The future of drug toxicity testing?

NanoBioDesign, a spin-out from the Imperial College London, UK, has received a £0.5m (€0.74m) investment from the Imperial Innovations Group and The Capital Fund to enable the company to start industrial production of its drug toxicity screening technology. P450 enzymes are probably the most important part of the primary drug metabolism pathway and if one drug prevents the metabolism of another drug this can lead to the accumulation of the second drug to dangerously high levels that can cause severe adverse events (AEs) to occur. P450-related AEs have been a common cause for post-approval drug withdrawals with drugs for a range of indications having been withdrawn due to such problems. According to Dr Stephen Collins, CEO of NanoBioDesign: "if you can fail a drug early, you can fail it cheaply - the FDA [US Food and Drug Administration] has estimated that a 10 per cent improvement in predicting drug development failure could save $100m [€73.4m] in the development of a drug."​ Dr Collins went on to mention that in 2004, Stephen Williams, Pfizer's head of global clinical technology, said​ the company had wasted $2bn over the last decade on drugs that failed in clinical trials or were withdrawn from the market due to liver toxicity problems. Currently, testing for interactions between drug candidates and P450 is carried out on drug lead candidates when a company has whittled the number of drugs of interest down to less than 100 entities. The drugs are incubated with P450 enzymes expressed in either insect cells or human liver microsomes before analysis by either the addition of fluorescent probe or by HPLC/MS (high performance liquid chromatography / mass spectrometry). According to Dr Collins, these studies take at least 90 minutes and can lead to significant bottlenecks in the screening process. NanoBioDesign aims to reduce this bottleneck by using self-sufficient recombinant human P450 enzymes that can be immobilised onto electrode surfaces that give more reliable results in about twenty minutes. The device will contain the 5 different P450 liver enzymes, 3A4, 2D6, 2C9, 2C19 and 2E1, which account for 97 per cent of drug metabolism. The company envisages that the technique could replace current fluorescence testing reduce the number of HPLC/MS experiments needed as only compounds with favourable kinetics will progress through to that stage. "The assay format is incredibly simple, the electrode is immobilised in a well in a three electrode system that after titrating the drug into the well generates Michaelis-Menten curves that allow you to calculate either Km [metabolism rate constant] or Ki [inhibition rate constant],"​ said Dr Collins. He continued by explaining that drug-drug interactions can be discovered by titrating two solutions into one well, keeping the drug candidate concentration constant, and seeing how the rate profile changes compared to that of the profile for the drug on its own. Dr Collins stressed that drug-drug interactions are not the only interactions that pharmaceutical companies need to study as food stuffs such as grapefruit or cranberry juice can also have an inhibitory effect on the metabolism of certain drugs. The company has also extended the approach to include five allelic variants of 2D6 and 2C9 which are of interest to the emerging pharmacogenetics market as different racial groups tend to have different variants of these enzymes. These enzyme variations can lead to dramatically altered drug responses that range from variations in the dosage that needs to be recommended to certain drugs, for example NitroMed's BiDil (isorbide dinitrate/hydralazine HCl), being prescribed for only one specific racial group. NanoBioDesign hope to close another funding round in the next few months and plan to have instruments at beta-test sites by the end of next year to allow pharmaceutical companies to use the device in parallel with their current toxicity measurement technologies.