MRI and PET images, and to some extent cerebrospinal fluid proteins, can help physicians predict which patients with mild cognitive impairment will likely develop Alzheimer’s disease, says a study recently published in the journal Radiology.
Researchers from North Carolina and New Mexico undertook a study to determine which Alzheimer’s disease biomarkers, when combined with clinical parameters, could aid in determining disease progression from mild cognitive impairment to Alzheimer’s disease. Their study included 97 patients with mild cognitive impairment who underwent Alzheimer’s Disease Neuroimaging Initiative (ADNI) baseline MR imaging and fluorine 18 fluorodeoxyglucose (FDG) PET. Cerebrospinal fluid proteins were analyzed for proteins.
The logistic regression models considered were all combinations of MR images, PET, and CSF markers with covariates (age, education, apolipoprotein E genotype, Alzheimer’s Disease Assessment Scale-Cognitive subscale score).
The researchers analyzed the MR imaging-derived gray matter probability maps and FDG PET images using independent component analysis. The loading parameters of the images plus the CSF proteins were entered into logistic regression models with a dependent variable of conversion to Alzheimer’s disease within four years.
Findings showed that by combining MR imaging and FDG-PET imaging, and CSF data with routine tests provided a significantly higher degree of accuracy when predicting which patients would go on to develop Alzheimer’s disease when compared with clinical testing alone. Patient misclassification was 28.4 percent with the combination, down from 41.3 percent with clinical testing.
The authors noted that FDG-PET contributed more, providing better prognostic information, than did MR imaging or the CSF findings.