. Predicting amyloid deposition using multiple neuropsychological measures. Human Amyloid Imaging 2011 Meeting Abstracts. 2011 Jan 15;


Objective: Many traditional neuropsychological measures cannot detect subtle cognitive changes at presumed preclinical stages of Alzheimer's disease characterized by PiB positivity. We administered FNAME, a challenging face-name memory association test, along with traditional neuropsychological measures, to investigate which tests showed the strongest ability to predict PiB retention in cognitively normal individuals.

Method: We studied 109 normal subjects (mean age = 73.5}8.1, education = 16.2}2.8; AMNART IQ = 122.6}9.3) with CDR scores = 0 and MMSE .28. PiB retention (DVR, cerebellar reference) was expressed as a dichotomous variable (PiB mean cortical DVR cutoff = 1.15). We transformed individual neuropsychological scores into z-scores (SRT, FCSRT, Trails A&B, FAS, 3 Categories, BNT), as well as subscores of the FNAME (e.g. face-name recall, face-occupation recall). Logistic regression was used to generate ROC curves, and the AUCs were used to compare the predictive accuracy for PiB positivity of FNAME alone vs FNAME with traditional neuropsychological measures. Both models included age and AMNART as independent variables.

Results: For the predictive equation generated by the first model, (FNAME immediate and delayed recall for names retained) the AUC was 0.73 (95% confidence interval [CI] = 0.61-0.85). For the predictive equation generated by the second model (model 1 variables plus Trails A and age retained), the AUC was 0.81 (95% CI = 0.71-0.91). No other cognitive measures were retained in the model.

Conclusions: The FNAME is a highly sensitive memory measure that has strong predictive accuracy for the presence of amyloid deposition alone. Moreover, the predictive ability of the FNAME was enhanced by a measure of processing speed (Trails A) and age when entered into a second model predicting PiB deposition. Taken together, results suggest that specific neuropsychological measures may help to identify the subset of clinically normal individuals who are amyloid positive.


Please login to recommend the paper.


No Available Comments

Make a Comment

To make a comment you must login or register.

This paper appears in the following:


  1. Miami: HAI Amyloid Imaging Conference Abstracts