The Challenge of Starving Cancer: Metabolic Flexibility vs. Oxidative Vulnerability

Metabolic adaptations that help tumor cells grow and survive make an attractive target for cancer therapy. The premise is seductive: if we could cut off the fuel supply that tumors need to grow and spread, could we starve them into submission? This appealing concept has spawned numerous dietary interventions, from ketogenic diets to fasting regimens, while motivating both the repurposing of existing drugs and the development of new ones designed to block specific metabolic pathways. However, this type of metabolic therapy faces two major obstacles. First, cancer cells are remarkably adaptable in how they fuel themselves, and different cancers, and even different cells within the same tumor, can use very different fuel mixes. Second, the line between a dose that harms cancer and a dose that harms healthy tissue is dangerously thin, because cancer cells and normal cells, including the immune cells that fight cancer, rely on many of the same nutrients. Mounting scientific evidence is now raising troubling questions about whether attempts to starve cancer through nutrient restriction might not only fail but potentially worsen outcomes. Three fundamental biological obstacles challenge this approach: cancer cells’ ability to utilize an astonishing variety of fuel sources, their remarkable ability to rapidly adapt when any single fuel is restricted, and the concerning possibility that the very nutrients we might withhold are equally essential for immune cells fighting cancer and healthy cells maintaining normal function. These limitations demand critical examination of whether metabolic restriction represents a viable therapeutic strategy or a dangerous oversimplification of cancer biology.

The Metabolic Versatility of Cancer Cells

Cancer cells are metabolic omnivores, capable of utilizing virtually any nutrient available in their environment. While the Warburg effect, discovered over a century ago, highlighted cancer’s preference for glucose, we now understand that tumors exploit a far more diverse metabolic repertoire.

Glucose remains a primary fuel via aerobic glycolysis, in which cancer cells convert glucose to lactate even when oxygen is abundant. This process, though energetically inefficient compared to complete oxidation, provides rapid energy and generates biosynthetic intermediates for nucleotides, amino acids, and lipids. Cancer cells overexpress glucose transporters, particularly GLUT1, and upregulate glycolytic enzymes to maximize glucose utilization. The reliance on glucose is so pronounced that it underpins PET imaging, which exploits elevated glucose uptake in tumors.

Beyond glucose, glutamine serves as another critical fuel source for many cancers. Often called the second most important nutrient after glucose, glutamine is metabolized via glutaminolysis to produce alpha-ketoglutarate, which replenishes the TCA cycle. This process provides both energy and nitrogen for nucleotide and amino acid synthesis. Many cancer types, including glioblastoma, renal cell carcinoma, and triple-negative breast cancer, exhibit glutamine addiction, becoming highly dependent on this amino acid despite it being nonessential in normal cells. The MYC oncogene directly upregulates glutamine transporters and glutaminase, coordinating increased glutamine metabolism with proliferation.

Fatty acids represent yet another fuel source that cancer cells exploit in multiple ways. Some tumors rely on de novo lipid synthesis, converting glucose and glutamine into fatty acids via upregulated enzymes such as fatty acid synthase and ATP citrate lyase. Other cancers, particularly cancer stem cells and metastatic cells, preferentially oxidize fatty acids to generate ATP. Prostate cancer exhibits particularly high fatty acid metabolism, with elevated expression of CD36, a fatty acid transporter that facilitates uptake from the bloodstream. The metabolic flexibility to switch between fatty acid synthesis and oxidation allows cancer cells to adapt to different microenvironmental conditions.

Recent research has revealed that lactate, once dismissed as merely a metabolic waste product, functions as an important fuel source. Cancer cells can import lactate via monocarboxylate transporters and convert it to pyruvate for mitochondrial oxidation. In human lung tumors, cancer cells actively consume lactate for energy, with studies showing that over half of the TCA cycle intermediates can be derived from lactate under glucose deprivation. This lactate metabolism exemplifies a phenomenon called metabolic symbiosis, in which glycolytic cancer cells export lactate, which more oxidative cancer cells and stromal cells import and oxidize.

Acetate provides yet another alternative fuel, particularly under metabolic stress. Cancer cells upregulate enzymes like ACSS2 that convert acetate into acetyl-CoA, which can then fuel both energy production and lipid synthesis. Under hypoxic conditions, acetate can provide up to half of the lipogenic acetyl-CoA. Sources of acetate include dietary intake, gut microbiota fermentation, and endogenous production. Multiple cancer types, including breast, prostate, glioblastoma, and hepatocellular carcinoma, show genetic amplification or upregulation of acetate-metabolizing enzymes, with high expression correlating with advanced disease and poor prognosis.

Ketone bodies, including beta-hydroxybutyrate and acetoacetate, present a more complex picture. While some have proposed ketogenic diets to starve tumors of glucose, evidence shows that certain cancer cells can utilize ketone bodies as alternative energy sources. The response appears highly heterogeneous depending on the expression of ketone-metabolizing enzymes. Some colorectal cancers and melanomas retain the capacity for ketone oxidation, potentially explaining why ketogenic diets have shown inconsistent effects across different tumor types.

Beyond these major fuel sources, cancer cells can use a wide range of amino acids. Serine metabolism supports one-carbon metabolism, which is essential for nucleotide synthesis, redox balance through glutathione production, and epigenetic regulation. Asparagine drives reverse transport of other amino acids and promotes metastasis in breast cancer models. Branched-chain amino acids like leucine, isoleucine, and valine provide both energy and signaling functions, particularly in activating mTORC1 to promote growth. Proline metabolism supports collagen synthesis and extracellular matrix remodeling. Cancer cells can even engage in macropinocytosis to scavenge extracellular proteins, breaking them down into constituent amino acids during nutrient stress.

This impressive metabolic versatility means that cancer cells are never dependent on just one fuel source. When a nutrient becomes scarce, tumors readily switch to numerous alternatives, thereby frustrating attempts at metabolic restriction.

The Extraordinary Metabolic Plasticity of Advanced Cancers

Metabolic plasticity refers to the ability of cancer cells to process nutrients in different ways or to utilize different nutrients to meet the same metabolic needs. This flexibility is maximized in advanced, metastatic cancers that have survived multiple selective pressures.

The mechanisms enabling this plasticity operate at multiple levels. Transcriptional regulation by factors such as HIF-1α, which is stabilized under tumor hypoxia, upregulates both glycolytic enzymes and alternative pathways, including reductive carboxylation of glutamine to produce fatty acids. The MYC oncogene coordinates the expression of glycolytic genes, glutaminase, glutamine transporters, and the pentose phosphate pathway to support rapid biosynthesis. PGC-1α promotes mitochondrial biogenesis and oxidative metabolism and exhibits dynamic expression during metastatic progression.

Post-translational signaling through pathways such as AMPK and mTOR enables rapid metabolic adaptation. AMPK senses energy stress via the AMP-to-ATP ratio, responding by inhibiting anabolic processes, activating autophagy, and increasing nutrient uptake. The mTOR complex integrates growth signals with nutrient availability, promoting biosynthesis when nutrients are abundant but allowing autophagy during starvation. Critically, most tumors retain functional nutrient-sensing pathways despite oncogenic mutations, preserving their ability to adapt.

Metabolic enzymes themselves can switch between different functional states. PKM2, for example, exists in dimeric and tetrameric forms that favor biosynthesis or energy production, respectively. Redundant pathways provide backup routes when one is blocked. When fatty acid desaturase SCD1 is inhibited, FADS2 can provide an alternative route for monounsaturated fatty acid synthesis, requiring dual inhibition to effectively block proliferation.

Metastatic cancers exhibit particularly pronounced metabolic flexibility because they must navigate drastically different environments during the metastatic cascade. Primary tumors exist in hypoxic, acidic environments with steep nutrient gradients. Cancer cells entering the circulation encounter high oxygen levels that generate reactive oxygen species, necessitating robust antioxidant systems fueled by alternative metabolic pathways. Different metastatic sites present unique metabolic landscapes. Lung metastases must adapt to high-oxygen, pyruvate-enriched conditions, often upregulating proline catabolism and glutaminolysis. Brain metastases experience low availability of glucose and serine, underscoring the importance of serine synthesis pathways. Liver metastases encounter variable oxygen and can utilize acetate and fructose as alternative fuels. Bone metastases modulate local metabolism by secreting lactate and serine, which support osteoclast activity and bone resorption.

Selection pressure during metastasis ensures that only cells with maximal metabolic adaptability successfully colonize distant organs. Studies show that inhibiting specific metabolic transporters, such as MCT1 or CD36, dramatically impairs metastasis while having minimal effects on primary tumors, revealing transient metabolic dependencies during specific metastatic steps.

When faced with nutrient restriction, cancer cells deploy remarkable adaptive responses. Under glucose deprivation, colorectal cancer cells survive by increasing glutamate dehydrogenase activity, thereby channeling glutamine-derived glutamate into the TCA cycle. Pancreatic cancer cells subjected to prolonged glutamine starvation stabilize glutamine synthetase despite nutrient shortage, enabling de novo glutamine synthesis from other amino acids through an mTORC1-mediated epigenetic adaptation. Sarcoma cells that initially depend on exogenous glutamine upregulate glutamine synthetase after an adaptation period, thereby replacing their external requirement with internal production.

This metabolic heterogeneity extends within individual tumors. Single tumor lesions display regional metabolic differences, with poorly perfused areas containing glycolytic cells and well-perfused regions harboring lactate-consuming oxidative cells. Cancer stem cells typically exhibit more oxidative metabolism than bulk tumor cells, relying heavily on fatty acid oxidation and displaying resistance to glycolysis-targeting therapies. Pre-existing resistant populations with distinct metabolic profiles await selection by therapy, emerging as dominant clones when treatment eliminates metabolically vulnerable cells.

Treatment itself drives metabolic evolution. Anti-angiogenic drugs initially reduce blood supply and induce hypoxia, but tumors adapt through enhanced glycolysis and HIF-1α-driven metabolic rewiring, eventually resuming growth despite continued treatment. mTOR inhibition in lung cancer initially blocks glycolysis, but compensatory upregulation of glutaminase through GSK3-MYC signaling increases glutaminolysis to maintain proliferation. These adaptive responses can be transient, disappearing when drug pressure is removed, but they permit tumor survival during therapy.

The Shared Metabolic Dependencies of Normal Cells and Cancer Cells

A central challenge in cancer therapy is the difficulty of selectively targeting malignant cells while sparing normal, healthy tissues. Cancer cells often share many biological pathways and surface markers with normal cells, making it hard for drugs, radiation, or even newer immunotherapies to distinguish between the two. This lack of cancer selectivity dramatically narrows the therapeutic window, the critical difference between doses that effectively kill tumor cells and doses that cause unacceptable toxicity to normal tissues. As a result, treatments potent enough to kill tumor cells often also harm rapidly dividing healthy cells in the bone marrow, gastrointestinal tract, skin, and other organs. This constrained therapeutic window forces doctors to balance tumor control against tolerability, often limiting therapeutic effectiveness since doses must be carefully restricted to avoid intolerable toxicity. The resulting side effects, including immunosuppression, fatigue, nausea, and tissue damage, not only compromise quality of life but also restrict how aggressively therapy can be pursued. In contrast, strategies that exploit vulnerabilities specific to cancer cells can widen this therapeutic window, enabling more effective tumor destruction while minimizing collateral damage to healthy tissues.

One of the most fundamental obstacles to metabolic cancer therapy is that normal cells, particularly immune cells critical for fighting cancer, rely on the same nutrients and metabolic pathways that cancer cells exploit. Normal quiescent cells primarily use oxidative phosphorylation for energy production, efficiently generating ATP through the complete oxidation of nutrients. However, during periods of rapid proliferation, such as tissue repair or immune activation, normal cells undergo metabolic reprogramming that is remarkably similar to that of cancer cells. They upregulate aerobic glycolysis, increase glucose and glutamine uptake, enhance amino acid transport, and activate biosynthetic pathways. This similarity creates a profound therapeutic challenge.

T cells provide perhaps the clearest example of this metabolic convergence. Naive and memory T cells maintain low metabolic activity with a preference for oxidative phosphorylation and fatty acid oxidation. However, upon activation through T cell receptor and CD28 co-stimulation, T cells undergo dramatic metabolic reprogramming, with five- to tenfold increases in glucose and glutamine uptake. They shift to aerobic glycolysis despite the availability of oxygen, directly paralleling the Warburg effect in cancer cells.

Glucose is absolutely critical for T cell effector function. Glucose deprivation blocks interferon-gamma production through a specific mechanism where GAPDH, a glycolytic enzyme, binds to interferon-gamma mRNA and prevents its translation. Studies using stable isotope tracing show that activated CD8 T cells in living organisms primarily use glucose for nucleotide and serine synthesis rather than energy production, emphasizing their biosynthetic importance. T cells also require glucose to maintain CD25 expression, the IL-2 receptor essential for expansion.

Glutamine is consumed by activated T cells at rates equal to or exceeding glucose consumption. It is essential for TH1 and TH17 differentiation, CD8 T cell expansion, aspartate synthesis for DNA and RNA production, and alpha-ketoglutarate generation for epigenetic regulation. In vivo tracing studies demonstrate that, in activated T cells, glutamine contributes more carbon to the TCA cycle than glucose, revealing its substantial role in oxidative metabolism.

Amino acid metabolism profoundly influences T cell function. Methionine is required for S-adenosylmethionine production and histone methylation. Methionine deprivation leads to loss of specific histone methylation marks, thereby diminishing STAT5 expression and IL-2 signaling, resulting in T cell dysfunction. Supplementing methionine improves T cell function in tumor-bearing mice. Arginine is essential for polyamine biosynthesis and T cell proliferation. Serine is required for optimal T cell expansion even when glucose is abundant.

Natural killer cells similarly depend on metabolic reprogramming for function. Cytokine-activated NK cells upregulate glucose uptake, increase glycolysis and oxidative phosphorylation, and enhance amino acid transport. Glucose is essential for interferon-gamma production and granzyme B expression. Inhibiting glycolysis with 2-deoxyglucose potently blocks NK cell cytotoxicity. Recent evidence indicates that fatty acid oxidation is crucial for NK cell function in viral infection and tumor immunity, supporting mitochondrial function, TCA cycle activity, aspartate production for nucleotide synthesis, and actin polarization for immune synapse formation.

Macrophages display polarization-dependent metabolic profiles. M1 pro-inflammatory macrophages with anti-tumor activity rely primarily on aerobic glycolysis, producing nitric oxide, reactive oxygen species, and pro-inflammatory cytokines. M2 anti-inflammatory macrophages depend on oxidative phosphorylation and fatty acid oxidation. Tumor-associated macrophages typically adopt an M2-like phenotype, but this polarization is metabolically driven. Tumor-derived lactate induces M2 polarization through HIF-1α stabilization. This creates a vicious cycle in which metabolic changes in the tumor microenvironment shape immune cell phenotypes that promote tumor growth.

The mounting evidence is making it unmistakably clear that tumor cells and immune cells share the same metabolic ecosystem, and many of the pathways we might want to shut down in cancer are also essential for T cells, NK cells, dendritic cells, and other immune populations. Consequently, aggressively blocking cancer metabolic fuels (e.g., glucose, glutamine, fatty acids) can backfire.

The tumor microenvironment functions as a metabolic checkpoint for immune responses. Landmark studies have demonstrated that tumor glucose consumption metabolically restricts T cells within the tumor microenvironment, dampening mTOR activity, reducing glycolytic capacity, and impairing interferon-gamma production. Remarkably, enhancing glycolysis in normally rejected tumors proved sufficient to override protective T cell responses, directly demonstrating that metabolic competition drives tumor progression. Checkpoint blockade therapies such as anti-PD-1 and anti-CTLA-4 act, in part, by restoring glucose availability in the tumor microenvironment, thereby enabling T-cell glycolysis and cytokine production.

The tumor microenvironment becomes hypoxic and acidic due to lactate accumulation, and is depleted of glucose, glutamine, and amino acids. Phosphoenolpyruvate, a glycolysis metabolite, serves as a metabolic checkpoint for anti-tumor T cell responses by sustaining calcium-NFAT signaling. Insufficient PEP due to glucose deprivation impairs T cell effector functions. Lactate-associated acidic pH directly inhibits T cell proliferation, cytokine production, and cytotoxicity.

Specific nutrient competition further undermines immunity. Tumors outcompete T cells for methionine via upregulated transporters, thereby lowering intracellular methionine levels and impairing T-cell function. Tumors expressing IDO deplete tryptophan while producing kynurenine, which suppresses T cell responses. Myeloid-derived suppressor cells and tumor-associated macrophages that express arginase-1 deplete arginine, thereby impairing T cell proliferation.

The Cachexia Paradox: Starving Patients While Trying to Starve Tumors

Before considering how nutrient restriction selects for aggressive tumor cells, we must confront an even more immediate danger. Cancer cachexia, a devastating muscle-wasting syndrome, affects 50-80% of patients with advanced cancer and directly causes 20-30% of cancer deaths. This syndrome cannot be reversed by conventional nutritional support and leads to progressive functional impairment, with death typically occurring when weight loss exceeds 25 to 30% of body weight. Cachectic patients at death have lost approximately 85% of total body fat and 75% of skeletal muscle, with respiratory failure from muscle wasting responsible for 48% of cancer patient deaths. Cachexia often kills patients before tumor progression does, making any intervention that worsens muscle loss potentially more dangerous than the cancer itself.

Aggressive nutrient restriction strategies risk catastrophically accelerating cachexia. The most striking evidence comes from a 2023 Cell Metabolism study showing that while a ketogenic diet delayed tumor growth in mice, it paradoxically caused accelerated onset of cancer cachexia and shortened survival. The mechanism involved systemic metabolic stress from NADPH depletion that impaired corticosterone production, leaving mice unable to adapt to the energetic demands of both tumor burden and dietary restriction. This exemplifies the fundamental problem: metabolic interventions that successfully limit tumor growth can simultaneously destroy the patient’s vital tissues. Multiple studies demonstrate that reduced food intake independently predicts worse survival in cancer patients, with severely reduced intake increasing the odds of severe weight loss nearly nineteen-fold. Clinical guidelines now explicitly warn that patients at risk for malnutrition or cachexia may not be candidates for fasting or severe dietary restriction, as further limiting nutrient intake may be unsafe even for short periods.

The metabolic mechanisms driving cachexia compound the danger of restriction approaches. Inflammatory cytokines activate the ubiquitin-proteasome system to degrade muscle proteins while simultaneously suppressing protein synthesis by inhibiting the mTOR pathway. Mitochondrial dysfunction impairs oxidative metabolism while hypothalamic inflammation reduces appetite. The result is a perfect storm where the patient experiences increased protein breakdown, decreased protein synthesis, reduced food intake, and elevated resting energy expenditure. Imposing additional caloric or nutrient restriction on top of this already catastrophic metabolic state risks pushing patients past the point of no return, where tissue loss becomes incompatible with survival regardless of the tumor response.

Cancer’s Ultimate Survival Strategy: Stealing Mitochondria from Neighboring Normal Cells

When nutrient restriction fails to starve cancer, tumors can deploy an even more sinister survival mechanism: they steal functional mitochondria directly from neighboring normal cells. This phenomenon, documented extensively in research published between 2017 and 2025, reveals that cancer cells acquire mitochondria from T cells, stromal cells, endothelial cells, and even neurons through tunneling nanotubes and other mechanisms. Critically, mitochondrial transfer rates increase dramatically under the very conditions that metabolic restriction strategies create: nutrient deprivation, hypoxia, and oxidative stress.

The most alarming findings come from studies of immune cell mitochondria. A 2023 Cancer Cell paper reported that 64.4% of bladder cancer cells acquired mitochondria from cocultured T cells, and that tumor hypoxia levels significantly correlated with mitochondrial transfer scores. Cancer cells that acquired T-cell mitochondria exhibited increased cell-cycle activity and poorer clinical outcomes. Even more concerning, a 2025 Nature study revealed that cancer cells transfer their damaged, mutated mitochondria to tumor-infiltrating lymphocytes, causing metabolic dysfunction and immune exhaustion, while simultaneously acquiring functional mitochondria from those same T cells. This bidirectional metabolic exchange empowers cancer cells while crippling the immune system’s ability to combat them.

Mitochondrial theft extends beyond immune cells. Bone marrow stromal cell transfer of functional mitochondria to acute myeloid leukemia cells occurs specifically under chemotherapy stress, conferring survival advantages and treatment resistance. Glioblastoma cells acquire mitochondria from mesenchymal stem cells during temozolomide chemotherapy, enabling a metabolic shift from glucose to glutamine utilization, thereby supporting continued growth. Most strikingly, a 2025 Nature study showed that breast cancer cells acquire neuronal mitochondria during metastasis, with 27.3% of lung metastases and 46.0% of brain metastases containing cancer cells enriched with neuronal mitochondria compared to just 5.4% of primary tumor cells. The pattern is consistent: when cancer cells face metabolic stress from nutrient deprivation, chemotherapy, or hostile microenvironments, they compensate by hijacking mitochondria from healthy neighboring cells.

The implications for metabolic restriction strategies are profound. Glucose deprivation, hypoxia, and oxidative stress do not simply stress cancer cells. These conditions actively trigger mechanisms that enable cancer cells to acquire functional mitochondria from the very cells patients need to survive and fight disease. A cancer cell, stripped of its mitochondrial function by metabolic inhibition, can obtain replacement mitochondria from a T cell, neuron, or stromal cell. Acquired mitochondria restore oxidative metabolism, increase ATP production, reduce drug-induced mitochondrial depolarization, and confer chemotherapeutic resistance. Rather than starving cancer into submission, severe metabolic restriction may inadvertently create selection pressure favoring cancer cells most adept at metabolic parasitism of normal tissues.

How Nutrient Restriction Can Select for More Aggressive Tumor Cells

Attempts to restrict nutrients and block metabolic pathways can create selection pressures that drive the evolution of more aggressive, treatment-resistant tumor phenotypes, ultimately making cancer worse.

Calorie-restriction studies provide direct evidence that severe energy restriction impairs antitumor immunity. In a 2023 study, calorie restriction slowed melanoma tumor growth in control mice, but CD8 T cell depletion had no effect on tumor growth in calorie-restricted mice, indicating that calorie restriction impaired CD8 T cell immune surveillance. Most strikingly, anti-PD-1 immunotherapy completely lost efficacy under conditions of calorie restriction. The mechanism involved altered mitochondrial activity and metabolic fitness in CD8 T cells. This research concluded that energy-restricted conditions in cancer patients may impair CD8 T cell immune surveillance and the efficacy of immunotherapy.

Methionine restriction demonstrates how dietary interventions can backfire through complex mechanisms. While methionine restriction inhibited cancer growth in immunodeficient mice, it enhanced tumor growth in immunocompetent mice. Methionine restriction reduced circulating CD8 T cells, decreased interferon-gamma and TNF-alpha-producing CD8 T cells, and impaired responses to anti-PD-1 and anti-CTLA-4 immunotherapy. The mechanism involved alterations in gut microbiota composition that reduced microbial hydrogen sulfide production, which normally potentiates T cell activation. This study demonstrated that an intervention effective against cancer cells in isolation proved counterproductive when immune function was taken into account.

Glutaminase inhibition illustrates how blocking specific metabolic pathways can harm immunity. A 2022 study of the glutaminase inhibitor CB-839 combined with anti-PD-1 therapy in lung cancer found that glutaminase inhibition reduced clonal expansion of CD8 T cells, blunted cytotoxic capacity, and impaired expansion of the effector T cell pool. The combination did not improve survival outcomes. Another study showed that prolonged glutaminase inhibition led to loss of the enhanced CAR T cell function observed with short-term treatment, underscoring the importance of timing and duration. Different approaches to blocking glutamine metabolism produced distinct outcomes: DON treatment reduced T cell persistence after adoptive transfer and impaired tumor control.

Beyond immune impairment, metabolic stress directly selects for cancer cells with enhanced adaptive capacity. Pre-existing resistant populations with distinct metabolic profiles exist before treatment. For example, cytarabine-resistant acute myeloid leukemia cells, characterized by elevated reactive oxygen species levels, increased mitochondrial mass, and oxidative metabolism, reside in the bone marrow vascular niche prior to drug exposure. Treatment selects for these metabolically distinct populations, allowing them to dominate.

Single-cell metabolic imaging reveals that tumors display increased heterogeneity after treatment, not from cell death or immune infiltration but from differential adaptation within surviving tumor cells. Some subpopulations successfully adapt through metabolic reprogramming while others remain vulnerable. The adapted survivors, having overcome metabolic stress, often exhibit more aggressive phenotypes and greater survival capacity under future stress.

The plasticity that enables adaptation means that blocking one pathway typically activates compensatory mechanisms. Inhibition of glycolysis prompts cells to switch to lactate consumption, upregulate glutaminolysis, and increase fatty acid oxidation. Glutaminase inhibition triggers activation of alternative glutamate sources, upregulation of glutamine synthetase, and makes asparagine conditionally essential. This metabolic compensation represents a form of non-genetic resistance that emerges rapidly under selection pressure.

Concerning evidence suggests that some metabolic interventions may actively promote metastasis. A 2024 study from Columbia University found that while the ketogenic diet suppressed primary breast tumor growth, it dramatically increased metastasis. The mechanism involves upregulation of the BACH1 protein, which enables cancer cells to escape glucose-deprived environments. This represents a major safety concern, as an intervention effective against the primary tumor promoted the process that actually kills most cancer patients.

Different cancer types can respond oppositely to the same metabolic intervention. Research from Memorial Sloan Kettering by Dr. Siddhartha Mukherjee found that the ketogenic diet accelerated certain leukemias while showing potential synergy with treatment in other cancer types. This context-dependence means that population-level metabolic recommendations could harm some patients even while benefiting others.

Metabolic stress can trigger epithelial-mesenchymal transition, a process associated with increased invasiveness, metastatic potential, and treatment resistance. Nutrient deprivation activates stress response pathways that can promote cancer stem cell phenotypes characterized by quiescence, metabolic flexibility, and resistance to conventional therapies. These stem-like cells often enter dormancy during metabolic stress, surviving chemotherapy that targets proliferating cells, only to re-emerge after treatment ends.

The tumor microenvironment provides metabolic support that can bypass restriction attempts. Cancer-associated fibroblasts produce lactate, ketones, and amino acids that cancer cells readily consume. They secrete lipids that support tumor growth. Stromal cells can synthesize glutamine to supply cancer cells, circumventing glutaminase inhibition. Adipocytes provide fatty acids via lipolysis, particularly relevant in breast and ovarian cancer metastasis to fat-rich environments. This metabolic crosstalk with the microenvironment means that even effective blockade of tumor cell-autonomous metabolism can be overcome by external nutrient sources.

The Gap Between Promise and Clinical Reality

The theoretical appeal of metabolic cancer therapy has motivated numerous clinical trials, but results have been disappointing, further illustrating the fundamental challenges of metabolic restriction approaches.

Ketogenic diet trials exemplify this gap. Despite extensive preclinical success across multiple mouse strains, demonstrating decreased tumor growth, prolonged survival, and reversal of cachexia, human trials have been equivocal at best. A comprehensive 2021 review concluded that despite promising preclinical effects and limited rigorous human trials, the effects of the ketogenic diet on cancer and as adjunctive therapy are essentially unknown due to a lack of high-quality clinical trials. Methodological limitations pervade the existing studies, including small, heterogeneous patient samples; the absence of randomization and control groups; poorly described protocols; inadequate assessment of dietary adherence; and short durations.

The few larger trials show limited benefit. An Osaka University study of 55 stage IV cancer patients found that 37 followed a ketogenic diet for at least three months, with a median overall survival of 25.1 months and a five-year survival of 23.9%. Propensity score analysis suggested a benefit from sustained adherence, but the study lacked adequate controls and statistical power. Memorial Sloan Kettering experts reviewing the evidence concluded that whole food plant-based diets are better than ketogenic diets for reducing cancer risk and supporting health after treatment.

Fasting and caloric restriction remain mostly preclinical. Despite a mechanistic rationale involving reduced insulin-like growth factor-1 and enhanced differential stress resistance between normal and cancer cells, few human trials have produced definitive results. Experts emphasize that fasting during cancer treatment carries risks, including malnutrition, weight loss contributing to fatigue, slowed healing, and difficulty managing treatment side effects. A 2024 review concluded that while fasting may hold promise as supportive therapy, there is currently insufficient evidence to support its use as a primary treatment modality.

Glutaminase inhibitors represent perhaps the most dramatic translational failure. CB-839, also known as telaglenastat, demonstrated potent antiproliferative activity across cell lines and xenograft models, prompting numerous clinical trials. However, the large, randomized Phase III CANTATA trial in metastatic renal cell carcinoma found that telaglenastat plus cabozantinib did not improve outcomes compared with cabozantinib alone and failed to meet the primary endpoint. The ENTRATA trial, combining telaglenastat with everolimus in advanced renal cell carcinoma, similarly showed no significant benefit. A Phase I/II study combining telaglenastat with nivolumab across melanoma, renal cell carcinoma, and non-small cell lung cancer enrolled 118 patients but found limited clinical activity, with investigators concluding that the combination did not show a pattern of efficacy across different study cohorts, despite a few exceptional responses in melanoma patients.

The glucose metabolism inhibitor 2-deoxy-D-glucose has been studied since the 1950s. A 2013 Phase I dose-escalation trial found that the maximum tolerated dose achieved only 32% stable disease, 3% partial response, and 66% progressive disease. Toxicity at therapeutic doses included reversible hyperglycemia, GI bleeding, QTc prolongation, and hypoglycemia-like symptoms. Experts concluded that the relative lack of clinical efficacy at tolerable doses has been echoed by most other attempts to directly target aerobic glycolysis, with clinical success remaining limited despite preclinical promise.

Among the few metabolic inhibitors to gain FDA approval, ivosidenib and enasidenib target specific IDH mutations in acute myeloid leukemia, representing a narrow indication that depends on a specific genetic context rather than on general metabolic targeting. This pattern of success only in specific genetic contexts highlights a critical lesson: effective metabolic therapy requires absolute metabolic dependency, not merely upregulation, along with limited compensation mechanisms and adequate therapeutic windows.

Why has metabolic targeting been so challenging clinically? A 2022 comprehensive review in Nature Reviews Drug Discovery noted that 100 years have passed since Warburg identified alterations in cancer metabolism, yet progress in targeting cancer metabolism therapeutically over the past decade has been limited. The intricacy and adaptability of metabolic networks hinder the effectiveness of metabolic therapies. Tumors rapidly activate alternative pathways through mTORC1-coordinated metabolic reprogramming, and the microenvironment provides metabolic substrates through cancer-associated fibroblasts, adipocytes, and other stromal cells.

The lack of a therapeutic window remains a critical problem. Normal proliferating cells in bone marrow, intestinal crypts, and hair follicles often have higher proliferation rates than cancer cells. Myelosuppression and gastrointestinal toxicity are frequently dose-limiting, preventing achievement of therapeutic drug levels. The metabolic profile of tumors depends on both genetic lesions and tissue type, creating context-specificity that frustrates one-size-fits-all approaches. Inadequate preclinical models that use 2D cell culture with supraphysiologic glucose and glutamine concentrations, and lack hypoxia, nutrient gradients, and stromal interactions, fail to recapitulate the complexity of human tumors.

Perhaps most fundamentally, strategies targeting intrinsic cancer cell metabolism often failed to account for the metabolism of non-cancer stromal and immune cells, which have pivotal roles in tumor progression and maintenance. The conceptual framework for drug design must consider metabolic vulnerabilities of non-cancer cells in the tumor immune microenvironment as well as those of cancer cells.

The Fundamental Limitations of Metabolic Restriction

Attempting to starve cancer through nutrient restriction may ultimately fail therapeutically from three interrelated biological realities. First, cancer cells are metabolic generalists that can utilize glucose, glutamine, fatty acids, lactate, acetate, ketone bodies, and numerous amino acids as fuel sources, with sophisticated mechanisms for switching among these nutrients as their availability changes. Second, advanced metastatic cancers possess extraordinary metabolic plasticity honed through survival of multiple selective pressures, enabling rapid adaptation through transcriptional reprogramming, post-translational signaling, enzyme-level switching, and exploitation of microenvironmental nutrient sources. Third, the very nutrients one might restrict are essential for normal cells, particularly for immune cells that fight cancer, creating a situation in which metabolic interventions can impair anti-tumor immunity more effectively than they impair tumor growth.

Clinical evidence reinforces these concerns. Severe calorie restriction abolishes immunotherapy efficacy. Methionine restriction enhances tumor growth in immunocompetent hosts. Glutaminase inhibition impairs T cell function and has failed in large randomized trials. Ketogenic diets show equivocal results in humans and may even promote metastasis in some contexts. After a century of metabolic cancer research, very few metabolism-based drugs have succeeded, with most failing due to metabolic compensation, insufficient therapeutic windows, microenvironment complexity, and unintended immune suppression.

Most concerning, nutrient restriction creates selection pressure that favors cancer cells with maximal metabolic adaptability. The pre-existing metabolically flexible subpopulations within tumors that survive metabolic stress are often more aggressive, more stem-like, more resistant to therapy, and more capable of metastasis than the bulk tumor cells eliminated. Rather than starving cancer, metabolic restriction may inadvertently select for the most dangerous cancer cells while simultaneously undermining the immune system that represents the body’s best defense.

This does not mean that metabolism is irrelevant to cancer therapy. Success requires moving beyond simple restriction strategies toward approaches that account for metabolic heterogeneity, identify absolute dependencies in specific genetic contexts, combine multiple pathway inhibitors to prevent compensation, optimize timing and dosing to minimize immune impairment, and develop tumor-targeted delivery to spare normal tissues. The goal should not be to indiscriminately starve cancer, but to exploit specific vulnerabilities while preserving the metabolic fitness of anti-tumor immunity. Until such sophisticated approaches mature, broad metabolic restriction strategies risk doing more harm than good, representing a cautionary tale of how appealing biological concepts can fail when confronted with the complexity of cancer and the body’s attempts to fight it.

Beyond Metabolic Restriction: The Inescapable Vulnerability of Oxidative Stress

While cancer cells escape nutrient restriction through metabolic flexibility, they cannot escape oxidative stress through the same adaptive mechanisms. Cancer cells operate near the upper limits of their tolerance for oxidative stress, with minimal spare capacity, whereas normal cells operate at less than half their maximum capacity. This creates a substantial therapeutic window that metabolic restriction approaches have not achieved. Cancer cells maintain markedly elevated baseline levels of reactive oxygen species due to cancer-driven metabolism and elevated fuel-burning rates, yet cannot reduce this oxidative burden without losing the growth signals that reactive oxygen species provide. This traps them in dependence on a primary antioxidant defense system that is already working at maximum capacity under normal conditions.

The quantitative differences between cancer and normal cells create a therapeutic opportunity that metabolic approaches cannot match. Cancer cells typically maintain reactive oxygen species levels that are 2-4 times higher than those of normal cells due to their altered metabolism and mitochondrial dysfunction. Despite upregulating their antioxidant systems, including elevated glutathione, thioredoxin, and constitutive NRF2 activation, cancer cells operate at 70-80% of their oxidative capacity just to manage baseline stress. Normal cells, by contrast, function at only 40-50% of their capacity, maintaining substantial reserves. This difference means that modest increases in oxidative stress that push cancer cells past their operational limits are easily tolerated by normal tissues, creating a 5-7-fold therapeutic window. Studies consistently demonstrate that normal cells can tolerate one 150-200% increase in baseline reactive oxygen species before experiencing cell death, while cancer cells succumb to increases of just 20-30%.

Unlike metabolic pathways, in which cancer cells can switch among numerous fuel sources, they cannot activate backup antioxidant systems when their primary defenses are overwhelmed. The structural constraints that prevent adaptation to oxidative stress are insurmountable. All antioxidant systems ultimately require NADPH for function, including glutathione reductase, thioredoxin reductase, and GPX4. Cancer cells already maximize their NADPH production through the pentose phosphate pathway and alternative sources to support both biosynthesis and antioxidant defense, and they cannot increase production beyond 30-40% without catastrophically compromising the anabolic processes required for proliferation. Additionally, cysteine availability limits glutathione synthesis, and the SLC7A11 transporter is already operating at maximal capacity in most cancers. The rate-limiting enzyme for glutathione synthesis cannot be rapidly upregulated, and alternative pathways provide insufficient flux. Most critically, GPX4 is the sole enzyme capable of reducing complex lipid hydroperoxides within membranes. Unlike other redundant antioxidant enzymes, GPX4 operates as a monomer with no backup systems, and cancer cells cannot rapidly synthesize more of it under acute stress.

The temporal dynamics of oxidative damage make adaptation impossible in ways that nutrient restriction does not face. While nutrient restriction allows gradual transcriptional adaptation over six to twenty-four hours, with cells temporarily arresting growth to adjust metabolism, oxidative stress causes damage within seconds. Lipid peroxidation is autocatalytic, with a single hydroxyl radical initiating chain reactions that generate hundreds of lipid hydroperoxides before antioxidant systems can intervene. This explosive kinetics means that damage accumulates faster than any transcriptional response can be activated. The level of oxidative stress increases directly with the rate at which cells burn fuel, regardless of which fuel they use, indicating that cancer cells cannot escape this vulnerability by changing energy sources, as they do when evading nutrient restriction. When reactive oxygen species production increases from external sources or when antioxidant defenses are depleted, cancer cells cross their threshold and undergo catastrophic failure, while normal cells with substantial reserve capacity tolerate the same stress.

Paradoxically, cancer stem cells that resist conventional therapies through metabolic flexibility and quiescence are more vulnerable to oxidative stress due to their dependence on iron and membrane lipids, thereby creating a therapeutic opportunity that nutrient restriction strategies cannot exploit. These cells exhibit lower reactive oxygen species levels than differentiated cancer cells due to enhanced antioxidant machinery, yet they accumulate higher iron levels than both normal stem cells and bulk cancer cells. This iron dependency, which is required for stemness maintenance and metabolic functions, sensitizes cancer stem cells to ferroptosis despite their robust antioxidant defenses. Mesenchymal-like cancer stem cells show elevated expression of ACSL4, enriching their membranes with polyunsaturated fatty acid phospholipids that serve as substrates for ferroptotic lipid peroxidation. Multiple studies demonstrate that therapy-resistant cancer stem cell populations, including those from triple-negative breast cancer and glioblastoma, preferentially succumb to ferroptosis. This represents a major therapeutic advance, as cancer stem cells evade conventional therapies through quiescence and drug efflux pumps, but cannot escape iron-dependent oxidative death once lipid peroxidation is initiated.

Conclusion

Metabolic adaptations that help tumor cells grow and survive make an attractive target for cancer therapy, yet this strategy faces two major obstacles: cancer cells are remarkably adaptable in how they fuel themselves, with wide variation from one tumor to another and even within the same tumor, and the margin between a dose that harms cancer and a dose that harms healthy tissue is dangerously thin, because cancer cells share many of the same nutrient needs as normal cells, including the immune cells that fight cancer. The fundamental contrast between cancer’s metabolic adaptability and its inability to withstand oxidative stress reveals why targeting oxidative stress represents a more promising therapeutic strategy than metabolic restriction. Cancer cells readily sidestep nutrient deprivation by switching to alternative fuel sources, rewiring their internal machinery, and drawing support from surrounding tissues. They can temporarily slow their growth, recycle their own components for energy, and emerge from metabolic stress even stronger and more adaptable than before. By contrast, oxidative damage above a certain threshold triggers irreversible destruction inside the cancer cell that cannot be bypassed through backup pathways or workarounds. The built-in limits of cancer’s antioxidant defenses, the explosive speed at which oxidative damage spreads through cell membranes, and the inability of cancer cells to quickly build replacement defenses create a vulnerability they cannot escape. While starvation-based strategies have largely failed in the clinic because cancer is so metabolically flexible and shares so many nutrient needs with healthy tissue, approaches that push cancer past its narrow oxidative tolerance offer a real therapeutic window rooted in biological limits that cancer cannot adapt its way out of.

References

1. Andrejeva G, Rathmell JC. Similarities and distinctions of cancer and immune metabolism in inflammation and tumors. Cell Metab. 2017;26(1):49-70.
2. Baracos VE, Martin L, Korc M, Guttridge DC. Cancer-associated cachexia. Nat Rev Dis Primers. 2018;4:17105.
3. Best SA, Gubser PM, Sethumadhavan S, et al. Glutaminase inhibition impairs CD8 T cell activation in STK11-/Lkb1-deficient lung cancer. Cell Metab. 2022;34(6):874-887.
4. Bian Y, Li W, Kremer DM, et al. Cancer SLC43A2-mediated methionine limitation supports major histocompatibility complex class I antigen presentation through mitochondrial metabolism. Nature. 2020;585(7824):277-282.
5. Brandl N, Seitz R, Sendtner N, Müller M, Gülow K. Living on the Edge: ROS Homeostasis in Cancer Cells and Its Potential as a Therapeutic Target. Antioxidants (Basel). 2025 Aug 16;14(8):1002.
6. Caro P, Kishan AU, Norberg E, et al. Metabolic signatures uncover distinct targets in molecular subsets of diffuse large B cell lymphoma. Cancer Cell. 2012;22(4):547-560.
7. Chang CH, Curtis JD, Maggi LB, et al. Posttranscriptional control of T cell effector function by aerobic glycolysis. Cell. 2013;153(6):1239-1251.
8. Chang CH, Qiu J, O’Sullivan D, et al. Metabolic competition in the tumor microenvironment is a driver of cancer progression. Cell. 2015;162(6):1229-1241.
9. Christen S, Lorendeau D, Schmieder R, et al. Breast cancer-derived lung metastases show increased pyruvate carboxylase-dependent anaplerosis. Cell Rep. 2016;17(3):837-848.
10. Ciccarese F, Ciminale V. Escaping Death: Mitochondrial Redox Homeostasis in Cancer Cells. Front Oncol. 2017 Jun 14;7:117.
11. Commisso C, Davidson SM, Soydaner-Azeloglu RG, et al. Macropinocytosis of protein is an amino acid supply route in Ras-transformed cells. Nature. 2013;497(7451):633-637.
12. DeBerardinis RJ, Mancuso A, Daikhin E, et al. Beyond aerobic glycolysis: transformed cells can engage in glutamine metabolism that exceeds the requirement for protein and nucleotide synthesis. Proc Natl Acad Sci U S A. 2007;104(49):19345-19350.
13. de Groot S, Lugtenberg RT, Cohen D, Welters MJP, Ehsan I, Vreeswijk MPG, Smit VTHBM, de Graaf H, Heijns JB, Portielje JEA, van de Wouw AJ, Imholz ALT, Kessels LW, Vrijaldenhoven S, Baars A, Kranenbarg EM, Carpentier MD, Putter H, van der Hoeven JJM, Nortier JWR, Kroep JR. Fasting mimicking diet as an adjunct to neoadjuvant chemotherapy for breast cancer in the multicentre randomized phase 2 DIRECT trial. Nat Commun. 2020;11:3083.
14. DeNicola GM, Karreth FA, Humpton TJ, Gopinathan A, Wei C, Frese K, Mangal D, Yu KH, Yeo CJ, Calhoun ES, Scrimieri F, Winter JM, Hruban RH, Iacobuzio-Donahue C, Kern SE, Blair IA, Tuveson DA. Oncogene-induced Nrf2 transcription promotes ROS detoxification and tumorigenesis. Nature. 2011 Jul 6;475(7354):106-9.
15. Du D, Qin M, Shi L, Liu C, Jiang J, Liao Z, Wang H, Zhang Z, Sun L, Fan H, Liu Z, Yu H, Li H, Peng J, Yuan S, Yang M, Xiong J. RNA binding motif protein 45-mediated phosphorylation enhances protein stability of ASCT2 to promote hepatocellular carcinoma progression. Oncogene. 2023;42(42):3127-3141.
16. Dung NT, Susukida T, Ucche S, He K, Sasaki SI, Hayashi R, Hayakawa Y. Calorie Restriction Impairs Anti-Tumor Immune Responses in an Immunogenic Preclinical Cancer Model. Nutrients. 2023 Aug 18;15(16):3638.
17. Ecker BL, Kaur A, Douglass SM, et al. Age-related changes in HAPLN1 increase lymphatic permeability and affect routes of melanoma metastasis. Cancer Discov. 2019;9(1):82-95.
18. Faubert B, Li KY, Cai L, et al. Lactate metabolism in human lung tumors. Cell. 2017;171(2):358-371.
19. Faubert B, Solmonson A, DeBerardinis RJ. Metabolic reprogramming and cancer progression. Science. 2020;368(6487):eaaw5473.
20. Fearon K, Strasser F, Anker SD, et al. Definition and classification of cancer cachexia: an international consensus. Lancet Oncol. 2011;12(5):489-495.
21. Fendt SM, Frezza C, Erez A. Targeting metabolic plasticity and flexibility dynamics for cancer therapy. Cancer Discovery. 2020;10(12):1797-1807.
22. Ferrer M, Mourikis N, Davidson EE, et al. Ketogenic diet promotes tumor ferroptosis but induces relative corticosterone deficiency that accelerates cachexia. Cell Metab. 2023;35(7):1147-1162.
23. Gao X, Sanderson SM, Dai Z, et al. Dietary methionine influences therapy in mouse cancer models and alters human metabolism. Nature. 2019;572(7769):397-401.
24. Gouda MA, Voss MH, Tawbi H, Gordon M, Tykodi SS, Lam ET, Vaishampayan U, Tannir NM, Chaves J, Nikolinakos P, Fan A, Lee R, McDermott D, Shapiro GI, Gandhi L, Bhatia S, Katragadda V, Meric-Bernstam F. A phase I/II study of the safety and efficacy of telaglenastat (CB-839) in combination with nivolumab in patients with metastatic melanoma, renal cell carcinoma, and non-small-cell lung cancer. ESMO Open. 2025;10(5):104536.
25. Harris IS, Treloar AE, Inoue S, Sasaki M, Gorrini C, Lee KC, Yung KY, Brenner D, Knobbe-Thomsen CB, Cox MA, Elia A, Berger T, Cescon DW, Adeoye A, Brüstle A, Molyneux SD, Mason JM, Li WY, Yamamoto K, Wakeham A, Berman HK, Khokha R, Done SJ, Kavanagh TJ, Lam CW, Mak TW. Glutathione and thioredoxin antioxidant pathways synergize to drive cancer initiation and progression. Cancer Cell. 2015 Feb 9;27(2):211-22.
26. Hayes JD, Dinkova-Kostova AT, Tew KD. Oxidative Stress in Cancer. Cancer Cell. 2020 Aug 10;38(2):167-197.
27. Ho PC, Bihuniak JD, Macintyre AN, et al. Phosphoenolpyruvate is a metabolic checkpoint of anti-tumor T cell responses. Cell. 2015;162(6):1217-1228.
28. Hoover G, Gilbert S, Curley O, Obellianne C, Lin MT, Hixson W, Pierce TW, Andrews JF, Alexeyev MF, Ding Y, Bu P, Behbod F, Medina D, Chang JT, Ayala G, Grelet S. Nerve-to-cancer transfer of mitochondria during cancer metastasis. Nature. 2025 Aug;644(8075):252-262.
29. Huang F, Zhao Y, Zhao J, et al. Upregulated SLC1A5 promotes cell growth and survival in colorectal cancer. Int J Clin Exp Pathol. 2014;7(9):6006-6014.
30. Ikeda H, Kawase K, Nishi T, Watanabe T, Takenaga K, Inozume T, Ishino T, Aki S, Lin J, Kawashima S, Nagasaki J, Ueda Y, Suzuki S, Makinoshima H, Itami M, Nakamura Y, Tatsumi Y, Suenaga Y, Morinaga T, Honobe-Tabuchi A, Ohnuma T, Kawamura T, Umeda Y, Nakamura Y, Kiniwa Y, Ichihara E, Hayashi H, Ikeda JI, Hanazawa T, Toyooka S, Mano H, Suzuki T, Osawa T, Kawazu M, Togashi Y. Immune evasion through mitochondrial transfer in the tumour microenvironment. Nature. 2025 Feb;638(8049):225-236. doi: 10.1038/s41586-024-08439-0. Epub 2025 Jan 22. Erratum in: Nature. 2025 Mar;639(8053):E5. Erratum in: Nature. 2025 Aug;644(8076):E33.
31. Jang C, Chen L, Rabinowitz JD. Metabolomics and isotope tracing. Cell. 2018;173(4):822-837.
32. Jeon YJ, Khelifa S, Ratnikov B, Scott DA, Feng Y, Parisi F, et al. Regulation of glutamine carrier proteins by RNF5 determines breast cancer response to ER stress-inducing chemotherapies. Cancer Cell. 2015;27(3):354-369.
33. Ji M, Xu X, Xu Q, Hsiao YC, Martin C, Ukraintseva S, Popov V, Arbeev KG, Randall TA, Wu X, Garcia-Peterson LM, Liu J, Xu X, Andrea Azcarate-Peril M, Wan Y, Yashin AI, Anantharaman K, Lu K, Li JL, Shats I, Li X. Methionine restriction-induced sulfur deficiency impairs antitumour immunity partially through gut microbiota. Nat Metab. 2023 Sep;5(9):1526-1543.
34. Johnson MO, Wolf MM, Madden MZ, et al. Distinct regulation of Th17 and Th1 cell differentiation by glutaminase-dependent metabolism. Cell. 2018;175(7):1780-1795.
35. Keckesova Z, Donaher JL, De Cock J, et al. LACTB is a tumour suppressor that modulates lipid metabolism and cell state. Nature. 2017;543(7647):681-686.
36. Kim LC, Cook RS, Chen J. mTORC1 and mTORC2 in cancer and the tumor microenvironment. Oncogene. 2017;36(16):2191-2201.
37. Leone RD, Zhao L, Englert JM, et al. Glutamine blockade induces divergent metabolic programs to overcome tumor immune evasion. Science. 2019;366(6468):1013-1021.
38. Li S, Zhang Y, Tong H, Sun H, Liao H, Li Q, Ma X. Metabolic regulation of immunity in the tumor microenvironment. Cell Rep. 2025 Nov 25;44(11):116463.
39. Liou GY, Storz P. Reactive oxygen species in cancer. Free Radic Res. 2010 May;44(5):479-96.
40. Lu J, Zheng X, Li F, et al. Tunneling nanotubes promote intercellular mitochondria transfer followed by increased invasiveness in bladder cancer cells. Oncotarget. 2017;8(9):15539-15552.
41. Luo C, Balsa E, Thomas A, et al. ERRα maintains mitochondrial oxidative metabolism and constitutes an actionable target in PGC1α-elevated melanomas. Mol Cancer Res. 2017;15(10):1366-1375.
42. Ma EH, Bantug G, Griss T, et al. Serine is an essential metabolite for effector T cell expansion. Cell Metab. 2017;25(2):345-357.
43. Ma EH, Verway MJ, Johnson RM, et al. Metabolic profiling using stable isotope tracing reveals distinct patterns of glucose utilization by physiologically activated CD8 T cells. Immunity. 2019;51(5):856-870.
44. Marlein CR, Zaitseva L, Piddock RE, et al. NADPH oxidase-2 derived superoxide drives mitochondrial transfer from bone marrow stromal cells to leukemic blasts. Blood. 2017;130(14):1649-1660.
45. Martin L, Birdsell L, Macdonald N, et al. Cancer cachexia in the age of obesity: skeletal muscle depletion is a powerful prognostic factor, independent of body mass index. J Clin Oncol. 2013;31(12):1539-1547.
46. Mashimo T, Pichumani K, Vemireddy V, et al. Acetate is a bioenergetic substrate for human glioblastoma and brain metastases. Cell. 2014;159(7):1603-1614.
47. Mayers JR, Torrence ME, Danai LV, et al. Tissue of origin dictates branched-chain amino acid metabolism in mutant Kras-driven cancers. Science. 2016;353(6304):1161-1165.
48. Moschoi R, Imbert V, Nebout M, et al. Protective mitochondrial transfer from bone marrow stromal cells to acute myeloid leukemic cells during chemotherapy. Blood. 2016;128(2):253-264.
49. Panieri E, Santoro MM. ROS homeostasis and metabolism: a dangerous liason in cancer cells. Cell Death Dis. 2016 Jun 9;7(6):e2253.
50. Pascual G, Avgustinova A, Mejetta S, et al. Targeting metastasis-initiating cells through the fatty acid receptor CD36. Nature. 2017;541(7635):41-45.
51. Pasquier J, Guerrouahen BS, Al Thawadi H, et al. Preferential transfer of mitochondria from endothelial to cancer cells through tunneling nanotubes modulates chemoresistance. J Transl Med. 2013;11:94.
52. Possemato R, Marks KM, Shaul YD, et al. Functional genomics reveal that the serine synthesis pathway is essential in breast cancer. Nature. 2011;476(7360):346-350.
53. Ramanathan A, Wang C, Schreiber SL. Perturbational profiling of a cell-line model of tumorigenesis by using metabolic measurements. Proc Natl Acad Sci U S A. 2005;102(17):5992-5997.
54. Recalcati S, Gammella E, Cairo G. Dysregulation of iron metabolism in cancer stem cells. Free Radic Biol Med. 2019 Mar;133:216-220.
55. Rohrig F, Schulze A. The multifaceted roles of fatty acid synthesis in cancer. Nat Rev Cancer. 2016;16(11):732-749.
56. Saxton RA, Sabatini DM. mTOR signaling in growth, metabolism, and disease. Cell. 2017;168(6):960-976.
57. Schug ZT, Peck B, Jones DT, et al. Acetyl-CoA synthetase 2 promotes acetate utilization and maintains cancer cell growth under metabolic stress. Cancer Cell. 2015;27(1):57-71.
58. Schumacker PT. Reactive oxygen species in cancer cells: live by the sword, die by the sword. Cancer Cell. 2006 Sep;10(3):175-6.
59. Sheppard S, Srpan K, Lin W, Lee M, Delconte RB, Owyong M, Carmeliet P, Davis DM, Xavier JB, Hsu KC, Sun JC. Fatty acid oxidation fuels natural killer cell responses against infection and cancer. Proc Natl Acad Sci U S A. 2024 Mar 12;121(11):e2319254121.
60. Sinclair LV, Rolf J, Emslie E, et al. Control of amino-acid transport by antigen receptors coordinates the metabolic reprogramming essential for T cell differentiation. Nat Immunol. 2013;14(5):500-508.
61. Stine ZE, Schug ZT, Salvino JM, Dang CV. Targeting cancer metabolism in the era of precision oncology. Nat Rev Drug Discov. 2022;21(2):141-162.
62. Su H, Yang F, Fu R, Li X, French R, Mose E, Pu X, Trinh B, Kumar A, Liu J, Antonucci L, Todoric J, Liu Y, Hu Y, Diaz-Meco MT, Moscat J, Metallo CM, Lowy AM, Sun B, Karin M. Cancer cells escape autophagy inhibition via NRF2-induced macropinocytosis. Cancer Cell. 2021;39(5):678-693.e11.
63. Su Z, Liu Y, Xia Z, Rustgi AK, Gu W. An unexpected role for the ketogenic diet in triggering tumor metastasis by modulating BACH1-mediated transcription. Sci Adv. 2024 Jun 7;10(23):eadm9481.
64. Sullivan MR, Danai LV, Lewis CA, et al. Quantification of microenvironmental metabolites in murine cancers reveals determinants of tumor nutrient availability. Elife. 2019;8:e44235.
65. Tannir NM, Agarwal N, Porta C, et al. Efficacy and safety of telaglenastat plus cabozantinib versus placebo plus cabozantinib in patients with advanced renal cell carcinoma: the CANTATA randomized clinical trial. JAMA Oncol. 2022;8(10):1411-1418.
66. Tisdale MJ. Mechanisms of cancer cachexia. Physiol Rev. 2009;89(2):381-410.
67. Trachootham D, Alexandre J, Huang P. Targeting cancer cells by ROS-mediated mechanisms: a radical therapeutic approach? Nat Rev Drug Discov. 2009 Jul;8(7):579-91.
68. Vander Heiden MG, Cantley LC, Thompson CB. Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science. 2009;324(5930):1029-1033.
69. von Haehling S, Anker SD. Prevalence, incidence and clinical impact of cachexia: facts and numbers—update 2014. J Cachexia Sarcopenia Muscle. 2014;5(4):261-263.
70. Watson MJ, Vignali PDA, Mullett SJ, et al. Metabolic support of tumor-infiltrating regulatory T cells by lactic acid. Nature. 2021;591(7851):645-651.
71. Weinberg F, Hamanaka R, Wheaton WW, et al. Mitochondrial metabolism and ROS generation are essential for Kras-mediated tumorigenicity. Proc Natl Acad Sci U S A. 2010;107(19):8788-8793.
72. Wise DR, DeBerardinis RJ, Mancuso A, et al. Myc regulates a transcriptional program that stimulates mitochondrial glutaminolysis and leads to glutamine addiction. Proc Natl Acad Sci U S A. 2008;105(48):18782-18787.
73. Yang L, Venneti S, Nagrath D. Glutaminolysis: a hallmark of cancer metabolism. Annu Rev Biomed Eng. 2017;19:163-194.
74. Yang Z, Follett J, Kerr MC, Clairfeuille T, Chandra M, Collins BM, Teasdale RD. Sorting nexin 27 (SNX27) regulates the trafficking and activity of the glutamine transporter ASCT2. J Biol Chem. 2018;293(18):6802-6811.
75. Yuneva MO, Fan TW, Allen TD, et al. The metabolic profile of tumors depends on both the responsible genetic lesion and tissue type. Cell Metab. 2012;15(2):157-170.
76. Zhang H, Yu X, Ye J, et al. Systematic investigation of mitochondrial transfer between cancer cells and T cells at single-cell resolution. Cancer Cell. 2023;41(10):1788-1802.
77. Zhang J, Pavlova NN, Thompson CB. Cancer cell metabolism: the essential role of the nonessential amino acid, glutamine. EMBO J. 2017;36(10):1302-1315.