Serum elementomic analysis indicates a panel of elements related with age

Abstract

Background: Elementomics, which includes metallic and non-metallic elements, is an emerging and promising research field for human diseases. Researchers are focusing on discovering the relationship between elements and various diseases; however, the changes in element concentrations during the process of aging remain unclear.

Materials and Methods: We performed elementomic analysis in the serum of 70 subjects aged 30 to 96 years using inductively coupled plasma mass spectrometry. The subjects were divided into 7 groups with an age range of 10 years. Random forest was used to estimate the variable importance of elements. Linear regression model and restricted cubic spline analysis were performed to screen for elements individually associated with age. Candidate elements were combined by corresponding multivariate linear regression coefficients to generate a risk score representing their collective effect on age.

Results: Among the 62 detected elements, lithium, boron, calcium, titanium, and selenium were identified as the most important predictors of age. There was an increase in lithium and boron as well as a decrease in calcium and titanium with increasing age. The concentration of selenium was elevated before the age of 60, but decreased thereafter. A formula of element risk score was constructed using the respective coefficients from a multivariate linear regression model for the above five elements. The formula = 4.5522 × lithium + 6.0575 × boron - 4.9990 × calcium - 7.0403 × titanium - 0.8849 × selenium.

Conclusion: Elementomics could be a novel and promising non-invasive biomarker for the assessment of senescence.

Keywords: Elementomics, serum, age, inductively coupled plasma mass spectrometry