The Hallmarks of Aging, in Plain Language

Aging is not one thing going wrong. It is a set of processes that unfold at different speeds in different tissues, interact with each other, and eventually converge on the same outcome: a body that is less able to repair itself.

In 2013, a group of biologists tried to impose order on that mess by naming a set of “hallmarks” of aging. A decade later, in Cell, the same core group published a revised version, expanding the list to twelve (López-Otín, Blasco, Partridge, Serrano and Kroemer, 2023). That framework is now the closest thing the field has to a shared map.

It is worth understanding for one specific reason: almost every longevity product on the market claims to target one of these hallmarks. Knowing what the hallmarks are is how you tell a real mechanism from a marketing hook.

The organizing idea

A hallmark is not just “something that changes with age.” Hair goes grey with age; nobody calls that a hallmark. To earn the label, a process has to do three things: appear with age, accelerate aging when you deliberately worsen it, and slow aging when you fix it. That third condition is the demanding one, and it is where most of the human evidence is missing.

A useful review in Frontiers in Cardiovascular Medicine groups the hallmarks into three tiers (Sanada, Hayashi and Morishita, 2025), and that grouping is the single most useful thing to carry away.

Primary hallmarks: the damage

These are the root causes — accumulated damage, with no upside.

Genomic instability. DNA takes hits constantly, from radiation, metabolic byproducts, and copying errors. Repair machinery catches most of it. Over decades, what it misses accumulates.

Telomere attrition. The protective caps on chromosome ends shorten each time a cell divides. When they get short enough, the cell stops dividing.

Epigenetic alterations. The chemical marks that tell a liver cell to behave like a liver cell drift over time. Cells become less certain about what they are. This drift is measurable, and it is the basis of the epigenetic clocks you may have seen sold as “biological age” tests.

Loss of proteostasis. Cells maintain an elaborate system for folding proteins correctly and disposing of the ones that misfold. That system degrades. Misfolded protein aggregates are central to Alzheimer’s and Parkinson’s disease.

Antagonistic hallmarks: responses that overstay

These start as protection and become the problem — the biological equivalent of a fire suppression system that floods the building.

Deregulated nutrient sensing. The pathways that read how much food is around (insulin signaling, mTOR, AMPK, sirtuins) shift with age toward a mode that favors growth over maintenance.

Mitochondrial dysfunction. The organelles that produce cellular energy become less efficient and leak more reactive byproducts.

Cellular senescence. When a cell is damaged beyond safe repair, it can stop dividing permanently rather than risk becoming cancerous. That is protective. But senescent cells do not politely disappear. They sit in tissue secreting inflammatory signals, and they accumulate.

Integrative hallmarks: the visible failure

These are what you actually notice.

Stem cell exhaustion. The reserve populations that replace worn-out cells run down. Wounds heal slower. Muscle rebuilds slower.

Altered intercellular communication. The signaling between cells and tissues degrades — hormonal, neural, immune.

Chronic inflammation. Low-grade, persistent inflammation without any infection to justify it. The field calls it “inflammaging,” and it is defined precisely that way in a 2023 review in Molecular Metabolism: low-grade chronic inflammation during aging, absent overt infection, associated with increased morbidity and mortality in older populations (Baechle et al., 2023).

The 2023 revision added two more that had been building evidence: disabled macroautophagy (the cell’s recycling system failing) and dysbiosis (disruption of the gut microbial community).

Why the framework matters more than the list

Memorizing twelve terms is not the point. The point is the structure.

The hallmarks are not a checklist of separate problems. They are a network, and the network has feedback loops.

The Molecular Metabolism review makes this explicit for inflammation, describing a bidirectional and cyclical relationship between chronic inflammation and age-related disease, and showing that inflammation both results from and drives dysfunction across the other hallmarks — genomic instability, telomere attrition, epigenetic change, proteostasis breakdown, mitochondrial dysfunction. The authors call it a vicious cycle.

This is why single-target interventions disappoint so consistently. If mitochondrial dysfunction drives inflammation and inflammation drives mitochondrial dysfunction, then a supplement that nudges one node of that loop is fighting the loop, not breaking it.

Where the human evidence actually is

Here is the part the marketing skips.

The Frontiers review sorts hallmark-targeting interventions by evidence strength, and the sorting is unflattering. On the stronger end, in humans: metformin, which has geroprotective signals and is already in wide clinical use, and SGLT2 inhibitors such as empagliflozin, which have demonstrated cardiovascular benefit in human populations. Senolytics — drugs designed to clear senescent cells — have reached humans, but the review characterizes the evidence as a first-in-human, open-label pilot study in pulmonary fibrosis. Open-label pilot studies are a starting point, not a result.

On the weaker end, described as promising in preclinical models only: telomerase gene therapy, NAD+ precursors, partial epigenetic reprogramming, and rapamycin.

That list is worth rereading if you have ever paid for an NAD+ product. The mechanism is real. The mouse data are real. The human outcome data are not there yet.

There is also a category the hallmarks framework tends to underweight, because it is not a drug: exercise, sleep, and food touch nearly every hallmark simultaneously — nutrient sensing, mitochondrial function, inflammation, stem cell activity. They are unglamorous precisely because they are not proprietary.

What to do with this

• Use the hallmarks as a filter, not a shopping list. When something claims to “reverse aging,” ask which hallmark, and then ask whether the evidence is in humans or in mice. The gap between those two is where most longevity spending disappears.
• Distrust single-target claims. The hallmarks feed each other. Anything that moves only one node has to overcome the rest of the network.
• Treat “biological age” tests as measurements of one hallmark. Epigenetic clocks measure epigenetic drift. That is a real hallmark. It is not the whole of aging, and the clocks have their own well-documented limits.
• Note which hallmarks you can already influence. Chronic inflammation, mitochondrial function, and nutrient sensing all respond to training, sleep, and body composition — with human outcome data behind them, not just mechanism.
• Watch senolytics, but do not buy them yet. The concept is among the best-supported in the field. The human trials are early and small. That combination should produce interest, not purchases.

The honest summary is that the biology of aging is now well mapped and poorly controlled. We know the mechanisms in unusual detail. We can reliably move very few of them in a living human. Both halves of that sentence are true, and anyone selling you the first half without the second is selling you something.