For decades, Alzheimer’s disease research, and the clinical approaches that derive from it, have been hyper-focused on amyloid-β and its deleterious effects. But despite billions spent, this line of work has yet to yield a single truly effective drug therapy.

That’s because there’s far more to the Alzheimer’s equation than amyloid plaques. In fact, by the time amyloid plaques are detectable, the underlying disease processes have been going on for years, if not decades.

According to functional neurologist David Perlmutter, MD, it’s time to expand the focus and recognize the role of the microglia—the resident macrophages of the brain and central nervous system—in the etiology of Alzheimer’s.

Putting it very simply, microglial cells have two main phenotypes: one form nourishes and protects neurons, the other form destroys them. It is this latter phenotype that contributes to the development of Alzheimer’s.

Speaking at the 2025 Integrative Healthcare Symposium, Dr. Perlmutter said: “The standard medical view of AD is as follows: ‘The cause of AD is amyloid plaque buildup. And the standard course of management is to wait until there are obvious symptoms of cognitive decline, and then put someone on an amyloid-targeting drug that is not likely to be effective.’ But all that we’re now learning about the risk of AD and other forms of dementia converge on one pathway: a shift in microglial cells from the M2 to the M1 phenotype.”

Microglial phenotypic expression is complex and not entirely binary, in that there is a variety of phenotypic permutations. But generally speaking, the M2 and M1 phenotypes are the most common. The cells have the same genetics, yet the expression and behavior change, explained Dr. Perlmutter, author of the popular book, Grain Brain, among others.

M2 microglia function like loving nurturers for neurons, he said. They phagocytize amyloid-β, remove dead and dying neurons, secrete neurotrophic factors like BDNF, NGF, IFG, regulate synaptic pruning, and maintain the blood-brain barrier.  M2 cells also secrete anti-inflammatory cytokines, and support synaptic repair. Metabolically, M2 microglia operate by oxidative phosphorylation.

M1 microglia spread misfolded proteins, phagocytize healthy neurons, increase reactive oxygen species, and promote inflammation by secreting inflammatory cytokines. Though they are activated by amyloid-β, they can actually impair amyloid-β clearance if activation persists. The M1 phenotype is associated with chronic inflammation, synaptic loss, and neurodegenerative conditions. In contrast to the M2 phenotype, the M1 cellular metabolism is based on glycolysis.

“All that we’re now learning about the risk of AD and other forms of dementia converge on one pathway: a shift in microglial cells from the M2 to the M1 phenotype.”

–David Perlmutter, MD

The key point is that under conditions of systemic metabolic dysregulation and inflammation, M2 microglia can shift their phenotypic expression and morph into the M1 form. As this happens, there’s a widespread shift toward glycolysis within the brain. And that’s not a good thing.

By the time symptoms of dementia begin to manifest, this microglial metabolic shift has been happening for years.

Metabolic Changes

At its core, the M2-to-M1 shift reflects a change in microglial cellular metabolism. In a paper published last year, researchers Viharkumar Patel and colleagues at the University of California-Davis, described Alzheimer’s as “an acquired mitochondropathy.” The change in microglial phenotype is primarily a consequence of impaired ability to make ATP, and the metabolic change occurs prior to microglial activation and subsequent neuroinflammation.

What drives this M2 to M1 shift? Many factors, said Dr. Perlmutter, whose forthcoming book, Brain Defenders, explores this subject in great detail. The presence of amyloid-β is certainly a trigger, but there are other factors, including poor diet, dysregulated glucose metabolism, leaky gut, elevated LDL, pollution exposure, smoking, trauma, excessive alcohol intake, lack of restorative sleep.

Researchers and clinicians alike are slowly realizing that AD and other forms of dementia, like cardiovascular disease, are end-stage manifestations of chronic inflammation.

“Systemic inflammation is a big driver of the M2 to M1 shift,” Perlmutter said, adding that advanced glycation end products (AGEs)—sugars covalently bound to proteins or fats—can bind to microglial cells causing them to shift from M2 to M1 expression. “This partly explains how high blood sugar drives AD and neurodegeneration.”

Is It Reversible?

The big question, of course, is whether the detrimental shift is reversible. To that, Dr. Perlmutter answers with a resounding “Yes!”

He contends that 80%-90% of AD cases can be delayed substantially or actually prevented. “Target the metabolism!! Don’t wait until there are amyloid plaques and then try to treat them.”

He noted that the brain is only 2-3% of the body’s total weight, but it uses 25% of the body’s resting metabolic energy. It’s the most metabolically active organ in the body. “We’ve been hyper-focused on the neurons, but we need to refocus on metabolism and metabolic patterns.”

Lifestyle factors are very relevant to how microglial cells function. “Obesity is the top modifiable risk factor for AD,” Perlmutter said. Citing a comprehensive 2024 research review on prevention of dementia published in The Lancet, he noted that many of the most significant risk factors are essentially metabolic and inflammation-related.

Diet and lifestyle changes to reestablish healthy glucose and lipid metabolism, and to reduce systemic inflammation are essential. There are some signals from animal studies that ketogenic diets may be particularly effective in decreasing microglial activation, and reducing pro-inflammatory cytokines (Fairley LH, et al. Front Immunol. 2021). This needs confirmation from human studies, but mechanistically, it makes sense.

Beyond dietary changes and exercise, Dr. Perlmutter says hyperbaric oxygenation therapies are showing promise. “They enhance mitophagy and oxidative phosphorylation. And that’s the key—to push oxidative phosphorylation.”

A Role for GLP1 Agonists

He also noted that some of the GLP1 agonist drugs are able to induce beneficial changes in mitochondrial function. Since there are GLP receptors on neurons, astrocytes, and microglia, these metabolic changes would presumably affect the brain in a good way.

Pointing to a 2023 study by a multicenter team of Spanish researchers, he said that, “Some GLP drugs do a lot of things that are good for the brain. They enhance mitochondrial function, increase oxygen consumption, and reduce IL-6 and other inflammatory cytokines.”

In a study of 204 mild AD patients, those randomized to treatment with liraglutide—one of the lesser-known GLP1 receptor agonists–showed a 50% reduction in tissue loss in several areas of the brain, compared with those on placebo. Clinical and cognitive measures of brain function also improved in the liraglutide-treated group. The downside is that this drug is not exactly user-friendly. Treatment requires daily subcutaneous injections.

Dr. Perlmutter stressed that not all of the GLP drugs have potential for prevention of dementia. For example, the popular semaglutide (aka Ozempic, Rybelsus, and Wegovy) is a relatively large molecule, and cannot possibly cross the blood-brain barrier.

Gamma Entrainment Shows Promise

Perlmutter said he is also enthusiastic about Gamma Entrainment Using Sensory stimulation (GENUS), an emerging non-invasive light-based therapy developed by researchers at Massachusetts Institute of Technology. The technique involves exposing someone to 40Hz light alone, or in combination with sound. Animal studies show that this can reduce amyloid and tau protein buildup tin the brain, and stimulate beneficial changes in microglial cells and astrocytes.

A phase II human clinical trial of GENUS showed that it can slow brain atrophy, preserve white matter, and improve some measures of cognitive performance. A nationwide phase III study is now underway.

As is the case with the Th1 and Th2 branches of the immune system, or the sympathetic and parasympathetic branches of the nervous system, both the M1 and M2 microglial cells play a role in overall brain health. The problems arise when systemic factors shift the balance toward the M1 expression pattern. The clinical goal, therefore, is not to “eliminate” or “suppress” the M1 phenotype, but to restore a healthy balance.