Brain insulin resistance · Alzheimer's

Brain Insulin Resistance in Alzheimer Disease: Why Type 3 Diabetes

Brain insulin resistance in Alzheimer disease means the brain both makes less insulin and stops answering it, a pattern found in donated brain tissue even from people without diabetes. That is why the disease was named type 3 diabetes.

The label came from pathologists holding brain tissue, not from a glucose meter. What they found explains why a normal blood sugar result says little about the brain’s insulin system.

Alzheimer’s disease earned a diabetes name in a pathology lab, not at a glucose meter. Brain insulin resistance in Alzheimer disease was documented in donated brain tissue, including tissue from people whose charts said they never had diabetes, yet routine care still screens the brain’s fuel problem with a blood sugar test. Measura [Cardiometabolic and Autonomic Health Analysis] cannot look inside a living brain, but it can measure the body-wide insulin picture that travels with it on fasting laboratory panels, years before anyone thinks to ask.

Where the name type 3 diabetes came from

In 2005, a team working in Suzanne de la Monte’s laboratory in Providence, Rhode Island, measured the genes that make insulin and its close relatives, the insulin-like growth factors, along with the receptors that receive those signals. Nerve cells normally make some of these hormones themselves. In brains with Alzheimer’s disease, production was sharply reduced, and so were the receptors and the relay proteins that carry the message inward. The researchers described the disease as a hormonal disorder of the brain that resembles diabetes without being identical to it, and they proposed the term type 3 diabetes.

The combination is the point. Diabetes fails in two ways, too little insulin or cells that stop answering it, and the Alzheimer’s brains were doing both at once.

What the tissue showed: a message that will not pass

Low production was only half the story. In 2012, a group led by Konrad Talbot and Steven Arnold studied the hippocampus, the brain’s memory center, from people who had died with Alzheimer’s disease. Its response to insulin signaling was markedly weakened. The breakdown sat at a relay protein called IRS-1, which sits just inside the cell and passes the insulin signal along. In Alzheimer’s tissue, that relay carried chemical tags that switch it off.

Picture a phone call that connects, rings once and goes dead at the switchboard. Insulin arrives, the receptor picks up, and the relay drops the line.

Three details lift this work above a curiosity. The people studied did not have diabetes. The off-switch tags rose step by step from normal brains to brains with mild cognitive impairment to brains with Alzheimer’s disease, whether or not the person carried the APOE4 risk gene. And the tags tracked worse memory during life even after the researchers accounted for plaques and tangles, the classic hallmarks of the disease. The same tags also rose with small clumps of amyloid, which suggests the damage runs both ways: amyloid may jam the relay, and a jammed relay may speed decline. Physicians can see the screening side of this in screening for insulin resistance past the glucose.

Insulin resistance in a brain without diabetes

This is the part a normal blood sugar result cannot address. Newer autopsy work from the Religious Orders Study, older members of Catholic religious orders in the United States who agreed to yearly testing and brain donation, sharpened the picture. Researchers examined the prefrontal cortex of 150 people, 75 with diabetes matched to 75 without, whose thinking had been tested for a mean of 9.4 years before death.

The surprising result: brains with more activity at a downstream insulin switch called AKT belonged to people whose memory and overall thinking had declined faster. A signal stuck in the on position is not a healthy signal. It behaves like a car alarm that never stops, which the neighbors soon learn to ignore. A 2025 follow-up in 116 members of the same study found that a growth-factor binding protein tied to Alzheimer’s disease partly carried that relationship, one more sign that the insulin system and the disease process are tangled together.

The evidence tier, stated once: autopsy tissue shows what was present at death and cannot prove which change came first. It does establish that the brain’s insulin machinery is abnormal in Alzheimer’s disease, and that the abnormality does not wait for a diabetes diagnosis. Research cohorts also favor clean diagnoses. The patient carrying fatty liver, high blood pressure, poor sleep and a thick waist at once, the one with the most insulin in the bloodstream, is the patient these studies were least built to see, so the mechanism has to be applied to that patient rather than waited on. Because the brain can still take up ketones when insulin signaling fails, how ketones reach an insulin-resistant brain is the natural next question.

The same pattern in living people, years earlier

Studies in living people use a stand-in called HOMA-IR, a calculation from fasting glucose and fasting insulin that estimates how hard the body is working to keep glucose normal. In Australian adults whose thinking was normal, higher HOMA-IR went with weaker verbal memory, executive function and overall cognition, and with more tau protein in spinal fluid. What that estimate means for your own body is covered in insulin resistance and metabolic health.

A Finnish study followed the question across two decades. Researchers selected 43 participants by their HOMA-IR at a mean age of 55.1 and tested them again late in life. Over a 5-year stretch ending at a mean age of 74.8, those who had been insulin resistant in midlife lost more episodic memory, the ability to recall events, than those who had been insulin sensitive, and amyloid build-up carried part of that link. For physicians, the broader list of changeable dementia risks is sorted by what a visit can measure in modifiable dementia risk in primary care.

Why the brain’s insulin problem goes unmeasured

For years I told patients with borderline blood sugar that their results were not too bad. The tissue work above is part of why I stopped. Two biological forces run under a reassuring glucose: a body pouring out extra insulin to hold glucose in range, and a brain whose insulin relay is being jammed by that flood and by amyloid. A third force is not biological at all. The food supply delivers acellular carbohydrate from breakfast to bedtime, the standard visit is built around glucose and hemoglobin A1c because those numbers define diabetes, and memory is handled in a different office from metabolism. Nobody is assigned to the space between them, where insulin resistance before diabetes quietly builds.

What a routine visit checks, and what a measurement can show

No test can see insulin signaling inside a living person’s hippocampus. The tissue studies were done after death, and the brain scans used in research are performed at imaging centers, not by Measura. What can be measured is the body-wide terrain the tissue keeps pointing to, plus a starting point for memory.

Blood work. Laboratory panels can include fasting insulin and fasting glucose, the two inputs to HOMA-IR, the estimate the cohort studies used, alongside hemoglobin A1c.
A thinking baseline. Cognitive assessment records memory, attention and executive function today, so a later score is compared with your own starting point. More is in memory and cognitive screening.

Measura measures; it does not diagnose Alzheimer’s disease, and results go to your physician. How a normal glucose can hide the insulin side is covered in type 3 diabetes and a normal blood sugar test.

The lesson in the name is timing. The tissue shows the insulin relay failing in people without diabetes, and the living cohorts show midlife insulin resistance predicting memory loss decades later. Muscle is the body’s largest glucose sink, and working muscle can pull glucose in through a route that needs less insulin, which is one physiological reason movement and a lighter carbohydrate load reduce the insulin your pancreas must produce. Those levers belong in a conversation with your physician, and they are easier to use when you know where your numbers start. Unmeasured is unmanaged. Questions worth bringing:

Has my fasting insulin ever been measured, or only my glucose?
Do we have a record of my memory and thinking to compare against later?

Frequently asked questions

Is type 3 diabetes an official diagnosis?

It is a research term, not a diagnosis that appears on a lab report or in your chart. Pathologists proposed it after finding that Alzheimer’s brain tissue made less insulin and responded poorly to it, features of both major kinds of diabetes. Researchers still debate how much it drives the disease. The practical point is that this insulin problem does not appear on a glucose test, as what insulin resistance looks like before diabetes explains.

Can the brain be insulin resistant if I do not have diabetes?

The tissue evidence says it can. The hippocampal samples with a blocked insulin relay came from people with Alzheimer’s disease who did not have diabetes, and the relay markers rose through mild cognitive impairment regardless of diabetes status. In living adults with normal thinking, higher insulin resistance tracked weaker memory. A fasting insulin value is one way to see the body-wide side of that problem, described in high insulin with normal blood sugar.

Does insulin resistance cause Alzheimer’s, or does Alzheimer’s cause insulin resistance?

The best reading is that both happen. Amyloid clumps can jam the insulin relay in brain cells, and a jammed relay was tied to memory decline even after plaques and tangles were accounted for. Midlife insulin resistance also predicted late-life memory loss in a Finnish cohort. Direction matters less for you than timing, because the insulin side is measurable long before symptoms. The physician’s view is in insulin resistance and dementia screening.

Why get a cognitive baseline if my memory is fine?

Because the insulin changes seen in brain tissue start before any memory complaint, and a score has meaning only when there is an earlier one to hold it against. Recording yours while you feel well lets your physician compare you with yourself instead of a population average. The reasoning is laid out in what a cognitive baseline is for.

How would I know whether my insulin resistance is improving?

By repeating the same measurements under the same fasting conditions rather than judging by how you feel. Fasting insulin can shift well before glucose does, and glucose may not move at all if it was already in range. Your physician sets the interval and decides what the numbers mean for your care. Practical detail on tracking change is in how to lower insulin resistance and tell whether it worked.

See your insulin side before symptoms

Request Measura testing for fasting insulin, a cognitive baseline and the rest of your metabolic picture, with the results sent to your physician.

Send a message

4477 Woodson Rd, Suite 201, St. Louis, MO 63134. Monday to Friday, 9:00 a.m. to 5:00 p.m. Please do not send symptoms, diagnoses or images through a web form — a website form is not a secure medical channel. Send your name and number and we will call you back.

References

Steen, E., Terry, B. M., Rivera, E. J., Cannon, J. L., Neely, T. R., Tavares, R., Xu, X. J., Wands, J. R., & de la Monte, S. M. (2005). Impaired insulin and insulin-like growth factor expression and signaling mechanisms in Alzheimer’s disease–is this type 3 diabetes? Journal of Alzheimer’s Disease, 7(1), 63-80. https://doi.org/10.3233/jad-2005-7107
Talbot, K., Wang, H. Y., Kazi, H., Han, L. Y., Bakshi, K. P., Stucky, A., et al., & Arnold, S. E. (2012). Demonstrated brain insulin resistance in Alzheimer’s disease patients is associated with IGF-1 resistance, IRS-1 dysregulation, and cognitive decline. The Journal of Clinical Investigation, 122(4), 1316-1338. https://doi.org/10.1172/JCI59903
Tong, H., Capuano, A. W., Carmichael, O. T., Gwizdala, K. L., Bennett, D. A., Ahima, R. S., Arnold, S. E., & Arvanitakis, Z. (2024). Brain Insulin Signaling is Associated with Late-Life Cognitive Decline. Aging and Disease, 15(5), 2205-2215. https://doi.org/10.14336/AD.2023.1117
Tong, H., Petyuk, V. A., Sendtner, M., Sood, A., Bennett, D. A., Capuano, A. W., & Arvanitakis, Z. (2025). Alzheimer’s disease-related cortical proteins modify the association of brain insulin signaling with cognitive decline. Journal of Alzheimer’s Disease, 104(3), 667-677. https://doi.org/10.1177/13872877251319463
Laws, S. M., Gaskin, S., Woodfield, A., Srikanth, V., Bruce, D., Fraser, P. E., et al., & Verdile, G. (2017). Insulin resistance is associated with reductions in specific cognitive domains and increases in CSF tau in cognitively normal adults. Scientific Reports, 7(1), 9766. https://doi.org/10.1038/s41598-017-09577-4
Pietilä, E., Karrasch, M., Helin, S., Snellman, A., Viitanen, M., Jula, A., Rinne, J. O., & Ekblad, L. L. (2026). Midlife insulin resistance and brain beta-amyloid accumulation as predictors of change in late-life cognitive function – A 20-year follow-up study. Neurobiology of Aging, 167, 29-41. https://doi.org/10.1016/j.neurobiolaging.2026.06.004