This blog is adapted from a guest post I wrote for The Cancer Blog for the European School of Oncology. You can find the original post here.
As someone who has spent a career in cancer drug discovery and development, in academic, biotech and big pharma settings, I find it very frustrating that we still lack the treatments we need to effectively control or cure many cancers.
So I welcomed the opportunity to give the opening presentation at the World Oncology Forum 2014, and to ask questions about what has to happen to translate the tremendous insight and knowledge we now have about the genetics and biology of cancer into effective treatments.
I opened on an optimistic note.
The success of the human genome project and its application to oncology has led over the past decade to an extraordinary increase in our understanding of the causation of cancer, and has opened up a new, ‘personalised’ approach to treating cancer.
There is a feeling of positivity among cancer researchers which I certainly share. Genetics and biology are being laid open, and ideas about rational therapies are still pouring out of labs.
It’s true that, while some of the new personalised therapies have been very successful, the overall impact so far has been disappointing. But an important point I stressed is that we do understand what the limitations are – and based on this we can now devise scientifically rational ways forward to move us to the next level.
We now know that cancers are able to develop resistance to molecularly targeted drug treatments – just as they did to the previous generation of cytotoxic drugs. This is because cancer cells evolve over time through a sort of Darwinian ‘survival of the nastiest’.
Such evolution leads to the high level of genetic and biological variation of cancer cells within a given patient. Even worse, the treatments given can themselves act as a ‘selective pressure’ to drive further variation and resistance.
So the task ahead is ideally to design therapies that will block cancer evolution. One class of drugs that does this that we at The Institute of Cancer Research have been heavily involved in developing – HSP90 inhibitors – is now in clinical trial. We need to find more targets of this type, for example those in epigenetic regulatory pathways.
Another important approach to overcome evolution and resistance is to develop drug combinations that can hit, for example, several different cancer targets or other alterations at the same time – a strategy which has worked well in controlling HIV.
But this is where we come up against a problem. Because, out of the 500 genes or so we have discovered that are responsible for causing cancer, we’ve only got drugs that will work against about 5% of them.
We still haven’t got any drugs that work, for instance, on RAS or on MYC or p53 – three of the big genes responsible for causing cancer. So we’re trying to find effective combinations when we have no drugs for 95% of genes that may be involved in resistance. There is a huge amount of work still to do in this key area.
Question 1
So the first question I posed at the World Oncology Forum was: how do we speed up the development of innovative drugs against more of these new cancer targets?
Part of the answer to this problem involves the technical challenges to find drugs for the less ‘druggable’ targets. But in fact there is another more systematic problem which lies in the ‘ecosystem’ that governs the way new therapies are discovered, developed, brought to market and paid for. It clearly isn’t functioning as well as it should for developing personalised cancer therapies.
Most companies are working on a relatively small number of targets, maybe 20 or 30. They all tend to invest in the same group of targets and often they prefer trying to develop a slightly better version of a drug that is already out there rather than going for a new target.
Taking a higher-risk, more innovative approach and working on currently undrugged targets, rather than developing ‘me-too’ drugs, is much more likely to deliver the sort of step changes in survival that we all want to see.
The risk-aversion that predominates in the pharma industry is understandable, from their point of view. They know that historically only 5–10% of drugs that enter phase I show enough activity to be approved, so they want to be very sure about the targets they work on.
But that thinking creates an innovation gap, or what we call rather more dramatically the ‘valley of death’. There is a gap – in fact a chasm – between early-stage research with real clinical implications, like the identification of a new target gene or the discovery of an early prototype drug candidate, and it being taken forward by companies to the point where a treatment can be trialled in patients.
Then there’s the problem that new drugs are being tested in settings where they are least likely to show a benefit, as single treatments given in late-stage disease which has already become drug resistant.
Even though we now have predictive biomarkers to select responsive patients, the benefit in heavily treated drug-resistant patients can be relatively small, which means that it takes more patients and a longer time to demonstrate in a clinical trial. This pushes up costs, which then means that bodies like NICE in the UK and its equivalents elsewhere may refuse to fund the drugs because they aren’t deemed sufficient value for money.
This is not helped by companies continuing to seek the maximum cost the market will bear. And once a drug has been turned down it becomes very difficult to find out what it might be capable of when used either earlier in the disease or in combination with other drugs. So great opportunities will be lost.
Question 2
So the second question I asked was, how can we incentivise the ecosystem to be more innovative in research and development?
I think the key to progress is all about decreasing risk aversion. Until now, the pharmaceutical industry has preferred to stick to tried and tested ways of discovering drugs and of steering them through the regulatory process – even though there may be strong scientific grounds for doing things differently when we’re dealing with rationally designed, targeted therapies and especially facing the challenge of cancer evolution and drug resistance.
It’s easy but fruitless to get into the blame game here. Pharmaceutical companies are not charities and they have a responsibility to their shareholders. Bodies like NICE have a responsibility to ensure health spending delivers value for money and regulators need to be convinced of the medical merits and safety of new therapies.
The question is how to find a way forward that works better for everybody.
If we accept that risk aversion is a big part of the problem – an assumption that is backed up by many conversations I have had and by my own experience in the industry – we have to find ways to bridge the innovation gap.
One way is to do more of the early high-risk drug discovery and development in the academic sphere, and in collaborations with Government, charities and industry.
Some organisations, such as Cancer Research UK and the National Institutes for Health in the US, are already pushing for this approach. They are beginning to provide ‘valley of death’-type funding to academic groups, including mine here at the ICR and MD Anderson and others, to help them get involved in early-stage, high-risk, innovative drug discovery and development. That can decrease the risk to industry in progressing a new approach.
I think this approach will really help, as exemplified by the breakthrough drug abiraterone that we discovered at the ICR and which was then subsequently licensed to industry and approved in late-stage prostate cancer.
We should be aware though that risk-aversion is not restricted to industry. There are many pressures at work in the academic sector – including hyper-competition, difficulties in accessing grant funding, and a preference for funding ‘safe’ rather than highly innovative approaches.
We also have to face tough questions about excessive requirements for publication which can actually slow progress and reduce efficiency. And conversely there are worries about the inability to replicate many academic discoveries in basic research which can then misinform choices about the best therapeutic targets to work on.
Question 3
So the final question I pushed very hard at the end of my presentation was really a challenge thrown down to the audience: How do we de-risk innovative drug development for companies? It’s a question from which a number of others flow: Where will the high-risk work be done? Who will fund it? And what should we expect from companies and academia?
It is going to require concerted efforts from all parts of the ecosystem to meet the challenges we face in overcoming cancer evolution and drug resistance. We need to work together as a community to find win-win solutions.
comments powered by