The diagnostic challenge of neuropsychiatric symptoms in Alzheimer’s disease

Neuropsychiatric symptoms (NPS) such as affective disorders, apathy or “frontal” behavior, are present in the vast majority (>80%) of Alzheimer’s disease (AD) patients. NPS are associated with a variety of negative clinical outcomes, including lower quality of life (QOL), faster disease progression, and higher dementia care costs. Caregivers consider behavioral and psychological manifestations of AD the most troublesome symptoms, leading to earlier institutionalization. Despite high prevalence and clear clinical impact, NPS in AD are currently underrecognized, and thus undertreated, in the memory clinic. Importantly, at present there are no effective disease modifying drugs available for AD, but there are evidence-based treatments and interventions that alleviate NPS. This highlights the need to improve early and accurate detection of NPS in AD, assess the effects of NPS treatment, understand the neurobiological underpinnings of NPS in AD, and disseminate knowledge to the wide clinical and scientific society. We address each of these points in this project, which will result in better recognition, treatment and knowledge on the etiology of NPS in AD, ultimately reducing associated psychosocial and financial burden. In WP1, we aim to improve early and accurate clinical detection of NPS in AD by 1) analyzing the presence of NPS in the memory clinic through NPI-Q scores in a large retrospective multicenter longitudinal dataset covering all clinical AD stages ranging from subjective cognitive decline (n=~750), mild cognitive impairment (n=~680) to AD dementia (n=~1500), and 2) applying innovative machine learning techniques (“natural language processing”) to gain insight in nomenclature used for NPS in patient charts and to examine potential clinical underrecognition op NPS in AD in the memory clinic. In WP2, we will identify a prospective cohort of AD patients with clinically relevant NPS who in clinical practice are currently not treated for their behavioral or psychological symptoms. We will provide a patient centered neuropsychiatric intervention. This entails first, the evaluation of clinically relevant NPS in AD, and second, a personalized multidisciplinary treatment advice by an experienced psychiatrist, clinical neuropsychologist, consulting patient and caregivers’ wishes and practical options. We will examine whether this (early) intervention has significant impact on QOL and clinical outcome (cognition and functional ability) at 6 and 12 months of follow-up. Furthermore, we will investigate the cost-effectiveness of systematic recognition and tailored treatment of NPS in AD in the memory clinic. In WP3, we will use the behavioral variant of AD (bAD) as a model to elucidate the neuroanatomical and pathological basis of NPS in AD. bAD patients present with early and predominant behavioral and personality changes but share the same underlying neuropathology (i.e. amyloid plaques and neurofibrillary tangles) as “typical” amnestic-predominant AD patients (tAD). We have a unique retrospective dataset of 75 bAD patients at our disposal (the largest sample worldwide) with neuroimaging (MRI and PET, n=75) and pathological (n=30) data available. We will test whether white matter hyperintensities, deep gray matter structures, glucose hypometabolism and functional connectivity are related to behavioral abnormalities, by comparing bAD with tAD and behavioral variant frontotemporal dementia (bvFTD) patients. Using post-mortem analyses, we will assess whether the distribution of amyloid-ß, tau and other types of pathology differ between bAD and tAD patients. Finally, we will investigate the density of a specific neuronal population (Von Economo neurons) that is only found in the anterior cingulate cortex and frontoinsular cortex and has previously been linked to behavioral changes in bvFTD. WP4 is designed to disseminate the knowledge acquired in WP1-3. First, if our work-up of systematic diagnostics and treatment of NPS in AD proves to have beneficial effects on QOL and appears to be cost effective, we will disseminate these results to other memory clinics by providing recommendations for diagnostics and treatment of NPS in AD. Second, we will organize a series of consensus meetings with international experts to establish a common framework for the classification of bAD, resulting in a publication of formal diagnostic criteria for bAD. With great expertise on AD, NPS, neuroimaging and pathology, strong clinical infrastructures and a large scientific and clinical network, this research is well situated at VUMC, Erasmus MC and collaborator UCSF. The partial use of retrospective data further enhances the feasibility of this ambitious project. Active patient and caregiver participation, cost-effectiveness analyses of NPS treatment and a solid knowledge dissemination plan ensures that, aside from its scientific merits, this research is highly relevant to all of its stakeholders.