Effects of controlled-release on the pharmacokinetics and absorption characteristics of a compound undergoing intestinal efflux in humans

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 231–239

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/ejps

Effects of controlled-release on the pharmacokinetics and absorption characteristics of a compound undergoing intestinal efflux in humans Marija Tubic a , Daniel Wagner a , Hildegard Spahn-Langguth a , Christine Weiler b , b ¨ Roland Wanitschke b , Wulf Otto Bocher , Peter Langguth a,∗ a

Biopharmacy und Pharmaceutical Technology, Institute for Pharmacy, Johannes Gutenberg-University, Staudinger Weg 5, 55099 Mainz, Germany b First Department of Internal Medicine, Johannes Gutenberg-University, Langenbeckstrasse 1, 55131 Mainz, Germany

a r t i c l e

i n f o

a b s t r a c t

Article history:

Objective: The number of active pharmaceutical ingredients (API) undergoing inhibitable and

Received 12 March 2006

saturable intestinal efflux is considerable. As a consequence, absorption and bioavailability

Accepted 2 April 2006

may depend on the intestinal concentration profile of the drug and may vary as a function

Published on line 26 April 2006

of dose and release rate of the drug from the dosage form. The impact of controlled versus immediate-release on the absorption of P-glycoprotein substrates is currently unknown.

Keywords:

Thus, the main focus of the present study was a comparison of the pharmacokinetics of

P-glycoprotein

the P-gp model substrate talinolol following administration of immediate-release (IR) and

Talinolol

controlled-release (CR) tablets to healthy human volunteers with a particular focus on the

Sustained release

absorption characteristics of the active pharmaceutical ingredients.

Controlled-release

Methods: Talinolol immediate-release (Cordanum® , 100 mg), one controlled-release (100 mg)

Pharmacokinetics

and two controlled-release tablets (200 mg) were administered as single doses to fasting

Absorption

healthy volunteers in a crossover design with a 1 week washout period between treatments. Sufficient blood and urine samples were drawn and analysed using a specific HPLC method with UV detection to describe the resulting plasma and urinary excretion versus time profiles. Results: The bioavailability of talinolol in term of AUC0→∞ for IR talinolol was approximately twice as high as compared to the administration of the same dose in a controlled-release dosage form. After administration of talinolol IR tablets, the drug was rapidly absorbed and reached maximum concentrations Cmax of 204.5 ng/ml ± 121.8 (means ± S.D.) 2 h after dosing. The terminal half-life of the drug averaged 19.8 h following IR administration in comparison to 32 h under CR dosing conditions. Following administration of the IR dosage form, significant secondary peaks were observed in one healthy subject. Secondary peaks were not clearly apparent in the CR plasma profiles. Conclusion: The present study demonstrates a considerable loss of bioavailability of drugs that are substrates of intestinal secretory transporters upon their administration in controlled-release dosage forms. © 2006 Elsevier B.V. All rights reserved.

∗

Corresponding author. Tel.: +49 6131 3925746; fax: +49 6131 3925021. E-mail address: [email protected] (P. Langguth). 0928-0987/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.ejps.2006.04.005

232

1.

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 231–239

Introduction

The number of active pharmaceutical ingredients (API) undergoing intestinal efflux is considerable, e.g. doxorubicin, daunorubicin, vincristine, vinblastine, etoposide, paclitaxel, cyclosporine A, salbutamol, talinolol, celiprolol, ranitidine. Several transporters may be involved in the intestinal secretion, e.g. organic cation transporters, organic anion transporters, MRP related transporters and P-glycoprotein (P-gp). Due to its broad substrate specificity and wide distribution in the intestine as well as in other organs, P-glycoprotein has a predominant role among the different transporters. The intestinal carrier-mediated secretion process is inhibitable by drugs, food constituents and some excipients and saturable at higher substrate doses. Consequently, absorption and bioavailability may depend on the intestinal concentration profile of the drug and may vary as a function of its dose and release rate from the dosage form. The impact of controlled versus immediate-release on the pharmacokinetics of P-glycoprotein substrates is currently unknown. In order to shed some light on the impact of the release rate of API on the pharmacokinetics of a P-gp substrate, the adrenoreceptor antagonist talinolol was chosen. It is commercially available as immediate-release capsules and tablets for oral administration and as injectable solution for intravenous infusion. It is used in oral doses of 50–300 mg daily for the treatment of cardiovascular diseases such as arterial hypertension, coronary heart disease and tachyarrhythmia (Schmitt, 1995). P-glycoprotein, an ATP driven efflux pump, is physiologically expressed in the apical membrane of mucosal cells of the small and large intestine as well as at the luminal membrane of proximal tubular cells in the kidney, the billiary canalicular membrane of hepatocytes, at the blood–brain barrier, in capillary endothelial cells of testis, the adrenal gland and the endometrium of pregnant uterus (Matheny et al., 2001). Talinolol as a suitable model compound for P-gp mediated transport was proposed by Wetterich at al., because of the following advantageous kinetic properties: broad therapeutic range, low plasma protein binding, minor enterohepatic recirculation and a very low metabolic clearance (Wetterich et al., 1996; Spahn-Langguth et al., 1998). About 99% of the amount systemically available is eliminated unchanged and less than 1% as hydroxylated metabolites. This ensures that intestinal drug efflux can be observed without an interference with biotransformation processes (Gramatte et al., 1996; Oertel et al., 1994). In the intestine P-glycoprotein can reduce bioavailability of drugs administrated orally, thus bioavailability of talinolol from immediate-release dosage forms reaches only 50–60% and is dose-dependent due to incomplete and nonlinear intestinal absorption (Wetterich et al., 1996). The differences in P-glycoprotein distribution within the gastrointestinal tract may cause site-dependence of talinolol absorption. Due to increasing P-glycoprotein expression levels talinolol is rapidly absorbed only from the upper parts of gastrointestinal tract (Gramatte et al., 1996), where lower P-glycoprotein expression was found when compared with jejunum, ileum and colon (Mouly and Paine, 2003; Brady et al., 2002).

Previous studies in healthy volunteers and patients have also described a large intra and intersubject variability in absorption after administration of immediate-release dosage forms of talinolol (Wetterich et al., 1996; Siegmund et al., 2003; Bogman et al., 2005). General, sustained release dosage forms are designed to reduce fluctuation and insure a more uniform plasma drug profile. The objective of this study was to evaluate the influence of a decrease in the release rate of talinolol from the dosage form on the intestinal P-glycoprotein mediated efflux and biopharmaceutic parameters in humans.

2.

Materials and methods

2.1.

Materials

All chemicals used for the preparation of the clinical batches of talinolol controlled-release tablets were of pharmaceutical grade. Talinolol active pharmaceutical ingredient (batch number 98010182/1) and film-coated immediate-release tablets containing 100 mg talinolol per tablet (Cordanum® 100, batch number 1H028A) were a gift from AWD Pharma (Dresden, Germany). Mannitol was supplied from Caelo (Hilden, Germany) ¨ and Eudragit® RL 100-55 was received from Rohm (Darmstadt, Germany). Carvedilol reference standard was kindly provided by Prof. H. Spahn-Langguth, Mainz. Solvents for HPLC analysis were purchased from Merck (Darmstadt, Germany) and Fluka Chemie (Buchs, Germany). The HPLC instrument consisted of an autosampler type AS 950 (Jasco Deutschland GmbH, GroßUmstadt, Germany), a pump type PU 980 (Jasco, Deutschland GmbH, Groß-Umstadt, Germany), a UV–vis detector model UV 975 (Jasco, Deutschland GmbH, Groß-Umstadt, Germany), a column oven type Jetstream Plus. Chromatograms were evaluated using BorwinTM software, version 3 (JMBS Development, Le Fontanil, France).

2.2.

Preparation of talinolol matrix tablets

Talinolol controlled-release matrix tablets were manufactured in compliance with Good Manufacturing Practice standards and facilities. Tablets with a content of 100 mg talinolol were directly compressed using standard 13 mm punch and die on a PW 20 GS manual tablet press (Paul Weber, Remshalden, Germany). The matrix tablet formulation was comprised of 40% talinolol, 40% Eudragit® L 100-55 as release-sustaining polymer and 20% mannitol as hydrophilic pore former. The tablets were characterized in terms of thickness, uniformity of mass, crushing strength and friability.

2.3. Dissolution of talinolol from immediate and controlled-release tablets Talinolol release from immediate and controlled-release tablets was carried out using a Pharma Test PTWS III (USP Apparatus II) dissolution tester. All tests were conducted in 1000 ml of dissolution media (0.1N hydrochloric acid pH 1.0; acetate buffer pH 4.5 or phosphate buffer pH 6.8) maintained at of 37 ± 0.5 ◦ C at a stirring speed of 50 rpm. Talinolol concentra-

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 231–239

233

tions in dissolution media were determined by UV spectrophotometry using a Lambda 20 UV–vis Spectrometer at 240 nm.

Heart rate, peripheral blood pressure and electrocardiograms were monitored during the study.

2.4.

2.6.

Subjects for clinical study

Seven healthy volunteers, aged between 22 and 44 years with a mean weight of 79.8 ± 11.3 kg were enrolled in the study. Five males and one female completed the study; one male volunteer completed only two out of the three arms of the study. The study was performed according to the declaration of Helsinki in its revised form from 1996. The study protocol was reviewed and approved by the Ethics committee of the ¨ Landesarztekammer Rheinland-Pfalz and written informed consent was obtained from each volunteer prior to the study. Exclusion criteria were heart rate at rest less than 55 beats per min, Q wave in ECG during rest higher than 220 ms, systolic blood pressure less than 120 mmHg, orthostatic dysregulation, blackouts, syncope, presence of significant diseases or recent surgery, history of asthma, clinical history of hypersensitivity to the excipients of the study medication, pregnancy or lactation, participation in another clinical trial within the last 3 months and dependence on drugs (except for caffeine or nicotine). No medication was taken during 15 days before and during the entire study. Participants underwent a medical examination and laboratory screening tests to confirm good general health.

2.5.

Study design

Volunteers reported to the study on the morning of the study after overnight (>10 h) fasting. Subjects received a standard meal 4 h following drug dosing, an evening meal at 7 p.m. and 5 p.m. snack. Each subject consumed the same meal on each of the study days. No other food was consumed during the treatment periods. Each formulation was administrated with 200 ml water. Treatment periods were separated by a washout period of at least 7 days. The dosage scheme is presented in the following scheme: Period I

Period II

Period III

A

B

C

C

A

B

B

C

A

A, one talinolol immediate-release tablet (Cordanum® ) with a drug content of 100 mg; B, one talinolol controlled-release tablet with a drug content of 100 mg; C, two talinolol controlled-release tablets with a drug content of 100 mg each.

Blank plasma and urine samples were obtained weekly prior to dosing. Blood samples of 7 ml each were collected at 1–5, 7, 10, 13, 24, 32 and 48 h post dosing. Over a period of 12 h a catheter was positioned in the forearm vein. Additional 24, 32 and 48 h blood samples were taken by direct venopuncture. The heparinized blood samples were immediately centrifuged at 600 g for 15 min to obtain plasma which was then kept frozen at −20 ◦ C until analysis. Total urine was collected every 2 h over a period of 48 h post dosing and stored at −20 ◦ C until analysis.

Talinolol analysis from plasma and urine

HPLC analysis of talinolol concentrations in plasma and urine was performed using a validated HPLC method, based on the reported procedure by Tannergren et al. (2001). Talinolol was extracted from urine and plasma by liquid–liquid extraction as described by Wetterich et al. using dichloromethane/isopropanol (95:5) as extraction fluid. The organic layer was separated and evaporated under a stream of nitrogen (Wetterich et al., 1996). The resulting residues were dissolved in 100 ␮l 0.1N hydrochloric acid and 100 ␮l mobile phase (formic acid/sodium formiate buffer pH 2.75, 55%, and acetonitrile, 45%, v/v). About 50 ␮l of sample solution was injected into the HPLC and analysed using a Lichrocart Superspher 100 RP-18 (250 mm × 4 mm) column (Merck, Darmstadt) under isocratic conditions at 40 ◦ C and a flow rate of 1 ml/min. Carvedilol was used as internal standard. Talinolol and carvedilol concentrations in the eluate were analysed by UV-absorption at 240 nm. The calibration curve of talinolol was prepared in a concentration ranging from 0.005 to 1 ␮g/ml for plasma and 0.05 to 40 ␮g/ml for urine as matrix. The correlation coefficient was always better than 0.99. The limit of quantification was 5 ng/ml. The intra-day and interday accuracy of the assay in both biological matrixes were in the range between −2.2% and 11.6% of the nominal values, precision was 1.27 to 7.9% for urine and 0.55 to 13.5% for plasma. The intra-day coefficient of variation (CV) and the corresponding inter-day CV of the assay in plasma and urine were less than 4% and 5%, respectively.

2.7.

Pharmacokinetic data analysis

Maximum plasma concentrations (Cmax ) and times of maximum plasma concentrations (tmax ) were obtained from the observed data. For calculation of areas under the plasma concentration–time curves (AUC) the linear trapezoidal rule was used with extrapolation to infinity (AUC0→∞ ). The elimination rate constant (z ) was calculated by linear regression of the terminal slope of the logarithm of plasma concentration versus time profile. The mean resident time (MRTtot ) was calculated from the ratio of the area under the first moment curve, AUMC0→∞ , and the AUC0→∞ . Cumulative urinary excretion versus time profiles, based on the fraction of talinolol excreted unchanged in urine, were calculated for each formulation and each volunteer individually. The total amount excreted (Ae∞ ) was calculated as the sum of the measured amounts excreted and the calculated amount to be excreted after the last sampling time, extrapolated to infinity. All values were reported as means ± S.D. The relative bioavailability was determined by comparing the mean AUC0→∞ of the tested formulation (one talinolol controlled-release tablet and two talinolol controlledrelease tablets) to that of the reference formulation (talinolol immediate-release tablet (Cordanum® )). Absorption model parameters, using the Loo–Riegelman method, are estimated independently for each volunteer after each treatment. The pharmacokinetic transfer rate con-

234

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 231–239

Fig. 1 – Talinolol dissolution from immediate-release tablets (Cordanum® ) at pH 1.0 and 6.8 (means ± S.D., n = 6).

stants k12 , k21 (k12 = 1.44; k21 = 0.67 h−1 ) after an i.v. dose of 30 mg/individual were taken from the doctoral thesis of Wetterich (1995). Based on the open two-compartment disposition model with two or three absorption sites noncompartmental pharmacokinetic parameters were assessed using Topfit version 2.0 software system. The fit of the disposition model is assessed by using Akaikes information criterion (AIC) (Ludden et al., 1994).

2.8.

Statistical data analysis

Comparison of pharmacokinetic parameters (logarithmically transformed AUC0→∞ and Cmax values) between the three oral treatments were compared using a one-way ANOVA model followed by the Tukey–Kramer Multiple Comparison Post test integrated in GraphPad PrismTM version 3 (GraphPad Software, San Diego, USA). The same model was used to compare tmax , MRTtot and t1/2 , z , Ae(0→∞) , absorption parameters (MRTabs , rate of absorption from different GIT segments, tlag(s) ). Differences were considered significant when the p-value was less than 0.05.

3.

Results

3.1.

In vitro release of talinolol from matrix tablets

The in vitro release profiles of talinolol from the immediaterelease tablet in two different dissolution media is shown in Fig. 1. At the pH of the empty stomach, the dose of talinolol was completely dissolved in less than 15 min, whereas at pH 6.8, the dissolution was significantly slower due to the pronounced pH-dependent solubility of the basic talinolol. For the controlled-release matrix tablet the dissolution profiles in various dissolution media are shown in Fig. 2. The release of talinolol from the matrix tablets was sustained without being affected by pH to a relevant extent. At pH 1.0 drug release within the initial 1–2 h was similar to the release in the other dissolution media, however at dissolution times > 2 h, release rates in 0.1N HCl dropped clearly below the rates measured in acetate and phosphate buffers. Considering

Fig. 2 – Talinolol dissolution from controlled-release matrix tablets at pH 1.0, 4.5 and 6.8 (means ± S.D., n = 6). Standard deviations were smaller than the particular icons of the dissolution curves.

relatively rapid gastric emptying in the fasted state, the differences in the in vitro release rates between 0.1N hydrochloric acid and the other dissolution media should not have any physiological implications. In acetate and phosphate buffers the dissolution was nearly completed within 8 h.

3.2.

Talinolol pharmacokinetics in plasma

Mean concentrations versus time profiles following administration of single dose of a talinolol 100 mg IR tablet, one talinolol 100 mg controlled-release tablet and two talinolol 100 mg controlled-release tablets are presented in Fig. 3. As expected, the controlled-release formulation exhibited significantly lower and delayed mean peak plasma levels Cmax as compared with the reference formulation Cordanum® . Talinolol concentrations in plasma generally were higher following administration of two controlled-release tablets compared to one CR tablet; however, the increases were less than proportional to the administered dose. The pharmacokinetic param-

Fig. 3 – Plasma concentrations (means ± S.E.M.) of talinolol following single doses of 100 mg talinolol immediate-release (quadrates), 100 mg talinolol controlled-release (triangles), and 200 mg talinolol controlled-release tablets (circles).

235

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 231–239

Table 1 – Pharmacokinetic parameters of talinolol following administration of immediate-release, controlled-release and two controlled-release tablets in seven healthy volunteers (means ± S.D.) Parameter Cmax (ng/ml) First tmax a (h) AUC(0→∞) (ng ml−1 h) t1/2 (h) MRTtot (h) Cl/F (ml/min) Ae∞ (mg) MRTexcreted (h) t1/2 (h)b ∗ a b

IR tablet Cordanum® 204.5 1.8 2933.9 19.8 27.7 612.3 27.0 18.1 12.2

± ± ± ± ± ± ± ± ±

121.8 0.8 738.4 10.5 15.2 209.1 10.7 3.2 2.9

One controlled-release tablet 43.9 6.1 1580.9 33.6 48.8 1262.3 16.1 29.4 21.5

± ± ± ± ± ± ± ± ±

*

11.6 1.5* 605.1* 22.4 31.7 679.5* 1.8* 9.6 7.9

Two controlled-release tablets 54.5 6.5 2230.1 32.2 47.6 913.0 17.1 30.9 20.3

± ± ± ± ± ± ± ± ±

12.4* 4.0* 1092.9 19.4 27.6 420.7 6.4 12.9 11.1

p < 0.05, compared to immediate-release tablet Cordanum® . Time to reach the first concentration maximum. Elimination half-lives (t1/2 ) from urinary excretion data.

eters following administration of the different study medications of talinolol are listed in Table 1. Administration of an immediate-release tablet resulted in a rapid increase in talinolol plasma concentrations which reached a maximum within 1.8 ± 0.8 h. Pharmacokinetic parameters after administration of IR tablets were comparable with the results obtained from previously conducted other human studies (Wetterich et al., 1996; Westphal et al., 2000; Giessmann et al., 2001). The plasma concentration versus time profiles were characterized by a prolongation of the peak plasma concentration, which was attained 6.1 and 6.5 h after administration of one and two matrix tablets, respectively. For the immediaterelease formulation a large standard deviation was found for Cmax , AUC0→∞ and the apparent total body clearance (Cltot /F), indicating a substantial interindividual variability of those parameters. AUC0→∞ and Cltot /F following administration of the CR tablet also varied from subject to subject but fluctuations in Cmax were less pronounced. The extent of absorption was significantly lower for the controlled-release as compared to the immediate-release formulation such that relative bioavailabilities were reduced to 53.9% (100 mg CR) and 38% (200 mg CR). The elimination half lives following administration of Cordanum® amounted to 19.8 ± 10.5 h and were shorter than following administration of the CR tablets (33.6 and 32.2 h) indicating a flip–flop situation in case of the controlled-release medication. MRTtot were significantly higher for the CR tablets when compared with the IR formulation likewise indicating a prolonged absorption of the API from the matrix tablets.

3.3.

Fig. 4 – Representative cumulative amounts of talinolol excreted unchanged vs. time for the three treatments in one volunteer.

sampling time of urine was too short to determine urinary excretion completely. Plots of cumulative amounts of talinolol excreted in urine versus time in one healthy subject following all three treatments are illustrated in Fig. 4. As expected, the data for the 48 h collection period of the respective study phases showed marked differences in total amounts of talinolol excreted

Cumulative excretion of talinolol in urine

Pharmacokinetic parameters from data analysis of urinary excretion are presented in Table 1. Following oral administration of IR tablets, the cumulative excretion of unchanged talinolol ranged between 13.3 and 32.7 mg, with a mean value of 27.0 ± 10.7% of the administered dose. The cumulative excretion levels after administration of one controlled-release tablet and two controlled-release tablets were 16.1 ± 1.8% and 8.6 ± 3.2% of the given dose, respectively. The urinary recovery of unchanged talinolol after administration of two controlled release tablets was lower that expected, we assume that the

Fig. 5 – Mean urinary excretion rates of talinolol following oral administration of 100 mg talinolol in commercially available tablet (Cordanum® ), controlled-release tablet and two controlled-release tablets (n = 6–7).

236

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 231–239

Fig. 6 – Mean cumulative amounts of unchanged talinolol absorbed following administration of IR and CR formulations by applying the Loo–Riegelman method.

between the immediate-release tablet and the controlledrelease tablet. The urinary excretion profiles of talinolol up to 48 h following one and two talinolol controlled-release tablets were found to be similar and are shown in Fig. 5. The urinary excretion rate profile following treatment with Cordanum® shows, that the maximal excretion rate (dAe /dt)max was achieved during the first 4 h. In the collection period from 0 to 6 h the urinary excretion rates were higher after application of two talinolol matrix tablets than following one talinolol matrix tablet. These excretion profiles after both treatments with one and two controlled-release tablets demonstrate that the matrix tablets exhibited comparable release kinetics with significantly delayed times for maximal excretion rate (tmax ) and a smooth and extended maximal excretion rate (dAe /dt)max .

3.4. Absorption characteristics of talinolol from IR and CR tablets The mean absorption profiles of talinolol following the administration of commercial and experimental formulations as calculated by the Loo–Riegelman method (Gibaldi and Perrier, 1982) are shown in Fig. 6. In order to further describe the input kinetics including the observed double peak phenomenon and shoulders a multisegment absorption model was applied (Mahmood, 1996). In this model absorption takes place from different segments of gastrointestinal tract. Each absorption segment (00;60;70) was defined by three parameters: the relative dose fraction, the

Fig. 7 – Talinolol plasma concentration vs. time profile for one subject expressing a biphasic drug input.

input absorption rate constant (K00,K60,K70) and the lag time. The total dose/F was split into fractions that enter the systemic circulation at different times after dosing (Langguth, 1994). A plasma concentration–time profile obtained from one volunteer after administration of IR talinolol tablets displays two significant peaks and is graphically presented in Fig. 7. Administration of controlled-release talinolol tablets lead to delays in absorption of a certain fraction of the dose, exhibiting broad plateau-like maxima and shoulders. Calculated absorption rates and lag times for each of the absorption sites were used to determine the absorption profile. In Table 2 the different absorption parameters are listed, based on curvefitting by the established multi-segmental absorption model. In general, the absorption rate from the controlled-release formulation was lower when compared with the immediaterelease formulation. Mean absorption times were 1.7 h for the IR formulation and were extended to 5–6 h following administration of the CR tablet.

4.

Discussion

The present study was designed to investigate the impact of modifications in the release rate of a P-glycoprotein substrate from its oral dosage form on the biopharmaceutics of the active pharmaceutical ingredient. Particular emphasis was attached to the change of absorption parameters between administration of an immediate-release versus a controlled-

Table 2 – Parameters of the multi-segmental input model (means ± S.D.) of talinolol following IR, CR and two CR tablets Parameter

Immediate-release −1

Absorption rate constant (h ) Rate of absorption from segment 00 (h−1 ) Rate of absorption from segment 60 (h−1 ) Rate of absorption from segment 70 (h−1 ) Mean absorption time (h) Lag time1 (h) Lag time2 (h) Lag time3 (h) ∗

3.69 ± 4.65 2.12 13.10 – 1.73 ± 1.35 0.51 ± 0.36 3.91 –

p < 0.05, compared to immediate-release tablet Cordanum® .

CR tablet 0.55 4.00 12.16 5.91 5.89 0.15 2.86 17.06

± ± ± ± ± ± ± ±

0.18 3.18 22.2 7.10 2.46* 0.27 0.84 10.80

Two CR tablets 1.68 4.09 2.5 7.34 6.92 0.50 2.62 12.33

± ± ± ± ± ± ± ±

1.52 3.99 2.16 12.09 2.65* 0.85 1.90 6.70

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 231–239

release dosage form. The ␤-adrenoceptor antagonist talinolol was chosen as a model drug due to its well documented intestinal secretion pathway mediated via P-glycoprotein. Traditional reasons for the creation of a modified release dosage form, e.g. extension of the terminal half-life of a short acting drug, do not apply for talinolol, since its elimination half-life in healthy human subjects has been reported to lie close to 12 h (Schmitt, 1995). In conventional immediate-release formulations talinolol is furthermore rapidly absorbed showing peak plasma levels within 2 h following administration. For these formulations, rapid dissolution of the immediaterelease tablet results in high talinolol concentrations in the upper parts of the gastrointestinal tract, i.e. in the stomach, duodenum and upper jejunum, such that the saturation level of the exsorptive transporter is exceeded. Therefore, the amount of drug undergoing intestinal efflux is reduced and absorption is enhanced in particular at high doses in IR formulations. In the rat, the talinolol permeability decreases along the intestine in parallel to an increase in P-gp mRNA levels and higher expression in the lower parts of the intestine (Wagner et al., 2001). Using human perfusion studies the regional absorption characteristics of talinolol has been assessed. Results reported by Gramatte et al., suggested significant site-dependence in talinolol’s intestinal permeability and point to the likelihood of the existence of an “absorption window” in the upper small intestine, i.e. a region of higher permeability (Gramatte et al., 1996). The absorption of P-gp substrates is assumed to decrease in regions with higher expression of secretory carriers. The results published by Mouly and Paine suggest regional differences in the functional expression of P-gp with higher activity in the distal ileum as compared to the jejunum (Mouly and Paine, 2003). These findings were in agreement with PCR experiments in the rat (Brady et al., 2002), where mdr1 mRNA levels increased in the order duodenum > jejunum > ileum. With these physiological constraints in mind, it was expected that a decrease of the release rate of a P-gp substrate from the pharmaceutical formulation will result in lower drug concentrations in the small and large intestine. Consequently, API concentrations will range below the saturation level of the transporter, and in this case an increased contribution of the drug efflux to the overall absorption will be expected, resulting in a decrease in bioavailability. These phenomena were clearly observed in the current study where the systemic exposure of talinolol following administration of 100 mg of talinolol in a controlled-release tablet resulted in 53.9% (41.2% based on AUC(0-48h) values) systemic availability when compared with the same dose administered in an IR dosage form. After administration of two controlled-release tablets simultaneously, the dose-corrected systemic exposure amounted to 38% (27.1% based on AUC(0-48h) values). This was somewhat unexpected and may be explained by taking into consideration some physicochemical properties of the drug. Talinolol belongs to class II of a modified version of the Biopharmaceutic Classification System, which includes the effects of efflux and absorptive transporters on oral drug absorption (Wu and Benet, 2005). Furthermore, the solubility of talinolol is pHdependent and lower at higher intestinal pH. In this case, the higher dose may not be completely soluble in the intestinal fluids in the distal intestine thus obstructing the effects antic-

237

ipated in the case of saturation of efflux transporters at higher doses only. Further modelling using advanced compartmental absorption and transit models may help to clarify this issue in the future. An alternative explanation for the lower bioavailability of talinolol following CR administration could be the existence of an uptake transporter in the upper small intestine (Wu and Benet, 2005; Weitschies et al., 2005). The multiple peaks and shoulders phenomenon which is characteristic for plasma concentration–time curves in many individuals after oral administration of, e.g. talinolol (Wetterich et al., 1996; Siegmund et al., 2003), ranitidine (Garg et al., 1983), cimetidine (Walkenstein et al., 1978; Langguth et al., 1994), furosemide (Hammarlund et al., 1984), acebutolol (Meffin et al., 1978) may be the result of several complex processes: absorption from multiple sites in the GI tract at different rates, enterohepatic and entero-enteral recycling, gastric emptying, storage and subsequent release of drug from a postabsorptive depot site. Enterohepatic cycling as a cause of the occurrence of double peaks can be excluded in the present case, even when double peak behaviour was observed following talinolol intravenous administration, since biliary elimination following talinolol administration was less than 10% of an intravenous dose (Terhaag et al., 1989). Double peaks were clearly less evident in the controlledrelease plasma profiles in the present study. These observations are in accordance with the findings of Gramatte et al., who reported double peaked blood levels in particular during the perfusion of proximal intestinal segments (Gramatte et al., 1996). In vitro tests (Wagner et al., 2003) showed that addition of surfactants can decelerate talinolol dissolution it may well be possible that physiological surfactants such as bile salts, cholesterol, phospholipids, small intestinal peptides and conjugated steroids could be held responsible for decrease in bioavailability as well. Furthermore, as reported previously for pafenolol (Lennernas and Regardh, 1993) and ranitidine (Suttle and Brouwer, 1994), an explanation for the double peak phenomenon could be the interaction between bile constituents and talinolol in the proximal part of the small intestine. A fraction of the released talinolol from the immediate-release tablet may be encapsulated in micellar complexes that dissociate only in the distal parts of the small intestine, where absorption of the residual amount of talinolol can take place and second peaks may thus appear. Double peaks were not always evident after IR administration in the present study, the explanation could be that after overnight fasting and administration of the dosage form in the fasted state, gall bladder contraction was not stimulated, and bile acid concentration in proximal intestine was not sufficiently high to form such complexes. Moreover, at high talinolol concentrations the binding to the complexes might be saturated and more drug would become available for absorption. When subjects received controlled-release tablets, talinolol release occurred mainly in the lower parts of the intestine, thus avoiding the formation of such nonabsorbable complexes. Deferme et al. reported that human intestinal fluid shows possible direct or indirect inhibitory effect on P-gp mediated efflux (Deferme et al., 2003). Furthermore, cholesterol, as a bile component, may directly inhibit the function of P-gp (Wang et al., 2000), due to occupation of the sub-

238

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 231–239

strate binding site of P-gp. The P-gp effect on talinolol, which is distally pronounced, may mask the human intestinal fluid influence on P-gp mediated secretion. As reported previously (Weitschies et al., 2005) the multiple peaks and shoulders may also result from talinolol storage within or following the intestinal barrier, absorption and subsequent release. The absorption of talinolol in some subjects after administration of controlled-release tablets was complex and could not be described by standard first-order and zero-order absorption rate. It was found, that oral absorption in this case was best described by a sequential zero-order/first-order absorption model. Absorption analysis shows that the first phase represents zero-order absorption kinetics, while the second phase corresponds to a first-order process (Zhou, 2003). A significant interindividual variability of pharmacokinetic parameters was observed in all study phases. Such results are in accordance with the literature (Wetterich et al., 1996; Siegmund et al., 2003). Bogman et al. reported that intersubject variation in intestinal transit time and pharmacogenetic variation in P-gp expression (Bogman et al., 2005; Mouly and Paine, 2003; Ayrton and Morgan, 2001; Lindell et al., 2003) might induce high intersubject variability. Multiple studies have demonstrated polymorphisms in the MDR1 gene that might have consequences on P-gp expression and function (Ayrton and Morgan, 2001; Lindell et al., 2003). In summary, talinolol controlled-release formulations resulted in lower and relatively constant plasma concentrations compared with the immediate-release treatment. This experimental proof is consistent with the theory of saturation of carrier-mediated intestinal secretion at higher doses and/or higher release rates of the active pharmaceutical ingredient from the dosage form. Thus, to attain the therapeutic benefit from controlled-release dosage forms with drugs undergoing intestinal efflux the influence of intestinal exsorptive transporters such as P-gp has to be considered in the development and design of controlled-release dosage forms. There is an incomplete absorption from more distal sites and talinolol and possible other P-gp substrates may not be suitable for use in sustained release formulations, unless potent inhibitors of the intestinal efflux system are coadministered with or incorporated into the dosage form.

references

Ayrton, A., Morgan, P., 2001. Role of transport proteins in drug absorption, distribution and excretion. Xenobiotica 31, 469–497. Bogman, K., Zysset, Y., Degen, L., Hopfgartner, G., Gutmann, H., Alsenz, J., Drewe, J., 2005. P-glycoprotein and surfactants: effect on intestinal talinolol absorption. Clin. Pharmacol. Ther. 77, 24–32. Brady, J.M., Cherrington, N.J., Hartley, D.P., Buist, S.C., Li, N., Klaassen, C.D., 2002. Tissue distribution and chemical induction of multiple drug resistance genes in rats. Drug. Metab. Dispos. 30, 838–844. Deferme, S., Tack, J., Lammert, F., Augustijns, P., 2003. P-glycoprotein attenuating effect of human intestinal fluid. Pharm. Res. 20, 900–903.

Garg, D.C., Weidler, D.J., Eshelman, F.N., 1983. Ranitidine bioavailability and kinetics in normal male subjects. Clin. Pharmacol. Ther. 33, 445–452. Gibaldi, M., Perrier, D., 1982. Pharmacokinetics. Marcel Dekker, New York. Giessmann, T., Zschiesche, M., Modess, C., Tschuschke, A., Hartmann, V., Wegner, D., Weitschies, W., Siegmund, W., 2001. Is the double peak phenomenon of talinolol caused by lymphatic transport? Poster at 5. EACPT Congress, Odense, Denmark. Gramatte, T., Oertel, R., Terhaag, B., Kirch, W., 1996. Direct demonstration of small intestinal secretion and site-dependent absorption of the beta-blocker talinolol in humans. Clin. Pharmacol. Ther. 59, 541–549. Hammarlund, M.M., Paalzow, L.K., Odlind, B., 1984. Pharmacokinetics of furosemide in man after intravenous and oral administration. Application of moment analysis. Eur. J. Clin. Pharmacol. 26, 197–207. Langguth, P., Lee, K.M., Spahn-Langguth, H., Amidon, G.L., 1994. Variable gastric emptying and discontinuities in drug absorption profiles: dependence of rates and extent of cimetidine absorption on motility phase and pH. Biopharm. Drug. Dispos. 15, 719–746. Lennernas, H., Regardh, C.G., 1993. Evidence for an interaction between the beta-blocker pafenolol and bile salts in the intestinal lumen of the rat leading to dose-dependent oral absorption and double peaks in the plasma concentration–time profile. Pharm. Res. 10, 879–883. Lindell, M., Karlsson, M.O., Lennernas, H., Pahlman, L., Lang, M.A., 2003. Variable expression of CYP and Pgp genes in the human small intestine. Eur. J. Clin. Invest. 33, 493– 499. Ludden, T.M., Beal, S.L., Sheiner, L.B., 1994. Comparison of the Akaike information criterion, the Schwarz criterion and the F test as guides to model selection. J. Pharmacokinet. Biopharm. 22, 431–445. Mahmood, I., 1996. Pharmacokinetic analysis of the absorption characteristics of diclofenac sodium in man by use of a multi-segment absorption model. J. Pharm. Pharmacol. 48, 1260–1263. Matheny, C.J., Lamb, M.W., Brouwer, K.R., Pollack, G.M., 2001. Pharmacokinetic and pharmacodynamic implications of P-glycoprotein modulation. Pharmacotherapy 21, 778–796. Meffin, P.J., Winkle, R.A., Peters, F.A., Harrison, D.C., 1978. Dose-dependent acebutolol disposition after oral administration. Clin. Pharmacol. Ther. 24, 542–547. Mouly, S., Paine, M.F., 2003. P-glycoprotein increases from proximal to distal regions of human small intestine. Pharm. Res. 20, 1595–1599. Oertel, R., Richter, K., Trausch, B., Gramatte, T., 1994. A quick, simple HPLC method for determination of talinolol in serum. Pharmazie 49, 291–292. Schmitt, J., 1995. Talinolol, ein klinisch bewahrter ß1-selektiver Betablocker-Eine Uebersicht zu pharmakokinetischen und pharmakodynamischen Eigenschaften. Perfusion 9, 291–296. Siegmund, W., Ludwig, K., Engel, G., Zschiesche, M., Franke, G., Hoffmann, A., Terhaag, B., Weitschies, W., 2003. Variability of intestinal expression of P-glycoprotein in healthy volunteers as described by absorption of talinolol from four bioequivalent tablets. J. Pharm. Sci. 92, 604–610. Spahn-Langguth, H., Baktir, G., Radschuweit, A., Okyar, A., Terhaag, B., Ader, P., Hanafy, A., Langguth, P., 1998. P-glycoprotein transporters and the gastrointestinal tract: evaluation of the potential in vivo relevance of in vitro data employing talinolol as model compound. Int. J. Clin. Pharmacol. Ther. 36, 16–24. Suttle, A.B., Brouwer, K.L., 1994. Bile flow but not enterohepatic recirculation influences the pharmacokinetics of ranitidine in the rat. Drug. Metab. Dispos. 22, 224–232.

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 2 9 ( 2 0 0 6 ) 231–239

Tannergren, C., Langguth, P., Hoffmann, K.J., 2001. Compound mixtures in caco-2 cell permeability screens as a means to increase screening capacity. Pharmazie 56, 337–342. Terhaag, B., Gramatte, T., Richter, K., Voss, J., Feller, K., 1989. The biliary elimination of the selective beta-receptor blocking drug talinolol in man. Int. J. Clin. Pharmacol. Ther. Toxicol. 27, 170–172. Wagner, D., Glube, N., Berntsen, N., Tremel, W., Langguth, P., 2003. Different dissolution media lead to different crystal structures of talinolol with impact on its dissolution and solubility. Drug. Dev. Ind. Pharm. 29, 891–902. Wagner, D., Spahn-Langguth, H., Hanafy, A., Koggel, A., Langguth, P., 2001. Intestinal drug efflux: formulation and food effects. Adv. Drug Deliv. Rev. 50 (Suppl. 1), S13–S31. Walkenstein, S.S., Dubb, J.W., Randolph, W.C., Westlake, W.J., Stote, R.M., Intoccia, A.P., 1978. Bioavailability of cimetidine in man. Gastroenterology 74, 360–365. Wang, E., Casciano, C.N., Clement, R.P., Johnson, W.W., 2000. Cholesterol interaction with the daunorubicin binding site of P-glycoprotein. Biochem. Biophys. Res. Commun. 276, 909–9016. Weitschies, W., Bernsdorf, A., Giessmann, T., Zschiesche, M., Modess, C., Hartmann, V., Mrazek, C., Wegner, D., Nagel, S.,

239

Siegmund, W., 2005. The talinolol double-peak phenomenon is likely caused by presystemic processing after uptake from gut lumen. Pharm. Res. 22, 728–735. Westphal, K., Weinbrenner, A., Giessmann, T., Stuhr, M., Franke, G., Zschiesche, M., Oertel, R., Terhaag, B., Kroemer, H.K., Siegmund, W., 2000. Oral bioavailability of digoxin is enhanced by talinolol: evidence for involvement of intestinal P-glycoprotein. Clin. Pharmacol. Ther. 68, 6–12. Wetterich, U., 1995. Doctoral Thesis. Wetterich, U., Spahn-Langguth, H., Mutschler, E., Terhaag, B., Rosch, W., Langguth, P., 1996. Evidence for intestinal secretion as an additional clearance pathway of talinolol enantiomers: concentration-and dose-dependent absorption in vitro and in vivo. Pharm. Res. 13, 514–522. Wu, C.Y., Benet, L.Z., 2005. Predicting drug disposition via application of BCS: transport/absorption/elimination interplay and development of a biopharmaceutics drug disposition classification system. Pharm. Res. 22, 11–23. Zhou, H., 2003. Pharmacokinetic strategies in deciphering atypical drug absorption profiles. J. Clin. Pharmacol. 43, 211–227.