12 Dec Duke and Duke-NUS Scientists Identify Metabolic Vulnerability in AML Using New Computational Approach

MedicalResearch.com Interview with:

Dr. Matthew Hirschey

Matthew Hirschey Ph.D.
Associate Professor of Medicine
Associate Professor of Cell Biology
Associate Professor in Pharmacology and Cancer Biology
Member of the Duke Cancer Institute
Member of Sarah W. Stedman Nutrition and Metabolism Center
Hirschey Lab in the Duke Molecular Physiology Institute,
Duke University

MedicalResearch.com: What is the background for this study? Would you briefly describe AML and why new therapeutic approaches are needed?

Response: Acute myeloid leukemia (AML) is an aggressive blood cancer that begins in the bone marrow and progresses rapidly. While recent advances, particularly the BCL-2 inhibitor venetoclax combined with other agents, have improved outcomes for some patients, many still relapse or don’t respond to treatment. The five-year survival rate remains below 30% overall, highlighting an urgent need for new therapeutic strategies.

We know that cancer cells rewire their metabolism to fuel rapid growth, and the mitochondria (the cell’s powerhouses) play a central role. However, understanding exactly how different metabolic pathways connect and depend on each other has been challenging. We wanted to develop better tools to map these connections and identify new vulnerabilities we could potentially target.

MedicalResearch.com: Your team developed a new computational approach called “pathway coessentiality mapping.” How does this differ from traditional methods, and why is this innovation important?

Response: Traditional approaches examine how individual genes relate to each other, but with roughly 20,000 genes in the human genome, that creates nearly 400 million possible gene pairs to evaluate. It’s like trying to understand traffic patterns by tracking every individual car rather than looking at how highways connect neighborhoods.

Our approach instead examines how entire biological pathways (groups of genes working together) depend on each other across hundreds of cancer cell lines. We use a statistical method called Canonical Correlation Analysis to find the strongest relationships between pathways. This allows us to see the bigger picture and identify connections that might be missed when looking gene-by-gene.

When we applied this to the electron transport chain (the series of protein complexes that generate cellular energy), we immediately noticed something unexpected: Complex II stood apart from the other complexes with a unique set of pathway connections, particularly to nucleotide and amino acid metabolism.

MedicalResearch.com: What are the main findings?

Response:  We discovered that Complex II, a component of the mitochondrial energy-production machinery, directly regulates the production of purines (essential building blocks for DNA and RNA) in AML cells. This was surprising because Complex II’s traditional role is to oxidize succinate as part of cellular respiration.

The mechanism involves a metabolic circuit centered on glutamine and glutamate. When cells build purines, they use glutamine to donate nitrogen atoms, producing glutamate as a byproduct. Complex II helps clear this glutamate by metabolizing it through the TCA cycle. When we block Complex II, glutamate accumulates and essentially jams the purine production machinery.

We validated this in mouse models of AML, where depleting Complex II caused rapid disease regression and significantly extended survival. In human data, we found that high expression of Complex II genes correlates with worse outcomes and resistance to venetoclax specifically in AML patients, but not uniformly across other cancer types. 

MedicalResearch.com: What should readers take away from your report?

Response: First, metabolic pathways are interconnected in ways we’re only beginning to understand. Complex II isn’t just an energy-production enzyme; it’s a critical regulator of nucleotide synthesis in certain cancers.

Second, AML has a specific metabolic vulnerability. Unlike many solid tumors, where Complex II mutations can actually promote cancer, AML cells depend heavily on Complex II activity. This tissue-specific dependency suggests there may be a therapeutic window to exploit.

Third, computational approaches that examine pathway-level relationships can reveal biology that’s invisible when studying genes one at a time. This study began with a computational prediction that we then validated extensively in the laboratory.

MedicalResearch.com: Your findings suggest Complex II inhibition could sensitize AML cells to venetoclax. What is the clinical significance of this?

Response: Venetoclax has transformed AML treatment, but resistance remains a major challenge. Our data show that high Complex II expression correlates with venetoclax resistance in patient samples, and that blocking Complex II sensitizes AML cells to venetoclax in the laboratory.

This suggests a potential combination strategy for patients whose disease doesn’t respond adequately to current venetoclax-based regimens. Safely targeting Complex II in patients will require careful drug development. We need compounds that effectively inhibit the enzyme in cancer cells while avoiding toxicity to normal tissues, particularly the nervous system. But the biological rationale is now clear. 

MedicalResearch.com: What recommendations do you have for future research as a results of this study?

Response: Several directions are important. First, we need to develop clinically viable Complex II inhibitors with acceptable safety profiles. Some existing compounds have neurological side effects, so medicinal chemistry efforts to create brain-impermeant inhibitors could be valuable.

Second, we should investigate whether other blood cancers showing Complex II dependency in our analysis, including certain lymphomas and B-cell leukemias, share this metabolic vulnerability.

Third, we identified OGDH (alpha-ketoglutarate dehydrogenase) as another potential target in this pathway. Since OGDH inhibition appears well-tolerated in animal models, it may offer an alternative or complementary approach.

Finally, clinical studies correlating Complex II expression with treatment response could help identify which patients might benefit most from metabolism-targeted therapies.

MedicalResearch.com: Is there anything else you would like to add? Any disclosures?

Response: I would like to emphasize that this was a true team effort involving multiple institutions: Duke, Duke-NUS in Singapore, INSERM in Paris, and Northwestern. The project combined computational biology, metabolomics, proteomics, and mouse modeling. The co-first authors, Derek Zachman and Amy Stewart, drove this project forward, and our collaboration with Alexandre Puissant and Kris Wood was essential.

MedicalResearch.com: Looking ahead, how do you see data-driven approaches changing the way scientists generate hypotheses and make discoveries?

Response: We’re at an inflection point. Large-scale datasets like the Cancer Dependency Map contain enormous amounts of information about how genes function across different cellular contexts, but extracting meaningful biology requires new analytical frameworks. Our pathway coessentiality approach is one example: it revealed a connection between Complex II and purine metabolism that wasn’t apparent from traditional analysis.

To make these tools accessible to the broader research community, we developed a web platform called Data-Driven Hypothesis (DDH) that allows researchers to explore gene dependencies and generate testable predictions without requiring programming expertise. This platform has now been commercialized by Heureka Labs (heurekalabs.co), a company working to accelerate scientific discovery by putting sophisticated computational tools into the hands of bench scientists.

I believe the future of biomedical research lies in this integration: using computation to guide where we look, then rigorous experimental work to validate what we find. The discoveries waiting in existing datasets are substantial; we just need better ways to find them.

Citation: Stewart, A.E.*, Zachman, D.K.*, et al. Pathway Coessentiality Mapping Reveals Complex II is Required for de novo Purine Biosynthesis in Acute Myeloid Leukemia. [Nature Metabolism, 2025]

—

The information on MedicalResearch.com is provided for educational purposes only, and is in no way intended to diagnose, cure, or treat any medical or other condition.
Some links may be sponsored. Products are not warranted or endorsed.

Always seek the advice of your physician or other qualified health and ask your doctor any questions you may have regarding a medical condition. In addition to all other limitations and disclaimers in this agreement, service provider and its third party providers disclaim any liability or loss in connection with the content provided on this website.

Last Updated on December 12, 2025 by Marie Benz MD FAAD