Editor's Note: In the Presidential Session of the 2026 European Hematology Association (EHA) Annual Meeting, Prof. Lars Bullinger from Charité – Universitätsmedizin Berlin delivered a keynote speech titled "From gene profiles to therapy profiles - a quarter century of molecular hematology." Combining his personal research experience with global major scientific breakthroughs, Prof. Bullinger provided a profound analysis of the evolutionary path of hematologic malignancies, represented by Acute Myeloid Leukemia (AML), from morphological classification to molecular precision diagnosis and treatment.01 From Morphology to Molecular Cytogenetics: Early Foundation for Defining AML
The origin of hematology research dates back to the mid-19th century. Prof. Lars Bullinger first paid tribute to Rudolf Virchow (founder of cellular pathology) and Otto Pappenheim. The former identified the abnormal growth of white blood cells through a microscope and coined the term “leukemia,” while the latter proposed the hematopoietic stem cell theory, laying the biological foundation for the study of AML origins. The real molecular revolution began in the 1970s. In 1972, Prof. Janet Rowley first described the balanced chromosomal translocation t(8;21) in AML and subsequently revealed the BCR-ABL fusion gene produced by t(9;22). These discoveries marked the entry of hematologic malignancies into the era of cytogenetics. In 1997, Prof. John Dick proved the existence of the Cancer Stem Cell (CSC) model, further clarifying the root of AML heterogeneity—that leukemia cells have a hierarchical differentiation structure and originate from the transformation of hematopoietic stem cells.
02 The Microarray Era: Gene Expression Profiling Opens the Door to Molecular Subtyping
Around 2000, the emergence of DNA Microarray technology completely changed the way tumor profiles were mapped. Todd Golub’s team proved in 1999 that machine learning algorithms based on Gene Expression Profiling (GEP) could accurately distinguish between AML and Acute Lymphoblastic Leukemia (ALL). Prof. Bullinger reviewed his research at Stanford University and the University of Ulm: • Mining of Molecular Markers: Through GEP analysis of Cytogenetically Normal AML (CN-AML), the research team identified the upregulation of the HOX gene family and MEIS1, which are highly associated with prognosis. This was later confirmed to be closely linked to the NPM1 mutation discovered in 2005. • Prognostic Prediction Models: Early studies were already able to predict the presence of key mutations such as FLT3-ITD through GEP. Although limited by technical bandwidth at the time, and some characteristics could not fully explain all clinical differences, research in this stage successfully decomposed AML from a single disease into a collection of subtypes with different molecular characteristics.
03 NGS and TCGA: Resolving AML Heterogeneity under a Whole-Genome Perspective
2008 was a watershed year in hematology history. Timothy Ley’s team completed the first Whole Genome Sequencing (WGS) of a tumor genome using shotgun sequencing, and the subject was an AML patient. • Discovery of Key Mutations: In 2009, researchers identified the IDH1 mutation through WGS, initiating a surge in research on epigenetic modification abnormalities in AML. • Milestone of the TCGA Project: The TCGA AML study published in 2013 performed Whole Exome Sequencing (WES) and WGS on 200 samples. Data showed that although the mutation frequency in AML is lower than in solid tumors, its molecular heterogeneity is extremely high, involving pathways such as DNA methylation-related genes (DNMT3A, TET2, IDH1/2), splicing factors (SRSF2, U2AF1, ZRSR2), chromatin remodeling factors (ASXL1), and cohesin complex mutations (RAD21, STAG2). This study officially established the AML Mutational Landscape.
04 Clonal Evolution and Clonal Hematopoiesis of Indeterminate Potential (CHIP)
Advances in molecular technology have allowed researchers to observe AML in the dimension of time. • Clonal Evolution: Dynamic sequencing shows that leukemia clones evolve under chemotherapy pressure. Some sub-clones sensitive to chemotherapy are cleared, while resistant clones carrying specific mutations (such as TP53) expand, leading to relapse. • CHIP and Aging: Research has found that healthy individuals accumulate mutations as they age (1-2 additional mutations every 10 years), which is termed CHIP. When hematopoietic stem cells carrying mutations such as DNMT3A or TET2 acquire a second transformation event (such as FLT3-ITD or chromosomal translocation), they evolve into acute leukemia. This discovery provides a new perspective for early monitoring and risk stratification of AML.
05 From Laboratory to Guidelines: HARMONY Project and the Evolution of ELN/WHO Standards
To address the extreme heterogeneity of AML, the EU-funded HARMONY project was launched in Europe. Prof. Bullinger participated as a core expert. Through big data analysis of over 1,500 AML cases, the research team found that the prognostic significance of a single mutation often depends on its co-mutation pattern. • Refinement of ELN 2022 Classification: Based on molecular big data, the ELN 2022 guidelines made important revisions to risk stratification. For example, the prognostic evaluation of NPM1 mutation with a high FLT3-ITD burden was changed from favorable to intermediate. This directly guides clinicians in precisely grasping the timing for allogeneic hematopoietic stem cell transplantation (allo-HSCT). • WHO 5th Edition and ICC Classification: Both the WHO and ICC standards released in 2022 emphasize molecular genetic characteristics, dividing AML into 18 distinct molecular entities. Prof. Bullinger pointed out that even at a scale of 10,000 samples, these molecular clusters remain highly robust, proving that molecular subtyping is the only scientific cornerstone of AML classification.
06 Current Status of the Therapy Profile: Full Implementation of Targeted and Precision Medicine
Since 2017, AML has entered a period of “blowout” growth in targeted drugs. Prof. Bullinger provided a detailed interpretation of treatment strategies based on molecular profiles: • FLT3 Inhibitors: Midostaurin significantly improved overall survival (OS) in FLT3-mutated patients in the RATIFY study, with an HR of 0.78 (P=0.009); subsequently, Quizartinib further established its role in first-line treatment in the QuANTUM-First study. • BCL-2 Inhibitors: Venetoclax combined with hypomethylating agents (HMA) has become the standard regimen for elderly patients or those intolerant to intensive chemotherapy (in the VIALE-A study, median OS was extended from 4.1 months to 14.7 months). • Menin Inhibitors: For patients with KMT2A rearrangement or NPM1 mutation, Revumenib (SNDX-5613) and Ziftomenib have shown significant CR/CRh rates in studies such as AUGMENT-101 (reaching over 30% in some populations). • CPX-351: For secondary AML (sAML) or AML with myelodysplasia-related changes, the OS benefit is significantly superior to the traditional “3+7” regimen.
07 Future Outlook: Long-Read Sequencing, Spatial Omics, and MRD Monitoring
In the final part of the speech, Prof. Bullinger explored the next frontiers of molecular hematology: • Nanopore Sequencing: This long-read sequencing technology enables real-time data analysis, allowing for rapid detection of copy number variation (CNV), structural variation (SV), and epigenetic modifications across the whole genome within 2-3 days, which is expected to replace traditional karyotyping. • Spatial Omics: Single-cell sequencing combined with spatial omics technology will reveal the interaction between tumor clones and immune cells in the bone marrow microenvironment, helping to understand the evasion mechanisms of resistant clones within microenvironmental barriers. • NGS-MRD: With increased detection sensitivity (10^-4 to 10^-6), Measurable Residual Disease (MRD) monitoring based on NGS, multi-parameter flow cytometry, or digital PCR will become a decisive indicator for adjusting treatment strategies (such as de-escalating treatment intensity or early intervention for relapse).
Summary Conclusion:
Prof. Lars Bullinger concluded that over the past 25 years, hematology has completed the leap from simple “morphological description” to “molecular systems biology.” Although the heterogeneity of AML remains a massive challenge, with the reduction in the cost of whole-genome sequencing and the intervention of new targeted drugs (such as Menin inhibitors, bispecific antibodies, and CAR-T), the diagnosis and treatment of AML are moving from “one-size-fits-all” toward an extremely precise “one clone, one strategy” approach. As the African proverb says: “If you want to go fast, go alone; if you want to go far, go together.” The next 25 years of molecular hematology still require profound global collaboration between clinical and basic research.
