Circulating Amyloid-β and Methionine-Related Metabolites to Predict the Risk of Mild Cognitive Impairment: A Nested Case-Control Study

Background: The high cost, limited availability, and perceived invasiveness of amyloid PET and cerebrospinal fluid biomarkers limit their use for the diagnosis of Alzheimer's disease. Objective: The present study aimed to assess the associations of mild cognitive impairment (MCI) with circulating amyloid-beta (A beta), methionine circulating metabolites (MCMs), and their downstream products, and to develop a nomogram based on these easily accessible blood indexes for the individualized prediction of MCI risk in older adults. Methods: In this nested case-control study, we recruited 74 MCI patients and, for each, 3 matched controls (n = 222) within the context of the Tianjin Elderly Nutrition and Cognition (TENC) cohort, a population-based prospective study in China. Concentrations of A beta, MCMs, and their circulating downstream factors (i.e., leukocyte telomere length and inflammatory cytokines) were evaluated in fasting blood sample using standard procedures. We constructed a nomogram for MCI harnessed multivariable logistic models incorporating variables selected in the Lasso regression. Results: Among the many biomarkers examined, the final prediction nomogram retained only 3 factors: A beta(42)/A beta(40) ratio, Hcy, and SAM/SAH ratio. The model achieved favorable discrimination, with a C-statistic of 0.75 (95% confidence interval 0.69-0.81) in internal validation after adjustment of optimism. The calibration accuracy was satisfactory; the Brier score of the model was 0.161 in internal validation after adjustment of optimism. Conclusion: This study presents an individualized prediction nomogram incorporating only three blood biomarkers (i.e., A beta(42)/A beta(40) ratio, Hcy, and SAM/SAH ratio), which can be conveniently utilized to facilitate early identification and the development of high-risk prevention strategies for MCI in older adults.

Automated summary

This study presents an individualized prediction nomogram based on easily accessible blood biomarkers for the identification and prediction of MCI risk in older adults.