Studies towards Radiolytically-activated Prodrugs of Kinase-inhibiting Anticancer Agents

Abstract

Cancer is one of the largest causes of mortality worldwide and is the leading cause of death in New Zealand. Hypoxia is a feature of all cancer types and is defined as a state of reduced oxygenation that alters the biological functions of a cell. Radiolytic fragmentation of the N-methoxyl bond of N-methoxyquinolinium (1.32), selectively in anaerobic conditions, was discovered by Dr Robert Anderson. This novel process results in elimination of an hydroxymethyl reducing radical, which in turn reduces more N-methoxylated compound in a chain reaction, leading to 'release' of the quinoline ring. A large number of tyrosine kinase inhibitors on the market contain a quinoline ring at their core (e.g. 1.38, 1.39, 1.47 and 1.48). While the inhibitors are effective anti-cancer agents; tyrosine kinases are present in healthy tissue and thus can lead to undesirable side effects. Development of radiation-activated prodrugs based on N-methoxyl bond fragmentation would aid in eliminating healthy cell toxicity by selective radiolytic release in hypoxia allowing the neutral cytotoxin to penetrate cells causing site selective kill. A proof of concept study was initially undertaken in which 28 N-methoxylated pyridiniums were successfully synthesised. Radiolysis studies were undertaken to evaluate what effects different substituents had on the radiolytic chain length of a compound. The results found nitrile and nitro substitution generally afforded shorter radiolytic chain lengths while phenyl, bromine and chlorine substitution generally afforded longer chain lengths. Computational studies were undertaken to determine if electron affinity or bond dissociation energies could be used to predict the radiolytic chain length of a given compound. It was found that the more electron affinic a compound was the longer a chain length it exhibited. Observation of nucleophilic substitution of 4-chloropyridin-1- ium (2.45) with methanol led to successful synthesis of 1-methoxy-4-(phenylamino)pyridin-1-ium (2.1). This observation is notable due to N-methoxy retention, for potential use at the final stage of synthesis of EGFR inhibitors (1.95). Synthesis targeting N-methoxylated analogues (1.95) of EGFR inhibitors encountered stability issues when isolating the 4-chloro-N-methoxyquinolinium series. A new synthetic route was developed which resulted in the successful synthesis of analogues 3.70 and 3.71. 3-Cyano-1,6,7-trimethoxy-4-(phenylamino)quinolin-1-ium (3.70) then underwent radiolysis and stability studies, where 3.70 exhibited a radiolytic chain length of 11 and was stable over a 24 hour period in α-MEM containing 5% FCS (92.4%). The observation of radiolysis initiating the chain reduction of 3.70 releasing 3.68, combined with 3.70 exhibiting excellent stability, indicates the potential of this class of compounds (1.95) to be successful radiation activated prodrugs.