Modeling Age-Related Disease Risk as Accumulation of Senescent Cells
Researchers here find that a simple model of senescent cell accumulation, with thresholds at which disease occurs, can be made to match the observed variations in risk of most age-related diseases. It is interesting to ask just how much of degenerative aging is driven by this accumulation of senescent cells, and the senescence-associated secretory phenotype that causes inflammation and disrupts tissue function. Clearly not all of aging, but the results in animal studies suggest that senescent cells contribute a large enough fraction of the whole to be a compelling target for rejuvenation therapies. Models such as the one produced here help to flesh out the observed data from animal and human studies.
Recent work on senescent cell dynamics with age used these dynamics to explain the distribution of death times in mice and humans. It was shown that senescent cells are produced and removed with a half-life of days in young mice, but their removal rate slows down in old mice to a half-life of weeks. These data, together with longitudinal measurement of senescent cells in mice, were used to develop a stochastic model for senescent-cell production and removal, called the saturated-removal (SR) model. The SR model shows that senescent cells slow their own removal rate, which leads to wide variations between individuals in the number of senescent cells at old ages. Assuming that death occurs when senescent cells exceed a threshold, it was shown that the SR model explains the distribution of times of death.
Since senescent cells are implicated in many age-related diseases, and since a threshold-crossing event of senescent cells in the SR model has an exponentially rising probability with age, we asked whether age-related diseases can be modeled as a threshold-crossing phenomenon in which senescent cells exceed a disease-specific threshold. To explain the drop in incidence at very old ages, we add to this model the epidemiological notion of heterogeneity, in which some people are more susceptible to the disease than others. We show that the SR model with differential susceptibility provides a model with 2 or 3 free parameters that can explain a wide range of age-related incidence curves. This includes the incidence of many types of cancer, major fibrotic diseases, and hundreds of other age-related disease states obtained from a large-scale medical record database.
This conceptual picture explains why different diseases have similar exponential rise in incidence and a drop at very old ages, based on a shared biological process, the accumulation of senescent cells. It also can be used to optimize the frequency of treatments that eliminate senescent cells, showing that even infrequent treatment starting at old age can reduce the incidence of a wide range of diseases.