- Email: [email protected]
Microspheres as nasal drug delivery systems
Advanced Drug Delivery Reviews 29 (1998) 185–194
L
Microspheres as nasal drug delivery systems Lena Pereswetoff-Morath 1 ¨ ¨ , Sweden Astra Arcus AB, S-151 85 Sodertalje Received 28 April 1997; accepted 21 June 1997
Abstract All types of microspheres that have been used as nasal drug delivery systems are water-insoluble but absorb water into the sphere’s matrix, resulting in swelling of the spheres and the formation of a gel. The building materials in the microspheres have been starch, dextran, albumin and hyaluronic acid, and the bioavailability of several peptides and proteins has been improved in different animal models. Also, some low-molecular weight drugs have been successfully delivered in microsphere preparations. The residence time in the cavity is considerably increased for microspheres compared to solutions. However, this is not the only factor to increase the absorption of large hydrophilic drugs. Microspheres also exert a direct effect on the mucosa, resulting in the opening of tight junctions between the epithelial cells. Starch and dextran microspheres have been administered repeatedly and can be classified as safe dosage forms. 1998 Elsevier Science B.V. Keywords: Nasal administration; Microspheres; Bioavailability; Absorption enhancer; Absorption mechanism; Ciliotoxicity; Mucociliary clearance; Mucosal toxicity
Contents 1. 2. 3. 4.
Introduction ............................................................................................................................................................................ Characteristics of microspheres ................................................................................................................................................ Mechanisms for improved drug absorption by microspheres ....................................................................................................... Improved nasal drug absorption by microspheres ....................................................................................................................... 4.1. Dextran microspheres ....................................................................................................................................................... 4.2. Degradable starch microspheres (DSM)............................................................................................................................. 4.3. Other types of microspheres.............................................................................................................................................. 5. Immunological aspects of microspheres as nasal drug delivery systems ....................................................................................... 6. Toxicological aspects of microspheres as nasal drug delivery systems ......................................................................................... 6.1. In vitro ciliotoxicity ......................................................................................................................................................... 6.2. In vivo mucociliary clearance ........................................................................................................................................... 6.3. In vivo mucosal toxicity ................................................................................................................................................... 7. Conclusions ............................................................................................................................................................................ References ..................................................................................................................................................................................
185 186 186 188 188 189 191 191 191 192 192 192 192 193
1. Introduction
1
Tel.: 1 46 8 55328129; fax: 1 46 8 55325080; e-mail: [email protected].
The nasal route for systemic drug delivery has mainly been investigated with large hydrophilic peptides and proteins in mind, although other type of
0169-409X / 98 / $19.00 1998 Elsevier Science B.V. All rights reserved. PII S0169-409X( 97 )00069-0
186
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194
drugs have also been investigated. Different types of absorption enhancers have been used to avoid the problem of low absorption. The use of dry-powder formulations containing bioadhesive polymers for nasal administration of peptides and proteins was first investigated by Nagai et al. in 1984 [1]. Waterinsoluble cellulose derivatives were mixed with insulin and the powder mixture was installed into the nasal cavity. The powder took up water, swelled and established a gel with a prolonged residence time in the nasal cavity as a consequence. Bioadhesive polymers have since then been reported to decrease the mucociliary transport rate ex vivo on frog palate [2] and in vivo in rats [3]. Several other groups have used similar formulations for the nasal delivery of drugs such as interferon, beclomethasone, and oxymetazoline [4–6]. The polymers used in those studies were microcrystalline cellulose (Avicel), hydroxypropyl cellulose (HPC) and hydroxypropyl methylcellulose (HPMC). Well-characterised bioadhesive microspheres as another way of prolonging the residence time in the nasal cavity were introduced by Illum et al. in 1987 [7]. Microspheres of albumin, starch and DEAEdextran absorbed water and formed a gel-like layer which was cleared slowly from the nasal cavity. Three hours after administration, 50% of the delivered amounts of albumin and starch microspheres and 60% of the DEAE-dextran microspheres were still present at the site of deposition. It was suggested that an increased contact time could increase the absorption efficiency of drugs.
2. Characteristics of microspheres The most frequently used microsphere system for nasal drug delivery is degradable starch microspheres (DSM), also known as Spherex . These microspheres are prepared by an emulsion polymerisation technique where starch is cross-linked with epichlorohydrine [8]. Most drugs of interest are incorporated into the sphere matrix due to the high exclusion limit, i.e. molecules with a molecular weight less than 30 000 Da can be incorporated into the swollen microsphere matrix, whereas larger molecules are totally excluded from the microspheres. The particle size of the microspheres used in
nasal drug delivery studies is about 45 mm in the swollen state. A microsphere system based on dextran crosslinked with epichlorohydrine has also been used for nasal drug delivery. These microspheres are commercially available under the brand names Sephadex and DEAE-Sephadex . In the latter, the glucose units in dextran are substituted with diethylaminoethyl groups, i.e. anion-exchange resins. The microspheres are available in particle sizes from 10 to 300 mm, although in nasal drug delivery only spheres less than 180 mm have been used. The degree of cross-linking differs between different grades. There is an inverse relationship between the degree of cross-linking and the degree of swelling. Consequently, the volume of liquid that the microspheres can hold in a swollen state is lower for spheres with a high degree of cross-linking than for spheres with a low degree of cross-linking. The exclusion limit ranges from 700 to 600 000 Da for different grades of dextran microspheres. For both the starch and the dextran systems, the drugs are loaded to the microspheres by a lyophilisation process [9,20]. A water solution of the drug is added to the microspheres, which takes up the fluid and a gel is formed. The gel is then freeze-dried and the resulting powder is passed through a sieve to yield a homogeneous dry formulation. Two different processes for the preparation of albumin microspheres have been described—a spraydrying technique with an aqueous solution of nicardipine and albumin [10] and an emulsification technique in which albumin propranolol microspheres were prepared [11]. The particle sizes in the swollen state were about 10 and 40 mm, respectively. Spraydrying and emulsification are also used for preparing polyacrylic acid microspheres [12,13]. The particles sizes obtained were in the range 1–10 mm in the dried state. Hyaluronic acid ester microspheres with a particle size of 6–8 mm were the result of an emulsification technique described by Illum et al. [14].
3. Mechanisms for improved drug absorption by microspheres Although the residence time in the cavity is considerably increased for microspheres that absorb
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194
water in contact with the mucosa and form a gel, it seems that this is not the only factor in the increased absorption of large hydrophilic drugs. The mechanism for the increased absorption of drugs administered with degradable starch microspheres (DSM) has been shown to be due to an effect on tight junctions between the epithelial cells [15]. Monolayers of Caco-2 cells were investigated in a transmission electron microscope prior to, immediately after, 15 min after and 180 min after the administration of dry microspheres. It was clearly shown that the tight junctions started to separate after only 3 min and that the separation continued for up to 15 min after administration (Fig. 1). Thus, when the microspheres absorb water from the mucus and swell, the epithelial cells are dehydrated and cause the tight junctions to separate. After 3 h, the tight junctions between the Caco-2 cells were comparable with the controls, indicating that the enhancing effect of DSM is rapid and reversible. Since the process is reversible, an increase in absorption for drugs that are transported via the paracellular pathway will take
187
place mainly during the short period when the tight junctions are separated. These findings are in agreement with the rapid absorption of insulin and the normalisation of glucose levels seen in animals when insulin was administered with DSM [16,17]. For an optimum effect, the drug has to be available for absorption, i.e. dissolved, when the spheres swell by taking water from the mucus layer and the underlying epithelial cells, resulting in the temporary widening of the tight junctions. Additional evidence for the validity of this suggested mechanism of action is that the absorption-enhancing effect of DSM is lost when the microspheres are administered in a pre-swollen state [18]. Another hypothesis for the mechanism of absorption enhancement of dry powders was recently put forward by Oechslein et al. [19]. The authors suggest that it is the calcium-binding capacity of the powder formulation which determines the absorption of the somatostatin analogue octreotide. Among the powder formulations that were investigated, dextran micro-
Fig. 1. TEM picture of an open tight junction between Caco-2 cells after exposure to DSM. Reprinted with permission [9].
188
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194
spheres (Sephadex ) were also included in the study. The peptide was mixed with the dry spheres, not loaded by lyophilisation as earlier described. The investigators found a correlation between the calcium-binding properties of the delivery system and the bioavailability of octreotide in rats. The Ca 21 binding capacity for dextran microspheres was initially quite low, although pre-swollen microspheres showed the highest calcium-binding capacity of the tested materials. Administration with these dextran microspheres resulted in the highest bioavailability for the test drug. It took 1 h to reach peak plasma concentration in vivo and the need for swelling before a good calcium-binding capacity was achieved was taken to be an explanation for this delay. However, this mechanism cannot explain the results obtained for the same type of microspheres in an absorption study in rats by Pereswetoff-Morath and Edman [20]. Insulin administered with dextran microspheres induced a maximum decrease in plasma glucose after 30–40 min. These findings are comparable to those published for insulin in degradable starch microspheres in rats [16]. In the latter study, the plasma peak concentration of insulin was reached about 10 min after administration, which would most likely also be the case for the dextran microspheres due to the similar plasma glucose concentration profile. If calcium binding had been crucial for the insulin absorption from dextran microspheres in this study, it would not have been possible to reach the peak concentration so soon after administration. Oechslein estimated the pre-soaking time needed to achieve calcium binding to be at least 30 min, and this would result in a much later plasma concentration peak for insulin in the study by Pereswetoff-Morath and Edman.
4. Improved nasal drug absorption by microspheres Microspheres of different materials have been evaluated in vivo as nasal drug delivery systems. Degradable starch microspheres (DSM) increase the absorption of insulin [16], gentamicin [21], human growth hormone [22], metoclopramide [23] and desmopressin [24]. Dextran microspheres have been used in vivo as a delivery system for insulin [20,25]
and octreotide [19] and have also been evaluated in vitro as a potential delivery system for nasally administered nicotine [26]. Hyaluronic acid ester microspheres increase the absorption of insulin [14] and albumin microspheres have been used to deliver propranolol [11]. Polyacrylic acid microspheres [12,13] and polyvinyl alcohol microspheres [27] have as yet only been evaluated in vitro as potential nasal drug delivery systems. Although microspheres of different materials appear to act in a similar manner, there are also differences.
4.1. Dextran microspheres In the study by Illum et al. [7], the slowest clearance was detected for DEAE-dextran, where 60% of the delivered dose was still present at the deposition site after 3 h. However, these microspheres were not successful in promoting insulin absorption in rats [25]. The insulin was too strongly bound to the DEAE groups to be released by a solution with an ionic strength corresponding to physiological conditions. In the same study, the dextran microspheres without the ion-exchange group induced a 25% decrease in plasma glucose about 40 min after the administration of insulin 1 IU / kg. The decrease in plasma glucose varied greatly between animals, probably due to the large size range of the microspheres (50–180 mm). In a later study, the same insulin dose was administered with microspheres of a particle size , 45 mm [20]. When the insulin was situated on the surface of the microspheres, a 52% decrease in plasma glucose was induced 30 min after administration to rats. However, when microspheres which included insulin in the sphere matrix were used, the maximum plasma glucose decrease was 30%, reached after 60 min (Fig. 2). Thus, dextran microspheres with insulin on the surface of the spheres were more effective in promoting insulin absorption than spheres with insulin distributed into the sphere matrix. It is possible that the limited amount of fluid in the nasal cavity is responsible for the differences observed. The microspheres have to be completely swollen to release all of the incorporated insulin. The limited amount of fluid in the nasal cavity will result in incomplete swelling. Thus, the release rate from microspheres with insulin incorporated into the
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194
189
Fig. 2. Changes in plasma glucose (mean6S.D.) after intranasal administration of 1 IU / kg insulin in dextran microspheres to rats. (d) Insulin incorporated into the sphere matrix; (s) insulin situated on the surface of the spheres.
sphere matrix will be more affected than for microspheres with insulin on the surface. Structural changes due to the lyophilisation process were observed in spheres with insulin incorporated, which probably further decreased the release rate. It was therefore concluded that the localisation of insulin is of great importance to the in vivo behaviour of dextran microspheres. This conclusion is not valid as a general rule for microspheres, since the absorptionenhancing effect of DSM with the insulin incorporated into the microsphere matrix is as good as for the dextran microspheres with insulin on the surface.
4.2. Degradable starch microspheres ( DSM) DSM is the most frequently used microsphere system for nasal drug delivery and has been shown to improve the absorption of insulin [16], gentamicin [21], human growth hormone [22], metoclopramide [23] and desmopressin [24]. The bioavailabilities are summarised in Table 1. Insulin administered in DSM to rats resulted in a peak plasma concentration after 7–10 min (Fig. 3) and a rapid dose-dependent decrease in blood glucose. The maximum reductions in blood glucose
Table 1 Nasal drug delivery in degradable starch microspheres (DSM) Species
Drug
Addition enhancer
Dose
t max (min)
Absolute bioavailability (%)
Rat
Insulin
Sheep
Insulin
Sheep
Gentamicin hGH
Sheep
Desmopressin
0.75 IU / kg 1.70 IU / kg 2 IU / kg 2 IU / kg 5 mg / kg 5 mg / kg 2.5 mg / kg 2.5 mg / kg 5 mg / kg 5 mg / kg
8 7 12 22 24 8 120 60 8 18
30 33 4.5 13.1 9.7 57.3
Sheep
— — — LPC — LPC — LPC — LPC
Relative bioavailability (%)
[16] 10.7 31.5
[28] [21]
2.7 14.4 4.7 9.6
Ref.
[22] [24]
190
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194
Fig. 3. Serum insulin concentrations (mean6S.D.) after intranasal administration of insulin in DSM to rats. (s) 0.75 IU / kg insulin; (d) 1.7 IU / kg; ( 3 ) DSM without insulin. Reprinted with permission [16].
level after doses of 0.75 and 1.7 IU / kg were 40 and 64%, respectively, and the absolute bioavailability was approx. 30%. The maximum decrease was reached about 40 min after administration, irrespective of the dose given [16]. In line with these results, a latter study by the same group [17] showed that an insulin dose of 1 IU / kg gave a 45–50% reduction in plasma glucose, 30–40 min after administration. DSM as a delivery system for insulin (2 IU / kg) has also been tested in sheep. The absolute bioavailability was 4.5% and the time to reach maximum effect, i.e. a 50% decrease in plasma glucose, was 60 min [28]. The effect could be further improved by incorporating the absorption enhancer lysophosphatidylcholine (LPC) into the microspheres together with insulin. The plasma glucose profile was similar, with a maximum glucose decrease of 60%, but the bioavailability was improved to 13%. The DSM system has also been used for other drugs. Gentamicin, a polar antibiotic compound, showed no to little absorption from the nasal cavity when administered as a plain solution to sheep [21].
When administered in DSM, the peak plasma concentration was reached 24 min after dosing and was maintained for up to 60 min. The absolute bioavailability was 10% and, as with insulin, the bioavailability could be further improved by incorporating LPC into the spheres. In this case the bioavailability increased to 57% and the absorption was more rapid, with a plasma concentration peak after 8 min. In the same animal model a similar trend was observed for biosynthetic human growth hormone (hGH) [22] and desmopressin [24]. However, the effect was not as pronounced as for gentamicin. The relative bioavailability of hGH was 3% when delivered in DSM and 14% when LPC was included. The plasma concentration peak was reached at 120 min post-dosing when DSM was used and 60 min when also LPC was incorporated into the spheres. This delay of the plasma concentration peak was attributed to the low solubility of hGH, which was probably increased when LPC was included and hence the peak was reached sooner. Desmopressin was rapidly absorbed, with a t max of 8 min for plain
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194
DSM and 18 min for DSM with LPC. The absolute bioavailabilities for desmopressin were 5 and 10%, respectively. Biodegradable starch microspheres have also been used in a clinical investigation of the nasal delivery of the antiemetic drug metoclopramide [23]. The bioavailability was 137% compared to a nasal spray, and the kinetics were the same. A mixture of metoclopramide and microcrystalline cellulose administered as a dry powder did not result in a increased absorption compared to the solution.
4.3. Other types of microspheres The group in Nottingham has also used hyaluronic acid ester microspheres to deliver insulin to sheep [14]. The relative bioavailability of 2 IU / kg insulin was the same as for the plain DSM, i.e. about 11%. However, the maximum decrease in plasma glucose of about 40–60% was reached after 90 min for the hyaluronic acid ester microspheres, compared to 60 min for the DSM. Albumin microspheres containing propranolol with extended drug release resulted in a prolonged plasma concentration plateau in dogs [11]. These microspheres were not administered as a dry-powder formulation, but in a suspension. About 50% of the incorporated propranolol was released in vitro within 4 h. The maximum plasma concentration was obtained after 90 min, and an effective plasma concentration was maintained for up to 12 h. The absolute bioavailability was approx. 84%.
5. Immunological aspects of microspheres as nasal drug delivery systems In the case of nasal drug delivery, a mucosal immune response is an unwanted effect that should be avoided. In a review by O’Hagan and Illum [29] caution is recommended when polymers and microspheres are used as nasal drug delivery vehicles since there are some reports of an adjuvant effect on the immune system. However, the intranasal delivery of a protein to rats in DSM resulted in a lower systemic immune response than intranasal administration of the protein in solution [18]. Intranasal protein delivery in dextran microspheres did not generate an IgA or IgG response in either plasma or nasal washings
191
[30]. Part of the explanation might be the low irritable and inflammatory properties of the delivery system used [31]. Inflammation or irritation attracts immunocompetent cells which would promote a local immunological response. It was therefore concluded that the lack of nasal mucosal immune response to exposure of an immunogen in a dextran microsphere system is evidence that this system does not act as an adjuvant for mucosal immune reactions. For an immunological response the antigen has to be presented to lymphoid tissue containing antibodyproducing cells, i.e. the nasal-associated lymphoid tissue (NALT). The antigen is then drained to the posterior cervical lymph nodes from where a local immune response can be evoked. The use of microspheres as delivery systems for mucosal vaccination has recently been described by several research groups. It is not within the scope of this paper to describe these types of microspheres, but it is worth mentioning that there are differences between the microspheres used for drug delivery and the ones used for vaccination. Microspheres prepared with poly(lactide) PLA [32,33] and poly(lactide-co-glycolide) PLG [34,35] have been used to deliver antigens of different kinds. These microspheres are smaller than the ones used for drug delivery, i.e. , 1 mm, and may be taken up as particles by NALT [36].
6. Toxicological aspects of microspheres as nasal drug delivery systems The effect on the ciliary system can be evaluated in vitro by measuring ciliary beat frequency on explants of ciliated mucosa or in vivo by measuring the mucociliary clearance. The effects on the ciliary beat frequency detected in vitro may be more pronounced than the effects on the ciliary clearance in vivo. In vitro experiments expose the ciliated epithelia directly to the test substance, whereas the cilia are protected by mucus in vivo. Histopathological examination of the nasal mucosa in animals after exposure to different enhancers has also been frequently used to determine damaging effects. The turn-over rate of epithelial cells in the respiratory tract is rather fast. The nasal septum of rabbits recovered in 5 days after total removal of the mucosa [37]. This can also reduce the damaging effect of a substance in vivo. Hence, in vitro ciliary beat
192
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194
frequency tests should be used as indicators of potential damage to the nasal epithelium rather than proof of such damage occurring in the in vivo situation.
6.1. In vitro ciliotoxicity Dextran microspheres had practically no effect on ciliary beat frequency (CBF) measured on explants from rat trachea [31]. It was not possible to detect CBF during the exposure due to the opaque nature of the swollen spheres; however, when the explants were rinsed with Hepes buffer after 15 and 30 min exposure, the cilia immediately started to beat, and after 15 min recovery in Hepes buffer the CBF was back to normal. The immediate recovery of the ciliary movement indicates that the cilia are not damaged by the dextran microspheres. The reason was believed to be that the periciliary fluid is held close to the cell surface and so it is not absorbed by the spheres or that the gel-like layer of dextran spheres formed maintains an environment that is sufficiently humid to keep the cilia intact.
6.2. In vivo mucociliary clearance DSM has been administered daily for 8 days to healthy volunteers [38]. The mucociliary clearance, analysed with a saccharine-dye test, varied between 3.6 and 15 mm / min prior to DSM administration. The clearance was unchanged on days 1 and 8, as was the geometry of the nasal cavity. Thus, no changes in mucociliary clearance could be found after repeated administration of DSM in vivo and the microspheres did not produce congestion or decongestion. These findings are in agreement with the results from the in vitro study previously described for dextran microspheres [31] with properties similar to the starch microspheres. The ciliary function seems to be unaffected by the administration of dry microspheres.
6.3. In vivo mucosal toxicity Administration of DSM twice daily for 8 weeks to rabbits [39] and dextran microspheres administered once daily for 4 weeks to rats [31] induced only slight hyperplasia in the mucosal membrane. Only minor findings were observed at the histopatho-
logical examination, most frequently in the anterior part of the nose. No findings related to the administered microspheres were observed in the middle and posterior parts of the nasal cavity. However, mild focal squamous metaplasia, i.e. replacement of columnar cells by squamous cells, was also detected in rats given dextran microspheres for 4 weeks. It is possible that the finding of squamous metaplasia is a sign of recent damage that has been repaired. It has been shown that following the removal of epithelial cells, the basal cells differentiate, flatten and migrate to establish cell-to-cell contact. This process rapidly covers the damaged zone with squamous epithelium [40].
7. Conclusions All types of microspheres that have been used as nasal drug delivery systems are water-insoluble but absorb water into the sphere’s matrix, resulting in swelling of the spheres. The dextran microsphere system was as effective as an absorption enhancer for insulin as degradable starch microspheres (DSM). The maximum decrease in plasma glucose and the kinetics of the effect curve were similar for both systems. These systems also have a similar effect on mucosal integrity and lack a of adjuvant effect on the immune system after repeated nasal administration. It is therefore likely that the mode of action for improved absorption found for starch microspheres is also applicable to dextran microspheres, i.e. separation of the tight junctions during the swelling process of the microspheres. An important factor for the absorption of large hydrophilic molecules is probably the initial availability of the drug during the period when the tight junctions are separated. The formulation of all kinds of microspheres as nasal drug delivery systems for peptides and proteins ought to take into account studies of the time for complete swelling of the spheres and the release rate of the incorporated drug under conditions resembling the situation in the nasal cavity. This is probably not as crucial for low-molecular weight drugs, which are absorbed to a larger extent without absorption enhancers and for which it is easier to achieve prolonged absorption by retaining the formulation in the nasal cavity for a longer period of time. The use of different types of microspheres as nasal
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194
drug delivery systems has been investigated mostly in rats and sheep, although dogs have also been used. In these animal models the absorption of peptides and proteins has been successfully improved. So far, there are no publications where the peptide and protein absorption-promoting effects of these systems have been investigated in humans. The only absorption study in humans that has been published with microspheres as a delivery system involved the low-molecular weight drug metoclopramide. In this study the bioavailability was improved with microspheres compared to an ordinary nasal spray. However, where peptides and proteins are concerned, there seems to be a tendency for the absorptionpromoting effect to decrease when larger animals are used as compared to rats. Accordingly, investigations in humans are necessary before any conclusions about the potential of microspheres as a nasal drug delivery system for the systemic absorption of peptides and proteins can be drawn.
References [1] T. Nagai, Y. Nishimoto, N. Nambu, Y. Suzuki, K. Sekine, Powder dosage form of insulin for nasal administration, J. Control. Release 1 (1984) 15–22. [2] S.Y. Lin, G.L. Amidon, N.D. Weiner, A.H. Goldberg, Viscoelasticity of anionic polymers and their mucociliary transport on the frog palate, Pharm. Res. 10 (1993) 411–417. [3] M. Zhou, M.D. Donovan, Intranasal mucociliary clearance of putative bioadhesive polymer gels, Int. J. Pharm. 135 (1996) 115–125. [4] Y. Maitani, T. Igawa, Y. Machida, T. Nagai, Plasma levels following intranasal and intravenous administration of human interferon-b to rabbits, Drug Design Delivery 4 (1989) 109–119. [5] T. Nagai, Y. Machida, Bioadhesive dosage forms for nasal administration, in: V. Lenaerts, R. Gurney (Eds.), Bioadhesive Drug Delivery Systems, CRC Press Inc., Boca Raton, FL, 1990, pp. 169–178. [6] P.E. Koochaki, Nasal pharmaceutical composition in powder form. European Patent Application EP 571671, 1993. [7] L. Illum, H. Jørgensen, H. Bisgaard, O. Krogsgaard, N. Rossing, Bioadhesive microspheres as a potential nasal drug delivery system, Int. J. Pharm. 39 (1987) 189–199. [8] B. Lindberg, K. Lote, H. Teder, Biodegradable starch microspheres—A new medical tool, in: S.S. Davis, L. Illum, J.G. McVie, E. Tomlinson (Eds.), Microspheres and Drug Therapy. Pharmaceutical, Immunological and Medical Aspects, Elsevier, Amsterdam, 1984, pp. 153–188. ¨ ´ Microspheres as a nasal [9] P. Edman, E. Bjork, L. Ryden, delivery system for peptide drugs, J. Control. Release 21 (1992) 165–172.
193
[10] U. Conte, P. Giunchedi, L. Maggi, M.L. Torre, Spray dried albumin microspheres containing nicardipine, Eur. J. Pharm. Biopharm. 40 (1994) 203–208. [11] S.P. Vyas, S. Bhatnagar, P.J. Gogoi, N.K. Jain, Preparation and characterization of HSA-propranolol microspheres for nasal administration, Int. J. Pharm. 69 (1991) 5–12. [12] P. Vidgren, M. Vidgren, J. Arppe, T. Hakuli, E. Laine, P. Paronen, In vitro evaluation of spray-dried mucoadhesive microspheres for nasal administration, Drug Dev. Ind. Pharm. 18 (1992) 581–597. [13] B. Kriwet, T. Kissel, Poly(acrylic acid) microparticles widen intercellular spaces of Caco-2 cell monolayers: An examination by confocal laser scanning microscopy, Eur. J. Pharm. Biopharm. 42 (1996) 233–240. [14] L. Illum, N.F. Farraj, A.N. Fisher, I. Gill, M. Miglietta, L.M. Benedetti, Hyaluronic acid ester microspheres as a nasal delivery system for insulin, J. Control. Release 29 (1994) 133–141. ¨ [15] E. Bjork, U. Isaksson, P. Edman, P. Artursson, Starch microspheres induce pulsatile delivery of drugs and peptides across the epithelial barrier by reversible separation of the tight junctions, J. Drug Target. 2 (1995) 501–507. ¨ [16] E. Bjork, P. Edman, Degradable starch microspheres as a nasal delivery system for insulin, Int. J. Pharm. 47 (1988) 233–238. ¨ [17] E. Bjork, P. Edman, Characterization of degradable starch microspheres as a nasal delivery system for drugs, Int. J. Pharm. 62 (1990) 187–192. ¨ [18] E. Bjork, Strach microspheres as a nasal delivery system for drugs. Ph.D Thesis, Uppsala University, Sweden, 1993. [19] C.R. Oechslein, G. Fricker, T. Kissel, Nasal delivery of octreotide: Absorption enhancment by particulate carrier systems, Int. J. Pharm. 139 (1996) 25–32. [20] L. Pereswetoff-Morath, P. Edman, Dextran microspheres as a potential nasal drug delivery system for insulin—in vitro and in vivo properties, Int. J. Pharm. 124 (1995) 37–44. [21] L. Illum, N. Farraj, H. Critchley, S.S. Davis, Nasal administration of gentamicin using a novel microsphere delivery system, Int. J. Pharm. 46 (1988) 261–265. [22] L. Illum, N.F. Farraj, S.S. Davis, B.R. Johansen, D.T. O’Hagan, Investigation of the nasal absorption of biosynthetic human growth hormone in sheep-use of a bioadhesive microsphere delivery system, Int. J. Pharm. 63 (1990) 207– 211. [23] N. Vivien, P. Buri, L. Balant, S. Lacroix, Nasal absorption of metoclopramide administered to man, Eur. J. Pharm. Biopharm. 40 (1994) 228–231. [24] H.S.S.D. Critchley, D. Farraj, L. Illum, Nasal absorption of desmopressin in rats and in sheep. Effect of bioadhesive microsphere delivery system, J. Pharm. Pharmacol. 46 (1994) 651–656. ´ P. Edman, Effect of polymers and microspheres on [25] L. Ryden, the nasal absorption of insulin in rats, Int. J. Pharm. 83 (1992) 1–10. [26] A.-L. Cornaz, A. De Ascentis, P. Colombo, P. Buri, In vitro charcateristics of nicotine microspheres for transnasal delivery, Int. J. Pharm. 129 (1996) 175–183. [27] T.-Y. Ting, I. Gonda, E.M. Gibbs, Microparticles of poly-
194
[28]
[29]
[30]
[31]
[32]
[33]
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194 vinyl alcohol for nasal delivery. I. Generation by spraydrying and spray-desolvation, Pharm. Res. 9 (1992) 1330– 1335. N.F. Farraj, B.R. Johansen, S.S. Davis, L. Illum, Nasal administration of insulin using bioadhesive microspheres as a delivery system, J. Control. Release 13 (1990) 253–261. D.T. O’Hagan, L. Illum, Absorption of peptides and proteins from the respiratory tract and the potential for development of locally administered vaccine, Crit. Rev. Ther. Drug. Carrier Syst. 7 (1990) 35–97. L. Pereswetoff-Morath, P. Edman, Immunological consequences of nasal drug delivery in dextran microspheres and ethyl(hydroxyethyl) cellulose in rats, Int. J. Pharm. 128 (1996) 23–28. ¨ R. Khan, M. Dahlin, P. L. Pereswetoff-Morath, S. Bjurstrom, Edman, Toxicological aspects of the use of dextran microspheres and thermogelling ethyl(hydroxyethyl) cellulose (EHEC) as nasal drug delivery systems, Int. J. Pharm. 128 (1995) 9–21. A.J. Almeida, H.O. Alpar, M.R.W. Brown, Immune response to nasal delivery of antigenically intact tetanus toxoid associated with poly(L-lactic acid) microspheres in rats, rabbits and guinea-pigs, J. Pharm. Pharmacol. 45 (1993) 198–203. H.O. Alpar, A.J. Almeida, Identification of some of the physico-chemical characteristics of microspheres which influence the induction of the immune response following
[34]
[35]
[36]
[37]
[38]
[39]
[40]
mucosal delivery, Eur. J. Pharm. Biopharm. 40 (1994) 198– 202. E.S. Cahill, D.T. O’Hagan, L. Illum, A. Barnard, K.H.G. Mills, K. Redhead, Immune responses and protection against Bordetella pertussis infection after intrenasal immunization of mice with filamentous haemagglutinin in solution or incorporated in biodegradable microparticles, Vaccine 13 (1995) 455–462. J.H. Eldridge, J.K. Staas, J.A. Meulbroek, J.R. McGhee, T.R. Tice, R.M. Gilley, Biodegradable microspheres as a vaccine delivery system, Mol. Immunol. 28 (1991) 287–294. R.M. Carr, C.M. Lolachi, R.G. Albaran, D.M. Ridley, P.C. Montgomery, N.L. O’Sullivan, Nasal-associated lymphoid tissue is an inductive site for rat tear IgA antibody responses, Immunol. Invest. 25 (1996) 387–396. Y. Ohashi, Y. Nakai, H. Ikeoka, H. Furuya, Regeneration of nasal mucosa following mechanical injury, Acta Otolaryngol. Suppl. 486 (1991) 193–201. ¨ K. Holmberg, E. Bjork, P. Edman, Influence of degradable starch microspheres on the human nasal mucosa, Rhinology 32 (1994) 74–77. ¨ ¨ P. Edman, Morphologic examination E. Bjork, S. Bjurstrom, of rabbit nasal mucosa after nasal administration of degradable starch microspheres, Int. J. Pharm. 75 (1991) 73–80. ¨ ¨ J.S. Erjefalt, I. Erjefalt, F. Sundler, C.G.A. Persson, Microcirculation-derived factors in airway epithelial repair in vivo, Microvascular Res. 48 (1994) 161–178.
L
Microspheres as nasal drug delivery systems Lena Pereswetoff-Morath 1 ¨ ¨ , Sweden Astra Arcus AB, S-151 85 Sodertalje Received 28 April 1997; accepted 21 June 1997
Abstract All types of microspheres that have been used as nasal drug delivery systems are water-insoluble but absorb water into the sphere’s matrix, resulting in swelling of the spheres and the formation of a gel. The building materials in the microspheres have been starch, dextran, albumin and hyaluronic acid, and the bioavailability of several peptides and proteins has been improved in different animal models. Also, some low-molecular weight drugs have been successfully delivered in microsphere preparations. The residence time in the cavity is considerably increased for microspheres compared to solutions. However, this is not the only factor to increase the absorption of large hydrophilic drugs. Microspheres also exert a direct effect on the mucosa, resulting in the opening of tight junctions between the epithelial cells. Starch and dextran microspheres have been administered repeatedly and can be classified as safe dosage forms. 1998 Elsevier Science B.V. Keywords: Nasal administration; Microspheres; Bioavailability; Absorption enhancer; Absorption mechanism; Ciliotoxicity; Mucociliary clearance; Mucosal toxicity
Contents 1. 2. 3. 4.
Introduction ............................................................................................................................................................................ Characteristics of microspheres ................................................................................................................................................ Mechanisms for improved drug absorption by microspheres ....................................................................................................... Improved nasal drug absorption by microspheres ....................................................................................................................... 4.1. Dextran microspheres ....................................................................................................................................................... 4.2. Degradable starch microspheres (DSM)............................................................................................................................. 4.3. Other types of microspheres.............................................................................................................................................. 5. Immunological aspects of microspheres as nasal drug delivery systems ....................................................................................... 6. Toxicological aspects of microspheres as nasal drug delivery systems ......................................................................................... 6.1. In vitro ciliotoxicity ......................................................................................................................................................... 6.2. In vivo mucociliary clearance ........................................................................................................................................... 6.3. In vivo mucosal toxicity ................................................................................................................................................... 7. Conclusions ............................................................................................................................................................................ References ..................................................................................................................................................................................
185 186 186 188 188 189 191 191 191 192 192 192 192 193
1. Introduction
1
Tel.: 1 46 8 55328129; fax: 1 46 8 55325080; e-mail: [email protected].
The nasal route for systemic drug delivery has mainly been investigated with large hydrophilic peptides and proteins in mind, although other type of
0169-409X / 98 / $19.00 1998 Elsevier Science B.V. All rights reserved. PII S0169-409X( 97 )00069-0
186
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194
drugs have also been investigated. Different types of absorption enhancers have been used to avoid the problem of low absorption. The use of dry-powder formulations containing bioadhesive polymers for nasal administration of peptides and proteins was first investigated by Nagai et al. in 1984 [1]. Waterinsoluble cellulose derivatives were mixed with insulin and the powder mixture was installed into the nasal cavity. The powder took up water, swelled and established a gel with a prolonged residence time in the nasal cavity as a consequence. Bioadhesive polymers have since then been reported to decrease the mucociliary transport rate ex vivo on frog palate [2] and in vivo in rats [3]. Several other groups have used similar formulations for the nasal delivery of drugs such as interferon, beclomethasone, and oxymetazoline [4–6]. The polymers used in those studies were microcrystalline cellulose (Avicel), hydroxypropyl cellulose (HPC) and hydroxypropyl methylcellulose (HPMC). Well-characterised bioadhesive microspheres as another way of prolonging the residence time in the nasal cavity were introduced by Illum et al. in 1987 [7]. Microspheres of albumin, starch and DEAEdextran absorbed water and formed a gel-like layer which was cleared slowly from the nasal cavity. Three hours after administration, 50% of the delivered amounts of albumin and starch microspheres and 60% of the DEAE-dextran microspheres were still present at the site of deposition. It was suggested that an increased contact time could increase the absorption efficiency of drugs.
2. Characteristics of microspheres The most frequently used microsphere system for nasal drug delivery is degradable starch microspheres (DSM), also known as Spherex . These microspheres are prepared by an emulsion polymerisation technique where starch is cross-linked with epichlorohydrine [8]. Most drugs of interest are incorporated into the sphere matrix due to the high exclusion limit, i.e. molecules with a molecular weight less than 30 000 Da can be incorporated into the swollen microsphere matrix, whereas larger molecules are totally excluded from the microspheres. The particle size of the microspheres used in
nasal drug delivery studies is about 45 mm in the swollen state. A microsphere system based on dextran crosslinked with epichlorohydrine has also been used for nasal drug delivery. These microspheres are commercially available under the brand names Sephadex and DEAE-Sephadex . In the latter, the glucose units in dextran are substituted with diethylaminoethyl groups, i.e. anion-exchange resins. The microspheres are available in particle sizes from 10 to 300 mm, although in nasal drug delivery only spheres less than 180 mm have been used. The degree of cross-linking differs between different grades. There is an inverse relationship between the degree of cross-linking and the degree of swelling. Consequently, the volume of liquid that the microspheres can hold in a swollen state is lower for spheres with a high degree of cross-linking than for spheres with a low degree of cross-linking. The exclusion limit ranges from 700 to 600 000 Da for different grades of dextran microspheres. For both the starch and the dextran systems, the drugs are loaded to the microspheres by a lyophilisation process [9,20]. A water solution of the drug is added to the microspheres, which takes up the fluid and a gel is formed. The gel is then freeze-dried and the resulting powder is passed through a sieve to yield a homogeneous dry formulation. Two different processes for the preparation of albumin microspheres have been described—a spraydrying technique with an aqueous solution of nicardipine and albumin [10] and an emulsification technique in which albumin propranolol microspheres were prepared [11]. The particle sizes in the swollen state were about 10 and 40 mm, respectively. Spraydrying and emulsification are also used for preparing polyacrylic acid microspheres [12,13]. The particles sizes obtained were in the range 1–10 mm in the dried state. Hyaluronic acid ester microspheres with a particle size of 6–8 mm were the result of an emulsification technique described by Illum et al. [14].
3. Mechanisms for improved drug absorption by microspheres Although the residence time in the cavity is considerably increased for microspheres that absorb
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194
water in contact with the mucosa and form a gel, it seems that this is not the only factor in the increased absorption of large hydrophilic drugs. The mechanism for the increased absorption of drugs administered with degradable starch microspheres (DSM) has been shown to be due to an effect on tight junctions between the epithelial cells [15]. Monolayers of Caco-2 cells were investigated in a transmission electron microscope prior to, immediately after, 15 min after and 180 min after the administration of dry microspheres. It was clearly shown that the tight junctions started to separate after only 3 min and that the separation continued for up to 15 min after administration (Fig. 1). Thus, when the microspheres absorb water from the mucus and swell, the epithelial cells are dehydrated and cause the tight junctions to separate. After 3 h, the tight junctions between the Caco-2 cells were comparable with the controls, indicating that the enhancing effect of DSM is rapid and reversible. Since the process is reversible, an increase in absorption for drugs that are transported via the paracellular pathway will take
187
place mainly during the short period when the tight junctions are separated. These findings are in agreement with the rapid absorption of insulin and the normalisation of glucose levels seen in animals when insulin was administered with DSM [16,17]. For an optimum effect, the drug has to be available for absorption, i.e. dissolved, when the spheres swell by taking water from the mucus layer and the underlying epithelial cells, resulting in the temporary widening of the tight junctions. Additional evidence for the validity of this suggested mechanism of action is that the absorption-enhancing effect of DSM is lost when the microspheres are administered in a pre-swollen state [18]. Another hypothesis for the mechanism of absorption enhancement of dry powders was recently put forward by Oechslein et al. [19]. The authors suggest that it is the calcium-binding capacity of the powder formulation which determines the absorption of the somatostatin analogue octreotide. Among the powder formulations that were investigated, dextran micro-
Fig. 1. TEM picture of an open tight junction between Caco-2 cells after exposure to DSM. Reprinted with permission [9].
188
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194
spheres (Sephadex ) were also included in the study. The peptide was mixed with the dry spheres, not loaded by lyophilisation as earlier described. The investigators found a correlation between the calcium-binding properties of the delivery system and the bioavailability of octreotide in rats. The Ca 21 binding capacity for dextran microspheres was initially quite low, although pre-swollen microspheres showed the highest calcium-binding capacity of the tested materials. Administration with these dextran microspheres resulted in the highest bioavailability for the test drug. It took 1 h to reach peak plasma concentration in vivo and the need for swelling before a good calcium-binding capacity was achieved was taken to be an explanation for this delay. However, this mechanism cannot explain the results obtained for the same type of microspheres in an absorption study in rats by Pereswetoff-Morath and Edman [20]. Insulin administered with dextran microspheres induced a maximum decrease in plasma glucose after 30–40 min. These findings are comparable to those published for insulin in degradable starch microspheres in rats [16]. In the latter study, the plasma peak concentration of insulin was reached about 10 min after administration, which would most likely also be the case for the dextran microspheres due to the similar plasma glucose concentration profile. If calcium binding had been crucial for the insulin absorption from dextran microspheres in this study, it would not have been possible to reach the peak concentration so soon after administration. Oechslein estimated the pre-soaking time needed to achieve calcium binding to be at least 30 min, and this would result in a much later plasma concentration peak for insulin in the study by Pereswetoff-Morath and Edman.
4. Improved nasal drug absorption by microspheres Microspheres of different materials have been evaluated in vivo as nasal drug delivery systems. Degradable starch microspheres (DSM) increase the absorption of insulin [16], gentamicin [21], human growth hormone [22], metoclopramide [23] and desmopressin [24]. Dextran microspheres have been used in vivo as a delivery system for insulin [20,25]
and octreotide [19] and have also been evaluated in vitro as a potential delivery system for nasally administered nicotine [26]. Hyaluronic acid ester microspheres increase the absorption of insulin [14] and albumin microspheres have been used to deliver propranolol [11]. Polyacrylic acid microspheres [12,13] and polyvinyl alcohol microspheres [27] have as yet only been evaluated in vitro as potential nasal drug delivery systems. Although microspheres of different materials appear to act in a similar manner, there are also differences.
4.1. Dextran microspheres In the study by Illum et al. [7], the slowest clearance was detected for DEAE-dextran, where 60% of the delivered dose was still present at the deposition site after 3 h. However, these microspheres were not successful in promoting insulin absorption in rats [25]. The insulin was too strongly bound to the DEAE groups to be released by a solution with an ionic strength corresponding to physiological conditions. In the same study, the dextran microspheres without the ion-exchange group induced a 25% decrease in plasma glucose about 40 min after the administration of insulin 1 IU / kg. The decrease in plasma glucose varied greatly between animals, probably due to the large size range of the microspheres (50–180 mm). In a later study, the same insulin dose was administered with microspheres of a particle size , 45 mm [20]. When the insulin was situated on the surface of the microspheres, a 52% decrease in plasma glucose was induced 30 min after administration to rats. However, when microspheres which included insulin in the sphere matrix were used, the maximum plasma glucose decrease was 30%, reached after 60 min (Fig. 2). Thus, dextran microspheres with insulin on the surface of the spheres were more effective in promoting insulin absorption than spheres with insulin distributed into the sphere matrix. It is possible that the limited amount of fluid in the nasal cavity is responsible for the differences observed. The microspheres have to be completely swollen to release all of the incorporated insulin. The limited amount of fluid in the nasal cavity will result in incomplete swelling. Thus, the release rate from microspheres with insulin incorporated into the
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194
189
Fig. 2. Changes in plasma glucose (mean6S.D.) after intranasal administration of 1 IU / kg insulin in dextran microspheres to rats. (d) Insulin incorporated into the sphere matrix; (s) insulin situated on the surface of the spheres.
sphere matrix will be more affected than for microspheres with insulin on the surface. Structural changes due to the lyophilisation process were observed in spheres with insulin incorporated, which probably further decreased the release rate. It was therefore concluded that the localisation of insulin is of great importance to the in vivo behaviour of dextran microspheres. This conclusion is not valid as a general rule for microspheres, since the absorptionenhancing effect of DSM with the insulin incorporated into the microsphere matrix is as good as for the dextran microspheres with insulin on the surface.
4.2. Degradable starch microspheres ( DSM) DSM is the most frequently used microsphere system for nasal drug delivery and has been shown to improve the absorption of insulin [16], gentamicin [21], human growth hormone [22], metoclopramide [23] and desmopressin [24]. The bioavailabilities are summarised in Table 1. Insulin administered in DSM to rats resulted in a peak plasma concentration after 7–10 min (Fig. 3) and a rapid dose-dependent decrease in blood glucose. The maximum reductions in blood glucose
Table 1 Nasal drug delivery in degradable starch microspheres (DSM) Species
Drug
Addition enhancer
Dose
t max (min)
Absolute bioavailability (%)
Rat
Insulin
Sheep
Insulin
Sheep
Gentamicin hGH
Sheep
Desmopressin
0.75 IU / kg 1.70 IU / kg 2 IU / kg 2 IU / kg 5 mg / kg 5 mg / kg 2.5 mg / kg 2.5 mg / kg 5 mg / kg 5 mg / kg
8 7 12 22 24 8 120 60 8 18
30 33 4.5 13.1 9.7 57.3
Sheep
— — — LPC — LPC — LPC — LPC
Relative bioavailability (%)
[16] 10.7 31.5
[28] [21]
2.7 14.4 4.7 9.6
Ref.
[22] [24]
190
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194
Fig. 3. Serum insulin concentrations (mean6S.D.) after intranasal administration of insulin in DSM to rats. (s) 0.75 IU / kg insulin; (d) 1.7 IU / kg; ( 3 ) DSM without insulin. Reprinted with permission [16].
level after doses of 0.75 and 1.7 IU / kg were 40 and 64%, respectively, and the absolute bioavailability was approx. 30%. The maximum decrease was reached about 40 min after administration, irrespective of the dose given [16]. In line with these results, a latter study by the same group [17] showed that an insulin dose of 1 IU / kg gave a 45–50% reduction in plasma glucose, 30–40 min after administration. DSM as a delivery system for insulin (2 IU / kg) has also been tested in sheep. The absolute bioavailability was 4.5% and the time to reach maximum effect, i.e. a 50% decrease in plasma glucose, was 60 min [28]. The effect could be further improved by incorporating the absorption enhancer lysophosphatidylcholine (LPC) into the microspheres together with insulin. The plasma glucose profile was similar, with a maximum glucose decrease of 60%, but the bioavailability was improved to 13%. The DSM system has also been used for other drugs. Gentamicin, a polar antibiotic compound, showed no to little absorption from the nasal cavity when administered as a plain solution to sheep [21].
When administered in DSM, the peak plasma concentration was reached 24 min after dosing and was maintained for up to 60 min. The absolute bioavailability was 10% and, as with insulin, the bioavailability could be further improved by incorporating LPC into the spheres. In this case the bioavailability increased to 57% and the absorption was more rapid, with a plasma concentration peak after 8 min. In the same animal model a similar trend was observed for biosynthetic human growth hormone (hGH) [22] and desmopressin [24]. However, the effect was not as pronounced as for gentamicin. The relative bioavailability of hGH was 3% when delivered in DSM and 14% when LPC was included. The plasma concentration peak was reached at 120 min post-dosing when DSM was used and 60 min when also LPC was incorporated into the spheres. This delay of the plasma concentration peak was attributed to the low solubility of hGH, which was probably increased when LPC was included and hence the peak was reached sooner. Desmopressin was rapidly absorbed, with a t max of 8 min for plain
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194
DSM and 18 min for DSM with LPC. The absolute bioavailabilities for desmopressin were 5 and 10%, respectively. Biodegradable starch microspheres have also been used in a clinical investigation of the nasal delivery of the antiemetic drug metoclopramide [23]. The bioavailability was 137% compared to a nasal spray, and the kinetics were the same. A mixture of metoclopramide and microcrystalline cellulose administered as a dry powder did not result in a increased absorption compared to the solution.
4.3. Other types of microspheres The group in Nottingham has also used hyaluronic acid ester microspheres to deliver insulin to sheep [14]. The relative bioavailability of 2 IU / kg insulin was the same as for the plain DSM, i.e. about 11%. However, the maximum decrease in plasma glucose of about 40–60% was reached after 90 min for the hyaluronic acid ester microspheres, compared to 60 min for the DSM. Albumin microspheres containing propranolol with extended drug release resulted in a prolonged plasma concentration plateau in dogs [11]. These microspheres were not administered as a dry-powder formulation, but in a suspension. About 50% of the incorporated propranolol was released in vitro within 4 h. The maximum plasma concentration was obtained after 90 min, and an effective plasma concentration was maintained for up to 12 h. The absolute bioavailability was approx. 84%.
5. Immunological aspects of microspheres as nasal drug delivery systems In the case of nasal drug delivery, a mucosal immune response is an unwanted effect that should be avoided. In a review by O’Hagan and Illum [29] caution is recommended when polymers and microspheres are used as nasal drug delivery vehicles since there are some reports of an adjuvant effect on the immune system. However, the intranasal delivery of a protein to rats in DSM resulted in a lower systemic immune response than intranasal administration of the protein in solution [18]. Intranasal protein delivery in dextran microspheres did not generate an IgA or IgG response in either plasma or nasal washings
191
[30]. Part of the explanation might be the low irritable and inflammatory properties of the delivery system used [31]. Inflammation or irritation attracts immunocompetent cells which would promote a local immunological response. It was therefore concluded that the lack of nasal mucosal immune response to exposure of an immunogen in a dextran microsphere system is evidence that this system does not act as an adjuvant for mucosal immune reactions. For an immunological response the antigen has to be presented to lymphoid tissue containing antibodyproducing cells, i.e. the nasal-associated lymphoid tissue (NALT). The antigen is then drained to the posterior cervical lymph nodes from where a local immune response can be evoked. The use of microspheres as delivery systems for mucosal vaccination has recently been described by several research groups. It is not within the scope of this paper to describe these types of microspheres, but it is worth mentioning that there are differences between the microspheres used for drug delivery and the ones used for vaccination. Microspheres prepared with poly(lactide) PLA [32,33] and poly(lactide-co-glycolide) PLG [34,35] have been used to deliver antigens of different kinds. These microspheres are smaller than the ones used for drug delivery, i.e. , 1 mm, and may be taken up as particles by NALT [36].
6. Toxicological aspects of microspheres as nasal drug delivery systems The effect on the ciliary system can be evaluated in vitro by measuring ciliary beat frequency on explants of ciliated mucosa or in vivo by measuring the mucociliary clearance. The effects on the ciliary beat frequency detected in vitro may be more pronounced than the effects on the ciliary clearance in vivo. In vitro experiments expose the ciliated epithelia directly to the test substance, whereas the cilia are protected by mucus in vivo. Histopathological examination of the nasal mucosa in animals after exposure to different enhancers has also been frequently used to determine damaging effects. The turn-over rate of epithelial cells in the respiratory tract is rather fast. The nasal septum of rabbits recovered in 5 days after total removal of the mucosa [37]. This can also reduce the damaging effect of a substance in vivo. Hence, in vitro ciliary beat
192
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194
frequency tests should be used as indicators of potential damage to the nasal epithelium rather than proof of such damage occurring in the in vivo situation.
6.1. In vitro ciliotoxicity Dextran microspheres had practically no effect on ciliary beat frequency (CBF) measured on explants from rat trachea [31]. It was not possible to detect CBF during the exposure due to the opaque nature of the swollen spheres; however, when the explants were rinsed with Hepes buffer after 15 and 30 min exposure, the cilia immediately started to beat, and after 15 min recovery in Hepes buffer the CBF was back to normal. The immediate recovery of the ciliary movement indicates that the cilia are not damaged by the dextran microspheres. The reason was believed to be that the periciliary fluid is held close to the cell surface and so it is not absorbed by the spheres or that the gel-like layer of dextran spheres formed maintains an environment that is sufficiently humid to keep the cilia intact.
6.2. In vivo mucociliary clearance DSM has been administered daily for 8 days to healthy volunteers [38]. The mucociliary clearance, analysed with a saccharine-dye test, varied between 3.6 and 15 mm / min prior to DSM administration. The clearance was unchanged on days 1 and 8, as was the geometry of the nasal cavity. Thus, no changes in mucociliary clearance could be found after repeated administration of DSM in vivo and the microspheres did not produce congestion or decongestion. These findings are in agreement with the results from the in vitro study previously described for dextran microspheres [31] with properties similar to the starch microspheres. The ciliary function seems to be unaffected by the administration of dry microspheres.
6.3. In vivo mucosal toxicity Administration of DSM twice daily for 8 weeks to rabbits [39] and dextran microspheres administered once daily for 4 weeks to rats [31] induced only slight hyperplasia in the mucosal membrane. Only minor findings were observed at the histopatho-
logical examination, most frequently in the anterior part of the nose. No findings related to the administered microspheres were observed in the middle and posterior parts of the nasal cavity. However, mild focal squamous metaplasia, i.e. replacement of columnar cells by squamous cells, was also detected in rats given dextran microspheres for 4 weeks. It is possible that the finding of squamous metaplasia is a sign of recent damage that has been repaired. It has been shown that following the removal of epithelial cells, the basal cells differentiate, flatten and migrate to establish cell-to-cell contact. This process rapidly covers the damaged zone with squamous epithelium [40].
7. Conclusions All types of microspheres that have been used as nasal drug delivery systems are water-insoluble but absorb water into the sphere’s matrix, resulting in swelling of the spheres. The dextran microsphere system was as effective as an absorption enhancer for insulin as degradable starch microspheres (DSM). The maximum decrease in plasma glucose and the kinetics of the effect curve were similar for both systems. These systems also have a similar effect on mucosal integrity and lack a of adjuvant effect on the immune system after repeated nasal administration. It is therefore likely that the mode of action for improved absorption found for starch microspheres is also applicable to dextran microspheres, i.e. separation of the tight junctions during the swelling process of the microspheres. An important factor for the absorption of large hydrophilic molecules is probably the initial availability of the drug during the period when the tight junctions are separated. The formulation of all kinds of microspheres as nasal drug delivery systems for peptides and proteins ought to take into account studies of the time for complete swelling of the spheres and the release rate of the incorporated drug under conditions resembling the situation in the nasal cavity. This is probably not as crucial for low-molecular weight drugs, which are absorbed to a larger extent without absorption enhancers and for which it is easier to achieve prolonged absorption by retaining the formulation in the nasal cavity for a longer period of time. The use of different types of microspheres as nasal
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194
drug delivery systems has been investigated mostly in rats and sheep, although dogs have also been used. In these animal models the absorption of peptides and proteins has been successfully improved. So far, there are no publications where the peptide and protein absorption-promoting effects of these systems have been investigated in humans. The only absorption study in humans that has been published with microspheres as a delivery system involved the low-molecular weight drug metoclopramide. In this study the bioavailability was improved with microspheres compared to an ordinary nasal spray. However, where peptides and proteins are concerned, there seems to be a tendency for the absorptionpromoting effect to decrease when larger animals are used as compared to rats. Accordingly, investigations in humans are necessary before any conclusions about the potential of microspheres as a nasal drug delivery system for the systemic absorption of peptides and proteins can be drawn.
References [1] T. Nagai, Y. Nishimoto, N. Nambu, Y. Suzuki, K. Sekine, Powder dosage form of insulin for nasal administration, J. Control. Release 1 (1984) 15–22. [2] S.Y. Lin, G.L. Amidon, N.D. Weiner, A.H. Goldberg, Viscoelasticity of anionic polymers and their mucociliary transport on the frog palate, Pharm. Res. 10 (1993) 411–417. [3] M. Zhou, M.D. Donovan, Intranasal mucociliary clearance of putative bioadhesive polymer gels, Int. J. Pharm. 135 (1996) 115–125. [4] Y. Maitani, T. Igawa, Y. Machida, T. Nagai, Plasma levels following intranasal and intravenous administration of human interferon-b to rabbits, Drug Design Delivery 4 (1989) 109–119. [5] T. Nagai, Y. Machida, Bioadhesive dosage forms for nasal administration, in: V. Lenaerts, R. Gurney (Eds.), Bioadhesive Drug Delivery Systems, CRC Press Inc., Boca Raton, FL, 1990, pp. 169–178. [6] P.E. Koochaki, Nasal pharmaceutical composition in powder form. European Patent Application EP 571671, 1993. [7] L. Illum, H. Jørgensen, H. Bisgaard, O. Krogsgaard, N. Rossing, Bioadhesive microspheres as a potential nasal drug delivery system, Int. J. Pharm. 39 (1987) 189–199. [8] B. Lindberg, K. Lote, H. Teder, Biodegradable starch microspheres—A new medical tool, in: S.S. Davis, L. Illum, J.G. McVie, E. Tomlinson (Eds.), Microspheres and Drug Therapy. Pharmaceutical, Immunological and Medical Aspects, Elsevier, Amsterdam, 1984, pp. 153–188. ¨ ´ Microspheres as a nasal [9] P. Edman, E. Bjork, L. Ryden, delivery system for peptide drugs, J. Control. Release 21 (1992) 165–172.
193
[10] U. Conte, P. Giunchedi, L. Maggi, M.L. Torre, Spray dried albumin microspheres containing nicardipine, Eur. J. Pharm. Biopharm. 40 (1994) 203–208. [11] S.P. Vyas, S. Bhatnagar, P.J. Gogoi, N.K. Jain, Preparation and characterization of HSA-propranolol microspheres for nasal administration, Int. J. Pharm. 69 (1991) 5–12. [12] P. Vidgren, M. Vidgren, J. Arppe, T. Hakuli, E. Laine, P. Paronen, In vitro evaluation of spray-dried mucoadhesive microspheres for nasal administration, Drug Dev. Ind. Pharm. 18 (1992) 581–597. [13] B. Kriwet, T. Kissel, Poly(acrylic acid) microparticles widen intercellular spaces of Caco-2 cell monolayers: An examination by confocal laser scanning microscopy, Eur. J. Pharm. Biopharm. 42 (1996) 233–240. [14] L. Illum, N.F. Farraj, A.N. Fisher, I. Gill, M. Miglietta, L.M. Benedetti, Hyaluronic acid ester microspheres as a nasal delivery system for insulin, J. Control. Release 29 (1994) 133–141. ¨ [15] E. Bjork, U. Isaksson, P. Edman, P. Artursson, Starch microspheres induce pulsatile delivery of drugs and peptides across the epithelial barrier by reversible separation of the tight junctions, J. Drug Target. 2 (1995) 501–507. ¨ [16] E. Bjork, P. Edman, Degradable starch microspheres as a nasal delivery system for insulin, Int. J. Pharm. 47 (1988) 233–238. ¨ [17] E. Bjork, P. Edman, Characterization of degradable starch microspheres as a nasal delivery system for drugs, Int. J. Pharm. 62 (1990) 187–192. ¨ [18] E. Bjork, Strach microspheres as a nasal delivery system for drugs. Ph.D Thesis, Uppsala University, Sweden, 1993. [19] C.R. Oechslein, G. Fricker, T. Kissel, Nasal delivery of octreotide: Absorption enhancment by particulate carrier systems, Int. J. Pharm. 139 (1996) 25–32. [20] L. Pereswetoff-Morath, P. Edman, Dextran microspheres as a potential nasal drug delivery system for insulin—in vitro and in vivo properties, Int. J. Pharm. 124 (1995) 37–44. [21] L. Illum, N. Farraj, H. Critchley, S.S. Davis, Nasal administration of gentamicin using a novel microsphere delivery system, Int. J. Pharm. 46 (1988) 261–265. [22] L. Illum, N.F. Farraj, S.S. Davis, B.R. Johansen, D.T. O’Hagan, Investigation of the nasal absorption of biosynthetic human growth hormone in sheep-use of a bioadhesive microsphere delivery system, Int. J. Pharm. 63 (1990) 207– 211. [23] N. Vivien, P. Buri, L. Balant, S. Lacroix, Nasal absorption of metoclopramide administered to man, Eur. J. Pharm. Biopharm. 40 (1994) 228–231. [24] H.S.S.D. Critchley, D. Farraj, L. Illum, Nasal absorption of desmopressin in rats and in sheep. Effect of bioadhesive microsphere delivery system, J. Pharm. Pharmacol. 46 (1994) 651–656. ´ P. Edman, Effect of polymers and microspheres on [25] L. Ryden, the nasal absorption of insulin in rats, Int. J. Pharm. 83 (1992) 1–10. [26] A.-L. Cornaz, A. De Ascentis, P. Colombo, P. Buri, In vitro charcateristics of nicotine microspheres for transnasal delivery, Int. J. Pharm. 129 (1996) 175–183. [27] T.-Y. Ting, I. Gonda, E.M. Gibbs, Microparticles of poly-
194
[28]
[29]
[30]
[31]
[32]
[33]
L. Pereswetoff-Morath / Advanced Drug Delivery Reviews 29 (1998) 185 – 194 vinyl alcohol for nasal delivery. I. Generation by spraydrying and spray-desolvation, Pharm. Res. 9 (1992) 1330– 1335. N.F. Farraj, B.R. Johansen, S.S. Davis, L. Illum, Nasal administration of insulin using bioadhesive microspheres as a delivery system, J. Control. Release 13 (1990) 253–261. D.T. O’Hagan, L. Illum, Absorption of peptides and proteins from the respiratory tract and the potential for development of locally administered vaccine, Crit. Rev. Ther. Drug. Carrier Syst. 7 (1990) 35–97. L. Pereswetoff-Morath, P. Edman, Immunological consequences of nasal drug delivery in dextran microspheres and ethyl(hydroxyethyl) cellulose in rats, Int. J. Pharm. 128 (1996) 23–28. ¨ R. Khan, M. Dahlin, P. L. Pereswetoff-Morath, S. Bjurstrom, Edman, Toxicological aspects of the use of dextran microspheres and thermogelling ethyl(hydroxyethyl) cellulose (EHEC) as nasal drug delivery systems, Int. J. Pharm. 128 (1995) 9–21. A.J. Almeida, H.O. Alpar, M.R.W. Brown, Immune response to nasal delivery of antigenically intact tetanus toxoid associated with poly(L-lactic acid) microspheres in rats, rabbits and guinea-pigs, J. Pharm. Pharmacol. 45 (1993) 198–203. H.O. Alpar, A.J. Almeida, Identification of some of the physico-chemical characteristics of microspheres which influence the induction of the immune response following
[34]
[35]
[36]
[37]
[38]
[39]
[40]
mucosal delivery, Eur. J. Pharm. Biopharm. 40 (1994) 198– 202. E.S. Cahill, D.T. O’Hagan, L. Illum, A. Barnard, K.H.G. Mills, K. Redhead, Immune responses and protection against Bordetella pertussis infection after intrenasal immunization of mice with filamentous haemagglutinin in solution or incorporated in biodegradable microparticles, Vaccine 13 (1995) 455–462. J.H. Eldridge, J.K. Staas, J.A. Meulbroek, J.R. McGhee, T.R. Tice, R.M. Gilley, Biodegradable microspheres as a vaccine delivery system, Mol. Immunol. 28 (1991) 287–294. R.M. Carr, C.M. Lolachi, R.G. Albaran, D.M. Ridley, P.C. Montgomery, N.L. O’Sullivan, Nasal-associated lymphoid tissue is an inductive site for rat tear IgA antibody responses, Immunol. Invest. 25 (1996) 387–396. Y. Ohashi, Y. Nakai, H. Ikeoka, H. Furuya, Regeneration of nasal mucosa following mechanical injury, Acta Otolaryngol. Suppl. 486 (1991) 193–201. ¨ K. Holmberg, E. Bjork, P. Edman, Influence of degradable starch microspheres on the human nasal mucosa, Rhinology 32 (1994) 74–77. ¨ ¨ P. Edman, Morphologic examination E. Bjork, S. Bjurstrom, of rabbit nasal mucosa after nasal administration of degradable starch microspheres, Int. J. Pharm. 75 (1991) 73–80. ¨ ¨ J.S. Erjefalt, I. Erjefalt, F. Sundler, C.G.A. Persson, Microcirculation-derived factors in airway epithelial repair in vivo, Microvascular Res. 48 (1994) 161–178.