To the Editor,

We read with great interest the manuscript by M. J. Rial and J. Sastre [1] about the geographical distribution of sensitization to non-specific lipid transfer proteins (ns-LTPs). In present day, this is an interesting topic, and therefore we would like to raise some issues and questions, especially concerning the prevalence of ns-LTP sensitization outside the Mediterranean area. The authors report that sensitization to ns-LTPs is infrequent in Central Europe and other non-Mediterranean regions; however, it appears that the authors have overlooked a large Belgian survey performed in 718 patients [2]. As a matter of fact, we demonstrated that the prevalence of sIgE reactivity towards ns-LTP(s) is demonstrable in about one-quarter of Belgian patients presenting with symptoms of a pollen and/or plant food allergy. In this survey, all patients were systematically screened for ns-LTP sensitization using a panel of six different ns-LTPs; four food ns-LTPs respectively rPru p 3 of peach (Prunus persica), rMal d 3 of apple (Malus domestica), rCor a 8 of hazelnut (Corylus avellana), and rAra h 9 of peanut (Arachis hypogaea) and two weed pollen ns-LTPs specifically nArt v 3 of mugwort (Artemisia vulgaris) and rPar j 2 of wall pellitory (Parietaria judaica). To the best of our knowledge, this study is the largest prevalence study focusing on ns-LTP sensitization in north-western Europe.

Moreover, this study also demonstrated that in a north-western European country, patients with ns-LTP sensitization can exhibit distinct phenotypes that are not readily predictable by the sIgE results. Although, similar to initial observations in the Mediterranean basin [3,4,5], some of our patients demonstrated systemic reactions, the majority of patients with sIgE reactivity towards ns-LTPs did not report any clinical reaction to the respective plant food(s). A possible explanation for the absence of overt allergy could be the high prevalence of sensitization to the major allergen of birch pollen, Bet v 1 (Betula verrucosa) [6,7,8]. However, for the time being, this explanation is highly speculative, but relies on observations from the Mediterranean basin on sensitization to Bet v 1 (PR10 molecule) to protect for ns-LTP-related allergies. In other words, patients co-sensitized to Bet v 1 and Bet v 1 homologues report milder clinical symptoms compared to patients without co-sensitization to PR10 molecules. Clearly, more studies are needed to fully elucidate the protective effect of PR10 molecules.

The exact reason(s) for the high prevalence of ns-LTP sensitization in our country remain(s) elusive. Although we cannot exclude our findings (in part) to reflect our methodology (usage of multiple sensitive single-plexed assays), we believe that in most patients, ns-LTP sensitization is genuine and might result from various sensitization routes that extend beyond pollen and plant-derived foods. Actually, we observed that Can s 3, the ns-LTP from Cannabis sativa, is a major allergen in cannabis allergy in our regions [9]. Moreover, it appears that sensitization to Can s 3 can result from both active and passive exposure to marijuana smoke [10] and that the Can s 3 cross-reactivity syndrome extends beyond fruits and vegetables but can also involve beverages and latex [11].

In conclusion, sensitization towards ns-LTP, although historically predominantly described in the Mediterranean basin, is not uncommon in north-western Europe and can result in clinically distinct phenotypes. Further collaborative studies are required to obtain insight into sensitization routes, clinical phenotypes, and the influence of pollen sensitization and to improve the predictive capacity of diagnostic tests for ns-LTP-related allergies.