PI3K-Like Protein Kinases
The role of the PI3K and TOR pathways in cancer
Dr David Fruman - University of California, Irvine, USA
The PI3 kinase mTOR signalling pathway really drives many of the classical and emerging hallmarks of cancer, as defined by Hanahan and Weinberg, so proliferation, survival, migration and invasion and metastasis, as well as emerging hallmarks like cancer metabolism and immune infiltration and inflammation. So all of these aspects involve the activation of PI3 kinase and mTOR by external factors, growth factors, as well as oncogenes.
How are these factors involved in leukaemogenesis?
If you think about leukaemia and lymphoma as many, many diseases, unlike a lot of the solid tumours you don’t often find, even among this broad range of diseases, you don’t often find activating mutations in the PI3 kinase genes themselves or mTOR. And p10 loss, which is the tumour suppressor that blocks this pathway, is fairly rare as well except in one specific T-cell leukaemia. Instead you find PI3 kinase and mTOR activated by two mechanisms, one is by specific oncogenes like the BCR-able oncogene or FLT3 but also by the microenvironment. So leukaemia cells grow in the bone marrow in association with stromal cells, lymphomas live in lymph nodes and spleen in association with nurse-like cells and in contact with macrophages and other cells. All of those cell-cell contact related interactions and local cytokines can activate the PI3 kinase mTOR pathway. So we like to view the PI3 kinase mTOR pathway in leukaemia as integrating signals from the oncogene as well as from the microenvironment.
Which components of the PI3 mTOR pathway are required for leukaemogenesis?
There are not very many mutations and activating mutations in PI3 kinase or a loss of p10 in leukaemias and lymphomas. However, it has been determined that the PI3 kinase delta isoform has a key role in the survival of B-cells in certain leukaemias and lymphomas. It’s not because of an activating mutation, it’s because the PI3 kinase delta isoform integrates signals from the microenvironment, from adhesion signals and survival signals, particularly in chronic lymphocytic leukaemia. A drug which is called idelalisib and is now marketed as Zydelig is a very selective inhibitor of PI3 kinase delta and has produced dramatic responses in a large percentage of patients with advanced CLL that is refractory to other treatments. It’s in the same pathway as ibrutinib, Imbruvica, which inhibits BTK which is the next step in the pathway. So patients with CLL as well as some patients with other indolent non-Hodgkin’s lymphomas are responding very well to inhibitors of PI3 kinase delta. Interestingly the mechanism of the response is not a direct killing of these cells but it causes their release from the lymph nodes or the spleen entering into the blood where they no longer get the survival signals and then eventually the cells die. So it’s an interesting response where initially the patients experience a large increase in leukaemia cells in the blood, which is kind of alarming to the doctors, but then over time they die and the lymph nodes shrink very quickly. So that’s a real excitement with this drug which is actually the first FDA approved PI3 kinase inhibitor, idelalisib.
There are some other leukaemias where PI3 kinase delta seems to play a role but there’s some redundancy with other isoforms. So in some T-cell leukaemias it appears as if dual targeting of PI3 kinase delta and PI3 kinase gamma may be required to achieve this sort of response. There are some other B-cell leukaemias and lymphomas as well where this kind of approach, where PI3 kinase delta plus PI3 kinase alpha may be useful. But at least this proves the concept that inhibiting PI3 kinase will interfere with the survival signals that leukaemia and lymphoma cells receive from the environment, which is a point that I was making earlier.
What is maintaining survival in cells that have lost this pathway?
One of the main ones in leukaemia and lymphoma cells is intrinsic to the mitochondria. So you have the pro-survival members of the BCL2 family which are often elevated in leukaemias and lymphomas, like BCL2 itself as well as MCL1. Then you have reduced expression of the proapoptotic members of the family like Bim and Puma and so forth. In some cases the loss of these proapoptotic proteins is controlled or silenced epigenetically. So the two main mechanisms of resistance are elevated expression by amplification of genes like BCL2 or epigenetic silencing of pro-death genes, either at the mitochondria or even death receptors and other extrinsic death factors.
Do you think that there is potential to target the pathway’s anti-cancer therapies?
I do. There’s a growing literature that if you combined inhibitors of the PI3 kinase mTOR pathway with direct antagonists of BCL2 family members such as ABT-199, ABT-263, that you can achieve synergistic killing of leukaemia and lymphoma cells. With some selectivity relative to normal lymphocytes, which is an important question, how do you specifically target the cancer cells. There’s also evidence that targeting the epigenome with HDAC inhibitors and potentially methylation pathway inhibitors that you can sensitise cells to these PI3 kinase inhibitors and mTOR inhibitors. There are many other mechanisms, for example if you’re thinking about an oncogene like BCR-able that activates the pathway in leukaemia, it’s also activating STAT signalling, it’s activating Ras, MAP kinase signalling, and those can maintain survival even when you block the PI3 kinase arm of the oncogenic signalling. So potentially targeting these other compensatory signalling pathways can give you an increased anti-cancer effect.
Is there any interaction with immunotherapy?
It is a very important question and that’s one that we’re very interested in my lab because if you’re really interested in developing small molecule targeted compounds against oncogenes, in today’s world you have to think about how they interact with immunotherapies. In many diseases immunotherapy is becoming the standard of care, whether we’re talking about antibody-based treatments like rituximab or Herceptin or checkpoint blockers or vaccines which are emerging as well. In the case of the antibody based therapeutics, one of the main mechanisms that they use, although not the only one, is optimisation of the cancer cell so that leads to direct killing by natural killer cells and monocytes. That activation of the killer cell is somewhat dependent on PI3 kinase. So there’s some concern that PI3 kinase inhibition might interfere with the efficacy of some of the antibody directed therapeutics.
Then, on the other hand, if you’re thinking about a cancer vaccine there’s evidence that using PI3 kinase inhibitors can actually stimulate innate immune responses and increase the activity of vaccine adjuvants like TLR agonists. Then there’s also some evidence that inhibiting specific PI3 kinase gene isoforms like PI3 kinase delta can change the balance of the T-cells in tumours so that they suppress the regulatory T-cells, the suppressor T-cells, and that frees up the more anti-tumour CD8 cells and CD4 cells to do their job. So it’s possible that you could enhance immunotherapies through targeting PI3 kinase but it’s also possible that you can block, interfere with, these and this is an area of ongoing investigation.
