Back

Evidence for an energetic trade-off model linking inflammaging and immunosenescence in the US Health and Retirement Study and UK Biobank

Aronoff, J. E.; Franck, M.; Cohen, A. A.; Trumble, B. C.

2025-10-17 evolutionary biology
10.1101/2025.10.17.682903 bioRxiv
Show abstract

Later life is characterized by the development of chronic inflammation, termed inflammaging, alongside changes in immune cell profiles, or immunosenescence. While these features contribute to health risk, they have also been interpreted as adaptive remodeling of the immune system in response to accumulating somatic damage. Here we consider a recently developed theoretical framework to understand these processes as interrelated: the Brain-Body Energy Conservation model of aging. This model views functional declines, such as immunosenescence, as part of an energy conserving response to the rising energy expenditure of inflammaging. This response promotes short term survival against somatic damage at the expense of future health risk. For example, naive T cells, which enhance defense against future infections, decline with age. We find evidence consistent with this model in the US Health and Retirement Study (HRS) and UK Biobank (UKB). TNFR1, a key marker of inflammaging, mediated 10% and 5% of the age-related declines in naive CD4T and CD8T cells respectively in the HRS (n = 8,261). Consistent with an impaired immune response to future infections, TNFR1 also mediated 16% of the age-related increased risk of hospitalization or death from COVID-19 in the UKB (n = 522 hospitalized or died, full sample n = 40,638). GDF15, which is produced in response to metabolic stress and has been found to induce immune tolerance in response to chronic inflammation, mediated 28% of the TNFR1-related COVID-19 health risk, as well as 38% of the age-related increased risk independent of TNFR1.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.