SEARCH

SEARCH BY CITATION

Keywords:

  • acrylamide;
  • cancer;
  • food chemistry;
  • epidemiology;
  • case-control

Abstract

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

The detection of acrylamide, classified as a probable human carcinogen, in commonly consumed foods created public health alarm. Thus far, only 2 epidemiologic studies have examined the effect of dietary acrylamide on cancer risk. Presently, we reanalyzed data from a large population-based Swedish case-control study of renal cell cancer. Food frequency data were linked with national food databases on acrylamide content, and daily acrylamide intake was estimated for participants. The risk of renal cell cancer was evaluated for intake of food items with elevated acrylamide levels and for total daily acrylamide dose. Adjusting for potential confounders, there was no evidence that food items with elevated acrylamide, including coffee (ORhighest vs. lowest quartile = 0.7; 95% CI = 0.4–1.1), crisp breads (ORhighest vs. lowest quartile = 1.0; 95% CI = 0.6–1.6) and fried potatoes (ORhighest vs. lowest quartile = 1.1; 95% CI = 0.7–1.7), were associated with a higher risk of renal cell cancer risk. Furthermore, there was no association between estimated daily acrylamide intake through diet and cancer risk (ORhighest vs. lowest quartile = 1.1; 95% CI = 0.7–1.8; p for trend = 0.8). The results of this study are in line with the 2 previous studies examining dietary acrylamide and suggest there is no association between dietary acrylamide and risk of renal cell cancer. © 2004 Wiley-Liss, Inc.

In April 2002, the Swedish National Food Administration reported the detection of unexpectedly high levels of acrylamide in commonly consumed fried and baked foods.1 Since that time, scientists around the world have detected and quantified acrylamide levels in several additional food items, including coffee.2, 3 Moreover, the mechanism of acrylamide formation in foods has been characterized and is likely to occur via a reaction between reducing sugars and amino acids at high temperatures.4

Because acrylamide is classified as a probable human carcinogen,5 the finding of acrylamide in foods generated substantial public health concern and led to claims that dietary sources of acrylamide could have global impacts on cancer incidence rates. However, the scientific evidence was based chiefly on laboratory studies, in which animals had been given doses several fold higher than those measured in dietary sources.5 Epidemiologic studies had been limited to occupational settings, with inconclusive results.4, 5

The first epidemiologic study of dietary acrylamide, undertaken in response to the Swedish findings, examined acrylamide in relation to cancer of the large bowel, bladder and kidney.6, 7 In this population-based case-control study, there was no evidence that food items with detectable levels of acrylamide or estimated daily acrylamide intake was associated with a higher risk of any of the studied cancers. Because limited statistical power to detect an association was a concern in our published study, we presently reanalyzed data from a larger Swedish case-control study of renal cell cancer.

MATERIAL AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Our data come from the Swedish component of an international collaborative population-based case-control study of renal cell cancer, described previously.8 Briefly, the study base was individuals aged 20 to 79 born in Sweden and residing in central Sweden between 1 June 1989 and 31 December 1991. Incident cases of renal cell cancer (ICD7 180.0) were identified through regional cancer registries, which are 99% complete. Controls were randomly selected from the study base through the register for total population, and frequency was matched by age and sex. Participation rates for cases and controls were 70% and 72%, respectively; 379 cases and 353 controls were interviewed. The study procedures were in line with the Declaration of Helsinki.

Data were obtained through structured questionnaires by trained interviewers using face-to-face interviews. A food frequency questionnaire asked about intake of selected food items and beverages about usual diet prior to 1987, disregarding recent changes. Information was collected on 11 food items with elevated acrylamide levels, including coffee, fried potatoes, biscuits, crisp and other breads. Acrylamide content in specific food items was ascertained through extensive food databases in Sweden1 and the United States.9 A summary measure of estimated acrylamide intake was determined for participants using median values in micrograms per kilogram of food multiplied by the daily intake of each of the food items.

We assessed cancer risk separately for food items reported to contain elevated acrylamide levels1, 2, 4 and for the total summary measure of dietary acrylamide. Multivariable unconditional logistic regression was employed to model odds ratios (ORs), an estimate of the rate ratio, and 95% confidence intervals (95% CIs) with adjustment for potential confounders: sex, age, smoking, education, body mass index and total energy.

Quartiles of daily food and acrylamide intake were created based on the distribution of the control group and modeled as categorical variables with the lowest quartile as referent group. Tests for trend were calculated using likelihood ratio tests. Since smoking contributes appreciably to acrylamide exposure, we examined possible differences in cancer risk among smokers and nonsmokers.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Estimated mean daily acrylamide exposure through dietary sources was 27.6 ± 0.6 μg among controls and 27.6 ± 0.7 μg among renal cell cancer cases. Among the controls, the major sources of acrylamide estimated in the diet were crisp bread (23.3%), coffee (21.3%) and fried potatoes (20.3%).

We found little difference in intake of specific food items with high acrylamide levels. Daily mean intake among cases and controls, respectively, was 34.3 ± 1.2 and 34.1 ± 1.4 g of crisp bread, 4.1 ± 0.1 and 4.2 ± 0.1 cups of coffee and 19.5 ± 1.1 and 19.6 ± 1.6 g of fried potatoes. Adjusting for potential confounders, there was consistently no evidence of a positive association for any of the measured food items with elevated acrylamide levels and risk of renal cell cancer (Table I). Comparing the highest and lowest quartiles, the relative risks were 1.0 (95% CI = 0.6–1.6) for crisp bread, 0.7 (95% CI = 0.4–1.1) for coffee and 1.1 (95% CI = 0.7–1.7) for fried potatoes.

Table I. Relative Risk of Renal Cell Cancer for Specific Food Items with Elevated Levels of Acrylamide, Sweden, 1989–1991
Daily intake of foods (quartiles)Number of cases/controlsAdjusted ORaAdjusted ORb (95% CI)
  • a

    Data were adjusted for matching factors age and sex.

  • b

    Data were also adjusted for smoking, education, body mass index and total energy.

  • c

    Reference group.

  • d

    Because of the distribution of data, we collapsed quartiles 3 and 4 together.

Crisp breads   
 Quartile 1, lowest102/911c1c
 Q2169/1640.90.9 (0.6–1.3)
 Q3/Q4d107/951.01.0 (0.6–1.6)
Coffee   
 Quartile 1, lowest95/761c1c
 Q268/730.70.7 (0.4–1.1)
 Q3135/1260.90.7 (0.5–1.1)
 Q480/750.90.7 (0.4–1.1)
Fried potatoes   
 Quartile 1, lowest117/1051c1c
 Q276/681.01.0 (0.7–1.5)
 Q3105/1130.80.8 (0.5–1.2)
 Q480/641.11.1 (0.7–1.7)
Other breads   
 Quartile 1, lowest69/731c1c
 Q2168/1371.31.3 (0.9–2.2)
 Q351/610.90.9 (0.5–1.5)
 Q490/791.21.2 (0.7–2.0)
Cereals   
 Quartile 1, lowest117/1211c1c
 Q252/491.11.1 (0.7–1.8)
 Q3136/1081.41.4 (1.0–2.1)
 Q473/721.01.1 (0.7–1.8)
Pancakes   
 Quartile 1, lowest101/991c1c
 Q218/181.01.0 (0.5–2.0)
 Q3228/2071.11.1 (0.8–1.5)
 Q431/261.11.1 (0.6–2.1)

Figure 1 presents quartiles of estimated dietary acrylamide daily intake and risk of renal cell cancer. Adjusting for potential confounders, there was no association between daily acrylamide intake and risk of renal cell cancer (p for trend = 0.8). The relative risk comparing the highest (> 31.9 μg per day) to the lowest quartile (< 20.1 μg per day) was 1.1 (0.7–1.8). With a similar comparison, there was no difference in the effect of acrylamide among smokers (OR = 1.0) vs. nonsmokers (OR = 1.0).

thumbnail image

Figure 1. Quartiles of estimated dietary acrylamide daily intake and risk of renal cell cancer. Odds ratios and 95% confidence intervals are presented on the log scale. Data were adjusted for age, sex, smoking, education, body mass index and total energy. Quartiles of acrylamide intake: Q1 < 20.1 μg per day (reference); Q2 = 20.1–25.9 μg per day; Q3 = 25.9–31.9 μg per day; Q4 > 31.9 μg per day.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Our results are in line with the first and substantially smaller epidemiologic study of dietary acrylamide that there is no evidence of a positive association between renal cell cancer risk on the one hand and dietary acrylamide on the other.6, 7 The only other epidemiologic study addressing this question to date was conducted by Pelucchi et al.,10 who examined fried potato consumption, an important source of dietary acrylamide, in relation to cancers of the oral/pharynx, esophagus, large bowel, breast and ovary. The authors found no evidence of an increased risk of any cancer associated with higher fried potato consumption. Indeed, they confirmed an inverse association with large bowel cancer noted in the earlier study by Mucci et al.6, 7

The validity of our data is supported by the study design, which was population-based and with data on the major renal cell cancer risk factors allowing adjustment to understand the extent of confounding. The food frequency questionnaire was distributed and collected by trained interviewers blinded to the study hypotheses.

One possible limitation of the study is statistical power. In the present study, we had 80% power to detect a relative risk of 1.23 for a one standard deviation increase in acrylamide intake. Recent criticisms of the first epidemiologic study of acrylamide6, 7 put forth that the sample size of the study was not sufficient to detect the relevant effect estimate.11 Using data extrapolated from the animal models, risk assessment models have determined an expected relative risk of cancer of 1.006–1.05 for the highest vs. lowest dose of acrylamide.11, 12 There are at least 2 notable limitations in making extrapolations from the animal data, namely, the acrylamide dosage and routes of administration in these studies. Notwithstanding these limitations, in epidemiology we lack the scientific means to document such a small effect beyond reasonable doubt.

Acrylamide formation occurs through a Malliard reaction between amino acids and reducing sugars at high temperatures.4 The amino acid asparagine appears to be a predominant participant in acrylamide formation. Variation in acrylamide between food items is related to the constituent amino acids and their concentrations. Potatoes contain high levels of asparagine, and thus fried potato products have some of the highest levels of acrylamide.4

Within food items, acrylamide formation is dependent on both the temperature at which the food is cooked and the duration of time for which cooking occurs. These factors account for the variability of acrylamide within a food item. We performed sensitivity analyses to assess what contribution within food variability could have on the study findings. We have undertaken 2 approaches, both employing a worst-case scenario. First, we assumed individuals in the highest quartile for acrylamide were more likely to consume specific food items at the higher end of the acrylamide range, whereas we assumed those in the lowest acrylamide quartile were more likely to consume a food item at the lower end. Conversely, we assumed the opposite, namely, that individuals in the lowest quartile were more likely to consume food items at the higher end. For each approach, we recalculated the estimated daily intake of acrylamide by using acrylamide contents at the higher or lower end of the range for individuals in the first and fourth quartiles of acrylamide intake. While overall estimated mean acrylamide intake differed, there remained no evidence of a positive association under either the first (ORhighest vs. lowest quartile = 1.1; 95% CI = 0.7–1.8) or second assumptions (ORhighest vs. lowest quartile = 0.8; 95% CI = 0.5–1.2) in the sensitivity analysis.

The debate on whether the amount of acrylamide taken in through diet is sufficient to increase risk of cancer continues. On the one hand, some experts have stated that there is no evidence that acrylamide in foods poses a risk, while others have taken a more conservative point in stating there is currently insufficient data on which to make a determination.13 Some have advocated employing the precautionary principle, i.e., given the suspicion of harm, there should follow preventive action.14 As additional data are revealed and epidemiologic studies examining multiple cancer sites are conducted, a clearer picture of the role of dietary acrylamide will become apparent.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Supported by a Swedish Foundation for International Cooperation in Research and Higher Education grant supporting collaboration between Karolinska Institutet and Harvard University (to L.A.M.).

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  • 1
    Swedish National Food Administration. Acrylamide in food. Uppsala: Swedish National Food Administration, 2002.
  • 2
    Food and Drug Administration, Center for Food Safety and Applied Nutrition. Exploratory data on acrylamide in foods. Washington: Food and Drug Administration, 2002.
  • 3
    World Health Organization/FAO. Acrylamide. Geneva: World Health Organization, 2003.
  • 4
    Friedman M. Chemistry, biochemistry and safety of acrylamide: a review. J Agricul Food Chem 2003; 51: 450426.
  • 5
    IARC. Monographs on the evaluation of carcinogen risk to humans. Some industrial chemicals. vol. 60. Lyon: IARC, 1994.
  • 6
    Mucci LA, Dickman PW, Steineck G, Adami HO, Augustsson K. Dietary acrylamide and cancer of the large bowel, kidney, and bladder: absence of an association in a population-based study in Sweden. Br J Cancer 2003; 88: 849.
  • 7
    Mucci LA, Dickman PW, Steineck G, Adami HO, Augustsson K. Dietary acrylamide and cancer risk: additional data on coffee. Br J Cancer 2003; 89: 7756.
  • 8
    Lindblad P, Wolk A, Bergstrom R, Adami HO. Diet and risk of renal cell cancer: a population-based case-control study. Cancer Epidemiol Biomarkers Prev 1997; 6: 21523.
  • 9
    Food and Drug Administration, Center for Food Safety and Applied Nutrition. Exploratory data on acrylamide in foods. Washington: Center for Food Safety and Applied Nutrition, 2002, 2003.
  • 10
    Pelucchi C, Franceschi S, Levi F, Trichopoulos D, Bosetti C, Negri E, La Vecchia C. Fried potatoes and human cancer. Int J Cancer 2003; 105: 55860.
  • 11
    Hagmar L, Törnqvist M. Inconclusive results from an epidemiological study on dietary acrylamide and cancer. Br J Cancer 2003; 89: 7745.
  • 12
    Dybing E, Sanner T. Risk assessment of acrylamide in foods. Toxicol Sci 2003; 75: 715.
  • 13
    Sharp D. Acrylamide in food. Lancet 2003; 361: 3612.
  • 14
    Vainio H. Acrylamide in heat-processed foods: a carcinogen looking for human cancer? Eur J Epidemiol, in press.