Some very few humans live for decades longer than the average, and the available evidence suggests that at very late age - and ever increasing frailty - genetic variation becomes an increasingly important determinant of longevity. To be clear, almost everyone with a superior genome dies before reaching a century of age: the odds of making it that far are tiny regardless of your genes. But all it takes is a small increase in those tiny odds to ensure that the present population of very old people is weighted in favor of those who are slightly more resilient. So don't imagine that this is something worth recapturing in a therapy. The study of genetics and natural variations in human longevity will, I suspect, be a transitory curio of our short era. We stand in a thin slice of history in which medical technology is advanced enough to catalog genetics and cellular biochemistry, but still too primitive to bring aging under medical control. There isn't much of a gap between those two thresholds of progress; the second follows quite quickly after the first. After rejuvenation therapies are developed, well within the lifetimes of most of those reading this today, after the creation of a comprehensive toolkit to repair all of the forms of cell and tissue damage that cause aging, few people will ever get to the point of being so damaged that their genes are relevant to how long they can survive in that reduced state. There will be little interest in the study of the condition of being aged. How many researchers study how genes affect the chance of surviving smallpox without modern treatments? Not many.
For today, however, the genetics of longevity is one of the more energetic parts of aging research, a field usually the neglected, poorly-funded stepchild of the medical research community. Genetics is a hot topic, and progress in biotechnology is causing both a dramatic fall in cost and dramatic increase in capabilities for the tools used by researchers. Great advances in gathering and understanding genetic data are made with each passing year. When you have a hammer, everything looks like a nail. Hence the existence of initiatives like Human Longevity, Inc., a well-funded young company mixing genetic data with personalized medicine and promising enhanced life spans on the horizon. I don't think they can deliver on that promise. I don't see that chasing human longevity-associated genetic variants is a path to anything other than producing larger databases with better maps of cellular biochemistry. That is an admirable course of action from a pure science perspective, but it won't lead to therapies capable of meaningfully moving the needle on human longevity. Useful treatments for aging, capable of adding decades or more of healthy life, are not going to emerge from understanding the networks of possibly hundreds of genes that make an individual slightly more rather than slightly less likely to live to 100. They are only going to arise from directly addressing the known forms of cell and tissue damage that cause aging.
As this open access paper points out, the genetic networks influencing survival at late ages are large and very complex. Studies attempting to map these networks tend to produce data that cannot be replicated in different populations, indicating very large numbers of relevant genes, each with a tiny individual effect, and all very dependent on one another and on environmental circumstances.
While the average life expectancy in the US is approaching 80 years, the mean life span of centenarians (and super-centenarians) is about 112. As a group, they represent a distinct region of the demographic distribution among the contemporary human populations. In other words, assuming that the average human generation time is about 25, centenarians are endowed with an extra human generation time. Besides, they are known to have a better health profile relative to the people with normal life span. These distinguishing features of centenarians have prompted an interest among demographers, health scientists and the general public alike, to explore the possibilities of extending the life span of cosmopolitan population to approach that of centenarians. Therefore, we consider their distinct features from an evolutionary, genetic, developmental and environmental perspective, as these factors have been suggested to influence quantitative traits universally. First, centenarians occur at a frequency of about 1.73 and 3.43 per 10000 individuals in the U.S. and Japan respectively; hence they are rare. Second, among genetic factors, certain genotypes/alleles that are known to influence longevity are enriched among centenarians (e.g., Apo C3-CC; FOXO3a-T; CETP-VV; AdipoQ-del/del; TSHr-G and IGFr). Third, others have suggested that longevity may be a function of genomic integrity. Although evidence on this important idea is relatively sparse on centenarians, research has reported low levels of chromosomal aberrations (an index of superior genomic integrity), relative to cosmopolitan populations. It is suggested that the relatively low level of chromosomal aberrations in the "oldest old" people may be both a consequence of their genomic stability and a contributing factor to their attainment of advanced age.
At the genomic level, anywhere between 300-700 genes (or perhaps more) may be influencing longevity. Although this appears to be a large number, in a recent study on human height, which is arguably a less complex trait than longevity, it was reported that 697 variants among 423 genomic regions may be influencing the trait, and speculated that perhaps thousands may be involved. A similar argument could be advanced for longevity, because longevity as a life history and as an indeterminate trait, is influenced by traits that contribute to both viability and reproductive fitness from zygotic stage through adult stages, till death. Note that a number of genes that influence human height also influence longevity (e.g., IGF1 and mTOR) and other life-history traits, such as body weight and sexual maturity, due to pleiotropy. Life history traits often display genetic correlation due to the underlying pleiotropic effects of genes. Further, life history traits maintain allometric relationships, and consequently show trade-offs in their functional aspects. Accordingly, genes that influence longevity could exert both differential and contextual influence on specific traits as well as correlated antecedent traits during the aging process, as shown by divergent patterns of methylation among age groups. These recent discoveries on the developmental regulation of aging, among contrasting age groups, using comparative gene expression, largely compliment previous reports on genotype-phenotype relationships.
The biological basis of exceptional health and longevity among centenarians has remained unclear. The general features of exceptional longevity, however, appears to run in families, and as a group they have a natural tendency to maintain good health much of their lives. Although centenarians are found to occur at higher frequencies in certain geographical locations, their life-style may not be significantly different from individual members of cosmopolitan populations who chose to lead a healthy life-style. It is likely that centenarians differ from each other just as individuals with normal life span do. Yet, individuals with exceptional longevity may interact with environmental and lifestyle factors differently than others. This unique feature may be interpreted as a form of genotype × environment interaction. As a parsimonious explanation, from a genomic perspective, exceptional longevity of centenarians may be attributed to their superior genomic integrity, specific polymorphisms among genes such as ApoC3-CC, FOXO3aT, and CETPVV, and associated molecular genetic and physiological homeostatic mechanisms. It is likely that centenarians arise and are maintained by negative frequency-dependent selection, as this mode of selection has been shown to have slightly superior physiological mechanisms relative to more common genotypes in general populations. There may be other mechanisms, however, and needs further investigation.