D2.207 - The Impact of Heat-Processing in Egg and Milk FPIES: A Multi-Center Study in the Southwest United States

Food protein-induced enterocolitis syndrome (FPIES) is a non-IgE-mediated food allergy characterized by delayed gastrointestinal symptoms after ingestion of trigger foods, with cow’s milk and egg among the most common. While most cases resolve over time, factors influencing tolerance development remain unclear. Emerging evidence suggests that protein denaturation through heat-processing may impact tolerance in FPIES. We aimed to evaluate clinical and immunologic differences between baked and unbaked egg and milk FPIES, and potential racial and ethnic disparities in tolerance acquisition.

A multi-institutional clinical registry was established for pediatric FPIES patients diagnosed since 2015. Data specific to egg and milk FPIES were extracted, and descriptive analyses were performed.

In the egg FPIES cohort (N=181), first reactions to baked egg (N=11) occurred later (median = 19 months) than unbaked egg (N=125, median = 8 months, p < 0.001). Baked egg patients had higher egg-specific IgE levels (median = 4.94 kU/L vs. 0 kU/L, p = 0.01). Oral food challenge (OFC) reaction rates were similar (13.9% baked vs. 12.0% unbaked, p = 0.813), though data were missing for 70.2% of cases. There was no significant difference in time to tolerance between baked and unbaked egg FPIES (p = 0.09). Asian children were more likely to develop tolerance compared to Black and White children (HR = 8.1, 95% CI: 1.57–41.7, p = 0.012).

In the milk FPIES cohort (N=169), first reactions to baked milk (N=3) occurred later (median = 16 months) than unbaked milk (median = 4 months, p = 0.0485). Milk-specific IgE levels did not differ significantly (median = 3.83 kU/L vs. 1.01, p = 0.54). OFC reaction rates were similar (14.3% baked vs. 10.6% unbaked, p = 0.346) with 69.2% missing data. Unbaked milk FPIES patients took longer to develop tolerance (HR = 0.3, 95% CI: 0.09–0.94, p = 0.039), and Asian children again showed a higher likelihood of tolerance (HR = 13.6, 95% CI: 1.81–102, p = 0.011).

These findings suggest differences in reaction timing between baked and unbaked egg and milk but no significant difference in OFC reaction rates; however, high missing data limits conclusions on tolerance development. Further research is needed to determine whether heat-processing affects FPIES tolerance development and resolution.