Chapter 1 · July 2026
What has evidence, and what just has hype
Start with the one thing that’s genuinely strong. In the SELECT trial of more than 17,000 obese patients without diabetes, semaglutide cut death from any cause by about 19% and major cardiovascular events by 20%. That’s the best human evidence any metabolic drug has ever produced for broad mortality benefit, and it’s why a 2025 Nature Biotechnology commentary asked, seriously, whether GLP-1s are the first real longevity drugs. A pill version, orforglipron, cleared Phase 3 and is heading to market — which means access is about to widen dramatically.
The reality checks
The same class also failed where it was hyped: semaglutide flatly did not slow Alzheimer’s in the large EVOKE trials, and the program was shut down. That pattern — metabolic benefit that doesn’t automatically become brain benefit — is the through-line of honest longevity reading. Meanwhile taurine, the viral 2023 “anti-aging biomarker,” was debunked by an NIH study in 2025 that found it doesn’t reliably decline with age at all.
The money and the moonshots
Funding stayed hot despite a broader biotech bust. Retro Biosciences raised at a $1.8 billion valuation; NewLimit closed a $435 million round at $3.1 billion; Altos remains best-funded at over $3 billion. The first-ever human trial of cellular reprogramming — partial Yamanaka factors, the David Sinclair lineage — got FDA clearance and began dosing in age-related eye disease. But note what that is: a first-in-human safety trial, not evidence you can act on. And the cautionary tale sits right beside it — AbbVie began winding down its decade-long, multi-billion-dollar Calico partnership with no approved product.
The unglamorous winners
Converging 2026 reviews keep landing on the same boring truth: sleep quality and physical activity are the strongest behavioral levers on healthy aging, and modest, combined changes beat any single heroic intervention. Aging “clocks” like DunedinPACE are getting better at measuring biological age, but a large 2025 comparison found no single clock is best across diseases — they’re research-grade, not ready to guide your personal choices. One clean policy shift worth knowing: the FDA moved to drop the decades-old “black box” warning on menopausal hormone therapy, reflecting evidence that for women starting within about ten years of menopause, the benefits outweigh the risks.
The open questions
Rapamycin: take it, or wait?
PromisingThe PEARL trial found low-dose rapamycin safe over 48 weeks with gains in lean muscle and reduced pain; it’s the most reliable lifespan extender in mice.
PEARL trial (Aging) Ahead of the evidenceNo human trial has ever shown it extends lifespan, PEARL missed its primary endpoint, and off-label use carries real risks like elevated glucose and impaired healing.
Geroevidence review NAD+ / NMN supplements: worth it?
The optimistic readMore than a dozen human trials confirm NMN and NR safely raise blood NAD+ levels, with early metabolic signals.
Trial review The skeptical readA 2025 meta-analysis of ten trials found no benefit for muscle, strength or function, and NIH scientists warn there’s no controlled data behind the pricey IV drips.
NPR Chapter 2 · July 2026
A first signal that a GLP-1 may slow the clock itself
The GLP-1 story picked up a genuinely new thread in June. Researchers at UC San Diego reported the first randomized, placebo-controlled evidence that semaglutide slows biological aging as measured by epigenetic “clocks.” In a 32-week trial, participants on the drug showed about a 9% slower pace of aging on the DunedinPACE clock and lower scores on PCGrimAge, a clock tied to all-cause mortality risk; a companion pilot found that roughly half of treated participants lengthened their telomeres and tended to walk faster.
Read the fine print
This is an early signal, not a verdict. The aging measures were a post hoc, exploratory analysis of a small trial — about 45 people on the drug and 39 on placebo — run in adults with HIV-associated lipohypertrophy, a specific population. And as this book’s first chapter noted, epigenetic clocks are still research-grade tools: a 2025 comparison found no single clock is best across diseases, so a clock moving is a lead to chase, not proof you should act. The larger test is already running — SURMOUNT-MMO, a roughly 15,000-person tirzepatide trial built around morbidity and mortality endpoints rather than weight — but its results are not yet in.
The open question
Does this change what anyone should do?
A real firstIt’s the first randomized, placebo-controlled hint that a widely used drug nudges multiple aging clocks in the right direction at once.
UC San Diego Easy to oversellA post hoc analysis of fewer than 90 people in one narrow population, on unvalidated surrogate markers, is a promising lead stretched thin when it becomes a longevity headline.
Longevity or Bullshit Chapter 3 · July 2026
The GLP-1 pill reaches the counter
On April 1 the FDA approved orforglipron, sold as Foundayo, as a once-daily tablet used with reduced calories and physical activity for adults with obesity, or adults who are overweight and have at least one weight-related condition. The FDA label says it may be taken with or without food and drink. A separate, time-limited CMS demonstration began July 1: eligible Medicare Part D beneficiaries using Foundayo for weight management can receive it through the Medicare GLP‑1 Bridge with a $50 copay. The Bridge operates outside the ordinary Part D payment flow and is scheduled to end December 31, 2027.
What wider access does and doesn’t prove
Easier access is not stronger evidence. Foundayo’s FDA indication is weight management, not aging, cardiovascular-risk reduction or lifespan extension. The CMS Bridge changes cost for eligible beneficiaries; it does not expand the approved use or answer what happens over decades. Evidence from injectable semaglutide cannot be assigned automatically to this oral drug.
The open question
What does easier access establish?
A practical changeA daily tablet and the temporary $50 Bridge remove specific dosing and cost barriers for eligible adults seeking weight management.
CMS A narrow evidence lineThe FDA approval rests on weight reduction and maintenance. It does not establish that orforglipron slows aging, prevents cardiovascular events or extends life.
FDA Chapter 4 · August 2026
The peptides get their hearing
On July 23 and 24 the FDA’s Pharmacy Compounding Advisory Committee met at the agency’s White Oak campus to consider seven peptide substances for the Section 503A Bulk Drug Substances List — the list of ingredients that traditional, state-licensed compounding pharmacies may use when making a medicine to a specific patient’s prescription. The seven names will be familiar to anyone who has spent time in longevity marketing: BPC‑157, KPV, TB‑500, MOTS‑c, Emideltide (better known as DSIP), Semax and Epitalon. The committee voted to recommend six of the seven for inclusion. Emideltide was not recommended.
Read what was actually evaluated
The FDA’s own agenda records the use it reviewed for each substance, and in most cases it is not the use these peptides are sold for. BPC‑157 was evaluated for ulcerative colitis. KPV for wound healing and inflammatory conditions; TB‑500 for wound healing. MOTS‑c — a mitochondrial-derived peptide widely marketed as an anti-aging compound — was evaluated for obesity and osteoporosis. Epitalon, sold in longevity circles as a telomere and pineal peptide, was evaluated for insomnia. Semax was evaluated for cerebral ischaemia, migraine and trigeminal neuralgia, and Emideltide for opioid withdrawal, chronic insomnia and narcolepsy. Not one of the seven was assessed for slowing aging. That gap between the indication on the FDA’s page and the claim on the sales page is the whole subject of this book.
And nothing has changed yet
The committee is advisory. It reviews evidence and recommends; the FDA keeps the decision, and advisory recommendations are non-binding. All seven substances remain off the 503A list today, are not the subject of USP or National Formulary monographs, and are not components of FDA-approved drug products — which is exactly why they remain ineligible for routine compounding. If the agency chooses to act, it would publish a notice of proposed rulemaking, take public comment for typically 60 to 90 days, and only then decide whether to issue a final rule. The recommendations also reach only 503A pharmacies, not 503B outsourcing facilities, which operate under a separate part of the law and would not be covered by any such change.
The open question
Does a favourable vote mean these peptides work?
What the vote establishesA federal advisory committee read the FDA’s own briefing documents on each substance and voted that six of the seven should be eligible for pharmacy compounding — a considerably higher bar than the grey market these peptides currently occupy.
LegitScript What it does notEligibility for compounding is not approval, the vote changes no rule today, and the uses reviewed were narrow clinical ones — wound healing, insomnia, ulcerative colitis. Nothing in the record establishes that any of these peptides slows aging.
FDA Chapter 5 · August 2026
The inflammation bet is tested at scale, and misses
On July 31 Novo Nordisk reported that ziltivekimab did not meet the primary endpoint of ZEUS. The trial enrolled more than 6,300 people who had atherosclerotic cardiovascular disease, chronic kidney disease and a marker of inflammation — high-sensitivity C-reactive protein at or above 2 mg/L — and gave them a monthly 15 mg injection of an antibody against interleukin‑6, or a placebo, on top of the care they were already receiving. The primary endpoint was how long people went before a major adverse cardiovascular event: cardiovascular death, non-fatal heart attack or non-fatal stroke. The hazard ratio came out at 0.99, with a 95 percent confidence interval of 0.88 to 1.11. There was no difference in death from any cause. Overall adverse event rates were similar to placebo, with a higher proportion of serious infections among those on the drug. The company said full results will be presented at a scientific meeting later in 2026, and flagged a non-cash impairment charge in the third quarter — which the announcement puts as a consequence and the company’s own quarterly filing, two weeks later, describes as something the outcome may result in.
Why this one matters to this book
Chronic low-grade inflammation that rises with age — the idea usually shortened to inflammaging — is one of the load-bearing propositions in longevity medicine, and a great deal of what gets sold on the strength of it. ZEUS is the kind of test that proposition rarely gets: a large, randomised, placebo-controlled trial with hard clinical endpoints rather than a biomarker. What makes the null result informative rather than merely disappointing is that the drug worked mechanically. Novo Nordisk reports that target engagement was confirmed, with the expected reductions in free interleukin‑6 and in C-reactive protein. Inflammation came down and the events did not follow. That is the same shape as the EVOKE failure in this book’s opening chapter, where a drug with strong cardiovascular evidence did not deliver in Alzheimer’s disease: a mechanism moving is not the same thing as an outcome changing.
What the result does not settle
Several limits belong on the record. These are headline results in a company announcement, not a peer-reviewed publication, and nobody outside the trial has seen the full dataset yet. The population was specific: people with established cardiovascular and kidney disease and raised inflammatory markers, not older adults in general. And the programme has not stopped. Two further outcomes trials of the same drug, HERMES in heart failure and ARTEMIS after myocardial infarction, are still running, with results anticipated in the first half of 2027. Separately, in its second-quarter report published on August 4, the company disclosed that it had terminated development of monlunabant, an obesity candidate, for portfolio reasons, and booked a 6.3 billion Danish kroner non-cash impairment of intangible pipeline assets, of which 4.0 billion related to that compound. Two programmes ended in a fortnight, one on evidence and one on portfolio grounds.
The open question
Does ZEUS close the case on lowering inflammation?
A clean answer to the question askedMore than 6,300 people, randomised and placebo-controlled, with confirmed target engagement and hard endpoints. A hazard ratio of 0.99 with a confidence interval of 0.88 to 1.11 leaves little room for a meaningful benefit in this population, and there was no mortality difference.
Novo Nordisk announcement One drug, one population, one endpointThe full data have not been published or peer-reviewed. The trial tested one antibody in people with established cardiovascular and kidney disease, not inflammation in ageing generally, and two further outcomes trials in different populations are still running to 2027.
Novo Nordisk announcement Novo Nordisk second-quarter report