Mortality and biomarkers of aging in heterogeneous stock (HS) mice

A longitudinal study was undertaken to evaluate the relationships among a battery of aging biomarkers and subsequent survival time in 319 genetically heterogenous stock (HS) mice. The biomarker variables chosen were selected from the broad domains of behavior, homeostatic physiology, oxidative defense, and immune function; biomarkers were measured at 45, 90, 360, 630, and 900 days of age. Sex differences were found in the survivor and mortality functions, with a mortality rate crossover occurring at about 525 days and a survival curve crossover at about 750 days of age. Females experienced lower initial mortality but had more sharply increasing mortality with age than did males. Survival analysis using Gompertz parametric models with biomarkers as time-varying covariates yielded significant biomarkers from each domain. Following backward elimination procedures, the final set of independent mortality predictors included headpokes in the File activity apparatus, maximum card drop time, weight, hematocrit, urine concentration, natural killer cell activity, and concanavalin A response.