Q&A with Ross Levine: From Biological Insights to Better Treatments for Myeloid Neoplasms

Over the past two decades, advances in understanding the pathogenesis and therapeutic vulnerabilities of myeloid cancers have led to increasingly precise and rational treatment strategies. Few scientists have influenced this landscape as deeply as Ross Levine, MD, whose work has defined key genetic drivers of myeloproliferative neoplasms (MPN) and acute myeloid leukemia (AML), and helped translate those discoveries into transformative clinical approaches. “We now have a strong understanding of genetic events in blood cancers,” he notes. “That gives us a roadmap—and a lot of work to do.” In this conversation with Blood Cancer Discovery’s editor Tanya Bondar, Dr. Levine reflects on the evolution of the field and the scientific priorities that will shape the next era of blood cancer research as he joins the journal as one of its Editors-in-Chief.What were the major influences that shaped your early career trajectory?I started my career planning for a career as a clinically focused investigator, and it was only when I was exposed to research and then to oncology that I realized that by spending my time in the lab, I might be able to contribute substantively in the longer term. I saw the need to understand the disease as a first principle and to apply that understanding as a second principle. That is where I have put the focus of my career.And what attracted you specifically to myeloid cancers?For me, it was taking care of people with leukemia at Dana-Farber during the first year of my fellowship. It was fulfilling, but also challenging, because there were no new diagnostics or therapeutics at that time. It was that clinical interaction and the realization that science was desperately needed that led to my decision.Did that experience also shape your style of research?I have a pretty open mind about how one follows science. My core interest is in trying to improve outcomes for patients with leukemia. When a new trainee joins the lab, we start with that. Sometimes that takes us into more basic research. When Aaron Vinny joined the lab, for example, we chose to investigate the biology of cohesin mutations and got interested in chromosomal looping and, fundamentally, how chromatin is regulated. Other folks in the lab have been more focused on medicines that are going into patients and what those medicines do and teach us about cancer biology. More than anything, I view our lab as a leukemia lab but am willing to go wherever the data take us. That is important for me, and I think important for Blood Cancer Discovery as well—to remain open-minded to anything of importance and high quality that our readers want to learn about. I believe there has to be a “big tent” with room for basic mechanistic work applicable to blood cancers, all the way to studies done directly in patients.Do you think this open-minded philosophy helps with training independent investigators?My view is that it takes people with different skills, roles, and interests to collaborate to achieve impact. The idea that the only path is to become an academic principal investigator is not true. I hope to provide an example and to offer opportunities for those who want that path, but I am equally thrilled by people who have come through the lab and go on to do other great things. I have been fortunate to work with trainees who have had exceptional academic success, like Omar Abdel-Wahab, and many others who have gone into industry or even healthcare investing. I am very proud of all of them. The most important thing is for people to understand that it is about what they want to accomplish, not about what is expected of them. It is, and always should be, fun! When I need inspiration, I walk into the lab and talk to people about their ideas and data. That is one of the great privileges of the job.In the beginning of your career, what helped you most in achieving your first discoveries?I was very fortunate to train with Gary Gilliland at a moment when he led an incredibly productive, vibrant lab that was making discoveries in everything from identifying new leukemia genes to figuring out how to target them. And so, I had a life-changing opportunity to work with an incredible mentor and role model. The lab was full of exceptional people, like Brian Huntley, Zuzana Tothova, Jan Cools, Thomas Mercher, Stefan Fröhling, and many others who are now world leaders.Working in a lab with people who had the same interests, focus, and commitment, and who were also kind and collaborative, was a special experience. I wish everyone could have the opportunity to train in such a stimulating and supportive environment. I look back at it as the single most important factor that allowed me to become the scientist I am.What were the challenges in discovering JAK–STAT pathway dysregulation as a central driver of myeloid neoplasms?A number of groups used complementary approaches to identify JAK2 and other driver mutations, so it was very much a team effort. When I was a postdoc in Gary Gilliland’s lab, he wanted to bring new tools into the lab to discover mutations. Some people pointed out that this would require new collaborations and new platforms and questioned how we were going to obtain samples and where we would sequence them. I was a little bit more open-minded and also naïve—so when he said this would be exciting, I felt the best way to make a difference was to try something different, even if high-risk.Another underappreciated point is that this was not something any lab could do at the time. So, it was also critical that this initiative occurred just as the Broad Institute was being established, particularly its cancer resequencing effort. Gary and I were fortunate to connect with Matthew Meyerson and Bill Sellers, who were setting up the first large-scale resequencing efforts in Boston—along with Washington University and the Sanger Institute—to be one of the three globally to really invest in cancer genomics and gene discovery.Looking back, many things aligned. At the time, we were figuring things out as we went. Matthew Meyerson and Bill Sellers deserve a lot of credit. Everything from large-scale resequencing to platform building was a team effort led by their vision.For me, this experience was incredibly important and instructive in learning how to do science, pursue new ideas, and work with others. Many things went right, and I was certainly the right person in the right place at the right time.What is resequencing?It is a term that is no longer used much. In the early 2000s, just after the Human Genome Project, “primary sequencing” referred to producing the first sequence of a genome. There was extensive work sequencing a spectrum of different organisms. Those efforts and platforms were built to sequence, initially, the human genome and then other genomes. “Resequencing” meant that, once you had the sequence, you would examine it across many individuals. Today, essentially everything we do is resequencing, but at that time, it was a meaningful distinction. There were groups discovering genomes de novo and groups analyzing variation across individuals and diseases.This helps to appreciate things we take for granted now.When I was a medical student in 1995, during my first lab experience, the idea that you could query a gene rather than mapping introns and exons manually was new.We now take for granted that we can look up gene structures instantly in our computer, but until the mid-1990s, you had to map your gene manually. As with everything in science, the things that are trivial now were not trivial then—and things that are difficult now will be trivial in 10 years.What was the journey like, translating mechanistic insights to targeted therapies for MPNs and AML?When we and others discovered JAK2 mutations in 2005 (1), the therapeutic implications were clear because it was a kinase and an activating mutation. Gary’s approach was to get as many people as possible working on it because we did not know who would ultimately develop a JAK therapeutic. Over the next few years, there were probably between 10 and 20 programs and eight clinical drugs emerging from those early efforts. In 2011, Incyte’s ruxolitinib was approved, and eventually three more drugs followed. It took a long time.This teaches us that drug development is rarely linear. A lot of unpredictable things happen along the way. The JAK2 story is not over. Current drugs are not good enough. I have spent a large part of my career, and even founded a company, trying to develop better medicines. That effort continues. My hope is that in the next 2 to 3 years, we will see a new wave of MPN therapeutics.But it is not just about JAK2. When I began my career, I wanted to find other mutations with previously unappreciated roles in leukemogenesis, understand them, and ultimately use that to inform diagnostics and therapeutics. That has been fun and interesting although not always straightforward. It is great when you have a mutation like JAK2 or IDH for which the discovery of the mutant gene or its function leads to a clear therapeutic strategy. But even then, it is not straightforward. As of now, we still do not have medicines for most mutations in leukemia and other cancers, and there is much work to do.“We still do not have medicines for most mutations in leukemia and other cancers, and there is much work to do.”Which targets seem most promising in the near term?I am appreciative of all the work already done but even more invigorated to ask how the field can get to the next level. Can we deliver a whole new set of therapeutics—not just small molecules? This could include antibodies and biologics, because mutations could be associated with surface proteins, or strategies aimed at DNA–RNA interactions. We now have a strong understanding of genetic events in blood cancers, particularly leukemias and lymphomas, and their roles in disease pathogenesis. That gives us a roadmap—and a lot of work to do. Although I am not a lymphoma expert, it is clear that therapeutics disrupting BCL6, for example, could be valuable. I am excited about these directions. They are not straightforward, but that is what makes them interesting. The most important value is what patients need: more effective and better-tolerated medicines.Are genetic drivers of leukemia generally good therapeutic targets? For example, once leukemia develops, does it remain addicted to the altered DNMT3A function?This is a core question in our lab and in others. We do not know whether initiating events are required throughout leukemogenesis and for maintenance of the clone. It is an important question because drug development is challenging, and determining which mechanistic events are truly therapeutically actionable is essential.In which areas is leukemia research currently leading the way? Could you reflect on the value of disease-specific versus pan-cancer approaches?In subtype-specific biology and therapeutics, leukemia is ahead of almost every human cancer. We have more drugs for more subtypes, and much of the current effort is figuring out how to bring them into the first line. For example, menin inhibitors were recently approved for MLL-rearranged and NPM1-mutated disease, and a randomized trial is opening to test venetoclax and azacitidine with or without a menin inhibitor in first-line treatment. That will be an important study. If we can integrate these targeted therapies into the first line, in which they are not expected to work alone, it could be transformative—similar to how adding BCR–ABL inhibitors to chemotherapy changed the outcome for patients with Philadelphia chromosome–positive B-cell acute lymphoblastic leukemia (ALL). The challenge is that these studies take time, and we must find ways to accelerate them, including working with regulatory agencies.The myeloid leukemia field has struggled with identifying and leveraging lineage-specific dependencies. CAR-T cells, antibodies, and BiTE therapies are used widely in ALL, lymphoma, and myeloma, but myeloid leukemia is behind. One reason is that myeloid antigens often share expression with normal stem and progenitor cells, which limits therapeutic windows. By contrast, targeting B cells or plasma cells is feasible although not without toxicity. You cannot do that if the target (e.g., CD123) is present on stem cells.My hope is that the field will identify new targets, or combinations of targets, that are more specific to leukemia. Recent work (2) showed that certain splicing mutations may create novel, mislocalized antigens. If these are disease-specific, strategies such as engineered antigen receptors may eventually exploit them. I am optimistic, but myeloid leukemia is behind lymphoid malignancies in this respect.Is leukemia at the leading edge of understanding cancer initiation through clonal hematopoiesis?On one level, it is a robust, exciting field with a lot of data and excellent mechanistic and translational work. Clinical trials are emerging. But conceptually, it is not entirely different from monoclonal B-cell lymphocytosis, monoclonal gammopathy of undertermined significance, atypical nevi, or colon polyps. The difference is access—peripheral blood is easy to sample, and many groups are studying it, so the field progresses quickly.The challenge in a fast-moving field is balancing between doing what everyone else is doing and choosing unique questions. One principle I follow is that senior investigators should try not to always do the same experiments as everyone else; rather, they should pursue high-risk studies that are harder to do.How well characterized is the spectrum of clonal hematopoiesis genetic drivers?For common single-nucleotide drivers of early stem cell expansion, we have probably found most of them. But for copy-number changes and larger rearrangements, the necessary technologies are only now emerging. RNA modifications are another area in which discoveries are likely.A key study (3) showed that copy-number alterations exist in stem cells and may predispose to lymphoid malignancies. We are not yet measuring these changes with high sensitivity or specificity. The jury is still out on how often these types of alterations occur and whether other blood cancers have analogous early events.Another area is understanding how clonal hematopoiesis (CH)–derived myeloid-biased clones influence other malignancies. Many groups are studying CH in lymphoma and myeloma. We are studying CH interactions with solid tumors. We are still early in that journey. The concept that tumor evolution occurs alongside evolving microenvironments or macroenvironments is likely to be important over the next decade.Do you think CH has a physiologic counterpart?I think there is natural variation in fitness within the stem cell pool. As people age, having a small number of stem cells with increased self-renewal and fitness—so long as they are not transformed—may be advantageous. When something occurs in 20% to 40% of the aging population, it will have both beneficial and adverse consequences. Using deep sequencing, probably by age 30 or 40 years, we can see mutant stem cells in almost everyone. The difference is that in about a third of people, the clones expand to detectable levels; in the other two thirds, they remain latent. Understanding not only which mutations arise but also which clones expand, and with what consequences, is a central question.Do you envision clinical interventions for CH as a premalignant state?That is at the top of the list of things I hope to see in my career. Once people develop overt myelodysplastic syndrome (MDS) or AML, in many patients the malignancy does not respond to treatments at all or stops responding over time. If we want to reduce deaths from leukemia and other blood cancers, intervening earlier must be our aspirational goal.“If we want to reduce deaths from leukemia and other blood cancers, intervening earlier must be our aspirational goal.”What is your philosophy on exploratory research versus hypothesis testing?I think there is real tension there. Large-scale profiling studies are powerful and needed although expensive. But sometimes they are emphasized, through grants or journals, at the expense of following up on agnostic discovery to understand what these discoveries actually mean. I have always argued that these studies should generate hypotheses, which must then be rigorously tested, and we need room in high-quality journals for discovery science and critical, mechanistic follow-up work.What topics excite you outside your own research scope?I appreciate the interplay between genetics and immunology and thinking about DNA, RNA, RNA modifications, and protein regulation. I am excited about advances in delivering molecules—large or small—to specific places and tissues. And I am excited about data science tools. We need to incorporate these tools broadly and also encourage a generation of data science–focused researchers to work on blood cancers. These are exciting, exploding areas of great science.My own work remains focused on developing better JAK2 therapeutics, understanding AML and how to target it, and the idea of preventing malignant transformation by intercepting CH. One of my approaches is to work on an area for a long time until we feel that we have addressed the core questions. We usually pursue new areas if we believe that the long-term impact could be significant, even if the path is long.How far into the future do you think about questions in your current area?I tend to think in the time frames that a trainee might take to address a question—pragmatically, 3 to 5 years. But sometimes that is only one chapter in a much longer story. Over time, it becomes easier to think longer term.What is your vision for Blood Cancer Discovery?Having been involved with different journals, my view is always to help make the journal a platform for the best and most exciting science and to allow a broad range of investigators to have a chance to showcase their work. Blood Cancer Discovery is at a great moment: it has established itself as an excellent platform for high-quality science in the field, and the amount of impactful work being produced is expanding. This creates an opportunity for the journal to grow and increase its impact, right as the field is growing. The two are intertwined, and that is what makes this exciting. That is why I am excited to join the team.