The slow dawn of the age of targeted therapies

This morning, the National Institute of Health and Clinical Excellence (NICE) reversed its original, preliminary decisions over two cutting-edge ‘targeted’ melanoma treatments – ipilimumab and vemurafenib.

Both drugs will now be available on the NHS to suitable patients throughout England, Wales and Northern Ireland.

(Scotland has its own drug approval body, the Scottish Medicines Consortium, which has so far given a ‘no’ to both drugs)

This is great news. It turns out that the manufacturers of both drugs have now provided sufficient information, and set up suitable discount schemes, to allow the drugs to be considered value for money on the NHS – essential in these straitened times. We hope they’ll resubmit this information to the Scottish regulators as soon as possible.

But how does this fit into the bigger picture of targeted therapies and personalised medicine?

We thought that in the light of this news, and of some recent critical pieces in the mainstream media, we’d take a look at the state-of-play regarding what many consider to be a new ‘era’ of cancer treatment.

‘Breakthrough’ drugs

But first, let’s look at the two drugs approved today. Both are examples of what are often described as ‘targeted’ treatments, in that they’re designed to block the action of specific molecules involved in cancer. But despite both being approved for advanced melanoma, they work in completely different ways.

The first, vemurafenib (also known by its brand name, Zelboraf), we’ve blogged about extensively before, and we’re proud that our research paved the way for its discovery. It’s considered to be a stand-out example of how tracking down faulty genes inside cancer cells leads directly to new treatments. We’ve already covered the story of its discovery in detail (see this blog post for the full picture), but as the animation below shows, it targets cancer cells that contain a particular fault in a gene called BRAF.

This fault causes cells to produce high levels of a mutant protein which, in turn, encourages the cancer cells to grow. Vemurafenib was precision-engineered to stick to a specific region of this mutant, preventing it from sending ‘grow’ signals.

Because of all this, melanoma patients can be tested for the presence of the faulty gene in their tumour. If the mutation is present – which it is in about half of all melanoma patients – it’s highly likely that they’ll respond to the drug.

And the response can be profound. Among some patients, their disease entirely disappears for weeks, and sometimes months. Doctors involved in the initial trials have spoken of it being a ‘miracle drug’, unlike anything they’ve seen before.

But here’s the kicker: in most patients, sooner or later the drug stops working and the disease comes back, sometimes even more aggressively. This is emphatically not a cure for melanoma. But it can offer relief from symptoms of the disease, and potentially extend life for weeks or months.

This phenomenon – of initial response followed by resistance and relapse – is a huge problem with the new generation of cancer drugs, but not an insurmountable one; we’ll come back to it later.

The second drug, ipilimumab (also called Yervoy), works in a completely different way. It doesn’t target the cancer at all – it targets a patient’s immune system.

Ipilimumab is an antibody, engineered to target a molecule called CTLA-4, found on the surface of immune cells called T-cells.

T-cells normally attack foreign cells in our bodies in a tightly controlled way. Years of research on the immune system led to the realisation that these attacks have an ‘off-switch’, to prevent damage to the rest of the body. CTLA-4 is that off-switch.

Crucially, it seems that certain cancers manage to cloak themselves in molecules that flip this off-switch, allowing them to hide from the body’s own defence mechanisms. Ipilimumab aims to block this from happening, allowing the immune system to destroy the tumour – at least in theory.

In practice, however, the drug only works well in about one in 10 patients. And just one patient in every 100 seems to benefit in a lasting way. And unlike vemurafenib – where there’s a gene test available – there’s currently no way to say with any certainty who will respond to ipilimumab, nor how well.

So we have two extremely expensive new drugs, both of which can have profound yet often short-lived effects, and for one we have very little idea about who will benefit. On top of this, both drugs can have significant side effects – rashes, skin problems, diarrhoea and, in the case of vemurafenib, non-melanoma skin cancers (which are unpleasant but easy to treat in most cases).

So why the optimism?

A new way of thinking

Both of these drugs, in their own way, encapsulate many of the difficulties we’re facing in cancer treatment at the moment. They’re expensive. Their effects are short lived. They don’t work in everyone. And they often need a complicated test to find out who to give them to.

So it’s no wonder that some have begun to question whether this strategy is the right one, or whether we’re just falling for the same hype that led US President Richard Nixon to proclaim, in the 1970s, that cancer “would be eradicated in a decade”. According to an article in The Guardian last weekend,

“Progress against cancer is stalling, with the latest targeted cancer drugs failing to live up to expectations and priced so high that treatment is becoming unaffordable even in rich countries, say experts… Only a few years ago, many cancer experts thought the arrival of targeted medicines, designed to attack the genetic makeup of the tumour, would make dramatic inroads into cancer deaths. That has not happened. Instead, these therapies have only bought a few extra months of life.”

And a report in The Independent, called ‘Are we losing the war on cancer’, stated:

“Most Americans thought a cure for cancer would be discovered within five years – emulating the technological success of landing a man on the Moon. But more than 40 years later, few experts talk of a single cure for the 200 known types of cancer. The optimism of the early 1970s has given way to the dogged determinism of a cancer community under siege from the growing global epidemic.”

At Cancer Research UK, we feel these articles paint an excessively gloomy picture (a sentiment echoed by others). Yes, there are still substantial hurdles to be overcome – we’ll discuss them below – but in our view, it’s never been a more exciting time for cancer research and treatment.

Biological barriers

The most serious obstacle to what we might call the ‘targeted age’ is something we alluded to above – cancer’s fearsome complexity.

Earlier this year, Cancer Research UK researchers showed how genetically diverse a single patient’s disease can be. No two samples from a kidney cancer patient’s primary tumour, or the secondaries, were the same.

And researchers are now finding evidence that even the individual cells that make up a tumour can be profoundly different from each other. It seems that tumours can ‘evolve’ as they grow, to become more and more disordered and complex.

This complexity goes a long way to explaining the limited success of the targeted drugs developed thus far. Cancers seem to evolve to evade treatment.

To illustrate this, consider a melanoma patient whose tumour has tested positive for the faulty BRAF gene targeted by vemurafenib. Since not all his or her tumour cells are identical, it’s possible that a small proportion of them don’t contain faulty BRAF, even though the test is positive. So the drug won’t affect these cells, and in time, they will continue to grow – and the cancer will come back.

Alternatively, suppose some of the cells – maybe just a handful among the millions that make up a tumour – carry a second gene fault alongside the BRAF mutation, allowing them to evade the drug’s effects. Again, most of the cancer cells will be eradicated, but their souped-up sisters can survive to grow back.

So how do you tackle an enemy that’s so hard to pin down? The answer, of course, is more research.

In the first scenario, the recurring tumour will be driven by something other than a faulty BRAF gene. Finding out what it is, and targeting that, should allow more treatment with a second drug. And hardly a week goes by without a new experimental drug, aimed at a particular mutation common in cancers, going into clinical trials. Once we have enough drugs like these, a cancer could theoretically be held at bay by identifying what makes it tick, and targeting that, each time it comes back.

In the second scenario, researchers are again making progress. For example, some of the key molecules that allow BRAF-driven cancer cells to evade vemurafenib’s effects have already been identified. Experimental drugs already exist to target these cells, for instance trametinib, which targets a protein called MEK. Combination trials are underway, and are showing promise.

In another example, vemurafenib was tested on bowel tumours that also bore the BRAF mutation. This didn’t work so well. But researchers have discovered why, and how to get around this – again it points to combining vemurafenib with another class of new drugs called EGFR inhibitors, which are already available.

The hope is that before too long, doctors will have a whole arsenal of targeted treatments, which they can give in different combinations – and alongside conventional chemotherapy – depending on the genetic make-up of a patient’s disease.

In this light, it’s hardly surprising that the single-drug regimes tested in recent years haven’t lived up to expectation – they almost certainly work much better in combination.

But this all needs to be proven in clinical trials. And if and when these trials do show which combinations work for which patients, we need to be able to pay for them.

Which brings us to the second set of barriers – the social, ethical and financial ones.

Social, ethical and financial barriers to targeted cancer therapy

As the story of vemurafenib and ipilimumab shows, there’s a gulf between what drugs companies expect (at least initially) and what health services can afford. Thankfully in this case a solution was reached, but it took nearly a year.

We urgently need a rethink of how this works, so patients aren’t left in limbo while regulators and pharma companies negotiate.

On the one hand, we do appreciate that the pharma industry needs to invest in new research and development, and is facing a squeeze as existing patents expire.

On the other, their marketing budgets have historically been very large, and the initial price they offer regulators is frequently too high for an immediate ‘yes’. We need to find a better way.

But aside from the crucial issue of cost – ever more pressing in these straitened times – there are other barriers, as discussed earlier this year in an excellent article New England Journal of Medicine article, ‘Preparing for Precision Medicine’. Its authors propose six key obstacles to be overcome:

1. Trial design and regulation. Testing new drugs in combination is going to require collaborations between drug companies which, historically, have been reluctant to work together. And setting up combination trials will also need to be done swiftly, with the minimum of bureaucratic fuss. We’ve been actively pushing for streamlined trial regulation on a UK and EU level for some time. On top of this, trials will be focused on smaller numbers of patients, and results will be expected quickly. Designing robust, ethical trials that can do this is no mean feat, but not impossible either.
2. Better classification of cancers. Cancers are classified according to where they originate in the body. But this 100-year-old system doesn’t take into account our new molecular knowledge. As noted above, some bowel cancers are driven by the same BRAF mutation that drives some melanomas. We need a system of recording cancers that recognises this new knowledge.
3. Improved electronic data management and patient records. To be able to realise fully the potential of molecular-targeted treatments, health services need to be able to link molecular test results to a patient’s record, and for these to be available across the NHS records system, not buried in a paper archive. Our Stratified Medicine Programme is aiming to pave the way here, but we’ve also seen some welcome signs that government is thinking seriously about this too.
4. Improved training of staff. These new systems and ways of working will need highly skilled doctors, nurses and pathology staff, who can operate and make best use of them.
5. Changing doctor/patient relationship. Over the next decade, many patients will be given targeted treatments in the context of clinical trials, as we carry out the experiments that could and should transform things. But trials require proper, informed consent, and detailed molecular information needs to be held and accessed in an ethical way.
6. Proof of success. Ultimately, this is going to be an expensive journey (at least initially), and to convince the public and the politicians that the future is ‘targeted’, we need to demonstrate some significant progress, and quickly.

Another issue, not mentioned in the article, is how to safely monitor a patient’s response to their treatment, and how their cancer reacts. Currently, this would mean doctors taking multiple tissue samples, potentially of the main tumour, but also where it comes back or spreads to. This isn’t ideal for the patient concerned. But again there’s hope – we’re trying to work out if a blood test to track a tumour’s DNA could be a replacement for repeated biopsies. (EDIT: They’ve recently shown that this is almost certainly possible – HS 04/13)

None of the above is impossible. But overcoming these barriers requires urgent and sustained action, from industry, government, researchers, the medical profession and others to hasten the dawn of the new era.

And finally…

Drug treatment for cancer tends to hog the headlines, for a whole range of reasons. But in all the excitement and hype about new blockbuster therapeutics, we mustn’t forget that – for all the undoubted promise – drug treatment for cancer isn’t the only show in town (although it’s possibly the most expensive).

We need to remember that almost all patients are operated on to remove their tumour, while 40 per cent of those who are cured receive radiotherapy.

Together, radiotherapy and surgery are extremely effective, and value for money. Very few patients would be cured without either of them. And yet they’re often forgotten in the rush to welcome new drugs, despite the latter not being ‘cures’ – at least not yet.

And we know that the UK’s radiotherapy services need serious investment if they’re to become world-class, while uptake of surgery around the UK is patchy.

So, while we rush to welcome the dawn of the new age of high-tech cancer drugs, we also need to make sure that there’s sufficient attention – and investment – given to all forms of cancer care.

The last decades have seen substantial improvements on many fronts in the war on cancer. The next decades offer the promise of many, many more – but it’ll be a while before we can say a new dawn has truly broken.

Henry