EUROPEAN lOURNAL OF DRUG METABOUSM AND PHARMACOKINETICS, 1991, Vol. 16, No.3, pp. 197-201
Pharmacokinetics of dipyridamole-Becyclodextrin complex in healthy volunteers after single and multiple doses 1 G. RICEVUTI 1, A. MAZZONE, D. PASOTII1, E. UCCELLI1, F. PASQUALI 1, G. GAZZANI2 and G.B. FREGNAN3 1Section ofMedical Pathology I, Department ofInternal Medicine and Therapeutics, University ofPavia, Pavia, Italy 2Department ofPharmaceutical Chemistry, University ofPavia, Pavia, Italy 3Research Center, EdmondPharma, Milan, Italy
Received for publication: October 13, 1989
Keywords: Dipyridamole, cyclodextrin complexes, molecular encapsulation, bioavailability, anti-aggregating agents, coronary dilator
SUMMARY Dipyridamole is a well known anti-aggregating agent characterized by poor water solubility as well as scant and variable bioavailability. Recently, the compound was complexed with JH:yclodextrin forming a molecular encapsulation resulting in better oral absorption and stronger biological activities in animals. In the present study, a randomized double blind cross-over comparison between dipyridamole-(3-cyclodextrin complex (dip-(3-CD) and dipyridamole was performed in 12 healthy subjects after single (75mg) and multiple oral treatments (75mg TID). Dip-(3-CD showed better bioavailability and less interindividual variability than dipyridamole either after single or multiple doses. In particular, dip-(3-CD had a greater AUC and C max, and a smaller Tmax even at the steady state. In addition, 100% of the subjects receiving a single dose of dip-(3-CD, as compared to 66.7% of those treated with dipyridamole, had plasma levels superior to I J.Lglml (which is the supposed anti-aggregating threshold level). In contrast, 0 and 33.03% of the subjects showed plasma levels superior to 2.5 ug/ml (which might cause the appearance of side-effects) on the 7th day of the multiple treatment with dip-(3-CD and dipyridamole, respectively. In fact, the subjects presenting higher levels after uncomplexed dipyridamole also complained of headache and/or dizziness on occasion. No adverse side effects were reported for dip-(3-CD.
INTRODUCTION The complexation of active drugs with cyclodextrins, may change their physico-chemical properties. Oral bioavailability, for example, is generally increased in animals and men when poorly water-soluble drugs are complexed. As a consequence, lag time and T max are shortened, Cmax is increased and prompter and stronger activity is observed. Occasionally, longer-lasting blood levels with more durable therapeutic effects are obtained. In the present study, Please send reprint requests to : Prof. G.B. Fregnan, Research Center, Edmond Pharma, v. Gradisca 8, 1-20151 Milan, Italy
the pharmacokinetic profile of dipyridamolefl-cyclodextrin complex (diP-B-CD) was determined in healthy volunteers after single and multiple oral treatments. Dipyridamole is a well known anti-aggregating agent widely used in the therapy of thrombo-embolic disorders. Unfortunately, the compound is poorly and variably absorbed after oral treatment (36--67%) (1) and also presents great interindividual variability (2, 3). Furthermore, its therapeutic index is very narrow in humans, not showing evident platelet anti-aggregating activity with plasma levels below 0.5 J.Lg!ml and evoking side-effects above 2.5 ug/ml (4). Previous studies have indicated that dipyridamole was able to form inclusion complexes with ~-cyclodextrin which are
Eur. J. DrugMetab. Pharmacokinet., 1991, No.3
more stable, more water soluble, more active and more bioavailable in animals (5-7). The main objective of the present investigation was to compare dip-13-CD and uncomplexed dipyridamole bioavailability in the same subjects, according to a randomized crossover design.
centrifuged and the separated plasma frozen at -20·e until assayed for dipyridamole. Urine was also collected immediately before, 0-6, 6--12 and 12-24 h after the morning dose on days 1 and 7. For each collection period, total urine voided was measured and 100 ml aliquots were lyophilized and stored at -20·e until assayed.
MATERIALS AND METHODS Sample analysis Subjects 12 healthy volunteers, 7 males and 5 females, 24 to 49 years old, 53 to 87 kg of body weight, took part in the study. Subjects had normal physical examination results and clinical laboratory profiles. Concomitant medications were excluded from one week prior to the study until its completion.
Trial design and treatment scheme A double blind cross over study was performed, with the subjects divided into two treatment groups (dip-13-CD and dipyridamole), according to a Latin square design, each of which received one or the other drug on the same day. After a two-week wash-out period the volunteers were assigned to the other treatment so as to make up a total of 12 individuals for each drug. The subjects were uninterruptedly treated for 7 days with each substance. On the Ist and 7th day the drugs were administered only once at 08.00 in the morning, while from the 2nd to 6th day they were administered thrice a day as close as possible to 08.00, 16.00 and 24.00. The morning dose was always taken with a glass of water after an overnight fast; light breakfast (cup of coffee plus a brioche) and lunch were given 2 and 5 h later. Each dose contained 75 mg of active principle, so that on the 1st and 7th day only 75 mglday were administered and from the 2nd to 6th day 225 mg/day (75 mg TID) were given.
Sample collection Two millilitres of blood were drawn into heparinized tubes just before dosing and 0.25,0.50,0.75, 1, 1.5,2, 3, 4, 5, 6, 8, 10, 12 and 24 h postdose on the first and seventh mornings. Blood samples were also collected . before and 2 h after each morning dose from the 2nd to 6th day of treatment, Samples were inunediately
After 1-2 weeks the biological samples were thawed. Plasma samples (0.2 ml) were placed in a 15 ml tube and 50 Jig (in 0.010 ml) of benzocaine, 1 ml of sodium hydroxide 1 N and 4 ml of terbutylmethylether were added. The samples were immediately extracted using a vortex mixer at high speed for 1 min, followed by centrifugation at 1000 g for 5 min to separate the aqueous and organic phases. The lower aqueous phase was frozen by immersing the tube in a dry ice-acetone bath and the upper organic phase decanted into another tube. The terbutylmethylether was evaporated under purified nitrogen at 4O-50·C. The residue was dissolved in 100 ul of the mobile phase and all or part of this volume was injected into a Beckman model UOA HPLC equipped with an injector and fitted with a Merck column CI8 (150 x3 nun i.d., particle size 5 urn). The mobile phase was a mixture of acetonitrile/monobasic potassium phosphate 0.01 M at pH 7 with addition of NaOH 0.1 N (45:55). The flow-rate was 0.6 mlImin. Fluorescence was measured with an Hitachi model F-1000 fluorometer. An excitation wavelength of 285 om was selected in conjunction with a 470 om emission ftlter. An aliquot (1-10 g) of lyophilized urine was brought to room temperature, 50 Jig of internal standard were added, mixed with appropriate volumes of sodium acetate 0.5 M at pH 4.7 and of 13-glucuronidase (5000 unitslml). The samples were incubated at 37·e for 12 hand extracted with adequate volumes of sodium hydroxide 1 M at pH 8.6 and of terbutylmethylether. Separation, exsiccation and chromatography of the organic phase were then performed as described for the plasma. Standard calibration curves were obtained by adding 1,3,20,30, 100,300 and 600 ng of dipyridamole and 50 ~g of internal standard (benzocaine) to 0.2 ml of control biological samples, which were assayed concurrently with the unknowns. Linear regression analysis was performed on the results obtained from the standard samples. The equation of the best fit for the standard curve was used to calculate the concentration of an unknown sample from the peak height ratio measured. Under the described conditions
G. Ricevuti et 01., Dip-~-CD complex
Pharmacokinetic and statistical analyses
dipyridamole and benzocaine had retention times of 3.5 ± 0.1 and 6.3 ± 0.2 min. respectively. The extraction procedure allowed for the determination of plasma concentrations of dipyridamole as low as 5-10 nglml of biological sample. By plotting concentration against peak. height ratio of standard curve. a good linearity was obtained from 0.005 to 3 J.1gIml with a correlation coefficient of 0.999 calculated by the least square method. The slope of the line was calculated to be 0.0062 with a y-intereept of 0.1008. The reproducibility of the assay was verified by extracting 10 samples of each biological specimen to which 100 nglml of dipyridamole had been added. The coefficient of variation of the peak height ratio was 4.5% (2.7-6.8) for the plasma and 5.2% (2.5-7.2) for urine. The percent recovery of dipyridamole from these standard samples was 85% (78-91) and 80% (74-89), respectively, with an average coefficient of variation of 4.3% (2.9-5.7) and 4.8% (3.5-6.3). 2
The pharmacokinetic parameters were assessed by the CSTRIP, a Fortran IV computer program (8). Statistical differences between treatments were calculated according to ANaVA or to Student's t-test for paired data and expressed for a P ~ 0.05.
RESULTS Single treatment The mean plasma-time curves and the pharmacokinetic parameter values are presented in Figure 1 and Table I, respectively. It is apparent from these average values that dip-~-CD behaved differently from dipyridamole. In particular, T max was
I .•.. l
.. ... . . .. . ... .. .. . . ~.
: I I l
III III I· I
Fig. 1 : Mean plasma concentrations ( ± SE) after single and multiple oral treatments with dip-~-CD (A) and with dipyridamole (B) Blood samples 0, 0.25, 0.50,0.75, I, 1.5,2,3,4,5,6,8, 10, 12 and 24 h after the morning dose on 1st and 7th day, 0 and 2 h after the morning dose from 2nd to 6th day of treatment • Dipyridamole was not detected in plasma p ~ 0.05 dip-J3-CD vs dipyridamole single (*) or multiple treatments (+)
Eur. J. Drug Metab. Pharmacokinet., 1991, No.3
Table I : Pharmacokinetic parameters after single and multiple oral treatments with
and dipyridamole. Means ± SE
Cmax (J.Lglml) T max (h) TlJ2~
5.47 ± 0.39*t 1.42 ± 0.06*t 1.04 ± 0.08*t 5.84±0.55
10.44 ± 0.41* 1.88 ±0.05* 1.27 ± 0.06* 4.87 ±0.47
4.13 ± 0.52t 0.93 ±0.13t 1.41±0.15 5.65 ±O.73
7.86 ± 0.46 1.37 ± 0.13 1.56 ± 0.08 5.34±0.51
*p S 0.05 dip- P-CD vs dipyridamoleafter single or multiple treatments t p S 0.05 single vs multipletreatments with dip-p-CDor dipyridamole
smaller while Cmax and AUC were greater for the former than for the latter drug, indicating prompter and better absorption, In addition, 100% of the subjects receiving dip-~-CD, as compared to 66.7% of those treated with dipyridamole, presented plasma . levels superior to 1 IJg!ml (which is the supposed anti-aggregating threshold level in man). By observing the standard errors of the mean values, it can also be seen that the interindividual variability was definitively reduced by the complexation of the active principle. Although the amount of dipyridamole excreted with the urine was very small (~ 1% of the administered dose), dip-~-CD gave significantly higher values than dipyridamole itself, further proof that the complexation improved the bioavailability of the active principle (Table II).
than for the latter drug. The plasma concentration-time curves (Fig. 1), some plasma phannacokinetic parameters (Table I) and the urinary excretion pattern (Table II) differed on the 7th day between treatments (generally, as reported on the 1st day). Significant differences were also observed between the two periods for each drug (e.g. higher AUC, Cmax and TlI2p, smaller interindividual variability and greater urinary excretion on the last than on the first day of each pharmacological treatment). In addition, 0 and 33.3% of the subjects presented plasma levels superior to 2.5 J.lg/mI on the 7th day of treatment with dip-~-CD and dipyridamole, respectively. Four subjects complained of headache and/or gastric distress on some occasions after the moming and/or night doses of dipyridamole. However, these troubles did not cause interruption of the treatment
Multiple treatments Figure 1 shows that after multiple treatments the steady state was obtained between the 3rd-4th or the 4th-5th day with dip-~-CD or dipyridamole, respectively, and that the morning plasma levels before treatment were generally greater for the former
DISCUSSION The present results prove that the oral pharmacokinetics of dipyridamole in healthy volunteers were influenced by the complexation of the
Table II: Comparative urinary excretion of dipyridamole after single or multiple oral treatments with dip-f3-CD and dipyridamole. Means±SE Drugs
J.Lg ofdipyridamole in urine at the following post-treatment periods (h) 6-12 12-24 0-24
% total recovery
355.3 ± 12.3*t 408.8 ± 15.8
260.9 ± 16.5*t 334.0 ± 20.9*
139.2 ± 10.2*t 213.6 ± 16.4
755.5 ± 23.9*t 948.3 ± 29.6*
262.6±44.6 357.0 ± 33.5
181.3 ± 22.9t 257.8 ± 20.7
89.5 ± 12.6t 162.2 ±23.7
533.5 ± 66.3t 777.0 ± 51.1
*p S 0.05 dip-P-COvs dipyridamole
tP S 0.05 1 vs 7 days of treatment with the same drug
G. Ricevuti et al., Dip·~·CD complex active principle with l3-cyclodextrin. Dip-~-CD was in. fact found to show higher plasma levels than dipyridamole itself. Higher Cmax and AUC plus smaller Tmax and interindividual variability, are proof of better bioavailability and prompter absorption, after either single or multiple treatments. These findings are fully in agreement, therefore, with previous ones demonstrating that dip-~-CD, as compared with the uncomplexed dipyridamole, gives faster, higher and longer-lasting blood levels in dogs, and possesses stronger vasodilating and platelet antiaggregating properties in several animal species (5, 6). Furthermore, most of the pharmacokinetic parameters were significantly changed for both drugs from day one to day seven. The greater values for AUC and Cmax at steady state are likely to be due to the presence of dipyridamole in circulation after the previous dose since Tll2fl did not change and urinary excretion increased at the same time for both dip-~-CD and dipyridamole. The following considerations might explain the different behaviour. First of all, it should be recalled that the complexation of dipyridamole with ~-cyclodextrin is achieved when the former (as whole or part) enters into the cavity of the latter compound. This process is called molecular encapsulation. The complex, so obtained, is characterized by physicochemical properties which are different from those of the individual substances (7, 9). In particular, dip-j3-CD appears to be more readily soludispersed in water and also in gastro-intestinal juices than dipyridamole itself. The j3-cyclodextrin, which is considered only a carrier agent, transports dipyridamole through the aqueous milieu of the gastro-intestinal tract directly to the absorption sites. The lipophilic intestinal membranes then absorb it, since they have a higher affmity for the guest molecule than ~·cyclodextrin does. The possibility also exists that some guest molecules might leave the host cavities either by exchanging with the environmental water or by enzymatic degradation of the cyclic oligosaccharidic structures. In any case, a
great number of single molecules of the active principle would come in contact with the villus membranes, involving a wide intestinal surface, with consequent facilitation of absorption. It may be concluded that the improved pharmacological behaviour in several animal species, the greater and more uniform bioavailability of dipyridamole after complexation, both in men and in dogs, allow us to predict a better use of the active principle in clinical practice. This means, of course, that dip-l3-CD should produce stronger and longer-lasting platelet anti-aggregating and coronary dilating effects with fewer side-effects (such as flushing, headache and dizziness) in patients affected by thrombo-embolic disorders.
REFERENCES 1. Nielsen-Kudsk F., Pedersen A.K. (1979): Pharmacokinetics of dipyridamole. Acta Phannaco!' Toxico!., 44, 391-399. 2. MellingerTJ., BohorfoushI.G. (1966): Blood levels of dipyridamole (persantin) in humans. Arch. Int. Phannacodyn., 163,471-480. 3. Manhoy C., Cox MJ., Bjomsson T.D. (1983): Plasma dipyridamole concentrations after two different dosage regimens in patients. I. Clin. Pharmaeol., 23, 123-126. 4. Rajah S.M., Crow MJ., Penny A.F., Alunad R., Watson D.A. (1977) : The effect of dipyridamole on platelet function: correlation with blood levels in man. Br, I. Clin. Pharmaeol., 4,129-133. 5. Fregnan G.B., Be~ F. (1990): Enhancement of specific biological activity of dipyridamole by complexation with fl-cyclodextrin. Phannacology,4O, 96-102. 6. Stracciari G.L., Malvisi I., Anfossi P., Fregnan G.B. (1989): Pharmacokinetics of dipyridamole-Bscyclodextrin complex in dogs. Arch. Int. Phannacodyn., 300, 7-13. 7. Tom G., Naggi A., Fregnan G.B., Trebbi A. (1990): Dipyridamole-lkyc1odextrin complex: preparation and characterization. Phannazie., 45, 193-196. 8. Sedman AJ., Wagner I.G. (1976): CS1RlP, a Fortran IV computer program for obtaining initial polyexponential parameter estimates. I. PhannacoI. Sci., 65, 1006-1010. 9. Fregnan G.B., Vandoni G., Tom G., Naggi A. (1989): Dipiridamolo-(l-ciclodestrina: un nuovo complesso con miglior biodisponibilitA. II Giomate italiane di biofannaceutica e fannacocinetica. Problematiche di rilevante interesse in biofarmaceutica e fannacocinetica. Roma, Italy, 21-22 March. Abstract p. 13.