Abstract
Purpose:
The aim of this study was to evaluate and compare the release of moxifloxacin from a variety of daily disposable (DD) contact lenses (CLs) under various conditions using a novel in vitro eye model.
Methods:
Four commercially available DD conventional hydrogel (CH) CLs (nelfilcon A, omafilcon A, etafilcon A, and ocufilcon B) and three silicone hydrogel (SH) CLs (somofilcon A, narafilcon A, and delefilcon A) were evaluated. These lenses were incubated in moxifloxacin for 24 hours. The release of the drug was measured using a novel in vitro model in three experimental conditions: (1) phosphate buffered saline (PBS); (2) artificial tear solution (ATS) containing a variety of proteins and lipids; and (3) ATS with mechanical rubbing produced by the device.
Results:
Overall, CH CLs had a higher drug release than SH CLs (P < 0.05) under all conditions. Typically, a higher drug release was observed in PBS than ATS (P < 0.05). For CH, drug release was found to be higher in ATS with rubbing than PBS or ATS (P < 0.05). For most lens types, ATS with rubbing produced higher drug release than ATS alone (P < 0.05). Generally, the release kinetics for all conditions were sustained over the 24-hour testing period, and no burst release was observed (P < 0.05).
Conclusions:
Moxifloxacin release from a CL into ATS is lower when compared to release into PBS. When mechanical rubbing is introduced, the amount of drugs released is increased.
Translational Relevance:
Results suggest that sophisticated in vitro models are necessary to adequately model on-eye drug release from CL materials.
Statistical analysis, repeated measures analysis of variance (RM-ANOVA) and posthoc Tukey's multiple comparison tests, was performed using Statistica 8 software (StatSoft, Tulsa, OK). Unless otherwise stated, all data are reported as mean ± standard deviation. Statistical significance was considered significant for a P value of < 0.05.
The total moxifloxacin released after 24 hours from seven DD CLs are summarized in
Table 3 and
Figure 2. The drug release profile over 24 hours for each experimental condition are shown in
Figures 3 to
5. Generally, the drug release from CLs was sustained over the 24-hour testing period for all conditions, and no burst release was observed (
P < 0.05). Overall, conventional hydroxyethyl methacrylate (HEMA)-based hydrogel CLs had a higher drug release than SH CLs (
P < 0.05) under all conditions (
P < 0.001).
Table 3 Release (μg/lens) Moxifloxacin after 24 Hours from CH and SH DD CLs in PBS, ATS, and ATS with Mechanical Rubbing
Table 3 Release (μg/lens) Moxifloxacin after 24 Hours from CH and SH DD CLs in PBS, ATS, and ATS with Mechanical Rubbing
Typically, a higher drug release was observed in PBS than ATS (P < 0.05). For CH CLs, drug release was found to be higher in ATS with rubbing than PBS or ATS (P < 0.05). For most lens types, ATS with rubbing produced higher drug release than ATS alone (P < 0.05). Total drug release in PBS varied between 27.9 ± 4.0 and 111.3 ± 12.9 μg/lens, in ATS drug release ranged between 7.0 ± 3.2 and 96.2 ± 4.4 μg/lens, and in ATS with rubbing the drug release ranged between 18.1 ± 4.3 and 164.3 ± 15.5 μg/lens.
To determine the components in ATS that may have led to a reduction in the observed drug release, a subsequent experiment was conducted with 2 CH (etafilcon A and ocufilcon B) and 1 SH (somofilcon A) in either ATS without proteins or ATS without lipids (
P < 0.05). As shown in
Figure 6, there were no differences for drug release from somofilcon A between the three tested solutions. For etafilcon A and ocufilcon B, there were differences in total drug release between the three solutions (
P < 0.05), but there was no conclusive trend. The dynamic viscosity of the ATS was 1.09 ± 0.03 mPa•s, which was slightly higher than the viscosity of PBS 0.99 ± 0.01 mPa•s at 23°C (
P < 0.001). The measured viscosity for PBS is similar to the reported viscosity of water.
33
Overall, materials that released the highest amounts of moxifloxacin were etafilcon A (1-Day Acuvue Moist) and ocufilcon B (Biomedics 1 Day), which are both HEMA-based, Food and Drug Administration (FDA) group IV materials with a high water content and are negatively charged. Omafilcon A (Proclear 1 Day), a HEMA-based, FDA group II material with a high water content but overall neutral charge, also showed high drug release. The CLs releasing the lowest amount of drug for each testing condition were nelfilcon A (Dailies AquaComfort Plus), a polyvinyl alcohol (PVA)-based, FDA group II material with a high water content and neutral charge, as well as all SH lenses. All lenses containing moxifloxacin were visually clear throughout the entire duration of the study.
The rate of drug release into an aqueous media is determined by its solubility, and the rate at which the drug reaches equilibrium between the solution and the lens. The release of moxifloxacin, which is highly hydrophilic,
34 would therefore be dependent on the rate of tear flow. In a previous study measuring the release of moxifloxacin from the same eye model, drug release from CLs was sustained for 24 hours for etafilcon A and ocufilcon B.
22 Omafilcon A reached a plateau within 12 hours, while narafilcon A and all SHs released moxifloxacin within 4 hours.
22 In the previous study, the flow rate was set at 3.33 μL/minute (4.8 mL/day). However, in this study, the flow rate was decreased to 2.08 μL/minute (3.0 mL/day), closer to physiological tear turnover rates,
21 and the release of moxifloxacin from CLs into PBS was sustained over the 24 hours for all lens types (
P < 0.05). In a vial, the release of moxifloxacin follows more closely to a burst-plateau profile within 1 hour.
22
Considering the prominent roles of proteins and lipids in the fouling of CLs,
35,36 using PBS as an elution solvent potentially overlooks key determinants in drug release. The effect of the composition of the release medium on therapeutics release from CLs has been previously reported; the release of phospholipids from CLs were five times faster in ATS than water.
23 Surprisingly, in this study the release of moxifloxacin was significantly lower when released in ATS than PBS (
P < 0.05), which markedly contrasts previous published results.
23 The differences in the observed release could be attributed to the differences in properties between phospholipids and moxifloxacin, the composition of ATS, and the in vitro model used to measure the release. Nevertheless, the underlying mechanisms resulting in a lower drug release in ATS than a PBS solution is unclear.
Drug release from CLs into ATS is significantly more complex to describe than PBS due to the presence of various salts, proteins, mucin, and lipids within ATS.
31 Any one of these factors, or a combination of them, could interact with the drug in the lens, leading to a lower or higher drug release. Furthermore, the mechanisms of protein and lipid deposition on the CLs
37 could also alter drug–lens interactions, and consequently drug release kinetics. However, the duration of the experiment was only 24 hours, and within this time frame it is expected that tear film deposition would not have a significant impact on drug release.
To elucidate the individual effects of proteins and lipids on drug release, three lenses (etafilcon A, somofilcon A, and ocufilcon A) were tested in ATS without either proteins or lipids. For somofilcon A, there were no differences in the amount of moxifloxacin release in ATS, ATS without proteins, or ATS without lipids (P > 0.05). For etafilcon A, ATS without proteins released less drug than ATS (P < 0.05), while for ocufilcon A, ATS without lipids released less drug than ATS (P < 0.05). These results suggest that each lens material interacts differently with ATS, resulting in slightly different drug release kinetics. Currently, there is no obvious trend to explain for the effects of protein or lipids on drug release from CLs.
Another explanation for a lower drug release from CLs is the differences in viscosity between PBS and ATS. The dissolution of a drug into the media is dependent on the media viscosity, where higher viscosity leads to slower drug dissolution.
38 However, the viscosity of ATS is only slightly higher than that of PBS. So while the higher viscosity of ATS may contribute to the lower drug release, it is likely that it is not the primary factor leading to the observed trend. The viscosity of ATS in this study is lower than the viscosity of human tears (2.33 mPa•s).
33 This is likely due to the fact that our ATS does not contain many of the lipids that contribute to the viscosity of natural tears.
33
Another important ocular parameter that can affect drug release from CLs is blinking. In this study, only the mechanical wear component of the blink was simulated. The CL sits tightly between the corneal eyepiece and eyelid piece. The rotation of the corneal eyepiece circularly causes the CL to rub against the eyelid piece. The rubbing occurred at a rate of 10 cycles every minute, which falls within the average range for blink rates, of 4.5 to 26 blinks/minute.
32 For most lens types, mechanical rubbing in ATS resulted in significantly higher drug release than ATS alone (
P < 0.05). This observation also has been reported in another study modelling the effects of blinking on drug release from hydrogels.
24 Furthermore, the tear film is shear-thinning (non-Newtonian) in which the viscosity is not constant. We speculate that the rubbing process between the eyelid and corneal eyepiece decreases the viscosity of the ATS,
33,39 which increases the amount of drugs released from the CL.
In general, CH lenses released more drugs than SH lenses, which is similar to previously reported results.
10,16,22 Etafilcon A (1-Day Acuvue Moist) and ocufilcon B (Biomedics 1 Day), two CH materials with a high water content that are negatively charged, released the highest amount of moxifloxacin. Nelfilcon A (Dailies AquaComfort Plus) was the only CH that behaved like a SH in regards to drug release. The lowered drug release for this lens type is unclear, but could be linked to the composition of the material, which is composed of PVA instead of the traditional HEMA-based materials found in other CH. Furthermore, the lens also contains wetting agents such as PVA and PEG, which are not cross-linked to the polymer. These agents occupy space within the matrix, which could interfere with the lens ability to absorb and release drugs. The two lenses with the lowest drug release for all tested conditions were two SH, delefilcon A (Dailies Total 1) and narafilcon A (1-Day Acuvue TruEye).
An important concern in CL drug delivery is whether the amount of drugs released by the CL is comparable to eye drops for treatment. In general, the amount of drugs that can be loaded and released from a CL is significantly lower in comparison to the amount in eye drops administered over the treatment duration. That being said, the idea behind drug delivery with CL is to achieve similar or better efficacy than eye drops at lower drug dosing. Results from several in vivo studies have validated the efficacy a CL drug delivery platform, despite releasing considerably less drugs.
15,40–45 For in vitro studies, the minimum inhibitory concentrations for 90% of bacterial isolates (MIC
90) can be used as relative measure of efficacy for release of antibiotics. All of the CLs, in all testing conditions, released enough moxifloxacin to meet the MIC
90 (0.047 μg/mL)
46 against common and susceptible pathogens over the 24-hour study period.
Based on the in vitro model used in this study, the release of moxifloxacin can be sustained for up to 24 hours. The drug release profiles are dependent on the properties of the CL. The amount of drug that is released is dependent on the composition of the elution solution, with lower amounts of drug being released in ATS than PBS. When mechanical rubbing is introduced to those lenses exposed to ATS, the amount of drug released from the CLs increased compared to ATS alone. In conclusion, certain parameters such as ATS can decrease the amount of drugs released, while other parameters, such as mechanical rubbing, increases the amount of drugs released. These results underline the importance of considering various ocular parameters when determining drug release from CLs. Another key parameter affecting drug release from CLs, the properties of the drug such as molecular size, chemical structure, polarity, and solubility,
47,48 was not investigated in this study. This factor, and the complex interactions between material, drug properties, and elution system on CL drug delivery will be investigated in future studies.