Principal component analysis of immunosenescent markers in older adults: The roles of age, sex, and cytomegalovirus

2017 
Phenotypic markers of immunosenescence have been associated with poor health outcomes in older adults. However, multiple phenotypic markers on natural killer (NK) and T cells may affect each other and be inter-related. We used principal component analysis (PCA) as a dimension-reduction technique to characterize the sets of markers that contribute independently to variability in immunosenescence among older adults ( N  = 140, 43% male, aged 60–92). Three principal components were identified (PC1-PC3). PC1 (49% variance explained) was characterized by indicators of NK cell exhaustion (NK56dim: NKG2C+, FCR γ -, NKG2C + CD57+, NKG2C + FCR γ -, and FCR γ -CD57+). PC2 (23% variance explained) was characterized by indicators of T cell senescence (CD8+: CD28-, CD57+, and CD56+). PC3 (10% variance explained) was characterized by NK56dimCD57+. Using PCA results, we examined relationships between the principal components and relevant descriptive and physical health variables that may affect immunological aging. PC1-PC3 were not significantly associated with chronological age, and PC1 and PC2 did not significantly differ by sex. However, males were significantly higher than females on PC3 (F(1,106) = 7.19, p  = 0.009). In addition, PC1 and PC2 were significantly associated with cytomegalovirus titers (PC1: b(SE) = 2.77(1.27), p  = 0.002; PC2: b(SE) = 5.98(2.13), p  = 0.006). Immunological aging may be driven less by chronological age (i.e., years lived) and more by overall health status and androgens. Furthermore, latent viral infections, particularly cytomegalovirus, may drive cellular aging in both NK and T cell dimensions in older adults.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []