in Acrylamide Foods: Toxicology, Epidemiology and Research Initiatives
Sara Hale Henry, P.D., Kathleen M. Koehler, Ph.D., MPH, Center for Food Safety and Applied Nutrition, College Park, MD Daniel R. Doerge, Natl. Ctr. for Toxicological Research, Jefferson, Arkansas U.S. Food & Drug Administration
Hemoglobin Adducts in Humans
Workers with neurotoxic symptoms
Chinese and Korean workers, polyacrylamide manufacture Swedish workers, tunnel construction > 500 to 18,000 picomoles/g globin Controls with no occupational exposure Smoking 116 picomoles/g globin Non-smoking 31 picomoles/g globin Laboratory workers Polyacrylamide gels 54 picomoles/g globin
Average Acrylamide per Serving*
Food Breakfast Cereal Brewed Coffee Cookies French Fries, (RF) French Fries, (OB) Potato Chips Soft Bread Toast Acrylamide, PPB or ug per kg food 133 8 222 322 698 548 44 213 Approximate Serving Size** 2 ounces 1 cup 1 ounce 2 ounces 2 ounces 1 ounce 1 slice 1 slice Acrylamide, ug per serving 7.3 2.0 6.6 22.5 48.8 16.4 2.2 9.8
*FDA exploratory data on acrylamide in foods through Feb 2003 **Nutrition Label Serving Sizes (21 CFR 101.12, Table 2)
Formation of Acrylamide in Food
Side reaction of browning reactions in food Non-enzymatic browning
Maillard reaction Reducing sugar (glucose, fructose) + amino acids
Acrylamide formation
Reducing sugar + asparagine Temperature > 100 deg C Low moisture
FAO/WHO Consultation Health Implications of Acrylamide
June 2002 Confirmed presence of acrylamide in foods Estimated average intake, 0.3 to 0.8 ug/kg bw/d No neurotoxic effects expected at these levels Carcinogenic potency of acrylamide in animals
Similar to other carcinogens in human diet Likely overall human intake is higher No evidence on carcinogenesis in humans Major concern based on induction of cancer and heritable mutations in animals
FAO/WHO Recommendations on Acrylamide in Food
Interim advice
Food should not be cooked excessively <But must be cooked thoroughly to destroy pathogens Eat a balanced and varied diet <Plenty of fruits and vegetables <Moderate consumption of fried and fatty foods Investigate possibilities for reducing acrylamide in
foods Establish an international network to share data and ongoing work
FDA Estimate of Mean Acrylamide Intake
CFSII 1994-96, 1998; Ages 2 and Older
Food Mean acrylamide intake (ug/kg bw/d) 0.056 0.049 0.044 0.041 0.040 0.027 0.023 0.020 0.070 Percent of total intake 14 13 11 10 10 7 6 5 18 Cumulative percent
French Fries (RF) French Fries (OB) Breakfast Cereal Potato Chips Cookies Brewed Coffee Toast Soft Bread Other Foods Tested
15 28 54 52 62 70 77 82 100
Total
0.37
100
100
Other Tested Foods Contributing to FDAs Estimate of Acrylamide Intake
Tested Foods Providing the Remaining 18% of Estimated Mean Acrylamide Intake
Corn snacks Crackers Pretzels Popcorn Baked beans Breaded chicken Peanut butter Soup mix Cocoa Crisp bread, matzo Instant coffee Bagels Chocolate Tortilla Breaded fish Doughnuts Almonds Nuts & seeds Taro Soy protein Pork rinds Malted drinks
FDA Preliminary Exposure Assessment Implications
Estimated average US exposure
Similar to international estimates 1000 x below animal neurotoxic or carcinogen levels
Exposure of small children
2-3 times greater per kg body weight Expected based on body size and food intake
Workers Exposed to AA in Three U.S. & One Dutch Chemical Plants
Cohort of 8508 workers potentially exposed to AA at 3 U.S. chemical plants & one plant in Netherlands 1925-89 (Collins et al, 1989) Data on cohort updated covering 11 yrs. (Marsh et al., 1999) Most definitive study of the human carcinogenic potential of exposure to AA conducted to date.
AA Worker Study (contd.)
Authors of study found increased SMRs of rectal, esophageal, pancreatic & kidney cancers for some categories of exposure to AA, but little evidence of exposure-response relations 2.26-fold risk (95% CI 1.03 4.29) found for pancreatic cancer among workers with cumulative exposure to AA >0.30 mg/m3 yrs, but no consistent exposure-response relation
Worker AA Study (contd.)
Authors noted limitations of study to be a large proportion of short term workers, low AA exposure, incomplete smoking data Low statistical power to detect cancers of brain & CNS, thyroid gland, testis & other male genital organs Good statistical power to detect respiratory cancers Authors concluded little evidence for a causal relation between exposure to AA and mortality from any cancer site
Epidemiology Study on Dietary Acrylamide and Cancer Risk
Case-control study, January, 2003 Men and women in Stockholm, Sweden Aged 51 to 77 Found no positive association between dietary acrylamide and incidence of cancer of large bowel, bladder, kidney
Evaluations of Workers Studies
In addition to evaluations by original authors, this chemical plant study has been evaluated by Sobel et al (1986), Hogan & Scott (1990) with reply by Collins et al. (1990). Also Granath et al. (2001) with reply by Marsh et al. (2001) Various authors disagree on conclusions, statistical power and limitations of original study
Epidemiology Study on Dietary Acrylamide and Cancer Risk
Dietary acrylamide
Food frequency questionnaire Swedish acrylamide food data Various fried and baked potatoes, crisp breads, soft breads, breakfast cereals, biscuits Did not include coffee or toasting of soft breads 538 controls 591 large bowel cancer 263 bladder cancer 133 kidney cancer
Source: Mucci et al., 2003
Subjects
Study Evaluation
Authors that
Acrylamide concluded intake through dietary sources may thus be effectively detoxified within the range of human exposure.
Strengths
This was a well-conducted study with strengths that reduced the possibility of selection bias and recall bias, and adjusted for major confounding variables.
Limitations
The limitations of the study could explain the observed lack of association, even if a true association exists
Limitations of Study
May prevent detection of an association if
there is a true association
Limited statistical power to detect a small risk if one
exists Possible uniformity of acrylamide intake Omission of some food sources of acrylamide Did not adjust for possible residual confounding by additional nutrients, food components Did not study possible excess risk of acrylamide intake at other cancer sites
Authors Recommendations
Determine cooking methods that avoid
acrylamide formation during food preparation Validation of food questionnaires for acrylamide intake should be a high priority Additional epidemiological evidence is needed
Other cancer sites Neurological and other disorders
Epidemiology Study on Fried/Baked Potatoes and Cancer Risk
Pelucchi et al., Int. J. Cancer, May, 2003 A group of coordinated case-control studies Men and women in Switzerland and Italy Aged up to 79 years Hospital-based control groups Found no positive association between
Intake of fried/baked potatoes Incidence of cancer of oral cavity, throat, larynx,
large bowel, breast, ovary
Implications of Current Knowledge of Acrylamide in Food
Average human exposure through food
About 1000 times lower per kg body weight than < Animal neurotoxicity < Animal genotoxicity and carcinogenesis Lack of data on human health risk at this exposure level Partial data on acrylamide levels in food < Types of foods < Processing, preparation methods
Advice for consumers should not create one
problem by solving another
Current FDA Advice to Consumers about Acrylamide
Until more is known, FDA recommends that
consumers
Eat a balanced diet Choose a variety of foods <Low in trans fat and saturated fat <Rich in high-fiber grains, fruits, vegetables
Acrylamide Web Page Links
FDA CFSAN Acrylamide web page
http://www.cfsan.fda.gov/~lrd/pestadd.html#acrylamide
JIFSAN Acrylamide INFONET
http://www.acrylamide-food.org/
FAO/WHO Acrylamide web page
http://www.who.int/fsf/Acrylamide/Acrylamide_index.htm
NIOSH Acrylamide Workers Study
Evaluate workers exposure to AA & congeners using ambient area & personal & dermal sampling, reported exp. Data & exp. Biomarkers (urinary metab., Hb adduct levels) Assess male reproductive health (semen quality, sperm DNA integrity, hormone levels, PSA levels & reported reproductive health history)
NIOSH Study (contd)
Assess neurobehavioral parameters (sensation-tactile, postural stability, grooved pegboard, simple reaction time tests) Assess relative sensitivities of reproductive & neurological effects. Project Officer: William Moorman. Investigators: S. Reutman, S. Schrader, T. Turner, L. Mickelson, E. Hitchcock, J. Kesner
Study has undergone peer review and HSRB. Final approval in process.
Studies at Center for Disease Control & Prevention
Objectives:
Obtain population-based data on AA exposure using NHANES popu...