Startin et al. Alzheimer's Research & Therapy (2019) 11:26 https://doi.org/10.1186/s13195-019-0477-0 RESEARCH Open Access Plasma biomarkers for amyloid, tau, and cytokines in Down syndrome and sporadic Alzheimer’s disease Carla M. Startin1,2,3* , Nicholas J. Ashton4,5,6,7, Sarah Hamburg1,2,3, Rosalyn Hithersay1,2,3, Frances K. Wiseman3,8, Kin Y. Mok3,9,10, John Hardy3,9,11, Alberto Lleó12, Simon Lovestone13, Lucilla Parnetti14, Henrik Zetterberg6,8,15,16, Abdul Hye4,5, LonDownS Consortium and André Strydom1,2,3 Abstract Background: Down syndrome (DS), caused by chromosome 21 trisomy, is associated with an ultra-high risk of dementia due to Alzheimer’s disease (AD), driven by amyloid precursor protein (APP) gene triplication. Understanding relevant molecular differences between those with DS, those with sporadic AD (sAD) without DS, and controls will aid in understanding AD development in DS. We explored group differences in plasma concentrations of amyloid-β peptides and tau (as their accumulation is a characteristic feature of AD) and cytokines (as the inflammatory response has been implicated in AD development, and immune dysfunction is common in DS). Methods: We used ultrasensitive assays to compare plasma concentrations of the amyloid-β peptides Aβ40 and Aβ42, total tau (t-tau), and the cytokines IL1β, IL10, IL6, and TNFα between adults with DS (n = 31), adults with sAD (n = 27), and controls age-matched to the group with DS (n = 27), and explored relationships between molecular concentrations and with age within each group. In the group with DS, we also explored relationships with neurofilament light (NfL) concentration, due to its potential use as a biomarker for AD in DS. Results: Aβ40, Aβ42, and IL1β concentrations were higher in DS, with a higher Aβ42/Aβ40 ratio in controls. The group with DS showed moderate positive associations between concentrations of t-tau and both Aβ42 and IL1β. Only NfL concentration in the group with DS showed a significant positive association with age. Conclusions: Concentrations of Aβ40 and Aβ42 were much higher in adults with DS than in other groups, reflecting APP gene triplication, while no difference in the Aβ42/Aβ40 ratio between those with DS and sAD may indicate similar processing and deposition of Aβ40 and Aβ42 in these groups. Higher concentrations of IL1β in DS may reflect an increased vulnerability to infections and/or an increased prevalence of autoimmune disorders, while the positive association between IL1β and t-tau in DS may indicate IL1β is associated with neurodegeneration. Finally, NfL concentration may be the most suitable biomarker for dementia progression in DS. The identification of such a biomarker is important to improve the detection of dementia and monitor its progression, and for designing clinical intervention studies. Keywords: Down syndrome, Alzheimer’s disease, Dementia, Biomarker, Plasma, Amyloid, Tau, Interleukin 1β, Cytokines * Correspondence: carla.startin.09@ucl.ac.uk 1 Department of Forensic and Neurodevelopmental Sciences, Institute of Psychiatry, Psychology and Neuroscience, King’s College London, 16 De Crespigny Park, London SE5 8AF, UK 2 Division of Psychiatry, University College London, London, UK Full list of author information is available at the end of the article © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Startin et al. Alzheimer's Research & Therapy (2019) 11:26 Page 2 of 12 Background [18–22]. Genome-wide association studies of sAD have Down syndrome (DS) is associated with an ultra-high implicated several genes involved in immune function risk of developing dementia caused by Alzheimer’s (including variants in CLU, ABCA7, CR1, and CD33) disease (AD) [1]. The lifetime dementia risk in DS is in [23–25], and there is an over-representation of genetic excess of 95%, with a median age of onset of 55 years associations with sAD in pathways involved in the innate [2, 3], and the majority of adults with DS who die after immune response [26]. the age of 36 have a diagnosis of dementia [4]. DS is To fully understand the development of AD in DS, it caused by chromosome 21 trisomy and is the most com- is therefore important to understand relevant molecular mon genetic cause of intellectual disability, with a UK differences between those with DS, those with sAD, and incidence of approximately 1 in 1000 live births [5]. AD controls. Similarities and group differences in plasma in DS is driven by the overexpression of genes on concentrations of molecules associated with AD and re- chromosome 21 due to their triplication, with triplica- lationships between these possible biomarkers, in tion of the amyloid precursor protein (APP) gene being particular Aβ peptides and tau, may help to identify of particular importance. The amyloid precursor protein markers of AD progression. Previous studies comparing encoded by the APP gene is cleaved by β and γ secre- plasma concentrations of Aβ peptides and tau in individ- tases to form amyloid-β (Aβ) peptides, with Aβ42 the uals with DS to age-matched controls have reported in- main component of the Aβ plaques characteristic of AD creased concentrations of Aβ40 and Aβ42 [17, 27–37] [6]. Such Aβ deposits are found in the brains of adults and total tau (t-tau) [28, 38] in DS. Results regarding the with full trisomy 21 by the mid-30s [1, 7]. Further, famil- Aβ42/Aβ40 ratio are less consistent; while several studies ial mutations in, or duplications of, the APP gene cause have reported this to be lower in those with DS [28, 30, early-onset AD independent of DS [1, 8], and individuals 31], others have reported this to be higher [33] or to with a partial trisomy of chromosome 21 not including show no difference [29] relative to controls. To date, few APP do not develop AD or show Aβ pathology [9, 10]. studies have compared plasma concentrations of Aβ Triplication of the APP gene and the subsequent over- peptides and tau in adults with DS to those with sAD. production of Aβ peptides are therefore likely to be cen- Two studies have reported increased Aβ40 and Aβ42 in tral to the development of AD in people with DS, and those with DS [29, 36], while another study reported DS may be considered a genetic cause of AD. lower concentrations of both Aβ42 and t-tau in adults The presence of Aβ plaques in both sporadic AD with DS [28] compared to individuals with sAD. (sAD) without DS and in people with DS, with the high In addition, a potential involvement of the immune risk of dementia in DS, indicates their development is a system in the development of AD is of particular interest key component of the development of AD. Indeed, the in DS due to immune dysfunction being common in DS, amyloid hypothesis of AD states the presence of Aβ pla- with an increased vulnerability to some types of infec- ques initiates a cascade of events that leads to cell death tions throughout life [39] and higher rates of auto- and cognitive decline [11, 12]. Other pathophysiological immune disorders [40] in DS. Further, in DS, there is an mechanisms are also important within this cascade. overexpression of immune genes found on chromosome Firstly, the development of neurofibrillary tangles, 21 (such as IFNAR1, IFNAR2, and IFNGR2, which all composed of hyperphosphorylated and misfolded tau encode interferon-γ (IFN-γ) receptor proteins) [41]. In protein, is another neuropathological feature of AD. adults and children with DS, higher blood (plasma or There is a stronger association between cognitive decline serum) concentrations of IL10 [31, 42], IL6 [31, 43, 44], and tau pathology than with Aβ plaques in both DS [13] and TNFα [31, 43, 45] have been reported compared to and sAD [14]. Secondly, neuronal damage within the age-matched controls, although lower concentrations of brain can be measured by the release of neurofilament IL10 [45], IL6 [42], and TNFα [42] in DS have also been light (NfL), a scaffolding cytoskeleton protein, into reported. A recent meta-analysis of 19 cytokine studies plasma in a number of neurodegenerative disorders in- in adults and children with DS suggested that IL1β, cluding sAD [15] and AD in DS [16, 17]. Finally, the in- TNFα, and IFN-γ (but not IL6 or IL10) concentrations flammatory response is of interest as a potential are raised by trisomy 21 [46]. contributory molecular mechanism to the development Further studies using newly developed ultrasensitive of AD. In sAD, the role of the immune system is assays are therefore needed to better understand differ- supported by epidemiological and genetic studies. The ences in concentrations of Aβ peptides, tau, and cyto- presence of systemic infections and increased blood kines in individuals with DS, sAD, and controls. We (plasma or serum) concentrations of cytokines including therefore compared plasma concentrations of Aβ40, interleukin 1β (IL1β), interleukin 10 (IL10), interleukin Aβ42, t-tau, IL1β, IL10, IL6, and TNFα between adults 6 (IL6), and tumour necrosis factor α (TNFα) have been with DS, adults with sAD, and controls age-matched to associated with sAD and predict cognitive decline the group with DS (to determine the effect of triplication Startin et al. Alzheimer's Research & Therapy (2019) 11:26 Page 3 of 12 of chromosome 21), and explored relationships between respectively) [50, 51]. All individuals in the sAD and molecular concentrations within each group. Given the control groups had Mini-Mental State Examination universal development of AD pathology in those with (MMSE) scores available. Where possible participants in DS and increased risk of dementia with increased age the sAD and control groups were matched in age and [2], we also explored relationships between molecular gender to participants with DS; age matching was not concentrations and age. In addition, in the group with fully possible for the group with sAD due to the older DS, we explored relationships with concentrations of age of this group (see Table 1). NfL. Our group and others have previously shown NfL For adults with sAD and controls, ethical approval was to be a potential biomarker for AD in DS, with higher obtained from the regional ethics committee. Plasma NfL concentration associated with increased age and de- and CSF samples were routinely collected in all subjects mentia development [16, 17], and so molecular concen- undergoing a diagnostic work-up for suspected neurode- trations that correlate with NfL may be particularly generative diseases, with informed written consent ob- informative. tained for their use in research from patients or their representatives. Methods Participants APOE genotyping Adults in the group with DS (n = 31, including 7 with a APOE genotype was determined using a Thermo Fisher clinical diagnosis of dementia) were recruited as a part Scientific TaqMan assay for SNPs rs7412 and rs429358 of the LonDownS Consortium’s studies in adults with (Waltham, MA, USA). DS [47]. Clinical diagnoses of dementia were made by each individual’s clinician after a comprehensive clinical Sample processing and assays assessment. Participants were included in the present Blood samples from individuals with DS were collected study if they lived within travelling distance of London in EDTA tubes after participants had undergone a cogni- to allow processing time for the blood sample, and they tive assessment [47] and processed as quickly as feasible agreed to have a blood test. The presence of an add- (within approximately 3 h). Plasma was prepared by cen- itional chromosome 21 was confirmed genetically using trifuging samples for 10 min at 2200 g, then the super- saliva or blood samples; following DNA extraction, natant was aliquoted and stored at − 80 °C. Samples genome-wide single nucleotide polymorphism genotyp- were only thawed immediately prior to analysis. Due to ing was performed using an Illumina OmniExpressEx- difficulties in obtaining blood samples from this popula- ome array (San Diego, CA, USA) at UCL Genomics, tion, we had to be pragmatic about the practicalities of then assembled and visually inspected in GenomeStudio. collecting these samples. Blood samples were collected All adults were observed to have trisomy 21, with clin- at a range of times throughout the day, with samples ical notes from two individuals confirming this was due collected between 9.30 am and 5.40 pm and the majority to a translocation. of samples collected after midday (mean time of collec- For adults with DS, ethical approval was obtained from tion = 2.33 pm, standard deviation = 1.99 h). It was not the North West Wales Research Ethics Committee (13/ possible to take fasting blood samples for research pur- WA/0194). Where individuals had the capacity to con- poses, though blood samples were typically taken at the sent for themselves, we obtained written informed con- end of a cognitive assessment that had lasted several sent. Where individuals did not have the capacity to hours, and participants did not eat during this time. In consent, a consultee was asked to approve the individ- addition, the amount of time samples were in the freezer ual’s inclusion based on their knowledge of the individ- before analysis varied as it took a number of months to ual and his/her wishes, in accordance with the UK collect enough blood samples for our analysis. During Mental Capacity Act 2005. this time, samples remained frozen at − 80 °C. To assess Adults in the sAD and control groups (n = 27 per potential relationships between the number of months group) were recruited from the multi-centre consortium samples were frozen before analysis and the concentra- European Medical Informatics Framework (EMIF) [48]. tions of each molecule, we performed correlational ana- All plasma samples used in this study were from two lyses; there were no significant relationships observed sites (Clinica Neurologica, Universita di Perugia, and (results not shown). Hospital de la Santa Creu i Sant Pau, Barcelona). All Blood sample collection and processing methods for participants in the sAD group had an AD diagnosis ac- adults in the sAD and control groups from the EMIF co- cording to standard criteria [49], and biological AD was hort have been previously described [48]. also confirmed using concentrations of cerebrospinal For all three groups, plasma concentrations of Aβ40, fluid (CSF) Aβ42 with a cutoff value indicating AD Aβ42, and t-tau (Human Neurology 3-Plex A assay (< 800 pg/ml and < 550 pg/ml for Perugia and Barcelona, (N3PA)), IL1β (Human IL-1β 2.0), and IL10, IL6, and Startin et al. Alzheimer's Research & Therapy (2019) 11:26 Page 4 of 12 Table 1 Group demographics DS sAD Controls Group comparison n 31 27 27 N/A Age (years) 46.77 ± 10.99 (23–67) 59.33 ± 4.04 (51–67) 49.26 ± 10.40 (24–64) F(2,82) = 14.84, p < 0.001a Sex 9 (29.0%) females, 22 (71.0%) 9 (33.3%) females, 18 (66.7%) 11 (40.7%) females, 16 (59.3%) Χ(2) = 0.89, p = 0.641 males males males Intellectual disability severityb 17 (54.8%) mild, 11 (35.5%) N/A N/A N/A moderate, 3 (9.7%) severe APOE ε4 carrier 8 (25.8%) carriers, 23 (74.2%) 10 (37.0%) carriers, 17 (63.0%) 5 (18.5%) carriers, 22 (81.5%) Χ(2) = 2.38, p = 0.304 non-carriers non-carriers non-carriers Mini-Mental State N/A 19.19 ± 4.42 (7–26) 28.89 ± 1.12 (27–30) t(52) = −11.06, p < 0.001c Examination (MMSE) score Values for age and MMSE score given as mean ± standard deviation (range) a sAD group older than DS and control groups (both p < 0.001), no difference between DS and control groups (p = 0.915) b Pre-dementia level, assessed via informant report based on everyday functional descriptions c sAD compared to controls only TNFα (Human Cytokine 3-Plex A) were measured in was performed for each group separately to determine as- duplicate using ultrasensitive Simoa immunoassays sociations between concentrations of molecules and with (Quanterix, Lexington, MA, USA) according to the man- age, in addition to associations with NfL concentration for ufacturer’s instructions at the Institute of Psychiatry, the group with DS. Due to the number of correlations per- Psychology and Neuroscience, King’s College London. formed, correlation coefficients were used to determine Samples from the three groups were blinded and rando- relationships rather than p values, with only strong and mised across two analytical plates for each assay. All du- moderate correlations considered; these were defined as plicate measures for all targets had an average absolute values of correlation coefficients of 0.70 and coefficient of variation (CV) of between 4 and 8%. above, and between 0.50 and 0.69, respectively. For the group with DS, concentrations of NfL were previously obtained from samples taken at the same time Results [16]. Blood samples collected in lithium heparin tubes Demographic information for individuals in the three were first layered over a similar amount of Ficoll (GE groups can be seen in Table 1, with median values and Healthcare, Little Chalfont, UK), centrifuged for 40 min ranges for molecule concentrations in Table 2 and Fig. 1. at 400 g, then the supernatant was aliquoted and stored As expected, the group with sAD was significantly older at − 80 °C until analysis. Plasma NfL concentration was than the group with DS and the control group. One measured using the Simoa immunoassay (Human plate for IL1β failed (containing 14 DS, 11 sAD, and 13 NF-light) (Quanterix, Lexington, MA, USA) according control samples), resulting in fewer samples measured to the manufacturer’s instructions at the Institute of for IL1β. Demographic information and median values Neurology, University College London. Values used in and ranges for molecule concentrations for adults with the present study had a median value of 27.09 pg/ml DS split into subgroups with and without clinical de- (range 10.97–112.60, n = 26). mentia can be found in Table 3 and Fig. 2. Statistical analysis was not performed to compare these subgroups Statistical analysis due to small sample sizes. SPSS version 22 was used for analyses. Age and demo- Concentrations of Aβ40, Aβ42, and IL1β were higher graphic factors were compared between groups using for individuals with DS compared to both those with ANOVAs, two-sample t tests, or chi-squared tests as ap- sAD and controls, with median concentrations of Aβ40 propriate. To compare concentrations of molecules be- and Aβ42 increased approximately two-fold and median tween groups of adults with DS, sAD, and controls, we concentration of IL1β increased over ten-fold in the performed ANCOVAs for log-transformed concentrations group with DS. The Aβ42/Aβ40 ratio was higher for con- (due to concentrations not being normally distributed) trols compared to both individuals with DS and those with age, sex, and the presence of an APOE ε4 allele in- with sAD (Table 2). cluded as covariates to adjust for their potentially con- All three groups showed strong positive associations founding effects, with η2 values determining the overall between Aβ40 and Aβ42 concentrations (see Table 4). effect size of group and post hoc pairwise comparisons Both the groups with DS and with sAD showed a mod- with Bonferroni corrections applied where appropriate. erate positive association between IL10 and TNFα con- Due to concentrations not being normally distributed, centrations. In addition, the control group showed a non-parametric Spearman’s rank correlational analysis moderate negative association between the Aβ42/Aβ40 Startin et al. Alzheimer's Research & Therapy (2019) 11:26 Page 5 of 12 Table 2 Concentrations of biomarkers for each group and group comparisons DS sAD Controls ANCOVA Post hoc Aβ40 312.00 (150.24–555.00) 160.80 (43.60–420.00) 144.40 (26.88–355.60) F(2,79) = 24.28, p < 0.001, η2 = 0.38 DS vs sAD p < 0.001 DS vs controls p < 0.001 sAD vs controls p = 0.506 Aβ42 24.48 (14.92–50.40) 13.32 (4.28–18.84) 14.76 (2.00–45.62) F(2,79) = 20.36, p < 0.001, η2 = 0.34 DS vs sAD p < 0.001 DS vs controls p < 0.001 sAD vs controls p = 0.710 Aβ42/Aβ40 0.09 (0.05–0.13) 0.08 (0.04–0.11) 0.10 (0.07–0.17) F(2,79) = 15.43, p < 0.001, η2 = 0.28 DS vs sAD p = 1.000 DS vs controls p < 0.001 sAD vs controls p < 0.001 Aβ42/t-tau 20.60 (1.17–93.33) 10.23 (0.77–52.00) 10.59 (1.14–82.25) F(2,79) = 2.30, p = 0.107, η2 = 0.06 N/A T-tau 1.45 (0.18–12.72) 1.00 (0.33–24.48) 1.49 (0.16–10.24) F(2,79) = 0.64, p = 0.529, η2 = 0.02 N/A IL1β a 2.35 (0.27–47.20) 0.18 (0.02–5.45) 0.09 (0.00–1.25) F(2,41) = 13.84, p < 0.001, η = 0.40 2 DS vs sAD p = 0.002 DS vs controls p < 0.001 sAD vs controls p = 1.000 IL10b 1.34 (0.36–93.20) 0.75 (0.27–8.76) 0.87 (0.27–4.48) F(2,76) = 2.38, p = 0.100, η2 = 0.06 N/A b IL6 2.12 (0.36–1024.00) 1.45 (0.50–9.56) 1.54 (0.07–8.70) F(2,76) = 2.54, p = 0.086, η2 = 0.06 N/A TNFα b 2.76 (1.25–328.80) 2.37 (1.36–20.44) 1.97 (0.56–10.50) F(2,76) = 2.42, p = 0.096, η = 0.06 2 N/A Concentrations (pg/ml) given are median (range). ANCOVA performed on log-transformed values and including age, sex, and the presence of an APOE ε4 allele as covariates a One plate failed (14 DS, 11 sAD, 13 controls) b Three samples failed (2 DS and 1 sAD) ratio and IL10 concentration. Adults with DS showed sev- group with DS, possibly reflecting a more activated im- eral additional associations between concentrations of mune system in DS which may be associated with higher molecules (Table 4 and Fig. 3). Firstly, within this group, rates of infection and/or autoimmune disorders [39, 40]. there was a moderate positive association between Aβ42 The group with DS also showed positive associations be- and t-tau concentrations. Secondly, within the cytokines tween concentrations of t-tau and both Aβ42 and IL1β, investigated, there was a strong positive association be- and a negative association between IL1β and the Aβ42/ tween IL1β and IL10, and a moderate positive association t-tau ratio, while the control group showed a negative between IL6 and TNFα. Finally, IL1β concentration association between concentration of IL10 and the Aβ42/ showed a moderate positive association with t-tau concen- Aβ40 ratio. tration and a moderate negative association with the Results relating to concentrations of Aβ peptides and Aβ42/t-tau ratio. In addition, in the group with DS, only their ratio indicate both likely differences and similarities NfL concentration showed a moderate positive association in the development of AD between those with DS and with age, as previously published [16]. There were no those with sAD. Similar to the results of previous stud- other moderate or strong associations for molecular con- ies, we observed higher concentrations of Aβ40 and Aβ42 centrations or with age across any group. in those with DS compared to those with sAD and con- trols [17, 27–37]. This observation is likely due to tripli- Discussion cation of the APP gene in adults with DS leading to We used ultrasensitive assays to compare plasma con- elevated production of Aβ peptides. Concentrations of centrations of Aβ peptides, tau, and selected cytokines Aβ40 and Aβ42 in adults with DS were approximately between adults with DS, adults with sAD, and controls double those in the other groups, rather than 1.5 times who were age-matched to the group with DS. We found higher as may be expected based on the presence of an significantly higher concentrations of Aβ40, Aβ42, and additional APP gene alone, suggesting that factor(s) IL1β in those with DS compared to those with sAD and other than APP triplication may contribute to increased controls. The Aβ42/Aβ40 ratio was higher in controls Aβ concentrations [52], such as a positive feedback loop compared to those with DS or sAD, indicating higher or an overloaded perivascular drainage system. Although plasma concentrations of Aβ42 relative to concentrations concentrations of Aβ40 and Aβ42 were higher in those of Aβ40 in controls. These group differences were ob- with DS compared to those with sAD, there was no dif- served when we accounted for age, sex, and the presence ference in the Aβ42/Aβ40 ratio between these two of an APOE ε4 allele. groups. This may suggest the relative processing and de- Several positive associations were found between con- position of Aβ40 and Aβ42 in the brain is similar between centrations of different cytokines, in particular for the the two groups. The higher concentrations of Aβ40 and Startin et al. Alzheimer's Research & Therapy (2019) 11:26 Page 6 of 12 Fig. 1 Concentrations of biomarkers for each group. a Aβ40, b Aβ42, c Aβ42/Aβ40, d Aβ42/t-tau, e t-tau, f IL1β, g IL10, h IL6, and i TNFα. Lines indicate median value (pg/ml) Aβ42 in adults with DS as compared to those with sAD, Adults with DS also showed significantly higher con- and no difference in the Aβ42/Aβ40 ratio between the centrations of IL1β compared to adults with sAD and two groups, is despite the younger age of the group with controls [46], and in the group with DS, IL1β concentra- DS compared to those with sAD. In contrast, the Aβ42/ tion was positively correlated with t-tau concentration Aβ40 ratio was higher in controls, indicating this group and negatively correlated with the Aβ42/t-tau ratio. has a higher relative plasma Aβ42 concentration and/ Given the lack of an association between concentrations or a lower relative plasma Aβ40 concentration com- of IL1β and Aβ42 in this group, it is likely the latter asso- pared to the other two groups. This may be due to ciation is driven by the association with t-tau. IL1β is a the build-up of Aβ plaques in the brains of those pro-inflammatory cytokine produced by activated mac- with DS and those with sAD, with Aβ42 a main com- rophages, and previous studies have indicated increased ponent, resulting in a relative decrease of free Aβ42 concentrations of IL1β are associated with AD [20] and a lower Aβ42/Aβ40 ratio. Similarly, previous stud- and precede cognitive decline [19]. The increase in ies have associated a lower Aβ42/Aβ40 ratio with a IL1β concentration in DS may reflect increased vul- higher risk of developing dementia [53] and with in- nerability to peripheral infections [39] and/or in- creased Aβ accumulation in the brain as measured creased prevalence of autoimmune disorders in those using Aβ positron emission tomography [54–57]. with DS [40]. Plasma t-tau concentration is associated Startin et al. Alzheimer's Research & Therapy (2019) 11:26 Page 7 of 12 Table 3 Demographic information and concentrations of biomarkers for adults with DS with and without clinical dementia Dementia No dementia n 7 24 Age (years) 52.00 ± 10.36 (40–67) 45.25 ± 10.90 (23–65) Sex 2 (28.6%) females, 5 (71.4%) males 7 (29.2%) females, 17 (70.8%) males Intellectual disability severitya 5 (71.4%) mild, 1 (14.3%) moderate, 1 (14.3%) severe 12 (50.0%) mild, 10 (41.7%) moderate, 2 (8.3%) severe APOE ε4 carrier 4 (57.1%) carriers, 3 (42.9%) non-carriers 4 (16.7%) carriers, 20 (83.3%) non-carriers Aβ40 320.80 (268.20–555.00) 304.80 (150.24–528.00) Aβ42 26.36 (15.16–42.40) 24.14 (14.92–50.40) Aβ42/Aβ40 0.08 (0.05–0.09) 0.09 (0.05–0.13) Aβ42/t-tau 14.76 (2.73–29.18) 20.86 (1.17–93.33) T-tau 1.88 (0.99–8.96) 1.27 (0.18–12.72) b IL1β 3.79 (1.65–5.10) 1.44 (0.27–47.20) IL10c 1.78 (0.62–6.33) 1.26 (0.36–93.20) c IL6 2.12 (0.44–58.40) 2.12 (0.36–1024.00) TNFαc 2.58 (1.25–29.84) 3.08 (1.30–328.80) d NfL 59.84 (16.36–112.60) 25.10 (10.97–55.45) Values for age given as mean ± standard deviation (range), and concentrations (pg/ml) given are median (range) a Pre-dementia level, assessed via informant report based on everyday functional descriptions b One plate failed (3 with dementia, 11 without dementia) c Two samples failed (both without dementia) d Five samples missing (1 with dementia, 4 without dementia) Fig. 2 Concentrations of biomarkers for adults with DS with and without clinical dementia. a Aβ40, b Aβ42, c Aβ42/Aβ40, d Aβ42/t-tau, e t-tau, f IL1β, g IL10, h IL6, i TNFα, and j NfL. Lines indicate median value (pg/ml) Startin et al. Alzheimer's Research & Therapy (2019) 11:26 Page 8 of 12 Table 4 Relationships between concentrations of biomarkers for each group with severity of neurodegeneration, thought to be due of pro-inflammatory cytokines may be a protective to neuronal damage causing tau release from the response to neuronal damage. Raised IL1β concentra- brain [58]. The association between IL1β and plasma tion in adults with DS may also have implications for t-tau in DS may therefore indicate that raised levels of the development of microbleeds or strokes in re- pro-inflammatory cytokines may contribute to neuro- sponse to cerebral amyloid angiopathy, as discussed degeneration in this group, or increased concentration in Buss et al. [59]. Fig. 3 Relationships between biomarkers for adults with DS. a Log Aβ42 and log t-tau, b log t-tau and log IL1β, and c log IL1β and log IL10 Startin et al. Alzheimer's Research & Therapy (2019) 11:26 Page 9 of 12 We did not find any associations with age and concen- cerebral β-amyloidosis has emerged [73], it is unknown trations of Aβ peptides, tau, or cytokines in any group. how well plasma concentrations of molecules reflect Previous studies have reported contrasting results regard- changes within the brain, and the relative contribution ing the association between Aβ42 and age in DS, with of different sources (including the brain and platelets) to some studies also finding no association [29, 32, 33, 60– plasma concentrations is unknown. To better under- 62], some finding a positive association [34, 63, 64], and stand concentrations of Aβ42 and tau in the brain of some finding a negative association [27, 30]. Previous adults with DS using blood samples, Hamlett et al. [74] studies have also suggested age shows a positive correl- measured concentrations from neuronal exosomes ation with concentrations of Aβ40 [33, 34] and t-tau [38] within the blood, finding higher concentrations of Aβ42 in DS. It has further been suggested Aβ42 and Aβ40 con- and phosphorylated tau in DS compared to controls. centrations may not show a linear relationship with age in These results indicate an alternative to measuring DS, being stable until age 50 and then decreasing [65]. As plasma concentrations. Future studies are needed to fur- reported previously, we found a positive association be- ther explore the relationships between plasma concen- tween age and NfL concentration in adults with DS [16]. trations of molecules and changes within the brain. Due to the strong association between age and the devel- Studies assessing blood biomarkers in DS typically opment of dementia in adults with DS, this may indicate consist of relatively small sample sizes [31, 38], in large NfL is a more suitable biomarker for dementia progres- part due to difficulties in obtaining blood samples from sion than Aβ or tau in this group. Further supporting the these individuals. Although our sample size is large use of NfL as a potential biomarker for dementia progres- enough to demonstrate group differences in plasma con- sion in DS, our group and others have previously shown centrations of some molecules, it is possible a larger that NfL concentration is significantly higher in adults sample would reveal additional group differences. It with DS and dementia compared to those with DS with- should also be noted we included individuals with vary- out dementia [16, 17]. ing stages of dementia progression in the group with DS, Identifying a biomarker for dementia in DS is import- which may have contributed to the presence of some re- ant for early detection and for monitoring disease pro- lationships in this group and not in the other groups, gression and is required for clinical intervention studies such as the positive associations between concentrations in combination with cognitive tests sensitive to detecting of t-tau and both Aβ42 and IL1β. Despite this, all adults cognitive decline [66]. In addition to NfL, several other with DS are thought to have Aβ neuropathology by their potential plasma or urinary biomarkers have been mid-30s [1, 7], and so the group with DS is likely similar proposed, including neopterin (a marker for activated in terms of the presence of Aβ neuropathology. Al- cellular immunity and inflammation) [67, 68], though some studies have suggested stage of dementia 3-methoxy-4-hydroxyphenylglycol (MHPG, a noradren- progression in DS may be associated with varying con- ergic metabolite) [69], dehydroepiandrosterone (DHEA, centrations of Aβ40 and Aβ42 [28, 31, 32, 63, 75], other a steroid hormone) [70], or molecules associated with studies have reported no differences in Aβ40 or Aβ42 oxidative stress, specifically superoxide dismutase en- concentrations in those with DS with and without de- zymes (SOD) [71] and iPF2alpha [72]. It has also been mentia [29, 31, 60–62, 76]. A recent study has further suggested that a combination of baseline concentrations suggested that plasma Aβ40 and Aβ42 concentrations and changes in Aβ peptides and inflammatory molecules show poor diagnostic performance for dementia in DS may be predictive of cognitive decline in adults with DS [17]. Given these differences in results, which may be [31]. Further longitudinal studies assessing multiple po- due to differences in study populations (including differ- tential biomarkers with large samples and multiple time ences in age, dementia stage, or dementia duration) and points starting prior to the onset of any cognitive decline assay sensitivity or sampling procedures, further larger, are needed to clarify the ideal biomarker for dementia in longitudinal studies with multiple time points starting DS. Blood samples are less invasive, easier, quicker, and prior to the onset of any cognitive decline and using ul- cheaper to obtain than CSF samples, with additional prac- trasensitive assays such as those used in the present tical considerations in those with DS often limiting the study are needed to clarify whether changes in Aβ40 or feasibility of obtaining CSF. The identification of a plasma Aβ42 concentrations are associated with the develop- biomarker for dementia progression in DS is therefore es- ment and progression of dementia in DS. sential. Such a biomarker will be of particular use in the Finally, considering the amyloid hypothesis of AD and population with DS due to the pre-morbid variability in associated changes in people with DS, our results support intellectual disability severity posing an additional chal- the triplication of APP leading to the overproduction of lenge for the clinical diagnosis of dementia. Aβ peptides and resulting in elevated concentrations of However, aside from plasma Aβ42 (and the Aβ42/Aβ40 Aβ40 and Aβ42 in people with DS compared to the sAD ratio), where a reasonably robust association with and control groups. Within the group with DS, the Startin et al. Alzheimer's Research & Therapy (2019) 11:26 Page 10 of 12 positive relationship between NfL concentration and age UK; Elizabeth Fisher, Department of Neurodegenerative Disease, Institute of (and also the previously published association between Neurology, University College London, London, UK; Dean Nizetic, Blizard Institute, Barts and the London School of Medicine, Queen Mary University higher NfL concentration and dementia development [16, of London, London, UK, and Lee Kong Chian School of Medicine, Nanyang 17]) indicates plasma NfL concentration reflects neuronal Technological University, Singapore, Singapore; John Hardy, Reta Lila Weston damage in the brain. We did not find group differences in Institute, Institute of Neurology, University College London, London, UK, and UK Dementia Research Institute at UCL, London, UK; Victor Tybulewicz, concentrations of t-tau or cytokines that may indicate al- Francis Crick Institute, London, UK, and Department of Medicine, Imperial tered concentrations with the development of dementia. It College London, London, UK; and Annette Karmiloff-Smith, Birkbeck is possible our relatively small sample size limited our University of London, London, UK (deceased). We would like to thank Tamara Al-Janabi (Division of Psychiatry, University College London, London, UK) for power to detect these differences. Alternatively, it is pos- managing the project as a whole. We would like to thank David Zhang for sible that the stage of dementia progression is relevant to support with genetic analysis for samples from adults with DS (Institute of detect group differences, and changes may occur in the Neurology, University College London, London, UK). earlier prodromal stage only. Again, larger, longitudinal Funding studies with multiple time points starting prior to the This work was funded by a Wellcome Trust Strategic Award (grant number: onset of any cognitive decline would be required to 098330/Z/12/Z) conferred upon The London Down Syndrome (LonDownS) determine associations between plasma molecular Consortium, and an award from the UCL ARUK network. Our work is further supported by the Baily Thomas Charitable Fund. HZ is a Wallenberg concentrations and the development of dementia. Academy Fellow and receives support from the European Research Council, the Swedish Research Council, and Frimurarestiftelsen. HZ is further supported by the Wolfson Foundation, the Wellcome Trust, and the UK Conclusions Dementia Research Institute. The study funders and sponsors had no role in We compared plasma concentrations of Aβ peptides, study design, data collection, data analysis, data interpretation, writing of the tau, and selected cytokines between adults with DS, report, or the decision to submit for publication. adults with sAD, and controls age-matched to the group with DS to investigate molecular mechanisms relevant Availability of data and materials The datasets used and/or analysed during the current study are available to the development of AD in DS using ultrasensitive as- from the corresponding author on reasonable request. says. Our results indicated the likely similar processing and deposition of Aβ40 and Aβ42 in those with DS and Authors’ contributions sAD, though those with DS showed much higher con- CMS, FKW, and AS designed the study and data analysis. CMS, SH, and RH contributed significantly to the recruitment and data collection of adults centrations of these molecules, despite their younger with DS. AL, SL, and LP provided the samples for adults with sAD and age. In addition, IL1β concentration is far higher in controls. NJA and AH performed the biomarker measurements. KYM and JH those with DS compared to those with sAD and con- contributed to the genetic analysis and interpretation of the genetic data for the samples from adults with DS. HZ contributed to the NfL analysis for the trols, and a positive association between IL1β and t-tau samples from adults with DS. CMS analysed the data and drafted the initial in those with DS may indicate IL1β is associated with version of the report. All authors contributed to the revision and editing of neurodegeneration in this group. Our results also indi- the report. All authors read and approved the final manuscript. cated that NfL concentration may be the most suitable Ethics approval and consent to participate biomarker for dementia progression in DS. Future longi- For adults with DS, ethical approval was obtained from the North West Wales tudinal studies to identify biomarker changes over time Research Ethics Committee (13/WA/0194). Where individuals had capacity to associated with pathological and clinical progression are consent for themselves we obtained written informed consent. Where individuals did not have capacity to consent a consultee was asked to approve needed to confirm these findings. the individual’s inclusion based on their knowledge of the individual and his/ her wishes, in accordance with the UK Mental Capacity Act 2005. Abbreviations For adults with sAD and controls, ethical approval was obtained from AD: Alzheimer’s disease; APP: Amyloid precursor protein; Aβ: Amyloid-β; the regional ethics committee. Plasma and CSF samples were routinely CSF: Cerebrospinal fluid; CV: Coefficient of variation; collected in all subjects undergoing a diagnostic work-up for suspected DHEA: Dehydroepiandrosterone; DS: Down syndrome; EMIF: European neurodegenerative diseases, with informed written consent obtained for Medical Informatics Framework; IFN-γ: Interferon-γ; IL10: Interleukin 10; their use in research from patients or their representatives. IL1β: Interleukin 1β; IL6: Interleukin 6; MHPG: 3-Methoxy-4- hydroxyphenylglycol; MMSE: Mini-Mental State Examination; N3PA: Neurology 3-Plex A assay; NfL: Neurofilament light; sAD: Sporadic Consent for publication Alzheimer’s disease; SOD: Superoxide dismutase enzymes; TNFα: Tumour Not applicable. necrosis factor α; t-tau: Total tau Competing interests Acknowledgements HZ has served at scientific advisory boards for Eli Lilly, Roche Diagnostics and The authors would like to thank all the participants in this study for their Wave, has received travel support from Teva, and is a co-founder of Brain time. This research was supported by the National Institute for Health Biomarker Solutions in Gothenburg AB, a GU Ventures-based platform company Research networks (mental health, dementias and neurology) and at the University of Gothenburg. The other authors declare that they have no participating NHS trusts. We would like to thank our NHS network of sites competing interests. that helped to identify participants. The LonDownS Consortium principal investigators are Andre Strydom (chief investigator), Department of Forensic and Neurodevelopmental Sciences, Publisher’s Note Institute of Psychiatry, Psychology and Neuroscience, King’s College London, Springer Nature remains neutral with regard to jurisdictional claims in published London, UK, and Division of Psychiatry, University College London, London, maps and institutional affiliations. Startin et al. Alzheimer's Research & Therapy (2019) 11:26 Page 11 of 12 Author details 15. Mattsson N, Andreasson U, Zetterberg H, Blennow K. Alzheimer’s disease 1 Department of Forensic and Neurodevelopmental Sciences, Institute of neuroimaging I. Association of plasma neurofilament light with Psychiatry, Psychology and Neuroscience, King’s College London, 16 De neurodegeneration in patients with Alzheimer disease. JAMA Neurol. 2017; Crespigny Park, London SE5 8AF, UK. 2Division of Psychiatry, University 74:557–66. College London, London, UK. 3The LonDownS Consortium (London Down 16. Strydom A, Heslegrave A, Startin CM, Mok KY, Hardy J, Groet J, Nizetic D, Syndrome Consortium), London, UK. 4Maurice Wohl Clinical Neuroscience, Zetterberg H, Consortium L. Neurofilament light as a blood biomarker for Institute of Psychiatry, Psychology and Neuroscience, King’s College London, neurodegeneration in Down syndrome. Alzheimers Res Ther. 2018;10:39. London, UK. 5NIHR Biomedical Research Centre for Mental Health, Biomedical 17. Fortea J, Carmona-Iragui M, Benejam B, Fernandez S, Videla L, Barroeta I, Research Unit for Dementia at South London, and Maudsley NHS Alcolea D, Pegueroles J, Munoz L, Belbin O, et al. Plasma and CSF Foundation, London, UK. 6Department of Psychiatry and Neurochemistry, biomarkers for the diagnosis of Alzheimer’s disease in adults with Down Institute of Neuroscience and Physiology, Sahlgrenska Academy at the syndrome: a cross-sectional study. Lancet Neurol. 2018;17:860–9. University of Gothenburg, Mölndal, Sweden. 7Wallenberg Centre for 18. Holmes C, Cunningham C, Zotova E, Woolford J, Dean C, Kerr S, Culliford D, Molecular & Translational Medicine, University of Gothenburg, Gothenburg, Perry VH. Systemic inflammation and disease progression in Alzheimer Sweden. 8Department of Neurodegenerative Disease, Institute of Neurology, disease. Neurology. 2009;73:768–74. University College London, London, UK. 9Department of Molecular 19. Holmes C, El-Okl M, Williams AL, Cunningham C, Wilcockson D, Perry VH. Neuroscience, Institute of Neurology, University College London, London, UK. Systemic infection, interleukin 1beta, and cognitive decline in Alzheimer’s 10 Division of Life Science, Hong Kong University of Science and Technology, disease. J Neurol Neurosurg Psychiatry. 2003;74:788–9. Hong Kong, SAR, People’s Republic of China. 11Reta Lila Weston Institute, 20. Swardfager W, Lanctot K, Rothenburg L, Wong A, Cappell J, Herrmann N. A meta- Institute of Neurology, University College London, London, UK. 12Memory analysis of cytokines in Alzheimer’s disease. Biol Psychiatry. 2010;68:930–41. Unit, Neurology Department, Hospital de la Santa Creu i Sant Pau, Barcelona, 21. Engelhart MJ, Geerlings MI, Meijer J, Kiliaan A, Ruitenberg A, van Swieten JC, Spain. 13Department of Psychiatry, University of Oxford, Oxford, UK. 14Centre Stijnen T, Hofman A, Witteman JC, Breteler MM. Inflammatory proteins in plasma for Memory Disturbances, Laboratory of Clinical Neurochemistry, Section of and the risk of dementia: the Rotterdam study. Arch Neurol. 2004;61:668–72. Neurology, Department of Medicine, University of Perugia, Perugia, Italy. 22. D'Anna L, Abu-Rumeileh S, Fabris M, Pistis C, Baldi A, Sanvilli N, Curcio F, 15 Clinical Neurochemistry Laboratory, Sahlgrenska University Hospital, Gigli GL, D'Anna S, Valente M. Serum interleukin-10 levels correlate with Mölndal, Sweden. 16UK Dementia Research Institute at UCL, London, UK. cerebrospinal fluid amyloid beta deposition in Alzheimer disease patients. Neurodegener Dis. 2017;17:227–34. Received: 28 August 2018 Accepted: 21 February 2019 23. Harold D, Abraham R, Hollingworth P, Sims R, Gerrish A, Hamshere ML, Pahwa JS, Moskvina V, Dowzell K, Williams A, et al. Genome-wide association study identifies variants at CLU and PICALM associated with References Alzheimer’s disease. Nat Genet. 2009;41:1088–93. 1. Wiseman FK, Al-Janabi T, Hardy J, Karmiloff-Smith A, Nizetic D, Tybulewicz 24. Hollingworth P, Harold D, Sims R, Gerrish A, Lambert JC, Carrasquillo MM, VL, Fisher EM, Strydom A. A genetic cause of Alzheimer disease: mechanistic Abraham R, Hamshere ML, Pahwa JS, Moskvina V, et al. Common variants at insights from Down syndrome. Nat Rev Neurosci. 2015;16:564–74. ABCA7, MS4A6A/MS4A4E, EPHA1, CD33 and CD2AP are associated with 2. McCarron M, McCallion P, Reilly E, Mulryan N. A prospective 14-year Alzheimer’s disease. Nat Genet. 2011;43:429–35. longitudinal follow-up of dementia in persons with Down syndrome. J 25. Lambert JC, Ibrahim-Verbaas CA, Harold D, Naj AC, Sims R, Bellenguez C, Intellect Disabil Res. 2014;58:61–70. DeStafano AL, Bis JC, Beecham GW, Grenier-Boley B, et al. Meta-analysis of 3. Sinai A, Mokrysz C, Bernal J, Bohnen I, Bonell S, Courtenay K, Dodd K, Gazizova 74,046 individuals identifies 11 new susceptibility loci for Alzheimer’s D, Hassiotis A, Hillier R, et al. Predictors of age of diagnosis and survival of disease. Nat Genet. 2013;45:1452–8. Alzheimer’s disease in Down syndrome. J Alzheimers Dis. 2018;61:717–28. 26. Jones L, Holmans PA, Hamshere ML, Harold D, Moskvina V, Ivanov D, 4. Hithersay R, Startin CM, Hamburg S, Mok KY, Hardy J, Fisher EMC, Pocklington A, Abraham R, Hollingworth P, Sims R, et al. Genetic evidence Tybulewicz VLJ, Nizetic D, Strydom A. Association of dementia with implicates the immune system and cholesterol metabolism in the aetiology mortality among adults with Down syndrome older than 35 years. JAMA of Alzheimer’s disease. PLoS One. 2010;5:e13950. Neurol. 2019;76:152–60. 27. Conti E, Galimberti G, Piazza F, Raggi ME, Ferrarese C. Increased soluble 5. Wu J, Morris JK. The population prevalence of Down’s syndrome in England APPalpha, Abeta 1-42, and anti-Abeta 1-42 antibodies in plasma from Down and Wales in 2011. Eur J Hum Genet. 2013;21:1016–9. syndrome patients. Alzheimer Dis Assoc Disord. 2010;24:96–100. 6. Mann DMA, Davidson YS, Robinson AC, Allen N, Hashimoto T, Richardson A, 28. Lee NC, Yang SY, Chieh JJ, Huang PT, Chang LM, Chiu YN, Huang AC, Chien Jones M, Snowden JS, Pendleton N, Potier MC, et al. Patterns and severity of YH, Hwu WL, Chiu MJ. Blood beta-amyloid and tau in Down syndrome: a vascular amyloid in Alzheimer’s disease associated with duplications and comparison with Alzheimer’s disease. Front Aging Neurosci. 2016;8:316. missense mutations in APP gene, Down syndrome and sporadic Alzheimer’s 29. Head E, Doran E, Nistor M, Hill M, Schmitt FA, Haier RJ, Lott IT. Plasma amyloid- disease. Acta Neuropathol. 2018;136:569–87. beta as a function of age, level of intellectual disability, and presence of 7. Mann DM. Alzheimer’s disease and Down’s syndrome. Histopathology. 1988; dementia in Down syndrome. J Alzheimers Dis. 2011;23:399–409. 13:125–37. 30. Mehta PD, Capone G, Jewell A, Freedland RL. Increased amyloid beta 8. Zis P, Strydom A. Clinical aspects and biomarkers of Alzheimer’s disease in protein levels in children and adolescents with Down syndrome. J Neurol Down syndrome. Free Radic Biol Med. 2018;114:3–9. Sci. 2007;254:22–7. 9. Doran E, Keator D, Head E, Phelan MJ, Kim R, Totoiu M, Barrio JR, Small GW, 31. Iulita MF, Ower A, Barone C, Pentz R, Gubert P, Romano C, Cantarella RA, Elia F, Potkin SG, Lott IT. Down syndrome, partial trisomy 21, and absence of Buono S, Recupero M, et al. An inflammatory and trophic disconnect Alzheimer’s disease: the role of APP. J Alzheimers Dis. 2017;56:459–70. biomarker profile revealed in Down syndrome plasma: relation to cognitive 10. Prasher VP, Farrer MJ, Kessling AM, Fisher EM, West RJ, Barber PC, Butler AC. decline and longitudinal evaluation. Alzheimers Dement. 2016;12:1132–48. Molecular mapping of Alzheimer-type dementia in Down’s syndrome. Ann 32. Schupf N, Patel B, Silverman W, Zigman WB, Zhong N, Tycko B, Mehta PD, Neurol. 1998;43:380–3. Mayeux R. Elevated plasma amyloid beta-peptide 1-42 and onset of 11. Hardy J, Allsop D. Amyloid deposition as the central event in the aetiology dementia in adults with Down syndrome. Neurosci Lett. 2001;301:199–203. of Alzheimer’s disease. Trends Pharmacol Sci. 1991;12:383–8. 33. Mehta PD, Dalton AJ, Mehta SP, Kim KS, Sersen EA, Wisniewski HM. 12. Selkoe DJ. The molecular pathology of Alzheimer’s disease. Neuron. 1991;6: Increased plasma amyloid beta protein 1-42 levels in Down syndrome. 487–98. Neurosci Lett. 1998;241:13–6. 13. Margallo-Lana ML, Moore PB, Kay DW, Perry RH, Reid BE, Berney TP, Tyrer 34. Mehta PD, Mehta SP, Fedor B, Patrick BA, Emmerling M, Dalton AJ. Plasma SP. Fifteen-year follow-up of 92 hospitalized adults with Down’s syndrome: amyloid beta protein 1-42 levels are increased in old Down syndrome but incidence of cognitive decline, its relationship to age and neuropathology. J not in young Down syndrome. Neurosci Lett. 2003;342:155–8. Intellect Disabil Res. 2007;51:463–77. 35. Cavani S, Tamaoka A, Moretti A, Marinelli L, Angelini G, Di Stefano S, 14. Johnson KA, Schultz A, Betensky RA, Becker JA, Sepulcre J, Rentz D, Mormino E, Piombo G, Cazzulo V, Matsuno S, Shoji S, et al. Plasma levels of amyloid Chhatwal J, Amariglio R, Papp K, et al. Tau positron emission tomographic beta 40 and 42 are independent from ApoE genotype and mental imaging in aging and early Alzheimer disease. Ann Neurol. 2016;79:110–9. retardation in Down syndrome. Am J Med Genet. 2000;95:224–8. Startin et al. Alzheimer's Research & Therapy (2019) 11:26 Page 12 of 12 36. Matsubara E, Ghiso J, Frangione B, Amari M, Tomidokoro Y, Ikeda Y, concentrations and stable isotope labeling kinetics of human plasma specific Harigaya Y, Okamoto K, Shoji M. Lipoprotein-free amyloidogenic peptides in to central nervous system amyloidosis. Alzheimers Dement. 2017;13:841–9. plasma are elevated in patients with sporadic Alzheimer’s disease and 57. Janelidze S, Stomrud E, Palmqvist S, Zetterberg H, van Westen D, Jeromin A, Down’s syndrome. Ann Neurol. 1999;45:537–41. Song L, Hanlon D, Tan Hehir CA, Baker D, et al. Plasma beta-amyloid in 37. Tokuda T, Fukushima T, Ikeda S, Sekijima Y, Shoji S, Yanagisawa N, Tamaoka Alzheimer’s disease and vascular disease. Sci Rep. 2016;6:26801. A. Plasma levels of amyloid beta proteins Abeta1-40 and Abeta1-42(43) are 58. Blennow K. A review of fluid biomarkers for Alzheimer’s disease: moving elevated in Down’s syndrome. Ann Neurol. 1997;41:271–3. from CSF to blood. Neurol Ther. 2017;6:15–24. 38. Kasai T, Tatebe H, Kondo M, Ishii R, Ohmichi T, Yeung WTE, Morimoto M, 59. Buss L, Fisher E, Hardy J, Nizetic D, Groet J, Pulford L, Strydom A. Chiyonobu T, Terada N, Allsop D, et al. Increased levels of plasma total tau Intracerebral haemorrhage in Down syndrome: protected or predisposed? in adult Down syndrome. PLoS One. 2017;12:e0188802. F1000Res. 2016;5:876. 39. Kusters MA, Verstegen RH, Gemen EF, de Vries E. Intrinsic defect of the 60. Jones EL, Hanney M, Francis PT, Ballard CG. Amyloid beta concentrations in older immune system in children with Down syndrome: a review. Clin Exp people with Down syndrome and dementia. Neurosci Lett. 2009;451:162–4. Immunol. 2009;156:189–93. 61. Prasher VP, Sajith SG, Mehta P, Zigman WB, Schupf N. Plasma beta-amyloid 40. Goldacre MJ, Wotton CJ, Seagroatt V, Yeates D. Cancers and immune and duration of Alzheimer’s disease in adults with Down syndrome. Int J related diseases associated with Down’s syndrome: a record linkage study. Geriatr Psychiatry. 2010;25:202–7. Arch Dis Child. 2004;89:1014–7. 62. Matsuoka Y, Andrews HF, Becker AG, Gray AJ, Mehta PD, Sano MC, Dalton AJ, 41. Wilcock DM, Griffin WS. Down’s syndrome, neuroinflammation, and Aisen PS. The relationship of plasma Abeta levels to dementia in aging Alzheimer neuropathogenesis. J Neuroinflammation. 2013;10:84. individuals with Down syndrome. Alzheimer Dis Assoc Disord. 2009;23:315–8. 42. Cetiner S, Demirhan O, Inal TC, Tastemir D, Sertdemir Y. Analysis of peripheral 63. Schupf N, Patel B, Pang D, Zigman WB, Silverman W, Mehta PD, Mayeux R. blood T-cell subsets, natural killer cells and serum levels of cytokines in Elevated plasma beta-amyloid peptide Abeta(42) levels, incident dementia, children with Down syndrome. Int J Immunogenet. 2010;37:233–7. and mortality in Down syndrome. Arch Neurol. 2007;64:1007–13. 43. Zaki ME, El-Bassyouni HT, Tosson AM, Youness E, Hussein J. Coenzyme Q10 64. Rafii MS, Wishnek H, Brewer JB, Donohue MC, Ness S, Mobley WC, Aisen PS, and pro-inflammatory markers in children with Down syndrome: clinical Rissman RA. The Down syndrome biomarker initiative (DSBI) pilot: proof of and biochemical aspects. J Pediatr. 2017;93:100–4. concept for deep phenotyping of Alzheimer’s disease biomarkers in Down 44. Corsi MM, Dogliotti G, Pedroni F, Palazzi E, Magni P, Chiappelli M, Licastro F. syndrome. Front Behav Neurosci. 2015;9:239. Plasma nerve growth factor (NGF) and inflammatory cytokines (IL-6 and 65. Matsubara E, Sekijima Y, Tokuda T, Urakami K, Amari M, Shizuka-Ikeda M, MCP-1) in young and adult subjects with Down syndrome: an interesting Tomidokoro Y, Ikeda M, Kawarabayashi T, Harigaya Y, et al. Soluble Abeta pathway. Neuro Endocrinol Lett. 2006;27:773–8. homeostasis in AD and DS: impairment of anti-amyloidogenic protection by 45. Nateghi Rostami M, Douraghi M, Miramin Mohammadi A, Nikmanesh B. lipoproteins. Neurobiol Aging. 2004;25:833–41. Altered serum pro-inflammatory cytokines in children with Down’s 66. Startin CM, Hamburg S, Hithersay R, Al-Janabi T, Mok KY, Hardy J, syndrome. Eur Cytokine Netw. 2012;23:64–7. LonDownS Strydom A. Cognitive markers of preclinical and prodromal 46. Zhang Y, Che M, Yuan J, Yu Y, Cao C, Qin XY, Cheng Y. Aberrations in Alzheimer’s disease in Down syndrome. Alzheimers Dement. 2019;15:245–57. circulating inflammatory cytokine levels in patients with Down syndrome: a 67. Coppus AM, Fekkes D, Verhoeven WM, Evenhuis HM, van Duijn CM. meta-analysis. Oncotarget. 2017;8:84489–96. Neopterin and the risk of dementia in persons with Down syndrome. 47. Startin CM, Hamburg S, Hithersay R, Davies A, Rodger E, Aggarwal N, Al- Neurosci Lett. 2009;458:60–4. Janabi T, Strydom A. The LonDownS adult cognitive assessment to study 68. Zis P, Strydom A, Buckley D, Adekitan D, McHugh PC. Cognitive ability in cognitive abilities and decline in Down syndrome. Wellcome Open Res. Down syndrome and its relationship to urinary neopterin, a marker of 2016;1:11. activated cellular immunity. Neurosci Lett. 2017;636:254–7. 48. Westwood S, Baird AL, Hye A, Ashton NJ, Nevado-Holgado AJ, Anand SN, 69. Dekker AD, Coppus AM, Vermeiren Y, Aerts T, van Duijn CM, Kremer BP, Liu B, Newby D, Bazenet C, Kiddle SJ, et al. Plasma protein biomarkers for Naude PJ, Van Dam D, De Deyn PP. Serum MHPG strongly predicts the prediction of CSF amyloid and tau and [(18)F]-flutemetamol PET scan conversion to Alzheimer’s disease in behaviorally characterized subjects result. Front Aging Neurosci. 2018;10:409. with Down syndrome. J Alzheimers Dis. 2015;43:871–91. 70. Landt J, Ball SL, Holland AJ, Hon J, Owen A, Treppner P, Herbert J. Age-related 49. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. changes in plasma dehydroepiandrosterone levels in adults with Down’s Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work syndrome and the risk of dementia. J Neuroendocrinol. 2011;23:450–5. Group under the auspices of Department of Health and Human Services 71. Zis P, Dickinson M, Shende S, Walker Z, Strydom A. Oxidative stress and Task Force on Alzheimer’s disease. Neurology. 1984;34:939–44. memory decline in adults with Down syndrome: longitudinal study. J 50. Majbour NK, Chiasserini D, Vaikath NN, Eusebi P, Tokuda T, van de Berg W, Alzheimers Dis. 2012;31:277–83. Parnetti L, Calabresi P, El-Agnaf OM. Increased levels of CSF total but not 72. Zis P, McHugh P, McQuillin A, Pratico D, Dickinson M, Shende S, Walker Z, oligomeric or phosphorylated forms of alpha-synuclein in patients Strydom A. Memory decline in Down syndrome and its relationship to diagnosed with probable Alzheimer’s disease. Sci Rep. 2017;7:40263. iPF2alpha, a urinary marker of oxidative stress. PLoS One. 2014;9:e97709. 51. Alcolea D, Martinez-Lage P, Sanchez-Juan P, Olazaran J, Antunez C, Izagirre 73. Ashton NJ, Scholl M, Heurling K, Gkanatsiou E, Portelius E, Hoglund K, A, Ecay-Torres M, Estanga A, Clerigue M, Guisasola MC, et al. Amyloid Brinkmalm G, Hye A, Blennow K, Zetterberg H. Update on biomarkers for precursor protein metabolism and inflammation markers in preclinical amyloid pathology in Alzheimer’s disease. Biomark Med. 2018;12:799–812. Alzheimer disease. Neurology. 2015;85:626–33. 74. Hamlett ED, Goetzl EJ, Ledreux A, Vasilevko V, Boger HA, LaRosa A, Clark D, 52. Wiseman FK, Pulford LJ, Barkus C, Liao F, Portelius E, Webb R, Chavez- Carroll SL, Carmona-Iragui M, Fortea J, et al. Neuronal exosomes reveal Gutierrez L, Cleverley K, Noy S, Sheppard O, et al. Trisomy of human Alzheimer’s disease biomarkers in Down syndrome. Alzheimers Dement. chromosome 21 enhances amyloid-beta deposition independently of an 2017;13:541–9. extra copy of APP. Brain. 2018;141:2457–74. 75. Coppus AM, Schuur M, Vergeer J, Janssens AC, Oostra BA, Verbeek MM, van 53. Lambert JC, Schraen-Maschke S, Richard F, Fievet N, Rouaud O, Berr C, Duijn CM. Plasma beta amyloid and the risk of Alzheimer’s disease in Down Dartigues JF, Tzourio C, Alperovitch A, Buee L, Amouyel P. Association of syndrome. Neurobiol Aging. 2012;33:1988–94. plasma amyloid beta with risk of dementia: the prospective Three-City 76. Schupf N, Zigman WB, Tang MX, Pang D, Mayeux R, Mehta P, Silverman W. Study. Neurology. 2009;73:847–53. Change in plasma AB peptides and onset of dementia in adults with Down 54. Fandos N, Perez-Grijalba V, Pesini P, Olmos S, Bossa M, Villemagne VL, syndrome. Neurology. 2010;75:1639–44. Doecke J, Fowler C, Masters CL, Sarasa M, Group AR. Plasma amyloid beta 42/40 ratios as biomarkers for amyloid beta cerebral deposition in cognitively normal individuals. Alzheimers Dement (Amst). 2017;8:179–87. 55. Nakamura A, Kaneko N, Villemagne VL, Kato T, Doecke J, Dore V, Fowler C, Li QX, Martins R, Rowe C, et al. High performance plasma amyloid-beta biomarkers for Alzheimer’s disease. Nature. 2018;554:249–54. 56. Ovod V, Ramsey KN, Mawuenyega KG, Bollinger JG, Hicks T, Schneider T, Sullivan M, Paumier K, Holtzman DM, Morris JC, et al. Amyloid beta