Elecsys Cerebrospinal Fluid Immunoassays Accurately Detect Alzheimer’s Disease Regardless of Concomitant Small Vessel Disease

INTRODUCTION

Alzheimer’s disease (AD), the most common cause of dementia, is a progressive neurodegenerative disease that is typically recognized by initial memory impairment and cognitive decline, followed by the deterioration of language and behavioral functions, visuospatial orientation, and the motor system [1]. AD is defined by neuropathological changes, namely neuritic plaques containing amyloid-β (Aβ) peptides and neurofibrillary tangles containing aggregated tau proteins, which ultimately lead to neuronal injury and degeneration [2]. While the underlying neuropathology of AD is well defined, the clinical presentation of the disease is heterogeneous; approximately 25% of AD cases do not conform to typical AD presentation [3]. Multiple stages of AD have been identified including pre-symptomatic stage AD, patients with subjective cognitive decline who exhibit AD biomarkers, mild cognitive impairment (MCI) due to AD, mild dementia due to AD, and moderate-to-severe AD [3–6]. MCI due to AD and mild dementia due to AD are often classed as “early AD” [7, 8].

The core AD biomarkers can be classified into two groups: biomarkers of Aβ peptide deposition, including decreased levels of Aβ₄₂ in cerebrospinal fluid (CSF) and amyloid positivity using positron emission tomography (PET), and biomarkers of neuronal injury and degeneration, including elevated levels of phospho-Tau181 (pTau181) and total Tau (tTau) in CSF, decreased fluorodeoxyglucose F18 (FDG) uptake on PET, and disproportionate atrophy on structural magnetic resonance imaging [9]. Numerous clinical studies have demonstrated that CSF biomarkers are associated with AD pathology and have demonstrated the ability to accurately identify AD at the stage of incipient dementia [10–12]. Thus, CSF biomarkers have been incorporated into multiple diagnostic frameworks for AD [13–15].

The accurate and early diagnosis of AD is critical to delivering stratified patient care and is a key consideration for current clinical trials evaluating novel treatments targeting AD neuropathology [16]. Many existing CSF immunoassays for AD are limited by lot-to-lot and interlaboratory variations, which have hindered the widespread introduction of CSF biomarkers into clinical practice [17]. The fully automated Elecsys^® CSF immunoassays (Roche Diagnostics International Ltd, Rotkreuz, Switzerland) accurately detect amyloid positivity by determining CSF pTau181/Aβ₄₂ and tTau/Aβ₄₂ biomarker ratios and have demonstrated superior interlaboratory variation (coefficient of variation: 4%) compared with existing manual CSF assays (coefficient of variation: >15%) [18]. The Elecsys CSF immunoassays can also accurately predict future disease progression and, thus, have the potential to support the diagnosis of early AD [18].

Though dementia due to AD is the most common form of dementia, less than half of cases are expected to be solely caused by AD and most cases are expected to be mixed dementias [19]. AD frequently coexists with other neurodegenerative co-pathologies, for example, vascular disease including cerebral small vessel disease (SVD) and large vessel disease [2, 20]. Vascular disease is thought to play a major role in the pathogenesis of AD [20]. SVD affects small arteries, arterioles, veins, and capillaries of the brain, and is the most frequent cause of vascular dementia [21]. The disease is characterized by white matter hyperintensities (WMH), small subcortical infarcts, lacunes, enlarged perivascular spaces, microbleeds, and brain atrophy [22]. WMH volume, particularly in parietal regions, is elevated among individuals with and at risk of AD [23]. Additionally, the presence of WMH increases the risk of cognitive decline and AD and has been shown to contribute to disease progression and severity [24–26]. Differentiating dementia due to SVD from dementia due to AD with concomitant SVD is challenging in clinical practice [27]. CSF biomarker levels could be altered due to impaired cerebral drainage caused by SVD [28]. In patients with early dementia, it is important to differentiate those with or without underlying AD, as the correct diagnosis is critical to delivering stratified patient care, particularly as novel disease-modifying treatments (DMTs) for AD (e.g., anti-Aβ drugs) are thought to be most effective in the early stages of the disease [29]. For example, a patient diagnosed with early dementia due to underlying AD may be optimally treated with anti-Aβ DMTs, whereas a patient with early dementia due to SVD may be optimally treated with a potential intervention for SVD such as endothelin antagonists, neurotrophins, or phosphodiesterase inhibitors [30].

This study aimed to characterize the results of Elecsys CSF immunoassays in patients diagnosed with early AD (based on core AD clinical criteria), with or without FDG-PET positivity for underlying AD pathophysiology, and with varying extent of SVD, and to identify a possible relationship between WMH and parameters of Elecsys CSF immunoassays.

METHODS

RESULTS

DISCUSSION

In patients with early AD, it is important to differentiate those with underlying AD pathophysiology regardless of the extent of SVD as the correct diagnosis is critical to delivering stratified patient care. We aimed to characterize the results of Elecsys CSF immunoassays for detecting underlying AD pathophysiology in patients with early AD (based on core AD clinical criteria) with varying extent of concomitant SVD (assessed by the volume of WMH). We also aimed to identify a possible relationship between WMH and parameters of Elecsys CSF immunoassays. In patients with early AD, the extent of WMH showed significant correlation with Aβ₄₂/Aβ₄₀ ratio, tTau, and tTau/Aβ₄₂ ratio, suggesting a possible confounding effect on the performance of Elecsys CSF immunoassays. The point estimates for sensitivity and specificity of the CSF immunoassays, compared with FDG-PET positivity as standard-of-truth, were generally comparable or greater in patients with high WMH versus patients with low WMH, suggesting that the extent of WMH is unlikely to affect the performance of Elecsys CSF immunoassays in detecting AD pathophysiology. For the association between Elecsys CSF immunoassay biomarker positivity with FDG-PET positivity, WMH were not a significant predictor and did not interact with biomarker positivity, which further highlights the robustness of the immunoassays to detect AD pathophysiology, regardless of extent of WMH. However, WMH did modify the association between pTau181 and tTau. In summary, our results demonstrate that the performance of the Elecsys CSF immunoassays in detecting AD is unlikely to be affected by the presence of concomitant SVD; therefore, they have potential to help differentiate between dementia due to SVD and AD with concomitant SVD.

Point estimates for specificity compared with FDG-PET positivity were relatively low across all WMH groups in comparison to higher published specificities for CSF biomarkers compared with amyloid PET positivity. Therefore, we conducted post-hoc analyses to compare the sensitivity and specificity of the Elecsys CSF immunoassays using an alternative standard-of-truth, Aβ₄₂/Aβ₄₀ positivity, which presumably develops earlier during the course of AD compared with FDG-PET positivity [40]. As expected, point estimates for sensitivity and specificity compared with CSF Aβ₄₂/Aβ₄₀ positivity were greater than those compared with FDG-PET positivity.

This study provides further evidence that CSF biomarkers have the potential to increase the accuracy of AD diagnosis, particularly at the earlier stages of disease, as well as in cases of atypical presentation and mixed pathology, i.e., AD with concomitant SVD or large vessel disease. Though patients were diagnosed with early AD based on expert opinion at a highly specialized center using core AD clinical criteria, almost 50% did not show a typical FDG-PET pattern for AD pathophysiology, underscoring the importance of using biomarkers to support the identification of the underlying neuropathology of dementia. Ongoing drug discovery efforts focus on developing DMTs that aim to delay the onset or progression of dementia and must be initiated early in the disease process [42]. Current AD treatments provide symptomatic benefit only; as such, it is important to identify patients with AD early in the disease process, when therapies are likely to be most effective [29]. Therefore, the accurate and early identification of underlying AD pathology is vital to ensure the right diagnosis and stratified treatment, i.e., symptomatic treatments or DMTs; it is also key for patient empowerment. There is growing evidence that patients and carers wish to reach a diagnosis as soon as possible, in order to reduce the anxiety caused by symptoms and to allow patients to initiate lifestyle changes and plan for the future, e.g., implement safeguarding procedures to prevent accidents [43–47]. In addition to improving diagnostic accuracy, CSF biomarkers have the potential to facilitate health care professionals in understanding AD etiology, including AD mixed pathologies.

Since CSF biomarkers have been incorporated into various diagnostic guidelines for AD, there is a demand for accurate and reliable methods for their measurement [12–15]. Notably, in 2018, the first-generation Elecsys β-Amyloid(1–42) CSF and Phospho-Tau (181P) CSF immunoassays were granted United States Food and Drug Administration Breakthrough Device Designation to support the improved diagnosis of AD in concordance with amyloid PET visual read result [48]; this was followed by United States Food and Drug Administration approval for the second-generation Elecsys β-Amyloid(1–42) CSF II and the updated Elecsys Phospho-Tau (181P) CSF immunoassays on December 8, 2022 [49].

SVD can contribute to the pathogenesis of both vascular dementia and AD. SVD has similar pathophysiological mechanisms to AD including oxidative stress, inflammation, mitochondrial disruption, and metabolic dysfunction, as well as similar risk factors including hypertension and diabetes [27]. For these reasons, differentiating between dementia due to SVD and AD with concomitant SVD is controversial and poses a difficult challenge. Vascular dementia and AD are both leading causes of dementia; SVD potentially interacts with pathophysiological processes in AD, either independently or through additive or synergistic effects on cognitive decline [20]. Differentiating between patients with AD and patients with dementia due to SVD or vascular dementia would benefit the patient in terms of treatment; though studies have reported that the treatment and prevention for vascular dementia and SVD may benefit patients with AD, there is still lack of evidence in clinical application of treatments that benefit both AD and SVD [50]. It is possible that CSF biomarkers may share a direct relationship with SVD and AD pathology. SVD may cause alterations in CSF biomarker levels due to impaired cerebral drainage [28]. Moreover, patients with normal pressure hydrocephalus, an expansion of the CSF-filled brain ventricles, have implicated impaired function of the glymphatic system [51], which is often associated with AD pathology [52]. Although there are a number of publications linking normal pressure hydrocephalus with AD biomarkers [53, 54], further studies are warranted to validate the association between them. In this study, alterations in CSF biomarker levels due to impaired cerebral drainage may have resulted in the establishment of WMH as a significant predictor for the pTau181 and tTau biomarker association.

While the monocentric design is a key strength of this study, reducing heterogeneity and variability, this did incur a relatively small sample size (n = 84), though the patients were highly characterized. This study is among the first to investigate differential diagnoses in patients with early AD using CSF biomarkers; however, as there is no independent biomarker for amyloid pathology, the accurate and reliable diagnosis of early AD was a limitation. Another limitation of this study was the use of a pre-analytical protocol that was not in accordance with Elecsys CSF immunoassay method sheets; however, this was mitigated by the robustness of the biomarker and biomarker ratio cut-offs [18].

There is now increasing recognition of the importance and value to the research and clinical communities of including underserved populations in AD biomarker studies [55, 56], particularly since some of those who are disadvantaged and/or underrepresented in clinical research may have greater risks for AD neuropathology or co-pathologies such as SVD [56–59]. Thus, future studies for the purposes of validating the results of this study should include: a heterogeneous population encompassing the full spectrum of racial, ethnic, and socioeconomic backgrounds; a larger number of patients with early AD (and at earlier disease stage, i.e., asymptomatic patients at risk); and a broader study for differential diagnoses, i.e., in patients with early dementia with or without underlying AD pathophysiology and with or without concomitant large vessel disease, and in patients with SVD without AD, which may be more reflective of a real-life clinical cohort. Validation of the results in a diverse population may clarify whether different populations require different interpretations of the results, as there is evidence suggesting differences in biomarker levels between race and age, particularly in pTau181 levels [60, 61].