Contemporary pre-clinical development of anticancer agents - What are the optimal preclinical models?

The successful identification of novel effective anticancer drugs is largely dependent on the use of appropriate preclinical experimental models that should possibly mimic the complexity of different cancer diseases. The huge number of targets suitable for the design of new anticancer drugs is producing hundreds of novel molecules that require appropriate experimental models to investigate their mode of action and antitumour activity in order to select for clinical investigation the ones with higher chances of being clinically effective. However, our ability to predict the clinical efficacy of a new compound in the clinic based on preclinical data is still limited. This paper overviews the in vitro/in vivo preclinical systems that are currently used to test either compounds with an unknown mechanism of action or compounds designed to hit cancer-specific or cancer-related molecular targets. Examples of experimental models successfully Used to identify novel compounds are provided. Xenografts are still the most commonly used in vivo models in drug development due to their high degree of reproducibility and because, in some cases, particularly when orthotopically transplanted, they maintain several biological properties of the human tumours they derive from. Genetic models are very useful for target validation, but are often not sufficiently reproducible to be used for drug evaluation. The variety of animal models can be effectively used to optimally test drugs that presumably act by a defined mode of action, but final success is highly dependent on the ability of drug development teams to integrate different expertises such as biology, chemistry, pharmacology, toxicology and clinical oncology into a clever and well orchestrated plan that keeps in consideration both the complexity of cancer diseases, involving alterations of different pathways, and the complexity of drugs whose pharmacological properties are crucial to obtain the desired effects. (C) 2009 Elsevier Ltd. All rights reserved.