This Doctor Thinks They Have The Longevity Protocol To Prevent Disease | Nir Barzilai

Barzilai leads with exercise as the obvious foundation before any pharmacological intervention. He notes that metformin does not prevent exercise gains on muscle function, only mass — so the combination is still beneficial. His framing: exercise improves cognitive function and immune function beyond muscle, just as metformin does beyond blood sugar. The two are somewhat complementary and the antagonism (on muscle mass) is at worst neutral for older adults who need functional quality more than bulk.

Barzilai's fasting practice reduces the daily eating window to approximately 6 hours without caloric counting. He notes the fasting already 'cuts lots of the stuff I shouldn't eat' without requiring willpower during the eating window — a behavioral benefit beyond the metabolic. He also notes he does not eat more in the eating window to compensate. His weight has remained stable. Intermittent fasting lowers IGF-1 by reducing insulin signaling — potentially part of why it mimics some centenarian biology.

Barzilai frames this as aspirational — 'more aspirational but I'm trying.' He contrasts with 10 years ago when his diet was quite different. His practical message: dietary quality matters but for most people with difficulty complying, metformin provides a bridge — particularly for lower-income populations who lack access to gyms and high-quality food.

Barzilai's comment that he sleeps exactly 6 hours 40 minutes on average every week reveals how consistent biological sleep need is — despite targeting 8 hours, his body settles at the same number. This tracking insight argues for accepting your natural sleep duration rather than lying in bed anxiously trying to reach an arbitrary target.

Barzilai is personally on metformin as a pre-diabetic. His nuanced position: metformin is extremely well-characterized (70+ years of use), FDA-approved, and cheap. For people who cannot or will not change lifestyle, it offers a cost-effective alternative. His current ranking: SGLT2 inhibitors are now his top pick based on AFAR evidence analysis.

Barzilai notes the anti-aging field enthusiastically pushed growth hormone for decades based on the 1990 NEJM study showing improved biomarkers and muscle mass. His centenarian data and UK Biobank analysis directly contradict this — the improvement in muscle mass and other biomarkers comes with downstream costs on cancer risk, insulin resistance, and aging acceleration that outweigh the benefits in older adults.

Barzilai's framework: 'I come with a longevity doctor and I say you cheating on me what should I try. And I said okay if I try x what should I test now and after what will be the decision to go on.' He tried other interventions beyond metformin — some had effects, some did not — and gathered data on what is measurable. His lesson: even a scientist cannot conduct reliable self-experiments without a structured decision process.

Barzilai's centenarian data adds further confirmation: 60% of centenarians carry genetic variants that impair growth hormone action — either a deletion of exon 3 in the growth hormone receptor or other functional mutations. Women centenarians in the lowest IGF-1 quartile live twice as long as other centenarians, with better cognitive function and no difference in physical activity. The centenarian with low IGF-1 has not traded away strength — muscle function is preserved, just muscle mass is reduced. This same logic applies to metformin: metformin lowers IGF-1 as part of its mechanism, which may be central to its longevity benefit.

The three hypotheses tested were: (1) perfect environment/lifestyle — disproved by the data above; (2) perfect DNA free of disease-causing variants — disproved when whole-genome sequencing of 44 centenarians found 230 pathogenic SNPs (5–6 per person that 'should' have made them sick), including ApoE4 homozygotes who by all genetics should be demented at 60 yet were 100 and cognitively sharp; (3) protective longevity genes that slow aging itself — supported. Barzilai himself has one of the published IGF receptor genotypes from his own centenarian studies. The medical cost data reinforces this: the CDC found that people who die after 100 incur one-third the medical costs in their final two years compared to those dying at 70.

The full ranked list from AFAR analysis: (1) SGLT2 inhibitors, (2) metformin, (3) bisphosphonates/etidronate, (4) GLP-1 agonists. Bisphosphonates surprised the field — originally osteoporosis drugs, they show decreased overall mortality in controlled studies beyond just preventing hip fracture death, though the mechanism is unknown (the bone marrow contains immune cells and the drugs may act on those). GLP-1 agonists rank fourth because they are caloric restriction mimetics — caloric restriction extends animal lifespan by 40% and GLP-1 mimics many of those effects; obesity accelerates aging, so GLP-1 creates a reversal of accelerated aging.

Barzilai frames this as the most compelling single example of metformin's pleotropic benefit outside diabetes. The logic: COVID kills by flooding the body with inflammation. Metformin doesn't just lower blood sugar — it suppresses the driver that causes genes to produce cytokines. Additionally, metformin's systemic body-conditioning effect (making a 75-year-old body more resilient to a severe systemic attack) may contribute beyond just cytokine suppression.

Barzilai's reframe: 'if they had smaller muscle and the same power then every gram of muscle is better.' The mechanism — mTOR inhibition — is the same pathway that limits muscle growth but drives cellular repair, autophagy, and anti-aging. Barzilai's position: metformin is not good for bodybuilders, but exercise improves cognitive function and immunity beyond muscle, and metformin similarly has benefits beyond muscle. For older adults, the anti-aging Hallmark improvements in every gram of muscle may outweigh the mass reduction. Some practitioners take metformin only on non-strength-training days to avoid the antagonism.

Barzilai started a company to develop humanin analogs as potential drugs — the challenge is making peptides stable and bioavailable enough to reach effective blood levels. The 8% East Asian variant finding validates humanin causally: population genetics shows reduced humanin activity → increased diabetes risk, the same pathway that genetic studies of drug targets now use as the gold standard for drug development (Regeneron's approach). Commercial peptide injections available online are concerning to Barzilai because manufacturing quality varies and antibody formation against injected peptides may have unknown adverse effects.

The trial design is efficient: 3,000 subjects (vs 12,000 for a cardiovascular outcomes trial) because every person over 65 will get at least one major age-related disease, so the event rate is high enough. The endpoint is a composite cluster, not a single disease, reflecting the geroscience insight that improving aging biology delays ALL diseases simultaneously. Barzilai is also scientific director of the American Federation for Aging Research (AFAR), which is running both TAME and the Superagers Initiative (recruiting 10,000 centenarians + offspring + spouses for genetic discovery) and the FAST biomarkers project.

The mechanism concern: many longevity compounds target mTOR or AMPK pathways. Stacking them could suppress these pathways below levels needed for normal immune function — plasma cells are entirely mTOR-dependent and exhaustion of mTOR inhibition at pharmaceutical doses could impair antibody production. Barzilai's advice: getting these compounds from food (bitter melon, saffron, berberine in diet, coffee chlorogenic acid) is unlikely to cause harm; taking them as supplements at pharmaceutical doses is a different risk category. The lack of clinical trials on combinations means no one actually knows whether combinations are additive, synergistic, or antagonistic.