Why a Low-Glycemic, Whole-Food, Plant-Based Diet Beats Keto and the Mediterranean Diet in Cancer

The case for the pattern with the strongest human evidence and the fewest trade-offs

Almost every person newly diagnosed with cancer asks some version of the same question: what should I eat? Three answers dominate the conversation. Ketogenic diets promise to starve tumors of glucose. The Mediterranean diet arrives with decades of respectable epidemiology. And then there is the least glamorous option, a low-glycemic, whole-food, plant-based (WFPB) pattern built on vegetables, legumes, intact whole grains, fruit, nuts and seeds, with animal products and refined carbohydrates minimized or eliminated.

The unglamorous option is the right one. Not because plants are morally superior, and not because sugar “feeds cancer” in the cartoon sense, but because when you line up the human evidence, not the mouse data and not the mechanism diagrams, the low-glycemic WFPB pattern is the only one of the three that is simultaneously supported by long-term outcome data, biologically coherent across multiple cancer pathways, and safe enough to sustain for the rest of a person’s life.

What the long-game human data actually shows

Start with the hardest evidence available: randomized trials with mortality endpoints.

The Women’s Health Initiative Dietary Modification trial randomized 48,835 postmenopausal women to a lower-fat, higher vegetable, fruit, and grain pattern. During the intervention, the results looked unimpressive. But at a median of 19.6 years of follow-up, deaths from breast cancer were significantly reduced (HR 0.79, 95% CI 0.64 to 0.97), as were deaths from any cause after a breast cancer diagnosis (HR 0.85, 95% CI 0.74 to 0.96). That is the single strongest randomized signal any dietary pattern has produced against cancer mortality, and it took two decades to surface.

Around that trial sits a remarkably consistent body of prospective data. Adherence to the World Cancer Research Fund and American Institute for Cancer Research recommendations, which are in substance a plant-forward, high-fiber, low-processed-food prescription, shows a dose-response relationship with survival: each one-point increase in adherence score corresponds to roughly 9% to 10% lower cancer-specific and overall mortality, and in the UK Biobank, 7% lower risk of cancer overall per point.

Then there are the specific findings that keep pointing in the same direction:

• Fiber: In 1,575 colorectal cancer patients followed after diagnosis, each additional 5 g of fiber per day was associated with 22% lower cancer-specific mortality and 14% lower all-cause mortality. Whole grains showed an independent effect.
• Immunotherapy response: Among 128 melanoma patients on checkpoint blockade, those eating at least 20 g of fiber daily had an 82% response rate versus 59% in the low-fiber group, with roughly 30% lower risk of progression or death for every additional 5 g per day. Probiotic supplement users did worse. Fiber is not a side dish; it is an immunological input.
• Prostate cancer: In more than 2,000 men with localized disease, the highest quintile of plant-based food intake was associated with a 47% lower risk of progression (HR 0.53, 95% CI 0.37 to 0.74). Ornish’s randomized lifestyle trial in low-risk prostate cancer found PSA fell 4% in the intensive plant-based arm and rose 6% in controls, with zero of 44 intervention patients requiring conventional treatment at one year versus six of 49 controls.
• Soy, the food patients are most often told to fear: In 5,042 breast cancer survivors, the highest soy protein intake was associated with 29% lower mortality and 32% lower recurrence, in ER-positive and ER-negative disease alike, and in women on tamoxifen.

Why “low-glycemic” is the load-bearing word

Plant-based alone is not enough, and this is where the argument gets sharper. White bread, fruit juice, and vegan cookies are plant-based. They are also a fast route to hyperinsulinemia.

Insulin and IGF-1 converge on the PI3K/AKT/mTOR pathway, activating the same signaling cascades as some of the most recurrent mutations in human cancer. In obesity and insulin resistance, muscle and liver become insulin resistant while tumor cells retain insulin sensitivity, which hands transformed cells a growth advantage.

This is not abstract. In CALGB 89803, stage III colon cancer patients in the highest quintile of dietary glycemic load had a 79% higher risk of recurrence or death, rising to 126% higher among those with a BMI of 25 or above. And the plant-based diet indices behave exactly as this model predicts: a healthful plant-based index tracks with lower breast and colorectal cancer risk, while an unhealthful plant-based index tracks with higher breast cancer risk.

So the prescription is specific: plants, yes, but intact ones. Legumes over flour, steel-cut oats over instant, whole fruit over juice.

The second lever: animal protein and IGF-1

Insulin is only half the growth-signal story. The other half is insulin-like growth factor 1, and IGF-1 answers to animal protein.

Higher circulating IGF-I predicts incident breast, prostate and colorectal cancer in the UK Biobank, and it signals through the same PI3K/AKT/mTOR axis as insulin while independently suppressing apoptosis, meaning damaged cells that should die instead survive and divide. It is one of the more direct dietary levers on cancer biology.

The cleanest human comparison comes from the EPIC-Oxford cohort. Vegan men had serum IGF-I roughly 9% lower than meat-eaters. Vegan women had IGF-I 13% lower, together with 20% to 40% higher IGFBP-1 and IGFBP-2, the binding proteins that sequester IGF-I and limit how much is biologically available, so the free, active fraction falls further still than the total.

The revealing detail is what happened to the vegetarians: they were essentially indistinguishable from the meat-eaters. What separated the vegans was not the absence of meat. It was the absence of dairy. Milk is, after all, a fluid whose evolved function is to make a young mammal grow fast, and milk intake is consistently associated with higher circulating IGF-I across studies. Cheese-heavy “healthy” diets are not IGF-1 neutral.

The outcome data point the same way. In NHANES III, adults aged 50 to 65 in the high-protein group had more than four times the cancer mortality of the low-protein group (HR 4.33, 95% CI 1.96 to 9.56), and the association was substantially attenuated once animal protein was accounted for, implicating the source of the protein rather than protein as such. That analysis has been fairly criticized as small and incompletely adjusted, and the relationship reversed in adults over 65, where protein adequacy protects against frailty. But it converges with the biomarker data rather than contradicting it.

This is precisely where the three diets separate. Keto lowers insulin while maximizing animal protein, pulling one lever while pushing the other. The Mediterranean pattern is comparatively generous with dairy and cheese. A low-glycemic whole-food plant-based diet is the only one of the three that turns both signals down at once. Its effect is not theoretical either: in the Ornish trial, serum drawn from men on the intensive plant-based program inhibited the growth of prostate cancer cells in culture by 70%, versus 9% for controls, a measurable change in the blood’s own growth-signaling environment.

Where the ketogenic diet fails

Keto’s appeal is that it targets the insulin problem directly, and it genuinely does lower fasting glucose and insulin. But it buys that one benefit at a high price, and after two decades it has never delivered the outcome that matters.

• No trial has shown a survival benefit: A systematic review covering 39 studies and 770 cancer patients concluded there was “no conclusive evidence for anti-tumor effects or improved overall survival”. The only randomized trial in glioma, ERGO2, was flatly negative on its primary endpoint, with 6-month progression-free survival of 20% versus 16%. The impressive-looking survival figures circulating online come from small single-arm trials in highly motivated patients compared against historical controls, which is the weakest design there is for a survival claim.
• Most patients cannot sustain it: Dropout in controlled trials runs about 25%, and in a 2026 pilot in metastatic renal cell carcinoma, only 40% of patients completed 12 months on the diet. A diet you abandon is not a treatment.
• The weight loss cuts the wrong way: In every randomized trial that measured it, keto produced significantly more weight loss than control, in one case 12.1 kg versus 0.5 kg. In an overweight survivor, that may help. In a patient at risk of cachexia or sarcopenia, it is a hazard. This is why ESPEN’s clinical nutrition guidelines discuss ketogenic diets under a section headed “Potentially harmful diets,” noting the absence of clinical trial evidence of benefit and cautioning against energy-restricting regimens in patients at risk of malnutrition.
• It loads up the amino acids tumors are hungriest for: Because keto is built on meat, eggs, cheese and fish, it delivers large amounts of methionine and cysteine. Many tumors are methionine-dependent, and restricting methionine sensitizes mouse cancer models to chemotherapy and radiation. Cysteine feeds glutathione synthesis, the antioxidant system tumors use to survive oxidative stress from treatment. Honesty requires flagging that the human arm of that landmark study measured metabolites in six healthy volunteers, and that clinical methionine restriction has not progressed beyond small phase I work. But the direction of travel in the research is toward less methionine, and keto goes the other way. A whole-food plant-based diet is the lowest-methionine way to eat adequately that exists.
• It maximizes dietary acid load: Sulfur amino acids yield sulfuric acid, and keto simultaneously strips out the potassium-rich and bicarbonate-rich plant foods that buffer it. Diet does not meaningfully change blood pH, and anyone claiming otherwise is overselling, but renal acid load is measurable. In the 43,570-woman Sister Study, the highest dietary acid load quartile carried 21% higher breast cancer risk overall and more than double the risk of triple-negative disease. Notably, in breast cancer survivors, acid load predicted recurrence only among women with elevated HbA1c, so acid load and glycemic control appear to compound each other.

In fairness to keto, it reliably reduces fat mass while preserving lean mass over short supervised courses, improves fatigue scores in pooled trial data, and has a plausible niche as an adjunct in glioblastoma and alongside PI3K inhibitors. Those are real. But they are surrogate endpoints and short horizons, and the metabolic goal keto pursues (low insulin) is achievable with a low-glycemic whole-food plant-based diet without the animal-protein IGF-1 signal, the methionine load, the acid load, the unresolved effects on LDL cholesterol, the attrition or the cachexia risk. ASCO’s guideline currently finds insufficient evidence to recommend ketogenic or low-carbohydrate diets for or against, and a 2022 JAMA Oncology review from Memorial Sloan Kettering concluded plainly that “currently available data support plant-based diets as opposed to KD as part of a lifestyle associated with reduced cancer risk.”

Where the Mediterranean diet falls short

The Mediterranean diet is not wrong. It is incomplete, and it works largely to the extent that it is plant-based.

The epidemiology is real but modest: pooled across 3.2 million people, high adherence associates with 13% lower cancer mortality, with the largest analyses attributing the effect chiefly to fruits, vegetables and whole grains. The famous randomized cancer finding, PREDIMED’s 68% reduction in breast cancer, rests on eight events in one arm versus seventeen in control, was a secondary endpoint of a cardiovascular trial, was not significant in the second intervention arm, and comes from a trial whose main paper was retracted and republished in 2018 over randomization failures.

More fundamentally, the pattern retains what the WFPB pattern removes. The official Mediterranean pyramid places dairy at two servings daily, with the IGF-1 consequences described above, and permits red meat weekly and processed meat weekly, a Group 2A probable carcinogen and a Group 1 carcinogen, with roughly 18% higher colorectal cancer risk per 50 g of processed meat daily. Most Mediterranean adherence scores award a point for moderate alcohol, another Group 1 carcinogen causally linked to at least seven cancers, with no established safe level. A diet score that rewards drinking wine and tolerates weekly charcuterie is not the ceiling of dietary oncology; it is a floor.

Nor is the Mediterranean diet inherently low-glycemic. White bread, pasta and potatoes sit comfortably within it.

The honest limitations

Several well-designed trials of plant-forward eating were null. WHEL raised vegetable intake by 65% in 3,088 breast cancer survivors and produced no difference in recurrence or mortality. The Polyp Prevention Trial found no reduction in adenoma recurrence. A pooled analysis of 1.8 million people found no overall cancer-incidence difference for vegetarians or vegans, with a colorectal signal that went the wrong way in a small vegan subgroup.

These deserve a straight answer rather than a dismissal. Three things explain most of it. The null trials ran four to seven years against endpoints that take decades to move, and WHI needed 19.6 years to show its effect. They compared a good diet to a moderately good diet rather than to a Western one. And none of them controlled glycemic load: WHEL added vegetables without removing refined carbohydrate, which is exactly the design this article argues against. Meanwhile, the beta-carotene trials, in which the isolated supplement increased lung cancer risk by 16% while carotenoid-rich foods track with lower risk, are the cleanest possible argument for eating whole foods rather than extracting their parts.

What this looks like on a plate

Legumes daily. Cruciferous and leafy vegetables daily. Intact whole grains rather than flour. Whole fruit, especially berries. Ground flaxseed, nuts and seeds. Little to no meat, dairy or eggs. Little to no refined carbohydrate, added sugar or alcohol. Soy foods are generally encouraged, not feared.

Before you change anything

This is general information, not medical advice, and diet is an adjunct to cancer treatment, never a substitute for it. Two cautions matter especially. For patients who are losing weight, in active treatment, at risk of cachexia, or have protein and calorie needs that a poorly planned plant-based diet can miss, an oncology dietitian should help build the plan. And anyone eating fully plant-based needs reliable vitamin B12, plus attention to vitamin D, iodine, iron, zinc and omega-3s.

The strongest claim the evidence supports is not that any diet cures cancer. It is that among the three patterns competing for a patient’s attention, one has randomized mortality data behind it, coheres with the insulin, IGF-1, methionine, fiber and immune biology all at once, and can be eaten safely for forty years. That is the low-glycemic, whole-food, plant-based diet.

References

1. Allen NE, Appleby PN, Davey GK, Kaaks R, Rinaldi S, Key TJ. The associations of diet with serum insulin-like growth factor I and its main binding proteins in 292 women meat-eaters, vegetarians, and vegans. Cancer Epidemiol Biomarkers Prev. 2002;11(11):1441-1448.
2. Allen NE, Appleby PN, Davey GK, Key TJ. Hormones and diet: low insulin-like growth factor-I but normal bioavailable androgens in vegan men. Br J Cancer. 2000;83(1):95-97.
3. Arends J, Bachmann P, Baracos V, et al. ESPEN guidelines on nutrition in cancer patients. Clin Nutr. 2017;36(1):11-48.
4. Bouvard V, Loomis D, Guyton KZ, et al. Carcinogenicity of consumption of red and processed meat. Lancet Oncol. 2015;16(16):1599-1600.
5. Chlebowski RT, Aragaki AK, Anderson GL, et al. Dietary modification and breast cancer mortality: long-term follow-up of the Women’s Health Initiative randomized trial. J Clin Oncol. 2020;38(13):1419-1428.
6. Dunneram Y, Lee JY, Watling CZ, et al. Vegetarian diets and cancer risk: pooled analysis of 1.8 million women and men in nine prospective studies on three continents. Br J Cancer. 2026;134(8):1218-1229.
7. Epner DE, Morrow S, Wilcox M, Houghton JL. Nutrient intake and nutritional indexes in adults with metastatic cancer on a phase I clinical trial of dietary methionine restriction. Nutr Cancer. 2002;42(2):158-166.
8. Estruch R, Ros E, Salas-Salvadó J, et al. Primary prevention of cardiovascular disease with a Mediterranean diet supplemented with extra-virgin olive oil or nuts. N Engl J Med. 2018;378(25):e34.
9. Gallagher EJ, LeRoith D. Hyperinsulinaemia in cancer. Nat Rev Cancer. 2020;20(11):629-644.
10. 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.
11. Gil-Lespinard M, Iglesias-Vázquez L, Jakszyn P. Healthful and unhealthful plant-based diets and site-specific cancer risk: a systematic review and meta-analysis of observational studies. Eur J Nutr. 2026;65(4):121.
12. Hopkins BD, Goncalves MD, Cantley LC. Insulin-PI3K signalling: an evolutionarily insulated metabolic driver of cancer. Nat Rev Endocrinol. 2020;16(5):276-283.
13. International Agency for Research on Cancer. IARC Monographs Volume 114: consumption of red meat and processed meat. Questions and answers. Lyon: IARC; 2015.
14. Knuppel A, Fensom GK, Watts EL, et al. Circulating insulin-like growth factor-I concentrations and risk of 30 cancers: prospective analyses in UK Biobank. Cancer Res. 2020;80(18):4014-4021.
15. Kordiak J, Bielec F, Jabłoński S, Pastuszak-Lewandoska D. Role of beta-carotene in lung cancer primary chemoprevention: a systematic review with meta-analysis and meta-regression. Nutrients. 2022;14(7):1361.
16. Levine ME, Suarez JA, Brandhorst S, et al. Low protein intake is associated with a major reduction in IGF-1, cancer, and overall mortality in the 65 and younger but not older population. Cell Metab. 2014;19(3):407-417.
17. Ligibel JA, Bohlke K, May AM, et al. Exercise, diet, and weight management during cancer treatment: ASCO guideline. J Clin Oncol. 2022;40(22):2491-2507.
18. Liu VN, Van Blarigan EL, Zhang L, et al. Plant-based diets and disease progression in men with prostate cancer. JAMA Netw Open. 2024;7(5):e249053.
19. Malcomson FC, Parra-Soto S, Ho FK, Celis-Morales C, Sharp L, Mathers JC. Abbreviated score to assess adherence to the 2018 WCRF/AICR cancer prevention recommendations and risk of cancer in the UK Biobank. Cancer Epidemiol Biomarkers Prev. 2024;33(1):33-42.
20. Meyerhardt JA, Sato K, Niedzwiecki D, et al. Dietary glycemic load and cancer recurrence and survival in patients with stage III colon cancer: findings from CALGB 89803. J Natl Cancer Inst. 2012;104(22):1702-1711.
21. Morze J, Danielewicz A, Przybyłowicz K, Zeng H, Hoffmann G, Schwingshackl L. An updated systematic review and meta-analysis on adherence to Mediterranean diet and risk of cancer. Eur J Nutr. 2021;60(3):1561-1586.
22. Office of the Surgeon General. Alcohol and cancer risk: the U.S. Surgeon General’s advisory. Washington, DC: US Department of Health and Human Services; 2025.
23. Ornish D, Weidner G, Fair WR, et al. Intensive lifestyle changes may affect the progression of prostate cancer. J Urol. 2005;174(3):1065-1070.
24. Park YM, Steck SE, Fung TT, et al. Higher diet-dependent acid load is associated with risk of breast cancer: findings from the Sister Study. Int J Cancer. 2019;144(8):1834-1843.
25. Pierce JP, Natarajan L, Caan BJ, et al. Influence of a diet very high in vegetables, fruit, and fiber and low in fat on prognosis following treatment for breast cancer: the Women’s Healthy Eating and Living (WHEL) randomized trial. JAMA. 2007;298(3):289-298.
26. Qin LQ, He K, Xu JY. Milk consumption and circulating insulin-like growth factor-I level: a systematic literature review. Int J Food Sci Nutr. 2009;60 Suppl 7:330-340.
27. Rolley C, Zidane M, Nedelcu C, et al. Feasibility and safety of a ketogenic diet during systemic therapy for metastatic renal cell carcinoma: results from the Cetorein pilot study. Nutrients. 2026;18(11):1712.
28. Römer M, Dörfler J, Huebner J. The use of ketogenic diets in cancer patients: a systematic review. Clin Exp Med. 2021;21(4):501-536.
29. Schatzkin A, Lanza E, Corle D, et al. Lack of effect of a low-fat, high-fiber diet on the recurrence of colorectal adenomas. N Engl J Med. 2000;342(16):1149-1155.
30. Shah UA, Iyengar NM. Plant-based and ketogenic diets as diverging paths to address cancer: a review. JAMA Oncol. 2022;8(8):1201-1208.
31. Shu XO, Zheng Y, Cai H, Gu K, Chen Z, Zheng W. Soy food intake and breast cancer survival. JAMA. 2009;302(22):2437-2443.
32. Solans M, Chan DSM, Mitrou P, Norat T, Romaguera D. A systematic review and meta-analysis of the 2007 WCRF/AICR score in relation to cancer-related health outcomes. Ann Oncol. 2020;31(3):352-368.
33. Song M, Wu K, Meyerhardt JA, et al. Fiber intake and survival after colorectal cancer diagnosis. JAMA Oncol. 2018;4(1):71-79.
34. Spencer CN, McQuade JL, Gopalakrishnan V, et al. Dietary fiber and probiotics influence the gut microbiome and melanoma immunotherapy response. Science. 2021;374(6575):1632-1640.
35. Toledo E, Salas-Salvadó J, Donat-Vargas C, et al. Mediterranean diet and invasive breast cancer risk among women at high cardiovascular risk in the PREDIMED trial: a randomized clinical trial. JAMA Intern Med. 2015;175(11):1752-1760.
36. Voss M, Wagner M, von Mettenheim N, et al. ERGO2: a prospective, randomized trial of calorie-restricted ketogenic diet and fasting in addition to reirradiation for malignant glioma. Int J Radiat Oncol Biol Phys. 2020;108(4):987-995.
37. Wu T, Hsu FC, Wang S, Luong D, Pierce JP. Hemoglobin A1c levels modify associations between dietary acid load and breast cancer recurrence. Nutrients. 2020;12(2):578.