Mediators in Allergy Diagnosis

Test related documents

Print Print icon Tell a friend Tell a friend icon
Professor Staffan Ahlstedt, Manager of Scientific Affairs, Pharmacia & Upjohn Diagnostics and Professor of Immunological Techniques at the Department of Clinical Immunology, University of Göteborg, Sweden. Published in ACI International 1998;10/2:37-44.
 
This article looks at the background and development of current in vitro tests, arguing that they are better standardized, in many cases more useful for examining the "allergy march," and often more convenient than in vivo tests. Specifically, the use of in vitro tests in determining IgE levels, inflammation and inflammatory mediators such as ECP, and therapy response is discussed, as is the need for careful interpretation of results.
 
Introduction
The allergic immune reaction and the associated inflammation involve many cells which release mediators contributing to the pathogenesis of the disorders. Antibodies and mediators have been used as diagnostic tools for many years. Mediators released from mast cells and eosinophilic granulocytes are becoming increasingly used as markers of the inflammation and are finding their place in clinical practical work, whereas others, in particular mediators from the immune system such as the cytokines, the interleukins (ILs), and the cell membrane associated markers (CDs), are still used only in research, as are candidates from the adhesion molecule family. Different kinds of assay systems are used to determine the presence and the levels of the mediators. For practical routine work, in vitro immunoassay systems may be the most convenient tools, and they also have the potential for a high degree of standardization.
 
For correct interpretation of results from tests such as those mentioned above, it must be emphasized that the development of allergic symptoms is genetically determined. Thus, the presence of allergic disease in both parents is associated with a high probability of their children developing disease, whereas allergic disease in only one parent or in other, more distant relatives is associated with a lower risk. There are also genetic factors involved in the regulation of the IgE antibody response and in the ability of an individual to become sensitized upon exposure to allergens. Similar genetic factors also regulate the inflammatory response and the level of reactivity, e.g., the state of hyperresponsiveness.
 
The development of allergic disease is a dynamic process over time involving IgE antibody formation, inflammatory reactions, and hyperresponsiveness to different stimuli. Thus, allergic diseases in infants show changing patterns in relation to age. Such a shift can be described as the "allergy march." This implies that many patients with atopic eczema subsequently develop allergic gastrointestinal problems and later asthma; many chronic asthmatics have a history of atopic dermatitis and gastrointestinal disorders during the first years of life. Sensitivity to food in infants is often associated with the appearance of respiratory allergy to inhalants later in life [1].
 
Allergen exposure is the important first step in initiating the atopic condition, followed later by inflammation and consequent hyperresponsiveness and symptoms. Thus, early sensitization in the gastrointestinal tract may express itself later as symptoms in the respiratory tract [2]. In different patients and on different occasions, the relationship between sensitization, inflammation, and hyperresponsiveness may be substantially different, leading to very heterogeneous patterns of clinical symptoms. This is made even more complicated in the various age groups following environmental exposure to allergens and pollutants. The sensitization process and the presence of symptoms may also be influenced by infections and treatment. This provides a platform for discussions regarding the clinical usefulness of diagnostic tests in daily work for the identification of the onset of allergic disease in different age groups — the "allergy march." This march is illustrated in Figure 1.
 


Fig. 1. Prevalence of allergic symptoms related to age. Eczema is the first symptom in the youngest age group, followed by gastrointestinal symptoms and respiratory tract symptoms, which often first appear as wheezing. [1]

 
The emerging question is: Can the "Allergy March" be identified and predicted, or even reversed? To answer this question, we need a careful and rigorous interpretation of test results to allow an efficient treatment strategy. The basis for correctly interpreting the levels of mediators is a careful patient history without which the information obtained by determining the levels of mediators can be misleading.
 
The Use of In Vitro Tests to Demonstrate the Presence of IgE Antibodies and Inflammation
 
Technical Considerations
In vitro diagnostic tests can demonstrate the presence or absence of mediators such as IgE antibodies; mediators such as ILs and CDs; inflammation markers from mast cells, such as tryptase and histamine, prostaglandins (PGs) and leukotrienes (LTs); from granulocytes, such as eosinophil cationic protein (ECP); and adhesion molecules from endothelial cells [3,4]. The test systems can be well standardized with a high reproducibility and accuracy for serum and other body fluids at the given sensitivity ranges.
 
Determination of specific IgE antibodies is currently the most widely used test system in clinical routine for diagnostic purposes. It is of utmost importance for such determinations to be specific and not to give elevated values in the presence of high total IgE. The reagents must be absolutely IgE specific without cross-reactivity for IgG. They must provide reproducible results over time which are related to a reference curve. In order to provide results in true quantitative terms, however, they need to be calibrated quantitatively against an international standard, preferably against the WHO lot standard. The test systems must also utilize allergen sources that are representative and that do not contain contaminating allergens or toxins from other sources. Allergen extracts must contain all important components in sufficient quantities in order to remain unaffected by the presence of high IgG antibody levels of the same specificity. Furthermore, the presence of nonspecific total IgE in some patients’ serum samples must not give rise to elevated signals with false positive results. Thus, positive and negative values must be reproducible. It should be emphasized that not every in vitro test system available fulfils the criteria of being standardized, accurate, and reliable [5].
 
In vivo testing such as skin tests must be well standardized to give accurate results. The traumatic effect of a needle being put into the skin may be one of the worst causes of inaccuracy between different test performers and of poor reproducibility between test occasions. Similarly, the location of the skin test sites has an impact on the results, as do chronobiological effects and medical treatments. Therefore, the evaluation and interpretation of in vitro test results vs. in vivo reactions must take into account the fact that environmental factors, pathogenic mechanisms, as well as pharmaceutical therapies may have different effects on in vitro compared with in vivoparameters [6].
 
In vitro tests that adequately measure the presence of IgE antibodies can be used in identifying atopic individuals, in determining the allergens to which they are sensitive, and in monitoring their risk for evolution of allergic disease [3, 7]. In vivo tests such as skin prick tests (SPT) may also be used for these purposes, although they are dependent on more complex interactions between IgE antibodies, activated mast cells, and other cells as well as on mere irritation in the skin.
 
Regardless of the test modality, an adequate case history is the first and essential step in the diagnosis of the patient, since it provides information about which allergens should be tested for using the in vitro test system. Test systems that utilize large constant panels of individual allergens may be misleading and difficult to interpret in the absence of adequate clinical information.
 
Similar considerations apply to the determination of cell mediators from the inflammatory system. Sensitivity is one important property. Thus, the test available for tryptase released in vivo from mast cells does not detect the normal levels of that mediator, allowing some uncertainty in the evaluation of a patient’s reaction unless positive tryptase levels arise. A new test has been developed that also detects normal values [8]. However, although the new test demonstrates mast cell tryptase levels also in normal individuals, it does not provide enhanced clinical sensitivity, since elevated tryptase levels in the serum can only be demonstrated after systemic anaphylactic reactions. Beside tryptase, histamine and methylhistamine are other mast-cell- or basophil-derived mediators. In the circulation, however, these mediators are less useful, because they exhibit half-lives of only a few minutes after the reaction is initiated.
 
For mediators from the eosinophils, assays have been developed and are now available, for example, for eosinophil cationic protein (ECP) and eosinophil protein X (EPX). In vivo release of these proteins can be determined in secretions, such as bronchial alveolar lavage (BAL), sputum, tear fluid, and urine. However, the serum levels of these mediators are very low and give little information. In contrast, useful clinical information is obtained if determined after ex vivo release from activated cells into serum during clotting.
 
In vitro cell activation assays for histamine and LTs are another option, although such tests do not provide direct information about IgE antibodies, but rather indicate that basophils may contribute to the pathogenesis of the disorder. Standardization of such assays in the laboratory also represents a challenge.
 
Factors Involved in Sensitization
Increased environmental allergen exposure combined with the proper genetic constitution may cause increased prevalence of allergic diseases. Such allergen exposure may begin very early: Maternal exposure to allergens may prime the fetus and represents an increased risk for later allergic diseases in the child [9]. Sensitization to food allergens seems to be one that appears the earliest, before any sensitization to inhalant allergens becomes evident. The factors influencing early sensitization and activation of the immune system for subsequent allergy development include allergen dose, allergen species, mucosal penetration, infections, exposure to tobacco smoke, and pollutants. The earliest contact routes are through the skin and the gastrointestinal tract, whereas later on inhalation may play the predominant role. Sensitization can to some extent be prevented by breast feeding, and avoidance of high allergen exposure as well as exposure to tobacco smoke.
 
Exposure of a genetically susceptible individual to environmental allergens and foods may turn the immune system on, specifically activating Th2 lymphocytes rather than Th1 lymphocytes and create the "atopic condition," i.e., a propensity to form specific IgE antibodies to common environmental allergens. The amount of IgE antibody formed is related to the degree of exposure to allergen as well as to confounding factors such as infections. Infections may also increase the permeability of the mucous membranes, resulting in further increased exposure to allergen and subsequent IgE antibody formation. This contrasts with the normal "nonallergy" situation, where the immune system is turned off and a state of tolerance induced to the most common environmental allergens. The phenomenon may be influenced by the normal gastrointestinal flora. This may be one reason for the different specific IgE antibody profiles in different age groups. Foods may be more important to start sensitization early in life. The earliest predictor for development of allergic disease seems to be IgE antibodies specific to hen’s egg white [7]. Such antibodies were recently found at 6 months of age to precede further IgE antibody formation at 5 years of age to inhalant allergens, such as mite allergen [10]. They indicate the development of an atopic constitution, with specific IgE antibody formation to food allergens in the very young child and to inhalant allergens later in life [7, 11, 12] (see Figure 2).
 

Fig. 2a. IgE antibody patterns in relation to age. IgE antibody formation often starts against food allergens and egg white, later followed by IgE antibodies against inhalant allergens. [2]
 
 

Fig. 2b. Symptoms of allergic disease catch up with IgE antibodies formed earlier in life. [2]
 
Such a sensitization pattern can be used to develop tests for the atopic condition, utilizing a mixture of food allergens in the infant and small child and the most common inhalant allergens in the older child and adult [11, 13–15]. The atopic condition, defined as an increased propensity to form IgE antibodies to the most common environmental allergens, can be detected by tests utilising mixtures of allergens. Such tests possess a predictive value of more than 90 % [11, 12].
 
Since the formation of IgE antibodies is regulated by a complex interplay between different immunological cells and cytokines, interleukins, and cell-membrane-derived markers (CDs), one interesting diagnostic possibility would be to analyze those mediators. Such attempts have been made, showing that IL-4 driving IgE antibody formation can be detected in small children developing allergic disease, whereas levels of interferon g (IFN-g) and CD23 were not related to the developing allergy.

IL-4 appeared before allergic symptoms, but its relation to developing IgE was not established, and any firm conclusion about its usefulness cannot yet be drawn [16].
 
Inflammation and Hyperresponsiveness
 
Detection and Monitoring of Inflammation
The ability of an individual to become sensitized to allergens, to develop allergic inflammation, and to develop irritation and hyperresponsiveness due to neurogenic mechanisms in some shock organs are inherited more or less independently from each other, although their successive appearance may be related.
 
Allergic diseases and their manifestations can be exemplified by asthma. All allergic conditions have the common feature of being variable over time. Airway inflammation involving several mechanisms and cell types also varies over time. There is also a variable airway obstruction with bronchospasm, reversible with b2-agonists. In most patients, airway hyperresponsiveness develops, which also varies over time. In any given patient, it is not possible to predict which of the mechanisms contribute most to the impairment of the lung function. Furthermore, these parameters have different kinetics. Thus, they may appear independently from each other, despite being related. The various measurements of airway inflammation, bronchospasm, and hyperresponsiveness should therefore not be expected to correlate closely. Interventions are necessary to cope with the progress of the disease and to prevent acute asthma episodes. Such interventions may include environmental control of allergen exposure to decrease the inflammatory process and the use of efficient drugs that interfere with the inflammatory process and/or relieve the symptoms [17].
 
Sensitization and renewed exposure to allergen lead to inflammatory reactions, usually symptomatic. The allergen is taken up and processed by antigen-presenting cells and subsequently presented to T cells of the immune system. T cells orchestrate the whole inflammatory reaction and the regulation of specific IgE antibodies primarily through a vast number of cytokines. IgE antibodies bound to mast cells and allergen in particular initiate the release of vasoactive substances and inflammatory mediators. The number and activity of mast cells are also regulated by the activity of T cells. With or without the allergen-specific IgE mast-cell reaction, T cells drive the evolution of the inflammatory process by orchestrating the number and activity of the eosinophils. The propensity of an individual to respond to tissue injury with inflammation is partly inherited and partly acquired. The inflammatory features involve lymphocytes, eosinophils, mast cells, macrophages, and epithelial cells, whereas neutrophils are less prominent. After activation, the cells also release mediators, which have deleterious effects on the bronchial epithelium, leading to epithelial cell shedding and tissue remodeling [18] (see Figure 3).
 

Fig. 3. Features of inflammation in asthma. Stimulating cells with IgE antibody receptors, activation of T-cells, and eosinophils result in epithelium shedding and fibrotic processes in the basal membrane (tissue remodeling).
 
Activated cells, such as eosinophils, as well as their mediators, such as ECP, are present in biopsies from the submucosal airway tissues, in bronchial alveolar lavage (BAL), as well as in the sputum of asthmatics. Exposing airways to allergen and thereby inducing inflammation results in the activation of T lymphocytes, recruiting in turn activated eosinophils. Such T-cell activation may result in the observed elevation of IL-2 receptor (IL-2R) levels in serum of asthmatics [19]. T cells activate eosinophils in the bone marrow and cause their migration as well as that of progenitor cells into the peripheral blood [20]. Such activated cells recovered from the peripheral blood have an increased propensity to release mediators such as ECP [21]. Therefore, serum ECP determination is a more efficient way to monitor eosinophil activity then mere eosinophil counts, where all cells are counted irrespective of their state of activation. Serum ECP can be used to measure allergen exposure [22, 23]. A similar situation exists for IL-2R serum levels and T-cell counts.
 
Late clinical reactions in asthma are closely associated with the recruitment of eosinophils releasing ECP in serum. Since ECP appearing in serum in the test tube originates from activated living eosinophils releasing ECP during serum preparation, its levels are dependent on time and temperature during the serum preparation [3]. Serum ECP can also be used as a measure of allergen provocation causing increased eosinophil activity [24, 25]. Allergen avoidance and the subsequent decline of eosinophilic inflammation, either by specific measures such as moving to a different area [22] or by decreasing presence of allergen in the environment [23], can also be monitored by serum ECP levels. Such variations can be utilized to assess the need for increase or reduction of therapy [26]. Serum IL-2R levels are not as sensitive to therapy effects [19].
 
One common feature of inflammation is edema. Experimental studies have demonstrated that this condition correlates well with the presence of the high molecular weight glycoprotein hyalorunane (HYA), which is produced by fibroblasts and present in high concentration in the lymph. This is reflected by elevated levels of HYA in BAL related to the severity of asthma [27].
 
Similar inflammatory activation processes occur in atopic dermatitis (AD) skin, with ECP release and attraction of eosinophils into the inflammatory focus [28]. Mediators from the cytokine system, such as TNF, have not shown any close association with inflammatory reactions seen in allergic disease processes [29]. Markers from the interleukin system such as sIL-2R have also been investigated and found to be elevated in both asthma and AD, although the relation to disease activity was less clear than for ECP; levels of IL-2R did not decline following corticosteroid therapy [19,30].
 
Results from studies on adhesion molecules in AD are promising from a pathogenic point of view. In one recent study in AD patients, measurements of ICAM-1 in serum reflected the clinical response to corticosteroids similarly to serum ECP [31]. However, since most results have been obtained in situ from biopsies utilizing immune histochemical techniques, the practical application of such determinations is still not clear.
 
Specific food challenges in food-intolerant patients result in release of ECP, methylhistamine, and albumin. The clinical correlates are discomfort, diarrhea, and bloating [32]. In some patients, there seems also to be a particular mechanism of local lymphedema and swelling of the waist circumference closely correlating with intraluminal concentrations of HYA [33].
 
Inflammatory Responses in Relation to Therapy
Patients with acute asthma exacerbation requiring emergency clinic visits usually have severe ongoing inflammation and increased serum ECP levels in relation to decreased lung function parameters [34]. With rescue medication, serum ECP levels decline rapidly and lung function improves, although in general more slowly. During maintenance therapy, serum ECP levels may remain low, but sometimes the values increase with or without a simultaneous drop in lung function; symptoms frequently accompany such ECP increases [35].
 
Patients treated with different doses of inhaled corticosteroids may demonstrate a dose-response relationship between the serum ECP level and the dose of inhaled corticosteroid administered. Therapeutic doses that are too low may result in poor control of the inflammation [36]. Other factors, such as smoking, may also significantly contribute to losing control of inflammation or to the development of a different form of asthma. Such situations can be monitored through lung-function and hyperresponsiveness parameters and better through inflammation parameters such as serum ECP and EPX [36].
 
It is often difficult to maintain a patient free of symptoms and inflammation while deliberately tapering off the dose of corticosteroids. Monitoring such patients with inflammatory mediators such as ECP has revealed that asthma deterioration is accompanied by an increase of inflammation and serum ECP levels. Increased serum ECP levels often precede symptoms, so that the increasing inflammation is easier to assess qualitatively and quantitatively in the clinical context than by symptoms and lung-function parameters. Elevated serum ECP levels correlate well with symptom scores [37].
 
Inflamed airways are more susceptible to asthma exacerbation by exposure to allergen, tobacco smoke, and physical stress. This provides a basis for the predictive value of serum ECP for subsequent symptoms and impairment of lung function. Patients with ongoing inflammation as assessed by elevated serum ECP levels have been reported to be more susceptible to late asthma reactions after challenges with low allergen doses [38] and exercise [39] than patients with normal serum ECP levels. Asthma patients who are symptom-free at blood sampling but develop asthma symptoms during a subsequent period of observation have higher serum ECP levels than do patients who remain symptom-free throughout. Thus, in the diagnosed asthma patient, serum ECP levels may predict the further development of asthma symptoms [40].
 
Similarly, pollen asthmatic patients entering the pollen season develop increased hyperresponsiveness to metacholine, even in the absence of deteriorated FEV1 values. They also develop increased symptom scores, require more b2-agonist treatment, and show deteriorated PEF values. Such patients can be monitored by serum ECP levels, and the efficacy of inhaled corticosteroid therapy can easily be demonstrated [23]. Longitudinal monitoring of chronic asthma patients has shown that, despite inhaled corticosteroid therapy, such patients may still have impaired lung function related to increased levels of serum ECP; they do respond to increased doses of corticosteroids. Patients on corticosteroid therapy but without increased serum ECP levels may also present symptoms and impaired lung function. They, however, do not respond well to increased doses of corticosteroids but rather to long-acting b2-agonists [41]. Such patients presumably suffer from airways hyperresponsiveness rather than from eosinophilic inflammation. In contrast, patients with increased serum ECP levels despite treatment with inhaled corticosteroids, and symptomatic ones with unstable disease who are prone to asthma exacerbation, should be considered as undertreated. They would benefit from increased corticosteroid treatment rather than from increased doses of b2-agonists [42].
 
Hyperresponsiveness in Relation to Inflammation
Based on the above findings, one may view the evolution of asthma as based on two parallel arms: inflammation and hyperresponsiveness. Intervention in the inflammatory process would involve minimization of allergen exposure and/or the use of anti-inflammatory agents such as glucocorticosteroids. In contrast, intervening in the process of hyperresponsiveness would result in lower amounts of irritating substances, including those originating from the inflammatory process. This may occur despite the relief of symptoms by bronchodilators and anticholinergics. The use of bronchodilators and b2-agonists can bring symptomatic relief immediately and so efficiently that any increase in the underlying inflammation is blunted. Corticosteroid therapy, on the other hand, may need more time for maximum effects, since it works on inflammation and has only indirect effects on subjective symptoms. Symptoms recognized by the patients and requiring medical treatment only represent the visible part of the disease process (see Figure 4).
 

Fig. 4. The symptoms in allergic disease and asthma correspond to the tip of an iceberg. Inflammatory processes and activation of cells, release of mediators, such as IgE antibodies and eosinophil cationic protein (ECP), may precede the appearance of the symptoms. All the processes are related to a genetic inheritance. The ability to produce IgE antibodies upon exposure, the propensity to develop inflammation, as well as the state of hyperreactivity are all related to genetic factors. Thus, in the individual situation, it is impossible to tell which type of pathology is predominating, making judgement about adequate treatment difficult.
 
Ongoing airway inflammation may also enhance reactivity to nonspecific irritants. Thus, a polluted environment increases the progress of the disease if it contains high loads of allergens or toxic and irritant substances such as nitrous and sulfur air pollutants or tobacco smoke. Such increased irritation of the airways may explain why patients may have impaired lung function values and symptoms requiring treatment even without increased serum ECP levels [43]. The presence of airways inflammation results in tissue reactions, including epithelial shedding and tissue remodeling. At any given time a close correlation between the presence of inflammation and hyperresponsiveness may not be evident. In cohorts of patients with moderate asthma, there may be a rough association between, on the one hand, bronchial hyperresponsiveness and symptom score and, on the other hand, serum ECP levels. Usually, however, the individual patients show symptoms, hyperresponsiveness, and inflammatory patterns that occur rather independently. If in a study the allergen exposure during the pollen season dominates the disease profile in the recruited patient population, a closer association between measures of hyperresponsiveness and inflammation may be expected [44, 45].
 
Both the propensity to react with specific inflammation and the nonspecific irritation to various stimuli are genetically determined and are enhanced upon long-lasting inflammation. Inflammation is dependent on Th2 cells, which predominantly mediate allergic reactions through activation of IgE antibody production, as well as regulation of mast cells and eosinophils and release of ECP. Inflammation elicits and drives the hyperresponsiveness.
 
At least part of the hyperresponsiveness is mediated through neurogenic mechanisms [46, 47]. One such mechanism for tissue irritation is the neurological enhancement of the reactions by stimulation of sensory nerves. When mast cells are triggered by an IgE antibody reaction, mast-cell-derived mediators subsequently trigger further mast cells remote from the allergen exposure site. The mechanism for the remote mast-cell triggering occurs through the sensory neural system and the release of neuropeptides [47]. Such reactivity may be nonspecifically mediated by irritants in allergen mixtures [48]. It may also be affected by other nonspecific exposure to irritating substances occurring in the environment. Among foods, some spices such as chili pepper exhibit such effects.
 
Thus, the clinical reactivity observed may or may not be related to a given allergic reaction. Nonallergen specific escalating mechanisms in the in vivo response and other factors such as hormones may contribute to the variability observed, including the variation of skin-prick tests. Also, medications, chronobiologic variation, age, and test organ may affect the in vivo results. In individual cases it is not possible to predict at any given point in time the respective mechanisms responsible for the clinical reaction. The contribution of specific IgE antibodies, activated inflammatory cells, neurogenic mechanisms, hyperresponsiveness, chronobiological factors, and hormonal balances in the individual can be expected [46, 47]. Such confounding effects are not detectable when only the presence of specific IgE antibodies is assessed by in vitro tests.
 
The Concept of "In Vitro Veritas"
 
Interpretation of IgE Antibody Results
What does the presence of specific IgE antibodies mean and how is one to interpret the results in clinical terms? The presence of IgE antibodies can be demonstrated with in vitro and in vivo tests. However, the results of the two test modalities do not always agree completely.
 
To achieve accurate measurements, all components in a test system have to be standardized separately as well as in combination. This applies to in vivo testing as well. Therefore, in the evaluation and interpretation of in vitro test results vs. in vivo reactions of the patients, it should be considered that different environmental factors, pathogenic mechanisms, and pharmaceutical therapies may have different effects on in vitro and in vivo parameters [6].
 
Even if the in vivo testing is performed with the highest possible standards using highly purified allergens, there may be substantial differences between in vitro and in vivo results. The specific IgE antibody level may show up to 100-fold differences in different individuals or at different occasions in the same individual, despite identical skin test results, proving that the test systems are affected differently by the individual propensity to react to allergen exposure [49]. The effects of irritants in the triggering of an in vivo response may also be demonstrated by a lack of reaction to isolated allergen or to isolated irritant, but a strong response to a combination between allergen and irritant [48]. This points to the need for standardization of in vivo testing, including both allergen and irritant components. However, since the levels and activities of neurogenic and other nonspecific components are often not known, proper standardization is difficult to achieve [48].
 
Variation of the relative allergen concentrations in test solutions can shift the results from more specific and less sensitive to more sensitive and less specific [50] (see Figure 5). Similar variations can be achieved with in vitro tests. If carefully standardized and performed, good test sensitivity and specificity can be obtained. Additional variation between allergen species can be seen. One approach to further analyze the usefulness of in vitro allergy testing is the so-called receiver-operating characteristic (ROC). This is done by plotting the percent of true positives (i.e., sensitivity) versus false negatives (i.e., 1 – false positives; or 1 – specificity) using different specific IgE values for the different allergens. The interpretation of the results should be done in accordance with the statement of the Executive Committee of the American Academy of Allergy, Asthma and Clinical Immunology: "quantitative reporting offers numerous benefits, including facilitation of the usual statistical criteria for threshold selection and determination of precision profiles." The use of ROC curves also allows a different approach to allergy screening — or to selective diagnosis. If a high specificity is needed for diagnosis, the value giving the highest possible performance can be selected, whereas if a high sensitivity is required, one can use the other end of the curve to find the best threshold for a sufficient sensitivity. If used at all, the ROC analysis must be interpreted cautiously, in particular since it has to be performed for each single allergen and possibly also for different age groups. For the Pharmacia CAP System™, the 0.35 kUA/l cut off and quantitative IgE measurements are the most important evaluation tools [4, 5, 51].
 

Fig. 5. The relation between sensitivity and specificity in a test like the skin prick test is rather simple. By increasing allergen concentration, one increases the sensitivity but decreases the specificity and vice versa. It is up to the investigator to decide whether the test should be more sensitive or more specific. The optimal sensitivity and specificity varies for different allergens and preparations. [50]
 
Accurate and reproducible detection of specific IgE antibodies by in vitro test systems can be demonstrated to correspond better to clinical diagnosis by allergy specialists than in vivo testing, which is affected by individual hyperresponsiveness. However, in vitro tests and in vivo tests are not absolutely interchangeable in the individual, and they should be regarded as complementary tests in the management of the allergic individual. As measured in more than 2500 patients, careful standardization of the in vitro CAP RAST® and ImmunoCAP systems has led to accurate and reproducible diagnosis of individual sensitivity of allergic individual patients, which is better than the agreement with in vivo skin prick test (see Table 1) [5]. Therefore, the demonstration of sensitization to an allergen and its magnitude requires in vitro demonstration of IgE antibodies.
 
Table 1.
Agreement between the Pharmacia CAP System, Expert Diagnoses, and Skin Prick Tests. [5]

 
Well-standardized determinations of IgE antibodies against allergens utilizing the Pharmacia CAP System have been shown to correspond very well to the diagnosis made by experienced allergy specialists. The agreement with the specialist diagnosis is even better than the agreement with the results obtained by skin prick test.
 
Interpretation of Results of Inflammatory Mediators such as ECP
What does the presence of ECP mean? The serum ECP in diagnosed asthma patients can be summarized in relation to need for therapy according to Figure 6. Increased serum levels of ECP in a patient with diagnosed asthma indicate an ongoing eosinophilic inflammation. Such inflammation may have been induced by allergens, infections in the environment, or by intrinsic factors. The ongoing eosinophilic inflammation should therefore be assessed as an increased risk of evolving symptoms and acute asthma exacerbation in case of further exposure to trigger factors. This has been shown by allergen challenge of symptom-free asthma patients with high serum ECP levels. The clinical judgment of need for therapy is obscured by the fact that the need for medication does not remain static, but varies according to changes in the activity of the disease. Because of environmental conditions and infections, there may be a need to adjust the therapy. However, if a patient becomes slightly symptomatic but has a low serum ECP, increased doses of corticosteroids may not be the first choice, but rather suggest a course of b2-agonist [41]. Measurements of severity of asthma regarding the inflammation and serum ECP levels cannot be based on single determinations [52]. Any interpretation of a serum ECP value should take account of the patient’s need for and response to therapy [43] (see Figure 6).
 

Fig. 6. Inflammation is a basic mechanism in allergic diseases, and asthma and often precedes the symptoms. In asthmatic patients, an increase of eosinophilic inflammation assessed as increased serum level of eosinophil cationic protein should be taken as a warning signal even in the absence of any obvious symptoms. Such a patient might require increased doses of corticosteroids. If the patient still has symptoms but no signs of inflammation, an increased treatment with b2-agonists rather than with corticosteroids may be considered.
 
The demonstration of ongoing inflammation requires an in vitro demonstration of an inflammatory mediator such as ECP. In vivo tests demonstrate sensitization, inflammation, and hyperresponsiveness as a mixture, so that the individual component cannot be assessed. The standardization of in vitro testing for the presence of atopy (e.g., Phadiatop®) and specific IgE antibodies has demonstrated the clinical efficacy. Such a test, if reproducible and accurate, corresponds very well to the clinical diagnosis made by the physician [5, 11, 15].
 
The presence of allergy to specific allergens impacts the treatment strategies, i.e., the presence of inflammation should be treated by allergen avoidance or minimization, or by anti-inflammation agents such as corticosteroids, whereas nonspecific release from neurologically stimulated mast cells demands antihistamines, bronchodilators, and smooth muscle relaxation. The latter drugs do not impair the demonstration of specific antibodies to the allergen even if they totally eliminate the in vivo reactions. Only anti-inflammatory therapy eliminates the presence of activated eosinophils and serum ECP. Thus, the in vitro test may be the test of choice for the presence of sensitization and inflammation, whereas in vivo tests are a good complement when information is needed concerning the overall picture: (a) the presence of sensitization; (b) the level of ongoing inflammation; and (c) the degree of hyperresponsiveness.
 

References:

    1. Saarinen UM, Kajosaari M. Lancet 1995; 346:1065–1069
    2. Sigurs N, Hattevig G, Kjellman B, Kjellman NIM, Nilsson L, Björkstén B. J Allergy Clin Immunol 1994; 94:757–763
    3. Ahlstedt S, Enander I, Peterson CGB, Lanner Å. Pract Allergy Immunol 1993; 8:149–154
    4. Pastorello EA, Incorvaia C, Ortolani C, Bonini S, Canonica GW, Romagnani S, Tursi A, Zanussi C. J Allergy Clin Immunol 1995; 96:580–587
    5. Yman L. Das wiss Magazin z Thema 1990; 1:18–22
    6. Nelson HS. Allergy Proc 1994; 15:265–268
    7. Hattevig G, Kjellman B, Bjorkstén B. Pediatr Allergy Immunol 1993; 4:182–186
    8. Schwartz LB, Bradford TR, Rouse C, Irani AM, Rasp G, van der Zwan JK, Van der Linden PWG. J Clin Immunol 1994; 14:190–204
    9. Miles EA, Warner JA, Lane AC, Jones AC, Colwell BM, Warner JO. Pediatr Allergy Imunol 1994; 5:202–208
    10. Sasai K, Furukawa S, Muto T, Baba M, Yabuta K, Fukuwatari Y. J Pediatr 1996; 128:834–840
    11. Eriksson NE. Allergy 1990; 45:285–292
    12. Crobach MJ, Kaptein AA, Kramps JA, Hermans JA, Ridderikhoff J, Mulder JD. Allergy 1994; 49:170–176
    13. Wüthrich B, Schindler C, Leuenberger P, Ackermann-Liebrich U. Int Arch Allergy Immunol 1995; 106:149–156
    14. Moneret-Vautrin DA, Frémont S, Kanny G, Déjardin G, Hatahet R, Nicholas JP. Allergie Immunol 1995; 27:2–6
    15. Lilja G, Kusoffsky E, Johansson SGO, Öman H. Allergy 1995; 50:316–321
    16. Björkstén B, Borres, MP, Einarsson R. Int Arch Allergy Immunol 1995; 107:34–36
    17. Sheffer AL, Buist AS, Busse W et al. Expert Panel Report 2:1
    18. Venge P. Allergy 1994; 49:1–8
    19. Park CS, Lee SM, Chung SW, Uh S, Kim HT, Kim YH. Chest 1994; 106:400–406
    20. Denburg JA, Inman MD, Leber B, Sehmi R, O’Byrne PM. Allergy 1996; 51:141–148
    21. Carlson M, Håkansson L, Kämpe M, Stålenheim G, Peterson C, Venge P. J Allergy Clin Immunol 1992; 89:131–139
    22. Boner AL, Peroni DG, Piacentini GL, Venge P. Clin Exp Allergy 1993; 23:1021–1026
    23. Vatrella A, Ponticiello A, Parrella R, Romano L, Zofra S, Di Leva A, Bariffi F. Allergy 1996; 51:547–555
    24. Van Bever HP, Moens MM, Bridts CH, de Clerck LS, Mertens AV, Bosmans E, Stevens WJ. Allergy 1993; 48:443–449
    25. Mapp CE, Plebani M, Faggian D, Maestrelli P, Saetta M, Calcagni P, Borghesan F, Fabbri LM. Clin Exp Allergy 1994; 24:730–736
    26. Kristjánsson S, Shimizu T, Strannegård IL, Wennergren G. Pediatr Allergy Immunol 1994; 5:223–239
    27. Nettelblad O, Bergh J, Schenholm M, Tengblad A, Hällgren R. Am Rev Respir Dis 1989; 139:759–762
    28. Kapp A. Allergy 1993; 48:1–5
    29. Sugai T, Sakiyama Y, Matumoto S. Clin Exp Allergy 1992;22:275–281
    30. Furue M, Sugiyama H, Tsukamoto K, Ohtake N, Tamaki K. J Dermatol Sci 1994; 7:89–95
    31. Kojima T, Ono A, Aoki T, Kameda-Hayashi N, Kobayashi Y. Ann Allergy 1994; 73:351–355
    32. Bengtsson U, Hanson LÅ, Ahlstedt S. Clin Exp Allergy 1996; 26:1387–1394
    33. Bengtsson U, Knutson TW, Knutson L, Dannæus A, Hällgren R, Ahlstedt S. Clin Exp Allergy 1996; 26:96–103
    34. Ädelroth E, Rosenhall L, Johansson SA, Linden M, Venge P. Am Rev Respir Dis 1990; 142:91–99
    35. Koller DY, Herouy Y, Götz M, Hagel E, Urbanek R, Eichler I. Arch Dis Child 1995; 73:413–417
    36. Pedersen B, Dahl R, Karlström R, Peterson CGB, Venge P. Am J Respir Crit Care Med 1996; 153:1519–1529
    37. De Baecker WA, Van Overveld FJ, Vermeire PA. Am J Respir Crit Care Med 1998; in press
    38. Venge P, Håkansson L, Peterson CGB. Int Arch Allergy Appl Immunol 1987; 82:333–337
    39. Venge P, Henriksen J, Dahl R. J Allergy Clin Immunol 1991; 88:699–704
    40. Miyoshi M, Sakurai T, Kodama S. Jpn J Allergol 1991; 40:1391–1392
    41. Wever AMJ, Wever-Hess J, Hensgens HESJ, Hermans J. Respir Med 1994; 88:613–621
    42. Wever AMJ, Wever-Hess J, Hermans J. Clin Exp Allergy 1997; 27:519–529
    43. Wever AMJ. Eur Respir J 1996; 6:15–16
    44. Rak S, Löwhagen O, Venge P. J Allergy Clin Immunol 1988; 82:470–480
    45. Rao R, Frederick JM, Gregson RK, Warner JA, Warner JO. Clin Exp Allergy 1996; 26:789–793
    46. Lundblad L, Lundberg JM, Änggård A, Zetterström O. Allergy 1987; 42:20–25
    47. Marshall JS, Waserman S. Clin Exp Allergy 1995; 25:102–110
    48. McLelland J, Shuster S, Matthews JNS. Arch Dermatol 1991; 127:1016–1019
    49. Witteman AM, Stapel SO, Perdok GJ, Sjamsoedin DHS, Jansen HM, Aalberse RC, van der Zee JS. J Allergy Clin Immunol 1996; 97:16–25
    50. Toogood JH. Ann R Coll Physicians Surg (Canada) 1983; 16:310
    51. Plebani M, Borghesan F, Basso D, Faggian D. Allergy 1996; 51:407–411
    52. Ferguson AC, Vaughan R, Brown H, Curtis C. J Allergy Clin Immunol 1995;
1998