Work

The tumor as a dynamic metabolic interaction network: From the Warburg effect to multi-node collapse as a therapeutic strategy

Year: 2026
Type: article
Abstract: Cancer metabolism research has long been shaped by a pathway-centric paradigm, in which individual biochemical routes—most prominently aerobic glycolysis (the Warburg effect)—are studied and targeted in relative isolation. Accumulating evidence from systems biology, multi-omics profiling, and computational oncology suggests, however, that this view is insufficient to explain the adaptive resilience that solid tumors routinely demonstrate against metabolic inhibitors. Here we propose a conceptual synthesis: the tumor dynamic metabolic interaction network, a graph-theoretic framework in which metabolic enzymes and transporters constitute nodes, substrate fluxes define weighted edges, and the system’s overall state is governed by topological properties such as betweenness centrality and eigenvector centrality. Within this framework, we elaborate the hypothesis of dynamic network collapse (DNC)—the notion that simultaneous perturbation of several high-centrality nodes may push the tumor metabolic network beyond a thermodynamic instability threshold, triggering irreversible functional disintegration that cannot be escaped through known compensatory rerouting. We review the three dominant metabolic axes of tumor survival—glucose (GLUT1/HK2/PKM2), glutamine (SLC1A5/GLS1), and lactate efflux (MCT1/MCT4)—and examine how each contributes to immunosuppression through metabolic competition with cytotoxic T lymphocytes and through the acidification of the tumor microenvironment. We further discuss genome-scale metabolic models, flux balance analysis, and graph neural network approaches that may render the DNC concept computationally actionable. The clinical implications of multi-node targeting are considered alongside the limitations of existing evidence and unresolved methodological challenges. Although the DNC hypothesis remains to be tested experimentally in relevant in vivo models, it provides a coherent rationale for the next generation of combination metabolic therapies and for patient-specific metabolic profiling as a decision tool.
Language: English
Subfield: Cancer Research
Open Access status: hybrid
OpenAlex ID: W7162698544
FWCI:
Cites: 45
Cited by:
Related to: 9