Lowe V, Jack C, Peller P, Tosakulwong N, Weigand S, Wiste H, Jordan L, Petersen R.
Biomarker Correlates in the ADNI AD Population Suggesting Dementia Unlikely Due to AD.
Human Amyloid Imaging Abstract. 2012 Jan 1;
Objective: ADNI includes subjects categorized with the clinical diagnosis of probable AD. Data from these subjects have been used as the basis for determining the utility of biomarkers in AD research. Recently, published modifications have been proposed to the clinical diagnostic characterization of AD by incorporating new biomarkers, especially in data used for research. We assessed the ADNI probable AD population to determine if any subjects are included that fit the proposed diagnostic classification of “Dementia unlikely to be due to AD” based on negative AB and neuronal injury biomarkers.
Methods: The ADNI AD population was searched for subjects with FDG PET and amyloid biomarkers. FDG was used as a neuronal biomarker surrogate. The subgroup was interrogated by searching for FDG findings that were negative for AD. Specifically, FDG patterns were assessed as being AD-like, FTD-like or other by blinded readers using Cortex ID. In addition we searched the sub-population for those who had negative AB biomarkers (CSF and/or Amyloid PET).
Results: Of the 92 subjects we reviewed, 58 (63%) had positive amyloid biomarkers and an AD pattern on FDG. Seven subjects (8%) were identified who had negative amyloid biomarkers. Of these 7, FDG PET demonstrated an AD pattern in 3, an FTD pattern in one and an “other” pattern in 3. Two (2/7) had both negative CSF and PiB. FDG PET findings uncharacteristic for AD were seen in an additional 27 subjects with positive amyloid studies.
Conclusion: Several subjects within the ADNI AD population fit the recently proposed characterization of “Dementia unlikely to be due to AD” based on negative AB and neuronal injury biomarkers. Several others have a single biomarker that is negative. These findings raise important considerations in research analysis of ADNI AD population data particularly with respect to the specificity of biomarkers and clinical diagnosis.