Scientists from the University of East Anglia, along with colleagues in the US, have published a paper today in the scientific journal Nature showing that a common arthritis drug could potentially be used to treat malignant melanoma – the most dangerous form of skin cancer.
But although this is an interesting and potentially important finding, the expert view is one of cautious optimism rather than all-out excitement.
Here’s a short audio comment from Cancer Research UK-funded melanoma expert Professor Richard Marais from The Institute of Cancer Research:
As Professor Marais says,
“This work represents a very interesting new experimental approach to melanoma, but obviously there’s a long way to go before we can start using these sorts of combinations in patients. We need to do a lot more preclinical studies and a lot more studies in appropriate models.”
It needs to be stressed that the research has only been done with animals (namely fish and mice) and is still a few years away from being available for cancer patients – and only if further tests prove that the drug is actually beneficial for people with melanoma.
And, based on this paper, there’s not nearly enough evidence to suggest that melanoma patients should take leflunomide at the moment to treat their disease.
Let’s look at the story in more detail.
From tiny fish to tumours
Melanoma rates are rising rapidly in the UK, thanks mainly to over-exposure to the sun and sunbeds. Although the chances of surviving the disease are good if it’s detected early, once it’s started to spread the outlook changes dramatically and there are very few effective treatment options available.
Thankfully, in recent years, there have been several important developments in melanoma research, from drugs designed to target the genetic faults in the tumour to using the power of a patient’s own immune system to attack the disease. But there’s still an urgent need to find new drugs, and to understand how melanoma cells originate and grow.
This week’s edition of Nature carries two papers – both led by Professor Leonard Zon at the Children’s Hospital in Boston – that shed new light on the genetic mistakes that drive melanoma, and how we can use this knowledge to find more effective treatments for the disease.
In these papers, the scientists describe how they recruited a new helper in the search for melanoma drugs –zebrafish. Although tumours and fish may seem very different, there are actually a number of important similarities that make them a very useful model for cancer research.
In the first paper, the scientists use zebrafish to show that two overactive proteins – SETDB1 and BRAF (of which we’ll hear a lot more later) – co-operate to drive melanoma. This is, in itself, an important step forward in our understanding of the faulty genes that lie at the heart of cancer, but it’s not the paper that’s hogged the headlines.
That distinction falls to the second paper, which reveals that a common arthritis drug may help to treat melanoma. It sound exciting, but what did the scientists actually do?
Modelling skin cancer
In humans, melanoma starts from specialised pigment cells in our skin called melanocytes. While we’re growing and developing in the womb, these melanocytes first appear in a region of the embryo known as the ‘neural crest’, before spreading out to our still-developing skin.
Thanks to evolution and our shared common ancestor, the neural crest isn’t just found in human embryos – it turns up in many other animals, including zebrafish.
As the first step in their research, the scientists, led by Dr Richard White from Harvard Medical School, studied fish carrying an overactive version of BRAF – a gene that Cancer Research UK scientists have discovered is overactive in more than half of all melanomas. These fish develop melanomas after a few months, which behave in a similar way to human tumours.
When the researchers looked at the pattern of gene activity in the fishes’ tumours, they found that it was very similar to that of healthy neural crest cells in the developing fish embryo. Intriguingly, this also mirrored the pattern of gene activity in samples from human melanomas, suggesting that these tumours are somehow reverting to a much earlier ‘neural crest-like’ state.
From genes to drugs
Given these similarities, the researchers figured that chemicals that can affect the development of the zebrafish’s neural crest would also stop the growth of melanoma. So they set about looking at the effects of 2,000 different chemicals on the zebrafish.
The best ‘hit’ was NSC210627, a chemical that blocks the activity of a protein called dihydroorotate dehydrogenase (or DHODH for short) which is involved in making the building blocks of DNA.
The interesting – and potentially exciting – thing about DHODH is that it’s already well-known to clinical researchers: it’s the target of an existing arthritis drug called leflunomide. Further experiments showed that leflunomide also affected neural crest development in zebrafish.
This gave researchers the evidence they needed to test the effect of leflunomide on human melanoma cells in the lab. As suspected, they found that it could slow their growth. Although this was in itself a promising result, the scientists wondered if they could get a more impressive effect by combining leflunomide with another drug.
The right combination
The researchers carried out their initial experiments using zebrafish carrying an overactive version of the BRAF gene, because – as we’ve already mentioned – around half of all melanomas have overactive BRAF. A number of drugs are currently being developed and tested that target this faulty protein, including a drug called PLX4032, which has shown “unprecedented” results in recent small-scale clinical trials and is currently being tested in large-scale, phase 3 trials.
To test whether blocking overactive BRAF could enhance the effect of leflunomide, the scientists tested it together with a near-identical drug to PLX4032 (called PLX4720) on mice carrying melanoma tumours.
After 12 days – an admittedly short time – the team found that the drug combination cut tumour growth by around half, compared with leflunomide alone. Although the result sounds impressive, it should be stressed that the tumours still showed some growth even over just a few days with the drug combination, and although this was less than with either drug alone, it certainly couldn’t be described as a “cure”.
Because leflunomide is already used to treat arthritis, and PLX4032 is already being tested in cancer patients, this means that it should be relatively quick to start clinical trials of the two together.
But combining drugs together – even if they are effective and safe on their own – can cause serious side effects. So based on the evidence presented in this paper, there’s still a fair bit more lab work to be done before we know whether the combination will be safe and effective in patients. For a start, it would be good to see the results of drug tests lasting longer than a fortnight.
Furthermore, there’s no guarantee that results of trials of the drugs combination will be positive, or more effective than other new treatments that are coming through the pipeline. And it takes time to set up and run even small-scale clinical trials. So at the least, it could be several years before we see the results of clinical trials of the two drugs together.
However, as Professor Marais pointed out in his comment earlier, there is a case for “cautious optimism” here.
If – and at the moment it’s still a big “if” – leflunomide can help to treat melanoma in patients, particularly in combination with BRAF-blocking drugs like PLX4032, then it could bring hope to people with the disease in future. But, as we’ve said already, these results don’t provide nearly enough evidence to suggest that patients currently suffering from the disease should take leflunomide.
Also on the positive side, the paper takes us another step towards to concept of “personalised” treatment for cancer. Increasingly, researchers are developing drugs that target specific gene faults in tumours, including BRAF-blocking drugs such as PLX4032. This means that once a doctor has genetically tested an individual person’s tumour, they can give them drugs that are most likely to work based on the genetic makeup of their cancer, rather than making ‘broader brush’ treatment decisions based solely on the type of cancer the patient has.
And finally, findings confirm that there’s probably a lot of progress to be made by researching different combinations of drugs for treating cancer rather than focusing on individual drugs, particularly when those combinations are created through investigating the genetic faults that drive the disease.
Although there’s still a lot more work to be done in the case of leflunomide and PLX4032, these results are another step forward in the global effort to beat cancer through painstaking and dedicated scientific research.
This research was funded by the Howard Hughes Medical Institute, the National Cancer Institute, Aid for Cancer Research, the American Society for Clinical Oncology, the National Institute of Arthritis and Musculoskeletal and Skin Diseases, and the Biotechnology and Biological Sciences Research Council.
Ceol, C. et al (2011). The histone methyltransferase SETDB1 is recurrently amplified in melanoma and accelerates its onset Nature, 471 (7339), 513-517 DOI: 10.1038/nature09806
White, R. et al (2011). DHODH modulates transcriptional elongation in the neural crest and melanoma Nature, 471 (7339), 518-522 DOI: 10.1038/nature09882