O1.1 - Discovering selective anticancer drugs: Succeeding slowly is better than failing fast

Abstract

ABSTRACT Most of the first generation of anticancer drugs act by preventing cellular proliferation. The fact that can induce remissions in the tumour without lethal toxicity to the patient is, at first sight, surprising. However anti-proliferative agents have been the mainstay of cancer treatment for 60 years and continue to play a major role. Perhaps we should view their lack of toxicities to non-proliferating tissues as an advantage rather than seeing only that their toxicity to proliferating tissues is a problem. Getting the right dose is important for all anticancer agents. In the case of carboplatin, the drug nearly failed in early development because of variability in patient tolerance leading to unpredictable toxicity. However it was possible to trace this variability back to pharmacokinetics and generate a dosing formula that has permitted the global use of the drug. Another widely used drug that almost failed in clinical development is pemetrexed. This drug targets the folate pathways and produced variable and severe toxicity in early trials. In this case the cause of the variability was pre-treatment folate status, but this could only be detected using the sensitive surrogate, homocysteine. In order to design an effective anticancer drug we clearly need a target that is specific to the tumour. In addition the target needs to be essential to the oncogenic process – targeting an epiphenomenon is less likely to be effective. In additions to specific driving mutations in oncogenes, fusion proteins and gene amplification have also provided some excellent targets. A further recent development has been that the phenomenon called synthetic lethality may be exploited for cancer treatment. Synthetic lethality occurs when blocking either one of two pathways on its own has no effect, but blocking both is lethal. BRCA related tumours have uniquely lost both alleles for one the BRCA proteins which are essential for homologous recombination repair (HR). Cells without HR function are uniquely sensitive to inhibition of poly(ADP-ribose) polymerase (PARP). Thus PARP inhibitors are active in tumours arising in a BCRA deficient background. However many cancers occurring with no genetic background have acquired loss of HR functionality, and these are also sensitive to PARP inhibitors. The challenge for further development is to generate a suitable predictive test for HR function.