Objectives: This study was undertaken to compare efficacy, tolerability, and pharmacokinetics of DM-1992, an extended-release formulation of carbidopa/levodopa (CD/l-dopa) with immediate-release (IR) CD/l-dopa in patients with advanced Parkinson's disease. Methods: This randomized, open-label, crossover study included a 3-d baseline and two 10-d treatment periods. Patients with daily OFF time of 2.5 h or more taking 400 mg or more l-dopa/d in four or more divided doses were titrated to stable regimens of DM-1992 2 times per day or CD/l-dopa IR 3 times to 8 times per day. Patients were allowed to take rescue CD/l-dopa as needed. Using home diaries, patients recorded OFF time and ON time with or without troublesome dyskinesia during baseline and treatment days 7 through 9. During 12-h clinic visits on day 10, plasma samples were collected for pharmacokinetics, and motor performance was assessed hourly. Results: Thirty-four patients were enrolled; mean baseline l-dopa dosage was 968 mg/d. After titration, CD/l-dopa IR was dosed 4.8 times per day and DM-1992, 2 times per day. Rescue CD/l-dopa IR was given 1.3 times during the DM-1992 arm and 0.2 times during the CD/l-dopa IR arm. The reduction from baseline in % OFF time was greater for DM-1992 compared with CD/l-dopa IR (−5.52% vs. +1.33%; P = 0.0471). At steady-state, compared with CD/l-dopa IR, DM-1992 exhibited a smoother plasma l-dopa concentration profile mostly because of a significantly higher (day 10) predose l-dopa concentration, associated with enhanced motor performance. Although more patients taking DM-1992 had one or more adverse events (AEs) than CD/l-dopa IR patients (35% vs. 15%), no pattern to the AEs was seen, nor any resulting discontinuations. Conclusions: DM-1992 was associated with a reduction in %OFF time compared with CD/l-dopa IR despite a reduced dosing frequency. Although the open-label study design and the greater number of rescue doses during the DM-1992 arm call for caution in interpreting the results, the elevated predose plasma l-dopa concentration (12 h after DM-1992 administration) lends objective support to our findings, suggesting that phase 3 studies are warranted. © 2015 International Parkinson and Movement Disorder Society.