Omega-3 for APOE4 Carriers: Why the Form Beats the Dose (and What I Actually Do)

Hi Phoenix friend,

A Phoenix member emailed me last week with the question I get more than any other. "Kevin, what's the best omega-3 for a carrier? Which one do you take, and how much?" She had just found out she's 4/4, same as me. I could feel the worry sitting underneath the question.

So I'm going to answer her the way I'd answer if she were across my kitchen table with a coffee, not the way a supplement label would. And it starts with a sentence I wish someone had told me on day one: most of the omega-3 you swallow never reaches our APOE4 brain.

Here's the one line to keep if you read nothing else. For a carrier, the form of your DHA matters more than the amount. Everything below is me showing you why, and then telling you exactly what I do about it.

The thing that finally made this simple for me: omega-3 has two jobs

For years I overcomplicated this. I'd stare at the fish-oil aisle doing dose math, like the goal was one big number. It isn't. The moment it clicked for me was when I stopped thinking about "omega-3" as one thing and started thinking about it as two jobs that happen to share a name.

Job one is your body. Your heart, your blood vessels, your triglycerides, your background inflammation. This job is real and it matters. It's also the easy one. You get omega-3 into your bloodstream, your levels climb, and the payoff basically tracks the total grams. EPA does a lot of the heavy lifting here.

Job two is your brain. This is the one with the bottleneck. Your brain is picky about how it lets DHA in, and for us that pickiness is the whole story. The amount you swallow matters far less than whether it arrives in the one form the brain door will actually open for.

Same nutrient. Two completely different destinations. And once you see that, a lot of the confusion melts, because you stop trying to solve both jobs with the same pill.

The way I picture it: the body job is a firehose, the brain job is a courier. For the body you want volume. For the brain you want one small package walked straight to the right door. You do not dose a courier the way you dose a firehose. Hold onto that, because it's the reason my brain dose is so small, and I'll come back to it.

The door most fish-oil conversations never mention: Mfsd2a

Your blood-brain barrier has a dedicated door for bringing DHA in. It has a name, Mfsd2a, and almost nobody talks about it. In 2014 a team at Duke-NUS in Singapore worked out exactly what it does, in a landmark Nature paper. Here's the part that changes everything:

In plain English: that door only opens for DHA when it shows up packaged as LPC-DHA (lysophosphatidylcholine-DHA). Hand it plain, "free" DHA, which is mostly what your fish-oil capsule becomes after you digest it, and the door stays shut. To prove the door mattered, they bred mice without it. Those mice got smaller brains, memory loss, and severe anxiety. The delivery route was gone, so the DHA never got where it needed to go. (That's animal data, and I'll flag every time we're leaning on animals, because it matters.)

Now the part that made me change what's on my own shelf. Also in 2017, a group at the University of Illinois fed adult mice the exact same dose of DHA in two different forms, free DHA versus LPC-DHA, and then measured the DHA in their brains:

Same dose. One form doubled brain DHA. The other did nothing. The only difference was the packaging. Two years later the same lab repeated it in rats and added a krill-style form, and the pattern held: the phospholipid and LPC forms got into the brain, the standard triglyceride form did not. Two species, several brain regions, same direction.

Why this hits carriers harder. In a 2017 PET study, carrier brains actually took up about 16% more DHA than non-carrier brains, right in the entorhinal cortex, the exact spot Alzheimer's tends to strike first. On the surface that sounds like good news, like we have no delivery problem at all. Rhonda Patrick's 2019 review offered an explanation for the paradox, and I want to be honest that this next part is a hypothesis, not settled fact: in carriers, the "outer" leaky route may let free DHA seep in fast (that 16%), while the orderly inner route your neurons actually rely on, the Mfsd2a route, is the one you want to be feeding on purpose. The 16% is hard data. The "leaky, not usable" reading of it is a well-argued idea, not proof. It's the biggest open question in this whole topic, and I act on it while being upfront that it isn't nailed down.

Put it together and you get the case for feeding the brain door directly: carrier brains grab more DHA but may misroute it, and lose it from the blood faster, and the one well-described door for the productive route only opens for LPC-DHA. That's the molecule the headlines never measured.

So why is my brain dose so small? Because direct delivery isn't flooding.

This is the question the friend at my table always asks next. "If DHA is that important, shouldn't I be taking a big dose of the LPC form? Like the 2,000 mg you see on the strong fish-oil bottles?"

No. And this is the courier-versus-firehose thing made concrete.

The big gram-level doses exist because normal DHA has to flood your entire bloodstream just to get a trickle across into the brain. You're filling a whole reservoir hoping a little seeps through the wall. That's a firehose strategy, and for the body job it's fine.

LPC-DHA doesn't work that way. It walks straight through the Mfsd2a door into the brain. You're not trying to raise your whole-body levels with it. You're couriering a small, correctly-addressed package to one destination. In the mouse work, it took only a modest, targeted dose of the LPC form to double brain DHA, while the same amount of free DHA did nothing. Efficiency, not volume.

So the label dose on a good LPC-DHA product looks tiny next to a gram of fish oil. That is not a weakness. That is the entire point. Comparing its milligrams to your fish-oil grams is comparing a courier to a firehose. Different job, different math.

And the flip side matters just as much for your wallet: for the body job, you do not need the fancy expensive form at all. Do not waste LPC-DHA on your triglycerides. Eating fatty fish, or taking a decent regular fish oil, covers the body job just fine. Save the precision tool for the job that actually needs precision.

That's the plan in one breath: eat a lot of fish (and regular fish oil if you like) for the body, take a small dose of LPC-DHA for the brain. Two lanes.

"But didn't a study just say fish oil is bad?" Yes. Let me reconcile it.

If you've seen a scary headline recently, you're not imagining it. Last month a paper landed (Liao and colleagues, 2026) showing that omega-3 supplement users declined faster on three standard cognitive tests. If you're a carrier who's dutifully taken fish oil since your genotype came back, that lands like a punch in the chest. It did for me.

So let me give the paper its due, then reconcile it, because a few years earlier the same dataset (the ADNI cohort) gave the near-opposite answer. Wei and colleagues, 2023: long-term omega-3 users had a 64% lower risk of Alzheimer's, and the protective effect showed up specifically in APOE4 carriers. Same data. Opposite-looking results. That's not a debunk, it's a puzzle with a clean solution, and it comes down to three things.

What they measured. Liao measured the slope of decline. Wei measured whether you actually crossed the Alzheimer's diagnosis line. You can slip a little faster on a test and still never cross that line.

How they measured your omega-3. This is the big one. Liao measured whether you said you take a supplement. Dose, form, brand, whether you actually swallow it, all unknown. Wei measured the omega-3 sitting in your red blood cell membranes, your actual tissue level. Real tissue beats a plasma snapshot beats "do you take a pill," every single time.

Who was in the study. Liao's group included people who were already declining, some of whom likely started a supplement because they'd gotten worried. That's cause and effect running backward. Wei's was a prevention population.

Here's the honest takeaway. Liao isn't wrong. Liao measured a fuzzy exposure in a mixed population, which limits what you can conclude for a healthy carrier trying to prevent. And neither study asked the question that matters most for us: what form of DHA was in those capsules. That's the hole the whole rest of this piece lives in.

The full ladder, if you're the type who wants it

You don't need this table to act. But a lot of you (I see you, fellow over-researchers) will want the whole ranking, so here it is, ordered by how well each form actually reaches the brain.

A few honest notes before anyone screenshots this as gospel:

• The LPC-DHA brain-doubling is animal data. Replicated across mice and rats and multiple papers, but no human brain-DHA trial of LPC-DHA exists yet. The door (Mfsd2a) is human. The exact size of the effect in people is inferred, not measured. A human trial in carriers, PreventE4, finished in 2024 with results still pending. When it reads out, it could shift this whole conversation, and I'll update you either way.

• Real fish beats a capsule in a way supplements can't copy. Fish naturally carry a small slice of their omega-3 as phospholipid, which your gut partly converts to that same brain-preferred LPC form. Fish-oil capsules carry none of it. The salmon on your plate has a structural head start.

The short version: fish oil is for your heart. Krill is a connecting flight. LPC-DHA flies direct.

What I actually take

For the body lane, food does most of the work. Fatty fish is my main source of protein. Not chicken, not steak, fish. Sardines every other day, a lot of salmon, mackerel in the rotation. It fits the keto cycling I already do, so my diet and my brain goal pull the same direction. And that fish is quietly doing double duty, because it sneaks a little of the brain-preferred form in the back door that no capsule can. If I want extra insurance on the body side, a plain certified fish oil is more than enough there. I don't reach for anything fancy for this job.

For the brain lane, I take Accentrate Omega Max every day, with a meal that has some fat in it. It's the LPC-DHA form, the one built to go through the Mfsd2a door. Here's the honest backstory, because I think you deserve it. I found the research first, the Nguyen Nature paper, the Sugasini brain-DHA studies, Patrick's review. Then I went looking for a product that actually used the LPC-DHA form, and Accentrate is what I landed on. I was taking it long before I ever spoke to the company.

Then something interesting happened: I kept noticing how many people in the Phoenix community were on it too. So we did the thing Phoenix is built to do. We reached out and made it an official partnership, so we could buy as a group and pull the price down for everyone. That's the entire point of a community like ours. On your own, you pay retail. Together, we negotiate. And it compounds: the more of us who order on the community code, the bigger the bulk discount we can unlock for the next carrier who joins.

Phoenix community gets 20% off with code PHOENIX: use this link

Two things I want to say plainly. First, I can't prove Accentrate has changed anything for me. I haven't run a clean before-and-after, and any honest n-of-1 across overlapping supplements is mostly noise. I take it for the mechanism, not because I measured a win on myself. Second, LPC-DHA is the category and Lysoveta is the technology. Accentrate is one product that uses it, not the only one that ever could. If a better LPC-DHA product shows up, I'll switch, and I'll tell you.

One data point from inside our own community, with its caveats attached. Across 1,757 supplement interventions our Phoenix members have logged, LPC-DHA is the #1-rated supplement by member self-report, at 4.83 out of 5, ahead of magnesium glycinate, creatine, and methylfolate. That is not a clinical trial. It's a satisfaction signal from a self-selected group of carriers, shaped by who bothered to log a result. A pattern worth flagging, not proof. It's exactly the kind of thing our Attribution Analysis exists to surface, and to keep us honest about.

How to know it's actually working: the Omega-3 Index

Don't guess. Measure. The one test that gives you a real answer is the RBC Omega-3 Index, and it has to be the red blood cell version, not a plasma test. Plasma wobbles with your last meal. Eat salmon at lunch and your plasma DHA is up by 3pm and back down by Wednesday. Red blood cells live about 120 days, so the RBC number is a four-month moving average of your true tissue level. It's the closest you get to "what does my brain actually see" without a spinal tap.

The threshold comes from Harris and von Schacky, who introduced the index in 2004: an Omega-3 Index of 8% or higher tracked with the greatest protection, 4% or lower with the least. Phoenix optimal is above 8%. The protection curve keeps climbing past 8, and given everything else I'm managing as a 4/4 carrier, I want the headroom.

Here's how I'd actually run it:

1. Order an RBC Omega-3 Index test. A finger-prick at home is fine (OmegaQuant, Quest, others). Usually $60 to $90.

2. Upload the result into the Phoenix Bloodwork module. One heads-up the internet keeps getting wrong: PDF upload is web and laptop only for now, the phone app doesn't do it yet. A current limitation, not a permanent one.

3. Run Attribution Analysis so you can see whether your number moved after a supplement or dose change, not whether it moved for a mouse. Not proof of cause. A link worth watching.

4. Retest in four months. That's how long your red blood cells take to turn over. Sooner is just meal noise. Later is procrastination.

5. Join a Pod if accountability helps, and for most of us it does. A couple of other carriers will keep your four-month retest honest.

And food first, always. The fish isn't the boring part of this plan. It's the foundation. The supplement is me topping up the brain lane on purpose, on top of a diet that's already doing the heavy lifting.

Remember though: blood Omega 3 level is not a good proxy for brain omega 3 levels. Unfortunately we do NOT have reliable ways of measuring if we have enough Omega 3s in our brain. So the best way here is to track functionally the benefits.
Within Phoenix we are also looking at the difference longitudinally between members who take LPC-DHA and those who don’t. Of course like with any other longitudinal studies, there are tons of confounders (e.g. people who decide to invest in their brain health by buying an expensive supplement are more likely to do a lot of other things for their brain health as well), but it is better than nothing.

Where omega-3 can backfire

Three places this can hurt you. I want them out in the open, not buried.

1. Atrial fibrillation at high doses. The big STRENGTH trial gave high-risk patients 4 g/day of omega-3 and saw no heart benefit, plus more new-onset AFib than placebo (2.2% vs 1.3%). REDUCE-IT showed a similar AFib signal at 4 g. The 1 to 2 g/day people use for the brain sits below where that signal is clearest, which is reassuring but not a guarantee. If you have AFib history, palpitations, structural heart disease, or you're on an antiarrhythmic, talk to your cardiologist before you start, not after.

2. Bleeding. Omega-3 thins the blood a little. At 1 to 2 g/day in a healthy adult, not a big deal. On warfarin, apixaban, or dabigatran, or in the week before any planned surgery, it matters. Ask your doctor exactly when to pause it.

3. Rancid oil is worse than none. Fish oil goes off, especially the cheap stuff. If you crack a capsule and it smells strongly fishy, that's oxidation, not "extra concentrated." Oxidized oil causes the exact oxidative stress you're taking omega-3 to fight. Buy certified (IFOS 5-star, low TOTOX), keep it cold, and skip the giant bottle you'll still be finishing eight months from now.

The bottom line

We started with a scary headline (omega-3 users declining faster) and its mirror image from the same data (64% lower Alzheimer's risk in carriers). Both are true. They asked different questions, in different people, with different precision. And neither one asked what form of DHA was in the capsule.

So here's what I'd actually tell my friend at the kitchen table, and it's what I do myself:

Feed the two jobs separately. Eat a lot of fatty fish for your body, add a plain fish oil there if you want, and don't spend a penny of your fancy budget on that lane. Then take a small, targeted dose of LPC-DHA for your brain, because that's the one form the brain door opens for, and because direct delivery is why the dose is small, not underpowered. Test your RBC Omega-3 Index so you're measuring, not hoping. Bring your result and your full medication list to your doctor. The supplement is the last step, not the first.

If someone I love were sitting across from me right now (and I know many of you have already lost the person I'm picturing), I wouldn't lead with a product. I'd say: eat the sardines and the salmon, test your number, talk to your doctor, and then have the LPC-DHA conversation. That's the order I'd give my mom. It's the order I give myself.

References

1. Liao Z-B, Hu Z-C, Zeng G-H, Chen J, Li X-P, Liu Y-H, Yao X-Q, Wang Y-R. (2026). The association between omega-3 supplementation and cognitive decline in older adults. The Journal of Prevention of Alzheimer's Disease. DOI: 10.1016/j.tjpad.2026.100569.^[DOI] URL: https://www.sciencedirect.com/science/article/pii/S2274580726000932

2. Wei B-Z, Li L, Dong C-W, Tan C-C, Xu W; for the Alzheimer's Disease Neuroimaging Initiative. (2023). The Relationship of Omega-3 Fatty Acids with Dementia and Cognitive Decline: Evidence from Prospective Cohort Studies of Supplementation, Dietary Intake, and Blood Markers. American Journal of Clinical Nutrition. PMID: 37028557^[PubMed]. DOI: 10.1016/j.ajcnut.2023.04.001.^[DOI] URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC10447496/

3. Nguyen LN, Ma D, Shui G, Wong P, Cazenave-Gassiot A, Zhang X, Wenk MR, Goh ELK, Silver DL. (2014). Mfsd2a is a transporter for the essential omega-3 fatty acid docosahexaenoic acid. Nature. PMID: 24828044^[PubMed]. DOI: 10.1038/nature13241.^[DOI] URL: https://pubmed.ncbi.nlm.nih.gov/24828044/

4. Sugasini D, Thomas R, Yalagala PCR, Tai LM, Subbaiah PV. (2017). Dietary docosahexaenoic acid (DHA) as lysophosphatidylcholine, but not as free acid, enriches brain DHA and improves memory in adult mice. Scientific Reports. PMID: 28900242^[PubMed]. DOI: 10.1038/s41598-017-11766-0.^[DOI] URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC5596017/

5. Patrick RP. (2019). Role of phosphatidylcholine-DHA in preventing APOE4-associated Alzheimer's disease. The FASEB Journal. PMID: 30289748^[PubMed]. DOI: 10.1096/fj.201801412R.^[DOI] URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC6338661/

6. Yassine HN, Croteau E, Rawat V, Hibbeln JR, Rapoport SI, Cunnane SC, Umhau JC. (2017). DHA brain uptake and APOE4 status: a PET study with [1-11C]-DHA. Alzheimer's Research & Therapy. PMID: 28335828^[PubMed]. DOI: 10.1186/s13195-017-0250-1.^[DOI] URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC5364667/

7. Sugasini D, Yalagala PCR, Goggin A, Tai LM, Subbaiah PV. (2019). Enrichment of brain docosahexaenoic acid (DHA) is highly dependent upon the molecular carrier of dietary DHA: Lysophosphatidylcholine is more efficient than either phosphatidylcholine or triacylglycerol. Journal of Nutritional Biochemistry. PMID: 31665653^[PubMed]. DOI: 10.1016/j.jnutbio.2019.108231.^[DOI] URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC6885117/

8. Yassine HN, Arellanes IC, Mazmanian A, et al. (2023). Baseline Findings of PreventE4: A Double-Blind Placebo Controlled Clinical Trial Testing High Dose DHA in APOE4 Carriers Before the Onset of Dementia. The Journal of Prevention of Alzheimer's Disease. DOI: 10.14283/jpad.2023.77.^[DOI] ClinicalTrials.gov: NCT03613844. URL: https://link.springer.com/article/10.14283/jpad.2023.77

9. Bantugan MA, Xian H, Solomon V, et al. (2023). Associations of ApoE4 status and DHA supplementation on plasma and CSF lipid profiles and entorhinal cortex thickness. Journal of Lipid Research. PMID: 36958720^[PubMed]. DOI: 10.1016/j.jlr.2023.100354.^[DOI] URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC10230261/

10. Schuchardt JP, Schneider I, Meyer H, Neubronner J, von Schacky C, Hahn A. (2011). Incorporation of EPA and DHA into plasma phospholipids in response to different omega-3 fatty acid formulations: a comparative bioavailability study of fish oil vs. krill oil. Lipids in Health and Disease. PMID: 21854650^[PubMed]. DOI: 10.1186/1476-511X-10-145.^[DOI] URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC3168413/

11. Nicholls SJ, Lincoff AM, Garcia M, et al. (2020). Effect of High-Dose Omega-3 Fatty Acids vs Corn Oil on Major Adverse Cardiovascular Events in Patients at High Cardiovascular Risk: The STRENGTH Randomized Clinical Trial. JAMA. PMID: 33190147^[PubMed]. DOI: 10.1001/jama.2020.22258.^[DOI] URL: https://pmc.ncbi.nlm.nih.gov/articles/PMC7667577/

12. Harris WS, von Schacky C. (2004). The Omega-3 Index: a new risk factor for death from coronary heart disease? Preventive Medicine. PMID: 15208005^[PubMed]. DOI: 10.1016/j.ypmed.2004.02.030.^[DOI] URL: https://pubmed.ncbi.nlm.nih.gov/15208005/