Infection, Immunity and Aging: What the Evidence Shows

The immune system is one of the clearest places where aging shows up in a blood test, and one of the most over-sold. This article sorts what is known about infection, immunity and aging into four labels: Established, Emerging, Tested in trials (including results that went the wrong way) and Hypothesis. Where I offer my own framework, it is labelled as mine.

At a Glance

How the evidence is labelled

• Established: consistent human data from large cohorts, outcome trials or consensus reviews.
• Emerging: real signals from observational studies, natural experiments or small trials; causation not settled.
• Tested in trials: randomised trials with hard outcomes, positive or negative.
• Hypothesis: reasoning that goes beyond the data, including mine.

Mouse and cell data are called mouse and cell data.

The immune system ages too

Established. With age the thymus, where T cells mature, shrinks and produces fewer new naive T cells. The immune repertoire narrows, memory cells take up more of the space, and responses to new antigens and vaccines weaken (Nikolich-Žugich 2018). Researchers have mapped T-cell aging onto the general hallmarks of aging: mitochondrial dysfunction, DNA damage, epigenetic change and senescence all show up in aged T cells (Mittelbrunn 2021).

Emerging. A modelling study found that the age-related fall in thymic T-cell output tracks the rise in incidence of many cancers and infections with age, sometimes better than models based on accumulated mutations alone (Palmer 2018). It is a model fitted to population data, not proof of cause, but it makes a good case for taking immune aging seriously. The peptide side of this question is covered in thymic involution and immune aging.

This is why immunosenescence sits close to two of the 12 hallmarks of aging: altered intercellular communication and chronic inflammation (López-Otín 2023).

Inflammaging: the slow burn

Established. The term inflamm-aging was coined in 2000 to describe the chronic, low-grade, largely sterile inflammation that accompanies aging and its link to immunosenescence (Franceschi 2000). It has since been tied to cardiovascular, metabolic and neurodegenerative disease (Franceschi 2018; Furman 2019). In older adults, higher IL-6 and CRP levels predicted higher mortality (Harris 1999).

Two caveats matter. First, immunosenescence and inflammaging are intertwined and partly adaptive; some of the change may help older people cope with lifelong antigen exposure (Fulop 2017). Second, the drivers are many: senescent cells, visceral fat, gut barrier changes, inactivity, poor sleep and persistent infections are all candidates (Furman 2019). For markers and practical levers, see the inflammaging guide.

Persistent infections: what the evidence shows

Some viruses, mostly from the herpes family, stay in the body for life after the first infection. The question is whether that lifelong presence wears down the immune system and speeds up aging.

Cytomegalovirus (CMV)

Emerging. In a study of twins, most variation in the human immune system came from non-heritable influences, and CMV infection was one of the largest single factors shaping it (Brodin 2015). In an older UK population cohort, CMV-positive people had higher mortality (Savva 2013). A 2024 review of observational studies found that most report a link between CMV and a lower CD4:CD8 ratio or expanded CD8 T-cell populations, while links to telomere length and inflammaging were weaker (Gadoth 2024).

What this does not show: that CMV causes aging, or that anything done about CMV in healthy people changes outcomes. CMV status is a research marker today.

Epstein-Barr virus (EBV) and multiple sclerosis

Strong for multiple sclerosis; emerging as a model for infection-driven chronic disease. In more than 10 million young adults serving in the US military, EBV infection preceded multiple sclerosis, and the risk of MS rose about 32-fold after infection (Bjornevik 2022). Antibodies from people with MS that bind both the EBV protein EBNA1 and the brain protein GlialCAM suggest molecular mimicry as one mechanism (Lanz 2022).

This is the clearest example of a common, lifelong virus setting up a chronic disease decades later. It is evidence about one disease. It does not show that EBV speeds up aging in general. Reactivation of latent viruses, EBV among them, is also one of the mechanisms studied in long COVID (Davis 2023); more on that in reactivated viruses.

Shingles vaccination and dementia

Emerging, contested. Varicella-zoster virus, which causes chickenpox and later shingles, also stays latent for life. Several studies now link shingles vaccination with less dementia:

• In Wales, eligibility for the shingles vaccine depended on exact date of birth. Comparing people just either side of the cut-off, vaccination reduced the probability of a new dementia diagnosis over seven years by 3.5 percentage points, about 20% in relative terms (Eyting 2025).
• Natural experiments in Ontario, Canada, found a similar direction of effect (Pomirchy 2026).
• In US health records, the recombinant vaccine was associated with lower dementia risk than the older live vaccine (Taquet 2024).

The critique: a 2026 analysis worked out the vaccine effect each natural experiment implies once you account for how many eligible people actually got vaccinated. Two estimates exceeded 100% effectiveness, which is impossible, and national dementia incidence did not fall after vaccination programmes began. The authors conclude that these studies “cannot be considered strong evidence of a clinically meaningful effect” (Kivimäki 2026).

My reading: the signal is interesting and worth a proper trial. Shingles vaccination is worthwhile for preventing shingles and its nerve pain. Dementia prevention is a possible bonus, not a reason to promise anything.

Herpes viruses and Alzheimer’s: the VALAD lesson

Tested in trials, negative. If latent herpes simplex virus contributed to Alzheimer’s disease, suppressing it should help. The VALAD trial tested exactly that: 120 people with early symptomatic Alzheimer’s disease and HSV antibodies received high-dose valacyclovir or placebo for 78 weeks. Cognition worsened more in the valacyclovir group, and the authors concluded it should not be used for this purpose (Devanand 2026).

The lesson reaches beyond Alzheimer’s. “Find the hidden virus and suppress it” sounds logical, and in the one rigorous test of the idea it made things worse. Antivirals belong to active, diagnosed infections.

Clonal hematopoiesis (CHIP)

Emerging as an aging mechanism; the risk association is well replicated. As we age, some blood stem cells pick up mutations that let them outgrow their neighbours. In a large genetic study, these clones were found in 9.5% of people aged 70 to 79 and 18.4% of those aged 90 and over. Carriers had about eleven times the risk of blood cancer, higher all-cause mortality and about double the risk of coronary heart disease (Jaiswal 2014). Follow-up work linked CHIP to atherosclerosis, with mouse data pointing to inflammatory signalling from the mutant cells (Jaiswal 2017; Polizio 2023).

CHIP links three things on this page: aging stem cells, inflammation and vascular disease. There is no proven treatment for CHIP itself in people without blood cancer; testing is mostly done in research or haematology settings.

Can inflammation be treated? Two outcome trials

Tested in trials. Two randomised trials showed that reducing inflammation, without lowering cholesterol, reduces cardiovascular events:

• CANTOS: canakinumab, an antibody against IL-1β, lowered recurrent cardiovascular events in people after a heart attack with raised hsCRP. It also caused more fatal infections than placebo (Ridker 2017).
• LoDoCo2: low-dose colchicine reduced cardiovascular events in chronic coronary disease. Deaths from non-cardiovascular causes were higher in the colchicine group (Nidorf 2020).

These trials establish inflammation as a modifiable cause of heart disease in these patients. They did not test aging, and they show the price of suppressing immune signalling: infections and other harms. Inflammation is part of defence, not only a problem.

Can immune aging be improved?

Emerging. In older adults, a short course of the mTOR inhibitor everolimus improved the antibody response to flu vaccination by about 20% (Mannick 2014), and a later phase 2a trial of low-dose TORC1 inhibition reported fewer infections (Mannick 2018). These are small trials with immune endpoints, not lifespan or healthspan outcomes.

Hypothesis-generating only. The TRIIM study gave nine men growth hormone, DHEA and metformin for a year and reported signs of thymic regeneration and a lower epigenetic age (Fahy 2019). With no control group and nine participants, it is a reason for a proper trial, not for treatment. For peptide angles, see thymosin alpha-1; for the senescent-cell side, see FOXO4-DRI, which has mouse data and no human trials.

My framework: infection- and immune-mediated aging

Hypothesis (Julian’s framework). My working model is that a lifetime of persistent and repeated infections adds to immune aging and inflammaging, and through them may speed up several hallmarks: senescence, vascular aging, possibly clonal hematopoiesis. I treat it as one contributor among many, next to genes, metabolism, environment and chance. It is not a claim that infection is the cause of aging.

Why I find it useful: it explains why vaccination, infection history and immune status belong in a longevity assessment, and it fits the clinical reality of a hospital that works with chronic infection and immune disorders. Why it is still a hypothesis: no trial has shown that preventing or controlling an infection slows aging as a whole. The VALAD result is a warning that plausible infection theories can fail when tested.

What would change my mind, in either direction: randomised data on vaccination and dementia or frailty, and long-term outcome data in people whose immune markers improve on an intervention.

What this means in practice

• Vaccination is the most defensible infection-related longevity measure. Follow your national schedule for older adults, which in Germany and many other countries includes shingles vaccination.
• Useful tests are those that change decisions: blood pressure, lipids, glucose, a blood count, and hsCRP read against its cardiovascular risk strata (below 1, 1 to 3, above 3 mg/L), repeated if high because any infection raises it (Pearson 2003).
• Not useful for healthy aging: broad viral antibody panels to look for “hidden infections”, or antivirals for latent viruses without an active, diagnosed infection.
• The basics: regular physical activity is linked to lower mortality (Ekelund 2019), and short or disturbed sleep to higher inflammatory markers (Irwin 2016).

None of this replaces an individual assessment. If you have a chronic infection or an immune disorder, the plan belongs with a physician who knows your history.

What we don’t know

• Whether any intervention aimed at immune aging improves hard outcomes in healthy older people.
• Whether the link between shingles vaccination and dementia survives a randomised test.
• Whether CMV or CHIP status should change clinical decisions outside research.
• How to measure immune age reliably in routine care.