We investigated the effect of immunotherapy (IT) on eosinophil (EOS) migration into the nasal cavity after nasal provocation with ragweed antigen and during seasonal exposure. In the first study, three groups of subjects participated: one group with no treatment (N = 19), one group with 10 months of IT, reaching maintenance at 2 micrograms of Amb a I (antigen E) (N = 15), and one group with 22 months of IT, reaching maintenance at 24 micrograms of Amb a I (N = 10). The percent of EOSs in nasal lavages performed during December before and 24 hours after nasal challenge with ragweed extract was determined. No significant difference between groups existed before challenge. The no treatment group demonstrated a significant increase in the percent of EOSs from 26% to 69.5% (p less than 0.008), whereas the treated groups demonstrated no significant change. In the second study, 45 patients were divided into four groups based on maintenance dose in micrograms of Amb a I and duration of treatment: (1) no treatment (N = 15), (2) 1 year at 2 micrograms (N = 13), (3) 2 years at 2 micrograms (N = 11), and (4) 3 years at 24 micrograms (N = 9). Nasal mucosal brushings were done during the ragweed season. A significantly smaller percentage of EOSs in 3-year IT-treated individuals was obtained compared to the control group (18 versus 8.4; p less than 0.04). The smaller dose of IT, regardless of duration, did not reveal a reduction compared to that in the no-treatment group.(ABSTRACT TRUNCATED AT 250 WORDS)