STUDY DESIGN: An approximately 24-week, randomized, double-blind, parallel-group, multicenter, placebo-controlled, allergy immunotherapy study of 344 pediatric patients evaluated the efficacy of GRASTEK. GRASTEK (n=149) or placebo (n=158) was administered as a daily sublingual tablet approximately 16 weeks prior to and continuing through GPS. Patients in both arms of the study were additionally permitted to take symptom-relieving allergy medications as needed. Both treatment groups were balanced with regard to baseline characteristics. Patients ranged from 5 through 17 years of age (mean age=12.3 years), were 65% male and 88% white, 7% African American, and 2% Asian, had a history of grass pollen-induced allergic rhinitis with or without conjunctivitis diagnosed by a physician and received treatment the previous GPS, and had sensitivity to Timothy grass as determined by positive skin prick and IgE tests. Twenty-six percent had mild, intermittent asthma, and FEV1 needed to be ≥70% predicted. Eighty-nine percent were sensitized to other allergens in addition to grass pollen, including weed, tree, pet dander, dust mites, and mold. Patients with a clinical history of symptomatic allergies to non-grass allergens that required treatment during GPS and patients with severe asthma were excluded from the study. The primary endpoint was TCS, which is the averaged sum of DSS and DMS, during the first entire GPS. DSS measured 6 symptoms (runny nose, stuffy nose, sneezing, itchy nose, gritty/itchy eyes, and watery eyes) on a scale of 0 (none) to 3 (severe), for a maximum daily total of 18. DMS measured use of standard open-label allergy medications, with oral antihistamine tablet or syrup scored as 6 points/dose, ocular antihistamine as 1.5 points/drop, steroid nasal spray as 4 points/spray for patients aged 5–11 years and 2 points/spray for patients aged 12–17 years, and oral steroid as 1.6 points/tablet and 1.6 x 2 points/tablet for Day 2 onward, for a maximum daily total of 36. Relative differences (% decreases) were calculated as follows: (GRASTEK–placebo)/placebo x 100 Estimated group means are reported and difference relative to placebo is based on estimated group means.