EACR 2016
Mouse models of lung cancer
Dr Anton Berns - Netherlands Cancer Institute, Amsterdam, Netherlands
Our work on mouse models, so I’m focused, my lab is focused, on cancer models for lung or thoracic cancers, I should say. So we are interested in some of the typical subtypes that we find so in the lung cancers you have more or less three major ones which is the adenocarcinomas, the squamous cell carcinomas and the small cell lung cancer. We do most of the work on small cell and the squamous cell carcinomas and so what we are interested in is to generate models in a mouse that really closely mimic what we see in humans because in a mouse we can study how the tumour starts, how it evolves over time. We would like to know from what is the cell of origin of these tumours so that we really know how this tumour development works because in humans we always are encountering patients that have already, usually, advanced tumours. That’s the reason why the prognosis for lung cancer patients is so bad. So we would like to know what are the initial phases, does that teach us new markers that could be used for early diagnosis, does it help us to design combination therapies that might be more successful? Because it’s very complicated to do that in advanced disease in humans. So we do a lot of, let’s say, quite basic cell biology or tumour biology on those tumours – how they develop, how heterogeneous they are, how they become metastatic and that we can study in a mouse very well and under very fine conditions.
What are the clinical benefits?
At the moment it’s too early. Of course we all hope that but the question is you have to understand the system. Of course the desire to help patients is enormous and I’m sure that every basic investigator, it would be great if he could see that what he did as research was benefitting patients. But at the same time just that desire is not enough, you have to understand the biology in order to do something useful and what we try to do is the things we see in a mouse we try to correlate, for instance, in humans. In other words, if we see abnormalities there then the question is do we see that in humans as well? If one applies or tries certain therapies in a mouse would that also work in patients? That’s what the stage… what we do a lot, like screens, so to say, to find out what might be the weak points of these mouse tumours that so much resemble the human tumours. If we find there are combinations, because we partly do that type of screens, we cannot all do that in mice so we do that also in cell lines derived from these mice. So once we have combinations then we will move it into the mouse models and if that would be successful that would be the moment that we would move it into human clinical trials.
But the other side effect might be, and we see some of that, that we find markers that have prognostic value, for example, or predictive value, so that might predict it’s responding to a particular therapy and so on. So we are also eager to look at those aspects because for some of the lung cancers it would really be great if you would have an easy test and also identify people at risk. Smokers are obviously at risk but the problem is early lesions, are they going to be malignant or not? It’s a major issue so you very easily if you don’t have really good biomarkers you end up with an enormous overtreatment and the examples are already there - overtreatment of breast cancer, overtreatment of prostate cancer and, interestingly, this overtreatment not necessarily results in more people surviving. So it just causes more misery so that’s something we have to be careful about.
