The method of assessment was a 9-point ocular itch scale measured in half-unit increments (0 none - 4 severe) with a higher score representing worse symptomology. Subjects were dosed once just prior to allergen chamber entry and again halfway through the chamber. The allergen chamber was 210 ten minutes in duration with the itching assessment being collected every 10 minutes from 110 to 210 minutes. The timepoints were assessed using mixed model repeated measures.

Change from baseline comparison of reproxalap to vehicle for ocular itching on a 0 to 4 scale ( 0 = none, 4 = severe). The treatment comparison was performed using a mixed model repeated measures approach for a crossover trial with two treatment periods. The model included time point, treatment, treatment period, treatment sequence, and interaction between time point and treatment as fixed effects, and subject nested within treatment sequence as a repeated measure.

Subject-reported ocular itching score area under the curve from 10 to 60 minutes post allergen challenge using a 0 to 4 scale (0 = none, 4 = severe) was assessed. The least squares mean was derived from analysis of covariance of area under the curve with baseline as a covariate and treatment as a fixed effect The possible range for area under the curve least squares mean is 0 to 100, where a lower score is better.

Mean change from baseline in ocular itching score using a 0 to 4 scale (0 = least, 4 = most) was assessed on days when peak pollen counts meet or exceed the American Academy of Allergy Asthma \& Immunology weed pollen scale 95th percentile (325 grains per cubic meter). The least squares mean (standard error) was derived from analysis of covariance with baseline, treatment, and day as factors.