Natural product ortho-quinones have long shown potential as anticancer compounds, being the active components of traditional anticancer therapies from across the globe. Their ability to futile redox-cycle and cause an intracellular surge in reactive oxygen species is a promising mechanism of action for cancer cell killing. This mechanism of metabolic interference, distinctly different to the drug-to-target binding paradigm, offers an alternative strategy for cancer treatment. Ortho-quinones can synergise with other therapeutics and with radiation therapy and can overcome cancer resistance mechanisms. The compounds show promise for treatment of NQO1+ solid tumours as well as acute leukaemias, where unpredictable cancer heterogeneity can increase the need for a non-specific metabolic approach to cancer killing. However, ortho-quinones have been largely abandoned from development pipelines as their high reactivity in vivo causes damaging redox-related haematological side effects, for example anaemia and methaemoglobinemia, that limit possible dosage. A method that can protect and target ortho-quinones in redox-inactive form, then trigger their activation in situ at the desired site of action has the potential to enable safe use of the compounds in the clinic. Prodrug strategies may be used to control the reactivity of a drug compound within the body by masking of drug structure through compound derivatisation with functionality that has controllable lability. Pro-moieties may be removed selectively at the site of drug action, to disclose the active drug. This thesis describes the development of a new method to protect ortho-quinones as para amino- or para-hydroxybenzyl ketols. These species were discovered to be self immolative upon deprotection of the para- protecting group and undergo a carbon carbon bond breaking 1,6-elimination to release the associated hydroquinone, which re-joins the redox cycle and subsequently oxidises to the quinone. By decoupling the pro-trigger from the drug attachment site, the benzyl ketol strategy can protect ortho-quinones with various pro moieties, for example, enzymatically cleavable peptides or sugars, and can facilitate addition of a targeting-moiety to allow specific localisation of an ortho-quinone drug within the body. Ortho-quinone protection strategies reported to date rely on cleavage of pro-moieties in acidic environments or via non-specific esterase-mediated hydrolysis. This work is the first report of a strategy that may be used to design ortho-quinone prodrugs activatable by action of more specifically located, cancer-associated enzymes. As part of the study, the mechanism, kinetics and rate-pH dependence of examples of fragmenting benzyl ketol derivatives were explored. Para-aminobenzyl ketols were found to fragment with a rate in the order of 10-5 s-1 and, notably, with an acidic rate-pH dependence. The maximum rate and exact rate-pH profile of the species were influenced by the quinone protected, with model ortho-quinone 9,10-phenanthrenequinone displaying a peak fragmentation rate of 2.4 × 10-5 s-1 at pH 5 in aqueous solution. Mechanistically, the fragmentation of para-aminobenzyl ketol linkers was predicted to be triggered upon formation of a minority species generated by protonation of the carbonyl in the benzyl ketol structure while the aniline nitrogen remains unprotonated. Para hydroxybenzyl ketol derivatives were found to fragment with varying rate in the order of 10-1-10-6 s-1, depending upon pH and the quinone species protected. In contrast with para-amino derivatives, they demonstrated a basic rate-pH dependence, due to their fragmentation being mechanistically dependent upon deprotonation of the para-hydroxy phenol. As an initial test of both linkers in a prodrug, cathepsin-B cleavable dipeptide and β-glucuronidase cleavable sugar prodrugs of ortho quinone β lapachone were synthesised and upon the relevant enzymatic action, quinone release was demonstrated. To test the potential of the protection strategy for the selective targeting of ortho-quinones for treatment of cancer, the strategy was used to build antibody-prodrug conjugates of ortho quinone natural product β lapachone for investigation against leukaemia models. The conjugates contained β-lapachone protected as a para-aminobenzyl ketol attached via a cathepsin-cleavable valine-citrulline linker to a CD33 targeting antibody. In vitro, the conjugate selectively released drug upon antibody internalisation inside target cells, with toxicity not observed for a non-internalising conjugate containing the payload at the same concentrations. Interestingly, a non-targeted small molecule prodrug with the same dipeptide pro-moiety also showed lower toxicity in vitro at comparable concentrations. Given the demonstrated acidic dependence of the fragmentation of para aminobenzyl ketol protected β-lapachone, with peak fragmentation rate observed at pH 3-4, this suggested that antibody-mediated internalisation leads to increased fragmentation of para-aminobenzyl species through effective localisation to acidic cellular compartments. This observation suggested an additional layer of protection for targeting the ortho-quinone in this manner, as if any linker deconjugates in circulation, even if its pro-trigger is cleaved, the benzyl ketol unit will undergo very limited release of quinone at neutral pH. Most notably, against a xenograft murine model of acute myeloid leukaemia, a CD33 targeting conjugate of β-lapachone with drug-antibody-ratio of two demonstrated significant antitumor activity without noticeable side effects. Treated animals survived for a mean of 52 days versus 24 days for untreated animals, with two out of the five treated animals alive at 76 days post treatment. This experiment provides proof-of-concept of the use of the benzyl ketol protection strategy to effectively deliver ortho-quinone payloads in vivo. Importantly, this work also demonstrates the first use of ortho-quinone payload in an antibody-conjugate for leukaemia treatment, something which is only possible with a redox-inactivating protection strategy due to potential reactivity of the quinone with the antibody carrier. Furthermore, the results validate the use of quinones as low-range potency (~500 nM IC50) payloads in ADC therapy of leukaemia. The ortho-quinone protection strategies developed in this thesis are modular and may be adapted as required for future applications. It is hoped that the discoveries begin to allow a finer control over the reactivity of these compounds and provide a foundation for further development of ortho-quinone prodrugs and targeted therapeutics.