Skip to main content
Download PDF

Association and multimodal model of retinal and blood-based biomarkers for detection of preclinical Alzheimer’s disease

Alzheimer's Research & Therapy volume 17, Article number: 19 (2025) Cite this article

Abstract

Background

The potential diagnostic value of plasma amyloidogenic beta residue 42/40 ratio (Aβ42/Aβ40 ratio), neurofilament light (NfL), tau phosphorylated at threonine-181 (p-tau181), and threonine-217 (p-tau217) has been extensively discussed in the literature. We have also previously described the association between retinal biomarkers and preclinical Alzheimer’s disease (AD). The goal of this study was to evaluate the association, and a multimodal model of, retinal and plasma biomarkers for detection of preclinical AD.

Methods

We included 82 cognitively unimpaired (CU) participants (141 eyes; mean age: 67 years; range: 56–80) from the Atlas of Retinal Imaging in Alzheimer’s Study (ARIAS). Blood samples were assessed for concentrations of Aβ42/Aβ40 ratio, NfL, p-tau181, and p-tau217 (ALZpath, Inc.) using Single molecule array (SIMOA) technology. The Spectralis II system (Heidelberg Engineering) was used to acquire macular centered Spectral Domain Optical Coherence Tomography (SD-OCT) images for evaluation of putative retinal gliosis surface area and macular retinal nerve fiber layer (mRNFL) thickness. For all participants, correlations (adjusted for age and correlation between eyes) were assessed between retinal and blood-based biomarkers. A subgroup cohort of 57 eyes from 32 participants with recent Aβ positron emission tomography (PET) results, comprising 18 preclinical patients (Aβ PET + ve, 32 eyes) and 14 controls (Aβ PET -ve, 25 eyes) with a mean age of 69 vs. 66, p = 0.06, was included for the assessment of a multimodal model to distinguish between the two groups. For this subgroup cohort, receiver operating characteristic (ROC) analysis was performed to compare the multimodal model of retinal and plasma biomarkers vs. each biomarker alone to distinguish between the two groups.

Results

Significant correlation was found between putative retinal gliosis and p-tau217 in the univariate mixed model (β = 0.48, p = 0.007) but not for the other plasma biomarkers (p > 0.05). This positive correlation was also retained in the multivariate mixed model (β = 0.43, p = 0.022). The multimodal ROC model based on retinal (gliosis area, inner inferior RNFL thickness, inner superior RNFL thickness, and inner nasal RNFL thickness) and plasma biomarkers (p-tau217 and Aβ42/Aβ40 ratio) had an excellent AUC of 0.97 (95% CI = 0.93–1.01; p < 0.001) compared to unimodal models of retinal and plasma biomarkers.

Conclusions

Our analyses show the potential of integrating retinal and blood-based biomarkers for improved detection and screening of preclinical AD.

Background

Alzheimer’s disease (AD) is a leading cause of disability and poor health (morbidity) in older adults, ranking among the top 10 leading causes of death in America [1, 2]. It is characterized by a complex interplay of pathological processes that begin decades before progressive cognitive decline and neurological dysfunction [3,4,5,6,7,8]. The hallmark pathology, which is the accumulation of amyloid-beta (Aβ) and tau triggers neuroinflammation, synaptic dysfunction, and neurodegeneration that leads to dementia. Globally, approximately 416 million individuals, accounting for 22% of all people aged 50 and above live with AD, including dementia, mild cognitive impairment (MCI) due to AD, and preclinical AD [9]. In the United States alone, an estimated 6.7 million Americans aged 65 and older are impacted, with a projected increase to 13.8 million by 2060 [1].

Neurodiagnostic methods such as Aβ positron emission tomography (Aβ-PET), tau-PET, cerebrospinal fluid (CSF) assessments of Aβ42/Aβ40 ratio, and CSF tau phosphorylated at threonine-181 (p-tau181) and threonine-217 (p-tau217) are valuable for diagnosis [10], but their extensive use for screening is hindered by their cost and invasiveness. Recent advancements in biomarker research offer promising avenues for early detection and screening, potentially enabling interventions during the preclinical stages when disease-modifying therapies may be most effective. Amidst various biomarker modalities, blood-based biomarkers and retinal biomarkers have recently received significant attention for their accessibility, scalability, and non-invasive potential to reflect underlying neuropathological changes associated with AD.

Due to extensive research concerning blood-based assays for total tau (t-tau) and phosphorylated tau (p-tau) isoforms as potential biomarkers for AD, elevated levels of p-tau181 and p-tau217 in blood plasma has been identified to be correlated with preclinical AD [11,12,13]. Similarly, elevated levels of Aβ42 or altered Aβ42/Aβ40 ratio in blood plasma have also been associated with AD pathology [14]. In addition to Aβ peptides and tau tangles, neurofilament light (NfL), which are structural proteins found in the neurons are released into the bloodstream and have been proven to be indicative of neuronal damage and axonal degeneration [15]. Hence, plasma biomarkers such as p-tau181, p-tau217, Aβ42/Aβ40 ratio, and NfL have been associated with AD pathology and may serve as a marker of neurodegeneration and disease progression.

Retinal imaging techniques such as blue autofluorescence fundus imaging, Spectral Domain Optical Coherence Tomography (SD-OCT), and OCT Angiography (OCTA), serve as non-invasive methods for assessing proteinopathy, neurodegenerative, and vascular changes in the retina of AD patients, respectively. Changes in retinal thickness, particularly thinning of the retinal nerve fiber layer (RNFL) at the macular (mRNFL) and the peripapillary region (pRNFL) have been observed in AD patients compared to cognitively unimpaired (CU) individuals [16, 17]. Alterations in retinal vascular parameters, including vessel caliber, fractal dimension, vessel tortuosity, and capillary free zones have been associated with AD pathology [18,19,20]. Prior studies have also investigated the presence of AD-related biomolecules, such as Aβ plaques and tau protein using immunohistochemical staining and molecular imaging in postmortem retina of AD patients [21, 22]. Our previous study showed that the surface area of putative retinal gliosis observed in vivo using en face SD-OCT imaging was larger in preclinical AD patients compared to controls, suggesting putative retinal gliosis as a novel biomarker of AD-related neuroinflammation in the retina [23].

Individually, blood-based and retinal biomarkers have shown adequate promise in identifying individuals at risk of AD. However, their association and integration have not yet been explored in preclinical AD. This concept introduces a novel and synergistic approach to improve screening accuracy and predictive capability for early AD detection. The objective of this study was to assess the correlation and develop a multimodal model using retinal and plasma biomarkers for detecting preclinical AD.

Methods

Study participants

Participants were retrospectively included from the Atlas of Retinal Imaging in Alzheimer’s Study (ARIAS) study [24], which was conducted in accordance with the principles outlined in the Declaration of Helsinki and received approval from the BayCare Health System (St. Petersburg, FL) Institutional Review Board. Written informed consent was obtained from all participants before the collection of experimental data. Inclusion criteria for participants involved the absence of or controlled hypertension (blood pressure < 140/90), hyperlipidemia (total cholesterol ≤ 240 mg/dl), and systemic diabetes (HbA1c ≤ 7). Eyes were excluded if there were any associated retinal pathologies predisposing individuals to glial changes such as glaucoma, diabetic retinopathy, retinal ischemic conditions, epiretinal membrane (ERM), age-related macular degeneration, macular hole, or other retinal procedures including pars plana vitrectomy, laser photocoagulation, and internal limiting membrane (ILM) peeling. Additional exclusion criteria encompassed unstable doses of antidepressants with significant anticholinergic side effects, large cataracts impeding imaging, current intake of retinotoxic drugs (e.g., chloroquine, hydroxychloroquine, cancer drugs), and other neurodegenerative diseases such as Parkinson’s disease or Multiple Sclerosis. All eyes included in the study had best-corrected visual acuity of ≥ 20/40 (~ Log MAR 0.30) and refractive errors ≤ ± 3.00 DS (spherical equivalent) or equivalent axial lengths (range: 22–24 mm) to prevent significant differences in retinal magnification as per the Bennett formula [25].

A total of 82 CU participants (141 eyes; mean age: 67 years; range: 56–80) from ARIAS with plasma biomarker data available were involved in this study. Out of the 82 CU older adults, 32 participants had recent Aβ PET results available. Hence, a cohort of 18 preclinical AD participants (4 males and 14 females; 32 eyes; mean age: 69 years; range: 62–80; Table 1) and 14 control participants (4 males and 10 females; 25 eyes; mean age: 66 years; range: 58–77; N = 57 eyes; Table 1) were included in the multimodal model involving AD blood-based and retinal biomarkers. This is an extension of our previously powered study [23]. There were no significant differences in age (t(30) = -2.0, p = 0.06; Table 1) or sex proportion (X2 (1, N = 32) = 0.17, p = 0.68; Table 1) between the two groups. Preclinical AD participants were CU and Aβ PET positive, while controls were CU and Aβ PET negative. CU participants had Montreal Cognitive Assessment (MoCA) scores of ≥ 26 [26, 27], and Repeatable Battery for the Assessment of Neuropsychological Status Update (RBANS-U) Delayed Memory Index (DMI) scores of ≥ 85 [28, 29].

Table 1 Plasma and retinal biomarkers compared between preclinical Alzheimer’s disease and control older adults for the multimodal model cohort
Full size table

Cognitive assessment

All enrolled individuals underwent comprehensive neuropsychological assessment using the MoCA [26, 27] and the RBANS-U [28, 29]. The MoCA is a 30-point evaluation test typically administered within 10 min, with a score of 26 or higher indicative of CU status. The test evaluates various cognitive domains, such as short-term memory, visuospatial abilities, executive function, attention, concentration, working memory, language, and orientation to time and place. Participants’ total MoCA scores were recorded for our study. The RBANS-U, designed for adults aged 20 to 89 [27, 29], is a brief neuropsychological battery comprising 10 subtests that yield five domain scores: immediate memory (immediate list learning and immediate story), visuoperceptual abilities (figure copy and line orientation), language (naming and semantic fluency), attention (digit span forward and digit-symbol coding), and delayed memory (delayed list memory with recognition, delayed story memory, and delayed figure memory). Administration of the RBANS-U typically takes around 30 min. In our study, we assessed the RBANS-U DMI scores of the participants.

Aβ PET acquisition

Thirty-two participants had recent Aβ PET neuroimaging, and these results were made available to the ARIAS study team. This procedure involved administering an intravenous injection of 18 F-florbetapir at a dose of 370MBq (10 mCi +/− 10%). Approximately 50 min post-injection, a 20-minute PET scan was performed, accompanied by a head computed tomography (CT) scan for attenuation correction. The images were obtained using a 128 × 128 matrix and subsequently reconstructed using iterative or row action maximization likelihood algorithms. The PET standardized uptake value (SUV) data were aggregated and normalized against the SUV of the entire cerebellum, yielding a ratio termed SUV ratio (SUVr). An SUVr threshold of ≥ 1.1 was deemed indicative of Aβ positivity, while an SUVr threshold < 1.1 denoted Aβ negativity, consistent with our previous protocols [30]. These SUVr calculations were conducted using the MIMneuro software, and Aβ positivity or negativity was confirmed in all cases following review by a board-certified radiologist.

SD-OCT volume scan image acquisition

Prior to imaging, all participants were dilated with two drops of tropicamide (Mydriacyl 1%) per eye. After a 15-minute waiting period, high-resolution macular-centered 20 × 20-degree (~ 6 × 6 mm) SD-OCT images were captured for both eyes of each participant. These images consisted of 512 B-scans, 512 A-scans per B-scan, spaced 12 microns apart, with an average of 5 frames per B-scan location. The imaging was performed using the Spectralis HRA + OCT system (Eye Explorer version 1.10.4.0; Heidelberg Engineering, Heidelberg, Germany). Signal quality was maintained at a minimum value of 30 to ensure optimal image quality.

Macular centered RNFL thickness

High-resolution SD-OCT images, centered on the macular and measuring 30 × 25-degree (~ 8.8 × 7.4 mm) in size, were acquired using the Heidelberg Spectralis OCT system [24]. These images consisted of 61 B-scans, with a spacing of 123 microns between B-scans, and an average of 10 frames per B-scan location. The Spectralis segmentation software, integrated into the system, which has been previously validated [31] was employed to segment the RNFL, following established protocols [24]. Manual adjustments to the segmentation of retinal layers and centration of the Early Treatment Diabetic Retinopathy Study (ETDRS) [32] map was performed when necessary. Subsequently, the average thickness of the RNFL in each of the nine sectoral regions delineated on the ETDRS map [32] was computed using the vendor’s automated software.

Gliosis definition and computation

Putative retinal gliosis, characterized by white or hyperreflective patchy structures at the ILM/RNFL boundary, typically aligned with regions showing hyperreflectivity lines on OCT B-scans. Gliosis quantification was performed as discussed in our previous study [23]. Briefly, the surface area of putative retinal gliosis was quantified using three en face slabs derived from dense volume SD-OCT scans using vendor software that included: (a) macrophage like cell (MLC) layer, segmented at the ILM at the vitreoretinal interface; (b) NFL, segmented from the ILM to the RNFL/retinal ganglion cell layer boundary; and (c) superficial vascular plexus (SVP) layer, segmented from the ILM to the inner plexiform layer/inner nuclear layer boundary. Subsequently, the images were processed using custom MATLAB scripts as explained in our previous study [23]. Briefly, MLC and NFL slabs were averaged, followed by subtraction of the SVP layer from the averaged image. Spatial filtering and binarization were then performed, and the count of white pixels was utilized to calculate the surface area of putative retinal gliosis. This calculation involved converting the total number of white pixels into mm2 based on the micron-to-pixel ratio in the x and y directions, obtained from the fiducial marks of the vendor software [23].

Plasma sampling and analysis

Fasting blood samples including a total of 17.5 ml of blood (10 ml for plasma and 7.5 ml for serum) were collected at the study site. Plasma EDTA tubes were gently inverted 5–10 times and centrifuged with a horizontal rotor for 10 min at 2000 x g within one hour of collection. Following centrifugation, 1.0 mL aliquots of plasma were transferred into polypropylene (cryovial) tubes and stored at -80⁰ C freezer within 2 h of collection. The blood samples were sent to a central biorepository for analysis using the new ultrasensitive immunoassay platform called the single molecule array (SIMOA platform) technology by Quanterix. Blood samples were analyzed for plasma p-tau217 (ALZpath p-tau217 assay), Aβ42/40 ratio, p-tau181, and NfL.

Statistical analyses

Statistical analyses and data visualization were conducted using R version 4.3.1 software (R Foundation for Statistical Computing, Vienna, Austria). Descriptive statistics are presented as mean ± SD. Individual univariate linear mixed-effects models that employed maximum likelihood estimation, adjusting for correlation between fellow eyes were used for the analysis. These models included age as a covariate to investigate the association between the surface area of putative retinal gliosis and blood-based AD biomarkers (p-tau217, Aβ42/40 ratio, p-tau181, and NfL; Table 2). A similar joint multivariate mixed-effects model was also utilized to evaluate the correlation of putative retinal gliosis with blood-based biomarkers (Table 2). The equations for the models were as follows:

Table 2 Univariate and multivariate joint mixed models for predicting putative retinal gliosis from blood-based biomarkers
Full size table
$$\:\text{G}\text{l}\text{i}\text{o}\text{s}\text{i}\text{s}\:\sim\:\text{p}-\text{t}\text{a}\text{u}217\:+\:\text{a}\text{g}\text{e}\:+\:\left(1\right|\text{s}\text{u}\text{b}\text{j}\text{e}\text{c}\text{t}\_\text{I}\text{D})$$
(1)
$$\:\text{G}\text{l}\text{i}\text{o}\text{s}\text{i}\text{s}\:\sim\:\text{s}\text{c}\text{a}\text{l}\text{e}(\text{A}{\upbeta\:}42/40\:\text{r}\text{a}\text{t}\text{i}\text{o})\:+\:\text{a}\text{g}\text{e}\:+\:\left(1\right|\text{s}\text{u}\text{b}\text{j}\text{e}\text{c}\text{t}\_\text{I}\text{D})$$
(2)
$$\:\text{G}\text{l}\text{i}\text{o}\text{s}\text{i}\text{s}\:\sim\:\text{p}-\text{t}\text{a}\text{u}181\:+\:\text{a}\text{g}\text{e}\:+\:\left(1\right|\text{s}\text{u}\text{b}\text{j}\text{e}\text{c}\text{t}\_\text{I}\text{D})$$
(3)
$$\:\text{G}\text{l}\text{i}\text{o}\text{s}\text{i}\text{s}\:\sim\:\text{N}\text{f}\text{L}\:+\:\text{a}\text{g}\text{e}\:+\:\left(1\right|\text{s}\text{u}\text{b}\text{j}\text{e}\text{c}\text{t}\_\text{I}\text{D})$$
(4)
$$\:\text{G}\text{l}\text{i}\text{o}\text{s}\text{i}\text{s}\:\sim\:\text{p}-\text{t}\text{a}\text{u}217\:+\:\text{s}\text{c}\text{a}\text{l}\text{e}(\text{A}{\upbeta\:}42/40)\:+\:\text{p}-\text{t}\text{a}\text{u}181\:+\:\text{N}\text{f}\text{L}\:+\:\text{a}\text{g}\text{e}\:+\:\left(1\right|\text{s}\text{u}\text{b}\text{j}\text{e}\text{c}\text{t}\_\text{I}\text{D})$$
(5)

Assumption tests for regression supported the appropriateness of these analyses. Specifically, normality and linearity were not violated based on a sample size of 141 eyes with skewness and kurtosis ≤ ± 3.50 and scatterplot assessment, respectively. For the subgroup multimodal model analysis, individual logistic regression models were constructed to determine predictive probabilities. This was done for predicting preclinical AD based on individual retinal (putative retinal gliosis, inner inferior, inner superior, and inner nasal RNFL thicknesses) [23] and blood-based biomarkers (p-tau217, Aβ42/40 ratio, p-tau181, and NfL) separately. Furthermore, other logistic regression models involving different combinations of gliosis, inner inferior, inner superior, and inner nasal RNFL thicknesses, along with p-tau217 and Aβ42/40 ratio, were developed to obtain predictive probabilities. The inner inferior, inner superior, and inner nasal RNFL thicknesses were chosen for the model based on our previous findings due to their proximity to significance on comparing them between groups as established from our previous study [23]. Receiver operating characteristic (ROC) curves with and without accounting for correlation between eyes were generated from various logistic regression scenarios to assess sensitivity, specificity, and the area under the curve (AUC) for distinguishing between groups (Table 3). Cutoffs were established using the maximum Youden’s index. The following AUC categorization was used for the study: 0.5–0.6 = unsatisfactory, 0.6–0.7 = satisfactory, 0.7–0.8 = good, 0.8–0.9 = very good, and 0.9-1 = excellent. A Bonferroni corrected p-value of < 0.01 (α = 0.05/5) was considered significant for the mixed models and a Bonferroni corrected p-value of < 0.003 (0.05/17) was considered significant for the ROC curves.

Table 3 Area under the curve and 95% CIs for the retinal and blood-based biomarkers to distinguish between preclinical AD and controls
Full size table

Results

Association between putative retinal gliosis and AD blood-based biomarkers

Among the univariate mixed models predicting putative retinal gliosis from blood-based biomarkers, a significant positive association was found between plasma p-tau217 and the surface area of putative retinal gliosis, while adjusting for correlation between eyes and including age as a covariate (β = 0.48, p = 0.007; Table 2; Fig. 1). Greater levels of p-tau217 in blood plasma are associated with increased gliosis. Specifically, while controlling for age, one pg/ml increase in p-tau217 is associated with a corresponding increase of 0.48 mm2 in gliosis. However, no significant associations were observed for the other univariate models for putative retinal gliosis vs. plasma Aβ42/40 ratio, p-tau181, and NfL (Table 2; Fig. 1). Figure 1 illustrates the univariate associations between putative retinal gliosis and plasma p-tau217, Aβ42/40 ratio, p-tau181, and NfL.

Fig. 1
figure 1

Univariate associations between putative retinal gliosis and blood-based biomarkers with the regression line represented in blue. In these analyses, the association between putative retinal gliosis and p-tau217 was significant (A). Other associations were not significant (B-D)

Full size image

Consistent with the findings of the univariate mixed model, the multivariate mixed model also revealed a positive association between plasma p-tau217 and putative retinal gliosis (β = 0.43, p = 0.022; Table 2). Figure 2 depicts increasing surface area of putative retinal gliosis with increasing levels of plasma p-tau217.

Fig. 2
figure 2

Macrophage-like cell (MLC) layer en face slabs and the corresponding binarized image depicting the area of putative retinal gliosis across the range of p-tau217. Increasing surface area of putative retinal gliosis is observed with increasing levels of plasma p-tau217

Full size image

ROC models of retinal and blood-based biomarkers to distinguish between preclinical AD and controls

The ROC models involving individual blood-based biomarkers with and without accounting for correlation between eyes to distinguish between preclinical AD and controls yielded unsatisfactory AUC (p-tau181 = 0.59, 95% CI = 0.44–0.74, p = 0.24; Aβ42/40 = 0.54, 95% CI = 0.38–0.70, p = 0.67; NfL = 0.50, 95% CI = 0.34–0.65, p = 0.96) except for the p-tau217 (AUC = 0.87, 95% CI = 0.78–0.97, p < 0.001; Table 3; Fig. 3a).

Fig. 3
figure 3

Receiver operating characteristic (ROC) curves showing unimodal and multimodal models of gliosis and blood-based biomarkers to distinguish between preclinical AD and controls. The unimodal ROC model including plasma biomarkers yielded an unsatisfactory area under the curve except for p-tau217 (A). A multimodal model involving putative retinal gliosis, inner inferior, inner superior, and inner nasal RNFL thickness along with plasma p-tau217 and Aβ42/40 yielded the maximum area under the curve that was excellent and statistically significant, AUC = 0.97 (95% CI = 0.93–1.01), p < 0.001(B)

Full size image

Various multimodal model combinations with and without accounting for correlation between eyes were assessed for its predictive ability for preclinical AD (Table 3; Fig. 3), and it was found that a multimodal model involving putative retinal gliosis, inner inferior, inner superior, and inner nasal RNFL thickness along with the p-tau217 and Aβ42/40 ratio yielded the maximum AUC that was excellent and statistically significant, AUC = 0.97 (95% CI = 0.93–1.01), p < 0.001 (Table 3; Fig. 3b). A Youden’s index of 0.858 was chosen to create a cut-off predictive value of ≥ 0.38. The cut-off predictive value corresponded to a gliosis surface area of ≥ 1.73 mm2, inner inferior RNFL thickness of ≥ 24 μm, inner superior RNFL thickness of ≥ 24 μm, inner nasal RNFL thickness of ≥ 20 μm, p-tau217 of ≥ 0.187 pg/mL, and Aβ42/40 ratio of ≥ 0.048 for the preclinical AD positive state with a sensitivity of 93.8% and a specificity of 92%.

Discussion

Our present study investigated the association between blood-based biomarkers of AD and putative retinal gliosis, providing valuable insights into the potential utility of retinal imaging for early detection of AD. The key finding of this study was the significant association between putative retinal gliosis and plasma p-tau217 in the univariate model but not in the multimodal model. In contrast, no significant associations were observed between putative retinal gliosis and plasma Aβ42/40 ratio, p-tau181, or NfL in both the univariate and multivariate models. Importantly, our study also demonstrated that a multimodal model combining retinal measures and blood-based biomarkers significantly improves diagnostic accuracy. In the initial model with only plasma p-tau217, the AUC was found to be 0.87. With the addition of Aβ42/40 ratio, the AUC increased to 0.93.The final model that incorporated a combination of putative retinal gliosis, RNFL thickness measures (inner inferior, inner superior, and inner nasal), plasma p-tau217, and Aβ42/40 ratio yielded an excellent AUC of 0.97 (95% CI = 0.93–1.01, p < 0.001), with high sensitivity (93.8%) and specificity (92%) compared to other biomarker combinations. This multimodal approach leverages the strengths of both retinal imaging and blood-based biomarkers, providing a comprehensive methodology for early AD detection.

While amyloid aggregation and tau tangle formation are widely recognized as hallmark pathologies associated with AD, tau proteins are known to hyperphosphorylate at multiple sites prior to aggregation into neurofibrillary tangles (NFTs) in AD brains [33, 34]. Among the various tau isoforms, p-tau181, p-tau217, and p-tau231 present in CSF and plasma are widely studied as diagnostic biomarkers for Alzheimer’s disease [35,36,37,38]. Studies report that plasma p-tau181 is increased during preclinical AD and further increases at the MCI and dementia stages of AD [11, 12]. Furthermore, studies have also recently established the use of p-tau217 levels in CSF and blood that are elevated in individuals with early AD, including those who are asymptomatic or have MCI due to AD [13, 39, 40]. The presence of elevated plasma p-tau217 has also been strongly correlated with amyloid plaque deposition in the brain of CU individuals as determined using Aβ-PET imaging [41].

Although the precise mechanisms behind the early elevation of p-tau in plasma are still under investigation, it is believed that Aβ pathology often coincides or initiates tau phosphorylation and the release of soluble tau through various pathways before NFTs develop [42,43,44,45]. Moreover, studies have also suggested that neuroinflammation might be the connecting link between initial Aβ pathology and the subsequent development of NFTs [46,47,48]. This is further supported by other studies that show stronger correlation between plasma p-tau217 with Aβ-PET, and earlier abnormalities in plasma p-tau217 preceding tau-PET changes [41, 49, 50]. Thus, plasma levels of p-tau217 are significantly increased during the preclinical stages of AD [13, 39, 40]. Building on this, the significant positive correlation between putative retinal gliosis and plasma p-tau217 suggests retinal gliotic changes may represent an early striking effect during the disease process before the NFTs are formed. A longitudinal study involving retinal imaging, Aβ and p-tau blood-based biomarker analysis, and PET imaging will be needed to investigate the above hypothesis.

Interpreting amyloid pathology further, it is well established that amyloid precursor protein (APP), a large, glycosylated membrane protein is cleaved by β- and γ-secretase, primarily producing the 40-residue Aβ (Aβ40) and a smaller, more amyloidogenic 42-residue form (Aβ42) [51]. The Aβ42 residue is more prone to aggregate into fibrils and deposit with age [52], whereas Aβ42/40 ratio is often assessed for AD because it provides a more accurate reflection of Aβ pathology than the former alone that could be influenced by individual variability and assay differences [53]. Multiple studies have demonstrated the ability of plasma Aβ42/40 ratio in reflecting brain amyloidosis as assessed with amyloid-PET [54, 55]. However, in the current study, since we did not find significant association between putative retinal gliosis and plasma Aβ42/40 ratio, it could be likely that inflammation could be a key factor driving the amyloid pathology itself as hypothesized by many other studies [56,57,58]. Another interpretation could also be that plasma Aβ42/40 ratio is not significantly elevated in preclinical AD and hence the lack of association. Again, a longitudinal study involving retinal imaging, Aβ blood-based biomarker analysis, and PET imaging will be needed to investigate the above hypotheses.

NfL which is a component of the axonal cytoskeleton is found primarily in larger myelinated axons [59] and has been established to be a sensitive marker for neuronal damage and axonal degeneration. Various studies demonstrate significant cross-sectional and longitudinal correlations of plasma NfL levels with brain atrophy, hypometabolism, and cognitive decline [15, 60, 61]. Although our study did not find significant associations between putative retinal gliosis and NfL, the observed inverse trend further supports the theory that glial activity may be an early indicator of neuroinflammation, potentially preceding irreversible axonal degeneration and cognitive decline.

Neuroinflammation involves not only microglia but also astrocytes and other immune cells. Elevated levels of glial fibrillary acidic protein (GFAP), an astrocytic cytoskeletal protein indicative of astrogliosis, have been measured in blood samples and are associated with preclinical AD [62, 63]. Furthermore, some studies suggest that astrogliosis may precede microglial activation [56, 64]. Therefore, it is crucial to examine the association between putative retinal gliosis and GFAP to determine if putative retinal gliosis predominantly represents astrogliosis, especially given the lack of significant association with the Aβ42/40 ratio which would have indicated microglial activity. This represents the next line of research for current study.

A single retinal biomarker may not be sensitive or specific to AD. This is because these retinal biomarkers are also affected by other retinal diseases [18, 23]. Even though we clinically excluded comorbid retinal conditions, it is difficult to distinguish preclinical manifestations of these comorbid retinal conditions from retinal changes that occur in preclinical AD. For example, putative retinal gliosis is indicated in other retinal conditions that have neuroinflammatory components such as glaucoma [65,66,67,68], diabetic retinopathy [69,70,71], and ERM [72]. While the above retinal diseases were clinically excluded from our study, it is possible that some of the retinal changes observed could be preclinical manifestations of these conditions. This may explain the smaller area AUCs of the retinal biomarkers in this study (putative retinal gliosis and RNFL thickness) compared to plasma p-tau217. It is however imperative to mention that we have previously shown larger surface area of putative retinal gliosis in preclinical AD patients as defined by Aβ PET positivity [23]. Our current findings of positive association between the surface area of putative retinal gliosis vs. plasma p-tau217 in both the univariate and multivariate mixed models provides further evidence of a strong association between this in vivo novel biomarker of retinal neuroinflammation and preclinical AD.

Unlike the retina which is a direct extension of the central nervous system (CNS), blood-based biomarkers are derived from peripheral sources via a downstream effect [73, 74]. This can lead to contamination of results and influence of genetic factors, diurnal variation, and renal function [73, 74]. Some critical limitations in measuring blood-based biomarkers for AD involve the relatively low concentration of these markers due to the reduced permeability through the blood-brain barrier, which prevents the free passage of molecules from the CNS to the blood [75]. In addition to the low concentration, the biomarker of interest present in the blood plasma may undergo proteolytic degradation which could lead to false positive or negative results [76, 77]. The robustness of plasma p-tau biomarkers ranges from small to large effect size from preclinical to dementia due to AD respectively, with plasma p-tau217 having a better diagnostic ability for the former compared to other plasma p-tau isoforms such as p-tau181 and p-tau231 [78, 79], indicating that some plasma-based biomarkers may do well in later rather than early disease. Consistent with previous studies [78, 79], our study found better diagnostic ability for plasma p-tau217 for distinguishing preclinical AD from controls compared to plasma p-tau181, Aβ42/40 ratio, and NfL. While retinal biomarkers may have some advantages over plasma-based biomarkers and vice-versa, the goal of our study was to leverage the strengths of both biomarkers with the hypothesis that a multimodal model incorporating these two non-invasive/low-cost alternatives to PET and CSF will have a better diagnostic ability than each biomarker alone. Consistent with our hypothesis, a multimodal model involving putative retinal gliosis, inner inferior, inner superior, and inner nasal RNFL thickness along with the plasma p-tau217 an Aβ42/40 yielded an excellent AUC = 0.97 (95% CI = 0.93–1.01) with a sensitivity of 93.8% and a specificity of 92% compared to each of these biomarkers alone.

One of the major limitations of this study was primarily its cross-sectional nature which limits the ability to infer causal relationships between plasma-based biomarkers and retinal gliosis. Longitudinal studies in large sample size more diverse population are required to establish temporal relationships and causality. This represents the next line of research for our work.

Conclusions

This study underscores the potential of retinal imaging as a non-invasive, accessible screening tool for AD when used in conjunction with plasma p-tau217. Our study also showed a significant positive association between putative retinal neuroinflammation, and preclinical AD as defined by plasma p-tau217 beyond our previous findings with Aβ PET positivity. This could revolutionize early AD detection, facilitating earlier intervention, and improved disease management. Future research should focus on validating these findings in larger more diverse populations and exploring the longitudinal impact of retinal changes in AD progression. Additionally, understanding the biological mechanisms linking AD pathology to putative retinal gliosis could provide deeper insights into AD pathology and therapeutic targets.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Abbreviations

AD:

Alzheimer’s disease

Aβ:

Amyloid–beta

MCI:

Mild cognitive impairment

PET:

Positron emission tomography

CSF:

Cerebrospinal fluid

p–tau181:

Tau phosphorylated at threonine–181

p–tau217:

Tau phosphorylated at threonine–217

t–tau:

Total tau

NfL:

Neurofilament light

SD–OCT:

Spectral domain optical coherence tomography

OCTA:

Optical coherence tomography angiography

RNFL:

Retinal nerve fiber layer

mRNFL:

Macular retinal nerve fiber layer

pRNFL:

Peripapillary retinal nerve fiber layer

CU:

Cognitively unimpaired

ERM:

Epiretinal membrane

ILM:

Internal limiting membrane

ARIAS:

Atlas of retinal imaging in alzheimer’s study

MoCA:

Montreal cognitive assessment

RBANS–U:

Repeatable battery for the assessment of neuropsychological status update

DMI:

Delayed memory index

CT:

Computed tomography

SUVr:

Standardized uptake value ratio

MLC:

Macrophage like cell

SVP:

Superficial vascular plexus

SIMOA:

Single molecule array

ROC:

Receiver operating characteristic

AUC:

Area under the curve

NFTs:

Neurofibrillary tangles

APP:

Amyloid precursor protein

Aβ40:

40–residue Aβ

Aβ42:

42–residue Aβ

GFAP:

Glial fibrillary acidic protein

CNS:

Central nervous system

References

  1. 2023 Alzheimer’s disease facts and figures. Alzheimers Dement. 2023;19(4):1598–1695.

  2. US Burden of Disease Collaborators, Mokdad AH, Ballestros K, Echko M, Glenn S, Olsen HE, Mullany E, Lee A, et al. The state of US Health, 1990–2016: Burden of diseases, injuries, and risk factors among US States. JAMA. 2018;319(14):1444–72.

    Article  Google Scholar 

  3. La Rue A, Jarvik LF. Cognitive function and prediction of dementia in old age. Int J Aging Hum Dev. 1987;25(2):79–89.

    Article  CAS  PubMed  Google Scholar 

  4. Linn RT, Wolf PA, Bachman DL, Knoefel JE, Cobb JL, Belanger AJ, et al. The’preclinical phase’of probable Alzheimer’s disease: a 13-year prospective study of the Framingham cohort. Arch Neurol. 1995;52:485–90.

    Article  CAS  PubMed  Google Scholar 

  5. Braak H, Braak E. Frequency of stages of Alzheimer-related lesions in different age categories. Neurobiol Aging. 1997;18(4):351–7.

  6. Snowdon DA, Kemper SJ, Mortimer JA, Greiner LH, Wekstein DR, Markesbery WR. Linguistic ability in early life and cognitive function and Alzheimer’s disease in late life. Findings from the Nun Study. JAMA. 1996;275(7):528–32.

    Article  CAS  PubMed  Google Scholar 

  7. Elias MF, Beiser A, Wolf PA, Au R, White RF, D’Agostino RB. The preclinical phase of alzheimer disease: a 22-year prospective study of the Framingham Cohort. Arch Neurol. 2000;57(6):808–13.

    Article  CAS  PubMed  Google Scholar 

  8. Gordon BA, Blazey TM, Su Y, Hari-Raj A, Dincer A, Flores S, et al. Spatial patterns of neuroimaging biomarker change in individuals from families with autosomal dominant Alzheimer’s disease: a longitudinal study. Lancet Neurol. 2018;17(3):241–50.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Gustavsson A, Norton N, Fast T, Frölich L, Georges J, Holzapfel D, et al. Global estimates on the number of persons across the Alzheimer’s disease continuum. Alzheimers Dement. 2023;19(2):658–70.

    Article  PubMed  Google Scholar 

  10. Jack CR Jr, Bennett DA, Blennow K, Carrillo MC, Dunn B, Haeberlein SB, et al. NIA-AA Research Framework: toward a biological definition of Alzheimer’s disease. Alzheimers Dement. 2018;14(4):535–62.

    Article  PubMed  Google Scholar 

  11. Karikari TK, Pascoal TA, Ashton NJ, Janelidze S, Benedet AL, Rodriguez JL, et al. Blood phosphorylated tau 181 as a biomarker for Alzheimer’s disease: a diagnostic performance and prediction modelling study using data from four prospective cohorts. Lancet Neurol. 2020;19(5):422–33.

    Article  CAS  PubMed  Google Scholar 

  12. Janelidze S, Mattsson N, Palmqvist S, Smith R, Beach TG, Serrano GE, et al. Plasma P-tau181 in Alzheimer’s disease: relationship to other biomarkers, differential diagnosis, neuropathology and longitudinal progression to Alzheimer’s dementia. Nat Med. 2020;26(3):379–86.

  13. Ashton NJ, Brum WS, Di Molfetta G, Benedet AL, Arslan B, Jonaitis E, et al. Diagnostic accuracy of a plasma phosphorylated tau 217 immunoassay for Alzheimer Disease Pathology. JAMA Neurol. 2024;81(3):255–63.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Fandos N, Pérez-Grijalba V, Pesini P, Olmos S, Bossa M, Villemagne VL, et al. Plasma amyloid β 42/40 ratios as biomarkers for amyloid β cerebral deposition in cognitively normal individuals. Alzheimers Dement (Amst). 2017;8:179–87.

    Article  PubMed  Google Scholar 

  15. Zhou W, Zhang J, Ye F, Xu G, Su H, Su Y, et al. Alzheimer’s Disease Neuroimaging Initiative. Plasma neurofilament light chain levels in Alzheimer’s disease. Neurosci Lett. 2017;650:60–4.

    Article  CAS  PubMed  Google Scholar 

  16. Paquet C, Boissonnot M, Roger F, Dighiero P, Gil R, Hugon J. Abnormal retinal thickness in patients with mild cognitive impairment and Alzheimer’s disease. Neurosci Lett. 2007;420(2):97–9.

    Article  CAS  PubMed  Google Scholar 

  17. Kesler A, Vakhapova V, Korczyn AD, Naftaliev E, Neudorfer M. Retinal thickness in patients with mild cognitive impairment and Alzheimer’s disease. Clin Neurol Neurosurg. 2011;113(7):523–6.

    Article  PubMed  Google Scholar 

  18. Arthur E, Ravichandran S, Snyder PJ, Alber J, Strenger J, Bittner AK, et al. Retinal mid-peripheral capillary free zones are enlarged in cognitively unimpaired older adults at high risk for Alzheimer’s disease. Alzheimers Res Ther. 2023;15(1):172.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Cabrera DeBuc D, Somfai GM, Arthur E, Kostic M, Oropesa S, Mendoza Santiesteban C. Investigating Multimodal Diagnostic Eye biomarkers of cognitive impairment by measuring vascular and neurogenic changes in the Retina. Front Physiol. 2018;9:1721.

    Article  PubMed  PubMed Central  Google Scholar 

  20. Arthur E, Somfai GM, Kostic M, Oropesa S, Santiesteban CM, DeBuc DC. Distinguishing cognitive impairment by using singularity spectrum and lacunarity analysis of the retinal vascular network. Neurophotonics. 2019;6(4):041109.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Koronyo Y, Biggs D, Barron E, Boyer DS, Pearlman JA, Au WJ, et al. Retinal amyloid pathology and proof-of-concept imaging trial in Alzheimer’s disease. JCI Insight. 2017;2(16):e93621.

    Article  PubMed  PubMed Central  Google Scholar 

  22. den Haan J, Morrema THJ, Verbraak FD, de Boer JF, Scheltens P, Rozemuller AJ, et al. Amyloid-beta and phosphorylated tau in post-mortem Alzheimer’s disease retinas. Acta Neuropathol Commun. 2018;6(1):147.

    Article  Google Scholar 

  23. Ravichandran S, Snyder PJ, Alber J, Kenny MR, Rothstein A, Brown K, et al. Quantifying putative retinal gliosis in preclinical Alzheimer’s Disease. Invest Ophthalmol Vis Sci. 2024;65(5):5.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Alber J, Arthur E, Sinoff S, DeBuc DC, Chew EY, Douquette L, et al. A recommended minimum data set framework for SD-OCT retinal image acquisition and analysis from the Atlas of Retinal Imaging in Alzheimer’s study (ARIAS). Alzheimers Dement (Amst). 2020;12(1):e12119.

    PubMed  Google Scholar 

  25. Bennett AG, Rudnicka AR, Edgar DF. Improvements on Littmann’s method of determining the size of retinal features by fundus photography. Graefes Arch Clin Exp Ophthalmol. 1994;232(6):361–7.

    Article  CAS  PubMed  Google Scholar 

  26. Smith T, Gildeh N, Holmes C. The Montreal Cognitive Assessment: validity and utility in a memory clinic setting. Can J Psychiatry. 2007;52(5):329–32.

    Article  PubMed  Google Scholar 

  27. Nasreddine ZS, Phillips NA, Bédirian V, Charbonneau S, Whitehead V, Collin I, et al. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc. 2005;53(4):695–9.

    Article  PubMed  Google Scholar 

  28. Karantzoulis S, Novitski J, Gold M, Randolph C. The repeatable battery for the Assessment of Neuropsychological Status (RBANS): utility in detection and characterization of mild cognitive impairment due to Alzheimer’s disease. Arch Clin Neuropsychol. 2013;28(8):837–44.

    Article  PubMed  Google Scholar 

  29. Randolph C, Tierney MC, Mohr E, Chase TN. The repeatable battery for the Assessment of Neuropsychological Status (RBANS): preliminary clinical validity. J Clin Exp Neuropsychol. 1998;20(3):310–9.

    Article  CAS  PubMed  Google Scholar 

  30. Snyder PJ, Johnson LN, Lim YY, Santos CY, Alber J, Maruff P, et al. Nonvascular retinal imaging markers of preclinical Alzheimer’s disease. Alzheimers Dement (Amst). 2016;4:169–78.

    Article  PubMed  Google Scholar 

  31. Wong BM, Cheng RW, Mandelcorn ED, et al. Validation of optical coherence tomography retinal segmentation in neurodegenerative disease. Transl Vis Sci Technol. 2019;8(5):6–6.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Grading diabetic retinopathy. From stereoscopic color fundus photographs–an extension of the modified Airlie House classification. ETDRS report number 10. Early Treatment Diabetic Retinopathy Study Research Group. Ophthalmology. 1991;98(5):786–806.

    Article  Google Scholar 

  33. Grundke-Iqbal I, Iqbal K, Tung YC, Quinlan M, Wisniewski HM, Binder LI. Abnormal phosphorylation of the microtubule-associated protein tau (tau) in Alzheimer cytoskeletal pathology. Proc Natl Acad Sci U S A. 1986;83(13):4913–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. Ihara Y, Nukina N, Miura R, Ogawara M. Phosphorylated tau protein is integrated into paired helical filaments in Alzheimer’s disease. J Biochem. 1986;99(6):1807–10.

    Article  CAS  PubMed  Google Scholar 

  35. Hansson O. Biomarkers for neurodegenerative diseases. Nat Med. 2021;27(6):954–63.

    Article  CAS  PubMed  Google Scholar 

  36. Leuzy A, Mattsson-Carlgren N, Palmqvist S, Janelidze S, Dage JL, Hansson O. Blood-based biomarkers for Alzheimer’s disease. EMBO Mol Med. 2022;14(1):e14408.

    Article  CAS  PubMed  Google Scholar 

  37. Teunissen CE, Verberk IMW, Thijssen EH, Vermunt L, Hansson O, Zetterberg H, et al. Blood-based biomarkers for Alzheimer’s disease: towards clinical implementation. Lancet Neurol. 2022;21(1):66–77.

    Article  CAS  PubMed  Google Scholar 

  38. Suárez-Calvet M, Karikari TK, Ashton NJ, Lantero Rodríguez J, Milà-Alomà M, Gispert JD, et al. ALFA Study. Novel tau biomarkers phosphorylated at T181, T217 or T231 rise in the initial stages of the preclinical Alzheimer’s continuum when only subtle changes in Aβ pathology are detected. EMBO Mol Med. 2020;12(12):e12921.

    Article  PubMed  PubMed Central  Google Scholar 

  39. Jonaitis EM, Janelidze S, Cody KA, Langhough R, Du L, Chin NA, et al. Plasma phosphorylated tau 217 in preclinical Alzheimer’s disease. Brain Commun. 2023;5(2):fcad057.

    Article  PubMed  PubMed Central  Google Scholar 

  40. Mattsson-Carlgren N, Janelidze S, Palmqvist S, Cullen N, Svenningsson AL, Strandberg O, et al. Longitudinal plasma p-tau217 is increased in early stages of Alzheimer’s disease. Brain. 2020;143(11):3234–41.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Therriault J, Vermeiren M, Servaes S, Tissot C, Ashton NJ, Benedet AL, et al. Association of phosphorylated tau biomarkers with amyloid Positron Emission Tomography vs Tau Positron Emission Tomography. JAMA Neurol. 2023;80(2):188–99.

    Article  PubMed  Google Scholar 

  42. Busciglio J, Lorenzo A, Yeh J, Yankner BA. beta-amyloid fibrils induce tau phosphorylation and loss of microtubule binding. Neuron. 1995;14(4):879–88.

    Article  CAS  PubMed  Google Scholar 

  43. Ferreira A, Lu Q, Orecchio L, Kosik KS. Selective phosphorylation of adult tau isoforms in mature hippocampal neurons exposed to fibrillar A beta. Mol Cell Neurosci. 1997;9(3):220–34.

    Article  CAS  PubMed  Google Scholar 

  44. Zheng WH, Bastianetto S, Mennicken F, Ma W, Kar S. Amyloid beta peptide induces tau phosphorylation and loss of cholinergic neurons in rat primary septal cultures. Neuroscience. 2002;115(1):201–11.

    Article  CAS  PubMed  Google Scholar 

  45. De Felice FG, Wu D, Lambert MP, Fernandez SJ, Velasco PT, Lacor PN, et al. Alzheimer’s disease-type neuronal tau hyperphosphorylation induced by a beta oligomers. Neurobiol Aging. 2008;29(9):1334–47.

    Article  PubMed  Google Scholar 

  46. Garwood CJ, Pooler AM, Atherton J, Hanger DP, Noble W. Astrocytes are important mediators of Aβ-induced neurotoxicity and tau phosphorylation in primary culture. Cell Death Dis. 2011;2(6):e167.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Kitazawa M, Yamasaki TR, LaFerla FM. Microglia as a potential bridge between the amyloid beta-peptide and tau. Ann N Y Acad Sci. 2004;1035:85–103.

    Article  CAS  PubMed  Google Scholar 

  48. Nisbet RM, Polanco JC, Ittner LM, Götz J. Tau aggregation and its interplay with amyloid-β. Acta Neuropathol. 2015;129(2):207–20.

    Article  CAS  PubMed  Google Scholar 

  49. Janelidze S, Berron D, Smith R, Strandberg O, Proctor NK, Dage JL, et al. Associations of plasma Phospho-Tau217 levels with tau Positron Emission Tomography in Early Alzheimer Disease. JAMA Neurol. 2021;78(2):149–56.

    Article  PubMed  Google Scholar 

  50. Mattsson-Carlgren N, Andersson E, Janelidze S, Ossenkoppele R, Insel P, Strandberg O, et al. Aβ deposition is associated with increases in soluble and phosphorylated tau that precede a positive tau PET in Alzheimer’s disease. Sci Adv. 2020;6(16):eaaz2387.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  51. Haass C, De Strooper B. The presenilins in Alzheimer’s disease–proteolysis holds the key. Science. 1999;286(5441):916–9.

    Article  CAS  PubMed  Google Scholar 

  52. Cirrito JR, May PC, O’Dell MA, Taylor JW, Parsadanian M, Cramer JW, et al. In vivo assessment of brain interstitial fluid with microdialysis reveals plaque-associated changes in amyloid-beta metabolism and half-life. J Neurosci. 2003;23(26):8844–53.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  53. Lehmann S, Delaby C, Boursier G, Catteau C, Ginestet N, Tiers L, et al. Relevance of Aβ42/40 ratio for detection of Alzheimer Disease Pathology in Clinical Routine: the PLMR Scale. Front Aging Neurosci. 2018;10:138.

    Article  PubMed  PubMed Central  Google Scholar 

  54. Li Y, Schindler SE, Bollinger JG, Ovod V, Mawuenyega KG, Weiner MW, et al. Validation of plasma Amyloid-β 42/40 for detecting Alzheimer Disease amyloid plaques. Neurology. 2022;98(7):e688–99.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  55. Doecke JD, Pérez-Grijalba V, Fandos N, Fowler C, Villemagne VL, Masters CL, AIBL Research Group, et al. Total Aβ42/Aβ40 ratio in plasma predicts amyloid-PET status, independent of clinical AD diagnosis. Neurology. 2020;94(15):e1580–91.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  56. Bjorkli C, Sandvig A, Sandvig I. Bridging the gap between fluid biomarkers for Alzheimer’s Disease, Model systems, and patients. Front Aging Neurosci. 2020;12:272.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  57. Kinney JW, Bemiller SM, Murtishaw AS, Leisgang AM, Salazar AM, Lamb BT. Inflammation as a central mechanism in Alzheimer’s disease. Alzheimers Dement (N Y). 2018;4:575–90.

    Article  PubMed  Google Scholar 

  58. Calsolaro V, Edison P. Neuroinflammation in Alzheimer’s disease: current evidence and future directions. Alzheimers Dement. 2016;12(6):719–32.

    Article  PubMed  Google Scholar 

  59. Schlaepfer WW, Lynch RG. Immunofluorescence studies of neurofilaments in the rat and human peripheral and central nervous system. J Cell Biol. 1977;74(1):241–50.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  60. Mattsson N, Andreasson U, Zetterberg H, Blennow K. Alzheimer’s Disease Neuroimaging Initiative. Association of plasma neurofilament light with neurodegeneration in patients with Alzheimer Disease. JAMA Neurol. 2017;74(5):557–66.

    Article  PubMed  PubMed Central  Google Scholar 

  61. Mattsson N, Cullen NC, Andreasson U, Zetterberg H, Blennow K. Association between Longitudinal plasma neurofilament light and neurodegeneration in patients with Alzheimer Disease. JAMA Neurol. 2019;76(7):791–9.

    Article  PubMed  PubMed Central  Google Scholar 

  62. Pereira JB, Janelidze S, Smith R, Mattsson-Carlgren N, Palmqvist S, Teunissen CE, et al. Plasma GFAP is an early marker of amyloid-β but not tau pathology in Alzheimer’s disease. Brain. 2021;144(11):3505–16.

    Article  PubMed  PubMed Central  Google Scholar 

  63. Chatterjee P, Pedrini S, Stoops E, Goozee K, Villemagne VL, Asih PR, et al. Plasma glial fibrillary acidic protein is elevated in cognitively normal older adults at risk of Alzheimer’s disease. Transl Psychiatry. 2021;11(1):27.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  64. Kumar A, Fontana IC, Nordberg A. Reactive astrogliosis: a friend or foe in the pathogenesis of Alzheimer’s disease. J Neurochem. 2023;164(3):309–24.

    Article  CAS  PubMed  Google Scholar 

  65. Ashimatey BS, King BJ, Swanson WH. Retinal putative glial alterations: implication for glaucoma care. Ophthalmic Physiol Opt. 2018;38(1):56–65.

    Article  PubMed  Google Scholar 

  66. Grieshaber MC, Orgul S, Schoetzau A, Flammer J. Relationship between retinal glial cell activation in glaucoma and vascular dysregulation. J Glaucoma. 2007;16(2):215–9.

    Article  PubMed  Google Scholar 

  67. Grieshaber MC, Moramarco F, Schoetzau A, Flammer J, Orguel S. Detection of retinal glial cell activation in glaucoma by time domain optical coherence tomography. Klin Monbl Augenheilkd. 2012;229(4):314–8.

    Article  CAS  PubMed  Google Scholar 

  68. Cheung H, King BJ, Gast TJ. Presumed activated retinal astrocytes and Müller cells in healthy and glaucomatous eyes detected by spectral domain optical coherence tomography. Ophthalmic Physiol Opt. 2020;40(6):738–51.

    Article  PubMed  PubMed Central  Google Scholar 

  69. Ong JX, Nesper PL, Fawzi AA, Wang JM, Lavine JA. Macrophage-like cell density is increased in proliferative Diabetic Retinopathy characterized by Optical Coherence Tomography Angiography. Invest Ophthalmol Vis Sci. 2021;62(10):2.

    Article  PubMed  PubMed Central  Google Scholar 

  70. Zhang NT, Nesper PL, Ong JX, Wang JM, Fawzi AA, Lavine JA. Macrophage-like cells are increased in patients with vision-threatening Diabetic Retinopathy and correlate with Macular Edema. Diagnostics (Basel). 2022;12(11):2793.

    Article  CAS  PubMed  Google Scholar 

  71. Wang W, Sun G, Xu A, Chen C. Proliferative diabetic retinopathy and diabetic macular edema are two factors that increase macrophage-like cell density characterized by en face optical coherence tomography. BMC Ophthalmol. 2023;23(1):46.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  72. Grondin C, Au A, Wang D, et al. Identification and characterization of Epivascular Glia using En face Optical Coherence Tomography. Am J Ophthalmol. 2021;229:108–19.

    Article  PubMed  Google Scholar 

  73. Galasko D, Golde TE. Biomarkers for Alzheimer’s disease in plasma, serum and blood-conceptual and practical problems. Alzheimers Res Ther. 2013;5:10.

    Article  PubMed  PubMed Central  Google Scholar 

  74. Roher AE, Esh CL, Kokjohn TA, Castaño EM, Van Vickle GD, Kalback WM, et al. Amyloid beta peptides in human plasma and tissues and their significance for Alzheimer’s disease. Alzheimers Dement. 2009;5:18–29.

    Article  CAS  PubMed  Google Scholar 

  75. Hampel H, O’Bryant SE, Molinuevo JL, Zetterberg H, Masters CL, Lista S, et al. Blood-based biomarkers for Alzheimer disease: mapping the road to the clinic. Nat Rev Neurol. 2018;14:639–52.

    Article  PubMed  PubMed Central  Google Scholar 

  76. Zetterberg H, Burnham SC. Blood-based molecular biomarkers for Alzheimer’s disease. Mol Brain. 2019;12:1–7.

    Article  CAS  Google Scholar 

  77. Zetterberg H. Blood-based biomarkers for Alzheimer’s disease—An update. J Neurosci Methods. 2019;319:2–6.

    Article  CAS  PubMed  Google Scholar 

  78. Mielke MM, Frank RD, Dage JL, Jeromin A, Ashton NJ, Blennow K, et al. Comparison of plasma phosphorylated tau species with amyloid and tau positron emission tomography, neurodegeneration, vascular pathology, and cognitive outcomes. JAMA Neurol. 2021;78:1108–17.

    Article  PubMed  Google Scholar 

  79. Ashton NJ, Janelidze S, Mattsson-Carlgren N, Binette AP, Strandberg O, Brum WS, et al. Differential roles of Aβ42/40, p-tau231 and p-tau217 for Alzheimer’s trial selection and disease monitoring. Nat Med. 2022;28:2555–62.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

The authors wish to acknowledge the ARIAS participants who were extremely generous with their time for this study to help advance research into the early detection of AD.

Funding

This study is part of the Atlas of Retinal Imaging in Alzheimer’s Study (ARIAS) supported by a generous grant from the Morton Plant Mease Health Care Foundation (Clearwater, FL, U.S.A.) to PJS and NIH/NIA R21AG079794 to EA. JA was supported by NIH/NIA R01AG079241, R21AG074153, and the Warren Alpert Foundation. PJS served as principal investigator for ARIAS.

Author information

Author notes

  1. Swetha Ravichandran and Peter J. Snyder contributed equally as first authors.

Authors and Affiliations

  1. School of Optometry, University of Alabama at Birmingham, Birmingham, AL, US

    Swetha Ravichandran & Edmund Arthur

  2. Department of Neurology, Alpert Medical School of Brown University, Providence, RI, US

    Peter J. Snyder

  3. Department of Biomedical and Pharmaceutical Sciences, University of Rhode Island, Kingston, RI, US

    Peter J. Snyder & Jessica Alber

  4. George and Anne Ryan Institute for Neuroscience, University of Rhode Island, Kingston, RI, US

    Jessica Alber

  5. Butler Hospital Memory & Aging Program, Providence, RI, US

    Jessica Alber

  6. Alzheimer’s Disease Research Center, Department of Neurology, University of Alabama at Birmingham, Birmingham, AL, US

    Charles F. Murchison

  7. Department of Biostatistics, University of Alabama at Birmingham, Birmingham, AL, US

    Charles F. Murchison

  8. Atlantic Biomarkers, LLC, Alachua, FL, US

    Lauren E. Chaby & Andreas Jeromin

Authors
  1. Swetha Ravichandran
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Peter J. Snyder
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Jessica Alber
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Charles F. Murchison
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Lauren E. Chaby
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Andreas Jeromin
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Edmund Arthur
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

EA conceptualized and formulated the initial hypothesis for the study. EA, PJS, and JA designed the study. EA and SR provided software, performed image analysis, as well as interpreted the results. EA, SR, and CFM performed statistical analysis and interpreted output. EA and SR wrote the original draft of the manuscript. EA, PJS, and JA collected data. PJS, JA, LEC, and AJ provided assays for plasma biomarker analysis. PJS, JA, LEC, AJ, and CFM revised the manuscript and provided detailed comments. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Edmund Arthur.

Ethics declarations

Ethics approval and consent to participate

The study adhered to the tenets of the Declaration of Helsinki, and informed consent from all subjects was obtained prior to experimental data collection after explanation of the nature and possible consequences of the study. The study was part of the Atlas of Retinal Imaging in Alzheimer’s Study (ARIAS; PJS served as principal investigator for ARIAS) which took place at the University of Rhode Island and Butler Hospital Memory and Aging Program, Providence, RI between 2020 and 2022, and was approved by the BayCare Institutional Review Board (IRB).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ravichandran, S., Snyder, P.J., Alber, J. et al. Association and multimodal model of retinal and blood-based biomarkers for detection of preclinical Alzheimer’s disease. Alz Res Therapy 17, 19 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13195-024-01668-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13195-024-01668-5

Keywords

Alzheimer's Research & Therapy

ISSN: 1758-9193