Your Body Does Not Age as One Number: What the Systems Age Blood Test Reveals

Edited and Approved by Stephen C. Rose, PhD, MS

A car dashboard does not compress fuel level, oil pressure, engine temperature, and tire pressure into one mysterious number. Yet most biological-age tests do something similar: they take signals from across the body and report one overall estimate. That can be useful, but it hides an obvious fact. Your heart, lungs, kidneys, brain, immune system, and metabolism do not necessarily change at the same rate.

A 2025 study in Nature Aging introduces a more detailed approach called Systems Age. From a single blood sample, the researchers created DNA-methylation scores for 11 physiological systems: heart, lung, kidney, liver, brain, immune, inflammatory, blood, musculoskeletal, hormone, and metabolic. They also produced one combined score for whole-body aging [1]. The work offers something closer to a dashboard—but it remains a research instrument, not a medical diagnosis or a proven guide to anti-aging treatment.

What DNA methylation can tell us

DNA methylation is a chemical marking system that helps cells regulate how genes are used. It does not rewrite the DNA sequence. Instead, small methyl groups attach at particular locations, often described as CpG sites, and their patterns shift with development, aging, cell identity, health, and environmental exposure. A laboratory can measure hundreds of thousands of these sites in blood and use an algorithm to recognize patterns associated with age or risk.

The first influential epigenetic clocks were trained mainly to estimate chronological age. Steve Horvath's 2013 clock showed that methylation patterns could estimate age across many human tissues and cell types [2]. Later clocks asked more clinically ambitious questions. PhenoAge was trained through a blood-based measure of health and mortality risk [3], while GrimAge incorporated methylation surrogates for smoking exposure and selected blood proteins to predict lifespan and disease [4].

Another tool, DunedinPACE, was designed to estimate how quickly someone is aging rather than how many biological years they appear to have accumulated. Its training target came from changes in 19 measures of organ-system integrity followed across two decades [5]. These clocks are related, but they are not interchangeable. Each answers the question built into its training data.

From a single clock face to 11 gauges

Systems Age starts from the idea that a person may have relatively preserved kidney function but accelerated metabolic aging, or strong physical function alongside a more vulnerable cardiovascular pattern. A global score could average those differences together. The new framework tries to preserve them. Importantly, it does not take cells from each organ. All 11 scores are inferred from methylation measured in circulating blood cells.

That makes the test practical for research, but the organ labels need to be interpreted carefully. A Brain score is not a direct reading of brain tissue, and a Heart score is not an echocardiogram. Each is a blood-methylation signature trained to reflect a collection of clinical measurements, functional indicators, diseases, and mortality patterns associated with that physiological system. The label describes what the model was designed to capture, not where the methylation was physically measured.

How the researchers built Systems Age

The team first grouped available clinical chemistry tests, blood measurements, functional assessments, and disease information from the Health and Retirement Study into physiological systems. They used principal component analysis to compress related measurements into underlying patterns. In plain language, this step finds combinations of variables that tend to move together, reducing a crowded spreadsheet to a smaller set of signals.

Next, machine-learning models were trained to reproduce those system patterns from blood DNA methylation. The researchers then used data from the Framingham Heart Study to combine the methylation-derived components according to their relationship with remaining life expectancy. Finally, the 11 system scores were integrated into a composite Systems Age score. Training and testing were separated across cohorts, an important protection against building a model that merely memorizes its original dataset [1].

What the scores predicted

The researchers tested specificity largely in independent Women's Health Initiative samples totaling about 5,600 participants. Several results followed the expected pattern. The Brain score had the strongest association with cognitive function, the Musculoskeletal score with physical function, and the Heart score with future coronary heart disease and heart attack. The Blood score was most strongly associated with future leukemia [1].

The scores were not perfectly isolated. Heart and Lung aging were highly correlated, as were Inflammation and Musculoskeletal aging. Lung and Heart scores both predicted lung cancer, probably capturing overlapping influences such as smoking and systemic vascular injury. Some unexpected associations also appeared; for example, the Musculoskeletal score was strongly associated with diabetes. These overlaps are biologically plausible, because body systems communicate constantly, but they argue against treating each score as a literal measurement of one organ's age.

Different routes through aging

Perhaps the most intriguing finding was that people of the same chronological age—and even the same overall Systems Age—could have very different system profiles. When the researchers clustered participants by these patterns, they identified nine aging subtypes. Two groups had elevated Lung aging and greater future lung-cancer risk, but one was distinguished by lower obesity prevalence while the other showed more future coronary disease. Another cardiovascular-risk group had lower Lung aging but higher Metabolic and Inflammation scores [1].

This fits a broader idea sometimes called an ageotype: individuals may show more change in metabolic, immune, liver, kidney, or other pathways. Earlier deep longitudinal profiling of 106 people also found distinct personal aging patterns, though that study used many molecular and clinical measurements rather than one methylation assay [6]. Systems Age scales the concept to larger cohorts and a more standardized blood test.

Why this could eventually matter

If independently validated, a system-level clock could make aging research more precise. A treatment might improve metabolic and inflammatory aging without meaningfully changing brain or musculoskeletal scores. A single global number could dilute that signal. Researchers could also enrich a clinical trial for people showing the type of biological vulnerability the intervention is intended to address.

Repeated testing might eventually help track whether a system-specific risk pattern is stable or changing. But this use requires excellent measurement reliability. Technical noise has produced differences of up to nine years between replicate measurements for some established clocks. Principal-component versions of those clocks substantially improved repeatability [7]. Systems Age uses a related principal-component strategy, but real-world longitudinal performance still requires careful independent testing.

What the test cannot tell you yet

Systems Age does not prove why a score is high. Smoking, medication, infection, chronic disease, blood-cell composition, social conditions, inherited variation, and laboratory effects can all contribute to methylation patterns. An elevated Kidney score does not diagnose kidney disease, and a favorable Brain score does not rule out cognitive impairment. Conventional medical evaluation remains necessary.

The study is observational. Its associations show that the scores contain information about present function and future disease, but they do not establish that changing a score will change health. The aging subtypes were discovered within one Women's Health Initiative cohort and need replication. Much of the validation involved middle-aged and older adults, with substantial representation from women, so performance in younger people and other populations cannot simply be assumed.

Even established clocks can disagree about an intervention. In the randomized CALERIE trial, two years of calorie restriction produced a small slowing on DunedinPACE but no significant change in PhenoAge or GrimAge [8]. That does not mean one clock was right and the others were wrong; they were trained to measure different constructs. A lower methylation score is therefore not automatically evidence of rejuvenation, improved organ function, or longer life.

How to read a commercial result

If a version of Systems Age is offered commercially, the result should be treated as contextual information rather than a diagnosis or treatment target. Ask whether the laboratory uses the published algorithm, what platform and quality controls are used, how repeatable the score is, which reference population defines "older" or "younger," and whether the exact report has been independently validated. Be cautious about expensive supplements or procedures sold primarily to improve a proprietary age score.

A sensible response to a concerning system score is not to chase the methylation number. It is to review established risk factors and appropriate clinical tests with a qualified professional: blood pressure and lipids for cardiovascular risk, glucose measures for metabolic health, kidney and liver laboratory values, lung function when indicated, vaccination and infection history, physical performance, sleep, smoking, exercise, diet, and medications. Those measures have direct clinical meaning today.

The bottom line

Systems Age is a clever and potentially important advance because it respects the uneven nature of human aging. One blood methylation assay generated 11 system-oriented views plus a composite score, and those views predicted relevant diseases and functions better than a single undifferentiated clock in many analyses. The finding that similar overall biological ages can conceal different vulnerabilities is especially compelling.

Still, the dashboard is experimental. The gauges are statistical inferences from blood, not direct inspections of 11 organs. Before Systems Age can guide individual care, researchers must show repeatability, population fairness, responsiveness to meaningful interventions, and—most importantly—that score changes correspond to better function, less disease, or longer healthy life. For now, it gives science a sharper map of aging heterogeneity, not a personal repair manual.

References

[1] Sehgal R, Markov Y, Qin C, et al. Systems Age: a single blood methylation test to quantify aging heterogeneity across 11 physiological systems. Nat Aging. 2025;5(9):1880-1896. doi:10.1038/s43587-025-00958-3.

[2] Horvath S. DNA methylation age of human tissues and cell types. Genome Biol. 2013;14(10):R115. doi:10.1186/gb-2013-14-10-r115.

[3] Levine ME, Lu AT, Quach A, et al. An epigenetic biomarker of aging for lifespan and healthspan. Aging (Albany NY). 2018;10(4):573-591. doi:10.18632/aging.101414.

[4] Lu AT, Quach A, Wilson JG, et al. DNA methylation GrimAge strongly predicts lifespan and healthspan. Aging (Albany NY). 2019;11(2):303-327. doi:10.18632/aging.101684.

[5] Belsky DW, Caspi A, Corcoran DL, et al. DunedinPACE, a DNA methylation biomarker of the pace of aging. eLife. 2022;11:e73420. doi:10.7554/eLife.73420.

[6] Ahadi S, Zhou W, Schüssler-Fiorenza Rose SM, et al. Personal aging markers and ageotypes revealed by deep longitudinal profiling. Nat Med. 2020;26(1):83-90. doi:10.1038/s41591-019-0719-5.

[7] Higgins-Chen AT, Thrush KL, Wang Y, et al. A computational solution for bolstering reliability of epigenetic clocks: implications for clinical trials and longitudinal tracking. Nat Aging. 2022;2(7):644-661. doi:10.1038/s43587-022-00248-2.

[8] Waziry R, Ryan CP, Corcoran DL, et al. Effect of long-term caloric restriction on DNA methylation measures of biological aging in healthy adults from the CALERIE trial. Nat Aging. 2023;3(3):248-257. doi:10.1038/s43587-022-00357-y.