A Ketogenic Diet Accelerated Intestinal Tumors — But the Culprit Is Not What We Thought!
In this article
Not Ketones, but Fat Burning
Nature published a new study examining, at a mechanistic level, the relationship between a popular eating pattern, the ketogenic diet, and intestinal tumor formation. The finding has two layers, and the second is more interesting than the first:
First: In mice that spontaneously develop intestinal adenomas (benign polyps with the potential to become cancerous), a ketogenic diet increased tumor burden and shortened survival.
Second — and the real surprise: The effect was not caused by the most discussed product of a ketogenic diet: ketone bodies. The researchers genetically both decreased and increased ketone production and also disrupted ketone use — tumor formation did not change at all. The culprit was elsewhere: the intracellular burning of dietary fat (fatty acid oxidation). This finding changes the framing of the ketogenic diet debate.
News like this easily spreads as "a ketogenic diet causes cancer." This interpretation would be wrong. Let us be clear:
- This is a mouse study. It is not a clinical study showing that a ketogenic diet increases intestinal cancer risk in humans.
- The mice used are genetically predisposed models — animals already programmed to develop intestinal polyps spontaneously. The same result has not been demonstrated in a healthy organism.
- The subject is tumor initiation (formation), not treatment of existing cancer. These are entirely different questions and may have different answers.
- A ketogenic diet has proven medical uses (especially refractory epilepsy), and this study does not invalidate them.
- No patient is advised to change their diet based on this article.
A ketogenic diet is an eating pattern that severely restricts carbohydrate intake while greatly increasing the proportion of fat. When carbohydrates decrease, the body turns to fat for energy, and the liver produces ketone bodies. This state is called ketosis .
The level of evidence varies greatly by use:
• Refractory epilepsy (especially in children): its most established and proven medical use. • Weight loss and blood sugar control: data suggest it can be effective in the short term; long-term results are debated. • Cancer: remains investigational. It is not a standard treatment and does not appear in guidelines as a cancer treatment.
This study concerns the third topic — but the aspect of tumor emergence, rather than treatment.
The research began with a genuine scientific contradiction.
On the one hand, it had previously been shown that a high-fat diet stimulates intestinal stem cells and increases tumor formation. This seems like bad news for a ketogenic diet — because it contains an even higher proportion of fat .
On the other hand, a ketogenic diet differs from a conventional high-fat diet in two important ways: it lowers blood insulin levels and triggers ketone production. Insulin is a powerful signal promoting cell proliferation; lowering it may theoretically be protective. Beneficial effects of ketones have also been proposed in some contexts.
On the harmful side: the very high fat content. On the protective side: low insulin and ketone production. The question was which would dominate. And this question was particularly relevant to people with familial adenomatous polyposis (FAP) and other genetic causes of numerous intestinal polyps and a high risk of small intestinal tumors; for these people, it had practical significance.
In mouse models that spontaneously develop intestinal adenomas, the researchers combined dietary, genetic, and metabolic interventions. The strength of the study lies in going beyond a single observation and dismantling and testing the mechanism step by step .
- In mice fed a ketogenic diet, tumor burden increased.
- Survival shortened.
- And crucially, this effect was independent of ketone metabolites.
How was the conclusion "independent of ketones" reached? This is the most elegant part of the study.
- Ketone production was reduced. The key enzyme in ketone production, HMGCS2, was genetically suppressed → tumor formation did not change.
- Ketone production was increased. The same enzyme pathway was enhanced → again, no change.
- Ketone use (ketolysis) was disrupted. In other words, cells were prevented from using ketones as fuel → again, no change.
The conclusion is clear: Whichever direction ketone levels were shifted, tumor formation was unaffected. The most discussed product of a ketogenic diet was a bystander in this picture.
- PPARα, PPARδ, and PPARγ — these are intracellular fatty acid sensors; they detect the presence of fat and adjust the genetic program accordingly. When all three were disabled together in the intestine, the ketogenic diet's effects on stem cell proliferation, division, and colony-forming ability weakened.
- CPT1A — this enzyme controls fatty acid transport into mitochondria for their burning and is the rate-limiting step. Loss of CPT1A meant that adenoma formation, especially under a ketogenic diet, was limited.
And the most subtle distinction: The problem with fat in the cell was not its accumulation, but its burning. Thus, the issue is not "fat is stored and causes harm," but "fat is burned, and this burning process puts stem cells into a proliferation program."
For years, discussions of ketogenic diets placed ketones at the center — among both advocates and critics. "Entering ketosis" was presented as a goal, and measuring ketones as a sign of success. This study shows, at least in the context of intestinal tumor formation, that ketones are not the main character in this story .
If the culprit is not ketones but the fat load and its burning, the implication is that the reasoning "I am entering ketosis, so I am safe" is not valid, at least in this context. The diet's total fat content and type may matter more than the presence of ketosis.
This finding also suggests that, in the future, targets may exist to reduce risk without changing the diet completely — such as selectively limiting the fatty acid oxidation pathway. However, it must be emphasized that this is a research idea, not an applicable treatment.
An honest assessment also requires showing evidence in the opposite direction. The picture in this field is genuinely complex — the same diet can produce opposite results in different contexts.
This study: intestinal adenomas increased and survival shortened in genetically predisposed mice. Previous studies: a high-fat diet had been shown to stimulate intestinal stem cells and increase tumor formation. In addition, a ketogenic diet was reported to increase kidney tumor growth in a tuberous sclerosis complex model. In human epidemiology, a high-fat Western-style diet has also been consistently associated with colorectal cancer risk.
Together with targeted treatment: a ketogenic diet was shown in mice to enhance the efficacy of drugs targeting the PI3K pathway — by suppressing the drug-induced rise in blood sugar and insulin. But one crucial detail: in the same study, a ketogenic diet did not affect tumor growth on its own. Thus, the benefit came not from the diet itself but from its partnership with the drug. Adding a ketogenic diet to standard treatment is also being investigated in glioblastoma; studies so far are predominantly at the stage of assessing safety and feasibility .
The effect of a ketogenic diet in cancer is context-dependent, not one-directional. The variables that appear decisive are:
• Cancer type and genetic background — different outcomes have been reported in intestinal, kidney, and brain tumors. • Prevention or treatment? — influencing tumor emergence and treating an existing tumor are separate questions. • Alone or in combination? — the most favorable data come from scenarios combining the diet with a particular drug. • Host susceptibility — outcomes may differ in a genetically high-risk setting.
And most importantly: in cancer, ketogenic diets still have very limited human data. Systematic reviews have found only a small number of small clinical studies; most focus on safety and feasibility, rather than providing survival evidence.
The group emphasized by the study itself is genetically predisposed people. Foremost among these are people with familial adenomatous polyposis (FAP): those who develop hundreds to thousands of intestinal polyps because of an inherited gene alteration and have a markedly increased risk of small intestinal tumors.
Because, in these people, the first spark of tumor formation is already present; what remains is environmental factors that accelerate or slow it. The mouse models used mimic exactly this situation. Therefore, the most reasonable population to which the finding might translate is not the general public, but people with inherited polyposis and similar high-risk syndromes. In their conclusion, the study authors also emphasize that dietary strategies need more careful evaluation in genetically predisposed populations. This means not "it should be banned," but "it should not be recommended blindly" .
"Scientists have proven: a ketogenic diet causes cancer! The popular weight-loss diet triggers intestinal cancer."
In mice and in models genetically predisposed to developing polyps, a ketogenic diet accelerated adenoma formation. This has not been demonstrated in humans. The study's main contribution is not a warning but a mechanistic discovery: the effect is caused not by ketones but by fatty acid oxidation. This may point to a future target for intervention without changing the diet.
- It is an animal study. Mouse and human intestines differ in metabolism, microbiota, and lifespan scale. Translation of preclinical findings to humans frequently fails.
- Genetically predisposed models were used; findings cannot be generalized directly to healthy individuals.
- The subject is tumor formation, not cancer treatment. The effect of a ketogenic diet in existing cancer is not the subject of this study.
- The endpoint is adenoma (polyp) burden. An adenoma is not cancer; it is a precursor lesion with the potential to become cancerous.
- The types of dietary fat were not distinguished. The separate effects of saturated, monounsaturated, and polyunsaturated fats cannot be inferred from this study — a major gap in translating the findings to humans.
- This study provides no cohort data examining ketogenic diets and intestinal cancer risk in humans.
- The proven medical indications for a ketogenic diet have not been invalidated (especially refractory epilepsy).
- This article does not recommend starting or stopping any diet.
If you follow a ketogenic diet and are healthy: This study alone is not sufficient reason to stop your diet — because it provides no human data. However, it is worth remembering that the long-term effects of sustained, very high-fat eating patterns on intestinal health are still not fully known. If you plan to follow any restrictive diet long term, doing so under the supervision of a doctor and dietitian is the best approach.
If you have an inherited predisposition to intestinal cancer, such as familial polyposis or Lynch syndrome: This finding concerns you directly. If you are considering a very high-fat diet, be sure to make the decision together with the doctor following you. More importantly, no diet replaces regular colonoscopy and polyp surveillance.
If you are receiving cancer treatment: A ketogenic diet is not a proven cancer treatment, and following a severely restrictive diet during treatment may be harmful because of the risk of weight loss and muscle wasting. Nutritional decisions during treatment should be made together with the oncology team and an oncology dietitian.
The measures with the strongest evidence for preventing intestinal cancer have not changed: a fiber-rich diet, eating vegetables and fruit, limiting processed meat, weight management, regular physical activity, smoking cessation, limiting alcohol, and timely screening. Indeed, data indicate that a Mediterranean-style diet reduces intestinal polyps. For details, see our articles on healthy eating and cancer and nutrition .
The ketogenic diet is one of the world's most widely followed restrictive diets and is often discussed in cancer circles. This study does more than observe the diet's effect on intestinal tumor formation: it genetically dissects the mechanism and shows which component is responsible. The result invalidates a common assumption in the field — "ketones are central to the effect."
In mouse models that spontaneously develop intestinal adenomas, a ketogenic diet increased tumor burden and shortened survival. Neither suppressing nor enhancing HMGCS2, nor disrupting ketolysis, changed tumor formation — meaning ketone metabolism is not decisive. In contrast, combined intestinal loss of PPARα/δ/γ weakened ketogenic diet-induced stem cell expansion and proliferation; for fatty acid oxidation's rate-limiting enzyme, CPT1A loss meant that adenoma formation, especially under a ketogenic diet, was limited. The effect arises not from lipid accumulation but from lipid oxidation .
It is a preclinical (mouse) study; translation to humans has not been demonstrated. Genetically predisposed models were used. The endpoint is adenoma burden — an adenoma is not cancer. Fat types were not distinguished. The subject is tumor formation, not treatment. No cohort data examining ketogenic diets and colorectal cancer risk in humans were provided.
It changes no guideline today. However, it provides reasonable grounds for caution about very high-fat diets in people with inherited polyposis syndromes: such diets should not be recommended blindly. The main clinical message is that patients often act on the assumption "I am entering ketosis, so I am safe" — this study does not support that assumption, at least in this context. Restrictive diets during cancer treatment also require caution because of the risk of weight and muscle loss.
Does the same effect occur in humans? Does fat type (saturated/unsaturated) change the result? Is the effect specific to the small intestine, or does it also occur in the colon? Could selectively targeting fatty acid oxidation offer protection without changing the diet? Could dietary intervention studies be conducted in patients with inherited polyposis? Why does the picture look different in the treatment of existing tumors?
- Shay JES, Chi F, Tzouanas CN, et al. Ketogenic diet mediates intestinal tumorigenesis through lipids not ketones. Nature. 2026. doi:10.1038/s41586-026-10779-y.
- Beyaz S, Mana MD, Roper J, et al. High-fat diet enhances stemness and tumorigenicity of intestinal progenitors. Nature. 2016;531(7592):53–58. (A foundational study in the field.)
- Hopkins BD, Pauli C, Du X, et al. Suppression of insulin feedback enhances the efficacy of PI3K inhibitors. Nature. 2018;560:499–503. (Beneficial effect of a ketogenic diet combined with targeted treatment; the diet alone did not affect tumor growth.)
- Klement RJ, Sweeney RA. Ketogenic diets in medical oncology: a systematic review with focus on clinical outcomes. Med Oncol. 2020;37(2):14. (The limitations of clinical evidence.)
- Noch EK, et al. Insulin feedback is a targetable resistance mechanism of PI3K inhibition in glioblastoma. Neuro Oncol. 2023;25(12):2165–2176.
- Related reading (drozdogan.com): How does a high-fat diet increase colorectal cancer risk? · How does diet affect tumor metabolism? · Western-style diet and colorectal cancer · The Mediterranean diet and intestinal polyps · What is healthy eating? · Cancer and nutrition.
Editor's note: This content is for general information only; it is not medical advice or a recommendation for diagnosis, treatment, or diet. All data presented have been verified against primary sources and conveyed without direct quotation. This is a preclinical (mouse) study, not a clinical investigation in humans; findings from animal models frequently fail to translate to humans. The animals used are models genetically predisposed to developing intestinal adenomas, and results cannot be generalized to healthy individuals. The study's endpoint is adenoma (polyp) burden; an adenoma is not cancer, but a precursor lesion with the potential to become cancerous. The study examines tumor formation; the effect of a ketogenic diet in treating existing cancer is not the subject of this research, and these two questions are independent. The dietary fat types were not distinguished. Human data on ketogenic diets in cancer remain very limited, and existing clinical studies predominantly focus on safety and feasibility; no survival benefit has been demonstrated. Conversely, favorable preclinical data exist in some contexts, such as in combination with particular targeted treatments — the effect is context-dependent and not one-directional. The proven medical indications for a ketogenic diet (especially refractory epilepsy) have not been invalidated by this study. This article does not recommend starting or stopping any diet. Restrictive diets during cancer treatment carry a risk of weight and muscle loss and must be assessed under the supervision of the oncology team and an oncology dietitian. In people with an inherited predisposition to intestinal cancer, no nutritional approach can replace regular colonoscopy and polyp surveillance . For every decision about your health, consult a doctor who knows you.
Was this article helpful?
Your feedback helps us improve our content.
