European Journal of Pharmaceutical Sciences 14 (2001) 331–337www.elsevier.nl / locate /ejps

Comparison of total oral bioavailability and the lymphatic transport ofhalofantrine from three different unsaturated triglycerides in

lymph-cannulated conscious ratsa a , b b a*´ ¨Rene Holm , Anette Mullertz , Egon Christensen , Carl-Erik Høy , Henning G. Kristensen

aDepartment of Pharmaceutics, The Royal Danish School of Pharmacy, Universitetsparken 2, DK-2100 Copenhagen, DenmarkbDepartment of Biochemistry and Nutrition, Technical University of Denmark, Anker Engelundsvej, building 224, DK-2800 Lyngby, Denmark

Received 16 March 2001; received in revised form 9 July 2001; accepted 14 August 2001

Abstract

The lymphatic transport and the portal absorption of the lipophilic drug halofantrine were investigated in a conscious rat model. Therats were dosed with 0.1 g with triolein, trilinolein or trilinolenin containing 2 mg halofantrine. Following oral administration of thetriglycerides, the mesenteric lymph and plasma samples were collected. The lymphatic transport for halofantrine was 11.161.2 afteradministration of trilinolein, 9.063.5 for trilinolenin and 8.662.2 for triolein and the total amount of halofantrine transported in the lymphwas linear proportional with the amount of triglyceride in the lymph. The absorption of halofantrine directly into the blood showed a trendtowards a higher AUC for trilinolien and trilinolenin compared to triolein, but no statistical difference could be found. The statisticallyanalysis of the mean total bioavailability therefore shows that the absorption of halofantrine was largely independent on triglycerideunsaturation. 2001 Elsevier Science B.V. All rights reserved.

Keywords: Lymphatic transport; Lipid; Halofantrine; Unsaturated triglycerides; Triolein; Trilinolein; Trilinolenin

1. Introduction 1985; Charman and Stella, 1986a,b,c; Charman et al.,1986; O’Driscoll et al., 1991), mepitionstan (Ichihashi et

The pharmaceutical advantages associated with the al., 1991, 1992), ontazolast (Hauss et al., 1998), pen-intestinal lymphatic transport of drugs comprise a higher clomedine (Meyers and Stella, 1992) and MK-386 (Kweioral bioavailability, due to the avoidance of hepatic first- et al., 1998).pass metabolism and the potential to selectively target Possible disadvantages and side effects can occur fordrugs to the lymphatic system (Porter and Charman, 1997). some lymphatic transported drugs, as has been seen forThe factors that control the transport of drugs into the halofantrine, where co-administration with food is con-intestinal lymphatic system are not completely understood. traindicated due to observations of prolongation of the QTIt has been suggested that the drug should have a log P in interval of the ECG in patients with a pre-existing car-excess of 4.6 and a solubility in the co-administered lipid diopathy (Monlun et al., 1995; Touze et al., 1996). This isgreater than 50 mg/ml (Charman and Stella, 1986c). As probably caused by delivery of a high concentration ofthese lipophilic drugs are generally transported in associa- halofantrine close to the heart, because the lymph aretion with the lipid core of the lipoproteins (Pocock and drained into the systemic blood circulation close to theVost, 1974), formulations that stimulate lipoprotein forma- hearth, at the junction of the jugular and the left subclaviantion, have been shown to enhance the lymphatic transport veins.of drugs (Hauss et al., 1994; Caliph et al., 2000). The Halofantrine was used as a model compound in thisintestinal lymphatic transport of numerous xenobiotics was study (log P58.5, solubility in triglycerides .50 mg/ml)shown to affect the total absorption, when administered to examine possible differences in the intestinal lymphaticwith lipids, including DDT (Sieber, 1976; Noguchi et al., transport as a function of co-administered vehicle, because

the intestinal lymphatic transport previously has beenfound to play a major role in the gastrointestinal absorption*Corresponding author. Tel.: 145-353-06-440; fax: 145-353-06-031.

¨E-mail address: amu@dfh.dk (A. Mullertz). of the drug, when co-administered with a lipid formulation

0928-0987/01/$ – see front matter 2001 Elsevier Science B.V. All rights reserved.PI I : S0928-0987( 01 )00186-5

332 R. Holm et al. / European Journal of Pharmaceutical Sciences 14 (2001) 331 –337

(Porter et al., 1996a,b; Caliph et al., 2000). Clinically, 2.3. Halofantrine formulationshalofantrine is being used as an antimalaria drug, againstdrug-resistant strains of Plasmodium falciparum (Bryson The concentration of halofantrine in the formulationsand Goa, 1992), one of the most prevalent and pathogenic was determined by an high-performance liquid chromatog-malarial parasites. raphy (HPLC) method previously described (Humberstone

The absorption of halofantrine in anaesthetised rats was et al., 1995), by dissolving the formulation in acetonitrile.seen to be affected by degree of unsaturation of co- The HPLC used for analysis included a Hewlett Packardadministered fatty acids (Holm et al., 2001). The rank 1100 system with a diode array detector. The separation oforder in terms of lymphatic transport of halofantrine was halofantrine was accomplished by using a reversed-phasefound to be linoleic.oleic.linolenic acid. In the same Luna C Column (4.63250 mm 5 mm ODS(2)), with a8

study, the blood data from the lymph cannulated rats dosed Phenomenex C , ODS 433.0 mm guard column, and the8

with the three fatty acids showed a trend towards a higher data were analysed using Hewlett Packard ChemStationabsorption into the systemic blood when halofantrine was software for LC. During the study, the formulations weredosed with linolenic acid, totally producing a higher inspected visually for physical changes as precipitation ofavailability. In terms of lipid formulation design, systems the drug, changes in the colour of the oil or phaseincorporating free fatty acids present potential stability separation of the intravenous formulation.problems, and consequently triglycerides are often usedinstead. The objective of the present study was therefore to 2.3.1. Solution in oilinvestigate whether this effect seen by Holm et al. (2001) The oil solutions were prepared by dissolving halofan-also applies when halofantrine is administered with differ- trine free base in triolein, trilinolein or trilinolenin to give aent unsaturated triglycerides in conscious rats. formulation containing 2 mg halofantrine base in 100 mg

triglyceride.

2. Materials and methods 2.3.2. Intravenous formulationHalofantrine was incorporated into an emulsion by using

2.1. Chemicals the method of El-Sayed and Repta (1983). Halofantrinefree base was dissolved in dimethylformamide (125 mg/

Halofantrine hydrochloride and the internal standard ml). Subsequently, 0.25 ml of this solution was slowly2, 4 - dichloro - 6 - trifluromethyl - 9h1 - [2 - (dibutylamino) - added, under aseptic conditions, to 20 ml of a rapidly

ethyl]jphenathrenemethanol hydrocloride were donated by stirred Intralipid 10% emulsion (containing 10% soybeanSmithKline Beecham Pharmaceuticals (Worthing, UK). oil, 1.2% lecithin and 2.25% glycerol in water, FreseniusTriolein 99% (1, 2, 3-Tri[cis-9-octadececenoyl]glycerol) Kabi, Copenhagen, Denmark)and Trilinolein 99% (1, 2, 3-Tri[(cis, cis)-9, 12-octadeca-dienoyl]glycerol) were obtained from Sigma Chemicals

2.4. Surgical procedures(St. Louis, MO, USA), Trilinolenin 99% (1, 2, 3-Tri[cis,cis, cis)-9, 12, 15-octadecatrienoyl]glycerol) was obtained

All surgical and experimental procedures were reviewed¨from Larodan Fine Chemicals (Malmo, Sweden). The

and approved by the local Animal Experimentation Ethicstriglycerides were used without further purification.

Committee. The animals were anaesthetised for the dura-Acetonitrile, tert.-butylmethylether and dichloromethane

tion of the surgery by intra-peritoneal injection of 3 ml /kgwere HPLC grade and sodium dodecyl sulphate was of a solution consisting of Hypnorm (fentanyl 0.2 mg/electrophoresis grade. The water used in all experiments ml, fluanisone 10 mg/ml, Janssen, Belgium), Hypnovelwas obtained from a Milli-Q-water purification system

(midazolam 5 mg/ml, Roche, Switzerland) and water(Millipore, Molsheim, France). All other chemicals were

(1:1:2). Additional injections were given if required.analytical reagent grade.

The mesenteric lymph duct was cannulated by a slightMale Sprague–Dawley rats (280–320 g) were purchased

modified method as previously described by Noguchi et al.from M&B (Lille Skendsved, Denmark) and maintained on

(1985) with a piece of PVC tubing (0.50 mm I.D., 0.80standard food and water ad libitum in our laboratory for at

mm O.D., Critchley Electrical Products, Australia), underleast 1 week prior to entering the experiment.

aseptic conditions. The cannula was externalised throughthe abdominal wall and the rats were equipped with a

2.2. Preparation of halofantrine free base collar and a jacket that held the collection bottle for thelymph, allowing free animal movement without interrupt-

The amorphous form of halofantrine free base was ing continuous lymph collection.prepared from the commercially available halofantrine The right external vein was cannulated with a Tygonhydrochlorid salt as described previously by Porter et al. Microbore tube, type S54-HL (Norton Performance Plas-(1996a). tics, USA) in such a way that the end of the tube was at the

R. Holm et al. / European Journal of Pharmaceutical Sciences 14 (2001) 331 –337 333

level of the right atrium. The cannula was exteriorised at were vortexed and centrifuged, and a 4 ml proportion wasthe back of the neck. removed into a glass centrifuge tube containing 100 ml of

The animals that received the intravenous formulation 0.005 M HCl (in acetonitrile). The content was evaporatedwere sham operated in terms of the mesenteric lymph duct to dryness under nitrogen atmosphere at 358C. The residuecannulation, and the animals applied for determination of was reconstituted with 100 ml acetonitril and 25 ml wasendogenous lipid production were sham operated in terms injected on to the LC column. The limit of quantificationof the vene cannula. by this procedure was 20 ng/ml. The assay was linear

between 20 and 4000 ng/ml, and the extraction efficiency2.5. Experimental procedures for halofantrine was greater than 90% across the con-

centration range.After completion of the surgical procedures, the animals

were transferred to clean cages and stabilised overnight(16–24 h), during which period the animals were fasted. 2.7. Analysis of triglycerides in lymphFree access to 5% dextrose in a lactated Ringer’s solutionwas permitted post-operatively and throughout the study. Lymph triglyceride concentrations were determined byThe animals were randomly assigned to receive 100 mg of an enzymatic colorimetric method (Boehringer Mannheim,one of the oils. The oils were orally administered via Mannheim, Germany, Cat. No. 701912) and assayed atgavage. 500 nm on a Cary 1Bio UV-vis spectrophotometer (Varian

The lymph was collected into polyethylene tubes con- Instuments, Mulgrave, Victoria, Australia).taining 100 ml of a 0.75% EDTA solution. The lymph was Lymphatic triglyceride transport due to exogenouslycollected 1 h before and 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, administered lipid was determined by subtracting the24 and 30 h after the drug administration. Blood samples endogenous lipid component from the mass of triglyceride(0.25 ml) were taken from the right external jugular vein at lipid determined in each collected lymph sample. Endo-25 min and 2, 4, 6, 8, 10, 12, 24 and 30 h after the drug genous lymphatic transport, which was zero order, wasadministration into a Eppendorf tube containing 20 ml of calculated to be 2.4360.80 mg/h determined from the200 units /ml heparin solution. Plasma was immediately slope of the regression of mean cumulative lymphatic

2harvested by centrifugation. Lymph and plasma samples triglyceride transport measured over 30 h (r .0.99) inwere stored at 2208C until analysis. seven rats.

In the intravenous group (n56), the animals receivedhalofantrine (0.46 mg in 0.3 ml emulsion) via the rightexternal jugular. Blood samples (0.25 ml) were taken from 2.8. Pharmacokinetic analysisthe right external jugular vein at 25, 5, 10, 20, 60 min and2, 4, 6, 8, 10, 12, 24 and 30 h post-dosing. Endogenous Plasma concentrations versus time data for halofantrinelymph triglyceride output over 30 h was determined from in individual rats were analysed by noncompartmentalrats receiving 1 ml of normal saline orally by gavage. estimations using the WinNonlin software version 2.1.

At the conclusion of the experiment, the animals were After intravenous administration, the AUC for halofantrinesacrificed by an overdose of sodium pentobarbitone given was determined using a two-compartment model in Win-in jugular catheter, and the integrity of the lymphatic Nonlin, and the area for the last measured point to infinitycannula was verified was calculated by the linear trapezoidal rule.

The total bioavailability after 30 h in the lymph cannu-2.6. Analysis of halofantrine in lymph and blood lated animals were calculated by addition of the cumula-

tive percent of halofantrine transported in the lymph withFrom the lymph samples, 100 ml of lymph was added to the percent of halofantrine found in the blood (Caliph et

5 ml acetonitrile. The sample was vortexed for 1 min. The al., 2000). The percent of halofantrine in the blood wasinsoluble protein-based components were removed by determined by normalising the AUC after oral and in-centrifugation, and the supernatant analysed by HPLC as travenous administration with the dose administered.previously described by Porter et al. (1996a). Recovery ofhalofantrine from blank lymph was found to be greaterthan 99% by this procedure. 2.9. Statistical analysis

The blood samples were analysed by a slightly modifiedmethod previously described by Humberstone et al. Statistical analysis was performed by one-way analysis(1995), by adding 100 ml plasma and 100 ml internal of variance using the Statgraphic software version 7.0, andstandard (2 mg/ml) to 0.5 ml acetonitrile in a 12 ml the Student–Newman–Keuls multiple comparison test waspolypropylene centrifuge tube. The sample was vortexed applied for analysing potential difference between thefor 2 min to precipitate plasma proteins, centrifuged, and 4 formulations. The results are considered significant if P,

ml of tert.-butyl methyl ether was added. The contents 0.05.

334 R. Holm et al. / European Journal of Pharmaceutical Sciences 14 (2001) 331 –337

3. Results and discussion

3.1. Intestinal lymphatic transport of halofantrine andtriglyceride

The cumulative appearance of halofantrine vs. time ispresented in Fig. 1. The lymphatic transport of halofan-trine, after 30 h expressed as percentage of the total dose,was 11.161.2 (mean % of dose6S.E., n55 for all groups)after administration of trilinolein, 9.063.5 for trilinoleninand 8.662.2 for triolein. There seems to be a tendency fortrilinolein to produce a higher and more reproducibleintestinal lymphatic transport of halofantrine, but there wasno significant difference (P,0.05) in the cumulative

Fig. 2. The cumulative transport of triglyceride (mean6S.E.; n55) inpercent of halofantrine transported for the three differentmesenteric lymph as a function time after oral administration of differenttriglycerides. The peak rates of halofantrine transportunsaturated triglycerides. Triolein (d), trilinolein (j) and trilinolenin

occurred 2–3 h post-dosing, and the absorption was almost (m).complete after 12 h, in accordance with Porter et al.(1996b).

After correction for endogenous contribution, the teric lymph, as a function of cumulative triglyceridecumulative transport of exogenous triglyceride into the absorption and time, for the three formulations is shown inmesenteric lymph is shown in Fig. 2. The rate and amount Fig. 3. The figure shows a linear relationship betweenof triglyceride transported was at the same level for the halofantrine and triglyceride transport into the lymph forthree groups, indicating that the degree of unsaturation has the first 8–9 h post-dosing. The slopes of the lines wereno impact on the efficiency or the capacity of triglyceride comparable for the three triglycerides examined, indicatingtransport of halofantrine. The maximal rates of triglyceride that there are no differences in the lymphatic drug trans-transport into the lymph occurred 2 h post-dosing. Here- port capacity of the lipids. A linear relationship betweenafter, the transport rate gradually declines. The uptake of triglyceride and drug transport was previously reported bytriglycerides into the lymph seems almost complete after McIntosh et al. (1999), who in fasted humans and beaglecollection of the 12–24 h fraction, which is consistent with dogs found a high correlation between the load of apolarthe observations of Porsgaard et al. (1999), who showed lipids and proportion of halofantrine in different plasma-that the composition of the fatty acids in rat lymph returns lipoproteins.to baseline value between 12 and 15 h post-dosing and that The halofantrine transport into the intestinal lymphthe fatty acids transported subsequently were derived decreases 8–9 h after dosing, whereas the triglyceridemostly from endogenous stores. transport continues. In accordance Khoo et al. (2001)

The cumulative halofantrine transport into the mesen- found a linear relation between triglyceride and halofan-

Fig. 1. The cumulative appearance of halofantrine (mean6S.E.; n55) inmesenteric lymph as a function of time after oral administration of 2 mg Fig. 3. Correlation between cumulative mass of triglycerides and cumula-halofantrine dissolved in three different triglycerides to conscious rats. tive halofantrine transport in the lymph after oral administration inTriolein (d), trilinolein (j) and trilinolenin (m). triolein (d), trilinolein (j) and trilinolenin (m).

R. Holm et al. / European Journal of Pharmaceutical Sciences 14 (2001) 331 –337 335

trine transport into the lymph in greyhound dogs, after to the two other fatty acids was not seen when halofantrinepost-prandial dosing of halofantrine, for the initial 2–3 h was dosed to conscious rats together with triglyceridespost-dosing. Thereafter, the transport of halofantrine de- with the same fatty acids on the glycerol backbone. Whencreased despite continuing lymphatic transport of tri- formulating lipid systems free fatty acids present potentialglycerides. The difference in transport times between these stability problem, and consequently triglycerides are oftentwo studies, is caused by the difference in the species used. used instead. The carboxyl acid on the free fatty acid canRats have constant bile flow independent of food intake, react with the hydroxy group on halofantrine and form anwhereas the intestinal environment in dogs differs in pre- ester-complex. This complex would differ in physicoand post-prandial states by different fluid volume and chemical and possible also biological characteristic com-different concentrations of bile salts, phospholipids, lip- pared to free halofantrine. The stability of halofantrine in

`ases, etc. (Carriere et al., 1993). The change in ratio the fatty acid formulations used by Holm et al. (2001) wasbetween halofantrine and triglyceride transport could be examined. The complex formation between halofantrinecaused by a reduced halofantrine concentration within the and the fatty acids occurred to some extent during thelipid-processing domains in the enterocyte, where the storage of the fatty acid formulation, but not at a level thatchylomicron syntheses are progressed. This is most likely was considered to interfere significantly with the purposedue to continuos partitioning of halofantrine into the portal or the results of the study (unpublished results). Theseblood, and to a decrease in the amount of halofantrine stability data indicated that the difference betweenremaining to be absorbed (Khoo et al., 2001). Another lymphatic transport of halofantrine found in this study andpossibility is the presence of a relative narrow ‘absorption in the study by Holm et al. (2001), was not introduced bywindow’ for halofantrine, which previously has been differences in the physical form of halofantrine. However,suggested to exist for cyclosporine by Ueda et al. (1983). the two studies could not be directly compared: (i) theHowever, these possibilities need further investigations. dosed triglycerides need to be digested before absorption,

The mass of lymph transported after dosing of the three whereas the free fatty acids can be absorbed directly; (ii)different triglycerides was 19.463.9 g (mean6S.E., n55 the presence of monoglycerides from the digested tri-for all groups) after administration of triolein, 26.264.8 g glyceride introduces the use of different biochemicalfor trilinolein and 31.663.7 g for trilinolein, suggesting pathways in the intracellular resynthesis of triglycerides inthat a higher amount of double bonds in the administered the two situations; and (iii) due to the use of differenttriglyceride enhances the lymph flow, but no statistically animal models in the two studies, as examined andsignificant difference could be found between the three discussed extensively by several authors (Charman andgroups. These results were partly consistent with the Stella, 1986b; Charman et al., 1986, Porter et al., 1996a,b).results from Nankervis et al. (1995), who found a differ-ence in the lymph flow after dosing isotretionin in cotton- 3.2. Total bioavailability of halofantrineseed, soybean or peanut oil to rats. The animals dosed withcottonseed and soybean oil showed a higher lymph flow Fig. 4 presents the plasma concentration–time profilesthan the animals dosed with peanut oil. This led the after oral administration of halofantrine dissolved inauthors to suggest that linoleate triglycerides enhance the triglycerides to lymph cannulated rats, and the totallymph flow, because cottonseed and soybean oil compriseof higher percentage of the trilinolein than peanut oil, 52,51 and 26%, respectively. Previous studies have shownthat the size of chylomicrons also are affected by thedegree of unsaturation (Feldman et al., 1983; Kalogerisand Story, 1992). The largest chylomicrons in these twostudies were produced when the rats where dosed withtriolein compared to trilinolein. The increased flow and theexpected difference in size seems to have very littleinfluence on the lymphatic triglyceride and drug transportin this study, indicated by their linear relationship. Thislack of relationship between lymph flow and the lymphatictransport of xenobiotic was also previously found byCharman and Stella (1986a) in anaesthetised rats afterdosing of DDT in either peanut oil or oleic acid.

Previously it was found that linoleic acid was sig-nificantly better than linolenic acid in terms of enhancing

Fig. 4. Plasma halofantrine concentration–time profiles (mean6S.E.; n5the lymphatic drug transport in anaesthetised rats (Holm et 5) as a function of formulation, following oral administration of 2 mgal., 2001). This difference found in both the efficacy and halofantrine dissolved in three different oils to conscious rats. Trioleinthe amount of drug transported for linolenic acid compared (d), trilinolein (j) and trilinolenin (m).

336 R. Holm et al. / European Journal of Pharmaceutical Sciences 14 (2001) 331 –337

Table 1 teric lymph for the first 8–9 h, indicating a dependence ofComparison of the calculated total bioavailability (mean6S.E., n55 for the kinetic processes in the lipid digestion, solubilisationall groups) for halofantrine determined after oral administration to lymph

and re-synthesis of triglyceride in the enterocyte on thecannulated ratslymphatic transport of drugs.

Triglyceride in Total bioavailability (%)oil solution a bBlood Lymph Total

(lymph1blood) AcknowledgementsTriolein 6.261.7 8.662.2 14.763.9Trilinolein 9.461.7 11.161.2 20.562.9 This work was financially supported by the DanishTrilinolenin 9.261.9 9.063.5 18.265.3 Medical Research Council, Centre for Drug Design and

a The percentage dose halofantrine absorbed directly into the blood was Transport. The authors would like to thank Dr C.H.J.0→30 h 0→8estimated based on plasma AUC relative to the AUC obtained Porter from department of Pharmaceutics at the Victorian

after i.v. administration (the AUC after i.v. administration of 0.46 mgCollege of Pharmacy, Melbourne, Dr G.P.Pedersen fromhalofantrine per rat was 10410.16680 ng?h/ml).

b Leo Pharmaceutics Products for useful discussions con-Cumulative mass of halofantrine recovered over 30 h in mesentericlymph, calculated as a percentage of dose. cerning this work and Irene Klausen for technical assis-

tance.

0→30bioavailability is presented in Table 1. The mean AUCfor trilinolein and trilinolenin is higher than triolein, Referencesindicating that the unsaturation of triglycerides has someeffect on the portal availability of halofantrine, but no Bryson, H.M., Goa, K.L., 1992. Halofantrine. A review of its antimalarial

activity, pharmacokinetic properties and therapeutic potential. Drugsstatistical difference can be found between the AUC’s of43, 236–258.the three triglycerides. Considering the total bioavail-

Caliph, S.M., Charman, W.N., Porter, C.J.H., 2000. Effect of short,ability, trilinolein has a non-significant advantage over the medium and long chain fatty acid-based vehicles on the absolute oralother two triglycerides, in stimulating the lymphatic trans- bioavailability and intestinal lymphatic transport of halofantrine andport and the blood absorption of halofantrine, as predicted assessment of mass balance in lymph-cannulated and non-cannulated

rats. J. Pharm. Sci. 89, 1073–1084.by the experiment with the fatty acids (Holm et al., 2001).`Carriere, F., Laugier, R., Barrowman, J.A., Douchet, I., Priymenko, N.,However no definitive conclusion can be given with the

Verger, R., 1993. Gastric and pancreatic lipase levels during test mealdata presented here. Consistent with the total bioavail- in dogs. Scand. J. Gastroenterol. 28, 443–454.ability reported in the present study, Caliph et al. (2000) Charman, W.N., Noguchi, T., Stella, V.J., 1986. An experimental systemfound the total bioavailability in lymph cannulated rats to designed to study the in situ intestinal lymphatic transport of lipophilic

drugs in anesthetized rats. Int. J. Pharm. 33, 155–164.be 22.764.0 after administration of halofantrine in peanutCharman, W.N., Stella, V.J., 1986a. Effect of lipid class and lipid vehicleoil.

volume on the intestinal lymphatic transport of DDT. Int. J. Pharm. 33,165–172.

Charman, W.N., Stella, V.J., 1986b. Testing potential dosage form4. Conclusions strategies for intestinal lymphatic drug transport: studies in the rat. Int.

J. Pharm. 33, 173–179.Charman, W.N., Stella, V.J., 1986c. Estimating the maximal potential forIn conclusion, this investigation suggested that mass of

intestinal lymphatic transport of lipophilic drug molecules. Int. J.lymph transported increased with increasing amount of Pharm. 34, 175–178.double bonds in the administered lipid. The change in El-Sayed, A.A.A., Repta, A.J., 1983. Solubilization and stabilization of anlymph flow does not seem to have an impact on the drug investigational antineoplastic drug (NSC-278214) in an intravenous

formulation using an emulsion vehicle. Int. J. Pharm. 13, 303–312.or triglyceride transport in the lymph, and a simpleFeldman, E.B., Russell, B.S., Hawkins, C.B., Forte, T., 1983. Intestinalrelationship to lymph mass transfer does therefore not

lymph lipoproteins in rats fed diets enriched in specific fatty acids. J.seem to exist. The influence of unsaturation on the Nutr. 133, 2323–2334.lymphatic transport of halofantrine previously seen when Hauss, D.J., Fogal, S.E., Ficorilli, J.V., Price, C.A., Roy, T., Jayaraj, A.A.,co-administered with free fatty acids, was not as pro- Keirns, J.J., 1998. Lipid based delivery systems for improving the

bioavailability and lymphatic transport of a poorly water soluble LTB4nounced when co-administered with triglycerides contain-inhibitor. J. Pharm. Sci. 87, 164–169.ing the same fatty acids on the glycerol backbone. Never-

Hauss, D.J., Mehta, S.C., Radebaugh, G.W., 1994. Targeted lymphatictheless, trilinolein turned up positively in both drug transport and modified systemic distribution of CI-976, a lipophilictransport pathways examined, which could suggest a small lipid-regulator drug, via formulation approach. Int. J. Pharm. 108,biopharmaceutical advantage for this triglyceride. How- 85–93.

¨Holm, R., Mullertz, A., Pedersen, G.P., Kristensen, H.G., 2001. Com-ever, the statistical analysis showed no significant impactparison of the lymphatic transport of halofantrine administered inon the lymphatic transport or the blood availability ofdisperse systems containing 3 different unsaturated fatty acids. Pharm.

halofantrine in any of the three examined triglycerides. A Res. (in press).linear relationship was found between the cumulative Humberstone, A.J., Currie, G.J., Porter, C.J.H., Scanlon, M.J., Charman,transport of halofantrine and triglycerides into the mesen- W.N., 1995. A simplified liquid chromatography assay for the quantita-

R. Holm et al. / European Journal of Pharmaceutical Sciences 14 (2001) 331 –337 337

tion of halofantrine and desbutylhalofantrine in plasma and identifica- vehicle on the intestinal lymphatic transport of isotretinoin in rat. Int.tion of a degradation product of desbutylhalofantrine formed under J. Pharm. 119, 173–181.alkaline conditions. J. Pharm. Biomed. Anal. 13, 265–272. Noguchi, T., Charman, W.N., Stella, V.J., 1985. Lymphatic appearance of

Ichihashi, T., Kinoshita, H., Takagishi, Y., Yamada, H., 1991. Intrinsic DDT in thoracic or mesenteric lymph duct cannulated rats. Int. J.lymphatic partition rate of mepitiostane, epitiostanol, and oleic acid Pharm. 24, 185–192.absorbed from rat intestine. Pharm. Res. 8, 1302–1306. O’Driscoll, C.M., Myers, R.A., Stella, V.J., 1991. Blood and lymph

Ichihashi, T., Kinoshita, H., Takagishi, Y., Yamada, H., 1992. Effect of transport of DDT after oral and parenteral administration to anaesthet-oily vehicles on absorption of mepitiostane by the lymphatic system in ised rats. Int. J. Pharm. 73, 177–183.rats. J. Pharm. Pharmcol. 44, 560–564. Pocock, D.M.E., Vost, A., 1974. DDT absorption and chylomicron

Kalogeris, T.J., Story, J.A., 1992. Lymph chylomicron composition and transport in rat. Lipids 9, 374–381.size are modified by level of intestinal infused cholesterol and Porsgaard, T., Straarup, E.M., Høy, C.-E., 1999. Lymphatic fatty acidtriglyceride source in rats. J. Nutr. 122, 1045–1055. absorption profile during 24 h after administration of triglycerides to

Khoo, S.-M., Edwards,G.A., Porter, C.J.H., Charman, W.N., 2001. A rats. Lipids 34, 103–107.conscious dog model for assessing the absorption, enterocyte-based Porter, C.J.H., Charman, S.A., Charman, W.N., 1996a. Lymphatic trans-metabolism and intestinal lymphatic transport of halofantrine. J. port of halofantrine in the triple-cannulated anesthetized rat model:Pharm. Sci. (in press). Effect of lipid vehicle dispersion. J. Pharm. Sci. 85, 351–356.

Kwei, G.Y., Novak, L.B., Hettrick, L.H., Reiss, E.R., Fong, E.K., Olah, Porter, C.J.H., Charman, S.A., Humberstone, A.J., Charman, W.N., 1996b.T.V., Loper, A.E., 1998. Lymphatic uptake of MK-386, a sterol Lymphatic transport of halofantrine in the conscious rat when adminis-5a-reductase inhibitor, from aqueous and lipid formulations. Int. J. tered as either the free base or the hydrochlorid salt: effect of lipidPharm. 164, 37–44. class and lipid vehicle dispersion. J. Pharm. Sci. 85, 357–361.

McIntosh, M.P., Porter, C.J.H., Wasan, K.M., Ramaswamy, M., Charman, Porter, C.J.H., Charman, W.N., 1997. Uptake of drugs into the intestinalW.N., 1999. Difference in the lipoprotein binding profile of halofan- lymphatics after oral administration. Adv. Drug Del. Rev. 15, 71–89.trine in fed and fasted human or beagle plasma are dictated by the Sieber, S.M., 1976. The lymphatic absorption of p, p-DDT and somerespective masses of core apolar lipoprotein lipid. J. Pharm. Sci. 88, structurally-related compounds in the rat. Phamacology 14, 443–454.378–384. Touze, J.E., Bernard, J., Keundjian, A., Viguier, A., Chaudet, H., Doury,

Meyers, R.A., Stella, V.J., 1992. Factors affecting the lymphatic transport J.C., 1996. Electrocardiographic changes and halofantrine plasma levelof penclomedine (NSC-338720), a lipophilic cytotoxic drug: Com- during acute falciparum malaria. Am. J. Trop. Med. Hyg. 54, 225–parison to DDT and hexachlorobenzene. Int. J. Pharm. 80, 51–62. 228.

Monlun, E., Le Metayer, P., Szwandt, S., Longy-Boursier, M., Horton, J., Ueda, C.T., Lemaire, M., Gsell, G., Nussbaumer, K., 1983. IntestinalLe Bras, M., 1995. Cardiac complications of halofantrine: A prospec- lymphatic absorption of cyclosporin A following oral administration intive study of 20 patients. Trans. R. Soc. Trop. Med. Hyg. 89, 430–433. an olive oil solution in rats. Biopharm. Drug Dispos. 4, 113–124.

Nankervis, R., Davis, S.S., Day, N.H., Shaw, P.N., 1995. Effect of lipid