Roles of biotransformation in the antitumour activity and toxicology of the hypoxia-activated pre-prodrug PR-104

Abstract

Hypoxia is a characteristic of tumours and an attractive therapeutic target. PR-104, a hypoxia-activated prodrug in clinical trial, is a phosphate ester which is rapidly converted to the alcohol PR-104A, and then activated by reduction to hydroxylamine (PR-104H) and amine (PR-104M) metabolites selectively in hypoxic cells. This thesis explores the pharmacology of PR-104. The primary objectives were to define relationships between metabolic transformation of PR-104, its toxicity to normal tissues, its clearance from the body, and its antitumour activity in preclinical models. DNA damage responsible for cytotoxicity of PR-104A was characterised by comparing sensitivity of repair-defective Chinese hamster ovary cell lines with their repair-competent counterparts. The activated metabolites PR-104H and PR-104M showed a repair profile similar to typical DNA interstrand crosslinking agents, including dependence on the ERCC1-XPF endonuclease (implicated in unhooking DNA interstrand crosslinks at blocked replication forks) and Rad51D (required for homologous recombination repair). PR-104A cytotoxicity under hypoxia is accounted for by DNA crosslinking by these reduced metabolites. This study showed that hypoxia, reductase activity, and DNA interstrand crosslink repair proficiency are key variables that interact to determine PR-104A sensitivity. PR-104 was rapidly excreted in mice, with approximately equal excretion in urine and faeces. Investigation of pathways of biotransformation in mice, rats, dogs and humans identified major species difference in biotransformation of PR-104 in that N-dealkylation and subsequent metabolites from the mercapturic acid pathway was major in rodents, while O-glucuronidation of PR-104A much more extensive in dogs and humans. The nitroreduction pathway was found in all species, with different patterns in that PR-104M was prominent in rodents but PR-104H was higher in humans. Overall, the results suggest that rodents may not be appropriate for modelling human biotransformation and toxicology of PR-104. A rapid and sensitive UHPLC-MS/MS method was developed for quantitation of the major metabolites of PR-104 in humans. Application to plasma samples from phase I clinical trials demonstrated approximately dose-linear pharmacokinetics but high variability between individuals in concentrations of the glucuronide. The species and individual difference in glucuronidation were further examined using a microsomal metabolism model. Glucuronidation was fastest in dog liver microsomes, followed by human then rodent microsomes. UGT2B7 appeared to be the only human isoform able to conjugate PR-104A, which was further confirmed by a strong correlation between UGT2B7 protein content and glucuronidation activity in microsomes from different individuals. The results suggest that differences in UGT2B7 activity may contribute to individual differences in PR-104A pharmacokinetics, and that the role of this enzyme in the toxicology of PR-104 warrants further investigation.