The inflammatory ageing clock (iAge) is based on the idea that as a person ages, their body experiences chronic, systemic inflammation because their cells become damaged and emit inflammation-causing molecules;
To develop iAge, a team including systems biologist David Furman and vascular specialist Nazish Sayed at Stanford University in California analysed blood samples from 1,001 people aged 8–96 who are part of the 1000 Immunomes Project, which aims to investigate how signatures of chronic, systemic inflammation change as people age. The researchers used the participants’ chronological ages and health information, combined with a machine-learning algorithm, to identify the protein markers in blood that most clearly signal systemic inflammation. In particular, they pinpointed the immune-signalling protein, or cytokine, CXCL9 as a top contributor; it is mainly produced by the inner lining of blood vessels and has been associated with the development of heart disease.
[…]
After developing it, the researchers tested iAge by collecting the blood of 19 people who had lived to at least 99 years old, and using the tool to calculate their biological age. On average, the centenarians had an iAge 40 years lower than their actual age, according to a press release — aligning with the idea that people with healthier immune systems tend to live longer.
[…]
When examining CXCL9 as a biomarker of systemic inflammation, Furman and his colleagues grew human endothelial cells, which make up the walls of blood vessels, in a dish and artificially aged them by letting them divide repeatedly. The researchers saw that high levels of the protein drove the cells into a dysfunctional state. When the team silenced expression of the gene that encodes CXCL9, the cells regained some function, suggesting that the protein’s harmful effects might be reversible.
19 is a pretty small sample. I guess it would have to be if you want the very old. But I’d be interested in what the margin of error is calculated to be.