PDF: SC LEVODOPA PD
(2026)Objective
To assess whether continuous subcutaneous infusion of levodopa/carbidopa (ND0612) or foslevodopa/foscarbidopa safely reduces daily OFF-time and improves quality of life compared to oral levodopa/carbidopa in Parkinson's disease patients.
Study Summary
• ND0612 reduced OFF-time by 1.42 h/day; foslevodopa/foscarbidopa reduced OFF-time by 2.76 h/day — both statistically significant
• PDQ-39 quality of life improved (p = 0.0003) and PDSS-2 sleep score improved (p = 0.02), though QoL improvement was statistically but not clinically significant
• Treatment-emergent adverse events were increased (p = 0.04), predominantly infusion site reactions
Intervention
Continuous subcutaneous infusion of levodopa/carbidopa (ND0612) or foslevodopa/foscarbidopa via wearable pump, administered over 14–24 hours per day, with or without supplementary oral levodopa depending on formulation
Inclusion Criteria
Randomized and non-randomized clinical trials of any duration comparing subcutaneous levodopa or foslevodopa infusion to oral levodopa/carbidopa in Parkinson's disease patients of any age or gender, with comparative data on at least one specified outcome
Study Design
Arms: Subcutaneous ND0612 (n=407) or foslevodopa/foscarbidopa (n=318) vs. oral levodopa/carbidopa with sham infusion (control, across 7 studies)
Patients per Arm: 725 total received subcutaneous infusions: ND0612 n=407; foslevodopa/foscarbidopa n=318
Outcome
• PDQ-39 QoL improved (p = 0.0003); PDSS-2 sleep improved (p = 0.02) — both moderate-quality evidence
• Increased treatment-emergent adverse events with subcutaneous infusion (p = 0.04), predominantly infusion site-related
Bottom Line
Subcutaneous infusion of levodopa or foslevodopa produces a clinically and statistically significant reduction of approximately 2 hours in daily OFF-time compared to oral levodopa in PD patients with motor fluctuations; it is a viable alternative to intestinal gel infusion that avoids surgical jejunostomy, though patients should expect increased infusion site adverse events.
Major Points
- Seven studies enrolling 725 patients on subcutaneous infusion were identified; moderate-quality evidence showed a mean 1.98 h/day reduction in OFF-time vs. oral levodopa (p = 0.0004)
- Formulation-specific reductions: ND0612 reduced OFF-time by 1.42 h/day; foslevodopa/foscarbidopa reduced OFF-time by 2.76 h/day — both statistically significant
- Health-related quality of life (PDQ-39) improved significantly (p = 0.0003) but the improvement was judged not clinically significant
- Sleep quality (PDSS-2) improved significantly (p = 0.02)
- Treatment-emergent adverse events were significantly more common with subcutaneous infusion (p = 0.04), predominantly infusion site reactions
- All seven included studies were funded by and had authors affiliated with the pharmaceutical manufacturer (NeuroDerm or AbbVie), introducing pervasive industry bias
- Subcutaneous infusion avoids the need for surgical percutaneous jejunostomy tube placement required for levodopa/carbidopa intestinal gel (LCIG), potentially broadening eligibility
Study Design
- Study Type
- Systematic review and meta-analysis
- Randomization
- No
- Blinding
- Not applicable (meta-analysis); individual RCTs used sham/dummy subcutaneous infusion as control
- Sample Size
- 725
- Follow-up
- 24 hours to 52 weeks across included studies
- Countries
- UK, USA, Europe, Israel, Australia, Russia
Primary Outcome
Definition: Daily duration of OFF-time in hours; measured by patient-reported home diary in all studies except Olanow et al., where measured by blinded clinician over 8 hours in clinic
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| Oral levodopa/carbidopa baseline or control arm | Subcutaneous infusion (ND0612 or foslevodopa/foscarbidopa) | - | p = 0.0004 |
Limitations & Criticisms
- All seven included studies were funded by and had authors affiliated with the pharmaceutical companies producing the studied products (NeuroDerm n=5; AbbVie n=2), introducing pervasive industry bias across all outcomes
- Three of seven studies were non-randomized (two dosing-regimen comparisons without independent control groups; one single-arm open-label), limiting causal inference
- Aldred et al. and Olanow et al. were judged at serious risk of bias due to lack of blinding
- Olanow et al. measured OFF-time via blinded clinician in clinic over 8 hours, differing from the patient-reported home diary used in all other studies — compromising cross-study comparability
- Aldred et al. compared efficacy at 52 weeks vs. baseline without a concurrent control, introducing confounding from disease progression (though this would likely underestimate, not overestimate, benefit)
- Giladi et al. rated as 'some concern' for risk of bias due to missing randomization process detail
- Fewer than ten studies precluded funnel plot analysis for publication bias
- PDQ-39 quality of life improvement was statistically but not clinically significant — the clinical relevance threshold was not met
- DIZ102, a third subcutaneous levodopa formulation, had no qualifying trials and was excluded from analysis
Citation
Burton M, Marsden D, Harish D, Morris P, Heneghan C, Butler A, Nunan D. Subcutaneous Levodopa in Parkinson's Disease: A Systematic Review and Meta-Analysis. Eur J Neurol. 2026;33:e70506.