Premium
O1‐07‐05: Mild cognitive impairment patients with suspected non‐alzheimer's disease pathology
Author(s) -
Wisse Laura,
Butala Nirali,
Das Sandy,
Yushkevich Paul,
Wolk David
Publication year - 2015
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2015.07.067
Subject(s) - neurodegeneration , cerebrospinal fluid , medicine , alzheimer's disease neuroimaging initiative , alzheimer's disease , pathophysiology , hyperintensity , biomarker , cognitive impairment , snap , white matter , dementia , pathology , neuroimaging , gastroenterology , disease , magnetic resonance imaging , psychiatry , radiology , biology , biochemistry , computer graphics (images) , computer science
Background: The combination of a positive amyloid burden with clinical diagnosis ofMCI has been used to support inclusion in clinical trials and to aid diagnosis. Yet despite amyloid confirmation, clinical progression varies widely. Using FDG PET, we characterized neurodegeneration in amyloid positive MCI subjects (MCI(Am+)) and compared it to baseline and longitudinal clinical endpoints. In particular, we sought to understand MCI(Am+) patients who do not exhibit typical AD hypometabolism, and how this might guide trial stratification and clinical care. Methods:We identified 79 Early MCI (EMCI) and 137 late MCI (LMCI) Am+ ADNI subjects having FDG, amyloid, and clinical measures. We applied to each FDG scan a previously validated machine-learning AD Progression Classifier that determines a numeric score (CV1) reflecting the degree to which a subject expresses a metabolic pattern corresponding to a progressive path toward AD dementia. Subjects were stratified by baseline CV1, and relationships with MMSE, CDRsb, ADAS-cog13, and Logical Memory examined with and without adjustment for baseline endpoints. We additionally examined regional glucose metabolism (rCMglu) in MCI(Am+) negative for AD-like pattern compared to NL(Am-) and EMCI(Am-), and as related to clinical status. Results: Up to 33% EMCI(Am+) and 16% LMCI(Am+) subjects were negative for an AD-like pattern. Baseline CV1 correlated with subsequent clinical worsening, which was least in pattern-negative subjects (p<0.001). Among MCI(Am+) subjects lacking an AD-like pattern, EMCI(Am+) rCMglu was