Objective: To study the long-term effects of selegiline in monotherapy and in combination with levodopa in the early phase of Parkinson disease (PD). Methods: One hundred fifty-seven de novo PD patients were randomized in a double-blind, placebo-controlled study of 7 years' duration. In the monotherapy part, selegiline significantly delayed the initiation of levodopa therapy vs placebo. The authors now report the results from the combination part of the study, in which 140 patients received selegiline or placebo in addition to individually tailored levodopa therapy. Results: Compared with placebo, selegiline slowed the progression of disease disability as measured by the Unified Parkinson Disease Rating Scale (UPDRS) total score (p = 0.003) or by motor (p = 0.002) and Activities of Daily Living (p = 0.0002) subscores. After 5 years in combination therapy, the mean difference in the UPDRS total score was nearly 10 points, with patients receiving placebo having 35% higher scores. Simultaneously, patients receiving placebo needed progressively higher doses of levodopa than patients receiving selegiline, after 5 years, the mean dosage of levodopa was 19% higher with placebo than with selegiline (p = 0.0002). Considering the entire (monotherapy and combination therapy) 7-year study time, there was a trend for selegiline to delay the start of wearing-off fluctuations (hazard ratio 0.55, p = 0.08). In both phases of the study, selegiline was safe and well tolerated. Conclusions: The results of this long-term study confirm earlier findings indicating that selegiline delays the progression of the signs and symptoms of Parkinson disease. Copyright © 2006 by AAN Enterprises, Inc.