INTERNATIONAL PROGRAMME ON CHEMICAL SAFETY
ENVIRONMENTAL HEALTH CRITERIA 210
PRINCIPLES FOR THE ASSESSMENT OF RISKS TO HUMAN HEALTH FROM
EXPOSURE TO CHEMICALS
This report contains the collective views of an international group
of experts and does not necessarily represent the decisions or the
stated policy of the United Nations Environment Programme, the
International Labour Organisation, or the World Health
Organization.
Published under the joint sponsorship of the United Nations
Environment Programme, the International Labour Organisation, and the
World Health Organization, and produced within the framework of the
Inter-Organization Programme for the Sound Management of Chemicals.
World Health Organization
Geneva, 1999
The International Programme on Chemical Safety (IPCS),
established in 1980, is a joint venture of the United Nations
Environment Programme (UNEP), the International Labour Organisation
(ILO), and the World Health Organization (WHO). The overall
objectives of the IPCS are to establish the scientific basis for
assessment of the risk to human health and the environment from
exposure to chemicals, through international peer review processes, as
a prerequisite for the promotion of chemical safety, and to provide
technical assistance in strengthening national capacities for the
sound management of chemicals.
The Inter-Organization Programme for the Sound Management of
Chemicals (IOMC) was established in 1995 by UNEP, ILO, the Food and
Agriculture Organization of the United Nations, WHO, the United
Nations Industrial Development Organization, the United Nations
Institute for Training and Research, and the Organisation for Economic
Co-operation and Development (Participating Organizations), following
recommendations made by the 1992 UN Conference on Environment and
Development to strengthen cooperation and increase coordination in the
field of chemical safety. The purpose of the IOMC is to promote
coordination of the policies and activities pursued by the
Participating Organizations, jointly or separately, to achieve the
sound management of chemicals in relation to human health and the
environment.
WHO Library Cataloguing-in-Publication Data
Principles for the assessment of risks to human health from exposure
to chemicals.
(Environmental health criteria ; 210)
1.Chemicals - toxicity
2.Chemicals - adverse effects
3.Risk assessment - methods
4.Environmental exposure
5.Toxicity tests
6.Dose-response relationship, Drug
7.No-observed-adverse effect level
I.International Programme on Chemical Safety
II.Series
ISBN 92 4 157210 8 (NLM Classification: QV 602)
ISSN 0250-863X
The World Health Organization welcomes requests for permission to
reproduce or translate its publications, in part or in full.
Applications and enquiries should be addressed to the Office of
Publications, World Health Organization, Geneva, Switzerland, which
will be glad to provide the latest information on any changes made to
the text, plans for new editions, and reprints and translations
already available.
(c) World Health Organization 1999
Publications of the World Health Organization enjoy copyright
protection in accordance with the provisions of Protocol 2 of the
Universal Copyright Convention. All rights reserved.
The designations employed and the presentation of the material in this
publication do not imply the expression of any opinion whatsoever on
the part of the Secretariat of the World Health Organization
concerning the legal status of any country, territory, city, or area
or of its authorities, or concerning the delimitation of its frontiers
or boundaries.
The mention of specific companies or of certain manufacturers'
products does not imply that they are endorsed or recommended by the
World Health Organization in preference to others of a similar nature
that are not mentioned. Errors and omissions excepted, the names of
proprietary products are distinguished by initial capital letters.
CONTENTS
PRINCIPLES FOR THE ASSESSMENT OF RISKS TO HUMAN HEALTH FROM EXPOSURE
TO CHEMICALS
PREAMBLE
ABBREVIATIONS
1. SUMMARY
2. INTRODUCTION
3. HEALTH HAZARD IDENTIFICATION
3.1. Introduction
3.2. Human data
3.2.1. Criteria for establishing causality
3.3. Animal studies
3.4. In vitro studies
3.5. Structure-activity relationships
4. DOSE-RESPONSE
4.1. Introduction
4.2. Considerations in dose-response assessment
4.2.1. Introduction
4.2.2. Inter- and intra-species considerations
4.2.2.1 Introduction
4.2.2.2 Species differences
4.2.2.3 Human variability
4.3. Non-neoplastic (threshold) effects
4.3.1. Characterization of threshold
4.3.1.1 No-observed-adverse-effect level (NOAEL)
4.3.1.2 Benchmark dose/concentration
4.3.1.3 Lowest-observed-adverse-effect level
4.3.2. Uncertainty factors
4.4. Quantitative risk assessment for neoplastic (non-threshold)
effects
4.4.1. Introduction
4.4.2. Linear extrapolation
4.4.3. Estimation of potency in the experimental range
4.4.4. Two-stage clonal expansion model
4.4.5. Proportional analyses - carcinogenic and
non-neoplastic effects
5. EXPOSURE ASSESSMENT
5.1. Definition of exposure and related terms
5.2. Exposure and dose
5.3. Approaches to quantification of exposure
5.3.1. Measurement at point of contact (personal
monitoring)
5.3.2. Scenario evaluation method (time activity and
monitoring/modelling)
5.3.3. Biomarkers of exposure/estimation of internal dose
5.4. Variability and uncertainty
5.4.1. Assessing uncertainty
5.5. Exposure settings
5.5.1. Exposure in the general environment
5.5.2. Occupational settings
5.5.3. Consumer products
6. RISK CHARACTERIZATION AND IMPLICATIONS FOR RISK MANAGEMENT
6.1. General considerations
6.2. Considerations in risk characterization
6.3. Considerations in risk management
6.3.1. Societal factors
6.3.2. Individual and population risks
6.3.3. Comparative risk
6.3.4. Risk perception
6.3.5. Risk and hazard communication
6.3.6. Economic factors
6.3.6.1Cost-benefit analyses
6.3.7. Political factors
6.3.8. Regulatory limits
6.4. Risk management options
6.4.1. Risk reduction
6.4.1.1 Technology-based criteria
REFERENCES
APPENDIX
RÉSUMÉ
RESUMEN
NOTE TO READERS OF THE CRITERIA MONOGRAPHS
Every effort has been made to present information in the criteria
monographs as accurately as possible without unduly delaying their
publication. In the interest of all users of the Environmental Health
Criteria monographs, readers are requested to communicate any errors
that may have occurred to the Director of the International Programme
on Chemical Safety, World Health Organization, Geneva, Switzerland, in
order that they may be included in corrigenda.
* * *
A detailed data profile and a legal file can be obtained from the
International Register of Potentially Toxic Chemicals, Case postale
356, 1219 Châtelaine, Geneva, Switzerland (telephone no. + 41
22 - 9799111, fax no. + 41 22 - 7973460, E-mail irptc@unep.ch).
* * *
This publication was made possible by grant number
5 U01 ES02617-15 from the National Institute of Environmental Health
Sciences, National Institutes of Health, USA, and by financial support
from the European Commission.
Environmental Health Criteria
PREAMBLE
Objectives
In 1973 the WHO Environmental Health Criteria Programme was
initiated with the following objectives:
(i) to assess information on the relationship between exposure
to environmental pollutants and human health, and to provide
guidelines for setting exposure limits;
(ii) to identify new or potential pollutants;
(iii) to identify gaps in knowledge concerning the health effects
of pollutants;
(iv) to promote the harmonization of toxicological and
epidemiological methods in order to have internationally
comparable results.
The first Environmental Health Criteria (EHC) monograph, on
mercury, was published in 1976 and since that time an ever-increasing
number of assessments of chemicals and of physical effects have been
produced. In addition, many EHC monographs have been devoted to
evaluating toxicological methodology, e.g., for genetic, neurotoxic,
teratogenic and nephrotoxic effects. Other publications have been
concerned with epidemiological guidelines, evaluation of short-term
tests for carcinogens, biomarkers, effects on the elderly and so
forth.
Since its inauguration the EHC Programme has widened its scope,
and the importance of environmental effects, in addition to health
effects, has been increasingly emphasized in the total evaluation of
chemicals.
The original impetus for the Programme came from World Health
Assembly resolutions and the recommendations of the 1972 UN Conference
on the Human Environment. Subsequently the work became an integral
part of the International Programme on Chemical Safety (IPCS), a
cooperative programme of UNEP, ILO and WHO. In this manner, with the
strong support of the new partners, the importance of occupational
health and environmental effects was fully recognized. The EHC
monographs have become widely established, used and recognized
throughout the world.
The recommendations of the 1992 UN Conference on Environment and
Development and the subsequent establishment of the Intergovernmental
Forum on Chemical Safety with the priorities for action in the six
programme areas of Chapter 19, Agenda 21, all lend further weight to
the need for EHC assessments of the risks of chemicals.
Scope
The criteria monographs are intended to provide critical reviews
on the effect on human health and the environment of chemicals and of
combinations of chemicals and physical and biological agents. As
such, they include and review studies that are of direct relevance for
the evaluation. However, they do not describe every study carried
out. Worldwide data are used and are quoted from original studies,
not from abstracts or reviews. Both published and unpublished reports
are considered and it is incumbent on the authors to assess all the
articles cited in the references. Preference is always given to
published data. Unpublished data are only used when relevant
published data are absent or when they are pivotal to the risk
assessment. A detailed policy statement is available that describes
the procedures used for unpublished proprietary data so that this
information can be used in the evaluation without compromising its
confidential nature (WHO (1990) Revised Guidelines for the Preparation
of Environmental Health Criteria Monographs. PCS/90.69, Geneva, World
Health Organization).
In the evaluation of human health risks, sound human data,
whenever available, are preferred to animal data. Animal and
in vitro studies provide support and are used mainly to supply
evidence missing from human studies. It is mandatory that research on
human subjects is conducted in full accord with ethical principles,
including the provisions of the Helsinki Declaration.
The EHC monographs are intended to assist national and
international authorities in making risk assessments and subsequent
risk management decisions. They represent a thorough evaluation of
risks and are not, in any sense, recommendations for regulation or
standard setting. These latter are the exclusive purview of national
and regional governments.
Content
The layout of EHC monographs for chemicals is outlined below.
* Summary -- a review of the salient facts and the risk evaluation
of the chemical
* Identity -- physical and chemical properties, analytical methods
* Sources of exposure
* Environmental transport, distribution and transformation
* Environmental levels and human exposure
* Kinetics and metabolism in laboratory animals and humans
* Effects on laboratory mammals and in vitro test systems
* Effects on humans
* Effects on other organisms in the laboratory and field
* Evaluation of human health risks and effects on the environment
* Conclusions and recommendations for protection of human health
and the environment
* Further research
* Previous evaluations by international bodies, e.g., IARC, JECFA,
JMPR
Selection of chemicals
Since the inception of the EHC Programme, the IPCS has organized
meetings of scientists to establish lists of priority chemicals for
subsequent evaluation. Such meetings have been held in: Ispra, Italy,
1980; Oxford, United Kingdom, 1984; Berlin, Germany, 1987; and North
Carolina, USA, 1995. The selection of chemicals has been based on the
following criteria: the existence of scientific evidence that the
substance presents a hazard to human health and/or the environment;
the possible use, persistence, accumulation or degradation of the
substance shows that there may be significant human or environmental
exposure; the size and nature of populations at risk (both human and
other species) and risks for environment; international concern, i.e.
the substance is of major interest to several countries; adequate data
on the hazards are available.
If an EHC monograph is proposed for a chemical not on the
priority list, the IPCS Secretariat consults with the Cooperating
Organizations and all the Participating Institutions before embarking
on the preparation of the monograph.
Procedures
The order of procedures that result in the publication of an EHC
monograph is shown in the flow chart. A designated staff member of
IPCS, responsible for the scientific quality of the document, serves
as Responsible Officer (RO). The IPCS Editor is responsible for
layout and language. The first draft, prepared by consultants or,
more usually, staff from an IPCS Participating Institution, is based
initially on data provided from the International Register of
Potentially Toxic Chemicals, and reference data bases such as Medline
and Toxline.
The draft document, when received by the RO, may require an
initial review by a small panel of experts to determine its scientific
quality and objectivity. Once the RO finds the document acceptable as
a first draft, it is distributed, in its unedited form, to well over
150 EHC contact points throughout the world who are asked to comment
on its completeness and accuracy and, where necessary, provide
additional material. The contact points, usually designated by
governments, may be Participating Institutions, IPCS Focal Points, or
individual scientists known for their particular expertise. Generally
some four months are allowed before the comments are considered by the
RO and author(s). A second draft incorporating comments received and
approved by the Director, IPCS, is then distributed to Task Group
members, who carry out the peer review, at least six weeks before
their meeting.
The Task Group members serve as individual scientists, not as
representatives of any organization, government or industry. Their
function is to evaluate the accuracy, significance and relevance of
the information in the document and to assess the health and
environmental risks from exposure to the chemical. A summary and
recommendations for further research and improved safety aspects are
also required. The composition of the Task Group is dictated by the
range of expertise required for the subject of the meeting and by the
need for a balanced geographical distribution.
The three cooperating organizations of the IPCS recognize the
important role played by nongovernmental organizations.
Representatives from relevant national and international associations
may be invited to join the Task Group as observers. While observers
may provide a valuable contribution to the process, they can only
speak at the invitation of the Chairperson. Observers do not
participate in the final evaluation of the chemical; this is the sole
responsibility of the Task Group members. When the Task Group
considers it to be appropriate, it may meet in camera.
All individuals who as authors, consultants or advisers
participate in the preparation of the EHC monograph must, in addition
to serving in their personal capacity as scientists, inform the RO if
at any time a conflict of interest, whether actual or potential, could
be perceived in their work. They are required to sign a conflict of
interest statement. Such a procedure ensures the transparency and
probity of the process.
When the Task Group has completed its review and the RO is
satisfied as to the scientific correctness and completeness of the
document, it then goes for language editing, reference checking, and
preparation of camera-ready copy. After approval by the Director,
IPCS, the monograph is submitted to the WHO Office of Publications for
printing. At this time a copy of the final draft is sent to the
Chairperson and Rapporteur of the Task Group to check for any errors.
It is accepted that the following criteria should initiate the
updating of an EHC monograph: new data are available that would
substantially change the evaluation; there is public concern for
health or environmental effects of the agent because of greater
exposure; an appreciable time period has elapsed since the last
evaluation.
All Participating Institutions are informed, through the EHC
progress report, of the authors and institutions proposed for the
drafting of the documents. A comprehensive file of all comments
received on drafts of each EHC monograph is maintained and is
available on request. The Chairpersons of Task Groups are briefed
before each meeting on their role and responsibility in ensuring that
these rules are followed.
PARTICIPANTS IN THE PLANNING AND TASK GROUP MEETINGS ON PRINCIPLES FOR
THE ASSESSMENT OF RISKS TO HUMAN HEALTH FROM EXPOSURE TO CHEMICALS
Members
Dr A. Aitio, Institute of Occupational Health, Laboratory of
Biochemistry, Helsinki, Finland a,b
Dr N. Aldrige, The Robens Institute of Industrial and Environmental
Health and Safety, University of Guildford, Guildford, Surrey, United
Kingdom (deceased)a,b
Dr D. Anderson, British Industry Biological Research Association
(BIBRA), Carshalton, Surrey, United Kingdoma,b
Professor C.L. Berry, Department of Morbid Anatomy, London Hospital
Medical College, London, United Kingdoma
Dr R. Burnett, Biostatistics and Computer Division, Environmental
Health Directorate, Health and Welfare Canada, Ottawa, Ontario,
Canadaa
Dr J.R.P. Cabral, Unit of Mechanisms of Carcinogenesis, International
Agency for Research on Cancer, Lyon, Francea
Dr E. Cardis, Unit of Biostatistics Research and Informatics,
International Agency for Research on Cancer, Lyon, Francea
Dr M. Cikrt, Institute of Hygiene and Epidemiology, Prague, Czech
Republica
Dr D.B. Clayson, Carp, Ontario, Canada
Mr D.J. Clegg, Pesticide Section, Toxicological Evaluation Division,
Food Directorate, Health Protection Branch, Tunney's Pasture, Ottawa,
Ontario, Canadaa
Professor E. Dybing, Department of Environmental Medicine, National
Institute of Public Health, Oslo, Norwayc
Dr R. Fielder, Department of Health, Elephant and Castle, London
United Kingdomb
Dr L. Fishbein, Fairfax, Virginia, USAc
Dr H. Gibb, US Environmental Protection Agency, Washington, DC,
USAa,b,d
Dr M. Goddard, Biostatistics and Computer Division, Environmental
Health Centre, Health and Welfare Canada, Tunney's Pasture, Ottawa,
Ontario, Canadab
Professor B. Goldstein, Rutgers Medical College, Busch Campus,
Pescataway, New Jersey, USAa
Dr R.F. Hertel, Federal Institute for Consumers, Health Protection and
Veterinary Medicine, FE-821 Bundesgesundheitsamt, BGVV, Berlin,
Germanyc,d
Dr J. Huff, Environmental Carcinogenesis Programme, National Institute
of Environmental Health Sciences, Research Triangle Park, North
Carolina, USAb
Professor M. Ikeda, Department of Environmental Health, Tohoku
University School of Medicine, Sendai, Japana
Dr D. Krewski, Biostatistics and Computer Division, Environmental
Health Directorate, Health and Welfare Canada, Ottawa, Ontario,
Canadaa
Professor R. Kroes, initially National Institute of Public Health
and Environmental Hygiene, Bilthoven, subsequently Research
Institute for Toxicology, University of Utrecht, Utrecht, the
Netherlandsa,c
Professor M. Lotti, University of Padua Medical School, Institute of
Occupational Medicine, Padua, Italya
Dr G.W. Lucier, Division of Biometry and Risk Assessment, National
Institute of Environmental Health Sciences, Research Triangle Park,
North Carolina, USAa
Dr L. Magos, Toxicology Unit, Medical Research Council Laboratories,
Carshalton, Surrey, United Kingdoma
Dr E. McConnell, Raleigh, North Carolina, USAa
Ms M.E. Meek, Environmental Health Directorate, Health Canada, Ottawa,
Ontario, Canadac
Dr R.L. Melnick, National Institute of Environmental Health Sciences,
Division of Biometry and Risk Assessment, Research Triangle Park,
North Carolina, USAa
Professor D.V. Parke, Department of Biochemistry, University of
Surrey, Guildford, Surrey, United Kingdoma
Dr J. Parker, Office of Health and Environmental Assessment, US
Environmental Protection Agency, Washington, DC, USAa
Dr O.E. Paynter, Hazard Evaluation Division, US Environmental
Protection Agency, Washington, DC, USAa
Dr P.K. Ray, Industrial Toxicology Research Centre, Lucknow, Indiaa
Dr A.G. Renwick, Clinical Pharmacology Group, University of
Southampton, Southhampton, Hampshire, United Kingdomc
Dr J. Sekizawa, Division of Information on Chemical Safety, National
Institute of Hygienic Sciences, Tokyo, Japanb
Dr J. Shaum, US Environmental Protection Agency, National Center for
Environmental Assessment, Washington, DC, USAd
Professor J.A. Sokal, Institute of Occupational Medicine and
Environmental Health, Sosnowiec, Polandc
Dr J. Steadman, Department of Health and Social Security, Elephant and
Castle, London, United Kingdoma
Dr L. Strayner, Division of Standards Development and Technology
Transfer, National Institute for Occupational Safety and Health,
Cincinnati, Ohio, USAb
Dr G.M.H. Swaen, Department of Occupational Medicine, University of
Limburg, Maastricht, the Netherlandsa,b
Dr A. Walker, Organisation for Economic Co-operation and Development,
Paris, Francea
Professor R. Walker, Food Safety Group, Division of Toxicology, School
of Biological Sciences, University of Surrey, Guildford, Surrey,
United Kingdomc
Dr J.E. Zejda, Department of Epidemiology, Institute of Occupational
Medicine and Environmental Health, Sosnowiec, Polandc
Observers
Professor G. Di Renzo, International Union of Toxicology, Department
of Neuroscience, Faculty of Medicine and Surgery, University of Naples
"Federico II", Naples, Italyc
Dr M. Jaroszewski, Health and Safety Directorate, Occupational
Medicine and Hygiene Unit, Commission of the European Community,
Luxembourgb
Dr C. Lally, European Council of Chemical Industry Federation (CEFIC),
Procter and Gamble, Strombbek Bever, Belgiumc
Professor A. Mutti, Institute of Clinical Medicine and Nephrology,
Parma, Italyc
Dr J. O'Donoghue (Representing AIHC) Corporate Health and Environment
Laboratories, Eastman Kodak Company, Rochester, New York, USAb
Dr M. Penman, ICI C & P Limited, Occupational Health Division, Wilton,
Middlesborough, Cleveland, United Kingdomc
Mrs M. Richold, European Centre for Ecotoxicology and Toxicology of
Chemicals (ECETOC), Unilever Research Laboratory, Environmental Safety
Laboratory, Sharnbrook, Bedford, United Kingdomc
Mr P. Verschuren, International Life Sciences Institute, Brussels
Belgiumc,b
Secretariat
Dr G.C. Becking, Inter-regional and Research Unit, International
Programme on Chemical Safety, World Health Organization, Research
Triangle Park, North Carolina, USAb
Dr K. Gutschmidt, International Programme on Chemical Safety, World
Health Organization, Geneva, Switzerlandd
Dr E. Smith, International Programme on Chemical Safety, World Health
Organization, Geneva, Switzerlandc
Dr M. Younes, International Programme on Chemical Safety, World Health
Organization, Geneva, Switzerlandd
a Participated in Planning and Working Groups on Scientific
Principles for the Assessment of Risks to Human Health from
Exposure to Chemicals.
b Participated in the WHO Task Group Meeting on the initial draft
of Principles for the Assessment of Risk from Exposure to
Chemicals (British Industry Biological Research Association
(BIBRA), Carshalton, Surrey, United Kingdom, March 1993).
c Participated in the WHO Task Group Meeting on the initial draft
of General Principles and Methods for Chemical Safety (Human
Health Protection (National Institute of Public Health and
Environmental Protection) (RIVM), Bilthoven, the Netherlands, 22-
25 March 1994).
d Participated in the WHO Finalizing Group Meetings on Principles
for the Assessment of Risks to Human Health from Exposure to
Chemicals (World Health Organization, Geneva, Switzerland, 2-5
September 1996 and 18-20 September 1997).
PRINCIPLES FOR THE ASSESSMENT OF RISKS TO HUMAN HEALTH FROM EXPOSURE
TO CHEMICALS
This monograph is an amalgamation of two draft documents
"Principles for the Assessment of Risk from Exposure to Chemicals" and
"General Principles and Methods for Chemical Safety (Human Health
Protection)".
Both documents were planned to cover different aspects of
chemical safety and risk assessment; one dealing with the basic
science for general readers, and the other providing more practical
approaches to risk assessment of chemicals for risk assessors.
However, they turned out to have a substantial amount of overlapping
information and it was therefore decided to use both drafts as a basis
for this new, comprehensive document. The more detailed draft on
"General Principles and Methods for Chemical Safety (Human Health
Protection)" will be published as a separate document for training
purposes.
This Environmental Health Criteria monograph is aimed at
furnishing a practical overview of chemical safety and at providing
the framework of risk assessment for regulatory and research
scientists, as well as risk managers. It is intended to complement
existing Environmental Health Criteria that address methodologies for
the assessment of risks from exposure to chemicals with a view towards
different end-points or to susceptible population groups. It is not
intended as a textbook on toxicology.
This monograph should not be considered as being of a
prescriptive nature. The chapters on exposure assessment and risk
characterization, in particular, provide rather some practical
guidance.
Several planning, working and Task Group meetings took place to
discuss and agree upon the structures and contents of both
Environmental Health Criteria documents.
A WHO Task Group on "Principles for the Assessment of Risk from
Exposure to Chemicals" met at the British Industrial Biological
Research Association (BIBRA), Carshalton, Surrey, United Kingdom, in
March 1993. Dr G.C. Becking, IPCS, welcomed the participants on behalf
of the Director, IPCS, and the three IPCS cooperating organizations
(UNEP/ILO/WHO), and the Task Group reviewed the draft document.
The main contributors to the first draft on Principles for the
Assessment of Risk from Exposure to Chemicals were Dr N. Aldridge,
Robens Institute of Industrial and Environmental Health and Safety,
United Kingdom, Dr H. Gibb, US Environmental Protection Agency, Dr J.
Huff, National Institute of Environmental Health Sciences, USA, Dr L
Stayner, National Institute for Occupational Safety and Health, USA.
A second WHO Task Group met to review the draft monograph on
General Principles and Methods for Chemical Safety (Human Health
Protection). This group met in at the National Institute of Public
Health and Environmental Protection (RIVM), Bilthoven, the
Netherlands, from 22 to 25 November 1995. Dr E. Smith, IPCS, welcomed
the participants on behalf of the Director, IPCS, and the three IPCS
cooperating organizations (UNEP/ILO/WHO), and the Task Group reviewed
the draft document.
The main contributors to the draft on Principles for the
Assessment of Risk from Exposure to Chemicals were Dr D.B. Clayson,
Carp, Canada, Professor E. Dybing, National Institute of Public
Health, Norway, Dr L. Fishbein, Fairfax, Virginia, USA, Dr A.G.
Renwick, University of Southampton, United Kingdom, Professor R.
Walker, University of Surrey, United Kingdom, and Professor J.A Sokal,
Institute of Occupational Health and Environmental Medicine,
Sosnowiec, Poland.
In addition to the Task Group meetings, meetings were held during
1996 and 1997 in Geneva to combine the two documents.
Dr E. Smith and Dr G. Becking, both members of the IPCS, were
responsible for the preparation of the initial draft documents. Dr M.
Younes (IPCS) was responsible for the overall scientific content of
the final monograph and Dr P.G. Jenkins (IPCS) for the technical
editing.
The efforts of all who helped in the preparation and finalization
of the document are gratefully acknowledged.
ABBREVIATIONS
ADD average daily dose
ADI acceptable daily intake
EPI exposure/potency index
GLP good laboratory practice
IARC International Agency for Research on Cancer
LOAEL lowest-observed-adverse-effect level
NOAEL no-observed-adverse-effect level
OECD Organisation for Economic Co-operation and Development
PBPK physiologically based pharmacokinetic
SAR structure-activity relationship
US EPA US Environmental Protection Agency
1. SUMMARY
Control of risks from exposure to chemicals (chemical safety)
requires first of all a scientific, ideally quantitative, assessment
of potential effects at given exposure levels (risk assessment). Based
upon the results of risk assessment, and taking into consideration
other factors, a decision-making process aimed at eliminating or, if
this is not possible, reducing to a minimum the risk to the
chemical(s) under consideration (risk management), can be started.
Risk assessment is a conceptual framework that provides the
mechanism for a structured review of information relevant to
estimating health or environmental outcomes. In conducting risk
assessments, the National Academy of Sciences risk assessment paradigm
has proven to be a useful tool (US NAS, 1983). This paradigm divides
the risk assessment process into four distinct steps: hazard
identification, dose-response assessment, exposure assessment and risk
characterization.
The purpose of hazard identification is to evaluate the weight of
evidence for adverse effects in humans based on assessment of all
available data on toxicity and mode of action. It is designed to
address primarily two questions: (1) whether an agent may pose a
health hazard to human beings, and (2) under what circumstances an
identified hazard may be expressed. Hazard identification is based on
analyses of a variety of data that may range from observations in
humans to analysis of structure-activity relationships. The result of
the hazard identification exercise is a scientific judgement as to
whether the chemical evaluated can, under given exposure conditions,
cause an adverse health effect in humans. Generally, toxicity is
observed in one or more target organ(s). Often, multiple end-points
are observed following exposure to a given chemical. The critical
effect, which is usually the first significant adverse effect that
occurs with increasing dose, is determined.
Dose-response assessment is the process of characterizing the
relationship between the dose of an agent administered or received and
the incidence of an adverse health effect. For most types of toxic
effects (i.e. organ-specific, neurological/behavioural, immunological,
non-genotoxic carcinogenesis, reproductive or developmental), it is
generally considered that there is a dose or concentration below which
adverse effects will not occur (i.e. a threshold). For other types of
toxic effects, it is assumed that there is some probability of harm at
any level of exposure (i.e. that no threshold exists). At the present
time, the latter assumption is generally applied primarily for
mutagenesis and genotoxic carcinogenesis.
If a threshold has been assumed (e.g., for non-neoplastic effects
and non-genotoxic carcinogens), traditionally, a level of exposure
below which it is believed that there are no adverse effects, based on
a no-observed-adverse-effect level (NOAEL) (approximation of the
threshold) and uncertainty factors, has been estimated. Alternatively,
the magnitude by which the no (lowest)-observed-adverse-effect level
(N(L)OAEL) exceeds the estimated exposure (i.e. the "margin of
safety") is considered in light of various sources of uncertainty. In
the past, this approach has often been described as a "safety
evaluation". Therefore, the dose that can be considered as a first
approximation of the threshold, i.e. the NOAEL, is critical.
Increasingly, however, the "benchmark dose", a model-derived estimate
(or its lower confidence limit) of a particular incidence level (e.g.,
5%) for the critical effect, is being proposed for use in quantitative
assessment of the dose-response for such effects.
There is no clear consensus on appropriate methodology for the
risk assessment of chemicals for which the critical effect may not
have a threshold (i.e. genotoxic carcinogens and germ cell mutagens).
Indeed, a number of approaches based largely on characterization of
dose-response have been adopted for assessment in such cases.
Therefore, the critical data points are those that define the slope of
the dose-response relationship (rather than the NOAEL, which is the
first approximation of a threshold).
The third step in the process of risk assessment is the exposure
assessment, which has the aim of determining the nature and extent of
contact with chemical substances experienced or anticipated under
different conditions. Multiple approaches can be used to conduct
exposure assessments. Generally, approaches include indirect and
direct techniques, covering measurement of environmental
concentrations and personal exposures, as well as biomarkers.
Questionnaires and models are also often used. Exposure assessment
requires the determination of the emissions, pathways and rates of
movement of a substance and its transformation or degradation, in
order to estimate the concentrations to which human populations or
environmental spheres (water, soil and air) may be exposed.
Depending on the purpose of an exposure assessment, the numerical
output may be an estimate of either the intensity, rate, duration or
frequency of contact exposure or dose (resulting amount that actually
crosses the boundary). For risk assessments based on dose-response
relationships, the output usually includes an estimate of dose. It is
important to note that the internal dose, not the external exposure
level, determines the toxicological outcome of a given exposure.
Risk characterization is the final step in risk assessment. It is
designed to support risk managers by providing, in plain language, the
essential scientific evidence and rationale about risk that they need
for decision-making. In risk characterization, estimates of the risk
to human health under relevant exposure scenarios are provided. Thus,
a risk characterization is an evaluation and integration of the
available scientific evidence used to estimate the nature, importance,
and often the magnitude of human and/or environmental risk, including
attendant uncertainty, that can reasonably be estimated to result from
exposure to a particular environmental agent under specific
circumstances.
The term "risk management" encompasses all of those activities
required to reach decisions on whether an associated risk requires
elimination or necessary reduction. Risk management strategies/or
options can be broadly classified as regulatory, non-regulatory,
economic, advisory or technological, which are not mutually exclusive.
Thus legislative mandates (statutory guidance), political
considerations, socioeconomic values, cost, technical feasibility,
populations at risk, duration and magnitude of risk, risk comparison,
and possible impact on trade between countries can generally be
considered as a broad panoply of elements that can be factored into
final policy or rule making. Key decision factors such as the size of
the population, the resources, costs of meeting targets and the
scientific quality of risk assessment and subsequent managerial
decisions vary enormously from one decision context to another. It is
also recognized that risk management is a complex multidisciplinary
procedure which is seldom codified or uniform, is frequently
unstructured, but which can respond to evolving input from a wide
variety of sources. Increasingly, risk perception and risk
communication are recognized as important elements, which must also be
considered for the broadest possible public acceptance of risk
management decisions.
Chemicals have become an indispensable part of human life,
sustaining activities and development, preventing and controlling many
diseases, and increasing agricultural productivity. Despite their
benefits, chemicals may, especially when misused, cause adverse
effects on human health and environmental integrity. The widespread
application of chemicals throughout the world increases the potential
of adverse effects. The growth of chemical industries, both in
developing as well as in developed countries, is predicted to continue
to increase. In this context, it is recognized that the assessment and
management of risks from exposure to chemicals are among the highest
priorities in pursuing the principles of sustainable development.
2. INTRODUCTION
Despite the societal benefits that accrue from the use of
chemicals, substantial potential hazards to health may be associated
with exposure during the production, use or disposal of the
approximately 100 000 unique chemicals or 4 million mixtures,
formulations and blends already in commercial use or the several
hundred new synthetic chemicals introduced each year (EC, 1990). This
monograph outlines the nature of the data available and their use in
the assessment of risk in a risk assessment/risk management framework.
It is hoped that scientists, risk assessors and health risk managers
will find this monograph helpful to decision-making in this area.
A number of national and international organizations and agencies
have developed guidance on assessment of exposure and various health
end-points (e.g., carcinogenicity, developmental toxicity, etc.). It
is not the purpose of this monograph to endorse particular approaches
but rather to acquaint the reader with relevant methodology and issues
for consideration.
It is also hoped that the reader will find this monograph useful
in the interpretation of risk assessments on specific chemicals. The
reader is referred to such sources for chemical-specific hazard
identification and, depending on the monograph, dose-response
information. A list of assessments produced by various national and
international agencies is included in ECETOC/UNEP (1996). These
sources do not, of course, provide the exposure information necessary
to characterize risk at the local level. Since exposure will vary
considerably under different circumstances, responsible authorities
are strongly encouraged to characterize risk on the basis of local
measured or predicted exposure scenarios. It is hoped that the general
approaches to exposure assessment described in this monograph will
assist the reader in characterizing risk in specific situations.
In the chapters of this monograph, the following four distinct
and essential components of the risk assessment paradigm are
addressed:
(1) hazard identification - identification of the inherent
capability of a substance to cause adverse effects;
(2) assessment of dose-response relationships involves
characterization of the relationship between the dose of an agent
administered or received and the incidence of an adverse effect;
(3) exposure assessment is the qualitative and/or quantitative
assessment of the chemical nature, form and concentration of a
chemical to which an identified population is exposed from all
sources (air, water, soil and diet);
(4) risk characterization is the synthesis of critically evaluated
information and data from exposure assessment, hazard
identification and dose-response considerations into a summary
that identifies clearly the strengths and weaknesses of the
database, the criteria applied to evaluation and validation of
all aspects of methodology, and the conclusions reached from the
review of scientific information.
The logical consequence of the process of assessment of potential
risk is the application of the information to the development of
practical measures (risk management) for the protection of human
health. Although not the principal focus of this monograph, the
importance of clear understanding and communication of the nature and
limitations of the scientific basis for risk assessment in risk
management is addressed in the final chapter.
In Appendix 1 to this monograph, an example of a hazard
identification scheme for carcinogenicity, developed by the
International Agency for Research on Cancer (IARC), is presented. In
Appendix 2, the currently available and draft guidelines of the
Organisation for Economic Cooperation and Development (OECD) for
testing of chemicals are presented. For sample exposure and risk
characterizations, readers are referred to IPCS (1994).
3. HEALTH HAZARD IDENTIFICATION
3.1 Introduction
The purpose of hazard identification is to evaluate the weight of
evidence for adverse effects in humans based on assessment of all
available data on toxicity and mode of action. It is designed to
address primarily two questions: (a) whether an agent may pose a
health hazard to humans, and (b) under what circumstances an
identified hazard may be expressed. Hazard identification is based on
analyses of a variety of data that may range from observations in
humans to analysis of structure-activity relationships.
In hazard identification, the weight of evidence is assessed on
the basis of combined strength and coherence of inferences
appropriately drawn from all of the available data. This entails
rigorous examination of the quantity, quality and nature of the
results of available toxicological and epidemiological studies and
structure-activity analyses and information on mechanisms of toxicity.
The latter is particularly important with respect to assessment of
relevance to humans.
Several classification schemes provide a framework for assessment
of the weight of evidence for various toxicological end-points (DFG,
1972; IPCS, 1986 (neurotoxicity); US EPA, 1986a, 1996a; IARC, 1987;
EC, 1992; Health Canada, 1994; IPCS, 1996 (immunotoxicity); IPCS, 1997
(delayed hypersensitivity)). An example (the IARC scheme) is presented
in Appendix 1 to illustrate the nature of criteria on which
classification of weight of evidence is based. Such classification
schemes have been helpful in standardizing and communicating the
assessment of hazard identification for particular end-points. In
addition to the classifications themselves, narrative statements to
summarize the nature of and confidence in the evidence based on
limitations and strengths of the database are helpful. Issues that are
often addressed include: the nature, reliability, validity and
consistency of data on response in humans and in laboratory animals,
current knowledge of the mechanistic basis for the response, and, in
the absence of human data, the relevance of responses in experimental
animals to humans.
The result of the hazard identification exercise is a scientific
judgement as to whether the chemical can cause an adverse effect in
humans.
The following is intended to provide the reader with an
appreciation of the complexity of considerations made in assessing
different types of data as a basis for hazard identification in risk
assessment. Fundamentals of epidemiology and toxicity testing are not
addressed here since they are considered in several other sources. An
Environmental Health Criteria monograph on the principles of exposure
assessment is currently in preparation (IPCS, in preparation).
Each source of information (e.g., human data, animal data,
structure-activity relationships) has its advantages and limitations
in contributing to an assessment of weight of evidence, but,
collectively, they permit characterization of potential adverse health
effects.
3.2 Human data
Well-documented observational and clinical epidemiological
studies have the clear advantage over studies in animals in providing
the most relevant information on health effects in the species of
interest, thus avoiding extrapolation from animals to humans. In
addition, epidemiological studies can address hazards to which humans
are exposed in their natural environment, in the presence of
concomitant risk factors such as diet and smoking.
Human populations are heterogeneous in their composition, and
studies of exposed populations are likely to include individuals of
differing susceptibility to the chemical of interest. This may be
viewed as an advantage relative to toxicological studies, which
involve genetically homogeneous populations of test animals.
The database for direct hazard identification in human
populations consists primarily of observational (epidemiological)
studies and case reports. Some information is also available from
ethically conducted human volunteer studies.
In observational studies, the investigator does not control
assignment of study subjects to either exposed or non-exposed groups.
Rather, such studies involve investigation of various individuals or
groups of subjects as they happen to have been exposed, and at no
stage of the study is the exposure of subjects influenced by the
research protocol. Although exposure scenarios are more realistic than
those in the experimental setting, owing to their observational nature
it is often difficult to control for "confounding factors", which may
be contributing to the etiology of the disease being investigated. For
example, variations in smoking between groups may confound
interpretation of observations concerning lung cancer.
Ethical experimental studies in human volunteers offer the
advantage of being better able to control for confounding factors. The
assignment of study subjects to exposure groups is made by the
investigator, who also controls the quality and quantity. Although
such investigations are generally reliable for the establishment of
both causality and exposure-response relationships, they are most
often restricted for ethical reasons to the examination of mild,
temporary effects (e.g., neurobehavioural or biochemical changes) of
short-term exposures in a limited number of subjects. They have
contributed considerably, particularly to our understanding of
kinetics and to the development of air quality guidelines and
standards for traditional pollutants.
Case reports describe a particular effect in an individual or
group of individuals who were exposed to a substance and often
observed by a single physician or group of physicians. These reports
are often anecdotal or highly selected in nature. Owing primarily to
their lack of statistical stability, they are of limited use for
hazard assessment, though helpful in generating hypotheses for further
study. However, reports of cases of the disease or effect of interest
can identify associations, particularly when there are unique features
such as an association with a rare disease or effect of interest
(e.g., vinyl chloride and angiosarcoma or methylmercury and Minamata
disease).
The major types of epidemiological (observational) studies are
analytical and descriptive or correlational studies. Each study type
has well-known strengths and weaknesses that affect interpretation of
study results (Lilienfeld & Lilienfeld, 1979; Mausner & Kramer, 1985;
Kelsey et al., 1986; Rothman, 1986). Analytical epidemiological
studies (that is, cohort and case-control studies), in which exposure
and outcome are examined in individuals rather than in populations,
are generally most reliable in hazard identification as a basis for
risk assessment since it is possible to adjust more rigorously for
confounding factors. The assessment of results of such studies is
based on several features of study design including estimation of
exposure, the role of confounding variables and the measurement of
outcome. Potential limitations, depending upon the nature of the
design, include lack of information on exposure, insufficient sample
size, short length of follow-up and potential bias and confounding.
These factors may limit the usefulness of particular studies for the
purposes of risk assessment.
Epidemiological data demonstrating dose-response, if available,
provide an advantageous basis for analysis, since concerns about
inter-species extrapolation do not arise. Adequacy of human exposure
data for quantification is an important consideration in deciding
whether epidemiological data are the best basis for analysis in a
particular case. If adequate exposure data exist in a well-designed
and well-conducted epidemiological study that detects no effects, it
may be possible to obtain an upper estimate of the potential human
risk to provide a check on plausibility of available estimates based
on animal tumour or other responses (e.g., do confidence limits on one
overlap the point estimate of the other?) (Stayner & Bailer, 1993; US
EPA 1996a).
3.2.1 Criteria for establishing causality
The first step in the evaluation of results of studies in humans
as a basis for hazard identification is the assessment of the
individual results of each separate report. The strengths and
weaknesses of each study must be considered along with potential for
the existence of bias (Gehlbach, 1982), with particular attention to
exposure data, criteria for definition of health outcome under study,
the size of the study population and the statistical power of the
analysis to detect adverse health effects. A set of standardized
criteria for assessing the weight of evidence of causality based on
assessment of the database has been developed (Hill, 1965; Susser,
1977).
Studies in which there is an apparent absence of evidence for a
hypothesized causal relationship between exposure and effect
("negative studies") need to be interpreted carefully (Hernberg,
1980). Such studies should be evaluated for dilution (the inclusion of
unexposed people in an allegedly exposed group of persons),
misclassification (Copeland et al., 1977), omissions, or premature
examination of subjects for diseases that may have long induction
(latency) periods. In addition, the statistical power of the study,
i.e. the probability that the study will be able to demonstrate the
presence of an effect, such as excessive disease or mortality, in a
population if the effect is actually present (Beaumont & Breslow,
1981), must be assessed.
There is no clear-cut criterion to distinguish positive from
negative studies. Although statistical significance has often been
used as the criteria, most epidemiologists believe that it is overly
simplistic to base decisions on arbitrary probability values (Rothman,
1986). For example, when a study fails to detect a statistically
significant effect, this may simply reflect inadequate sample size or
other aspects of study design. Conversely, when the results of a study
are statistically significant, the seemingly positive results may
still be due to confounding or even chance.
A positive association between an agent and an effect may be
interpreted as implying causality, to a greater or lesser extent, if
the following criteria are met: (a) there is not identifiable positive
bias; (b) the possibility of positive confounding has been considered;
(c) the association is unlikely to be due to chance alone; (d) the
association is strong; and (e) there is a dose-response relationship
(IARC, 1990). The following criteria for inferring causality from the
results of epidemiological studies have been developed by Hill (1965):
(a) The strength of the association as measured by the relative risk
In general, epidemiologists have more confidence in their results
when the magnitude of the relative risk is large. However, relative
risks of small magnitude do not necessarily imply lack of causality
and may be important if the disease under study is common (IARC,
1990). In evaluating relative risks, it is important to note the
actual numbers of observed and expected cases.
(b) The consistency of the association
The case for causal inference is strengthened by repetition of
findings "by different investigators, in different places,
circumstances and times" (Hill, 1965). The reproducibility of findings
constitutes one of the strongest arguments for the existence of
causality. If there are discordant results among investigations,
possible reasons such as differences in exposure should be considered
in assessing the results, and data from studies judged to be of high
quality given greater weight than data from studies judged to be
methodologically less sound (IARC, 1990).
(c) The temporal relationship between cause and effect
This principle may be simply restated as exposure must precede
illness. When latency is a factor, exposures must have occurred
sufficiently early to have produced an effect by the time of the
study.
(d) The biological gradient of the association
The evidence for causality is strengthened when the risk of
disease is shown to increase with levels of exposure. Because there
are many possible reasons that an epidemiological study may fail to
detect an exposure-response relationship (e.g., poor exposure data,
lack of adequate exposure gradient), the absence of a dose-response
relationship does not necessarily imply that the relationship is not
causal (IARC, 1990). Strong evidence for causality is provided when a
change in exposure brings about a change in disease frequency
(Hernberg, 1980), e.g., the decrease in risk of lung cancer that
follows cessation of smoking (Doll & Hill, 1956).
(e) the specificity of the association
A highly specific association is one in which the disease under
study is only induced by a particular agent. Specificity of cause is
common in infectious diseases but less common in chronic diseases that
often have a multi-factorial etiology. However, a specific association
may be observed for certain chronic diseases such as between exposure
to crocidolite asbestos and mesothelioma or vinyl chloride and
angiosarcoma. Although the presence of specificity seems to imply
causality, its absence does not exclude it (Fralick, 1983).
(f) biological plausibility of the association
Hill (1965) stated strongly that a proposed causal relationship
should not seriously conflict with knowledge of the biology and
pathophysiology of a disease under study. An epidemiological inference
of causality may be strengthened by data from experimental studies
showing consistency with biological mechanisms. For example, exposure
to ionizing radiation causes cancer in many animal species. However,
the lack of mechanistic or positive animal bioassay data to support an
association observed in an epidemiological study is not, in itself,
sufficient reason to reject causality.
3.3 Animal studies
Owing to the lack of adequate epidemiological data for most
substances, toxicological studies in animal species play an important
role in hazard identification for risk assessment. Toxicity studies
vary widely in purpose, design and conduct, and range from relatively
well-standardized and widely accepted test methods for assaying
various types of toxicity to large numbers of basically
research-oriented investigations employing specialized study designs.
The design, conduct and completeness of reporting of experimental
findings in toxicological studies on mammalian species are of critical
importance in determining the validity and relevance of results.
Toxicological results from adequate animal systems signal anticipated
effects in humans. Thus, negative results cannot be assessed from an
inadequate study, and full evaluation of a positive effect is
confounded by incomplete reporting from poorly designed or poorly
conducted studies. However, positive findings cannot be ignored.
Studies should be of good scientific quality and follow standard
guidelines and recognized good laboratory practices (GLPs) wherever
possible.
Information on the design of specific bioassays, including those
that address acute, short-term, sub-chronic, chronic and developmental
and reproductive toxicity, immunotoxicity and carcinogenicity, are not
presented here but are available in test guidelines, for which
principles of GLP are also specified (IARC, 1986; OECD, 1987, 1998;
Chhabra et al., 1990). A list of currently available OECD Guidelines
is included in Appendix 2. In this section, examples of factors to be
taken into account in assessing these various aspects of study design
for hazard identification are described.
Major end-points in toxicity studies can be grouped into the
following categories (IPCS, 1987a):
* Functional manifestations (weight loss, laxative effects, etc.);
* non-neoplastic lesions with morphological
manifestations/organ-directed toxic effects;
* neoplastic/carcinogenic manifestations.
In addition, a number of specific end-points may require targeted
testing strategies. Such end-points include skin and eye irritation,
reproductive/developmental manifestations, immunotoxicity and
neurotoxicity (including neurodevelopmental effects).
It is important to recognize that there are two types of data
generated in such studies; those in which response is graded, such as
enzyme inhibition (i.e. continuous data), and those in which the
response occurs or does not occur in a single animal, such as a
particular tumour (i.e. quantal data).
In assessing the relevance of various toxicological studies to
hazard identification and risk assessment, several features of study
design are considered, including the purity of the compound
administered, physico-chemical properties (volatility, stability,
solubility), homogeneity of distribution in inhalation experiments,
the size of the study (i.e. the number of exposed and control
animals), whether the study adhered to the principles of GLP, the
relevance of the route of exposure to that of humans, duration of
exposure, the number and suitability of the dose levels administered,
the extent of examination of various toxicological end-points and the
statistical analysis of the data. The types, site, incidence and
severity of effects and the nature of the exposure- or dose-response
relationship are also taken into account. Where data indicate that
there are significant differences in absorption, distribution,
metabolism and elimination of the compound in different animal
species, wherever possible, studies in which the species and strain of
animal are most similar to Homo sapiens in this regard are used
(where relevant human data are available). The consistency of the
results of the principal studies are also considered in the assessment
of the weight of evidence for an effect (e.g., whether similar effects
have been observed in studies in other species or whether such effects
would have been expected based on the structure or properties of the
chemical).
For example, the size of each exposure and concurrent control
group should be large enough for thorough toxicological and
statistical evaluation. The number of animals considered sufficient
depends on the variability, sensitivity and nature (e.g., quantal or
continuous) of the end-point being evaluated. For example, it is
commonly 50 per group in carcinogenicity bioassays where the responses
of interest are quantal in nature and 10 per group in subchronic
studies, where many of the examined end-points are continuous.
Studies in which the route of exposure is similar to that of
humans are most relevant to hazard identification for risk assessment.
For substances of low toxicity, it is important to ensure that when
administered in the diet, the quantities of the substance do not
interfere with normal nutritional needs.
Studies designed and conducted with 3-5 dosed groups plus a
vehicle control group of animals will yield reasonable dose-response
data relevant to hazard identification. The highest concentration of
the chemical should be one that induces a recognizable effect in the
animals such as changes in body or organ weights, enzyme changes or
minor histological changes. Changes such as mortality, gross
pathological changes, and painful or stressful conditions should be
avoided as they may confound the results of the study and may not be
in compliance with national and local animal welfare regulations.
Intermediate dose(s) should be targeted to produce minimally
observable toxic effects. Dose levels should be selected to produce
graded responses; too large intervals may complicate accurate
estimations of the lowest-observed-effect level (LOEL). Ideally, the
lowest dose should not demonstrate any toxicity (e.g., a NOAEL).
To assess fully the toxicological potential of a chemical for
local and systemic effects, all major organ systems should be examined
for dose-related effects and adverse effects in various organs should
be evaluated and described.
3.4 In vitro studies
Isolated cells, tissues and organs can be prepared and maintained
in culture by methods that preserve their in vivo properties and
characteristics. Increasing concern about the ethics of animal
experimentation has served to catalyse efforts leading to the possible
replacement or reduction in the use of animals, and the refinement of
test methods to minimize the stress and suffering to animals (ECETOC,
1989; Gelbke, 1993). In vitro testing contributes particularly to
the assessment of genotoxicity, permitting a decision concerning the
need for further testing.
Over the last decade, in vitro tests have been proposed as a
pre-screen or as an alternative method for other end-points, such as
prenatal toxicity, eye irritation, dermal irritation, tumour promotion
and target organ toxicity (Purchase, 1986; Tennant et al., 1987;
Anderson, 1990; Frazier, 1993; Atterwill, 1995). There has been
particular emphasis on validation programmes for skin and eye
irritation, but most of the tests mentioned above have not yet been
sufficiently validated and the results of validation studies,
especially in the past, have been lacking in consistency. The results
have failed to meet the need for reproducibility and high correlation,
ideally with sound human data but usually, for practical reasons, with
existing animal tests, which they are intended to replace.
Aspects that are important in assessing the adequacy of
in vitro studies include:
* the range of exposure levels, taking into account the toxicity of
the substance in the bacteria/cells, its solubility and, where
appropriate, its effects on the pH and osmolality of the culture
medium;
* whether, in the case of volatile substances, precautions were
taken to ensure the maintenance of effective concentrations of
the substance in the test medium;
* whether (when necessary) an appropriate exogenous metabolism mix
(e.g., S9 from induced rat or hamster liver) was used;
* whether appropriate negative and positive controls were included;
and
* whether there was an adequate number of replicates (within the
tests and of the tests).
Clearly, greater mechanistic understanding would facilitate
moving from purely empirical/correlative approaches to more
mechanistic-based tests. This is likely to facilitate greatly the
chances of adequate validation and acceptance of alternatives for
regulatory purposes.
3.5 Structure-activity relationships
Where epidemiological and toxicological data are not available,
the use of structure-activity relationships (SARs) may be considered.
SARs are based on the assumption that chemical substances that reach
and interact with target sites by the same mechanism do so as a result
of their similar chemical properties.
At present, SAR techniques, particularly those of a quantitative
nature, are not well developed in relation to mammalian toxicity. They
are primarily of value in predicting toxicokinetic properties and in
priority setting for research and evaluation.
4. DOSE-RESPONSE
4.1 Introduction
Approaches to quantification of dose-response vary according to
the scope and purpose of assessments. However, for most types of toxic
effects (i.e. organ-specific, neurological/behavioural, immunological,
non-genotoxic carcinogenesis, reproductive or developmental), it is
generally considered that there is a dose or concentration below which
adverse effects will not occur (i.e. a threshold). For other types of
toxic effects, it is assumed that there is some probability of harm at
any level of exposure (i.e. that no threshold exists); this currently
applies primarily for mutagenesis and carcinogenesis. Some have
restricted the non-threshold assumption to genotoxic carcinogens.
The distinction in approaches for genotoxic carcinogens and other
types of toxic effects is based primarily on the premise that simple
events such as in vitro activation and covalent binding may be
linear over many orders of magnitude. Though it is not possible to
demonstrate experimentally the presence or absence of a threshold,
differences in approach to the dose-response assessment of genotoxic
versus non-genotoxic carcinogens have been adopted in some countries.
However, simple pragmatic distinction on this basis is increasingly
problematic. For example, it is likely that there are thresholds for
aneugenic genotoxic effects.
If a threshold has been assumed (e.g., for non-neoplastic effects
and non-genotoxic carcinogens), traditionally, a level of exposure
below which it is believed that there are no adverse effects, based on
a no-observed-adverse-effect level or NOAEL (approximation of the
threshold) and uncertainty factors, has been estimated (section 4.3).
Alternatively, the magnitude by which the N(L)OAEL exceeds the
estimated exposure (i.e. the "margin of safety"), is considered in
light of various sources of uncertainty (Commission Regulation (EC)
No. 1488/94; Council Regulation (EEC) 793/93) (EC, 1993, 1994). In the
past, this approach has often been described as "safety evaluation".
Therefore, the dose that can be considered as a first approximation of
the threshold, i.e. the NOAEL, is critical. Increasingly, however, the
"benchmark dose", a model-derived estimate (or its lower confidence
limit) of a particular incidence level (e.g., 5%) for the critical
effect, is being proposed for use in quantitative assessment of the
dose-response for such effects.
At present, there is no clear consensus on appropriate
methodology for the risk assessment of chemicals for which the
critical effect may not have a threshold (i.e. genotoxic carcinogens
and germ cell mutagens). Indeed, a number of approaches based largely
on characterization of dose-response have been adopted for assessment
in such cases (section 4.4). Therefore, the critical data points are
those that define the slope of the dose-response relationship (rather
than the NOAEL, which is the first approximation of a threshold).
In North America and some European countries, cancer risks have
traditionally been assessed by mathematical modelling of the
dose-response data in the observable range to estimate the risk at
much lower human intakes or exposures (low dose risk extrapolation).
It should be noted, however, that quantitative estimation of such
risks, particularly those orders of magnitude below the experimental
range (i.e. low dose risk estimation), is uncertain. Owing to this
uncertainty, some countries have chosen not to adopt this approach as
the basis for their regulatory actions for genotoxic carcinogens, and
other countries are increasingly adopting alternative measures of
dose-response. In Canada and the USA, for example, there is,
currently, increasing reliance on specification of the margin between
potency in the experimental range and exposure as the measure of risk
for carcinogens (Health Canada, 1994; US EPA, 1996b). In the United
Kingdom, dose-response for genotoxic carcinogens is not quantified;
instead the goal in risk management is to eliminate exposure or to
reduce levels to as low as is reasonably practical (UK DOH, 1991).
Owing to the increasing reliance on modelling in the experimental
range to characterize dose-response for tumours, which is essentially
similar to the benchmark dose being used increasingly to characterize
dose-response for non-neoplastic effects, approaches to quantitative
risk estimation for carcinogenic and non-neoplastic effects are
converging.
4.2 Considerations in dose-response assessment
4.2.1 Introduction
In considering toxic effects at various dose levels, the dose range of
interest is generally the low-dose range, since it usually reflects
the human exposure situation. Often, however, data on dose-response
are available for higher doses only, and are often derived from animal
experiments only. Therefore, the uncertainty in the dose-response
assessment is larger than the uncertainty in hazard identification, as
it requires extrapolation both from animal to human and from high-dose
to low-dose levels. In certain instances, a distinction is made
between response and effect, with a response being quantal and counted
(e.g., the incidence of a tumour) and an effect being graded and
measured (e.g., relative liver weight).
4.2.2 Inter- and intra-species considerations
4.2.2.1 Introduction
The strains and species of laboratory animals exposed in toxicity
studies have been selected to show minimum inter-individual
variability. In contrast to laboratory animals, humans represent a
very heterogeneous population with both genetic and acquired
diversity.
Therefore, two principal areas are considered when interpreting
data on toxicity acquired in animal species in relation to human risk:
a) Inter-species consideration: comparison of the data for animals
with a representative healthy human. Species differences result
from metabolic, functional and structural variations.
b) Intra-species or inter-individual consideration: comparison of
the representative healthy human with the range of variability
present within the human population in relation to the relevant
parameter(s).
For each of these areas, there are two aspects to be considered
in assessing risk, i.e. toxicokinetics (the delivery of the compound
to the site of action) and toxicodynamics (the inherent sensitivity of
the site of action to the chemical). Any approach that allows for the
incorporation of adequate data on toxicokinetic or toxicodynamic
differences between test animal and humans, or between different
humans, will increase the scientific validity of risk assessment.
Sources of inter-species and inter-individual variations in
toxicokinetics include differences in anatomy (e.g., gastrointestinal
structure and function), physiological function (e.g., cardiac output,
renal and hepatic blood, glomerular filtration rate and gastric pH),
and biochemical differences in, for example, enzymes involved in
xenobiotic metabolism. Sources of inter-species and inter-individual
differences in toxicodynamics (or inherent sensitivity) also include
anatomy. For example, the effect may occur in an organ of questionable
relevance to humans, such as the rodent forestomach. Physiological
differences, such as the hormonal control of the target organ, and
biochemical differences, e.g., species differences in key biochemical
components such as alpha2u-globulin, may also play a role (Flamm &
Lehman-McKeeman, 1991).
In some cases, it may be possible to conclude that effects
detected in animals are unlikely to be relevant to humans. In other
cases, there may be data to indicate that humans are likely to be more
or less sensitive than animal species; this information is important
for consideration in selection of critical effects.
If compound-specific toxicokinetic data are introduced into risk
assessment, then it is essential that these are related to the
species, protocol and active chemical entity (e.g., parent compound or
metabolite) involved in the toxicity that is the basis for the hazard
identification (Monro, 1990, 1993; Renwick, 1993a).
4.2.2.2 Species differences
Metabolism and structural/functional variations are both
important determinants of species differences. Common areas of
metabolic variation between species are digestive tract enzymes,
levels of circulating enzymes, liver enzymes and detoxification
processes.
In extrapolating between species, three aspects need to be
considered: the first relates to differences in body size, which
requires dose normalization or scaling (often done by expressing the
dose per kg body weight). The second relates to differences in
toxicokinetics, particularly bioactivation and/or detoxification
processes. The third aspect concerns the nature and severity of the
target for toxicity. Inter-species normalization (or scaling) is
generally based on physical characteristics (e.g., body weight, body
surface area), although occasionally it is based on caloric demand or,
where there are data in four species, multiple species regression.
When clearance of the parent substance is limited by enzyme
activity rather than blood flow or when metabolites are the toxic
agents, more sophisticated physiologically based pharmacokinetic
models are more appropriate, provided that adequate data are
available. Currently, such data are available for only a small number
of substances.
4.2.2.3 Human variability
Although data from animal studies may provide limited information
on inter-individual variability within the test species, it is the
greater potential variability in the human population that must be
addressed in risk assessment. Sources of inter-individual variability
in human populations include, for example, variations in genetic
composition, nutrition, disease state and lifestyle.
Inter-individual variability may occur in both the toxicokinetics
of the chemical and the sensitivity of the target for toxicity.
4.3 Non-neoplastic (threshold) effects
Although specific aspects vary, comparable schemes have been
developed by various national and international agencies and
organizations to derive levels of exposure considered to present
minimal or no risk for non-neoplastic effects to the general
population. These include: Reference Dose/Concentrations (US
Environmental Protection Agency), Tolerable Daily
Intakes/Concentrations (Health Canada), Minimal Risk Levels (US
ATSDR), Tolerable/Acceptable Daily Intakes (IPCS, 1987a,b, 1990a,b,
1994). In evaluating dose-response for non-neoplastic effects, the
European Union does not derive tolerable intakes; instead effect
levels are compared to estimated exposures ("margin of safety").
In the case of substances for which the critical effect is not
carcinogenicity, it is generally assumed that there is a level of
exposure below which the probability for an adverse effect to occur is
minimal, if not zero (i.e. a threshold). The mechanism underlying this
assumption is that multiple cells (or cell components) must be
irreversibly injured before an adverse effect becomes evident, and
that cellular defence and repair mechanisms are overwhelmed by the
rate at which injury occurs.
4.3.1 Characterization of threshold
For toxic effects, other than heritable mutations and genotoxic
carcinogenicity, considered to have a threshold, i.e. a dose below
which there would be no detectable effect, a number of different
estimates may be used as an approximation of the biological threshold.
4.3.1.1 No-observed-adverse-effect level (NOAEL)
This is a simple estimate of the highest dose in which the
incidence of a toxic effect or change in target organ weight,
histopathology etc., was not significantly different from the
untreated group (from a statistical and biological assessment). It is
based on toxic effects of functional importance or pathological
significance rather than adaptive responses, and is defined as the
highest observed dose or concentration of a substance at which there
is no detectable adverse alteration of morphology, functional
capacity, growth, development or life span of the target (IPCS, 1994).
The NOAEL will depend on the sensitivity of the methods used, the
sizes of the exposed groups and the differences between estimated
exposures or doses. The NOAEL is an observed value which does not take
into account the nature or steepness of the dose-response curve.
In consequence, the NOAEL is not the same as the biological
threshold and may either underestimate or overestimate the true
no-effect level. Though such limitations are recognized and have been
the basis for criticism of the use of the NOAEL (Leisenring & Ryan,
1992; Calabrese & Baldwin, 1994), dose-response relationships are
often so poorly characterized that the NOAEL or LOAEL is the only
quantitative value available as the basis for characterization of
dose-response.
4.3.1.2 Benchmark dose/concentration
This is an alternative method of defining the lower end of the
dose-response curve in the area of the observed threshold
(Crump, 1984). The benchmark dose is the effective dose (or its lower
confidence limit) that produces a certain increase in incidence above
control levels (e.g., 1% or 5% of the maximum toxic response). The
benchmark dose is derived by modelling the data in the observed range
and selecting the point on the curve (or its upper confidence limit)
corresponding to a specified increase in the incidence of an effect.
Any model that fits the empirical data well is likely to provide a
reasonable estimate of the benchmark dose, and choice of the model may
not be critical since estimation is within the observed dose range.
The advantages of the benchmark dose are that it takes into account
the slope of the dose-response curve, the size of the study groups and
the variability in the data. It should be recognized that unless there
are a sufficient number of dose levels at which effects have been
observed, the benchmark dose/concentration offers little advantage
over effect levels as an approximation of the biological threshold.
Statistical modelling of continuous data as a basis for developing
benchmark doses/concentrations is also currently problematic.
4.3.1.3 Lowest-observed-adverse-effect level (LOAEL)
In some studies, there is a significant effect compared to
controls in the lowest dose group. In such cases, there is no NOAEL
and an alternative approach must be adopted. These include estimation
of a benchmark dose or threshold estimate (if the dose-response data
approach zero response) or application of an additional uncertainty
factor.
4.3.2 Uncertainty factors
In deriving tolerable intakes (or RFDs or ADIs), the N(L)OAEL or
benchmark dose/concentrations are divided by uncertainty factors to
account for variabilities and uncertainties. Principal factors applied
relate to extrapolation from animal studies to the human situation and
to inter-individual variability within the response for the human
population. Traditionally, default factors of 10 have been applied to
account for each of these variations. Additional uncertainty factors
have been applied to account for the inadequacy of the database, for
extrapolation from subchronic to chronic exposure and from LOAEL to
NOAEL, and for the severity of a given effect.
Knowledge of actual inter-species differences and
inter-individual variability in the biokinetic behaviour of a given
compound (toxicokinetics) and its target organ (toxicodynamics) would
enable the development of full biologically based dose-response models
or physiologically based pharmacokinetic models. In the absence of
full biological understanding, several approaches have been developed
to incorporate as much scientific information as possible in the
development and application of uncertainty factors. Indeed, a formal
approach to the development of data-derived uncertainty factors has
been developed by Renwick (1993a,b) and proposed by IPCS (IPCS, 1994).
It is presented here as an example of a flexible but structured
approach to the selection of uncertainty factors which reflects the
nature and extent of the database (Lewis, et al., 1990; Renwick,
1993b).
The scheme retains the two 10-fold default uncertainty factors
(for inter-species and inter-individual variation) as the cornerstone
of the structure, in the absence of specific and relevant data on
toxicokinetics or mechanism of action (Renwick, 1993a). However, it
allows for the division of the two default uncertainty factors (for
inter- and intra-species variation) to account for toxicokinetics and
toxicodynamics. The default components of these two factors can then
be replaced by actual quantitative data, when available. This reduces
the extent of uncertainty by allowing the incorporation of appropriate
data on the compound of interest in one or both of these aspects,
where they exist (Fig. 1). There would be very few databases in which
adequate information was available to account quantitatively for both
aspects of either inter-species or of inter-individual differences.
Incorporation of data on one aspect only (e.g., inter-species
toxicokinetics) requires the use of a default factor for the
uncertainty associated with the remaining undefined aspect (e.g.,
inter-species toxicodynamics).
Uncertainty factors often address:
a) Nature of toxicity
Some bodies, e.g., the FAO/WHO Joint Meeting on Pesticide
Residues (JMPR), have used an additional "safety factor" in cases
where the NOAEL is derived for a critical effect that is a severe and
irreversible phenomenon, such as teratogenicity or non-genotoxic
carcinogenicity, especially if the dose-response relationship is
shallow (IPCS, 1987a,b, 1990a,b). This additional factor (of up to 10)
has been applied in such cases to provide a greater margin between the
intake/exposure of any particularly susceptible humans and the
dose-response curve for such toxicity demonstrable in animals.
However, for other types of toxic effect, for example, changes in
organ weight or histopathology, a value of 1 (no further correction)
would be appropriate.
b) Adequacy of the database
A minimum dataset that is considered adequate for risk assessment
is generally established. This will vary according to the purpose of
the assessment (e.g., screening level or full). Additional
deficiencies in a toxicity database that increase the uncertainty of
the extrapolation process have also been recognized by the use of an
additional uncertainty factor. A value of 1 would be applied to an
appropriate and complete database, but a higher factor would be
considered necessary for barely adequate databases.
c) LOAEL to NOAEL extrapolation
In situations where a NOAEL has not been achieved but data are of
sufficient quality to be the basis of the risk assessment, then an
extra uncertainty factor may be applied (Dourson & Stara, 1983). The
magnitude of this factor (e.g., 3 or 10) should be based on the
dose-response data.
d) Inter-species extrapolation
The inter-species uncertainty factor is not necessary if the NOAEL or
risk assessment is based on human data. Where an assessment is based
on data in animals, however, and in situations where there are
appropriate compound-specific toxicokinetic and/or toxicodynamic data,
the relevant default uncertainty factor for inter-species variation
would be replaced by the data-derived factor (Renwick, 1993b). Data on
physiologically based pharmacokinetic (PBPK) modelling should be
included wherever possible; however, such information is available
currently for only a small number of substances. If a data-derived
factor is introduced, then the commonly used 10-fold factor would be
replaced by the product of that factor and the remaining default
factor.
The composite default value of 10 has been criticized as
inadequate, for example, to allow for metabolic processes in mice
which can be related to body surface area (Calabrese et al., 1992);
the introduction of data-derived uncertainty factors would allow the
logical future development of more appropriate species specific
defaults.
e) Inter-individual variability in humans
In situations where appropriate toxicokinetic and toxicodynamic
data exist for a particular compound in humans, then the relevant
uncertainty factor should be replaced by the data-derived factor
(Renwick, 1993b). Data on PBPK modelling may also be able to
contribute to this assessment. If a data-derived factor is introduced,
then the commonly used 10-fold factor would be replaced by the product
of the data-derived factor and the remaining default factor.
Although the 10-fold default uncertainty factor is reasonable for
most cases (Dourson & Stara, 1983), it has been criticised as
inadequate for human variability especially when genetically
determined differences in a bioactivation process may be involved
(Calabrese, 1985; Goldstein, 1990). This concern reinforces the
importance of using an approach that allows the incorporation of data
on human variability in either toxicokinetics of the compound or the
sensitivity to its mechanism of action.
In addition to approaches aimed at incorporating as much
biological data as possible in the derivation of uncertainty factors,
probabilistic approaches have been investigated for the
characterization of uncertainty (Baird et al., 1996; Price et al.,
1997). Distributions can be developed on the basis of empirical
relationships observed for, for example, variations between LOAELs and
NOAELs and effect levels in subchronic versus chronic studies. Monte
Carlo techniques can be used to integrate probabilities for the
various areas of uncertainty.
4.4 Quantitative risk assessment for neoplastic (non-threshold)
effects
4.4.1 Introduction
A number of approaches have been adopted for characterization of
dose-response in the assessment of genotoxic neoplastic effects,
including quantitative extrapolation by mathematical modelling of the
dose-response curve to estimate the risk at likely human intakes or
exposures (low-dose risk extrapolation). Traditionally, where
dose-response has been extrapolated into the low-dose range, this has
been accomplished by the use of the linearized Armitage-Doll
multi-stage model. Dose-response may also be estimated in a two-step
process by straight linear extrapolation into the low-dose range from
a modelled point on the dose-response curve. Other measures of
dose-response include estimation of carcinogenic potency in the
experimental range and division of effect levels by a margin of
protection. In more recently developed biological models, different
stages in the process of carcinogenesis have been incorporated and
time to tumour has been taken into account (Moolgavkar et al., 1988),
although currently data are sufficient for application in only a
limited number of cases. In some cases where data permit, the dose
delivered to the target tissue has been incorporated into the
dose-response analysis (PBPK modelling) (IPCS, 1993).
In the same way as approaches adopted for non-neoplastic
(threshold) effects, there are increasingly attempts to incorporate
more of the scientific data in adopted approaches. For example, the
proposed cancer guidelines issued by the US EPA (1996b), updating the
previous guidelines (US EPA, 1986a), put emphasis on the full
integration of mechanistic information and dose-response data.
Depending on the mode of action, linear extrapolation into the
low-dose range or, alternatively, a margin of exposure would be
presented. The adequacy of the latter approach must be judged by
criteria similar to those used in developing tolerable
intakes/exposures for non-cancer effects.
4.4.2 Linear extrapolation
Where data on the mechanism of tumour induction are not
available, as a default, risks are often linearly extrapolated into
the low-dose range. Previously (e.g., US EPA, 1986a) the linearized
multistage model was widely adopted for such extrapolations for data
from studies in animal species, whereas data from epidemiological
studies were generally modelled using a multistage model with a linear
term. More recently, curve fitting within the range of observation
with extrapolation from the lower 95% confidence limits on a dose
associated with a 10% extra risk (the LED10) has been recommended (US
EPA, 1996a). Linear extrapolation is considered to be appropriate if
available evidence supports a mode of action that is anticipated to be
linear or, as a science policy default, there is no evidence of either
linearity or non-linearity.
Other approaches to linear extrapolation have been described in
the literature. Gross et al. (1970) suggested a method based on
discarding data at the upper end of the dose range until a linear
model provides an adequate description of the remaining data. Van
Ryzin (1980) suggested the use of any model that fits the data
reasonably well to estimate the dose producing an excess risk of 1%,
and then using simple linear extrapolation to lower doses. Gaylor &
Kodell (1980) proposed fitting a model to the available data and then
using linear extrapolation below the lowest dose at which observations
were taken. Since the estimates at the lower doses might be unduly
influenced by the choice of the model used in the experimental dose
range, Farmer et al. (1982) suggested linear extrapolation below the
lowest dose or the dose corresponding to an estimated risk of 1%,
whichever was larger.
A model-free procedure based on linear extrapolation below the
lowest dose showing an increased (not necessarily statistically
significant) risk has been proposed by Krewski et al. (1984, 1986)
using linear extrapolation from all doses for which there were no
statistically significant increases in tumour incidence above the
baseline level, and selecting the smallest slope for low-dose risk
estimation. Similarly, Gaylor (1987) considered the smallest slope
obtained from all the possible combinations of data from the doses
where the lowest dose was in the convex portion of the dose-response
curve. In both cases, upper confidence limits on the slopes were used.
A number of arguments have been advanced in support of the
hypothesis of low-dose linearity (Krewski et al., 1986; Murdoch et
al., 1987). For example, the class of additive background models
considered by Crump et al. (1976) predicts low-dose linearity provided
only that the response increases smoothly with dose. However, it is
difficult to prove or disprove low-dose linearity experimentally even
in bioassays involving extremely large numbers of animals (Gaylor et
al., 1985). Indeed, dose-response curves for different types of
tumours in mice following exposure to 2-acetylaminofluorene (2-AAF) in
an ED01 study varied considerably.
Often, linear extrapolation is criticized as being too
conservative. For example, Bailar et al. (1988) demonstrated that a
significant fraction of bioassays conducted for the National
Toxicology Program indicate that, at high experimental doses, observed
response rates are higher than those predicted by a linear model. They
argue that, at low doses, the one-hit model may thus not be
conservative in some cases. However, these observations are not
necessarily inconsistent since, at low doses, the linear term
predominates. Crump et al. (1976), Peto (1978) and Hoel (1980) argue
that low-dose linearity occurs when substances augment existing
carcinogenic processes. The formation of DNA adducts, which may be
predictive of certain tumours induced by genotoxic carcinogens, has
often been observed to be linear at very low doses (Poirier & Beland,
1987). Based on these considerations, it is unclear whether an
estimate based on a linear approximation over- or under-estimates the
true risk.
The outcome of low-dose extrapolation is the resulting lifetime
cancer risk associated with estimated exposure for a particular
population. In view of the considerable uncertainties in extrapolating
results over several orders of magnitude, in the absence of
information on mechanisms of tumour induction, specification of risks
in terms of predicted incidence or numbers of excess deaths per unit
of the population implies a degree of precision that is considered
misleading by some (e.g., Health Canada, 1994).
4.4.3 Estimation of potency in the experimental range
For assessment of Priority Substances under the Canadian
Environmental Protection Act (CEPA), e.g., for genotoxic carcinogens,
a Tumorigenic Dose or Concentration05 (TD5) has been adopted as the
measure of dose-response (Health Canada, 1994; Meek et al., 1994). It
is the intake or concentration associated with a 5% incidence of
tumours in experimental studies on animals or epidemiological studies
on human populations. It serves as the basis for development of an
Exposure/Potency Index (EPI) which is the estimated daily human intake
or exposure divided by the TD5. A calculated EPI of 10-6 represents
a one million fold difference between human exposure and that at the
lower end of the dose-response curve, on which the estimate of potency
is based.
Any model that fits the empirical data well is likely to provide
a reasonable estimate of the TD5. Choice of the model may not be
critical since estimation is within the observed dose range, thereby
avoiding the numerous uncertainties associated with low-dose
extrapolation. Wherever possible, and if considered appropriate,
information on pharmacokinetics, metabolism and mechanisms of
carcinogenicity and mutagenicity is incorporated into the quantitative
estimates of potency derived particularly from studies in animals (to
provide relevant scaling of potency for human populations). The value
of 5% is arbitrary; selection of another value would not affect the
relative potencies for each of a range of compounds. Indeed, in the
literature, others have proposed the TD50 (Peto et al., 1984) and the
TD25 (Allen et al., 1988; Dybing & Huitfeldt, 1992; Dybing et al.,
1997). The Committee on Carcinogenicity of Chemicals in Food, Consumer
Products and the Environment in the United Kingdom has concluded that
the TD50 is the most practical quantitative estimate of carcinogenic
potency for the ranking of genotoxic carcinogens (UK DOH, 1995).
If there is no evidence for linearity, and there is sufficient
evidence to support an assumption of non-linearity for the
carcinogenic response, US EPA (1996a) recommends estimation of a
margin of exposure, which is the LED10 or other point of departure
divided by the environmental exposure of interest. It should be noted,
however, that this contrasts with the approach in Canada and Europe,
where characterization of potency within the experimental range is
considered appropriate for carcinogens, whereas the default in the USA
is linear. Indeed the Committee on Carcinogenicity of Chemicals in
Food, Consumer Products and the Environment in the United Kingdom
concluded that potency indices are not appropriate for the ranking of
non-genotoxic carcinogens. Rather for non-genotoxic compounds, the
emphasis should be on understanding mechanisms and their relevance to
humans.
4.4.4 Two-stage clonal expansion model
This approach is based on the two-stage model of carcinogenesis,
in which it is hypothesized that chemical carcinogenesis occurs in two
steps. Cells are initiated following the occurrence of genetic damage
in one or more cells in the target tissue. Such initiated cells may
then undergo malignant transformation to give rise to a cancerous
lesion. The rate of occurrence of such lesions may be increased by
subsequent exposure to a promoter, which serves to increase the pool
of initiated cells through mechanisms that result in clonal expansion.
Mathematical formulations of this process have been presented by
Moolgavkar et al. (1988) and Chen & Farland (1991). This stochastic
birth-death-mutation model assumes that two mutations, each occurring
at the time of cell division, are necessary for a normal cell to
become malignant. Initiating activity may be quantified in terms of
the rate of occurrence of the first mutation. The overall rate of
occurrence of the second mutation describes progression to a fully
differentiated cancerous lesion. Promotional activity is measured by
the difference in the birth and death rates of initiated cells. In the
absence of promotional effects and variability in the pool of normal
cells, the two-stage birth-death-mutation model reduces to the
classical two-stage model.
It should be noted, however, that there are currently few cases
where data are sufficient to permit application of such a model.
4.4.5 Proportional analyses - carcinogenic and non-neoplastic effects
There have been several investigations of the possibility of
predicting potency for particular types of toxicity from data on other
types of toxicity, including work by Tennant et al. (1987), Portier
(1988), Travis et al. (1990a,b, 1991), Zeiger et al. (1990) and
Haseman & Clark (1990). Such approaches have been necessary due, for
example, to the high cost and degree of difficulty of long-term or
carcinogenic bioassays. However, it is important to note that
correlations between potencies for different types of effects may be
artificially strengthened by dose selection (e.g., the top dose in
carcinogenic bioassays is often the maximum tolerated dose, selected
to elicit small reductions in body weight).
5. EXPOSURE ASSESSMENT
The objective of exposure assessment is to determine the nature
and extent of contact with chemical substances experienced or
anticipated under different conditions. Approaches for assessing
exposure and characterizing uncertainties/variability in resulting
estimates presented here are derived primarily from the Exposure
Assessment Guidelines (US EPA, 1986b, 1992).
5.1 Definition of exposure and related terms
Although there is reasonable agreement that human exposure means
contact with the chemical or agent (Allaby, 1983; Environ, 1988;
Hodgson et al., 1988), there has not yet been widespread agreement as
to whether this means contact with (a) the visible exterior of the
person (skin and openings into the body such as mouth and nostrils),
or (b) the so-called exchange boundaries where absorption takes place
(skin, lung, gastrointestinal tract). These different definitions have
led to some ambiguity in the use of terms and units for quantifying
exposure. In 1992, The US EPA published Guidelines (US EPA, 1992)
defining exposure as taking place at the visible external boundary, as
in (a) above.
Under this definition, it is helpful to think of the human body
as having a hypothetical outer boundary separating inside the body
from outside the body. This outer boundary of the body is the skin and
the openings into the body such as the mouth, the nostrils, and
punctures and lesions in the skin. Exposure to a chemical is the
contact of that chemical with the outer boundary. An exposure
assessment is the quantitative or qualitative evaluation of that
contact, which includes consideration of the intensity, frequency and
duration of contact, the route of exposure (e.g., dermal, oral or
respiratory), rates (chemical intake or uptake rates), the resulting
amount that actually crosses the boundary (a dose), and the amount
absorbed (internal dose). The Commission of the European Communities
(EC, 1996) presented a similar definition for exposure assessment: the
determination of the emissions, pathways and rates of movement of a
substance and its transformation or degradation, in order to estimate
the concentrations/ doses to which human populations or environmental
spheres (water, soil and air) are or may be exposed.
Depending on the purpose of an exposure assessment, the numerical
output may be an estimate of the intensity, rate, duration and
frequency of contact exposure or dose (the resulting amount that
actually crosses the boundary). For risk assessments based on
dose-response relationships, the output usually includes an estimate
of dose.
5.2 Exposure and dose
Most of the time, the chemical coming into contact with the outer
boundary of the body is contained in air, water, soil, a product or a
transport or carrier medium; the chemical concentration in these media
at the point of contact is the concentration, on which exposure
estimates are based. Exposure over a period of time can be represented
by a time-dependent profile of the exposure concentration. The area
under the curve of this profile is the magnitude of the exposure, in
concentration-time units (Lioy, 1990; US NRC, 1990):
where E is the magnitude of exposure, C(t) is the exposure
concentration as a function of time, and t is time, t2-t1 being the
exposure duration (ED). If ED is a continuous period of time (e.g., a
day, week, year, etc.), then C(t) may be zero during part of this
time. Integrated exposures are done typically for a single individual,
a specific chemical, and a particular pathway or exposure route over a
given time period.
The integrated exposures for a number of different individuals (a
population or population segment, for example), may then be displayed
in a histogram or curve (usually, with integrated exposure increasing
along the abscissa or x-axis, and the number of individuals at that
integrated exposure increasing along the ordinate or y-axis). This
histogram or curve is a presentation of an exposure distribution for
that population or population segment.
Applied dose is the amount of a chemical at the absorption
barrier (skin, lung, gastrointestinal tract) available for absorption.
Usually, it is very difficult to measure the applied dose directly, as
many of the absorption barriers are internal to the human and are not
localized in such a way as to make measurement easy. An approximation
of applied dose can be made, however, using the concept of potential
dose (Lioy, 1990; US NRC, 1990). Potential dose is simply the amount
of the chemical ingested, inhaled or in material applied to the skin.
For the dermal route, potential dose is the amount of chemical
applied or the amount of chemical in the medium applied, e.g., as a
small amount of particulate deposited on the skin. It should be noted
that as not all of the chemical in the particulate is in contact with
the skin, this differs from exposure (the concentration in the
particulate multiplied by the time of contact) and applied dose (the
amount in the layer actually touching the skin).
The applied dose, or the amount that reaches the exchange
boundaries of the skin, lung or gastrointestinal tract, may often be
less than the potential dose if the material is only partly
bioavailable. This will depend, for example, on the form in which the
compound is administered (e.g., neat or in vehicle on skin). Where
data on bioavailability are known, adjustments to the potential dose
to convert it to applied dose and internal dose may be made. For
example, chemicals reaching their target through the gastrointestinal
tract can be metabolized in the anaerobic conditions of the lower
colon prior to absorption. Bioavailability via various routes of
exposure may also vary. For example, intestinal absorption results in
a first pass effect that may lead to metabolic detoxication or
activation by the liver.
The amount of a chemical that has been absorbed and is available
for interaction with biologically significant receptors is called the
internal dose. Once absorbed, the chemical can undergo metabolism,
storage, excretion or transport within the body. The amount
transported to an individual organ, tissue or fluid of interest is
termed the delivered dose. The delivered dose may be only a small part
of the total internal dose. The biologically effective dose, or the
amount that actually reaches cells, sites or membranes where adverse
effects occur (US NRC, 1990), may only be a part of the delivered
dose. Currently, most risk assessments dealing with environmental
chemicals (as opposed to pharmaceutical assessments) use dose-response
relationships based on potential (administered) dose or internal dose,
since the pharmacokinetics necessary to base relationships on the
delivered dose or biologically effective doses are not available. This
may change in the future, as more becomes known about the
pharmacokinetics of environmental chemicals.
Doses are often presented as dose rates, or the amount of a
chemical dose (applied or internal) per unit time (e.g., mg/day), for
instance, as dose rates on a per-unit-body-weight basis (e.g., mg/kg
per day).
The general equation for potential dose for intake processes,
e.g., inhalation and ingestion, is simply the integration of the
chemical intake rate (concentration of the chemical in the medium
multiplied by the intake rate of the medium, C x IR) over time:
where Dpot is potential dose and IR(t) is the ingestion or inhalation
rate.
The quantity t2-t1, as before, represents the period of time
over which exposure is being examined, or the exposure duration (ED).
The exposure duration may contain times where the chemical is in
contact with the person, and also times when C(t) is zero. Contact
time represents the actual time period where the chemical is in
contact with the person. For cases such as ingestion, where actual
contact with food or water is intermittent, and consequently the
actual contact time may be small, the intake rate is usually expressed
in terms of a frequency of events (e.g., 8 glasses of water consumed
per day) multiplied by the intake per event (e.g., 250 ml of water per
glass of water consumed). Intermittent air exposures (e.g., 8 h
exposed/day multiplied by one cubic metre of air inhaled/hour) can
also be expressed easily using exposure duration rather than contact
time. Hereafter, the term exposure duration will be used in the
examples below to refer to the term t2-t1, since it occurs
frequently in exposure assessments and it is often easier to use.
Equation 2 can also be expressed in discrete form as a summation
of the doses received during various events i:
where EDi is the exposure duration for event i. If C and IR are
nearly constant (which is a good approximation if the contact time is
very short), equation 4-3 becomes:
_
where ED is the sum of the exposure durations for all events, and C
__
and IR are the average values for these parameters. Equation 4 will
not necessarily hold in cases where C and IR vary considerably. In
those cases, equation 3 can be used if the exposure can be broken out
into segments where C and IR are approximately constant. If even this
condition cannot be met, equation 2 may be used.
For risk assessments, estimates of dose should be expressed in a
manner that can be compared with available dose-response data.
Frequently, dose-response relationships are based on potential dose
(called administered dose in animal studies), although dose-response
relationships are sometimes based on internal dose.
Doses may be expressed in several different ways. Solving
equations 2, 3 or 4 for example, gives a total dose accumulated over
the time in question. The dose per unit time is the dose rate, which
has units of mass/time (e.g., mg/day). Because intake and uptake can
vary, dose rate is not necessarily constant. An average dose rate over
a period of time is a useful number for many risk assessments.
Exposure assessments take into account the time scale related to
the biological response studied, unless the assessment is intended to
provide data on the range of biological responses (US NRC, 1990). For
developmental toxicity effects, a single short-term exposure can cause
the adverse health effects. For many non-cancer effects, risk
assessments consider the period of time over which the exposure
occurred, and often, if there are no excursions in exposure that would
lead to acute effects, average exposures or doses over the period of
exposure are sufficient for the assessment. These averages are often
in the form of average daily doses (ADDs) expressed, for example, in
mg/kg body weight per day.
An ADD can be calculated from equation 2 by averaging Dpot over
body weight and an averaging time, provided the dosing pattern is
known so that the integral can be solved. It is unusual to have such
data for human exposure and intake over extended periods of time, so
some simplifying assumptions are commonly used. Using equation 4
instead of 2 or 3 involves making steady-state assumptions about C and
IR, but this makes the equation for ADD easier to solve. For intake
processes, then, using equation 4, this becomes:
where ADDpot is the average daily potential dose, BW is body weight,
and AT is the time period over which the dose is averaged (converted
_
to days). As with equation 4, the exposure concentration C is best
expressed as an estimate of the arithmetic mean regardless of the
distribution of the data. Again, using average values for C and IR in
equation 5 assumes that C and IR are approximately constant.
For effects such as cancer, where the biological response is
usually described in terms of lifetime probabilities, even though
exposure does not occur over the entire lifetime, doses are often
presented as lifetime average daily doses (LADDs). The LADD takes the
form of equation 6, with lifetime (LT) replacing the averaging time
(AT):
5.3 Approaches to quantification of exposure
Exposure (or dose) is assessed generally by one of the following
approaches:
a) The exposure can be measured at the point of contact (the outer
boundary of the body) while it is taking place, measuring both
exposure concentration and time of contact and integrating them
(point-of-contact or personal measurement);
b) The exposure can be estimated by separately evaluating the
exposure concentration and the time of contact, then combining
this information (scenario evaluation);
c) The exposure can be estimated from dose, which in turn can be
reconstructed through internal indicators (biomarkers, body
burden, excretion levels, etc.) after the exposure has taken
place (reconstruction).
These three approaches to quantification of exposure (or dose)
are independent, as each is based on different data. This offers the
opportunity of checking the accuracy of exposure estimated by one
approach through use of an independent approach, where data permit.
The independence of the three methods is a useful concept in verifying
or validating results. Each of the three has strengths and weaknesses;
using them in combination can considerably strengthen the credibility
of an exposure or risk assessment.
5.3.1 Measurement at point of contact (personal monitoring)
Point-of-contact exposure measurement evaluates the exposure as
it occurs, by measuring the chemical concentrations at the interface
between the person and the environment as a function of time,
resulting in an exposure profile. The best known example of the
point-of-contact measurement is the radiation dosimeter. This small
badge-like device measures exposure to radiation as it occurs and
provides an integrated estimate of exposure for the period of time
over which the measurement has been taken. Another example is the
Total Exposure Assessment Methodology (TEAM) studies (US EPA, 1987a)
conducted by the EPA and similar multimedia exposure studies in Canada
(Otson et al., 1996). In the TEAM studies, a small pump with a
collector and absorbent was attached to a person's clothing to measure
his or her exposure to airborne solvents or other pollutants as it
occurred. A third example is the carbon monoxide (CO) point-of-contact
measurement studies where subjects carried a small CO measuring device
for several days (US EPA, 1984). Dermal patch studies and duplicate
meal studies are also point-of-contact measurement studies. In all of
these examples, the measurements are taken at the interface between
the person and the environment while exposure is occurring. Use of
these data for estimating exposures or doses for periods that differ
from those for which the data are collected (e.g., for estimates of
lifetime exposures) will require some assumptions.
The strength of this method is that it measures exposure
directly, and providing that the measurement devices are accurate, is
likely to give the most accurate exposure value for the period of time
over which the measurement was taken. It is often expensive, however,
and measurement devices and techniques do