Oncology keeps returning to first principles: the cell is the unit of disease and therapy. When normal cells become malignant, they do more than divide faster; they rewire how they use nutrients, hoarding glucose, rerouting amino acids and lipids, and finding new fuel pathways when others are blocked. Modern cancer treatment has targeted DNA, signaling pathways and immune cells, but metabolism—the cell’s immediate supply chain—remains underused as a therapeutic target.
A century after Otto Warburg identified cancer’s tendency to consume glucose and produce lactate even in oxygen, metabolism faded behind genetics. Yet tumors are adaptable: they rewire metabolic pathways to resist drugs. That adaptability suggests a different approach—tumor-informed metabolism—where dietary interventions are matched to the biology of a patient’s tumor, the drugs used and the patient’s physiology.
Mukherjee offers a clinical example: a woman in her 50s with hormone receptor–positive, HER2‑negative breast cancer and a PIK3CA mutation receives a PI3K inhibitor with endocrine therapy. Initial response is followed by progression linked to drug-induced rises in glucose and insulin, which open a metabolic escape route for the tumor. Paired with a tailored dietary plan designed to blunt postprandial glucose and insulin spikes—timed around dosing, calibrated to her physiology and monitored with metabolic biomarkers—the drug response deepens and lasts longer. The improvement, he argues, stems from combining drug and diet as a single, coordinated therapy.
Tumor-informed metabolism treats the tumor’s fuel use as clinical information. Tumors compete with their microenvironment and use different fuels depending on their location and treatment pressures. Asparagine depletion works in acute lymphoblastic leukemia because that tumor relies on the amino acid; serine and glycine can be critical in some breast and colorectal cancers; methionine restriction can alter one-carbon metabolism and sensitize certain tumors. Interventions can range from reducing refined carbohydrates to limiting specific amino acids during chemotherapy or radiotherapy, or adding calories and protein when weight loss would undermine treatment.
Mukherjee and colleagues at Faeth Therapeutics are developing prospective regimens that pair diets with pathway-directed drugs—PI3K/AKT/mTOR inhibitors in endometrial cancer, amino acid–restricted diets in rectal cancer—each plan tied to a specific mechanism and treatment window. He calls for more companies to create rigorous nutrition–drug regimens and for sponsors and regulators to treat diet as a prespecified element of clinical protocols rather than an unmeasured background factor.
The field is early. Preclinical and some clinical signals are encouraging when diets are mechanism-directed and time-limited; results have been mixed when diets are generic, prolonged or disconnected from therapy. Many patients lose weight during chemotherapy, some have metabolic comorbidities, and others undertake unmonitored fasting that weakens them. Tumor-informed metabolism, Mukherjee says, should be delivered as a targeted adjuvant with safeguards for weight, strength and metabolic health—not as ideology.
To make this standard practice will require prospective, controlled studies with objective endpoints—response rates, survival, preserved dose intensity and reduced toxicity—and clinical infrastructures that can act on the evidence. Needed components include multimodal regimens that limit collateral immune damage, precision nutrition scripted to complement drug mechanisms, and what Mukherjee calls a metabolic operating system: computational models that predict metabolic flux, anticipate resistance and test combinations in silico before patient implementation.
Skepticism is expected: metabolism is plastic and variable, and its clinical impact may be modest in some settings. Equity is also a concern. If metabolism becomes a precision therapeutic tool, it must be accessible and adaptable to diverse cultures and socioeconomic circumstances, not restricted to concierge care.
Mukherjee frames the case philosophically: if the cell is the unit of life, metabolism is its first action. Oncology has expanded what counts as therapy before—surgery, poisons, immune modulation—and should now integrate metabolism as another fundamental stratum of biology. The proposal is practical: prescribe macronutrient and micronutrient targets, timing and contraindications with the same specificity as a drug order, so that clinicians can “feed the person and starve the cancer” on purpose.
Siddhartha Mukherjee, M.D., D.Phil., is an oncologist, researcher and author. He is associate professor of medicine at Columbia University, co-founder of biotech ventures including Faeth Therapeutics and Manas AI, and author of The Emperor of All Maladies. A new edition of that book is now available with four additional chapters.
Leave a Reply