Comparison of the In Vitro Tolerance and In Vivo Efficacy of Traditional Timolol Maleate Eye Drops versus New Formulations with Bioadhesive Polymers

Abstract

Purpose.: To assess the in vitro tolerance and in vivo efficacy of new unpreserved formulations of timolol maleate (TM) in aqueous solutions of bioadhesive polymers used for dry eye treatment and to compare them with three traditional TM formulations: unpreserved Timabak (Thea, Madrid, Spain), benzalkonium chloride (BAK)-preserved Timoftol (Frosst Laboratories, Madrid, Spain), and BAK-preserved Timolol Sandoz (Frosst Laboratories).

Methods.: New formulations were composed of TM (0.5%) and carboxymethyl cellulose (0.5%), hyaluronic acid (0.2%), or hydroxypropylmethyl cellulose (0.3% or 0.5%). In vitro tolerance was determined in human corneal-limbal epithelial cells and normal human conjunctival cells. The ocular hypotensive effect was evaluated measuring IOP in rabbit eyes for 8 hours.

Results.: In all cases, cell survival after exposure to the formulations was greater in the new unpreserved TM formulations than in the traditional TM solutions (BAK-preserved and unpreserved). In addition, the new formulations were demonstrated to maintain the hypotensive effect of TM in different magnitudes. The maximum hypotensive effect was reached by TM 0.5% in carboxymethyl cellulose 0.5% (32.37%).

Conclusions.: The results demonstrated that new unpreserved formulations of TM with bioadhesive polymers decreased IOP in rabbits and reached values closer to those reached by traditional solutions. Furthermore, new formulations presented a significantly higher in vitro tolerance than the same compound in traditional formulations. Although unpreserved formulations are usually more expensive, preservative-free antiglaucoma eye drops should improve compliance and adherence in the medical treatment of glaucoma. Bioadhesive polymers could be part of antiglaucoma formulations to reduce ocular toxicity, improve drug efficacy, and protect the ocular surface in long-term therapies.

Glaucoma is a chronic and multifactorial optic neuropathy characterized by progressive optic nerve damage, which can eventually lead to visual field loss and, if left untreated, blindness. ¹ Glaucoma is the second-leading cause of blindness in the world, and it is expected that around 79.6 million people will suffer from it in 2020. ²  

Most forms of glaucoma are characterized by high intraocular pressure (IOP). Some risk factors for the development of glaucoma include advanced age, family history, or race; however, elevated IOP is considered the greatest risk. All the currently approved topical treatments are indicated for decreasing eye pressure, thereby preventing further damage to the optic nerve. 

The ocular barriers and the continuous turnover of tears limit drug absorption, ^3,4 so even though eye drops are easily applied treatments, they must be instilled frequently or at high concentrations to achieve therapeutic levels in the tissues. Most patients undergo the instillation of eye drops one to two times per day, and it is not unusual to find therapies that combine two or more drugs; moreover, at present, therapies must be continued throughout the lifetime of the patient. The combination of all these situations can induce lack of patient compliance. ^{5 –7}  

Prolonged antiglaucoma treatments are usually associated with allergic reactions and adverse effects. ^8,9 The drug and other components of the formulations (mainly preservatives) can be damaging to the ocular surface in long-term therapies ^{10 –13} and thus are often responsible for the secondary effects produced by the formulations. A good example is dry eye syndrome, which frequently develops in patients with prolonged use of antiglaucoma medications. ^{14 –16}  

Most eye drop formulations are used in multidose bottles that require the inclusion of an antimicrobial preservative in the solutions. A direct correlation has been reported between the presence of preservatives in the formulations and the symptoms experienced during antiglaucoma treatments. ^{17 –21} Even though the concentration of preservatives is generally low, their repeated application can cause serious effects. ^{22 –25} The cationic detergent benzalkonium chloride (BAK) is the most commonly used preservative in topical ophthalmic formulations (0.025%–0.004% ²⁶ ) and is especially toxic to the ocular surface cells. ^27,28  

Many studies are conducted to solve complications related to the long-term use of preserved antiglaucoma medications. Some research groups try to find new active substances and combinations with high ocular hypotensive efficacy and few adverse effects, ^{29 –34} and other studies show how mixtures of already-commercialized hypotensive agents help avoid multiple applications. ^{35 –38} In addition, several strategies have been developed to increase the bioavailability of topical antiglaucoma drugs by prolonging the contact time of the drug on the ocular surface. ^{39 –42} Considering the available options, an optimal formulation for long-term treatments should combine good properties for the eye (without preservatives, if possible) with a high therapeutic efficacy of the active compound. 

The introduction of unpreserved formulations to be used in the treatment of ocular surface impairments is an important contribution to the formulation of ophthalmic solutions. Nevertheless, the use of unpreserved formulations has two drawbacks. First, due to the disruption of the hydrophobic barrier of the corneal epithelium that some preservatives produce, it is believed that there is an enhancement of the active compounds' penetration and, subsequently, an improvement of their efficacy. Although this fact has not been completely proved, ²⁶ it is one of the reasons that the use of preservatives has been maintained in eye drop formulations. Second, to protect the solutions from microbial contamination, unpreserved eye drops have to be formulated in special bottles (unit-dose or ABAK system) that are usually more expensive than multidose bottles. 

Because of the ability of bioadhesive polymers to enhance the viscosity of the ophthalmic vehicles, they are able to reduce the drainage rate of the drugs and subsequently to increase their therapeutic efficacy. ^{43 –45} Furthermore, some of the polymers are protective of the eye, and so their use is more than appropriate for chronic treatments, such as the therapy for dry eye syndrome. Most bioadhesives studied for drug delivery adhere to epithelial tissue and possibly to the mucosal surface of these tissues. ⁴⁶ One currently accepted mechanism to explain the attachment of some polymers to mucin is a physical entanglement of the polymer chains with mucin when the polymer undergoes swelling in water. ⁴⁷ However, there are also other nonbiological mechanisms that try to explain the process. ⁴⁰  

Common bioadhesive polymers are cellulose polymers and hyaluronic acid (HA); both have been demonstrated to extend the ocular residence time of ophthalmic formulations, and they are also widely used in artificial tears. ^45,48,49 Carboxymethyl cellulose (CMC) and hydroxypropylmethyl cellulose (HPMC) have film-forming properties, and HPMC, specifically, is able to interact with the tear film, increasing its stability. ⁵⁰ HA is also involved in wound healing ⁵¹ and has water-retention properties, providing a long-lasting hydration to the ocular surface. ⁵²  

The purpose of this work was to assess the in vitro tolerance and in vivo efficacy of novel, unpreserved formulations of timolol maleate (TM, 0.5%) in aqueous solutions of bioadhesive polymers, and to compare them with three traditional TM formulations: unpreserved Timabak, BAK-preserved Timoftol, and BAK-preserved Timolol Sandoz. We decided to employ polymers and polymer concentrations that are usually part of existing commercial artificial tear products. ⁵³ The polymers used as vehicles were aqueous solutions of CMC (0.5%), HA (0.2%), and HPMC (0.3% and 0.5%). In this study, we assessed the in vitro tolerance of the formulations in human corneal-limbal epithelial cells and normal human conjunctival cells. Moreover, we used in vivo experiments to determine the hypotensive effect of the formulations after topical application in rabbit eyes. 

Materials and Methods

Compounds

CMC (400–800 cps, 2% solution at 20°C), HA (Mw 400,000–800,000) and HPMC (1390 cps, 2% solution at 20°C) were purchased from Abaran Materias Primas (Madrid, Spain). Timabak (TB) was purchased from Thea (Madrid, Spain); Timoftol (TF) and Timolol Sandoz (TL) were supplied by Frosst (Madrid, Spain). TM powder was provided by Sigma-Aldrich (Madrid, Spain). Phosphate-buffered saline (PBS), rendered isotonic with NaCl (pH 7.2), was prepared with ultrapure water (milliQ; Millipore Iberica SA, Madrid, Spain). 

Preparation of TM Solutions

Four formulations containing TM 0.5% and a mucoadhesive polymer (CMC 0.5%, HA 0.2%, HPMC 0.3%, or HPMC 0.5%) were prepared. The polymers were dissolved in isotonic PBS (pH 7.2) and, in all cases, were sterilized first by filtration (0.20-μm filter pore) and then by autoclaving. Stock solutions containing TM were used for the preparation of the formulations. TM was dissolved in ultrapure water and then diluted with isotonic PBS (pH 7.2) to get a final TM concentration of 10 mg/mL. Solutions were filtrated by a 0.20-μm filter pore. To prepare the final formulations, TM stock solutions were diluted 1:2 with the corresponding polymer. The dose of TM was quantified by the spectroscopic method described below. The compositions of the new TM solutions prepared and the traditional TM formulations used for the study are shown in Table 1. 

Table 1.

 

Composition of the New TM Solutions and the Traditional TM Formulations Used for the Study

Formulation	Composition	Preservative
(1) TM/CMC	TM 0.5%, CMC 0.5%, sodium dihydrogen phosphate dihydrate, disodium hydrogen phosphate dodecahydrate, sodium chloride, double-distilled water.	Unpreserved
(2) TM/HA	TM 0.5%, HA 0.2%, sodium dihydrogen phosphate dihydrate, disodium hydrogen phosphate dodecahydrate, sodium chloride, double-distilled water.	Unpreserved
(3) TM/HPMC03	TM 0.5%, HPMC 0.3%, sodium dihydrogen phosphate dihydrate, disodium hydrogen phosphate dodecahydrate, sodium chloride, double-distilled water.	Unpreserved
(4) TM/HPMC05	TM 0.5%, HPMC 0.5%, sodium dihydrogen phosphate dihydrate, disodium hydrogen phosphate dodecahydrate, sodium chloride, double-distilled water.	Unpreserved
(5) TB	TM 0.5%, sodium dihydrogen phosphate dihydrate, disodium hydrogen phosphate dodecahydrate, water for injection.	Unpreserved
(6) TF	TM 0.5%, monobasic and dibasic sodium phosphate, sodium hydroxide, water for injection.	BAK 0.01%
(7) TL	TM 0.5%, disodium phosphate dodecahydrate, sodium phosphate dihydrate, water for injection.	BAK*

*  Concentration not specified. All the information is available in the package leaflet.

Spectroscopic Quantitation of TM

Quantitation analyses were performed on a spectrophotometer (λ = 294 nm; 1700 UV-Visible; Shimadzu, Kyoto, Japan). Accompanying software (UV Probe 2.0; Shimadzu) was used for all the absorbance signals and treatment of data. TM was quantified by a method described earlier by Nevin Erk. ⁵⁴  

Characterization of Formulations

pH and Osmolarity.

The pH of the formulations was measured with a pH meter (GLP 21; Crison; Alella, Barcelona, Spain) equipped with an Ag/AgCl combined glass electrode (Crison 52-02; Alella). The pH measurements were performed in triplicate at 25°C (room temperature). 

Osmotic activities were analyzed by vapor pressure measurements with an osmometer (model K-7000; Knauer, Berlin, Germany). Before the analyses the osmometer was calibrated with 400 mOsM NaCl. The measurements were made in triplicate at 33°C (corneal surface temperature). ^55,56  

Rheology.

Rheological properties were assessed with a rheometer (Haake RheoStress RS-1; Thermo Scientific, Düsseldorf, Germany) using a plate–plate measuring geometry (diameter of 60 mm, gap 0.5 mm). The flow curves, which represent viscosity as a function of the shear rate, were used to study the rheologic behavior of the formulations. A variable shear rate 50 to 1000 seconds⁻¹ was applied. The viscosity of the samples was calculated at a shear rate of 100 seconds⁻¹, which was within the linear viscoelastic range for all the formulations. Each formulation was measured at 33°C in triplicate. 

Viability Assays

The viability assays were performed with two different cell lines: immortalized human corneal-limbal epithelial (HCLE) cells (Schepens Eye Research Institute, Harvard Medical School, MA) and normal human conjunctival cells (IOBA-NHC; Instituto de Oftalmobiología Aplicada, Valladolid University, Valladolid, Spain). 

Cytotoxicity studies were assessed by the mitochondrial-dependent reduction of the tetrazolium salt, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma-Aldrich) to formazan (MTT method). ^57,58  

HCLE cell cultures were plated in 15-mL culture flasks in a medium nutritionally optimized for proliferation of keratinocytes ⁵⁹ : keratinocyte serum free medium (SFM) supplemented with 0.5 mL CaCl₂ 0.3M, 1.25 mL bovine pituitary extract, and 40 μL epidermal growth factor (EGF). The cultures were grown at 37°C in a 5% carbon dioxide atmosphere. Trypsin was inactivated by neutralizing the mixture with the same volume of Dulbecco's modified Eagle's medium:nutrient mixture F-12; (DMEM/F12), which contained 10% calf serum and 2% penicillin/streptomycin. All reagents were purchased from Invitrogen-Difco (Barcelona, Spain). The mixture was centrifuged (3000 rpm for 4 minutes at room temperature), and the cells were resuspended in SFM. 

IOBA-NHC cells were plated in 15-mL culture flasks in DMEM/F-12 medium supplemented with 10% calf serum, 2% penicillin-streptomycin, 2.5 μg/mL amphotericin B, 1 μg/mL bovine pancreas insulin, 0.5 μg/mL hydrocortisone, 0.1 μg/mL cholera toxin, and 0.2 ng/mL EGF. The cultures were grown at 37°C in a 5% carbon dioxide atmosphere. The mixture was neutralized with the same volume of DMEM/F12, which contained 10% calf serum and 2% penicillin/streptomycin. Then, it was centrifuged (1000 rpm for 5 minutes at room temperature) and finally, the cells were resuspended in the culture medium. All reagents were purchased from Invitrogen-Difco (Barcelona, Spain). 

For the cytotoxicity studies, the cells were seeded into 96-well culture plates (50,000 cells/well). After adhering to the plates (37°C for 24 hours in an atmosphere of 5% CO₂/95% air), the medium was removed and the testing formulation was added. 

The cells were exposed to seven formulations: (1) TM 0.5% in CMC 0.5% (TM/CMC), (2) TM 0.5% in HA 0.2% (TM/HA), (3) TM 0.5% in HPMC 0.3% (TM/HPMC03), (4) TM 0.5% in HPMC 0.5% (TM/HPMC05), (5) TB, (6) TF, and (7) TL. 

Taking into account that a conventional topical ophthalmic formulation is eliminated from the ocular surface in only 5 minutes, short contact times (15 minutes) are enough to determine tolerance. Nevertheless, to simulate long-term therapies and taking into consideration that polymers increase the contact time of the formulations on the ocular surface, we also used contact times of 1 and 4 hours in the assay. After that, the medium was carefully removed and the MTT solution (5 mg/mL in PBS) was added to the plates that were then incubated for 3 hours at 37°C. After careful aspiration of the medium, cells were solubilized with 100 μL/well of dimethyl sulfoxide (DMSO; Sigma-Aldrich). The extent of the reduction of MTT to formazan within cells was quantified by the measurement of optical density at 550 nm, using a microplate reader (model 6010152EU; DigiScan, Eugendorf, Austria). 

Viability was set as 100% in untreated cells. The vehicle used as the positive control was BAK 0.005% (Sigma-Aldrich). Cytotoxicity data were obtained from three different experiments by testing seven wells per sample. 

Animals

Male New Zealand White rabbits weighing 3 to 4 kg were used. The animals were kept in individual cages with free access to food and water. They were maintained under controlled 12 hour/12 hour light-dark cycles. All the protocols herein complied with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were also in accordance with the European Communities Council Directive (86/609/EEC). 

Intraocular Pressure Measurements

Intraocular pressure (IOP) was measured by means of a rebound tonometer (Tonovet; Medicalmix, Barcelona, Spain). A portion (25 μL) of each of the formulations was applied to the New Zealand rabbit eyes (n = 10). As a control for noncommercial TM formulations, we used the corresponding vehicle without TM. PBS was used as a control for TB, TF, and TL. 

To study the time course of the effect of the formulations, we obtained two IOP measurements before any compound was administered (30 minutes and just before the instillation). Then, IOP determinations were performed once every hour over a period of 8 hours. 

Statistical Analysis

Data are expressed as the mean ± SD (n > 3, indicated in each case). Statistical differences between two mean values were evaluated by two-tailed student's t-test. If necessary an analysis of variance (ANOVA) was used. Results were taken as significantly different at P < 0.05. The plotting and fitting of dose–response curves was performed with the computer program (OriginPro 8 SR2; Originlab, Northampton, MA). 

Results

Quantitation of TM in New Formulations

The spectroscopic method of quantifying TM was validated with respect to linearity, accuracy, and reliability in the range of concentrations between 0.5 and 50 μg/mL. Then, the amount of the active substance in TM/CMC, TM/HA, TM/HPMC03, and TM/HPMC05 was quantified. The results are summarized in Table 2. In all cases, nonsignificant differences (P > 0.05) were observed between the theoretical TM concentration and the values obtained after the analyses of the formulations. 

Table 2.

 

Quantitation of TM by Spectrophotometry

Formulation	TM
(1) TM/CMC	99.94 ± 2.04
(2) TM/HA	100.70 ± 1.23
(3) TM/HPMC03	101.58 ± 0.82
(4) TM/HPMC05	101.70 ± 0.91

 λ = 294 nm. Data are represented as total amount of TM in percentage ± SD.

 Initial solutions of 10 mg/ml of TM in ultrapure water were always employed to prepare the formulations.

Characterization of TM Formulations

pH and Osmolarity.

The parameters were measured for TM/CMC, TM/HA, TM/HPMC03, and TM/HPMC05 and for the traditional TM formulations (TB, TF, and TL); the corresponding data are shown in Table 3. All the formulations that we tested showed values nearly neutral (P > 0.1 in all cases). In osmolarity, all of them were within the range of isotonicity. 

Table 3.

 

pH, Osmolarity, and Viscosity Data

Formulation*	pH	Osmolarity (mOsm/Kg)	Viscosity (cps)
(1) TM/CMC	6.7 ± 0.03	314.6 ± 2.46	6.00 ± 0.60
(2) TM/HA	6.6 ± 0.02	301.9 ± 1.53	1.62 ± 0.12
(3) TM/HPMC03	6.5 ± 0.01	310.6 ± 0.66	2.72 ± 0.41
(4) TM/HPMC05	6.7 ± 0.02	316.3 ± 1.19	14.48 ± 0.99
(5) TB	6.9 ± 0.01	294.3 ± 1.5	0.77 ± 0.06
(6) TF	6.9 ± 0.02	309.6 ± 2.6	0.82 ± 0.06
(7) TL	6.7 ± 0.01	318.7 ± 2.1	0.97 ± 0.02

 Data were calculated from n = 3 and are expressed as the mean ± SD.

*  (1) TM 0.5% in CMC 0.5%, (2) TM 0.5% in HA 0.2%, (3) TM 0.5% in HPMC 0.3%, (4) TM 0.5% in HPMC 0.5%, (5) Timabak, (6) Timoftol, and (7) Timolol Sandoz.

Viscosity.

A variable shear rate of 50 to 1000 seconds⁻¹ was applied to study the rheologic behavior of the formulations. For TB, TF, and TL, the viscosity was independent of the shear rate, showing Newtonian behavior. For TM/CMC, TM/HA, TM/HPMC03, and TM/HPMC05, the viscosity decreased as the shear rate increased, that is, these systems showed a shear thinning or a pseudoplastic behavior. The viscosity of the samples was measured at 100 seconds⁻¹ (Table 3). TM/HPMC05 showed the highest viscosity (14.48 cps) and TB the lowest (0.77 cps). 

Viability Assays.

According to this protocol, HCLE and IOBA-NHC cells were exposed to the seven formulations: TM/CMC, TM/HA, TM/HPMC03, TM/HPMC05, TB, TF, and TL. The results are shown in Figures 1 and 2. 

Figure 1.

 

HCLE cells viability in percentage (± coefficient of variation) for (1) TM/CMC, (2) TM/HA, (3) TM/HPMC03, (4) TM/HPMC05, (5) TB, (6) TF, and (7) TL. BAK 0.005% (B) was used as a positive control. *P < 0.05–0.01, **P < 0.01–0.001, ***P < 0.001.

Figure 2.

 

IOBA-NHC cells viability in percentage (± coefficient of variation) for (1) TM/CMC, (2) TM/HA, (3) TM/HPMC03, (4) TM/HPMC05, (5) TB, (6) TF, and (7) TL. BAK 0.005% (B) was used as positive control. *P < 0.05–0.01, ***P < 0.001.

The formulations TM/CMC, TM/HA, TM/HPMC03, and TM/HPMC05 were observed to be the least cytotoxic in all cases, giving the highest HCLE and IOBA-NHC cell viability values among all the formulations, after short- and long-term exposures. The lowest viability values were always obtained with TB (unpreserved), TF (BAK-preserved), and TL (BAK-preserved). However, TB proved to be the least cytotoxic formulation among the commercialized TM formulations in all cases. 

Effect of TM Formulations on IOP in Rabbits

All the formulations were able to maintain a hypotensive effect, providing different maximum effects, as is shown in Figures 3 and 4. The maximum percentage of IOP reduction, the time of maximum effect (t _max), and the area under the ΔIOP (%) time curve from 0 to 8 hours (AUC) were calculated for all the formulations (Table 4). In addition, several other IOP parameters were considered: mean IOP, mean of the minimum and maximum IOP, and IOP fluctuation (Table 5). 

Figure 3.

 

Ocular hypotensive effect of (1) TM/CMC, (2) TM/HA, (3) TM/HPMC03, (4) TM/HPMC05, (5) TB, (6) TF, and (7) TL. Vehicles CMC 0.5%, HA 0.2%, HPMC 0.3%, and HPMC 0.5% were used as the control for formulations 1, 2, 3, and 4, respectively. Isotonic PBS (pH 7.2) was the control for formulations 5, 6, and 7. Data are expressed as the mean ± SEM (n = 10).

Figure 4.

 

Maximal hypotensive effect (% ± SEM) of (1) TM 0.5% in CMC 0.5%, (2) TM 0.5% in HA 0.2%, (3) TM 0.5% in HPMC 0.3%, (4) TM 0.5% in HPMC 0.5%, (5) Timabak, (6) Timoftol, and (7) Timolol Sandoz.

Table 4.

 

Maximum IOP Reduction over 8 Hours

Formulation*	Maximum IOP Reduction	t max	AUC
(1) TM/CMC	32.37 ± 2.05	2.5	135.17 ± 11.98
(2) TM/HA	25.18 ± 2.68	1.0	76.12 ± 6.19
(3) TM/HPMC03	27.59 ± 2.65	1.0	134.76 ± 14.29
(4) TM/HPMC05	28.92 ± 2.85	2.0	117.27 ± 19.29
(5) TB	26.35 ± 2.35	1.5	138.70 ± 11.23
(6) TF	21.99 ± 2.36	2.0	101.31 ± 12.93
(7) TL	29.20 ± 4.22	2.5	99.81 ± 11.46

 Maximum IOP reduction is expressed as the percentage of reduction ±SEM; t _max is the median time of maximum effect; and AUC is the ΔIOP (%) time curve from 0 to 8 hours ± SEM.

*  Formulations are as described in Table 3.

Table 5.

 

IOP Parameters

Formulation*	Mean IOP	Mean of Minimum IOP	Mean of Maximum IOP	IOP Fluctuation
(1) TM/CMC	10.59 ± 1.53	8.30 ± 0.97	12.70 ± 0.97	4.10 ± 0.88
(2) TM/HA	11.06 ± 1.41	8.70 ± 1.25	12.20 ± 0.42	3.50 ± 1.35
(3) TM/HPMC03	10.74 ± 1.54	9.00 ± 1.15	12.70 ± 0.67	3.70 ± 1.06
(4) TM/HPMC05	10.76 ± 1.72	8.60 ± 1.35	12.60 ± 1.07	4.00 ± 1.15
(5) TB	10.48 ± 1.53	8.90 ± 0.88	12.40 ± 1.35	3.50 ± 1.08
(6) TF	10.74 ± 1.32	9.30 ± 1.06	12.30 ± 1.06	3.00 ± 0.82
(7) TL	10.64 ± 1.56	8.20 ± 1.32	12.00 ± 0.67	3.80 ± 1.75

 Data are expressed as the mean ± SEM.

*  Formulations are as described in Table 3.

The maximum hypotensive effect was reached by TM/CMC (32.37%). The rest of the formulations achieved significantly lower values (P < 0.05 in all cases). In general, the time of maximum effect was longer for the traditional TM formulations (from 1.5 to 2.5 hours). Among the new TM formulations prepared, the longest t _max observed belonged to TM in CMC 0.5% (2.5 hours). Regarding AUC, TB reached the highest value (138.70). However, there were no significant differences among TB and the TM/CMC and TM/HPMC formulations (P = 0.37 and P = 0.40, respectively). No significant differences were found among the highest IOP fluctuation values, which were reached by TL, TM/CMC, and TM/HPMC05 (5, 4.1, and 4 mm Hg, respectively; P = 0.19). 

Discussion

One of the consequences of chronic antiglaucoma therapy is ocular surface modification caused by the frequent exposure of the eye to drugs and excipients. ¹⁰ All active agents, in varying degrees, can produce several types of ocular side effects and induce systemic effects once the drug is absorbed into the systemic circulation. ^{60 –63} Furthermore, these effects are worsened when formulations are preserved. ^{18 –21}  

Preservatives can compromise the ability of the tear film to provide protection and trophic factors to the cornea, ⁶⁴ damage the corneal epithelial cells, and inhibit the growth of the trabecular meshwork. ¹³ These reactions, combined with the fact that most patients must instill eye drops one to two times per day, can induce a lack of patient compliance. ^{6, 7}  

Most complications can be relieved by removing preservatives from the preparations or by increasing the drugs' bioavailability, and so it is possible to reduce the number of applications of the formulation to the eye. The toxicity of preservatives on the ocular surface is well documented, so patients who require the application of eye drops in a long-term therapy should use only unpreserved formulations. Nevertheless, preservative-free formulations are available in single-use vials or in special devices (such as the ABAK system) that, although still affordable, are more expensive. Another option is related to a reduction of the number of applications of the formulations by increasing the drug's bioavailability. The use of bioadhesive polymers is a good alternative that enhances drug corneal penetration. Bioadhesive polymers can stay in the precorneal area for an extended period, thus reducing the drainage rate of the drug from the eye and consequently improving its therapeutic efficacy. ⁴⁶  

In this work, we prepared novel unpreserved formulations of TM 0.5% with aqueous solutions of CMC 0.5% (TM/CMC), HA 0.2% (TM/HA), and HPMC 0.3% and 0.5% (TM/HPMC03 and TM/HPMC05, respectively). We decided to employ polymers and polymer concentrations that are usually part of existing commercial artificial tear products. ⁵³ The formulations developed satisfied the given requirements of pH and osmolarity for the ophthalmic solutions. ^65,66  

Cytotoxicity studies in HCLE and IOBA-NHC cells were performed to determine the in vitro tolerance of the formulations prepared, as well as the tolerance of three traditional eye drops used for glaucoma treatment: TB (unpreserved), TF (BAK-preserved), and TL (BAK-preserved). The results showed that, in both cell lines and after all exposure times, new unpreserved TM formulations developed with bioadhesive polymers were less cytotoxic than traditional solutions. Although it was expected that the formulations that we developed (without BAK) would be better than the traditional BAK-preserved formulations, we additionally found that our formulations were better than unpreserved TB at any exposure time in both corneal and conjunctival cell lines (Figs. 1, 2). Although further studies are needed, these results may suggest that the polymers CMC, HA, and HPMC are able to protect the cells from TM's toxicity in these experimental conditions. In addition, a decrease in cell viability values was shown in commercial formulations at longer exposure times (unpreserved and BAK-preserved), possibly due to the toxic effect of TM on the cell lines. TF and TL showed higher cytotoxicity than the positive control at short contact times. These results were in accordance with BAK content, since TF contains 0.01% BAK and the positive control 0.005% BAK (BAK concentration in TL is not available). 

The in vivo efficacy of all the formulations was determined in rabbit eyes. The maximum hypotensive effect was reached by TM/CMC (32.37%). Concerning t _max, traditional solutions generally showed longer values than the novel TM eye drops (from 1.5 to 2.5 hours). Nevertheless, TM/CMC t _max proved to be the closest to that of the traditional solutions (2.5 hours). Concerning IOP fluctuation, the greatest values were attained in rabbits treated with TL (5 mm Hg), TM/CMC (4.1 mm Hg), and TM/HPMC05 (4 mm Hg). 

Previous studies demonstrated that the most significant relative improvement in ocular bioavailability has been observed for vehicles in the viscosity range 1 to 15 cP. ^43,67 Our best results were obtained with viscosity values of 6 and 14.50 cP (CMC 0.5% and HPMC 0.5%, respectively). 

The use of polymers to increase humidity and to improve lubrication and protection of the ocular surface has limitations. The polymers that we used to develop our formulations are currently used in ophthalmology, and secondary effects related to blurry vision have been reported. ^68,69 Future studies should be conducted to determine the polymer concentration that provides the best ratio of secondary effects to drug efficacy. 

Increasing the viscosity of ophthalmic formulations not only enhances the ocular absorption of timolol but reduces its systemic absorption. ³⁰ Several studies have attempted to demonstrate the efficacy of other possible methods to reduce the systemic absorption of timolol. For instance, the use of prodrugs demonstrated to be effective in that they improved the corneal penetration, but only when a small instilled dose volume was used. ²⁹ Another approach, the instillation of timolol in a viscous vehicle with a low vasoconstrictor concentration, has also been studied. ^30,70 Nevertheless no data regarding long-term efficacy of vasoconstriction and clinical efficacy were available. We developed formulations with timolol and bioadhesive polymers able to decrease IOP in rabbit eyes and reach IOP values closer to traditional solutions; however, the ocular and systemic-absorption of these formulations needs to be studied further. 

We have developed four unpreserved formulations of a traditional hypotensive agent (TM) and bioadhesive polymers widely used in artificial tear solutions (CMC, HA, and HPMC). TM formulated with bioadhesive polymers presented a significantly higher tolerance (in vitro toxicity studies in culture cells) than the same compound in traditional formulations (unpreserved TB, BAK-preserved TF, and BAK-preserved TL). Furthermore, the new formulations of TM were able to decrease IOP in rabbit eyes and reach IOP values closer to traditional solutions; among the new formulations, TM/CMC was demonstrated to be the most effective at reducing IOP (32.37%). Despite the drawbacks regarding a higher, though still affordable, cost of the formulations, preservative-free antiglaucoma eye drops should improve compliance and adherence in the medical treatment of glaucoma, especially for patients with corneal or conjunctival damage. Our results show that bioadhesive polymers could be proposed to be part of glaucoma formulations to reduce ocular toxicity, improve drug efficacy, and protect the ocular surface, especially in patients with dry eye symptoms. 

Footnotes

 Supported by the Spanish Ministry of Health (RETICSnet RD07/0062/2002, RD07/0062/0000), Spanish Health Research Fund (FIS PI07/0043 and PI07/0012; MAT 2007–65288), and Complutense University Research Group UCM-920415.

Footnotes

 Disclosure: V. Andrés-Guerrero, None; M. Vicario-de-la-Torre, None; I.T. Molina-Martínez, None; J.M. Benítez-del-Castillo, None; J. García-Feijoo, None; R. Herrero-Vanrell, None

The authors thank Beatriz de las Heras and Natalia Girón for performing the MTT assays.