Loading
Old version — revision 1
This is a fixed snapshot of Immunosenescence, saved by Import as part of the initial corpus import. It is not edited and it is not updated; the article may have changed since.
Edit summary: Initial import of content/immunosenescence.md — the filesystem corpus, unchanged. Not an edit.
The age-related remodelling of the immune system, in which responses to unfamiliar antigens weaken while chronic activation against familiar ones persists.
Immunosenescence is the set of changes that accumulate in the immune system with age: a shrinking capacity to respond to antigens the body has not met before, alongside persistent low-grade activation against antigens it has. It is the immune half of biological aging, and it accounts for a substantial part of the mortality gradient with age: infection kills disproportionately in the old, latent viruses reactivate, and immune control of transformed cells is one of the defences against Cancer that weakens.
The thymus, where T cells are selected and licensed, begins involuting in childhood and is largely replaced by fat by middle age. Nothing else in the body loses so much of an organ so early with so little clinical notice, and the consequence is that the supply of new naive T cells falls steeply while the existing pool must be maintained by division rather than replacement.
The result is a compositional shift. Naive T cells decline, memory and terminally differentiated effector cells expand, and the diversity of the T-cell receptor repertoire contracts. Deep sequencing of receptor repertoires in donors from childhood to old age found diversity tracking the naive T-cell fraction, which fell roughly linearly to about the age of 70.1 A narrower repertoire is a narrower set of things that can be recognised, which is the mechanistic link between an aged immune system and a novel pathogen.
The B-cell compartment shifts similarly, with reduced output of new naive B cells and less effective germinal-centre reactions, so antibody responses to new antigens are lower and less durable. Innate immunity changes rather than simply weakening: neutrophil and macrophage function declines while baseline inflammatory signalling rises, the process described under Inflammaging. The two are usually treated together because the 2023 revision of the hallmarks added chronic inflammation as a hallmark in its own right, and because senescent cells, which accumulate through the mechanism set out in Cellular senescence, secrete much of the signal involved. Falling output from the haematopoietic compartment is the same phenomenon described under Stem cell exhaustion, seen from the immune side.
Most of the world's population carries cytomegalovirus for life, and controlling it consumes a disproportionate share of the T-cell compartment. In some older people, virus-specific clones occupy a large fraction of the entire CD8 pool.
Swedish longitudinal studies of octogenarians and nonagenarians identified a cluster of markers that predicted death within two to six years: an inverted CD4/CD8 ratio, poor T-cell proliferative response, and cytomegalovirus seropositivity. The cluster was named the immune risk phenotype and is the closest thing the field has to a validated prognostic immune signature.2 The same research group later reported that individuals who survived to 100 did not show the profile, which is consistent either with the phenotype being a cause of death or with it being a marker of people already failing.
Decline or remodellingThe word "immunosenescence" implies degeneration, and the field is not unanimous that this is the right description. On the alternative reading the aged immune system is remodelled rather than broken: it is optimised for the antigens it has already met, at the cost of the ones it has not, which is a reasonable allocation for an organism past reproduction. Cytomegalovirus is the sharpest version of the argument. It correlates with mortality in Swedish cohorts and does not in some other populations, which suggests its effect depends on context rather than being a fixed cost.
There is no accepted assay for immune age, and this is the field's practical bottleneck. Two approaches have been taken.
The first builds composite scores from immune-cell phenotyping. A nine-year longitudinal study of healthy adults produced a trajectory-based metric, IMM-AGE, that predicted all-cause mortality in an independent cohort after adjustment for established risk factors, and that tracked immune status better than chronological age.3 Composite scores built from DNA methylation and from circulating inflammatory proteins are parallel efforts in the same direction, and share the interpretive difficulty set out under epigenetic clocks: a score that predicts mortality is not thereby a measure of the process that causes it.
The second uses function rather than composition, and the standard functional readout is vaccine response. A quantitative review of thirty-one studies found substantially lower rates of seroconversion and seroprotection after influenza vaccination in older adults than in younger ones across all three vaccine components.4 Vaccine response has the advantage of measuring what actually matters and the disadvantage of being an indirect and noisy measure of the underlying state. Neither approach has been accepted by a regulator as a surrogate endpoint, which is the general problem set out under Aging biomarkers.
The clearest is infection. Age-specific mortality from respiratory infection rises steeply and roughly log-linearly through later life, a gradient that was measured with unusual precision during the COVID-19 pandemic and that held across countries with very different health systems.5 That gradient is the reason pandemic risk and longevity are not separate subjects.
Reactivation of latent viruses is the second. Herpes zoster is the visible case: the virus is controlled by cell-mediated immunity that declines with age, and its reactivation rate rises accordingly.
The third is tumour surveillance. Immune control of transformed cells is real, as checkpoint blockade demonstrated by removing a brake on it, and a declining capacity to exercise that control is one proposed contributor to the steep rise in cancer incidence with age. The contribution has not been quantified in humans, and the age-incidence curve is adequately explained by mutation accumulation alone, so this remains a plausible mechanism rather than a measured one.
Vaccine formulation is where the field has delivered. High-dose influenza vaccine was more efficacious than standard dose in a randomised trial in adults 65 and over.6 The adjuvanted recombinant zoster vaccine reached efficacy above 90% against shingles in adults over 50, far exceeding the live-attenuated vaccine it replaced.7 Neither reverses immunosenescence. Both work around it by presenting more antigen or better adjuvant to a system that responds poorly to less.
Drug intervention has one suggestive human result. A randomised trial of low-dose mTOR inhibitors in older adults reported improved influenza vaccination response, upregulated antiviral gene expression, and fewer self-reported respiratory infections over the following year.8 The compound was later taken into a larger trial that did not meet its primary endpoint, so the result stands as the most interesting human signal in geroscience and not as a demonstrated therapy. It is the main reason Rapamycin is discussed as an immune-restoring agent rather than only as an immunosuppressant.
Exercise is associated with better vaccine responses and lower infection mortality in observational data, with the causal direction hard to establish in a population where illness reduces activity.
Thymic regeneration is the most direct approach. A small uncontrolled study combining growth hormone with two diabetes drugs reported regrowth of thymic tissue on imaging in nine men, along with a reversal of epigenetic clock readings that attracted more attention than the immune data.9 Nine participants with no control arm cannot establish an effect, and the follow-up study has not reported an outcome that changes that.
Other routes are earlier. Engineering thymic tissue is one of the objectives of Tissue engineering work, and allogeneic cultured thymus tissue is already approved for children born without the organ. Rejuvenating haematopoietic stem cells by transient reprogramming is under investigation at companies including NewLimit, entirely in cells and mice so far. Clearing senescent T cells with Senolytics has a rationale and no human efficacy data. Immunity is one of the three functional domains scored by the XPRIZE Healthspan competition, which is a useful forcing function precisely because the field cannot yet agree what to measure.
Whether immunosenescence is a driver of aging or one of its outputs is unsettled, and the question is not merely academic: it determines whether restoring immune function would extend Healthspan broadly or only reduce infection deaths. The Geroscience hypothesis predicts the former and has no human evidence for it.
The measurement gap is more tractable and more urgent. Without a validated marker, a trial of an immune-restoring intervention has to be powered on clinical infection or vaccine response, which makes it large, slow, and confined to endpoints regulators already accept. The interventions with the best human evidence remain the ones that route around the aged immune system rather than repairing it, and there is no clear reason yet to expect that to change soon.
paperBritanova, O. V. et al. "Age-Related Decrease in TCR Repertoire Diversity Measured with Deep and Normalized Sequence Profiling." The Journal of Immunology, 2014.↩Thirty-nine healthy donors sampled cross-sectionally, so the decline is inferred across people rather than followed within them.
paperWikby, A. et al. "An Immune Risk Phenotype, Cognitive Impairment, and Survival in Very Late Life." Journals of Gerontology: Series A, 2005; and Strindhall, J. et al. "No Immune Risk Profile among individuals who reach 100 years of age." Experimental Gerontology, 2007.↩Both draw on the same small Swedish cohorts, and the phenotype has not replicated consistently in other populations.
paperAlpert, A. et al. "A clinically meaningful metric of immune age derived from high-dimensional longitudinal monitoring." Nature Medicine, 2019.↩Derived from 135 healthy adults and validated against mortality in the Framingham cohort; it has not been used as a trial endpoint.
paperGoodwin, K., Viboud, C. and Simonsen, L. "Antibody response to influenza vaccination in the elderly: a quantitative review." Vaccine, 2006.↩Pools antibody responses across 31 studies; antibody titre is a correlate of protection, not protection itself.
paperO'Driscoll, M. et al. "Age-specific mortality and immunity patterns of SARS-CoV-2." Nature, 2021. ↩
paperDiazGranados, C. A. et al. "Efficacy of High-Dose versus Standard-Dose Influenza Vaccine in Older Adults." New England Journal of Medicine, 2014. ↩
paperLal, H. et al. "Efficacy of an Adjuvanted Herpes Zoster Subunit Vaccine in Older Adults." New England Journal of Medicine, 2015.↩Efficacy was sustained across age strata in this trial, which is unusual; a companion trial extended the finding to adults over 70.
paperMannick, J. B. et al. "TORC1 inhibition enhances immune function and reduces infections in the elderly." Science Translational Medicine, 2018; and Mannick, J. B. et al. "Targeting the biology of ageing with mTOR inhibitors to improve immune function in older adults: phase 2b and phase 3 randomised trials." The Lancet Healthy Longevity, 2021.↩The infection endpoint in the 2018 trial was self-reported over the following year, and the later phase 3 did not meet its primary endpoint.
paperFahy, G. M. et al. "Reversal of epigenetic aging and immunosenescent trends in humans." Aging Cell, 2019.↩An uncontrolled study of nine men, run to test thymus regeneration rather than to test an intervention against a clock.