Type 3 diabetes is caused by insulin resistance in the brain. Unlike type 2 diabetes, where cells throughout the body stop responding properly to insulin, type 3 diabetes refers specifically to neurons losing their ability to use insulin, which leads to cell death, chronic inflammation, and the buildup of the toxic plaques associated with Alzheimer’s disease. The term isn’t an official medical diagnosis yet, but it reflects a growing body of evidence that Alzheimer’s may be, at its core, a metabolic disease of the brain.
How Brain Insulin Resistance Develops
Your brain depends on insulin not just for managing blood sugar but for keeping neurons alive, forming memories, and clearing cellular waste. Insulin binds to receptors on neurons and triggers a chain of signals that power mitochondria, protect synapses, and regulate cell survival. When those receptors stop responding normally, the entire system breaks down.
The breakdown starts when inflammatory signals, oxidative stress, or excess fat molecules cause a chemical modification to the docking proteins that relay insulin’s signal inside the cell. Instead of passing the message along efficiently, these proteins become inhibited. The result is that neurons can’t produce enough energy, toxic byproducts accumulate, and the connections between brain cells begin to degrade. This is the same basic problem that occurs in type 2 diabetes, just happening inside the skull rather than in muscle or liver tissue.
The Amyloid Connection
One of the most important consequences of brain insulin resistance involves a cleanup enzyme called insulin-degrading enzyme (IDE). This enzyme has two jobs: it breaks down insulin, and it breaks down amyloid-beta, the protein fragment that clumps into the plaques found in Alzheimer’s brains. The problem is that IDE can only do so much work at once. When insulin levels are chronically high, as they are in people with insulin resistance, the enzyme gets monopolized by insulin and can’t keep up with amyloid-beta clearance.
Research in mice genetically engineered to lack IDE confirmed this competition directly. When IDE was absent, amyloid-beta levels rose. When excess insulin was added to normal mice, amyloid-beta degradation dropped to the same low level seen in the IDE-deficient animals. The implication is straightforward: chronic high insulin levels in the body can starve the brain of its amyloid cleanup capacity, allowing plaques to accumulate over years or decades.
At the same time, insulin resistance removes the brain’s natural brake on an enzyme that adds chemical tags to tau, another protein involved in Alzheimer’s. Without insulin signaling to keep this enzyme in check, tau becomes overloaded with these tags, causing it to twist into tangled fibers that destabilize the internal scaffolding of neurons. Plaques on the outside, tangles on the inside: both hallmarks of Alzheimer’s trace back to the same insulin signaling failure.
Inflammation as Both Cause and Consequence
Inflammation doesn’t just result from brain insulin resistance. It actively drives it. A key inflammatory molecule called TNF-alpha, released when amyloid-beta oligomers (small toxic clusters of the protein) irritate brain tissue, directly blocks the insulin signaling machinery inside neurons. Research published in Cell Metabolism demonstrated this loop in both mice and monkeys: amyloid-beta oligomers triggered TNF-alpha release, which activated stress enzymes that shut down insulin receptor signaling, which caused synapse loss and memory impairment.
This creates a vicious cycle. Insulin resistance allows amyloid-beta to accumulate, amyloid-beta triggers inflammation, and inflammation deepens insulin resistance. Fat-derived molecules called ceramides add another layer. These lipids, which increase with obesity and metabolic dysfunction, activate inflammatory pathways in the brain’s immune cells and simultaneously disable a key survival signal in neurons. Ceramides essentially bridge what’s happening in the body (metabolic syndrome, excess fat) with what’s happening in the brain (neuroinflammation, tau tangles).
Metabolic Syndrome and Diet
The most significant modifiable risk factor is metabolic syndrome, the cluster of conditions that includes abdominal obesity, high blood pressure, elevated blood sugar, high triglycerides, and low HDL cholesterol. A study of nearly two million people published in Neurology found that metabolic syndrome was associated with a 24% higher risk of young-onset dementia overall and a 12% higher risk of Alzheimer’s specifically. The risk was particularly elevated in people aged 40 to 49 and in women.
Diet plays a direct role. Chronic high-fructose consumption has been shown to disrupt insulin signaling in the brain, impair synaptic plasticity (the mechanism behind learning and memory), and increase the risk of Alzheimer’s disease. These effects aren’t limited to extreme diets. The pattern of high sugar intake, processed food, and excess calories that drives type 2 diabetes appears to simultaneously set the stage for insulin resistance in the brain.
Genetic Risk: The APOE4 Factor
Not everyone with metabolic problems develops Alzheimer’s, and not everyone with Alzheimer’s has metabolic problems. Genetics help explain the gap. The APOE4 gene variant, the strongest known genetic risk factor for Alzheimer’s, directly interferes with insulin signaling in neurons. Research published in Neuron showed that the apoE4 protein physically interacts with insulin receptors and traps them inside cellular compartments called endosomes, preventing them from reaching the cell surface where they need to be. With fewer functional insulin receptors available, neurons can’t respond to insulin properly, even when insulin levels are normal.
This effect worsens with age. As the brain ages, apoE4 protein becomes more prone to clumping, and the endosomal system that recycles receptors becomes less efficient. The combination means that APOE4 carriers face a progressively steepening disadvantage in brain insulin signaling as they get older, which may explain why APOE4 carriers tend to develop Alzheimer’s earlier than non-carriers.
How Brain Insulin Resistance Is Detected
Brain insulin resistance can’t be measured with a standard blood test. The most established method is a type of brain scan called FDG-PET, which tracks how much glucose different brain regions are consuming. In people with early cognitive impairment, reduced glucose metabolism appears first in the posterior cingulate cortex and hippocampus, two areas critical for memory. A meta-analysis of 24 studies found this scan correctly predicted which people with mild cognitive impairment would go on to develop Alzheimer’s with 88% sensitivity and 84% specificity. These metabolic changes can appear years before noticeable memory problems.
Treatments Targeting Brain Insulin Resistance
Because type 3 diabetes isn’t yet an official clinical diagnosis, there are no approved treatments specifically for it. But several approaches are in active testing. Intranasal insulin, which delivers insulin directly to the brain through the nasal passages, has reached Phase 2/3 trials for Alzheimer’s. A trial completed in September 2024 found that insulin treatment was associated with improvement on a composite cognitive measure and increased white matter volume in 42 participants, with no treatment-related side effects or blood sugar changes.
GLP-1 receptor agonists, the class of drugs that includes semaglutide (used for type 2 diabetes and weight loss), are also being investigated. These drugs can cross into the brain and appear to locally improve insulin sensitivity, restoring energy balance in neural circuits. A trial combining intranasal insulin with semaglutide in people with mild cognitive impairment and metabolic syndrome began in January 2024, with results expected by December 2027. A separate trial showed that one of these drugs engaged insulin and cell-survival signaling pathways in the brains of patients with Parkinson’s disease, suggesting the approach has broad potential across neurodegenerative conditions.
The combination of a diabetes drug with a tau-lowering agent is also being explored. In the completed 2024 trial, the diabetes drug empagliflozin, with or without intranasal insulin, decreased levels of tau protein in cerebrospinal fluid, a potential sign that the tangle-forming process was being slowed.