The sensitisation profile at a molecular level in plant food allergy is complex. Several allergens may be involved, which imply different potential for reaction severity.1–4 Bet v 1 homologues and profilins are mainly involved in pollen-food allergy syndrome, which in most cases consists of exclusively oral symptoms due to a cross-reactivity phenomenon with pollen proteins.5–7 In contrast, storage proteins and lipid transfer proteins (LTP) have been traditionally associated with the potential to trigger systemic more severe reactions.8–10 LTPs are considered the most relevant family of plant food allergens in Mediterranean countries.11

The potential of multiplexed molecular-based diagnostic tools, such as the ImmunoCAP ISAC microarray, to accurately characterise plant-food allergic patients and help clinicians in individual risk assessment has led to an increase in its use in clinical practice.12,13 However, certain observations make results interpretation complex and challenging for the clinician.14,15 For instance, in LTP-sensitised patients allergic manifestations can range from mild symptoms to severe anaphylaxis.11 The reason for such variability is unclear and only partially explained by LTP-specific immunoglobulin E (sIgE) levels.16–18 In addition, the range of plant-foods involved can be extremely wide, including clinical reactivity to several taxonomically unrelated plant-foods in an individual, whereas others closely related are tolerated. Likewise, sensitisation to a wide panel of plant-foods due to cross-reactivity (either by skin prick test or specific IgE) is common in LTP-sensitised patients, although its clinical significance is not always clear.18–20 Finally, most evidence comes from studies in adult populations, whereas less is known about LTP sensitisation/allergy in children.

In this context, we sought to describe the clinical pattern and sensitisation profile of children with plant-food allergy and LTP sensitisation from our Mediterranean area in Spain. In a previous study21 we described a cohort of adults from the same geographical area with LTP syndrome, i.e. clinical reactivity to taxonomically unrelated plant-foods which was exclusively explained by LTP sensitisation, in the absence of sensitisation to other plant-food allergens. Anaphylaxis and symptoms exacerbation by cofactors such as exercise or non-steroidal anti-inflammatory drugs were common. We hypothesised that a similar clinical and sensitisation profile might be found in children.

A total of 130 patients were recruited for the present study. The mean age was 10.8 years (SD; range: 3.95; 3–18). Demographics and other allergic co-morbidities of patients recruited are described in Table 1.

To our knowledge this is amongst the largest studies in children addressing molecular-based sensitisation in plant-food allergy, with special insight into LTPs. In agreement with previous studies on LTP sensitisation,11,21,25–28 symptoms severity was highly variable and a wide spectrum of plant-foods were involved in reactions. Accordingly, we have shown that LTP syndrome can occur from a very young age. However, in contrast to our previous study in adults from the same geographical area, sensitisation to other plant-food panallergens –especially storage proteins – was very common in our children. Interestingly, the clinical significance of LTP and/or storage protein sensitisation was unclear in a significant proportion of cases. Cross-reactivity might help explain this in some patients, for instance rCor a 8 or rAra h 9 sensitisation in children with stone-fruit allergy but tolerant to nuts. In addition to LTPs, 65% of cases were sensitised to storage proteins, which was associated with experiencing anaphylaxis and nut allergy. This finding raises the question whether LTP sensitisation is relevant in some patients, for instance in those exclusively allergic to nuts who are sensitised to storage proteins. Whether this likely silent LTP sensitisation might become clinically relevant over time and/or in the presence of cofactors, – or simply resolve – requires further prospective longitudinal studies. Our findings are concordant with a previous observation of frequent asymptomatic rPru p 3 sensitisation in an older cohort from Madrid, Spain, in which only 9% of patients had food allergy (median age: 27 years, range: 15–41).27 Likewise, 60% of patients without any nut or seed allergy were sensitised to storage proteins, whose clinical relevance in such a context is also unknown. Since asymptomatic sensitisation to LTP and/or storage protein was common (along with the fact that different allergenic components may explain clinical reactivity), the specific IgE testing by microarray failed to accurately discriminate allergic versus tolerant individuals to the plant-foods evaluated for the present study (i.e. peach, peanut, hazelnut, walnut). Similarly, a study in Italian children highlighted the limited value of specific IgE levels to foods other than Rosaceae to predict clinical allergy.17

With regards to reactions severity, specific IgE levels to LTPs did not correlate with reaction severity on history in our series, as previously described for peach allergy.17,29–31 Additionally, sensitisation to storage proteins was associated with anaphylaxis, suggesting that these allergens might involve higher potential for severe reactions than LTPs. In agreement with that, in an Italian adult series, storage-proteins sensitisation -rather than LTPs- was found more relevant for anaphylaxis and clinical reactivity to nuts.32,33 Nonetheless, in our study 31% of children exclusively sensitised to LTPs (and not to storage proteins) had experienced anaphylaxis (including anaphylaxis to nuts). Although this is a much lower proportion than that reported in our previous adult series (which was 75%), the potential of LTP to trigger anaphylaxis in children should not be underestimated.

Co-factors have been reported to enhance reactions severity in adults with LTP allergy.21,34,35 In our previous adult series co-factors were involved in reactions in 40% of cases. In contrast, we identified a cofactor (exercise) for reactions only in three children. We cannot exclude that this difference may be due to difficulties in evaluating cofactors in children, since parents might not report the coincidence of exercise (jumping, running) or intense emotions (crying, excitement), as these situations can be regarded as “normal life”. Further intervention studies would be required to determine the role of cofactors in LTP allergy in children, i.e., their ability to increase reactions severity and/or reduce patients’ reactivity threshold. Co-existent pollinosis has been reported to soften LTP-allergy symptoms.36,37 This was not reproduced in our study, since anaphylaxis rates were similar in children with or without sensitisation to pollen LTPs. Additionally, pollinosis and sensitisation to Art v 3 have been reported to contribute to LTP poly-sensitisation.16,17,31 In agreement with this, we observed that subjects sensitised to pollen LTPs had sIgE for a broader spectrum of plant-food LTPs than those who were not. Prospective studies are required to assess whether this phenomenon corresponds to a different primary sensitisation mechanism (i.e. via inhalation versus via oral/skin route) and to which extent sensitisation to pollen LTP contributes to “IgE spreading” or, alternatively, if it merely reflects broader cross-reactivity. We observed that sensitisation to rCor a 8 and rAra h 9 occurred almost exclusively in children sensitised to rPru p 3 and nJug r 3. These associations might suggest how sensitisations to the different plant-food LTPs develop overtime, which needs confirmation with prospective studies. Our observations are in agreement with two previous studies supporting Pru p 3 as the primary sensitiser in peach allergic adults with co-existent peanut38 and hazelnut allergy, respectively.39

In view of current evidence in the field of molecular-based diagnosis in plant-food allergy, it is difficult to determine to what extent these new techniques help the clinician in routine decision-making in terms of individual risk assessment or discrimination between allergy and tolerance.

Finally, we acknowledge several limitations in our study. First, it is a retrospective study, although we completed clinical data by reviewing all patients. Blinded food challenges -the gold standard for food allergy diagnosis- were not performed to confirm the diagnosis of plant-food allergy or to systematically determine allergy versus tolerance to other plant-foods.

In conclusion, plant-food allergy in the context of LTP sensitisation in children from Northeast Spain involves a diverse clinical picture (including anaphylaxis) and a broad range of triggers, as previously described in adults. However, co-sensitisation to storage proteins is common, which are associated to anaphylaxis and nut allergy. Sensitisation to LTP and/or storage proteins without clear clinical relevance is relatively common. Prospective studies are required to evaluate the role of these silent sensitisations in the future and/or when coinciding with cofactors. Caution is still required when interpreting the results of molecular-based diagnostic tools in clinical practice.

The authors declare that they do not have any potential conflict of interest. The study was self-funded.