Shared mechanisms for cognitive impairment and physical frailty: A model for complex systems

Area of Science:

• Gerontology
• Biomarkers
• Machine Learning in Aging Research

Background:

• Cognitive impairment and physical frailty, traditionally studied separately, share common underlying etiologies.
• Emerging research supports the construct of cognitive frailty as a complex aging condition.
• A theoretical framework is proposed to advance the study of interconnected aging conditions.

Purpose of the Study:

• To develop a population predictive model for understanding the biological mechanisms linking physical frailty and cognitive impairment.
• To provide empirical support for the emerging construct of cognitive frailty.
• To investigate shared etiologies between physical and cognitive decline in older adults.

Main Methods:

• Utilized data from the longitudinal "Aging in Chianti" (InCHIANTI) study (n=1453).
• Employed a machine learning (ML) technique (tree boosting) for supervised learning analysis.
• Tested 132 candidate single nucleotide polymorphisms (SNPs) and 155 protein biomarkers.

Main Results:

• Developed two highly accurate predictive models (AUCs of 0.88 and 0.86).
• Identified dysregulation in cellular processes, including genetic alterations, nutrient/lipid metabolism, and inflammation, as drivers of cognitive frailty.
• Results indicate a multifactorial biological basis for cognitive frailty.

Conclusions:

• Established a foundation for understanding the biological mechanisms underlying cognitive decline and physical frailty.
• Highlighted the need for further examination of molecular pathways associated with predictive biomarkers.
• Framework aligns with proposed biological underpinnings of Alzheimer's disease, including genetic factors, immune dysfunction, and neuroinflammation.