Starvation on a Petri Dish: How Fasting Rewires Cancer Cells and Exposes Their Weaknesses

A systematic review published in Medical Oncology has pulled together the laboratory evidence behind one of oncology’s most tantalizing ideas: that fasting, or drugs that mimic it, can reprogram the metabolism of cancer cells in ways that make them easier to kill. The review, led by Rihab Bensalek and colleagues at Ibn Tofail University in Kenitra, Morocco, followed PRISMA guidelines to identify in vitro studies in which cancer cell lines were exposed to fasting-mimicking conditions, serum starvation, or nutrient restriction. Eleven studies made the final cut, spanning a range of tumor types, and their findings converge on a strikingly consistent mechanistic picture. When nutrients run low, malignant cells appear to lose the metabolic flexibility that normal tissues retain, and the molecular consequences ripple through the very signaling networks that drive tumor growth.

The central logic of the fasting-cancer hypothesis rests on a fundamental difference between healthy and malignant tissue. Cancer cells are locked into an anabolic mode: they consume glucose voraciously, channel it through glycolysis even when oxygen is abundant, and divert the resulting carbon skeletons into the biosynthesis of lipids, proteins, and nucleotides needed for relentless division. This metabolic inflexibility, often described as a hallmark of cancer, is precisely what fasting is thought to exploit. Normal cells can switch fuel sources, suppress growth signaling, and enter protective maintenance modes during nutrient scarcity. Transformed cells, wired for growth by oncogenic mutations, often cannot. The review’s authors set out to determine whether the published cell-culture evidence actually supports this systems-level vulnerability, and whether the molecular details hold together across independent laboratories and models.

What emerged from the eleven eligible studies was a remarkably coherent signaling signature. Across diverse cancer models, fasting conditions consistently reduced signaling through IGF-1, the insulin-like growth factor that acts as a master growth driver, and this reduction cascaded downstream into inhibition of the PI3K/AKT/mTOR pathway. That pathway is the cell’s central anabolic command center: when active, it promotes protein synthesis, lipid production, cell growth, and survival. Suppressing it effectively pulls the metabolic rug out from under the tumor cell. In parallel, the studies documented suppression of glycolysis, the glucose-hungry pipeline that many cancers depend upon, and activation of AMPK, the AMP-activated protein kinase that functions as the cell’s energy sensor. When AMPK detects a falling energy charge, it shuts down biosynthetic programs and initiates stress responses, forcing the cell to conserve rather than build.

The functional consequences of this reprogramming were measurable in the culture dish. Cancer cells exposed to nutrient restriction showed impaired proliferative capacity, growing more slowly and in some cases undergoing apoptosis as their anabolic demands collided with dwindling supplies. More clinically interesting, however, was the second finding: under fasting-mimicking conditions, cancer cells became more sensitive to chemotherapeutic agents. This sensitization effect is the crux of the translational promise. Chemotherapy works best when tumor cells are already stressed, unable to mount their usual repair and survival responses. By depleting the metabolic headroom that malignant cells use to buffer damage, fasting conditions appear to lower the threshold at which drug-induced stress becomes lethal.

The individual studies assembled in the review illustrate how this principle plays out across different drugs and tumor contexts. Work on pancreatic cancer models showed that fasting cycles potentiate the efficacy of gemcitabine, a mainstay chemotherapy for that disease. In breast cancer, fasting-mimicking diets have been reported to induce regression in combination with hormone therapy and to block the escape of triple-negative cancer stem cells, an especially stubborn cell population implicated in relapse. Gallbladder cancer cells were resensitized to gemcitabine through fasting-induced expression of RNF152, which inhibits mTORC1-mediated glycolysis. In prostate cancer, caloric restriction enhanced antiandrogen therapy by inhibiting the translation of the androgen receptor. Studies in oral cancer reported that fasting-mimicking conditions enhanced the efficacy of EGFR-targeted tyrosine kinase inhibitors, while work in hepatocellular carcinoma showed that fasting potentiated the anticancer activity of sorafenib. Starvation-induced activation of the ATM/Chk2/p53 DNA damage response pathway has been shown to sensitize cells to cisplatin, and combined intermittent fasting with ERK inhibition amplified chemotherapy effects through the GSK3β-SIRT7 axis.

Not all of the mechanisms are cell-autonomous. Some of the cited work points to effects on the tumor immune microenvironment, including evidence that fasting-mimicking diets reduce heme oxygenase-1 to promote T cell-mediated tumor cytotoxicity, and that their benefit in oral cancer involves crosstalk with tumor-associated macrophages. Other studies implicate non-coding RNA regulation, such as the lncRNA RBM5-AS1/GCN5 axis, whose targeting under fasting conditions reprograms glycolysis and induces apoptosis in ovarian cancer cells. This breadth matters because it suggests that fasting does not act through a single lever but through a coordinated, multi-layered reprogramming that touches growth signaling, energy sensing, stress responses, epigenetic regulation, and immune interactions simultaneously. It is this systems-level character that the review’s authors emphasize as the distinguishing feature of the approach.

The review is candid about its limitations, and these deserve attention. The evidence base is entirely in vitro, meaning the findings come from cells in dishes rather than living organisms, and cell culture strips away the hormonal, immune, and metabolic complexity of a fasting body. The eleven included studies also showed considerable methodological heterogeneity: different cell lines, different fasting-mimicking protocols, different durations of nutrient deprivation, and different endpoints. Risk of bias was assessed across predefined methodological domains, but the authors acknowledge that standardization is lacking. Serum starvation in a dish, for example, is a crude approximation of the physiological fasting state, in which specific nutrients, growth factors, and metabolites fluctuate in coordinated patterns. Translating a dish-based sensitization effect into a safe clinical protocol requires demonstrating that the tumor-selective stress can be achieved in patients without harming normal tissues, a question that only animal models and carefully designed trials can answer.

Even so, the mechanistic convergence across independent models is the review’s most persuasive contribution. Eleven studies, conducted in different laboratories on different cancers, arrived at overlapping conclusions: reduced IGF-1 signaling, inhibited PI3K/AKT/mTOR activity, suppressed glycolysis, and AMPK-mediated stress responses. In a field where individual preclinical findings often fail to replicate, this kind of convergence strengthens the case that fasting-induced metabolic reprogramming is a real and reproducible biological phenomenon rather than an artifact of any single experimental system. It also provides a mechanistic roadmap for the next generation of experiments, identifying the specific nodes, IGF-1, mTORC1, AMPK, and glycolytic flux, where combination strategies could be rationally designed.

The clinical context gives the work its urgency. Cancer remains a leading cause of death worldwide, with GLOBOCAN estimates for 2022 documenting millions of new cases across 185 countries, and metabolic targeting represents one of several fronts in the effort to improve outcomes. Dietary interventions are attractive because they are conceptually simple and potentially complementary to existing drugs, and fasting-mimicking diets have already been tested alongside hormone therapy and chemotherapy in preclinical and early clinical settings. But the review’s authors are careful to frame fasting as a strategy that warrants standardized translational investigation, not a ready-made treatment. The gap between a sensitized cancer cell in a nutrient-poor dish and a safely fasted patient remains substantial, and premature enthusiasm for unsupervised fasting in cancer patients carries real risks, including weight loss and cachexia, which are themselves associated with worse outcomes.

What the study ultimately delivers is a disciplined synthesis of a fast-moving literature, organized around mechanism rather than hype. It establishes that, at the cellular level, nutrient deprivation exposes intrinsic anabolic dependencies of malignant cells and that this exposure can be quantified in terms of defined signaling pathways and drug interactions. The next steps it implies are clear: standardized fasting-mimicking protocols, rigorous in vivo validation of the sensitization effects, identification of the tumor types most dependent on the affected pathways, and clinical trials that pair metabolic interventions with the chemotherapies and targeted agents the cell-culture work has flagged as most promising. If those steps succeed, the humble act of withholding nutrients, or pharmacologically mimicking it, could become a genuine component of combination cancer therapy rather than a speculative idea confined to the laboratory bench.

Subject of Research: Fasting-induced metabolic reprogramming and chemosensitization in cancer cell models

Article Title: Fasting-induced metabolic reprogramming as a therapeutic vulnerability in cancer: a systematic mechanistic review of in vitro evidence

Article References: Bensalek, R., Hassani Idrissi, H., Ait Baha, A., Hicham, M., Akhouayri, O., Habbane, M., & Benazzouz, B. (2026). Fasting-induced metabolic reprogramming as a therapeutic vulnerability in cancer: a systematic mechanistic review of in vitro evidence. Medical Oncology, 43(11), Article 301. https://doi.org/10.1007/s12032-026-03391-7

Image Credits: AI Generated

DOI: 10.1007/s12032-026-03391-7

Keywords: fasting, cancer metabolism, fasting-mimicking diet, IGF-1, PI3K/AKT/mTOR, AMPK, glycolysis, chemotherapy sensitization, systematic review, in vitro studies, nutrient restriction, tumor metabolism