In contrast, several studies [55,64,65] have now reported correlations between 11C-PIB amyloid binding and memory scores. Similarly, Rosenberg and colleagues [61] examined cognitive performance in the cohort of subjects described by Wong and colleagues [26] and found a significant correlation between florbetapir F 18 binding and seriously ADAS-cog (Alzheimer’s Disease Assessment Scale Cognitive Sub-scale) performance by normal elderly controls. Park and colleagues [66] have also recently reported a relationship between florbetapir PET amyloid binding and working memory performance in cognitively normal aging subjects. It is not surprising that the strength of correlation between PET result and cognitive performance, and/or the magnitude of the difference in cognitive performance between cognitively normal subjects with A??-positive and A??-negative PET scans, was modest and sometimes variable.

At least three factors work to limit the magnitude of effect that can be obtained in cognitively normal subjects. First, the range of cognitive performance in cognitively normal subjects is constrained by the criteria used to separate cognitively impaired subjects from cognitively normal. The earlier and more aggressively the diagnosis of impairment is made, the less potential for variance within the normal group as a function of amyloid level, as subjects with greater amyloid burden, and more advanced impairment, may be classified as cognitively impaired. Second, the outcome may depend on the difficulty of the cognitive tests used. More difficult tests are more likely to uncover deficits that may otherwise go unnoticed [64].

Finally, the relationship between amyloid binding and cognitive performance can be modified by the subject’s education/cognitive reserve [64,65]. Subjects with high education/high Brefeldin_A cognitive reserve appear to maintain cognitive function in the normal range for a longer period or in the face of greater PET amyloid binding than subjects selleck chemicals Gemcitabine with lower cognitive reserve. The Pike and colleagues [55] and the Rentz and colleagues [64] reports above both include scatterplots of cognitive performance as a function of amyloid binding (SUVR). Rather than a preferential distribution of abnormally low memory scores in association with high amyloid binding, the scatterplots are notable for the relative absence of high memory scores in the high amyloid group. It is tempting to speculate that this kind of distribution is the result of the limiting factors discussed above.