BAYESIAN NETWORK ANALYSIS INDICATES A HIGH PROBABILITY OF SIMULTANEOUS SENSITIZATION TO MAJOR GRASS ALLERGENS

Abstract

<h3>Introduction</h3> Grass pollen allergy impacts many patients characterized by high allergen cross-reactivity. The primary elicitor of grass allergy is Phl p 1. However, IgE to Phl p 5 or Phl p 2 also serve as markers of true grass pollen sensitization. Understanding the patterns of sensitization to grasses in certain population remains important for accurate prevention and treatment of allergy to Poaceae pollen. <h3>Methods</h3> Data obtained by multiplex allergy testing Alex² was collected from 20,333 patients. Bayesian Network analysis was used to build probabilistic patterns of patient sensitization to majpr grass pollen allergens. <h3>Results</h3> Grass pollen sensitive individuals constituted 6170 patients or 30.34 % of the entire database; 3935 (19.35 %) were sensitive to Phl p 1; 3460 (17.02 %) – to Lol p 1; 2842 (13.98 %) – to Cyn d 1; and 1772 (8.71 %) – to Phl p 2. Bayesian Network analysis indicated that sensitization to Phl p 1 was associated with sensitivity to Cyn d 1, Lol p 1, Phl p 2 and Sec c_pollen with 94.70 % probability. Sensitization to Lol p 1 was connected with sensitivity to Cyn d 1, Pas n extract, Phl p 2, Phl p 5 with 91.96 % probability. Cyn d 1 sensitization could be seen together with Cyn d, Sec c_pollen, Phr c extracts and Phl p 6 sensitization in 86.58 % of cases. <h3>Conclusion</h3> Bayesian Network analysis suggests that group I allergens are highly cross-reactive and can be a good target for AIT in the Ukrainian population sensitive to grass pollen.