Heated Starches and Cancer Risk

National Food Administration, Box 622, SE-751 26 Uppsala, tel: +46 18 17 55 00, fax: +46 18 10 58 48

Summary - Acrylamide in Heat-Processed Foods

A scientific group at the University of Stockholm, headed by Prof. Margareta Törnqvist, has found that acrylamide is formed during heating of starch-rich foods to high temperatures.

Acrylamide is a "probable human carcinogen", according to international risk assessments.

The Swedish National Food Administration has developed a new, rapid LC/MS/MS-method for the analysis of acrylamide in foods. Analysis has shown that acrylamide is present in a large number of foods, including many regarded as staple foods. The levels of acrylamide differ widely within each food group analysed.

Using information on the levels in different foods and Swedish food consumption data, it seems reasonable to conclude that a significant number, perhaps several hundred, of the annual cancer cases in Sweden can be attributed to acrylamide. The risks associated with acrylamide in foods are not new - we have probably been exposed to acrylamide in food for generations. The new, emerging knowledge may make it possible to reduce the risks that we have so far accepted without discussion. This is a very positive development.

Acrylamide in food is a global problem that requires international action. The National Food Administration has therefore informed the European Commission, other food safety agencies and international organisations about the findings. It is important to obtain much more information through international co-operation in research, in order to be able to reduce the risks associated with acrylamide in foodstuffs.

Detailed studies on the levels of acrylamide in different foods, an extended risk characterisation and knowledge of the mechanisms of formation are vital. Possible specific risk groups in the population have to be identified.

Due to lack of data, it is not possible at present to issue detailed and specific recommendations to consumers or to the food industry and trade. However, since acrylamide is most probably formed during heating of starch-rich foods to high temperatures, some preliminary general recommendations can be made to the food industry and trade and to consumers.

Toxicological aspects of acrylamide

Acrylamide and polyacrylamide are used in the industry for the production of plastics.

It has been supposed that the main exposure for acrylamide in the general population has been through drinking water and tobacco smoking. The exposure via drinking water is small and the EU has determined maximum levels of 0.1 microgram per liter water.

Recent analyses have indicated that the exposure to acrylamide is probably considerably higher (for non-smokers) from consumption of certain foods that have been heated.

Acrylamide is water soluble and is quickly absorbed in the digestive tract. The excreation via the urine is fast and half of the acrylamide is cleared from the body in a few hours.

The toxicological effects of acrylamide are well known. It causes DNA damage and at high doses neurological and reproductive effects have been observed. Glycidamide, a metabolite of acrylamide, binds to DNA and can cause genetic damage. Prolonged exposure has induced tumours in rats, but cancer in man has not been convincingly shown. The International Agency for Research on Cancer (IARC) has classified acrylamide as a "probably carcinogenic to humans" (Group 2A).

Acrylamide has been shown to induce gene mutations in cultured animal cells and also in animals treated in vivo. Thus it is assumed that exposure also to very low doses of acrylamide increases the risk for mutation and cancer.

High doses of acrylamide have been applied in the toxicological studies, which is an accepted practice. 25-50 mg per kg body weight is the lowest dose that has been shown to increase the mutation frequency in mouse. Recent studies in the laboratory of the Swedish Food Adminstration have shown that chromosome aberrations are induced in mice at 10-20 times lower doses.

Among the acrylamide metabolites glycidamide is considered the most likely candidate for causing genetic damage. Glycidamide has been found in mice and rats, and also in humans exposed to acrylamide.

Neurological damage was observed when rats were given acrylamide in their drinking water. The lowest effective dose was 2 mg/kg body weight and day, and the highest no-effect dose was 0.5 mg/kg body weight and day. Also humans exposed to high doses of acrylamide have shown neurological damage, e.g. some workers occupied in the building of the tunnel at Hallandsåsen. It is difficult to assess the highest acrylamide dose in humans that does not cause neurological effects (NOEL). The level is probably several times higher than the average acrylamide intake from food.

Decreased fertility was observed in rats exposed to 5-10 mg acrylamide/kg body weight and day.

Acrylamide - Cancer studies and comparisons of risk

Epidemiological studies in man have not shown a correlation between exposure to acrylamide and increased cancer rate. These studies have been criticised because the number of studied persons was too low considering the expected effect.

Two long-term studies in rat have shown a substantial increase of tumours in different organs when the animals were exposed to acrylamide in drinking water. Similar studies have been made in mice. The lowest effective dose was 2 mg/kg body weight and day .

In the studies with rat the increase of tumours was most evident in specific organs, e.g. mammary gland, uterus, adrenal gland, scrotal mesothelium. In mice there was an increase of lung and skin tunours. These cancer studies have been used for the assessment of the risk of cancer in humans due to acrylamide exposure.

It should be noted that the genotoxic studies have indicated that there is no threshold value for the risk of cancer induced by acrylamide, i.e. there is no dose of acrylamide so low that it does not increase the risk of cancer. In making these assessments it is assumed that man and rat have the same sensitivity for cancer induction by acrylamide.

The results of the risk assessments are somewhat different since they are based on different mathematical models. By consumption of 1 microgram acrylamide/kg body weight and day the lifetime risk for cancer has been calculated to

• 4.5 per 1000 (U.S. EPA)
• 0.7 per 1000 (WHO)
• 10 per 1000 (Granath et al. 1999, Stockholm University)

These figures for risk should be compared to the total lifetime risk for cancer in Sweden, i.e. one person of three will have cancer during his/her life. For mostly unknown reasons 45 000 Swedes get cancer every year (most cases at high ages). It is assumed that a third of all cases of cancer, are due to the diet. For comparison one can mention other assessments of lifetime cancer risks: 3 per 1000 for ionizing radiation (general background excluding radon), 0.01 per 1000 for aflatoxin (a fungus toxin found in e.g. peanuts) at the EU maximum level, and this last figure is also representative for most carcinogens found in food.

Acrylamide in foodstuffs, consumption and intake

When foodstuffs were analysed at the Swedish National Food Administration (NFA) in Uppsala and at AnalyCen AB in Lidköping it was found that some foodstuffs, which had been heated, could contain relatively high levels of the substance acrylamide. In total, more than 100 food samples have been analysed at the NFA. The food survey comprised bread, pasta, rice, fish, sausages, meat (beef and pork), biscuits, cookies, breakfast cereals and beer, etc as well as some ready-made dishes such as pizza and products based on potatoes, maize and flour.

The levels of acrylamide vary considerably between single foodstuffs within food groups, but potato crisps and French fries generally contained high levels compared to many other food groups. The average content in potato crisps is approximately 1000 mikrogram/kg and in French fries approximately 500 mikrogram/kg. Other food groups which may contain low as well as high levels of acrylamide are crisp bread, breakfast cereals, fried potato products, biscuits, cookies and snacks, e.g. popcorn.

Foodstuffs which are not fried, deep fried or oven-baked during production or preparation are not considered to contain any appreciable levels of acrylamide. No levels could be detected in any of the raw foodstuffs or foods cooked by boiling investigated so far (potato, rice, pasta, flour and bacon).

According to the NFA food survey "Riksmaten 1997-98", which is based on approximately 1200 individuals between the age of 17 to 70 who recorded their food consumption during one week, an average intake of acrylamide of approximately 25 mikrogram per day (maximum intake is approximately six times higher) is obtained, based on the food groups shown below. The remaining food groups are estimated to account for approximately 10-15 mikrogram of acrylamide; in total an average intake of 35-40 mikrogram. The percentage contribution based on an intake of 40 mikrogram akrylamid per day results in:

• potato products: 36 % (French fries 16 %, fried potatoes 10 %, crisps 10 %)
• bread: 16 %
• biscuits, cookies and wafers: 5 %
• breakfast cereals: 3 %
• remaining foodstuffs groups, basically not investigated yet: 40 %

It should be emphasized that, considering our present knowledge, the contribution from different food groups is extremely uncertain since only foodstuffs which were expected to contain acrylamide have been examined so far. There is also a considerable variation in the measured levels of acrylamide.

Young adults (17 to 34 years of age) have, according to "Riksmaten", a higher consumption of snacks (nuts, chips and popcorn) than other adults. For children under 17 years of age newer data are lacking. In the food survey "Ungdom mot år 2000" (Samuelson et al 1996), which was carried out 1993-94 among 15-year olds in Uppsala and Trollhättan, the consumption of snacks was comparable to that of young adults in Riksmaten. Children have a lower average body weight than the 70 kg generelly assumed when carrying out risk assessments. This implies that the food intake per kg body weight and the exposure to various substances could be even larger for those groups of individuals compared to adults. According to Riksmaten, 10 per cent of the adult population consumes 90 per cent of the snacks consumed in Sweden.

An alternative way of estimating the intake of acrylamide is by adduct measurement, that is to measure a reaction product of acrylamide with the protein of the blood, the hemoglobin (Törnqvist et al 1997). This reaction product seems to occur in all investigated humans at approximately the same levels and is furthermore a measurement of the continously administered dose of acrylamide. The reason is unknown in this case, but workers who were exposed to acrylamide at the tunnel accident at Hallandsåsen in Sweden had higher levels of this reaction product in their blood.

In the general population, although not in smokers (who have a level of this adduct 2-3 times the background level), the background level has been estimated to account for a daily intake corresponding to approximately 100 mikrogram per day.

Other sources than foodstuffs (estimated average intake of 35-40 mikrogram/day), e.g. cosmetics, drinking water, and a possible endogenous formation in the body of acrylamide, could, to a lower extent contribute to the background level. Estimated administered amount of acrylamide for the formation of the background level together with levels of acrylamide in foodstuffs are, however, presently extremely uncertain.

Analytical methodology and survey results for acrylamide in foods

Findings of a specific haemoglobin adduct of acrylamide in human subjects, later also in rats fed with fried feed, lead scientists at the University of Stockholm to the discovery that acrylamide can be formed in foods during cooking. At a later stage they also demonstrated high acrylamide levels in some foodstuffs that were heated in laboratory experiments.

Based of this information the SwedishNational Food Administration decided to try to develop a new analytical procedure for acrylamide detection in foods. One important purpose was to enable verification of the presence of acrylamide in food by using alternative methodology, i.e. to confirm the identity of acrylamide and to show that it was not formed as an artefact during the analytical procedure. Moreover, there was a need for a simple and fast method that could be used for extensive investigations of acrylamide in a wide range of foods.

The development work,started in January 2002 and has resulted in a new and significantly improved analytical method. The procedure is presently under validation and accreditation has been requested from the national accreditation authority SWEDAC. The method utilizes liquid chromatography coupled to two-stage mass spectrometry (LC-MS-MS). This modern state-of-the-art technology is considered highly reliable for chemical structure identification. Quantification is aided by spiking each sample prior to extraction with an internal standard consisting of an acrylamide molecule where three of the hydrogen atoms have been substituted by the isotope deuterium.

As a complement to the research on acrylamide formation at the University of Stockholm, the National Food Administration has started a broad survey of acrylamide levels in common foods. In order to speed up the investigations, a number of samples were initially sent to a private laboratory (AnalyCen AB, Lidköping, Sweden), also engaged by the University of Stockholm for acrylamide analysis. The analytical method employed was a modification of a procedure originally developed for analysis of acrylamide in water. The major disadvantages compared to the new method include a lower sample throughput and the fact that it is based on indirect detection of acrylamide through volatile derivative.

Up to this date more than a hundred food samples have been collected by the National Food Administration and analysed for acrylamide. No levels could be detected in any of the raw foodstuffs or foods cooked by boiling investigated so far (potato, rice, pasta, flour and meat). A summary of the other results is given in Table 1. Various fried and oven-cooked foods based on meats and flour respectively, corn crisps, muesli, as well as some of the breakfast cereals and crisp breads, were in the concentration interval up to 100 mg/kg. French fries and other fried, deep fried or oven-baked potato products, together with some crisp breads, biscuits, crackers and breakfast cereals, were in the interval 100-1000 mg/kg. About half of the potato crisp samples were also in this group while the levels above 1000 mg/kg were detected in the remaining half.

Individual results for all tested samples are given in Table 2 (link at the bottom of this page). It must be emphasized that the concentration figures in most cases refers only to one single randomly selected package of each specific product. The uncertainty intervals given in the table are statistical estimations of the random errors of the analytical procedure only. Moreover the choice of products within each group is far from complete.

Consequently we would like to stress that knowledge that could provide guidance in consumers' choice between different products and brands within a certain food type is still lacking.

 The primary objectives so far have been to provide concentration data for intake estimations of acrylamide from foods and to identify the most important food types. The investigation is in full progress and more data will be published when available.

We are convinced that information now presented by the University of Stockholm and the Swedish National Food Administration will initiate considerable research and investigation activities at official institutions and food industries world-wide. Further studies on the occurrence and formation of acrylamide in foods will hopefully lead to modifications of production methods and raw materials that will eliminate or limit the formation of acrylamide in foods.

Recommendations regarding acrylamide in food

The Swedish National Food Administration (NFA) and researchers from Stockholm University have found that acrylamide is formed in many types of food prepared/cooked at high temperatures. No acrylamide formation has been detected in boiled foodstuffs. The NFA has informed the European Commission and other international authorities and organisations about the findings in order to initiate international collaboration in the handling of the problem. Moreover, international initiatives to commence multidisciplinary research are urgently needed, since it is very likely that there are ways of decreasing the formation of acrylamide during the cooking process.

Present knowledge does not allow for a balanced analysis of risks and benefits of staple foods containing acrylamide. The Swedish NFA can currently only issue general advice regarding the risk management of acrylamide to the food industry and consumers.

Manufacturers

Current knowledge suggests that acrylamide is formed in carbohydrate-rich foods at high temperatures. It is probable that research into the mechanisms behind this formation may lead to changes in food processing/cooking that will decrease the formation of acrylamide. The Swedish NFA recommends the food industry to investigate whether acrylamide is formed during the production of their food products. If acrylamide is formed, the industry should determine the levels in their products and initiate research to find measures to reduce the concentrations as much as possible.

Acrylamide is probably formed in many types of food, many of which have not yet been analysed. It is therefore not possible at present to say which types of food pose the greatest problem. The amount to which the consumer is exposed is governed both by the level of acrylamide in the food and the amount of the food consumed. If eaten in large quantities, foods with low levels of acrylamide may contribute substantially to the total exposure. Foods with high levels may contribute less if the consumption is low. It is therefore not possible to give detailed advice about proper measures for consumer protection. The NFA recommends that wholesalers and retailers collaborate with the authorities and the food industry in their efforts to find means to decrease the formation of acrylamide in food.

Restaurants and consumers

Current research indicates that some simple measures may reduce the acrylamide exposure of the consumer. Cigarette smoking is a source of acrylamide; exposure can be reduced by refraining from smoking. Research on food indicates that more acrylamide is formed at higher temperatures and longer cooking times during frying, deep-frying, broiling and grilling. No acrylamide has been found in boiled foodstuffs.

More knowledge is needed before the dietary advice issued by the NFA can be changed. Currently the NFA advises the public to eat more fruit, vegetables, cereal products and bread, and to reduce consumption of fat-rich products. From a nutritional perspective, boiling is more advantageous than cooking at high temperatures. Avoid burning food during frying, deep-frying, broiling and grilling and do not eat burned food.

Last updated 27 April 2002