Neuroscientists Found the Fastest Way to Reverse Brain Aging (And It's Not What You Think)

Episode Overview

• Dr. Mark Hyman (#1 NY Times best-selling author, family physician, and author of The UltraMind Solution and The 100: What to Do Today to Live 100 Healthy Years) interviews Dr. Paul Thompson (Professor of Neurology, Psychiatry, Radiology & Engineering at USC, and leader of the ENIGMA Consortium and AI4AD project) on how big data, neuroimaging, and blood biomarkers are transforming brain health.
• The conversation explores insights from the ENIGMA Consortium—a global collaboration of 2,500 scientists across 47 countries that has analyzed 200,000 brain scans across over 30 brain disorders—moving psychiatry and neurology away from symptom-based categorization (the DSM-5) toward objective biological mechanisms.
• The narrative traces how systemic metabolic health, obesity, hormones, nutrition, and sleep physically reshape brain structure and function, revealing why traditional single-target pharmaceutical approaches to Alzheimer's and mental illness often fall short.
• Listeners learn how emerging tools—including quantitative "Brain Charts," proteomics, genetics, and AI—combine with lifestyle, nutritional, and novel therapeutics to halt neurodegeneration and physically regrow brain tissue.

Key Concepts

• Speaker Disclosures, Funding, and Episode Sponsors: Dr. Hyman discloses his roles as co-founder of Function Health (a platform with 500+ employees and over 500,000 members charging $365/year for 160 blood tests and offering brain/body MRIs via Ezra and NeuroQuant) and Clinical Board member at Seed. Dr. Thompson discloses receiving NIH grants, collaborating on a clinical trial with Eli Lilly comparing a newer antipsychotic against Haldol, and noting his former lab manager Chris Whelan works at Biogen and Pfizer, where he organized the Plasma Proteomics Project (PPP) using UK Biobank data. Episode sponsors include BON CHARGE (15% off with code HYMAN), Paleovalley (15% off first order with code HYMAN), Timeline/Mitopure (up to 20% off first order with Subscribe & Save), Seed DS-01 (24-strain synbiotic, 25% off first month with code 25HYMAN), BIOptimizers Magnesium Breakthrough Advanced 10 (10-form magnesium, 15% off with code HYMAN), and fatty15 (C15:0 supplement, 15% off a 90-day starter kit).
• The Limits of Symptom-Based Diagnosis vs. Big-Data Neuroimaging: Citing former NIMH director Thomas Insel, Hyman notes that the DSM-5 has 100% accuracy in categorizing people by symptoms, but 0% validity in identifying underlying biological mechanisms. Early neuroimaging studies of schizophrenia had sample sizes of only about a dozen people (12) and were met with skepticism when reporting changes in the hippocampus. By scaling to 200,000 brain scans across over 30 disorders (using 100,000 brain scans to compile reference "Brain Charts," and AI to analyze millions of scans), the ENIGMA Consortium overcomes small-sample noise to map the physical reality of brain diseases.
• Alzheimer's Epidemiology, Progression, and Pathology: Among 70-year-olds, about 5% have Alzheimer's disease, and its incidence doubles every 5 years thereafter—reaching 10% at age 75 and 20% at age 80. Time-lapse MRI scans taken 1 or 2 months apart reveal a "wildfire of tissue loss" beginning in the memory system, spreading to language systems, and eventually hitting frontal planning and decision-making regions. A brain scan can show an Alzheimer's patient is 10% below the normal volume range for their age and background, and imaging can detect brain changes 30 years before symptoms appear. However, Hyman notes that billions of federal dollars spent on hundreds of amyloid-targeting studies have largely failed, with current drugs only delaying nursing home placement by a few months. Thompson explains that autopsies and AI imaging reveal many dementia patients do not have amyloid plaques at all; instead, they have Lewy bodies (synuclein) or a recently discovered condition called LATE caused by TDP-43 protein, which attacks different brain regions than classic Alzheimer's.
• Metabolic Psychiatry and the Cardiometabolic-Brain Connection: Brain disorders are deeply tied to systemic bodily health. Thompson notes that 60% of people diagnosed with dementia have cardiometabolic issues that are to some extent treatable. Furthermore, ENIGMA data shows that people who are overweight or obese have brains that look 8 years older on "brain age" measures. In psychiatric conditions like schizophrenia, bipolar disorder, and depression, Thompson (citing colleague Tomas Hajek) explains that antipsychotic medications often cause weight gain and obesity, and patients with high BMI or cardiometabolic problems exhibit far worse brain tissue loss on scans than from the psychiatric illness alone. Conversely, treating their systemic health improves both their symptoms and brain scans. Hyman highlights that schizophrenia patients have inflamed brains and mitochondrial dysfunction, and notes that researchers like Harvard's Chris Palmer and Iain Campbell are using ketogenic diets in metabolic psychiatry to resolve schizophrenia and bipolar disease without medication.
• Brain Networks, Structure vs. Function ("Gas Pedal" and "Brakes"): While structural MRI maps physical tissue volume, functional MRI (fMRI)—discovered around 30 years ago when researchers had subjects tap their fingers or look at a flashing light in a scanner to detect tiny differences in blood oxygenation—maps network activity and "traffic flow." Across ENIGMA, psychiatric disorders group into two network patterns: (1) Internalizing/depressive disorders (depression, anxiety, bipolar depression), which show structural deficits and metabolic/glutamate abnormalities in the deep limbic system (the emotional "gas pedal"); and (2) Compulsive/externalizing disorders (OCD, ADHD, mania, and addiction to alcohol, meth, opiates, or cocaine), which show damage or abnormalities in the frontal lobes (the behavioral "brakes"). While direct brain stimulation like TMS initially raised safety questions due to its radical effect on the brain, it has proven effective for treatment-resistant depression by targeting these networks.
• Normal Brain Aging vs. True Neuroregeneration: Human brains grow during childhood, reach peak volume in their 20s, and experience a gentle shrinking of tissue throughout life. Contrary to medical school dogma that adult brains cannot regrow tissue, ENIGMA proved that severe structural loss is reversible. Out of all psychiatric disorders studied, eating disorders (anorexia) caused the greatest negative effect on the brain (massive tissue loss); yet when patients were hospitalized and their nutrition was restored, all of those brain systems were completely restored on follow-up scans. Similarly, during sleep, neuroimaging shows the glymphatic system acting like a "car wash," sending waves of fluid to cleanse toxic waste, and ENIGMA's sleep group found that improving sleep restores brain biomarkers. In anEli Lilly clinical trial comparing a newer antipsychotic to Haldol in schizophrenia patients, scans taken at baseline and 3 months later proved that the newer drug slowed the physical rate of tissue loss.
• Genetics, APOE Variants, and Stem-Cell Drug Repurposing: Purely genetic developmental brain diseases (like Down, Fragile X, and Williams syndromes) are rare, and only 2% of Alzheimer's cases are purely genetic (caused by presenilin or amyloid precursor protein gene mutations). For the other 98%, genes merely tilt the scales. About 75% of people do not carry the APOE4 risk variant; around 25% carry 1 copy (tripling risk), and about 6% carry 2 copies (increasing risk 9 to 10 times). However, APOE4 risk depends on ancestry: it is worst in Japanese and Chinese ancestry, intermediate in European ancestry, and matters much less in African ancestry because East Africans carry a nearby genetic variation that neutralizes APOE4. Conversely, the rarer APOE2 variant protects against Alzheimer's, though Thompson cautions that carriers often have cardiometabolic problems. To find treatments for genetic risks, Boston University neuroscientist Gyungah Jun engineers induced pluripotent stem cells (iPSCs) expressing Alzheimer's risk genes that turn into astrocytes and neurons, testing existing FDA-approved drugs in vitro to see if they suppress toxic protein products.
• Multi-Omics, Proteomics, and Composite Biomarker Diagnostics: Human biology encompasses 20,000 genes, 5 million genetic variations, 6,000 metabolites, tens of thousands of proteins, and a microbiome of 40 trillion cells containing 10 times as much genetic material as our own genes. Historically, dementia trials relied on a single endpoint: the volume of the hippocampus (a seahorse-shaped structure the size of two teaspoons of gray matter, where being in the lower 5% on a percentile chart flagged disease). Thompson emphasizes that a single brain measure or blood marker is insufficient and brain scans alone do not tell everything. Inspired by pediatric growth charts used when his daughter was born prematurely at 30 weeks, Thompson's team built reference "Brain Charts" using tools like NeuroQuant. While 11,000-protein proteomic panels and metabolomics are currently research tools (such as the UK Biobank's 500,000-person dataset enhanced by the Plasma Proteomics Project) and not yet in routine clinical use, Thompson predicts that within 5 to 10 years, combining brain scans with blood markers will diagnose all major neurological and psychiatric conditions.

Quotes

• At 0:01:51 - "He said, it has 100% accuracy, but 0% validity." - Dr. Mark Hyman quoting former NIMH director Thomas Insel on why the DSM-5 accurately groups psychiatric patients by symptoms but fails to explain the biological cause of their conditions.
• At 0:05:00 - "There was this wildfire of tissue loss in the brain, and you could take a scan, a little like a Polaroid, you'd take a scan, you know, a month later, two months later, and you could make a time-lapse movie where you could see the propagation of the disease." - Dr. Paul Thompson explaining how time-lapse MRI scans visualize the physical spread of Alzheimer's disease from memory centers to language and frontal areas.
• At 0:06:42 - "You could see physical changes in the brain tissue, and if someone was scanned today and in three months, the people that were taking that particular medication would have a slower rate of tissue loss." - Dr. Paul Thompson describing an Eli Lilly schizophrenia trial where neuroimaging proved a newer antipsychotic slowed brain tissue loss compared to Haldol.
• At 0:10:49 - "An X-ray is pretty good. If you break your leg, we go have an X-ray. But the brain's complicated. And if you were a pilot flying over Los Angeles and you were trying to figure out what was going on, I mean, the map isn't all there is to it. You want to have a look at the traffic flow and if is anything blocked or, you know, what's going on." - Dr. Paul Thompson contrasting structural MRI (the physical map) with functional MRI and tractography (the traffic flow of brain networks).
• At 0:19:44 - "60% of people that are diagnosed with dementia have cardiometabolic issues and problems that, to some extent, are treatable. They may be very treatable." - Dr. Paul Thompson highlighting the massive overlap between preventable systemic metabolic dysfunction and dementia diagnoses.
• At 0:21:14 - "And just to be really precise, the amount of difference in their brain on a brain scan was far greater if there was metabolic disease or if there was cardiovascular disease." - Dr. Paul Thompson explaining ENIGMA findings that metabolic and cardiovascular issues cause greater brain tissue loss in psychiatric patients than the psychiatric disorders themselves.
• At 0:25:51 - "And the frontal lobes are kind of like the brakes. So if the limbic system is the gas pedal, the frontal lobe is a little bit like the brakes. And if you've damaged the brakes, there's often a, you know, an out-of-control aspect to behavior." - Dr. Paul Thompson illustrating how frontal lobe deficits drive compulsive disorders like OCD, ADHD, mania, and addiction.
• At 0:29:39 - "When we looked at brain scans of people that were maybe obese, overweight, their brain would look eight years older." - Dr. Paul Thompson sharing ENIGMA data on how excess body weight accelerates structural brain aging.
• At 0:35:40 - "And my cortical thickness, which is sort of the size of your brain, essentially, like went up 8%. And my white matter lesions, which is the level of inflammation, which tends to get worse as you get aged, it went down 22%." - Dr. Mark Hyman detailing his NeuroQuant MRI results 3 months after undergoing an ibogaine treatment in Mexico at age 66.
• At 0:36:56 - "When you're asleep, there's these sort of waves, almost like a car wash, where the brain fluid is cleansing out toxic products from the brain." - Dr. Paul Thompson describing how neuroimaging captures the glymphatic system clearing metabolic waste during sleep.
• At 0:38:38 - "We were looking with brain imaging at people with eating disorders, people with anorexia. And it had the greatest effect on the brain of any disorder that we've studied in psychiatry." - Dr. Paul Thompson revealing that severe nutritional deprivation shrinks the brain more than schizophrenia, bipolar disorder, or depression.
• At 0:38:57 - "When they were treated, when they went to the hospital, when their nutrition was back in balance, all of these brain systems were restored." - Dr. Paul Thompson providing proof from anorexia recovery scans that the human brain can fully regenerate lost tissue when given proper nutrition.
• At 1:03:40 - "I really believe that within the next five to 10 years, we're going to be able to at least diagnose all of the major neurological and psychiatric conditions using a combination of scans, the brain scans we talked about, and blood markers." - Dr. Paul Thompson predicting the near-term clinical integration of neuroimaging and blood proteomics for objective brain diagnostics.

Takeaways

• Optimize routine systemic blood biomarkers to slow brain aging: Track and correct clinical markers that ENIGMA and clinical data directly link to brain structure and function—specifically vitamin D, iron, omega-3 fats, heavy metals, fasting insulin, thyroid hormone, sex hormones (estrogen and testosterone), homocysteine, and methylmalonic acid (a marker of B-vitamin status). Supplementing with folate and B vitamins to lower elevated homocysteine has been shown in clinical trials to physically improve brain scans.
• Aim for optimal—not just "normal"—metabolic targets: Maintain healthy body weight and blood sugar control to prevent accelerated brain aging and dementia risk. While the standard normal cutoff for fasting blood sugar is 100 mg/dL, Hyman notes that levels over 85 mg/dL are associated with a linear increase in cardiovascular disease risk.
• Treat systemic and gut health to resolve neurodevelopmental and psychiatric symptoms: Look beyond psychiatric medications to address root physiological causes. As demonstrated in Hyman's case study of a 12-year-old boy with severe ADHD (who was on Ritalin and 7 medications with asthma, anxiety, insomnia, gut and skin issues, and lead exposure), transitioning from junk food to a whole-foods diet, restoring the gut, supplementing nutrients, and detoxifying lead completely resolved his severe dysgraphia (illegible handwriting) within 2 months without handwriting lessons.
• Consider metabolic and nutritional interventions for mental health and brain restoration: Because brain tissue loss is reversible when nutrition and metabolism are restored (as seen in anorexia recovery scans), explore nutritional frameworks—such as ketogenic diets being researched in metabolic psychiatry for bipolar disorder and schizophrenia—and treat any cardiometabolic side effects induced by psychiatric medications.
• Prioritize sleep to activate glymphatic brain cleansing: Protect deep sleep and treat insomnia or chronic stress so the brain's glymphatic "car wash" can clear toxic metabolic proteins and restore healthy brain biomarkers.
• Utilize emerging dementia blood panels for early risk assessment: Take advantage of newly available blood biomarkers to assess neurodegeneration risk decades before symptoms start, including the FDA-approved beta-amyloid (Aβ42/40) ratio, p-tau217, Neurofilament Light Chain (NfL for neuronal integrity), and GFAP (Glial Fibrillary Acidic Protein for astrocytic health). While Hyman notes we do not yet know the exact optimal age to begin baseline testing, he suggests considering it around age 50 or 60, or earlier if you have a family history or APOE4 genetic risk.
• Pair quantitative brain MRI with longitudinal blood testing: When getting a brain MRI, use quantitative tools like NeuroQuant to plot your brain regions on percentile "Brain Charts" against 100,000 reference scans, and pair imaging with comprehensive longitudinal blood panels (such as Function Health's 160-test annual panel, compared to the US government's NHANES study of 40 tests in 5,000 people) to track whether your interventions are working.
• Track emerging neuro-regenerative therapies with objective before-and-after scans: If exploring novel therapeutics, use quantitative neuroimaging to verify physical brain changes—just as Hyman did at age 66 (inspired by former 15-year Texas Governor Rick Perry) by running NeuroQuant MRIs before, 1 week after, and 3 months after ibogaine treatment in Mexico to document an 8% increase in cortical thickness and a 22% reduction in white matter lesions.