Background
While All-cause dementia (ACD) may often be characterized by the abnormal deposition of extracellular β-amyloid (Aβ) in the brain cortex and hyperphosphorylated tau (p-tau) as neurofibrillary tangles intracellularly, there is a need to identify early metabolic changes that may accompany these pathological changes.
Objective
This study evaluated the predictive value of fatty acids in cerebrospinal fluid (CSF) fractions in differentiating cognitively unimpaired (CU), mild cognitive impairment (MCI), and ACD participants.
Methods
CSF fatty acid profiles were analyzed from CU (n=68), MCI (n=38), and ACD (n=37) individuals aged 77.3 ± 7.7 years, sourced from the Huntington Medical Research Institutes (HMRI). Multivariable binary logistic regression identified the most effective CSF fatty acid biomarkers for distinguishing CU, MCI, and ACD groups. The top-performing CSF fatty acid biomarkers were combined with Aβ 42 , tau, and the Aβ 42 /tau ratio to evaluate their collective diagnostic performance. The model was adjusted for covariates, including age, sex, smoking status, hypertension, diabetes, and APOE genotype. Receiver operating characteristic (ROC) curves were generated for the top 10 CSF fatty acids ranked by area under the curve (AUC), sensitivity, and specificity, and their significance was assessed using the DeLong test.
Results
The top 10 fatty acids in CSF fractions demonstrated superior discrimination between CU individuals and those with MCI compared to traditional markers such as Aβ 42 , tau, and the Aβ 42 /tau ratio. Furthermore, incorporating a panel of these fatty acid biomarkers alongside Aβ 42 /tau significantly improved diagnostic accuracy. Age, sex, smoking, hypertension, APOE genotype, and diabetes did not significantly influence the model’s performance.
Conclusion
This study suggests that changes in fatty acid metabolism occur in early ACD pathology. Thus, strategies that regulate fatty acid metabolism may prevent cognitive decline in an older population.
Highlights
This study revealed that CSF fractions’ fatty acids have a better discriminatory power for CU from MCI than CSF Aβ 42 , tau, and Aβ 42 /tau ratio.
A panel of CSF fraction fatty acids combined with Aβ 42 /tau remarkably improved its diagnostic performance for differential diagnosis of CU, MCI, and ACD.
Clinical evaluation of these fatty acids will strengthen the detection of early cognitive impairment. A prospective large cohort multicenter study of the diagnostic utility of CSF fractions’ fatty acids will provide robust evidence before extensive clinical usage of these fatty acids.