The Impact of Kidney Function on Alzheimer’s Disease Blood Biomarkers: Implications for Predicting Amyloid-β Positivity

doi:10.21203/rs.3.rs-5180714/v1

Background

Impaired kidney function has a potential confounding effect on blood biomarker levels, including biomarkers for Alzheimer’s disease (AD). Given the imminent use of certain blood biomarkers in the routine diagnostic work-up of patients with suspected AD, knowledge on the potential impact of comorbidities on the utility of blood biomarkers is important. We aimed to evaluate the association between kidney function, assessed through estimated glomerular filtration rate (eGFR) calculated from plasma creatinine and AD blood biomarkers, as well as their influence over predicting Aβ-positivity.

Methods

We included 242 participants from the Translational Biomarkers in Aging and Dementia (TRIAD) cohort, comprising cognitively unimpaired individuals (CU; n = 124), mild cognitive impairment (MCI; n = 58), AD dementia (n = 34), and non-AD dementia (n = 26) patients all characterized by [¹⁸F] AZD-4694. Plasma samples were analyzed for Aβ42, Aβ40, glial fibrillary acidic protein (GFAP), neurofilament light chain (NfL), tau phosphorylated at threonine 181 (p-tau181), 217 (p-tau217), 231 (p-tau231) and N-terminal containing tau fragments (NTA-tau) using Simoa technology. Kidney function was assessed by eGFR in mL/min/1.73 m², based on plasma creatinine levels, age, and sex. Participants were also stratified according to their eGFR-indexed stages of chronic kidney disease (CKD). We evaluated the association between eGFR and blood biomarker levels with linear models and assessed whether eGFR provided added predictive value to determine Aβ-positivity with logistic regression models.

Results

Biomarker concentrations were highest in individuals with CKD stage 3, followed by stages 2 and 1, but differences were only significant for NfL, Aβ42, and Aβ40 (not Aβ42/Aβ40). All investigated biomarkers showed significant associations with eGFR except plasma NTA-tau, with stronger relationships observed for Aβ40 and NfL. However, after adjusting for either age, sex or Aβ-PET SUVr, the association with eGFR was no longer significant for all biomarkers except Aβ40, Aβ42, NfL, and GFAP. When evaluating whether accounting for kidney function could lead to improved prediction of Aβ-positivity, we observed no improvements in model fit (Akaike Information Criterion, AIC) or in discriminative performance (AUC) by adding eGFR to a base model including each plasma biomarker, age, and sex. While covariates like age and sex improved model fit, eGFR contributed minimally, and there were no significant differences in clinical discrimination based on AUC values.

Conclusions

We found that kidney function seems to be associated with AD blood biomarker concentrations. However, these associations did not remain significant after adjusting for age and sex, except for Aβ40, Aβ42, NfL, and GFAP. While covariates such as age and sex improved prediction of Aβ-positivity, including eGFR in the models did not lead to improved prediction for any biomarker. Our findings indicate that renal function does not seem to have a clinically relevant impact when using highly accurate blood biomarkers, such as p-tau217, in a biomarker-supported diagnosis.

eGFR

kidney impairment

amyloid

p-tau

NfL

GFAP

The quest to identify reliable biomarkers for early detection of Alzheimer’s disease (AD) and monitoring disease progression has gained significant momentum in recent years (1). Established AD biomarkers included positron emission tomography (PET) (2–4) and cerebrospinal fluid (CSF) measurements (2, 5, 6), which reflect the neuropathological hallmarks of AD. Blood-based biomarkers that are widely validated against CSF and imaging measures hold great promise for enabling early intervention and improved patient management (7). Certain amyloid-beta (Aβ) plasma assays demonstrate good agreement with Aβ-PET, but their small reduction in Aβ-positive compared to Aβ-negative patients, peripheral expression and confounding by medication use may limit their utility (8, 9). Phosphorylated tau (p-tau) biomarkers, detectable through various assays (10) and epitopes, such as p-tau181 (11), p-tau217 (12), and p-tau231 (13), are increased in Aβ-positive patients and their levels increase even further with advanced clinical stages and evident tau pathology (14). Unlike p-tau, N-terminal containing tau fragments (NTA-tau) (15, 16) measures soluble tau fragments regardless of their phosphorylation state, which increase during mid-to-late AD and are tightly associated with tau pathology. In contrast, plasma neurofilament light (NfL) acts as a general marker of neuro-axonal degeneration, which occurs across several neurodegenerative diseases (17), whereas plasma glial fibrillary acidic protein (GFAP) is elevated early in the course of AD, showing an association with Aβ pathology (18) but also increased on other neurogenerative disorders.

In anticipation of forthcoming disease-modifying therapies (DMT) across the globe, such as lecanemab (19) and donanemab (20), and prior to the integration of blood-based biomarkers into clinical practice, it is imperative to delineate the confounding factors that may exert a substantial influence on blood biomarker measurements (21). As the field of AD biomarker research progresses, it becomes increasingly important to account for such factors that could impact the accuracy and interpretation of fluid biomarkers, particularly p-tau217, which is anticipated to be the most useful in this context. One such confounding factor that has garnered attention is impaired kidney function (22). The kidneys play a pivotal role in the clearance of waste products and various substances from the bloodstream, including promising AD biomarkers. Impaired kidney function, as reflected by low estimated glomerular filtration rate (eGFR), can disrupt the equilibrium of substances in the bloodstream, potentially affecting the concentrations of analytes measured in peripheral tissues such as blood. For proteins such as β2-microglobulin (B2M) (23) and immunoglobulin light chains (free light chains) (24), which are routinely measured in clinical chemistry laboratories, their levels can change due to impaired kidney function and decreased clearance. Given this, it is relevant to investigate the extent to which kidney impairment influences the potential clinical utility of AD blood biomarkers.

It has been previously shown that various medical comorbidities, including chronic kidney disease (CKD), are associated with plasma levels of AD biomarkers, such as Aβ, p-tau species and NfL (22, 25, 26). However, many studies have lacked data on eGFR (22, 26, 27), a standardized measure of kidney function (28), and most have focused on only a narrow selection of plasma AD biomarkers (29–31). Studying the relationship between eGFR and AD biomarker concentrations, along with evaluating the potential enhancement of amyloid prediction models through the incorporation of eGFR, is essential. This could provide valuable insights on whether kidney function has clinically relevant implications for the incorporation of these biomarkers in the routine diagnostic work-up of AD and for the inclusion of individuals with preclinical AD in clinical trials.

In this study, we aimed to investigate the impact of kidney impairment on a broad range of AD plasma biomarkers, including Aβ40, Aβ42, GFAP, NfL, p-tau181, p-tau217, p-tau231, and NTA-tau. We first began by examining the concentrations of these biomarkers across different stages of CKD. Next, we incorporated eGFR into multiple regression models to examine how these biomarkers associate with eGFR. Finally, we assessed whether including eGFR in clinical prediction models to determine Aβ-positivity would lead to enhanced model fit or higher discriminative ability.

Participants & ethics

This research involved participants from the Translational Biomarkers in Aging and Dementia (TRIAD) observational cohort, which was designed to represent the AD continuum. Clinical diagnoses were conducted with no prior knowledge of the biomarker results. All subjects underwent clinical evaluations, which included the Clinical Dementia Rating (CDR) (32), Mini-Mental State Examination (MMSE) (33), and an assessment of cerebrovascular disease risk using the Hachinski Ischemic Scale (34). Participants were excluded from the study if they had uncontrolled systemic conditions despite being on a stable medication regimen. Additional exclusion criteria included ongoing substance abuse, recent head trauma, recent major surgery, or contraindications for magnetic resonance imaging (MRI)/PET safety. All participants provided informed consent, and the study protocols were approved by the appropriate ethical review boards. TRIAD was approved by the Montreal Neurological Institute PET working committee and the Douglas Mental Health University Institute Research Ethics Board (IUSMD16-61, IUSMD16-60). A detailed description of the cohort can be found in the supplementary material.

Imaging

Within the TRIAD cohort, Aβ-PET scans were conducted using [18F]-AZD4694. Amyloid positivity (referred to as A+) was determined when the [18F]-AZD-4694 standardized uptake value ratio (SUVR) exceeded 1.55. For participants in TRIAD, the meantime intervals between the initial plasma sample collection and the PET scans were approximately 0.44 years for Aβ-PET. Detailed imaging methods can be found in the supplementary methods.

Measurement of plasma creatinine

Blood samples were collected according to protocols that have been previously established and described (11). Plasma creatinine levels were determined using the Roche Cobas chemistry analyzer (Roche Diagnostics in Indianapolis, IN) at the Department of Psychiatry and Neurochemistry, University of Gothenburg. The analysis was conducted through the enzymatic assay known as creatinine PlusVer.2 (REF 03263991), which is designed by Roche and operated on the c501 platform. Detailed method can be found in the Supplementary Methods.

Calculation of estimated Glomerular Filtration Rate

Glomerular barrier function was evaluated by eGFR (mL/min/1.73 m²) using the plasma creatinine, age, and gender (28). CKD has been categorized into five stages in accordance with the 2012 Guidelines set forth by the Kidney Disease: Improving Global Outcomes (KDIGO-CKD) organization (35): CKD-stage 1, characterized by eGFR ≥ 90 mL/min/1.73 m²; CKD-stage 2, with eGFR ranging from 60-89 mL/min/1.73 m²; CKD-stage 3, encompassing eGFR levels of 30-59 mL/min/1.73 m²; CKD-stage 4, involving eGFR values within the range of 15-29 mL/min/1.73 m²; and CKD-stage 5, defined by eGFR < 15 mL/min/1.73 m². Due to the limited number of patients in CKD stages 4 and 5, participants in this cohort were grouped into three categories based on their eGFR: CKD Stage 1 (eGFR ≥ 90; n = 147, 61.00%), CKD Stage 2 (eGFR 60-89; n = 88, 36.51%), and CKD Stage 3 or higher (eGFR < 60; n = 6, 2.49%) (Table 2).

Plasma Biomarkers

Plasma samples were analyzed at the Department of Psychiatry and Neurochemistry, University of Gothenburg. The quantification of plasma Aβ 42/40, GFAP, and NfL was carried out using the commercially available Neurology 4-plex E kit (#103670, Quanterix).Plasma p-tau181 (11), p-tau231 (13), and NTA-tau (15) were analyzed using in-house Simoa assays developed at the University of Gothenburg. Plasma p-tau217 was quantified using the commercially available AlzPath Simoa assay (12), as previously described. All measurements were performed on the automated Simoa HD-X platform (Quanterix, MA, USA).

Statistical Analysis

Demographic information was summarized with median (Q1-Q3) for continuous variables and as count and percentages for categorical variables. We initially plotted plasma biomarker concentrations according to CKD stage and examined between-group differences with linear models adjusted for age and sex. Given the small number of individuals in CKD stage 3, we focused further analyses on continuous eGFR measures. We computed the Spearman correlation between plasma biomarkers and eGFR. Then, we used linear models with blood biomarkers as the outcome variable to evaluate the magnitude of the association between eGFR and plasma biomarkers, plotting the standardized eGFR β-estimate and 95% confidence intervals with three models per plasma biomarker. The first model included only eGFR as a predictor; the second included eGFR, age, and sex; the third, eGFR, age, sex, and Aβ-PET SUVr. This was to determine whether eGFR would remain associated with blood biomarkers independently of demographics and AD pathology. Afterwards, we evaluated whether eGFR would contribute to the prediction of Aβ-PET positivity. We built logistic regression models with Aβ-PET status as the outcome and three schemes of predictors: (i) plasma biomarker alone, (ii) plasma biomarker, age and sex, (iii) plasma biomarker, age, sex, and eGFR. Within each biomarker, we then compared the area under the curve (AUC) between the models and calculated the delta Akaike Information Criterion (AIC) from the models compared to the plasma biomarker alone. A difference of -2 or greater would be considered significant based on previous work (36). These logistic and linear models were always fitted separately for each biomarker, given our goal was to evaluate the influence of eGFR on each biomarker rather than developing a combination panel of biomarkers. Statistical significance was defined as an alpha of 0.05, and analyses were all performed in R Statistical Software (version 4.2.1).

Participant characteristics

The study encompassed 242 participants, with the following median ages across groups: CKD-stage 1 (median = 67.1 years, IQR: 62.5–72.1), CKD-stage 2 (median = 73.0 years, IQR: 69.0–77.3), and CKD-stage 3 (median = 78.2 years, IQR: 74.9–79.2). There were 160 females (66.4%) in the total cohort. 147 (61%), 88 (36.5%), and 6 (2.5%) participants were assigned to CKD-stage 1, stage 2, and stage 3 groups, respectively. eGFR levels were highest in the CKD-stage 1 group, while participants in the CKD-stage 3 group were the oldest and exhibited higher levels of plasma biomarkers. The detailed demographic, clinical information, and plasma biomarker results of participants are described in Table1 and Table 2.

Plasma biomarker concentrations after stratified by eGFR

Plasma Aβ concentrations altered across the various CKD stages. Additionally, adjustments for age, sex, and amyloid status were carefully applied to account for potential confounding factors. Aβ40 levels were significantly elevated in CKD stage 3 (median (Q1–Q3): 124.00 (113–143) pg/mL, p < 0.001), followed by stage 2 (median (Q1–Q3): 93.80 (84.8-105) pg/mL, p = 0.002), and stage 1 (median (Q1–Q3): 89.60 (77.2-100) pg/mL, p = 0.001) (Table 2, Supplementary Table 1, Figure 1). Similarly, Aβ42 concentrations were higher in CKD stage 3 (median (Q1–Q3): 8.50 (6.76-10.3) pg/mL, p = 0.004) compared to stage 1 (median (Q1–Q3): 6.17 (4.86-7.15) pg/mL, p = 0.021). However, no statistically significant difference was observed between stages 2 and 3 (p = 0.076). Furthermore, the Aβ42/Aβ40 ratio did not show any statistically significant differences across CKD stages (p > 0.05) (Table 2, Supplementary Table 1, Figure 1).

All plasma p-tau species displayed elevated concentrations in CKD stage 3 compared to earlier CKD stages, but none of these differences reached statistical significance after adjustment for age, sex, and amyloid status. P-tau181 levels were highest in stage 3 (median (Q1–Q3): 11.60 (10.5-15.5) pg/mL), followed by stage 2 (median (Q1–Q3): 8.46 (6.73-12.5) pg/mL), and stage 1 (median (Q1–Q3): 7.98 (5.62-11.9) pg/mL). However, the differences between stages did not achieve statistical significance. Both p-tau217 and p-tau231 levels followed a similar trend, with higher median values in stage 3 (p-tau217: 0.57 pg/mL; p-tau231: 23.80 pg/mL) compared to earlier stages, though these differences also failed to reach statistical significance after adjusting for age, sex, and amyloid status (p > 0.05) (Table 2, Supplementary Table 1, Figure 1). Plasma NTA-tau levels also showed a slight increase in CKD stage 3 (median 0.28 pg/mL) compared to stages 1 and 2. However, these differences were not statistically significant (p > 0.05), even after adjusting for age, sex, and amyloid status (Table 2, Supplementary Table 1, Figure 1).

NfL levels demonstrated a significant increase in CKD stage 3 (median (Q1–Q3): 40.40(31.7-48.7) pg/mL, p < 0.001), with progressively lower levels in stage 2 (median (Q1–Q3): 26.80 (18.5-32.7) pg/mL, p < 0.001) and stage 1 (median (Q1–Q3): 18.80 (14.4-27.1 pg/mL, p = 0.001). In contrast, GFAP levels did not show any statistically significant differences between CKD stages (p > 0.05), even after adjusting for age, sex, and amyloid status (Table 2, Supplementary Table 1, Figure 1).

The relationship between eGFR and plasma AD biomarker concentrations

Associations between eGFR and plasma biomarkers

A significant inverse correlation was observed between eGFR and multiple AD biomarkers, including Aβ42 (rho = -0.23, p = 2e-04), Aβ40 (rho = -0.43, p < 1e-04), p-tau181 (rho = -0.22, p = 3e-04), p-tau217 (rho = -0.34, p < 1e-04), p-tau231 (rho = -0.24, p < 1e-04), NfL (rho = -0.52, p < 1e-04), and GFAP (rho = -0.40, p < 1e-04) (Figure 2a). In contrast, a positive correlation was observed between the Aβ42/Aβ40 ratio and eGFR (rho = 0.15, p = 0.0144). However, NTA-tau did not show a significant correlation with eGFR (rho = -0.047, p = 0.4535) (Figure 2a).

When dividing the cohort into two subgroups based on amyloid status (Aβ-negative and Aβ-positive), distinct patterns emerged. In the Aβ-negative group, significant inverse correlations were observed between eGFR and several biomarkers, including Aβ42 (rho = -0.27, p = 4e-04), Aβ40 (rho = -0.5, p < 1e-04), p-tau181 (rho = -0.24, p = 0.0021), p-tau217 (rho = -0.4, p < 1e-04), p-tau231 (rho = -0.24, p = 0.0019), NfL (rho = -0.6, p < 1e-04), and GFAP (rho = -0.55, p < 1e-04) (Figure 2). Additionally, the Aβ42/Aβ40 ratio showed a positive correlation with eGFR (rho = 0.23, p = 0.003). No significant correlations were found for NTA-tau (rho = -0.02, p = 0.8006) (Figure 2b).

In contrast, in the Aβ-positive group, fewer significant correlations were observed. For Aβ40, a weaker but still significant inverse correlation was seen (rho = -0.24, p = 0.0179). However, the correlations for Aβ42 (rho = -0.34, p = 0.001) and the Aβ42/Aβ40 ratio (rho = -0.19, p = 0.0604), though trending towards significance, did not reach statistical significance. Additionally, no significant correlations were found for p-tau181 (rho = -0.073, p = 0.4765), p-tau217 (rho = -0.049, p = 0.6431), p-tau231 (rho = -0.096, p = 0.3492), NfL (rho = -0.31, p = 0.0016), or GFAP (rho = -0.062, p = 0.5418), indicating a weaker relationship between these biomarkers and eGFR in this group. As previously observed, NTA-tau did not show significant correlations with eGFR (rho = 0.025, p = 0.8048) (Figure 2b).

Multivariable Regression Analysis of eGFR and Plasma Biomarkers

A significant inverse association was observed between standardized eGFR and multiple plasma biomarkers across different models. In the univariate analysis, lower eGFR was associated with the levels of Aβ42 (β = -0.29, 95% CI: -0.40 to -0.18, p < 0.001), Aβ40 (β = -0.53, 95% CI: -0.63 to -0.43, p < 0.001), p-tau217 (β = -0.15, 95% CI: -0.26 to -0.05, p = 0.005), p-tau231 (β = -0.16, 95% CI: -0.28 to -0.05, p = 0.007), p-tau181 (β = -0.24, 95% CI: -0.36 to -0.12, p < 0.001), NfL (β = -0.49, 95% CI: -0.59 to -0.38, p < 0.001), and GFAP (β = -0.41, 95% CI: -0.52 to -0.30, p < 0.001). However, the Aβ42/Aβ40 ratio exhibited a positive association with eGFR (β = 0.23, 95% CI: 0.11 to 0.34, p < 0.001), while NTA-tau did not show a significant relationship with eGFR (β = -0.07, 95% CI: -0.19 to 0.05, p = 0.24) (Table 3, Figure 3).

After adjusting for age and sex, the associations remained significant for Aβ42 (β = -0.54, 95% CI: -0.71 to -0.38, p < 0.001), Aβ40 (β = -0.49, 95% CI: -0.64 to -0.34, p < 0.001). GFAP (β = -0.19, 95% CI: -0.35 to -0.04, p = 0.016) remained significant after adjustment. The associations with p-tau217 (β = -0.06, 95% CI: -0.18 to 0.07, p = 0.38), p-tau231 (β = -0.11, 95% CI: -0.28 to 0.07, p = 0.23), and p-tau181 (β = -0.13, 95% CI: -0.30 to 0.04, p = 0.14) weakened and became non-significant. NfL showed a weaker but still significant association with eGFR (β = -0.40, 95% CI: -0.55 to -0.24, p < 0.001), while the Aβ42/Aβ40 ratio also became non-significant (β = -0.14, 95% CI: -0.30 to 0.02, p = 0.10). NTA-tau remained non-significant (β = -0.14, 95% CI: -0.32 to 0.04, p = 0.13) (Table 3, Figure 3).

After further adjustment for Aβ-PET status, the associations for Aβ42 (β = -0.51, 95% CI: -0.66 to -0.35, p < 0.001) and Aβ40 (β = -0.43, 95% CI: -0.59 to -0.28, p < 0.001) remained robust, while GFAP continued to show a significant association (β = -0.19, 95% CI: -0.33 to -0.05, p = 0.007). The association between eGFR and NfL remained significant (β = -0.38, 95% CI: -0.54 to -0.23, p < 0.001), while p-tau217 (β = -0.11, 95% CI: -0.25 to 0.03, p = 0.12), p-tau231 (β = -0.11, 95% CI: -0.28 to 0.05, p = 0.18), and p-tau181 (β = -0.12, 95% CI: -0.28 to 0.04, p = 0.14) were not significantly associated. The Aβ42/Aβ40 ratio remained non-significant (β = -0.13, 95% CI: -0.28 to 0.02, p = 0.09). NTA-tau remained non-significant across all models (β = -0.15, 95% CI: -0.32 to 0.04, p = 0.44) (Table 3, Figure 3).

Assessing the Incremental Value of eGFR in Predicting Aβ-PET Positivity Using Plasma Biomarkers

To further assess the predictive value of plasma biomarkers for Aβ-PET positivity, logistic regression analyses were conducted to evaluate the AUC and AIC across different models. The AUC for each biomarker was calculated for univariate models, models adjusted for age and sex, and fully adjusted models that included eGFR.

The ΔAIC values revealed that while the addition of age and sex to plasma biomarkers significantly improved model fit, the inclusion of eGFR had minimal impact. For instance, in predicting Aβ-PET positivity, the AIC for the Aβ42/40 ratio improved by 26.3 points when age and sex were added, but further inclusion of eGFR led to only a slight change (ΔAIC = 0.78). Similarly, for Aβ40, the AIC improved by 19.3 points with age and sex, while adding eGFR resulted in a negligible difference (ΔAIC = -0.23). For p-tau217, the improvement in AIC was 26.7 points after adjusting for age and sex, with eGFR contributing minimally to model improvement (ΔAIC = 0.53). In the case of NfL, the AIC improved by 19.6 points with age and sex, while the inclusion of eGFR made no notable impact (ΔAIC = 0.74). Similarly, NTA-tau showed an improvement in AIC by 22.8 points with age and sex, with a negligible change when eGFR was added (ΔAIC = -0.05)(Table 4, Figure 4).

In terms of AUC, the addition of eGFR resulted in minimal improvements in clinical discrimination. For instance, the AUC for the Aβ42/40 ratio increased from 0.75 in the univariate model to 0.76 after adjusting for age and sex, with no further improvement when eGFR was added (AUC = 0.76). Similarly, for Aβ40, the AUC increased from 0.57 to 0.63 with age and sex and marginally improved to 0.63 with eGFR. For p-tau217, the AUC increased from 0.80 to 0.86 with age and sex and to 0.87 when eGFR was added. The AUC for NfL increased from 0.57 to 0.65 with age and sex and showed a slight improvement to 0.67 with the addition of eGFR. Finally, NTA-tau's AUC improved from 0.64 to 0.71 after adjusting for age and sex, with no further improvement when eGFR was added (AUC = 0.71) (Table 4, Figure 4).

These findings suggest that while age and sex substantially improve the predictive power of plasma biomarkers for Aβ-PET positivity, the inclusion of eGFR adds little to no further benefit in terms of model fit or discrimination ability across all evaluated biomarkers, including p-tau181 and NTA-tau.

This study examined the relationship between kidney function, as measured by eGFR, and the concentrations of peripheral AD biomarkers, as well as the role of kidney impairment in predicting amyloid positivity. The levels of all AD biomarkers were increased the most in CKD stage 3, followed by CKD stage 2 and CKD stage 1. Furthermore, the AD biomarkers investigated were initially found to be significantly associated with eGFR. However, after adjusting for age, sex, and Aβ positivity, the associations for the Aβ42/40 ratio, p-tau181, p-tau217, and p-tau231 were no longer significant. Plasma NTA-tau remained non-significantly associated with eGFR across all models. Despite these adjustments, significant associations remained for other biomarkers, including NfL, Aβ42, Aβ40 and GFAP. We also observed that adding eGFR to the prediction models for Aβ positivity did not enhance their performance, whereas the inclusion of age and sex significantly improved model fit when combined with plasma biomarkers. Taken together, our study provides evidence that blood AD biomarkers increase as kidney function declines and are associated with eGFR. However, when incorporated into amyloid positivity prediction models, these biomarkers did not improve the models' predictive performance.

Blood-based biomarkers hold significant potential for diagnosing and predicting the progression of AD (12). However, before they can be widely adopted, it is crucial to thoroughly understand the biological and technical factors that could compromise their diagnostic accuracy. Moreover, for inclusion in clinical trials for AD, biomarkers should accurately identify participants who are in the preclinical phase of the disease (7, 37, 38) and track treatment efficacy (19, 20). Additionally, any drifts in biomarkers that are independent of amyloid pathology could lead to misclassification of the current pathology (8, 9, 39), inappropriate selection of participants for clinical trials, and misinterpretation of treatment responses (7). Recently, the Global CEO Initiative on Alzheimer’s Disease convened a blood biomarker (BBM) workgroup and made recommendations on the acceptable performance of BBM tests for amyloid pathology (40). The importance of using a two cut-off approach was also emphasized, along with the observation that BBMs are associated with kidney dysfunction(40, 41). Notably, BBM ratios, such as Aβ42/40 and p-tau217 to non-phosphorylated tau, were highlighted for their consistent performance and reduced susceptibility to kidney dysfunction (40), as demonstrated in several previous publications (25, 26, 42). In the recently revised criteria for the biological diagnosis and staging of AD, NfL and GFAP are not included in the core biomarker category to aid in diagnosis (43). However, they are mentioned as useful for staging, prognosis, and as indicators of biological treatment effects, serving as markers of neurodegeneration and inflammation, respectively (43). Therefore, it is important to determine the effect of kidney dysfunction on these biomarkers, along with the core biomarkers.

Given this context, it is important to note that CKD represents a global public health burden, with early stages often being asymptomatic. As a result, CKD is commonly diagnosed at later stages. The global estimated prevalence of CKD is 13.4% (44). Studies have reported that in the general population, particularly among older adults, the rate of decline in eGFR can range from about 0.75 to 1.2 mL/min/1.73 m² per year (45–47). Given that the most common form of sporadic AD typically occurs around the age of 60 (48), the influence of CKD should be considered. Therefore, previous studies have investigated the effect of kidney impairment on AD biomarkers (25, 26, 49). Our study adds to this growing body of evidence and is in line with previous studies that investigated the effect of kidney dysfunction on plasma biomarkers via eGFR or creatinine (22, 25, 26, 31, 42, 49, 50). For example, among NfL, GFAP, and p-tau217, NfL showed the strongest association with eGFR, followed by GFAP and p-tau217. Even after adjusting for age, sex, and amyloid positivity, plasma NfL and GFAP remained significantly associated with kidney impairment, suggesting that the influence of kidney dysfunction should be considered when assessing neuronal injury and astrocyte activation in patients with AD. These findings are consistent with previous studies (25–27), underscoring the importance of considering kidney function as a factor that can affect biomarker interpretation (26). In our study, we used eGFR instead of creatinine, and this methodological difference, along with our relatively smaller sample size, may explain any discrepancies in our findings. Janalidze et al. have shown that using ratios for p-tau species can mitigate the influence of kidney impairment on blood p-tau levels (42). Although plasma NTA-tau did not show a significant association with eGFR, we observed similar findings for the p-tau species in our study. Before adjusting for covariates, p-tau181, p-tau217, and p-tau231 were significantly associated with eGFR. However, after adjusting for covariates, these associations were no longer significant. Building on these findings, incorporating covariates, such as eGFR, in a two-step model of p-tau217 could reduce the misclassification rate, particularly in intermediate categories; however, further research is needed in this area.

Additionally, we compared different predictive models, including a univariate model, an adjusted model (age + sex), and a full model (age + sex + eGFR), to assess whether the inclusion of eGFR alongside other covariates could enhance the prediction of amyloid positivity. We then evaluated model fit by examining the impact of adding eGFR to the age + sex model using AIC. The findings revealed that eGFR does not contribute to the prediction of Aβ-PET positivity, its effect is relatively modest compared to that of age and sex. Similar to the existing literature (51), where most studies did not include eGFR in their prediction models (8, 52), our findings also underscore the dominant role of demographic factors and biomarker levels, given no improvement in model fit nor in discrimination were observed by adding eGFR. These results indicate that while age and sex significantly add predictive value for Aβ-PET positivity when combined with plasma biomarkers, eGFR may be less relevant.

This study has several strengths and limitations. First, we investigated a broad range of blood biomarkers and their associations with kidney function, which had not yet been done in such extent. Moreover, we utilized eGFR to objectively assess renal function and its relationship with plasma biomarkers, rather than relying on self-reported kidney dysfunction or a medical history of CKD. Furthermore, our study included prediction models where eGFR, as a proxy for kidney impairment, was incorporated into the analyses. However, due to the cross-sectional design of the study, we cannot infer a causal relationship between decreased renal function and elevated plasma biomarker levels. Previous studies (53, 54) have suggested that a possible kidney-brain axis may contribute to the development of dementia. Nonetheless, as mentioned earlier, we are unable to draw such causal conclusions from our findings. On top of that, incorporating cystatin C into the eGFR calculation could provide a more accurate estimation, as demonstrated in a large Swedish study (55), particularly in older populations. However, since cystatin C data was not available in this study, we were unable to assess the potential influence of cystatin C on eGFR calculation and its impact on the prediction models. Additionally, the small number of participants in CKD stage 3 could have led to an underestimation of the observed effects. Finally, while our findings were based on a well-defined cohort, more real-world effects might have been observed if these variables were investigated in a community-based cohort.

In summary, this study offers a detailed examination of how kidney dysfunction affects blood biomarker levels, utilizing a wide array of plasma biomarkers. Although all the biomarkers studied were initially associated with eGFR, further adjustments for additional covariates revealed the need for more precise evaluations. Future studies with population-based cohorts are necessary to explore the relationship between renal function and other AD plasma biomarkers, considering a broader range of potential confounders. These studies should aim to validate the findings across diverse populations and incorporate covariates, such as eGFR, into two-step decision models. Ultimately, diagnostic and prognostic algorithms for AD based on plasma biomarkers may not be improved by adding eGFR into the models despite the apparent increase in biomarker levels as kidney function declines.

Alzheimer’s disease

AD

Positron emission tomography

PET

Cerebrospinal fluid

CSF

Amyloid-beta

Aβ

Phosphorylated tau

p-tau

N-terminal containing tau fragments

NTA-tau

Neurofilament light

NfL

Glial fibrillary acidic protein

GFAP

Disease-modifying therapies

DMT

Estimated glomerular filtration rate

eGFR

β2-Microglobulin

B2M

Chronic kidney disease

CKD

Translational Biomarkers in Aging and Dementia

TRIAD

Clinical Dementia Rating

CDR

Mini-Mental State Examination

MMSE

Magnetic resonance imaging

MRI

Kidney Disease: Improving Global Outcomes

KDIGO-CKD

Area under the curve

AUC

Akaike Information Criterion

AIC

Blood biomarker

BBM

Ethics approval and consent to participate: All participants provided informed consent, and the study protocols were approved by the appropriate ethical review boards. TRIAD was approved by the Montreal Neurological Institute PET working committee and the Douglas Mental Health University Institute Research Ethics Board (IUSMD16-61, IUSMD16-60).

Consent for publication: Not applicable

Availability of data and materials: The present study includes no data deposited in external repositories. Bulk Anonymized data can be shared upon reasonable request from a qualified academic investigator for the sole purpose of replicating procedures and results presented in the article, providing data transfer is in agreement with EU legislation and decisions by the institutional review board of each participating research centre.

Competing interests: NJA has given lectures, produced educational materials, and participated in educational programs for Eli-Lily, BioArtic, Quanterix. KB has served as a consultant and at advisory boards for Abbvie, AC Immune, ALZPath, AriBio, BioArctic, Biogen, Eisai, Lilly, Moleac Pte. Ltd, Neurimmune, Novartis, Ono Pharma, Prothena, Roche Diagnostics, Sanofi and Siemens Healthineers; has served at data monitoring committees for Julius Clinical and Novartis; has given lectures, produced educational materials and participated in educational programs for AC Immune, Biogen, Celdara Medical, Eisai and Roche Diagnostics; and is a co-founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program, outside the work presented in this paper. HZ has served at scientific advisory boards and/or as a consultant for Abbvie, Acumen, Alector, Alzinova, ALZPath, Annexon, Apellis, Artery Therapeutics, AZTherapies, CogRx, Denali, Eisai, Nervgen, Novo Nordisk, Optoceutics, Passage Bio, Pinteon Therapeutics, Prothena, Red Abbey Labs, reMYND, Roche, Samumed, Siemens Healthineers, Triplet Therapeutics, and Wave, has given lectures in symposia sponsored by Cellectricon, Fujirebio, Alzecure, Biogen, and Roche, and is a co-founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program (outside submitted work).

Funding: This research is supported by the Weston Brain Institute, Canadian Institutes of Health Research (CIHR) [MOP-11-51-31; RFN 152985, 159815, 162303], Canadian Consortium of Neurodegeneration and Aging (CCNA; MOP-11-51-31 -team 1), the Alzheimer’s Association [NIRG-12-92090, NIRP-12-259245], Brain Canada Foundation (CFI Project 34874; 33397), the Fonds de Recherche du Québec – Santé (FRQS; Chercheur Boursier, 2020-VICO-279314; 2024-VICO-356138). TAP, P.R-N and SG are members of the CIHR-CCNA Canadian Consortium of Neurodegeneration in Aging. Colin J. Adair Charitable Foundation. HZ is a Wallenberg Scholar and a Distinguished Professor at the Swedish Research Council supported by grants from the Swedish Research Council (#2023-00356, #2022-01018 and #2019-02397), the European Union’s Horizon Europe research and innovation programme under grant agreement No 101053962, Swedish State Support for Clinical Research (#ALFGBG-71320), the Alzheimer Drug Discovery Foundation (ADDF), USA (#201809-2016862), the AD Strategic Fund and the Alzheimer's Association (#ADSF-21-831376-C, #ADSF-21-831381-C, #ADSF-21-831377-C, and #ADSF-24-1284328-C), the European Partnership on Metrology, co-financed from the European Union’s Horizon Europe Research and Innovation Programme and by the Participating States (NEuroBioStand, #22HLT07), the Bluefield Project, Cure Alzheimer’s Fund, the Olav Thon Foundation, the Erling-Persson Family Foundation, Familjen Rönströms Stiftelse, Stiftelsen för Gamla Tjänarinnor, Hjärnfonden, Sweden (#FO2022-0270), the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 860197 (MIRIADE), the European Union Joint Programme – Neurodegenerative Disease Research (JPND2021-00694), the National Institute for Health and Care Research University College London Hospitals Biomedical Research Centre, and the UK Dementia Research Institute at UCL (UKDRI-1003). KB is supported by the Swedish Research Council (#2017- 00915 and #2022-00732), the Swedish Alzheimer Foundation (#AF-930351, #AF-939721 and #AF- 968270), Hjärnfonden, Sweden (#FO2017-0243 and #ALZ2022-0006), the Swedish state under the agreement between the Swedish government and the County Councils, the ALF-agreement (#ALFGBG- 715986 and #ALFGBG-965240), the European Union Joint Program for Neurodegenerative Disorders (JPND2019-466-236), the Alzheimer’s Association 2021 Zenith Award (ZEN-21-848495), and the Alzheimer’s Association 2022-2025 Grant (SG-23-1038904 QC). ALB is supported by the Swedish Alzheimer Foundation (grant #AF-940262), Stiftelsen för Gamla Tjänarinnor and the Alzheimer’s Association Research Fellowship (grant #AARFD-22-974564).

Authors' contributions: B.A, WB, N.JA, A.L.B: Conceptualization, Methodology, Formal analysis, Data curation, Visualization, Writing, Original draft preparation.N.JA., A.L.B, HZ, K.B, P.R.N, S.G: Supervision, Writing, Reviewing, I.P, K.T, J.L.R: Sample preparation and assay of the samples. J.T, N.R, J.S, S.S., P.V, M.M, J.K, A.C.M, C.T.: Data preparation for the TRIAD cohort and project administration. TRIAD Cohort provided all demographic and clinical data analyzed in this study.

Acknowledgements: The authors would like to express their most sincere gratitude to all TRIAD participants and relatives, without whom this research would have not been possible. The authors would like to thank the diverse funding sources that made this work possible.

Hansson O. Biomarkers for neurodegenerative diseases. Nat Med. 2021;27(6):954–63.
Palmqvist S, Zetterberg H, Mattsson N, Johansson P, Initiative ADN, Minthon L, et al. Detailed comparison of amyloid PET and CSF biomarkers for identifying early Alzheimer disease. Neurology. 2015;85(14):1240–9.
Schindler SE, Li Y, Buckles VD, Gordon BA, Benzinger TL, Wang G, et al. Predicting symptom onset in sporadic Alzheimer disease with amyloid PET. Neurology. 2021;97(18):e1823–34.
Therriault J, Lussier FZ, Tissot C, Chamoun M, Stevenson J, Rahmouni N et al. Amyloid beta plaque accumulation with longitudinal [18F] AZD4694 PET. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring. 2023;15(3):e12391.
Hansson O, Seibyl J, Stomrud E, Zetterberg H, Trojanowski JQ, Bittner T, et al. CSF biomarkers of Alzheimer's disease concord with amyloid-β PET and predict clinical progression: A study of fully automated immunoassays in BioFINDER and ADNI cohorts. Alzheimer's Dement. 2018;14(11):1470–81.
Alcolea D, Pegueroles J, Munoz L, Camacho V, López-Mora D, Fernández‐León A, et al. Agreement of amyloid PET and CSF biomarkers for Alzheimer's disease on Lumipulse. Ann Clin Transl Neurol. 2019;6(9):1815–24.
Hansson O, Blennow K, Zetterberg H, Dage J. Blood biomarkers for Alzheimer’s disease in clinical practice and trials. Nat aging. 2023;3(5):506–19.
Benedet AL, Brum WS, Hansson O, Initiative ADN, Karikari TK, Zimmer ER, et al. The accuracy and robustness of plasma biomarker models for amyloid PET positivity. Alzheimers Res Ther. 2022;14(1):26.
Brum WS, Docherty KF, Ashton NJ, Zetterberg H, Hansson O, McMurray JJ, et al. Effect of neprilysin inhibition on Alzheimer disease plasma biomarkers: a secondary analysis of a randomized clinical trial. JAMA Neurol. 2024;81(2):197–200.
Ashton NJ, Keshavan A, Brum WS, Andreasson U, Arslan B, Droescher M et al. The Alzheimers Association Global Biomarker Standardization Consortium (GBSC) plasma phospho-tau Round Robin study. medRxiv. 2024:2024.08. 22.24312244.
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.
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.
Ashton NJ, Pascoal TA, Karikari TK, Benedet AL, Lantero-Rodriguez J, Brinkmalm G, et al. Plasma p-tau231: a new biomarker for incipient Alzheimer’s disease pathology. Acta Neuropathol. 2021;141:709–24.
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.
Lantero-Rodriguez J, Salvadó G, Snellman A, Montoliu-Gaya L, Brum WS, Benedet AL, et al. Plasma N-terminal containing tau fragments (NTA-tau): a biomarker of tau deposition in Alzheimer’s Disease. Mol Neurodegeneration. 2024;19(1):19.
Lantero-Rodriguez J, Tissot C, Snellman A, Servaes S, Benedet AL, Rahmouni N, et al. Plasma and CSF concentrations of N‐terminal tau fragments associate with in vivo neurofibrillary tangle burden. Alzheimer's Dement. 2023;19(12):5343–54.
Arslan B, Zetterberg H. Neurofilament light chain as neuronal injury marker–what is needed to facilitate implementation in clinical laboratory practice? Clinical Chemistry and Laboratory Medicine (CCLM). 2023;61(7):1140–9.
Benedet AL, Milà-Alomà M, Vrillon A, Ashton NJ, Pascoal TA, Lussier F, et al. Differences between plasma and cerebrospinal fluid glial fibrillary acidic protein levels across the Alzheimer disease continuum. JAMA Neurol. 2021;78(12):1471–83.
Van Dyck CH, Swanson CJ, Aisen P, Bateman RJ, Chen C, Gee M, et al. Lecanemab in early Alzheimer’s disease. N Engl J Med. 2023;388(1):9–21.
Sims JR, Zimmer JA, Evans CD, Lu M, Ardayfio P, Sparks J, et al. Donanemab in early symptomatic Alzheimer disease: the TRAILBLAZER-ALZ 2 randomized clinical trial. JAMA. 2023;330(6):512–27.
Self WK, Holtzman DM. Emerging diagnostics and therapeutics for Alzheimer disease. Nat Med. 2023:1–13.
Mielke MM, Dage JL, Frank RD, Algeciras-Schimnich A, Knopman DS, Lowe VJ, et al. Performance of plasma phosphorylated tau 181 and 217 in the community. Nat Med. 2022;28(7):1398–405.
Jamba A, Javzan T, Boldbaatar K, Amarbold U-E. # 1372 Association of plasma beta-2-microglobulin with risk of cardiovascular disease in patients with chronic kidney disease. Nephrol Dialysis Transplantation. 2024;39(Supplement1):gfae069–1466.
Sprangers B, Claes K, Evenepoel P, Kuypers D, Poesen K, Delforge M, et al. Comparison of 2 serum-free light-chain assays in CKD patients. Kidney Int Rep. 2020;5(5):627–31.
Syrjanen JA, Campbell MR, Algeciras-Schimnich A, Vemuri P, Graff‐Radford J, Machulda MM, et al. Associations of amyloid and neurodegeneration plasma biomarkers with comorbidities. Alzheimer's Dement. 2022;18(6):1128–40.
Pichet Binette A, Janelidze S, Cullen N, Dage JL, Bateman RJ, Zetterberg H, et al. Confounding factors of Alzheimer's disease plasma biomarkers and their impact on clinical performance. Alzheimer's Dement. 2023;19(4):1403–14.
Akamine S, Marutani N, Kanayama D, Gotoh S, Maruyama R, Yanagida K, et al. Renal function is associated with blood neurofilament light chain level in older adults. Sci Rep. 2020;10(1):20350.
Inker LA, Eneanya ND, Coresh J, Tighiouart H, Wang D, Sang Y, et al. New creatinine-and cystatin C–based equations to estimate GFR without race. N Engl J Med. 2021;385(19):1737–49.
Dittrich A, Ashton NJ, Zetterberg H, Blennow K, Zettergren A, Simrén J et al. Association of Chronic Kidney Disease With Plasma NfL and Other Biomarkers of Neurodegeneration: The H70 Birth Cohort Study in Gothenburg. Neurology. 2023.
Lehmann S, Schraen-Maschke S, Vidal J-S, Delaby C, Blanc F, Paquet C, et al. Plasma phosphorylated tau 181 predicts amyloid status and conversion to dementia stage dependent on renal function. J Neurol Neurosurg Psychiatry. 2023;94(6):411–9.
Wu J, Xiao Z, Wang M, Wu W, Ma X, Liang X, et al. The impact of kidney function on plasma neurofilament light and phospho-tau 181 in a community-based cohort: the Shanghai Aging Study. Alzheimers Res Ther. 2024;16(1):32.
Morris JC. The Clinical Dementia Rating (CDR) current version and scoring rules. Neurology. 1993;43(11):2412–a.
Arevalo-Rodriguez I, Smailagic N, i Figuls MR, Ciapponi A, Sanchez‐Perez E, Giannakou A et al. Mini‐Mental State Examination (MMSE) for the detection of Alzheimer's disease and other dementias in people with mild cognitive impairment (MCI). Cochrane database of systematic reviews. 2015(3).
Moroney J, Bagiella E, Desmond D, Hachinski VC, Molsa P, Gustafson L, et al. Meta-analysis of the Hachinski Ischemic Score in pathologically verified dementias. Neurology. 1997;49(4):1096–105.
Stevens PE, Levin A, Members* KDIGOCKDGDWG. Evaluation and management of chronic kidney disease: synopsis of the kidney disease: improving global outcomes 2012 clinical practice guideline. Ann Intern Med. 2013;158(11):825–30.
Burnham KP, Anderson DR. Multimodel inference: understanding AIC and BIC in model selection. Sociol methods Res. 2004;33(2):261–304.
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 Reviews Neurol. 2018;14(11):639–52.
Hampel H, Hu Y, Cummings J, Mattke S, Iwatsubo T, Nakamura A, et al. Blood-based biomarkers for Alzheimer’s disease: Current state and future use in a transformed global healthcare landscape. Neuron. 2023;111(18):2781–99.
Karikari TK, Ashton NJ, Brinkmalm G, Brum WS, Benedet AL, Montoliu-Gaya L, et al. Blood phospho-tau in Alzheimer disease: analysis, interpretation, and clinical utility. Nat Reviews Neurol. 2022;18(7):400–18.
Schindler SE, Galasko D, Pereira AC, Rabinovici GD, Salloway S, Suárez-Calvet M et al. Acceptable performance of blood biomarker tests of amyloid pathology—recommendations from the Global CEO Initiative on Alzheimer’s Disease. Nat Reviews Neurol. 2024:1–14.
Brum WS, Cullen NC, Janelidze S, Ashton NJ, Zimmer ER, Therriault J, et al. A two-step workflow based on plasma p-tau217 to screen for amyloid β positivity with further confirmatory testing only in uncertain cases. Nat Aging. 2023;3(9):1079–90.
Janelidze S, Barthélemy NR, He Y, Bateman RJ, Hansson O. Mitigating the Associations of Kidney Dysfunction With Blood Biomarkers of Alzheimer Disease by Using Phosphorylated Tau to Total Tau Ratios. JAMA Neurol. 2023;80(5):516–22.
Jack CR Jr, Andrews JS, Beach TG, Buracchio T, Dunn B, Graf A et al. Revised criteria for diagnosis and staging of Alzheimer's disease: Alzheimer's Association Workgroup. Alzheimer's & Dementia. 2024;20(8):5143-69.
Lv J-C, Zhang L-X. Prevalence and disease burden of chronic kidney disease. Ren fibrosis: Mech Ther. 2019:3–15.
Toyama T, Kitagawa K, Oshima M, Kitajima S, Hara A, Iwata Y, et al. Age differences in the relationships between risk factors and loss of kidney function: a general population cohort study. BMC Nephrol. 2020;21:1–9.
Cohen E, Nardi Y, Krause I, Goldberg E, Milo G, Garty M, et al. A longitudinal assessment of the natural rate of decline in renal function with age. J Nephrol. 2014;27:635–41.
Santos-Araújo C, Mendonça L, Carvalho DS, Bernardo F, Pardal M, Couceiro J, et al. Twenty years of real-world data to estimate chronic kidney disease prevalence and staging in an unselected population. Clin Kidney J. 2023;16(1):111–24.
Vermunt L, Sikkes SA, Van Den Hout A, Handels R, Bos I, Van Der Flier WM, et al. Duration of preclinical, prodromal, and dementia stages of Alzheimer's disease in relation to age, sex, and APOE genotype. Alzheimer's Dement. 2019;15(7):888–98.
Dittrich A, Ashton NJ, Zetterberg H, Blennow K, Zettergren A, Simrén J, et al. Association of chronic kidney disease with plasma NfL and other biomarkers of neurodegeneration: the H70 birth cohort study in Gothenburg. Neurology. 2023;101(3):e277–88.
Zhang B, Zhang C, Wang Y, Chen L, Qiao Y, Wang Y, et al. Effect of renal function on the diagnostic performance of plasma biomarkers for Alzheimer’s disease. Front Aging Neurosci. 2023;15:1150510.
Arranz J, Zhu N, Rubio-Guerra S, Rodríguez-Baz Í, Ferrer R, Carmona-Iragui M, et al. Diagnostic performance of plasma pTau217, pTau181, Aβ1–42 and Aβ1–40 in the LUMIPULSE automated platform for the detection of Alzheimer disease. Alzheimers Res Ther. 2024;16(1):139.
Hu Y, Kirmess KM, Meyer MR, Rabinovici GD, Gatsonis C, Siegel BA, et al. Assessment of a plasma amyloid probability score to estimate amyloid positron emission tomography findings among adults with cognitive impairment. JAMA Netw open. 2022;5(4):e228392–e.
Kjaergaard AD, Johannesen BR, Sørensen HT, Henderson VW, Christiansen CF. Kidney disease and risk of dementia: a Danish nationwide cohort study. BMJ open. 2021;11(10):e052652.
Bugnicourt J-M, Godefroy O, Chillon J-M, Choukroun G, Massy ZA. Cognitive disorders and dementia in CKD: the neglected kidney-brain axis. J Am Soc Nephrol. 2013;24(3):353–63.
Fu EL, Carrero J-J, Sang Y, Evans M, Ishigami J, Inker LA, et al. Association of low glomerular filtration rate with adverse outcomes at older age in a large population with routinely measured cystatin C. Ann Intern Med. 2024;177(3):269–79.

Table 1. Demographic and Clinical Characteristics of Study Participants

	CU (N=124)	MCI (N=58)	AD (N=34)	Non-AD (N=26)
Age, median (Q1-Q3)	70.3 (65.7-74.7)	72.2 (66.3-75.9)	66.9 (61.1-73.5)	64.0 (57.5-69.6)
Female, n (%)	89 (71.8%)	35 (60.3%)	21 (61.8%)	15 (57.7%)
YOE, median (Q1-Q3)	15.0 (12.8-18.0)	16.0 (12.0-17.8)	15.0 (13.3-16.0)	15.0 (12.0-17.0)
MMSE, Median (Q1-Q3)	29.0 (28.0-30.0) [11]	29.0 (27.0-29.0) [8]	23.0 (20.0-25.8) [4]	26.0 (23.8-29.0) [6]
APOE ε4 carriers, No. (%)	29 (23.4%) [0]	26 (44.8%) [2]	20 (58.8%) [0]	3 (11.5%) [1]
Amyloid status, No. (%)	30 (24.2%) [0]	37 (63.8%) [0]	33 (97.1%) [0]	0 (0%) [0]
eGFR Median (Q1-Q3)	92.8 (84.2-97.3)	93.1 (83.1-97.2)	92.3 (87.8-97.5)	100 (88.6-104)
Plasma p-tau217, median (Q1-Q3), pg/mL	0.274 (0.170-0.405) [13]	0.677 (0.428-0.940) [7]	1.31 (0.671-1.83) [3]	0.273 (0.190-0.353) [0]
Plasma p-tau181, median (Q1-Q3), pg/mL	7.12 (5.46-10.7) [1]	8.71 (6.82-11.8) [1]	13.6 (9.21-16.9) [0]	7.49 (4.74-10.6) [0]
Plasma p-tau231, median (Q1-Q3), pg/mL	13.0 (10.0-16.7) [0]	18.5 (13.8-23.7) [2]	22.8 (15.9-28.6) [1]	13.4 (11.1-19.8) [0]
Plasma NTA-tau, median (Q1-Q3), pg/mL	0.177 (0.111-0.308) [5]	0.242 (0.112-0.432) [1]	0.522 (0.384-0.635) [1]	0.195 (0.148-0.473) [0]
Plasma GFAP, median (Q1-Q3), pg/mL	154 (109-195) [1]	186 (134-232) [0]	270 (190-312) [0]	140 (72.7-189) [0]
Plasma NfL, median (Q1-Q3), pg/mL	19.4 (14.5-26.9) [1]	23.6 (16.7-30.1) [0]	31.2 (22.0-36.2) [0]	23.3 (14.4-38.4) [0]
Plasma Aβ₄₂, median (Q1-Q3), pg/mL	6.65 (5.68-7.83) [1]	6.56 (5.52-7.19) [0]	5.46 (4.32-6.38) [0]	6.84 (5.23-8.33) [0]
Plasma Aβ₄₀, median (Q1-Q3), pg/mL	92.0 (80.5-101) [1]	91.6 (85.7-106) [0]	97.0 (78.9-103) [0]	86.7 (75.5-103) [0]
Plasma Aβ_42/40, median (Q1-Q3), pg/mL	0.0725 (0.0630-0.0828) [1]	0.0664 (0.0616-0.0778) [0]	0.0601 (0.0551-0.0658) [0]	0.0767 (0.0685-0.0854) [0]

Abbreviations: CU = Cognitively Unimpaired; MCI = Mild Cognitive Impairment; AD = Alzheimer's Disease; Non-AD = Non-Alzheimer's Disease; YOE = Years of Education; MMSE = Mini-Mental State Examination; APOE = Apolipoprotein E; eGFR = Estimated Glomerular Filtration Rate; p-tau = phosphorylated tau; NTA-tau = N-terminal containing tau fragments; GFAP = Glial Fibrillary Acidic Protein; NfL = Neurofilament Light chain; Aβ = Amyloid Beta.

Categorical variables are reported as counts and percentages, while continuous variables are presented as median values with interquartile ranges (Q1-Q3). Missing values are reported in brackets next to the respective variables.

Table 2. Participant Characteristics and Plasma Biomarker Levels across CKD Stages

	CKD Stage-1 (N=147)	CKD Stage-2 (N=88)	CKD Stage-3 (N=6)
Age, median (Q1-Q3)	67.1 (62.5-72.1)	73.0 (69.0-77.3)	78.2 (74.9-79.2)
Female, n(%)	97 (66.0%)	61 (69.3%)	2 (33.3%)
YOE, median (Q1-Q3)	15.0 (12.0-17.0)	15.0 (12.0-18.0)	18.0 (17.3-19.5)
MMSE, Median (Q1-Q3)	29.0 (27.3-30.0) [21]	28.5 (26.0-30.0) [8]	29.0 (28.3-29.8) [0]
APOE ε4 carriers, No. (%)	50 (34.0%) [3]	26 (29.5%) [0]	2 (33.3%) [0]
Amyloid status, No. (%)	61 (41.5%)	37 (42.0%)	2 (33.3%)
Plasma p-tau217, median (Q1-Q3), pg/mL	0.305 (0.188-0.693) [10]	0.418 (0.263-0.812) [13]	0.572 (0.550-0.890) [0]
Plasma p-tau181, median (Q1-Q3), pg/mL	7.98 (5.62-11.9) [1]	8.46 (6.73-12.5) [1]	11.6 (10.5-15.5) [0]
Plasma p-tau231, median (Q1-Q3), pg/mL	14.1 (10.6-20.2) [2]	16.5 (12.5-20.6) [1]	23.8 (19.7-31.5) [0]
Plasma NTA-tau, median (Q1-Q3), pg/mL	0.213 (0.118-0.410) [4]	0.251 (0.123-0.399) [3]	0.282 (0.220-0.447) [0]
Plasma GFAP, median (Q1-Q3), pg/mL	149 (104-218) [0]	181 (143-252) [1]	236 (190-257) [0]
Plasma NfL, median (Q1-Q3), pg/mL	18.8 (14.4-27.1) [0]	26.8 (18.5-32.7) [1]	40.4 (31.7-48.7) [0]
Plasma Aβ42, median (Q1-Q3), pg/mL	6.17 (4.86-7.15) [0]	6.84 (5.79-7.77) [1]	8.50 (6.76-10.3) [0]
Plasma Aβ40, median (Q1-Q3), pg/mL	89.6 (77.2-100) [0]	93.8 (84.8-105) [1]	124 (113-143) [0]
Plasma Aβ42/40, median (Q1-Q3), pg/mL	0.0682 (0.0608-0.0789) [0]	0.0732 (0.0611-0.0828) [1]	0.0679 (0.0636-0.0811) [0]

Abbreviations: CKD = Chronic kidney disease; YOE = Years of Education; MMSE = Mini-Mental State Examination; APOE = Apolipoprotein E; eGFR = Estimated Glomerular Filtration Rate; p-tau = phosphorylated tau; NTA-tau = N-terminal containing tau fragments; GFAP = Glial Fibrillary Acidic Protein; NfL = Neurofilament Light chain; Aβ = Amyloid Beta.

Categorical variables are reported as counts and percentages, while continuous variables are presented as median values with interquartile ranges (Q1-Q3). Missing values are reported in brackets next to the respective variables.

Table 3. Standardized eGFR β-Estimates for Plasma Alzheimer's Disease Biomarkers: Univariate and Multivariable Models

Biomarker	Model	β Estimate	95% CI (Lower)	95% CI (Upper)	p-value
Plasma Aβ42	Univariate	-0.29	-0.40	-0.18	9.92e-7
Plasma Aβ42	Adjusted (age + sex)	-0.54	-0.71	-0.38	5.86e-10
Plasma Aβ42	Adjusted (age + sex + Aβ-PET)	-0.51	-0.66	-0.35	3.11e-10
Plasma Aβ40	Univariate	-0.53	-0.63	-0.43	3.98e-21
Plasma Aβ40	Adjusted (age + sex)	-0.49	-0.64	-0.34	7.93e-10
Plasma Aβ40	Adjusted (age + sex + Aβ-PET)	-0.47	-0.62	-0.32	4.75e-9
Plasma Aβ42/40	Univariate	0.23	0.11	0.34	< 0.001
Plasma Aβ42/40	Adjusted (age + sex)	-0.14	-0.30	0.03	0.100
Plasma Aβ42/40	Adjusted (age + sex + Aβ-PET)	-0.13	-0.28	0.02	0.097
Plasma p-tau181	Univariate	-0.24	-0.36	-0.12	5.91e-5
Plasma p-tau181	Adjusted (age + sex)	-0.13	-0.30	0.04	0.136
Plasma p-tau181	Adjusted (age + sex + Aβ-PET)	-0.12	-0.28	0.04	0.143
Plasma p-tau217	Univariate	-0.25	-0.38	-0.13	4.87e-5
Plasma p-tau217	Adjusted (age + sex)	-0.11	-0.28	0.06	0.216
Plasma p-tau217	Adjusted (age + sex + Aβ-PET)	-0.11	-0.25	0.03	0.113
Plasma p-tau231	Univariate	-0.16	-0.28	-0.05	0.007
Plasma p-tau231	Adjusted (age + sex)	-0.11	-0.28	0.07	0.232
Plasma p-tau231	Adjusted (age + sex + Aβ-PET)	-0.11	-0.28	0.05	0.183
Plasma NTA-tau	Univariate	-0.07	-0.19	0.05	0.242
Plasma NTA-tau	Adjusted (age + sex)	-0.14	-0.32	0.04	0.132
Plasma NTA-tau	Adjusted (age + sex + Aβ-PET)	-0.15	-0.32	0.02	0.085
Plasma NfL	Univariate	-0.49	-0.59	-0.38	1.21e-17
Plasma NfL	Adjusted (age + sex)	-0.40	-0.55	-0.24	8.64e-7
Plasma NfL	Adjusted (age + sex + Aβ-PET)	-0.38	-0.54	-0.23	1.55e-6
Plasma GfAP	Univariate	-0.41	-0.52	-0.30	1.66e-12
Plasma GfAP	Adjusted (age + sex)	-0.19	-0.35	-0.04	0.016
Plasma GfAP	Adjusted (age + sex + Aβ-PET)	-0.19	-0.33	-0.05	0.007

Abbreviations: eGFR, estimated glomerular filtration rate; Aβ42, amyloid-beta 42; Aβ40, amyloid-beta 40; p-tau181, phosphorylated tau at threonine 181; p-tau217, phosphorylated tau at threonine 217; p-tau231, phosphorylated tau at threonine 231; NTA-tau, N-terminal containing tau fragments; NFL, neurofilament light chain; GFAP, glial fibrillary acidic protein; Aβ-PET, amyloid-beta positron emission tomography.

This table presents the β-estimates, 95% confidence intervals (CI), and p-values for the association between standardized eGFR and various plasma AD biomarkers, including Aβ42, Aβ40, p-tau217, p-tau231, NfL, GFAP, and the Aβ42/Aβ40 ratio. The results are shown for three different models: univariate (no adjustments), adjusted for age and sex, and adjusted for age, sex, and Aβ-PET status. The p-values indicate the statistical significance of these associations, with values below 0.05 considered significant.

Table 4. Summary of AUC and AIC Metrics for Plasma Biomarkers in Predictive Models of Aβ-PET Positivity

			Confidence Interval 95%
Biomarker	Model	AUC	Lower Limit	Upper Limit	AIC	AICc	p value (de long)	ΔAICc	ΔAICc with eGFR
Plasma Aβ42	Univariate	0.659	0.594	0.724	336.07	336.16		0
Plasma Aβ42	Adjusted (+ age + sex)	0.747	0.689	0.805	309.63	309.86	0.001	-26.30
Plasma Aβ42	Full (+ age + sex + eGFR)	0.754	0.696	0.811	310.32	310.64	0.000	-25.52	0.78
Plasma Aβ40	Univariate	0.568	0.498	0.637	356.74	356.83		0
Plasma Aβ40	Adjusted (+ age + sex)	0.634	0.568	0.701	337.28	337.51	0.137	-19.32
Plasma Aβ40	Full (+ age + sex + eGFR)	0.635	0.569	0.702	338.95	339.27	0.126	-17.56	1.76
Plasma Aβ42/40	Univariate	0.747	0.689	0.806	311.21	311.30		0
Plasma Aβ42/40	Adjusted (+ age + sex)	0.764	0.707	0.821	304.31	304.54	0.235	-6.77
Plasma Aβ42/40	Full (+ age + sex + eGFR)	0.764	0.707	0.821	306.22	306.54	0.225	-4.76	2.01
Plasma p-tau181	Univariate	0.758	0.701	0.814	329.48	329.57		0
Plasma p-tau181	Adjusted (+ age + sex)	0.745	0.687	0.804	317.53	317.77	0.513	-11.81
Plasma p-tau181	Full (+ age + sex + eGFR)	0.745	0.686	0.803	318.29	318.61	0.526	-10.96	0.85
Plasma p-tau217	Univariate	0.896	0.853	0.939	236.73	236.83		0
Plasma p-tau217	Adjusted (+ age + sex)	0.900	0.860	0.940	234.13	234.38	0.613	-2.44
Plasma p-tau217	Full (+ age + sex + eGFR)	0.899	0.858	0.940	233.36	233.72	0.781	-3.11	-0.66
Plasma p-tau231	Univariate	0.754	0.694	0.814	323.00	323.10		0
Plasma p-tau231	Adjusted (+ age + sex)	0.763	0.707	0.820	307.14	307.38	0.673	-15.72
Plasma p-tau231	Full (+ age + sex + eGFR)	0.771	0.715	0.826	306.70	307.02	0.482	-16.07	-0.35
Plasma NTA-tau	Univariate	0.644	0.573	0.714	342.93	343.02		0
Plasma NTA-tau	Adjusted (+ age + sex)	0.708	0.646	0.770	316.48	316.71	0.084	-26.30
Plasma NTA-tau	Full (+ age + sex + eGFR)	0.710	0.648	0.772	316.95	317.28	0.083	-25.74	0.56
Plasma NfL	Univariate	0.664	0.600	0.727	351.77	351.86		0
Plasma NfL	Adjusted (+ age + sex)	0.631	0.565	0.697	337.23	337.46	0.290	-14.40
Plasma NfL	Full (+ age + sex + eGFR)	0.638	0.572	0.704	338.14	338.46	0.421	-13.39	1.00
Plasma GfAP	Univariate	0.795	0.743	0.847	294.03	294.12		0
Plasma GfAP	Adjusted (+ age + sex)	0.796	0.744	0.848	292.48	292.71	0.938	-1.41
Plasma GfAP	Full (+ age + sex + eGFR)	0.801	0.749	0.854	291.14	291.46	0.623	-2.67	-1.26

This table presents a detailed comparison of area under the curve (AUC), Akaike Information Criterion (AIC), and ΔAIC values across different predictive models for Aβ-PET positivity, using various plasma biomarkers including Aβ42, Aβ40, the Aβ42/40 ratio, P-tau181, P-tau217, P-tau231, and NfL. The models considered include univariate models (plasma biomarkers alone), models adjusted for age and sex, and fully adjusted models that also include eGFR.

For each biomarker, the AUC values are provided along with their confidence intervals (95% CI) to indicate the model's discriminative ability, AIC and AICc values are presented to assess model fit, with ΔAIC showing the improvement gained by adding age, sex, and eGFR to the models. The p-values from DeLong's test for comparing AUCs are also included to assess the statistical significance of the differences between models.

eGFR, estimated glomerular filtration rate; Aβ42, amyloid-beta 42; Aβ40, amyloid-beta 40; p-tau181, phosphorylated tau at threonine 181;p-tau217, phosphorylated tau at threonine 217; p-tau231, phosphorylated tau at threonine 231; NTA-tau, N-terminal containing tau fragments; NFL, neurofilament light chain; GFAP, glial fibrillary acidic protein; Aβ-PET, amyloid-beta positron emission tomography.

Competing interest reported. NJA has given lectures, produced educational materials, and participated in educational programs for Eli-Lily, BioArtic, Quanterix. KB has served as a consultant and at advisory boards for Abbvie, AC Immune, ALZPath, AriBio, BioArctic, Biogen, Eisai, Lilly, Moleac Pte. Ltd, Neurimmune, Novartis, Ono Pharma, Prothena, Roche Diagnostics, Sanofi and Siemens Healthineers; has served at data monitoring committees for Julius Clinical and Novartis; has given lectures, produced educational materials and participated in educational programs for AC Immune, Biogen, Celdara Medical, Eisai and Roche Diagnostics; and is a co-founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program, outside the work presented in this paper. HZ has served at scientific advisory boards and/or as a consultant for Abbvie, Acumen, Alector, Alzinova, ALZPath, Annexon, Apellis, Artery Therapeutics, AZTherapies, CogRx, Denali, Eisai, Nervgen, Novo Nordisk, Optoceutics, Passage Bio, Pinteon Therapeutics, Prothena, Red Abbey Labs, reMYND, Roche, Samumed, Siemens Healthineers, Triplet Therapeutics, and Wave, has given lectures in symposia sponsored by Cellectricon, Fujirebio, Alzecure, Biogen, and Roche, and is a co-founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program (outside submitted work).

Arslanetalsupplement.docx

The Impact of Kidney Function on Alzheimer’s Disease Blood Biomarkers: Implications for Predicting Amyloid-β Positivity

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Introduction

Methods

Results

Discussion

Abbreviations

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1