The Performance of Memory Scales for Predicting Alzheimer’s Disease Biomarkers in Amnestic Mild Cognitive Impairment Patients

Abstract

AbstractBackgroundAnti‐amyloid‐β therapies have been approved to treat mild Alzheimer’s disease (AD) and amnestic mild cognitive impairment (aMCI) due to AD. Therefore, diagnosis of AD pathologies is crucial for selecting patients for treatment. Positron emission topography (PET) is used to detect biomarkers for β‐amyloid (A) and tau (T). However, the availability of PET is still limited. Neuropsychological tests may help to select patients who are likely to have AD pathologies. This study aims to select a scale with high performance for AD biomarkers in aMCI patients.MethodParticipants with aMCI were recruited from the Memory Clinic at King Chulalongkorn Memorial Hospital. PET was used to detect AD biomarkers. The Wechsler Memory Scale, Fourth Edition (WMS‐IV) and the Montreal Cognitive Assessment (MoCA) were carried out. We selected four scales which are frequently used to assess AD patients including Visual Reproduction (VR), Logical Memory (LM), and Verbal Paired Associates (VPA) from the WMS‐IV and the Memory Index Scale (MIS) from the MoCA to analyze as predictors. Patients with A+T+ were defined as aMCI with AD biomarkers (aMCI‐AD). Patients with A+T‐ and A‐T‐ were defined as aMCI with non‐pathophysiology of AD (aMCI‐npAD). Multivariable logistic regression was used to construct the predictive model. Area under receiver operating characteristic curve (AUROC) was used to identify the best model in predicting A+T+.ResultAmong 35 participants enrolled (18 aMCI‐AD and 17 aMCI‐npAD), 25 (71.4%) were female. The mean age and education were 71.3±8.1 and 13.8±4.4 years, respectively. The predictive model using LM demonstrated the highest AUROC of 0.84 (95%CI 0.70–0.93), with a sensitivity, specificity, positive predictive value, and negative predictive value of 0.83, 0.71, 0.75, and 0.80, respectively, at the cut‐point with highest accuracy of 77.1%; followed by VR (AUROC 0.78; 95%CI 0.61–0.95), MIS (0.74; 95%CI 0.57–0.91), and VPA (0.68; 95%CI 0.50–0.87).ConclusionLM has the highest performance for AD pathologies among the 4 scales to predict AD biomarkers in aMCI. We propose that before undergoing PET, LM may be an appropriate test for selected aMCI patients with a high chance of having AD pathologies.