Biological Age Estimation Based On Multi-domain Biomarkers In Older Adults

Biological age (BA) provides valuable information about age-related risk of adverse health outcomes and the extent to which aging deviates from chronological age (CA). Aging spans multiple domains; yet, the accuracy of BA estimates using multi-domain biomarkers remains uncertain. PURPOSE: This study derives the sex-specific measures of BA in a population of Korean older adults based on six-domain biomarkers that are commonly available in clinical practice: physical, body composition, cognitive function, psychological and sensory abilities, and blood biomarkers. METHODS: We used data of 2,400 older adults aged 70-84 years (mean age: 76.5 ± 3.9 years; women: 52.8%) from the Korean Frailty and Aging Cohort Study at baseline survey. Seventy-two biomarkers from six domains were drawn from blood samples, sensory assessments, physiological tests, physical and cognitive function evaluations, and body composition assessed using dual-energy X-ray absorptiometry. We calculated the sex-specific BA estimate using the Klemera and Doubal (KD) method. Additionally, we assessed its efficacy in predicting incident physical frailty and all-cause mortality over a 4-year period, validating the utility of BA in these outcomes. RESULTS: Using Boruta algorithm selection procedures, we identified a final selection of 19 biomarkers in men and 21 biomarkers in women that are significantly associated with CA. The BA estimates from all six domains were correlated with CA, with the highest correlation observed in sensory abilities for both men (r = 0.90) and women (r = 0.95). In predicting the onset of physical frailty, both physical abilities (area under the curve [AUC] 0.817 for men and AUC 0.776 for women) and psychological abilities (AUC 0.796 for men and AUC 0.877 for women) among the six domains of BA estimates showed significant associations. For mortality, BA estimated using overall biomarkers demonstrated the strongest association (AUC 0.914 for men and AUC 0.860 for women) compared to BA based on the six-domain biomarkers. CONCLUSION: Our results suggest that the effectiveness of sex-specific BA models, based on a comprehensive panel of biomarkers reflecting a multisystem approach, can be optimized for determining associations with health-related outcomes in older adults. This work was supported by a grant from the Research Program funded by the National Institute of Health, Korea Disease Control and Prevention Agency (2021-ER0605-00); and the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIT) (No. 2022R1A2C1012305).