Abstract
Cancers remain one of the most debilitating diseases and are associated with a high morbidity and mortality. Current chemotherapy treatments are associated with severe adverse events arising from off-target toxicities which can limit the dose given and so reduce the efficacy of treatment. The loading of these cytotoxic drugs into particles < 1000 nm in diameter is an established approach to minimize the off-target toxicity and increase efficacy. These loaded particles are able to limit systemic exposure of the free drug and can accumulate at the tumour site, increasing local concentrations of the drug. However, limited clinical benefit over the free drug has been seen with the current generation of particles, with research into a new generation of polymeric particles with triggerable and targeted behaviour ongoing.
In this thesis block-co-polymers (BCPs) were synthesised using poly(ethylene glycol) (PEG) and responsive hydrophobic monomers. These BCPs were hypothesised to form particles that would be able to encapsulate cytotoxic drugs with minimal leakage until triggered by a tumour specific stimulus. The nature of the responsive units was such that upon triggering, toxic compounds would be generated in the tumour environment via self-immolation adding to the cytotoxicity produced by the encapsulated chemotherapy. The self-assembly of the BCPs was investigated in order to gain control over particle morphology, followed by the evaluation of responsive behaviours. Drug loading was then performed followed by in vitro and in vivo evaluation of these formulations.
The synthesis of the BCPs was optimised using an activators regenerated by electron transfer (ARGET) atom transfer radical polymerisation (ATRP) reaction that avoided degradation of the aryl azide capping group by reducing the copper content and conducting the reaction at room temperature (Chapter 2). This allowed the reaction to proceed without modification between monomers giving a good polydispersity (1.16 – 1.37). The monomer units contained a self-immolative group that was masked by the responsive moiety, either aryl azide or aryl boronate, which was anticipated to give responsive behaviour towards hydrogen sulfide (H2S) and hydrogen peroxide (H2O2) respectively. Linking the self-immolative group to the ethyl methacrylate was either a carbamate or thiocarbamate bond.
i
The self-assembly of the BCPs was conducted via nanoprecipitation using different organic solvents (DMF, THF, DMSO) and the anti-solvent PBS, with particle analysis via dynamic light scattering (DLS), cryo-transmission electron microscopy (TEM), cryo- tomography and small angle x-ray scattering (SAXS) (Chapter 3 and 4). In vitro triggering of particles was conducted using the H2S salt, NaSH, for the aryl azide BCPs (Chapter 3) or H2O2 for the aryl boronates (Chapter 4). The outcome of triggering on the particles was dependent upon the relative polarity of the bilayer and the pH of the triggering solution, with the more non-polar BCPs forming cross-linked structures at pH 7.4. The triggering of thiocarbamate containing aryl boronate BCP particles was found to generate H2S in the presence of carbonic anhydrase, a ubiquitous endogenous enzyme.
Loading of the particles with the cytotoxic drugs was performed using passive and active loading techniques. The aryl azide BCP particles were found to be unable to entrap useful amounts of either drug. The aryl boronate BCP particles were able to entrap 10-fold higher amounts of doxorubicin. These particles were evaluated in vitro which revealed an increased IC50 for the thiocarbamate containing particles. In vivo toxicity evaluation with Balb/c mice was piloted. The tested carbamate containing aryl boronate particles was found to be toxic to the mice.
This thesis explored the use of two endogenous triggers with self-immolative responsive groups. It was found that the aryl azide required a polar environment to trigger effectively, which limited loading and as such, alternate polymer structures would be required for use with aryl azide triggering. Aryl boronate particles achieved loading and triggering, although concentration of the formulation was required to reach clinically relevant concentrations of doxorubicin, which led to toxicity. Loading of the aryl- boronate particles therefore needs to be increased or used with a more potent drug. The thiocarbamate containing aryl boronate particles did however lead to the generation of H2S, and with a more potent azide-capped prodrug, polymer toxicity may be avoided. Alternatively, these particles could find use in pathologies characterised by oxidative stress, such as ischemia/reperfusion in cardiac or cerebral tissue, where generation of H2S has been shown to be protective.