How accurate is the clinical diagnosis of Alzheimer disease (AD), and can we identify factors that predict other etiologies for dementia? In this issue of JAMA Neurology, Monsell et al1 report that 25% of individuals clinically diagnosed as having mild to moderate Alzheimer dementia with a Mini-Mental State Examination score of 16 to 26 at their last visit at a National Institute on Aging鈥揻unded AD research center had no more than sparse neuritic amyloid plaques on postmortem examination. The percentage of individuals with low amyloid levels was much higher in apolipoprotein E 蔚4 (APOE4) noncarriers (37%) vs carriers (13%). The finding of low amyloid levels in a substantial segment of individuals diagnosed as having AD in experienced centers is consistent with a series of recent reports from the National Alzheimer鈥檚 Coordinating Center neuropathology database and biomarker evidence from large phase 3 clinical trials.2-5 Serrano-Pozo et al2 found that 22 of 161 individuals (14%) diagnosed as having mild to moderate AD (Mini-Mental State Examination score of 16-26) in the National Alzheimer鈥檚 Coordinating Center neuropathology data set had no more than sparse neuritic plaques at autopsy. In 2 large phase 3 clinical trials of bapineuzumab for mild to moderate Alzheimer dementia, Salloway et al4 reported that 36% of APOE4 noncarriers and 6% of APOE4 carriers did not meet the amyloid positron emission tomography (PET) cutoff for amyloid positivity, percentages nearly identical to the neuropathological results reported by Monsell and colleagues.