Early- vs late-start levodopa relieved symptoms but did not affect disease progression in Parkinson disease

Abstract

Question In patients with early Parkinson disease (PD), does early start of levodopa modify disease progression compared with delayed start? Methods Design Randomized controlled trial (RCT) (Levodopa in Early Parkinson s Disease [LEAP] trial). ISRCTN30518857. Allocation Concealed.* Blinding Blinded* {patients, clinicians, data collectors, and outcome adjudicators}. Follow-up period 80 weeks. Setting 57 community and academic hospitals in The Netherlands. Patients 445 adults 30 years of age (mean age 65 y, 70% men, mean Unified Parkinson s Disease Rating Scale [UPDRS] score 29) who were diagnosed with PD in the past 2 years by an experienced neurologist based on standard clinical criteria and were not disabled enough to warrant treatment with anti-Parkinson medication. Exclusion criteria included past treatment with anti-Parkinson medication, tremor as the most prominent symptom, dementia, secondary parkinsonism or atypical disease, or life expectancy 2 years. Intervention Early-start (n =222) or delayed-start (n =223) oral combination levodopacarbidopa 100 mg25 mg 3 times/d. For the first 40 weeks (phase 1), the early group received levodopacarbidopa and the delayed group received placebo. For the next 40 weeks (phase 2), both groups received levodopacarbidopa (unblinded). Outcomes Primary outcome was change in disease severity (UPDRS) at 80 weeks. Secondary outcomes included change in disease severity at 40 weeks, change in disease severity per week between 4 and 40 weeks and between 44 and 80 weeks, and adverse events. 334 patients would provide 80% power to detect a 4-point difference on the UPDRS (2-sided =0.05) for the primary outcome. Patient follow-up 94% (intention-to-treat analysis). Main results The efficacy results are in the Table. 11% of patients in the early-treatment group and 39% in the delayed-treatment group developed disabilities involving activities of daily living during phase 1 and were switched to phase 2. The early and delayed groups did not differ at 80 weeks for PD symptom severity or motor complications of dyskinesia and motor fluctuations. Conclusion In patients with early Parkinson disease, early- vs delayed-start levodopa relieved symptoms but did not affect disease course. Early- vs delayed-start levodopa for change in disease progression (UPDRS) in patients with early Parkinson disease Assessment periods Mean change in UPDRS score Absolute mean difference (95% CI) Early-start Delayed-start Baseline to 80 wk 1.0 2.0 1.0 (1.5 to 3.5) Baseline to 40 wk 3.1 2.0 5.1 (7.2 to 2.9) Mean change in UPDRS score/wk 4 to 40 wk 0.04 0.06 0.02 (0.07 to 0.03) 44 to 80 wk 0.10 0.03 0.07 (0.03 to 0.10) UPDRS = Unified Parkinson s Disease Rating Scale; CI defined in Glossary. UPDRS score range 0 to 176; higher scores = more severe disease. Negative between-group difference favors early-start levodopa. 90% CI reported. The prespecified noninferiority margin was 0.055; difference between groups did not meet the criterion for noninferiority. Commentary The goal of Verschuur and colleagues RCT was to determine whether levodopa alters PD progression. Both the cause of disease as well as the cause of progression remain unknown. However, a popular theory is that progression results from oxidative stress induced by free radicals generated by the synthesis of dopamine from levodopa. The ELLDOPA trial (1) compared disease progression in patients with early, previously untreated PD who received therapy with 1 of 3 doses of levodopa or placebo for 40 weeks. After a 2-week washout period, the group with the highest levodopa dose showed the least progression of symptoms. The authors of ELLDOPA concluded that levodopa had no effect on progression, but the results were not definitive. One confounder was that although most patients lose the benefit of levodopa within hours, some maintain it for weeks. Also, few data were available on the effects of the drug on the dopamine-transporter scan. The LEAP trial used a delayed-start design, treating half of the patients with active drug and half with placebo for 40 weeks. All were then treated similarly with the active drug for the next 40 weeks. The assumption was that maximal benefit of levodopa is achieved within days to weeks of initiation; if the drug had no effect on progression, then at week 80 it should not matter if the drug was started at week 0 or week 40. No differences were found, supporting the interpretation of the ELLDOPA trial that levodopa does not alter progression. Will this make a difference in clinical practice? Probably not. However, it is reassuring because levodopa remains the gold standard and most widely used drug. The results are also important because they support the belief that dyskinesias and clinical fluctuations are influenced more by disease duration than by duration of levodopa use (2), so levodopa should be initiated early.