ORIGINAL ARTICLE
JOURNAL OF OCULAR PHARMACOLOGY AND THERAPEUTICS Volume 00, Number 00, 2014 ª Mary Ann Liebert, Inc. DOI: 10.1089/jop.2013.0164
Dorzolamide Cyclodextrin Nanoparticle Suspension Eye Drops and Trusopt in Rabbit ´ sgrı´msdo´ttir,4 Gauti Jo´hannesson,1,2 Maria D. Moya-Ortega,3,4 Gudrun Marta A 5 6 3 Bjarni A. Agnarsson, Sigru´n H. Lund, Thorsteinn Loftsson, and Einar Stefa´nsson1,4
Abstract
Purpose: Dorzolamide nanoparticle g-cyclodextrin eye drops may prolong the effect of dorzolamide on intraocular pressure. We test whether the nanoparticle drops have an irritating or toxic effect on the eye in an in vivo rabbit model. Methods: Eighteen pigmented rabbits were divided into 4 groups receiving dorzolamide nanoparticle g-cyclodextrin eye drops · 1/day or · 2/day, Trusopt (dorzolamide HCl) · 3/day, and untreated controls that received no drops. The rabbits received treatment for 1 month. After sacrifice, 33 eyes and 25 Harderian glands were evaluated for histopathology in a masked way. Results: Mild inflammation was seen in 19/31 eyes and 13/23 Harderian glands. The difference in inflammation (n = eyes/n = glands)between the g-cyclodextrin nanoparticle eye drops · 1/day (n = 5/5), · 2/day (n = 5/3), Trusopt (n = 7/4), or untreated control (n = 2/0) groups was nonsignificant in both eyes and glands (P = 0.87 and P = 0.92) Acute inflammation was seen in 1 Harderian gland that received g-cyclodextrin nanoparticle eye drops · 2/day. The difference in conjunctival injection between the groups was nonsignificant (P = 0.30). Conclusions: Dorzolamide g-cyclodextrin nanoparticle eye drops are no more locally toxic or irritating to the eye than Trusopt.
number of drops per day from 3 to 1 would not only provide comfort for the patient but would also improve the compliance, that is, how well the patient follows the prescription. Before testing dorzolamide cyclodextrin nanoparticle eye drops in a long-term phase II human study, the toxicology profile of dorzolamide g-cyclodextrin nanoparticle eye drops needs to be assessed. Thus, this 1-month in vivo comparison of clinical and histological irritation between Trusopt and dorzolamide nanoparticles was performed. Our hypothesis was that the effect of the dorzolamide g-cyclodextrin nanoparticle eye drops on ocular tissues would be no more than for Trusopt.
Introduction
g-C
yclodextrin nanoparticles have been shown to be an effective vehicle to transport lipophilic drugs through biological membranes, including the eye.1 Recent in vitro and in vivo studies indicate that dorzolamide is applicable to the g-cyclodextrin nanoparticle drug delivery platform.2 A pilot study in humans has also showed that dorzolamide nanoparticle g-cyclodextrin eye drops had a longer IOP lowering effect than Trusopt and could possibly be administered once daily compared with 3 times a day for Trusopt.3 Topically delivered carbonic anhydrase inhibitor dorzolamide is a widely used drug to treat glaucoma. The commercially available dorzolamide, Trusopt, is administered to the eye 2 to 3 times per day. Treatment compliance worsens if a medication has to be taken twice per day (compared to once per day) and it worsens further if the dose frequency is increased to 3 times per day.4,5 The possibility of reducing the
Methods In vivo rabbit model This study was performed in accordance with the ARVO declaration for the use of laboratory animals in research.
Departments of Ophthalmology1 and Pathology,5 National University Hospital, University of Iceland, Reykjavik, Iceland. 2 Department of Clinical Sciences, Ophthalmology, Umea˚ University, Umea˚, Sweden. 3 Faculty of Pharmaceutical Sciences, University of Iceland, Reykjavik, Iceland. 4 Oculis ehf., Reykjavik, Iceland. 5 Department of Pathology, National University Hospital, University of Iceland, Reykjavik, Iceland. 6 Department of Public Health Faculty of Medicine, University of Iceland, Reykjavik, Iceland.
1
´ HANNESSON ET AL. JO
2
The study was approved by the National Animal Research Committee of Iceland. The study was performed at The Institute for Experimental Pathology at the University of Iceland. Pigmented rabbits (n = 18) were fed on a regular diet and were placed unanesthetized in restraint boxes. The rabbits were divided into 4 groups: The first group (n = 5) received gcyclodextrin nanoparticle dorzolamide eye drops once daily in the morning, the next group (n = 5) received g-cyclodextrin nanoparticle dorzolamide eye drops twice daily, the third group (n = 5) received Trusopt 3 times daily, and finally, an untreated control group (n = 3) did not receive any drops. One eye from each control rabbit was used for comparison in this study. Both eyes of treated rabbits received the same treatment. The dorzolamide g-cyclodextrin nanoparticle eye drops used in this study were identical to the drops produced and aimed for use in humans. The eye drops consist of unionized dorzolamide (3% w/v; pH 7.2) in an aqueous vehicle of lower viscosity (71 cP) containing g-cyclodextrin and hydroxylpropylmethylcellulose as stabilizing agents, and other excipients as benzalkonium chloride, disodium edetate, monobasic sodium phosphate, dibasic sodium phosphate, and tyloxapol. Dorzolamide/g-cyclodextrin complexes selfassemble together resulting in nano- or microparticles, which enhance and prolong drug delivery to the eye.2 The rabbits were sacrificed after 34 days of drug administration by an intravenous injection of sodium pentobarbital. The eyes were enucleated and extra ocular tissues removed immediately after death. Before sacrifice, the rabbits’ eyes were examined by an ophthalmologist to identify any local inflammatory adverse effects.
Sample preparation for histopathological examination The eyes were trimmed and fixed in Bouin solution containing 700 mL of a saturated solution of picric acid (1.2%), 250 mL of formaldehyde (37%), and 50 mL of acetic acid glacial, for 24 h. The eyes where then rinsed twice with 50% ethanol aqueous solution. After that, the eyes were kept in 70% ethanol aqueous solution · 3 for 4 days. The eyes were then processed by routine paraffin embedding, and 4-
Table 1.
mm sections were cut, mounted, and stained with hematoxylin–eosin for their analysis. Histopathological examination was performed according to a 4-grade scoring system (minimal; mild; moderate; severe) in a masked way by one observer where all specimens were evaluated individually and compared with control samples.
Statistics The null hypothesis of equal number of incidences of inflammation between treatment groups was tested with the Fisher’s exact test.
Results Seventeen rabbits completed the study. One rabbit (dorzolamide g-cyclodextrin nanoparticle · 2/day) was sacrificed after 2 weeks due to breathing problems. Autopsy revealed that the rabbit had pneumonia. Its eyes were examined before it was put to death and no adverse effects were seen. The clinical eye examination was normal except for mild (1 out of 4) conjunctival injection seen in both eyes of 1 control rabbit, in both eyes of 1 rabbit that had received gcyclodextrin nanoparticle dorzolamide · 1/day, and finally in 1 eye that had received g-cyclodextrin nanoparticle dorzolamide · 2/day. The number of incidences between treated and untreated groups was nonsignificant (P = 0.30). Histological examination revealed inflammation (Table 1) in 19 out of 31 eyes. The difference in the number of inflamed eyes between the groups was nonsignificant (P value = 0.87) (Table 1). The inflammation was always located in the limbal area where the conjunctiva and sclera connect to each other. Figure 1a shows a normal histological section of the cornea and Fig. 1b shows mild inflammation in the corneal epithelium. The Harderian gland was missing in 8 specimens, including 2 of the control eyes resulting in only one control Harderian gland for comparison. Histopathological examination revealed minimal inflammation (Fig. 2) in 13 out of 23 Harderian glands (Table 1). The difference in inflammation between the treated groups was nonsignificant (P value = 0.92) Acute inflammation was seen in 1 Harderian gland (Fig. 3) from a rabbit receiving g-cyclodextrin nanoparticle dorzolamide · 2/day.
Distribution and Amount of Inflammation in Eyes and Harderian Glands in All Groups Inflammation
Eyes DorzNP · 1/day DorzNP · 2/day Trusopt · 3/day Control Harderian gland DorzNP · 1/day DorzNP · 2/day Trusopt · 3/day Control
-
(+)
+
++
+++
Focal
Missing
5 3 3 1
1 2 — —
4 3 6 1
— — 1 1
— — — —
— — — —
— — — —
3 3 3 1
5 3 4 —
— — — —
— — — —
— — — —
— 1 — —
2 1 3 2
DorzNP, g-cyclodextrin nanoparticle dorzolamide; - , no inflammation; ( + ), minimal inflammation (lymphocytes); + , mild inflammation (lymphocytes); + + , moderate inflammation (lymphocytes); + + + , severe inflammation (lymphocytes); focal, acute inflammation (neutrophils + ); missing, missing specimen.
COMPARABLE TOXICITY OF DORZOLAMIDE DRUGS
3
FIG. 3. Acute focal inflammation (arrow) with neutrophils in the Harderian gland of one of the rabbits receiving dorzolamide g-cyclodextrin nanoparticle eye drops 2 times daily.
cytoplasm and therefore it was not possible to show a histological difference between control specimens and treated specimens.
FIG. 1. (a) Histological section of a normal cornea from a rabbit receiving Trusopt. (b) Mild inflammation with lymphocyte infiltration (arrows) in the corneal epithelium from a rabbit receiving Trusopt. In all cases of mild Harderian gland inflammation (Fig. 2b), the gland had dual histology, that is, one part of the gland had acini with cells containing cytoplasm with vacuoles and another part had acini with granular cytoplasm. In all cases where inflammation was found, it was limited to the latter part. The control gland had only vacuolated
FIG. 2. A histological section of the Harderian gland, which shows the dual histology of the gland in a rabbit receiving Trusopt. Lower section shows acini with vacuoles in the cytoplasm, while the upper part contains granular cytoplasm. Arrows show lymphocytes.
Discussion The toxicology profile of dorzolamide g-cyclodextrin nanoparticle is comparable to Trusopt and untreated controls as to the clinical and histological effect of the drug on ocular tissues. The dosing of dorzolamide g-cyclodextrin nanoparticle in humans is expected to be · 1/day. For safety reasons, we also doubled the dose in this toxicology study. The histopathological evaluation showed no significant difference in ocular inflammation between the groups, lymphocytes were seen in most preparations, including control samples, and no acute inflammation with neutrophil infiltration was seen. In all cases, the inflammation was located in the limbal area, and no specimen showed inflammation in deeper structures of the eye. In the Harderian glands, a dual histology was seen where one part of the gland had vacuoles in the cytoplasm, while the other part had granular cytoplasm. Lymphocytes were seen in glands in all treated groups and they were always restricted to the granular area. Unfortunately, only one control sample contained a Harderian gland and this sample contained no granular cytoplasm. It might be possible that the presence of lymphocytes in the granular cytoplasm is a normal finding. However, Davis6 describes this dual histology in the Harderian gland, but does not mention lymphocytes as being a normal finding in either part of the gland. During the course of the study, 1 rabbit in the dorzolamide g-cyclodextrin nanoparticle · 2/day group got sick with breathing problems. The rabbit was put to death and autopsy revealed that it suffered from pneumonia. It is therefore unlikely that the eye drops might have caused the breathing problems. In conclusion, this study shows that dorzolamide gcyclodextrin nanoparticle given once daily has a similar effect on the ocular tissues as Trusopt. Thus, dorzolamide g-cyclodextrin nanoparticle is safe to use in human subjects for a long-term study.
´ HANNESSON ET AL. JO
4
Acknowledgments This study was supported by unrestricted grants from the Icelandic research council, Research to prevent blindness in Iceland, Technology Development Fund, Swedish Medical Society, Kempe Fund, and Oculis ehf. The authors thank ´ lo¨f Sigur+ardo´ttir for help with sample preparation and O Eggert Gunnarsson and Elvar Rı´khar+sson for taking great care of the animals.
Author Disclosure Statement The financial disclosures of the authors are the following: Gauti Jo´hannesson–none; Maria D. Moya-Ortega–employee ´ sgrı´msdo´ttir–employee (Oculis (Oculis ehf ); Gudru´n M. A ehf ); Bjarni A. Agnarsson–none; Sigru´n H. Lund–none; Thorsteinn Loftsson–patent, equity owner (Oculis ehf ); Einar Stefa´nsson–patent, equity owner (Oculis ehf ).
References 1. Loftsson, T., Vogensen, S.B., Brewster, M.E., and Konradsdottir, F. Effects of cyclodextrins on drug delivery through biological membranes. J. Pharm. Sci. 96:2532– 2546, 2007. 2. Jansook, P., Stefansson, E., Thorsteinsdottir, M., Sigurdsson, B.B., Kristjansdottir, S.S., Bas, J.F., Sigurdsson, H.H., and Loftsson, T. Cyclodextrin solubilization of carbonic anhydrase inhibitor drugs: formulation of dorzolamide eye drop
3.
4. 5. 6.
microparticle suspension. Eur. J Pharm. Biopharm. 76:208– 214, 2010. Gudmundsdottir, B.S., Dyrleif Petursdottir, D., Asgrimsdottir, G.M., Gottfredsdottir, M.S., Hardarson, S.H., Jo´hannesson, G., Kurkov, S.V., Jansook, P., Loftsson, T., and Stefansson, E. gCyclodextrin nanoparticle eye drops with dorzolamide—effect on intraocular pressure in man. J. Ocul. Pharmacol. Ther. 30:35–41, 2014. Buchan, J.C., Siddiqui, S., and Gilmour, D. Once daily drop regimes help reduce involuntary non-compliance. Graefes Arch. Clin. Exp. Ophthalmol. 245:327–328, 2007. Buller, A.J., Morgan, L.H., and Hercules, B.L. Patients prefer once-daily glaucoma drops. Graefes Arch. Clin. Exp. Ophthalmol. 245:293–294, 2007. Davis, F.A. The anatomy and histology of the eye and orbit of the rabbit. Trans. Am. Ophthalmol. Soc. 27:400, 402–441, 1929.
Received: August 28, 2013 Accepted: March 10, 2014 Address correspondence to: Dr. Gauti Jo´hannesson Department of Ophthalmology National University Hospital University of Iceland IS 101 Reykjavı´k Iceland E-mail:
[email protected]
