Main Menu

Hitting the right target – validating new approaches for innovative cancer drugs


Therapeutic innovation and clinical benefit require the discovery of truly novel drugs that act on previously unexploited molecular targets.

Posted on 10 October, 2014 by Professor Paul Workman

Major therapeutic innovation and the ensuing significant clinical benefit require the discovery of truly novel drugs that act on molecular targets that are previously unexploited.

As estimated previously by my colleagues and I here at The Institute of Cancer Research, despite major progress leading to the identification of 500 or so known cancer-causing proteins, we have approved drugs available for only five per cent of these.

This fact is one of the biggest motivations for our 200-strong team of cancer therapeutics researchers at ICR. Only by discovering drugs that tackle all cancer proteins and pathways can we truly hope to defeat cancer with personalised medicines. And this full range of drugs will be essential to build the spectrum of creative drug combinations that are needed to meet the major challenge of overcoming drug resistance.

A new “current opinion” article – which I co-authored with my ICR colleague Professor Julian Blagg – provides an updated roadmap for target validation, which is a critical bottleneck in innovative drug discovery across all diseases. 

Target validation and druggability

Our Cancer Research UK Cancer Therapeutics Unit at the ICR is responsible for designing, synthesising and testing the finely tuned chemical compounds that will ultimately become innovative cancer drugs.

The most effective drugs will have a high affinity with their target protein – binding strongly to it, like a key in a lock, without also binding strongly elsewhere. And they will bind in the right place, which must be a site on the target protein which controls its behaviour or prevents it interacting with other molecules to drive cancer.

However, before drug discovery begins it is critical to identify those targets that if modulated are the most likely to deliver a strong therapeutic impact. This requires a robust package of evidence – known as validation – not only for the causal involvement of the molecular target in the biological initiation and maintenance of cancer, but also of the potential for its modulation by small-molecule chemical compounds – so-called ‘druggability’ – even if this is technically challenging.

With the costs of clinical trials and their chances of failure still so high – and with a plethora of potential targets arising from cancer genome sequencing and other high-throughput initiatives – validating the best possible targets has never been more important.

Guidance for researchers

Building on the earlier work from the ICR our new article also provides updated guidance – with recent examples from the work of ourselves and others – on critical concepts and approaches to improve target validation. A key recommendation is the use of carefully designed and selected small-molecule chemicals as a key step in target validation.

We stress use of small-molecule ‘tool’ or ‘probe’ compounds alongside the complementary technological approach of genetic modification of the same target. If both techniques support the importance of the target in cancer this builds confidence that the drug discovery project is likely to be successful – and also that it is feasible to modulate the target with a small-molecule drug. Both as aspects are important.

In fact there can be good reasons why different results are obtained with genetic versus chemical validation approaches. Genetic changes are not always able to mimic the effects of a chemical modulator. And interesting results with a chemical compound can lead to new biological insights. In certain cases the small-molecule approach is the only practical option.

As part of the roadmap for target validation, we emphasise the importance of using at least two different classes of chemical inhibitor (or chemotypes) that act on the same potential target – so as to increase the likelihood that the anticancer effects that are seen are ‘on target’ and not due to unrelated and undesired effects.

We also stress the value of including in the research work a closely related form of the compound that is deliberately designed to be inactive – this can further strengthen the link between target modulation and the cellular consequences, and avoid misleading ‘off-target’ effects.

Validating a new target being exploited in the clinic

For several years, my colleagues and I at the ICR have been exploring the potential of targeting a protein called heat-shock protein 90 (Hsp90) as a cancer treatment. Many outside the ICR had questioned its importance as a drug target and also suspected that agents targeting Hsp90 would cause too much collateral damage in healthy cells. 

Validation of Hsp90 as a target – in the lab and then in the clinic – was in fact based initially on natural product inhibitors and subsequently by synthetic prototype drugs with improved ‘fitness factors’.  My colleague Ian Collins and I recommended these fitness factors as a means to ensure that the chemical probes used will give the correct results in target validation, again reducing the risk of misleading conclusions.

After producing specific inhibitors of Hsp90 with improved fitness factors – underpinned by structural insights to ensure that the chemical compound ‘key’ provided the best fit to the protein ‘lock’ – our ICR team discovered in collaboration with colleagues at Vernalis a drug called AUY922 which has now shown very promising results in phase II clinical trials by Novartis.

Our work on Hsp90 led to an interdisciplinary team of ICR scientists led by myself and my former ICR colleague Professor Laurence Pearl receiving the Translational Cancer Research Prize from Cancer Research UK.  More important is the benefit seen in cancer patients, including those with breast and lung cancers resistant to previous targeted therapies.

A strong learning point here is that validation of Hsp90 as a cancer target by genetic means has proved very difficult and it was the use of small-molecule probes that proved crucial in building confidence in the target.

Missing the target

As well as stressing fitness factors for chemical probes, our new article also emphasises the importance of providing direct evidence – using biomarkers of biochemical pathways – that the anticancer effect of a chemical probe or drug is indeed due to modulation of the target in the cancer cell. One example we highlight where this was not done is iniparib, which underwent expensive late-stage clinical trials in breast cancer before it was exposed as having a completely different mechanism of action to the one originally believed.

Researchers trialled it in the belief that it belonged to a class of drug called PARP inhibitors, several of which are currently undergoing clinical trials as a targeted cancer treatment for women with mutations to the BRCA1 and BRCA2 genes. But recent studies of iniparib in the laboratory have shown that it has a very different mechanism of action, not in fact involving PARP inhibition at all.

Not only did that mistake lead to the failure of a phase III clinical trial – affecting around 2,500 women with breast cancer – it also undermined for a period the potential of ‘genuine’ PARP inhibitors to treat BRCA-mutated cancer. True PARP inhibitors – such as olaparib – have undergone pioneering trials by ICR researchers at our partner hospital, The Royal Marsden, and much of the underpinning science proving their potential in BRCA-mutant cancers took place here over a period of almost 30 years.

Much of that work could have gone to waste had the molecular behaviour of iniparib not ultimately been investigated much more thoroughly, once the trial had failed.

Improving the rigour of chemical probes

There has been a welcome increase in the availability of high-quality chemical tools and the use of biomarker assays to confirm ‘on-target’ effects. But sadly there continue to be frequently published examples where researchers use unsuitable compounds, often suffering from clearly undesirable chemical structures and properties – and frequently also ignoring the need for essential biomarker evidence of specific target engagement.

The roadmap we propose will increase the rigour and value of chemical probes in driving therapeutic innovation and will avoid misleading biological conclusions and time wasted on unsuitable targets and drugs.

Most importantly, the guidelines will help ensure that the very best cancer targets are taken forward for drug discovery – and that the most innovative and effective cancer drugs are produced for the benefit of cancer patients.


HSP90 inhibitors
comments powered by Disqus