Shows NCAC Main Header

Toxicology Of Chemical Mixtures

National Capital Area Chapter Society of Toxicology

Spring Symposium

May 24, 2005

Lister Hill Auditorium

National Library of Medicine

Bethesda, MD


Chemical Mixtures Toxicology and Risk Assessment: Guidance and Methods

John C. Lipscomb

U.S. Environmental Protection Agency, Cincinnati, OH

Humans are exposed to at least scores of environmental contaminants daily. This may occur in the form of mixtures of chemicals, where multiple chemicals occur in a given environmental medium, or as a cumulative exposure, where multiple chemicals are encountered from multiple environmental media via multiple exposure routes. Once inside the body, chemicals can interact so that tissue disposition (toxicokinetics; TK) is altered, and/or so that the response at the organ or cellular level (toxicodynamics; TD) is altered. When effects are measured and subsequently expressed at the level of the encountered concentration/dose, separation of TK and TD is not performed. Whether benefit may be gained from such a more intensive investigation should be assessed on a case-by-case basis. Regardless, human chemical exposure is complex, and risk assessors may rely on available guidance and methods to assess the risks of chemical mixtures.  Risk assessments for chemical mixtures should address the spectrum of insults possible; site or organ concordance is not a part of the risk assessment approach for the U.S. EPA (U.S. EPA does not develop organ-specific reference levels for humans). Thus, effects at the level of the whole-organism are assessed. This presents some special challenges.  In addition to identifying the affected tissues, organs or systems and the dose-response associated with them, advanced information on mode of action or mechanism of action is critical. For component-based risk assessment methods, information on mode or mechanism of action will determine how chemicals in the mixture should be grouped and which mixtures additivity method to apply. Chemicals with the same or similar mode of action are grouped into a common mechanism group (CMG), and the toxicity of these chemicals is expressed as a dose function, e.g., based on a selected index chemical from the group. For chemicals in the same CMG, risk is assessed via dose addition; relative potency factors represent one commonly used form of dose addition. Risks from chemicals with different modes or mechanisms of action are often assessed through response addition. Though it is recognized that toxic interactions of chemicals (where mixtures of chemicals produce a toxicity not adequately described by dose-addition) in a mixture may represent potentiation, synergistic, or antagonism, the default position for U.S. EPA methods is to assume an interaction-based hazard index approach. When the available toxicity or epidemiologic information was derived from whole-mixture exposures, that information implicitly reflects such interactions. When the risk assessment activity addresses a complex mixture of chemicals, information from other mixtures deemed sufficiently similar (i.e., mixtures of drinking water disinfection byproducts) may be used to develop risk information (i.e., cancer slope factors, reference levels). This lecture will present the fundamentals of chemical risk assessment, and briefly summarize guidance available from the U.S. EPA.

Slide presentation available (Adobe Acrobat)

Toxicology of Complex Mixtures of Disinfection Byproducts

Jane Ellen Simmons

U.S. Environmental Protection Agency, Research Triangle Park, NC

Chemical disinfection of water is a major public health advance that has decreased dramatically water-borne disease.   Chemical disinfectants react with naturally occurring organic and inorganic matter in water to produce a wide variety of disinfection byproducts (DBPs). DBP number, chemical type and concentration are dependent on source-water and treatment-scenario characteristics.  Although more than 500 DBPs have been identified, ~50% of the total organic halide (TOX) mass formed during chlorination remains unidentified.  Some epidemiological investigations have suggested associations, albeit weak, between human consumption of chlorinated drinking water and adverse health outcomes such as developmental and reproductive effects, and bladder, colon and rectal cancer. The health effects observed in some epidemiological studies are unexpected based on the available data from experimental-animal single-chemical DBP studies.  Understanding the human health risk(s) associated with consumption and use of chemically disinfected water will require relevant toxicological information on individual DBPs, defined DBP mixtures of known composition and complex, environmentally realistic mixtures of DBPs. Individual DBP assessments are essential but do not account for potential interactions that influence toxicity.  Component-based assessment of simple, defined mixtures are needed as four trihalomethanes (THMs, chloroform, bromoform, bromodichloromethane and chlorodibromomethane) and five haloacetic acids (HAAs, monochloro-, dichloro-, trichloro-, monobromo-, and dibromoacetic acid) are regulated together, respectively, under a total THM and a total HAA standard. Defined mixture data provide important information, but are not by themselves sufficient because a significant portion of the DBP mixture mass remains unidentified.  Methods are needed to determine the portion of any observed toxicity attributable to the unidentified fraction of the mixture. This talk will summarize recent data on individual DBPs and both defined and complex mixture of DBPs.  (This abstract does not represent U.S. EPA policy.)

Slide presentation not available

Tools and Data Needed to Assess Multiple Chemical Exposures

Chris T. De Rosa

U.S. Agency for Toxic Substances and Disease Registry, Atlanta, GA

Literally thousands of chemicals, mostly as mixtures, are found in the environment.  Several attempts have been made by various federal agencies to prioritize mixtures so as to accomplish their missions and to meet the needs of their specific mandated programs.  These mixtures could be simple or complex in their content or composition. The toxicity and risk assessments of chemicals or their mixtures should represent all the available integrated scientific evidence on their plausible toxicities. According to convention such assessments are performed using the so-called “NAS paradigm” consisting of four steps: hazard identification, dose-response, exposure assessment, and risk characterization.   Mostly, single chemical assessments are performed, and only when data lend themselves toxicity or risk assessments are performed for mixture.  This is a fundamental deficiency of all the assessments that are performed.  Dependent upon the availability of data, three broad approaches are often available for the toxicity evaluation of mixtures.  The potency weighted dose or response addition approach is most often used because it utilizes the available single chemical data.  This approach until recently did not allow the integration of interaction data in the overall toxicity assessment of the mixture.  In the 1990’s a weight of evidence method was developed which allows a qualitative, and if data are available, a quantitative factor to include an interaction factor.  With advancements in computational techniques and computer capabilities advances are being made to move from the basic default methodologies to more sophisticated tools that will help advances for the assessment methods for mixtures.  Research data needs have to be identified and filled as we move towards development and identification of these advanced tools that must be supported by credible science.

Slide presentation available (Adobe Acrobat)

Occupational Exposure to Chemical Mixtures

Frank J. Hearl

National Institute for Occupational Safety and Health, Washington, DC

Workers are exposed to multiple agents, either as intrinsically complex mixtures or as separate simultaneous exposures to a variety of substances or stressors. Some intrinsically complex mixtures routinely encountered in occupational settings are diesel exhausts, welding fumes, coke oven emissions, and metal working fluids. Other workplace combinations that result in biological interactions are less obvious, such as the combined action of certain organic solvents and noise exposure, which results in hearing loss to an extent greater than would be predicted by either exposure alone. Although the regulatory agencies and consensus standard setting bodies have recognized the existence of combined effects from mixed chemical exposures, and have proposed dose-additivity formulas for adjusting an occupational exposure limit (OEL), in practice most exposures are regulated or controlled as if they occurred independent of any other substance exposures. Little information or guidance is available to assist practicing industrial hygienists for the application of a modified OEL to account for mixed exposures. Research is needed to provide a sound scientific basis to describe interactions, and to assist practitioners in applying appropriate algorithms for controlling exposures where antagonistic, additive, or synergistic effects may be predicted and expected.

Slide presentation available (Adobe Acrobat)

The Use of Interaction Data in the Joint Toxicity of Chemical Mixtures

Moiz M. Mumtaz and Chris T. De Rosa

U.S. Agency for Toxic Substances and Disease Registry, Atlanta, GA

Health risk assessment is the practice for evaluating the degree of danger associated with environmental exposures to chemicals and other stressors.  The recent national report on human exposure to environmental chemicals, through the analyses of blood and urine samples, indicates that over 100 chemicals are found in the U.S. human population.  Consequently, risk assessments are performed for chemical mixtures most often using the hazard index approach.  The presence of multiple chemicals within a biological system increases the potential for interactions that could enhance or diminish the toxicity of other chemical(s).   However, interpretation of the interaction data poses a challenge due to (1) data limitation on chronic exposure to mixtures, (2) lack of higher order mixtures data, (3) lack of statistical power, and (4) equivocal evidence from epidemiological studies. The dilemma of lack of information versus perception of high risk from exposures to mixtures by the exposed population poses an enormous challenge for the risk assessment community.  In order to address this challenge, joint toxicity assessment methods have evolved from the initial default generic approaches to the most sophisticated physiologically based pharmacokinetic modeling and in silico methods that try to define threshold for interactions.  Results from limited but well designed experimental studies will be presented that indicate that environmental level short term exposures to mixtures can be adequately characterized using additivity approaches.

Slide presentation available (Adobe Acrobat)

Toxicology of Polycyclic Aromatic Hydrocarbon (PAH) Mixtures

Lynn Flowers

U.S. Environmental Protection Agency, Washington DC

The Integrated Risk Information System (IRIS) Program is undertaking a health assessment for PAH mixtures. The IRIS database contains entries for 15 individual PAHs, but these assessments do not consider the environmental occurrence of PAHs as complex mixtures. The PAH mixtures assessment considers three approaches that have been defined for conducting the assessment of health risks of chemical mixtures: the comparative potency, surrogate and relative potency approaches as outlined in the Guidance for the Health Risk Assessment of Chemical Mixtures (U.S. EPA, 1986, 2000). These approaches utilize data pertaining to the mixture of concern, toxicologically similar mixtures, or the mixture’s component chemicals, respectively. The comparative potency approach assumes that toxicological modes of action are the same for similar mixtures and that the potency of both mixtures in in vivo or in vitro bioassays is directly proportional to the potency in humans. The surrogate approach estimates the potency of the PAH fraction of a mixture of concern, based on the assumption that the cancer risk of this fraction is proportional to the level of an indicator PAH in the mixture. An assumption must be made that the composition of the PAH mixture of concern is sufficiently similar to a surrogate PAH mixture. The relative potency factor approach provides a cancer risk estimate for the whole mixture by summing the carcinogenic potential of individual PAHs relative to an index compound (e.g., benzo[a]pyrene). This approach is outlined in the Provisional Guidance for Quantitative Risk Assessment of PAHs (U.S. EPA, 1993) and is extensively utilized for the estimation of risk from exposure to PAH mixtures. The provisional guidance, however, does not reflect the most recent research findings on PAHs and PAH mixtures, nor does it consider some PAHs with carcinogenic potential (e.g., fjord-region PAHs). The PAH mixtures health assessment will encompass all the approaches with a particular emphasis on the relative potency factor approach. (The views expressed are those of the author and do not necessarily reflect the views or policies of the U.S. EPA).

Slide presentation available (Adobe Acrobat)

Understanding Mechanisms of, and Mechanistic Models for, Chemical Mixtures

Harvey Clewell

Chemical Industry Institute of Toxicology, Research Triangle Park, NC

Because humans are exposed to a wide variety of environmental compounds, it is necessary to consider the potential for interactions that could result in a cumulative toxicity different from the default assumption of additivity.  Physiologically based pharmacokinetic (PBPK) models that incorporate information on the mechanism of these interactions can provide useful insights on the health effects of mixed exposures such as gasoline vapors, contaminated food or drinking water.  Once validated on the basis of data from animal exposures, a mechanistic PBPK mixture model can be used to provide quantitative predictions of the interactions expected in humans at environmental exposure levels.  Examples will be provided for both simple (pesticide) and complex (gasoline) mixtures, illustrating mechanistic modeling of one important kind of interaction: competitive inhibition by substrates for the same enzyme system.  Interactions of this kind are probably among the most common in environmental mixtures.  Other possible interactions that can be investigated using the PBPK approach include depletion of metabolic cofactors, induction of enzyme synthesis, and destruction of enzyme.  For all of these cases, the interaction can lead to either suppression or potentiation of a particular toxic effect, depending on the relationship of the toxic event to the affected metabolic step.  In general, interactions are more likely to be seen at the high exposures typical of animal experiments than at the much lower concentrations of concern for human exposures, with the exception of drug-drug interactions and the effect of alcohol consumption on chemical toxicity in humans.  As mechanistic modeling of representative interactions accumulates, it will become increasingly possible to draw reliable conclusions about the human risk associated with exposures to mixtures of chemicals.

Slide presentation available (Adobe Acrobat)





NCAC-AGT-JHU Reception Tues. March 12 from 6:30 pm–9:00 at Pratt St. Ale House, Baltimore. Please RSVP by March 10.

The NCAC 10/15/2018 Fall Newsletter is now available; click here to access.

NCAC Fall Symposium Predictive Toxicology Roadmaps Fall Symposium—slides now available.