Purpose: The treatment of ocular surface disease is an important therapeutic area of interest. Although biomolecules are now the prevalent new-therapeutic agents, delivery of biomolecules to treat ocular surface disease is rare. This is due to the negative impact of the rapid speed of ocular surface clearance and the instability of proteins in aqueous ophthalmic formulations. A stable formulation for the delivery of proteins to the ocular surface would expand ophthalmic treatment options. In this study, insulin was chosen as a model protein. The delivery profile of an insulin solution was compared to that of insulin in a partially hydrated PROLOC® gel.
The availability of insulin to the corneal surface was studied using a dynamic diffusion model (DiffEr). The model uses specially adapted spherical diffusion Franz Cells maintained at 34°C. This model simulates ocular surface clearance by reflux and normal tear turnover. The membranes used were fresh, mature rabbit corneas or dialysis membrane with a cut-off of 6-8 kDa. The dose was 60 µL of either 0.25% Insulin in PBS or 0.22% Insulin in PROLOC® gel. In this model, the membrane surface is flushed immediately after dosage at a flow rate that simulates reflex tear flow. The flow rate was then reduced to a basal rate for the duration of the experiment. To evaluate drug retention and release profile, fluid removed from the pre-corneal layer was collected and analyzed by HPLC. At the end of the 5-hour study, the rinsate from the donor chamber was collected and tested by HPLC for insulin.
This study compared the drug retention and release profile of an insulin solution to insulin in a partially hydrated PROLOC® gel. The study showed that a large amount of insulin was delivered and quickly cleared from the ocular surface when the insulin solution formulation was dosed. For the solution formulation, no significant concentrations were observed after 60 minutes. The insulin in PROLOC® gel formulation delivered insulin at a slower, more consistent rate throughout the 5-hour study. Furthermore, when the cells were rinsed with PBS at the end of the study, insulin was detected in the cells containing the PROLOC® formulation, but not in the cells dosed with solution formulation.
The total mass of insulin collected in the clearance or flushing fluid changed depending on formulation and type of membrane. This is shown in Figure 1. The amount of insulin recovered for both formulations on cornea was less than that recovered using a synthetic membrane. The total amount of insulin cleared from the ocular surface using the solution formulation was greater than the amount cleared from the PROLOC® formulation. Small amounts of the insulin in PROLOC® gel were visible at the end of the experiment. Rinsing the chamber with PBS and analyzing the fluid confirmed insulin remained in the gel at the end of the study. Notably, no insulin was found diffusing across the membranes.
There is clearly a difference in the release profile for the solution and PROLOC® gel. Figure 2 displays the difference in the clearance profile between the solution and PROLOC® gel. It shows the solution formulation is cleared from the membrane surface in a higher concentration of drug over a shorter time. The PROLOC® gel formulation provided a sustained release of insulin over the 5 hours studied.
The PROLOC® gel formulation continued to deliver drug at a steady rate throughout the 5-hour period studied while the solution formulation was rapidly removed from the ocular surface. Notably, some of the PROLOC® gel, which still contained insulin, was retained on the membrane surface until the end of the 5-hour study. The partially hydrated PROLOC® gel, therefore, may be able to deliver drug for more than 5 hours. Further studies should investigate PROLOC® gel’s full potential.
The differences between solution and PROLOC® Gel formulation were notably independent of what membrane was chosen for the study. Significant differences were noted, however, in the recovery of insulin dependent on the membrane used. The lower recovery of insulin from the cornea studies indicates that there is a protein tissue interaction that impacts the results. This highlights the importance of membrane selection based on the focus of the study. Here the synthetic membrane should be chosen for the evaluation and comparison of the product release profile. The corneal membrane should be chosen to directly mimic the interaction of the drug with the target tissue.
PROLOC® has previously been shown to provide sustained release of an antibiotic from an ocular mini tablet. This study indicates in a gel form it is suitable for ocular delivery of a protein. Since PROLOC® can be supplied as a dry ocular mini tablet or as a partially hydrated gel it has the potential to provide both product stability and constant ocular drug delivery.
Amanda Goode– Chemist, Encompass Pharmaceutical Services
Emily Mayville– University of Georgia
Angela Thomas– Sr. Chemist, Encompass Pharmaceutical Services, Georgia
Eric Rustin– Sr. Chemist, Encompass Pharmaceutical Services, Georgia