Aflatoxin, Mould and Safe Sourcing

Table of Contents

1. Why Peanuts Specifically
2. What Aflatoxin Is
3. How It Was Discovered
4. How It Causes Liver Cancer
5. The Hepatitis B Multiplication
6. The Global Burden, and Where It Falls
7. Children, Growth and Immunity
8. Regulatory Limits: US, EU and Codex
9. How a Regulated Supply Chain Removes It
10. What Works Where Regulation Cannot Reach
11. What You Can Actually Do at Home
12. What Does Not Work
13. Who Should Be Most Careful
14. Key Research Papers
15. Connections
16. Featured Videos

Why Peanuts Specifically

The peanut plant does something no other major crop does: it buries its own fruit. The flowers appear above ground and pollinate themselves, and then the base of the fertilised flower elongates into a stalk called a peg, or gynophore, which turns downward, drives several centimetres into the soil, and only there does the tip swell into a pod. Botanists call this geocarpy. Studies of gynophore and pod development describe the surface changes as the developing fruit sits in the earth, in continuous contact with soil organisms, for weeks.

That habit is why peanuts have an aflatoxin problem and, say, apples do not. The mould responsible, Aspergillus flavus, is a common soil saprophyte. It lives in the ground the peanut is ripening in. It cannot easily invade a healthy, undamaged kernel — but it readily colonises one that is cracked by insect damage, split during lifting, or shrunken and physiologically stressed by drought. A drought-stressed peanut crop is the classic high-aflatoxin crop, because the same dryness that stresses the plant also favours the mould over its competitors.

The second window is after harvest. Peanuts are lifted and left to cure, and if they are stacked damp, dried too slowly, rained on in the windrow, or stored in a warm humid shed, the mould that was already present multiplies and produces toxin. In a mechanised system with forced-air drying and moisture monitoring, that window is short. In a smallholder system where the crop dries on the ground and is stored in a sack in a hut, it is long.

Back to Table of Contents

What Aflatoxin Is

Aflatoxins are a family of related compounds produced by Aspergillus flavus and Aspergillus parasiticus. Four occur naturally in food — B1, B2, G1 and G2, named for whether they fluoresce blue or green under ultraviolet light. Aflatoxin B1 is the most abundant and by far the most toxic. A fifth, M1, appears in the milk of animals fed contaminated feed, which is why dairy is monitored separately and why contaminated groundnut cake used as cattle feed is a public-health problem even for people who never eat peanuts.

The International Agency for Research on Cancer classifies naturally occurring mixtures of aflatoxins in Group 1 — carcinogenic to humans, the category reserved for agents where the human evidence is considered sufficient rather than merely suggestive. That is the same category as tobacco smoke and asbestos. The classification describes the strength of the evidence that it causes cancer, not the size of the risk from any particular exposure, and the distinction matters: a substance can be an unambiguous carcinogen and still pose a negligible risk at the levels a given person actually meets.

Two properties make aflatoxin awkward to deal with in food. It is heat-stable — roasting reduces it somewhat but does not eliminate it, and no domestic cooking method destroys it. And it is extremely unevenly distributed. Contamination sits in a small number of badly affected kernels rather than spreading through a batch, so a single mouldy nut can carry more toxin than the rest of the bag combined. That unevenness is the reason sampling a lot for testing is a statistical science in its own right, and the reason throwing out a few bad-looking kernels is disproportionately effective.

Back to Table of Contents

How It Was Discovered

In 1960 something killed more than a hundred thousand turkey poults on English farms in a matter of months. The cause was unknown, the syndrome was named Turkey X disease, and the investigation converged on a shipment of Brazilian groundnut meal used in the feed. In 1962 a paper in Nature by Nesbitt and colleagues identified toxic metabolites of Aspergillus flavus in that meal — the compounds that would be named aflatoxins, after Aspergillus flavus toxin.

That episode is the origin of modern mycotoxin regulation. Within a decade aflatoxin had been linked to liver tumours in laboratory animals, testing methods had been standardised, and importing countries had begun setting limits on groundnut consignments. It is a rare case where the food-safety system now in place can be traced to a single identifiable outbreak, and it is worth knowing because it explains why the peanut trade in high-income countries is built around aflatoxin testing to a degree that no other contaminant commands.

Back to Table of Contents

How It Causes Liver Cancer

Aflatoxin B1 is not itself very reactive. It becomes dangerous in the liver, where cytochrome P450 enzymes — the same system that processes drugs and alcohol — convert it into aflatoxin B1-8,9-epoxide. That epoxide is highly electrophilic and binds directly to guanine bases in DNA.

Most of that damage is repaired. What matters is the fraction that is not. Hsu and colleagues reported in Nature in 1991 that hepatocellular carcinomas from regions of high aflatoxin exposure carried a striking mutational hot spot in the TP53 tumour suppressor gene — a specific G-to-T transversion at codon 249. TP53 is the gene that normally halts a damaged cell's division and, failing that, instructs it to die. Knock it out and a cell with accumulating damage keeps dividing. The codon 249 signature became one of the first molecular fingerprints in cancer epidemiology: a mutation whose pattern points back to a specific environmental cause.

Kensler and colleagues' review of five decades of aflatoxin research traces this whole arc — from animal bioassays through metabolic activation, DNA adducts and biomarkers to human prevention trials — and it remains the clearest single account of how the mechanism was established. Two features are worth carrying away. First, the process is cumulative: risk accrues with lifetime exposure, not with any one meal. Second, the biomarkers are good. Aflatoxin-albumin adducts in blood and aflatoxin metabolites in urine give a measurable, individual-level record of exposure over recent weeks, which is why the human epidemiology here is unusually strong for a dietary carcinogen.

Back to Table of Contents

The Hepatitis B Multiplication

The single most important fact about aflatoxin risk in an individual is that it is not the same for everybody, and the reason is chronic hepatitis B.

Ross and colleagues followed a cohort in Shanghai, measuring urinary aflatoxin biomarkers and hepatitis B surface antigen status, and reported that liver cancer risk rose with aflatoxin exposure, rose with chronic hepatitis B infection, and rose far more steeply in people who had both than either factor alone would predict. The interaction is multiplicative rather than additive. Chronic viral infection keeps liver cells in a continuous cycle of injury and regeneration; a dividing cell has less opportunity to repair DNA damage before copying it. Aflatoxin supplies the damage, hepatitis supplies the proliferation, and the combination is much worse than the sum.

This is also why the geography of aflatoxin-attributable liver cancer maps so closely onto the geography of chronic hepatitis B, and why the two great interventions against it are not primarily food-safety measures at all: hepatitis B vaccination of infants, and reducing dietary exposure. Vaccination programmes in East and West Africa and East Asia have already changed the picture for the generations born under them.

Back to Table of Contents

The Global Burden, and Where It Falls

Liu and Wu published a quantitative risk assessment in Environmental Health Perspectives in 2010 estimating the number of hepatocellular carcinoma cases attributable to aflatoxin worldwide each year, combining exposure estimates by region, hepatitis B prevalence and cancer potency factors. Their central estimate placed aflatoxin's contribution in the range of a few per cent to well over a quarter of all global liver cancer cases, with the enormous width of that range reflecting genuine uncertainty in regional exposure data rather than vagueness about the mechanism. The essential finding is not the exact number. It is the distribution: almost the entire burden falls in sub-Saharan Africa, Southeast Asia and China, in populations eating home-grown maize and groundnuts that never pass a testing laboratory.

Wild and Gong made the broader case that mycotoxins are a substantially under-recognised global health problem, precisely because their effects are chronic and statistical rather than acute and visible. Nobody in a village notices aflatoxin. There is no outbreak, no cluster of sick people on a single day — just liver cancer appearing at rates several times higher than elsewhere, decades later. Acute aflatoxicosis outbreaks do occur, most notoriously in Kenya, where large numbers of people fell ill and many died after eating heavily contaminated home-stored maize, but these are the exception; the ordinary harm is slow.

For a reader in a country with enforced limits, the honest framing is this: you are being protected by an infrastructure, and the infrastructure is the reason your risk is low. Not because peanuts are inherently clean.

Back to Table of Contents

Children, Growth and Immunity

Cancer decades later is not the only concern. Gong and colleagues measured aflatoxin-albumin adducts in young children in Benin and Togo and reported a strong association between higher adduct levels and impaired growth — both stunting and being underweight — in a cross-sectional study of hundreds of children. The association held after adjustment for the obvious confounders, and a follow-up strengthened the picture.

Association is not proof of causation, and children with high aflatoxin exposure differ from children with low exposure in many ways, poverty first among them. But the finding is biologically plausible — aflatoxin impairs protein synthesis and damages the gut and immune system in animal models — and it has been reproduced across several African settings. It matters here because it reframes aflatoxin from a distant cancer risk into a possible contributor to childhood undernutrition in exactly the places where groundnut is a key weaning food, including in the ready-to-use therapeutic pastes made from peanut that are used to treat severe acute malnutrition. Those products are made under aflatoxin specifications for this reason.

Children are also more exposed per kilogram of body weight than adults for the same portion, and drink more milk, which is where aflatoxin M1 arrives.

Back to Table of Contents

Regulatory Limits: US, EU and Codex

Different jurisdictions have set different numbers, and the gap between them is one of the more instructive facts in food regulation.

• United States. The Food and Drug Administration applies an action level of 20 parts per billion for total aflatoxins in food for human consumption, with a separate and much lower limit for aflatoxin M1 in milk. Product above the action level is subject to enforcement.
• European Union. Substantially stricter, and split by intended use. For groundnuts intended for direct human consumption the maximum levels are in the low single digits of parts per billion for aflatoxin B1 and only slightly higher for total aflatoxins; groundnuts destined for sorting and further processing before consumption are allowed higher levels, on the reasoning that the processing will remove the contamination.
• Codex Alimentarius, the international reference standard, sits between the two and is what many countries adopt where they lack their own testing capacity.

The EU's tighter standard has been argued about for two decades. Analyses of the trade effects concluded that the difference between the European and international standards produced only a very small reduction in projected health risk in European consumers while measurably reducing African groundnut exports — an argument that the strictest attainable number is not automatically the best public-health policy when it prices a whole region's crop out of a market and leaves it to be eaten, untested, at home. Reasonable people disagree about the conclusion. What is not in dispute is that the limits differ by roughly an order of magnitude, and that both are far below the levels associated with harm in the epidemiological studies.

Always check the current official text rather than a number on a web page, including this one. Limits are revised. The authoritative sources are the FDA's compliance policy guides, the EU's contaminants regulation, and the relevant Codex standard.

Back to Table of Contents

How a Regulated Supply Chain Removes It

Understanding the steps is what makes the risk assessment concrete rather than a matter of trust.

1. Agronomy. Irrigation where available, because drought stress is the biggest single field-level driver; timely harvest, because leaving a mature crop in the ground invites both insect damage and moisture cycling; and resistant or early-maturing varieties.
2. Rapid drying. Getting kernel moisture down below roughly 8–10 per cent quickly, and keeping it there, stops the mould growing. Mechanical forced-air drying does in hours what sun-drying on the ground does unreliably in days.
3. Sampling and testing. Because contamination is so uneven, a valid sample is large and taken from many points in the lot, then ground and subsampled. This is the step that most distinguishes a regulated from an unregulated chain, and it is expensive.
4. Physical sorting. The workhorse. Contaminated kernels are disproportionately the shrivelled, discoloured, split and immature ones, so removing them removes most of the toxin. Screening by size, density separation, and electronic colour sorting — machines that photograph every kernel in a falling stream and blow the off-colour ones aside — between them can cut aflatoxin in a lot by an order of magnitude.
5. Blanching. Removing the skin also removes damaged surface material and further reduces measured toxin.
6. Segregation by destination. Lots that fail the direct-consumption limit are diverted to crushing for oil — refining removes aflatoxin from the oil — or rejected, rather than blended into food.

The result is that surveys of retail peanut butter and roasted peanuts in regulated markets typically find aflatoxin undetectable or at the very low end of the parts-per-billion scale. Peanut butter is, somewhat counter-intuitively, one of the better-monitored foods on the shelf: it is made in large industrial batches from tested lots by companies with real legal exposure, and it is routinely sampled by regulators precisely because of this history.

Back to Table of Contents

What Works Where Regulation Cannot Reach

The interesting public-health work is in the places where none of the above exists, and there are two approaches with real evidence behind them.

Post-harvest handling. Turner and colleagues ran a village-level intervention in Guinea — hand sorting, thorough drying on mats rather than bare ground, storage on wooden pallets in natural-fibre sacks, and insecticide on the storage floor — and measured aflatoxin-albumin adducts in blood before and after. The intervention villages showed a large reduction in adduct levels over the storage season compared with control villages. What makes this study important is that it used a biomarker in people, not just a measurement in grain, and that every component of the package was cheap and locally available. No new technology was required.

Biological control. Not all strains of Aspergillus flavus produce aflatoxin. Atoxigenic strains are common, and if enough of them are established in a field they crowd out the toxin-producing ones by competing for the same niche. Atehnkeng, Ojiambo, Cotty and Bandyopadhyay reported field trials in which a mixture of atoxigenic vegetative compatibility groups applied to maize substantially reduced aflatoxin in the harvested crop. Products built on this principle — a carrier grain coated with local atoxigenic strains, broadcast in the field — have since been registered in several African countries under the Aflasafe name and are used on groundnut as well as maize. It is one of the more elegant food-safety interventions in agriculture: it uses the organism against itself, and it is applied once per season by a farmer with no equipment.

Groopman, Kensler and Wild reviewed the whole spectrum of protective strategies, including chemoprevention agents that increase the body's own detoxification of aflatoxin, tested in randomised trials in high-exposure Chinese populations. Those trials shifted biomarkers; the primary prevention wins have come from the boring measures — drying, sorting, storage and hepatitis B vaccination.

Back to Table of Contents

What You Can Actually Do at Home

Six rules, in rough order of how much they matter.

1. Buy from a supply chain that tests. Established brands and retailers in countries with enforced limits test their lots because they are legally and commercially exposed if they do not. This single decision does most of the work, and nothing else you do comes close to it.
2. Be cautious with unbranded bulk and open-bin peanuts, especially loose in-shell peanuts of unknown age and origin, and especially in hot, humid climates or from informal markets. This is where the residual risk in an ordinary diet actually lives.
3. Look at what you are eating and discard the bad ones. Throw out any kernel that is shrivelled, discoloured, dusty, grey-green, visibly mouldy or broken open. Because contamination is concentrated in a handful of damaged nuts, hand sorting is far more effective than it feels — it is the same principle the electronic sorters use, just slower.
4. Trust bitterness. A peanut that tastes bitter, musty, sharp, sour or simply “off” should be spat out and the batch discarded. Rancid oil and mould are the two things that taste wrong in a peanut, and neither is worth swallowing. This applies to peanut butter too: it should taste of roasted peanuts and nothing else.
5. Store cool, dry, sealed — and buy less, more often. Mould needs moisture and warmth. Shelled raw peanuts keep about six months airtight in the refrigerator and about a year in the freezer with no meaningful quality loss; roasted peanuts keep a few weeks at room temperature. In a warm kitchen, refrigerate the peanut butter. Never decant nuts into a container that is not completely dry.
6. Vary the diet. Aflatoxin risk is about cumulative exposure, so the person eating peanuts and maize as the dietary foundation every day faces a different question from the person eating a handful of peanuts among many other foods. Rotating through almonds, walnuts, pumpkin seeds and sunflower seeds is good practice for several reasons and this is one of them.

Back to Table of Contents

What Does Not Work

Some widely repeated advice is useless, and a few pieces are worse than useless because they create false confidence.

• Cooking, roasting or boiling. Aflatoxin survives ordinary cooking temperatures. Roasting reduces it modestly; it does not solve the problem. “I'll just cook the questionable ones” is exactly the wrong instinct.
• Washing. Aflatoxin is inside the kernel, not on it.
• Cutting off the mouldy bit. Reasonable for a firm cheese, wrong for a nut. The toxin diffuses beyond the visible growth.
• Assuming organic means lower. Aflatoxin is produced by a mould, not applied by a farmer. Organic certification says nothing about it either way, and organic and conventional peanut products have both tested clean and tested high in surveys. What matters is drying, storage and sorting.
• Assuming raw is safer. If anything the reverse, marginally: raw peanuts have been through fewer processing steps and no blanching.
• Detox protocols, binders and “anti-mould” supplements. There is no consumer product that meaningfully removes aflatoxin from a person, and the supplement market around mycotoxins is an unregulated one with its own hazards — see dangerous supplements. The intervention that works is not eating the contaminated food.

One more piece of context that belongs here: aflatoxin is not a peanut-specific problem. Maize is the single largest source of human exposure worldwide. Tree nuts, dried figs, spices, dried chilies, rice and cottonseed all carry limits for the same reason, and the same mould that colonises groundnuts colonises them. The broader picture is on mould and mycotoxins. A person who avoids peanuts and eats untested maize daily has not reduced their exposure.

Back to Table of Contents

Who Should Be Most Careful

People with chronic hepatitis B or C, or any chronic liver disease. The multiplicative interaction described above makes this the group for whom sourcing genuinely matters. The advice is not to avoid peanuts — it is to buy from a tested supply chain, store them properly, discard anything questionable without hesitation, and be vaccinated against hepatitis B if not already infected. See liver disease and liver function tests.

Young children. Higher exposure per kilogram for the same portion, and the growth data above. Buy the same brands you would buy for yourself, do not use bulk-bin peanuts for a toddler's food, and pay attention to how peanut butter is stored in a warm house.

People eating peanuts as a dietary staple — several servings a day, groundnut paste as a main protein and fat source — rather than as a snack. Cumulative dose is what counts, and sourcing deserves proportionately more attention.

Anyone buying peanuts in an informal market in a hot, humid climate. This is the highest-risk ordinary consumer situation there is. Sort by hand, reject anything discoloured, and store dry.

Who should not worry unduly: a person in a country with enforced limits, eating a handful of peanuts or a couple of tablespoons of peanut butter from an ordinary shop, storing it sensibly and throwing out anything that tastes odd. That person's aflatoxin exposure is low and well characterised, and the measured cardiovascular and mortality benefits of eating nuts regularly are on the other side of the ledger — see heart health and mortality. Avoiding peanuts over aflatoxin, in that setting, is a bad trade.

Back to Table of Contents

Key Research Papers

1. Nesbitt BF, O'Kelly J, Sargeant K, Sheridan A. Aspergillus flavus and Turkey X disease: toxic metabolites of Aspergillus flavus. Nature. 1962;195(4846):1062-1063. — doi:10.1038/1951062a0
2. Hsu IC, Metcalf RA, Sun T, Welsh JA, Wang NJ, Harris CC. Mutational hot spot in the p53 gene in human hepatocellular carcinomas. Nature. 1991;350(6317):427-428. — doi:10.1038/350427a0
3. Ross RK, Yu MC, Henderson BE, Yuan JM, Qian GS, Tu JT, et al. Urinary aflatoxin biomarkers and risk of hepatocellular carcinoma. The Lancet. 1992;339(8799):943-946. — doi:10.1016/0140-6736(92)91528-G
4. Kensler TW, Roebuck BD, Wogan GN, Groopman JD. Aflatoxin: a 50-year odyssey of mechanistic and translational toxicology. Toxicological Sciences. 2011;120(Suppl 1):S28-S48. — doi:10.1093/toxsci/kfq283
5. Liu Y, Wu F. Global burden of aflatoxin-induced hepatocellular carcinoma: a risk assessment. Environmental Health Perspectives. 2010;118(6):818-824. — doi:10.1289/ehp.0901388
6. Wild CP, Gong YY. Mycotoxins and human disease: a largely ignored global health issue. Carcinogenesis. 2010;31(1):71-82. — doi:10.1093/carcin/bgp264
7. Groopman JD, Kensler TW, Wild CP. Protective interventions to prevent aflatoxin-induced carcinogenesis in developing countries. Annual Review of Public Health. 2008;29:187-203. — doi:10.1146/annurev.publhealth.29.020907.090859
8. Gong YY, Cardwell K, Hounsa A, Egal S, Turner PC, Hall AJ, et al. Dietary aflatoxin exposure and impaired growth in young children from Benin and Togo: cross sectional study. BMJ. 2002;325(7354):20-21. — doi:10.1136/bmj.325.7354.20
9. Turner PC, Sylla A, Gong YY, Diallo MS, Sutcliffe AE, Hall AJ, et al. Reduction in exposure to carcinogenic aflatoxins by postharvest intervention measures in west Africa: a community-based intervention study. The Lancet. 2005;365(9475):1950-1956. — doi:10.1016/S0140-6736(05)66661-5
10. Atehnkeng J, Ojiambo PS, Cotty PJ, Bandyopadhyay R. Field efficacy of a mixture of atoxigenic Aspergillus flavus Link:Fr vegetative compatibility groups in preventing aflatoxin contamination in maize. Biological Control. 2014;72:62-70. — doi:10.1016/j.biocontrol.2014.02.009
11. Webb AJ, Hansen AP. Histological changes of the peanut (Arachis hypogaea) gynophore and fruit surface during development. Annals of Botany. 1989;64(3):351-357. — doi:10.1093/oxfordjournals.aob.a087851

Authoritative References

1. International Agency for Research on Cancer. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans — aflatoxins are classified in Group 1 (carcinogenic to humans). monographs.iarc.who.int
2. US Food and Drug Administration. Compliance policy guides on aflatoxin action levels in human food and animal feed. fda.gov
3. World Health Organization. Aflatoxins — food safety digest and fact sheets. who.int
4. Codex Alimentarius Commission. General standard for contaminants and toxins in food and feed, including maximum levels for aflatoxins in groundnuts. Codex Alimentarius

Live PubMed Searches

1. PubMed: aflatoxin in groundnuts and prevention
2. PubMed: aflatoxin and hepatitis B interaction
3. PubMed: atoxigenic biocontrol of aflatoxin
4. PubMed: sorting and processing to reduce aflatoxin

Back to Table of Contents

Connections

• Peanuts — the main topic page, including buying and storage
• Peanuts — Benefits Deep Dive — the hub for these four articles
• Peanuts: History and Origins — including the 1960 Turkey X outbreak
• Mould and Mycotoxins — the wider picture beyond peanuts
• Toxins — the section index
• Hepatitis B — the factor that multiplies aflatoxin risk
• Hepatitis C — chronic liver injury from another cause
• Liver Disease — why hepatic clearance matters here
• Liver Function Tests — reading a liver blood panel
• Corn — the largest single source of human aflatoxin exposure worldwide
• Dangerous Supplements — why “detox” products are not the answer
• Heart Health and Mortality — the other side of the risk ledger

Back to Table of Contents

Featured Videos