Elizabeth Coker, PhD Student in the Computational Biology Team
Telling someone that you’re doing a PhD in cancer research is a sure-fire way of stopping a conversation in its tracks. One moment you can be merrily chatting about last night’s TV, but as soon as someone finds out what you do for a living, their expression invariably becomes solemn. They then almost always ask this question – when do you think they’ll find a cure for cancer?
It’s true that research into cancers and how to treat them has made great advances over the past few decades: for example, research into cervical cancer has identified the significant role infection with the human papillomavirus (HPV) plays in developing the disease, and as a result of thisteenage girls in the UK are now routinely offered the HPV vaccine. My own establishment, The Institute of Cancer Research, London, has made huge contributions to the quest cures: for example, in recent years scientists at The Institute of Cancer Research (ICR) discovered abiraterone, a drug used to treat advanced metastatic prostate cancer, and uncovered some of the basic biology of cancer, including the discovery that mutated forms of the gene BRAF drive over half of all melanomas. However, the seemingly simple question “when will we find a cure for cancer?” has the following answer: it’s complicated.
When I say it’s complicated, I’m not trying to avoid answering the question. One of the major problems faced by biological research is that of complexity, a mathematical concept regarding the behaviour of systems with large numbers of elements within it. The Greek philosopher Aristotle sums up the problem nicely in his work Metaphysics: “The whole is greater than the sum of its parts.” Understanding how the components in a system work in isolation is often little help when trying to predict how the whole system will behave, because of the complex interactions that are present between the components. Cancer is arguably the most complex biological system of all.
Let me explain. Each cell within a tumour may have hundreds or thousands of genetic mutations. However, there is no one single mutation or chemical reaction that causes cancer: it is the combination of hundreds of molecular events and the failure of numerous biological ‘safety mechanisms’ that results in a tumour. A tumour may contain tens of thousands of cells, all of which will behave slightly differently to their neighbours. Between some neighbours there may be quite large differences - for example if one cell contains mutations that make it grow and multiply rapidly while another continues to grow at the same rate as a healthy cell. Other differences can be more subtle: when a cell divides in two, the two daughter cells may, by chance, produce very slightly different amounts of important chemical signals. This can cause noise in signalling pathways, which can also make the cells behave in subtly different manners.
Cells within a tumour will also interact with healthy cells in neighbouring tissues, bones, blood cells, connective tissues and so on. The shape and structure of a tumour can also affect how cells within the tumour, and hence the tumour itself, behave: cells can be compressed by other rapidly dividing cells, or cells at the middle of the tumour can be starved of oxygen and so have to activate new metabolic pathways, for example. There are likely to be hundreds of thousands of factors that act in combination to make a tumour behave the way it does, and while some are obvious, many are likely to be incredibly subtle and therefore incredibly difficult for scientists to detect.
All of the differences outlined above can affect how the cells within the tumour grow and how these cells will respond to drug treatment. It can therefore be extremely challenging to identify which molecular features are the most important to the cancer cells’ survival, and so are the best elements to target to kill the cells. These targets should also ideally be unique to the cancer to minimise any damage to healthy cells, which can cause side-effects such as hair loss.
In addition to this, one patient’s tumour may have an almost completely different set of mutations compared with that of another patient with cancer of the same type, and tumours are often highly heterogeneous, with different areas having different mutations and behaviours to others. Metastases can also behave extremely differently to the original tumour from which its cells have spread. It can therefore be almost impossible to generalise and predict how the system of the tumour will respond to treatment, and hence the best strategy to use to kill it.
However, thanks to advances in data collection techniques and computational data analysis, the scientific community is finally managing to overcome the hurdle of complexity of cancer. I strongly believe that in order to understand how the system of a tumour works, we must study the tumour as a whole in situ, not just individual pathways or mutations in cells grown in Petri dishes. This is now becoming possible thanks to the development of techniques such as high-throughput, affordable DNA sequencing. By collecting and analysing the ‘big data’ generated using these technologies (for example, ‘proteomics’ data sets with tens of thousands of measurements of protein levels across multiple sections of a tumour), we can begin to model the complex reactions that occur in cancers. Whereas previously research has been limited by the amount of data that can be collected and processed, we are now able to understand tumour behaviour and dynamics at an infinitely higher resolution than ever before. This so-called ‘systems biology’ approach is being used by many groups at the ICR, and is at the centre of my work for my own PhD project. In the next few decades it is likely that these large-scale analysis techniques will be regularly applied in clinical settings, as part of ‘personalised cancer medicine’, in which drugs and treatments tailored to the exact behaviour of the tumour will be used instead of a ‘one size fits all’ approach.
I hope it’s becoming clear why it’s so challenging to understand how cancers ‘work’. Complexity, which is part of cancer’s very nature, is something that even many scientists do not fully appreciate. I find complexity fascinating, and it’s what truly excites me about cancer biology. Finding cures for cancers is certainly not an easy task, but I am confident we will get there eventually as scientists embrace technology to unravel the complexity of cancer.
Image: Visual Complexity, by Casey Hussein Bisson / Creative Commons BY-NC-SA