IgE Antibody Associations With Allergic Disease Phenotypes Using ISAC and ALEX Assays

Over the past two decades, molecular allergen research has significantly advanced allergy diagnostics 1. This progress has deepened our understanding of IgE-mediated responses, establishing clear links between specific allergens and clinical reactions. For example, studies have shown associations such as polcalcins with asthma 2 and profilin reactivity with rhinitis 3. Additionally, outcomes in food allergy vary based on sensitisation to lipid transfer proteins (LTPs), PR-10 or seed storage protein 4. Multiplex diagnostic assays now enable comprehensive profiling of individual reactivity across various conditions, including allergies and hyper-IgE syndromes 5. These assays can evaluate up to 300 molecular allergens simultaneously. For instance, the severity of LTP hypersensitivity often correlates with the number of reactive molecules 6. Similarly, research on PR-10 has distinguished patients with pollen allergy from those also reacting to food 7. This study compared the ISAC112 (Thermo Fisher Scientific) and ALEX-2 (Macro Array Diagnostics) platforms in a clinical setting, assessing their diagnostic effectiveness in identifying associations between IgE reactivities and clinical patterns. Both assays followed manufacturer protocols, with results reported in standardised units or IgE response units 8. We aimed to determine whether these methods are interchangeable for clinical allergists or if they have unique characteristics that should be known before their routine use. Patients seen at the IDI-IRCCS Allergy Unit in Rome, reflecting sensitisation patterns prevalent across the Mediterranean region, were included. This cross-sectional study documented demographic profiles and clinical manifestations (food reactions, respiratory and dermatological symptoms). Patients were enrolled from January 2015 to December 2019 for the ISAC test and from January 2020 to April 2024 for the ALEX-2 test. Eligible patients, aged 1 year or older, presented with adverse reactions to foods, allergic rhinitis, bronchial asthma and/or atopic dermatitis. Symptom classification 9 preceded IgE testing, with physicians blinded to assay results. Ethical approval was obtained, and all patients provided informed consent according to the Declaration of Helsinki. The case groups comprised 9368 patients out of 53,600 individuals exhibiting notable reactivity detected by ISAC or ALEX assays. On average, over 900 patients underwent multiplex proteomic assessment annually. ISAC primarily tested 112 molecular allergens, whereas ALEX included 300 extracts and purified allergens from various sources. Allergens were categorised into subsets, including lipocalin, profilin, tropomyosin, LTP, PR-10, arginine kinase, parvalbumin, serum albumin, 2S-albumin and polcalcin, based on reactivity to at least one molecule from these allergen families. Statistical analyses were conducted using IBM-SPSS (version 29). Mann–Whitney U-tests compared continuous IgE values. Each variable was then dichotomised into negative (0.3) categories to examine proportions. Pearson's chi-square examined independence, and multinomial logistic regression evaluated associations between symptoms and allergen reactivity, adjusting for other variables. A total of 6292 and 3124 participants were screened using the ISAC test and the ALEX-2 test, respectively. ISAC patients were younger (33.4 ± 19.8 vs. 36.5 ± 25.5 years, p < 0.01); females prevailed in both subsets examined (58.5% and 62.4%, respectively; p < 0.01). Additional information about study findings are available in the following repository: https://osf.io/76ywj/. The correlation between molecular reactivity and clinical presentation was categorised by respiratory symptoms (rhinitis or asthma) or food allergy symptoms (local or moderate-to-systemic reactions) 9. Participants from 2015 to 2019 had a lower incidence of local reactions (9.6% vs. 13.5%, p < 0.01) but a higher prevalence of moderate-to-severe food reactions (39.3% vs. 20.5%, p < 0.01) compared to those from 2020 to 2024. ISAC patients showed less allergic rhinitis (46% vs. 52.4%, p < 0.01) but more frequent bronchial asthma (19.3% vs. 10.8%, p < 0.01). This pattern also reflected substantial heterogeneity of co-occurring respiratory and/or food symptoms in the two patient groups. For single food molecules identified by ISAC, local reactions were linked to kiwi Act d 1, cashew Ana o 2, birch-PR-10, walnut-LTP and profilins from birch and latex. Severe reactions were associated with Anisakis, 2S-albumins, cod parvalbumin, LTPs, rubber latex, peach-PR-10 and wheat-ω-5-gliadin. ALEX corroborated ISAC findings, linking local reactions to kiwi-thaumatin, hazelnut and strawberry-PR-10, melon-profilin and shrimp-arginine kinase. Severe reactions were associated with LTPs, Anisakis simplex, cashew, walnut, macadamia and sesame SSPs, shrimp arginine kinase, soybean PR-10, swordfish parvalbumin, wheat Tri a 19 and peach GRP. Regarding respiratory species-specific molecules tested by ISAC, rhinitis was linked to dust mites, cypress, wall pellitory and grasses. Asthma was associated with Alternaria, cat and polcalcin. ALEX identified rhinitis links to Alternaria, cypress, house dust mite Der p 23, male dog, beech PR-10, cat uteroglobin, wall pellitory and grasses. Asthma was associated with animal lipocalins, serum albumins, Fel d 1, Der p 23 and Alternaria. Food allergen family reactivity demonstrated by ISAC indicated that local reactions were linked to PR-10, profilins and LTPs, while systemic reactions correlated with parvalbumin, nsLTPs, 2S-albumins and tropomyosin. ALEX confirmed these findings, highlighting parvalbumin and tropomyosin as additional indicators (Figure 1). For respiratory protein family, ISAC identified PR-10 and profilin associations with rhinitis, while lipocalins and polcalcins were linked to asthma. ALEX validated these connections, with additional associations with serum albumins (Figure 1). This study highlights the importance of understanding specific molecular allergen associations in clinical practice. Both ISAC and ALEX-2 systems showed comparable abilities in linking specific molecules with allergic responses, despite differences in antigen availability and methodologies. As the assays were used in different patient populations and periods, common associations are most informative, while differences should be cautiously interpreted. It is also notable that the allergen content of the ISAC test has undergone several changes since 2019. These findings support tailored diagnostics and treatments, emphasising the unique features of each tool for optimal outcomes. E.S., V.V. and E.C. conducted the experiments and collected the data. E.S., M.G., E.C.G., M.L., G.M., L.P., D.Q. and A.Z. recruited the patients. E.S., T.S. and D.A. performed the statistical analysis. E.S. conceived the study and assisted in data interpretation. All authors reviewed, including LC, DV and RA, edited and approved the final manuscript. All authors consent to publication, confirm that this manuscript is original, has not been published previously, is not under consideration for publication elsewhere and has not been submitted to a preprint server. All authors have read and agreed to the published version of the manuscript. We extend our gratitude to Mr. Mauro Helmer Citterich for conducting the in vitro tests, and also to the nursing and medical staff. Their daily dedication and selflessness made it possible for us to provide care to every patient who visited our centre. The research was conducted in accordance with ethical standards outlined in the World Medical Association Declaration of Helsinki. The study protocol underwent review and approved by the IDI–IRCCS's committee, with approval number IDI-IRCCS CE|495/1. Enrico Scala has received consultant arrangements and participated in speakers' bureaus for Stallergenes and Thermo Fisher Scientific. The remaining authors declare that they have no relevant conflicts of interest. The data supporting the findings of this study are available upon request from the corresponding author. However, please note that the data are not publicly available due to privacy or ethical restrictions.