Plasma NfL Research Articles

Plasma neurofilament light and brain volumetric outcomes among middle‐aged urban adults

AbstractBackgroundElevated plasma neurofilament light chain (NfL) is associated with dementia, but the specific underlying mechanisms are largely unknown. We examined cross‐sectional and longitudinal relationships of plasma NfL with selected brain structural magnetic resonance imaging (sMRI) prognostic markers of dementia, namely global and cortical brain, hippocampal and white matter lesion volumes (WMLV).MethodData were extracted from three prospective visits within the Healthy Aging in Neighborhoods of Diversity Across the Life Span (HANDLS) study, selecting participants with complete v1 (2004‐2009) and v2 (2009‐2013) plasma NfL exposure and ancillary sMRI data at vscan (2011‐2015, n = 179, mean v1 to vscan time: 5.4y). Multivariable‐adjusted linear regression models were conducted, overall, by sex, and by race, correcting for multiple testing with q‐values.ResultNfL(v1) (Loge transformed) had a moderately strong association with larger WMLV at follow‐up vscan visit, after a period of 5‐6 years of follow‐up, overall (β = +2.131±0.660, b = +0.29, p = 0.001, q = 0.0029). Moreover, a 1 unit increase in NfLv2 was linked to a left hippocampal volume that was lower by 125 mm3 (p = 0.004, q = 0.015), using minimally adjusted models and mainly among males. Associations were largely unaltered by further adjustment for downstream health‐related factors.ConclusionIn summary, plasma NfL may be an independent medium‐term (within 5 years) marker for increase in WMLV, and a short term (within 1 year) marker for smaller hippocampal volumes normalized to the intracranial volume, independently of other confounding factors. Therefore, plasma NfL holds promise as a blood‐based marker to predict future subclinical brain pathologies in middle‐aged adults.

Prognostic value of plasma biomarkers in a clinical trial of mild‐to‐moderate Alzheimer’s Disease

AbstractBackgroundInformative, readily accessible plasma biomarkers of pathology, neuroinflammation, and neurodegeneration in Alzheimer’s disease (AD) could enhance targeted approaches to AD trial design and treatment. In post‐hoc analyses, we examined whether plasma biomarkers (Aβ1‐40, Aβ1‐42, total tau, p‐tau‐181, NfL, and GFAP) predicted 48‐week change in cognitive (ADAS‐Cog11), global function (CDR‐SB), and volumetric MRI (6 cortical regions and 2 whole brain measures) outcomes in T2 Protect AD, a phase‐2 placebo‐controlled RCT of troriluzole in mild‐to‐moderate AD (NCT03605667).MethodAt least one baseline biomarker measurement was available for 319 of 350 trial participants, forming the analytic sample. No significant treatment effects were found in T2 Protect AD (Feldman 2021, JPAD 8(4):s46); thus, treatment arms were pooled for this analysis. Baseline levels of, and 48‐week changes in, plasma biomarkers were assessed for association with 48‐week change in outcomes using linear regression adjusted for relevant covariates. Combinations of baseline plasma biomarkers that best predicted 48‐week decline on the ADAS‐Cog11 and CDR‐SB were identified using LASSO regression. Statistical significance level was 5%; p‐values were Bonferroni corrected for multiple comparisons.ResultParticipant characteristics are presented in Table 1. Higher baseline plasma NfL predicted greater 48‐week decline on ADAS‐Cog11 (p = 0.026) and CDR‐SB (p = 0.048; Fig. 1A). LASSO revealed that the combination of baseline plasma NfL, total tau, and Aβ42/40 ratio best predicted 48‐week decline on ADAS‐Cog11, whereas baseline NfL alone best predicted 48‐week decline on CDR‐SB. Regarding MRI outcomes, baseline NfL predicted increase in ventricular volume (p = 0.018; Fig 1A). Baseline NfL, GFAP, and p‐tau‐181 each predicted 48‐week decline in mid‐temporal cortical volume (all p<0.02; Fig.1B); LASSO results were similar. The only significant association between 48‐week biomarker change and clinical outcomes was between increased plasma NfL and worsening CDR‐SB.ConclusionElevated baseline plasma NfL predicted greater 48‐week decline on cognitive, global function, and MRI measures in a clinical trial of mild‐to‐moderate AD. Furthermore, greater increase in plasma NfL over time was associated with greater clinical decline. Plasma NfL is an easily accessible biomarker that may enhance AD clinical trial design and treatment strategies.Acknowledgement: Trial funding by Biohaven Pharmaceuticals; data and trial coordination by the Alzheimer’s Disease Cooperative Study (ADCS).

Relationship between neurofilament light chain and cognitive decline moderated by amyloid‐β status

AbstractBackgroundIn this study, we identified whether amyloid‐β (Aβ) moderates the relationship between the plasma level of neurofilament light chain (NfL) and the change of cognition or brain volume.MethodOf 723 non‐demented participants from the Alzheimer’s Disease Neuroimaging Initiative (mean age 72.1), 341 (47.2%) had cerebral Aβ deposition (Aβ [+]) and 382 (52.8%) had not (Aβ [‐]) determined by Florbetapir PET. Longitudinal changes of cognition was assessed by Alzheimer’s Disease Assessment Scale‐Cognitive Subscale (ADAS‐Cog) score, and hippocampal and white matter hyperintensity volume by brain MRI.ResultAβ status moderated the relationship between plasma NfL and ADAS‐Cog score (beta = 0.002, P = 0.029), hippocampal volume (beta = ‐0.073, P = 0.004), and white matter hyperintensity volume (beta = 0.673, P = 0.048).ConclusionThese results suggest the possible interactive effect between plasma NfL and brain Aβ on the longitudinal cognition and brain structure.

The generalizability of proposed reference values for plasma neurofilament light chain in cognitively healthy adults in Thailand

AbstractBackgroundDespite being an established biomarker for quantifying neuroaxonal injury in research, the clinical utility of neurofilament light chain (NFL) for assessing individual patients is limited, primarily due to the confounding effect of aging on its levels. Although many studies reported age‐specific cutoff values in neurologically healthy population of European descent in high‐income countries, evidence suggests that the value of neurodegenerative biomarkers may differ in ethnically and socioeconomically underrepresented populations. In this study, we aimed to evaluate the acceptability of the reported NFL cutoff values for local clinical use by examining specimens from healthy Thai cohorts.MethodThis study enrolled participants without neurological disorder and normal cognitive examination from two distinct cohorts in Bangkok, Thailand, ranging in age from 40 to 90 years old. Plasma levels of NFL were quantified using the single molecule array. By utilizing reference data from previous studies (Simrén et al., 2022; Bornhorst et al., 2022), the proportion of participants exceeding the proposed cutoff values were estimated and compared to the expected values. Furthermore, the cohort was stratified by age group and methods similar to prior studies were employed to determine age‐specific cutoff values.ResultOne hundred and three participants were included, comprising of 69 females (67%), with a median age of 65.4 years (interquartile range (IQR): 56.7, 70.3) and a cognitive z‐score of 0.11 (IQR: ‐0.44, 0.63). Using the reference threshold from prior studies, plasma NFL were elevated in 4 participants (3.9%, 95% confidence interval: 1.0‐9.6%) which is consistent with the expected prevalence of 5% (as 95th percentile cutoffs were used). Subsequently, we determined age‐specific plasma NFL cut‐off values for each decade through quantile regression, values ranging from 7.2 pg/mL for 40‐50 years to 58.2 pg/mL for >80 years.ConclusionOur initial findings suggest that the previously established cutoffs for plasma NFL may be applicable in Thailand. Nonetheless, further studies with a larger and more diverse sample size may be necessary in order to fully validate the generalizability of these cutoffs and potentially establish more specific thresholds for local use.

A CLU/APOJ GWAS AD risk variant suppresses the astrocytic response to plaques and reduces axonal and neuritic damage in 5xFAD mice

AbstractBackgroundGenome‐Wide Association Studies (GWAS) implicate clusterin (CLU, also known as apolipoprotein J) as a risk factor for late‐onset Alzheimer’s disease (LOAD), with elevated expression of CLU being associated with increased risk of LOAD. The region surrounding the SNP risk allele rs2279590 is poorly conserved within the mouse genome. Consequently, to model the effect of the rs2279590 risk allele on development of AD‐related pathology in mice we developed mice with a partially humanized allele of Clu.MethodCRISPR was used to substitute a 2kb region of human CLU (intron 7 to exon 9) containing the enhancer region and the rs2279590 SNP variant for the orthologous region of Clu within the C57BL/6J mouse genome. Clu‐rs2279590_h2kb mice were crossed with 5xFAD mice and aged to 4 and 12. Coronal brain sections were immunolabeled to visualize dense‐core plaques (ThioS), microglia (IBA1), astrocytes (S100β and GFAP), axonal (NfL) and neuritic (LAMP1) damage, confocal images of subiculum and cortical regions were taken and analyzed using Imaris software. In addition, brain Aβ and plasma NfL levels were quantified using meso scale discovery technology.Result5xFAD;Clu‐rs2279590_h2kb showed significantly reduced microglial density compared to 5xFAD mice in both brain areas at 4 months. By 12 months of age, microglial density was comparable between genotypes in cortex while in subiculum, elevated IBA1 cell density was observed in 5xFAD;Clu‐rs2279590_h2kb mice. Clu‐rs2279590_h2kb variant in 5xFAD background increased the density of S100β astrocytes in subiculum at 4 months and decreased the total S100β volume at 12 months. Reduced GFAP (reactive astrocytes) levels were found in cortex but not subiculum at 12 months. Despite comparable plaque load and insoluble Aβ42 levels, neuritic and axonal damage was decreased in the brain of 5xFAD;Clu‐rs2279590_h2kb compared to 5xFAD mice at 12 months. Plasma NfL was also significantly decreased.ConclusionDespite the minimal impact on Aβ plaque deposition, humanization of 2kb of Clu allele and introduction of rs2279590 risk SNP suppresses astrocytic response to plaques and is protective against axonal and neuritic damage which is mirrored by the reduced levels of plasma NfL.

Effect of calcineurin/NFAT inhibitors on biomarkers and cognition in a canine model of Alzheimer disease

AbstractBackgroundCalcineurin/NFAT signaling increases in Alzheimer disease (AD) and is associated with neurodegeneration, neuroinflammation, amyloid‐β (Aβ) production, and cognitive decline. Hence, calcineurin/NFAT inhibitors (e.g., tacrolimus and Q134R) are attractive potential therapeutics to improve/prevent AD‐associated cognitive impairments. In AD mouse models, both tacrolimus and Q134R improved cognition and ameliorated neuroinflammation and synapse dysfunction.We hypothesized that 1) chronic low‐dose administration of tacrolimus and Q134R will delay cognitive decline over time in middle‐aged beagles, a model of human brain aging and AD; and 2) canine plasma and CSF biomarkers will be improved with treatment.MethodsCognitive outcomes were evaluated in 37 middle‐aged beagles at baseline and during follow‐up (years 1‐3) by assessing spatial learning [landmark‐discrimination (1, 2 and 4cm)], visual discrimination learning, reversal learning (executive function), and spatial working memory (20‐, 70‐, and 110‐second delays). Dogs were ranked for cognition at baseline and randomized into treatment groups: Control (n = 12; placebo), Q134R (n = 12; 8mg/day), and tacrolimus (n = 13; 0.075mg/kg/day). Concentrations of Aβ40, Aβ42, and NfL in plasma and CSF were measured using the Quanterix SIMOA.ResultsAccuracy at all landmark‐discrimination distances declined significantly among controls 20% faster than tacrolimus, and 15% faster than Q134R dogs, at the most difficult distance of 4cm. Accuracy on spatial working memory task declined 21% faster/year among controls than tacrolimus (years 2&3); and 18% faster than Q134R dogs (year 1). Performance on visual discrimination and reversal tasks declined with age but there was no treatment effect.Biomarkers in plasma tended to increase over time with no significant differences between treatment groups. In CSF, relative decreases in Aβ40 and Aβ42 in tacrolimus (2.9 and 2, respectively) and Q134R (5.4 and 3.7, respectively) were less than in control dogs. Finally, relative increases of NfL in tacrolimus and Q134R were less than in control dogs (2.3 and 3.7, respectively).ConclusionBoth tacrolimus and Q134R showed beneficial preventative effects on age‐related decline in spatial functions and memory but not on associative learning nor executive function in a canine model of AD. Finally, only CSF biomarkers tended to improve relative to treatment and are suggestive of less Aβ deposition in brain and higher clearance.

Physiological Confounders of Blood‐Based Alzheimer’s Disease Biomarkers in Middle‐Aged Asymptomatic Individuals

AbstractBackgroundBlood‐based biomarkers represent a promising tool for Alzheimer’s disease (AD) clinical diagnosis and patient management. However, the identification of physiological factors that may influence their concentration levels in plasma is fundamental to warrant their clinical use, but more data is needed on whether such possible confounders result in clinically relevant changes. Here, we aim to identify physiological confounders of blood‐based AD biomarkers in a middle‐aged asymptomatic population.MethodThe Progression of Early Subclinical Atherosclerosis (PESA) study is a longitudinal study following >4000 asymptomatic middle‐aged individuals. This study included 537 PESA participants, 370 of them with follow‐up data 4.7±0.6 years later. The concentrations of Aβ40, Aβ42, p‐tau181, GFAP and NFL in plasma were quantified using the Simoa pTau‐181 V2 and Neurology 4‐Plex E Advantage Kits. Potential confounders were also measured and included body mass index (BMI), as a proxy of blood volume; creatinine, as of kidney function; and aspartate (AST) and alanine aminotransferase (ALT), as of liver function. Stepwise regression was used to determine which confounders (predictors) are associated with plasma biomarkers (outcomes), adjusting for age, sex and APOE, both cross‐sectionally and in annual change rates (p<0.05). Participants were classified into normal and abnormal, based on standard cut‐off values for all potential confounders. Between‐group differences in plasma biomarkers were sought, and the abnormal group’s fold‐change was calculated.ResultThe studied sample had a mean age of 50.4±4.3 years, 441 were males and 88 APOE‐ε4 carriers (Figure 1). All blood‐based biomarkers but Aβ42/40 were negatively correlated with BMI and positively correlated with creatinine (Figure 2). ALT and AST were associated with Aβ40, Aβ42 and p‐tau181 but with opposite effects. Increases in BMI were associated with a decline in NFL and GFAP, while an increase in AST with a decline in amyloid and tau accumulation in plasma. Abnormally high BMI was associated with lower biomarker plasma concentrations, except for in Aβ42/40, with a maximum fold‐change of ‐12% for GFAP (Figure 3).ConclusionBMI, kidney and liver function are associated with plasma brain biomarkers levels in asymptomatic middle‐aged individuals. However, the magnitude of their effect is minor, with limited impact for their clinical applicability in individual patients.

Association Between Neuroimaging Biomarkers, Plasma NfL, and Memory Performance in Adults With Down Syndrome With and Without Dementia

AbstractBackgroundDown syndrome (DS) is associated with early development of Alzheimer’s disease pathology. Both the severity of tau pathology and the increase of neurofilament light chain (NfL) protein are correlated with cognitive decline in the neurotypical populations. Here, we assessed whether tau accumulation in the hippocampus and entorhinal cortex and plasma NfL concentrations were associated with memory performance in adults with DS.MethodWe used 18F‐AV‐1451 (FTP) positron emission tomography (PET) to assess tau accumulation in 44 participants enrolled in the multi‐site Alzheimer’s Disease in Down syndrome (ADDS) study (age 50.72 + 6.21). FTP‐PET scans were partial volume‐corrected, and weighted standardized uptake value ratio (SUVR) were calculated for each of the following regions of interests (ROIs): entorhinal cortex, hippocampus, and the precentral gyrus (control). Participants' plasma NfL concentration (in pg/mL) was measured on a single plex plate using the ultra‐sensitive single molecule array (Simoa) technology platform HD‐X. Memory performance was measured with the modified Cued Recall Test. We evaluated the relationship between FTP SUVRs, NfL, and memory performance using linear regression analysis. Sex and site were used as covariates.ResultIncreased FTP SUVR in the hippocampus and entorhinal cortex were associated with lower scores on free recall (hippocampus: r2 = 0.53, p <0.001; entorhinal: r2 = 0.35, p <0.001) and total recall (hippocampus: r2 = 0.46, p <0.001; entorhinal: r2 = 0.29, p <0.01). As expected, no association between FTP SUVRs and memory performance was found in the precentral gyrus. Additionally, higher plasma NfL concentrations were associated with lower free recall (r2 = 0.35, p <0.001) and total recall (r2 = 0.29, p <0.01). Further, increased levels of plasma NfL were associated with increased FTP SUVRs in the hippocampus (r2 = 0.38, p <0.001), entorhinal cortex (r2 = 0.28, p<0.01), and precentral gyrus (r2 = 0.19, p<0.05).ConclusionIncreased levels of both in‐vivo tau in medial temporal ROIs and NfL in plasma were associated with worse memory performance, consistent with observations in the neurotypical population. Future work will investigate the impact of tau accumulation and NfL concentration on longitudinal cognitive decline in older adults with DS.

Differentiation of clinically diagnosed patients with aMCI and dementia from cognitively unimpaired older adults using plasma and serum p‐tau 181 and NfL

AbstractBackgroundBlood‐based biomarkers give the possibility of having a less invasive, more acceptable to patients, and cost‐ and time‐efficient screening tool. Therefore, blood‐based biomarkers represent an ideal addition to cerebrospinal fluid (CSF) and neuroimaging biomarkers. Standard examination of neurocognitive disorders, such as lumbar punction or brain PET scan, cannot be performed within general practice. Given the high prevalence of these diseases, alternatives such as blood‐based biomarkers would represent a significant advance. The aim of this study was to investigate whether levels of p‐tau 181 and neurofilament light chain (NfL) in serum and plasma can differentiate cognitively unimpaired (CU) elderly individuals from individuals in the stage of aMCI and dementia.MethodA total of 75 biomarker‐defined individuals from the Czech Brain Aging Study were recruited for the study, including the CU, individuals in the aMCI stage, and individuals in the stage of dementia syndrome with various etiology (AD, FTLD or LBD). Commercial ELISA kits were used for the assessment of standard biomarkers from CSF (Aß42, Aß40, pTau, tTau). Plasma and serum samples were analyzed using a Quanterix SR‐X analyzer with Simoa technology. Receiver operating characteristic analyses of the concentrations of p‐tau 181 and NfL in plasma and serum were performed using the R statistical software and GraphPad Prism.ResultPlasma NfL, AUC 0.9861 (95% CI, 0.9439 to 1.000), and serum NfL with AUC 0.9583 (95% CI, 0.8774 to 1.000) differentiated aMCI and dementia from CU with high predictive accuracy. The results were significant for plasma p‐tau 181 too with AUC 0.7869 (95% CI, 0.6474 to 0.9265), along with serum p‐tau 181 with AUC 0.7623 (95% CI, 0.6292 to 0.8954).ConclusionPlasma and serum p‐tau 181 and NfL may be potentially useful and cost‐effective biomarkers for differentiating cognitively unimpaired elderly individuals from those at aMCI and dementia stages in general health care settings.

Plasma biomarkers of Alzheimer´s disease in the continuum of dementia with Lewy bodies

AbstractBackgroundPatients with dementia with Lewy bodies (DLB) may have Alzheimer´s disease (AD) pathology. Plasma biomarkers of beta‐amyloid (Aβ), phosphorylated tau (pTau), and neurodegeneration are sensitive to AD neuropathologic changes in AD dementia. While these plasma biomarkers are well tested in the AD continuum, their performance for concomitant AD pathology in the DLB continuum is still unclear.MethodWe included patients with isolated REM sleep behavior disorder (iRBD; n = 15), with mild cognitive impairment with Lewy bodies (MCI‐LB; n = 37), and with DLB (n = 70) from the Mayo Clinic Alzheimer’s Disease Research Center. Cognitively unimpaired individuals (CU; n = 100) included from the Mayo Clinic Study of Aging were balanced on age and sex. A panel of plasma biomarkers of Aβ40, Aβ42, p‐tau181, neurofilament light chain (NfL), and glial fibrillary acidic protein (GFAP) measures was obtained. We calculated the ratio of plasma Aβ40/42. Lower Aβ40/42 indicates increased Aβ pathology. Plasma Aβ40, Aβ42, GFAP, and NfL were measured using the Simoa® Neurology 4‐Plex E Advantage kit and pTau‐181 was measured with the Simoa® pTau‐181 Advantage V2 kit. Both were run on a Quanterix HD‐X analyzer. AD imaging biomarkers were assessed on PET using Pittsburgh compound B (PiB) for Aβ and 18F‐Flortaucipir for tau.ResultHigher levels of plasma GFAP were observed in the MCI‐LB (p = 0.014) and DLB (p<0.001) groups compared to CU (Figure). DLB patients also showed higher levels of plasma pTau181 (p<0.001) and NfL (p<0.001). Higher amyloid and tau PET standardized uptake value ratio (SUVr) in the DLB continuum were associated with higher levels of plasma p‐tau181, NfL, and GFAP (p<0.001). After accounting for age, plasma Aβ40/42 did not correlate with amyloid or tau PET SUVr.ConclusionIn the DLB continuum, abnormal levels of GFAP in plasma can be detected as early as the prodromal stages of the disease. Plasma pTau181 reached abnormal levels in the DLB stage. Higher plasma GFAP, NfL, and p‐tau181 are associated with higher amyloid and tau PET SUVr in the DLB continuum. Plasma biomarkers have the potential to contribute to screening and early diagnosis in the DLB continuum, which has implications in designing new treatments.

The Association of Plasma Biomarkers and Survival in the Oldest Old

AbstractBackgroundPrevious research showed that plasma NfL (neurofilament light chain) associated with survival. Here we perform exploratory analyses to associate a broader range of plasma biomarkers Amyloid‐βeta‐40 (Aβ40), Amyloid‐βeta‐42 (Aβ42), phosphorylated Tau‐181 (pTau‐181), NfL and glial fibrillary acidic protein (GFAP), with survival after reaching 100 years.MethodThe longitudinal Dutch 100‐plus study cohort comprises of self‐reported cognitively healthy centenarians. Biomarkers were measured in 202 centenarians (137 females, ISCED education level median [IQR] 3 [2–4]) on the Simoa platform. Information on survival was obtained via family report until January 2023. Centenarians who were alive were censored at date of last visit or date last confirmed alive. We investigated the relationship between biomarkers and survival by cox regression analyses, adjusting for age at blood draw, sex and cognitive functioning. We performed maximally selected log‐rank statistics between all plasma biomarkers and survival rate to find an optimal cut‐off, and constructed Kaplan‐Meier survival curves.ResultOf 202 centenarians (median age [IQR] at blood draw 101.4 [100.4‐102.9]), 173 (85.6%) were deceased (median age [IQR] at death 103.2 [102.0‐104.3]). The median (IQR) survival time after blood draw was 14.4 (5.5‐25.1) months. All biomarkers Aβ40, Aβ42, pTau‐181, NfL and GFAP intercorrelated (rs = 0.17–0.79; all P‐values<0.05). Cox regression analyses showed that for each standard deviation increase in NfL, adjusted for age, sex, and MMSE score, the risk of mortality increased by 28% (HR 1.28; 95% CI, 1.11–1.47; P<0.001), Table 1. Maximally selected log‐rank statistics split up NfL into survival groups high and low (cut‐off 70.34 pg/mL, M = 3.803, P = 0.003, Figure 1). The median (IQR) shorter survival time of high versus low NfL was 7.8 (5.38‐9.27) months. No statistically significant cut‐off values were identified for the remaining biomarkers.ConclusionOur results indicate that elevated levels of NfL, but not Aβ40, Aβ42, pTau‐181 and GFAP, associate with lower survival. Furthermore, we were able to find an optimal cut‐off for NfL at which survival in centenarians could be split twofold.

Plasma biomarkers as prognostic markers in Alzheimer’s disease: Influence of age at onset

AbstractBackgroundLittle is known about the influence of age at onset (AAO) on plasma biomarkers and their use as prognostic biomarkers in Alzheimer’s disease (AD).MethodWe selected patients with AD diagnosis with available neuropsychological testing (NPS) at time of diagnosis and two years later, and plasma biomarkers at baseline.NPS battery included Free and Cued Selective Reminding Test (FCSRT), Landscape test (visual memory), Boston Naming Test, Semantic Fluency, BDAE auditory comprehension, Constructional and Ideomotor Praxis, Visual Object and Space Perception (VOSP) Incomplete Letters and Number Location subtests, Trail Making Test (TMT) A and B, Phonemic Fluency, and Digit Span Forward and Backward. NPS scores were compared by AAO: early‐onset AD (EOAD; <65 years) vs. late onset AD (LOAD; >65y).We analyzed plasma biomarkers phosphorilated‐tau181 (p‐tau181), total tau (t‐tau), neurofilament light chain (NfL), glial fibrillary acidic protein (GFAP) and ubiquitin C‐terminal hydrolase L1 (UCHL‐1) using the Quanterix Simoa p‐tau181 Advantage V2 and Neurology 4‐Plex A assays. Group comparisons and linear regressions adjusted by years of education (YOE) were performed in Stata/IC 16.1.ResultForty‐two participants were included, 23 LOAD and 19 EOAD. Plasma p‐tau181 and GFAP levels were higher in LOAD (Table 1). We did not find differences between LOAD and EOAD in NPS tests at baseline or +2 years (Table 2).Plasma ptau‐181 was associated with progression in MMSE globally, VOSP‐Incomplete letters globally and in EOAD. Plasma NfL were associated to Boston Naming test globally and in EOAD, Semantic fluency test globally, VOSP‐incomplete letters in EOAD, and Free and Total Learning of FCSRT in LOAD. Plasma GFAP was associated to MMSE globally and in EOAD, Free learning of FCSRT in LOAD and VOSP‐Incomplete letters and number location globally. Plasma UCHL‐1 was associated to Semantic fluency test in LOAD (Table 3). Praxis and attention and executive function tests loss were not associated to plasma biomarkers.ConclusionPlasma p‐tau18, NfL, GFAP and UCHL‐1 were associated to pogression in memory, language and visual tests. They were predominantly associated to memory loss in LOAD and language and visual function loss in EOAD. Results need to be interpreted cautiously due to small sample size.

Towards the future of ADRD diagnosis: combination of a brief digital cognitive battery and plasma biomarkers for diagnosis and prognosis of disease

AbstractBackgroundScalable and accurate diagnostic markers of Alzheimer’s disease and related dementias (ADRD) are urgently needed. A combination of a sensitive brief digital cognitive assessment and plasma biomarkers could represent a promising scalable alternative to standard diagnostic procedures. We evaluated the performance of the TabCAT Brain Health Assessment Cognitive Score (BHA‐CS) and plasma Aβ42/40, pTau‐181, NfL, and GFAP at predicting concurrent Aβ positivity on CSF or PET, concurrent disease severity, and future functional decline.MethodParticipants were 309 subjectively or objectively impaired older adults enrolled in longitudinal observational studies at UCSF (Table 1), who completed a blood draw and TabCAT‐BHA at baseline. Biomarker concentrations were measured using commercially available Quanterix kits. Receiver operating characteristic analyses were used to determine the accuracy of combined BHA‐CS and plasma biomarkers in discriminating between Aβ+/‐ status on CSF/PET at baseline. Multiple regressions were used to examine baseline associations of BHA‐CS and plasma markers with disease severity measured by Clinical Dementia Rating Sum of Boxes (CDR‐SB). Linear mixed effect models were used to examine the associations of baseline BHA‐CS and biomarker values with longitudinal decline on CDR‐SB. All analyses controlled for age, sex, education, and baseline diagnosis.ResultpTau‐181 was the strongest predictor of Aβ positivity (AUC = .745; 95%CI: .656‐.833), with no significant improvements in the AUC with the addition of any combination of markers (Figure 1). In a combined model with all markers included simultaneously, BHA‐CS (B = ‐0.678, P<.001) and plasma NfL (B = 0.625, P = .01) were the only unique significant predictors of concurrent CDR‐SB. In a combined mixed model with all markers included simultaneously, baseline BHA‐CS, pTau‐181, NfL, and GFAP were unique significant predictors of CDR‐SB decline (Figure 2), and together accounted for 27% of longitudinal variance in CDR‐SB.ConclusionOur findings indicate unique and complementary roles of scalable digital cognitive and plasma biomarkers as diagnostic and prognostic markers of ADRD. At baseline, whereas pTau‐181 was most accurate at predicting Aβ positivity, a combination of BHA‐CS and NfL was most accurate at predicting concurrent disease severity. In longitudinal analyses, a model that combined baseline cognitive testing results and plasma pTau‐181, NfL, and GFAP was most accurate at predicting future decline.

Association of Cardiometabolic Multimorbidity with Dementia and Alzheimer’s Plasma Biomarkers in Older Adults: a population‐based study

AbstractBackgroundIndividual cardiometabolic disorders are associated with an increased dementia risk. However, the relationship of cardiometabolic multimorbidity (CMM) with dementia has not been studied in Chinese older adults and the underlying mechanisms are poorly understood.MethodThis population‐based study included 5665 rural‐dwelling adults (age ≥60 years, 57.3% women) in the baseline examination (March‐September 2018) of the Multimodal Interventions to Delay Dementia and Disability in Rural China study. Cardiometabolic diseases (CMDs), which included heart disease, stroke, type 2 diabetes, and obesity, were assessed through clinical and neurological examinations. CMM was defined as the concurrent presence of two or more CMDs. Amyloid‐β(Aβ), total tau, and neurofilament light chain (NfL) were measured using Simoa platform in a subsample (n = 1434). Dementia, Alzheimer’s disease (AD), and vascular dementia (VaD) were clinically diagnosed following international criteria. Data were analyzed using the general linear and logistic regression and mediation models.ResultOf the 5665 participants, 1091 (19.26%) were detected with CMM, and 302 (5.33%) were diagnosed with dementia (193 with AD and 98 with VaD). The multivariable‐adjusted odds ratios of CMM (vs. no CMD) were 3.00 (95% confidence interval [CI] 2.19‐4.11) for all‐cause dementia, 1.66 (1.11‐2.47) for AD, and 12.5 (6.47‐24.18) for VaD. There was a significant interaction between CMM and sex on the likelihood of dementia and subtypes (P for interaction<0.05), such that the associations of CMM with dementia and subtypes were stronger in men than in women. In the biomarker subsample, CMM was significantly associated with higher plasma NfL (demographic‐adjusted β coefficient = 0.059; 95% CI 0.028‐0.091), Aβ40 (0.034; 0.018‐0.049), and Aβ42 (1.068; 0.636‐1.501); the observed association remained significant even in the multivariable‐adjusted model (P<0.001). The mediation analysis showed that plasma NfL could mediate 19.46% of the association between CMM and AD (P<0.05), whereas plasma Aβ and total tau showed no significant mediation (P>0.05).ConclusionCMM is associated with increased likelihoods of dementia, AD, and VaD in older adults, especially in men. Neurodegenerative pathway might play a part in the association of CMM with AD.

Refinement of Neurofilament light Dynamics in CSF and Blood for familial Alzheimer’s Disease

AbstractBackgroundDisease‐modifying therapies for Alzheimer’s Disease (AD) are likely most beneficial when initiated in the pre‐symptomatic phase. To track success of such interventions fluid biomarkers became instrumental, with neurofilament light chain (NfL) showing particular promise. We previously reported that serum NfL increases in pre‐symptomatic phases of familial (autosomal‐dominantly inherited) AD (fAD) and that within‐person rate‐of‐change in NfL is an earlier predictor of fAD symptom onset compared to cross‐sectional levels. The aim of the study at hand was to extend these initial findings in a larger cohort.MethodParticipants enrolled in the Dominantly Inherited Alzheimer Network (DIAN) with matched cross‐sectional and longitudinal cerebrospinal fluid (CSF; n = 962) as well as plasma (n = 1294) samples were used. NfL measurements have been performed on the SIMOA platform using commercially available assay‐kits. To investigate the influence of physiological factors independent from AD, we first analyzed whether age and body mass index (BMI) could account for observed variation in inter‐individual CSF and plasma NfL levels considering non‐carrier family‐members (NC) as a healthy control group. Next, utilizing estimated years to symptom onset (EYO) and previously published methods (Preische et al., 2019), we determined the point in disease course when baseline and longitudinal CSF and plasma NfL concentrations started to increase in mutation carriers (MC) relative to NC, after adjusting for age and BMI.ResultOur results reveal a tight correlation of CSF and blood NfL values within MC as well as NC (figure 1A‐B). However, within NC, after correction for age and BMI, some unexplained variability remained (figure 1C). Further, a decrease of the NfL plasma/CSF ratio over age was found after correcting for BMI (figure 1D). The discrimination of MC from NC was equally good for plasma compared to CSF, being possible on the group level as early as around ‐15 to almost ‐20 EYO depending on the modelling (figure 2).ConclusionOur results support plasma equivalently to CSF NfL as a clinically useful biomarker to longitudinally track neurodegeneration in fAD. The systemic factors influencing physiological NfL values in blood should be investigated in future studies to further improve interpretation of this biomarker.

Comparison of the diagnostic performance of blood‐based biomarkers using two distinct commercially available assays in a prospective memory clinic cohort

AbstractBackgroundBlood‐based biomarkers have recently emerged as minimally‐invasive, accessible and relatively inexpensive diagnostic and prognostic tools for people with cognitive impairment. Before being routinely implemented in clinical practice, the diagnostic performance of distinct commercially available assays should be studied in real‐world cohorts. We aimed to study and compare the diagnostic accuracy of different plasma biomarkers measured using two different assay platforms in a memory clinic cohort.MethodParticipants were selected from a prospective memory clinic cohort; all had Alzheimer’s disease (AD) CSF biomarkers performed. Plasma p‐tau181, GFAP and NfL were measured using Simoa (Quanterix), while plasma p‐tau181, Aβ1‐40 and Aβ1‐42 were quantified using Lumipulse G (Fujirebio). Clinical diagnoses were made according to published criteria, blinded to plasma biomarkers. Aβ status (‐/+) was defined by CSF Aβ concentration using local cutoffs.ResultOne hundred and ten participants were included (mean age [standard deviation] 66 [7.8] years, 56% women). Diagnostic categories included 10 cognitively unimpaired controls, 24 with suspected non‐neurodegenerative cause of cognitive impairment (SND), 53 AD, 9 Lewy body disease (LBD, 4 Aβ+) and 14 frontotemporal dementia (FTD, 1 Aβ+).Plasma p‐tau181Quanterix and Aβ1‐42/Aβ1‐40 had the highest diagnostic accuracy (Figure 1) to discriminate between SND and AD (AUC [CI] 0.94 [0.89‐0.99] and 0.94 [0.85‐1]), followed by GFAP (0.93 [0.87‐0.99]), p‐tau181Fujirebio (0.90 [0.82‐0.98]) and Aβ1‐42 (0.71 [0.58‐0.85]). Plasma NfL performed the best to differentiate FTD from SND and AD (AUC 0.95 [0.88‐1] and 0.85 [0.71‐0.99], respectively).For Aβ status discrimination (Figure 2), p‐tau181Quanterix had an AUC [CI] of 0.91 [0.85‐0.96], followed by p‐tau181Fujirebio (0.86 [0.79‐0.93]), Aβ1‐42/Aβ1‐40 (0.85 [0.76‐0.93]) and GFAP (0.84 [0.77‐0.92]) with no statistically significant differences in AUCs. Balanced (Youden index) cut‐offs were calculated to study diagnostic performance, resulting in sensitivities of 79‐83%, specificities of 74‐83% and accuracies of 76‐83%. No combination of plasma biomarkers resulted in a significantly increased discriminative accuracy for Aβ status. All plasma biomarkers were moderately correlated with p‐tau181Quanterix (ρ = 0.40‐0.75, Figure 3).ConclusionIn our cohort, p‐tau181Quanterix, p‐tau181Fujirebio, Aβ1‐42/Aβ1‐40 and GFAP had a high diagnostic performance to discriminate CSF‐defined Aβ status. Plasma NfL identified individuals with FTD. Further studies comparing different plasma biomarkers are needed before implementation in clinical practice.

Plasma neurofilament light chain predicts Alzheimer’s disease in patients with subjective cognitive decline and mild cognitive impairment: a longitudinal study

AbstractBackgroundSubjective cognitive decline (SCD) and mild cognitive impairment (MCI) represent a target population for early detection of Alzheimer’s Disease (AD). Nevertheless, both MCI and SCD are common and heterogeneous conditions. We aim to evaluate the accuracy of plasma neurofilament light chain (NfL) in predicting AD and the progression of cognitive decline in patients with SCD and MCI.Method140 patients (45 SCD, 73 MCI, and 22 AD dementia [AD‐D]) underwent plasma NfL and CSF biomarker measurement at baseline. A subgroup of patients also underwent amyloid‐PET and 18F‐FDG‐PET brain scans. They were rated according to the A/T/N system and followed up for a mean time of 2.72±0.95 years. Forty‐eight (19 SCD, 29 MCI) had plasma NfL measurement also after two years from baseline.ResultNfL levels were higher in AD‐D than MCI and in MCI than SCD. In each group, patients with ATN profiles consistent with AD (A+/T+/N+ or A+/T+/N‐ [defined as ATN+]) had higher NfL levels (F[5,103] = 13.50, p<0.001, η2 = 0.396) than patients classified as ATN‐ (including patients with normal AD biomarkers, isolate Aβ pathology [A+] and non‐AD pathologic changes). NfL can distinguish ATN‐ and ATN+ patients with high accuracy (AUC = 0.82) and cut‐off values of 19.45 pg/mL in SCD and 20.45 pg/mL in MCI provide the highest Youden’s index. During the follow‐up, nine (30%) SCD patients progressed to MCI (p‐SCD) and 14 (29.79%) MCI patients developed AD‐D (p‐MCI). The previously defined cut‐off values identified p‐SCD with an 80.00% [95%C.I. = 65.69:94.31] accuracy. The rate of NfL change per year was higher in p‐MCI (3.52±4.06 pg/mL) as compared to np‐SCD (0.81±1.25 pg/mL) and np‐MCI (‐0.13±3.24 pg/mL) and np‐MCI (‐0.13±3.24 pg/mL) patients and in ATN+ (1.53±3.60 pg/mL) as compared to patients in the A+ group (‐0.97±1.66 pg/mL). A rate of change lower than 1.64 pg/mL can exclude progression to AD with a 92.86% (95%C.I. = 83.32:100) accuracy.ConclusionNfL concentration and change over time are increased in SCD and MCI patients with a biological demonstration of AD and can predict the progression of cognitive decline. Despite being considered a non‐specific neurodegeneration biomarker, if applied to selected populations, NfL may be a reliable, non‐invasive tool to detect the early stages of AD.

Plasma biomarkers of Alzheimer’s Disease associate with gray matter volume and cerebral glucose metabolism in older adults with Mild Cognitive Impairment

AbstractBackgroundAlzheimer’s Disease (AD) pathology in vivo has traditionally been measured in the cerebrospinal fluid or by positron emission tomography (PET). Developments in measuring plasma biomarkers present less invasive, cheaper ways to quantify AD pathology. However, the associations between these new plasma markers and neuroimaging markers of AD are not fully understood. We examined the relationship between plasma biomarkers of AD and measures of whole brain gray matter volume (GMV) and cerebral glucose metabolism in both cognitively unimpaired (n = 83) and impaired (n = 30) individuals.MethodWe collected data from individuals within the University of Kansas Alzheimer’s Disease Research Center’s (KU ADRC) Clinical Cohort who underwent MRI and FDG‐PET scans and had available plasma data. Participants included 64 females, and age of the neuroimaging cohort was 76.0 ± 6. Plasma biomarkers were measured in EDTA plasma using the Quanterix Simoa® HD‐X, and included NfL, Aβ42, Aβ40, GFAP, and p‐tau181. GMV was estimated using MPRAGE MRI processed in FreeSurfer. Cerebral glucose metabolism was estimated from FDG‐PET data using SUVR values calculated through SPM12, with standardization to pons. We used linear regression to examine the relationship between each biomarker to whole brain GMV and glucose metabolism. ANCOVA was used to evaluate differences in the plasma biomarkers between diagnosis groups. Analyses included sex and age as covariates.ResultWhole brain GMV (F = 8.4, p = .004) and cerebral glucose metabolism (F = 10.9, p = .002) differed between diagnostic groups. In the linear regression, we observed an interaction effect between cognitive diagnosis and each biomarker, such that the slopes of regressions between each plasma biomarker and neuroimaging marker was significant in the impaired group, but not significant in the unimpaired. In the impaired group, the strongest association observed was between plasma NFL and whole brain GMV (β = ‐1.4, p = .020) and cerebral glucose metabolism (β = ‐1.7, p = .014). Additionally, all plasma biomarkers were found to differ between diagnostic groups.ConclusionWe show that plasma markers of AD pathology associate with whole brain GMV and cerebral glucose metabolism in individuals with cognitive impairment. Further work will examine the relationship between biomarkers and regional neuroimaging measures

Plasma p‐tau 181 outperforms FDG‐PET and plasma‐NfL in the diagnosis of biological AD

AbstractBackgroundTo compare the performance of plasma p‐tau181, plasma‐NfL and FDG‐PET in the diagnosis of biological AD.MethodWe included 787 participants (cognitively unimpaired (CU, n = 251), mild cognitive impairment (MCI, n = 412), and Alzheimer’s dementia (AD, n = 124) from the ADNI database. Participants underwent measures of plasma p‐tau181, plasma‐NfL, CSF P‐tau181, [18F]FDG‐PET, amyloid‐PET and cognitive screening tests. ROC analyses were used to determine diagnostic accuracy of plasma P‐tau181, NfL and [18F]FDG‐PET using clinical diagnosis and core AD biomarkers (amyloid‐PET and CSF P‐tau181) as reference standards. We assessed whether plasma P‐tau181 and [18F]FDG‐PET diagnostic accuracies were significantly different with a bootstrap‐based test (pROC package in R). Second, correlations of [18F]FDG‐PET, plasma‐NfL, plasma P‐tau181 with P‐tau181 CSF, Aβ‐PET, and cognitive performance were evaluated. Correlation coefficients were compared via Cocor package, a statistical framework for comparing associations between intercorrelated measurements.ResultAcross the total sample we observed that plasma P‐tau181, plasma‐NfL, and [18F]FDG‐PET were able to identify individuals with CSF evidence of AD as well as participants who are amyloid‐PET+. In the MCI group, plasma P‐tau181 outperformed [18F]FDG‐PET and plasma‐NfL in identifying AD pathophysiology measured via CSF P‐tau181 (p = 0.0007), and amyloid‐PET (p = 0.001). We also observed that both plasma P‐tau181, plasma‐NfL and [18F]FDG‐PET were associated with AD pathophysiology measured by core AD biomarkers (CSF and amyloid‐PET). However, [18F]FDG‐PET was more closely associated with cognitive outcomes than plasma P‐tau181 and plasma NfL (MoCA: p < 0.0001; MMSE: p< 0.0001; CDR‐SB: p < 0.0001)ConclusionThis study investigated the diagnostic properties of plasma P‐tau181 and two neurodegenerative biomarkers (plasma‐NfL and [18F]FDG‐PET) as well as their to identify biological AD. While plasma P‐tau181, plasma‐NfL concentrations and [18F]FDG‐PET were associated with AD pathophysiology measured by core AD biomarkers (CSF and amyloid‐PET), plasma P‐tau181 outperformed [18F]FDG‐PET and plasma‐NfL in identifying individuals with AD pathophysiology. However, we also observed that [18F]FDG‐PET was more strongly associated with neuropsychological assessments than plasma P‐tau181 and plasma‐NfL. Taken together, our study suggests that plasma P‐tau181 may aid in the evaluation of individuals by identifying those with underlying AD pathophysiology.

Evaluating brain age as a marker of autosomal dominant Alzheimer disease progression

AbstractBackgroundEstimates of “brain‐predicted age” quantify apparent brain age compared to normative trajectories of neuroimaging features. Brain‐predicted age is elevated in sporadic Alzheimer disease (AD), but is underexplored in autosomal dominant AD (ADAD). We used a previously‐trained model (DeepBrainNet [DBN], Bashyam et al., 2020), which requires minimal MRI pre‐processing, to predict brain age in the Dominantly Inherited Alzheimer Network (DIAN) and evaluated brain age as a marker of ADAD progression.MethodWe used DBN to predict brain age from minimally‐processed, whole‐brain T1 structural MRI scans in 265 ADAD mutation‐carriers (MCs) and 183 non‐carrier controls (NCs). We then tested whether the brain age gap (BAG; difference between predicted and true age) differed as a function of mutation and cognitive status, or estimated years until symptom onset (EYO), and whether it was associated with established markers of amyloid (PiB PET, CSF amyloid‐ß‐42/40), phosphorylated tau (CSF and plasma pTau‐181), neurodegeneration (CSF and plasma neurofilament light [NFL]), and cognition (global neuropsychological composite and CDR‐sum of boxes). Finally, we compared BAG to other MRI measures, and examined heterogeneity in BAG as a function of mutation variants (APP vs. PSEN1 before or after codon 200), APOE E4 carrier status, sex, and education.ResultDBN accurately predicted age in DIAN (Figure 1A). BAG was elevated in symptomatic and asymptomatic MCs, compared to NCs, beginning around ‐10 EYO (Figure 1B,C). BAG, hippocampal volume, and a cortical thickness summary followed similar trajectories across EYO (Figure 1D). BAG was positively associated with PiB PET, CSF and plasma pTau, and CSF and plasma NFL, and was associated with worse cognitive scores, especially in symptomatic MCs (Figure 2). BAG was most elevated in MCs with a PSEN1 mutation before codon 200, but did not differ as a function of APOE, sex, or education (Figure 3).ConclusionWe extend prior work using BAG from sporadic AD to ADAD, noting consistent results. BAG tracks well with markers of pTau, neurodegeneration, and cognition, but to a lesser extent, amyloid, in ADAD. BAG may capture similar signal to established MRI measures. However, DBN‐based BAG offers unique benefits in simplicity of data processing and interpretation.