma concentrations (CmaX) were reached after about 3 . 5 4 h following the capsule formulations and after about 2-3 h after the syrup formulations with mean levels of about 30MOO pg ml-I. The syrup formulations showed higher C,,, (p < 0-005) values compared to the capsule formulations, although intersubject variability was substantial. The disposition of ketotifen showed a tendency to be biphasic after absorption although it was not possible in individual subjects to clearly discriminate between a mono- and biphasic profile. No significant differences in AUC and C,,, were found when comparing the two capsule formulations. Confidence intervals (90 per

260

Pg/ml

400

350

300

250

200

150

100

50

0

A. GRAHNEN ET AL.

I

0 5 10 15 20 25 30 35 40 Hours

Figure 2. Mean f SEM (n = 20) plasma ketotifen concentration-time profiles following single oral dose of administration 2 mg ketotifen (A Ketotifen syrup, A Zaditen" syrup, 0 Ketotifen capsules,

Zaditen@ capsules)

Table 1. Ketotifen pharmacokinetic parameters

Parameter Capsules Syrup Ketotifen Zaditen" Ketotifen Zaditen"

Cm,x(Pg m1- I) 281 f 8 4 270 * 66 421 f 206 355 f 147 T,,X(h) 3.9 f 2.2 3.6 f 1.6 2.9 f 1.9 2.3 f 1.2 AUC,-,(pg ml-'*h) 3008 f 1556 3390 f 1260 3984 f 2359 3387 f 2103

AUCo-,(pgml-'*h) 4703 f2057* 4941 f 1813* 4970f2607 3815f 1513

t1/2(h) 13.1 f6.0* 18.3f6.7* 12.2f4.5 12.7f5.6

Mean f SD (n = 20). * n = 11-15.

KETOTIFEN KINETICS 261

cent) were 93-1 15 per cent and 74104 per cent for C,,, and AUC, respectively. When comparing the syrup formulations, small (20 per cent) but statistically significant differences were observed in favour of the Hexal@-Pharma formula- tion. Confidence intervals (90 per cent) for the syrup formulations were 1041 34 per cent and 102-133 per cent for C,,, and AUC, respectively. The elimination half-life could be accurately estimated in approximately half of the subjects during capsule treatment and in all subjects during syrup treatment. Terminal half-life varied between 7 and 27 h for these subjects with a mean of about 12 h.

DISCUSSION

The analytical GC-MS method as previously described by Julien-Larose’ was modified in this study and the limit of quantitation was found to be 40 pg ml- or lower for all processed analytical batches. Acceptable reproducibility and accuracy (in light of the low levels) were confirmed from the analysis of quality control samples. The levels of unchanged ketotifen in plasma deter- mined from this study do not agree with previously published results. The levels reported from the study by Schmidt-Redemann utilizing RIA9 were about a magnitude of order higher. Using the data obtained from this single dose study, multiple dose steady state levels (Cmax/Cmin) would be expected to be (on average) in the order of 330/190 pg rn-l for a daily dose of 1 mg bid assuming linear kinetics. The levels reported from earlier GC-MS analysis were of the same order although our results show somewhat lower levels after a 2 mg oral dose. The reason for the discrepancy with the study by Schmidt- Redemann is probably due to poor selectivity of the reported RIA method in which metabolites of ketotifen may have been co-determined. The low levels of unchanged ketotifen in plasma found in this study and high amounts of metabolites found in earlier studies suggest that ketotifen is subject to substan- tial first-pass metabolism rather than to incomplete absorption. The lack of an intravenous reference precludes any definite conclusion regarding the absol- ute bioavailability of ketotifen.

Ketotifen seems to be fairly rapidly absorbed with maximum concentrations appearing after 3 to 4 h following capsule and somewhat faster (2 h) following syrup administration. The disposition of ketotifen showed a tendency to be biphasic although it was not possible in individual subjects to discriminate between a monophasic and biphasic profile. The terminal elimination half-life was estimated to be on average 12 h with a substantial individual variability (7 to 27 h). To our knowledge, no published data have appeared in which the elimination half-life for ketotifen has been accurately estimated. Our results are somewhat different from the results published in. a recent review claiming a terminal half-life of more than 20 h.6J The discrepancy in reported pharmacok- inetic parameters between different studies is most probably due to the inherent

262 A. GRAHNEN ET A L .

problems of accurate bioanalysis of low levels of ketotifen in plasma. The minor pharmacokinetic differences between dosage forms observed in this study has probably no clinical significance.6

REFERENCES

1. U. Martin, D. Romer, The pharmacological properties of a new, orally active antianaphylactic compound: ketotifen, a benzocycloheptathiophene. Monographs in Allergy, 28,770-782 (1978).

2. L. Craps, The prophylaxis of bronchial asthma with ketotifen. Five years of clinical investi- gation. Acta Therupeuliru, 6, 1-16 (1980).

3. D. G. Tinkelman, B. A. Moss, S. C. Bukantz, A. L. Sheffer, J. H. Dobken et al., A multicenter trial of the prophylactic effect of ketotifen, theophylline and placebo in atopic asthma. J . Allerg. Clin. Immunol., 16,487497 (1985).

4. R. D. Monie, A. P. Smith, G. Anderson, B. H. Danies, G. 0. Thomas, A double-blind clinical trial of ketotifen and disodium chromoglycate in bronchial system. Br. J . Dis. Chest.,

5. A. R. Tauer and J. Elmes, A controlled trial of ketotifen in exercise - induced asthma. Br. J . Dis. Chesl, 74,398402 (1980).

6. S. Grant, K. L. Goa, A. Fitton and E. M. Sorkin, Ketotifen. A review of its pharmacodynamic and pharmacokinetic properties, and therapeutic use in asthma and allergic disorders. Drugs, 40(3),41248(1990).

7. C . Julien-Larose, M. Guerett, D. Lavene and J. R. Kiechel, Quantification of ketotifen and its metabolites in human plasma by gas chromatography mass spectrometry. Biomed. Mass Spectrometry, 10, 136142 (1983).

8. G. R . Kennedy, Metabolism and pharmacokinetics of ketotifen in children. Research and Clinical Forums, 4, 17-20 (1982).

9. B. Schmidt-Redemann, P. Brenneisen, W. Schmidt-Redemann and S. Gonda, The determi- nation of pharmacokinetic parameters of ketotifen in steady state in young children. Inr. J . Clin. Pharmacol. Ther. Toxicol., 24,49&498 (1986).

10. D. Schuirmann, A comparison of the two one-sided test procedure and the power approach

76,383-389 (1980).

for assessing the equivalence of average bioavailability. J . Pharmarokinet. Biopharm., 15,657- 680(1987).