A Three-Dose Prophylactic Vaccine Against Food Allergy in Early Childhood: A Mechanistic, Bayesian, and Falsifiable Conceptual Framework

Food allergy (FA) affects 6--8% of children globally, with no approved prophylactic intervention. This manuscript proposes a comprehensive, pre-clinically grounded, and falsifiable framework for a three-dose prophylactic vaccine administered at 2, 4, and 12 months of age, integrated into the WHO Expanded Program on Immunization (EPI). The vaccine utilizes hypoallergenic peptides from major allergens---peanut (Ara h 2), egg (ovomucoid), and milk (\( \alpha_{s1} \)-casein)---encapsulated in GE11-functionalized lipid nanoparticles (LNPs) and adjuvanted with Pam3Cys + MPLA to promote regulatory T cell (Treg)-mediated tolerance. Key aspects addressed include biocompatibility, antigen dosing, LNP stability, lymphatic trafficking, and immunological non-interference in the multivalent formulation. A mechanistic ordinary differential equation (ODE) model captures the temporal dynamics of discrete dosing through a pharmacokinetically informed antigen input function \( A(t) \), adjusted for maternal antibody interference. Bayesian inference calibrates the model using data from the LEAP and EAT trials via a bridging parameter \( \phi \), with validation through global sensitivity analysis. The proposal is completed with a WHO-aligned cost-benefit analysis and a phased clinical roadmap (Phases I--III). No funding was received for this work.