BACKGROUND: Treatment of Parkinson's disease (PD) with levodopa results in a sustained reduction of symptoms. Although the plasma half-life of levodopa is short, it elicits a lasting effect, the long-duration levodopa response (LDR). A decrease in LDR as PD progresses has been linked to motor complications, but long-term data on the LDR and its clinical implications remain scarce.

OBJECTIVES: The aim is to analyze the magnitude and impact of the LDR over time using data from the Parkinson's Disease Progression Marker Initiative (PPMI).

METHODS: First, therapy-naïve Movement Disorder Society-Unified Parkinson's Disease Rating Scale part III (MDS-UPDRS III) scores were predicted using a mixed linear model (MLM) from n = 245 untreated people with PD (PwPD). This model yielded an increase of MDS-UPDRS III scores of 2.65 points per year. Using this model, we then calculated LDR and short-duration response in longitudinal data of 148 initially therapy-naïve PwPD. Symptom progression was analyzed using correlation analyses and MLMs.

RESULTS: In the 98 PwPD with observed LDR, the LDR accounted for approximately half of the total levodopa response. No significant change in LDR magnitude was observed over up to 10 years (analysis of variance, P = 0.14; generalized estimating equations, P = 0.26). The LDR magnitude was not associated with the onset of motor complications. PwPD with absent LDR (n = 50) progressed faster than PwPD with observed LDR in several motor and non-motor domains.

CONCLUSIONS: The LDR is a stable component of the levodopa response and needs to be considered in clinical trials. These findings argue against a declining LDR as a major driver of motor fluctuations in PD.