Background and aims: Cerebral white matter hyperintensities (WMH) have been suggested to contribute to progression in Alzheimer's disease (AD). Quantification of WMH in patients can be performed both manually, where WMH is categorized according to the Fazekas scale, and automatically using software which calculates WMH based on a FLAIR MRI sequence. The aim of this study was to investigate to which extent the WMH-burden affects progression in a mixed population of clinical AD and prodromal AD. Furthermore, we assessed whether manual rating and automatic segmentation of WMH provide equal information on progression. Methods: Patients with clinical diagnosis of AD and MCI patients suspected of having early AD were included. Evaluation of progression was performed by an experienced clinician at a 12-month follow-up visit. Manual evaluation (Fazekas scale) of WMH was performed by an experienced neuroradiologist and automatic segmentation was performed as previously described (Koikkalainen et al, 2016, Neuroimage). Patients were examined for the association between WMH-burden at baseline and progression in disease after 12 months and stratified by diagnosis of AD without CVD and AD with CVD. Results: There was no significant difference between WMH-burden and progression status in either AD without CVD (p=0.122) or AD with CVD (p=0.159). However, there was a trend for a higher WMH-burden in progressed vs. stable patients diagnosed with AD with CVD. Conclusion: WMH-burden seems to have an impact on progression in AD only when present in large amounts. We are currently investigating the prognostic value of manual and automatic WMH-burden measurements. Background and aims: According to international criteria, amyloid-biomarkers are diagnostic elements for Alzheimer's disease (AD). Controversies about their accuracy for early and differential diagnosis may be explained by distinct pathological processes or by between-center variability in amyloid-markers. We pretended to compare the agreement between amyloid CSF-biomarkers, [11C]-Pittsburgh Compound Positron Emission Tomography (PIB-PET) and Florbetapir (18F) and the accuracy of these biomarkers for AD diagnosis. Methods: 96 patients with at least two amyloid markers were included. The clinical course was considered the diagnostic gold standard. We used locally established cutoffs of amyloid CSF-AD biomarkers-Aβ42<610pg/mL; Aβ42/Aβ40 ratio<0.068. PIB-PET and 18F were evaluated qualitatively. Results: There was entire agreement between PIB-PET and 18F. Amyloid Imaging Markers (AIM) agreed with Aβ42 in 77% cases and with Aβ42/Aβ40 ratio in 74%. Discrepancies were found in 10 clinically non-AD patients (9 with CSF-AD profile and negative AIM; 1 with positive AIM and a non-AD CSF profile) and 12 AD patients (11 with non-AD CSF and positive AIM; 1 with CSF-AD profile and negative AIM). CSF Aβ42 had, in this cohort, a sensitivity of 71% and a specificity of 66% for the diagnosis of AD and MCI-AD, with an overall diagnostic accuracy of 69%. AIM achieved a sensitivity of 85%, a specificity of 92%, with 88% diagnostic accuracy. Conclusion: Agreement between CSF amyloid and AIM indicates that these measures are not fully equivalent as surrogates of AD-pathology. CSF amyloid seems to have a moderate sensitivity and specificity for AD-related pathologies, while AIM achieves a higher diagnostic accuracy. These results deserve a neuropathological confirmation. Disclosure: Nothing to disclose
Dyremose, N., Bruun, M., Frederiksen, K., Gjerum, L., Rhodius-Meester, H., Baroni, M., … Santana, I. (2018). Impact of White Matter Hyperintensities on progression in Alzheimer’s disease Comparison of Amyloid Biomarkers in Alzheimer’s Disease-a Monocentric Study. European Journal of Neurology, 25.