For most of the last century, longevity research looked for a single lever. Caloric restriction. Resveratrol. Rapamycin. Metformin. Each was pitched in its moment as the intervention that would slow aging. Each produced interesting data and an enduring research community. None delivered the transformative effect early enthusiasts hoped for.
Sometime between 2013 — when Carlos López-Otín and colleagues published "The Hallmarks of Aging" in Cell — and 2020, the field quietly changed its mind. Aging, the consensus now holds, is not one process with one lever. It is a dozen partly overlapping biological failures that reinforce each other. Metabolic dysregulation accelerates senescent cell burden. Senescent cells release signals that disrupt metabolism. Epigenetic drift erodes the regulation that holds both in check. An intervention that targets any one of these hallmarks may produce meaningful but bounded benefit. An intervention that targets several, synergistically, is the only strategy that plausibly addresses aging as a system.
This is the logic behind what researchers and practitioners now call the longevity stack: a layered framework that combines interventions acting on different hallmarks of aging. In this article we lay out a 2026-practical version of that stack — three layers, what each layer does, which interventions exist today, and where the evidence is honest versus where it is aspirational.
The Three-Layer Model
The stack we will describe has three layers, each targeting a different biological failure mode:
- Signaling Layer (Peptides). Modulate metabolic, inflammatory, and tissue-repair signals. These are the fast-acting, reversible interventions. Examples include GLP-1 analogs, MOTS-c, GHK-Cu, and BPC-157. They are the layer closest to conventional pharmacology.
- Damage-Clearance Layer (Senolytics). Selectively kill or modulate senescent cells — damaged cells that have stopped dividing but refuse to die and that release pro-inflammatory factors. Examples include dasatinib plus quercetin (D+Q), fisetin, and UNITY Biotechnology's UBX-series drugs.
- Reset Layer (Partial Reprogramming). Restore youthful patterns of gene expression by transiently reactivating developmental transcription factors — typically the Yamanaka factors OCT4, SOX2, KLF4, and sometimes c-MYC — without turning cells back into stem cells. This is the most speculative layer and the one most shaped by delivery technology.
These layers are not independent. Senescent cells secrete factors that disrupt metabolic signaling. Metabolic signals regulate epigenetic marks. Reprogramming may alter the burden of damaged cells. The layers are separable intellectually but not physiologically.
| Layer | Primary hallmarks of aging targeted | Mechanism | Reversibility | Evidence strength in humans (2026) |
|---|---|---|---|---|
| 1. Peptide signaling | Deregulated nutrient sensing, altered intercellular communication, mitochondrial dysfunction | Acute modulation of receptors and pathways | Hours to weeks | Strong for specific indications (GLP-1); mixed for longevity-specific peptides |
| 2. Senolytics | Cellular senescence, chronic inflammation, stem cell exhaustion | Selective clearance of senescent cells | Weeks to months | Emerging; small positive trials, many failures |
| 3. Partial reprogramming | Epigenetic alterations, loss of proteostasis, genomic instability | Transient resetting of gene expression patterns | Potentially durable | Preclinical only; no validated human protocols |
Layer 1: Peptide Signaling
The first layer is the most mature. Peptides have been clinical products for a century (insulin was approved in 1923) and are the only layer with interventions that currently have regulatory approval for chronic use in humans. The question for longevity is not whether peptides work — we know they do for specific conditions — but which peptides influence aging biology versus just a downstream disease.
Three subcategories matter:
Metabolic Peptides
GLP-1 receptor agonists — semaglutide, tirzepatide, and successors — began as diabetes drugs, became obesity drugs, and are now being studied for cardiovascular protection, chronic kidney disease, metabolic-dysfunction-associated liver disease, and neurodegeneration. The SELECT trial (Lincoff et al., 2023, NEJM) showed semaglutide reduced major adverse cardiovascular events by ~20% in patients with overweight/obesity and established cardiovascular disease — a benefit that emerged before substantial weight loss differences, suggesting effects beyond pure weight change.
From a longevity framework, GLP-1 peptides directly target deregulated nutrient sensing and altered intercellular communication. They do not, on current evidence, appear to directly modulate senescent cell burden or reset epigenetic age. Their contribution is metabolic and inflammatory, which matters but doesn't address the whole system.
Mitokines and Mitochondrial Peptides
MOTS-c is a small peptide encoded by the mitochondrial genome itself. Discovered by Pinchas Cohen's lab (Lee et al., 2015, Cell Metabolism), it is released into circulation and acts on skeletal muscle and adipose tissue to regulate metabolic homeostasis in response to stress. Circulating MOTS-c declines with age. Administration in mice improves glucose tolerance, reduces inflammation, and extends healthspan markers.
MOTS-c is interesting precisely because it targets mitochondrial dysfunction — a hallmark that GLP-1 drugs do not directly address. It is not an approved drug in any jurisdiction. Clinical evidence in humans is limited. A handful of longevity-focused practitioners use it experimentally; the field is watching for well-controlled trials.
Tissue-Repair and Regeneration Peptides
GHK-Cu (copper tripeptide) has decades of cosmetic use and some evidence for skin wound healing. BPC-157, derived from a gastric protection protein, is widely discussed in the longevity and sports-medicine community but remains unapproved as a drug anywhere in the world and has very limited human clinical data. Thymosin alpha-1 and thymosin beta-4 have specific immune and regenerative indications.
From a hallmarks perspective, these peptides address stem cell exhaustion, loss of proteostasis, and altered intercellular communication in tissue repair. The honest status: the mechanistic rationale is real, but the clinical evidence base is thin and anyone citing these peptides as "longevity drugs" is going beyond what published human data supports.
Layer 2: Senolytics
The senolytics story starts with a specific observation: senescent cells — cells that have stopped dividing but refuse to undergo apoptosis — accumulate with age, secrete a pro-inflammatory mixture of cytokines, chemokines, and proteases called the senescence-associated secretory phenotype (SASP), and appear to drive age-related dysfunction in nearby tissue. Baker et al. (2011, Nature) showed that genetically clearing senescent cells in mice delayed age-related pathology. The subsequent question became: can we clear them pharmacologically?
The modern senolytic field dates to Kirkland, Tchkonia, and colleagues at Mayo Clinic (Zhu et al., 2015, Aging Cell), who identified dasatinib (a tyrosine kinase inhibitor) plus quercetin (a flavonoid) as a combination that selectively kills senescent cells in multiple models. That combination (D+Q) has since been tested in small human trials:
- Idiopathic pulmonary fibrosis (Justice et al., 2019, EBioMedicine): a 14-person open-label pilot showed improved physical function.
- Diabetic kidney disease (Hickson et al., 2019, EBioMedicine): reduced senescent-cell burden in adipose tissue and skin.
These are small, early, mechanistic studies — not definitive proof of longevity benefit. Larger trials are underway.
Fisetin, another flavonoid, is in clinical trials for multiple indications (including the AFFIRM-LITE trial at Mayo). UNITY Biotechnology took a different approach: localized delivery. UBX1325, an inhibitor of the anti-apoptotic protein BCL-xL, is being tested for diabetic macular edema (BEHOLD trial, 2023–2024) with promising early data in specific endpoints and failures in others. UNITY's stock chart reflects the reality that senolytics have not yet produced a clean clinical win.
| Senolytic | Mechanism | Clinical stage | Notable evidence | Caveats |
|---|---|---|---|---|
| Dasatinib + quercetin | TKI + flavonoid, selective killing | Small human trials | Kirkland et al. pilot studies | Dasatinib has known toxicities |
| Fisetin | Natural flavonoid | Phase 2 trials | Mouse healthspan data (Yousefzadeh et al., 2018) | Bioavailability concerns |
| UBX1325 (foselutoclax) | BCL-xL inhibitor, local injection | Phase 2 DME trials | Mixed results | Not systemic |
| Navitoclax | BCL-2/BCL-xL inhibitor | Oncology + senolytic research | Strong preclinical | Thrombocytopenia |
| Senomorphics (e.g., rapamycin) | Modulate SASP without killing | Various | mTOR inhibition | Not true senolytics |
The honest read on the senolytic layer in 2026: mechanistically compelling, clinically promising in narrow contexts, not yet proven as a general longevity intervention. Exercise — free, accessible, and evidence-backed — produces many of the same inflammatory and metabolic benefits and is probably the most defensible "senolytic-adjacent" intervention any human can adopt today.
Layer 3: Partial Reprogramming
The third layer is where longevity research meets the most speculative biology — and also where the most conceptually exciting data lives. The foundational finding: transient expression of the Yamanaka factors OCT4, SOX2, KLF4, and c-MYC (OSKM) can partially reset age-related epigenetic marks without erasing cellular identity.
The seminal paper is Ocampo et al. (2016, Cell), from Juan Carlos Izpisúa Belmonte's lab at Salk. They used doxycycline-induced, cyclic OSKM expression in progeroid mice and demonstrated extended lifespan and reversal of several aging phenotypes. The key refinement was dose control: continuous OSKM expression caused teratomas; cyclic expression did not. Partial reprogramming was born as a concept.
Lu et al. (2020, Nature), from David Sinclair's lab at Harvard, showed that delivering just three of the factors — OCT4, SOX2, KLF4, omitting c-MYC — via AAV to the eye could restore visual function in aged mice and a glaucoma model, with reversal of DNA methylation age markers in retinal ganglion cells. This was the first reasonably clean demonstration that partial reprogramming could reverse a functional deficit in a living adult mammal.
Zhou et al. (2009, Cell Stem Cell) had years earlier shown that protein-based delivery of Yamanaka factors — fused to cell-penetrating peptides — could induce pluripotency without any DNA entering cells. This is the foothold for peptide-delivered reprogramming: a delivery modality that avoids the cancer risk of permanent genomic expression and allows precise dose-and-wait control.
Companies and Academic Groups
- Altos Labs launched in 2022 with ~$3B in funding and recruited Juan Carlos Izpisúa Belmonte, Shinya Yamanaka, Wolf Reik, Steve Horvath, and other central figures. Its public statements describe working on "cellular rejuvenation programming" as a long-horizon research endeavor.
- Life Biosciences has advanced a program (sometimes referred to as RG-1001) delivering reprogramming factors via AAV for ophthalmic indications, following the Lu et al. eye precedent.
- Retro Biosciences (backed by Sam Altman) works on partial reprogramming and cellular rejuvenation.
- Turn Biotechnologies is developing mRNA-based partial reprogramming.
- Rejuvenate Bio (co-founded by George Church) pursues gene therapy for age-related phenotypes in companion animals and humans.
The honest status of this layer in 2026: zero human clinical evidence for longevity endpoints. No approved protocols. No validated biomarkers of successful reprogramming in living humans. Everything we know about partial reprogramming in mammals comes from rodents and from in vitro cell biology. The field is promising, richly funded, and may or may not deliver the transformative interventions its advocates envision.
How the Layers Interact
The theoretical appeal of a layered stack is that the layers address different problems and may be synergistic:
- Peptides handle acute signaling dysregulation. Metabolic and inflammatory pathways can be tuned in days to weeks. Necessary but insufficient.
- Senolytics clear the accumulated damaged inventory. Senescent cells generate chronic signals that peptides can only mask, not remove.
- Reprogramming resets the regulatory state. If the epigenome has drifted, neither pharmacological signaling nor cell clearance fixes the underlying control problem.
A plausible sequence might be: first stabilize metabolism and inflammation with peptides, then periodically clear senescent cells with a senolytic pulse, then — when safe reprogramming protocols exist — occasionally reset regulatory state. The layers reinforce each other in principle. In practice, no human has ever been treated with all three in a clinical protocol, and nothing is known about interactions.
The Pragmatic 2026 Stack
Here is the honest version of what is actually available to a thoughtful, well-informed adult in 2026 — and what isn't.
| Layer | What's clinically available | Evidence tier | What's experimental | Honest gap |
|---|---|---|---|---|
| Peptide signaling | GLP-1/GIP agonists (semaglutide, tirzepatide) for qualifying patients | Strong (regulatory approval) | MOTS-c, BPC-157, GHK-Cu | Regulatory limits to qualifying patients; research-use-only status for longevity-specific peptides |
| Senolytics | Fisetin (as a supplement, not a drug) | Weak to moderate | D+Q off-label, UBX-series in trials | No approved senolytic for general use |
| Partial reprogramming | Nothing | N/A | Academic and corporate research | Zero approved protocols; mouse evidence only |
| Supporting interventions | Exercise, resistance training, sleep, diet, smoking cessation | Very strong | Rapamycin, metformin, NAD+ precursors | Widely discussed, uneven evidence |
The gap between the three-layer framework and what you can actually do today is large. The framework is directionally right. The implementations are partial. This is precisely why careful writers in this field emphasize that exercise, sleep, and avoiding obvious harms remain the highest-leverage interventions any individual can make — not because the biology is less interesting, but because those interventions work, are safe, and are available today.
Who Is Actually Using This Framework?
Academic research groups at Mayo Clinic (Kirkland, Tchkonia — senolytics), Harvard (Sinclair — reprogramming and epigenetic clocks), Salk (the former Izpisúa Belmonte lab, now distributed), UCLA (Horvath — epigenetic clocks), and Oviedo (López-Otín — hallmarks) shape the intellectual foundations. Companies — Altos, Retro, Life Biosciences, Turn Bio, Rejuvenate Bio, UNITY, Alkahest, BioAge, Insilico — pursue individual layers or combinations. Longevity-oriented clinical practices in several countries offer peptides and some senolytic protocols under supervision, often outside the conventional approval system.
No integrated clinical protocol combining all three layers has been run as a registered trial in humans. The framework exists because the biology points toward it, not because any combined clinical experiment has validated it.
Risks of Stack-Building
Individuals tempted to assemble their own longevity stacks should take the following seriously:
- Interactions are unknown. Combining an investigational peptide with an off-label senolytic produces a drug-drug interaction risk profile that no trial has characterized.
- Compounding products vary. Peptides sold through compounding pharmacies and research chemical suppliers vary enormously in purity, identity, and endotoxin contamination.
- Supervision matters. Any protocol involving injectable or systemic agents should be under qualified medical supervision with baseline and follow-up labs.
- Biomarker enthusiasm can mislead. Epigenetic age tests and other longevity biomarkers remain research tools. A change in a clock reading is not proof of improved biology.
- Opportunity cost. Time and money spent on speculative stacks is time and money not spent on interventions with strong evidence — exercise, sleep, diet, smoking cessation, blood pressure control, cancer screening.
The framework is valuable as a map. It is dangerous as a prescription.
Key Takeaways
- Modern longevity science has moved from single-lever thinking to layered, combinatorial frameworks.
- The three-layer stack — peptide signaling, senolytic damage clearance, and partial reprogramming — corresponds roughly to the hallmarks of aging classification.
- Layer 1 (peptides) is the most clinically mature, with GLP-1 drugs providing strong evidence for metabolic-signaling effects that extend beyond weight loss.
- Layer 2 (senolytics) has promising mechanistic support and limited but growing clinical data; fisetin, D+Q, and UBX-series drugs are the leading candidates.
- Layer 3 (partial reprogramming) is preclinical only in 2026, with important rodent data (Ocampo 2016, Lu 2020) but no validated human protocols.
- No integrated clinical trial has tested all three layers together in humans.
- The gap between framework and practice is large; exercise, sleep, and conventional risk-factor management remain the highest-leverage interventions for most people.
- The stack is a useful map, not yet a prescription.
Frequently Asked Questions
Is there a proven longevity stack I can follow today?
No. There is a conceptual framework supported by diverse research, but no validated clinical protocol exists. The strongest interventions any adult can adopt with confidence remain exercise, sleep, diet quality, smoking cessation, and conventional risk factor management.
Do GLP-1 drugs actually slow aging?
They reduce cardiovascular events and improve metabolic markers in large trials. Whether those effects translate to slower biological aging is an open question. GLP-1 drugs are one of the more interesting signaling-layer interventions because they produce large, reversible effects on several hallmarks of aging simultaneously.
Is fisetin a proven senolytic in humans?
Not yet. Fisetin has compelling mouse data (Yousefzadeh et al., 2018) and is in multiple human trials, but as of 2026 no completed large trial has shown the kind of benefit you would need to call it proven in humans.
Has partial reprogramming been tested in humans?
Not for longevity endpoints. Some disease-specific programs (notably for eye diseases) are in or near clinical development, but no registered human trial has used partial reprogramming as a longevity intervention.
Can you combine peptides, senolytics, and reprogramming?
Theoretically yes. Clinically, no combined protocol has been tested or validated in humans. Combining multiple experimental interventions compounds unknown risks.
Is exercise really competitive with these interventions?
For most people today, yes. Exercise improves almost every hallmark of aging, has no serious downside when done sensibly, and is the intervention with the strongest effect-size evidence in humans. Any discussion of a longevity stack that does not start with exercise is incomplete.
Further Learning
- Hallmarks of Aging Explained — the foundational framework that underlies the stack.
- Epigenetic Clocks and Biological Age — how researchers measure the effects of interventions.
- Yamanaka Factors and Partial Reprogramming — the science behind the reset layer.
- Yamanaka Factors and Peptide Reprogramming Frontier — how peptide delivery changes the reprogramming equation.
- What Are Peptides? A Complete Guide — foundational guide to the signaling layer.
