The work stands apart: nearby in the archive are studies on predicting kidney failure, cellular aging, and risk calculation, but there are no direct neighbors on urinary aging clocks. This is the first attempt to explicitly separate the aging signal from the noise of kidney filtration in the urinary peptidome.
Urine was collected from 1811 people with healthy kidneys, and a model was trained to predict age from 12 019 peptides. Then, in a cohort of 7798 people with varying kidney function, a calibration surface was built that adjusts the prediction depending on filtration rate and albuminuria. The adjustment is applied only when kidney function is impaired, leaving healthy individuals unchanged.
The k-UPBioAge clock explained 87.6% of the variation in age with an error of 4.91 years. An acceleration of aging by one standard deviation increased the risk of death by 48% after accounting for all factors. Acceleration also predicted heart attack, heart failure, and progression of kidney disease. In people with diabetes and kidney disease, age acceleration reached 13.1 years compared to healthy individuals.
Unlike neighboring works that either ignore the kidney or exclude kidney peptides, here the kidney was turned into an axis of calibration. This changes the interpretation: urinary aging clocks can now be read as a systemic signal, not as a reflection of the filter. This approach can be transferred to other biofluids produced by organs.
It is possible to test whether the clocks respond to interventions that slow aging. There is already evidence that diabetes and blood pressure medications change the urinary peptide profile. If these changes shift the clock backward, a cheap way to evaluate the effectiveness of anti-aging therapy will emerge.
On the epidemiology of aging, nephrology, and preventive medicine. Urinary clocks could become a tool for risk stratification in population screenings.
To verify the reproducibility of the clocks on other platforms and in ethnically diverse populations. To assess whether age acceleration changes with treatment.
Connection to an unsolved problem in gerontology: how to separate biological age from comorbidities. This work shows that accounting for the organ that produces the biofluid is essential, otherwise the aging signal is lost in the noise of disease.
The strongest single peptide in the clock comes from the protein PECAM-1, which sits on the surface of vascular cells. Its contribution to the age estimate is 0.634 years per one standard deviation.
The idea of a 'urinary clock' echoes Philip K. Dick's novel 'Do Androids Dream of Electric Sheep?', where age and health were determined by biological markers. Here urine becomes a similar window into biological age, only without the fantastic assumptions.
Not published on arXiv. Here are its page, the PDF and the formalization: you can check it yourself, not only read our account.