Timothy T. Iyaniwura obtained his doctorate in pharmacology in 1985 from Ninewells Hospital, Department of Pharmacology and Neuroscience of the University of Dundee Teaching Hospital and Medical School, Scotland. He completed postdoctoral training in toxicology at the Laboratory for General Toxicology, National Institute of Public Health and Environmental Hygiene, in Bilthoven, The Netherlands. He returned to Ninewells Hospital in 1998 on a Wellcome fellowship with an appointment as a senior research fellow in toxicology. In 2003, he emigrated to Canada, where his current address is: Apt. 1002, 3301 Uplands Drive, Ottawa, Ontario, K1V 9V8, Canada. His e-mail address is firstname.lastname@example.org. Please address comments regarding this paper to the author.
Individual and Subpopulation Variations
By Timothy T. Iyaniwura, Ph.D.
Senior Research Fellow,
This paper reviews factors that should receive attention in defining the toxicity of chemicals within a population setting, with emphasis on those factors that may confer the greatest susceptibility to toxic effects. The review is intended to aid in identifying at-risk populations and the indices of toxicity that characterize them, thereby helping to establish toxicological endpoints that will allow for individual variations in susceptibilities to chemical exposures and will be representative of the general population.
Assessments of the risk of toxic chemicals to human health require the establishment of toxicological endpoints that serve as guidelines for exposure limits for the various chemicals. Variations exist in the way individuals within the population respond to chemical exposures. Among the factors most relevant in assessing human responses to toxic chemicals are age, gender, pregnancy status, dietary/nutritional status, disease/health status, and physiologically based parameters such as toxicokinetics and toxicodynamics. These factors are discussed in this paper with emphasis on the contributions they could make to individual susceptibility to toxic chemicals and, by extension, to subpopulation responses to the effects of toxic chemicals.
As for other human health risk assessments, a toxicological risk assessment consists of four principal steps:
(1) Hazard identification--the identification from animal and human studies, in vitro studies, and structure activity relationships of adverse health effects associated with exposure to a specific chemical.
(2) Hazard characterization--the determination of the quantitative potency of any adverse effect of the chemical, usually by dose-effect relationships, mechanism of action, and variations in subpopulation responses.
(3) Exposure assessment--the measurement or prediction of the intake of the chemical, in terms of magnitude, duration, and frequency of exposure, of the general population, a subpopulation, or an individual.
(4) Risk characterization--the integration of hazard identification, hazard characterization, and exposure assessment to determine the probability of occurrence and the severity of risk to human health from an exposure to a chemical or chemicals.
The quality of the risk assessment depends, of course, on the accuracy of the toxicological data included in the assessment. The following discussion shows how that accuracy is influenced by the various susceptibility factors listed above.
Age is an important susceptibility factor, with the groups most vulnerable to toxic exposures being the very young and the very old.
Infants (newborns) have a unique susceptibility to chemical exposure due to (a) the premature development of most of their bodies’ drug detoxifying mechanisms, such as the liver and drug-metabolizing enzymes; (b) their extremely small body weight; and (c) their considerable dietary restrictions. Breast-feeding babies, for instance, have been shown to have a higher level of dioxin per body weight than adults (Lawrie, 1998). (Note: The fetus or developing embryo is considered separate from the newborn infant, as discussed below under ”Pregnancy Status.”)
Because young children have higher nutritional requirements and smaller body masses, they may be especially susceptible to chemical exposure. Their energy protein and liquid requirements are much higher than those of adults, and their food consumption is less varied. For example, their consumption of dairy products, soft drinks, and some fruits and vegetables may be as much as sixteen times higher than in adults (Lawrie, 1998). Thus the permitted dietary levels of contaminants could be grossly exceeded. For example, intakes of the sweetener saccharin have been found to be much higher in children than in the normal population (Lawrie, 1998; Gregory et al., 1995).
Body fat is low in children, leading to a concentration of lipid-soluble chemicals in smaller volumes. In addition, plasma protein binding is reduced due to lower plasma albumin concentration; thus plasma chemical levels in children may be higher than normal and provoke toxicity.
The blood brain barrier in children is also less developed, leading to increased risk of toxicity of neurotoxic and centrally acting agents. Hepatic drug metabolizing enzymes are also immature and less developed, again increasing risk of chemical toxicity. Renal efficiency--filtration, secretion, and absorption--is low at birth but increases early in life.
Finally, children are at special risk because they have little control over what they eat or are exposed to. They are less informed than the adult population and rely mainly on adults for protection and precautions. Children are clearly a highly vulnerable group to toxicities, and efforts should be made to identify chemical hazards that are more peculiar to them.
At the other end of the age spectrum, the vulnerability of the elderly to chemical toxicity may be influenced by the physical, mental, social, psychological and economic changes associated with aging. Many older people experience social isolation and cognitive impairment; thus their ability to interpret and recognize environmental cues that are associated with chemical exposure may be limited. Advanced age is itself invariably a predictor of many disabling diseases.
Sometimes the effects of age and disease are indistinct. Combination drug use is common in the elderly, and the potential for drug-chemical interaction is quite high. Kidney and liver function is impaired with age, limiting the body’s ability to detoxify chemicals. Hepatic biotransformation, hepatic blood flow and glomerular filtration rate decrease; thus overall drug and chemical metabolism is impaired (Denham and George, 1990; Powell, 1997; Woodhouse, 1997). Toxicokinetics (i.e., drug absorption, distribution and elimination) change with aging and disease, resulting in higher plasma concentrations of chemicals and increased toxicity.
Homeostatic mechanisms are also less efficient in the elderly, reducing their ability to compensate for incidences such as unsteadiness and postural hypotension. The central nervous system becomes more sensitive to hypnosedative agents, especially with cognitive impairment. Aging also causes changes in the immune response that can predispose to allergic reactions with chemicals.
Thus, like the very young, the elderly are especially vulnerable to the effects of chemical exposure and toxicity and constitute special at-risk populations.
There are great differences in the responses of men and women to chemical exposures, much of which is due to differences in body size and composition (Hattis, 1996a,b). It has been shown from toxicological data in animal studies that there are over 200 chemical agents for which important differences exist between males and females (Calabrese, 1986). Other studies have shown that women are more likely than men to develop lung and oral cancer from smoking (Muscat et al., 1996) and are more likely to be adversely affected by certain psychoactive drugs (Yonkers et al., 1992, 1993; Yonkers and Hamilton, 1995).
The difference in dietary patterns and habits between men and women also impact their susceptibilities (EPA, 1996). Silvaggio and Mattison (1994) have shown that differences in absorption, distribution, metabolism, and elimination of chemicals between men and women may have a great influence on toxicity.
With these and other differences between the sexes, it is apparent that risk assessment studies should not concentrate on male populations, as is often the case with occupational studies.
A woman’s exposure to chemicals during pregnancy can inflict harm on the growing fetus and change the maternal physiology. Agents of small molecular weight can cross the placenta by passive diffusion, the rate of diffusion depending on the concentration of the chemical and its lipid solubility. Agents that reduce placental blood flow can reduce the birth weight, as has been shown for beta-blockers (Koren et al., 1998; Rubin, 1998). Xenobiotics could accumulate in the neuroectoderm in the developing fetus. In early pregnancy, the blood brain barrier is not fully developed, and the fetus` central nervous system may be especially sensitive to developmental toxins.
A detailed data collection is required for incidences of chemical exposure during pregnancy. The distribution of agents that are water soluble may be increased during pregnancy due to the increased volumes of blood and body water. This may lead to longer retention of agents in the pregnant woman and consequently to greater toxicity because of the longer residency time (longer half-lives) of the chemicals.
There is enzyme induction in the pregnant liver due to increased hormonal levels, and this may increase the rate of elimination, but for agents that are bioactivated in vivo, their toxicity may be enhanced. However, there is increased drug excretion during pregnancy due to increased renal blood flow and glomerular filtration rate in the pregnant kidney.
Exposure to social poisons has been shown to adversely affect both the developing fetus and the pregnant mother. Chronic alcoholism can lead to spontaneous abortion, mental retardation, congenital heart disease, and impaired growth. Cigarette smoking has been associated with spontaneous abortion, premature delivery, and low birth weight. Cocaine causes congenital abnormalities, low birth weight, and slow neurological and behavioral development (Koren et al., 1998).
Obviously, incidences of chemical exposure during pregnancy should be closely monitored and should be given priority as much as possible in epidemiological surveys and risk assessments of toxic chemicals.
Exposure to a chemical present in food is estimated by combining chemical residue data that give information on the occurrence and concentration of additives and contaminants of interest in foods with consumption data that provide information on the food consumed. The nature of the hazard should be considered whether the chemical in question has an acute or a chronic toxic effect.
Dietary exposure is usually assessed against safety limits or toxicological endpoints that have been established on the basis of toxicological tests of the chemical in animals, with uncertainty factors that allow the extrapolation of data from animals to man. The emphasis should be on protecting at-risk populations. These groups may be especially susceptible to the specific toxic effects of a food chemical or may consume greater quantities of a food containing the chemical--fish polluted with heavy metal, for instance.
The most reliable exposure estimates require accurate food consumption data obtained through surveys of individual consumers, with emphasis on at-risk populations such as infants, school children, those with special dietary restrictions (for example, diabetics or vegetarians), or consumers of highly contaminated food (such as fish from polluted waters or crops from soils that have been highly treated with chemicals). It is possible to have localized pockets of contaminated food so that those living in the contaminated areas may be at higher risk.
The nutritional status of individuals within a population exposed to toxic chemicals may influence their response, particularly for the young, the elderly, and the infirm (Iyaniwura, 1990). Malnutrition may make an individual less able to cope with chemical exposure. Thus care should be taken to identify the most critical consumers and to focus exposure assessments on them. That is, those with less varied diets, infants, breast-feeding babies, or high consumers of potentially contaminated food should be identified and given prominence in surveillance studies so that toxicological endpoints or safety guidelines make sufficient allowance for them.
Certain pathological states can influence the way in which the body responds to chemicals, and these should be borne in mind. Gastrointestinal, cardiac, renal, liver, and thyroid disorders all influence the behavior of drugs in the body.
Diseases that alter bowel evacuation can influence responses to the oral exposure of chemical agents. Hypoperfusion resulting from heart failure will reduce the elimination of chemicals with consequential toxicity due to reduced liver perfusion from a reduction in hepatic blood flow. Tissue hypoperfusion could also increase lactic acidosis due to hypoxia. Chronic renal failure will reduce excretion and metabolism of chemicals and increase their toxicity. Nephrotic syndrome reduces chemical distribution due to reduced binding to albumin, and this increases the overall chemical level in the plasma and toxicity. Liver disease impairs metabolism and distribution of chemicals due to portal systemic shunting and hypoalbuminaemia. Hypothroidism increases sensitivity to certain chemicals such as opiods, and hyperthyroidism increases sensitivity to certain agents such as warfarin.
Considerable effort has centered on lung cancer, with some studies showing a decreased risk of developing lung cancer with poor metabolizers and others showing a slightly higher risk with extensive metabolizers (Rostami-Hodjegan et al., 1998); however, this observation has not been widely confirmed. Besides, it illustrates drug metabolizing enzyme polymorphism rather than a disease state. The impact of genetic factors on chemical toxicity is discussed later in this paper.
Chemical toxicity in the human population may be mediated through the immune system as a result of several types of immunotoxicological reactions. There could be immunosuppression leading to decreased resistance to infections or immunosuppression leading to the development of tumors. There also could be autoimmunity, where the chemical induces an immune reaction to “self” components and causes a hypersensitivity reaction.
Immunosuppressive properties of a chemical may be detected by studies in rodents such as rabbits, guinea pigs, rats or mice. The allergenic or autoimmune effects would require more specialized studies, as these may be more individualized and have a genetic input. Besides, animal data would be insufficient to provide an adequate interpretation of the human situation.
The important question is to establish whether it is possible to define toxicological endpoints that will adequately provide for immune response variations within the population. The method used should be sufficiently sensitive to identify those that are genetically susceptible to immunotoxicological reactions. For example, patients with atopic disease may be at increased risk of immune reactions, especially those mediated by immunoglubulin (Coombes and Gell, 1968; Wilson and Duff, 1995). Patients with connective tissue diseases may also be at increased risk of autoimmune reactions. Immunodeficient patients, such as transplant patients on immunosuppressive drug therapy or those with the human immunodeficiency virus, are also at a higher risk of immunologically mediated reaction. The challenge is to identify the at-risk populations and their locations in surveys in order to ensure that toxicological endpoints make sufficient allowance for them.
Genetic differences contribute significantly to individual variations in responses to toxic chemicals. Mendelian traits have been identified for drug metabolism, with most of the data being available in clinical medicine where individualization of drug therapy is now the goal. The enzymes responsible for the metabolism of chemicals in man show a wide interindividual variation. The thought that individuals handle chemicals differently has been with us for more than 40 years, and medical science has known for some time that individual responses could be inherited (Kalow, 1989).
Genetic polymorphisms of cytochrome P450 ((CYP) 2D6), the hepatic microsomal drug metabolizing enzymes, have received a lot of attention. Particular attention has been given to those related to the carbon oxidation of chemicals, which include cytochrome P450 (CYP) 2D6, acetylation by N-acetyltransferase, S-methylation by thiopurine, methyl transferase, and ester hydrolysis by pseudocholinesterase.
A good illustration of the genetic polymorphism of a drug (or chemical) metabolizing enzyme is CYP 2D6 (Lennard, 1993; Tucker, 1994). Individuals who are homozygous for an autosomal recessive trait of the enzyme gene are poor metabolizers (about 7-10 % of Caucasians). The rest of the population are extensive metabolizers with wide interindividual variation up to a subgroup of ultra rapid metabolizers (Lennard, 1993; Tucker, 1994). For instance, for a particular dose of the beta-blocking drug metoprolol, poor metabolizers develop plasma concentrations that are six times higher than those of extensive metabolizers and this produces a more pharmacological effect in the poor metabolizers (Lennard, 1993). The two phenotypes suffer differential toxicity from drugs and chemicals. A recent evaluation of the literature (Rostami-Hodjegan et al., 1998) showed that poor metabolizers have a small but significantly reduced risk of developing lung cancer.
Genetically variable metabolism is a major source of interindividual differences in response to chemicals, and genetic polymorphisms of drug metabolizing enzymes can have important consequences for the toxicity of chemicals. With progress in clinical medicine and human genomics, it will be possible to identify individuals and sub-populations with genetic traits that may confer increased risk of toxicity from exposure to chemicals. More work is clearly required in this area. With improved technology of genotyping and micro array, the classification of toxic chemicals and the identification of those with genetic traits that confer increased susceptibilities to the toxic effects of the chemicals would not be an impossible task.
Toxicokinetic and Toxicodynamic Considerations
The importance of differences in the toxicokinetics of individuals--i.e., differences in their absorption distribution and excretion of toxic chemicals--has been discussed in some of the preceding paragraphs. There are also differences in the toxicodynamics of individuals--i.e., differences in the way their organs and tissues respond when a toxic chemical reaches critical target sites within the person. Such differences are likely to be determined in part by genetic factors leading to heterogeneity in both toxicokinetics and toxicodynamics. The differences are also likely to be affected by other factors like age, sex, immunity, nutrition, disease, and pregnancy.
The data that are available to address these factors are limited, which introduces a lot of uncertainties to the risk assessment process. Most mechanistic data are obtained from animal studies. For some of the studies, especially in vivo, this is the only way it can be done, as ethical considerations restrict toxicological studies in man to retrospective and prospective epidemiological surveys. Although in vitro studies are becoming increasingly feasible with cultures of human cells and genomics, toxicological assessments will continue to rely on animal data for the foreseeable future, especially for mechanistic studies. However, physiologically based pharmacokinetic modeling and the increasing possibility of utilizing the large data base on human medicine to predict the toxicity and behavior of chemicals could make the task of toxicological risk assessment somewhat easier. Selective and individualized responses that have been identified in clinical medicine would make the task of classifying subpopulation responses and individual susceptibilities less formidable. More so since the body handles these chemicals in almost identical manners, as the therapeutic agents and the metabolic pathways are essentially the same. The mammalian toxicologist at least would remain closely in liaison with his/her clinical counterpart in clinical pharmacology, since for him at least toxicology would not cease to be applied pharmacology.
The world is filled with toxic chemicals, and the reduction and/or prevention of chemical toxicities for various subpopulations is dependent on how well the risks involved can be determined. Critical to such determinations is the establishment of toxicological endpoints that will allow for individual variations in susceptibilities to exposures and can be extended to subpopulations. Once these endpoints have been established, risks can be accurately characterized through well-known risk assessment processes. Risk communication and risk management techniques can then be employed to inform and protect at-risk populations.
Calabrese, E. J., “Sex Differences in Susceptibility to Toxic Industrial Chemicals,” Brit J. Ind. Med. 43, 577-579 (1986).
Coombes, R. R. A., and Gell, P. G. A., “Classification of Allergic Reactions Responsible for Clinical Hypersensitivity and Disease,” pp 575-596 in Clinical Aspects of Immunology, P. G. H. Gell, editor, Oxford University Press, Oxford, UK (1968).
Denham, M. J., and George, C. F., “Drugs in Old Age: New Perspectives,” Brit. Med. Bull. 46, 1-299 (1990).
Exposure Factors Handbook, Vols. 1-3, Washington, D.C., U.S. Environmental Protection Agency (1996).
Gregory, J., Collins, D., Davies, P., Hughes, J., and Clarke, P., National Diet and Nutrition Survey: Children Aged 1-1/2 and 4-1/2 years, London, UK, HMSO (1995).
Hattis, D., “Variability in Susceptibility--How Big, How Often, for What Responses, to What Agents?” Environ. Toxicol. Pharmacol. 2,135-145 (1996).
Hattis, D., “Human Interindividual Variability in Susceptibility to Toxic Effects: From Annoying Detail to a Central Determinant of Risk.,” Toxicol. 111, 5-14 (1996).
Iyaniwura, T. T., “Dietary Factors in Mammalian Toxicity of Chemicals,” Vet. Hum. Toxicol. 32(2), 106-110 (1990).
Kalow, W., “Race and Therapeutic Drug Response,” New Eng. J. Med. 320, 588-590 (1989).
Koren, G., and Pastuszak, Ito S., “Drugs in Pregnancy,” New Eng. J Med. 338,1128-1136 (1998).
Lawrie, C. A., “Different Dietary Patterns in Relation to Age and the Consequences for Intake of Food Chemicals,” Food Addit. Contam. 15(Suppl),75-81 (1998).
Lennard, M. S., “Genetically Determined Adverse Drug Reactions,” Drug Safety 9, 60-77 (1993).
Muscat, J. E., Richie, J. P., Thompson, S., and Wynder, E. L., “Gender Differences in Smoking and Risk for Oral Cancer, “ Cancer Res. 56, 5192-5197 (1996).
Powell, C., “Frailty: Help or Hindrance,” J. Roy. Soc. Med. Suppl. 90, 23-36 (1997).
Rostami-Hodjegan, A., Lennard, M.S., Woods, H. F., and Tucker, G. T., “Meta Analysis of Studies of the CYP2D6 Polymorphism in Relation to Lung Cancer and Parkinson`s Disease,” Pharmacogenetic 8, 227-238 (1998).
Rubin, P. C., “Drug Treatment During Pregnancy,” Brit. Med. J. 317, 1503-1506 (1998).
Silvaggio, T., and Mattison, D.R., “Setting Occupational Health Standards: Toxicokinetic Differences Among and Between Men and Women,” J. Occup. Med. 36, 849-854 (1994).
Tucker, G. T.,”Clinical Implications of Genetic Polymorphism in Drug Metabolism,” J. Pharm. Pharmacol. 46, 417-424 (1994).
Wilson, A. G., and Duff, G. W., “Genetics Traits in Common Disease,” Brit. Med. J. 310, 1482-1483 (1995).
Woodhouse, K. W., “Age and Self Poisoning: The Epidemiology in Newcastle–upon- Tyne in the 1980s,” Human Toxicol. 6, 511-515 (1987).
Yonkers, K. A., Kando J. C., Cole, J. O., and Blumenthal, S., “Gender Differences in Pharmacokinetics and Pharmacodynamics of Psychotropic Medication,” Am. J. Psychiat. 149, 587-595 (1992).
Yonkers, K. A., Kando, J. C., Cole, J.O., and Blumenthal, S., “Gender Differences in Pharmacokinetics and Pharmacodynamics of Psychotropic Medication--Reply,” Am. J. Psychiat. 150, 679 (1993).
Yonkers, K. A., and Hamilton, J.A., “ Psychotropic Medicine,” in American Psychiatric Press Review of Psychiatry, Vol. 14, pp. 307-332 (J. Oldham and A. Riba, Editors), Washington D.C., USA, American Psychiatric Press (1995).
*Copyright is retained by the author.
Published July 1, 2004.