Venom immunotherapy (IT) is a very effective method for the treatment of Hymenoptera venom allergy. We compared safety, efficacy, and modulation of specific immunologic parameters in 70 patients sensitized to Apis mellifera, treated for > or = 5 years with standardized quality (SQ) aqueous IT, either with a rush (n = 20) or with a cluster (n = 20) induction protocol, or with an SQ depot extract and a cluster induction protocol (n = 30). We made an open, noncontrolled study. Side effects were monitored and the effects of field stings during the maintenance phase of the treatment and after its interruption were recorded. Skin reactivity to Apis was measured by end point dilution and specific serum immunoglobulin E (IgE) were measured by a solid-phase-based assay. The depot IT was better tolerated than aqueous IT with rush induction. This was caused by mainly the lower frequency in the induction phase of systemic side effects (3.4% versus 36.8% [p < 0.0041] on a "per patient" and 0.1% versus 0.9% [p = 0.0092] on a "per dose "basis, respectively). The cluster protocol with the aqueous extract tended to be better tolerated that the rush protocol. About one-half of patients from each group were re-stung during the study and all suffered only minor discomfort. Reduction of skin reactivity and of serum-specific IgE was significant in the three groups (p < 0.02 in all cases). SQ Depot IT to Apis venom allergy administered with a cluster protocol induces less side effects and is equally effective then IT with SQ aqueous extract administered with a rush protocol.