The Thymus Is Still on Duty: What a New AI Study Suggests About Adult Health and Longevity

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

The thymus has spent decades being treated like the immune system's childhood training center: essential early, then mostly retired. A new study in Nature challenges that familiar story. Using artificial intelligence to analyze routine chest CT scans from 27,612 adults, researchers found that people whose thymus looked healthier tended to live longer and had lower risks of lung cancer and cardiovascular death [1].

That is an arresting result, but it is not proof that a healthier-looking thymus causes a longer life. The study was observational. It found durable associations after accounting for many competing risks, yet it could not determine whether the thymus protects health, whether healthier people simply preserve more thymic tissue, or whether both reflect deeper biological advantages. The evidence is strong enough to change how researchers think about the adult thymus, but preliminary for personal medical decisions.

What the thymus actually does

The thymus is a small organ behind the breastbone, above the heart. Its best-known job is preparing T cells, immune cells that recognize infected or abnormal cells while learning not to attack the body's healthy tissues. Young T cells enter the thymus as unfinished recruits. There they are tested, selected, and released with a broad range of receptors that can respond to unfamiliar threats.

Beginning relatively early in life, the thymus undergoes involution: working tissue gradually shrinks and is increasingly replaced by fat. The result is lower production of new naive T cells and a narrower immune repertoire. This age-related process is established biology, although its rate and functional consequences vary widely among individuals [2]. The thymus does not simply vanish, and residual adult tissue can remain active.

Earlier CT research showed that adult thymic appearance differs with age, sex, body mass index, and smoking history. In one study of 2,540 generally healthy adults, 74% showed complete fatty replacement by conventional visual scoring, while women tended to retain more soft tissue and people with higher BMI or greater smoking exposure tended to have more fatty change [3]. The new study asked whether a more sensitive computer model could detect meaningful differences that human visual categories miss.

How an AI model estimated thymic health

The researchers trained a deep-learning system on 5,674 CT scans to locate the thymic bed and measure radiographic features associated with preserved tissue. They then applied it to two independent, prospectively assembled cohorts: 25,031 participants in the National Lung Screening Trial, or NLST, and 2,581 participants in the Framingham Heart Study, or FHS [1].

The resulting thymic-health score was a percentile ranking within each study population, not a universal clinical measurement. Participants were grouped as low, average, or high: the bottom quarter, middle half, and top quarter. Importantly, the score was an imaging-based proxy for thymic function. The researchers did not directly measure each person's output of new T cells, and a score from one scanner or population cannot yet be interpreted using a universal cutoff.

The survival and disease signals

In the NLST, which consisted of older current or former heavy smokers, the 12-year estimated mortality was 13.4% in the high-thymic-health group and 25.5% in the low group. The unadjusted hazard ratio was 0.49, meaning the rate of death over follow-up was roughly half as high in the high group. The relationship persisted in models accounting for age, sex, smoking exposure, and numerous medical conditions [1].

Lung-cancer patterns moved in the same direction. At six years, estimated incidence was 3.4% in the high group versus 5.3% in the low group. Estimated lung-cancer mortality was 1.1% versus 2.0%. Cardiovascular mortality at 12 years was 2.9% in the high group and 7.5% in the low group. These are group-level estimates, not predictions for an individual patient.

The independent Framingham cohort strengthened the central finding. Higher thymic health was again associated with lower cardiovascular mortality after adjustment for age, sex, and smoking. The association with death from all causes pointed in the same direction but was no longer statistically significant after full adjustment in this smaller cohort. That partial replication is encouraging, but it also shows why the most dramatic numbers should not be treated as settled clinical facts.

Why the adult thymus may matter

The biology is plausible. A thymus that continues producing diverse T cells could help the immune system respond to new infections and notice emerging cancer cells. It might also support better immune regulation. A 2023 study offers indirect human evidence: adults whose thymus had been surgically removed experienced higher five-year mortality and cancer risk than matched patients who underwent similar chest surgery without thymectomy [4]. Those patients also showed lower production of new T cells and higher inflammatory cytokines years later. Surgery is not the same as natural involution, but the findings argue against calling the adult thymus useless.

The new Nature study also connected lower thymic health with higher fasting glucose, triglycerides, blood pressure, obesity, smoking exposure, frailty, and reduced physical activity. In a smaller proteomic subgroup, 16 of 68 measured inflammatory proteins were negatively associated with thymic health. Among 1,156 Framingham participants with repeated C-reactive protein tests, persistently elevated inflammation was linked to a lower score [1].

This could represent a two-way relationship. Chronic inflammation and metabolic strain might accelerate thymic decline; reduced thymic function might then weaken immune regulation and amplify inflammation. Researchers sometimes describe the combination of declining immune effectiveness and persistent low-grade inflammation as immunosenescence and inflammaging [5]. The current study fits that framework, but it cannot reveal which process comes first.

What the study cannot establish

Several cautions matter. Most participants were white, so the model needs validation in more diverse populations. The NLST was composed of people aged 55 to 74 with substantial smoking histories, which limits how directly its results apply to younger adults and never-smokers. The Framingham cohort was broader but much smaller. The researchers also used population-specific score thresholds because CT scanners and imaging protocols differed.

The inflammatory protein measurements involved only 317 people and were obtained, on average, 10.4 years before their CT scans. That time gap makes the biological link suggestive rather than definitive. Some analyses involved numerous comparisons without adjustment for multiple testing, increasing the chance that a few findings could occur by chance. Most fundamentally, no intervention changed thymic health and then demonstrated improved survival.

A healthier score could therefore be a cause, a consequence, or a companion marker of healthy aging. People who avoid smoking, remain physically active, and maintain better metabolic health may preserve the thymus while independently lowering their disease risk. Statistical adjustment helps, but it cannot eliminate every difference between those groups.

What this means for you

There is currently no standard thymic-health screening test, no validated consumer cutoff, and no approved therapy proven to rejuvenate the thymus and extend life. A chest CT should not be ordered solely to obtain this experimental score; CT uses ionizing radiation and should have a sound medical indication. Supplements marketed as thymus boosters are not supported by this study.

The practical message is familiar but useful. Avoiding tobacco, staying active, managing blood pressure and blood sugar, and maintaining a healthy body composition were associated with better thymic health and already have established benefits independent of the thymus. These behaviors are reasonable ways to protect overall health, not guaranteed methods for rebuilding thymic tissue.

The larger scientific shift may be more important. An organ dismissed as a childhood relic appears to retain information about immune aging, cancer risk, cardiovascular health, and longevity. If future studies validate the AI score across populations and connect it to direct measures of T-cell production, routine scans obtained for other reasons might reveal a useful immune-health marker. Intervention trials would then need to show that preserving or restoring thymic function actually changes outcomes. Until that happens, the adult thymus deserves renewed attention, but not exaggerated promises.

References

[1] Bernatz S, Prudente V, Pai S, et al. Thymic health consequences in adults. Nature. 2026;652:986-994. doi:10.1038/s41586-026-10242-y.

[2] Gui J, Mustachio LM, Su DM, Craig RW. Age-related thymic involution: mechanisms and functional impact. Aging Cell. 2022;21(9):e13671. doi:10.1111/acel.13671.

[3] Araki T, Nishino M, Gao W, et al. Normal thymus in adults: appearance on CT and associations with age, sex, BMI and smoking. Eur Radiol. 2016;26(1):15-24. doi:10.1007/s00330-015-3820-y.

[4] Kooshesh KA, Foy BH, Sykes DB, Gustafsson K, Scadden DT. Health consequences of thymus removal in adults. N Engl J Med. 2023;389(5):406-417. doi:10.1056/NEJMoa2302892.

[5] Thomas R, Wang W, Su DM. Contributions of age-related thymic involution to immunosenescence and inflammaging. Immun Ageing. 2020;17:2. doi:10.1186/s12979-020-0173-8.