Sensitivity, specificity and positive predictive value of ALEX2 multiplex examination in patients suffering from atopic dermatitis and reaction to tomatoes

ABSTRACT

The aim of the study is to evaluate the clinical reactions to tomatoes in atopic dermatitis (AD) patients, and to analyse the results of specific IgE to allergens of tomatoes and to molecular components of pollen, seeds, fruits, vegetables and latex with the use of ALEX2 Allergy Xplorer test. According to our results, 15% of AD patients suffer from clinical reactions to tomatoes; we confirmed the low sensitivity and low positive predictive value of specific IgE to molecular component Sola l 6 and allergen extract Sola l, the specificity of this test is higher. The significantly higher sensitisation to other nonspecific lipid transfer proteins is observed in patients sensitised to tomato allergens. An elimination diet and an exposure test are recommended to detect a reaction to tomatoes. ALEX2 Allergy Xplorer test gives us a comprehensive picture of sensitisation and the possibility of evaluation of cross-reacting allergens.

KEYWORDS:

• Clinical reaction to tomatoes
• ALEX2 Allergy Xplorer
• atopic dermatitis
• specific IgE

Introduction

Although reactions to tomatoes are relatively common, there are no published papers on the investigation of the specific IgE aganist molecular components of tomatoes in patients with atopic dermatitis (AD), the common chronic skin condition, affecting up to 20% of the population and characterised by intense pruritus and epidermal barrier dysfunction (Bergmann et al., 2013).

Food allergens are, in addition to inhaled allergens, the important factors contributing to allergic inflammation in the pathogenesis of AD; these patients also show much higher rates of food sensitisation than controls, particularly to food proteins that are cross-reactive with airborne allergens (Bergmann et al., 2013; Maintz et al., 2006). The allergenic properties of tomatoes have been mainly neglected in the past, because of the potential to induce unspecific histamine liberation, which means non-allergic (non-immunological) hypersensitivity, food intolerance (Ballmer-Weber & Hoffmann-Sommergruber, 2011). Dermatologic sequels can be rashes, itch, urticaria, angioedema and worsening of AD (Maintz et al., 2006). On the other hand, food allergy is a reaction of the immune system to a certain food component (usually a glycoprotein). Tomato (Solanum lycopersicum L.) belongs to the Solanaceae family and is the second most important fruit or vegetable crop next to potato. Despite the beneficial role of tomatoes in the daily diet, they have been confirmed as one of the most prevalent allergenic vegetables (Ballmer-Weber & Hoffmann-Sommergruber, 2011; Włodarczyk et al., 2022). Patients with AD often react with clinical reactions to multiple foods and sometimes it is very difficult to detect food aggravating the skin lesion. In patients with severe and moderate AD, it is therefore recommended to test specific IgE against food allergens, but these results can often be false positive or false negative. It is therefore advisable to implement an elimination diet and to perform exposure tests with foods that may contribute to the worsening of skin findings (Ballmer-Weber & Hoffmann-Sommergruber, 2011; Włodarczyk et al., 2022). Although the gold standard for diagnosis of food allergy is an oral food challenge, clinically relevant biomarkers of IgE sensitisation, including serum-specific IgE and skin prick testing, can aid in diagnosis (Bousquet et al., 2016; Campana et al., 2017; European Task Force on Atopic Dermatitis, 1993; Hanifin & Rajka, 1980; Heffler et al., 2018; Jensen-Jarolim et al., 2017; Matricardi et al., 2016; Passalacqua et al., 2013; Patelis et al., 2016; van Hage et al., 2017; WHO/IUIS Allergen Nomenclature Sub-Committee Allergen Nomenclature, 2021). Advances in biotechnology and allergen research at the molecular level have made a major contribution to the diagnosis of IgE-mediated allergic conditions. The examination of molecular allergens focuses on the determination of allergen-specific IgE antibodies in the serum against molecular components prepared by recombinant technology (recombinant) or purified from natural sources (native) (Bousquet et al., 2016; Campana et al., 2017; Heffler et al., 2018; Jensen-Jarolim et al., 2017; Matricardi et al., 2016; Passalacqua et al., 2013; Patelis et al., 2016; van Hage et al., 2017; WHO/IUIS Allergen Nomenclature Sub-Committee Allergen Nomenclature, 2021). Some diagnostic modalities combine molecular allergens and extract testing in the forms of ImmunoCAP® Solid-phase Allergen Chip (ISAC) multiplex testing and ALEX® Allergy Xplorer which utilises microarray technology to perform multiple simultaneous analyses (Bousquet et al., 2016; Campana et al., 2017; Heffler et al., 2018; Jensen-Jarolim et al., 2017; Matricardi et al., 2016; Passalacqua et al., 2013; Patelis et al., 2016; van Hage et al., 2017; WHO/IUIS Allergen Nomenclature Sub-Committee Allergen Nomenclature, 2021).

Tomatoes allergenic molecules are recorded in Table 1. From these allergenic molecules, allergen extract Sola l and molecular component Sola l 6 are examined in ALEX2 Allergy Xplorer. Sola l 6 is non-specific lipid transfer protein type 2. Other molecular components are Sola l 2 (beta-fructofuranosidase), Sola l 3 (non-specific lipid transfer protein type 1), Sola l 4 (pathogenesis-related protein, PR-10, Bet v 1 family member), Sola l 5 (Cyclophilin), Sola l 6 (non-specific lipid transfer protein type 2) and Sola l 7 (non-specific lipid transfer protein type 1).

Table 1. Tomatoes allergenic molecules (Włodarczyk et al., 2022).

Molecular component	Biochemical name Protein family
Sola l 1	Profilin
Sola l 2	Beta-fructofuranosidase
Sola l 3	Non-specific lipid transfer protein type 1 (nsLTP1)
Sola l 4	Pathogenesis-related protein, PR-10, Bet v 1 family member, TSI-1
Sola l 5	Cyclophilin
Sola l 6	Non-specific lipid transfer protein type 2 (nsLTP2)
Sola l 7	Non-specific lipid transfer protein type 1 (nsLTP1)

Molecular component examined in ALEX2 Allergy Xplorer is shown extra bold.

To date, there are no studies evaluating the occurrence of tomato allergy in AD patients and evaluating the sensitivity of specific IgE antibody testing in ALEX2 Allergy Xplorer test in these patients. Allergy to tomatoes is also linked to other allergies, such as grass pollen and latex allergy (Ballmer-Weber & Hoffmann-Sommergruber, 2011; Włodarczyk et al., 2022). Some findings point towards a high rate of clinically silent sensitisation to tomato in pollen-allergic patients and an IgE response towards clinically not relevant cross-reactive allergens (Ballmer-Weber & Hoffmann-Sommergruber, 2011; Włodarczyk et al., 2022). For this reason, we not only focused our work on the analysis of specific IgE against molecular components of tomato, but also performed the analysis of specific IgE against other molecular components that may play a role in patients with a clinical reaction to tomatoes and in patients with sensitisation to tomatoes.

The aim of the study

The aim of our study is to evaluate the occurrence of clinical reactions to tomatoes and to evaluate the sensitivity, specificity and positive predictive value of tomatoes allergens Sola l and Sola l 6. Our other aim is to perform the analysis, if there is some relation between the clinical reaction to tomatoes, sensitisation to Sola l, Sola l 6 and molecular components of latex (Hev b 1, Hev b 3, Hev b 5, Hev b 6.02, Hev b 8, Hev b 11), birch (Bet v 1), celery (Api g 1, Api g 2, Api g 6), carrot (Dau c, Dau c 1), apple (Mal d 1, Mal d 2, Mal d 3), potatoes (Sol t), timothy (Phl p 1, Phl p 2, Phl p 5, Phl p 6), allergen extract of cultivated rye (Secc pollen), peach (Pru p 3), peanuts (Ara h 9), hazelnuts (Cor a 9), walnuts (Jug r 3), pellitory of the wall (Par j), mugwort (Art v 3) and sunflower (Hel a).

Methods

Patients and methods

According to the Hanifin-Rajka criteria (Hanifin & Rajka, 1980), we made the diagnosis of AD. In the period 2018–2020, we included 100 patients suffering from AD in our study; in all of them the complete dermatological and allergological examination was performed. The study took place in the Department of Dermatology, Faculty Hospital Hradec Králové, Charles University, Czech Republic. Exclusion criteria were systemic therapy (cyclosporin, systemic corticoids, biological therapy), pregnancy, breastfeeding. Patients with AD having other systemic diseases were excluded from the study as well. This study was approved by Ethics Committee of Faculty Hospital Hradec Králové, Charles University of Prague, Czech Republic.

Dermatological examination

A complete dermatological examination was performed on all patients included in the study. All these patients were examined in the Department of Dermatology, Faculty Hospital Hradec Králové, Charles University, Czech Republic. The diagnosis of AD was determined according to Hanifin–Rajka's diagnostic criteria. The severity of atopic dermatitis was scored in agreement with SCORAD (Scoring of atopic dermatitis), with assessment of topography items (affected skin area), intensity criteria and subjective parameters (European Task Force on Atopic Dermatitis, 1993).

To measure the extent of AD, the rule of nines was applied on a front/back drawing of the patient’s inflammatory lesions. The extent was graded 0–100 points. The intensity part of the SCORAD index consists of six items: erythema, oedema/papulation, excoriations, lichenification, crusts and dryness. Each item was graded on a scale 0–3. The subjective items included daily pruritus and sleeplessness. Both subjective items were graded on a 10-cm visual analogue scale and the maximum subjective score was 20 points. All items were filled out in the SCORAD evaluation form. The SCORAD index formula was A/5 + 7B/2 + C. In this formula, A is defined as the extent (0–100 points), B as the intensity (0–18 points) and C as the subjective symptoms (0–20 points). The severity of AD was evaluated with SCORAD as a mild form to 25 points, as moderate over 25–50 points, as a severe form over 50 points (European Task Force on Atopic Dermatitis, 1993).

Inclusion criteria: (1) age 14 years and over (2) AD as defined by the criteria of Hanifin and Rajka.

Exclusion criteria: pregnancy, breastfeeding, systemic therapy (cyclosporin, systemic corticoids, biologics).

Allergological examination

A complete allergological examination was performed in all patients included in the study evaluating the occurrence of bronchial asthma and allergic rhinitis.

Bronchial asthma

The diagnosis of bronchial asthma (AB) was determined according to the guidelines of the Global Initiative for Asthma (GINA) at the allergy outpatient clinic of the Institute of Clinical Immunology and Allergology, Faculty Hospital Hradec Kralove, Czech Republic (Global Initiative for Asthma. Global Strategy for asthma management and prevention—Update 2015. www.ginasthma.com).

Allergic rhinitis

The evaluation of allergic rhinitis (AR) was made according to the allergy testing and personal history of the Institute of Clinical Immunology and Allergology, Faculty Hospital Hradec Kralove, Czech Republic. AR was defined as a process that included three cardinal symptoms: sneezing, nasal obstruction, and mucus discharge. Symptoms occur with allergen exposure in the allergic patient.

Examination of specific IgE to allergen reagents

For the detection of specific IgE against molecular components of tomatoes and other allergens, we used the ALEX2 Allergy Xplorer multiplex method (MacroArray Diagnostics GmbH, Vienna, Austria). ALEX² is an immunoassay test based on enzyme-linked immunosorbent assay (ELISA) (Hanifin & Rajka, 1980; Heffler et al., 2018). This array contains 295 allergen reagents (117 allergenic extracts and 178 molecular components), with a large majority of aeroallergen families and cross-reactive food allergens being represented. The ALEX2 Allergy Xplorer measuring range for specific IgE is 0.3–50 kU_A/L (quantitative) and for total IgE is 1–2500 kU/L (semiquantitative). The sample requirement is 100 µL serum or plasma. The results are expressed as Class 0 (< 0.3 kU_A/L) – negative, Class 1 (0.3–1 kU_A/L) – low positivity, Class 2 (1–5 kU_A/L) – moderate positivity, Class 3 (5–15 kU_A/L) – high positivity, and Class 4 (> 15 kU_A/L) – very high positivity. As far as molecular (recombinant) allergens are concerned, these allergens are “produced” by the Escherichia coli; there are their own clones (and their own biobank) for each of these allergens. The exact amount of allergen in one batch varies a bit because allergens associate with magnetic particles. Next, the reactivity of this activated allergen is measured, which ultimately determines the final concentration. The volume of each allergen (regardless of whether it is an extract or a recombinant) is 20 nL. The ALEX2 Allergy Xplorer is commercially available, having attained CE certification, which assures that the quality of the assay, regarding LoD, precision and repeatability as well as specificity and linearity, is in line with in vitro diagnostic features. There is no significant interference from high total IgE, haemoglobin, bilirubin or triglycerides. The patients were examined on a batch of kit (LOT) 02ADH01.

ALEX2 Allergy Xplorer-working procedure

Initially, allergens are coupled to activated nanoparticles, for coupling individual and combinatorial optimisation. Each allergen is attached reflecting its biochemical properties and specific requirements for stability, thereby preserving the full epitope complexity. The nanoparticles multiply the surface of the solid-phase presenting the allergen during the immunoassay, enabling highly sensitive detection. In the next step, the allergen-bearing nanoparticles are deposited onto a solid-phase matrix, forming a macroscopic array of individual assay parameters. The different allergens and components, spotted onto a nitrocellulose membrane as immunosorbent in a cartridge chip, are incubated with 0.5 mL of a 1:5 dilution of serum under agitation, the serum diluent containing a cross-reactive carbohydrate determinants (CCDs) inhibitor. After incubation for 2 h, the chips are extensively washed. A pretitered dilution of anti-human IgE labeled with alkaline phosphatase is added and incubated for 30 min. Following another washing cycle, the enzyme substrate is added, and after a few minutes, the reaction is complete. After the membranes are dried, the quantification of this colorimetric enzyme assay is performed with an easy-to-use and affordable image explorer. The image acquisition and analysis of a single test takes only a few seconds. The assay time is 3.5 h, and tests per run are up to 50 per operator, with manual processing.

Confirmation of clinical reaction to tomatoes

The clinical reactions to tomatoes were confirmed in patients with the clinical symptoms in the open exposure test or in the history (early and/or late reactions). Food allergy to tomatoes was confirmed in patients with clinical reaction to tomatoes and with positive results of specific IgE to molecular component Sola l 6 and/or to allergen extract Sola l. Food intolerance was confirmed in patients with positive reaction to tomatoes and with negative specific IgE. Sensitisation to tomatoes was confirmed in patients with positive results in ALEX2 Allergy Xplorer and with the negative results in the open exposure test.

Recommendation of the open exposure test with tomatoes

The open exposure test (OET) with tomatoes after the diagnostic hypoallergenic diet was recommended in patients suffering from moderate and severe form of AD. In patients with a mild form of AD, this test was recommended according to the positive results in ALEX 2 Allergy Xplorer test; in these patients, we recommended the elimination diet with the elimination of tomatoes in the period of 4 weeks. In patients with early reactions after tomatoes ingestion in their history, the open exposure test was not performed because of anaphylactic reaction danger.

Diagnostic hypoallergenic diet

In the period of 4 weeks, we recommended these foods: gluten-free foods, potatoes, rice, meat-beef, pork, vegetable, and fruits only after thermal modification, but parsley, celery, and seasoning were not allowed. The patients are allowed to drink only drinking water, mineral water, or black tea.

Open exposure test

Before performing the exposure test, medications that could affect the results of the examination (antihistamines), were stopped for at least 5 days; other local skin treatments remained unrestricted. If the patient did not have an acute systemic anaphylactic reaction in the past and tomatoes were part of the normal diet, the test was performed in the home environment. Patients with a more severe form of bronchial asthma and/or after experiencing a systemic allergic reaction were admitted for short-term hospitalisation to perform OET with a suspected food. OET was not performed in patients with a history of an early allergic reaction after peanut ingestion due to the risk of an anaphylactic reaction.

Open exposure test was performed after this diet in intervals without symptoms or during a consistent period with regard to AD (not in pollen season in patients with pollen allergy). In general, the open exposure test was recommended to perform with three doses of examined foods given during two days. One fresh tomato was given in incremental dosages in 10 min intervals during 1 h. The food challenge results were scored as positive if one or more of the following objective and subjective clinical reactions were noted: itching, rash, urticaria, angioedema, vomiting, wheezing, abdominal pain, diarrhoea, pruritus or worsening of atopic dermatitis. Early reactions were defined as clinical symptoms within 2 h after the ingestion of food and late symptoms if occurring after more than 2 h. In the case the physician or the patient recorded worsening of the atopic dermatitis or some other reactions during the open exposure test, the patient continued in the elimination of suspected food and the severity of atopic dermatitis was evaluated every 3 months during 1 year. If the open exposure test was negative, the patient introduced tomatoes in the diet regimen. The severity of atopic dermatitis was evaluated during the ingestion of tomatoes over a period of 3 months during 1 year (Čelakovská et al., 2020).

Control group

When introducing the test into the clinical examination, we also examined 15 healthy volunteers – blood donors (equivalent to age, male and female representation). All of these blood donors had in the multiplex examination (ALEX 2–Allergy Xplorer) the specific IgE negative – expressed as Class 0 (< 0.3 kU_A/L).

Statistical analysis

We compared the results of specific IgE to allergen reagents of tomatoes (Sola l, Sola l 6) in the examination of ALEX2 Allergy Xplorer testing and clinical reactions to tomatoes in the open exposure test (patient’s history) test. The sensitivity, specificity and positive predictive value were calculated (we show rates and 95% confidence intervals value, lower, upper).

Sensitivity was calculated as the proportion of positive IgE antibody tests in patients with allergic reactions to tomatoes. Sensitivity (true positive rate) is the probability of a positive test result, conditioned on the individual truly being positive. Sensitivity refers to the test's ability to correctly detect ill patients out of those who do have the condition. Specificity (true negative rate) is the probability of a negative test result, conditioned on the individual truly being negative. Specificity refers to the test's ability to correctly reject healthy patients without a condition. Positive predictive value is the ratio of patients truly diagnosed as positive to all those who had positive test results (including healthy subjects who were incorrectly diagnosed as patients). This characteristic can predict how likely it is for someone to truly be patient, in case of a positive test result. Fisher's Exact test was used to measure the relationship.

We calculated the relation between the result of specific IgE to other molecular components and specific IgE to Sola l, Sola l 6 and clinical reaction to tomatoes; for this evaluation, we used Fisher's Exact test.

We calculated the AUC (area under the curve) for specific IgE of Sola l, Sola l 6 and other molecular components significantly associated with sensitisation to tomatoes.

We evaluated the relation between the occurrence of clinical reaction to tomatoes, positive results of specific IgE to molecular components Sola l and Sola l 6 (sensitisation) and the severity of AD with Fisher's Exact test in the contingency tables.

We used statistical software: NCSS 2021 Statistical Software (2021). NCSS, LLC. Kaysville, Utah, USA, css.com/software/ncss. The level of significance was set to α = 0.05.

Results

Characteristic of patients

A total of 100 AD patients were included in the study (49 men and 51 women with the average age 40.6 years: min. age 14 years, max. age 67 years and with the average SCORAD 39 points, SD 13.5 points). The mild form of AD was recorded in 14 patients (14%), moderate form in 61 patients (61%), severe form in 25 patients (25%); 55 patients (55%) suffer from bronchial asthma and 74 patients (74%) suffer from allergic rhinitis.

Results of specific IgE to Sola l and Sola l 6 expressed in Class 0–Class 4.

In Table 2, we show the results of specific IgE to molecular component Sola l 6 and allergen extract Sola l. The positive results of specific IgE to Sola l were confirmed in 10 patients (10%), low positivity was confirmed in 6 patients (6%), moderate positivity in 3 patients (3%) and high positivity in 1 patient (1%). The positive results of specific IgE to Sola l 6 were confirmed in five patients (5%), low positivity was confirmed in one patient (1%), moderate positivity in three patients (3%) and high positivity in one patient (1%).

Table 2. The level of specific IgE to Sola l and Sola l 6 in classes 0, 1, 2, 3, 4 in 100 AD patients (=100%).

	The number of patients (%) according to the level of specific IgE in classes 0–4 (100 patients included in the study = 100%)
	Class 0	Class 1	Class 2	Class 3	Class 4
Sola l	90.0%	6.0%	3.0%	1.0%	0.0%
Sola l 6	95.0%	1.0%	3.0%	1.0%	0.0%

Class 0 – negative (< 0.3 kU_A/L); Class 1 – low positivity (0.3–1 kU_A/L); Class 2 – moderate positivity (1–5 kU_A/L); Class 3 – high positivity (5–15 kUA/L); Class 4 – very high positivity (> 15 kU_A/L). The positive results of specific IgE to Sola l were confirmed in 10 patients (10%), low positivity was confirmed in 6 patients (6%), moderate positivity in 3 patients (3%) and high positivity in 1 patient (1%). The positive results of specific IgE to Sola l 6 were confirmed in 5 patients (5%), low positivity was confirmed in 1 patient (1%), moderate positivity in 3 patients (3%) and high positivity in 1 patient (1%).

Clinical reactions to tomatoes

The reaction to tomatoes with clinical symptoms was confirmed in 15 patients (15%) according to the convincing history and/or open exposure test. We recorded in seven patients (7%) early reaction such as oral allergy syndrome, maculopapulous rash and pruritus of the skin, in eight patients (8%), we recorded a combined reaction (early reaction with pruritus of the skin and the late reaction with the worsening of AD).

Sensitivity, specificity and positive predictive value of allergen reagents in ALEX 2 Allergy Xplorer test in patients with clinical reaction to tomatoes

We compared the results in the examination of ALEX2 Allergy Xplorer testing and really confirmed clinical reactions in the open exposure (patient’s history) test; according to these results, the sensitivity, specificity and positive predictive were calculated with Fisher's Exact test. In patients with clinical reaction to tomatoes, we confirmed the positive results to specific IgE in one patient (1%) to Sola l and in another one patient (1%) to Sola l 6. The results of specific IgE to molecular components of tomatoes and the occurrence of clinical reactions to tomatoes are recorded in Table 3. For molecular component Sola l 6, the sensitivity is 6.7% (0.17–31.9%), the specificity 95.3% (88.4–98.7%), positive predictive value 20.0% (5.1–71.4%). For allergen extract Sola l, the sensitivity is 6.7% (0.17–31.9%), the specificity 89.4% (80.1–95.0%), positive predictive value 10.0% (2.5–44.5%). According to these results, we confirmed food allergy to tomatoes in 2 patients (2%) and food intolerance in 13 patients (13%). In complement to Table 3, we show area under the curve (AUC) for allergens Sola l and Sola l 6.

Table 3. Sensitivity, specificity and positive predictive value of ALEX2 Allergy Xplorer.

Results of specific IgE	Number of patients – clinical reaction to tomatoes		p-value	Test characteristics	Rates and 95% confidence intervals
	Yes 15 (15%)	No 85 (85%)			Value	Lower	Upper
Sola l positive	1	9	1.000	Sensitivity	0.0667 (6.67%)	0.0017 (1.70%)	0.3195 (31.95%)
Sola l negative	14	76		Specificity	0.8941 (89.41%)	0.8085 (80.85%)	0.9504 (95.04%)
				Positive predictive value	0.1000 (10.0%)	0.0025 (0.25%)	0.4450 (44.5%)
Sola l 6 positive	1	4	0.564	Sensitivity	0.0667 (6.67%)	0.0017 (1.70%)	0.3195 (31.95%)
Sola l 6 negative	14	81		Specificity	0.9529 (95.29%)	0.8839 (88.39%)	0.9870 (98.70%)
				Positive predictive value	0.2000 (20.0%)	0.0051 (0.51%)	0.7164 (71.64%)
Complement to this table: area under the curve (AUC) for allergens Sola l and Sola l 6.
			Upper
Criterion	Count	AUC	Standard error	1-Sided	95% Confidence Limits
				P-Value	Lower	Upper
Sola_l	100	0.5184	0.0746	0.4024	0.3577	0.6493
Sola_l_6	100	0.4996	0.0374	0.5042	0.4229	0.5692

Definitions: Criterion is the Criterion Variable containing the scores of the individuals. Count is the number of the individuals used in the analysis. AUC is the area under the ROC curve using the empirical (trapezoidal) approach. Standard Error is the standard error of the AUC estimate. P-Value is the probability level to Test AUC > 0.5. The lower and upper confidence limits form the confidence interval for AUC.

Sensitisation to tomatoes

We evaluated the sensitisation to tomatoes (=no clinical reaction to tomatoes, but positive results of specific IgE to molecular component and/or to allergen extract of tomatoes). We confirmed the positive results of specific IgE to allergen extract Sola l in 9 patients (9%), to Sola l 6 in 4 patients (4%), altogether in 10 patients (10%).

The relation between sensitisation to tomatoes, reaction to tomatoes and the specific IgE to molecular components of birch, vegetables, fruits, seeds and nuts

We show the relation between the sensitisation to molecular components of birch, vegetables, fruits, seeds, nuts and sensitisation to molecular component of tomatoes Sola l 6 and allergen extract Sola l and clinical reaction to tomatoes (Table 4).

Table 4. The significant relation (p-value <0.05) between the occurrence of clinical reactions to tomatoes, results of specific IgE to Sola l and Sola l 6 (sensitisation) and the results of specific IgE to other molecular components and allergens.

	Specific IgE to allergen Sola l and molecular component Sola l 6		Clinical reaction to tomatoes
	Sola l	Sola l 6 (nsLTP typ 2)
Specific IgE to allergens and molecular components of pollen, seeds, fruits and vegetables and latex
Hev b 1(Rubber Elongation Factor)
Hev b 3 (Small rubber particle proteins)
Hev b 5 (Acidic structural protein)
Hev b 6.02
Hev b 8 Profilin (actin-binding protein)
Hev b 11 (Chitinase class I)
Bet v 1 (PR 10 protein)
Api g 1 (PR-10 protein)
Api g 6 (Non-specific lipid transfer protein type 2)	<0.05	<0.05
Api g 2 (Non-specific lipid transfer protein, type 1)	<0.05	<0.05
Dau c (Carrot)
Dau c 1 (PR 10 protein)
Mal d 1 (PR 10 protein)			<0.05
Mal d 2 (Thaumatin-like protein)
Mal d 3 (Non-specific lipid transfer protein type 1)	<0.05	<0.05
Sol t (Potatoes)	<0.05
Phl p 1 (Beta-expansin)
Phl p 2 (Grass group II/III)
Phl p 5 (Timothy)
Phl p 6 (Timothy)
Secc pollen (Cultivated rye)
Pru p 3 (Non-specific lipid transfer protein)	<0.05	<0.05
Ara h 9 (Non-specific lipid transfer protein)	<0.05	<0.05
Jug r 3 (Non-specific lipid transfer protein)	<0.05	<0.05
Par j (Pellitory of the wall)
Art v 3 (Non-specific lipid transfer protein)	<0.05	<0.05
Cor a 9 (11S globulin)
Hel a (Sunflower)	<0.05

In patients with the positive results of specific IgE to Sola l and Sola l 6 (=sensitisation), we confirmed significantly higher sensitisation to molecular components Api g 2, Api g 6, Mal d 3, Pru p 3, Ara h 9, Jug r 3, Art v 3; in addition in patients with sensitisation to allergen extract Sola l, we confirmed significantly higher sensitisation to allergen extract Sol t (potatoes) and Hel a (Sunflower). The significant relation was confirmed between the results of specific IgE to Mal d 1 and patients with reaction to tomatoes.

In patients with sensitisation to Sola l and Sola l 6, we confirmed significantly higher sensitisation to molecular components Api g 2, Api g 6, Mal d 3, Pru p 3, Ara h 9, Jug r 3, Art v 3; in addition in patients with sensitisation to allergen extract Sola l, we confirmed significantly higher sensitisation to allergen extract Sol t (potatoes) and Hel a (Sunflower), (Table 4). These significant associations are demonstrated in graphs. Figure 1 shows the significant relationship between the results of specific IgE to molecular allergen Sola l and specific IgE to other molecular components. Figure 2 shows the significant relationship between the results of specific IgE to molecular component Sola l 6 and specific IgE to other molecular components. Values are the percentage of true positives and true negatives within Sola l and Sola l 6, respectively.

Figure 1. The graph shows the significant relationship between the results of specific IgE to molecular allergen Sola l and the results of specific IgE to other molecular components. Values are the percentage of true positives (blue) and true negatives (orange) within Sola l.

Figure 2. The graph shows the significant relationship between the results of specific IgE to molecular component Sola l 6 and the results of specific IgE to other molecular components. Values are the percentage of true positives (blue) and true negatives (orange) within Sola l 6.

According to our result, the significantly lower sensitisation to Mal d 1 (apple, PR 10 protein) is confirmed in patients with clinical reaction to tomatoes. No significant relation between the occurrence of clinical reaction to tomatotes and the results of specific IgE to molecular components of latex, birch, celery, carrot, potatoes, timothy, rye, peach, peanuts, walnuts, hazelnuts, pellitory of the wall, mugwort and sunflower was confirmed (Figures 3 and 4; Table 4).

Figure 3. In patient with positive results of specific IgE to Sola l, we show the significant associations with other molecular components. Sensitivity vs 1-specificity plots – curves of parameters.

Figure 4. In patient with positive results of specific IgE to Sola l 6, we show the significant associations with other molecular components. Sensitivity vs 1-specificity plots – curves of parameters.

The calculation of AUC (area under the curve)

Empirical AUC (area under the curve) with confidence intervals for each component with a significance level of the AUC = 0.5 test against the alternative that AUC > 0.5 is recorded in Table 5 for Sola l and in Table 6 for Sola l 6. Based on the AUC value, it can predict whether a patient is likely to react to foods containing allergens associated with sensitisation to Sola l – Table 5 and with sensitisation to Sola l 6 – Table 6.

Table 5. In patients with positive results of specific IgE to Sola l, we show empirical AUC (area under the curve) with confidence intervals for each component with a significance level of the AUC = 0.5 test against the alternative that AUC > 0.5.

Upper
			Standard	1-Sided	95% Confidence limits
Criterion	Count	AUC	Error	P-Value	Lower	Upper
Api_g_2	100	0.7833	0.0763	0.0001	0.5829	0.8939
Api_g_6	100	0.7283	0.0917	0.0064	0.4947	0.8638
Mal_d_3	100	0.7300	0.0882	0.0046	0.5070	0.8614
Sol_t	100	0.7639	0.0866	0.0011	0.5357	0.8881
Pru_p_3	100	0.7322	0.0854	0.0033	0.5174	0.8602
Ara_h_9	100	0.8083	0.0834	0.0001	0.5719	0.9208
Jug_r_3	100	0.7772	0.0880	0.0008	0.5390	0.9003
Art_v_3	100	0.6711	0.1028	0.0480	0.4188	0.8272
Cor_a_9	100	0.6783	0.0911	0.0251	0.4584	0.8200
Hel_a	100	0.7472	0.0882	0.0025	0.5193	0.8759

AUC is the area under the ROC curve using the empirical (trapezoidal) approach. Standard error is the standard error of the AUC estimate. P-Value is the probability level to Test AUC > 0.5. The lower and upper confidence limits form the confidence interval for AUC.

Table 6. In patients with positive results of specific IgE to Sola l 6, we show empirical AUC (area under the curve) with confidence intervals for each component with a significance level of the AUC = 0.5 test against the alternative that AUC > 0.5.

Upper
			Standard	1-Sided	95% Confidence limits
Criterion	Count	AUC	Error	P-Value	Lower	Upper
Api_g_2	100	0.8158	0.1588	0.0234	0.2103	0.9689
Api_g_6	100	0.9389	0.0618	0.0000	0.6084	0.9919
Mal_d_3	100	0.8779	0.1045	0.0001	0.4407	0.9785
Pru_p_3	100	0.7832	0.1287	0.0139	0.3807	0.9362
Ara_h_9	100	0.7600	0.1409	0.0325	0.3295	0.9289
Jug_r_3	100	0.8758	0.1082	0.0003	0.4195	0.9788
Art_v_3	100	0.8147	0.1209	0.0046	0.4103	0.9514
Cor_a_9	100	0.7421	0.1388	0.0406	0.3360	0.9155

AUC is the area under the ROC curve using the empirical (trapezoidal) approach. Standard error is the standard error of the AUC estimate. P-Value is the probability level to Test AUC > 0.5. The lower and upper confidence limits form the confidence interval for AUC.

The relation between the severity of AD and the clinical reaction to tomatoes and the results of specific IgE to molecular components Sola l 6 and Sola l

In Table 7, we show the association between the severity of AD and the clinical reaction to tomatoes and the association between the severity of AD and the results of specific IgE to molecular components Sola l 6 and Sola l. The association between the severity of AD and the clinical reaction to tomatoes and positive results of specific IgE to Sola l 6 and Sola l (sensitisation) was not confirmed.

Table 7. The relation between the occurrence of clinical reaction to tomatoes, positive results of specific IgE to molecular components Sola l and Sola l 6 (sensitisation) and the severity of AD.

	Severity of atopic dermatitis – number of patients
	Mild	Moderate	Severe	Total
Clin. reaction +	2 (14.29%)	10 (16.39%)	3 (12.0%)	15 (15%)
Clin. reaction −	12 (85.71%)	51 (83.61%)	22 (88.0%)	85 (85%)
Total	14 (100%)	61 (100%)	25 (100%)	100 (100%)
p-Value	0.921
Sola l +	0 (0%)	7 (11.48%)	3 (12.0%)	10 (10%)
Sola l −	14 (100%)	54 (88.52%)	22 (88.0%)	90 (90%)
Total	14 (100%)	61 (100%)	25 (100%)	100 (100%)
p-Value	0.510
Sola l 6 +	0 (0%)	5 (8.2%)	0 (0%)	5 (5%)
Sola l 6 −	14 (100%)	56 (91.8%)	25 (100%)	95 (95%)
Total	14 (100%)	61 (100%)	25 (100%)	100 (100%)
p-Value	0.317

The association between the severity of AD and the clinical reaction to tomatoes and positive results of specific IgE to Sola l 6 and Sola l (sensitisation) was not confirmed. Sola l+: positive results of specific IgE, Sola l−: negative results of specific IgE, Sola l 6 +: positive results of specific IgE, Sola l 6 −: negative results of specific IgE.

Discussion

Studies dealing with food allergy to tomatoes focus usually on flare-ups in the general population and do not focus on AD patients. There are no studies investigating the benefit of ALEX2 Allergy Xplorer test in AD patients with reactions to tomatoes. In general population, the tomato has been confirmed as one of the most prevalent allergenic vegetables (Ballmer-Weber & Hoffmann-Sommergruber, 2011; Burney et al., 2014; Geroldinger-Simic et al., 2011; Ortolani et al., 1989; Petersen et al., 1996). According to some studies, the prevalence of tomato allergy is approximately 1.7–9.3% in different populations of Europe with an average of 4.9% (Burney et al., 2014; Ortolani et al., 1989; Petersen et al., 1996), up to 16% of the adults in Italy (Ortolani et al., 1989). This allergy mostly affects the population of southern Europe, so the majority of studies are dedicated to Italy and Spain (Bencivenni et al., 2012; López-Matas et al., 2015).

In our study, we evaluated clinical reactions to tomatoes in AD patients and the sensitivity, specificity and positive predictive value of ALEX2 Allergy Xplorer test. Results of our study show, that clinical reactions to tomatoes are quite common – in 15% of AD patients. These reactions manifest as early reactions such as oral allergy syndrome (in 7% of patients), maculopapulous rash, pruritus of the skin and new eczematic lesions with worsening of AD (in 8% of patients). The food allergy (positive results of specific IgE to molecular component Sola l 6 and/or to allergen extract Sola l and clinical reaction to tomatoes) was confirmed in 2% of patients. We calculated the characteristics of the test, our results show the low sensitivity and low positive predictive value of specific IgE to molecular component Sola l 6 and allergen extract Sola l in ALEX2 Allergy Xplorer test. The specificity of this test is good (for Sola l is 89.4%, for Sola l 6 it is 95.3%). The explanation for this is that most patients may either have a reaction of food intolerance or other molecular components of tomatoes play a role. The explanation of the low sensitivity of Sola l and Sola l 6 may be in fact, that the levels of specific IgE antibodies needed to trigger an allergic reaction can vary among individuals. Some individuals might have low levels of IgE antibodies to Sola l and Sola l 6, which are below the threshold required to elicit a clinical reaction. For this reason, we recommend that patients with moderate and severe AD should be put on an elimination diet and undergo a tomato exposure test, even if the specific IgE against tomato allergens Sola l and Sola l 6 is negative. In our study, we were able to observe that after eliminating tomatoes from the diet of patients with a clinical reaction to tomatoes, skin findings improved and skin itching decreased.

Another explanation for the low sensitivity of Sola l and Sola l 6 may be in the cross-reactivity. It is possible that individuals with clinical reactions to tomatoes might have cross-reactive IgE antibodies that primarily target different allergens but still result in tomato-related symptoms.

Cross-reactivity to tomato was described for up to 28% of latex-allergic patients (Beezhold et al., 1996). Foetisch et al. (2001) showed that 9% of birch-pollen-allergic patients had adverse reactions to tomato. In another study, published by de Martino et al. (1988), 39% of grass pollen-allergic children were sensitised to tomato. Allergy to tomatoes is also linked to other allergies, such as grass pollen and latex allergy.

Genetic factors play a role in determining an individual's propensity to develop allergies and how their immune system responds to specific allergens. Genetic variation could contribute to the varying results observed in IgE testing for Sola l and Sola l 6.

On the other hand, results of our study show a relatively frequent positivity of specific IgE to Sola l and/or Sola l 6 in AD patients without clinical reactions to tomato. It is a mere sensitisation without clinical manifestations, but patients are educated and are cautious when consuming tomatoes. We also investigated whether there is a significant relationship between the reaction to tomatoes and the results of specific IgE to other molecular components, but we did not confirm in patients suffering from clinical reaction to tomatoes the significantly higher occurrence of positive results of specific IgE to molecular components of latex, birch, celery, carrot, potatoes, timothy, rye, peach, peanuts, walnuts, hazelnuts, pellitory of the wall, mugwort and suflower. The significantly lower sensitisation to Mal d 1 (apple, PR 10 protein) is confirmed in patients with clinical reaction to tomatoes.

In patients with sensitisation to Sola l and Sola l 6 without clinical reaction to tomatoes, we confirmed significantly higher sensitisation to molecular components of non-specific lipid transfer protein (nsLTP) – Api g 2, Api g 6, Mal d 3, Pru p 3, Ara h 9, Jug r 3, Art v 3; in addition, in patients with sensitisation to allergen extract Sola l we confirmed significantly higher sensitisation to allergen extract Sol t (potatoes) and Hel a (Sunflower). Some of these patients with sensitisation to Sola l and Sola l 6 report clinical reactions to peaches, celery and peanuts. After ingestion of these foods, patients observed oral allergy syndrome, pruritus of the skin and worsening of AD. The explanation is in the fact, that molecular components Sola l 6, Api g 2, Api g 6, Mal d 3, Pru p 3, Ara h 9, Jug r 3, Art v 3 belong to the group of non-specific lipid transfer protein (nsLTP). These allergens have the potential to sensitise and provoke symptoms to a large number of plant foods, including those which are raw, cooked or processed (Cubells-Baeza et al., 2017; Skypala et al., 2021). Sensitisation to nsLTP can occur in several different ways. It is unknown why some sensitised individuals develop clinical symptoms to foods whereas others do not, or what other allergens besides Pru p 3 may be primary sensitising allergens. In early life, sensitisation to foods such as milk, egg and peanut may occur through cutaneous exposure due to skin barrier disruption and exposure via the epi-cutaneous route (Jiménez-Saiz et al., 2019; Kong et al., 2015). Sensitisation to nsLTP may also occur through the cutaneous route; it is reported that peach-induced contact urticaria has been linked to nsLTP sensitisation (Asero, 2011). Sensitisation to foods via the gastrointestinal tract is a common cause of food-allergic reactions. Tordesillas et al. (2013) confirmed that Pru p 3 has a high capacity to cross the gastric barrier. Tomás-Pérez et al. studied the different sensitisation patterns in patients who are allergic to potatoes and/or tomatoes. The allergen sensitisation patterns were similar in all the patients that had been studied and a new allergen involved in the allergy to solanaceae plants has been detected (Tomás-Pérez et al., 2019). Welter et al. carried out the study to test the hypothesis that individual variability to tomatoes is based on differential reactions of single subjects to particular allergens in tomato fruits of plants with certain genetic backgrounds or cultivated under distinct conditions. The hypothesis that high interindividual differences in allergic reactions are based on the interactions between the IgE of allergic subjects with particular allergens was rejected; five proteins with putative clinical relevance as tomato allergens were identified (Mascheri et al., 2015; Welter et al., 2013).

In our study, all patients with Sola l and Sola l 6 positivity were advised to be cautious when consuming other foods containing allergens containing nsLTP.

In majority of AD patients with clinical reaction of tomatoes, we assume that these are non-allergic reactions such as food intolerance, or other molecular components probably may be involved in tomatoes allergy. Patients may be sensitised to molecular components such as Sola l 1 (Profilin), Sola l 2 (Beta-fructofuranosidase), Sola l 4 (PR 10 protein) and Sola l 5 (Cyclophilin). The first evidence of recognising β-fructofuranosidase as a potential allergen was shown in the research of Petersen et al. (1996). In this research, the allergenicity of Sola l 2 was tested by checking the IgE reactivity to tomato fruits (Petersen et al., 1996). The tomato profilin is a minor allergen in tomato fruit; it shows biological activity, as confirmed by in vitro histamine release assays with human basophils and thereby has the potential to account for clinical symptoms in tomato-allergic patients (Petersen et al., 1996; Westphal et al., 2004). On the other hand, Hossin et al. (2012) investigated the anti-allergic potential of red tomato peel and flesh extracts and yellow tomato peel extract. This study confirmed that the red tomato peel extract has a more effective anti-allergic activity compared to other analysed extracts. In the peel of red tomato, the level of lycopene is much higher than in the flesh or in the peel of yellow tomatoes. These findings lead to the conclusion that certain antioxidants may have an anti-allergic effect (Hossin et al., 2012). Even though the antioxidants from tomatoes may have anti-allergic potential, they cannot counteract the allergy for tomatoes. The antioxidants mentioned above may help with the allergenic reaction to house dust or mites, but both the flesh and the tomato skin are, at the same time, the sources of tomato allergens (Hossin et al., 2012).

According to some studies, 22% of tomato-allergic patients are sensitised to Sola l 1 (profilin) (Westphal et al., 2004); approximately 17% of the patients are sensitised to Sola l 2 (β-fructofuranosidase), (Asero et al., 2008; Westphal et al., 2004) and up to 15% of Italian tomato-allergic patients were mono-sensitised to the non-specific lipid transfer protein from tomato (Sola l 3), (Palomares et al., 2005). Pectinesterase, polygalacturonase 2 A, superoxidedismutase (Westphal et al., 2004), an osmotin-like protein, glucanase, and chitinase, pepsin-resistant protein are potential tomato allergens (Diaz-Perales et al., 1999; López et al., 2011; Palomares et al., 2005).

According to Asero, who examined 96 adults with plant food allergy, tomato allergy occurs via sensitisation towards different proteins and component-resolved diagnosis helps to define clinical subgroups with different risk levels. In their study, IgE levels were not associated with tomato allergy in any of the subgroups. Similarly, birch and peach-specific IgE levels were not associated with tomato allergy in PR-10/profilin or in LTP reactors, respectively (Asero, 2013). Hegde et al. studied the association between grass pollen sensitisation and food allergy to tomato in Indian population. They investigated 246 allergic rhinitis / asthma patients; tomato profilin appears to be an important cross-sensitising panallergen in respiratory allergic patients in the Indian subcontinent (Hegde et al., 2018).

In Dölle’s study, patients with a history of tomato allergy were recruited for skin allergy tests. Tomatoes carrying distinct genotypes were grown under various cultivation conditions and harvested at different maturation stages. According to their genetic background and maturity level, some tomato cultivars elicit positive reactions in tomato-allergic patients in the skin allergy test (Dölle et al., 2011). López-Matas et al. characterised the allergenic profile of 6 commonly ingested varieties of tomato and to identify and quantify the most representative allergens. Their results demonstrate differences in the antigenic and allergenic profiles of the analysed varieties and variations in their allergenic composition, affecting the diagnostic capacity (López et al., 2011).

We also evaluated the relation between the occurrence of clinical reaction to tomatoes, positive results of specific IgE to molecular components Sola l and Sola l 6 (sensitisation) and the severity of AD. The association between the severity of AD and the clinical reaction to tomatoes and positive results of specific IgE to Sola l 6 and Sola l (sensitisation) was not confirmed. On the other hand, in our previous study, the significant dependence between the severity of AD and the occurrence of food hypersensitivity reactions was confirmed; 96% of patients with a severe form of atopic dermatitis suffer from food reactions. In evaluating the single foods, the significant dependence was found between the severity of AD and the reactions to nuts, apples and fishes (Čelakovská & Bukač, 2015).

Conclusion

According to our results, 15% of AD patients suffer from clinical reaction to tomatoes; the food allergy to tomatoes was confirmed with ALEX2 Allergy Xplorer test only in a minority of them. The reaction of food intolerance or food allergy to other molecular components of tomatoes may play a role. The positive results of specific IgE against molecular components of tomato without clinical reaction are observed in another 10% of AD patients. In these patients, the significantly higher sensitisation to other non-specific lipid transfer proteins is observed. Our results show the low sensitivity and low positive predictive value of specific IgE to molecular component Sola l 6 and allergen extract Sola l in ALEX2 Allergy Xplorer test, the specificity of this test is higher. For this reason, we recommend that patients with moderate and severe AD be placed on an elimination diet and that a tomato exposure test be performed. In our study, we were able to observe that after eliminating tomatoes from the diet of patients with a clinical reaction to tomatoes, skin findings improved and skin itching decreased.

ALEX2 Allergy Xplorer test is valuable tool because it gives us a comprehensive picture of sensitisation and the possibility of evaluation of cross-reacting allergens.

Acknowledgements

This study was approved by Ethics Committee of the Faculty Hospital Hradec Králové, Charles University of Prague, Czech Republic. Reference number is: 2018 07 S 04 P. The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board-Ethics committee of the Faculty Hospital Hradec Králové, Charles University of Prague, Czech Republic. Data of Approval: 3 June 2018. Compliance with ethics guidelines: The study was conducted in accordance with the Helsinki Declaration of 1964 and all subsequent amendments, and all patients provided written informed consent. Patient-level data used for this analysis were de-identified.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Additional information

Funding

This work was supported by Charles University of Prague Cooperation [grant number 207034].