Standardizing research food challenges

A pressing need for standards of conduct for research-related double-blind, placebo-controlled, oral food challenges (OFC) provided the impetus for a consensus report by Sampson et al in this month’s issue (J Allergy Clin Immunol 2012;130:1260-1274).

After a brief review of the historical literature on diagnosing food allergy, Sampson et al present topical discussion summaries for major issues in the conduct of research OFC, noting that the guidance and recommendations apply equally to clinically performed challenges. The authors emphasize that patient/subject safety was the pre-eminent consideration. The following are highlights from their discussions:

·      Pre-challenge assessments. The authors cover important evaluations and practices prior to conducting OFC. These include case history, current food avoidance, comorbid conditions that impact safety such as atopic dermatitis and chronic disease, the utility of surrogate test results and their intrinsic problems, and the requirement for an optimal challenge setting. They also discuss the issue of differential severity of allergens as an important safety consideration prior to OFC.

·      Challenge procedures and assessments. Sampson et al discuss dosing schedules and their necessary variability tied to outcomes. General recommendations for dosing include low starting doses, separate active and placebo challenge days, dosing intervals no less than 20 minutes and appropriate, low fat matrices. Recording important, objective clinical parameters, such as respiratory symptoms and other physical reactivity symptoms, is covered. Sampson et al emphasize that scoring and stopping criteria must be pre-specified in light of the critical role of clinical judgment in OFC conduct. Delayed reaction considerations and subjective symptom issues are also reported. Tables of their recommendations are provided for reference.

·      OFC analysis and reporting. To ensure a reliable evidence base, Sampson et al provide guidance and recommendations for statistical analysis and reporting results, particularly from randomized trials.