Exploring multimodal biomarkers on the horizon of the behavioral variant of frontotemporal dementia: Addressing the remaining diagnostic challenges with primary psychiatric disorders

Summary of findings In part I (fluid biomarkers) chapter 2, we successfully developed a robust method to isolate, characterize, lyse and quantify CSF-derived extracellular vesicles. Aiming to detect and quantify TDP-43 inside the CSF-EVs in a cohort of pathologically confirmed FTD patients we found that in the majority of subjects the TDP-43 levels were below the lower limit of detection of the Simoa assay. Using the automated Western Blot, the antibodies showed aspecific binding of the TDP-43 antibodies, which made the results not reliable. In chapter 3 we found that in the majority of the cohort of FTD and PPD patients, NfL concentrations were not detectable in urine and there was no correlation between the urine and serum NfL levels. As expected, serum NfL levels were higher in FTD, compared to PPD and controls, and ROC analysis yielded high diagnostic accuracies In chapter 4 we found that lower NPTX2 concentrations were specific to bvFTD, whereas higher concentrations of SNAP25 and Ng in particular seemed AD specific. The synaptic markers had however limited added value to NfL in diagnostic distinction of bvFTD from PPD and AD and did not correlate with social cognition or disease severity in bvFTD. In part II (neuroimaging biomarkers) chapter 5, we found subtle but distinct metabolic patterns in bvFTD and PPD. The dorsal anterior cingulate cortex (dACC) presented as a key hypometabolic region, being most strongly affected in bvFTD compared to PPD and controls. Additionally, the degree of hypometabolism was related to impaired social cognition, compulsiveness and executive dysfunctioning. In a subsequent series of analyses, we further demonstrated that the extent of reduced metabolic activity in the dACC was linked to compromised facial emotion recognition, a hallmark symptom of bvFTD. In part III (physiological biomarkers) chapter 6, we found that the end stages of bvFTD and yoAD are characterized by a broad spectrum of clinical symptoms and extend far beyond the initial behavioral and cognitive features, including severe autonomic dysfunction. In chapter 7, we found that FTD patients have widespread autonomic symptoms and smaller hypothalamic volumes compared to PPD and controls, as well as an association between autonomic symptoms and regional hypothalamic volumes. In part IV (distinction of bvFTD with other clinical entities: The bvFTD phenocopy syndrome) chapter 8, we report on an autopsy confirmed phenocopy bvFTD case without FTLD pathology which showed that the etiology in this phFTD patient was explained by the combination of a low average IQ, psychiatric traits of autism spectrum disorder and/or personality cluster C with a passive coping strategy and a tendency for somatization, in addition to the small size and neuropathological changes in the amygdala. In chapter 9, we found that phFTD patients performed evidently better at facial emotion recognition tests which provided the most optimal diagnostic distinction, and phFTD reported more depressive symptoms compared to bvFTD patients. Caregiver-reported questionnaires showed more symptoms in phFTD than as reported by the bvFTD cases themselves.