Abstract

BACKGROUND: Little is known about the development and resolution of early peanut allergy (PA). OBJECTIVE: We examined the natural history and biomarkers of PA longitudinally in 3 cohorts. METHODS: PA development was examined in the Enquiring About Tolerance (EAT), Learning Early About Peanut (LEAP), and Peanut Allergy Sensitization (PAS) cohorts. Early PA was defined by skin prick test result of >4 mm by 12 months or oral food challenge at study entry. PA was confirmed by oral food challenge at study end point (36 months for EAT, 60 months for LEAP/PAS). Four groups were defined: early PA development with persistence (EP); early PA development with resolution (ER); late PA development (LA); and never peanut allergic. Clinical characteristics and biomarkers were compared between the groups. RESULTS: A total of 56.3% of peanut-allergic children developed PA by 12 months; 32.1% had early PA resolution by study end point. The rate of early PA resolution was 54.2% in EAT, 41.4% in LEAP, and 18.6% in PAS cohorts. Median skin prick test wheals for EP, ER, and LA were 6, 2, and 0 mm at baseline, and 10, 0, 9 mm at study end point. Median peanut-specific IgE (sIgE) levels for EP, ER, and LA were 5.9, 0.4, and 0.3 kUA/L (P < .001) at baseline; 4.7, 1.3, and 0.9 kUA/L (P < .001) at 12 months; and 20.1, 0.2, and 5.1 kUA/L (P < .001) at study end point. LA had slower component expansion (number of components Ara h 1-sIgE, Ara h 2-sIgE, Ara h 3-sIgE > 0.1 kUA/L) compared to EP. ER showed component expansion from baseline to 12 months but component retraction by study end point. Absence of eczema and egg allergy, low peanut-sIgE, or skin prick test result were predictive of PA resolution. CONCLUSION: A significant proportion of PA resolves in early childhood. Different phenotypes of PA display different biomarkers trajectories.