What does Nutrition & Cancer Risk do in the body?

Finuwell

Mine Ask

Reading room/ConditionsReader

Story

Nutrition & Cancer Risk

Your saved stories

Last updated September 2026

In one pass A single food on its own rarely decides whether you get cancer.

Educational content, not medical advice — consult a clinician.

Chapter 1

Food only shifts the odds

A single food on its own rarely decides whether you get cancer. Even processed meat and alcohol — which the International Agency for Research on Cancer (IARC) lists as Group 1 carcinogens (the evidence that they cause cancer is certain; the label says nothing about how large the harm is) — raise the odds through the amount taken over many years, not through any one meal.

This plays out at the level of cells: cancer is cells multiplying out of control after years of accumulated genetic damage, and cancer is an umbrella term for hundreds of different diseases — colorectal, breast and liver cancers all differ in mechanism. What diet actually does is slowly shift the probability of disease across a population: long-term eating patterns make certain cancers somewhat more or somewhat less likely.

What this means for you: do not expect one food to save you or doom you. Put your effort into long-term patterns, where the payoff is real but modest; if anyone guarantees that eating something will keep you free of cancer, treat the claim with suspicion from the start.

Background · Why it is odds, not a switch

Saying that diet only shifts the odds is not a way of letting diet off the hook. It follows from the shape of cancer itself.

Cancer is not a single event. It is one cell collecting several pieces of genetic damage over many years, and slipping past the clean-up the body should have carried out, before it slowly goes out of control. What any one meal or any one drink can do is make one step in that process happen a little more or a little less often — so its effect only shows up across many people and many years, and for a single person it shows up as probability.

That is also why two popular sayings fail in the same place:

"Superfoods prevent cancer" and "sugar feeds cancer" both turn a question of probability into a cause-and-effect switch. The first makes people think eating more of something switches the risk off; the second makes people think eating less of something starves a tumor. Both are wrong.The evidence that carries weight comes from pooling large population cohorts together with mechanistic studies — for example, the Third Expert Report of the World Cancer Research Fund and the American Institute for Cancer Research (WCRF/AICR). This kind of evidence speaks of risk, not guarantees: cohort studies see associations, and a firmer conclusion needs a mechanism that makes sense plus several studies pointing the same way.
So what the rest of this story does is keep what the evidence says apart from what it does not say.

Chapter 2

What the evidence supports

Filter out the marketing noise, and what has stronger evidence comes down to a few plain conclusions (WCRF/AICR 2018; IARC):

Excess body fat is a clear risk: the IARC Working Group judged the evidence sufficient that people without excess body fat have a lower risk of 13 cancers (Lauby-Secretan 2016). This comes from a large body of population studies plus a mechanism that makes sense, and it is the firmest piece of diet-related evidence.Alcohol is a Group 1 carcinogen: there is no such thing as a safe amount of drinking; the more you drink, the higher the risk of cancers of the mouth, throat, esophagus, liver, breast and other sites (Bagnardi 2015; WHO 2023).Processed and red meat: IARC classifies processed meat (sausages, bacon, ham) as a Group 1 carcinogen and red meat as Group 2A (probably carcinogenic), mainly in connection with colorectal cancer (Bouvard 2015; Norat 2005). These grades describe how certain the evidence is, not how big the harm is: Group 1 does not mean the harm is as large as smoking's.Dietary fiber and whole grains: people who eat more of them have a lower risk of colorectal cancer; this is an observed association (WCRF/AICR).
What this means for you: none of this is mysterious — keep your weight in check, drink less, eat less processed meat, and eat more whole grains and fiber. The same steps are good for the heart and for metabolism too.

Evidence · Who is on the list, and what grades mean

Start by laying the list out. The group the IARC Working Group named includes colorectal, postmenopausal breast, endometrial, kidney, liver and pancreatic cancer, esophageal adenocarcinoma, and others. It is not a random pile of organs — the members of the list share common threads. The two threads below are the leading explanations; they come from mechanistic studies and population data, and are not yet a causal chain verified link by link.

One thread is insulin. When you carry extra body fat, muscle and liver become less responsive to insulin, so the pancreas has to release more of it, and the blood spends years soaking in raised insulin and the growth signals related to it. What those signals tell a cell is simple: grow, don't stop, don't self-destruct. For a healthy cell that is no big deal; for a cell that already carries damage, that message helps it dodge the clean-up that should have happened. The gut, liver and pancreas members of the list are thought to follow mainly this thread.

The other thread is estrogen. After menopause the ovaries largely stop making estrogen, but fat tissue can still convert androgens into estrogen — so the more fat, the more estrogen in the blood after menopause. Breast and endometrial cells happen to be estrogen-sensitive: they divide when they receive the signal. That is why the list specifically says postmenopausal breast cancer. Before menopause the ovary is the main producer and fat's small output is drowned out; after menopause the ovary steps aside, fat becomes the main producer, and the difference shows.

Then the word that unsettles people most: Group 1 carcinogen. This classification answers exactly one question — how sure we are that it can cause cancer, not how dangerous it is. Processed meat and smoking are in the same group, which means both have been confirmed to cause cancer, not both are equally deadly. The first nudges colorectal cancer risk up a little; the second multiplies lung cancer risk many times over. Reading certainty as size of harm is the most common misuse of this classification, and it is why bacon gets argued over every year.

Finally, sugary drinks. One of the ten WCRF/AICR cancer prevention recommendations is to limit sugar-sweetened drinks. They raise risk not by the direct-damage route but by a detour: sugar in liquid form brings almost no feeling of fullness, so it easily pushes a day's total energy up without your noticing; more total energy means more body fat, and more body fat lights up both threads above at once. So the effect is indirect — but indirect does not mean it does not count. It feeds straight into the firmest piece of evidence in this whole story.

Chapter 3

Why these few things

These conclusions are not arbitrary; each has a mechanism behind it:

Obesity: too much fat tissue (especially visceral fat, around the organs) acts like a low-grade inflammation factory; it also pushes up insulin and (a growth signal that drives cells to multiply), and converts androgens into estrogen (Lauby-Secretan 2016; Tchernof and Després 2013).Alcohol: in the body, alcohol is first broken down into acetaldehyde. Acetaldehyde is a proven genotoxin: it binds to DNA to form adducts and triggers mutations, especially in the upper digestive tract (IARC 2023).Processed meat: curing and processing produce N-nitroso compounds, and together with the heme iron in red meat, they may damage the DNA of cells lining the gut (Bouvard 2015).Fiber: gut bacteria ferment fiber into short-chain fatty acids (fuel that maintains the gut wall); fiber also shortens the time stool spends in the gut and dilutes carcinogens (Sonnenburg 2014).
What this means for you: once you see the mechanisms, most of the benefit of lowering cancer risk turns out to be a by-product of metabolic health plus less exposure to proven carcinogens — not the magic of some mysterious ingredient.

Mechanism · Why alcohol hits the upper digestive tract

The chain from alcohol to cancer is usually told in two steps: alcohol becomes acetaldehyde, and acetaldehyde damages DNA. The step that really needs telling is the third one — where the acetaldehyde goes.

Once alcohol (ethanol) is in the body, an enzyme first converts it into acetaldehyde. Acetaldehyde is the troublemaker: it is chemically reactive and latches directly onto DNA bases to form adducts — like sticking a lump onto a book that is being copied over and over. The next time the cell copies its DNA and reaches that lump, it may copy a letter wrong; if the error stays, it is a mutation. That is why acetaldehyde is classed as a genotoxin, not merely an unpleasant metabolic intermediate.

The next step is the key one. Normally another enzyme immediately converts acetaldehyde into acetate, which is sent off to be burned as fuel. Acetaldehyde does not stay around long, so few lumps get the chance to stick. The enzyme that does this step is aldehyde dehydrogenase.

And in East Asian populations this step has a very common twist. A large share of people carry a gene variant that almost wipes out aldehyde dehydrogenase activity (a common variant of the ALDH2 gene). The flushed face, red neck, racing heart and pounding head after a drink are this event showing up on the skin: acetaldehyde has been made, but nothing clears it in time, so it builds up in the blood and widens the small vessels of the skin. Flushing is not a sign of low tolerance. It is this clearance step running slow.

So why does this chain land especially on the esophagus and the mouth and throat? Because the first stretch of the route alcohol travels is the mouth, throat and esophagus, and the lining of that stretch is bathed directly in the drink, so local acetaldehyde levels are far higher than anywhere else in the body; bacteria in the mouth also turn ethanol into acetaldehyde on the spot, right on the lining's surface. By this mechanism, in someone whose clearance is slow, the same sip lingers longer on that stretch at a higher concentration, and more lumps stick. So the same amount of alcohol does not run up the same bill for everyone, and the bill concentrates on this entry corridor.

This also turns there is no safe amount of alcohol from a population statistic into something you can reason through for your own body: as long as you keep drinking, you keep making acetaldehyde; if it is made and cleared slowly, the lumps keep adding up. People who flush when they drink especially need to know this — the reaction is not the vague idea that your constitution is not suited to alcohol. It is a concrete, visible biochemical signal.

Mechanism · What fat tissue releases

Treating fat as surplus stored grain is the easiest way to misunderstand this whole chain. Fat tissue is actually a living organ that releases things into the blood, and what it releases reaches cells all over the body.

First: it is quietly inflamed. When fat cells are packed too full they become stressed and draw immune cells in, and both release inflammatory signals into the blood. This inflammation is not red and painful like a sprain; it is low-grade and never stops. A constant inflammatory environment has two consequences. One: tissue is damaged and repaired over and over, and every repair is a cell division, and every division is a chance to copy DNA — the more copies, the more copying errors in absolute terms. Two: inflammation itself tilts cells toward staying alive rather than dying when they should.

Second: it pushes insulin up. When there is more fat, muscle and liver become less responsive to insulin, so the pancreas has to release more just to hold blood glucose down, and the blood spends years soaking in raised insulin and its related growth signals. Those signals tell a cell the same thing every time: grow, don't stop, don't self-destruct. For a cell that already carries damage, that message helps it dodge the clean-up that should have happened.

Visceral fat (the kind wrapped around the organs in the belly) matters especially, because what it releases passes through the liver first and only then into the general circulation — the liver gets the undiluted share. That is why waist size often tells you more than body weight.

Third: it makes its own estrogen. Fat tissue has an enzyme that can convert androgens into estrogen. Before menopause the ovary is the main producer and fat's small output is drowned out; after menopause the ovary largely stops, and fat is promoted to main producer. Breast and endometrial cells are estrogen-sensitive — they divide when they get the signal — so people with more body fat receive more orders to divide after menopause. That explains a detail that looks odd at first: the list says postmenopausal breast cancer, not all breast cancer.

Put the three together and they point to one picture: more damage, more orders to divide, less clean-up. Cancer is what these three things add up to over years, and excess body fat pushes all three up a little at once. By this picture, weight loss should move risk slowly and modestly, but in a clear direction — what you pull down is these same three things.

Mechanism · What meat and fiber do in the colon

The last two chains happen in the same place — the lining of the large intestine — so they are clearest side by side. That lining is only one cell thick, yet it has to keep everything in the gut apart from your body; it stays intact by constantly renewing itself, renewal means cell division, and division means copying DNA. Anything that makes these cells take a little more damage, or divide a few more times, pushes in the same direction.

The processed-meat chain: curing and processing introduce a class of compounds called N-nitroso compounds; some come directly from the processing, and some are formed on the spot in the gut from ingredients the meat brings with it. What these molecules share is that they modify DNA bases, doing much the same thing as acetaldehyde — leaving a site that is easy to copy wrong.

The red-meat chain: the red color comes from heme iron, the iron set into hemoglobin. In the gut it acts like a small catalyst, driving fat oxidation and the formation of nitroso compounds, so the surface of the lining is exposed to more reactive molecules. Note that this is iron inside the gut, not iron in the blood — which is why red meat is a good source of iron and red meat is linked to colorectal cancer can both be true: they are about iron in two different places.

The fiber chain runs the other way, and it works two ways. The first is chemical: bacteria in the large intestine ferment fiber into short-chain fatty acids, and one of them happens to be the fuel colon cells like best — in effect delivering a meal straight to the cells that renew most often, so the barrier holds up better. The second is physical: fiber holds water and adds bulk, so the contents move faster and are diluted more, and both the concentration of those reactive molecules at the lining and the time they spend there fall together.

Lay the three side by side and you can reason it out yourself: processed and red meat add reactive molecules to the gut; fiber dilutes them, speeds up the clear-out, and feeds the cells that guard the gate. That also explains why the advice always comes in a pair — less processed meat and more whole-grain fiber are not two separate recommendations. They act on the same surface, in opposite directions.

Chapter 4

Sugar, detox and megadose claims

The most popular claims are exactly the ones that hold up worst:

"Sugar feeds cancer": an oversimplification. Every cell in the body runs on glucose, and the brain can use almost nothing else; even if you eat no sugar at all, the body makes its own blood glucose, so cutting sugar cannot starve a tumor. The real link is a detour: sugar (especially sugary drinks) leads to weight gain and raised insulin, and that is what indirectly raises risk (WCRF).Alkaline or detox diets prevent cancer: there is no evidence. A systematic review found no causal link between the acid load of the diet, or alkaline water, and cancer (Fenton 2016); commercial detox diets have no controlled trials behind them either (Klein and Kiat 2015). Your kidneys and liver are already doing the detoxing.High-dose antioxidant supplements: not just useless, possibly harmful — this is the most important warning here. Two large randomized trials, CARET and SELECT, put the idea of taking more to a real test, and risk went up: beta-carotene and lung cancer in smokers, vitamin E and prostate cancer.A single cancer-preventing superfood: blueberries, broccoli and turmeric may be good, but each is only one part of a balanced diet; none of them on its own keeps you from getting cancer.
What this means for you: the simpler the promise — cut out one thing and you will not get cancer, take one thing and it will cure it — the more wary you should be. Sometimes high-dose supplements are more dangerous than taking nothing.

Myth · What the Warburg effect does and does not say

Cutting sugar cannot starve a tumor is a sentence worth unpacking until you can explain the reason to someone else.

Start with the true fact that keeps getting cited: many tumor cells really do eat sugar especially greedily. When oxygen is plentiful, a normal cell sends glucose into its mitochondria and burns it slowly and thoroughly; many tumor cells, even with plenty of oxygen, prefer the fast, inefficient road — they take in large amounts of glucose, extract only a fraction of the energy, and dump the rest. That preference is called the Warburg effect. It is real, real enough to hunt tumors with: in the clinic, a patient is injected with a little labeled glucose look-alike and then scanned, and the spots that light up are often tumors, because that is where sugar is being eaten hardest.

The problem is the next step of reasoning. From tumors eat a lot of sugar you cannot get to eat less sugar and the tumor has nothing to eat, because one fact sits in between: blood glucose is not set by how much sugar you ate today.

Blood glucose is a quantity the body defends hard. If you eat no sugar, the liver breaks down glycogen and makes glucose on the spot from amino acids and glycerol; the moment blood glucose starts to fall, the brain is the first to object — it can use almost nothing else. So in a healthy person, however the diet changes, blood glucose is held in a fairly narrow range. A tumor bathes in that defended bloodstream, not in your plate. To starve a tumor through diet, you would first have to push blood glucose to a level the brain cannot tolerate — which is neither an achievable goal nor one worth pursuing.

So is there any link between sugar and cancer? Yes, but by a detour. Sugar (especially sugar you drink) easily pushes a day's total energy up without your noticing, body fat rises, and the three things described in the chapter Why these few things (chronic inflammation, raised insulin, and estrogen made by fat) all light up at once. That path is real, and it feeds straight into the firmest piece of evidence in this whole story.

Telling the two paths apart matters, because it decides what you do: drinking fewer sugary drinks has a reason, and the reason is energy and weight; fearing every staple food in order to starve a tumor has no reason — and in someone who is already ill and already losing weight, that fear can do direct harm.

Evidence · Why CARET and SELECT backfired

These two trials deserve their own telling, because they are two of the most expensive lessons in nutrition science, and they teach exactly the kind of judgment this story is trying to teach.

The starting point was a very reasonable hypothesis. Observational studies kept seeing the same thing: people with more beta-carotene in their blood had less lung cancer, and people who ate more antioxidants had less cancer. The mechanism made sense too — free radicals damage DNA, antioxidants neutralize free radicals, so antioxidant supplements should reduce the damage. It held in the test tube, population observations backed it, and there seemed to be no reason it would fail.

CARET put it to the test. The researchers recruited the people at highest risk (smokers and people exposed to asbestos) and randomized them: one group took high-dose beta-carotene plus vitamin A, the other a placebo, and then the researchers waited to see who got less lung cancer. The supplemented group got more lung cancer, and the trial was stopped early (Omenn 1996).

SELECT tested it again, with a different nutrient and a different cancer. It was a large randomized trial in men, comparing vitamin E with placebo and looking at prostate cancer. Again the result ran the wrong way: the supplemented group had a relative increase in prostate cancer of about 17% (Klein 2011). Later analyses that pooled the antioxidant-supplement trials concluded that these supplements do not lower mortality (Bjelakovic 2013).

Why did it flip? Honestly, there is no agreed answer. One idea that often comes up: oxidative signals are themselves one of the ways a cell notices that it has gone wrong, and holding them too low for too long may make it easier for already-damaged cells to slip through. But that is only an explanation, not a proven conclusion — admitting we do not know matters more here than inventing a mechanism that sounds right.

What to take away is not the two results but this shape: a mechanism that holds at the level of cells, plus a stack of observational data pointing the same way, can still produce the opposite result in people. That is exactly why exist; and it is why the amount that comes naturally in food and the amount concentrated into a pill have to be discussed separately — beta-carotene in carrots has never caused trouble; the trouble came from pulling it out on its own, multiplying the dose, and taking it for years.

One last thing that often gets mixed up: these two trials are about healthy people taking high-dose antioxidant supplements long-term to prevent cancer. They do not carry over directly to a supplement a doctor prescribes because a deficiency has been confirmed — that is a different question, decided by lab tests and a doctor's orders, not by this story.

Chapter 5

The overall pattern, plus screening

Put both sides together and land on what you can actually do — and it is the overall pattern, not a single ingredient: keep a healthy weight, eat mainly plants, limit alcohol, eat less processed meat and ultra-processed food, and move regularly; then add screening by the guidelines, the vaccines you should have, and not smoking.

Of these, keeping a healthy weight matches the firmest evidence in this story; for alcohol, the less the better, and there is no shortcut in which moderate drinking protects the heart (WHO 2023). And for colorectal, cervical, breast and other cancers, guideline-based screening, vaccines (such as HPV and hepatitis B) and not smoking do far more to lower how often cancer occurs and how many people die of it than any single food.

What this means for you: it is plain, but it is everything the evidence can support — and that is already very useful.

Evidence · What kind of evidence backs each step

All of these are things worth doing, but the evidence behind them is not all of one kind. Item by item:

A plant-forward diet: plenty of vegetables and fruit, whole grains, legumes and nuts. The strongest evidence for a Mediterranean-style diet is about the heart: in the PREDIMED randomized trial, among adults at high cardiovascular risk, a Mediterranean diet with added extra-virgin olive oil or nuts reduced major cardiovascular events (Estruch 2018). It was not a cancer trial; it is here because the overall dietary pattern is associated with lower disease risk (WCRF/AICR).Limit alcohol: the less the better; there is no shortcut in which moderate drinking protects the heart (WHO 2023). The case that alcohol causes cancer rests on IARC's Group 1 classification and on risk rising with the dose.Less processed meat and ultra-processed food: in France's NutriNet-Santé cohort, people who ate more ultra-processed food had a higher risk of cancer (Fiolet 2018); this is an observed association and cannot show that ultra-processed food was the cause.Keep a healthy weight and move regularly: this maps directly onto the firmest evidence in the story (Lauby-Secretan 2016), a judgment the IARC Working Group made from a large body of population studies and mechanisms.
The way to read this list: the randomized trial proves a heart benefit of the Mediterranean diet, while the points about cancer rest mainly on population studies plus mechanisms. They are enough to act on, but do not read them as eat right and you will not get cancer.

Chapter 6

Lowering risk is not treatment

Finally, the boundaries have to be completely clear:

This story is about lowering risk at the population level; it is not a way to diagnose or treat cancer. Diet can help nudge the odds down a little, but it cannot treat cancer, and it cannot replace proper cancer treatment such as surgery, chemotherapy, radiotherapy, targeted therapy or immunotherapy.If you have warning symptoms (unexplained weight loss, ongoing bleeding from the digestive tract or black stools, difficulty swallowing, a persistent cough or coughing up blood, an unusual lump, bleeding after menopause, and so on) → see a doctor promptly; do not wait it out with "food therapy".
Disclaimer: this site is education, meant to help you understand the mechanisms and make better decisions together with your doctor; it is not medical advice and does not replace professional diagnosis and care.

Clinical · If you are already diagnosed

If you are already diagnosed or in treatment: follow your oncology team's plan. Online claims about anti-cancer diets, starving tumors by cutting sugar, or high-dose supplements are useless at best, and at worst delay treatment or interact with your medicines — always ask your treating doctor and a clinical dietitian first.For someone who is already ill and already losing weight, the fear behind starving a tumor by cutting sugar can do direct harm; the chapter Sugar, detox and megadose claims gives the full reason why cutting sugar cannot starve a tumor.Whether you need screening, from what age, and how often depends on your personal risk and family history — talk it through with your doctor.

References · 17

World Cancer Research Fund / American Institute for Cancer Research. (2018). Diet, Nutrition, Physical Activity and Cancer: a Global Perspective (Third Expert Report). WCRF/AICR. Ten evidence-based Cancer Prevention Recommendations: maintain a healthy weight, be physically active, eat wholegrains/vegetables/fruit/beans, limit fast foods and ultra-processed foods, limit red and processed meat, limit sugar-sweetened drinks, limit alcohol, and do not rely on supplements for cancer prevention. The meat recommendation reads: eat no more than moderate amounts of red meat and little, if any, processed meat. For red meat it gives a quantity - no more than about three portions a week, equivalent to about 350-500 g (12-18 oz) COOKED weight. For processed meat it deliberately gives NO gram figure, only the wording consume very little, if any. Evidence grading for colorectal cancer: processed meat is graded convincing, red meat probable. www.wcrf.org/diet-activity-and-cancer
Lauby-Secretan, B., Scoccianti, C., Loomis, D., Grosse, Y., Bianchini, F., & Straif, K. (IARC Handbook Working Group). (2016). Body fatness and cancer — viewpoint of the IARC Working Group. New England Journal of Medicine, 375(8), 794-798. Sufficient evidence that the absence of excess body fatness lowers the risk of 13 cancers, including colorectal, postmenopausal breast, endometrial, kidney, liver, pancreatic and esophageal adenocarcinoma. 10.1056/NEJMsr1606602
Bagnardi, V., Rota, M., Botteri, E., Tramacere, I., Islami, F., Fedirko, V., et al. (2015). Alcohol consumption and site-specific cancer risk: a comprehensive dose-response meta-analysis. British Journal of Cancer, 112(3), 580-593. 572 studies, 486,538 cancer cases. Heavy drinkers vs non- and occasional drinkers: oral and pharyngeal cancer RR 5.13, oesophageal squamous cell 4.95, laryngeal 2.65, breast 1.61, colorectal 1.44, all with a clear dose-risk relationship; also stomach 1.21, liver 2.07, gallbladder 2.64, pancreas 1.19, lung 1.15 (abstract, PMID 25422909). 10.1038/bjc.2014.579
World Health Organization. (2023). No level of alcohol consumption is safe for our health. The Lancet Public Health, 8(1), e6-e7. www.who.int/europe/news/item/04-01-2023-no-level-of-alcohol-consumption-is-safe-for-our-health
Bouvard, V., Loomis, D., Guyton, K. Z., Grosse, Y., Ghissassi, F. E., Benbrahim-Tallaa, L., et al. (2015). Carcinogenicity of consumption of red and processed meat. Lancet Oncology, 16(16), 1599-1600. IARC classifies processed meat as Group 1 carcinogen, red meat as Group 2A. 10.1016/S1470-2045(15)00444-1
Norat, T., Bingham, S., Ferrari, P., Slimani, N., Jenab, M., Mazuir, M., et al. (2005). Meat, fish, and colorectal cancer risk: the European Prospective Investigation into Cancer and Nutrition (EPIC). Journal of the National Cancer Institute, 97(12), 906-916. 10.1093/jnci/dji164
Tchernof, A., & Despres, J. P. (2013). Pathophysiology of human visceral obesity: an update. Physiological Reviews, 93(1), 359-404. Visceral (android) fat is roughly twice as high in men as women at matched adiposity and is metabolically detrimental, whereas gluteo-femoral (gynoid) fat is relatively protective. 10.1152/physrev.00033.2011
International Agency for Research on Cancer. (2023). The IARC perspective on alcohol reduction or cessation and cancer risk. New England Journal of Medicine, 389(26), 2486-2494. Ethanol in alcoholic beverages and its metabolite acetaldehyde are Group 1 carcinogens; acetaldehyde forms DNA adducts that drive mutations in the upper aerodigestive tract, and reducing or stopping alcohol lowers the risk of oral cavity and esophageal cancers. 10.1056/NEJMsr2306723
Sonnenburg, E. D., & Sonnenburg, J. L. (2014). Starving our microbial self: the deleterious consequences of a diet deficient in microbiota-accessible carbohydrates. Cell Metabolism, 20(5), 779-786. 10.1016/j.cmet.2014.07.003
World Cancer Research Fund International. (n.d.). Ask the expert: does sugar feed cancer? WCRF. All cells in the body use glucose for energy, and there is no evidence that dietary sugar directly feeds tumors or that cutting sugar starves cancer; the real pathway is excess energy intake leading to overweight/obesity, which raises the risk of several cancers. www.wcrf.org/about-us/news-and-blogs/ask-the-expert-does-sugar-feed-cancer
Fenton, T. R., & Huang, T. (2016). Systematic review of the association between dietary acid load, alkaline water and cancer. BMJ Open, 6(6), e010438. Found no evidence supporting a causal link between dietary acid load or alkaline water and cancer. 10.1136/bmjopen-2015-010438
Klein, A. V., & Kiat, H. (2015). Detox diets for toxin elimination and weight management: a critical review of the evidence. Journal of Human Nutrition and Dietetics, 28(6), 675-686. Very little clinical evidence supports detox diets: a handful of clinical studies reported enhanced liver detoxification or removal of persistent organic pollutants, but they were hampered by flawed methods and small samples; evidence for particular foods (coriander, nori, olestra) is preliminary and mostly from animals; and, verbatim, 'no randomised controlled trials have been conducted to assess the effectiveness of commercial detox diets in humans' (abstract, PMID 25522674). 10.1111/jhn.12286
Omenn, G. S., et al. (1996). Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease (CARET). NEJM, 334(18), 1150–1155. 10.1056/NEJM199605023341802
Klein, E. A., Thompson, I. M., Tangen, C. M., Crowley, J. J., Lucia, M. S., Goodman, P. J., et al. (2011). Vitamin E and the risk of prostate cancer: the Selenium and Vitamin E Cancer Prevention Trial (SELECT). JAMA, 306(14), 1549-1556. Among 35,533 men, vitamin E 400 IU/day significantly increased prostate cancer risk (about 17% relative increase; 76 vs 65 cases per 1,000). 10.1001/jama.2011.1437
Bjelakovic, G., Nikolova, D., & Gluud, C. (2013). Antioxidant supplements to prevent mortality. JAMA, 310(11), 1178–1179. 10.1001/jama.2013.277028
Estruch, R., Ros, E., Salas-Salvadó, J., Covas, M.-I., Corella, D., Arós, F., et al. (2018). Primary prevention of cardiovascular disease with a Mediterranean diet supplemented with extra-virgin olive oil or nuts. New England Journal of Medicine, 378(25), e34. PREDIMED RCT (~7,447 high-risk adults); a Mediterranean diet supplemented with mixed nuts (or EVOO) reduced major cardiovascular events versus a control low-fat diet. (First published 2013; retracted in 2018 over randomization issues and republished the same year with corrected data.). 7,447 adults aged 55-80 (57% women) at high cardiovascular risk; median 4.8 years. Events 3.8% (olive oil), 3.4% (nuts), 4.4% (control); adjusted HR 0.69 (0.53-0.91) and 0.72 (0.54-0.95). Results similar after omitting 1,588 participants whose assignments departed from protocol (household members enrolled without randomisation, some participants at 1 of 11 sites assigned without randomisation, inconsistent use of randomisation tables at another site) (abstract, PMID 29897866). 10.1056/NEJMoa1800389
Fiolet, T., Srour, B., Sellem, L., Kesse-Guyot, E., Allès, B., Méjean, C., et al. (2018). Consumption of ultra-processed foods and cancer risk: results from NutriNet-Santé prospective cohort. BMJ, 360, k322. NutriNet-Santé prospective cohort (observational): per 10% more ultra-processed food in the diet, overall cancer HR 1.12 (1.06-1.18; 2228 cases) and breast cancer HR 1.11 (1.02-1.22; 739 cases), still significant after adjusting for markers of diet quality (abstract, PMID 29444771). 10.1136/bmj.k322