Diets in Oncology

WHAT THIS DIET IS AND WHERE IT CAME FROM. The low-bacterial diet grew out of simple reasoning: if a patient after high-dose chemotherapy has very few granulocytes, cut off the bacteria that come with food. For decades it was standard practice on haematology and transplant wards. Typical restrictions include:

• raw vegetables and fruit, especially those that cannot be peeled,
• mould-ripened and soft ripened cheeses, unpasteurised products,
• raw or undercooked meat, fish and eggs,
• loose nuts, sprouts, unpasteurised honey.

WHAT THE EVIDENCE SHOWS

A 2025 systematic review covered 5 randomised trials and a total of 510 adults after high-dose therapy and haematopoietic stem cell transplantation; none of the analysed endpoints - infections, gastrointestinal complications, mortality, length of hospital stay - favoured the neutropenic diet ^[1].

A second 2025 systematic review, published together with a clinical practice guideline, pooled 8 randomised trials in children with cancer and haematopoietic cell transplant recipients. The data were consistent: food restrictions lack clinically significant benefit in preventing infections, and the panel issued a conditional recommendation AGAINST their use ^[2].

A single randomised trial in paediatric oncology illustrates this best: among 150 participants, infection during neutropenia occurred in 33% of those following food safety guidelines alone and in 35% of those additionally following the neutropenic diet - no statistically significant difference, while the restrictive arm reported a clearly greater adherence burden ^[5].

An earlier 2019 meta-analysis reached the same conclusion in adults: no statistically significant difference in major infections or in bacteraemia and fungaemia between the neutropenic and the regular diet, with a slightly HIGHER infection risk on the restrictive diet in the stem cell transplant subgroup. The authors conclude that there is no evidence to support the use of a neutropenic diet or other food restrictions in neutropenic patients with cancer, and refer instead to food-handling guidelines ^[4].

A 2023 systematic review (12 reports, 3,469 participants) adds the quality-of-life dimension: the low-bacterial diet did not reduce infection or mortality rates compared with a free diet, and was associated with poorer quality of life resulting from a limited and less palatable menu ^[6].

STRENGTH OF EVIDENCE

Moderate and consistent in direction. In its favour: the data come from randomised trials rather than observations, and several independent systematic reviews converge on the same conclusion ^[1][2][4][6]. Against full certainty: the individual trials are small, the definition of the low-bacterial diet varies between centres, and the paediatric guideline recommendation is CONDITIONAL rather than strong ^[2]. The honest summary is therefore: no benefit has been demonstrated - not "the diet has been proven harmful".

DOES LIBERALISING THE DIET HELP BY ITSELF

The honest answer is: on its own, not much. In a randomised trial in patients undergoing haematopoietic stem cell transplantation, a liberalised diet encouraging fresh fruit and vegetables did not improve energy intake during neutropenia (678 versus 724 kcal per day; p=0.46) or protein intake ^[3]. Lifting the restrictions is therefore not by itself a remedy for malnutrition - planned nutritional support is what is needed.

WHAT THESE RESULTS DO NOT SAY

They do not say that food hygiene no longer matters. Safe food-handling rules - thorough washing, avoiding raw meat, fish and eggs, respecting use-by dates, separate boards and utensils for raw produce, avoiding food of uncertain origin - remain recommended regardless of whether a list of banned products is used ^[4][7]. The difference concerns restrictive lists, not hygiene itself.

WHO THIS MATTERS FOR

Primarily patients with haematological malignancies treated with high-dose chemotherapy and stem cell transplantation, because that is the population studied ^[1][2]. For patients with solid tumours who become neutropenic after standard chemotherapy, evidence of this quality does not exist - which does not mean the neutropenic diet helps them, only that it has not been tested in them.

WHAT TO DISCUSS WITH THE TREATING TEAM

Centre practice varies and some units still apply restrictions. The scope of the diet during neutropenia is set by the treating team, which knows the patient's nutritional status and the ward's epidemiological situation.

WHY THIS IS A TOPIC AT ALL

A supplement suggests risk; a fruit does not. A patient who honestly lists every preparation they take will usually not mention the daily glass of grapefruit juice. Yet this is one of the best-documented food-drug interactions in all of pharmacology, and in oncology it concerns drugs taken by mouth, daily, for many months.

MECHANISM: The furanocoumarins present in grapefruit inhibit the cytochrome P450 enzyme CYP3A4 ^[1]. CYP3A4 metabolises a great many anticancer drugs, so inhibiting it changes how much drug reaches the bloodstream. The key point is that this change happens OUTSIDE dosing control: the patient takes the prescribed dose, while the body behaves as if it had received a different one.

WHAT HAS BEEN MEASURED — THE NUMBERS:

• NILOTINIB: in a clinical trial, combination with grapefruit juice increased the area under the concentration curve by 29% and peak serum concentration by 60% ^[1];
• SUNITINIB: a clinical trial observed increased bioavailability after grapefruit juice exposure ^[1];
• IMATINIB: a clinical review indicates the juice may raise plasma levels by inhibiting CYP3A4 and trigger organ toxicity ^[1];
• ETOPOSIDE: the OPPOSITE direction — after grapefruit juice pretreatment, bioavailability fell from about 73% to 52%, corresponding to a 26% reduction in area under the curve ^[1].

WHY THE DIRECTION CAN REVERSE

This is the most commonly misunderstood part. The popular belief is that grapefruit makes drugs stronger. With etoposide the opposite occurred, and that situation is the more dangerous one, because lower exposure to an anticancer drug means weaker treatment than planned and produces no symptom by which it could be noticed. Toxicity is visible; loss of efficacy is not.

NOT ONLY GRAPEFRUIT

Alongside grapefruit, NCI names fruits with comparable interaction potential: Seville orange (an ingredient of marmalades and some liqueurs), pomelo and lime ^[1].

WHAT THIS ENTRY DOES NOT SAY

It neither recommends nor discourages eating grapefruit for any individual and gives no safe quantities — NCI publishes interaction tables but makes no explicit recommendation about avoiding grapefruit during cancer therapy ^[1], and the answer depends on the specific drug a patient is taking. Most intravenous chemotherapy is unaffected; the issue concerns chiefly oral drugs metabolised by CYP3A4.

WHAT TO TELL THE TREATING TEAM

About regular consumption of grapefruit juice or these fruits — ideally before oral treatment begins. The question to ask the doctor or clinical pharmacist is simply: is my drug metabolised by CYP3A4. That single question settles the matter, and the answer is in the summary of product characteristics.

Sources

1. [1] National Cancer Institute — Cancer Therapy Interactions With Foods and Dietary Supplements (PDQ), Health Professional Version (furanokumaryny i CYP3A4; nilotynib AUC +29% i Cmax +60%; sunitynib; imatynib; etopozyd 73% do 52%, AUC -26%; pomarancza sewilska, pomelo, limonka): cancer.gov ↗

WHAT IARC CONCLUDED

In 2016 an IARC Working Group reviewed more than 1000 observational and experimental studies covering over 20 cancer sites and found the evidence for the carcinogenicity of coffee drinking INADEQUATE. Coffee was moved from Group 2B (possibly carcinogenic), where it had sat since 1991, to Group 3 — not classifiable ^[1]. IARC notes that the earlier classification was a valid reading of the data available in 1991; the change reflects accumulated evidence, not a past error.

WHAT THE WCRF GRADING SAYS

The World Cancer Research Fund and the American Institute for Cancer Research grade the evidence that coffee DECREASES risk as probable for two cancers: hepatocellular carcinoma and cancer of the uterine corpus (endometrium) ^[2]. In the WCRF scheme, "probable" is the second strongest grade, after "convincing". For endometrial cancer the association was seen for both caffeinated and decaffeinated coffee, which argues against attributing it to caffeine alone ^[2].

WHERE THE REAL RISK LIES — TEMPERATURE, NOT THE DRINK. The same IARC evaluation classified drinking VERY HOT beverages in Group 2A (probably carcinogenic to humans) in relation to squamous cell carcinoma of the oesophagus; this concerns beverages consumed at around 65°C and above ^[1]. IARC points out that in most countries coffee and tea are not drunk at such temperatures — the habit is common in parts of South America, Central Asia and East Africa. The practical content of this evaluation therefore differs from the headline "hot drinks cause cancer": what was assessed is temperature, not coffee, tea or mate.

WHAT THESE GRADINGS DO NOT SAY

An IARC classification describes the STRENGTH OF EVIDENCE for carcinogenicity, not the size of the risk — Group 2A does not mean an agent is as hazardous as every other agent in that group. The WCRF grading, in turn, describes associations observed across populations; it does not support the conclusion that drinking coffee will protect a given person from liver cancer, or that someone who does not drink coffee should start. Neither grading concerns the TREATMENT of cancer.

WHAT THIS MEANS FOR A PATIENT IN TREATMENT

The data above concern the risk of developing cancer in the general population, not management during therapy. Caffeine can interact with some medicines and affect sleep, hydration and gastrointestinal symptoms, and dietary advice during chemotherapy, radiotherapy and after surgery is set individually. Decisions about diet during cancer treatment are made by the treating physician together with the care team.

SOURCES:

• ^[1] IARC Monographs Volume 116: Drinking Coffee, Mate, and Very Hot Beverages (2016/2018) — IARC news release and materials.
• ^[2] World Cancer Research Fund / American Institute for Cancer Research, Continuous Update Project — coffee and liver cancer, coffee and endometrial cancer.

WHY THIS LOCATION IN PARTICULAR

NCI states explicitly that nutrition complications are usually most notable and severe with tumors involving the digestive tract or head and neck, owing to mechanical obstruction or dysfunction ^[1]. For this location NCI lists dysphagia and odynophagia, xerostomia and taste changes as typical consequences, and for radiotherapy of the head and neck region — xerostomia, mucositis and taste changes, with thick saliva as a later effect ^[1].

THE SCALE OF THE PROBLEM

A systematic review published in 2026, covering 31 studies, reports that malnutrition affects 30–50% of patients at diagnosis and up to 90% during therapy ^[2]. This is why nutritional status is treated here as part of the treatment plan rather than a side issue: malnutrition directly affects whether the planned radiotherapy or chemoradiotherapy can be delivered without interruption.

WHAT THE EVIDENCE SAYS ABOUT EACH FORM OF SUPPORT. The same review compares three approaches ^[2]:

• oral nutritional supplements — early use was associated with fewer treatment interruptions, particularly before escalation to enteral feeding;
• prophylactic percutaneous endoscopic gastrostomy — associated with reduced weight loss, fewer unplanned hospitalisations and higher treatment completion rates, although the authors explicitly caution that these findings must be interpreted carefully because the available evidence is observational;
• nasogastric tube feeding — associated with fewer device-related complications and lower long-term dependence, but heterogeneity across studies and the predominance of gastrostomy-focused evidence prevented direct comparison.

WHAT THE EVIDENCE DOES NOT SETTLE

The review authors put it plainly: comparative evidence remains insufficient to define the optimal enteral approach, and mucositis, dysphagia and impaired oral intake remained the most significant outcomes DESPITE the implementation of multiple nutritional support measures ^[2]. In other words, nutritional support mitigates the consequences of treatment but does not remove them, and the choice between gastrostomy and nasogastric tube has no randomised-trial answer today. High-quality randomised trials integrating oral health-related outcomes are needed ^[2].

STRENGTH OF EVIDENCE

Moderate. In favour: the data come from a systematic review conducted in accordance with PRISMA, with methodological quality appraisal of the included studies, and the malnutrition figures are consistent and high ^[2]. Against full certainty: the key observations on the benefit of prophylactic gastrostomy come from observational studies rather than randomisation, and are described as such by the authors ^[2].

WHAT THIS ENTRY DOES NOT SAY

It does not indicate which method of feeding any individual patient should choose, or when a gastrostomy should be placed. That decision is made by the treating team — radiation oncologist, head and neck surgeon, clinical dietitian and speech and language therapist — on the basis of nutritional status, the radiotherapy plan and symptom severity. The service does not interpret individual clinical situations.

Sources

1. [1] National Cancer Institute — PDQ: Nutrition in Cancer Care (Health Professional Version): cancer.gov ↗
2. [2] Nutritional management and oral health-related outcomes in head and neck cancer treated with radiotherapy or chemoradiotherapy: a systematic review, BMC Oral Health 2026 (PMID 42010533): pubmed.ncbi.nlm.nih.gov ↗

WHAT ORAL NUTRITIONAL SUPPLEMENTS ARE

They are ready-made liquid or powdered products registered as foods for special medical purposes. In a small volume they provide concentrated energy, protein, fat and carbohydrate, and often fibre, vitamins and minerals. NCI stresses that they are not intended to serve as the sole source of nutrition — they supplement what the patient eats ^[1].

WHAT THE EVIDENCE SHOWS

The strength of evidence is moderate and concerns nutritional indicators rather than survival.

• Multinutrient, high-protein supplements significantly improved total energy and protein intake and reduced the incidence of complications ^[1].
• In head and neck cancer treated with radiotherapy or chemoradiotherapy, early use of supplements was associated with fewer treatment interruptions, particularly before escalation to tube feeding; the review authors note that comparative data remain too limited to define the optimal approach ^[2].
• In perioperative care (a review of patients undergoing prostate cancer surgery), supplements form part of the recommended pathway alongside early screening for malnutrition ^[3].
• A review of 38 studies found NO benefit from fish-oil-enriched supplements in the treatment of cachexia in advanced cancer ^[1].

WHAT SUPPLEMENTS DO NOT SOLVE

A supplement will not by itself halt progressive cancer cachexia, nor replace treatment of the cause of appetite loss. It is not an anticancer drug and there are no data showing an effect on the outcome of cancer treatment.

A PRACTICAL LIMITATION PATIENTS THEMSELVES RAISE

A review of patient experiences after colorectal surgery found that experiences with oral nutritional supplements vary widely, and that palatability largely determines whether they are consumed ^[4]. This is not a minor detail: a supplement that is not drunk does not nourish, and unacceptable taste is a common reason for abandoning advice.

WHERE TO SEEK HELP

Choosing a supplement (energy density, protein content, form, fibre content, suitability in diabetes or renal impairment) is the responsibility of the treating physician and a clinical dietitian. This site does not give dietary advice and does not recommend specific products.

Sources

1. [1] National Cancer Institute — Nutrition in Cancer Care (PDQ), Health Professional Version (oral nutrition supplements; multinutrient high-protein products; systematic review of 38 studies on fish oil in cachexia): cancer.gov ↗
2. [2] Nutritional management and oral health-related outcomes in head and neck cancer treated with radiotherapy or chemoradiotherapy: a systematic review. BMC Oral Health 2026 (PMC13267656): europepmc.org ↗
3. [3] Evidence-based perioperative nutritional management for patients undergoing prostate cancer surgery: a systematic review. Gland Surgery 2026 (PMC13264766): europepmc.org ↗
4. [4] Patient Experiences of Nutrition in Enhanced Recovery After Colorectal Surgery: A Systematic Review. Nutrients 2026 (PMC13258613): europepmc.org ↗

WHAT IT IS

Enteral nutrition means delivering a prepared nutrient formula directly into the stomach or intestine through a tube — a narrow catheter passed through the nose or through the abdominal wall. Parenteral nutrition means delivering nutrients intravenously, bypassing the digestive tract altogether. Both serve the same purpose: meeting the needs of a patient who cannot meet them by eating.

WHEN IT IS CONSIDERED AT ALL

The National Cancer Institute PDQ summary ties nutrition support to the situation of a patient who is malnourished and unable, for a prolonged period, to take in adequate nutrients by mouth ^[1]. It is therefore not a dietary top-up "for strength", but a response to a specific, ongoing problem with food intake.

WHY THE ENTERAL ROUTE COMES FIRST

PDQ states the principle plainly: if the patient has a functioning gastrointestinal tract, enteral nutrition is the route used ^[1]. Feeding through the gut keeps it working and avoids the complications associated with central venous access. Parenteral nutrition is reserved for situations in which the gut does not work or is obstructed.

HAEMATOPOIETIC CELL TRANSPLANTATION — THE ONE EXPLICIT TIME THRESHOLD. For patients undergoing haematopoietic cell transplantation, PDQ gives a concrete criterion: nutrition support is given to those who are malnourished and expected to be unable to ingest or absorb adequate nutrients for a prolonged period, defined as more than 7-14 days ^[1]. This is the only figure the source states in this context, and it applies only to this group of patients.

WHAT IT DOES NOT SOLVE

Nutrition support is not a cancer treatment and does not replace anticancer therapy. PDQ presents no evidence for routinely adding it to standard chemotherapy or radiotherapy in patients who are eating adequately; for that period it puts symptom control and practical coping strategies first ^[1].

WHAT IS OUTSIDE THE SCOPE OF THIS PAGE

Detailed starting thresholds, choice of formula, rate of delivery and prevention of refeeding complications are matters for clinical guidelines and individual assessment — they belong to the treating team and the responsible physician, not to an information page. This page describes what these methods are and the logic by which they are ordered, not when any particular person should use them.

SOURCES.

• ^[1] National Cancer Institute, PDQ Nutrition in Cancer Care (health professional version)

Sources

1. National Cancer Institute — PDQ Nutrition in Cancer Care (health professional version): cancer.gov ↗

WHAT THE EVIDENCE SHOWS

A meta-analysis published in Nutrients (2026) on adherence to the Mediterranean diet AFTER a cancer diagnosis, with emphasis on colorectal cancer, included eight studies ^[2]. Six prospective cohort studies (6,697 participants in total) assessed recurrence: higher adherence was associated with a lower risk of recurrence or cancer-specific death (HR 0.83; 95% CI 0.70-0.99; I2 = 49%, i.e. moderate heterogeneity) ^[2]. Two randomised trials (76 participants in total) assessed cancer-related fatigue and found no significant effect (MD 0.29; 95% CI -0.58 to 1.16 — the interval includes zero) ^[2]. The authors themselves note that both results are limited by the small number of studies and methodological heterogeneity, and describe the fatigue evidence as exploratory and statistically imprecise ^[2].

STRENGTH OF EVIDENCE

Moderate and largely observational. That distinction matters most here: in cohort studies, people who eat this way are also more likely to be physically active, not to smoke and to have better access to care — each of which improves prognosis on its own. The result must therefore not be read as "the Mediterranean diet lowers recurrence risk by 17%" ^[2]. For primary prevention the NCI position is more cautious still: population studies have not yet shown definitively that any dietary component causes or protects against cancer ^[1].

WHAT IS WELL DOCUMENTED

One element of this pattern is limiting alcohol, and here the evidence is strong and points the other way: according to the NCI, heavy or regular alcohol consumption INCREASES the risk of cancers of the oral cavity, pharynx, larynx, oesophagus, liver, breast, colon and rectum ^[1]. If one dietary decision has the best-documented bearing on cancer risk, it is limiting alcohol — not selecting "anticancer" foods.

WHO IT SUITS

The pattern aligns with general healthy-eating advice and as such is safe for most people, including during and after cancer treatment. Exceptions need individual assessment: patients with malnutrition or cachexia need an energy- and protein-dense diet rather than a restrictive one, and patients with neutropenia, mucositis, diarrhoea or recent gastrointestinal surgery receive separate advice from their treating team.

WHAT THIS DIET DOES NOT DO

It does not treat cancer and does not replace cancer treatment. None of the studies cited tested the diet as anticancer therapy, and diet is never a reason to delay or modify treatment.

Sources

1. National Cancer Institute — Risk Factors: Diet, cancer.gov 2024: cancer.gov ↗
2. Papamichael D, Felekkis K, Andreou EP — Post-Diagnosis Adherence to the Mediterranean Diet and Cancer Recurrence and Fatigue Outcomes in Cancer Survivors, with Emphasis on Colorectal Cancer: A Systematic Review and Meta-Analysis, Nutrients 2026 (PMID 41829977): europepmc.org ↗

WHAT IT IS

An international expert panel (Lancet Oncology, 2011) defined cancer cachexia as a multifactorial syndrome of ongoing loss of skeletal muscle mass — with or without loss of fat mass — that CANNOT be fully reversed by conventional nutritional support and leads to progressive functional impairment ^[1]. It is driven by a negative protein and energy balance arising from a variable combination of reduced food intake and abnormal metabolism ^[1]. That last part is the key to this entry: cachexia is not simply undereating. The National Cancer Institute puts it plainly: wasting "can occur even when you are eating well" ^[4].

HOW IT IS DIAGNOSED

The agreed criterion is weight loss greater than 5%, or greater than 2% in people already depleted by body-mass index (below 20 kg/m²) or by skeletal muscle mass (sarcopenia) ^[1]. The consensus describes stages — precachexia, cachexia and refractory cachexia — and states that assessment should cover four domains: anorexia or reduced food intake, catabolic drive, muscle mass and strength, and functional and psychosocial impairment ^[1].

HOW COMMON, AND HOW OFTEN MISSED. A retrospective record analysis of 76 547 cancer patients (Journal of Cachexia, Sarcopenia and Muscle, 2026) found cachexia in 2.42% of the whole cohort but in 29.3% of 6946 patients with colorectal, pancreatic or bronchial/lung cancer ^[2]. What matters most is what the authors found inside the records themselves: ICD coding identified 620 patients and free-text analysis a further 1507; in addition, 1340 patients had a BMI below 20 or more than 5% weight loss yet were not coded as cachectic ^[2]. The authors conclude that standard coding substantially underestimates the prevalence of cachexia ^[2].

WHY IT MATTERS TO THE PATIENT

A systematic review of 37 studies covering 52 053 patients with solid tumours (Journal of Cachexia, Sarcopenia and Muscle, 2026) found cachexia associated with worse physical function (a statistically significant association in 25 of 31 studies, 80.6%), worse health-related quality of life (22 of 24, 91.7%), worse performance status (14 of 16, 87.5%) and worse activities of daily living (6 of 6) ^[3]. The same review exposed a real limitation of the literature: eleven different definitions of cachexia were used, and the 2011 consensus criteria in only 45% of studies, so comparisons between studies carry uncertainty ^[3].

THE SCOPE OF WHAT WE WRITE HERE

We deliberately do NOT give calorie targets, grams of protein per kilogram, or enteral and parenteral feeding regimens. Those decisions belong to the treating team and clinical dietitian, because they depend on the diagnosis, the stage of treatment, gut function and comorbidities. Nor do we know of — and we did not find in reference-class sources — any diet or supplement that can honestly be said to reverse cachexia; the consensus definition itself states that conventional nutritional support does not ^[1]. Offers of "anti-wasting diets" and muscle-rebuilding preparations therefore call for particular caution: a promise that the best-documented nutritional intervention does not keep will not be kept by a product sold without such documentation.

WHAT FOLLOWS IN PRACTICE

Early recognition matters, because the consensus describes a precachexia stage preceding the full syndrome ^[1]. Weight loss is measurable and worth recording — the record-based data show it is often present in measurements yet never reaches the diagnosis ^[2]. Raising unintended weight loss with the treating team is therefore reasonable even when the patient "is eating normally" ^[4].

Sources

1. Fearon K i wsp. — Definition and classification of cancer cachexia: an international consensus, The Lancet Oncology 2011: europepmc.org ↗
2. Colardelle Y i wsp. — Cancer Cachexia Prevalence Is Underestimated in Medical Records of Patients in a Regional Tertiary Hospital, Journal of Cachexia, Sarcopenia and Muscle 2026: europepmc.org ↗
3. Crawford J i wsp. — Functional Outcomes and Quality of Life for Patients With Cachexia and Solid Tumour Cancers: Findings of a Systematic Literature Review, Journal of Cachexia, Sarcopenia and Muscle 2026: europepmc.org ↗
4. National Cancer Institute — Nutrition During Cancer Treatment, cancer.gov 2024: cancer.gov ↗

WHAT THE INSTITUTIONS RECOMMEND

In Diet, Nutrition, Physical Activity and Cancer (2018) the World Cancer Research Fund recommends that wholegrains, non-starchy vegetables, fruit and pulses form the basis of the everyday diet. The numerical goals are at least 30 g of dietary fibre from food per day and at least five portions, totalling a minimum of 400 g, of a variety of non-starchy vegetables and fruit per day ^[1]. Non-starchy vegetables are vegetables other than potatoes, sweet potatoes and other starchy tubers.

HOW STRONG IS THE EVIDENCE - THE DISTINCTION THAT USUALLY GETS LOST. WCRF grades its evidence, and for vegetables and fruit it did not award the highest grade:

• evidence in the strong category (probable) applies to the combined effect of non-starchy vegetables and fruit on aerodigestive cancers considered as one group;
• for individual cancer sites the grade falls to limited-suggestive. Non-starchy vegetables: mouth, pharynx and larynx, nasopharynx, oesophagus (adenocarcinoma and squamous cell carcinoma), lung in people who smoke or used to smoke, and oestrogen-receptor-negative breast cancer. Fruit: squamous cell carcinoma of the oesophagus and lung cancer in people who smoke or used to smoke ^[1];
• the limited-suggestive category means the findings are broadly consistent in direction but rarely sufficient to justify recommendations for reducing cancer risk ^[1].

WHAT NCI SAYS ABOUT CRUCIFEROUS VEGETABLES

In its fact sheet on cruciferous vegetables (broccoli, cabbage, Brussels sprouts, cauliflower) the National Cancer Institute states that human studies have shown mixed results ^[2]. For prostate cancer, large cohort studies found little or no association, although some case-control studies suggested a lower risk. For colorectal, lung and breast cancer the picture is similarly inconsistent, and a meta-analysis in breast cancer found no association. NCI presents cruciferous vegetables as part of overall vegetable intake rather than as a separate risk-reducing measure ^[2].

WHY THE RECOMMENDATION IS NEVERTHELESS FIRM

The value of this food group does not rest solely on a direct effect on cancer risk. A diet rich in vegetables, fruit and wholegrains makes it easier to maintain a healthy body weight, and excess body weight is one of the best-documented nutritional risk factors for cancer ^[1]. Separate, strong evidence links dietary fibre and wholegrains to a lower risk of colorectal cancer ^[1] - this catalogue covers that in its own entry.

WHAT THIS ENTRY DOES NOT SAY

There is no evidence that vegetables or fruit treat cancer or replace any form of treatment. Nor is there a basis for singling out any individual product as anticancer - the evidence concerns the overall dietary pattern rather than single components. Nutrition during cancer treatment, in particular in malnutrition, swallowing disorders, neutropenia or after gastrointestinal surgery, is determined by the treating physician together with a clinical dietitian.

WHAT THE RECOMMENDATION COVERS

In Diet, Nutrition, Physical Activity and Cancer (2018) the World Cancer Research Fund frames it as limiting consumption of processed foods high in fat, starches or sugars, including foods described as fast foods ^[1]. It concerns energy-dense products that make it easy to exceed energy requirements, rather than one category of dish or any particular restaurant chain.

HOW STRONG IS THE EVIDENCE

• Strong evidence: diets containing greater amounts of fast foods and other processed foods are a cause of weight gain, overweight and obesity, because they increase the risk of excess energy intake ^[1].
• Probable evidence: consuming fast foods is a cause of weight gain, overweight and obesity; the same applies to a Western-type diet as an overall dietary pattern ^[1].
• Probable evidence, this time for CANCER directly: a high dietary glycaemic load is a cause of endometrial cancer ^[1].

WHY THE RECOMMENDATION STILL MATTERS ALTHOUGH THE LINK IS INDIRECT. For most cancers the effect of this food group has not been demonstrated directly - it runs through body weight. That indirect route is not a weak one: excess body weight is among the best-documented nutritional risk factors for cancer and is linked to more than a dozen diagnoses. This catalogue covers it in a separate entry that lists the specific cancers together with their qualifications (for example, breast cancer only after the menopause). Limiting highly processed foods is therefore a tool for maintaining body weight, not a standalone anticancer measure.

WHAT THIS ENTRY DOES NOT SAY

There is no evidence that a single fast-food meal increases cancer risk, or that avoiding one protects against the disease. The recommendation concerns a sustained dietary pattern, not isolated episodes. The opposite situation deserves separate attention: in patients with malnutrition or cancer cachexia, restricting energy intake can be harmful, and nutrition during cancer treatment is determined by the treating physician together with a clinical dietitian - this catalogue describes prevention at population level, not management of an individual person.

WHAT THIS ENTRY COVERS

This is about nutritional management delivered by the treating team around planned cancer surgery, especially of the gastrointestinal tract. It comprises three distinct things that should not be conflated: assessment of nutritional risk before surgery, shortening the fasting period and returning to oral intake early after surgery, and so-called immunonutrition formulas (enriched with arginine, omega-3 fatty acids and nucleotides, among others). The first two are part of standard perioperative care; the third is an add-on intervention, and it is the subject of most of the studies cited here.

WHAT THE EVIDENCE SHOWS

An umbrella review of 54 meta-analyses in patients undergoing cancer surgery (BMC Cancer, 2026) reported, in favour of immunonutrition:

• overall complications RR 0.79 (95% CI 0.70–0.88)
• infectious complications RR 0.61 (0.58–0.65)
• surgical-site infection RR 0.66 (0.59–0.74)
• mortality HR 0.86 (0.74–0.99)
• length of hospital stay shorter by a mean of 1.75 days (–2.09 to –1.41)

The ESPEN guideline on clinical nutrition in surgery (2025 update) states the general principles: start nutritional therapy as soon as nutritional risk becomes apparent, avoid long periods of preoperative fasting, and re-establish oral feeding, the preferred route, as early as possible after surgery.

WHERE THE SOURCES DIVERGE — AND WHY WE SAY SO PLAINLY. A meta-analysis of 9 randomised trials in colorectal cancer (1,198 patients, Frontiers in Nutrition, 2026) confirms the reduction in infections (OR 0.48; 0.34–0.66) but does NOT confirm an effect on anastomotic leak, postoperative ileus or length of stay — precisely the endpoints where the umbrella review sees benefit. We resolve this in favour of the more cautious reading, on two methodological grounds. First, an umbrella review pools meta-analyses that partly rest on the same primary trials, so a single result can be counted several times. Second, its length-of-stay estimate has heterogeneity of I² = 90.3%, meaning the individual studies disagree strongly and the averaged figure is unstable. One effect is reproducible across both sources: fewer infectious complications.

WHAT THESE RESULTS DO NOT SAY

They do not say that an immunonutrition formula replaces ordinary food, or that it improves cancer prognosis in itself — the endpoints measured concern the perioperative course, not cure of the tumour. Nor do they establish when to start: the observation that postoperative administration outperforms preoperative administration comes from a subgroup analysis and is therefore a hypothesis, not a finding. Finally, they say nothing about people who are not undergoing surgery.

SAFETY AND INTERACTIONS

Immunonutrition is a medical product used under hospital care, not a supplement bought independently; its composition (notably high arginine doses) has been questioned in patients with sepsis and severe organ failure. After gastrointestinal surgery, the route of delivery (oral, nasoenteric tube, jejunostomy) and the rate of volume escalation are part of the surgical plan. Whether, when and in what form nutritional support is introduced is decided by the treating team — the surgeon together with a clinical dietitian.