Identification and validation of Alzheimer’s disease-related metabolic brain pattern in biomarker confirmed Alzheimer’s dementia patients

Abstract

Metabolic brain biomarkers have been incorporated in various diagnostic guidelines of neurodegenerative diseases, recently. To improve their diagnostic accuracy a biologically and clinically homogeneous sample is needed for their identification. Alzheimer’s disease-related pattern (ADRP) has been identified previously in cohorts of clinically diagnosed patients with dementia due to Alzheimer’s disease (AD), meaning that its diagnostic accuracy might have been reduced due to common clinical misdiagnosis. In our study, we aimed to identify ADRP in a cohort of AD patients with CSF confirmed diagnosis, validate it in large out-of-sample cohorts and explore its relationship with patients’ clinical status. For identification we analyzed 2-[18F]FDG PET brain scans of 20 AD patients and 20 normal controls (NCs). For validation, 2-[18F]FDG PET scans from 261 individuals with AD, behavioral variant of frontotemporal dementia, mild cognitive impairment and NC were analyzed. We identified an ADRP that is characterized by relatively reduced metabolic activity in temporoparietal cortices, posterior cingulate and precuneus which co-varied with relatively increased metabolic activity in the cerebellum. ADRP expression significantly differentiated AD from NC (AUC = 0.95) and other dementia types (AUC = 0.76–0.85) and its expression correlated with clinical measures of global cognition and neuropsychological indices in all cohorts.

Automated differential diagnosis of dementia syndromes using FDG PET and machine learning

“…Patients with AD fulfilled diagnostic criteria for probable AD dementia (amnestic presentation; McKhann et al, 2011 ) and had the diagnosis confirmed by cerebrospinal fluid (CSF) analysis of Aβ42 or Aβ42/Aβ40, p-tau, and t-tau according to the recent National Institute of Aging-Alzheimer’s Association research framework ( Jack et al, 2018 ). A detailed CSF procedure has been described previously ( Perovnik et al, 2022a ). Patients with DLB were diagnosed according to the 4th Consensus report of the DLB Consortium ( McKeith et al, 2017 ), followed by a diagnosis confirmation at an office visit conducted at least 12 months after the onset of symptoms.…”

“…Patients with FTD were diagnosed with possible FTD according to the International Behavioral Variant FTD Consortium criteria ( Rascovsky et al, 2011 ) and had their diagnosis confirmed at a follow-up visit conducted at least 12 months after the onset of symptoms. AD and DLB cohorts and their FDG PET scans included in this study have appeared previously ( Perovnik et al, 2022a , b ). NCs completed clinical, neuropsychological, and FDG PET brain imaging for purposes of an earlier research project ( Tomše et al, 2017 ).…”

“…Multivariate analysis approaches, such as scaled subprofile model/principal component analysis (SSM/PCA) applied to FDG PET images, can reveal characteristic metabolic brain patterns, which expressions can be prospectively quantified on a single case basis ( Spetsieris and Eidelberg, 2011 ). AD-related pattern (ADRP; Perovnik et al, 2022a ), DLB-related pattern (DLBRP; Perovnik et al, 2022b ) and FTD-related pattern (FTDRP; Rus et al, 2021 ) have been identified and validated in different clinical cohorts in the past. Similarly, a pattern of default mode network (DMN)—a dominant resting-state network in healthy individuals, which is affected also in the pathogenesis of neurodegenerative dementias—has been characterized with FDG PET ( Spetsieris et al, 2015 ).…”

“…Similarly, a pattern of default mode network (DMN)—a dominant resting-state network in healthy individuals, which is affected also in the pathogenesis of neurodegenerative dementias—has been characterized with FDG PET ( Spetsieris et al, 2015 ). Based on the expression of metabolic brain patterns, we can very accurately distinguish between patients with dementia and normal controls (NCs; Perovnik et al, 2022a , b ). However, it was also observed that the information of a single pattern’s expression score may not be sufficient to distinguish between different neurodegenerative dementias accurately ( Katako et al, 2018 ).…”

Diagnosis of autism spectrum disorder based on functional brain networks and machine learning

“…The same sliding workflow was considered with samplings of 10, 20, 30, 50, 124, and 188 patients. The choice of such different sizes was based on previous neuroscience studies that used fMRI of similar sample sizes, respectively [119][120][121][122][123][124].…”