The monoamine oxidase type B inhibitor deprenyl (selegiline) has been demonstrated to delay the emergence of disability in early untreated Parkinson's disease. Lazabemide (RO19-6327) is a short-acting, reversible, highly selective inhibitor of monoamine oxidase type B which, unlike deprenyl, is not metabolized to active compounds. We conducted a randomized, double-blinded clinical trial to assess the short-term tolerability of lazabemide in subjects who had early, untreated Parkinson's disease. Two hundred and one patients were enrolled at 14 centers and randomized to receive 100 mg/day, 200 mg/day, or 400 mg/day of lazabemide or matching placebo. Subjects were followed for 8 weeks including a randomized, double-blinded withdrawal of lazabemide for 2 or 4 weeks. The primary measure of tolerability was the proportion of treated subjects who were able to complete the study on their originally assigned treatment. Clinical features were assessed by the Unified Parkinson's Disease Rating Scale (UPDRS). Lazabemide treatment was as well tolerated as placebo and was not attended by serious adverse experiences. A significant improvement in the activities of daily living component of the rating scale was found after 4 weeks of lazabemide treatment, although other subscale scores did not change significantly. The overall safety and benefits of lazabemide observed in this short-term study justify further long-term investigations to determine if this monoamine oxidase type B inhibitor can slow the clinical progression of Parkinson's disease.