- Email: [email protected]
In vitro skin permeation of Sinitrodil, a member of a new class of nitrovasodilator drugs
European Journal of Pharmaceutical Sciences, 7 (1999) 231–236
In vitro skin permeation of Sinitrodil, a member of a new class of nitrovasodilator drugs a, a a b P. Minghetti *, A. Casiraghi , L. Montanari , M.V. Monzani a
Istituto di Chimica Farmaceutica e Tossicologica, Universita` degli Studi di Milano, viale Abruzzi 42, 20131 Milano, Italia b Italfarmaco Research Centre, Via dei Lavoratori 54, 20092 Cinisello B. ( MI), Italia Received 27 October 1997; accepted 11 May 1998
Abstract Clinical trials have shown the potential of benzoxazinones, a new class of organic nitrates, in cardiovascular therapy. In particular Sinitrodil possesses a coronary vascular selectivity greater than that of Nitroglycerin and Isosorbide dinitrate. The objective of this study was a preliminary evaluation of the ability of these new organic nitrate derivatives to reach therapeutical steady-state plasma concentrations following a transdermal administration. In vitro permeation studies through human stratum corneum and epidermis have been conducted on two members of this class: Sinitrodil (ITF 296) and ITF 1129. Comparative studies have also been carried out with Nitroglycerin, Isosorbide dinitrate and Nicorandil. Two different fixed concentrations were tested: 0.08% w / v solution and saturated solution. Sinitrodil could be considered a good candidate for transdermal administration on the basis of the in vitro permeation results and of the known therapeutical plasma concentration. 1999 Published by Elsevier Science B.V. All rights reserved. Keywords: Sinitrodil; ITF 296; Skin permeation; Benzoxazinones
1. Introduction For many years Nitroglycerin (NTG) and other organic nitrates have been the mainstay of cardiovascular therapy (Abrams, 1985) and of particular benefit in the treatment of angina pectoris (Gunnar and Fisch, 1990), unstable angina (Parker, 1987) and the early stages of acute myocardial infarction (Conti, 1987). They have shown real efficacy in coronary atherosclerosis, hypercholesterolemia or other associated blood vessel wall disorders involving endothelial dysfunction (McGregor, 1983) and reduced vasodilator capacity of the coronary arteries (Williams et al., 1975). This group of drugs, termed nitrovasodilators, has a unique spectrum of activity, in that they preferentially dilate, at low doses, the large coronary conductance vessels and, as the dose is increased, they also dilate resistance vessels. Members of a new, recently described (Benedini et al., 1995) class of organic nitrates (benzoxazinones) have been shown to exert marked effects on large coronary arteries at doses below those active on peripheral vascular resistances (Fitzgerald, 1995). Among these compounds, the best pharmacological and toxicological results were obtained by *Corresponding author. Tel.: 139 2 29405002; fax; 139 2 29514197; e-mail: [email protected] 0928-0987 / 99 / $ – see front matter PII: S0928-0987( 98 )00030-X
3-[2-(nitrooxy) ethyl]-2H-1,3 benzoxazin 4(3H)-one Sinitrodil (ITF 296) and 6-methyl-3-[2-(nitrooxy) ethyl]-2H-1,3 benzoxazin 4(3H)-one, ITF 1129 (Fig. 1). In particular ITF 296 possesses a coronary and peripheral vascular selectivity greater than that of Nitroglycerin and Isosorbide dinitrate (Fitzgerald, 1995). Studies with ITF 296 in healthy volunteers confirmed a dose-dependent reduction of systolic and diastolic blood pressure with a minor increase in heart rate and a better tollerability than other nitrates. Although the oral pharmacokinetic profile of ITF 296 showed a first-pass effect less marked than other organic nitrates, this phenomenon was still important (Monzani et al., 1998). The objective of this study was a preliminary evaluation of the ability of these new organic nitrate derivatives to reach therapeutical steady-state plasma concentrations following a transdermal administration, considering this route suitable for the administration of these drugs. In vitro permeation studies through human stratum corneum and epidermis were carried out on ITF 296 and ITF 1129 and, under the same experimental conditions, on Nitroglycerin (NTG), Isosorbide dinitrate (ISDN) and Nicorandil (NR), three of the most widely used organic nitrates.
1999 Published by Elsevier Science B.V. All rights reserved.
P. Minghetti et al. / European Journal of Pharmaceutical Sciences 7 (1999) 231 – 236
232
Italy). All substances and other chemicals were used as obtained.
2.2. Methods 2.2.1. Determination of drug solubility ( S) The solubility of each drug in water:propylene glycol (80:20 v / v) was obtained by equilibrating large excess of the solute and vehicle for 24 h. Each solution was stirred vigorously with a magnetic bar throughout the experiment and the temperature was maintained at 32618C. After equilibration, the sample was filtered quickly with a membrane filter (pore size 0.22 mm, Millex -GV Millipore, Ireland). The filtrate was adequately diluted with distilled water and analysed by HPLC, with the method described below.
Fig. 1. Chemical structures of the tested organic nitrates.
Two different concentrations were tested: 0.08% w / v solution and saturated solution. The first one was chosen to compare the permeation profile of the drugs at the same concentration. Then the saturated solutions were used to compare the maximum steady-state flux obtained in vitro from the same vehicle. As vehicle a mixture of water and propylene glycol (80:20 v / v) was used. Propylene glycol was added to water to increase the solubility of the considered drugs. It is well known that it could alter the barrier properties of the skin. However at the considered concentration the propylene glycol should not increase in a very significant way the flux of the drugs (Megrab et al., 1995a; Goodman and Barry, 1992). The in vitro permeation data together with the solubility and the systemic clearance of each substance allow to calculate a preliminary predicted plasma steady-state concentration that can be compared with the therapeutical plasma concentration. This information can be useful to assess the possibility of transdermal administration.
2. Materials and methods
2.1. Materials ITF 296 and ITF 1129 were synthesised at Italfarmaco (Milano, Italy). Nitroglycerin (NTG) and Isosorbide dinitrate (ISDN) adsorbed on lactose (respectively 10% w / w and 40% w / w) were obtained from Dipharma (Udine, Italy). Nicorandil (NR) was a gift from Bracco (Milano,
2.2.2. Determination of partition coefficient ( P) The partition coefficient of ITF 296 was determined in octanol / pH 7.4 phosphate-buffered saline solution (PBS, Ph. Eur., ed. 1997, p. 274). The shake flask method was used. After vigorous mixing for 2 h at room temperature the phases were centrifuged, appropriately diluted and the aqueous phase was assayed by HPLC, with the method described below. The partition coefficient of ITF 1129 was calculated adding the contribution of the methyl group on the aromatic ring (Krogsgaard-Larsenand and Bundgaard, 1991) to the experimental value of ITF 296. 2.2.3. In vitro permeation studies Human abdomen skin, obtained with surgical operation, has been used within 24 h from removal. Human stratum corneum and epidermis (SCE) was prepared by immersion of the skin in distilled water at 60618C for 1 min and peeling it from the derma. SCE membranes were dried in a desiccator at approximately 25% RH, wrapped in aluminium foil and stored at 4618C until use. Dried SCE samples were rehydrated at room temperature by immersion in distilled water for about 16 h before use. Each membrane was carefully mounted on a modified Franztype diffusion cell of approximately 5 ml receiver capacity and fastened with a rigid clamp. These cells had a diffusion area of 0.636 cm 2 . Each cell was individually calibrated with respect to its receiver volume and diffusion area. At the start of the experiment, 1 ml of water:propylene glycol (80:20 v / v) solution containing 0.08% w / v of one of the tested drugs (first series) or a saturated solution with large excess of the same drugs (second series) was applied to the diffusion cell as donor phase. In the first series of experiments, ITF 1129 was not used because of its low solubility in the vehicle, while the second series was performed except for the NTG because of its potential danger. The receiver medium was constituted with
P. Minghetti et al. / European Journal of Pharmaceutical Sciences 7 (1999) 231 – 236
PBS:propylene glycol (80:20 v / v) solution containing 100 mg / ml streptomycin (Sigma Chem. Co, USA) as preservative. The receiver medium was continuously stirred with a small magnetic bar and thermostated at 37618C, so that the skin surface temperature was 32618C. At predetermined times (0.5, 1, 2, 3, 4, 5, 6, 7, 8 and 24 h) 0.2 ml samples were withdrawn from the receiver compartment and replaced with fresh receiver medium. Degradation products in solution were less than 5% after 24 h. Sink conditions were maintained throughout the experiments. Samples were analysed by HPLC, with the method described below. Each value represents the average of three sample readings.
2.2.4. Drug assay The HPLC system was equipped with a high-pressure pump (320 System–Kontron Instruments, Italy), an autosampler (Autosampler 460–Kontron Instruments, Italy), a variable wavelength UV detector (HP 1100 Series–Hewlett Packard, USA) and an integrator-recorder (HP 3349–Hewlett Packard, USA). Samples (20 ml) were injected at room temperature on a reverse-phase column (C18, 5 mm Spherisorb ODS2, 20 cm–Shandon HPLC, UK.). The chromatographic conditions (mobile phase composition, flow rate and UV wavelength) were adjusted in order to obtain the best separation for each compound and are reported in Table 1. All the mobile phases were filtered and degassed before using. The standard curve was constructed by plotting the peak area against the concentration of six known concentrations of substance ranging from 1–100 mg / ml and prepared using HPLC-grade methanol. 2.2.5. Data analysis The cumulative amount permeated through the SCE per unit area was calculated from the concentration of each substance in the receiving medium and plotted as a function of time. Each data point on the plot represents a mean of triplicate permeation experiments. The flux (J) was determined as the slope of the linear portion of the plot.
Drug
Mobile phase
Flow rate (ml / min)
Wavelength (nm)
ITF 296
Methanol:water:tetrahydrofuran (54:43:3) Methanol:water:tetrahydrofuran (54:43:3) Methanol:water:tetrahydrofuran (54:43:3) Methanol:water (45:55) Water:acetonitrile (60:40)
0.9
264
1.2
318
0.9
220
0.9
220
0.9
254
NTG ISDN NR
The permeability coefficient was calculated following the Fick’s first law of diffusion: Kp 5 Jmax /S, where Kp is the permeability coefficient (cm / h), Jmax is the maximum flux (mg / cm 2 / h) and S is the drug donor concentration, corresponding to the drug solubility in the vehicle (mg / ml). The predicted plasma steady-state concentration was obtained using the following equation (Calpena et al., 1994): Css 5 Jmax ? TTS area / Cl where Css is the plasma steady-state concentration (mg / l), TTS area is the hypothetical area of the transdermal therapeutic system (cm 2 ) and Cl is the systemic clearance after iv administration (l / h).
3. Results and discussion In Table 2 are reported the relevant physicochemical parameters of the tested substances. It is well documented that the molecular weight (MW) of a substance directly affects its diffusion across simple or complex membranes (Potts and Guy, 1992). Because of the narrow MW range of the selected compounds it is possible to presume that this parameter is not significant for the explanation of the difference of permeation observed through human skin. In Table 2 are also reported the melting point (MP) of the compounds, as this parameter is often related to the transdermal transport of the drug (Izumoto et al., 1992). On the other hand, the studied compounds presented large differences in their hydrophobicity (Log P). Considering that the chemical structures of ITF 296 and ITF 1129 are similar, it was noteworthy the relevant increase of the hydrophobicity due to insertion of a methyl group on the aromatic ring of ITF 296. This simple modification also caused a strong decrease of the solubility of ITF 1129 in the selected vehicle because of its high hydrophobicity. Table 2 Relevant physicochemical parameters of the drugs
Table 1 HLPC conditions for the analysis of drugs
ITF 1129
233
Drug
MW (g / mole)
MP (8C)
Log P
S (mg / ml)
ITF 296 ITF 1129 ISDN NR NTG
238 252 236 211 227
55.5–56.5 80–81 70 92–93 13.5
1.85 2.41 1.34 a 21.02 b 2.05 a
2219* 505* 1415* 35747* 1730**
Note: molecular weight (MW), melting point (MP), log partition coefficient (Log P), solubility (S). a Drayton, 1990. b Hatanaka et al., 1992. * Solubility in water:propylene glycol (80:20 v / v) at 32618C. ** Solubility in water at 208C (Mark, 1965).
234
P. Minghetti et al. / European Journal of Pharmaceutical Sciences 7 (1999) 231 – 236
Fig. 2. Permeation profiles of the drugs through SCE (diluted solution) (n53; error bars are mean6S.D.).
The cumulative amounts of drugs permeated in 24 h through SCE prepared from the same human donor are shown in Figs. 2 and 3, respectively related to the experiments conducted using as donor phase the diluted solution (0.08% w / v) and the saturated one. A mixture water:propylene glycol (80:20 v / v) was used as donor vehicle to obtain a higher thermodynamic activity than the value of the pure water. The addition of this amount of propylene glycol, that can be considered as an enhacer, should not alter in a significant way the barrier properties of the human SCE (Megrab et al., 1995b). A steady-state skin flux was attained within the first hours of application of the drug solutions as reflected by the short diffusion lag time. No significant drug depletion from the donor compartment was observed, and the steadystate skin flux was maintained over the entire period of the diffusion experiment (24 h). The NR permeation profile, when the diluted solution was used, is not reported in Fig. 2, as the amount permeated was not detectable. When the diluted solution at the same concentration were used, the amount permeated increased in order of ISDN.NTG.ITF 296. In the second phase of the study the permeation experiments were performed using saturated drug solutions to ensure unit thermodynamic activity (Fig. 3). In this case NR permeated through the SCE even if in small amounts.
In the case of the ITF substances the permeation studies have been conducted with SCE from two different human donors. The flux values differed of a very small amount for ITF 296 (3.5260.18; 3.3660.61), while the variation was bigger for ITF 1129, even if it was still not significant (8.7761.63; 5.4861.02). Comparing data obtained with SCE of the same donor, ITF 296 showed the highest flux and consequently the highest amount permeated at 24 h in both the experiments (diluted or saturated donor solutions), as shown in Table 3. ITF 296 presented the best balance between hydrophobicity and solubility in this group of organic nitrates. Although ITF 1129 showed the highest permeability coefficient due to its high hydrophobicity it was characterised by lower Jmax compared with ITF 296 and ISDN because of its low thermodynamic activity in the used vehicle. On the other hand Nicorandil, which is very soluble in the donor phase, showed a very low Jmax due to its low hydrophobicity that affected the permeability coefficient of the drug. In the case of ITF 296 and ISDN the Kp were calculated from either the data of the diluted solutions (ITF 296 Kp 54.260.76310 23 ; ISDN Kp 52.3160.07310 23 ) or the saturated solutions (Table 3). The numbers are in reasonable agreement that is within the expected limits given that the saturated solutions will not be behaving as if they were ideal. Delivery of Nitroglycerin through human skin has been shown to well correlate with the in vivo data (Hadgraft et al., 1993). Furthermore it has been suggested that the same correlation may apply to other substances provided that there is no significant difference in the transfer rate of the drug through dead skin compared to viable skin. It is therefore necessary an in vivo study to evaluate the skin metabolism of ITF 296 in order to ascertain that the compounds do not encounter metabolic events through the viable skin. The therapeutical plasma concentrations (Ct ) of ITF 296 were estimated considering the effects of the drug on arterial compliance. A dose-response study showed that significant effects on this parameter are obtained in the Table 3 Permeation parameters of the drugs obtained from diluted (0.08% w / v) or saturated solutions Drug
ITF 296 ITF 1129 NTG ISDN NR Fig. 3. Permeation profiles of the drugs through SCE (saturated solution) (n53; error bars are mean6S.D.).
Diluted solution
Saturated solution
J (mg cm 22 h 21 )
Jmax (mg cm 22 h 21 )
Kp 310 3 (cm h 21 )
3.3660.61 – 2.8560.51 1.8560.06 –
7.9161.14 5.4861.02 6.1661.09* 6.3661.35 1.8760.41
3.5660.96 10.8562.02 – 4.4960.95 0.0560.01
Note: flux (J), maximum flux (Jmax ), permeability coefficient (Kp ). * Calculated value. (Jmax 5Kp ?S; where S is solubility.)
P. Minghetti et al. / European Journal of Pharmaceutical Sciences 7 (1999) 231 – 236 Table 4 Calculated plasma steady-state concentration of the drugs from experimental permeation studies: Css 5 Jmax ?TTS area / Cl Drug
Cl (l / h)
Css (ng / ml)
Ct /Css
ITF 296 ISDN NR NTG
138.6 a 224.4 b 52.2 c 300 d
2.0260.37 0.4460.01 1.0760.23 0.6160.11*
2.47 10.23 140.19 1.15
Note: Plasma steady-state concentrations (Css ), clearance (Cl), therapeutical plasma concentrations (Ct ). * Css was obtained using calculated Jmax . a Monzani et al., 1998. b Abshagen et al., 1985. c Frydman, 1992. d McNiff et al., 1981.
presence of plasma concentrations below 5 ng / ml (Sardina, 1996). The therapeutical plasma levels of NTG are between 0.2 and 1.2 ng / ml (Rossi and Bolognese, 1989), those of ISDN are between 1 and 8 ng / ml (Cohn and Rittinghausen, 1985) and those of NR between 100 and 200 ng / ml (Frydman, 1992). Since the in vitro SCE permeation data allow to estimate the rate of input of the drug into the systemic circulation, it should be possible to evaluate the achievable plasma levels combining the in vitro permeability data with the systemic clearance of the drug reported in the literature. Assuming for the TTS area a reasonable limit of 30 cm 2 and knowing the clearance value it is possible to estimate the plasma steady-state concentrations (Css ), from the in vitro experiments conducted with the human skin from the same donor (Table 4). Css could not be calculated for ITF 1129 because its clearance in humans is unknown. ITF 296 and NTG estimated steady-state plasma concentrations that were very similar to their respective therapeutic concentrations. The value of the Ct /Css ratio has been reported as it could be useful as an indication of the suitability of the drug for transdermal administration. ISDN, which presented a Ct /Css ratio of 10.23, is still suitable for the transdermal administration, as demonstrated by the presence on the market of TTS containing the drug. On the contrary the predicted steady-state plasma concentration was considerably less than the therapeutic concentration for NR, as indicated by the high Ct /Css ratio (Table 4).
4. Conclusion The considered benzoxazinones, ITF 296 and ITF 1129, showed promising in vitro skin permeability. The different permeation behaviour of these compounds is particularly determined by the contribution of both their solubility in the vehicle and hydrophobic characteristic. The values of Log P of both compounds are in the range of
235
values that are generally considered suitable for transdermal administration. ITF 296, that possesses the largest solubility in the vehicle, showed the highest flux and the highest amount permeated at 24 h either when diluted or saturated solutions were used as donor phase. The value of the Ct /Css ratio for ITF 296 was 2.47. As this value is in the same order of magnitude of Nitroglycerin ratio (1.15), it is a positive indication that ITF 296 is suitable for transdermal administration. Therefore, ITF 296 appears to be a good candidate for the development of transdermal patch. However, the flux of ITF 296 from the patch through the human skin might be different from the flux obtained in this work. A larger flux might be obtained due to the solubility of ITF 296 in the selected matrix, to the occlusive conditions obtained with TTS and to the possibility of adding adequate enhancers.
References Abrams, J., 1985. Haemodynamic effects of nitroglycerin and long-acting nitrates. Am. Heart J. 110, 216–314. Abshagen, U., Betzien, G., Endele, R., Kaufmann, B., Neugebauer, G., 1985. Pharmacokinetics and metabolism of isosorbide-dinitrate after intravenous and oral administration. Eur. J. Clin. Pharmacol. 27, 637–644. Benedini, F., Bertolini, G., Cereda, R., et al., 1995. New anti-anginal nitroesters with reduced hypotensive activity. Synthesis and pharmacological evaluation of 3-nitrooxyalkyl-2H-1,3-benzoxazin-4(3H)-ones. J. Med. Chem. 38, 130–136. Calpena, A.C., Blanes, C., Moreno, J., Oback, R., Domenech, J., 1994. A comparative in vitro study of transdermal absorption of antiemetics. J. Pharm. Sci. 83 (1), 29–33. Cohn, J., Rittinghausen, R., 1985. Mononitrate. Springer Verlag, Berlin Heidelberg New York Tokyo, pp. 147–153. Conti, C.R., 1987. Use of nitrates in unstable angina pectoris. Am. J. Cardiol. 60, 31H–34H. Drayton, C.J., 1990. Comprehensive Medicinal Chemistry. Pergamon Press, Oxford UK, pp. 6, 608, 715. Fitzgerald, D., 1995. Pharmacologic properties of new nitric oxide donor ITF 296. J. Cardiovasc. Pharmacol. 26 (supp. 4), S1–S66. Frydman, A., 1992. Pharmacokinetic profile of nicorandil in humans: an overview. J. Cardiovasc. Pharmacol. 20 (3), S34–S44. Goodman, M., Barry, B.W., 1992. Action of penetration enhancer on human skin as assessed by the permeation of model drug 5-fluorouracil and oestradiol. I. Infinite dose technique. J. Invest. Dermatol. 91, 323–327. Gunnar, R.M., Fisch, C., 1990. Special Report: ACC /AHA guidelines for the early management of patients with acute myocardial infarction. Circulation 82, 664–707. Hadgraft, J., Beutner, D., Wolff, H.M., 1993. In vitro–in vivo comparisons of the transdermal delivery of nitroglycerin. Intl. J. Pharm. 89, R1–R4. Hatanaka, T., Inuma, M., Sugibayashi, K., Morimoto, Y., 1992. Prediction of skin permeability of drugs. II. Development of composite membrane as a skin alternative. Intl. J. Pharm. 79, 21–28. Izumoto, T., Aioi, A., Uenoyama, K., Azuma, M., 1992. Relationship between the transference of a drug from a transdermal patch and the physicochemical properties. Chem. Pharm. Bull. 40 (2), 456–458.
236
P. Minghetti et al. / European Journal of Pharmaceutical Sciences 7 (1999) 231 – 236
Krogsgaard-Larsenand, P., Bundgaard, M., 1991. A Textbook of Drug Design and Development. Harwood Academy, Chur, Switzerland. Mark (Ed.), 1965. Kirk-Othmer Encyclopedia of Chemical Technology, 2nd ed., vol. 8. Interscience Publishers, New York, USA. McGregor, M., 1983. Pathogenesis of angina pectoris and role of nitrates in relief of myocardial ischemia. Am. J. Cardiol. 74 (6B), 21–27. McNiff, E.F., Jacobi, A., Young-Chang, F.M., Golden, L.M., Goldfarb, A., Ho-Leung, F., 1981. Nitroglycerin pharmacokinetics after intravenous infusion in normal subjects. J. Pharm. Sci. 70, 1054–1058. Megrab, N.A., Williams, A.C., Barry, B.W., 1995a. Oestradiol permeation through human skin and silastic membrane: effetcs of propylene glycol and supersaturation. J Control. Release 36, 277–294. Megrab, N.A., Williams, A.C., Barry, B.W., 1995b. Oestradiol permeation across human skin, silastic and snake skin membranes: the effects of ethanol / water co-solvent systems. Intl. J. Pharm. 116, 101–112.
Monzani, M.V., Coltro, G., Sala, A., Sardina, M., 1998. Pharmacokinetic of ITF 296 (Sinitrodil) a novel organic nitrate, in healthy volunteers. Eur. J. Pharm. Sci., in press. Parker, J.O., 1987. Nitrate therapy in stable angina pectoris. New Engl. J. Med. 316, 1635–1642. Potts, R.O., Guy, R.H., 1992. Predicting skin permeability. Pharm. Res. 9 (5), 663–669. Rossi, P., Bolognese, L., 1989. I nitrati in cardiologia. Ed. Centro Scientifico Torinese. Sardina, M., 1996. ITF 296–Italfarmaco Investigator’s Brochure, Edition No. 2. Williams, D.O. Amsterdam, E.A., Mason, D.T., 1975. Haemodynamic effects of nitroglycerin in acute myocardial infarction: decrease in ventricular preload at the expense of cardiac output. Circulation 51, 421–427.
In vitro skin permeation of Sinitrodil, a member of a new class of nitrovasodilator drugs a, a a b P. Minghetti *, A. Casiraghi , L. Montanari , M.V. Monzani a
Istituto di Chimica Farmaceutica e Tossicologica, Universita` degli Studi di Milano, viale Abruzzi 42, 20131 Milano, Italia b Italfarmaco Research Centre, Via dei Lavoratori 54, 20092 Cinisello B. ( MI), Italia Received 27 October 1997; accepted 11 May 1998
Abstract Clinical trials have shown the potential of benzoxazinones, a new class of organic nitrates, in cardiovascular therapy. In particular Sinitrodil possesses a coronary vascular selectivity greater than that of Nitroglycerin and Isosorbide dinitrate. The objective of this study was a preliminary evaluation of the ability of these new organic nitrate derivatives to reach therapeutical steady-state plasma concentrations following a transdermal administration. In vitro permeation studies through human stratum corneum and epidermis have been conducted on two members of this class: Sinitrodil (ITF 296) and ITF 1129. Comparative studies have also been carried out with Nitroglycerin, Isosorbide dinitrate and Nicorandil. Two different fixed concentrations were tested: 0.08% w / v solution and saturated solution. Sinitrodil could be considered a good candidate for transdermal administration on the basis of the in vitro permeation results and of the known therapeutical plasma concentration. 1999 Published by Elsevier Science B.V. All rights reserved. Keywords: Sinitrodil; ITF 296; Skin permeation; Benzoxazinones
1. Introduction For many years Nitroglycerin (NTG) and other organic nitrates have been the mainstay of cardiovascular therapy (Abrams, 1985) and of particular benefit in the treatment of angina pectoris (Gunnar and Fisch, 1990), unstable angina (Parker, 1987) and the early stages of acute myocardial infarction (Conti, 1987). They have shown real efficacy in coronary atherosclerosis, hypercholesterolemia or other associated blood vessel wall disorders involving endothelial dysfunction (McGregor, 1983) and reduced vasodilator capacity of the coronary arteries (Williams et al., 1975). This group of drugs, termed nitrovasodilators, has a unique spectrum of activity, in that they preferentially dilate, at low doses, the large coronary conductance vessels and, as the dose is increased, they also dilate resistance vessels. Members of a new, recently described (Benedini et al., 1995) class of organic nitrates (benzoxazinones) have been shown to exert marked effects on large coronary arteries at doses below those active on peripheral vascular resistances (Fitzgerald, 1995). Among these compounds, the best pharmacological and toxicological results were obtained by *Corresponding author. Tel.: 139 2 29405002; fax; 139 2 29514197; e-mail: [email protected] 0928-0987 / 99 / $ – see front matter PII: S0928-0987( 98 )00030-X
3-[2-(nitrooxy) ethyl]-2H-1,3 benzoxazin 4(3H)-one Sinitrodil (ITF 296) and 6-methyl-3-[2-(nitrooxy) ethyl]-2H-1,3 benzoxazin 4(3H)-one, ITF 1129 (Fig. 1). In particular ITF 296 possesses a coronary and peripheral vascular selectivity greater than that of Nitroglycerin and Isosorbide dinitrate (Fitzgerald, 1995). Studies with ITF 296 in healthy volunteers confirmed a dose-dependent reduction of systolic and diastolic blood pressure with a minor increase in heart rate and a better tollerability than other nitrates. Although the oral pharmacokinetic profile of ITF 296 showed a first-pass effect less marked than other organic nitrates, this phenomenon was still important (Monzani et al., 1998). The objective of this study was a preliminary evaluation of the ability of these new organic nitrate derivatives to reach therapeutical steady-state plasma concentrations following a transdermal administration, considering this route suitable for the administration of these drugs. In vitro permeation studies through human stratum corneum and epidermis were carried out on ITF 296 and ITF 1129 and, under the same experimental conditions, on Nitroglycerin (NTG), Isosorbide dinitrate (ISDN) and Nicorandil (NR), three of the most widely used organic nitrates.
1999 Published by Elsevier Science B.V. All rights reserved.
P. Minghetti et al. / European Journal of Pharmaceutical Sciences 7 (1999) 231 – 236
232
Italy). All substances and other chemicals were used as obtained.
2.2. Methods 2.2.1. Determination of drug solubility ( S) The solubility of each drug in water:propylene glycol (80:20 v / v) was obtained by equilibrating large excess of the solute and vehicle for 24 h. Each solution was stirred vigorously with a magnetic bar throughout the experiment and the temperature was maintained at 32618C. After equilibration, the sample was filtered quickly with a membrane filter (pore size 0.22 mm, Millex -GV Millipore, Ireland). The filtrate was adequately diluted with distilled water and analysed by HPLC, with the method described below.
Fig. 1. Chemical structures of the tested organic nitrates.
Two different concentrations were tested: 0.08% w / v solution and saturated solution. The first one was chosen to compare the permeation profile of the drugs at the same concentration. Then the saturated solutions were used to compare the maximum steady-state flux obtained in vitro from the same vehicle. As vehicle a mixture of water and propylene glycol (80:20 v / v) was used. Propylene glycol was added to water to increase the solubility of the considered drugs. It is well known that it could alter the barrier properties of the skin. However at the considered concentration the propylene glycol should not increase in a very significant way the flux of the drugs (Megrab et al., 1995a; Goodman and Barry, 1992). The in vitro permeation data together with the solubility and the systemic clearance of each substance allow to calculate a preliminary predicted plasma steady-state concentration that can be compared with the therapeutical plasma concentration. This information can be useful to assess the possibility of transdermal administration.
2. Materials and methods
2.1. Materials ITF 296 and ITF 1129 were synthesised at Italfarmaco (Milano, Italy). Nitroglycerin (NTG) and Isosorbide dinitrate (ISDN) adsorbed on lactose (respectively 10% w / w and 40% w / w) were obtained from Dipharma (Udine, Italy). Nicorandil (NR) was a gift from Bracco (Milano,
2.2.2. Determination of partition coefficient ( P) The partition coefficient of ITF 296 was determined in octanol / pH 7.4 phosphate-buffered saline solution (PBS, Ph. Eur., ed. 1997, p. 274). The shake flask method was used. After vigorous mixing for 2 h at room temperature the phases were centrifuged, appropriately diluted and the aqueous phase was assayed by HPLC, with the method described below. The partition coefficient of ITF 1129 was calculated adding the contribution of the methyl group on the aromatic ring (Krogsgaard-Larsenand and Bundgaard, 1991) to the experimental value of ITF 296. 2.2.3. In vitro permeation studies Human abdomen skin, obtained with surgical operation, has been used within 24 h from removal. Human stratum corneum and epidermis (SCE) was prepared by immersion of the skin in distilled water at 60618C for 1 min and peeling it from the derma. SCE membranes were dried in a desiccator at approximately 25% RH, wrapped in aluminium foil and stored at 4618C until use. Dried SCE samples were rehydrated at room temperature by immersion in distilled water for about 16 h before use. Each membrane was carefully mounted on a modified Franztype diffusion cell of approximately 5 ml receiver capacity and fastened with a rigid clamp. These cells had a diffusion area of 0.636 cm 2 . Each cell was individually calibrated with respect to its receiver volume and diffusion area. At the start of the experiment, 1 ml of water:propylene glycol (80:20 v / v) solution containing 0.08% w / v of one of the tested drugs (first series) or a saturated solution with large excess of the same drugs (second series) was applied to the diffusion cell as donor phase. In the first series of experiments, ITF 1129 was not used because of its low solubility in the vehicle, while the second series was performed except for the NTG because of its potential danger. The receiver medium was constituted with
P. Minghetti et al. / European Journal of Pharmaceutical Sciences 7 (1999) 231 – 236
PBS:propylene glycol (80:20 v / v) solution containing 100 mg / ml streptomycin (Sigma Chem. Co, USA) as preservative. The receiver medium was continuously stirred with a small magnetic bar and thermostated at 37618C, so that the skin surface temperature was 32618C. At predetermined times (0.5, 1, 2, 3, 4, 5, 6, 7, 8 and 24 h) 0.2 ml samples were withdrawn from the receiver compartment and replaced with fresh receiver medium. Degradation products in solution were less than 5% after 24 h. Sink conditions were maintained throughout the experiments. Samples were analysed by HPLC, with the method described below. Each value represents the average of three sample readings.
2.2.4. Drug assay The HPLC system was equipped with a high-pressure pump (320 System–Kontron Instruments, Italy), an autosampler (Autosampler 460–Kontron Instruments, Italy), a variable wavelength UV detector (HP 1100 Series–Hewlett Packard, USA) and an integrator-recorder (HP 3349–Hewlett Packard, USA). Samples (20 ml) were injected at room temperature on a reverse-phase column (C18, 5 mm Spherisorb ODS2, 20 cm–Shandon HPLC, UK.). The chromatographic conditions (mobile phase composition, flow rate and UV wavelength) were adjusted in order to obtain the best separation for each compound and are reported in Table 1. All the mobile phases were filtered and degassed before using. The standard curve was constructed by plotting the peak area against the concentration of six known concentrations of substance ranging from 1–100 mg / ml and prepared using HPLC-grade methanol. 2.2.5. Data analysis The cumulative amount permeated through the SCE per unit area was calculated from the concentration of each substance in the receiving medium and plotted as a function of time. Each data point on the plot represents a mean of triplicate permeation experiments. The flux (J) was determined as the slope of the linear portion of the plot.
Drug
Mobile phase
Flow rate (ml / min)
Wavelength (nm)
ITF 296
Methanol:water:tetrahydrofuran (54:43:3) Methanol:water:tetrahydrofuran (54:43:3) Methanol:water:tetrahydrofuran (54:43:3) Methanol:water (45:55) Water:acetonitrile (60:40)
0.9
264
1.2
318
0.9
220
0.9
220
0.9
254
NTG ISDN NR
The permeability coefficient was calculated following the Fick’s first law of diffusion: Kp 5 Jmax /S, where Kp is the permeability coefficient (cm / h), Jmax is the maximum flux (mg / cm 2 / h) and S is the drug donor concentration, corresponding to the drug solubility in the vehicle (mg / ml). The predicted plasma steady-state concentration was obtained using the following equation (Calpena et al., 1994): Css 5 Jmax ? TTS area / Cl where Css is the plasma steady-state concentration (mg / l), TTS area is the hypothetical area of the transdermal therapeutic system (cm 2 ) and Cl is the systemic clearance after iv administration (l / h).
3. Results and discussion In Table 2 are reported the relevant physicochemical parameters of the tested substances. It is well documented that the molecular weight (MW) of a substance directly affects its diffusion across simple or complex membranes (Potts and Guy, 1992). Because of the narrow MW range of the selected compounds it is possible to presume that this parameter is not significant for the explanation of the difference of permeation observed through human skin. In Table 2 are also reported the melting point (MP) of the compounds, as this parameter is often related to the transdermal transport of the drug (Izumoto et al., 1992). On the other hand, the studied compounds presented large differences in their hydrophobicity (Log P). Considering that the chemical structures of ITF 296 and ITF 1129 are similar, it was noteworthy the relevant increase of the hydrophobicity due to insertion of a methyl group on the aromatic ring of ITF 296. This simple modification also caused a strong decrease of the solubility of ITF 1129 in the selected vehicle because of its high hydrophobicity. Table 2 Relevant physicochemical parameters of the drugs
Table 1 HLPC conditions for the analysis of drugs
ITF 1129
233
Drug
MW (g / mole)
MP (8C)
Log P
S (mg / ml)
ITF 296 ITF 1129 ISDN NR NTG
238 252 236 211 227
55.5–56.5 80–81 70 92–93 13.5
1.85 2.41 1.34 a 21.02 b 2.05 a
2219* 505* 1415* 35747* 1730**
Note: molecular weight (MW), melting point (MP), log partition coefficient (Log P), solubility (S). a Drayton, 1990. b Hatanaka et al., 1992. * Solubility in water:propylene glycol (80:20 v / v) at 32618C. ** Solubility in water at 208C (Mark, 1965).
234
P. Minghetti et al. / European Journal of Pharmaceutical Sciences 7 (1999) 231 – 236
Fig. 2. Permeation profiles of the drugs through SCE (diluted solution) (n53; error bars are mean6S.D.).
The cumulative amounts of drugs permeated in 24 h through SCE prepared from the same human donor are shown in Figs. 2 and 3, respectively related to the experiments conducted using as donor phase the diluted solution (0.08% w / v) and the saturated one. A mixture water:propylene glycol (80:20 v / v) was used as donor vehicle to obtain a higher thermodynamic activity than the value of the pure water. The addition of this amount of propylene glycol, that can be considered as an enhacer, should not alter in a significant way the barrier properties of the human SCE (Megrab et al., 1995b). A steady-state skin flux was attained within the first hours of application of the drug solutions as reflected by the short diffusion lag time. No significant drug depletion from the donor compartment was observed, and the steadystate skin flux was maintained over the entire period of the diffusion experiment (24 h). The NR permeation profile, when the diluted solution was used, is not reported in Fig. 2, as the amount permeated was not detectable. When the diluted solution at the same concentration were used, the amount permeated increased in order of ISDN.NTG.ITF 296. In the second phase of the study the permeation experiments were performed using saturated drug solutions to ensure unit thermodynamic activity (Fig. 3). In this case NR permeated through the SCE even if in small amounts.
In the case of the ITF substances the permeation studies have been conducted with SCE from two different human donors. The flux values differed of a very small amount for ITF 296 (3.5260.18; 3.3660.61), while the variation was bigger for ITF 1129, even if it was still not significant (8.7761.63; 5.4861.02). Comparing data obtained with SCE of the same donor, ITF 296 showed the highest flux and consequently the highest amount permeated at 24 h in both the experiments (diluted or saturated donor solutions), as shown in Table 3. ITF 296 presented the best balance between hydrophobicity and solubility in this group of organic nitrates. Although ITF 1129 showed the highest permeability coefficient due to its high hydrophobicity it was characterised by lower Jmax compared with ITF 296 and ISDN because of its low thermodynamic activity in the used vehicle. On the other hand Nicorandil, which is very soluble in the donor phase, showed a very low Jmax due to its low hydrophobicity that affected the permeability coefficient of the drug. In the case of ITF 296 and ISDN the Kp were calculated from either the data of the diluted solutions (ITF 296 Kp 54.260.76310 23 ; ISDN Kp 52.3160.07310 23 ) or the saturated solutions (Table 3). The numbers are in reasonable agreement that is within the expected limits given that the saturated solutions will not be behaving as if they were ideal. Delivery of Nitroglycerin through human skin has been shown to well correlate with the in vivo data (Hadgraft et al., 1993). Furthermore it has been suggested that the same correlation may apply to other substances provided that there is no significant difference in the transfer rate of the drug through dead skin compared to viable skin. It is therefore necessary an in vivo study to evaluate the skin metabolism of ITF 296 in order to ascertain that the compounds do not encounter metabolic events through the viable skin. The therapeutical plasma concentrations (Ct ) of ITF 296 were estimated considering the effects of the drug on arterial compliance. A dose-response study showed that significant effects on this parameter are obtained in the Table 3 Permeation parameters of the drugs obtained from diluted (0.08% w / v) or saturated solutions Drug
ITF 296 ITF 1129 NTG ISDN NR Fig. 3. Permeation profiles of the drugs through SCE (saturated solution) (n53; error bars are mean6S.D.).
Diluted solution
Saturated solution
J (mg cm 22 h 21 )
Jmax (mg cm 22 h 21 )
Kp 310 3 (cm h 21 )
3.3660.61 – 2.8560.51 1.8560.06 –
7.9161.14 5.4861.02 6.1661.09* 6.3661.35 1.8760.41
3.5660.96 10.8562.02 – 4.4960.95 0.0560.01
Note: flux (J), maximum flux (Jmax ), permeability coefficient (Kp ). * Calculated value. (Jmax 5Kp ?S; where S is solubility.)
P. Minghetti et al. / European Journal of Pharmaceutical Sciences 7 (1999) 231 – 236 Table 4 Calculated plasma steady-state concentration of the drugs from experimental permeation studies: Css 5 Jmax ?TTS area / Cl Drug
Cl (l / h)
Css (ng / ml)
Ct /Css
ITF 296 ISDN NR NTG
138.6 a 224.4 b 52.2 c 300 d
2.0260.37 0.4460.01 1.0760.23 0.6160.11*
2.47 10.23 140.19 1.15
Note: Plasma steady-state concentrations (Css ), clearance (Cl), therapeutical plasma concentrations (Ct ). * Css was obtained using calculated Jmax . a Monzani et al., 1998. b Abshagen et al., 1985. c Frydman, 1992. d McNiff et al., 1981.
presence of plasma concentrations below 5 ng / ml (Sardina, 1996). The therapeutical plasma levels of NTG are between 0.2 and 1.2 ng / ml (Rossi and Bolognese, 1989), those of ISDN are between 1 and 8 ng / ml (Cohn and Rittinghausen, 1985) and those of NR between 100 and 200 ng / ml (Frydman, 1992). Since the in vitro SCE permeation data allow to estimate the rate of input of the drug into the systemic circulation, it should be possible to evaluate the achievable plasma levels combining the in vitro permeability data with the systemic clearance of the drug reported in the literature. Assuming for the TTS area a reasonable limit of 30 cm 2 and knowing the clearance value it is possible to estimate the plasma steady-state concentrations (Css ), from the in vitro experiments conducted with the human skin from the same donor (Table 4). Css could not be calculated for ITF 1129 because its clearance in humans is unknown. ITF 296 and NTG estimated steady-state plasma concentrations that were very similar to their respective therapeutic concentrations. The value of the Ct /Css ratio has been reported as it could be useful as an indication of the suitability of the drug for transdermal administration. ISDN, which presented a Ct /Css ratio of 10.23, is still suitable for the transdermal administration, as demonstrated by the presence on the market of TTS containing the drug. On the contrary the predicted steady-state plasma concentration was considerably less than the therapeutic concentration for NR, as indicated by the high Ct /Css ratio (Table 4).
4. Conclusion The considered benzoxazinones, ITF 296 and ITF 1129, showed promising in vitro skin permeability. The different permeation behaviour of these compounds is particularly determined by the contribution of both their solubility in the vehicle and hydrophobic characteristic. The values of Log P of both compounds are in the range of
235
values that are generally considered suitable for transdermal administration. ITF 296, that possesses the largest solubility in the vehicle, showed the highest flux and the highest amount permeated at 24 h either when diluted or saturated solutions were used as donor phase. The value of the Ct /Css ratio for ITF 296 was 2.47. As this value is in the same order of magnitude of Nitroglycerin ratio (1.15), it is a positive indication that ITF 296 is suitable for transdermal administration. Therefore, ITF 296 appears to be a good candidate for the development of transdermal patch. However, the flux of ITF 296 from the patch through the human skin might be different from the flux obtained in this work. A larger flux might be obtained due to the solubility of ITF 296 in the selected matrix, to the occlusive conditions obtained with TTS and to the possibility of adding adequate enhancers.
References Abrams, J., 1985. Haemodynamic effects of nitroglycerin and long-acting nitrates. Am. Heart J. 110, 216–314. Abshagen, U., Betzien, G., Endele, R., Kaufmann, B., Neugebauer, G., 1985. Pharmacokinetics and metabolism of isosorbide-dinitrate after intravenous and oral administration. Eur. J. Clin. Pharmacol. 27, 637–644. Benedini, F., Bertolini, G., Cereda, R., et al., 1995. New anti-anginal nitroesters with reduced hypotensive activity. Synthesis and pharmacological evaluation of 3-nitrooxyalkyl-2H-1,3-benzoxazin-4(3H)-ones. J. Med. Chem. 38, 130–136. Calpena, A.C., Blanes, C., Moreno, J., Oback, R., Domenech, J., 1994. A comparative in vitro study of transdermal absorption of antiemetics. J. Pharm. Sci. 83 (1), 29–33. Cohn, J., Rittinghausen, R., 1985. Mononitrate. Springer Verlag, Berlin Heidelberg New York Tokyo, pp. 147–153. Conti, C.R., 1987. Use of nitrates in unstable angina pectoris. Am. J. Cardiol. 60, 31H–34H. Drayton, C.J., 1990. Comprehensive Medicinal Chemistry. Pergamon Press, Oxford UK, pp. 6, 608, 715. Fitzgerald, D., 1995. Pharmacologic properties of new nitric oxide donor ITF 296. J. Cardiovasc. Pharmacol. 26 (supp. 4), S1–S66. Frydman, A., 1992. Pharmacokinetic profile of nicorandil in humans: an overview. J. Cardiovasc. Pharmacol. 20 (3), S34–S44. Goodman, M., Barry, B.W., 1992. Action of penetration enhancer on human skin as assessed by the permeation of model drug 5-fluorouracil and oestradiol. I. Infinite dose technique. J. Invest. Dermatol. 91, 323–327. Gunnar, R.M., Fisch, C., 1990. Special Report: ACC /AHA guidelines for the early management of patients with acute myocardial infarction. Circulation 82, 664–707. Hadgraft, J., Beutner, D., Wolff, H.M., 1993. In vitro–in vivo comparisons of the transdermal delivery of nitroglycerin. Intl. J. Pharm. 89, R1–R4. Hatanaka, T., Inuma, M., Sugibayashi, K., Morimoto, Y., 1992. Prediction of skin permeability of drugs. II. Development of composite membrane as a skin alternative. Intl. J. Pharm. 79, 21–28. Izumoto, T., Aioi, A., Uenoyama, K., Azuma, M., 1992. Relationship between the transference of a drug from a transdermal patch and the physicochemical properties. Chem. Pharm. Bull. 40 (2), 456–458.
236
P. Minghetti et al. / European Journal of Pharmaceutical Sciences 7 (1999) 231 – 236
Krogsgaard-Larsenand, P., Bundgaard, M., 1991. A Textbook of Drug Design and Development. Harwood Academy, Chur, Switzerland. Mark (Ed.), 1965. Kirk-Othmer Encyclopedia of Chemical Technology, 2nd ed., vol. 8. Interscience Publishers, New York, USA. McGregor, M., 1983. Pathogenesis of angina pectoris and role of nitrates in relief of myocardial ischemia. Am. J. Cardiol. 74 (6B), 21–27. McNiff, E.F., Jacobi, A., Young-Chang, F.M., Golden, L.M., Goldfarb, A., Ho-Leung, F., 1981. Nitroglycerin pharmacokinetics after intravenous infusion in normal subjects. J. Pharm. Sci. 70, 1054–1058. Megrab, N.A., Williams, A.C., Barry, B.W., 1995a. Oestradiol permeation through human skin and silastic membrane: effetcs of propylene glycol and supersaturation. J Control. Release 36, 277–294. Megrab, N.A., Williams, A.C., Barry, B.W., 1995b. Oestradiol permeation across human skin, silastic and snake skin membranes: the effects of ethanol / water co-solvent systems. Intl. J. Pharm. 116, 101–112.
Monzani, M.V., Coltro, G., Sala, A., Sardina, M., 1998. Pharmacokinetic of ITF 296 (Sinitrodil) a novel organic nitrate, in healthy volunteers. Eur. J. Pharm. Sci., in press. Parker, J.O., 1987. Nitrate therapy in stable angina pectoris. New Engl. J. Med. 316, 1635–1642. Potts, R.O., Guy, R.H., 1992. Predicting skin permeability. Pharm. Res. 9 (5), 663–669. Rossi, P., Bolognese, L., 1989. I nitrati in cardiologia. Ed. Centro Scientifico Torinese. Sardina, M., 1996. ITF 296–Italfarmaco Investigator’s Brochure, Edition No. 2. Williams, D.O. Amsterdam, E.A., Mason, D.T., 1975. Haemodynamic effects of nitroglycerin in acute myocardial infarction: decrease in ventricular preload at the expense of cardiac output. Circulation 51, 421–427.