Is LIfespan Really Half Genetic?

Longevity genetics is usually sold as a tidy family drama: Grandma lived to 98, Uncle Al made it to 101, therefore the ancestral dice are loaded in your favor. Or the opposite: Dad had a heart attack at 54, Aunt Ruth developed dementia, therefore the DNA verdict has been read aloud. Both reactions are emotionally understandable and scientifically premature.

For decades, the standard estimate was humbling. Classic twin studies suggested that genes explain only about 20% to 25% of the variation in human longevity [1]. Then a large family-tree analysis made the genetic story look even smaller, arguing that lifespan resemblance inside families had been inflated by shared culture, geography, socioeconomic status, habits, and the deeply unromantic statistical nuisance that people tend to marry people who resemble them in relevant ways [2]. Families pass down DNA, yes. They also pass down neighborhoods, recipes, doctors, risk tolerances, and food customs with suspicious amounts of butter.

Then the argument got louder. A 2026 Science paper by Shenhar and colleagues claimed that if you correct for extrinsic mortality, the heritability of intrinsic human lifespan is not modest at all. Their estimate was above 50% [3]. That is the sort of number that makes headlines breathe into a paper bag. It sounds like the return of genetic destiny, wearing a lab coat and carrying a spreadsheet. But that is not what it means.

Heritability is one of those words that seems obvious until you ask it to behave. It does not mean that 50% of your personal lifespan is written in DNA and the other 50% is negotiated over oatmeal, gym shoes, and statins. Heritability is a population statistic. It asks how much of the variation among people, in a given population and environment, is statistically associated with genetic differences. Change the environment and the number can change. Biology has many virtues, but producing clean little pie charts for reassurance is not one of them.

Imagine two worlds. In one, everyone has similar sanitation, medical care, food security, smoking exposure, and occupational hazards. Genes may explain more of the remaining lifespan variation because the environmental noise has been narrowed. In another, some people have vaccines, clean water, blood pressure medication, and safe housing, while others do not. There, environment can flatten the genetic signal like a piano dropped from a roof. The genes did not disappear. They just lost the argument to circumstance.

This is why the phrase "intrinsic lifespan" matters. The 2026 paper is trying to separate deaths that reveal something about aging biology from deaths that mostly reveal external hazard [3]. A young death in a car crash says little about whether the person's arteries, immune system, brain, mitochondria, or cancer defenses were aging slowly. A late-life disease death is more entangled with aging biology. The authors argue that earlier studies mixed these piles together and diluted the genetic signal.

The older work was not naive. It was asking a different, and still important, question. The Danish twin study followed 2,872 like-sex twin pairs born between 1870 and 1900 and estimated longevity heritability at 0.26 in men and 0.23 in women [1]. Twin studies are useful because identical twins share nearly all their DNA, while fraternal twins share about half on average. When identical twins resemble each other more in lifespan than fraternal twins do, genes are implicated. But the conclusion of that study was moderate inheritance, not a genetic monarchy.

Ruby and colleagues came at the question from another angle, using enormous public Ancestry family trees [2]. They found that lifespan resemblance extended in ways that were awkward for a purely genetic explanation, including through in-laws. In-laws are useful for Thanksgiving tension, but they are not usually treated as genetic evidence. The point was that assortative mating and shared social worlds can make families look genetically similar for lifespan when part of the resemblance is cultural, behavioral, or economic. The family tree contains DNA, but it also contains county lines, schools, occupations, and spouses chosen from the same social pond.

The 2026 Science paper pushes back by saying: wait, you are measuring the wrong pile [3]. Shenhar and colleagues used mathematical modeling and twin-cohort data, including twins raised together and apart, to estimate the heritability of lifespan after correcting for deaths less tied to intrinsic aging biology. Once that correction was made, they estimated intrinsic lifespan heritability above 50%. If that is right, genes are not bit players in the aging drama. They are more like members of the orchestra who were previously hidden behind the curtain because the fire alarm kept going off.

The finding also fits a familiar clue: exceptional longevity clusters in families. In the New England Centenarian Study, Perls and colleagues studied 444 centenarian families and found that siblings of centenarians had a striking survival advantage compared with the U.S. 1900 birth cohort. Male siblings were at least 17 times as likely to reach 100, and female siblings were at least 8 times as likely [4]. That does not prove the advantage is purely genetic. Siblings can share childhood nutrition, infections, income, regional exposures, health beliefs, and the family rule that nobody discusses symptoms until a limb is nearly detachable. Still, the clustering deserves respect.

The frustrating part is that finding the actual genes has been harder than noticing the family patterns. Genome-wide association studies scan the genome for variants that occur more often in long-lived people than in comparison groups. A 2019 meta-analysis of more than 13,000 long-lived individuals identified several longevity-associated signals, including regions connected to APOE and other genes [5]. Longevity is not controlled by one royal gene ordering the kidneys to behave. It is a many-variant, many-pathway, many-context trait, with common variants, rare variants, gene interactions, environmental interactions, and chance all elbowing each other in the hallway.

APOE is the perfect warning label. APOE variants influence lipid transport and Alzheimer disease risk, and APOE repeatedly appears in longevity genetics. But APOE is not a fortune teller. Some people with risk-associated APOE variants live very long lives. Some people with more favorable versions die earlier from cancer, injury, infection, cardiovascular disease, or sheer biological bad luck. Genes bend probabilities. They do not issue calendar appointments.

The broader human-aging genetics literature lands in the same complicated place. Melzer, Pilling, and Ferrucci reviewed the field and emphasized both the progress and the limitations: genetic studies have identified loci related to lifespan and age-related diseases, but aging traits are measured through imperfect proxies and are shaped by environmental exposures across a lifetime [6]. That is a very polite scientific way of saying that human aging is not a vending machine. You do not insert a genotype and receive a predictable expiration date.

The new estimate should therefore be taken seriously, but not swallowed whole without chewing. A model that separates intrinsic from extrinsic mortality depends on assumptions about what counts as external, how causes of death are classified, and how much hidden biology sits underneath events that look external. Infection is not merely a germ meeting a body; immune resilience matters. Accidents are not purely random; cognition, occupation, frailty, sleep, alcohol use, and risk-taking can shape exposure to danger. Even the category labeled "external" has biology sneaking in through the side door.

None of this makes the paper unimportant. It makes it interesting for the right reason. The useful question is not, "Am I doomed or blessed by my genes?" That is a bad question, though a very human one. The better question is, "How much information about aging biology is hiding in families that age unusually slowly?" If intrinsic lifespan is more heritable than older estimates suggested, long-lived families become more valuable as biological maps. They may point toward pathways that protect against several late-life diseases at once: lipid handling, inflammation, immune function, DNA repair, proteostasis, metabolism, vascular resilience, and other systems that aging likes to rough up.

For an individual reader, the practical answer remains beautifully unsatisfying: know your family history, but do not worship it. If relatives routinely lived into their 90s with delayed disease, that is meaningful. If relatives died early from heart disease, dementia, diabetes, stroke, or cancer, that is meaningful too. But neither story cancels prevention. Blood pressure control, exercise, not smoking, sleep, vaccination, screening, social connection, and competent medical care still matter. A higher heritability estimate means lifestyle acts on a body whose starting conditions differ from person to person.

So is lifespan half genetic? The careful answer is that intrinsic lifespan may be much more heritable than many older estimates implied, and the 2026 Science paper makes a serious argument for that [3]. But your personal lifespan is not half locked in. Heritability describes variation across populations, not destiny inside one person. The exciting possibility is not resignation. It is translation: use genetic luck as a clue, learn the mechanisms, and turn some of those mechanisms into interventions that help people who did not win the family-history lottery.

References

1. Herskind AM, McGue M, Holm NV, Sorensen TIA, Harvald B, Vaupel JW. The heritability of human longevity: a population-based study of 2872 Danish twin pairs born 1870-1900. Hum Genet. 1996;97(3):319-323. doi:10.1007/BF02185763. PMID: 8786073.
2. Ruby JG, Wright KM, Rand KA, Kermany A, Noto K, Curtis D, et al. Estimates of the Heritability of Human Longevity Are Substantially Inflated due to Assortative Mating. Genetics. 2018;210(3):1109-1124. doi:10.1534/genetics.118.301613. PMID: 30401766.
3. Shenhar B, Pridham G, Lopes De Oliveira T, Raz N, Yang Y, Deelen J, et al. Heritability of intrinsic human life span is about 50% when confounding factors are addressed. Science. 2026;391(6784):504-510. doi:10.1126/science.adz1187. PMID: 41610249.
4. Perls TT, Wilmoth J, Levenson R, Drinkwater M, Cohen M, Bogan H, et al. Life-long sustained mortality advantage of siblings of centenarians. Proc Natl Acad Sci U S A. 2002;99(12):8442-8447. doi:10.1073/pnas.122587599. PMID: 12060785.
5. Deelen J, Evans DS, Arking DE, Tesi N, Nygaard M, Liu X, et al. A meta-analysis of genome-wide association studies identifies multiple longevity genes. Nat Commun. 2019;10:3669. doi:10.1038/s41467-019-11558-2. PMID: 31413261.
6. Melzer D, Pilling LC, Ferrucci L. The genetics of human ageing. Nat Rev Genet. 2020;21(2):88-101. doi:10.1038/s41576-019-0183-6. PMID: 31690828.