Scientists from Duke-NUS Medical School and their international collaborators have developed a new computational tool that maps how gene pathways interact in complex biological systems. Using this novel algorithm, the team discovered a previously unknown protein pathway that, when blocked, triggers the death of blood cancer cells.
The discovery could pave the way for new therapies and drugs that could reduce disease severity and improve survival rates in patients with aggressive acute myeloid leukemia (AML).
Published in Nature Metabolism, the study was a multi-institutional collaboration between Duke-NUS, Duke University, Northwestern University, and INSERM (France).
Understanding acute myeloid leukemia
Globally, blood cancers account for 6% of all cancer cases, affecting an estimated 1.24 million people each year. AML is one of the most aggressive forms of cancer that affects the blood and bone marrow, and is marked by poor survival rates and limited treatment options.
To tackle this key challenge in treating AML, scientists needed to understand how different cellular pathways work together. Past research mainly focused on single-gene relationships.
New mapping tool reveals key pathways
With their new mapping tool, the team analyzed about 3,000 gene sets covering nearly 15,000 genes, revealing a clearer, more complete picture of how gene pathways interact—particularly those that help cancer cells generate energy.
This new approach revealed an unexpected connection between energy metabolism and DNA building blocks in leukemia. The team discovered that a protein complex, Complex II, allows cancer cells to multiply by producing purines—essential building blocks of the cells’ DNA and RNA.
As cancer cells need energy and nutrients to keep growing rapidly, blocking Complex II in the body prevents them from creating enough purines, causing them to die. Normal cells, on the other hand, find alternative ways to acquire nutrients and will continue surviving.
Associate Professor Matthew Hirschey from the Cardiovascular & Metabolic Disorders Program at Duke-NUS and Duke University School of Medicine, is one of the senior authors of the paper. He explained, “Cancer cells need energy and building blocks to grow rapidly. Complex II acts like a metabolic traffic controller, managing the flow of nutrients that cells use to build components of their DNA. DNA functions as genetic instructions for cell development, and if inhibited, will affect cells’ ability to reproduce.”
Another senior author, Associate Professor Kris Wood, Department of Pharmacology and Cancer Biology at Duke University School of Medicine, stated, “Complex II represents a metabolic Achilles’ heel in acute myeloid leukemia. By disrupting this specific metabolic circuit, we can selectively kill cancer cells while sparing normal tissues.”
In particular, high-risk AML patients, or those who do not respond well to current therapies, stand to benefit from this discovery. For instance, patients with drug-resistant leukemia might benefit from combining Complex II inhibition in their treatment regiment, which has been shown to enhance cancer cell death.
Dr. Alexandre Puissant, French National Institute of Health and Medical Research (INSERM) and a senior author of the paper, said, “Our pre-clinical trials showed dramatic tumor regression and extended survival when targeting Complex II, suggesting that this could translate into meaningful clinical benefits.”
The research also demonstrates that higher Complex II levels are linked to increased treatment resistance and poorer patient survival, which may potentially make them useful biomarkers in patient screening and treatment plans.
Next steps and future applications
In the next stages of development, the researchers are hoping to identify new drug targets, investigate how purine synthesis can be better inhibited, and reveal more metabolic vulnerabilities in other blood cancers and solid tumors.
The researchers also hope to develop a web app and machine learning tools that can predict which patients are likely to benefit from metabolic treatments, based on the unique gene pathway patterns in their tumors.
Professor Patrick Tan, Senior Vice-Dean for Research at Duke-NUS, remarked, “By combining computational science with molecular biology, we can uncover new ‘blind spots’ in cancer metabolism—potentially identifying dozens of new treatment targets in far less time than conventional methods. Our new approach moves us closer to more effective therapies that can help patients live longer, healthier lives.”
More information
Amy E. Stewart et al, Pathway coessentiality mapping reveals complex II is required for de novo purine biosynthesis in acute myeloid leukaemia, Nature Metabolism (2025). DOI: 10.1038/s42255-025-01410-x
Journal information:
Nature Metabolism
Key medical concepts
The content is provided for information purposes only.