There is much that excites U.S. physician/scientist, Kimberly Stegmaier, about advances in AML research. The Professor of Pediatrics at Harvard Medical School, Boston, was one of the international speakers at the Leukaemia Foundation-hosted New Directions in Leukaemia Research conference, in Brisbane in 2018. We spoke to her following her presentation, ‘Clinical translation of AML genomics’.

How did you become interested in leukaemia?

As a child I was always interested in cancer. When I was 12 and in a gifted program at school, I did a research project on paediatric leukaemia and wrote a fictional story about a child with leukaemia. In college, on a pre-med track, I vividly remember my excitement in figuring out the answer to a Cell Biology test question about receptor tyrosine kinase signaling in cancer. Moreover, when I secured a research fellowship at Harvard Medical School, I had an amazing year in the lab of a well-known leukaemia researcher and became hooked on science. I started a paediatrics residency and did an oncology rotation as early as I could. I absolutely loved it. It was then that I knew I would pursue a career in paediatric oncology with a focus on leukaemia.

What is it about AML that particularly interests you?

In medical school I worked on acute lymphoblastic leukaemia (ALL). However, when I did my clinical training in paediatric haematology/oncology, I was very struck by how toxic the therapy was for children with AML and how poorly we were doing treating these children. It was horrible. When I went back to the lab later in my fellowship training, I decided to work on AML. I felt it was a very strong unmet need, both for children and adults, and felt the work had potential for a broad impact on both paediatric and adult patients.

What’s the focus of research at the lab you direct at the Dana-Farber Cancer Institute?

Our primary focus is finding new therapeutic targets and ultimately new drugs for children with cancer. Half of my laboratory works on leukaemia while the other half works on paediatric solid tumors. We have been using both chemical screening and functional genomic approaches to identify new therapeutic targets in these cancers. Most recently, we have been using a

powerful new technology called CRISPR-Cas9 to systematically delete every gene in the human genome in leukaemia and other cancer cells to identify the Achilles heels of these cells. Our hope is that these Achilles heels will inform new drug targets for these cancers.

Have you found anything in particular that you are excited about?

We are excited about a number of new targets that we have discovered in the lab. The project that’s gone deepest is our work on SYK (an enzyme), which has been translated to clinical trials. This work has benefitted from the pharmaceutical industry’s interest in the enzyme as a target for autoimmune disorders and cancer. Two drugs are being actively tested in adults with AML, with some complete responses observed in patients, including with the SYK inhibitor entospletinib. Published data indicates that the mixed lineage leukaemia (MLL) rearranged AML group may be particularly sensitive to these molecules. That is very exciting because the MLL gene is involved in both AML and ALL in children and adults. Infants (children less than one) is one particularly high-risk patient group with MLL rearrangements. MLL rearrangements are particularly common in infants with leukaemia, and this leukaemia subset is particularly resistant to standard chemotherapy, so new therapeutic approaches are very much needed for these children. Our own data also suggests that patients with FLT3 mutations, another poor prognostic group, may also be sensitive to SYK inhibitors. It is exciting that these molecules may have potential for patients most in need of new therapies.

What technologies are you excited about?

Cutting-edge technologies, such as CRISPR and shRNA screening, enable us to find new drug targets more rapidly by systematically deleting or repressing genes in order to figure out the Achilles heels in these cancer cells. We couldn’t do that 10 years ago. The ability to incorporate information about AML mutations and functionally perturb every gene in a cell is very powerful. It gives me a lot of optimism that by combining these different technologies with new approaches in chemistry, we will make inroads into novel AML therapies more rapidly than before. There was a long gap in any FDA approvals for AML, but now three targeted therapies awere recently approved – that’s exciting. There’s still a ton of work to do. Rather than extending survival by a few months, we’re looking to cure patients; that’s our goal.

Are mutations the problem with AML?

If we take all cancers, rank order them by how many mutations are present in any one cancer, AML has amongst the least! Cancers such as melanoma or lung cancer tend to have a high mutational burden. AML, however, tends to have shockingly few mutations. What we do see, however, is heterogeneity (a great diversity) from patient-to-patient and even within an individual patient’s leukaemia. Not all of the leukaemia cells have the same mutations. There are cell-to-cell differences which can be a challenge. For example, in the case of FLT3-mutated AML, we know that not every leukaemia cell within a person’s AML has a FLT3 mutation. The acquisition of FLT3 mutations is a later event in the evolution of the leukaemia. So, you can potentially eliminate those FLT3-mutated AML cells with a FLT3 inhibitor drug, but there are frequently AML cells that lack the mutation. That’s a problem.

Another challenge is that many of the mutational events that occur in AML don’t make for easy drug targets. For example, a common event in AML is the presence of a cancer-promoting fusion oncoprotein. These fusions occur when there are breaks in chromosomes and then the chromosomes fuse together to generate abnormal genes encoding cancer-promoting proteins. Often these fusion oncoproteins involve a class of proteins called transcription factors. These DNA binding proteins have been notoriously difficult to make drugs against. In other cases, the mutations lead to an absence of the protein, such as is the case of cohesin complex mutations. It is really difficult to replace missing proteins.

What is also quite puzzling is that while many patients with AML will go into remission, the majority of adults with AML will relapse. Why? These patients don’t typically have new mutations in the AML cells at the time of relapse. There are hypotheses about the stem cell and how these cells may be more immune to the effects of chemotherapy because they are not dividing as much as the bulk of the leukaemia cells. Arguably, the answer to this question is still largely unknown. Our lab has been thinking about the right types of studies to understand the differences in those cells that lead to relapse for those that are effectively killed by chemotherapy drugs.

What is GSK-3 alpha?

A fundamental problem in AML is that the cells stay in a very immature state. One of our research goals is to figure out how to trigger the cells to mature. One of the targets that I discussed during the NDLR conference was glycogen synthase kinase 3 alpha (GSK-3 alpha). This is an enzyme involved in a number of important activities within cells. We were excited to discover this target several years ago as a candidate for promoting AML differentiation. However, we were limited in our studies because we didn’t have selective molecules to distinguish between the GSK-3 alpha and GSK-3 beta isoforms. By targeting both at the same time, you unfortunately also stabilise a protein called beta-catenin, which can promote leukaemic stem cells. Thus, we needed molecules to specifically only target GSK-3 alpha because selective targeting of only the alpha isoform does not stabilise beta-catenin. In order to accomplish this goal, our laboratory joined forces with a chemistry team at the Broad Institute to develop some important molecules that inhibit GSK-3 alpha. We have found that these compounds promote AML cell differentiation and also impair the leukaemia-initiating cells preventing the establishment of the leukaemia in models of AML tested in the lab.

Can you tell us more about the new GSK-3 alpha inhibitors?

The molecule that we have validated, BRD0705, is still a tool compound; it’s not a drug yet. The big goal of the project, which we have recently published in Science Translational Medicine, was to find chemicals (small molecules) that target the alpha and not the beta isoforms of GSK-3. In my opinion, BRD0705 is the best molecule we have today that does just that. The next step is to use medicinal chemistry to make that molecule a proper drug. Academic groups are good at finding lead molecules but many millions of dollars and a lot of effort is necessary to optimise a lead compound for delivery to human patients. BRD0705 has been licensed to a company, and that work is underway. The information we published included the compound structures so that chemists can synthesise the molecule themselves and study them too. That’s one of the beauties of sharing and having data in the public domain; it gives that information to the community broadly to expedite progress for the patients we treat.

How does your clinical patient care as a paediatric oncologist inform your discoveries in the lab?

In addition to running my laboratory, I take care of children with cancer, most typically leukaemia, at the Dana-Farber Cancer Institute and Boston Children’s Hospital. My clinical work is a constant reminder of why I’m doing what I’m doing in the laboratory. The bedside care is an inspiration like no other to work harder in the lab. These children are in urgent need of better therapies. Right now, I’m involved in leading a multi-institution study that’s right at the edge of the clinic and the lab. We are sequencing samples from patients across the U.S. with relapsed or refractory leukaemia or very high-risk subsets of newly diagnosed leukaemia. The goal of the study is to determine the feasibility of rapidly sequencing leukaemia promoting genes in individual patient samples and matching these mutations to targeted drugs. For a number of the children in the study, this type of sequencing has informed a change in their therapy.

In your own lab, what are you working on now that’s new, different and promising?

I am very excited about using the new and very powerful CRISPER technology to systematically define the Achilles heels in AML cells. Our lab is now aggressively validating candidate drug targets that have emerged from these genome-wide CRISPR screens.

What’s the one thing you’d like to achieve in your research career?

For me the Holy Grail would be that a discovery made in our lab has a strong impact on curing patients with cancer.

Anything else?

For me, there is nothing more gratifying than working with children with cancer. It is a real honour. It is truly remarkable now that I have been in this profession for many years, to see them grown up – adults, married with their own children and pursuing exciting careers. These are moments to treasure.

I am very excited about new opportunities in cancer discovery. It has been an amazing time in terms of advancements in technologies. From sequencing genomes to deleting genes one by one with CRISPR, I believe that these advances will enable discoveries that change how we treat patients with leukaemia. I am also excited about new classes of drugs that perturb the epigenome. And, of course, while not a focus of my own work, there is a lot of excitement about the promise of other new therapeutic approaches such as the application of immunotherapy.