Nicotinamide riboside, an NAD supplement, did not slow early Parkinson's and did slightly worse than placebo

Nicotinamide riboside is sold in pharmacies and online as a way to top up a molecule that cells run on. For people with Parkinson’s disease it carried a larger hope: that feeding struggling nerve cells might slow the disease itself, which no existing treatment does. Small early trials seemed to back the idea.

The large trial has now reported, and the answer is no. In a study of 410 people with early Parkinson’s, published in JAMA, a leading medical journal, on October 8, a year of the supplement did not improve symptoms. The group taking it ended up slightly worse off than the group taking a placebo. The difference was small, smaller than patients would be expected to notice, but it was in the wrong direction.

Why an NAD supplement was tested in Parkinson’s disease

The molecule at the center of this is nicotinamide adenine dinucleotide (NAD). Cells need it to turn food into energy, and it takes part in several other housekeeping jobs. The university behind the trial describes NAD as essential for cellular energy metabolism and several other vital processes, including DNA repair and regulation of inflammation.

Nicotinamide riboside (NR) is a form of vitamin B3 that the body converts into NAD. The paper calls it a nicotinamide adenine dinucleotide precursor, a precursor being a raw material.

The link to Parkinson’s runs through the cell’s power supply. In Parkinson’s disease the nerve cells that make dopamine, a chemical messenger needed for smooth movement, gradually die. The energy-producing parts of those cells appear to work poorly. Current medicines replace or mimic dopamine. As one account of the trial put it, available treatments primarily manage symptoms rather than halt neurodegeneration, neurodegeneration meaning the progressive loss of nerve cells.

So a compound that might protect the cells themselves was an attractive prospect. The paper sets out the starting point: nicotinamide riboside has been proposed as a potential disease-modifying therapy for Parkinson disease based on preclinical and early studies. Disease-modifying means changing the course of the illness, as opposed to easing its symptoms. Preclinical means laboratory and animal work.

How the nicotinamide riboside trial was run

The trial, called NOPARK, was led by Charalampos Tzoulis of Haukeland University Hospital and the University of Bergen. It was a randomized, double-blind, placebo-controlled, phase 3 clinical trial conducted at 11 centers in Norway. Double-blind means that neither the participants nor the study teams knew who was taking what.

The participants were at an early stage. They were patients aged 35 years or older with Parkinson disease diagnosed within 2 years before enrollment, already on stable treatment. Each had a brain scan to confirm the diagnosis. The registry entry requires a DaTscan confirming nigrostriatal degeneration, that is, a scan showing loss of the dopamine pathway.

Of 537 screened individuals, 410 were randomized, between May 2020 and June 2024. Patients were randomized to receive oral nicotinamide riboside, 500 mg twice daily (n = 206), or matching placebo (n = 204) for 52 weeks. The final analysis included 393 of them. Their average age was 65.6, and about a third were women.

They were checked every three months. The registry sets out the schedule: participants will be assessed at baseline, 13, 26, 39 and 52 weeks.

The yardstick was a standard rating scale for Parkinson’s. A clinician and the patient score a long list of items covering movement, daily activities and symptoms such as mood, sleep and pain. The total used in this trial runs from 0 to 236, and a higher score means more severe disease.

The scale comes with a rule of thumb for what counts as a real change. The paper gives the clinically important differences for improvement and worsening as 6.7 and 5.2 points. A shift smaller than that is one a patient would probably not feel.

What the Parkinson’s trial found after a year

At 52 weeks	Nicotinamide riboside	Placebo
Change in total symptom score (0 to 236, higher is worse)	2.45 points worse	0.27 points better
Participants with a serious adverse event	17 of 205 (8.3%)	29 of 205 (14.1%)

The supplement group got a little worse and the placebo group stayed about where it was. The difference between them, after statistical adjustment, was 2.72 points, with a plausible range of 0.47 to 4.98. The paper reports it as an adjusted mean difference of 2.72 points favoring placebo. Because the whole of that range lies on one side of zero, chance is an unlikely explanation.

Five other measures had been named in advance as the most important secondary ones. Four of them, including the separate parts of the rating scale and a scan of the dopamine system, showed no clear difference between the groups. The fifth went against the supplement. In the paper’s words, nonmotor symptom burden worsened more with nicotinamide riboside. Nonmotor symptoms are the ones that do not involve movement, such as sleep problems, low mood and constipation. On a scale running from 0 to 360, the gap was 4.22 points.

The scans are the nearest thing the trial has to a direct look at the disease. The university summarized them this way: brain imaging showed no evidence that NR either slowed or accelerated changes in the brain’s dopamine system.

Safety, in the usual sense, was not the problem. Serious adverse events occurred in 17 (8.3%) of 205 participants receiving nicotinamide riboside and 29 (14.1%) of 205 receiving placebo. Those counts use the 205 people in each group who made up the trial’s safety population. An adverse event is any medical problem recorded during a trial, whether or not the treatment caused it.

The authors’ conclusion is blunt. Among patients with early Parkinson disease, nicotinamide riboside treatment for 52 weeks did not improve clinical outcomes and was associated with worse clinical outcomes than placebo.

How large the difference from placebo was

Two things are true of the 2.72-point gap, and both matter.

It is real in the statistical sense. A trial of this size would rarely produce a difference that large by accident.

It is also small. It amounts to a little over 1% of the scale’s range, and it is about half of the 5.2 points taken as a worsening a patient would notice. The university’s own summary makes the point, calling the difference modest and below the commonly used threshold for a clinically important difference.

That leaves two readings. One is that the supplement does a small amount of harm in early Parkinson’s. The other, which the statistics make less likely, is that it does nothing and the placebo group happened to have a good year. The scans, which showed no difference either way, do not help choose between them. Neither reading is evidence of benefit.

Tzoulis did not soften the result. “We had a strong biological rationale for testing this treatment, and earlier, smaller clinical studies had produced encouraging findings. NR clearly increased NAD metabolism, but this did not translate into clinical improvement,” Tzoulis said in the university’s statement.

That second sentence is the heart of it. The supplement did what it was supposed to do chemically. Raising the molecule was the easy part. It made no difference to the disease that anyone could measure, except a slight one in the wrong direction.

How earlier, smaller trials of the supplement compare

The same Bergen group ran the two trials that led here, and both looked promising.

The first, in 2022, was tiny and brief. Thirty newly diagnosed, treatment-naive patients received 1,000 mg NR or placebo for 30 days. Brain scans showed that NAD levels rose in most of those taking the supplement, and in those patients the change was associated with mild clinical improvement. The authors wrote then: “Our findings nominate NR as a potential neuroprotective therapy for PD, warranting further investigation in larger trials.”

The second, in 2023, tested a dose three times higher for safety. It involved 20 individuals with PD, PD being Parkinson’s disease, for four weeks. Symptom scores improved in the supplement group there too, but the authors flagged a snag. In their words, this change was also associated with a shorter interval since the last levodopa dose. Levodopa is the main Parkinson’s medicine, and a patient examined soon after a dose scores better. The apparent benefit may have been a matter of timing.

Both were phase 1 trials, built to check safety and biology in a few dozen people. Neither could show whether the disease was slowed. The new trial is more than ten times the size of the first and ran twelve times as long, and the signal did not survive.

It is not the only place where the promise of NAD boosters has faded on closer testing. An independent review last year pooled randomized trials of NR and a related compound, NMN, in older adults, looking at muscle. Its conclusion was that current evidence does not support NMN and NR supplementation for preserving muscle mass and function in adults with mean age of over 60 years. One of the trials it covered found a result that echoes the new one: among people with mild memory problems, physical performance scores were lower on NR than on placebo.

What the trial leaves open about NAD supplements

One dose, one year. The trial used 1,000 mg a day for 52 weeks. As one report on it noted, the result suggests that increasing NAD availability through NR supplementation, at the tested dose and treatment duration, is insufficient to produce a meaningful clinical benefit. The group’s own safety trial went up to 3,000 mg a day. Whether a higher dose would act differently is untested in a trial of this kind.

Early disease only. Everyone had been diagnosed within two years. The result says nothing direct about later stages.

Subgroups. An average can hide people who respond differently. The researchers plan to look, with the aim of learning whether different groups of patients may have responded differently to the treatment. Analyses of that kind, done after the main result is known, generate ideas more than answers.

Why the supplement group did worse. The trial does not explain it, and a difference this small may not have a single explanation.

Other conditions. NAD supplements are sold for many purposes. The university is careful on this: the result does not rule out the possibility that NAD-boosting treatments could be beneficial in other diseases. Each would need its own trial.

One country. All 11 centers were in Norway.

On Parkinson’s itself the researchers are direct. According to the university, they do not recommend that people with Parkinson’s disease use NR or other NAD-boosting supplements outside controlled clinical studies. The trial was sponsored and run by a public hospital; the university describes it as sponsored and coordinated by Haukeland University Hospital.

Whether any supplement makes sense alongside Parkinson’s treatment depends on a person’s medicines and condition, which is assessed by their doctor.

A year of nicotinamide riboside did not slow early Parkinson’s disease in a 410-person placebo-controlled trial, and the group taking it scored 2.72 points worse on a 236-point symptom scale, a difference probably too small for patients to feel, and no evidence of benefit from the dose tested.