EACR 2016

'Rewiring' tumour susceptibility

Dr Michael Yaffe - MIT, Cambridge, USA

What I plan to present is some data on two aspects of cancer therapy. When we think about how tumours respond to therapy it depends on two types of rewiring, one we call dynamic rewiring and one we call static rewiring. Dynamic rewiring is the way that tumour cells respond to therapy in a very acute, quick manner and it’s usually the way you think of tumour cells developing resistance to certain types of treatments. So, for example, a big treatment of melanoma is to use a BRAF inhibitor and that works well for a couple of weeks or months and then the tumour cells develop resistance and that ability to rapidly acquire resistance after treatment is what we call dynamic rewiring. Among many other things one of the things that my lab has been interested in is whether you could use dynamic rewiring therapeutically, that is instead of giving tumour cells a drug that they then rewire to become resistant to, could you treat tumour cells in a certain way that would then make them become sensitive to some second treatment. So I’m going to show data, some of which is published and some of which isn’t, that for triple negative breast cancer and non-small cell lung cancer if you treat these tumours with two drugs, both of which are clinically approved but have never been used in quite this way, you can make them very sensitive and kill them. The treatment involves first treating them with an EGF receptor inhibitor and waiting and if you wait somewhere between 8 and 48 hours those cells will dynamically rewire their death pathways and then you can treat them with a conventional DNA damaging agent like doxorubicin. You’ll kill about 500% more tumour cells than you will if you just give the two drugs at the same time; that’s because of this dynamic rewiring. The treatment with the EGF receptor inhibitor for a long period of time, that is somewhere between 8 and 48 hours, causes them to rewire their death pathways and as a result a treatment that only kills a fraction of the cells now kills a much larger fraction.

I’ll show some unpublished data that this approach which, as I said, works in triple negative breast cancers, a subset of triple negative breast cancers and a subset of non-small cell lung cancers, can also be used to treat head and neck tumours, head and neck cancers. In this case what works best seems to be a combination of inhibiting the EGF receptor and the FGF receptor and then again waiting between 4 and 48 hours and then treating the cells with either cisplatinum or doxorubicin. Again we think it works the same way – you’re unmasking a death pathway that is otherwise masked but it requires that dynamic rewiring and that time stagger between when you inhibit the growth factor receptor and then when you induce the DNA damage. So that’s dynamic rewiring.

There’s another thing which is called static rewiring. Static rewiring is how we think tumour cells arise in the first place, that is they lose a tumour suppressor gene or they gain an oncogene where they do something that now changes their baseline wiring and promotes the development of tumours. One of the things that has been well-known is that many tumour cells are defective in p53 and p53 function, the function of the p53 gene. As a consequence of that loss of p53 they can’t respond normally to DNA damage. What our lab showed was that those cells have now recruited through static rewiring an alternative pathway that they use to survive DNA damage. This pathway is a pathway that normal cells don’t need to survive DNA damage. Normal cells use this pathway to respond to various types of cell stress; tumour cells use it for cell stress but they also require it to survive DNA damage. As a result, if you inhibit this pathway in tumour cells you can enhance your ability to kill them with things like doxorubicin, cisplatinum, irinotecan, camptothecin, a wide variety of drugs. In contrast, because normal cells don’t require this pathway, inhibiting this pathway in normal cells does not enhance their sensitisation so what you’ve done is you’ve increased the therapeutic window by which you can kill tumour cells with chemotherapy without increasing the damage to normal cells. This pathway goes through the p38 MAP kinase pathway and involves one critical kinase, MAPKAP kinase 2, which we call MK2. So MK2 is an excellent target to sensitise p53 defective tumour cells to combination chemotherapy.

Is static rewiring a similar concept to synthetic lethality?

Absolutely. It is synthetic; if you target MK2 it is synthetic lethality. You’re inhibiting a pathway and then activating another pathway and those two pathways together result in… the inhibition of one pathway and the activation of another work together in a synthetic way to increase tumour cell killing.

In dynamic rewiring are you trying to control the way cancer behaves?

That’s our hope. The interesting thing about the type of dynamic rewiring that we’ve been able to study so far is that it only seems to be effective if you can figure out what the driver is for that type of cancer. So in the 40% of triple negative breast cancers that respond to this type of therapy and in the subset of non-small cell lung cancers that respond, that’s because those cells in the baseline state are being driven by the EGF receptor. As a result, if you inhibit that now you can rewire them dynamically to make them sensitive to other agents. The 60% of triple negative breast cancers that aren’t being driven by the EGF receptor don’t show synergy when you block the EGF receptor together with chemotherapy and, as a result, it means that you have to figure out in order to use this time-staggered dynamic rewiring approach what it is that’s driving the tumour in the first place. If you look at HER2 driven breast cancers you might infer that I bet HER2 is doing the driving and in fact we’ve shown now that if you block the HER2 oncogene you can dynamically rewire that subset of breast cancer cells to make them more sensitive to chemotherapy. Similarly, with head and neck cancer the reason that you need to inhibit both the EGF receptor and the FGF receptor for a subset of those tumours is because if you inhibit the EGF receptor alone the FGF receptor can take over as a tumour driver. In those cells that use two different tumour drivers you need to inhibit both of those. The real clinical limitation for us at the moment is being able to identify which subset of patients have tumours that are being driven by which particular oncogene and for the EGF receptor you can’t use simple techniques like gene expression analysis or proteomics to figure out which lines or which tumours are being driven, you need some sort of a dynamic measurement. In our hands that turns out to be immunohistochemistry with a phospho antibody against the EGF receptor but translating that into a practical therapeutically useful biomarker is a little challenging.

What are the processes of dynamic rewiring?

I can tell you that it appears that with dynamic rewiring in our hands, with the approach we’ve used with the EGF receptor inhibitor and the FGF receptor inhibitor, it seems that that cancer driver is turning on a Ras and Myc driven gene expression signature. In order to wean cells off of that Ras and Myc oncogene signature you need to inhibit the receptor for somewhere between 4 to 48 hours, that’s that time window. Somehow doing that affects the ability of a pathway that involves the death caspase, caspase 8, to become activated. So that as long as the cells are driven by this Ras and Myc gene signature caspase 8 is supressed; when you wean cells off of this dependency on Ras and Myc then this second death pathway that goes through caspase 8 reawakens so the DNA damage can kill the cells not simply through caspase 9 and 3, a well-recognised mechanism of cell death, but somehow can also activate caspase 8. So you get the combined effects of caspase 9 and 3 and caspase 8 and that combination of death caspases seems to be primarily responsible for the enhanced death.