Editor's Note: At the 2026 European Hematology Association (EHA) Annual Meeting, the association presented the "EHA Research Excellence Award" to Professor George Vassiliou from the University of Cambridge and the Wellcome Sanger Institute, in recognition of his pioneering contributions to the fields of biology, pathogenesis, and early intervention in myeloid malignancies. In the subsequent award lecture, Professor Vassiliou reviewed the landmark research conducted by his team over the past 15 years regarding the identification of driver genes in acute myeloid leukemia (AML), the evolutionary logic of clonal hematopoiesis (CH), and the development of clinical prediction tools. 01 NPM1 Mutation Mouse Models: Revealing the “Two-Hit” Requirement for Myeloid Leukemia Development
Professor Vassiliou’s academic exploration began with research into the most common subtype of acute myeloid leukemia (AML)—NPM1 mutations. During his collaboration with Professor Allan Bradley at the Sanger Institute, his team successfully constructed the world’s first NPM1 mutation mouse model (Cell Reports, 2011). The study observed that although the bone marrow progenitor and stem cells of NPM1-mutant mice exhibited significant proliferative advantages, only about 1/3 of the mice developed AML after a long latency period. This key phenomenon suggested that a single NPM1 mutation is insufficient to drive leukemia development; subsequent synergistic mutations must exist. To identify these “synergistic factors,” the team utilized a transposon-mediated genetic screening platform and successfully identified known driver genes such as FLT3 and NUP98, as well as a series of novel collaborative genes whose functions were unclear at the time. This laid a solid genetic foundation for the molecular subtyping of AML.
02 UTX/UTY Genes: Discovering the First Y-Chromosome-Specific Tumor Suppressor Gene
In the study of genetic differences in myeloid tumors, Professor Vassiliou’s team conducted an in-depth investigation into the X-linked gene UTX (a histone lysine demethylase), which is frequently mutated in AML, T-ALL, and various solid tumors. The study found that female mice lacking two copies of the UTX gene rapidly developed AML, but male mice lacking their only copy of the UTX gene did not. This puzzle was ultimately solved by identifying the homologous gene UTY on the Y chromosome. The study confirmed that although UTY lacks demethylase activity, it can prevent the occurrence of leukemia by recruiting tumor suppressor proteins to the genome. This discovery not only confirmed that UTY is the first confirmed Y-chromosome-specific tumor suppressor gene but also explained the gender differences in the incidence of myeloid tumors from an epigenetic perspective.
03 Translational Research Highlight: METTL3 Inhibitors Move from CRISPR Screening to the Clinic
To systematically explore therapeutic targets for myeloid tumors, Professor Vassiliou’s team conducted the world’s first genome-wide CRISPR-Cas9 screen in acute myeloid leukemia (Nature Genetics, 2017). The screen not only validated several known targets but also discovered a series of potential druggable targets, the most prominent of which was the RNA methyltransferase METTL3. The study revealed that METTL3, as an RNA modification gene, maintains the over-expression of key proteins required by AML cells. When METTL3 is inhibited, the survival rate of leukemia cells significantly decreases. Based on this mechanism, a small molecule inhibitor jointly developed by the team officially entered Phase I clinical trials in 2022. This marks a complete translational cycle from basic genetic screening to new drug development, providing a brand-new therapeutic path for refractory myeloid tumors.
04 Clonal Hematopoiesis (CH): Mapping the Dynamics from Health to Malignant Transformation
One of Professor Vassiliou’s most significant recent contributions is the in-depth analysis of clonal hematopoiesis (CH). Through retrospective deep sequencing of blood samples from over 500,000 participants in the EPIC prospective cohort, the team discovered that high-frequency, large-clone mutations in genes such as DNMT3A, TET2, and ASXL1 already existed in the blood of patients years before the diagnosis of AML (Nature, 2018). Subsequently, using big data from 450,000 individuals in the UK Biobank, the team further clarified the natural history of CH: • Age Interaction: The clonal expansion of different genes is significantly affected by age. For example, DNMT3A dominates in the middle-aged population around 50 years old, while splicing factor mutations such as SRSF2 often show significant clonal growth only at older ages. • Gender Bias: The study found that DNMT3A mutations are more common in females, while the splicing factor SRSF2 mutation is significantly biased toward males, which highly aligns with the distribution characteristics of myeloid tumor subtypes between the two genders. • Bidirectional Selection of Telomere Length: DNMT3A clones tend to expand in individuals with polygenically longer telomeres; conversely, splicing factor mutations (such as SRSF2) and PPM1D mutations show significant selective advantages in individuals with shorter telomeres.
05 MN-predict Tool and Precision Prevention: The Potential Chemical Intervention Value of Metformin
Based on multi-dimensional genetic data, Professor Vassiliou’s team developed a prediction tool called “MN-predict.” This tool integrates age, gender, mutation profile, clone size, and basic blood cell parameters to accurately predict an individual’s risk of developing AML, myelodysplastic syndromes (MDS), or myeloproliferative neoplasms (MPN) over the next 5 to 10 years. In the field of preventive medicine, the team made a breakthrough discovery: cells with DNMT3A mutations are highly dependent on the oxidative phosphorylation pathway. Through interdisciplinary collaboration, the study confirmed that the common anti-diabetic drug Metformin can effectively inhibit the growth of such clones. Retrospective data from the UK Biobank further validated this hypothesis: in populations taking Metformin over the long term, the incidence of DNMT3A-driven clonal hematopoiesis was significantly reduced. This discovery provides a highly cost-effective candidate for the early chemical prevention of myeloid malignancies.
[Summary and Outlook]
In concluding the lecture, Professor Vassiliou pointed out that with the establishment of CH clinics across the UK and Europe, hematology research has officially entered the era of “pre-leukemia” monitoring. Through high-sensitivity molecular testing, predictive algorithms, and low-toxicity intervention methods like Metformin, clinical practice is expected to shift from “treating late-stage tumors” to “preventing malignant transformation.” Professor George Vassiliou’s outstanding contributions lie not only in revealing genetic truths but also in his mapping of a future blueprint for the prevention and treatment of myeloid malignancies through cross-scale, large-scale population studies.
