What we study in my research team is how cancer begins in malignant brain tumours. So the idea is that this is a cancer that is essentially incurable; it’s a death sentence, its prognosis has not improved in thirty years. So the question is, since we have not yet developed good, effective therapies through all of the traditional available ways, might we be missing something? So the idea is to make mouse models of human GBM. How do we do this? Well, after a survey of many human GBMs we arrived at the provisional conclusion that three well-known cancer associated genes were prevalent in the formation of GBM, glioblastoma multiforme malignant brain cancer. So we used genetic engineering technologies to specifically create these mutations in the brains of mice. Lo and behold, with 100% penetrance we have a colony of mice that spontaneously develop malignant glioblastoma multiforme.
So what this now permits us to do, because we have mice that we know will develop cancer of the type that we want to study, is we can begin to ask questions that are impossible to ask in patients. We can look early before the mice become ill and ask the question what happens first? When? How? These kinds of questions lead to novel information and to novel ideas that then give us strategies that were before unavailable to try to better tackle these cancers.
So, for example, traditionally it was thought that malignant brain tumours arose in glia. Glia are one of two major subtypes of cells that constitute the brain, one is neurons, nerve cells, and the other is glia which means glue but in fact they’re more than glue. They’re important and they have all sorts of important functions in the brain but it was thought that they were the source of malignant brain tumours. Our evidence strongly suggests that a majority, if not all, malignant brain tumours normally arise in stem cells, a relatively newly discovered population of cells in the brain so that, therefore, the precursor to product relationship is between a stem cell and a cancer cell, not between some terminally differentiated specialised cell and a cancer cell. What this now permits us to do, because it was not envisioned before, is to ask how does a stem cell become a cancer cell, which is very different than how does an astrocyte become a cancer cell. So it’s these kinds of novel insights that we believe are going to lead us in new directions for understanding how cancer happens but also how to treat it.
Does this tie in with the work of others in looking at stem cells in cancer treatment?
I think it does. I think that this is a still controversial field and it’s controversial for a variety of reasons that don’t mean… controversy is not a bad thing, controversy means that people are thinking deeply about a question. In liquid tumours, that is to say tumours of the haematopoietic system or of the blood and immune system, the idea of stem cells leading to cancer is far more advanced and the reason is that that field is far more advanced in many ways and it’s also much easier to isolate cells in liquid than it is in a solid tumour mass where there are all kinds of things going on and one can’t even be sure which are the tumour cells and which aren’t. So this is where mouse models have greatly advanced the field. So ultimately we now have two powerful tools in cancer research today that were not available that long ago. One is physiologically relevant genetically engineered mouse models. What that means is that we can actually identify mutations in cancer, create them in a mouse and recapitulate the same cancer in a mouse. This gives us a strength of certainty that we’re actually making something that’s meaningful and that if we study it we will learn something about how it happens in humans.
The other is genomic technology which now permits us to sequence entire cancer genomes and to identify all of the possible mutations that are there. To be able to couple these two things is going to be incredibly synergistic and it’s going to be a great advancement for cancer research.
Where are you so far with this research?
What this has now permitted us to do in our small group in our small area is to take the genomic information and use it to create the appropriate mouse models then to identify in those mouse models that cancer isn’t coming from any old cell in the brain but from a very specific subset of stem cells in the brain. Then to figure out and develop the techniques to culture these cells readily and then to use them to screen for compounds that specifically kill these cells and not normal cells. The upshot of this is that we have great excitement for the future that this is going to be a strategy that would be markedly different from classical anti-cancer therapies. Classical anti-cancer therapies, also called chemotherapy, is a misnomer, the reason is that classical anti-cancer therapy does not distinguish between a cancer cell and a non-cancer cell; it distinguishes between a dividing cell and a non-dividing cell. It turns out that many normal cells in our body divide, that’s why chemotherapy has so many terrible side effects and particularly in those tissues that have many dividing cells like hair, gut, the blood system, the immune system. So it’s about killing the tumour without killing the patient. However, now that we understand the very nature of these tumour cells, we can actually devise screens that will truly discern between a cancer cell and a normal cell and not target cell division. And it’s promising; it’s early days but it’s very promising.
Why is GBM so lethal?
If we knew, maybe it wouldn’t be so lethal. But I would say that the two cancers that are the most lethal are GBM and pancreatic adenocarcinoma. These are among the many cancers that we’ve made the least progress in understanding. GBM is even more difficult to treat than pancreatic adenocarcinoma because at least if you get the pancreatic tumour out while it’s still in the pancreas you can actually cure the patient. In GBM these tumours are very disseminating, they’re very invasive, so you might imagine ice-cream with raisins in it and the raisins are the GBM. How do you get all the raisins out of the ice-cream? That’s why it makes it surgically intractable from very early times. That’s one of the reasons why it’s so malignant, that you can’t really ever get all of it.
But many of the other reasons why it’s malignant is because simply we don’t understand it enough and we hope that through using these genetically engineered mouse models we are gaining a far greater understanding and therefore will be able to more rationally attack it.
