Convection-Enhanced Delivery of Carboplatin PLGA Nanoparticles for the Treatment of Glioblastoma

Abstract

We currently use Convection-Enhanced Delivery (CED) of the platinum-based drug, carboplatin as a novel treatment strategy for high grade glioblastoma in adults and children. Although initial results show promise, carboplatin is not specifically toxic to tumour cells and has been associated with neurotoxicity at high infused concentrations in pre-clinical studies. Our treatment strategy requires intermittent infusions due to rapid clearance of carboplatin from the brain. In this study, carboplatin was encapsulated in lactic acid-glycolic acid copolymer (PLGA) to develop a novel drug delivery system. Neuronal and tumour cytotoxicity were assessed in primary neuronal and glioblastoma cell cultures. Distribution, tissue clearance and toxicity of carboplatin nanoparticles following CED was assessed in rat and porcine models. Carboplatin nanoparticles conferred greater tumour cytotoxicity, reduced neuronal toxicity and prolonged tissue half-life. In conclusion, this drug delivery system has the potential to improve the prognosis for patients with glioblastomas.

Fig 1. Characterisation of carboplatin NP.

(A) Carboplatin NP show a consistent spherical shape, less than 200nm in size. Morphological analysis of the nanoparticles was conducted using TEM scaled at 200nm. (B) Approximately 100% of carboplatin is released within 24 hours. Release profile of carboplatin from nanoparticles was conducted by analysing platinum content by ICP-MS at defined timepoints.

Fig 2. In vitro cytotoxicity and uptake of carboplatin NP.

(A) Carboplatin NP showed increased cytotoxicity over 48 hours. The cytotoxic effect of the carboplatin nanoparticles (Carboplatin NP) were assessed by MTT assay in UPAB and SNB19 human glioblastoma multiforme (GBM) cell lines. For UPAB, 0.18mg/ml and for SNB19, 0.03mg/ml carboplatin was used as these concentrations represent the IC50 after 72 hours. Paired t-test statistical analysis comparing cytotoxicity revealed significant differences for SNB19 at 24hours (p = 0.001) and 48 hours (p = 0.004), indicated by asterix. (B) Uptake of fluorescein-labelled carboplatin NP (green) occurs within 24 hours of dosing. Cells were dosed and fixed after defined periods of culture. Cells were stained with phalloidin (red), to visualise actin cytoskeleton and DAPI (blue) for cell nuclei. Uptake into cells is indicated by white arrows.

Fig 3. In vitro neurotoxicity in primary rat hippocampal cultures.

Carboplatin NP are less toxic to neurones compared to the free drug. Primary rat brain hippocampal cultures were dosed with either carboplatin (0.03mg/ml) or carboplatin NP (1mg/ml) and assayed after 72 hours of culture. (A) MTT analysis shows significant increase in cell viability with carboplatin NP (p<0.001). (B) Carboplatin alone causes deregulation of the neurones and loss of the glial cells, whilst carboplatin NP retained neuronal connections and glial cells. Immunofluorescent analysis of the neurones (B3tubulin; green) and glial cells (GFAP; red). Cells were counterstained with DAPI to visualise cell nuclei (blue) (Scale bar; 100μm).

Fig 4. CED of carboplatin NP into the striatum of rat brains.

Immediately after CED, nanoparticles distribute throughout the striatum. (A) CED of aCSF alone (5μl volume) or fluorescein-labelled carboplatin NP into the striatum of rat brain was conducted (1mg/ml; 5μl volume). These were analysed by sectioning around the needle track and counterstaining with DAPI (Scale bar; 250μm). (B) Carboplatin NP show increased tissue retention after 24 and 48 hours. Following CED of carboplatin NP (1mg/ml; 5μl volume), punches of the brain tissue obtained around the site of infusion were analysed by ICP-MS. Paired t-test analysis revealed a significant increase in the tissue half-life of carboplatin NP at 24 hours (indicated by asterix).

Fig 5. Toxicity analysis after CED into the striatum of rat brains.

CED of aCSF (negative control), carboplatin (0.72mg/ml; 5μl volume) (or carboplatin NP (1mg/ml; 5μl volume) was conducted and toxicity in rat striatum assessed. Dual IHC analysis of neurons (NeuN) and Glial cell (GFAP) demonstrated minimal toxicity localised to the needle track. No glial or neuronal cell loss was observed elsewhere. Scale bar: 100μm.

Fig 6. Toxicity assessment after dose-escalation study in pigs.

A dose-escalation study comparing CED of carboplatin NP with aCSF (co-infused with 0.2% Gd) was conducted in a porcine model. (A) Widespread putamenal distribution was observed using real-time MR imaging (B) Fluorescence immunohistochemistry demonstrates intact neuronal (E & L) and glial (D & K) networks at the site of infusion. Repeat dosing of carboplatin NP was not associated with increased macrophage activation compared to control (G & N). Minimal toxicity was also confirmed by maintenance of cellular architecture (H & O) (Scale bar; 250μm in magnified images, 100μm in images showing entire section (B&I)