Shifting our focus to blood-based biomarkers of Alzheimer’s disease: the promise of ceramides and sphingomyelins

The public health burden of Alzheimer’s disease (AD) threatens to explode in the middle of this century. The longevity of the population of the USA and other countries is increasing and the prevalence of AD, comprising 60–70% of all dementia cases, doubles every decade after the age of 65 years [1]. Barring the development of effective preventative measures or treatments for AD, it is expected that 115 million people globally will have dementia by 2050, with the preponderance of cases in developing countries.

As evidenced by the many failed treatment trials for AD, there is no treatment benefit in the fully symptomatic stage of the disease, partly because the treatments may be administered too late in the disease process. AD pathophysiology (amyloid-β deposition as evidenced from amyloid imaging or cerebrospinal fluid [CSF] amyloid, and neuronal injury as evidenced by quantitative MRI, ¹⁸FDG PET or CSF tau) are thought to begin several years, probably decades, before the emergence of clinical symptoms [2]. Amyloid pathology has been hypothesized to precede neuronal injury by up to 20 years. Neuronal injury and neurodegeneration also precede cognitive changes, but are temporally closer. In recognition of the need to identify individuals at the earliest stages of the AD pathophysiological process, workgroups of the National Institute on Aging and the Alzheimer’s Association recently published new guidelines for the identification of the preclinical and mild cognitive impairment phases of AD [3,4]. These guidelines, presently only for research, place an unprecedented emphasis on biomarkers of AD pathophysiology. Continued exploration and validation of the temporal relationships of these imaging and CSF biomarkers with the emergence and progression of clinical symptoms are ongoing and could lead towards the use of these markers in clinical practice. It is therefore appropriate to consider the role of these biomarkers in the population for the early detection of pathology and preclinical AD diagnosis, prediction of progression to clinical symptoms, use as surrogate markers of disease modification and/or confirmation of AD diagnosis.

The proposed CSF and neuroimaging biomarkers will be helpful in the differential diagnosis of dementia and to determine which asymptomatic individuals may be at risk of AD pathology, particularly if potential treatments geared towards preclinical AD become available. However, the collection of CSF and neuroimaging will not be ideal for the first-line screening of populations for the presence of AD pathology due to practicality, cost, invasiveness and patient burden. Persons in rural communities and developing countries, where the majority of cases will occur in the next 50 years, will have limited access to these technologies. Furthermore, CSF and neuroimaging biomarkers will also not be ideal if repeated measures are needed to determine rate and progression of the pathology. For these purposes, the identification of a blood-based biomarker would be a giant leap forward as it would meet expert criteria of being “noninvasive, simple to perform and inexpensive” [5] and would be superior to more invasive and costly CSF or brain imaging markers. Specifically, a blood-based biomarker could be utilized as a routine screening tool; easily incorporated into patient care at the general practitioner level and in developing countries.

Despite the noted advantages, however, blood-based biomarkers for AD have often been criticized because it is difficult to ascertain direct links between peripheral markers and brain processes. Notably, it is also difficult to determine exactly how changes in CSF markers relate to brain cellular processes, but CSF markers are in closer proximity to the brain and transportation across the blood–brain barrier is not a concern. Several promising blood-based measures are under investigation as potential diagnostic and/or prognostic AD biomarkers [6]. More recently, multiple studies have suggested that peripheral sphingolipids could be promising candidate biomarkers of AD progression [7–10].

Sphingolipids, including ceramides, sphingomyelins (SMs), gangliosides and sphingosine-1-phosphate, have important structural roles in cell membranes. Importantly, ceramides and sphingosine-1-phosphate also function as second messengers to modulate a wide variety of signaling events including cellular differentiation and proliferation, apoptosis, cytokine production and synaptic plasticity. Lipidomic, metabolomic and targeted approaches have identified pathways and products of sphingolipid metabolism that are altered early in the course of AD and contribute to the neuropathological alterations associated with AD [7,11,12]. For example, multiple studies have demonstrated both direct and indirect connections between ceramides and amyloid-β production [13–16]. Additionally, ceramides and SMs strongly affect membrane dynamics and are critical components of specialized microdomains known as lipid membrane rafts, where the processing of amyloid precursor protein by β- and γ-secretases occurs. Initial translation of these findings to humans showed that ceramide and SM levels in brain tissue and CSF, as well as gene expression patterns of enzymes participating in the sphingolipid metabolism pathway, varied with AD severity [11–13,17,18].

In the first studies to examine peripheral sphingolipid levels, we also reported that blood ceramides varied by AD severity [8–10]. Furthermore, high ceramide levels, regardless of the stage of AD, were predictive of cognitive decline. Elevated levels of serum long-chain ceramides predicted memory impairment in a population-based study of community-dwelling, cognitively normal women [8]. Remarkably, no-one in the lowest tertile of ceramide C22:0 developed memory impairment over 9 years of follow-up. Among mild cognitive impairment cases, high levels of plasma ceramides, particularly C22:0 and C24:0, predicted memory decline and hippocampal volume loss [9]. Hippocampal volume loss is a well-known and well-characterized neuropathological feature of AD; this association suggests that peripheral sphingolipids can indeed have an impact on brain processes and contribute to neurodegeneration. Lastly, high plasma ceramide levels predicted faster rates of cognitive decline among AD patients [10]. Decreasing ceramide levels via sphingomyelinase inhibitors may slow the rate of progression among AD patients. In these studies, a targeted mass spectrometry (multiple reaction monitoring) approach was used to quantitatively measure individual sphingolipid species [9]. A recent study using a shotgun lipidomics approach confirmed our findings of specific alterations in blood ceramides in AD patients [7]. Thus, the findings are robust across different populations and methods.

How can peripheral sphingolipids best be utilized as AD biomarkers? Notably, in studies to date, there is a high degree of overlap in levels of blood ceramides between cognitively normal individuals and mild cognitive impairment and AD patients [7,9]. Thus, peripheral sphingolipids are not going to be a diagnostic marker of AD. Rather, the results suggest that they could be utilized as biomarkers of disease progression from presymptomatic stages to clinical AD. It is also possible that blood sphingolipids could be used for therapeutic purposes, to predict the progression of cognitive impairment for the purposes of trial enrichment, to predict drug response or to serve as surrogate markers in trials. Indeed, sphingomyelinase inhibitors are being investigated as potential AD treatments.

A critical next step in determining whether blood ceramides can be utilized as specific AD biomarkers will be to determine the relationship between blood and CSF ceramides, and blood ceramides and CSF amyloid-β and tau. If there is a strong correlation between blood and CSF ceramide levels, as well as blood ceramides and CSF amyloid-β or tau, blood ceramides can be utilized as a direct measure of brain ceramides and/or as indicators of AD pathology. However, if a direct correlation is not found, this does not mean that blood ceramides will not be helpful in the prognosis of AD. Despite the dearth of research directly relating SMs and ceramides to cognitive impairment and AD, several basic science and clinical studies have suggested the importance of ceramides in the development and progression of vascular conditions [19]. High levels of plasma SMs and ceramides have been associated with subclinical atherosclerosis [20,21]. Ceramides are also important in the development of insulin resistance and diabetes. All of these vascular outcomes are known to increase the risk of AD and also affect the rate of disease progression after an AD diagnosis. Thus, even if future research does not find that peripheral and brain ceramide levels directly correlate, this research will be important to identify other mechanisms, such as vascular pathways, by which ceramides can increase the risk of AD and affect the rate of disease progression. While there is still a lot of work to do, the current results suggest that peripheral sphingolipids do appear to be important in the pathophysiological process of AD and may be useful blood-based biomarkers of disease progression.