Interaction of naproxen with noncrystalline acetyl β- and acetyl γ-cyclodextrins in the solid and liquid state

Interaction of naproxen with noncrystalline acetyl β- and acetyl γ-cyclodextrins in the solid and liquid state

European Journal of Pharmaceutical Sciences 15 (2002) 21–29 www.elsevier.nl / locate / ejps Interaction of naproxen with noncrystalline acetyl b- and...

739KB Sizes 0 Downloads 8 Views

European Journal of Pharmaceutical Sciences 15 (2002) 21–29 www.elsevier.nl / locate / ejps

Interaction of naproxen with noncrystalline acetyl b- and acetyl g-cyclodextrins in the solid and liquid state a b, b a c Giampiero Bettinetti , Paola Mura *, Maria Teresa Faucci , Milena Sorrenti , Massimo Setti a

Dipartimento di Chimica Farmaceutica, Universita` di Pavia, Viale Taramelli 12, I-27100 Pavia, Italy b Dipartimento di Scienze Farmaceutiche, via G. Capponi 9, I-50121 Firenze, Italy c Dipartimento di Scienze della Terra, Universita` di Pavia, Via Ferrata 1, I-27100 Pavia, Italy Received 12 February 2001; accepted 13 September 2001

Abstract Randomly acetylated, amorphous b-cyclodextrin (AcbCd) and g-cyclodextrin (AcgCd), having an average substitution degree per anhydroglucose unit, respectively, of 1.1 and 0.95 (|7.7 acetyl residues per macrocycle), were investigated for their interactions in the solid and liquid state with naproxen (NAP). Differential scanning calorimetry (DSC), supported by X-ray powder diffractometry (XRD), of NAP–AcbCd and NAP–AcgCd blends revealed an apparent decrease in drug crystallinity which was related to a heating-induced solid-state interaction between the drug and each carrier. A solubility of |0.40 NAP mass fraction in amorphous AcbCd and amorphous AcgCd at room temperature was determined. Phase-solubility analysis at 25, 37, and 458C accounted for A L -type inclusion complexation of NAP with AcbCd (K1:1,258C 54.5(4)310 3 l mol 21 ) and AcgCd (K1:1,258C 50.80(7)310 3 l mol 21 ) and revealed a solubilizing efficiency of AcbCd toward NAP |4 times that of AcgCd. Equimolar drug–carrier combinations prepared from the respective blends by grinding, kneading, coevaporation and freeze-drying were characterized by DSC and XRD and tested for dissolution rate of NAP using the dispersed amount and continuous flow through methods. The best performance in terms of dissolution rate enhancement (|23 times and |10 times the dissolution efficiency of pure drug in the dispersed amount and continuous flow through tests, respectively) was displayed by the NAP–AcbCd colyophilized product.  2002 Elsevier Science B.V. All rights reserved. Keywords: Naproxen; Acetyl b-cyclodextrin; Acetyl g-cyclodextrin; Solid-state interaction; Inclusion complexation; Thermal analysis; X-ray diffraction; Dissolution rate

1. Introduction Various kinds of cyclodextrin (Cd) derivatives have been prepared in order to modulate the physicochemical properties of the parent Cd. A number of methylated, hydroxyalkylated and branched Cds, for example, are currently used as solubilizing agents for poorly soluble drugs (Szejtli, 1994). Esterification of the primary 6hydroxyl groups on bCd with various acyl chlorides has been reported to disrupt the crystallinity of bCd, thereby enhancing its aqueous solubility and increasing its ability to solubilize water-insoluble compounds (Liu et al., 1992). Substitution of all seven hydroxyl groups at the 3-position of heptakis (2,6-di-O-methyl)bCd by acetyl groups led to a water-soluble derivative with superior bioadaptability and inclusion ability than the parent Cd (Hirayama et al., 1999). On the other hand, triacetylated bCd (Matsubara et *Corresponding author. Tel.: 139-55-275-7292; fax: 139-55-240-776. E-mail address: [email protected] (P. Mura).

al., 1994; Nakanishi et al., 1997) and gCd (Matsubara et al., 1994), as well as peracylated aCds (Nakanishi et al., 1998), bCds (Nakanishi et al., 1998; Uekama et al., 1994; Hirayama et al., 1995) and gCds (Nakanishi et al., 1998), have been proposed as bioabsorbable sustained-release carriers for hydrophilic drugs. In the present work randomly acetylated amorphous bCd (AcbCd) and gCd (AcgCd), having an average substitution degree per anhydroglucose unit, respectively, of 1.1 and 0.95 (i.e. |7.7 acetyl residues per macrocycle), were investigated for their interactions in the solid and liquid state with naproxen (NAP) as a model drug (Melani et al., 1995; Bettinetti et al., 1999; Mura et al., 1995). Differential scanning calorimetry (DSC) supported by X-ray powder diffractometry (XRD) was used for physicochemical characterization of NAP–AcbCd and NAP–AcgCd blends containing 0.98 to 0.40 NAP mole fraction, as well as of equimolar drug–carrier combinations prepared from the respective blends by grinding, kneading, coevaporation or freeze-drying. Phase-solubility analysis at 25, 37, and 458C was carried out in order to

0928-0987 / 02 / $ – see front matter  2002 Elsevier Science B.V. All rights reserved. PII: S0928-0987( 01 )00199-3

22

G. Bettinetti et al. / European Journal of Pharmaceutical Sciences 15 (2002) 21 – 29

evaluate the solubilizing power of AcbCd and AcgCd toward NAP and to determine the apparent stability constants of the equimolar complexes, along with the relevant thermodynamic parameters. Equimolar NAP– AcbCd and NAP–AcgCd combinations (blends and coground, kneaded, coevaporated or freeze-dried products) were tested for dissolution rate (dispersed amount and continuous flow through methods) of NAP with the aim of showing possible implications of the method of manipulation of blends on the dissolution properties of the drug.

2.3. Differential scanning calorimetry ( DSC) Temperature and enthalpy values were measured with a METTLER STARe product equipped with a DSC821 e Module on 3–5-mg (Mettler M3 Microbalance) samples in uncovered aluminium pans under static air. The heating rate was 10 K min 21 over the 30–1808C range for NAP and its combinations with AcbCd and AcgCd and the 30–3008C range for the individual AcbCd and AcgCd. Heat of fusion measurements on NAP, NAP–AcbCd and NAP–AcgCd blends were carried out in triplicate.

2. Materials and methods

2.4. Thermogravimetric analysis ( TGA)

2.1. Materials

Mass losses were recorded with a Mettler TA 4000 apparatus equipped with a TG 50 cell at the heating rate of 10 K min 21 on 7–10-mg samples in open alumina crucibles in the 30–3008C temperature range under static air.

Naproxen (NAP) ((S)-(1)-6-methoxy-a-methyl-2-naphthaleneacetic acid) from Sigma Chemical Company (St. Louis, MO, USA) was recrystallized twice from ethanol. AcbCd and AcgCd with an average degree of substitution per anhydroglucose unit, respectively, of 1.1 and 0.95, which correspond to |7.7 acetyl substituents per macrocycle, were kindly provided by Wacker Chemie GmbH ¨ (Munchen 70, G). Their physicochemical features were (a) AcbCd (CAVASOL W7 A): average molecular weight 1459; water content by TGA (see below) 2.860.2% (w / w) (three runs), corresponding to |2.3 mol H 2 O per AcbCd mol; apparent density 240 kg m 23 ; aqueous solubility 2600 21 g l at 258C and (b) AcgCd (CAVASOL W8 A): average molecular weight 1616.5; water content by TGA (see below) 3.260.2% (w / w) (three runs), corresponding to |3 mol H 2 O per AcgCd mol. All other materials and solvents were of analytical reagent grade.

2.2. Preparation of NAP–cyclodextrin binary systems Blends of NAP (,180 mm sieve granulometric fraction) with AcbCd or AcgCd (,180 mm sieve granulometric fraction) ranging from 0.98 to 0.40 NAP mole fraction were prepared by simple homogenization of the powders by turbula mixing for 10 min. The equimolar blends (|1 g) were: (a) ball-milled in a vibrational mill (Retsch, GmbH, ¨ Dusserdolf, Germany) for 60 min (coground product); (b) wetted in a mortar with the minimum volume of an ethanol–water (50 / 50 v / v) mixture and ground thoroughly with a pestle to obtain a paste which was then dried under vacuum at room temperature up to constant weight (kneaded product); (c) dissolved in an ethanol–water (50 / 50 v / v) solution which was evaporated in a rotary evaporator at 608C (coevaporated product); (d) dissolved in aqueous ammonia solutions at 2508C and 1.3310 22 mmHg and freeze-dried (Lyovac GT2, Leybold-Heraeus) (colyophilized product). The 75–150 mm sieve granulometric fractions of each preparation were collected (Mura et al., 1995) and used in the following experiments.

2.5. X-ray diffractometry ( XRD) X-ray powder diffraction patterns were taken at ambient temperature and atmosphere with a computer-controlled Philips PW 1800 / 10 apparatus equipped with a specific PC-APD software with powdered samples or tablets placed ˚ Cu in Al holders. Wavelengths: Cu K a,1 51.54060 A, ˚ K a,2 51.54439 A. Scan range: 2–508 2u. Scan speed: 0.028 21 2u s . Monochromator: graphite crystal.

2.6. Phase-solubility analysis Solubility measurements of NAP were carried out by adding 100 mg of drug to 10 ml of unbuffered water (pH|6) or aqueous solutions of AcbCd or AcgCd, respectively, in the 5–25 mmol l 21 concentration range (pH 4.8–4.5) in sealed glass containers equilibrated upon electromagnetical stirring at constant temperature (2560.58C, 3760.58C, 4560.58C) for 3 days. After equilibrium was attained (pH 4.3–4.0), aliquots were withdrawn, filtered (pore size 0.45 mm) and spectrophotometrically (Perkin Elmer Spectrophotometer Mod. 552S) assayed for drug content by a second derivative ultraviolet absorption method at 274 nm (Melani et al., 1995). Each experiment was performed in triplicate (coefficient of variation CV,5%). The apparent 1:1 binding constants (K1:1 ) of the NAP–AcbCd and NAP–AcgCd complexes were calculated from the slope of the straight lines of the phase-solubility diagrams and the aqueous solubility of NAP (Melani et al., 1995).

2.7. Dissolution rate studies Dispersed amount experiments were performed at 3760.58C by adding 60 mg of NAP or NAP equivalent

G. Bettinetti et al. / European Journal of Pharmaceutical Sciences 15 (2002) 21 – 29

23

(75–150 mm sieve granulometric fraction) to 75 ml of unbuffered water (pH|6, NAP solubility |40 mg ml 21 ) in a 100-ml beaker, where a glass three-blade propeller was centrally immersed and rotated at 100 rev. / min (non-sink conditions). At appropriate time intervals suitable aliquots were withdrawn with a filter-syringe (pore size 0.45 mm) and analyzed for drug concentration as in phase-solubility analysis. A correction was calculated for the cumulative dilution caused by replacement of the sample with an equal volume of original medium. Each test was repeated four times (coefficient of variation CV,2%). Continuous flow through experiments were carried out using a Sotax CE1 Flow Through Cell (USP apparatus 4), equipped with a 3252 Flow Cell unit specially designed for powders (Sotax AG, Switzerland). The dissolution medium (pH|6 unbuffered water deaerated by ultrasonication) was pumped

to the cell via the Sotax CY1 piston pump at a flow rate of 8 ml min 21 . The temperature of the flow cell unit was kept at 3760.58C. A bed of |600 mg glass beads (f 51 mm) was placed in the bottom of the cell and 60 mg of NAP or NAP equivalent were put into the dissolution chamber. Millipore AP25 filters (Millipore, USA) were used to filter the eluate which was continuously collected (open method, sink conditions). Samples were collected every 5 min up to 60 min and spectrophotometrically analyzed for drug concentration as in Section 2.6. The cumulative amount of drug dissolved as a function of time was calculated (Abdou, 1989). Each test was repeated three times (coefficient of variation CV,1.5%). Student’s t-test was used to

Fig. 1. DSC (upper curves) and TGA (lower curves) of individual components.

Fig. 2. DSC curves of NAP–AcbCd and NAP–AcgCd blends of various compositions (NAP mole fraction on the curves).

G. Bettinetti et al. / European Journal of Pharmaceutical Sciences 15 (2002) 21 – 29

24

compare the effect of both the preparation method and the cyclodextrin type in terms of drug dissolution efficiency (calculated according to Khan (1975)).

3. Results and discussion

3.1. Solid-state studies Thermal analysis indicated the crystalline, anhydrous state of NAP (Bettinetti et al., 1999) (T onset 5 155.960.148C, T peak 5156.860.178C, enthalpy of melting5142.561.3 J g 21 (six runs)) and the amorphous nature of both AcbCd (glass transition at T onset 5 211.9(5)8C and T midpoint 5216.7(4)8C (three runs)) and AcgCd (glass transition at T onset 5226.6(6)8C and T midpoint 5229.3(4)8C (three runs)) (Fig. 1). TGA mass losses over the 30–1008C range of |3% (w / w) for both carriers (see Section 2.1) were due to evaporation of loosely bound water, whereas those at higher temperatures (|2678C and |2778C, respectively, for AcbCd and AcgCd) to sample decomposition. As can be seen in Fig. 2 and Table 1, in the NAP–AcbCd and NAP–AcgCd systems both temperature and specific enthalpy of melting (i.e. enthalpy per drug unit mass in blends) of NAP were lower than those of the pure drug. A heating-induced drug–carrier interaction during DSC scans, which apparently results in a loss of NAP crystallinity as already observed for combinations of NAP with other amorphous Cds (Melani et al., 1995; Mura et al., 1995), can be postulated. Thermal cycles were performed on blends by heating to 2008C, cooling down to 308C, and heating a second time to 2008C (Fig. 3). Solidification and remelting of NAP, whose thermal stability and tendency to re-

crystallize from the melt are known (Bettinetti et al., 1991a), occur in combinations containing $0.81 mole fraction of NAP, which corresponds to |0.40 mass fraction of NAP. This value can be considered the solubility of NAP in amorphous AcbCd and amorphous AcgCd at room temperature. XRD studies of the NAP–AcbCd and NAP– AcgCd blends, respectively, at 0.86 and 0.88 NAP mole fraction, which correspond to the 1:1 (w / w) drug-to-carrier ratio, produced patterns that were due to the superimposition of those of crystalline drug (Melani et al., 1995) (not shown) and amorphous carrier (Fig. 4). No additional new diffraction lines were detected in the XRD patterns of blends as prepared or kept in a hot air oven at |1508C for 15 min. Therefore no apparent degradation or new compound formation had been induced by thermal treatments, in agreement with DSC data. The decrease in relative intensity of some diffraction peaks can be attributed to a parallel heating-induced decrease in NAP crystallinity as a consequence of its interaction with the carrier. Weakening of some O–H . . . O type hydrogen bonds involved in molecular packing of NAP (Kim et al., 1987), possibly mediated by elimination of water present in the carrier, can be postulated. Such an effect is particularly pronounced in extremely fine NAP crystals, that can be either dispersed onto the surface or included in the Cd’s cavity as crystallites showing a lower melting point and heat of fusion than intact NAP crystals.

3.2. Solubility and dissolution rate studies Phase-solubility diagrams (Fig. 5) and thermodynamic parameters (Table 2) accounted for specific interactions of NAP with both AcbCd and AcgCd in aqueous medium, which was accompanied by an increase in solvent entropy

Table 1 DSC data for naproxen (NAP) alone and blended with acetyl b-cyclodextrin (AcbCd) or acetyl g-cyclodextrin (AcgCd) (standard deviations in parentheses refer to last decimal digit) NAP mole fraction

T onset (8C) AcbCd

1.00 0.98 0.96 0.94 0.91 0.88 0.86 0.81 0.75 0.73 0.62 0.50 0.42 0.36 0.34 0.27 0.25

155.9(1) 153.8(5) 152.6(5) 144.1(4) 147(2) 144(2) 144.6(5) 143.0(5) 142.4(2) 144.0(8) 141.7(4)

T peak (8C) AcgCd 155.4(3) 153.1(4) 152.0(2) 148.8(5) 147.8(7) 147(1) 145.9(2) 145.6(5) 145.5(5) 146(1) 146(1)

142.5(1)

AcbCd 156.8(2) 155.6(3) 155.0(5) 152(1) 154.2(3) 152.8(9) 150.7(1) 151.8(2) 151.3(2) 152.1(2) 151.2(3)

AcgCd 156.9(2) 155.8(6) 156.0(4) 154.6(2) 154.8(5) 152.5(7) 152.9(2) 152.7(1) 152.9(3) 152.5(5) 152.9(4)

151.2(8) 146.7(2)

143(2)

Enthalpy per NAP unit mass (J g 21 ) AcbCd 143(1) 138(9) 137(4) 135(4) 134(9) 123(4) 114(4) 104(8) 101(6) 72(9) 70(2)

141(1) 136(3) 134(2) 127(1) 116(9) 121(5) 105(4) 98(4) 84(6) 66(9) 56(7)

47(9) 153.2(4)

151.4(9)

AcgCd

19(9) 17(8)

G. Bettinetti et al. / European Journal of Pharmaceutical Sciences 15 (2002) 21 – 29

25

Fig. 3. Cyclic heat-cooling DSC runs of NAP–AcbCd and NAP–AcgCd blends (NAP mole fraction on the curves). The sharp exotherms peaking at |1008C during cooling and the broad exotherms just before melting during reheating indicate NAP crystallization.

associated with hydrophobic bond formation (Mura et al., 1998). The higher affinity for NAP of native bCd with respect to gCd (and aCd) (Bettinetti et al., 1989, 1991b) was maintained in AcbCd, which shows a stability constant value (K1:1,258C 54.5(4)310 3 l mol 21 ) 5.5 times higher than that of AcgCd (K1:1,258C 50.80(7)310 3 l mol 21 ). Physicochemical characterization by DSC (Fig. 6) and XRD (Fig. 7) of differently treated equimolar NAP– AcbCd and NAP–AcgCd blends, that respectively correspond to drug-to-carrier 1 / 6.3 (w / w) and 1 / 7 (w / w) ratios, revealed the presence of fully amorphous NAP in coground and colyophilized products. In dispersed amount experiments the rank order in terms of dissolution efficiency was colyophilized product (70.861.4)4coground

Fig. 4. XRD powder patterns of AcbCd, AcgCd, and the respective 1:1 (w / w) blends with NAP as prepared or heated to |1508C in a hot air oven for 15 min.

product (55.361.1).coevaporated product (48.360.9)¯ kneaded product (46.860.9).blend (40.360.8) for the NAP–AcbCd system, and colyophilized product (47.660.8)4coground product (26.660.5). coevaporated product (20.060.4).kneaded product

26

G. Bettinetti et al. / European Journal of Pharmaceutical Sciences 15 (2002) 21 – 29

Fig. 5. Phase-solubility diagrams of NAP in aqueous solution of AcbCd (open symbols) and AcgCd (filled symbols) at 258C (h, j), 378C (s, d), and 458C (^, m).

Fig. 7. XRD powder patterns of equimolar NAP–AcbCd and NAP– AcgCd combinations. (a) Blend; (b) coground product; (c) kneaded product; (d) coevaporated product; (e) colyophilized product.

Fig. 6. DSC curves of equimolar NAP–AcbCd and NAP–AcgCd combinations. (a) Blend; (b) coground product; (c) kneaded product; (d) coevaporated product; (e) colyophilized product.

(19.360.4).blend (14.060.3) for the NAP–AcgCd system (Fig. 8). The colyophilized products showed the faster initial dissolution rates, which were characterized by a supersaturation state followed by an apparent decline in the amount of dissolved drug. The coground products, despite their fully amorphous state, showed instead dissolution profiles similar to those of other combinations (i.e. no supersaturation effect) and dissolution efficiencies distinctly lower than those of colyophilized products. Better wettability of colyophilized products and possible particle aggregation phenomena (e.g. massive aggregation of fine powders which usually accompanies the grinding of solids (Shakhtshneider and Boldyrev, 1999) in coground products) are probably responsible for the observed dissolution behaviour. An analogous trend in the rank order of dissolution efficiency was observed in the continuous flow through tests, which showed similar discriminatory abilities among the various preparations as the dispersed

G. Bettinetti et al. / European Journal of Pharmaceutical Sciences 15 (2002) 21 – 29

27

Table 2 Solubilizing efficiency (S.E.), apparent stability constants (310 3 ), and derived thermodynamic parameters for the interaction of naproxen (NAP) with acetyl b-cyclodextrin (AcbCd) and acetyl g-cyclodextrin (AcgCd) (standard deviations in parentheses refer to last decimal digit) Cyclodextrin

AcbCd AcgCd a

S.E.a

76 20

Apparent stability constant (310 3 ), K1:1 (l mol 21 ) 258C

378C

458C

4.5(4) 0.80(7)

4.1(4) 0.75(6)

3.4(3) 0.70(6)

DG8 258C kJ mol 21

DH8 kJ mol 21

DS8 258C J mol 21 K 21

221(3) 217(3)

25.9(9) 23.6(6)

51(9) 44(8)

At 258C in the presence of 0.025 mol l 21 Cd.

amount experiments and confirmed both the greatest effectiveness of colyophilized products and the better performance of AcbCd compared to AcgCd as dissolution rate enhancer for NAP (Fig. 9). The relative increases in dissolution rate of NAP were however less pronounced in the continuous flow through than in the dispersed amount test due to the higher dissolution rate of pure drug under sink conditions. Statistical comparison (unpaired t-test) of the data obtained from both sets of dissolution rate experiments showed that within each drug–Cd system the dissolution efficiency values of the various preparations were significantly different from one another (P,0.05), except for the kneaded and coevaporated products. The

dissolution efficiency values of the NAP–AcgCd preparations were all significantly lower (P,0.001) than the respective preparations based on AcbCd.

4. Conclusions The feature of ‘tailored partner’ for NAP of native bCd with respect to gCd was maintained in AcbCd, which forms an inclusion complex 5.5 times stronger than that with AcgCd. Therefore, AcbCd is a suitable solubilizing agent for NAP, which in its unionized form is poorly and erratically absorbed from solid dosage forms due to its

Fig. 8. Dispersed amount method: dissolution curves (top) and dissolution efficiencies* (bottom) of NAP and its equimolar combinations with AcbCd (A) and AcgCd (B). (d) NAP; (h) blend (BL); (m) kneaded product (KN); (j) coevaporated product (COE); (s) coground product (GR); (^) colyophilized product (COL). The dotted lines on the top plots indicate the NAP concentration attainable at the equilibrium. (*DE 60 calculated from the area under the dissolution curve at 60 min expressed as % of the area of the rectangle described by 100% dissolution in the same time (Khan, 1975)).

28

G. Bettinetti et al. / European Journal of Pharmaceutical Sciences 15 (2002) 21 – 29

Fig. 9. Continuous flow through method: dissolution curves (top) and dissolution efficiencies* (bottom) of NAP and its equimolar combinations with AcbCd (A) and AcgCd (B). (d) NAP; (h) blend (BL); (m) kneaded product (KN); (j) coevaporated product (COE); (s) coground product (GR); (^) colyophilized product (COL). (*DE 60 calculated from the area under the dissolution curve at 60 min expressed as % of the area of the rectangle described by 100% dissolution in the same time (Khan, 1975)).

extremely low aqueous solubility (|27 mg ml 21 in unbuffered water (pH|6) at 258C). The dissolution efficiency of the NAP–AcbCd colyophilized product of equimolar composition (i.e. containing |14% by weight of drug) is |23 times and |10 times that of pure drug in the dispersed amount and continuous flow through tests, respectively. Actually, the development of AcbCd-based formulations of NAP for solid dosage forms (powders, compressed tablets) is advisable from both the biopharmaceutical and technological point of view (Bettinetti et al., 2000).

Acknowledgements The authors wish to thank Dr. P. Carraro (WackerChemie Italia SpA) for his kind cooperation. Financial support from FAR (Fondo di Ateneo per la Ricerca Scientifica) is gratefully acknowledged.

References Abdou, H.A., 1989. In: Dissolution, Bioavailability and Bioequivalence. Mack Publishing Company, Easton, PA, pp. 128–130. Bettinetti, G.P., Mura, P., Liguori, A., Bramanti, G., Giordano, F., 1989. Solubilization and interaction of naproxen with cyclodextrins in aqueous solution and in the solid state. Farmaco 44, 195–213. Bettinetti, G.P., Mura, P., Giordano, F., Setti, M., 1991a. Thermal behaviour and physicochemical properties of naproxen in mixtures with polyvinylpyrrolidone. Thermochim. Acta 199, 165–171. Bettinetti, G.P., Melani, F., Mura, P., Monnanni, R., Giordano, F., 1991b. Carbon-13 NMR study of naproxen interaction with cyclodextrins in solution. J. Pharm. Sci. 80, 1162–1170. Bettinetti, G.P., Sorrenti, M., Negri, A., Mura, P., Faucci, M.T., Setti, M., 1999. Differential scanning calorimetry analysis of crystallinity changes of naproxen in ground mixtures with maltohexaose, the non cyclic analog of alpha-cyclodextrin. In: Torres Labandeira, J.J., VilaJato, J.L. (Eds.), Proceedings 9th International Symposium on Cyclodextrins. Kluwer Academic, The Netherlands, pp. 367–370. Bettinetti, G.P., Rossi, S., Ferrari, F., Sorrenti, M., Carraro, P., 2000. Solid-state interactions of naproxen with noncrystalline acetyl b- and g-cyclodextrins: biopharmaceutical and technological implications. In:

G. Bettinetti et al. / European Journal of Pharmaceutical Sciences 15 (2002) 21 – 29 Szejtli, J. (Ed.), Proceedings 10th International Cyclodextrin Symposium CD, in press. Hirayama, F., Yamanaka, M., Horikawa, T., Uekama, K., 1995. Characterization of peracylated b-cyclodextrins with different chain lengths as a novel sustained release carrier for water-soluble drugs. Chem. Pharm. Bull. 43, 130–136. Hirayama, F., Mieda, S., Miyamoto, Y., Arima, H., Uekama, K., 1999. Heptakis(2,6-di-O-methyl-3-O-acetyl)-b-cyclodextrin: a water-soluble cyclodextrin derivative with low hemolytic activity. J. Pharm. Sci. 88, 970–975. Khan, K.A., 1975. The concept of dissolution efficiency. J. Pharm. Pharmacol. 27, 48–49. Kim, Y.B., Song, H.J., Park, I.Y., 1987. Refinement of the structure of naproxen, (1)-6-methoxy-a-methyl-2-naphthaleneacetic acid. Arch. Pharm. Res. 10, 232–238. Liu, F.-Yu., Kildsig, D.O., Mitra, A.K., 1992. Complexation of 6-acyl-Ob-cyclodextrin derivatives with steroids — Effects of chain length and substitution degree. Drug Dev. Ind. Pharm. 18, 1599–1612. Matsubara, K., Irie, T., Uekama, K., 1994. Controlled release of the LHRH agonist buserelin acetate from injectable suspensions containing triacetylated cyclodextrins in an oil vehicle. J. Control. Release 31, 173–180. Melani, F., Bettinetti, G.P., Mura, P., Manderioli, A., 1995. Interaction of naproxen with a-, b-, and g-hydroxypropyl cyclodextrins in solution and in the solid state. J. Incl. Phenom. 22, 131–143.

29

Mura, P., Bettinetti, G.P., Melani, F., Manderioli, A., 1995. Interaction between naproxen and chemically-modified b-cyclodextrins in the liquid and solid state. Eur. J. Pharm. Sci. 3, 347–355. Mura, P., Bettinetti, G.P., Manderioli, A., Faucci, M.T., Bramanti, G., Sorrenti, M., 1998. Interactions of ketoprofen and ibuprofen with b-cyclodextrins in solution and in the solid state. Int. J. Pharm. 166, 189–203. Nakanishi, K., Masukawa, T., Nadai, T., Yoshii, K., Okada, S., Miyajima, K., 1997. Sustained release of flufenamic acid from a drug-triacetyl-bcyclodextrin complex. Biol. Pharm. Bull. 20, 66–70. Nakanishi, K., Masatoshi, N., Masukawa, T., Nadai, T., Miyajima, K., 1998. Sustained release and intestinal absorption of drug from the hydrophobic a-, b- and g-cyclodextrin complexes. In: Torres Labandeira, J.J., Vila-Jato, J.L. (Eds.), Proceedings 9th International Symposium on Cyclodextrins. Kluwer Academic, The Netherlands, pp. 317–320. Shakhtshneider, T.P., Boldyrev, V.V., 1999. Mechanochemical synthesis and mechanical activation of drugs. In: Reactivity of Molecular Solids. The Molecular Solid State, Vol. 3. J. Wiley, p. 271. Szejtli, J., 1994. Medicinal applications of cyclodextrins. Med. Res. Rew. 14, 353–386. Uekama, K., Horikawa, T., Yamanaka, M., Hirayama, F., 1994. Peracylated b-cyclodextrins as novel sustained-release carriers for a water-soluble drug, Molsidomine. J. Pharm. Pharmacol. 46, 714–717.