Feature: Leukaemia CSI

After identifying an interesting class of precancerous stem cells (pre-CSCs) found in the thymus, Dr Matthew McCormack from the Walter and Eliza Hall Institute (WEHI) and his team then looked at the gene expression profile of the pre-CSCs after isolating them from the thymus. Interestingly, they found a similar expression pattern to normal thymocytes, with only about 80 genes that were differentially expressed.

“This really highlighted an advantage of these pre-CSCs, because the number of gene expression changes from normal thymus cells is much lower than you would normally find with overt cancer stem cells, which can have up to thousands of mutations, and it is hard to determine which ones are key for driving the cancer.

“It is a bit like arriving at a crime scene a year after the fact – the clues might still be there, but too much else has happened to see what is key and what happened leading up to the crime.” As such, this finding enabled the team to really home in on the key cancer-causing pathways much quicker than expected.

“All of this work was done in the mouse models, and now we are extending the work using an in vitro model we have developed to study these pre-CSCs in more detail,” says McCormack. “We want to find the key pathways necessary for Lmo2 to turn these thymus cells bad.

“One important candidate seems to be a homeobox pathway. We know that Lmo2 activates a homeobox gene called Hhex, which is expressed very early in the life of the pre-CSCs.”

Further gene expression experiments in these cells confirmed that this Hhex gene seemed to be doing a lot of the work in inducing self-renewal in the thymus cells – one of the hallmarks of becoming a cancer stem cell – implying that it could be a critical downstream component of the Lmo2-induced gene expression program leading to cancer.

“So that is really one avenue down which we are extending that work,” he says. “Looking at the importance of pathways such as that involving Hhex in the maintenance of both pre-CSCs and overt tumours.”

The other aspect is looking at the implications of these cells for therapy. McCormack and Curtis realised very quickly that identifying the pre-CSCs could yield an important target for the development and testing of new treatments for T-ALL.

“Children with T-ALL currently receive extended therapy over two to three years in an attempt to stop a relapse. More targeted therapy on the thymus cells could reduce the length and toxicity of treatment and prevent relapse. So, this has also become a major focus of our work.

“There is some evidence that when you treat leukaemia, relapsed cancer often shows a different gene signature from that of the primary disease. That suggests that treated cancers might actually revert to this pre-cancerous state in some cases and then those cells, rather than overt cancer stem cells, give rise to the relapse, but with a different set of mutations – like going back to a sort of checkpoint.” Eradicating these pre-CSCs could therefore be a way to prevent relapse of leukaemia.

“Specifically, we are looking at their susceptibility to conventional cancer therapies (radiation and chemotherapies),” explains McCormack. “Then we are also trying to generate more targeted therapies to eliminate these cells because despite the known genetic heterogeneity, current therapies for most cases of acute leukaemia are essentially identical. A better understanding of how these chromosomal translocations cause leukaemia will provide rational approaches for identifying new therapeutic strategies.”

---P--- Targeting the cancer cell niche

The main avenue being taken by the team in Melbourne to develop therapies against these pre-CSCs is to target the specific niche within the thymus where these cells seem most comfortable. One of the important factors in this niche may be the generic and multi-functional Notch signalling pathway.

Notch is expressed by thymic epithelial cells and is required at several stages of T cell development in the thymus. Notch signalling is also dysregulated in many cancers including T-ALL.

“The pre-CSCs self-renew at a stage of T cell development that normally requires both Notch signals and growth factors, including stem cell factor and Interleukin-7. Therefore inhibiting each of these pathways is a potential therapeutic strategy to specifically eliminate these T-ALL-causing cells.”

Some pan-Notch inhibitors, called gamma-secretase inhibitors, are already in clinical trials for T-cell leukaemia, and they have reported efficacy in T-ALL cell lines and in xenograft models. However, these agents also have the adverse side effect of gut toxicity, limiting their use.

According to McCormack, there are more specific drugs now emerging that specifically target the Notch receptor molecule involved in T-cell leukaemia, and this same receptor may be critical for keeping the T-ALL pre-CSCs alive in the thymus.

“Our hope, then, is that specific therapies targeting cell niche molecules will be more effective against some cases of T-ALL. Currently we are testing a range of such drugs in our in vitro model and also in the lineage tracing model that we used originally to discover and characterise the pre-CSC population.”

Using this assay, the team will look specifically for the import of tracer-positive cells into the thymus following treatment as the readout of therapeutic efficacy of whatever drug they are testing. This would indicate that the thymus has regenerated from the normal bone marrow source and not from the abnormally programmed pre-CSCs.

McCormack is looking forward to his move to WEHI and foresees that it will be relatively seamless, having been affiliated already with the Cancer and Haematology division there over the past few years. “My research focus will be essentially the same and I will continue to collaborate with David Curtis, but I also plan to extend the work into two other areas,” he says.

Firstly, McCormack is keen to pursue some basic blood cell biology based on what he knows from the pre-CSC work. “We know that the stem cell pathways that are active in our T-ALL pre-CSCs are quite similar to pathways expressed in normal haematopoietic stem cells.

So we hope that studying these cells could also tell us something more basic about self-renewal in normal stem cells – what are the requirements of these particular pathways in normal haematopoietic stem cells and are the same pathways required more widely.

“Down the track, I also want to model pre-CSCs in other leukaemias. We have shown them in T cell leukaemia, but that is not actually the disease with the most unmet clinical need. We will therefore try to develop models in B cell lymphoma and also in acute myeloid leukaemia using similar cell-fate tracing strategies to try to identify the presence of the same type of pre-cancerous stem cell.”

If successful, that original analysis of the sick mice from Cambridge could have a huge impact on patients with blood cancers everywhere.