Alzheimer’s disease pathophysiology: Investigating biological heterogeneity in cerebrospinal fluid

Currently, approximately 50 million individuals have dementia, and this number is expected to rise due to increased life expectancy. Responsible for approximately 70% of dementia cases, Alzheimer’s disease (AD) is the leading cause of dementia. AD is caused by accumulation of amyloid plaques and tau tangles in the brain, resulting in neuronal death and cognitive impairment. AD has several stages: first, there is a long preclinical period, followed by a prodromal or mild cognitive impairment (MCI) stage, which finally results in an AD-type clinical dementia stage wherein the affected individual ultimately can no longer live independently. Until recently, a diagnosis of AD relied on clinical criteria. However, pathological studies showed that clinical symptoms do not always accurately reflect the underlying pathology. The current research framework definition of AD now requires biomarkers that reflect amyloid and tau pathology, regardless of cognitive state. Such biomarkers include cerebrospinal fluid (CSF) levels of amyloid-beta 1-42 (amyloid), phosphorylated tau (p-tau) and total tau (t-tau). While amyloid and tau are important in diagnosing AD, an unresolved issue is that despite many clinical trials targeting amyloid and tau proteins, no disease-modifying treatments are available. The limited success of clinical trials targeting amyloid and tau may reflect that AD pathophysiology is heterogeneous, with patients differing in the precise underlying causes. The first studies that analysed biological heterogeneity focussed on the core AD biomarkers, and showed subgroups of individuals with AD dementia with different tau levels. It remains largely unknown if tau subgroups also exist in earlier clinical stages of AD, and if they differ in other biological processes. Another approach to study biological heterogeneity in AD is with new CSF proteomic technologies, which can analyse large numbers of proteins simultaneously. So far, CSF proteomic studies mostly compared clinical AD-type dementia to controls. However, for many proteins, these proteomic studies showed conflicting results, and it remains unclear whether this was due to study design factors or due to disease heterogeneity within AD. Another potential explanation of the conflicting results is that most of the previous studies relied on clinical status, while cognitively normal individuals can also have abnormal amyloid (preclinical AD). Additionally, CSF protein levels may differ depending on factors such as APOE ε4 carriership (i.e., the strongest genetic risk factor for AD), age and sex. A concomitant investigation of these factors is necessary to understand which processes are involved in early AD pathogenesis. Finally, memory dysfunction is the hallmark symptom of AD, and previous pathological studies have shown memory impairment is related to synaptic loss. It remains largely unknown when processes related to synaptic functioning start to alter during the development of AD, and whether such synaptic changes are related to the earliest changes in memory functioning in preclinical AD and MCI. In this thesis, we aimed to study biological heterogeneity in AD using CSF proteomic techniques. In the first part of this thesis, we aimed to identify and characterize AD subgroups based on CSF t-tau and p-tau levels. In chapter 2 and 3, we identified CSF tau subgroups in two independent cohorts in a data-driven manner, and studied whether the subgroups were related to specific clinical and biological characteristics, including biomarkers for amyloid metabolism, synaptic integrity and axonal damage. In the second part of this thesis, we investigated AD-related processes with CSF proteome-wide analyses. We investigated which proteins are consistently associated with AD (chapter 4), studied the effects of age, amyloid, APOE ε4 genotype and sex in normal cognition (chapter 5), and investigated how the synaptic proteome associates with memory decline in individuals without dementia (chapter 6).