Chapter 1 - Preclinical Drug Development: Translating Basic Research into Clinical Work
The development of a new anti-cancer agent begins with the evaluation of its anti-tumour activity against a panel of malignant cell lines. These tests help to identify compounds that deserve further evaluation in animal models. Several methods exist to detect the anti-tumour effect of a new agent against tumour cell lines, such as anti-proliferative assays using incorporation of radioactive nucleotides like [3H] thymidine, direct cell counting, or colony formation. Others assess viability or growth using colorimetric assessment.
A screening model that has been very useful for the evaluation of hundreds of compounds is represented by the US National Cancer Institute screening model (NCI-60). It is composed of 60 different cell lines derived from the major human tumours and provides a source for rapid evaluation of the in vitro anti-tumour activity of new compounds. Every new agent is tested against each of the cell lines in order to evaluate its ability to inhibit growth or cause cell death. Based on parameters such as the concentrations of the drug that cause: growth inhibition of 50% of the cells (GI50), or total growth inhibition (TGI), or cytotoxic killing of 50% of the cells (LC50), a specific fingerprint is produced for each compound that can be compared with the activity of others with the same or different mechanisms of action.
Although the empirical screening of natural products has discovered anti-cancer drugs such as paclitaxel and trabectedin, the current strategy of drug discovery favours approaches that are rationally and biologically driven to develop agents that inhibit specific molecular targets involved in tumour formation and progression. This can be achieved through high-throughput screening of small-molecule libraries or through a more sophisticated structure-guided discovery approach that leads to the identification of compounds that interfere with specific molecular targets. Lead compounds are subjected to specificity evaluations to test their ability to engage and interact with their putative molecular targets. Studies in vitro are therefore designed not only to show that a new agent has inhibitory or cytotoxic activity against cell lines, but also to demonstrate that it is able to produce target inhibition to support its underlying mechanism of action.
The assessment of target inhibition of a new agent in vitro (in cell-based and non-cell-based assays) is generally based on the concentration of the drug necessary to inhibit the activity of its target. For many new agents that target specific enzymes, this is assessed by measurement of the concentration of the drug needed to produce 50% of enzymatic inhibition (IC50). Drugs that cause inhibition at low doses in vitro (i.e. IC50 at low nanomolar range) are preferred, as they more likely result in favourable therapeutic index in clinical studies. The IC50 against other enzymes in the same family must also be determined in order to define the specificity of the agent against its target. If a drug inhibits several enzymes at low IC50, it may act by modulating different targets. This finding must be taken into consideration when studying the mechanism of anti-tumour activity of a new drug, and it may also raise the possibility of undesirable side effects due to multiple target inhibition.
While in vitro studies using tumour cell lines represent an initial step in the evaluation of anti-tumour activity and the elucidation of the mechanism of action of a new anti-cancer agent, they have limitations in predicting positive effects in animal models and, more importantly, in patients with cancer. In fact, cell lines present important biological differences from the tumours they derive from and they do not reflect the intricacy of human cancers and the complex interplay between cancer cells and their micro-environment. It has now become clear that important mechanisms of resistance to treatment depend on the relationships between cancer cells and the surrounding stromal cells and these conditions are not easy to reproduce in preclinical studies. The use of ex vivo models (i.e. using cells or tissues taken directly from patients and tested in an external environment with minimal artificial alterations), co-culture of tumour cells together with stroma cells and the development of in vitro models with tumour spheroids or multi-layered cells represent some of the possibilities to better reproduce in vitro the complexity of human tumours and the relationships with their micro-environment.
Recently, the discovery that specific genetic changes are responsible for the development of particular tumour types has permitted the anti-tumour activity of some new agents to be evaluated in tumour cell lines expressing these genetic changes. Furthermore, comparison of the behaviour of these agents in wild-type cell lines of the same tumour types can be informative. An example is represented by the anti-tumour activity of PARP inhibitors in the context of BRCA1/2-deficient cell lines in comparison with lack of activity in cell lines with heterozygous or wild-type BRCA1/2. The evidence from such in vitro studies was the basis for the subsequent clinical development of several PARP inhibitors for tumours bearing homozygous mutations in BRCA1/2 genes. The in vitro evaluation of a new agent’s anti-tumour activity against tumours with specific genetic changes may therefore provide a strong rationale to support its clinical development in a genetically selected patient population.