Background: Standardized allergen vaccines are tested for potency by manufacturers by using assays proposed in their license applications and approved by the Center for Biologics Evaluation and Research, which reviews and verifies the test results before lot release. The current lot-release limits for mite and grass pollen allergen vaccines impose statistical equivalence to the national reference extract; thus the limits are primarily based on assay variability. Objective: We sought to establish a clinical basis for lot-release limits for the relative potency of allergen vaccines and to evaluate alternative specifications. Methods: We performed literature selection and review, linear and logistic regression analyses of selected studies, and analysis of lots submitted to the Food and Drug Administration for approval since 1995. Results: Therapeutic equivalence is achieved over a 10-fold range of allergen concentration. Safety equivalence is more difficult to assign, but on the basis of injection data, a 4-fold increase in allergen concentration is associated with a 5% to 10% increase in adverse reaction rates. The SD in log relative potency for the submitted allergenic products was determined to be 0.090 for grasses and 0.061 for mites, compared with 0.079 for competition ELISA. Conclusions: Current lot-release limits are well within literature-based estimates of therapeutic, diagnostic, and safety equivalence ranges for the clinical use of allergen vaccines. In addition, the aggregate consistency of the submitted products is comparable with the precision of the assay that is used to assess the products. These results support expanded release limits for verification of relative potency, provided the submitted lots of material remain at their present level of consistency. (J Allergy Clin Immunol 2000;105:468-74.)