Component testing for pollen-related, plant-derived food allergies

INTRODUCTION — Advances in the identification of clinically relevant allergens and the development of recombinant proteins allow for assessment of immunoglobulin E (IgE) binding to individual proteins within an allergenic food. This type of testing is known as component-resolved diagnosis (CRD). Increased sensitivity and specificity can be achieved by assessing IgE binding to separate proteins, either purified native or recombinant, thereby providing improved diagnostic accuracy for predicting clinical reactivity. CRD may also provide additional prognostic information regarding the severity or persistence of food allergies.

CRD testing for pollen-related, plant-derived food allergies is reviewed here. Component testing for animal-derived food allergies is discussed separately. An overview of testing for food allergies is also presented separately. (See "Component testing for animal-derived food allergies" and "Diagnostic evaluation of IgE-mediated food allergy".)

OVERVIEW — Allergies to plant-derived foods may occur in pollen-sensitized individuals due to pollen allergens that cross-react with food allergens, such as profilins or pathogenesis-related class 10 (PR-10) proteins that are homologs of the major white birch pollen antigen (Betula verrucosa 1 [Bet v 1]). This type of allergy is associated with symptoms that are generally limited to the oropharyngeal area (oral allergy syndrome/pollen-food allergy syndrome). In the absence of sensitization to pollens, allergies to plant-derived foods are the result of sensitization to more stable proteins, such as the seed storage or nonspecific lipid transfer proteins (nsLTPs). In these cases, reactions are more often systemic, and there is a higher risk for anaphylaxis [1]. (See "Pathogenesis of oral allergy syndrome (pollen-food allergy syndrome)" and "Clinical manifestations and diagnosis of oral allergy syndrome (pollen-food allergy syndrome)".)

Component-resolved diagnosis (CRD) entails measurement of specific IgE responses to individual allergens as opposed to measuring IgE responses to allergen extracts that contain a mixture of proteins, including ones that may have greater or less clinical relevance. The pattern of specific IgE reactivity to defined allergens can help determine which patients are at higher risk for allergic reactions versus those who are sensitized but clinically tolerant. It may also help distinguish between those who are at risk for more severe reactions versus those who are likely to have milder symptoms.

The prime example of the utility of CRD in the management of food allergy is peanut allergy [2]. Several studies have noted that having detectable levels of specific IgE to the seed storage proteins are associated with more severe, persistent peanut allergy [3-7], whereas exclusive sensitization to a Bet v 1 homolog is more often associated with a low risk of reactivity to peanut [8,9]. (See "Peanut, tree nut, and seed allergy: Diagnosis" and 'Peanut' below.)

ASSAY SYSTEMS — Two different types of immunoassays are commercially available for the measurement of IgE levels to individual allergens:

●Measurement of specific IgE to individual allergens or components – Quantitative results of specific IgE levels to purified native or recombinant allergens can be obtained using a fluorescent enzyme immunoassay, such as the ImmunoCAP. Components may be selected for individual allergens with ImmunoCAP, allowing for individualized testing to further characterize a patient's allergies.

●Measurement of specific IgE to multiple allergen components in one test panel simultaneously – A microarray-based immunoassay such as the ImmunoCAP ISAC (Immuno Solid-phase Allergy Chip) provides results for multiple components from different allergen sources that cover a wide spectrum of food and environmental allergens using a small quantity of blood. This is a semiquantitative assay and, as such, may be better suited as a screening tool rather than a test for diagnosis and management of allergies [10].

Technical differences in fixing allergen to the assay platforms result in variability of specific IgE binding. Thus, IgE values obtained with ImmunoCAP are not equivalent to the corresponding values obtained with the ISAC microarray system [11,12]. Different units of measurement are used in the two assays (kU_A/L for ImmunoCAP versus ISAC standardized units [ISU] for ISAC).

In the United States, several of the individual component tests are approved by the US Food and Drug Administration (FDA), whereas others are not. This can impact the likelihood that the test will be covered by third-party payers. Thus, patients and/or clinicians may wish to determine if the tests are covered before they are ordered.

ALLERGEN-SPECIFIC USE/INTERPRETATION

Peanut — Several peanut (Arachis hypogaea, Ara h) proteins have been characterized and are US Food and Drug Administration (FDA) approved and commercially available for component testing (table 1). IgE levels to components, similar to whole-peanut IgE, provide a likelihood of clinical reactivity but, on their own, are not diagnostic of allergy or severity of reactions. Guidelines on when to use component testing for peanut allergy are available and are reviewed below (table 2) [13]. (See 'Peanut testing approach' below.)

Types of peanut proteins — Ara h 1, 2, and 3 are seed storage proteins and are the major allergens for peanut. Ara h 6 is closely related to Ara h 2. Ara h 8 is a birch allergen (Bet v 1) homolog associated mainly with milder symptoms localized to the oropharyngeal area. Ara h 9 is a nonspecific lipid transfer protein (nsLTP). Sensitization to this peanut protein is predominantly seen in the Mediterranean area [14-16]. Ara h 10 and 11 are oleosins, and small studies have shown that some peanut-allergic patients have detectable IgE directed towards these proteins [17]. However, their clinical relevance is unclear. Ara h 12 and 13 are defensins that have antifungal activity. In one small series, they were associated with severe anaphylactic reactions [18].

Diagnostic performance — IgE binding to Ara h 2 outperforms whole-peanut extract and other components in predicting clinical allergy [3-7], although optimal cutoff values for Ara h 2 with high positive predictive values have not yet been determined. A range of cutoffs with varying sensitivities and specificities have been reported in several studies, probably due at least in part to differing patient selection criteria and geographic locations, reflecting inherent differences in patient characteristics such as genetics, diet, and environmental exposures [3-6,9].

Component-resolved diagnosis (CRD) testing can facilitate patient selection of oral food challenges. A stepwise approach for peanut IgE and Ara h 2 component testing can significantly reduce the need for oral food challenges in patients who are highly likely to be allergic [5,6]. Data also suggest the utility of measuring IgE levels to other peanut components. Isolated sensitization to Ara h 8, for example, is associated with a high likelihood of tolerance to peanut [8,9]. Thus, testing for peanut components prior to a food challenge can be beneficial for risk stratification, enhancing the ability to select patients who are likely to pass a diagnostic food challenge.

The following studies are illustrative:

●A 2010 study using a population-based birth cohort of 933 children in the United Kingdom demonstrated a high rate of false-positive test results when using skin prick tests (SPTs) and specific IgE testing to whole peanut [3]. This study concluded that assessing IgE binding to Ara h 2 had higher accuracy than whole-peanut extract and other components in predicting clinical reactivity to peanut.

One-hundred ten (110) children (11.8 percent) had positive testing to peanut at eight years of age. Approximately two-thirds of these children (n = 66) had no reaction at oral food challenge to peanut; approximately one-fifth were allergic (12 by convincing history without challenge, 7 by challenge); and the remainder (n = 6) were deemed inconclusive.

Approximately 15 percent were misclassified when a peanut-specific IgE level of ≥15 kU_A/L or a SPT wheal size of ≥8 mm was used for a cutoff, whereas the misclassification rate was reduced to approximately 7 percent when results from a microarray that included multiple recombinant allergens from peanut, tree pollens, grass pollens, and cross-reactive carbohydrate determinants (CCDs) were used. Subjects with peanut allergy tended to have higher-fold change values (calculated expression level estimates of the sample against the negative controls) to the major peanut components Ara h 1 to 3, whereas the peanut-tolerant subjects had higher values to CCDs and grass components (Phl p 1, Phl p 4, and Phl p 5).

●Findings from a Danish study suggest that additional peanut components not routinely included in test panels may play important roles for certain individuals. This study reported that 2 percent of positive challenges had undetectable Ara h 2 [4], whereas another European study found that 26 percent of subjects who were tolerant to peanut showed IgE binding to Ara h 2 [19]. In addition, isolated sensitization to Ara h 8 is sometimes associated with positive challenges with systemic symptoms [8,20]. As an example, a case was reported of a child with undetectable IgE to Ara h 2 and elevated IgE to Ara h 8 who experienced anaphylaxis at oral food challenge to peanut. After performing additional testing, he was found to have an elevated IgE to Ara h 6, a homolog of Ara h 2 [21].

●One study has examined whether IgE levels to peanut components have prognostic value for persistence of disease. Sensitization to Ara h 1 to 3 at one year of age was associated with persistent peanut allergy at 13 years of age, but it did not provide independent predictive value in addition to the information provided by IgE levels to whole peanut [22].

●A United States study examined threshold levels of Ara h 2 in the diagnosis of symptomatic peanut allergy in a general pediatric allergy referral population [9]. Measurement of specific IgE to Ara h 2 had a diagnostic sensitivity of 96 percent and specificity of 54 percent when sensitization was defined as >0.1 kU_A/L (sensitivity 88.5 percent and specificity 71.4 percent using a threshold of 0.35 kU_A/L).

Peanut testing approach — In a patient with a history consistent with IgE-mediated allergy to peanut, we suggest first testing peanut-specific IgE.

The following scenarios illustrate some instances where targeted testing with peanut components may or may not be useful (table 2) [2,13]:

●Component testing can be informative for those with low peanut IgE levels (<25 kU_A/L), known birch pollen sensitization, and/or mild or no history of reaction to peanut. Having detectable IgE to Ara h 1, 2, or 3 is associated with higher risk of systemic reactions to peanut, and these patients would be advised to continue peanut avoidance.

●Peanut component testing is probably useful in adults with seasonal allergies who had tolerated peanuts earlier in life but later reported mild symptoms with peanut ingestion. Isolated Ara h 8 sensitization (without detectable IgE to Ara h 1, 2, or 3) suggests low risk for systemic reactions to peanut. The risk of a reaction on oral food challenge is decreased in this population, but a challenge is probably still warranted to confirm lack of systemic reactivity.

●For a young child who has experienced immediate systemic symptoms upon peanut ingestion, component testing is unlikely to provide additional information that would influence management.

●When considering a food challenge for peanut, we typically obtain component tests because of the potential for improved predictive value compared with the standard peanut test. Interpretation requires evaluation of the component-specific IgE concentrations and pattern of sensitization. As an example, isolated positive testing to Ara h 8 suggests unlikely allergy, whereas high levels of IgE to Ara h 2 may indicate probable allergy. Sensitization to multiple stable proteins (eg, Ara h 1, 2, and 3) may indicate a greater chance of reaction compared with isolated sensitization to just one of these proteins. The risk of reaction based upon testing and other patient factors (eg, asthma) is assessed when deciding whether to offer a medically supervised feeding test and also potentially where to perform the procedure (eg, office versus hospital). (See "Peanut, tree nut, and seed allergy: Diagnosis" and "Oral food challenges for diagnosis and management of food allergies".)

Tree nuts — Component testing is available for several tree nuts. The most well studied ones are listed below, although the studies are in small, selected populations with different outcomes and test cutoff points. Their utility in routine clinical practice is still unclear. Further studies are needed to determine optimal specific test cutoffs and in which patients testing will be most helpful.

Cashew — One of the major cashew (Anacardium occidentale, Ana o) allergens, Ana o 3, is predictive of cashew allergy [23-26]. Ana o 3 belongs to the 2S albumin protein family, a group of proteins that have highly stable intrinsic protein structure. Several studies have shown that Ana o 3 discriminates between allergic and tolerant children better than cashew-specific IgE (area under the curve of 0.89 to 0.94 versus 0.76 to 0.79) [24-26]. The optimal cutoff for sensitivity and specificity for Ana o 3 was 0.70 kU_A/L (sensitivity 85 percent; specificity 88 percent) in one study that confirmed cashew allergy by oral food challenge [26]. Another study found that the optimal cutoff was 0.16 kU_A/L [23]. A positive decision point for a sensitivity of at least 95 percent was 2.0 to 2.3 kU_A/L [24,26].

A systematic review and analysis of six previously published studies modeled outcomes for SPT (lower and upper cutoffs <3 mm and ≥12 mm, respectively), cashew specific IgE ([sIgE] ≤0.1 and ≥8.5 kU_A/L), Ana o 3 (≤0.1 and ≥0.35 kU_A/L), or a combination of sIgE and Ana o 3, with home reintroduction at or below the lower cutoff, avoidance at or above the upper cutoff, and oral food challenge for patients in between [27]. False positives were lowest with SPT alone (0.5 percent) compared with Ana o 3 or sIgE plus Ana o 3 (7 and 8.1 percent, respectively), and false negatives were lowest with sIgE alone (0.4 percent) compared with Ana o 3 or sIgE plus Ana o 3 (1.1 percent). However, these approaches required oral food challenges to determine allergy or tolerance in a higher percentage of patients (58.4 and 74.5 percent, respectively) compared with screening with Ana o 3 alone (12.2 percent) or combination sIgE plus Ana o 3 (11.4 percent). One suggested testing approach that should minimize false positive and negative tests and limit the number of food challenges needed is shown in the algorithm (algorithm 1).

Hazelnut — Several hazelnut (Corylus avellana, Cor a) allergens have been characterized. Cor a 1 is a heat-labile protein that is homologous with the major birch pollen allergen, Bet v 1. Sensitization to Cor a 1 is generally associated with tolerance to hazelnut or mild oropharyngeal symptoms [28,29]. Sensitization to Cor a 8, an nsLTP, has been linked with more severe symptoms in patients from the Mediterranean area [29,30]. Cor a 9 is an 11S globulin, and data have implicated sensitization to Cor a 9 in severe reactions in children in the United States and Europe [28,31,32]. Sensitization to Cor a 14, a 2S albumin, is also highly specific for more severe reactions at food challenge [32].

In a study of 26 children from a birch-endemic area (Spain) who had positive hazelnut IgE and underwent double-blind, placebo-controlled food challenge (DBPCFC), sensitization to Cor a 8 was found to be a good discriminator of severe allergic reactions to hazelnut [30].

A 2013 Dutch study investigated specific IgE binding to hazelnut components Cor a 9 and Cor a 14 as indicators for a more severe hazelnut allergy phenotype [32]. Diagnostic cutoffs of IgE levels ≥1 kU_A/L to native Cor a 9 in children and adults or ≥5 kU_A/L to recombinant Cor a 14 in children and ≥1 kU_A/L in adults had a specificity of >90 percent. This association was confirmed in a 2014 United States study of 116 patients who underwent hazelnut oral food challenge. Cor a 9 and 14 were highly sensitive and specific for predicting clinical reactivity to hazelnut.

Component testing for hazelnut is commercially available and can be informative for patients who test positive to hazelnut but have known birch pollen sensitization and/or mild or no history of reaction to hazelnut. Having detectable IgE to Cor a 8, 9, and/or 14 is associated with a higher likelihood of clinical reactivity to hazelnut, and these patients would be advised to continue hazelnut avoidance.

Walnut — The major English walnut (Juglans regia, Jug r) allergens include the seed storage proteins Jug r 1 (2S albumin), Jug r 2 and Jug r 6 (vicilins), and Jug r 4 (11S globulin) and an nsLTP, Jug r 3. Allergens associated with pollen allergies include Jug r 7 (profilin) and Jug r 5 (Bet v 1 homolog).

In one European study of 91 patients with walnut allergy and 24 tolerant controls, 96 percent were sensitized to at least one walnut component compared with 87 percent to walnut extract [33]. Sensitization to Jug r 5 was associated with mild symptoms and negative in vitro testing to walnut extract. Patients who had a history of severe reactions, most of whom developed the allergy before 14 years of age, had higher specific IgE levels to Jug r 1, Jug r 4, Jug r vicilin-L, and Jug r vicilin-H, but not Jug r 3 or Jug r 5, compared with patients with a history of mild reactions.

Another study found that Jug r 1 and/or Jug r 4 had the best diagnostic accuracy for walnut allergy and were superior to whole walnut extract and other components for predicting allergy versus asymptomatic sensitization [34]. No association was seen between IgE to individual components and type of symptoms or severity of reaction during an oral food challenge.

Wheat — Several groups of allergens have been described for wheat (Triticum aestivum, Tri a). Some of these allergens are associated with specific forms of wheat allergy, such as baker's asthma or wheat-dependent, exercise-induced anaphylaxis. Others are associated with severe allergic reactions. However, detection of multiple wheat allergens concurrently does not identify all patients with wheat allergy [35]. Thus, further research is needed to confirm the relevance of these proteins in larger populations and to assess the utility of measuring IgE to these individual allergens in the diagnosis of wheat allergy. These allergens are discussed in greater detail separately. (See "Grain allergy: Allergens and grain classification", section on 'Wheat'.)

Soy — Several soy (Glycine max, Gly m) allergens have been characterized [36]. However, sensitization to these specific allergens is not consistently associated with the diagnosis of clinical allergy or severity of reactions. Thus, further studies are needed to determine the clinical utility of measuring IgE to these components.

Birch pollen-related soy allergy is attributed to sensitization to Gly m 4, a Bet v 1 homolog [37]. Gly m 5 (beta-conglycinin) and Gly m 6 (glycinin) are storage proteins that are associated with severe reactions to soybean [38,39]. However, IgE levels to these allergens are not diagnostic for soy allergy, since nonsymptomatic individuals also have detectable IgE to these proteins. Although Gly m 4 sensitization is generally associated with milder symptoms or oral allergy, it is also related to severe, generalized symptoms in the absence of sensitization to Gly m 5 and 6 [40].

The storage protein 2S albumin (Gly m 2S albumin or Gly m 8) was reported to be an important allergen in Japanese children [41]. In this study, symptomatic children had significantly higher levels of IgE to Gly m 2S albumin than nonsymptomatic children. A study from the Netherlands also demonstrated that IgE to Gly m 2S albumin is important in adults with soy allergy. However, the diagnostic value of IgE to Gly m 2S albumin was comparable with SPT or IgE to soy extract [42]. In another study, the sensitivity of IgE to Gly m 8 was better than IgE to soy extract (78 versus 44 percent), but the specificity was lower (78 versus 87 percent) [43].

Fruits — Fruits more commonly cause oral allergy syndrome (pollen-food allergy syndrome) than systemic allergic reactions. A number of fruit allergens have been identified that cross-react with pollen allergens such as Bet v 1. Sensitization to other allergens, particularly nsLTPs, is seen in patients with fruit allergy who do not also have pollen allergy. Testing for most of these allergens is not available outside of the research setting. In addition, the utility of testing is generally not clear. Fruit and vegetable allergens are discussed in greater detail separately. (See "Pathogenesis of oral allergy syndrome (pollen-food allergy syndrome)".)

Measurement of IgE to specific kiwi (Actinidia deliciosa) allergens (Act d 1, 2, 5, and 8) on microarray testing is commercially available. Act d 5 is homologous with Bet v 1. Thus, determination of IgE reactivity to these proteins may distinguish those who have primary kiwi allergy and are at high risk for systemic reactions from those with secondary kiwi allergy due to pollen cross-reactivity [44].

Testing for IgE to several peach (Prunus persica) allergens is commercially available. Pru p 1 is a Bet v 1 homolog, Pru p 3 is an LTP, and Pru p 4 is a profilin. Some studies suggest utility in measuring IgE to these proteins in the characterization of peach reactions [45,46], but one study noted that sensitization to Pru p 3 was not associated with systemic symptoms nor was there an association with severity of reactions [47].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Food allergy".)

SUMMARY

●Advances in the identification of relevant allergens and the development of recombinant proteins allow for assessment of immunoglobulin E (IgE) binding to individual proteins within an allergenic food. This type of testing is known as component-resolved diagnosis (CRD). (See 'Introduction' above.)

●CRD holds promise for distinguishing patients with clinical allergy as opposed to those sensitized, but tolerant to, plant-derived foods. It also has the potential to provide additional information regarding the severity of symptoms that may occur in clinically reactive individuals, with those sensitized to pollen-related proteins having a lower risk of severe systemic reactions with exposure. (See 'Overview' above.)

●Two different types of immunoassays are commercially available: those that measure specific IgE to individual allergens (components) and those that measure specific IgE to multiple components simultaneously (microarrays). (See 'Assay systems' above.)

●For peanut allergy, data indicate that CRD can provide useful information for discriminating between pollen-related symptoms as opposed to primary peanut allergy, particularly in older individuals. This information can be used to facilitate patient selection of oral food challenges (table 2). (See 'Peanut' above.)

●Further studies in larger populations are needed to determine the clinical utility of CRD for each different food. Oral food challenge remains the gold standard for food allergy diagnosis. (See 'Allergen-specific use/interpretation' above.)