Omega-3 Brain Trial SHOCKS Scientists. Now What?

Omega-3 Brain Trial SHOCKS Scientists. Now What?

Last Updated:

We have a bit of a problem, and we need to work this out. A really well done study, published in one of the Lancet journals, looked at giving omega-3 to people at high risk of developing dementia. The results weren't what we expected.

Now, one of the theories is that if we supplement with omega-3 and have healthy B vitamin levels, we can improve cognitive performance and slow brain shrinkage. In this trial of 365 participants, even though both groups supplemented with B vitamins, and we could see clear evidence that DHA, which is one of the types of omega-3, increased in the fluid surrounding the brain, that's called the CSF, cerebrospinal fluid, there was no benefit over placebo for either cognitive performance or brain volume [1].

But other studies, such as the VITACOG trial, which I've talked about a lot on my YouTube channel, seem to show that B vitamins, combined with good omega-3 levels, can slow brain shrinkage. So what's going on? Why is there conflicting data here?

Table of Contents

Where the Idea Came From

This link between brain health and omega-3 really kicked into overdrive in 2006, because we could observe that populations that ate more fish, particularly fish that had omega-3, seemed to have better cognitive performance and lower dementia rates. So that gave the idea that maybe if we supplemented with omega-3, we could improve the health of the brain.

That year, the Framingham study found that people with the most DHA in their blood had about half the risk of developing dementia [2].

And this made sense from a biological perspective. DHA is one of the building blocks of the walls of our brain cells, and it helps brain cells pass signals to each other. So higher DHA, in theory, would help the structure and function of our brain cells [1].

But there's a problem with this observational data about people who eat more fish. People who eat more fish are different in lots of other ways too. The only way to know if omega-3 itself helps is to give it to people at random.

The Trials, and the B Vitamin Idea

So scientists then ran the trials, where they gave omega-3 or a placebo to different populations, and the results were mixed.

One review of 24 of the larger omega-3 trials found that only 5 trials showed improved cognition. The rest mostly found nothing. And all four trials in people who already had Alzheimer's found nothing [3, 1].

Now, one of the theories from those mixed results, and this comes from the VITACOG trial as well, is that if we just supplement with omega-3, we may not have a clinical benefit for the brain. Instead, what we might need to do is also supplement with B vitamins.

Why would the two work as a pair? B vitamins keep a substance in our blood called homocysteine low. When homocysteine builds up, it jams a chemical cycle in the liver, and that blocks an enzyme called PEMT (phosphatidylethanolamine N-methyltransferase). PEMT is one of the main ways the liver packs DHA into a fat called phosphatidylcholine, which carries DHA through the blood to the rest of the body [4, 5, 6].

Mice without this enzyme have much less DHA in their blood. And in a small study, people with Alzheimer's had less DHA in these fats, and the more of the homocysteine by-product they had, the less DHA there tended to be. So the idea is that without enough B vitamins, less DHA gets delivered to our cells, and the VITACOG researchers suggest that includes the brain [7, 5, 8].

VITACOG gave older people with mild memory problems high-dose B vitamins, that's folic acid, B6 and B12, or a placebo, for two years. Brain shrinkage, measured on MRI scans, was about 30 percent slower overall with the B vitamins [9].

When researchers went back and measured omega-3 in the blood, the B vitamins only slowed shrinkage in people whose omega-3 levels were already high. In that group, it was 40 percent slower. In people with low omega-3, the B vitamins made no measurable difference [10].

The New Trial

The VITACOG finding came from an analysis done after the trial was over, and the researchers themselves said the next step was a trial of the two together [8].

That's the theory that this new trial tested.

It was a randomised controlled trial where half of the participants got omega-3 and the other half got a placebo. Now, what's crucial here is that these people were aged between 55 and 80, they didn't have a history of dementia, and they had low dietary intake of DHA. So this was a population where the theory should work, because these people truly had low DHA.

Plus about half of them were ApoE4 carriers. One of the populations that we hoped this theory would generate benefits for is ApoE4 carriers, because they have a much higher risk of developing Alzheimer's disease.

Up to one in four people carry at least one copy of this gene, and between two and five in every hundred carry two. In large studies of people of European descent, one copy came with about three times the odds of Alzheimer's, and two copies with about fifteen times. Chris Hemsworth learned while making his series Limitless that he carries two copies [11, 12, 13].

The ApoE4 gene also speeds up how fast DHA is broken down, and people with Alzheimer's who carry it have lower DHA in their blood and in the fluid around their brain [1].

The omega-3 group got a high dose, 2 grams of DHA a day. That's more than twenty times what they were getting from their diet. Earlier trials had mostly used 1 gram a day or less, and one explanation for their null results was that not enough DHA was getting into the brain, so this new trial opted for a higher dose so that absorption wasn't an issue. The placebo was corn and soybean oil. The trial was paid for by the US National Institute on Aging and an Alzheimer's research foundation, and the capsules were donated by DSM, a company that makes omega-3 [1, 14].

To check the DHA was actually reaching the brain, about half of the participants had a lumbar puncture, where a needle takes a small sample of the fluid from around the spinal cord, at the start and again after six months. And the group that had the DHA supplements did indeed have higher levels of DHA in this fluid. So we know that absorption is not the problem here.

And it didn't matter whether people carried the ApoE4 gene or not. DHA rose just as much in both [1].

But what we could see is that over the 2-year follow-up period, there were no treatment differences in brain volumes or cognitive performance, which seems to completely fly in the face of the theory that if we raise omega-3, we can have these clinical benefits, including slowing the brain shrinkage.

That held for the hippocampus, the memory centre of the brain, and for the thickness of the brain's outer layer, the cortex. And it was true whether people carried ApoE4 or not [1].

So what are the potential reasons as to why we didn't see a positive outcome here? And what does that mean for omega-3 supplements and brain health?

Why No Difference?

Remember, everyone in this trial, the placebo group included, took a B vitamin supplement for the full two years, so high homocysteine or low B vitamin status is unlikely to be the reason for this disappointing result [1].

Instead, look at how these people were doing. On the main thinking test, scores went up by almost 3 points in both groups. For omega-3 to show a benefit, the placebo group needs to decline, so there's something to slow down. So this study had little chance to test the theory. And it was the same for brain shrinkage: the authors describe the shrinkage of the hippocampus as minimal [1].

And 38 percent dropped out, mostly during COVID [1].

What the trial does show is that, with B vitamins on board, DHA does indeed rise in the fluid around the brain, but two years may not have been long enough to see a cognitive difference for this relatively healthy group [1].

Why did the VITACOG trial show a benefit then? Those people already had memory problems, and the B vitamins helped most in the people with high homocysteine [15].

The LipiDiDiet Trial

But it's not only the VITACOG trial where we see signs of positive effects for B vitamins and omega-3 on brain health. Take a look at the LipiDiDiet trial, which this new trial references. The LipiDiDiet trial gave 311 people either a daily drink that had omega-3s, B vitamins and other nutrients, or a placebo for 2 years.

They were 71 on average, they had early memory problems, and their scans or spinal fluid showed signs of Alzheimer's. Six in ten carried ApoE4 [16].

We have to look closely at the results and not just read the study abstract. While LipiDiDiet didn't reach its primary endpoint, this was likely because it was underpowered, possibly for a similar reason to the new study we looked at. As the trial authors put it, cognitive decline in this population was much lower than expected, and that left the main result underpowered [16].

The researchers had planned for the control group's main thinking score to fall by 0.4 over two years, and they worked out that 300 people would be enough to spot the drink slowing that decline by 40 percent. But the control group's score fell by only about a quarter of that. With so little decline to slow down, a trial of that size had little chance of showing a clear difference [16].

But if we have a look at where the data was trending, which again, we need to be careful not to overinterpret, it is interesting to see the trend. The cognitive decline in the treatment group trended less compared to the control group, and we can see that in a couple of measurements of brain shrinkage, the active group had less brain shrinkage, which reached statistical significance compared to the control group [16].

The hippocampus shrank 26 percent less with the drink. The ventricles, the fluid-filled spaces in the middle of the brain that get bigger as the brain shrinks, grew 16 percent less [16].

The clearest difference was on a rating doctors use to track how someone is coping, both their thinking and their everyday function. That got worse 45 percent more slowly with the drink [16].

The memory scores and total brain volume also leaned towards the drink, though another thinking score didn't, and in the main analysis none of these differences were statistically significant. These were secondary results, so they point to something worth testing rather than prove it. And about the same number of people went on to be diagnosed with dementia in both groups, 41 percent with the drink and 37 percent without [16].

The researchers did keep going to three years, and by then the main thinking score was also significantly better with the drink, 60 percent less decline. The hippocampus shrank a third less. But only 81 people had three-year results, and the company that makes the drink helped fund that part and helped write the report [17].

So What Does This Mean?

Here's my take. In studies such as VITACOG and the LipiDiDiet trial, where there was brain shrinkage in the placebo groups, we see some evidence that the brain shrinkage is less when people have both B vitamins and omega-3 on board, and cognitive scores trended better [8]. Whereas in the new omega-3 trial, where the placebo group's thinking didn't decline and their brain shrinkage was minimal, there wasn't an effect. So while we wait for a longer, larger trial that can definitively answer these questions, to me it's reasonable to aim for healthy B vitamin and omega-3 levels, but not to megadose, because high doses carry their own risks [18, 19]. Plus, there's also the question of whether, in addition to omega-3 and B vitamins, things like creatine and TMG (trimethylglycine) and choline may also contribute to brain health and performance [20, 21].

The researchers behind the new trial made one more point. Many of their volunteers had risk factors for their blood vessels: about half had high blood pressure, and seven in ten weren't physically active. And they suggested that tackling those things together, like exercise and blood vessel health, may help the brain make better use of DHA than a single supplement can [1].

Other studies

A few more findings from the research behind this article, which didn't make it into the video:

  • The World Health Organization's 2026 dementia guideline recommends against (conditionally) B vitamin, omega-3 or multivitamin supplements to reduce the risk of cognitive decline or dementia in people without a diagnosed deficiency [22].
  • The first big omega-3 trial in Alzheimer's, OmegAD (2006), gave 174 people with mild to moderate disease 1.7 g of DHA and 0.6 g of EPA (eicosapentaenoic acid, another omega-3) a day and did not slow their cognitive decline overall; a slower decline showed up only in a subgroup with very mild disease [23].
  • Two other large trials found no cognitive benefit from omega-3: VITAL-Cog (1 g a day, over 2 to 3 years) and the omega-3 arm of MAPT (over 3 years). A 2020 meta-analysis of 38 randomised trials with 49,757 people concluded that omega-3 "probably has little or no effect" on new cognitive impairment or dementia [24, 25, 26].
  • Not every trial shows the VITACOG pattern. In a later analysis of the FACIT trial, folic acid helped cognition in people with lower omega-3 status, the opposite direction to VITACOG [27].
  • VITACOG's competing interests: its lead author, A. David Smith, is named as an inventor on two University of Oxford patents on using folic acid to treat Alzheimer's disease; the tablets were donated and the trial supported by Meda AB/Recip AB; and two authors had past honoraria from Recip and Axis-Shield [9].
  • A 2014 meta-analysis of 11 trials in about 22,000 people found that B vitamins lowered homocysteine without changing cognition [28].
  • In the new trial, serious adverse events were 2.8% with DHA and 7.2% with placebo; there was one case of atrial fibrillation in the DHA group [1].

References

    1. https://doi.org/10.1016/j.ebiom.2026.106316

    2. https://doi.org/10.1001/archneur.63.11.1545

    3. https://doi.org/10.3233/JAD-231467

    4. https://pmc.ncbi.nlm.nih.gov/articles/PMC5836397/

    5. https://doi.org/10.1016/j.neurobiolaging.2006.08.003

    6. https://pmc.ncbi.nlm.nih.gov/articles/PMC3023562/

    7. https://pubmed.ncbi.nlm.nih.gov/14608048/

    8. https://pmc.ncbi.nlm.nih.gov/articles/PMC4927899/

    9. https://pmc.ncbi.nlm.nih.gov/articles/PMC2935890/

    10. https://pubmed.ncbi.nlm.nih.gov/25877495/

    11. https://www.nia.nih.gov/health/genetics-and-family-history/alzheimers-disease-genetics-fact-sheet

    12. https://doi.org/10.1001/jama.1997.03550160069041

    13. https://www.irishtimes.com/culture/film/2022/11/21/chris-hemsworth-to-take-time-off-from-acting-after-genetic-risk-discovered-on-superpowers-tv-show

    14. https://www.jpreventionalzheimer.com/7539-baseline-findings-of-prevente4-a-double-blind-placebo-controlled-clinical-trial-testing-high-dose-dha-in-apoe4-carriers-before-the-onset-of-dementia.html

    15. https://pubmed.ncbi.nlm.nih.gov/23690582/

    16. https://doi.org/10.1016/S1474-4422(17)30332-0

    17. https://pmc.ncbi.nlm.nih.gov/articles/PMC7821311/

    18. https://pubmed.ncbi.nlm.nih.gov/34612056/

    19. https://pubmed.ncbi.nlm.nih.gov/6308447/

    20. https://doi.org/10.1093/nutrit/nuac064

    21. https://pmc.ncbi.nlm.nih.gov/articles/PMC3610948/

    22. https://bigg-rec.bvsalud.org/en/recommendations/0c0d64fb9fb5910c0b0c788898e0335dcd7e5f5b

    23. https://pubmed.ncbi.nlm.nih.gov/17030655/

    24. https://pubmed.ncbi.nlm.nih.gov/35415212/

    25. https://pubmed.ncbi.nlm.nih.gov/28359749/

    26. https://pubmed.ncbi.nlm.nih.gov/32305302/

    27. https://pubmed.ncbi.nlm.nih.gov/33564824/

    28. https://pubmed.ncbi.nlm.nih.gov/24965307/

About Dr Brad Stanfield

Dr Brad Stanfield

Dr Brad Stanfield is a General Practitioner in Auckland, New Zealand, with a strong emphasis on preventative care and patient education. Dr Stanfield is involved in clinical research, having co-authored several papers, and is a Fellow of the Royal New Zealand College of General Practitioners. He also runs a YouTube channel with over 319,000 subscribers, where he shares the latest clinical guidelines and research to promote long-term health.

YouTube LinkedIn
Back to Articles