Genetic study finds uneven links between organ aging and disease

A genetic analysis published October 5 in Clinical Epigenetics finds that relationships between organ-specific aging measures and disease differ across organs and by the direction being tested, complicating simple causal interpretations.

Researchers used European-ancestry genetic summary data covering nine UK Biobank-derived organ-age measures and 19 disease-related traits. Their bidirectional Mendelian randomization analysis prioritized 31 associations among 306 estimable comparisons.

Cardiovascular age measures were linked with coronary disease and large-artery stroke. Genetic susceptibility to type 2 diabetes was associated with metabolic, cardiovascular, brain and liver age measures in reverse analyses.

The authors identified substantial variation between genetic variants, weaknesses in some analyses and uneven statistical power. Only some associations persisted when multiple exposures were considered together.

What an organ-age gap measures

The foundation for interpreting these findings is the distinction between chronological age and an estimated biological age. An organ-age gap is the difference between an age predicted from biological measurements and a person’s actual age. It is a model output, whose meaning depends on the measurements used to construct it.

A related Nature Aging study published in June 2024 characterized these measures across nine organ systems using data from 377,028 UK Biobank participants of European ancestry. Its models incorporated imaging, physical characteristics and physiological measurements. The systems included the brain, eyes, heart and circulation, liver, immune system, metabolism, muscles and skeleton, lungs, and kidneys.

That earlier work identified 393 genetic-region–age-gap pairs. Many genetic associations were specific to the organ being measured, while others connected multiple systems. This provides useful context: an estimated older kidney and an estimated older brain need not represent the same biological process.

It also explains why a single number cannot capture every question researchers might ask about aging. The choice of organ, measurements and model matters when interpreting what an age estimate represents.

Genetic evidence has conditions

Mendelian randomization uses inherited genetic variation to investigate whether a risk factor may contribute to an outcome. Because genetic variants are fixed before disease develops, this approach can reduce some problems that complicate ordinary observational comparisons, including reverse causation and certain forms of confounding.

But it depends on assumptions. A methodological guide in The BMJ explains that the genetic variants must reliably track the exposure being investigated and must not affect the outcome through an unrelated pathway. Variants with effects on several biological processes can make those conditions difficult to satisfy.

Weak genetic predictors can also produce unreliable estimates. In analyses combining separate genetic datasets, overlapping participants and differences between populations require attention. Sensitivity analyses help researchers assess these problems, but a collection of statistical checks does not automatically establish causality.

Another distinction matters for longevity medicine: genetic estimates may reflect differences operating across a lifetime. They cannot simply be translated into the effect of changing a biomarker with a drug or lifestyle intervention later in life. The BMJ guide therefore emphasizes interpreting such findings alongside evidence from other study designs.

The practical question remains clinical

For healthspan research, the useful next question is whether a proposed intervention preserves function or prevents disease. An aging score can help formulate that question, but demonstrating a change in the score and demonstrating a health benefit remain separate tasks.