Development of an in vitro eye model to better represent in vivo physiological conditions

Abstract

Purpose The purpose of this thesis was to optimize a novel in vitro blink model, the OcuBlink, to best mimic the physiological characteristics of the human eye. By improving this in vitro model, the results from in vitro studies using this model can be more representative of in vivo studies. Methods The first experimental chapter of this thesis, Chapter 3, explores the results of active lysozyme deposition on contact lenses using the blink model. An ex vivo active lysozyme deposition study was referenced to determine the ideal flow rate for the blink model. These parameters were used to determine the active lysozyme deposition data for several other lens materials. The blink model was directly compared to a simple vial at 8 hours of incubation to show the difference between the two in vitro models. The experiments in Chapter 3 led to several developmental improvements to the blink model design. Chapter 4 explores the changes to the blink model from its initial design through different iterations to incorporate a heated system to simulate ocular temperatures. Chapter 5 uses the new blink model improvements to study contact lens dehydration. The effect of incubation temperature, incubation solution, and in vitro model design were explored. The comparison between the vial system and the blink model showed a difference in water content patterns over time. Results With ex vivo data as a reference, the blink model was able to replicate the active lysozyme deposition data on etafilcon A lenses. The parameters of the blink model were used to determine active lysozyme deposition on other contact lenses. This study provided the expected range for ex vivo active lysozyme deposition for these lens materials as this data was not available in the literature at the time of the study. The comparison of the in vitro models showed that contact lens material plays a large role in active lysozyme deposition patterns over time. The different iterations of the blink model show how different materials and designs can improve and progress in vitro testing of ocular studies. The most significant addition to the blink model was the incorporation of a heated element to allow studies to be conducted at ocular temperatures. The improved blink model showed a decrease in water content for all lenses for an increase in incubation temperature. The incubation solution did not have an effect on water content for most tested lenses, however, lens material played a major role. All lenses decreased in water content after the first hour of incubation. With the blink model, lenses showed a recovery in water content over 16 hours, however the lenses showed a plateau effect in a simple vial model. The recovery in water content on the blink model has not been seen on other in vitro or in vivo studies to date and will need further testing to better understand the phenomenon. Conclusion Although the data has yet to be validated with in vivo data, this thesis shows that the blink model has promise as a predictive tool for in vivo studies. The advanced in vitro blink model can be adjusted per study as required to fulfil experimental requirements. The ultimate goal of the blink model is to produce results that are more representative of in vivo data compared to more simplistic in vitro models while minimizing the cost and time of animal models and clinical trials where appropriate.