- Email: [email protected]
Examination of oral absorption and lymphatic transport of halofantrine in a triple-cannulated canine model after administration in self-microemulsifying drug delivery systems (SMEDDS) containing structured triglycerides
European Journal of Pharmaceutical Sciences 20 (2003) 91–97 www.elsevier.com / locate / ejps
Examination of oral absorption and lymphatic transport of halofantrine in a triple-cannulated canine model after administration in self-microemulsifying drug delivery systems (SMEDDS) containing structured triglycerides a, *, ¨ Rene´ Holm a ,1 , Christopher J.H. Porter b , Glenn A. Edwards c , Anette Mullertz a b Henning G. Kristensen , William N. Charman a
b
Department of Pharmaceutics, The Royal Danish School of Pharmacy, Universitetsparken 2, 2100 Copenhagen, Denmark Department of Pharmaceutics, Victorian College of Pharmacy, Monash University, 381 Royal Parade, Victoria 3052, Australia c Department of Veterinary Science, University of Melbourne, 250 Princes Highway, Werribee, Victoria 3030, Australia Received 19 February 2003; received in revised form 13 June 2003; accepted 23 June 2003
Abstract The potential for lipidic self-microemulsifying drug delivery systems (SMEDDS) containing triglycerides with a defined structure, where the different fatty acids on the glycerol backbone exhibit different metabolic fate, to improve the lymphatic transport and the portal absorption of a poorly water-soluble drug, halofantrine, were investigated in fasted lymph cannulated canines. Two different structured triglycerides were incorporated into the SMEDDS; 1,3-dioctanoyl-2-linoleyl-sn-glycerol (C8:0-C18:2-C8:0) (MLM) and 1,3-dilinoyl-2octanoyl-sn-glycerol (C18:2-C8:0-C18:2) (LML). A previously optimised SMEDDS formulation for halofantrine, comprising of triglyceride, Cremophor EL, Maisine 35-1 and ethanol was selected for bioavailability assessment. The extent of lymphatic transport via the thoracic duct was 17.9% of the dose for the animals dosed with the MLM SMEDDS and 27.4% for LML. Also the plasma availability was affected by the triglyceride incorporated into the multi-component delivery system and availabilities of 56.9% (MLM) and 37.2% (LML) were found. These data indicate that the pharmaceutical scientist can use the structure of the lipid to affect the relative contribution of the two absorption pathways. The MLM formulation produced a total bioavailability of 74.9%, which is higher than the total absorption previously observed after post-prandial administration. This could indicate the utility of disperse lipid-base formulations based on structured triglycerides for the oral delivery of halofantrine, and potentially other lipophilic drugs. 2003 Elsevier B.V. All rights reserved. Keywords: Halofantrine; Structured triglyceride; Lymphatic transport; Bioavailability; Canine; SMEDDS
1. Introduction The oral absorption of a number of potent lipophilic compounds is limited by their physico-chemical properties and has shown low and variable availability. This has led to a growing effort to develop pharmaceutical formulations enhancing the oral bioavailability of these lipophilic compounds. An increasingly popular approach to overcome the low oral bioavailability is the incorporation of the active lipophilic component into lipid vehicles such as oil solutions, self-emulsifying drug delivery systems (SEDDS) (Humberstone and Charman, 1997; Pouton, 2000). The application of lipid based delivery systems has *Corresponding author. Tel.: 145-35-30-6440; fax: 145-35-30-6031. ¨ E-mail address: [email protected] (A. Mullertz). 1 Present address: H. Lundbeck A / S, Ottilavej 9, 2500 Valby, Denmark. 0928-0987 / 03 / $ – see front matter 2003 Elsevier B.V. All rights reserved. doi:10.1016 / S0928-0987(03)00174-X
gained considerable interest after the commercial success of lipid based formulations of cyclosporine A (Sandimmun Neoral姠), saquinavir (Fortovase) and ritonavir (Norvir). SEDDS are isotropic mixtures of oil, a surfactant, and possibly one or more hydrophilic solvents or co-surfactants, which form fine oil-in-water emulsions or microemulsions (SMEDDS) when exposed to aqueous media under conditions of gentle agitation (Constantinides, 1995). SEDDS typically produce emulsions with a particle size between 100 and 300 nm, while SMEDDS form transparent microemulsions with a particle size of less than 100 nm. These properties have been suggested to make SEDDS and SMEDDS a good formulation alternative for the oral delivery of lipophilic drugs (Charman et al., 1992). Khoo et al. (1998) have pursued the potential for SEDDS to improve the oral bioavailability of halofantrine,
92
R. Holm et al. / European Journal of Pharmaceutical Sciences 20 (2003) 91–97
a highly lipophilic antimalaria agent, in fasted beagle dogs, and showed a trend towards a higher bioavailability when dosed in SMEDDS based on long-chain triglyceride. The lymphatic transport of halofantrine has been found to play a major role in the absorption from the intestine when co-administered with a lipid formulation in rats (Porter et al., 1996; Caliph et al., 2000). Recently Khoo et al. (2003) showed that halofantrine was also transported by the lymphatic systems in dogs when dosed in a human size long-chain triglyceride based SMEDDS formulation described previously (Khoo et al., 1998). Generally, compounds processed by the intestinal lymph are transported to the systemic circulation in association with the lipid core of lipoproteins (Pocock and Vost, 1974), and as such require co-administered lipid to stimulate lipoprotein formation. Short- and medium-chain fatty acids (with a carbon chain length shorter than 12 carbon atoms) are transported to the systemic circulation by the portal blood and are not incorporated to a great extent in chylomicrons (Kiyasu et al., 1952). In contrast, long-chain fatty acids and monoglycerides are re-esterified to triglycerides within the intestinal cell, incorporated into chylomicrons and secreted from the intestinal cell by exocytosis into the lymph vessels. In addition to the stimulation of lymphatic transport, administration of lipophilic drugs with lipids may enhance drug absorption into the portal blood when compared to nonlipid formulations (Caliph et al., 2000). In a recent study Holm et al. (2002) demonstrated that the intestinal lymphatic transport and the portal absorption of halofantrine when co-administered with a structured triglyceride (a triglyceride where the glycerol backbone is esterified with different fatty acid chains (Fig. 1)), was affected compared to a natural long chain triglyceride, sunflower oil, in lymph cannulated rats. However, there is limited additional information in the literature detailing the positional benefits in terms of drug delivery of these unique triglyceride vehicles. Furthermore, it is not known if these triglycerides can be incorporated into a pharmaceutical formulation without losing their beneficial properties. Moreover, it also remains unexplored how the absorption paten of a lipophilic drug would be affected in a non-rodent animal species with one of these structured triglycerides. The objective of this study was therefore to
Fig. 1. Schematic presentation of the structured lipids used in this study. M represents a medium-chain fatty acid (C 8 – 10 ), and L represents a long chain fatty acid (C 18 ).
investigate the impact of structured triglycerides with varying intramolecular structures and chain lengths incorporated into a pharmaceutical formulation, on the intestinal lymphatic transport and absorption into the blood of halofantrine, in lymph cannulated conscious canines.
2. Materials and methods
2.1. Chemicals and reagents Crystalline halofantrine free base and the internal standard 2,4-dichloro-6-trifluoromethyl-9h1-[2-dibutylamino)ethyl]jphenathrenemethanol hydrochloride were donated by SmithKline Beecham Pharmaceuticals (Mysore, India). The structured triglycerides, 1,3-dioctanoyl-2-linoleyl-snglycerol (C8:0-C18:2-C8:0) (MLM) and 1,3-dilinoyl-2octanoyl-sn-glycerol (C18:2-C8:0-C18:2) (LML), were manufactured at the Department of Biotechnology at the Technical University of Denmark as previously described (Mu et al., 1998). The fatty acid compositions of the structured triglycerides were previously reported by Holm et al. (2002). Cremophor EL was donated by BASF ´ (Ludwigshafen, Germany) and Maisine 35-1 by Gattefosse (Saint-Priest, France). R.P. Scherer (Australia) kindly provided 1 ml air-filled gelatine capsules. Acetylpromazine maleate (Delvet Pty. Ltd., Australia), propofol (ScheringPlough, Australia), carprofen (Pfizer, Australia), cephazolin (Sigma Pharmaceuticals, Australia) and polydioxinone sutures (Ethicon, USA) were used during the surgery. Normal saline (0.9%) and Hartmans solution were obtained from Baxter Healthcare (Australia). Acetonitrile and tert-butylmethylether were HPLC grade and sodium dodecyl sulphate was electrophoresis grade. The water used in all experiments was obtained from a Milli-Q-water purification system (Milipore, Milford, MA). All other chemicals were analytical reagent grade.
2.2. Halofantrine formulations The composition of the formulations was based on a previous published SMEDDS which was developed for halofantrine by Khoo et al. (1998). The long-chain triglyceride used by Khoo and co-workers was, however, substituted with the structured triglycerides. The SMEDDS formulations consisted of halofantrine: triglyceride: Maisine-35-1: Cremophor EL: absolute ethanol (5:29:29:30:7) (% w / w). Approximately 4 g of each formulation was prepared the day before use, by weighing halofantrine base into a 12 ml teflon-lined screw-capped glass conical tube, followed by addition of Cremophor EL, Maisine 35-1 and the structured triglycerides (MLM or LML). The components were mixed by gentle stirring, and heated in a 50 8C water bath, until all the halofantrine had dissolved. The mixture was then cooled to ambient temperature and the ethanol added. One gram of the formula-
R. Holm et al. / European Journal of Pharmaceutical Sciences 20 (2003) 91–97
tion was filled into a soft gelatine capsule using a syringe and a needle. A disintegration experiment was performed each day to assess the efficacy of the self-emulsification using a standard dissolution apparatus (Erweka, Germany) with the paddle set at 50 rpm at 37 8C. The droplet size of the emulsions was determined by photo correlation spectroscopy using a Zetasizer 3000 (Malvern Instruments, UK). Both formulations produced a clear slightly bluish microemulsion with a mean particle size of 47.9 nm (MLM microemulsion) and 50.3 nm (LML microemulsion). The concentration of halofantrine in the formulations was determined prior to administration by high-performance liquid chromatography (HPLC), as described by Humberstone et al. (1995), thereby determining the exact dose of halofantrine administered.
2.3. Surgical procedures
93
1, 1.5, 2, 2.5, 3, 3.5, 4, 5, 6, 7, 8, 10, 12 and 24 h after the drug administration. Blood samples were collected into individual tubes containing dipotassium EDTA. The plasma was harvested immediately by centrifugation for 10 min (1000 g) and stored at 220 8C until further analysis. Lymph was collected continuously into 50 ml tubes containing 75 mg dipotassium EDTA, and collection tubes were changed hourly. The lymph was collected for 12 h post-dosing. The amount of lymph collected in each interval was determined gravimetrically. Lymph samples were stored at 5–8 8C prior to analysis (within 16 h). At the conclusion of the experiment the animals were sacrificed by an overdose of sodium pentobarbitone given intravenously.
2.5. Analysis of halofantrine in blood and lymph
All surgical and experimental procedures were reviewed and approved by the local Institutional Animal Experimentation Ethics Committee (University of Melbourne, Australia). Studies were conducted in male greyhound dogs (28–35 kg), and their health status was verified by a veterinarian prior to the study. The surgery was performed as described by Khoo et al. (2001). Each dog was fed a small lipid meal prior to surgery to facilitate subsequent identification of the thoracic lymph duct. The dogs received a pre-anaesthetic by subcutaneous injection of acetylpromazine maleate (0.5 mg / kg) and were subsequently anaesthetised with an intravenous injection of propofol (3–6 mg / kg). The surgical anaesthesia was maintained by delivery of halothane and oxygen. The dogs received intravenous infusion of saline during surgery, as well as post-operative injection of antibiotic (cephazolin, 20 mg / kg) and analgesic (carprofen 4 mg / kg). Following surgery, the dogs were allowed to recover unrestrained in a closed run for 15–19 h. In the initial period of recovery, fluids were administered intravenously to ensure adequate hydration and to prevent hypoproteinaemia. The dogs were allowed to return to normal ambulatory movements prior to drug administration. An intravenous catheter was inserted into the cephalic vein immediately prior to dosing to facilitate sampling of systemic blood during the study.
For analysis of halofantrine in the lymph, 100 ml of lymph was added to 10 ml acetonitrile and the sample was vortexed for 1 min. Insoluble protein-based components were removed by centrifugation (1000 g) and the supernatant analysed by HPLC as previously described (Porter et al., 1996). Recovery of spiked halofantrine from blank lymph was greater than 99%. As all the lymph draining from the thoracic lymph duct was collected, the mass of drug transported was calculated by multiplying the lymph drug concentration by the corresponding mass of lymph from each collection period. Blood samples were analysed using a validated HPLC method (Humberstone et al., 1995), by adding 200 ml plasma and 100 ml internal standard (2 mg / ml) to 1.0 ml acetonitrile in a 12 ml polypropylene centrifuge tube. Samples were subsequently vortexed for 2 min to precipitate plasma proteins then centrifuged (800 g), and 8 ml aliquot of tert-butyl methyl ether was added. The tube was vortexed and centrifuged (1000 g) and 8 ml of the upper organic layer was removed into a polypropylene centrifuge tube containing 100 ml of 0.005 M HCl (in acetonitrile). The organic phase was then evaporated to dryness under nitrogen at 35 8C. The residue was reconstituted with 100 ml acetonitrile and 25 ml was injected on to the LC column. The limit of quantification by this procedure was 20 ng / ml. The assay was linear between 20 ng / ml and 4000 ng / ml, and the extraction efficiency for halofantrine was greater than 90% across the concentration range.
2.4. Experimental procedures
2.6. Analysis of lymph triglyceride
The fasted dogs were orally administered a single soft gelatine capsule containing 1 g of the SMEDDS formulation (and 50 mg halofantrine) with 50 ml of water. To prevent dehydration, 25 ml of Hartmann’s solution was administered at hourly intervals by intravenous bolus injections. Water was available ad libitum, however, food was withheld until after collection of the 12 h sample. Blood samples (2.5 ml) were taken at 25 min and 0.5,
The triglyceride concentration in lymph was measured using automated clinical chemistry analyser (Roche Cobas Mira, Basle, Switzerland) and an enzyme-based colorimetric assay kit (Boehringer Mannheim, Germany).
2.7. Lymph distribution of halofantrine The lymph collected for the first 6 h was separated as
94
R. Holm et al. / European Journal of Pharmaceutical Sciences 20 (2003) 91–97
described by Raub et al. (1992), using a Beckman SW-60 rotor. The chylomicrons were firstly separated by layering 1 ml of lymph under a sodium chloride solution (d5 1.0063 g / ml) followed by centrifugation at 44 100 rpm for 1 h and 20 min (262 000 g) at 15 8C. After centrifugation, the bottom of the tube was pierced using a needle to enable the remaining lymph to be removed, leaving only the chylomicron fraction which forms a white semi-solid plug at the top of the tube. After fractionation, the chylomicron fraction was dissolved in 10 ml acetonitrile and the proportion of halofantrine associated with chylomicrons determined by HPLC.
2.8. Pharmacokinetic analysis Plasma concentrations versus time data for halofantrine for individual dogs were analysed by standard non-compartment analysis using WinNonlin software (version 2.1). The total bioavailability was calculated by adding the cumulative amount of halofantrine transported in the lymph with the percent of halofantrine found in the systemic circulation. An estimate of the percent absorption of halofantrine into the systemic circulation was determined by comparison of the dose normality obtained in lymph cannulated canines after oral administration in the SMEDDS formulation with the AUC after intravenous administration from previous data by Khoo et al. (2002).
2.9. Statistical analysis Statistical analysis was performed by one-way analysis of variance (SigmaStat for Windows version 2.0). The Student–Newman–Keuls multiple comparison test was subsequently applied to analyse potential difference between the formulations. The results were considered significant if P,0.05.
on the absorption of the highly lipophilic antimalarial halofantrine.
3.1. Intestinal lymphatic transport of halofantrine The lymphatic transport of halofantrine, expressed as the cumulative percentage of the administered dose, is presented in Fig. 2. The extent of lymphatic transport of halofantrine after 12 h (mean% dose6S.E.) was statistically different and was 27.461.3 (n54) after administration in the LML and 17.961.3 (n55) in the MLM. Khoo et al. (2003) reported a lymphatic transport of halofantrine in a similar formulation containing a natural long-chain triglyceride (soybean oil) on 28.367.6 (n54), findings that are equal to the lymphatic transport found in the LML, demonstrating that this structural formation of the triglyceride initiates a lymphatic transfer at a high level Both examined SMEDDS formulations in this study introduced a lymphatic transport higher than the transport reported when dosed in a SMEDDS formulated with medium-chain triglycerides in lymph cannulated canines (Khoo et al., 2003). These data indicate that the small amount of lipid dosed in a soft gelatine capsule (here 580 mg of absorbable lipid) was sufficient to trigger the biochemical processes and initiate lymphatic transport of halofantrine, consistent with data previously reported by Khoo et al. (2003), also when structural triglycerides are used as a vehicle. The canines were dosed with a lipid meal before the surgery to facilitate the identification and cannulation of the thoracic lymph duct, which potentially could influence the content of halofantrine found in the lymph. However, Khoo et al. (2001) previously reported a lymphatic transport of halofantrine, in a study with a similar experimental setup as the one used in the present study, of 1.360.2% in
3. Results and discussion Rats are often used to investigate the intestinal lymphatic transport of lipophilic drugs and for examining the effect of lipidic formulations on the lymphatic transport and bioavailability due to their low cost and relatively facile accessibility. The conduction in rats is limited by the volume of formulation that can be administered to each animal and their continuous bile flow. Unlike the rat, dogs can be administered formulations similar to the formulations intended for use in humans. In a previous study Khoo et al. (2003) have demonstrated that the relative small amount of lipid contained in a SMEDDS formulation can significantly increase lymphatic transport of halofantrine. In this study we have therefore used lymph cannulated canines to evaluate the impact of a structured triglyceride incorporated into a pharmaceutical SMEDDS formulation
Fig. 2. Cumulative percentage dose of halofantrine (mean6S.E.) collected in the thoracic lymph as a function of time after oral administration of 50 mg halofantrine contained in SMEDDS formulation containing two different structured triglycerides to conscious canines. MLM (n55, j) and LML (n54, •).
R. Holm et al. / European Journal of Pharmaceutical Sciences 20 (2003) 91–97
fasted animals. This transport could be considered as the basal lymphatic transport initiated by the endogenous lipid turnover as well as the potential interaction of the food dosed before the surgery. With a very limited lymphatic transport found in the base situation the potential influence of food dosed before surgery was not considered to have an impact on the results in this study with the used experimental setting.
3.2. Lymphatic transport of triglycerides and lymph flow The cumulative transport of C 18 triglyceride into the thoracic lymph is presented in Fig. 3. Lipid collected in the thoracic lymph 12 h after dosing (expressed as the mean6S.E., n54–5) was 5.261.2 g for animals dosed with LML and 2.660.7 g for animals dosed with MLM. Statistically, the two groups are not significantly different, due to large variation, but the mean amount of triglyceride found in the lymph suggests a difference as a function of the dosed lipid. In fasted dogs, triglyceride transport has previously been reported by Khoo et al. (2001) at 0.560.2 g (mean6S.E.), which reflected basal transport levels arising from endogenously derived lipid. Both SMEDDS formulations examined in this study had a higher mean triglyceride transport confirming that the available lipid in the SMEDDS is sufficient to trigger the biochemical processes involved in the lymphatic transport of lipids. Previous data suggest that the lymphatic transport of medium chain lipids is small (Kiyasu et al., 1952). MLM was expected to have a lower lymphatic triglyceride transport when compared to the LML formulation, because of the lower molar amount of long chain lipids in the first case than in the second, as previously discussed by Holm et al. (2002). Previous reports (Hauss et al., 1998; Holm et al., 2001) have described correlation between the
95
lymphatic triglyceride and drug transport, probably reflecting a partition between an aqueous phase (portal absorption) and a lipid phase (lymphatic absorption). The lower amount of halofantrine found in the lymph when the dogs were dosed with the MLM formulation was therefore probably derived from this correlation between the triglyceride and the drug transport in the intestinal lymphatics. Lymph flow was 539.76120.0 g for the animals dosed with MLM and 648.1637.3 g for the animals dosed with LML, representing a much less pronounced difference than the mass of triglyceride transported in the two situations. No significant difference could be found between the two groups, and the lymph flow was intermediate to the flow previously reported by Khoo et al. (2001) for animals in the pre- (2796230 g) and post-prandial (10476124 g) state. The lack of relationship between lymph flow and drug transport is consistent with previous reports, where the intestinal lymphatic transport was examined in lymph cannulated rats (Charman et al., 1986; Holm et al., 2001).
3.3. Lymph distribution of halofantrine Highly lipophilic compounds transported by the intestinal lymphatics are typically associated with the triglyceride core of the lipoproteins, primarily the chylomicrons (Pocock and Vost, 1974). The distribution of halofantrine to chylomicrons isolated from the thoracic lymph is shown in Table 1, for the first 6 h post-dosing, during which more than 85% of the total absorbed dose was absorbed. The major proportion of lymphatically transported halofantrine was associated with the chylomicron fraction, irrespective of the co-administered lipid, and the lower lymphatic transport of halofantrine in the MLM group did not affect the distribution of halofantrine in the chylomicron fraction. These observations were consistent with previous findings in rats in terms of extent of halofantine associated with the chylomicrons (Porter et al., 1996; Holm et al., 2002), but it is interesting to note that Khoo et al. (2001) reported an association greater than 95% when halofantrine was dosed with a large quantity of lipid (approximately 34 g fat). Tso et al. (1987) examined the lipid transporting capacity of chylomicrons and VLDLs after intraduodenal infusion of Table 1 Fractional distribution of halofantrine into the chylomicron fraction of thoracic lymph over the time course of the study (mean6S.E., n54–5) Lipid
Fig. 3. Cumulative mass (mg) of triglycerides (mean6S.E.) recovery in thoracic lymph after oral administration of SMEDDS formulations containing different structured triglycerides, MLM, (n55, j), and LML (n54, •).
0–1 h 1–2 h 2–3 h 3–4 h 4–5 h 5–6 h
Percentage distribution of halofantrine in chylomicrons MLM
LML
81.2620.3 75.665.2 62.2612.1 66.963.3 63.066.7 64.2616.9
94.066.1 79.263.2 77.564.8 75.0614.0 66.2612.4 76.8621.7
R. Holm et al. / European Journal of Pharmaceutical Sciences 20 (2003) 91–97
96
Fig. 4. Plasma concentration time profile of halofantrine (mean6S.E.) in lymph-cannulated animals after oral administration of 50 mg halofantrine dissolved in SMEDDS containing two different triglycerides to conscious lymph cannulated canines. MLM (n54, j) and LML (n54, •).
different lipid volumes and found that the lipid transported in VLDL was relatively constant whereas the lipid transported by the chylomicrons increased significantly during the infusion. The higher association of halofantrine with the chylomicron fraction found by Khoo et al. (2001) compared to the findings reported in this study likely reflects the larger amount of lipid dosed to the animals compared to the present study.
3.4. Absorption of halofantrine into the portal blood and total bioavailability in lymph-cannulated animals Halofantrine plasma concentration profiles in lymphcannulated animals are presented in Fig. 4. The AUC 0→24 h for MLM was 18906212 ng?h / ml and 12696212 ng?h / ml for LML. Statistically, it was not possible to separate the AUC 0→24 h data for the two formulations (P50.053), however, a trend towards elevated plasma concentrations was evident after administration of the MLM formulation, indicating that a higher content of medium-chain fatty acid components in the triglyceride has a beneficial effect on the portal absorption of halofantrine. Khoo et al. (2003)
did not find a similar positive effect from the mediumchain triglyceride components in a SMEDDS formulation dosed to lymph cannulated canines, suggesting that the effects seen in this study are somehow affected by the presence of both medium- and long-chain fatty acids and difficult to extrapolate to pure medium- or long-chain formulation systems. However, the findings presented in the present study were consistent with previous data, where halofantrine dissolved in both MLM and LML was orally dosed to conscious rats, and MLM showed a trend towards a higher plasma availability when compared to both LML and LLL (Holm et al., 2002). These consistencies could make the structured triglycerides an interesting choice for a lipid based formulation. Realising the problems with cross study comparisons, the total bioavailability of halofantrine for the lymphcannulated animals is shown in Table 2 compared to previous intravenous kinetics parameters published by Khoo et al. (2002). Although no statistical difference could be found between the two formulations or the SMEDDS formulation containing LLL (P50.064) reported by Khoo et al. (2003), there was a definite trend towards higher total bioavailability for the animals dosed with the MLM formulation. None of the examined formulations reached a lymphatic transport at the same level as after post-prandial administration (Khoo et al., 2001), but this was compensated for by a higher plasma availability, most pronounced for the animals dosed with the MLM. As the maximum oral bioavailability of halofantrine in dogs has previously been reported to be approximately 80% by Humberstone et al. (1996), the current data demonstrates the possibility of obtaining an oral bioavailability when dosed in SMEDDS at the same level as when dosed post-prandial, in the case of the MLM containing formulation. Holm et al. (2002) argued that the combination of medium- and long-chain fatty acids in the structured triglycerides combines the characteristic absorption profiles of the medium- and long-chain triglycerides, where the medium-chain fatty acid enhances the absorption into the systemic blood circulation and the long-chain fatty acid enhances the lymphatic transport. The results reported here are consistent with this hypothesis and show that the
Table 2 Halofantrine bioavailability (mean% dose6S.E., n54–5) in lymph-cannulated canines after oral administration of SMEDDS formulation containing structured triglycerides Lipid in SEDDS
Lymphatic transport a
Plasma availability b
Total bioavailability c
% of total bioavailability due to lymphatic transport
MLM LML
17.961.3 d 27.461.3 d
56.965.5 37.266.2
74.966.0 64.667.1
24.461.9 43.363.1
a
Percentage of halofantrine dose recovered over 12 h in the thoracic lymph. The percentage of halofantrine absorbed into the systemic circulation was calculated by dose normalisation of the AUC 0→24 h after oral administration relative to the AUC 0→24 h obtained after intravenous administration of 2 mg / kg halofantrine to lymph-cannulated canines (AUC was 69936330 ng?h / ml) published by Khoo et al. (2002a). c Total bioavailability was calculated as percentage transported by the lymph plus percentage absorbed directly into the blood. d Statistically significant difference between lymphatic transport in the two groups (P,0.05). b
R. Holm et al. / European Journal of Pharmaceutical Sciences 20 (2003) 91–97
pharmaceutical scientist can direct the drug into one of the two transport routes, by the choice of the excipient, and demonstrates that structured triglycerides can be used in SMEDDS to enhance the oral absorption of a lipophilic compound.
4. Conclusions In conclusion, this study has demonstrated that both the lymphatic transport and the absorption via the portal system of halofantrine was affected after administration of halofantrine in two different self-microemulsifying drug delivery systems based on a structured triglyceride, containing both medium- and long-chain fatty acids on the same glycerol backbone. The lipid content in the two formulations was sufficient to trigger a lymphatic transport of halofantrine, however, the MLM structure produced a smaller transport than the LML structure. This was nevertheless compensated for by an elevated portal absorption of halofantrine when dosed in the MLM vehicle, producing a trend towards a higher total bioavailability of the MLM SEDDS when compared to the LML SMEDDS. This indicates that the MLM structured lipid produces an absorption profile, which combines the characteristics from the medium- and long-chain triglyceride. The use of different structures of the co-administered triglyceride in a SMEDDS makes it possible for the pharmaceutical scientist to manipulate the relative contribution of the two absorption pathways within certain limits. It is thereby possible to enhance the lymphatic transport for a given compound where it contributes significantly without compromising the availability.
Acknowledgements This work was financially supported by the Danish Medical Research Council (Center for Drug Discovery and Transport) and the Augustinus Foundation. Majella Ryan is gratefully acknowledged for technical assistance during the sampling period and Dr. Shui-Mei Khoo for useful discussions regarding this work.
References Caliph, S., Charman, W.N., Porter, C.J.H., 2000. Effect of short-, medium- and long-chain fatty acid-based vehicles on the absolute oral bioavailability and intestinal lymphatic transport of halofantrine and assessment of mass balance in lymph-cannulated and non-cannulated rats. J. Pharm. Sci. 89, 1073–1084. Charman, S.A., Charman, W.N., Rogge, M.C., Wilson, T.D., Dutko, F.J., Pouton, C.W., 1992. Self-emulsifying drug delivery systems: formulation and biopharmaceutic evaluation of an investigational lipophilic compound. Pharm. Res. 9, 87–93. Charman, W.N., Noguchi, T., Stella, V.J., 1986. An experimental system
97
designed to study the in situ intestinal lymphatic transport of lipophilic drugs in anesthetized rats. Int. J. Pharm. 33, 155–164. Constantinides, P.P., 1995. Lipid microemulsions for improving drug dissolution and oral absorption: physical and biopharmaceutical aspects. Pharm. Res. 12, 1561–1572. Hauss, D.J., Fogal, S.E., Ficorilli, J.V., Price, C.A., Roy, T., Jayaraj, A.A., Keirns, J.J., 1998. Lipid based delivery systems for improving the bioavailability and lymphatic transport of a poorly water soluble LTB4 inhibitor. J. Pharm. Sci. 87, 164–169. ¨ Holm, R., Mullertz, A., Christensen, E., Høy, C.-E., Kristensen, H.G., 2001. Comparison of the lymphatic transport of halofantrine from 3 different unsaturated triglycerides in lymph-cannulated conscious rats. Eur. J. Pharm. Sci. 14, 331–337. ¨ Holm, R., Porter, C.J.H., Mullertz, A., Kristensen, H.G., Charman, W.N., 2002. Structured triglyceride vehicles for oral delivery of halofantrine: examination of intestinal lymphatic transport and bioavailability in conscious rats. Pharm. Res. 19, 1354–1361. Humberstone, A.J., Currie, G.J., Porter, C.J.H., Scanlon, M.J., Charman, W.N., 1995. A simplified liquid chromatography assay for the quantitation of halofantrine and desbutylhalofantrine in plasma and identification of a degradation product of desbutylhalofantrine formed under alkaline conditions. J. Pharm. Biomed. Anal. 13, 265–272. Humberstone, A.J., Porter, C.J.H., Charman, W.N., 1996. A physicochemical basis for the effect of food on the absolute oral bioavailability of halofantrine. J. Pharm. Sci. 85, 525–529. Humberstone, A.J., Charman, W.N., 1997. Lipid-based vehicles for the oral delivery of poorly water soluble drugs. Adv. Drug Del. Rev. 15, 103–128. Khoo, S.-M., Humberstone, A.J., Porter, C.J.H., Edwards, G.A., Charman, W.N., 1998. Formulation design and bioavailability assessment of lipidic self-emulsifying formulations of halofantrin. Int. J. Pharm. 167, 155–164. Khoo, S.-M., Edwards, G.A., Porter, C.J.H., Charman, W.N., 2001. A conscious dog model for assessing the absorption, enterocyte-based metabolism, and intestinal lymphatic transport of halofantrine. J. Pharm. Sci. 91, 1599–1607. Khoo, S.-M., Prankerd, R.J., Edwards, G.A., Porter, C.J.H., Charman, W.N., 2002. A physicochemical basis for the extensive intestinal lymphatic transport of a poorly lipid soluble antimalarial, halofantrine hydrochloride, after postprandial administration to dogs. J. Pharm. Sci. 91, 647–659. Khoo, S.-M., Shackleford, D., Porter, C.J.H., Edwards, G.A., Charman, W.N., 2003. Intestinal lymphatic transport of halofantrine occurs after oral administration of a unit dose lipid-based formulation to fasted dogs. Pharm. Res., in press. Kiyasu, J.Y., Bloom, B., Chaikoff, I.L., 1952. The portal transport of absorbed fatty acids. J. Biol. Chem. 199, 415–419. Mu, H., Xu, X., Høy, C.-E., 1998. Production of specific structured triacylglycerols by lipase-catalyzed interesterification in a laboratory scale continuous reactor. J. Am. Oil. Chem. Soc. 75, 1187–1193. Pocock, D.M.E., Vost, A., 1974. DDT absorption and chylomicron transport in rat. Lipids 9, 374–381. Porter, C.J.H., Charman, S.A., Charman, W.N., 1996. Lymphatic transport of halofantrine in the triple-cannulated anesthetized rat model: Effect of lipid vehicle dispersion. J. Pharm. Sci. 85, 351–356. Pouton, C.W., 2000. Lipid formulations for oral administration of drugs: non-emulsifying, self-emulsifying and ‘self-microemulsifying’ drug delivery systems. Eur. J. Pharm. Sci. 11, S93–S98. Raub, T.J., Douglas, S.L., Melchior, G.W., Charman, W.N., Morozowich, W., 1992. Methodologies for assessing intestinal lymphatic transport. In: Charman, W.N., Stella, V.J. (Eds.), Lymphatic Transport of Drugs. CRC Press, Boca Raton, pp. 113–179. ¨ ¨ Tso, P., Lindstrom, M.B., Borgstrom, B., 1987. Factors regulating the formation of chylomicrons and very-low-density lipoproteins by the sat small intestine. Biochim. Biophys. Acta 922, 304–313.
Examination of oral absorption and lymphatic transport of halofantrine in a triple-cannulated canine model after administration in self-microemulsifying drug delivery systems (SMEDDS) containing structured triglycerides a, *, ¨ Rene´ Holm a ,1 , Christopher J.H. Porter b , Glenn A. Edwards c , Anette Mullertz a b Henning G. Kristensen , William N. Charman a
b
Department of Pharmaceutics, The Royal Danish School of Pharmacy, Universitetsparken 2, 2100 Copenhagen, Denmark Department of Pharmaceutics, Victorian College of Pharmacy, Monash University, 381 Royal Parade, Victoria 3052, Australia c Department of Veterinary Science, University of Melbourne, 250 Princes Highway, Werribee, Victoria 3030, Australia Received 19 February 2003; received in revised form 13 June 2003; accepted 23 June 2003
Abstract The potential for lipidic self-microemulsifying drug delivery systems (SMEDDS) containing triglycerides with a defined structure, where the different fatty acids on the glycerol backbone exhibit different metabolic fate, to improve the lymphatic transport and the portal absorption of a poorly water-soluble drug, halofantrine, were investigated in fasted lymph cannulated canines. Two different structured triglycerides were incorporated into the SMEDDS; 1,3-dioctanoyl-2-linoleyl-sn-glycerol (C8:0-C18:2-C8:0) (MLM) and 1,3-dilinoyl-2octanoyl-sn-glycerol (C18:2-C8:0-C18:2) (LML). A previously optimised SMEDDS formulation for halofantrine, comprising of triglyceride, Cremophor EL, Maisine 35-1 and ethanol was selected for bioavailability assessment. The extent of lymphatic transport via the thoracic duct was 17.9% of the dose for the animals dosed with the MLM SMEDDS and 27.4% for LML. Also the plasma availability was affected by the triglyceride incorporated into the multi-component delivery system and availabilities of 56.9% (MLM) and 37.2% (LML) were found. These data indicate that the pharmaceutical scientist can use the structure of the lipid to affect the relative contribution of the two absorption pathways. The MLM formulation produced a total bioavailability of 74.9%, which is higher than the total absorption previously observed after post-prandial administration. This could indicate the utility of disperse lipid-base formulations based on structured triglycerides for the oral delivery of halofantrine, and potentially other lipophilic drugs. 2003 Elsevier B.V. All rights reserved. Keywords: Halofantrine; Structured triglyceride; Lymphatic transport; Bioavailability; Canine; SMEDDS
1. Introduction The oral absorption of a number of potent lipophilic compounds is limited by their physico-chemical properties and has shown low and variable availability. This has led to a growing effort to develop pharmaceutical formulations enhancing the oral bioavailability of these lipophilic compounds. An increasingly popular approach to overcome the low oral bioavailability is the incorporation of the active lipophilic component into lipid vehicles such as oil solutions, self-emulsifying drug delivery systems (SEDDS) (Humberstone and Charman, 1997; Pouton, 2000). The application of lipid based delivery systems has *Corresponding author. Tel.: 145-35-30-6440; fax: 145-35-30-6031. ¨ E-mail address: [email protected] (A. Mullertz). 1 Present address: H. Lundbeck A / S, Ottilavej 9, 2500 Valby, Denmark. 0928-0987 / 03 / $ – see front matter 2003 Elsevier B.V. All rights reserved. doi:10.1016 / S0928-0987(03)00174-X
gained considerable interest after the commercial success of lipid based formulations of cyclosporine A (Sandimmun Neoral姠), saquinavir (Fortovase) and ritonavir (Norvir). SEDDS are isotropic mixtures of oil, a surfactant, and possibly one or more hydrophilic solvents or co-surfactants, which form fine oil-in-water emulsions or microemulsions (SMEDDS) when exposed to aqueous media under conditions of gentle agitation (Constantinides, 1995). SEDDS typically produce emulsions with a particle size between 100 and 300 nm, while SMEDDS form transparent microemulsions with a particle size of less than 100 nm. These properties have been suggested to make SEDDS and SMEDDS a good formulation alternative for the oral delivery of lipophilic drugs (Charman et al., 1992). Khoo et al. (1998) have pursued the potential for SEDDS to improve the oral bioavailability of halofantrine,
92
R. Holm et al. / European Journal of Pharmaceutical Sciences 20 (2003) 91–97
a highly lipophilic antimalaria agent, in fasted beagle dogs, and showed a trend towards a higher bioavailability when dosed in SMEDDS based on long-chain triglyceride. The lymphatic transport of halofantrine has been found to play a major role in the absorption from the intestine when co-administered with a lipid formulation in rats (Porter et al., 1996; Caliph et al., 2000). Recently Khoo et al. (2003) showed that halofantrine was also transported by the lymphatic systems in dogs when dosed in a human size long-chain triglyceride based SMEDDS formulation described previously (Khoo et al., 1998). Generally, compounds processed by the intestinal lymph are transported to the systemic circulation in association with the lipid core of lipoproteins (Pocock and Vost, 1974), and as such require co-administered lipid to stimulate lipoprotein formation. Short- and medium-chain fatty acids (with a carbon chain length shorter than 12 carbon atoms) are transported to the systemic circulation by the portal blood and are not incorporated to a great extent in chylomicrons (Kiyasu et al., 1952). In contrast, long-chain fatty acids and monoglycerides are re-esterified to triglycerides within the intestinal cell, incorporated into chylomicrons and secreted from the intestinal cell by exocytosis into the lymph vessels. In addition to the stimulation of lymphatic transport, administration of lipophilic drugs with lipids may enhance drug absorption into the portal blood when compared to nonlipid formulations (Caliph et al., 2000). In a recent study Holm et al. (2002) demonstrated that the intestinal lymphatic transport and the portal absorption of halofantrine when co-administered with a structured triglyceride (a triglyceride where the glycerol backbone is esterified with different fatty acid chains (Fig. 1)), was affected compared to a natural long chain triglyceride, sunflower oil, in lymph cannulated rats. However, there is limited additional information in the literature detailing the positional benefits in terms of drug delivery of these unique triglyceride vehicles. Furthermore, it is not known if these triglycerides can be incorporated into a pharmaceutical formulation without losing their beneficial properties. Moreover, it also remains unexplored how the absorption paten of a lipophilic drug would be affected in a non-rodent animal species with one of these structured triglycerides. The objective of this study was therefore to
Fig. 1. Schematic presentation of the structured lipids used in this study. M represents a medium-chain fatty acid (C 8 – 10 ), and L represents a long chain fatty acid (C 18 ).
investigate the impact of structured triglycerides with varying intramolecular structures and chain lengths incorporated into a pharmaceutical formulation, on the intestinal lymphatic transport and absorption into the blood of halofantrine, in lymph cannulated conscious canines.
2. Materials and methods
2.1. Chemicals and reagents Crystalline halofantrine free base and the internal standard 2,4-dichloro-6-trifluoromethyl-9h1-[2-dibutylamino)ethyl]jphenathrenemethanol hydrochloride were donated by SmithKline Beecham Pharmaceuticals (Mysore, India). The structured triglycerides, 1,3-dioctanoyl-2-linoleyl-snglycerol (C8:0-C18:2-C8:0) (MLM) and 1,3-dilinoyl-2octanoyl-sn-glycerol (C18:2-C8:0-C18:2) (LML), were manufactured at the Department of Biotechnology at the Technical University of Denmark as previously described (Mu et al., 1998). The fatty acid compositions of the structured triglycerides were previously reported by Holm et al. (2002). Cremophor EL was donated by BASF ´ (Ludwigshafen, Germany) and Maisine 35-1 by Gattefosse (Saint-Priest, France). R.P. Scherer (Australia) kindly provided 1 ml air-filled gelatine capsules. Acetylpromazine maleate (Delvet Pty. Ltd., Australia), propofol (ScheringPlough, Australia), carprofen (Pfizer, Australia), cephazolin (Sigma Pharmaceuticals, Australia) and polydioxinone sutures (Ethicon, USA) were used during the surgery. Normal saline (0.9%) and Hartmans solution were obtained from Baxter Healthcare (Australia). Acetonitrile and tert-butylmethylether were HPLC grade and sodium dodecyl sulphate was electrophoresis grade. The water used in all experiments was obtained from a Milli-Q-water purification system (Milipore, Milford, MA). All other chemicals were analytical reagent grade.
2.2. Halofantrine formulations The composition of the formulations was based on a previous published SMEDDS which was developed for halofantrine by Khoo et al. (1998). The long-chain triglyceride used by Khoo and co-workers was, however, substituted with the structured triglycerides. The SMEDDS formulations consisted of halofantrine: triglyceride: Maisine-35-1: Cremophor EL: absolute ethanol (5:29:29:30:7) (% w / w). Approximately 4 g of each formulation was prepared the day before use, by weighing halofantrine base into a 12 ml teflon-lined screw-capped glass conical tube, followed by addition of Cremophor EL, Maisine 35-1 and the structured triglycerides (MLM or LML). The components were mixed by gentle stirring, and heated in a 50 8C water bath, until all the halofantrine had dissolved. The mixture was then cooled to ambient temperature and the ethanol added. One gram of the formula-
R. Holm et al. / European Journal of Pharmaceutical Sciences 20 (2003) 91–97
tion was filled into a soft gelatine capsule using a syringe and a needle. A disintegration experiment was performed each day to assess the efficacy of the self-emulsification using a standard dissolution apparatus (Erweka, Germany) with the paddle set at 50 rpm at 37 8C. The droplet size of the emulsions was determined by photo correlation spectroscopy using a Zetasizer 3000 (Malvern Instruments, UK). Both formulations produced a clear slightly bluish microemulsion with a mean particle size of 47.9 nm (MLM microemulsion) and 50.3 nm (LML microemulsion). The concentration of halofantrine in the formulations was determined prior to administration by high-performance liquid chromatography (HPLC), as described by Humberstone et al. (1995), thereby determining the exact dose of halofantrine administered.
2.3. Surgical procedures
93
1, 1.5, 2, 2.5, 3, 3.5, 4, 5, 6, 7, 8, 10, 12 and 24 h after the drug administration. Blood samples were collected into individual tubes containing dipotassium EDTA. The plasma was harvested immediately by centrifugation for 10 min (1000 g) and stored at 220 8C until further analysis. Lymph was collected continuously into 50 ml tubes containing 75 mg dipotassium EDTA, and collection tubes were changed hourly. The lymph was collected for 12 h post-dosing. The amount of lymph collected in each interval was determined gravimetrically. Lymph samples were stored at 5–8 8C prior to analysis (within 16 h). At the conclusion of the experiment the animals were sacrificed by an overdose of sodium pentobarbitone given intravenously.
2.5. Analysis of halofantrine in blood and lymph
All surgical and experimental procedures were reviewed and approved by the local Institutional Animal Experimentation Ethics Committee (University of Melbourne, Australia). Studies were conducted in male greyhound dogs (28–35 kg), and their health status was verified by a veterinarian prior to the study. The surgery was performed as described by Khoo et al. (2001). Each dog was fed a small lipid meal prior to surgery to facilitate subsequent identification of the thoracic lymph duct. The dogs received a pre-anaesthetic by subcutaneous injection of acetylpromazine maleate (0.5 mg / kg) and were subsequently anaesthetised with an intravenous injection of propofol (3–6 mg / kg). The surgical anaesthesia was maintained by delivery of halothane and oxygen. The dogs received intravenous infusion of saline during surgery, as well as post-operative injection of antibiotic (cephazolin, 20 mg / kg) and analgesic (carprofen 4 mg / kg). Following surgery, the dogs were allowed to recover unrestrained in a closed run for 15–19 h. In the initial period of recovery, fluids were administered intravenously to ensure adequate hydration and to prevent hypoproteinaemia. The dogs were allowed to return to normal ambulatory movements prior to drug administration. An intravenous catheter was inserted into the cephalic vein immediately prior to dosing to facilitate sampling of systemic blood during the study.
For analysis of halofantrine in the lymph, 100 ml of lymph was added to 10 ml acetonitrile and the sample was vortexed for 1 min. Insoluble protein-based components were removed by centrifugation (1000 g) and the supernatant analysed by HPLC as previously described (Porter et al., 1996). Recovery of spiked halofantrine from blank lymph was greater than 99%. As all the lymph draining from the thoracic lymph duct was collected, the mass of drug transported was calculated by multiplying the lymph drug concentration by the corresponding mass of lymph from each collection period. Blood samples were analysed using a validated HPLC method (Humberstone et al., 1995), by adding 200 ml plasma and 100 ml internal standard (2 mg / ml) to 1.0 ml acetonitrile in a 12 ml polypropylene centrifuge tube. Samples were subsequently vortexed for 2 min to precipitate plasma proteins then centrifuged (800 g), and 8 ml aliquot of tert-butyl methyl ether was added. The tube was vortexed and centrifuged (1000 g) and 8 ml of the upper organic layer was removed into a polypropylene centrifuge tube containing 100 ml of 0.005 M HCl (in acetonitrile). The organic phase was then evaporated to dryness under nitrogen at 35 8C. The residue was reconstituted with 100 ml acetonitrile and 25 ml was injected on to the LC column. The limit of quantification by this procedure was 20 ng / ml. The assay was linear between 20 ng / ml and 4000 ng / ml, and the extraction efficiency for halofantrine was greater than 90% across the concentration range.
2.4. Experimental procedures
2.6. Analysis of lymph triglyceride
The fasted dogs were orally administered a single soft gelatine capsule containing 1 g of the SMEDDS formulation (and 50 mg halofantrine) with 50 ml of water. To prevent dehydration, 25 ml of Hartmann’s solution was administered at hourly intervals by intravenous bolus injections. Water was available ad libitum, however, food was withheld until after collection of the 12 h sample. Blood samples (2.5 ml) were taken at 25 min and 0.5,
The triglyceride concentration in lymph was measured using automated clinical chemistry analyser (Roche Cobas Mira, Basle, Switzerland) and an enzyme-based colorimetric assay kit (Boehringer Mannheim, Germany).
2.7. Lymph distribution of halofantrine The lymph collected for the first 6 h was separated as
94
R. Holm et al. / European Journal of Pharmaceutical Sciences 20 (2003) 91–97
described by Raub et al. (1992), using a Beckman SW-60 rotor. The chylomicrons were firstly separated by layering 1 ml of lymph under a sodium chloride solution (d5 1.0063 g / ml) followed by centrifugation at 44 100 rpm for 1 h and 20 min (262 000 g) at 15 8C. After centrifugation, the bottom of the tube was pierced using a needle to enable the remaining lymph to be removed, leaving only the chylomicron fraction which forms a white semi-solid plug at the top of the tube. After fractionation, the chylomicron fraction was dissolved in 10 ml acetonitrile and the proportion of halofantrine associated with chylomicrons determined by HPLC.
2.8. Pharmacokinetic analysis Plasma concentrations versus time data for halofantrine for individual dogs were analysed by standard non-compartment analysis using WinNonlin software (version 2.1). The total bioavailability was calculated by adding the cumulative amount of halofantrine transported in the lymph with the percent of halofantrine found in the systemic circulation. An estimate of the percent absorption of halofantrine into the systemic circulation was determined by comparison of the dose normality obtained in lymph cannulated canines after oral administration in the SMEDDS formulation with the AUC after intravenous administration from previous data by Khoo et al. (2002).
2.9. Statistical analysis Statistical analysis was performed by one-way analysis of variance (SigmaStat for Windows version 2.0). The Student–Newman–Keuls multiple comparison test was subsequently applied to analyse potential difference between the formulations. The results were considered significant if P,0.05.
on the absorption of the highly lipophilic antimalarial halofantrine.
3.1. Intestinal lymphatic transport of halofantrine The lymphatic transport of halofantrine, expressed as the cumulative percentage of the administered dose, is presented in Fig. 2. The extent of lymphatic transport of halofantrine after 12 h (mean% dose6S.E.) was statistically different and was 27.461.3 (n54) after administration in the LML and 17.961.3 (n55) in the MLM. Khoo et al. (2003) reported a lymphatic transport of halofantrine in a similar formulation containing a natural long-chain triglyceride (soybean oil) on 28.367.6 (n54), findings that are equal to the lymphatic transport found in the LML, demonstrating that this structural formation of the triglyceride initiates a lymphatic transfer at a high level Both examined SMEDDS formulations in this study introduced a lymphatic transport higher than the transport reported when dosed in a SMEDDS formulated with medium-chain triglycerides in lymph cannulated canines (Khoo et al., 2003). These data indicate that the small amount of lipid dosed in a soft gelatine capsule (here 580 mg of absorbable lipid) was sufficient to trigger the biochemical processes and initiate lymphatic transport of halofantrine, consistent with data previously reported by Khoo et al. (2003), also when structural triglycerides are used as a vehicle. The canines were dosed with a lipid meal before the surgery to facilitate the identification and cannulation of the thoracic lymph duct, which potentially could influence the content of halofantrine found in the lymph. However, Khoo et al. (2001) previously reported a lymphatic transport of halofantrine, in a study with a similar experimental setup as the one used in the present study, of 1.360.2% in
3. Results and discussion Rats are often used to investigate the intestinal lymphatic transport of lipophilic drugs and for examining the effect of lipidic formulations on the lymphatic transport and bioavailability due to their low cost and relatively facile accessibility. The conduction in rats is limited by the volume of formulation that can be administered to each animal and their continuous bile flow. Unlike the rat, dogs can be administered formulations similar to the formulations intended for use in humans. In a previous study Khoo et al. (2003) have demonstrated that the relative small amount of lipid contained in a SMEDDS formulation can significantly increase lymphatic transport of halofantrine. In this study we have therefore used lymph cannulated canines to evaluate the impact of a structured triglyceride incorporated into a pharmaceutical SMEDDS formulation
Fig. 2. Cumulative percentage dose of halofantrine (mean6S.E.) collected in the thoracic lymph as a function of time after oral administration of 50 mg halofantrine contained in SMEDDS formulation containing two different structured triglycerides to conscious canines. MLM (n55, j) and LML (n54, •).
R. Holm et al. / European Journal of Pharmaceutical Sciences 20 (2003) 91–97
fasted animals. This transport could be considered as the basal lymphatic transport initiated by the endogenous lipid turnover as well as the potential interaction of the food dosed before the surgery. With a very limited lymphatic transport found in the base situation the potential influence of food dosed before surgery was not considered to have an impact on the results in this study with the used experimental setting.
3.2. Lymphatic transport of triglycerides and lymph flow The cumulative transport of C 18 triglyceride into the thoracic lymph is presented in Fig. 3. Lipid collected in the thoracic lymph 12 h after dosing (expressed as the mean6S.E., n54–5) was 5.261.2 g for animals dosed with LML and 2.660.7 g for animals dosed with MLM. Statistically, the two groups are not significantly different, due to large variation, but the mean amount of triglyceride found in the lymph suggests a difference as a function of the dosed lipid. In fasted dogs, triglyceride transport has previously been reported by Khoo et al. (2001) at 0.560.2 g (mean6S.E.), which reflected basal transport levels arising from endogenously derived lipid. Both SMEDDS formulations examined in this study had a higher mean triglyceride transport confirming that the available lipid in the SMEDDS is sufficient to trigger the biochemical processes involved in the lymphatic transport of lipids. Previous data suggest that the lymphatic transport of medium chain lipids is small (Kiyasu et al., 1952). MLM was expected to have a lower lymphatic triglyceride transport when compared to the LML formulation, because of the lower molar amount of long chain lipids in the first case than in the second, as previously discussed by Holm et al. (2002). Previous reports (Hauss et al., 1998; Holm et al., 2001) have described correlation between the
95
lymphatic triglyceride and drug transport, probably reflecting a partition between an aqueous phase (portal absorption) and a lipid phase (lymphatic absorption). The lower amount of halofantrine found in the lymph when the dogs were dosed with the MLM formulation was therefore probably derived from this correlation between the triglyceride and the drug transport in the intestinal lymphatics. Lymph flow was 539.76120.0 g for the animals dosed with MLM and 648.1637.3 g for the animals dosed with LML, representing a much less pronounced difference than the mass of triglyceride transported in the two situations. No significant difference could be found between the two groups, and the lymph flow was intermediate to the flow previously reported by Khoo et al. (2001) for animals in the pre- (2796230 g) and post-prandial (10476124 g) state. The lack of relationship between lymph flow and drug transport is consistent with previous reports, where the intestinal lymphatic transport was examined in lymph cannulated rats (Charman et al., 1986; Holm et al., 2001).
3.3. Lymph distribution of halofantrine Highly lipophilic compounds transported by the intestinal lymphatics are typically associated with the triglyceride core of the lipoproteins, primarily the chylomicrons (Pocock and Vost, 1974). The distribution of halofantrine to chylomicrons isolated from the thoracic lymph is shown in Table 1, for the first 6 h post-dosing, during which more than 85% of the total absorbed dose was absorbed. The major proportion of lymphatically transported halofantrine was associated with the chylomicron fraction, irrespective of the co-administered lipid, and the lower lymphatic transport of halofantrine in the MLM group did not affect the distribution of halofantrine in the chylomicron fraction. These observations were consistent with previous findings in rats in terms of extent of halofantine associated with the chylomicrons (Porter et al., 1996; Holm et al., 2002), but it is interesting to note that Khoo et al. (2001) reported an association greater than 95% when halofantrine was dosed with a large quantity of lipid (approximately 34 g fat). Tso et al. (1987) examined the lipid transporting capacity of chylomicrons and VLDLs after intraduodenal infusion of Table 1 Fractional distribution of halofantrine into the chylomicron fraction of thoracic lymph over the time course of the study (mean6S.E., n54–5) Lipid
Fig. 3. Cumulative mass (mg) of triglycerides (mean6S.E.) recovery in thoracic lymph after oral administration of SMEDDS formulations containing different structured triglycerides, MLM, (n55, j), and LML (n54, •).
0–1 h 1–2 h 2–3 h 3–4 h 4–5 h 5–6 h
Percentage distribution of halofantrine in chylomicrons MLM
LML
81.2620.3 75.665.2 62.2612.1 66.963.3 63.066.7 64.2616.9
94.066.1 79.263.2 77.564.8 75.0614.0 66.2612.4 76.8621.7
R. Holm et al. / European Journal of Pharmaceutical Sciences 20 (2003) 91–97
96
Fig. 4. Plasma concentration time profile of halofantrine (mean6S.E.) in lymph-cannulated animals after oral administration of 50 mg halofantrine dissolved in SMEDDS containing two different triglycerides to conscious lymph cannulated canines. MLM (n54, j) and LML (n54, •).
different lipid volumes and found that the lipid transported in VLDL was relatively constant whereas the lipid transported by the chylomicrons increased significantly during the infusion. The higher association of halofantrine with the chylomicron fraction found by Khoo et al. (2001) compared to the findings reported in this study likely reflects the larger amount of lipid dosed to the animals compared to the present study.
3.4. Absorption of halofantrine into the portal blood and total bioavailability in lymph-cannulated animals Halofantrine plasma concentration profiles in lymphcannulated animals are presented in Fig. 4. The AUC 0→24 h for MLM was 18906212 ng?h / ml and 12696212 ng?h / ml for LML. Statistically, it was not possible to separate the AUC 0→24 h data for the two formulations (P50.053), however, a trend towards elevated plasma concentrations was evident after administration of the MLM formulation, indicating that a higher content of medium-chain fatty acid components in the triglyceride has a beneficial effect on the portal absorption of halofantrine. Khoo et al. (2003)
did not find a similar positive effect from the mediumchain triglyceride components in a SMEDDS formulation dosed to lymph cannulated canines, suggesting that the effects seen in this study are somehow affected by the presence of both medium- and long-chain fatty acids and difficult to extrapolate to pure medium- or long-chain formulation systems. However, the findings presented in the present study were consistent with previous data, where halofantrine dissolved in both MLM and LML was orally dosed to conscious rats, and MLM showed a trend towards a higher plasma availability when compared to both LML and LLL (Holm et al., 2002). These consistencies could make the structured triglycerides an interesting choice for a lipid based formulation. Realising the problems with cross study comparisons, the total bioavailability of halofantrine for the lymphcannulated animals is shown in Table 2 compared to previous intravenous kinetics parameters published by Khoo et al. (2002). Although no statistical difference could be found between the two formulations or the SMEDDS formulation containing LLL (P50.064) reported by Khoo et al. (2003), there was a definite trend towards higher total bioavailability for the animals dosed with the MLM formulation. None of the examined formulations reached a lymphatic transport at the same level as after post-prandial administration (Khoo et al., 2001), but this was compensated for by a higher plasma availability, most pronounced for the animals dosed with the MLM. As the maximum oral bioavailability of halofantrine in dogs has previously been reported to be approximately 80% by Humberstone et al. (1996), the current data demonstrates the possibility of obtaining an oral bioavailability when dosed in SMEDDS at the same level as when dosed post-prandial, in the case of the MLM containing formulation. Holm et al. (2002) argued that the combination of medium- and long-chain fatty acids in the structured triglycerides combines the characteristic absorption profiles of the medium- and long-chain triglycerides, where the medium-chain fatty acid enhances the absorption into the systemic blood circulation and the long-chain fatty acid enhances the lymphatic transport. The results reported here are consistent with this hypothesis and show that the
Table 2 Halofantrine bioavailability (mean% dose6S.E., n54–5) in lymph-cannulated canines after oral administration of SMEDDS formulation containing structured triglycerides Lipid in SEDDS
Lymphatic transport a
Plasma availability b
Total bioavailability c
% of total bioavailability due to lymphatic transport
MLM LML
17.961.3 d 27.461.3 d
56.965.5 37.266.2
74.966.0 64.667.1
24.461.9 43.363.1
a
Percentage of halofantrine dose recovered over 12 h in the thoracic lymph. The percentage of halofantrine absorbed into the systemic circulation was calculated by dose normalisation of the AUC 0→24 h after oral administration relative to the AUC 0→24 h obtained after intravenous administration of 2 mg / kg halofantrine to lymph-cannulated canines (AUC was 69936330 ng?h / ml) published by Khoo et al. (2002a). c Total bioavailability was calculated as percentage transported by the lymph plus percentage absorbed directly into the blood. d Statistically significant difference between lymphatic transport in the two groups (P,0.05). b
R. Holm et al. / European Journal of Pharmaceutical Sciences 20 (2003) 91–97
pharmaceutical scientist can direct the drug into one of the two transport routes, by the choice of the excipient, and demonstrates that structured triglycerides can be used in SMEDDS to enhance the oral absorption of a lipophilic compound.
4. Conclusions In conclusion, this study has demonstrated that both the lymphatic transport and the absorption via the portal system of halofantrine was affected after administration of halofantrine in two different self-microemulsifying drug delivery systems based on a structured triglyceride, containing both medium- and long-chain fatty acids on the same glycerol backbone. The lipid content in the two formulations was sufficient to trigger a lymphatic transport of halofantrine, however, the MLM structure produced a smaller transport than the LML structure. This was nevertheless compensated for by an elevated portal absorption of halofantrine when dosed in the MLM vehicle, producing a trend towards a higher total bioavailability of the MLM SEDDS when compared to the LML SMEDDS. This indicates that the MLM structured lipid produces an absorption profile, which combines the characteristics from the medium- and long-chain triglyceride. The use of different structures of the co-administered triglyceride in a SMEDDS makes it possible for the pharmaceutical scientist to manipulate the relative contribution of the two absorption pathways within certain limits. It is thereby possible to enhance the lymphatic transport for a given compound where it contributes significantly without compromising the availability.
Acknowledgements This work was financially supported by the Danish Medical Research Council (Center for Drug Discovery and Transport) and the Augustinus Foundation. Majella Ryan is gratefully acknowledged for technical assistance during the sampling period and Dr. Shui-Mei Khoo for useful discussions regarding this work.
References Caliph, S., Charman, W.N., Porter, C.J.H., 2000. Effect of short-, medium- and long-chain fatty acid-based vehicles on the absolute oral bioavailability and intestinal lymphatic transport of halofantrine and assessment of mass balance in lymph-cannulated and non-cannulated rats. J. Pharm. Sci. 89, 1073–1084. Charman, S.A., Charman, W.N., Rogge, M.C., Wilson, T.D., Dutko, F.J., Pouton, C.W., 1992. Self-emulsifying drug delivery systems: formulation and biopharmaceutic evaluation of an investigational lipophilic compound. Pharm. Res. 9, 87–93. Charman, W.N., Noguchi, T., Stella, V.J., 1986. An experimental system
97
designed to study the in situ intestinal lymphatic transport of lipophilic drugs in anesthetized rats. Int. J. Pharm. 33, 155–164. Constantinides, P.P., 1995. Lipid microemulsions for improving drug dissolution and oral absorption: physical and biopharmaceutical aspects. Pharm. Res. 12, 1561–1572. Hauss, D.J., Fogal, S.E., Ficorilli, J.V., Price, C.A., Roy, T., Jayaraj, A.A., Keirns, J.J., 1998. Lipid based delivery systems for improving the bioavailability and lymphatic transport of a poorly water soluble LTB4 inhibitor. J. Pharm. Sci. 87, 164–169. ¨ Holm, R., Mullertz, A., Christensen, E., Høy, C.-E., Kristensen, H.G., 2001. Comparison of the lymphatic transport of halofantrine from 3 different unsaturated triglycerides in lymph-cannulated conscious rats. Eur. J. Pharm. Sci. 14, 331–337. ¨ Holm, R., Porter, C.J.H., Mullertz, A., Kristensen, H.G., Charman, W.N., 2002. Structured triglyceride vehicles for oral delivery of halofantrine: examination of intestinal lymphatic transport and bioavailability in conscious rats. Pharm. Res. 19, 1354–1361. Humberstone, A.J., Currie, G.J., Porter, C.J.H., Scanlon, M.J., Charman, W.N., 1995. A simplified liquid chromatography assay for the quantitation of halofantrine and desbutylhalofantrine in plasma and identification of a degradation product of desbutylhalofantrine formed under alkaline conditions. J. Pharm. Biomed. Anal. 13, 265–272. Humberstone, A.J., Porter, C.J.H., Charman, W.N., 1996. A physicochemical basis for the effect of food on the absolute oral bioavailability of halofantrine. J. Pharm. Sci. 85, 525–529. Humberstone, A.J., Charman, W.N., 1997. Lipid-based vehicles for the oral delivery of poorly water soluble drugs. Adv. Drug Del. Rev. 15, 103–128. Khoo, S.-M., Humberstone, A.J., Porter, C.J.H., Edwards, G.A., Charman, W.N., 1998. Formulation design and bioavailability assessment of lipidic self-emulsifying formulations of halofantrin. Int. J. Pharm. 167, 155–164. Khoo, S.-M., Edwards, G.A., Porter, C.J.H., Charman, W.N., 2001. A conscious dog model for assessing the absorption, enterocyte-based metabolism, and intestinal lymphatic transport of halofantrine. J. Pharm. Sci. 91, 1599–1607. Khoo, S.-M., Prankerd, R.J., Edwards, G.A., Porter, C.J.H., Charman, W.N., 2002. A physicochemical basis for the extensive intestinal lymphatic transport of a poorly lipid soluble antimalarial, halofantrine hydrochloride, after postprandial administration to dogs. J. Pharm. Sci. 91, 647–659. Khoo, S.-M., Shackleford, D., Porter, C.J.H., Edwards, G.A., Charman, W.N., 2003. Intestinal lymphatic transport of halofantrine occurs after oral administration of a unit dose lipid-based formulation to fasted dogs. Pharm. Res., in press. Kiyasu, J.Y., Bloom, B., Chaikoff, I.L., 1952. The portal transport of absorbed fatty acids. J. Biol. Chem. 199, 415–419. Mu, H., Xu, X., Høy, C.-E., 1998. Production of specific structured triacylglycerols by lipase-catalyzed interesterification in a laboratory scale continuous reactor. J. Am. Oil. Chem. Soc. 75, 1187–1193. Pocock, D.M.E., Vost, A., 1974. DDT absorption and chylomicron transport in rat. Lipids 9, 374–381. Porter, C.J.H., Charman, S.A., Charman, W.N., 1996. Lymphatic transport of halofantrine in the triple-cannulated anesthetized rat model: Effect of lipid vehicle dispersion. J. Pharm. Sci. 85, 351–356. Pouton, C.W., 2000. Lipid formulations for oral administration of drugs: non-emulsifying, self-emulsifying and ‘self-microemulsifying’ drug delivery systems. Eur. J. Pharm. Sci. 11, S93–S98. Raub, T.J., Douglas, S.L., Melchior, G.W., Charman, W.N., Morozowich, W., 1992. Methodologies for assessing intestinal lymphatic transport. In: Charman, W.N., Stella, V.J. (Eds.), Lymphatic Transport of Drugs. CRC Press, Boca Raton, pp. 113–179. ¨ ¨ Tso, P., Lindstrom, M.B., Borgstrom, B., 1987. Factors regulating the formation of chylomicrons and very-low-density lipoproteins by the sat small intestine. Biochim. Biophys. Acta 922, 304–313.