Fine-Tuning the Biological Aging Clock

A collage with a clock, vascular system, and inflammation.

iAge, an inflammation aging clock, aims to predict your risk of disease and death.

Why do some people live longer, healthier, and more active lives while others their same age struggle with lifelong chronic pain and suffer maladies up to their dying day—which comes much earlier than others? 

This basic longevity question has been nagging physicians for ages. The importance of lifestyle factors such as diet, exercise, stress, and epigenetic processes like lifestyle and exposure to environmental hazards have been called into account to explain this divergence, and now a team of researchers from Stanford’s Cardiovascular Institute Division of Vascular Surgery and the Buck Institute for Research on Aging believes they found the answer. 

Rather than the biological age, they say a better predictor of health and longevity is a person’s inflammation age. Aided by artificial intelligence and machine learning, the researchers have concluded that epigenetic effects of inflammation processes, particularly on the cardiovascular and neurological level, are connected with much of the morbidity and mortality associated with aging.

Led by Nazish Sayed, a professor of vascular surgery, the group includes Yingxiang Huang, a co-author on the paper and David Furman, a pioneer in the field of immunosenescence, director of the Stanford 1000 Immunomes Project, and chief of the Center for AI and Data Science of Aging at the Buck Institute. Their findings were published in the journal Nature Aging.

Too much of a good thing

The idea that inflammation, an innate defensive mechanism shared by humans with mammals and other organisms, can be an agent of premature death is a complex thought to wrap one’s mind around. After all, inflammation is the natural process by which our bodies fight invading pathogens. Nevertheless it also seems to drive aging, though the hows and whys of inflammation’s impacts on the aging process still escape researchers’ understanding.

“Not all humans age biologically at the same rate,” said Furman in a press release. “Some older people are extremely disease-prone, while others are the picture of health.”

One can have too much of a good thing, goes the saying. And the too much refers not only to excesses but also to chronic, low-level lifestyle-linked sources of inflammation. And therein lies the nexus between aging and inflammation. Chronic inflammation has been linked to cancer and neurodegenerative diseases and is strongly associated with increased cellular senescence.

Cascades of proteins called leukocytes, chemokines, interleukins, and antibodies are induced by inflammation bursts caused by bacteriological and viral vectors and environmental stressors, and they often have the beneficial effect of ameliorating disease or ferreting out an invading pathogen. For example, inflammation causes fever to fight bacteria, mobilizes phagocytes, and induces secretion of immunological factors, tweaking the body’s homeostasis and rendering it unwelcoming to diseases. However, over time and with aging, these processes can become harmful.

In order to understand how the connection between aging and inflammation takes place and what markers of inflammation must be targeted for neutralization, the researchers created iAge, a new tool to gauge a person’s degree of chronic inflammation, and to understand whether that person’s “immunological aging” will lead to disease later in life. 

Spinning off the iAge technology, Furman and others founded Edifice Health, a Silicon Valley startup that announced $12 million in Series A funding in April, led by investments from Leaps by Bayer. This summer the company announced an upcoming randomized clinical study that will test whether taking dietary supplements based on one’s iAge profile will reduce chronic inflammation.

“The idea that inflammation can be used to create an aging clock is interesting, but there are already many ‘aging clocks,’ and we now realize that they capture different dimensions of aging,” says Luigi Ferrucci, director of the National Institutes of Health’s National Institute on Aging. “What iAge does is [to go] within the mechanism of inflammation using the extraordinary power of AI and look for better, more precise target biomarkers. More needs to be done, but this is very valuable work of high quality that opens a door in the right direction,” says Ferrucci, who was not involved in the study.

Among the fifty or so markers tested, a protein emerged as a culprit of increased inflammation. Called CXCL9, it may foretell healthy versus unhealthy aging. A protein that regulates the activation of immune cells, CXCL9 is secreted by the aging endothelium—the walls of the body’s blood vessels—and is an optimal biomarker for age-related vascular inflammation and levels of cardiovascular decay.

For instance, studies have shown that CXCL9 is clinically significant in hypertension and plays a key role in the health outcomes for people with left ventricular dysfunction. In addition, the iAge team, validating that CXCL9 is an indicator of cardiovascular pathology regardless of a person’s age, verified that its presence could predict asymptomatic levels of cardiovascular aging in otherwise healthy individuals, suggesting that targeting CXCL9 with drugs or treatments may prevent the deterioration of the vascular tree caused by age.

“Machine learning told us that CXCL9 was higher in people with a higher inflammation age,” says Sayed. “In vitro experiments demonstrated that when we knocked down CXCL9, the aging process was delayed, and the cells showed much less senescence.”

Almost seventy percent of the centenarians scored low on the iAge index.

To attain the metrics to create iAge, the researchers collected blood samples from a cohort of 1,001 individuals aged 8–96, interviewed 902 of them to obtain overall clinical health data, and conducted comprehensive cardiovascular phenotyping of 97 healthy individuals that, which besides measuring factors such as arterial stiffness and degree of aortic and ventricular decay, involved also obtaining a sample of their blood vessels via venipuncture and carefully measuring the genomic and proteomic contents. Then aided by artificial intelligence—to pinpoint recurring patterns—and machine learning, they extracted a predictive model of age-related inflammation events. In the end, this allowed them to develop an inflammation clock of age-related chronic inflammation conditions, which they dubbed iAge, that can predict critical characteristics of cardiovascular aging.

“Eventually, iAge could be a powerful diagnostic marker—besides cholesterol and hemoglobin levels or white cell count—for health professionals to identify healthy older adults who risk developing early cardiovascular aging as well as younger people at risk for life-threatening diseases,” says Sayed. 

Interestingly, applied to the evaluation of an independent cohort of centenarians, iAge confirmed their exceptional longevity. Almost seventy percent of the centenarians scored low on the iAge index. Furthermore, their levels of CXCL9 were lower than that of a control group of individuals aged 59 to 79.  

Sixty-eight percent (13 out of 19) centenarians were in the low rank group (P=0.028), whereas only 31 percent (6 out of 19) were in the high rank group. In contrast, there were 77 percent (14 out of 18) of controls in the high rank versus 23 percent in the low rank group.

“In the end, all we hope is that iAge becomes part of a person’s annual check-up, another indicator that contributes to making aging healthy,” concludes Sayed.

Go Deeper