Alzheimer’s disease (AD) is a progressive neurodegenerative disease that represents a global health problem. AD is the leading cause of dementia, representing about 70% of cases. In the world, there are 50 million people with dementia, and it is estimated that this figure will triple in the next two decades [1]. AD mainly affects older adults, generating a slow and progressive deterioration of cognitive ability and interfering with normal brain functions. Due to the fact that neural damage initially affects areas of the brain involved in memory, language, and thinking, early symptoms of AD can be manifested as difficulty in remembering recent conversations, names, or events. Depression and apathy are also early manifestations. Subsequently, problems begin in communication, judgment, and confusion. Eventually, brain damage covers areas involved in basic functions such as walking, swallowing, and speaking [2]. The neurodegenerative process begins years before the appearance of the first symptoms. In clinical settings, diagnosis is probable or possible, and it is made by clinical criteria, mainly those of the National Institute on Aging and Alzheimer’s Associations (NIA-AA) and the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) criteria ^[3][4][3,4]. Frequently, the symptoms are subtle and are usually ignored, diminished, and undiagnosed, which delays proper treatment and care [2]. At the stage in which it is typically diagnosed, there has already been significant neuronal loss in many brain areas. Therefore, the detection of biological markers is necessary to increase the opportunity for early diagnosis of AD.

2. The Need for Alzheimer’s Disease Biomarkers

The National Institutes of Health (NIH) defines a biomarker as a characteristic that is objectively measured and evaluated as an indication of normal biological processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention [5]. Biomarkers are useful tools to screen, characterize, rule out, diagnose, stage, and monitor diseases, as well as to inform prognosis, individualize therapeutic interventions, predict adverse drug reactions, and identify cell types [6]. One of the most important challenges of neurodegenerative diseases such as AD is the subjective approaches to perform the diagnosis that delays the opportunity for adequate and timely management and treatment. Thus, the detection of biomarkers at the early stages of the disease is indispensable to increase the opportunity to delay AD progression or even mild cognitive impairment (MCI), which are markers of early detection. The NIA-AA criteria for AD are classified into three categories: (1) probable AD dementia, (2) possible AD dementia, and (3) probable or possible AD dementia with evidence of biomarkers for the pathophysiological process of AD. The first two are established in clinical settings, and the last one is established for research purposes [4]. Biomarkers for the pathophysiological process of AD included in these criteria are subdivided into two main classes: markers of amyloid deposition and markers of neuronal damage. The former includes a decrease in cerebrospinal fluid (CSF) amyloid beta levels and positive positron emission tomography (PET) amyloid imaging. Increased CSF tau (including total tau and phosphorylated tau) decreased fluorodeoxyglucose positron emission tomography (FDG-PET) consumption in the temporoparietal cortex, and disproportionate atrophy in the medial, basal, and lateral temporal lobe temporoparietal cortex and in the medial parietal cortex are markers of neuronal damage. Although the presence of these biomarkers increases the certainty of the diagnosis of AD, the use of these biomarkers as a routine diagnostic test is not recommended [4]. One of the reasons is the limited access to these biomarkers in common clinical settings. Table 1 summarizes the biomarkers recognized in the diagnostic criteria of the NIA-AA and DSM-5, as well as their use and interpretation in clinical practice.

* Use and interpretation of biomarkers in diagnostic criteria.

3. Brain Imaging Markers

Structural imaging studies in AD show atrophy in many brain areas. This atrophy begins in the medial temporal lobe, in areas including the entorhinal cortex and hippocampus. Structures of the limbic system, such as the olfactory bulb tract, thalamus, and amygdala, are also affected. Subsequently, there is a loss of the gray matter of the parahippocampus and medial temporal gyrus. The frontal, parietal, and temporal areas of the brain also suffer volume reduction [7]. These structural changes can be evaluated by magnetic resonance imaging (MRI). MRI is a non-invasive imaging procedure that does not require the injection of a contrast agent or radiation exposure. However, the analysis and rating of these images are qualitative and subjective, causing different results between evaluators. Although there are several MRI techniques, only traditional structural imaging is currently recommended for routine use in clinical settings. Automated methods that measure hippocampal volume have been proposed to decrease the variability of results, achieving the correct classification rate for AD and cases without cognitive impairment in about 80% and 65% for stable and progressive MCI cases [8]. PET imaging studies use radiotracers that bind to insoluble fibrillar forms of Aβ40 and Aβ42 to visualize amyloid deposition in the brain in vivo [9]. PET amyloid detection is highest in the frontal cortex, followed by the parietal, temporal, and occipital cortex, and striatum (Table 2) [10]. The amyloid regional distribution detected with PET correlates to the distribution of amyloid deposits observed in autopsies ^[11][12][13][11,12,13]. This Table summarizes the findings regarding the most accepted and widely described biomarkers, such as amyloid-β and tau proteins in blood and cerebrospinal fluid (CSF), postmortem human brains, and PET images. In addition, in biological processes altered in Alzheimer’s disease, molecules such as calmodulin and organelles as primary cilia are shown as sources of potential biomarkers in different human fluids and tissues, as well as in transgenic animal models of AD. ↑: increased; ↓: decreased; *: in blood exosomes. FDG-PET detects the metabolism of glucose in patients’ brains. In patients with early AD, the areas of glucose hypometabolism have been commonly observed in the parietotemporal association cortices, posterior cingulate cortex, and the precuneus [9]. In a meta-analysis that included 119 studies, FDG-PET was found to have a sensitivity of 90% and a specificity of 89% for identifying AD when compared to controls without dementia. In addition, FDG-PET was able to discriminate AD dementia from other causes of dementia with a sensitivity of 92% and a specificity of 78% [30].

4. Postmortem Markers

A diagnosis of definite Alzheimer’s disease requires histopathologic confirmation obtained from a biopsy or autopsy: the presence of neuritic plaques composed of Aβ aggregates and neurofibrillary tangles (NFTs) formed from hyperphosphorylated tau protein (Table 2) ^[15][31][15,31].

5. Markers in Biological Fluids (Cerebrospinal Fluid and Plasma)

The CSF is in direct contact with the CNS and is considered to reflect the pathological processes that occur in the brain. For these reasons, CSF has been studied as a resource to study biomarkers in AD. The main proposed biomarkers are amyloid beta peptide (specifically Aβ42), total tau (t-tau), and phosphorylated tau (p-tau181 and p-tau231) [32]. Various studies have shown that whereas Aβ levels are decreased in the CSF of patients with AD, tau (t-tau and p-tau) levels are, in contrast, increased (Table 2) [14]. CSF biomarkers are strongly associated with AD diagnosis and cortical Aβ deposition [33]. Correlations exist between CSF p-tau concentrations and scores of NFTs and hyperphosphorylated-tau load in all neocortical regions [34]. The reduction in Aβ42 is about 50% in AD patients compared to age-matched controls without dementia [35]. The value of Aβ42 in the CSF was analyzed in a systematic review where four studies were included, and a median specificity of 64% and sensitivity of 81% were reported [36]. Total tau CSF levels are about three times higher in AD patients than in controls [32]. Accuracy increases when both CSF biomarkers are combined, as shown by a study with neuropathological confirmation of the diagnosis, which showed that the p-tau/Aβ42 ratio has a sensitivity of 91.6% and a specificity of 85.7% for AD diagnosis [32]. In clinical settings, biomarkers are particularly relevant for both the diagnosis and follow-up of patients with suspected AD. In the very near future, they will also be decisive as markers of response and prognosis in the face of new treatments based on immunotherapy against amyloid peptides ^[37][38][37,38]. Biomarkers for diagnostic support are essential to be directly related to key aspects of the AD pathological process, i.e., abnormal processing of amyloid beta peptide and total and phosphorylated tau protein. As mentioned above, CSF determination of Aβ peptide fractions 1–40 and 1–42, together with total tau and phosphorylated tau proteins, currently offer high diagnostic accuracy, both in terms of sensitivity, specificity, and predictive values. Recently, diagnostic biomarkers in plasma have also been developed as a method to facilitate and reduce the costs of population screening. The performance of the Aβ1-42/Aβ1-40 index and 181p-tau together with plasma glial fibrillary acidic protein (GFAP) has generated great interest [39]. Compared to necropsy as the gold standard, PET and CSF-related biomarkers in combination showed high concordance with the pathological stage; alone, they showed high accuracy in discriminating autopsy-confirmed AD. However, the plasma-related biomarkers alone showed better discriminative performance only when combined with the apolipoprotein E (APO) Eε4 genotype [40]. In this context, biomarkers that accurately reflect disease progression and have prognostic capabilities are also needed. In AD, the cognitive reserve accumulated throughout life, based on the level of education, professional qualifications, hobbies, and the intensity of social relationships, cushions the severity of the clinical manifestations, and it is not uncommon for individuals who, despite bearing a high lesional burden typical of the advanced stages of the disease, show few symptoms with little impact on their daily lives [41]. Due to the amyloid beta peptide level reaching a plateau relatively early in both PET and CSF in the continuum of AD, it was not considered a good marker of progression, but it is an excellent biomarker to evaluate the response to new anti-amyloid drugs [42]. Other proteins, such as GFAP, particularly in plasma [39], or elevated CSF levels of proinflammatory cytokines MCP-1 or chitinase-3-like protein 1 (YKL-40), are currently under evaluation as markers of disease progression [43]. In the current context dominated by recent anti-amyloid therapies, it seems likely that an amyloid beta signature biomarker will become mandatory and that combination with other biomarkers of neurodegeneration or neuroinflammation will become standard diagnostic practice.