Large Genome-Wide Study Finds Only a Few Genetic Influences on Human Longevity

The influence of genetic variants on natural variations in human longevity is a very complex matter. The evidence to date supports a model in which thousands of genes have individually tiny, conditional effects. Near all associations identified in any given study population have failed to appear in any of the other study populations, and effect sizes for the very few longevity-associated genes that do appear in multiple studies are not large in the grand scheme of things. These variants provide a small increase in the odds of living to be very old, but the individuals bearing them are still diminished and damaged by aging. The genetics that determine how cellular metabolism gives rise to variations in aging are of great scientific interest, but there is nothing here that can act as the foundation for therapies that will help people to live significantly longer.

The extent of the role of genetic variation in human lifespan has been widely debated, with estimates of broad sense heritability ranging from around 25% based on twin studies to around 16.1% based on large-scale population data. One very recent study suggests it is much lower still (less than 7%), pointing to assortative mating as the source of resemblance amongst kin. Despite this modest heritability, extensive research has gone into genome-wide association studies (GWAS) finding genetic variants influencing human survival. Only two robustly replicated, genome-wide significant associations (near APOE and FOXO3) have been made to date, however.

An alternative approach is to study lifespan as a quantitative trait in the general population and use survival models to allow long-lived survivors to inform analysis. However, given the incidence of mortality in middle-aged subjects is low, studies have shifted to the use of parental lifespans with subject genotypes, circumventing the long wait associated with studying age at death in a prospective study. In addition, the recent increase in genotyped population cohorts around the world, and in particular the creation of UK Biobank, has raised GWAS sample sizes to hundreds of thousands of individuals, providing the statistical power necessary to detect genetic effects on mortality. A third approach is to gather previously published GWAS on risk factors thought to possibly affect lifespan, such as smoking behaviour and cardiovascular disease (CVD), and estimate their actual independent, causal effects on mortality.

Here, we blend these three approaches to studying lifespan and perform the largest GWAS on human lifespan to date. First, we leverage data from UK Biobank and 26 independent European-heritage population cohorts to carry out a GWAS of parental survival. We then supplement this with data from 58 GWAS on mortality risk factors. Finally, we use publicly available case-control longevity GWAS statistics to compare the genetics of lifespan and longevity and provide collective replication of our lifespan GWAS results.

We identified 11 novel genome-wide significant associations with lifespan and replicated six previously discovered loci. We also replicated long-standing longevity SNPs near APOE, FOXO3, and 5q33.3/EBF1 - albeit with smaller effect sizes in the latter two cases - but found evidence of no association (at effect sizes originally published) with lifespan for more recently published longevity SNPs near IL6, ANKRD20A9P, USP42, and TMTC2. Despite studying over 1 million lives, our standard GWAS only identified 12 variants influencing lifespan at genome-wide significance. This contrasts with height (another highly polygenic trait) where a study of around 250,000 individuals found 423 loci.

This difference can partly be explained by the much lower heritability of lifespan (0.12 versus 0.8 for height), consistent with evolution having a stronger influence on the total heritability of traits more closely related to fitness and limiting effect sizes. In addition, the use of indirect genotypes reduces the effective sample size to 1/4 for the parent-offspring design. When considering these limitations, we calculate our study was equal in power to a height study of only around 23,224 individuals, were lifespan to have a similar genetic architecture to height. Under this assumption, we would require a sample size of around 10 million parents (or equivalently 445,000 nonagenarian cases, with even more controls) to detect a similar number of loci.

Individual genetic variants that increase dementia, cardiovascular disease, and lung cancer - but not other cancers - explain the most variance in lifespan. We hoped to narrow down the search and discover specific genes that directly influence how quickly people age, beyond diseases. If such genes exist, their effects were too small to be detected in this study. The next step will be to expand the study to include more participants, which will hopefully pinpoint further genomic regions and help disentangle the biology of ageing and disease.



This is good and bad. The bad is that there is no easy win for anybody. The good is that if there is a win it will apply to everybody

Posted by: cuberat at January 17th, 2019 10:57 AM

The results of this study do not square with the results of long-lived family studies and twin studies of lifespan. I don't think we have heard the last word on the genetics of lifespan yet.

Posted by: Biotechy Marcks at January 17th, 2019 11:36 AM

OTOH personalized genomic medicine is a win for everyone, and it will be easy soon - or at least that's what many are saying. There was a very interesting report recently about COMT, vitamin E supplementation, and cancer risk:

In a pre-SENS era, knowing what interventions to take and what exposures to avoid could do a lot for some people - perhaps less for others, but knowing is a good thing in itself (some may dispute that point).

I watched the George Church talk you linked to (thanks) and had a much more positive impression of it than you seemed to have had. The Nebula Genomics project sounds really promising*. Perhaps I just like George Church (even though I disagree with him on a few things); maybe it's his voice I find so charming, or more likely, it's that he says what I want to hear.

* more on Nebula Genomics:

Posted by: CD at January 18th, 2019 11:27 AM

Just got a notification from 23andme - confirmation that I'm heterozygous for the Familial Mediterranean Fever gene. I had come across it previously when doing some of my own checking with Promethease, but couldn't believe what I was seeing - I already have a bunch of bad SNPs, what are the odds? I really am the poster child for Muller's Ratchet.

That's something a lot of people who are proponents of the 'just get back to a more ancestral diet (however they define it) and lifestyle (we can all agree it involved a lot more exercise and probably fasting) and you won't need supplements or drugs' camp don't seem to understand - some of us aren't like our ancestors. Without a warm bed to retreat to and all the other supports modern life has to offer, I would never have made it to adulthood, much less 50.

Posted by: CD at January 18th, 2019 1:10 PM

A genome-wide association study doesn't look at the whole genome. If they looked at a million SNPS, they looked at a small fracture of the 3 billion base pairs making up the whole human genome. Whole genome sequencing is only beginning to be cheap enough to do on large groups of people.

Posted by: NY2LA at January 19th, 2019 4:33 PM

With diet if I saw people on one diet living to 140, and normal people living to 75 or whatever hte life expectancy is, then I would take it more seriously. But from what I've seen we are talking a matter of a few years at best gained.

The longevity genes seem the same. Its not like some guy from a long lived family is living to 200. We are talking a few years difference in the rate of aging.

Posted by: aa3 at January 20th, 2019 5:27 AM


Agreed, it seems if you're of normal/average weigh and exercise, plus don't smoke, those seem to be the biggest influence to basic good health.

So many different diets. Just eat plant based foods with nuts and you're good to go, IMO.

Posted by: Robert at January 20th, 2019 12:10 PM

Only 1 in 250,000 make it to 105, and only 1 in 50 million make it to 110, so you are really looking at the far extremes of human lifespan.

Posted by: Biotechy Marcks at January 20th, 2019 2:13 PM

@NY2LA: Good point. For the longevity gene FOXO3A, there are about 15 known SNP's that have longevity benefits other than the SNP studied in many research studies. That is probably true for many genes, that probably have other nearby SNP that also benefit longevity. Also, genetics becomes more and more of a factor in lifespan the older you become. So at 100 genetics may constitute 40%, while at 105 it is 50%, and is thought to increase as a factor at even older ages rather than plateau.

Posted by: Biotechy Marcks at January 20th, 2019 2:52 PM

re: diet
No, diet won't get you radical life extension, but here are cases where it could mean as much as 30 years difference and at this point in history, that could be a big deal. The vegan advocate, Michael Greger, often uses the example of his grandmother, who was at death's door due to CVD in her 60's, but switched to a vegan diet and lived to her 90's.

However, I suspect there's a lot of inter-individual variation in response to diet with respect to specific diseases. Greger also eschews all supplements except B12, D, and algal DHA, stating that everyone should be able to get what they need from a whole foods plant-based vegan diet, which ignores inter-individual variability (the point I was trying to make about some diet/lifestyle advocates). Overall I think Greger gives excellent advice, but for his absolutist stance on supplements as well as his stance on fish consumption (he has tried to draw a causal connection between fish consumption and Alzheimer's Disease when fish consumption is associated with lower AD risk. Also, fish consumption - despite all the contamination problems - is associated with lower all-cause mortality).

Posted by: CD at January 21st, 2019 10:44 AM

Eight times as many women as men make it to 100, which is quite a genetic advantage. In my family tree, 3 males and 3 females made 100, with the oldest making 106. I like the 1 to 1 male to female odds in my family compared to the 1 to 8 odds in the general population!

Posted by: Biotechy Marcks at January 21st, 2019 4:48 PM

Here's an article for you, and it's not paywalled :)

In mice at least the longevity gap can be decreased by acarbose. It would be interesting to know if xylitol has the same effect since it also inhibits alpha glucosidase and alpha amylase; unfortunately it has the same gastrointestinal side effects.

Posted by: CD at January 22nd, 2019 12:59 PM

@CD: Thanks for the excellent research article addressing the genetic longevity differences between men and women, and showing the limitations of the GWAS studies on human longevity, especially genetic sex differences. The GWAS longevity studies are not the be all and end all of the genetics of longevity!

Posted by: Biotechy Marcks at January 22nd, 2019 7:49 PM

You are most welcome.
You might also like this talk by Brendan Frey on deep learning and genomics:

Posted by: CD at January 22nd, 2019 8:44 PM

I like to see research papers that show you real results that will be useful from your own lifespan perspective, which that Zeng paper does on gender related longevity with data produced on several important SNP's. That way I am able to look up in 23andme DNA database to see if I have the longevity alleles. Promthease is helpful as well, and other information is possible to easily obtain such as epigenetic age estimates, telomere lengths of leucocytes, etc.

Posted by: Biotechy Marcks at January 23rd, 2019 10:30 AM

Post a comment; thoughtful, considered opinions are valued. Comments incorporating ad hominem attacks, advertising, and other forms of inappropriate behavior are likely to be deleted.

Note that there is a comment feed for those who like to keep up with conversations.