BACKGROUND: Sublingual immunotherapy (SLIT) with monomeric carbamylated allergoid proved to be well tolerated, safe and effective in patients with respiratory allergy. Standard administration regimens are expected to require a long time before clinical benefit can be appreciated. We investigated whether pre-seasonal and perennial regimens differently affect the clinical efficacy of grass pollen SLIT. METHODS: Adult patients with allergic rhino-conjunctivitis with/without mild intermittent asthma due to grass pollen were included into this open prospective study and randomised to receive SLIT with a continuous regimen (Group 1: 1,000 AU/week for the entire study period) or a pre-seasonal regimen (Group 2: 5,000 AU/week for 10 weeks/year for 2 years), or on demand drug therapy alone (Group 3) for two years. At entry (November 2005), at the end of the first and second pollen season, a Visual Analogue Scale (VAS) was used to assess patients' well-being. Symptom score and drug consumption were evaluated during the seasons. Methacholine challenge was performed at study entry and conclusion. Adverse events were recorded along the whole study duration. RESULTS: Thirty-two patients were divided into Group 1 (n = 10), Group 2 (n = 11) and Group 3 (n = 11). A significant VAS improvement was observed in both SLIT groups, after the first and second pollen season, compared to baseline and to Group 3 (p < 0.05). Less symptoms and need for medications resulted during the second season (p < 0.05). No relevant variations in bronchial hyper-reactivity have been observed between the three groups. Only 2 patients experienced local or mild reactions in SLIT groups. CONCLUSION: Both pre-seasonal and continuous regimen of SLIT with monomeric allergoid turned out effective and safe, suggesting that a pre-seasonal course with 5,000 AU/week for 10 weeks may represent a convenient option in patients with grass pollen allergic rhinitis with/without mild intermittent asthma. Further research is urgently needed to consolidate these preliminary evidences.