Dose-Response Curves and Toxicity Testing

**DOSE - RESPONSE CURVES AND TOXICITY TESTING** **6.1 INTRODUCTION TO DOSE-RESPONSE CURVES** 'Dose-response' and 'dose-response relationship' describe the effect on an organism caused by differing levels of exposure (or dose) to a stressor (usually a chemical). The dose-response curve is a crucial...

**DOSE - RESPONSE CURVES AND TOXICITY TESTING** **6.1 INTRODUCTION TO DOSE-RESPONSE CURVES** 'Dose-response' and 'dose-response relationship' describe the effect on an organism caused by differing levels of exposure (or dose) to a stressor (usually a chemical). The dose-response curve is a crucial tool to understand the levels at which chemicals, drugs or pollutants begin to exert harmful effects and the degree of harm to be expected at various levels. Data relating to the amount of drug or pollutant can be plotted on a graph against the response of the organism. The resulting curve can then be used to show a number of points including: - the 'no-effect level', where no effect occurs or no effect is detectable - the threshold-dose of the 'stressor', the level at which the effect starts to occur and - the levels at which the effect occurs in a set percentage or all of the organisms The dose response of a population is that proportion of the population which experiences a specific effect following exposure of the total population to specified harmful contaminant. The correlation of the response with estimates of the dose provides a dose-response relation, which is normally expressed as a graph, with percentage of population affected on the y axis and estimated dose on the x axis (see Figure 6.1). There are a number of terms that are used to describe particular points on the dose-response curve including: - LD50 -- Lethal Dose, 50% - the dose that kills 50% of the test animals. The units used are in milligrams of substance per kilogram body weight of the test animal. - LC50 -- Lethal Concentration, 50% - the concentration of a gas or vapour that kills 50% of the test animals. 58 - TD50 - Toxic dose, 50% -- the dose at which 50% of the test animals show a particular effect - TC50 - Toxic concentration, 50% -- the concentration of a gas or vapour at which 50% of the test animals show a particular effect **6.1.1 No Observed Adverse Effect Level** The "no observed adverse effect level" (NOAEL) is an experimentally derived value and reflects the dose at which no adverse effects were observed in the studies available. The robustness of the NOAEL depends on many factors including the type(s) of study and its design (number of animals, experimental protocols etc). Effects, particularly adverse effects, are generally manifestations of the change in an organ and particularly the cells of the organ. In toxicology, the NOAEL is specifically the highest tested dose or concentration of a substance at which no adverse effect is observed in the exposed test species (usually animals or cells). The NOAEL plays an important role in the risk assessment of the substance. Another important toxicological concept is "lowest observed adverse effect level" (LOAEL) or the lowest dose or concentration that causes any observed adverse effect. Thus by definition the NOAEL is less than the LOAEL. Other terms sometimes encountered include in toxicity testing include LDLO and LCLO which are the lowest doses or concentrations at which death occurs and TDLO and TCLO which are the lowest doses or concentrations at which a test animal shows a particular effect. These terms are effectively the same as the lowest observed adverse effect level (LOAEL). As these determinations of exposure and effect have generally been established in species other than humans, various safety or uncertainty 59 factors are applied before this data is used in the establishment of workplace exposure standards. In many instances factors of 10 times have been used to take into account inter-species variation and a further factor of 10 times to take into account variability within humans. However, care must be taken with this type of approach as there is limited scientific basis for these factors and specialist toxicological advice should always be sought. **6.1.2 Threshold** The term \"threshold\" is used in toxicology to describe the dividing line between no-effect and effect levels of exposure. It may be considered as the maximum quantity of a chemical that produces no effect or the minimum quantity that does produce an effect. Every effect produced by a chemical, whether it is beneficial, indifferent, or harmful, has a threshold. For a given population, as illustrated by the dose response relationship (Figure 6.1), it is clear that thresholds exist because it can be determined experimentally that certain low levels of exposure will produce no detectable effect, and that as the dosage is increased the effect appears. The precise threshold for a given effect can, and usually does, vary within certain limits with different species, as well as between individuals within a species. Indeed, within a population there will be variability -- not everybody will have the same response. It can be shown that each point on a dose-response curve represents a normal distribution of responses within the sample populations. 60 (Source: BP International -- reproduced with permission) ***Figure 6.1 Typical dose-response curve*** Since the dose-response relationship is a continuum, somewhere between the experimental no-effect and effect levels is the turning point known as the threshold. Dose-response curves typical of those plotted from data obtained in chronic toxicity experiments exist for a number of contaminants. It is very important to recognise that such a curve is drawn from only several points, one for each exposure group in the experiment. The greater the number of exposure groups, the greater the number of points, and hence, the greater the accuracy of the curve that is drawn. But without an infinite number of points, the precise shape of the dose-response curve cannot be known. The curve is interpreted as follows: with chronic exposure of increasing doses up to the threshold, no effect is detectable because some biochemical or physiological mechanism handles the chemical in a manner that prevents an effect from occurring. At the threshold, the defence mechanism is saturated, or in some manner overwhelmed, for the more susceptible individuals and the effect begins to appear. With increasing 61 doses, increasing numbers of individuals show the effect until finally a dose is reached where all of the members of the population show the effect (ceiling level). The threshold concept is of great importance to toxicologists because it permits them to make judgements about the potential risk, or lack thereof, to humans from exposure to chemicals. Another question relates to the shape of dose-response curves for carcinogens as they approach zero doses. The inability of toxicology to answer this question by experiment has given rise to a scientific controversy concerning whether or not there is a threshold (no-effect level) for carcinogenic effects. If there is no threshold, extension of the experimentally derived dose-response curve to zero effect would yield a line that would go through the origin (zero dose). If there is a threshold, the extended line would meet the abscissa at some point greater than zero dose. In regard to carcinogens, it is important to note that it is rare to have any data except for high doses, so the estimate of the shape of the dose response curve below the lowest actual data point must typically cover many orders of magnitude. If a threshold cannot be identified, then approaches to this problem vary around the world depending on national policies. For example, in the US a quantitative risk assessment approach using mathematical modelling is used, whereas in the UK a limit is established using a risk management approach (control to as low a level as is reasonably practicable). For respiratory sensitisers, the same problems of lack of information on NOAEL and dose-response relationships can exist and so a similar approach may be required as to that for non-threshold carcinogens. It is very important, as background to all considerations of the threshold, to recognise that detectable biological effects are not universally adverse. 62 What should be recognised is that in any group of test subjects there are some susceptible individuals (hypersensitive) who are affected at low concentrations of the test contaminant and there are also some highly resistant individuals (hyposensitive) who are not affected at high concentrations but there are the vast majority of "average" individuals in the middle (Figure 6.2). It is important to recognise that some hypersensitive individuals may be in a work group and that they may suffer adverse health effects at exposures below the recognised exposure standard. (Source: AIOH 2007 -- *Reproduced with permission*) ***Figure 6.2 -- Variability of Human Exposure to Dose*** For each substance, no matter how toxic, there exists a dose level called the threshold of intoxication, which the human body is capable of accepting and detoxifying without injury to itself. It is this principle that the major exposure standards used within the western world are based upon. **6.1.3 Slope of curve** Another characteristic of the shape of the dose-response curve that may be examined is the slope of the curve. Knowledge of the shape and slope of 63 the dose-response curve is extremely important in predicting the toxicity of a substance at specific dose levels. Major differences among toxicants may exist not only in the point at which the threshold is reached but also in the percent of population responding per unit change in dose Dose-response curves for some substances show a steep curve (rise rapidly from the threshold point to the ceiling level). This indicates that particular care may need to be taken to prevent excessive exposure as the undesired consequences of that exposure will occur at levels only slightly above threshold levels. On the other hand, a relatively flat slope suggests that the effect of an increase in dose is generally minimal and that there is much greater variation in the likelihood of the effect occurring in the whole exposed population. **6.2 TOXICITY TESTING** **6.2.1 Types of toxicity testing** In order to make judgements on likely risks and the appropriate measures required to mitigate against them, it is necessary to gather information on the hazardous toxicological properties of chemical substances. Clearly we are interested in the effects of substances on human health and so of course toxicological data from human exposure in principle provides the most useful information. Data are available from studies looking at populations of humans exposed, for example, in the course of their work (this is covered further in section 7). In general, though, these do not provide the broad range of information required to understand all the potential toxicological hazards of a substance. Consequently, a range of approaches have been developed to investigate the various areas of toxicological hazard; these are covered briefly below. 64 A toxicologist will use all the information available to develop a picture of the toxicity profile of a substance. A starting point will often be the physico-chemical properties such as the pH of the substance (if this is very high or low then in general it will be assumed that the substance is corrosive) and a consideration of its structure. If a substance has a structure very similar to another substance or substances (e.g. is part of a chemical group or family) then it may be possible and reasonable to compare the toxicity of the two (or across the group) where data already exist (this is called structure activity relationship, SAR, or group approaches). Usually, however, having considered these possibilities, it is likely that more information will be required and some experimental testing will be needed. The use of animals in these experiments, though in some parts of the world a controversial issue, remains the main approach for gathering information. In some areas of toxicity testing, though, there have been some significant advances in recent years to replace animal tests, reduce the numbers used or refine experiments to reduce suffering. Approaches replacing animals generally use cells or tissues in "culture" (e.g. test tubes and Petri dishes) and are known as *in vitro* (Latin for "within the glass") methods. Where animal testing is carried out (often for regulatory purposes) they are usually conducted to established international guidelines (e.g. OECD Guidelines). This is to ensure that they meet certain agreed standards and that the results are acceptable to different regulatory authorities around the world. It is by no means the case that such studies are the only source of toxicity information and toxicological research not conducted to these protocols but published in established scientific journals can contribute significantly to the database for any substance. In general, most toxicity studies are carried out in rodents, particularly special bred laboratory rats, mice, guinea pigs and rabbits. Some studies may use dogs and non-human primates, though this is generally not common for workplace chemicals. 65 The following sections give a brief overview of testing for the different types of toxicity in the following broad categories: toxicokinetics, acute toxicity, including skin and eye irritation, sensitisation, genotoxicity, repeated dose (sub-acute, sub-chronic and chronic) toxicity, reproductive and developmental toxicity and carcinogenicity. **6.2.2 Toxicokinetic studies** Although not testing for a toxic effect, the study of how a substance is absorbed, distributed, metabolised and excreted from the body often provides important information to help understand its potential to induce toxicity. These studies are known as toxicokinetic studies though the term pharmacokinetic is also sometimes used (this is more often in relation to pharmaceutical compounds). In general toxicokinetic studies involve administering a substance (which is often labelled in some way e.g. with a radioactive atom as part of the molecular structure) to animals, via one of the main routes of uptake. The presence of the substance (or its metabolites or the radiolabel) in various tissues and excreta is then measured for a period (up to a few weeks sometimes) of time to determine the fate of the chemical. Specialised studies may be carried out *in vitro* (e.g. using extracts of cells, often liver cells) to look at how the substance is metabolised by cell enzymes. The aim is to build up a picture of how much chemical is taken up by different routes, where it gets to in the body, how the body metabolises it and what metabolites are formed (as some of these may be toxic whereas the chemical itself is not) and how much, and by what routes, is excreted from the body over the observation period. Some methods have been developed whereby uptake across the skin can be measured in the laboratory using pieces of animal or human skin (the latter where national ethical considerations allow). 66 There have also been considerable developments in computer aided modelling of toxicokinetics (called physiologically-based pharmacokinetic, PBPK-modelling) to try to help predict how a substance will behave in the body. These models, based on biological principles, describe the body and its organs and tissues mathematically and use these principles to describe and predict the fate of chemicals in the body. They can be used as a means of extrapolating from one species or route of exposure to another and/or across a range of doses and exposure patterns. **6.2.3 Acute toxicity studies** The term acute toxicity in the testing context usually refers to experiments to determine the effects seen rapidly following a single dose of a chemical. **Acute systemic toxicity** - the aim of these types of studies is to provide information on what general toxicity might be seen if someone was exposed to a single, relatively high dose of a chemical (e.g. an accidental poisoning). In these studies a substance is administered to groups of animals as single doses. For solids and liquids this is often via the oral route and for gases, vapours and dusts by inhalation over a fixed period of time (usually 4 hours). A dermal test can also be conducted whereby the substance is applied to the skin usually under a dressing to keep it in place. Following administration the animals are observed, usually for at least two weeks afterwards in order to look for signs of toxicity. At the end of the observation period the animals will have a general examination of their body organs to look for signs of organ and tissue damage as markers of toxic effects. Historically, these studies were designed to identify a statistically derived value known as the LD50 or LC50-- the dose or concentration that would on average result in the death of 50% of the test population of animals over a specified time period. This crude measure was and is useful as a way of 67 making relative comparisons of the acute toxicity of substances in order to rank them in terms of their severity in inducing acute toxicity. However, such studies often overlooked more critical information such as toxic signs which may be useful indicators as to how a chemical exerts its effect. Other methods (such as the "Fixed Dose Procedure" and "Acute Toxic Class Method") have been developed which use fewer animals and depend more on signs of toxicity rather than just mortality. In general one will still find LD50 or LC50 values cited (e.g. in Safety Data Sheets) but the results from the other methods are becoming more readily available and are now part of regulatory schemes having established international guidelines. **Irritation studies** - some substances can cause localised inflammation on contact with the skin, eye and respiratory tract -- in the worst cases this damage can be extremely severe leading to corrosive destruction of the tissue. Clearly if it can be predicted that a substance is likely to be corrosive (e.g. strong acids or alkalis) in an undiluted form then it would not be reasonable or ethical to undertake any animal testing. Traditionally, skin and eye irritation tests use animals, usually specially bred laboratory rabbits. For skin irritation, the test substance is applied to the rabbit skin under a dressing for a few hours, with moistening if required for solid substances, and then the dressing is removed. The skin is then assessed by looking for redness (erythema) and swelling (oedema) and a scoring system based on the severity of the reaction seen is used to provide a semi-objective method for judging the degree of the response. Observations will be made for up to 14 days after the treatment to look for reversibility of response. Regulatory schemes use the severity and duration of the scores to formally classify substances as irritants. For eye irritation, the substance is instilled into the lower eyelid of the eye sack and the severity of the response judged by scoring the redness and swelling of the conjunctiva and eyelids, the opacity of the cornea and effects on the iris over a period of up to 21 days. 68 Significant efforts have been made to find alternative non-animal test methods for assessing the skin and eye irritation potential of substances. These efforts have resulted in internationally validated methods becoming available which use *in vitro* systems for skin irritation and corrosion. Although not routinely investigated, the ability of substances to cause irritation of the respiratory tract can be tested both in animals and humans. Studies in humans are usually designed to measure sensory irritation -- that is subjective feelings of irritation (itching, soreness) in the eyes and upper airways. Such studies are often used as the basis for establishing, for example, occupational exposure limits, but care needs to be taken when interpreting such experiments. Usually, this type of study involves exposing volunteers to a test atmosphere and requiring them to record what they experience. Animal experiments can also be undertaken to investigate sensory irritation using changes in breathing rates as objective measures of response. These breathing changes are related to the way sensory nerves in the respiratory tract respond to stimulation by the substance involved. In some repeated exposure experiments (see section below) signs of irritation of the respiratory tract may also be observed. **6.2.4 Sensitisation studies** Testing whether a substance is able to induce a sensitisation response is normally carried out in animals. For skin sensitisation, traditionally guinea pigs have been used. There are a number of different experimental protocols that have been developed but all basically depend upon giving the guinea pigs a series of exposures via the skin. After a break from this "induction" phase (inducing the immune system to respond) another part of the animal's skin is then exposed to a lower (non-irritant) dose of the chemical. If the substance has sensitising properties then this is manifest through a skin reaction which can be scored for severity (similar to the 69 approach used for skin irritation). A substance will be considered as a skin sensitiser depending upon the proportion of test animals responding at a specified level of response. The above approach can sometimes be unpleasant for the guinea pigs, particularly where a second substance is used to stimulate the immune system. Consequently, another approach has been developed in mice which uses fewer animals and results in less severe responses. This test depends on measuring the proliferation (rate of cell division) in cells of the immune system involved in the induction of the sensitised state, rather than testing to see if the animals respond to a subsequent challenge to show that they are sensitised as in the guinea pig test. In this type of experiment the skin on the mouse ear is treated with the substance of interest and then the animals are given radioactively-labelled precursors of DNA molecules to help detect increased cell division. After the induction period, lymph nodes are collected and the radioactivity measured to judge the increase in cell division and if it meets a specified level (usually at least a three-fold increase) then the substance is judged to have the potential to be a skin sensitiser. For humans who exhibit signs of skin sensitisation (e.g. rashes typically induced by these kinds of substance) then "challenge" tests can be performed under medically supervised conditions by injecting substance under the skin or placing it on the skin surface to see if a response is induced. This is clearly useful for helping to diagnose the condition and identifying the causative agent. Attempts have been made over many years to develop tests for predicting if a substance has respiratory sensitisation potential but it has been difficult to obtain consistency in appropriate animal models. The guinea pig has been used most often and the approach has been to expose them repeatedly to attempt to develop the immune sensitised state and then subsequently with a challenge dose with measurement of, for example, breathing rates and 70 blood antibodies being made. However, no single method has emerged as the basis for establishing a recognised, validated guideline. As for skin sensitisation, humans with occupational asthma can undertake medically supervised challenge tests for diagnostic purposes in order to determine causative agents and severity of response. **6.2.5 Repeated dose toxicity studies** In most workplace situations exposure to a chemical usually will occur repeatedly at relatively low levels over relatively long periods of time. Testing for the effects of repeated exposure to a chemical is carried out in animals, most often rats, over varying periods of time. The most often used exposure periods are 28 days (sub-acute testing), 90 days (sub-chronic) and for one or two years (chronic), the latter often being part of testing for the potential of a chemical to cause cancer (carcinogenicity). Exposure is most commonly undertaken by the oral route, either by a tube (gavage) directly into the stomach but alternatively, particularly for chronic exposure periods, with the chemical incorporated into the animals' diet or drinking water. Less commonly, animals may be exposed via inhalation to vapours, gases or dusts if this is appropriate (e.g. that is the most likely and/or important route of exposure). Inhalation testing is technically very demanding and expensive which is why it is less commonly used. The least often method of repeated exposure is by application to the skin. Three dose levels are usually used with an unexposed (i.e. receiving the same treatment procedure but without the chemical present) group acting as controls. Normally the top dose is designed to induce clear toxic effects and the other two doses less effect and no effects. The shorter term experiments can act as a guide for dose selection for the longer term experiments for the same chemical. 71 Where experiments have only been performed via one route of exposure then when assessing systemic toxicity it may be necessary to make judgements about relative uptake (dose) by different routes and allow for the fact that local effects may occur. At the end of the overall exposure period a range of measurements will be made on the animals, ranging from body and organ weights, assessment of blood levels of tissue parameters (e.g. liver and kidney associated enzymes, red and white blood cells) and macro- and microscopic examination of samples of tissues from the main organs of the body (the range of organs and tissues examined increases with increasing length of exposure period). In the longer term studies, extra groups of animals may be exposed and examined during the exposure period so that development of effects and their underlying mechanisms can be more thoroughly studied. Some studies may be designed to investigate specific effects, for example toxicity to the nervous system. In general, the shorter term studies use fewer animals in smaller group sizes with groups of both male and females being used. The longer term experiments use larger group sizes (to allow for some losses during the time of the experiment) and, particularly for the two year studies, to increase the statistical power of the studies to detect substance-related effects. The main outcome of these types of experiments is to produce information on the type of toxicity the substance possesses, potential target organs and tissues (both locally and systemically), the dose-response relationship (i.e. the variation in the magnitude of the effect with increasing dose) and if possible the No Observed Adverse Effect Level (NOAEL) -- the dose producing no observed adverse effects. Note that some changes may occur at the NOAEL but these may not be considered as adverse by the toxicologist and so are not thought of as contributing to the toxicity of the substance. It is also worth noting that in some cases the effects seen may or may not be induced in humans. 72 Sometimes the mechanism leading to the effect may not be possible in humans because of species differences and in other cases may be less severe. However, adverse effects seen are considered as markers that a substance possesses the ability to induce toxicity on repeated exposure and without any further detailed information this is taken as being of concern for human health. The above describes briefly the way standardised toxicity testing studies conducted to internationally validated and accepted guidelines are carried out, usually in relation to regulatory requirements. However, one may often find that, for a particular chemical, experiments have been published in scientific journals that are not necessarily conducted to these protocols but have investigated specific issues of interest relating the toxicity of the substance. These will have been through standard peer review procedures and can add usefully to the knowledge base of the toxicity profile of the substance and the understanding of its relevance to humans. **6.2.6 Genotoxity studies** There is a relatively wide range of tests available to look at the potential of a substance to induce genetic damage. Some look at the ability of a chemical to generally damage the genetic material or attempts by the cell to repair it, whereas others look at a substance's ability to cause specific mutations that can be measured by changes in cell function (e.g. the ability of the cell to make or use certain enzymes). The availability of this wide range of tests has lead to the development of strategic approaches where experiments on cells *in vitro* are used to screen initially for a chemical to be genotoxic. The *in vitro* studies usually include tests in bacterial cells (called the Ames test after the person who first developed them) which detect mutations and cells derived from mammals that are used to detect damage, its repair or mutations. The Ames test is based on the assumption that a substance that is 73 genotoxic (mutagenic) to the strain of Salmonella bacteria used in the test may also be a carcinogen i.e. cause cancer. While some carcinogens do not give a positive Ames test (and vice-versa) the ease and low cost of the test make it valuable as a screening test. Some experiments use yeast cells or fruit flies (*Drosophila melanogaster*). The *in vitro* experiments include exposing the cells to the just the chemical itself and also to it in the presence of a liver extract intended to investigate the possibility that its metabolism may generate genotoxic metabolites. If this battery of tests is negative then usually the substance is considered as not possessing genotoxic potential and no further testing is required. If these experiments in cells demonstrate a substance to have genotoxic potential, then studies can be performed to determine whether this activity would be expressed in animals. Generally these studies are performed in mice (other rodent species can be used) which receive one or two doses of the chemical and then measurements made in bone marrow derived cells to look for signs of genetic damage. Positive results in these animal studies would be taken to mean that the substance has the ability to induce genetic damage and in somatic cells this might mean that it is potentially a cancer causing agent. If genetic damage is detected in the normal (somatic) cells of animals then further tests can be carried out to see if mutations could be formed in germ cells by the chemical and these passed on to offspring. This is usually carried out by dosing parental animals and then looking for damage to the genetic material of germ cells (in sperm, as this is easier to detect) or for specific changes in the offspring which are known to be due to mutations being passed on from the parents. If positive responses are observed then this is taken as evidence that the substance has the potential to induce heritable genetic damage. The type of strategic approach outlined above has been developed over many years. However, this step-wise approach is used mainly in regulatory 74 settings and it is possible to find studies published in scientific journals that use these techniques (and others) but not necessarily in a step-wise fashion. As for other toxic properties, these published studies provide useful information to add to the overall picture about genotoxic potential. **6.2.7 Reproductive and developmental toxicity studies** Some chemicals may have the potential to cause toxicity to the male and/or female reproductive systems and functions and/or to the developing offspring. Useful information on toxicity to reproduction may be obtained by examination of the reproductive organs in repeated dose studies. Testing for a chemical's ability to cause toxicity to reproduction is normally carried out in rats or mice. The approach used is relatively straightforward in that groups of male and female animals of reproductive age are dosed with a range (usually up to three) of doses of the substance of interest and then allowed to breed in pairs: a control group that does not receive the chemical but otherwise undergoes the same procedures is also included. Dosing is repeated over a number of weeks. Effects on reproduction can be observed, by alterations in mating behaviour or by reductions in the numbers of offspring produced compared to control groups. In some cases, some of the offspring can be used to continue dosing and these then mated in pairs to look at the effects through more than one generation of animals (these are called multi-generation studies). More complex studies can be undertaken to investigate specific aspects and mechanisms by which the chemical may be acting on reproductive organs. For example, only male or female animals might be dosed or special measurements made of how the offspring develop in their learning ability. As for repeat dose studies, such experiments will provide information on dose-response relationships and no effect levels. 75 Although effects on the developing offspring can be observed in the above types of experiments some tests can be carried out that are specifically designed to investigate these aspects. These experiments usually use rats or rabbits as the test species where groups of pregnant female animals are dosed for a fixed number of days over the period from implantation of the developing offspring to just before birth to cover development and growth in the uterus. The highest dose is designed to induce some level of general toxic effects in the pregnant females. At the end of the experiment (usually just before the animals would normally give birth) the offspring and females are examined. If the treatments are toxic to the developing offspring then a range of effects may be observed. For example, normal development may have ceased. In some cases the chemical may have affected the normal development of for example the limbs of the animals resulting in malformations, an effect called teratogenesis (thalidomide is a typical example of a substance that has this property). Sometimes the effects may be very subtle and can just be small changes in growth rate or in sexual or intellectual development. In situations where effects are only seen in the presence of toxicity to the mother careful and specialist interpretation may be needed in order to determine whether the effects were secondary to this or directly by the substance on the developing offspring. **6.2.8 Carcinogenicity studies** Testing for carcinogenic properties is basically an extension of the chronic two-year repeated dose studies. Larger groups (50 male and female) of animals (usually rats and/or mice) receive two years (in some cases it may be lifetime) exposure to doses of chemical; a control group will received the same treatment but without the chemical present. The highest dose is usually designed to cause a small toxic effect but not enough to cause excessive numbers of deaths. Extra groups may be used to look at 76 different times to studies any possible mechanisms of cancer induction, such as increased cell proliferation in expected target tissues. The animals are observed for the duration of the experiment and any that die are examined for the presence of tumours. At the end of the exposure period, all remaining animals are examined for the development of any tumours and its type (e.g. benign, malignant) is determined. A wide range of body organs and tissues are also examined macro- and microscopically. If there is a significant increase in the number of tumours appearing in treated animals compared to an unexposed control group then this may be indicative of the substance having carcinogenic properties in humans. However, this interpretation is dependent upon a number of factors. For example, although the increase may be statistically significant at one dose, no dose-response relationship may be seen and so the biological significance may be questionable. Knowledge of the pattern of tumour formation in the species being used is also important since some strains of laboratory rodents may be prone to develop certain types of tumours when stressed and the substance used may not be critical. Also tumours may occur in rodents in organs not possessed by humans or through mechanisms that would not be expected to occur in man. Overall, therefore, the interpretation of such studies is complex and requires specialised knowledge and experience. **6.3 ALLERGY ASSESSMENT METHODS IN HUMANS** There are a number of different assessment methods that are available for determining whether or not a person is allergic to a particular substance. These include lung function tests (or spirometry), challenge testing, skin prick testing and blood IgE testing. 77 **6.3.1 Lung function tests** Lung function tests (sometimes called pulmonary function tests) are undertaken to evaluate how well the lungs are working. They are used to assess and differentiate conditions such as asthma, pulmonary fibrosis and chronic obstructive pulmonary diseases including emphysema and chronic bronchitis. The most common test uses a spirometer which measures the amount (volume) and the speed (flow rate) of air that can be inhaled and exhaled. They are undertaken breathing into a mouthpiece attached to the spirometer. Typically, in the most common test, the person is required to take the deepest breath they can and exhale as hard as possible for as long as possible. One limitation of lung function testing is that it is highly dependent on patient co-operation and effort. The test is also normally repeated at least three times to ensure reproducibility. The results of the lung function tests are usually given in terms of the basic data i.e. litres of air and litres per second, and also as a percentage of the typical or "predicted values" for people of similar characteristics (age, gender, height etc). Interpretation of the results must be undertaken by suitably qualified medical professionals, however, results within about 20% of the predicted values are usually considered "normal". Some common terms used in lung function tests are defined below: - FVC -- Forced vital capacity -- the total amount of air in litres that can be blown out after maximum inhalation - FEV1 -- Forced expiratory volume in 1 second -- the amount of air in litres that can be blown out in one second 78 - FEV1/FVC -- FEV1% - the ratio of FEV1 to FVC. In healthy adults this should be approximately 75 -- 80% - PEF -- Peak expiratory flow -- the maximum flow of air in litres per second at the start of the exhalation **6.3.2 Challenge tests** Spirometry can also be part of a bronchial challenge test. This may be undertaken to determine whether sudden contraction of the bronchioles (or bronchospasm) is as a result of exposure to a particular substance or environmental condition. It is useful to confirm the specific substance that is causing the sensitisation, but it should only be undertaken if other ways of diagnosing the sensitivity are ineffective. Starting with a small amount, the challenge involves exposure to increasing doses of the substances in question. Challenge tests are only undertaken under full medical supervision as it is possible that a severe reaction may occur. **6.3.3 Skin prick allergy tests** Skin prick testing is usually the first test recommended when an allergy is suspected. The main advantages are that it is a simple, quick (providing results within about 20 -- 30 minutes) and inexpensive form of testing. It can give useful information in all forms of allergy including allergies to substances that are inhaled or ingested. The test is undertaken by suitably qualified medical personnel. Substances thatare suspected cause of the allergy are mixed with liquid to make a solution. The arm is marked with a marker pen to identify the point of application of each drop. Up to 20 different allergen solutions may be tested at a time by applying a drop of each solution to the marked position on the forearm. 79 The skin beneath the drop is then pricked with a needle. This is usually not painful as only the top surface of the skin is pricked. However, this is sufficient to introduce a very small amount of the substance into the skin. The skin is then observed for a reaction, this will occur within a short period of time. A positive reaction is when the skin at the point of test becomes red and itchy within a few minutes. The area then becomes red and swollen with a "weal" in the centre, much like the reaction to a nettle sting. This normally reaches a maximum size after about 15 -- 20 minutes and usually clears within about an hour. **6.3.4 Patch testing** Skin prick testing is a way that suspected allergens are introduced into the body. It is therefore used to tests for allergies that do not necessarily occur on the skin. Patch testing is different in that it places substances on the surface of the skin and aims to identify skin allergens. The test involves the application of various test substances to the skin under adhesive tape that are left in place for 48 hours. The sites are examined after this period and again a further 48 hours later. The patches are usually applied to the upper back. Any reaction to the substance is classified to established criteria (see Section 2.5.3). The distinction between allergic and irritant reactions is of major importance. An irritant reaction is most prominent immediately after the patch is removed and fades over the next day. An allergic reaction takes a few days to develop, so is more prominent on day five than when the patch is first removed. 80 **6.3.5 Serological tests** Serological tests involve analysis of the blood serum for the presence and quantification of specific IgE antibodies in circulating blood serum. There are a range of different techniques such as RAST (radioallergosorbent test), UniCAP system and ELISA. The details of the various techniques are beyond the scope of the course. However, it is important to ensure that any laboratories used to undertake these tests take part in suitable external proficiency testing programmes. Serological tests are useful: - As a quantitative measure of specific IgE antibody in a particular individual - For individuals who are taking antihistamine drugs which suppress skin prick test reactions - For individuals with such extensive skin disease that skin prick tests are difficult to undertake However, serological tests are invasive requiring blood sampling, is relatively costly and the results are not available immediately. The skin prick test is less invasive, provides immediate results and is generally less expensive. 81