Doctors, scientists, insurers, biotech companies. and patients themselves have long sought to be able to accurately predict (and extend) an individual’s longevity. And given the right situations, such estimates could have health care benefits.
Over the past several years, scientists have identified four genetic and molecular biomarkers that potentially predict human and animal longevity. The first is the rate at which an individual’s telomeres shorten in length. There is increasing evidence from both human and animal studies that the slower the rate of telomere shortening, the longer that individual is likely to live. The second is the rate of gene methylation, indicating an increased level of methylation was correlated with shortened longevity. The third is the polygenic risk. A recently reported genetic analysis can identify “10 percent of people with the most protective genes, who will live an average of five years longer than the least protected 10 percent,” according to a statement from a scientist who developed the method, in a press release.
The fourth approach was described in a study this year that identified 14 blood-based biomarkers of metabolism that when combined into a predictive score was statistically associated with predicting the end of life. Screening individuals using these metabolite profiles appeared to be predictive of a high risk of mortality within 10 years. In this study, scientists included more 44,000 people from 12 cohorts who were followed between three and 17 years to establish a correlation with these blood metabolites and longevity. There are no comparable studies that have examined such a large population using telomere length or gene methylation as longevity predictors.
Each of the four approaches to predicting longevity raises several scientific and ethical concerns that need to be addressed. First, the metabolite biomarker study examined cohorts of only European descent. To be a valid predictor for human longevity, it will be essential to sample diverse cohorts from other populations, as other studies have done within different European cohorts. Second, the predictability factor using metabolite biomarkers was only increased slightly, from 78 percent using traditional markers such as weight and cholesterol to 83 percent with the biomarkers. The aim must be to achieve a “zero-error” prediction. Third, we need to examine the relative effects of telomeres and gene methylation when combined with the metabolite biomarkers in order to make longevity predictions more robust and to determine which specific factors add the most to the predictive values. Finally, future studies using blood metabolite biomarkers, telomere length, or gene methylation should examine whether diet, genetics, or pharmacological interventions affect predictability.
The blood-based biomarker studies differ from current clinical end-of-life predictors, such as blood pressure and cholesterol levels, because there are established behavioral and drug interventions to reduce blood pressure and cholesterol levels. Were biomarkers to be developed for clinical applications, we propose that they should only be used if they provide actionable results. We should be cautious in applying both premature and unproven longevity results in a clinical situation that has such serious implications.
Several companies are already offering consumers tests to assay their telomere length. We would not be surprised if in the future companies will use other biological or genetic predictors to assess human longevity or offer ways to reverse our biological clocks. We also caution consumers against seeking out such longevity predictions should they be offered direct to the public, unless companies present the results to the consumer by a certified genetic counselor, as the psychological effect from these data could be devastating. A responsible model system for genetic counselling may be the All of Us precision medicine project, led by the National Institutes of Health. The health technology company Color will gather and present the results of genome sequencing from, and to, 1 million volunteers. Here, concern regarding proper informed consent is essential. Importantly, the right to know as well as the right not to know their results should be upheld.
Furthermore, the unintended consequences of using end-of-life predictions based on these preliminary studies can be unsettling. Do we want our life insurance agents canceling any policy or raising rates based on our biomarkers? In fact, YouSurance is the first company to use epigenetic biomarkers to assess life insurance applicants’ health and lifespan.
In conclusion, we have not reached the point when it is ethical and scientifically valid to use biomarkers to predict longevity. Assessing molecular or biochemical outcomes that accurately predict longevity will best be applied when we have defined preventive measures or treatments to accompany the predictors that together are likely to improve longevity.
