The Phoenix CommunityThe Phoenix CommunityFree guide for APOE4 carriersGet the Guide
Free guide · For APOE4 carriers

Your brain may already be running low on fuel, and you cannot feel it.

You have heard glucose and insulin matter for your health. Here is the sharper finding: in 1996, researchers scanned people who were completely healthy, sharp memory, normal on every test, and carried two copies of APOE4. Their brains were already burning less glucose in the exact regions Alzheimer’s hits first, the same pattern seen in Alzheimer’s patients, decades before any symptom [Reiman et al., 1996]. Not a diagnosis. A measurable vulnerability you cannot feel yet. This free guide walks through what a continuous glucose monitor can and cannot prove, what fasting can and cannot do for a carrier brain, and the honest verdict on Ozempic, Mounjaro, and the diabetes drugs now linked to lower dementia risk, sorted honestly by how strong the evidence actually is.

Instant PDF · Join thousands of APOE4 carriers · No spam
The APOE4 Metabolic Guide

The brain-fuel gap that shows up decades before symptoms, and the honest verdict on the drugs everyone’s asking about.

The Reiman brain-glucose gradient · CGM, proven or not · the fasting caution nobody prints · Ozempic and Mounjaro, honestly.

Dr. Kevin Tran, PharmD20 pages · PubMed-cited
Brain regions losing fuel72 months, ADNI cohort
23 vs 4 regions of declinecarriers vs non-carriers, Paranjpe et al. 2019
The reframe almost nobody makes

It is not a diabetes problem. It is a brain-fuel problem, decades early.

Most metabolic advice treats glucose as generically bad for you, like sugar or stress. The brain-imaging data is more specific than that, and the honest version is more useful than the scary one.

Three scans, one gradient

More copies, less fuel reaching your brain

In 1996, cognitively normal adults with two copies of APOE4 already showed reduced glucose metabolism in the exact regions Alzheimer’s hits first. A 2005 study found a clean dose-response: zero copies, one copy, two copies, each one turning the brain’s fuel supply down another notch.

Mechanism, not metaphor

Your brain runs a tighter fuel budget than most

The brain burns roughly a fifth of the body’s glucose, most of it through a transporter called GLUT1. Insulin resistance throttles that delivery, a pattern some researchers now call “type 3 diabetes.” APOE4 narrows a fuel line that was already tight.

Regions of significant glucose decline over 72 months · Paranjpe et al., 2019 · ADNI cohort

The gap between carriers and non-carriers was not close.

Non-carriers4 regions
APOE4 carriers23 regions
What’s inside

Sorted honestly, by the strength of the evidence.

Every claim is PubMed-cited, sample sizes included, animal studies clearly labeled.

01

The brain-glucose gradient

Why cognitively normal carriers already burn less fuel in the exact regions Alzheimer’s hits first, and why two copies is worse than one.

The spine
02

The gap that grows

What a 72-month scan study found when it tracked exactly where glucose use kept dropping.

Longitudinal
03

Should you wear a CGM

What a continuous glucose monitor is proven to do, and the one outcome no study has tested.

The gadget
04

Fasting, with a caution

The evidence for a 16:8 window, and the same review’s warning about windows that go too far.

The tension
05

GLP-1s: the trial that failed

What the EVOKE trials actually found in a room that was 60% APOE4 carriers, and what they did not test.

Ozempic, honestly
06

SGLT2 inhibitors and your numbers

The diabetes drug linked to a third less dementia, the honest catch in the randomized data, and the five markers worth testing.

The targets
And the myths worth dropping

Ozempic “failed.” Normal glucose “means you’re fine.” Neither is exactly true.

The guide runs six popular claims against the actual trial data, calmly, one at a time.

The Phoenix Community

Why I built this

Dr. Kevin Tran, PharmD · APOE4/4 carrier · Founder

“I found out I was APOE4 4/4 in December 2024. For two years before that, I tracked my own glucose and insulin numbers in a spreadsheet, alone, at night, trying to work out which claims about my brain were actually real. I am a pharmacist, and I still second-guessed every fasting protocol and every headline about Ozempic. This guide is the spreadsheet I wish I had on day one. What is proven, what is promising, and what is just noise.”

A failed treatment trial is not the same sentence as “GLP-1s do nothing for your brain.” Getting that distinction right is the whole skill.
From the guide · GLP-1s, honestly reported
You don’t have to track it alone

Turn the numbers into a habit you can see.

Inside Phoenix, your glucose and insulin numbers are tracked against APOE4-aware targets, alongside a pod of carriers doing the same thing.

Your own carrier pod

Matched to fellow APOE4 carriers, so a CGM experiment or a fasting question does not have to happen alone at midnight.

Metabolic tracking that speaks APOE4

Log fasting glucose, fasting insulin, and HOMA-IR, and see your own trend against carrier-specific targets instead of guessing.

Bloodwork that speaks APOE4

Track 27 APOE4-aware biomarkers against carrier-specific targets, not generic “normal.”

A community that has done the homework

Hundreds of carriers who have already had the exact CGM, fasting, and Ozempic conversations in this guide.

89%
report biomarker gains in 3 months
27
APOE4-optimized biomarkers
664+
members beating the odds
4.9/5
member rating
Common questions

Frequently asked questions.

Does having the APOE4 gene actually change how my brain uses glucose?

Yes, and it shows up before any symptom does. A 1996 scan study found that cognitively normal adults with two copies of APOE4 already had reduced glucose metabolism in the exact brain regions Alzheimer’s hits first [Reiman et al., 1996]. A 2004 follow-up found the same pattern in one-copy carriers as young as their twenties [Reiman et al., 2004], and a 2005 study found a clean dose-response: more copies of the gene, less glucose reaching those regions [Reiman et al., 2005].

Is a continuous glucose monitor worth wearing if I don’t have diabetes?

The tool itself is proven, in people with diabetes. A 2024 meta-analysis of 25 trials and nearly 3,000 adults, including people without diabetes, found CGM feedback lowered HbA1c by about 0.28% and increased healthy-range time by about 7% [Richardson et al., 2024], though 44% of those trials had a financial conflict of interest with a CGM company. No study has tested whether wearing a CGM changes brain energy or cognition in a healthy APOE4 carrier. The guide frames it as a 2-week self-experiment, not a validated prevention device.

Did the Ozempic Alzheimer’s trial actually fail?

For treating existing Alzheimer’s, yes. The EVOKE and EVOKE+ trials tested oral semaglutide in 3,808 people with early, biomarker-confirmed Alzheimer’s disease and found a flat tie against placebo, a result The Lancet called "not efficacious in slowing clinical progression" [Cummings et al., 2026]. About 60% of participants carried APOE4, so this was close to a real test in people who look like us genetically, and the result did not improve. But EVOKE never tested prevention in healthy people. That trial has not been run, and separate observational cohorts in people with diabetes keep finding lower dementia rates in GLP-1 users [Lin et al., 2025; Tang et al., 2025].

Is intermittent fasting safe for someone my age?

The evidence is genuinely mixed, and the guide does not smooth that over. A 2026 review found a 16:8 eating window had the strongest evidence for weight loss in adults 60 and older [Couto-Alfonso et al., 2026], but the same review’s narrative synthesis linked very restrictive windows of 10 hours or less, and fasts beyond about 12.5 hours, to worse outcomes, including a 58% rise in cardiovascular mortality. An 8-hour window sits inside both the praised protocol and the caution zone. A gentler 12 to 13-hour overnight fast sits outside the flagged range, a reasonable starting point, ideally discussed with your doctor first, especially if you are a premenopausal woman.

What lab numbers should I actually ask my doctor for?

Five: fasting glucose, HbA1c, fasting insulin, HOMA-IR, and your triglyceride-to-HDL ratio, which you can calculate from a standard lipid panel you may already have. HOMA-IR combines fasting glucose and fasting insulin from the same blood draw into one insulin-resistance score, and it is often one of the earliest markers to shift, sometimes years before HbA1c moves. Many standard panels skip fasting insulin entirely, so you may need to ask for it by name.

Get the free guide

Fuel is a lever. Pull the ones that are proven.

Enter your email and we’ll send the APOE4 Metabolic Guide straight to your inbox, plus the research and member experiments we publish each week.

Instant PDF · No spam · Unsubscribe anytime

This guide is educational and not medical advice. Several findings are observational cohort studies that show a link, not proof of cause, and some mechanisms are demonstrated only in animal models, always labeled as such. GLP-1 and SGLT2 medications are not approved or indicated for Alzheimer’s disease; any decision to start, stop, or change them belongs to you and your physician. Fasting protocols beyond a gentle overnight window are not settled science for older adults, especially premenopausal women, and are worth a direct conversation with your doctor before you start.