It is now more than eight years since Swedish researchers announced their discovery that a potentially carcinogenic contaminant, acrylamide, could be produced in significant amounts during high temperature processing of starchy foods like bread, French fries and crisps. The finding was greeted with alarm by food safety experts worldwide and led to an urgent flurry of research activity as scientists attempted to determine the extent of any threat to human health, how the contaminant was formed, and how it could be controlled. A lot of progress has been made since then, but there is still a lot we don’t know about acrylamide and recently published data shows that European consumers are still exposed to the chemical in their diets. Some experts believe that current levels of acrylamide in food present no threat, but others are not so sure.
In April 2002, researchers at the Swedish National Food Administration published a report that took the majority of food safety experts completely by surprise. The report revealed that some carbohydrate-rich foods processed at temperatures above 120oC, such as baked and fried cereal and potato products, contained appreciable levels of acrylamide, a synthetic vinyl compound usually associated with the manufacture of polyacrylamide gels for the water treatment, mining, textile and paper industries. The discovery was made only after an earlier study into adverse health effects suffered by construction workers who used polyacrylamide gels revealed unexplained exposure to acrylamide amongst people in the control group. One hypothesis to explain how this exposure occurred was that acrylamide might be present in the diet, and so it proved on closer investigation.
Since acrylamide can cause cancer in animals, its presence in foods caused immediate concern and a prompted a rush to confirm the Swedish results, to find out how the contaminant got into food and, most importantly, to establish the significance of the findings for public health.
Researchers investigate the extent of the problem
It quickly became apparent that the Swedish researchers had stumbled upon something significant and widespread, as their results were confirmed by new studies conducted all over the world. Acrylamide seemed to be produced as a natural by-product of high temperature processing of starchy foods and varying levels of the contaminant were detected in a wide range of products. The highest levels were generally found in potato crisps, French fries and potato-based snacks, but acrylamide was also present in breakfast cereals, bread and bakery products, roast and ground coffee, biscuits and crackers, chocolate and dried fruits. It could also be found in home-cooked foods as well as processed products, but meat and foods cooked at lower temperatures tended not to contain significant levels. It was quickly realised that this potentially toxic chemical had previously gone unnoticed as a contaminant in food, probably since human beings first began to cook.
It was clear that acrylamide was somehow being produced during the processing of the contaminated foods, but the exact mechanism by which this was happening was a mystery at first. Research continued at a rapid pace and later in 2002 scientists in the UK, Switzerland, Canada and the USA all independently established a link with the Maillard browning reactions that occur in foods cooked at higher temperatures. Maillard reactions occur between amino acids and reducing sugars and are important in the development of the flavour and appearance of many cooked foods. The researchers noted a similarity between the chemical structures of acrylamide and those of the amino acid asparagine, and so began to investigate the hypothesis that acrylamide might be produced from asparagine. Sure enough, it transpired that acrylamide could indeed be produced as a by-product of the reaction between free asparagine and reducing sugars, such as fructose and glucose, during Maillard browning. Other mechanisms that may contribute to acrylamide contamination in food have since been suggested, but the consensus remains that the reaction between asparagine and reducing sugars is the main source of the contaminant in foods.
The next question requiring an urgent answer was whether acrylamide in food had health implications for consumers. The chemical is a neurotoxin at large concentrations and can cause numbness in hands and feet of industrial workers exposed to high levels. It has also been shown to be genotoxic in animal studies. But the amounts present in food were considered very unlikely to be sufficient to cause either neurological damage or reproductive toxicity. More worrying was the possibility that lower levels could be carcinogenic. This had been demonstrated in animal studies and the International Agency on Research in Cancer (IARC) classified acrylamide as “probably carcinogenic to humans (IARC Group 2A).”
This concern prompted the Joint FAO/WHO Expert Committee on Food Additives (JECFA) to review all the available toxicity and dietary intake data for acrylamide and conduct a risk assessment for the effect on human health. They did this in 2005 and concluded that, despite considerable uncertainty in estimating the risk, exposure to acrylamide in the diet might indeed be a concern, particularly in view of the relatively low safety margin – in comparison with other carcinogens – between possible human exposure and the level known to increase the risk of cancer in rats.
The problem facing JECFA, the food industry and worldwide health authorities, was that acrylamide is an inadvertent contaminant present in a wide range of staple foods. Legislating to control it was not a practical option at that stage and neither was advising consumers to avoid such basic foods as bread and fried potatoes. All that could be done was to recommend that the food industry work towards lowering acrylamide levels in processed foods and that consumers be advised on how they might limit its production in foods they cooked themselves. In other words acrylamide in certain foods may be undesirable, but it is also inevitable.
Concerns about possible links between acrylamide in the diet and carcinogenicity also gave rise to a number of longer term research projects looking for evidence of associations between diet and a range of common cancers. The results of some of these studies have been published and their conclusions are mixed so far. The first real alarm bells were sounded by a Dutch study carried out by researchers at Maastricht University and published in late 2007. Using the data from a long-term cohort study on diet and cancer, which asked 120,000 people between 55-70 years of age for details about their diet, the Maastricht team were able to identify a link between higher levels of acrylamide intake (approximately 40 micrograms per day) in women and a two-fold increase in the risk of developing womb and ovarian cancer. Although they found no increased risk for breast cancer, a Danish study published early in 2008 and using more accurate methods to estimate acrylamide exposure, suggested that there was a link between acrylamide in food and a certain type of breast cancer. Interpretation of these studies is difficult and other factors such as smoking need to be taken into account – tobacco smoke contains acrylamide – but these findings at the very least point to a need for further research.
On the other hand, earlier large epidemiological studies in Sweden and the USA showed no link between dietary acrylamide intake and several common cancers, including colorectal, breast and renal cancers. More recently, the Maastricht team has been using the data from the Dutch cohort diet and cancer study to look for links between acrylamide and other cancers. Their most recent findings were published in 2009 and revealed no detectable link with brain cancer, or with lung cancer. There was even a suggestion that acrylamide might reduce the risk of lung cancer in women, but this finding has yet to be confirmed.
Much of the funding for acrylamide research in Europe has come from a large EU-funded ‘strategic project’ called HEATOX, which ran from November 2003 to February 2007. This 24-partner research project looked not only at acrylamide, but also at other potential toxins generated by heat processing. HEATOX covered all aspects of the subject including toxicity and health effects, exposure assessment, risk assessment, methods of analysis, mitigation techniques and intervention strategies for acrylamide formation in food and the priorities for further research. The project concluded that, “Increasing toxicological evidence suggests that acrylamide in food might be a cancer risk factor.”
So many research projects looking at aspects of acrylamide generation in food have been initiated worldwide that the WHO has set up the ‘Acrylamide Infonet’ to help researchers share information and provide updates on ongoing studies. The European Commission has also set up the ‘Acrylamide Information Base’ to fulfil a similar role.
Prevention is better than cure
As soon as the mechanism for acrylamide production during food processing had been explained, researchers and the food industry began to look at means of combating the problem and reducing levels of the contaminant in susceptible foods. Various approaches were suggested, but most could be included in one of two broad categories.
- Changes in product formulation – an obvious means of influencing acrylamide formation in processed foods is to minimise the amounts of asparagine and reducing sugars present before cooking. A number of methods for achieving this have been tested successfully, including using low sugar and low asparagine varieties of potatoes and wheat, controlling potato storage conditions to minimise sugar production, avoiding fructose use in baked products and replacing ammonium carbonate with other raising agents. The use of asparaginases (enzymes that convert asparagine to aspartate) is a novel approach the problem. Inclusion of commercial asparaginases, such as PreventASe(tm) from DSM Food Specialities, in formulations of bread and baked products has been used successfully to give reductions of up to 90%, although it would not be suitable for products with no mixing stage.
- Modification of processes – The ‘thermal input’ of the process is directly linked to the amount of acrylamide produced and reducing the cooking time and/or temperature can be beneficial, although this will inevitably affect the sensory characteristics of the product. Lighter colours – golden, not brown – for baked and fried products will generally equate to lower levels of acrylamide. This approach is also applicable to home cooked foods, where consumers can be educated to avoid overcooking fried and baked foods.
These techniques have been brought together by the Confederation of the Food and Drink Industries of the EU (CIAA) as an ‘Acrylamide Toolbox’ of practical techniques for limiting acrylamide generation, designed specifically for processors. The Toolbox has recently been updated and now includes information from the US food industry compiled by the Grocery Manufacturers Association. The updated version of the Toolbox is freely available to download from the CIAA web site (see the URL below). The European Commission, in collaboration with the CIAA, has also published a series of pamphlets aimed at specific food manufacturing sectors, which provide simple practical tips for reducing acrylamide. So far, individual pamphlets are available for biscuits and crackers, fried potato crisp products, French fries, bread and breakfast cereals and they too can be downloaded from the Internet at the relevant URL listed below.
Despite the impressive range of techniques now available for controlling acrylamide formation, the HEATOX project estimated that, in the case of the Dutch population, a theoretical reduction in dietary acrylamide intake of 40% was the best that could be achieved using known methods. Clearly there is still some way to go before the problem can be completely eliminated.
How are we doing so far?
A substantial amount of testing activity since 2002 has generated a huge quantity of data on the levels of acrylamide in foods from all over the world. Much of this data has been collected in the WHO/FAO Acrylamide Infonet Analytical Database and is available on-line. This data provides a valuable baseline against which progress in reducing acrylamide levels can be measured. The European food industry has done a great deal in terms of research and development aimed at cutting acrylamide production during processing, but how effective has that effort been so far?
That question can be answered in part by the European Food Safety Authority (EFSA), which has published Results on the monitoring of acrylamide levels in food, a series of scientific reports presenting data from foods sampled in 2007 and 2008, the first two years of a three-year monitoring programme requested by the European Commission. Approximately 2000 foods were sampled in ten categories: French fries; potato crisps; potato products for home cooking; bread; breakfast cereals; biscuits; roasted coffee; jarred baby foods; processed cereal-based baby foods; and other products. Mean levels of acrylamide in the samples varied from 23 µg/kg for ‘bread not specified’ and 44 µg/kg for jarred baby foods up to 628 µg/kg for potato crisps, and 1124 µg/kg for ‘substitute coffee, but some individual samples contained much higher levels, with the maximum value for both biscuits and potato crisps being around 4,200 µg/kg.
Comparing the data from 2008 with that collected in 2007 and in earlier surveys provided some welcome news, with an apparent trend towards lower acrylamide values over time. But this trend did not apply across the board and both the ‘potato crisps’ and ‘instant coffee’ categories actually showed statistically higher levels of acrylamide in 2008 than in 2007, whereas in several other categories the reverse was the case. Nevertheless, an estimated 30% drop in overall dietary exposure to acrylamide has been achieved since 2003, based on detailed consumption data from two EU countries. That is not far away from the 40% ‘best achievable’ reduction predicted by the HEATOX project and represents significant progress. Unfortunately, this positive outlook is somewhat overshadowed by the rising acrylamide levels in some categories. When one takes into account the fact that an apparent drop in levels in roasted coffee might be the result of earlier overestimates – there are no practical measures to reduce levels in coffee as yet – the overall picture looks a little less rosy. In fact the most recent EFSA report, while acknowledging a downward trend in acrylamide exposure, says, “…it may be appropriate to assume that the application of the acrylamide toolbox was effective only in a limited number of food groups.”
These findings seem to suggest that certain sectors of the industry have made good progress in reducing acrylamide in their products, while others have had less success. This may be because effective strategies have yet to be developed for some foods, but it seems more likely that the poor rate of progress in some sectors is the result of inertia and a failure to take action. The EFSA reports might act as a wake up call for manufacturers who have yet to tackle acrylamide. If not, it would be no great surprise if the European Commission were to propose legislation to force processors to act.
Finally it is worth noting that the HEATOX project did not confine itself solely to acrylamide and concluded that other chemicals produced by similar mechanisms might also be risk factors. In fact the project produced a database of over 800 heat-induced compounds in food and estimated that as many as 50 of these might also be potential carcinogens and would need to be assessed. As an illustration used in the project report itself rather ominously implies, acrylamide may just be the tip of the iceberg.
Results on the monitoring of acrylamide levels in food (EFSA report for 2008)
This article is an updated version of one first published in the journal Food Engineering and Ingredients