Op22. modelling the fate of interstitially delivered therapies to brain malignancies

Advances in therapy delivery systems are providing new opportunities for the treatment of malignancies of the central nervous system. There is opportunity for locally administered therapies that overcome the Blood Brain Barrier (BBB) to be implanted at the point of tumour resection surgery into or around the cavity left behind. This is particularly the case for Glioblastoma Multiforme (GBM). The pharmacodynamics of a therapy delivered in this way into brain tissue is of key importance to optimizing these therapies as the correct dose reaching the micro-deposits of malignant cells unreachable by surgery is thought to be key in slowing disease progression. We examine the particular case of a drug loaded poly(lactic-co-glycolic acid)/poly(ethylene-glycol) (PLGA/PEG) polymer particle paste designed to release drug to surrounding brain tissue post surgical implantation. By experimentally parameterizing the distribution, release and uptake kinetics we have modeled the biodistribution of methotrexate conjugated to a fluorescent moiety fluorescein isothiocyanate (FITC). We have also used micro-computerized tomography (µ-CT) to determine the microstructure of the PLGA/PEG construct and assess how the drug progresses out from the interior of the construct. The model consists of a system of reaction-diffusion equations describing the time evolution of spatially dependent concentrations of a drug in three different materials: normal brain tissue, cerebrospinal fluid (CSF) and the PLGA/PEG implant, each have their own diffusion tensors that describe the anisotropy of the brain. The model was built upon simple one-dimensional time course experiments measuring the rate at which a drug molecule can pass through a standardized volume of PLGA/PEG paste and the rate at which the drug is taken up by brain tissue. We simulated computationally a gap of CSF between the PLGA/PEG construct and the brain tissue to ascertain whether good tissue apposition was key to drug delivery. Known release rates of drugs from previous studies were also fitted to the model, allied to this their known logD values were used to test the model. We have been able to see that the chemical properties of the compounds studied, in particular the logD values, have a great effect on the simulated release, passage across the CSF and uptake in the brain tissue. Hence not having a CSF barrier between Allying these drug release and distribution models to models of tumour growth, the interaction of the polymer with the tissue and the polymer and the drug we aim to better select drugs according to chemical properties, simulate repurposed drugs and predict drug coverage where it is required to improve the control of GBM as a disease.