Nicotinamide riboside remains an experimental approach to Parkinson’s disease in the trials reviewed here. NADPARK showed that oral NR could increase brain NAD-related measurements in some participants. NR-SAFE tested short-term tolerability at a higher exposure. Neither early trial established that NR slows the disease or replaces established symptom treatment. [1][2]
These are studies of an oral precursor, not injected NAD⁺. Our NR explainer describes that distinction. It matters because a positive finding for one molecule and route cannot validate a different commercial treatment.
Two early trials with different jobs
| Study and participants | Design and outcome | Finding | Main limit |
|---|---|---|---|
| NADPARK; n = 30 [1] | Phase I, randomized, double blind; newly diagnosed participants; thirty days | Variable increase in brain NAD; biological activity and tolerability findings. | Short duration; not a disease-progression efficacy trial. |
| NR-SAFE; n = 20 [2] | Randomized, double blind; four weeks; safety primary endpoint | No moderate/severe adverse events; blood NAD⁺ up to fivefold higher. | Small and short; symptom change confounded by levodopa timing. |
| NOPARK registry [3] | Larger randomized clinical-progression study; 52-week design | Record listed completed, 410 enrolled, no registry results posted at check. | Study completion and protocol availability are not efficacy results. |
NADPARK: did the intervention reach the target?
NADPARK randomized 30 newly diagnosed, treatment-naive people with Parkinson’s disease to oral NR or placebo for thirty days. It was double blinded and phase I. Researchers measured brain NAD using magnetic resonance spectroscopy and examined related metabolites and other biological responses. NR was generally well tolerated and produced a significant but variable increase in cerebral NAD. [1]
Participants with an increase in brain NAD also showed altered cerebral metabolism, associated with mild clinical improvement. That association is interesting, but selecting the people whose biomarker responded is not equivalent to demonstrating a durable treatment effect in the full randomized population. Thirty days also cannot establish slower progression of a chronic neurodegenerative disease.
The trial’s achievement was to show biological activity worth pursuing. It answered more than whether a pill can raise a blood number, because it included brain measurements. It still left the patient-centered efficacy question open. Related cognition trials illustrate why biological engagement and clinical benefit can diverge.
NR-SAFE: what four weeks of monitoring showed
NR-SAFE randomized 20 participants, ten per group, to higher-dose NR or placebo for four weeks. All completed the study. There were no moderate or severe adverse events, and mild adverse-event frequency did not differ significantly. Blood NAD⁺ increased by as much as fivefold. [2]
The primary endpoint was safety, not proof of efficacy. The study also observed an improvement in total MDS-UPDRS, a scale covering Parkinson-related symptoms and function. However, that change was associated with a shorter interval since the last levodopa dose. Medication timing can influence a symptom assessment, so it is an important alternative explanation rather than a minor footnote.
Researchers monitored laboratory markers as well as reported symptoms. A small initial rise in serum homocysteine was observed, while measures of the methyl-donor pool remained intact. The authors supported further monitored clinical testing. That conclusion should not be converted into a recommendation to copy the study exposure at home. [2]
What the later registry can tell us
The larger NOPARK study was designed to assess clinical progression over 52 weeks. Its ClinicalTrials.gov record was checked on October 6, 2026: the study was listed as completed, with 410 actual participants and no results posted in the registry. A completed study is not itself a positive result. A protocol or statistical-analysis plan describes the intended test; it cannot tell us whether treatment worked. [3]
Registry entries and journal publications can appear at different times. The status statement here refers specifically to the record checked, not a claim that no other report can exist. Our trial-registry guide shows how to inspect both the study record and its linked publications.
Safety, sponsorship and the next evidentiary step
ChromaDex supplied NR and matching placebo for NR-SAFE. The paper disclosed patent applications concerning NR treatment for Parkinson’s disease, with investigators Charalampos Tzoulis and Christian Dölle named as inventors. Those were the disclosures at publication; they are not a statement about current patent status. Industry involvement does not by itself invalidate a finding, but complete reporting, prespecified outcomes and independent replication help readers judge confidence. [2]
A convincing progression claim needs a sufficiently large randomized comparison, meaningful follow-up, careful accounting for background Parkinson medication and prespecified clinical outcomes. It also needs more safety information than twenty people observed for four weeks can provide. The general safety guide explains why the absence of events in a small short trial cannot establish universal safety.
These studies justify research interest, not a change to prescribed Parkinson’s treatment. A neurologist can put an experimental intervention in the context of symptoms, medicines and trial eligibility. There is no injection-shopping recommendation embedded in these oral-NR findings.