BACKGROUND Inhaled corticosteroids are the most commonly used antiinflammatory agents for asthma. There is no simple way to compare objectively the relative potency of inhaled corticosteroids. The allergen-induced late asthmatic response (LAR) can be suppressed by a single dose of inhaled corticosteroid. OBJECTIVE This study was undertaken to evaluate LAR as a model for the determination of the relative potency of single doses of inhaled corticosteroids. METHODS We compared doses of 200 and 800 microg of a highly active inhaled corticosteroid (budesonide) with placebo and a marginally active investigational inhaled corticosteroid (D5159). Ten atopic patients with asthma completed a randomized, double-blind, double-dummy, multicenter, four-way, crossover trial. A standardized allergen challenge with the identical dose of allergen was performed 10 minutes after each of four blinded, single-dose treatments: 200 microg of budesonide, 800 microg of budesonide, 8 mg of D5159, and placebo, all administered from Turbuhaler. The LAR was recorded as the maximum percent fall in FEV1 between 4 and 7 hours, and the allergen-induced increase in methacholine airway responsiveness at 24 hours was recorded as the A log PC20 from the day before to the day after allergen challenge. RESULTS There were no significant differences in the early asthmatic responses during the 4 days; the mean maximum percent in FEV1 fall ranged between 19.5% and 22%. D5159 produced a slight inhibition of the LAR with maximum percent fall in FEV1 recorded as 28.8% +/- 5.0% for D5159 versus 34.1% +/- 4.8% for placebo (p < 0.05). There was a greater reduction recorded after administration of the two doses of budesonide. The mean LAR was 15.1% +/- 3.8% for 200 microg of budesonide and 11.2% +/- 2.3% for 800 microg of budesonide (p < 0.01 compared with placebo and D5159). The two doses of budesonide were not statistically different. Airway responsiveness to methacholine increased by 1.07 doubling doses 24 hours after allergen challenge. This increased airway responsiveness was slightly, but not significantly, reduced by the three active treatments (0.6 to 0.91 doubling doses). CONCLUSIONS The allergen-induced LAR model was able to differentiate a single dose of an active inhaled corticosteroid from placebo and a highly potent inhaled corticosteroid from a weak inhaled corticosteroid. The model did not differentiate between 2 fourfold doses of the highly active inhaled corticosteroid (at the doses used in this study), neither for the fall in FEV1 nor for the increase in airway hyperresponsiveness.