2023 Volume 46 Issue 7 Pages 921-928
Intravitreal injection therapy of anti-vascular endothelial growth factor (VEGF) antibody or steroids is the mainstream for patients with age-related macular degeneration (AMD). However, since intravitreal injection is invasive administration, side effects such as endophthalmitis are major problems. In this study, we selected eye drops as a non-invasive treatment method, and aimed to develop eye drops that can deliver TAK-593 (VEGF receptor tyrosine kinase inhibitor) to the posterior segment of the eye. Since TAK-593 is a poorly water-soluble drug, the TAK-593 emulsion was formulated. The solubility of TAK-593 in various oils was measured, and the oil used for the emulsion was selected. Furthermore, viscosity enhancers were added to the emulsion in order to improve the drug delivery into the eye. As viscosity enhancer, xanthan gum was selected based on the properties and the viscosity of the emulsion. The delivery of TAK-593 to the posterior eye was increased by the formulation concentration and the addition of viscosity enhancers. In the laser-induced choroidal neovascularization model, TAK-593 emulsion eye drops showed the same angiogenesis-suppression efficacy as anti-VEGF antibody intravitreal injection. From these results, it was revealed that TAK-593 with an effective drug concentration can be delivered to the posterior eye by non-invasive eye drop administration.
Ocular neovascular diseases, including age-related macular degeneration (AMD), diabetic retinopathy (DR) and retinal vein occlusion (RVO), are the highly prevalent causes of irreversible visual impairment in developed countries.1,2) In the foreseeable future, the incidence of these pathologies is expected to rise due to population aging and increasing life expectancy. By 2040, the number of individuals in Europe with early AMD will range between 14.9 and 21.5 million, and for late AMD between 3.9 and 4.8 million.3)
Currently, intravitreal injection therapy of anti-vascular endothelial growth factor (VEGF) drugs (ranibizumab, aflibercept, etc.) or steroids is the mainstream for patients with AMD and diabetic macular edema (DME).4–7)
Anti-VEGF drugs are antibodies developed to bind VEGF and avoid the interaction of VEGF with its receptors. Aflibercept (Eylea®) is a recombinant fusion protein of specific domains from human VEGFR1, VEGFR2, IgG1.8) It was approved in 2011, and is designed to target VEGFA, VEGFB, and placental growth factor. Its recommended dosing interval is once every 4 weeks (Q4) for the first 3 months, followed by a dosing interval of once every 8 weeks (Q8). In VIEW studies, patients treated with aflibercept had no retinal fluid at weeks 52 and 96.9) Recent subgroup analyses of the VIEW trials suggested a superior morphologic efficiency of aflibercept in reducing intraretinal and subretinal fluid as well as reducing RPE elevation, which suggest a superior anatomic efficacy of aflibercept compared with ranibizumab.
However, since intravitreal injection is invasive and frequent administration, the risk of endophthalmitis and the mental and economic burden on the patient are major problems.10–12) By developing eye drops for the treatment of these posterior ocular diseases, the burden on the patient can be reduced and treatment options can be provided.13)
Many tyrosine kinase inhibitors (TKIs) having anti-VEGFR activity have been developed as anticancer agents.14) Treatment of AMD with tyrosine kinase inhibitors has already been reported. These reports report not only invasive treatment methods such as intravitreal administration of sunitinib,15) but also non-invasive treatment methods such as oral administration of vorolanib16) and eye drops of sorafenib, regorafenib, and pazopanib.17–20) However, clinical development did not show efficacy and was terminated.21)
TAK-593 is an anti-VEGF/platelet-derived growth factor (PDGF) drug developed by Takeda Pharmaceutical Company Limited, and has been developed as an anti-cancer agent.22) Since the activity of TAK-593 is extremely high,23) it can be expected that a sufficient concentration showing a medicinal effect can be delivered by eye drop administration. Emulsions are used as formulations for the intraocular transfer of hydrophobic small molecule compounds for ophthalmic delivery, improving bioavailability and patient compliance. Examples of ophthalmic emulsion products include Restasis® (cyclosporine ophthalmic emulsion) and Durezol® (difluprednate ophthalmic emulsion), which are used to treat keratoconjunctivitis sicca (or dry eye) and anterior uveitis, respectively. In addition, Durezol® has been reported to deliver difluprednate to the posterior eye of the retina, choroid, etc., and the emulsion can be expected as a preparation capable of delivering small molecule compounds to the posterior eye.24) In addition, it has been reported that liposomes,25) lipoprotein-based nanoparticles26) and poly(D,L-lactide-co-glycolide) nanoparticles27) can also deliver small molecule compounds into retina by eye drop administration.
In this study, an emulsion with xanthan gum was selected as a preparation for delivering TAK-593 to the posterior segment of the eye by eye drops. Xanthan gum acts as a viscosity enhancer. Furthermore, Faradi et al. reported that the topical administration of xanthan gum does not cause damage to corneal surface.28) Therefore, we confirmed the pharmacokinetics and pharmacological effects of the designed emulsion eye drops were confirmed, and the potential as a therapeutic agent for age-related macular degeneration was evaluated.
TAK-593 was provided by Takeda pharmaceutical Company Limited (Tokyo, Japan). Castor oil and Liquid paraffin was purchased from Maruishi Pharmaceutical Company (Osaka, Japan). Soybean oil, Corn oil, Cotton oil, Olive oil and Rapeseed oil and Polyethylene glycol 400 was purchased from Nacalai Tesque (Kyoto, Japan). Polysorbate 80 was purchased from NOF Corporation (Tokyo, Japan). Hydroxypropylmethyl cellulose (HPMC, 65SH-4000) and Methylcelluloce (MC, SM-50) was purchased from Shin-Etsu Chemical Company (Tokyo, Japan). Glycerin was purchased from Sakamoto Yakuhin Kogyo (Osaka, Japan). Water was purified with a Milli-Q purification system (Millipore, Japan). Other reagents were of HPLC grade or the highest grade commercially available. Aflibercept (Eylea®) was purchased from Bayer Yakuhin (Osaka, Japan).
Measurement of TAK-593 Solubility in OilsExcess amount of TAK-593 was added to the oils. The mixture was stirred at 25 °C for about 24 h. While maintained at the same temperature, the mixture was filtered through a 0.45 µm membrane filter in order to remove undissolved drug particles in the saturated solution. The resulting filtrate was measured the content of TAK-593. The concentration of TAK-593 was determined by HPLC (Shimadzu Prominence system (Kyoto, Japan) equipped with UV detector). Separation was carried out at 40 °C using a reverse-phase C18 column (CAPCELL PAK C18 MG II, 3 µm, 4.6 × 100 mm, Shiseido, Tokyo, Japan). The mobile phase consisted of 20 mM phosphate buffer solution (pH 5.0) and acetonitrile (13 : 7, v/v). The detection wavelength was 230 nm, and a flow rate of 1.0 mL/min was employed. A sample volume of 20 µL was injected.
Preparation of TAK-593 Emulsions and SolutionsOil-in-water emulsions composed of TAK-593 (0.02 or 0.06% (w/v)) as a lipophilic drug, castor oil (5% (w/v)) as a lipid phase, and polysorbate 80 (4% (w/v)) as an emulsifying agent were prepared. The emulsions were prepared in two steps; in the first step, polysorbate 80 (8% (w/v)), glycerin (4.4% (w/v)), and disodium edetate (0.04% (w/v)) were added to water and mixed, and sodium acetate (0.1% (w/v)) and boric acid (0.2% (w/v)) was then dissolved in the solution as a buffering agent. TAK-593 was added to castor oil and dissolved at 70 °C, and this mixture was then added to the solution preheated to 70 °C and emulsified by a homogenizer (Robomics, PRIMIX Corporation, Hyogo, Japan) at 8000 rpm for 15 min. The mixture was then cooled to room temperature and adjusted to pH 5.5. In the second step, the coarse emulsion was treated with a high-pressure emulsifier (Starburst, Sugino Machine Limited, Toyama, Japan) at 20 °C with an inlet pressure of 240 MPa. The coarse emulsion was processed through the Starburst with 20 passes, and collected into glass beakers. The emulsion was then cooled to room temperature. Viscosity enhancers were dissolved in acetate buffer at pH 5.5 to prepare 1% solutions, respectively. Viscosity enhanced emulsion was prepared by mixing and stirring the same volumes of emulsion and 1% viscosity enhancer solution. TAK-593 solutions and suspensions were composed of benzyl alcohol (2% (w/v)), polysorbate 80 (5% (w/v)), polyethylene glycol 400 (5% (w/v)).
Characterization of These EmulsionsMean particle sizes of emulsions were measured by photon correlation spectroscopy (PCS) using a dynamic light scattering particle size analyzer (Nano-ZS, Malvern Instruments, U.K.) with 100-fold dilution by distilled water at 25 °C. Viscosity was determined by a cone plate viscometer (TVE-25, TOKI SANGYO Co., Ltd., Tokyo, Japan) at 20 °C. The system was calibrated using standard viscosity fluids (Nippon-Grease Co., Ltd.). The appearance of TAK-593 emulsions before and after centrifugation (20000 × g, 20 min) was observed to evaluate whether creaming of emulsion was occurred or not.
AnimalsAll animals used in this study were treated in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. The protocols were approved by the Institutional Animal Care and Use Committee of Research Laboratories at Senju Pharmaceutical Co., Ltd. (Permit Nos. 20140516-01 and 20140924-01). Eight-week-old male Brown Norway rats were obtained from Charles River Japan (Yokohama, Japan). Male Japanese White rabbits (Kbs:JW) weighing approximately 1.5–1.99 kg were obtained from Kitayama Labes Co., Ltd. (Nagano, Japan). The animals were maintained in conventional animal rooms and individually housed in plastic cages in an air-conditioned room with a temperature of 22 ± 3 °C, 55 ± 10% relative humidity, and a 12 h light/dark cycle, and fed ad libitum.
Pharmacokinetics StudyRabbits were given a single instillation of 50 µL of different formulations in the left eye only. At 1 h after the instillation, about 1 mL of blood was drawn from the ear vein using a syringe fitted with a needle containing heparin. The animals were then euthanized by intravenous administration of 100 mg/kg of pentobarbital injection (Somnopentyl injection, Kyoritsu Seiyaku, Japan) through the auricular vein, immediately after which their eyes were washed with saline. Just after euthanasia, both eyes were enucleated from each animal. Posterior retina was isolated from the eye, and TAK-593 concentrations in the tissues were determined by LC-tandem mass spectrometry (LC-MS/MS) (Shimadzu Prominence system).
HPLC was performed using Shimadzu Prominence system equipped with binary pump, in-line degasser, column oven, and thermostatically controlled autosampler. A triple quadrupole mass spectrometer API4000 LC/MS/MS System (AB Sciex, Framingham, MA, U.S.A.) was coupled to the HPLC system. TAK-593 were separated using Luna C18 (2) (2.0 mm × 50 mm) column with 5 µm particle size (Phenomenex, Torrance, CA, U.S.A.) at 50 °C. The analytes were eluted with a mobile phase consisting of 0.1% formic acid (mobile phase A) and 100% methanol with 0.1% formic acid (mobile phase B), pumped at a flow rate of 0.2 mL/min. The gradient started at 40% mobile phase B and then was linearly increased to 90% B over 2 min. Subsequently, the eluent composition was maintained for 3 min before it was decreased to 40% mobile phase B. It was returned to its initial conditions for a minute and then equilibrated 3 min at the end.
For tissue drug levels, the tissue was homogenized with 50% methanol solution. After mixing the solution by aspiration and dispensing the samples were transferred onto the Isolute® SLE+ 200 µL plate. The sample was left to load onto the phase for 7 min. Each well was eluted with 1200 µL of tert-butanol into each well of a 2 mL 96 deep well collection plate. The eluent was evaporated to dryness with nitrogen at 30 °C then reconstituted in 300 µL of 75% methanol solution.
Laser-Induced Choroidal Neovascularization ModelRats were anesthetized with a 5 mL/kg body weight intraperitoneal injection of a mixture containing medetomidine hydrochloride (Domitor, Nippon Zenyaku Kogyo, Japan), Midazolam (Dormicum Injection 10 mg, Astellas, Japan) and butorphanol tartrate (Meiji Pharma, Japan). Pupils were dilated with topical tropicamide/phenylephrine (Midrin-P, Santen Pharmaceutical, Japan). Visualization of the fundus was aided by placing a cover glass on the eye over hydroxy ethylcellulose. Ophthalmic Laser (Ultima2000SE, Lumenis, Japan) was used to encircle the optic nerve on the retina of the right eye with six laser spots (100 µm each, 532 nm, 120 mW, 0.1 s). Breakage of Bruch’s membrane was confirmed by bubble formation and used as an end point for treatment. Animals with severe hemorrhages in the retina were excluded from the study. After the laser photocoagulation, the different formulations (10 µL) were instilled into the right eye four times a day. In intravitreal (IVT) administrations, 40 mg/mL aflibercept and vehicle solution were injected 5 µL, 1 time at the day of choroidal neovascularization (CNV) induction. After euthanasia on day 14 following laser photocoagulation, the eyes were enucleated and prefixed with 10% formaldehyde for 30 min. Retina-RPE-choroid complexes were microsurgically isolated from the prefixed eyes and further fixed with 10% formaldehyde for 1 h. The retina was removed from RPE-choroid complexes. The RPE-choroid complexes were then washed with phosphate buffered saline (PBS) and incubated for 30 min with PBS blocking buffer containing 0.5% Triton X-100 and 1% bovine serum albumin (BSA). To visualize neovascularization, RPE-choroid complexes were incubated overnight at 4 °C with 0.01% fluorescein isothiocyanate (FITC)-conjugated isolectin B4 derived from Griffonia (Bandeiraea) simplicifolia agglutinin (Vector Laboratories, Peterborough, U.K.) diluted with PBS containing 0.5% Triton X-100. After the RPE-choroid complexes were washed with PBS and sealed with VECTASHIELD (Vector Laboratories), CNV was observed with a fluorescence microscope (BX-51; Olympus, Japan). Image visualization software (Image pro® plus) was used for determination the size of CNV.
Statistical AnalysisAll results were reported as means ± standard deviation, and n values represented the number of animals used. Statistical analyses of the data were performed using one way ANOVA followed by Dunnett’s test and Tukey’s test for multiple comparisons, or paired t-test and Student’s t-test for two samples. p < 0.05 was considered statistically significant.
The solubility of TAK-593 in vegetable oils and liquid paraffin at 25 °C was determined to select the suitable oil for the preparation of emulsions. The results are summarized in Table 1. Vegetable oils is often used for parenteral applications29); however, TAK-593 showed low solubility at 0.009% in olive oil, and TAK-593 solubility in the other oils was under 0.008%. TAK-593 was most soluble in castor oil in comparison with the other vegetable oils and liquid paraffin, with 0.311%. From these results, we selected castor oil as lipid phase for TAK-593 ophthalmic lipid emulsion.
| Oils | Solubility |
|---|---|
| Castor oil | 0.311% |
| Olive oil | 0.009% |
| Corn oil | 0.008% |
| Cotton oil | 0.008% |
| Rapeseed oil | 0.007% |
| Soybean oil | 0.007% |
| Liquid paraffin | N.D. |
The description and appearance of each formulation after the addition of various viscosity enhancers is shown in Table 2 and Fig. 1. The addition of HPMC, MC, PVA and sodium alginate caused creaming after 2 to 3 h. Further, creaming of emulsion was accelerated by centrifugation. On the other hand, viscosity enhancer-free formulation and formulation containing xanthan gum did not occur creaming. Creaming of the viscosity enhancer-free formulation was accelerated by centrifugation but creaming was not observed for the xanthan gum containing formulation. These results showed that xanthan gum inhibited creaming. Figure 2 shows the viscosity of emulsions to which various viscosity enhancers were added at a concentration of 0.5%. The xanthan gum contained formulation was more viscous than the other formulations and exhibited pseudo-plastic behavior. Presumably, emulsions with xanthan gum did not creaming because xanthan gum is a higher viscosity than other enhancers in the rest state. The results of appearance and viscosity evaluations indicated that xanthan gum was the suitable viscosity enhancer as an additive to TAK-593 emulsions.
| Viscosity enhancers | Particle size | Creaming*1 | |||
|---|---|---|---|---|---|
| Z-average (PdI) | D10 | D90 | Before centrifugation | After centrifugation | |
| None | 124.9 nm (0.183) | 72.9 nm | 260 nm | − | ++ |
| 0.5% HPMC | 128.0 nm (0.171) | 85.9 nm | 244 nm | + | ++ |
| 0.5% MC | 124.2 nm (0.207) | 68.3 nm | 278 nm | ++ | ++ |
| 0.5% PVA | 124.5 nm (0.202) | 71.3 nm | 278 nm | + | ++ |
| 0.5% Sodium alginate | 112.2 nm (0.259) | 56.3 nm | 298 nm | + | ++ |
| 0.5% Xanthan gum | 132.2 nm (0.284) | 63.3 nm | 322 nm | − | − |
*1; A homogeneous emulsion was rated as homogenous (−), an emulsion in which the upper and lower parts had different shades of a color as Creaming (+), and an emulsion with a creaming phase as Creaming (++).
The left side is the before centrifugation and the right side is the after centrifugation (20000 × g, 20 min).
Concentration of each viscosity enhancers uniformed at 0.5%. The viscosity in the range of shear rate from 5 to 200 s−1 was measured at 20 °C.
Stability of TAK-593 emulsion with xanthan gum was investigated in glass ampules. The results of stability are summarized in Table 3. Residual percentage of TAK-593 was 94.0, 93.5 and 93.6 at 40 °C for 2 months, 60 °C for 2 weeks and 4 weeks, respectively, and the degradation of TAK-593 was slight. This was probably due to degradation of TAK-593 dissolved in water phase of the emulsion, although TAK-593 solubility in water phase was low (under 1 µg/mL).
| Storage condition | Residual percentage of TAK-593 | Particle size | Viscosity (at 200 s−1) | ||
|---|---|---|---|---|---|
| Z-average (PdI) | D10 | D90 | |||
| Initial | 100.0% | 127.8 nm (0.232) | 67.6 nm | 308 nm | 26.33 mPa·s |
| 25 °C 2 months | 102.5% | 128.5 nm (0.234) | 65.3 nm | 297 nm | 26.38 mPa·s |
| 40 °C 4 weeks | 103.3% | 124.9 nm (0.225) | 68.7 nm | 284 nm | 27.05 mPa·s |
| 40 °C 2 months | 94.0% | 121.3 nm (0.231) | 65.5 nm | 285 nm | 26.32 mPa·s |
| 60 °C 2 weeks | 93.5% | 197.0 nm (0.176) | 126 nm | 386 nm | 28.17 mPa·s |
| 60 °C 4 weeks | 93.6% | 262.8 nm (0.193) | 154 nm | 557 nm | 24.92 mPa·s |
Residual percentage of TAK-593 (%) = TAK-593 content at each time/initial TAK-593 content.
Neither separation nor change in particle size under storage condition at 25 and 40 °C for 2 months was observed, but the z-average particle size was slightly increased from 127.8 to 262.8 nm at 60 °C for 4 weeks.
Pharmacokinetics StudyFigure 3 shows the concentration of TAK-593 in choroid, retina and plasma at 1 h after instillation. The concentration of TAK-593 in these tissues increased depending on the dosage of TAK-593. These tissues concentration of TAK-593 of 0.02% TAK-593 emulsion was similar to that of 0.02% TAK-593 aqueous solution. Furthermore, the local intraocular delivery into choroid and retina evaluated as the difference between the tissue concentration of the instilled eye and that of the contralateral eye is shown in Fig. 4. The locally-distributed TAK-593 concentrations of 0.06% TAK-593 emulsion and 0.06% TAK-593 emulsion with xanthan gum was estimated to be 10.22 and 19.64 ng/ng in retina. The addition of xanthan gum to 0.06% TAK-593 emulsion tended to increase the local delivery of TAK-593 into the retina, but there was no significant difference.
Each formulation was administered in right eye of rabbits at doses of 50 µL/eye. Data represent the mean ± standard deviation (S.D.) (n = 4). *: p < 0.05 compared with contralateral group (paired t-test). Tukey’s test was used for statistical analysis comparing each formulation group.
The local delivery was calculated by subtracting the tissue concentration of the contralateral eye from that of the instilled eye. Data represent the mean ± S.D. (n = 4). *: p < 0.05 comparing each formulation group (Tukey’s test).
Purpose of this study was to evaluate the efficacy of instillation of 0.06% TAK-593 emulsion with xanthan gum on CNV in rats. The control and formulation groups were compared for each administration methods to account for the effect of the administration method on the CNV model. Figure 5 shows representative images of retinal flat mounts with isolectin staining in the laser induced CNV rat model. The CNV levels were 58947 ± 9700 and 60048 ± 19452 pixel in the vehicle eye drops and in the vehicle IVT injections. Instillation of 0.06% TAK-593 emulsion showed the significant preventive effect on CNV when compared to animals treated with vehicle. Moreover, the effect of 0.06% TAK-593 emulsion was similar to that of 0.2% TAK-593 suspension. Zero point zero six percent TAK-593 emulsion was expected to show the efficacy of 86% of aflibercept.
The retina of the rat’s right eye was irradiated with six lasers to surround the optic nerve (a). The bar indicates 1 mm. (b)–(f) are representative images of laser spots observed on day 14 laser photocoagulation. Vehicle solution (b), 0.2% TAK-593 suspension (c), and 0.06% TAK-593 emulsion (d) were instilled into the laser irradiated eye four times a day. Vehicle solution (e) and afliberpect (f) were injected once into the vitreous body of the laser irradiated eye.
Eye drop administration is limited due to the unique physiology and anatomy of the eye, providing low bioavailability.30) Therefore, improving bioavailability into the eye is important for compounds such as TKIs which may have serious systemic side effects. A lipid emulsion has improved intraocular penetration of lipophilic drugs compared to the suspension.24) Since, lipophilic drugs must be dissolved in tears in order to delivery into the eye, a lipid emulsion is more effective in delivery lipophilic drugs into the eye than the suspension.
Since the solubility of TAK-593 in castor oil is high compared to that in water (38 µg/mL),23) most of TAK-593 is distributed in castor oil in emulsions. In addition, stability of TAK-593 in castor oil was very high (residual percentage was 96.4% at 60 °C for 2 weeks). Castor oil is an ingredient used as a lipid phase in emulsion eye drops.24) Therefore, by selecting an emulsion as the dosage form, we found the possibility of stable eye drop for TAK-593.
Further, by increasing the viscosity of ophthalmic solution, the residence time on the ocular surface can be extended and the bioavailability into the eye can be improved. Usually, some polymers are added to increase the viscosity of the ophthalmic solution. The addition of HPMC, MC, PVA, and sodium alginate increased the viscosity of the ophthalmic solution, but caused creaming (Fig. 1). However, by adding xanthan gum, a highly viscous emulsion without creaming could be obtained. The creaming is the aggregation of emulsion particles. Polymers such as HPMC, MC, PVA, and sodium alginate adsorbed to surface of emulsion particles by slight surface activity and promoted aggregation by bridging between emulsion particles. In contrast, xanthan gum has no surface activity and doesn’t adsorb to the emulsion surface. In addition, xanthan gum solution has the property of pseudoplasticity, which has effect of reducing the mobility of emulsion particles. So, we speculated that xanthan gum inhibited aggregation of emulsion particles and didn’t cause creaming.
In small animals, the drug concentration in the posterior eye after topical instillation has a high ratio of delivery from systemic circulation. Since the ocular tissue concentration of the contralateral eye is the drug concentration delivered from the systemic circulation, the local intraocular delivery can be evaluated as the difference between the tissue concentration of the instilled eye and that of the contralateral eye.
The choroidal and retinal concentration of 0.06% TAK-593 emulsion with xanthan gum was significantly higher in the instilled eyes than in contralateral eyes (Fig. 3b). Therefore, TAK-593 emulsion is locally delivered into the posterior segment of the eye.
Viscosity enhancers and mucoadhesive agents improve the retention of eye drops on the ocular surface and improve delivery of drug into the eye. Since xanthan gum has increased formulation viscosity and mucoadhesive property,31) it is considered that the intraocular delivery of TAK-593 has been improved by adding xanthan gum to the emulsion. In addition, Mizuno et al. reported that drug delivery to the ocular posterior segment proceeded mainly through the periocular and the transposterior scleral route.32) Furthermore, Shikamura et al. used in silico simulations to prove that drugs deliver to the posterior segment of the eye by the periocular and the transposterior scleral route.33) TAK-593 was estimated to deliver through the periocular and the transposterior scleral route because 0.06% TAK-593 emulsion with xanthan gum were locally deliver into the choroid.
Sigurdsson et al. reported that local delivery ratio into the retina was about 60% in rabbits instilled with dexamethasone eye drops.34) Since TAK-593 concentration of retina of 0.06% TAK-593 emulsion was 15.43 and 5.21 ng/g in the instilled eye and the contralateral eye, the ratio of local delivery into retina was 66%, indicating that TAK-593 emulsions had the local delivery ratio equal to previously reported. On the other hand, in the 0.06% TAK-593 with xanthan gum, the ratio of local delivery into retina was 78%, which was improved compare to 0.06% TAK-593 emulsion without xanthan gum and previously reported. The local delivery of 0.06% TAK-593 with xanthan gum into retina is higher than IC50 of VEGFR2 (0.42 ng/g), suggesting that effective concentrations can be delivery into the retina in humans.
The model of laser-induced CNV relies on laser injury to perforate Bruch's membrane, resulting in sub-retinal blood vessel recruitment from the choroid. The histopathological progression of angiogenesis post-laser induction was found to mimic neovascular AMD. For these reasons, this model has been used extensively in studies of the exudative form of AMD. The laser-induced CNV model has also been used in preclinical trials of aflibercept.35) Aflibercept also showed a strong tendency to suppress CNV in this study.
The preventive effect of 0.06% TAK-593 emulsion was similar to that of 0.2% TAK-593 suspension on CNV in rats (Fig. 6). Actually, the retinal TAK-593 concentration of the instilled eye after administration of 0.2% TAK-593 ophthalmic suspension was 27 ng/g, comparable to that of the 0.06% formulation. Since TAK-593 is a tyrosine kinase inhibitor and the tyrosine kinase inhibitor may be of concern for systemic toxicity, it is very useful that the dosage of TAK-593 is reduced. In addition, 0.06% TAK-593 emulsion was expected to show the efficacy of 86% of aflibercept. Aflibercept shows the high therapeutic effect in the age-related macular degeneration patient. These results show that the emulsion with xanthan gum is an effective formulation, since it is able to deliver TAK-593 to posterior segment of the eye inhibiting laser induced CNV. Therefore, TAK-593 emulsion eye drops should have a high therapeutic effect on age-related macular degeneration patients.
The CNV area was determined using image visualization software (Image pro® plus) and normalized using the average area of the vehicle administration group in topical instillations and IVT administrations. Data represent the mean ± S.D. (Eye drops: n = 6–10, IVT: n = 3–5), *: p < 0.05 compared with vehicle group (Eye drops: one-sided Dunnet test, IVT: student’s t-test).
We have developed the high viscosity ophthalmic emulsion for the treatment of AMD. The ophthalmic emulsion with xanthan gum showed high stability against physical stimulation and no creaming occurred. In pharmacokinetics study, the choroidal and retinal concentration of 0.06% TAK-593 emulsion with xanthan gum was significantly higher in the instilled eyes than in contralateral eyes, suggesting that 0.06% TAK-593 emulsion can be locally delivered to the posterior segment of the eye by topical instillation. Since, TAK-593 emulsion eye drops show the same efficacy as intravitreal aflibercept injection in laser-induced CNV model, it can be expected to show a high therapeutic effect on AMD patients.
The authors are grateful to Senju Pharmaceutical Co., Ltd. for allowing to use the data.
Yasuhiro Mori, Akifumi Yamamoto, Ayumi Nakagawa, and Akiharu Isowaki are employees of Senju Pharmaceutical Co., Ltd. Tomohiro Hikima declares no conflict of interest.
