Expert Review of Proteomics
ISSN: 1478-9450 (Print) 1744-8387 (Online) Journal homepage: http://www.tandfonline.com/loi/ieru20
Diagnostic function of the neuroinflammatory biomarker YKL-40 in Alzheimer’s disease and other neurodegenerative diseases Filippo Baldacci, Simone Lista, Enrica Cavedo, Ubaldo Bonuccelli & Harald Hampel To cite this article: Filippo Baldacci, Simone Lista, Enrica Cavedo, Ubaldo Bonuccelli & Harald Hampel (2017) Diagnostic function of the neuroinflammatory biomarker YKL-40 in Alzheimer’s disease and other neurodegenerative diseases, Expert Review of Proteomics, 14:4, 285-299, DOI: 10.1080/14789450.2017.1304217 To link to this article: http://dx.doi.org/10.1080/14789450.2017.1304217
Accepted author version posted online: 10 Mar 2017. Published online: 20 Mar 2017. Submit your article to this journal
Article views: 27
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ieru20 Download by: [93.144.75.233]
Date: 30 March 2017, At: 13:44
EXPERT REVIEW OF PROTEOMICS, 2017 VOL. 14, NO. 4, 285–299 http://dx.doi.org/10.1080/14789450.2017.1304217
REVIEW
Diagnostic function of the neuroinflammatory biomarker YKL-40 in Alzheimer’s disease and other neurodegenerative diseases Filippo Baldaccia,b, Simone Listab, Enrica Cavedob,c, Ubaldo Bonuccellia and Harald Hampelb a
Department of Clinical and Experimental Medicine, University of Pisa, Pisa, Italy; bAXA Research Fund UPMC Chair, Sorbonne Universités, Université Pierre et Marie Curie (UPMC) Paris 06, Inserm, CNRS, Institut du cerveau et de la moelle (ICM), Département de Neurologie, Institut de la Mémoire et de la Maladie d’Alzheimer (IM2A), Hôpital Pitié-Salpêtrière, Boulevard de l’hôpital, Paris, France; cIRCCS Istituto Centro San Giovanni di Dio-Fatebenefratelli, Brescia, Italy ABSTRACT
ARTICLE HISTORY
Introduction: Neuroinflammation is a crucial mechanism in the pathophysiology of neurodegenerative diseases pathophysiology. Cerebrospinal fluid (CSF) YKL-40 – an indicator of microglial activation − has recently been identified by proteomic studies as a candidate biomarker for Alzheimer’s disease (AD). Areas covered: We review the impact of CSF YKL-40 as a pathophysiological biomarker for AD and other neurodegenerative diseases. CSF YKL-40 concentrations have been shown to predict progression from prodromal mild cognitive impairment to AD dementia. Moreover, a positive association between CSF YKL-40 and other biomarkers of neurodegeneration – particularly total tau protein − has been reported during the asymptomatic preclinical stage of AD and other neurodegenerative diseases. Albeit preliminary, current data do not support an association between APOE-ε4 status and CSF YKL-40 concentrations. When interpreting the diagnostic/prognostic significance of CSF YKL-40 concentrations in neurodegenerative diseases, potential confounders – including age, metabolic and cardiovascular risk factors, diagnostic criteria for selecting cases/controls – need to be considered. Expert opinion/commentary: CSF YKL-40 represents a pathophysiological biomarker reflecting immune/inflammatory mechanisms in neurodegenerative diseases, associated with tau protein pathology. Besides being associated with tau pathology, CSF YKL-40 adds to the growing array of biomarkers reflecting distinct molecular brain mechanisms potentially useful for stratifying individuals for biomarker-guided, targeted anti-inflammatory therapies emerging from precision medicine.
Received 21 January 2017 Accepted 6 March 2017
1. Background Neuroinflammation is a key mechanism in the pathophysiology of primary neurodegenerative diseases [1] – which presents a continuous genetic, biochemical, pathological and clinical spectrum and can collectively be conceptualized as protein misfolding disorders or proteinopathies of the central nervous system (CNS). Misfolded proteins can act as catalysts for the activation of microglial cells, the main resident cells of the brain involved in the maintenance of the neural environment, injury, and repair [2]. Neuroinflammation has been extensively studied in relation to Alzheimer’s disease (AD) – a multifactorial and polygenic neurodegenerative disease [3] mainly related to the abnormal accumulation of amyloid beta (Aβ) plaques [4]. Extracellular Aβ aggregates and intracellular hyperphosphorylated tau protein in neurofibrillary tangles constitute the two major pathophysiological hallmarks of AD [5,6]. Other less-specific alterations of AD include diffuse neuronal and axonal injury, synaptic disruption, reactive gliosis, and chronic neuroinflammatory changes [7]. Growing evidence indicates that neuroinflammation is involved in AD initiation [2,8–12] and in the self-perpetuating deposition of Aβ plaques [9,13]. Nonetheless, the question as to whether
KEYWORDS
YKL-40; neuroinflammation; biomarkers; cerebrospinal fluid; Alzheimer’s disease; mild cognitive impairment; dementia; neurodegenerative diseases
neuroinflammation is a cause, a contributor, a reactive phenomenon, or even a protective factor [2,11] in AD remains open [14,15]. The role of neuroinflammation in the pathogenesis of AD is supported by genetic studies [16,17] and pathophysiological evidence of its occurrence in the AD brain. In this regard, activation of microglia, astrocytes [13,18,19] and the complement system [20], increased release of chemokines and cytokines [13,21,22], caspase activation [13,23], and higher nitric oxides and reactive oxygen species production [24] have all been reported in AD. In addition, 9 genes/loci that govern the immune response have been previously linked to AD susceptibility [16,17]. Another line of evidence in support of neuroinflammatory mechanisms underlying AD is derived from neuroimaging studies. To date, the most important in vivo marker of neuroinflammation is the brain uptake of the 11CPK11195 radiotracer – a positron emission tomography (PET) marker of the translocator protein (TSPO) located on microglia and astroglia [9,19]. Increased brain glial activation as reflected by elevated 11C-PK11195 uptake has been reported in AD and amnestic MCI subjects [25]. Despite some conflicting data [26], imaging evidence of neuroinflammation has been identified in
CONTACT Harald Hampel
[email protected] Département de Neurologie, AXA Research Fund & UPMC Chair, Sorbonne Universités, Université Pierre et Marie Curie (UPMC) Paris 06, Inserm, CNRS, Institut du cerveau et de la moelle (ICM), Institut de la Mémoire et de la Maladie d’Alzheimer (IM2A), Hôpital Pitié-Salpêtrière, 47 Boulevard de l’Hôpital, CEDEX 13, 75651, Paris, France © 2017 Informa UK Limited, trading as Taylor & Francis Group
286
F. BALDACCI ET AL.
areas of Aβ deposition [26–28]. Growing evidence also indicates that neuroinflammation is not limited to AD but may also occur in other neurodegenerative diseases. For example, genetic polymorphisms in genes involved in the inflammatory response – including tumor necrosis factor (TNF) and TNF receptor 1, TREM2, IL-1β, and CD14 – have been associated with the risk of Parkinson’s disease (PD) [10]. Intriguingly, a TREM2 genetic variant has been also related to frontotemporal dementia (FTD) [29]. Epidemiological data also suggest that chronic use of anti-inflammatory drugs is associated with a reduced incidence of PD [30]. Glial activation as reflected by 11 C-PK11195 uptake on PET imaging has been reported in PD and other atypical Parkinsonisms [31], Huntington disease (HD) [32], amyotrophic lateral sclerosis (ALS) [33], and FTD [34]. Interestingly, evidence of activated microglia and astroglia has been associated with accumulation of misfolded proteins in patients with Parkinsonian syndromes, ALS, and HD [10,11,32]. An increased production of proinflammatory cytokines has been reported in clinical cohorts of PD, ALS, and HD patients [10,11]. Owing to the central role of neuroinflammation in the pathophysiology of AD and other neurodegenerative diseases, interest in novel neuroinflammatory biomarkers in the asymptomatic preclinical, prodromal, and symptomatic disease stages has gained momentum. However, a reliable and individual marker that accurately reflects brain inflammation has not yet been identified. A recent study based on the use of two-dimensional differential gel electrophoresis (2-D DIGE) in conjunction with liquid chromatography-tandem mass spectrometry (LCMS/MS) has suggested that YKL-40 may serve as a potential novel CSF neuroinflammatory biomarker of AD [35]. YKL-40 is a 40-kDa heparin- and chitin-binding glycoprotein also known as chitinase-3-like protein 1 (CHI3L1) [36]. The abbreviation YKL-40 is based on the one-letter code for the first three N-terminal amino acids − tyrosine (Y), lysine (K), and leucine (L) − and its molecular weight of 40 kDa [37,38]. The human CHI3L1 gene encoding for YKL-40 has been localized to chromosome 1q31q32 [39]. Notably, YKL-40 shows a chitin-binding ability but does not possess any chitinase activity [40,41]. YKL-40 plays a role in various cellular responses − including proliferation, differentiation, and survival [41,42] − but its modulatory effects on inflammation have not been entirely elucidated [41–44]. YKL-40 is expressed in macrophages and can be considered a specific differentiation marker for this cell type [36,42,45]. Other cells that are able to express YKL-40 include mast cells [46], chondrocytes, fibroblasts, vascular smooth muscle cells, endothelial cells [42], astroglia, and microglia [35,47–49]. Among neurological pathologies other than non-primary neurodegenerative diseases, CSF YKL-40 concentrations have been studied in stroke [50–53], multiple sclerosis [54–60], and traumatic brain injury [47] a well-known possible contributor to AD and dementia developing [2]. Particularly, in multiple sclerosis (MS), the most important neuroinflammatory disease model in humans, CSF YKL-40 helped to predict the conversion to clinically definite MS in subjects with clinically isolated syndromes [59,60] and discriminate primary progressive from relapsing remitting MS [57]. Moreover, serum YKL-40 might represent a potential biomarker for response to immunotherapy in MS [57,58]. Recently, the CSF concentrations of YKL-40 have been
investigated in patients with AD, subjects with MCI or in the prodromal phase of the AD, as well as in the asymptomatic/ preclinical phases of the disease. Generally, most – but not all [61] – published reports demonstrated increased YKL-40 concentrations in AD. A few studies have also focused on CSF concentrations of YKL-40 in AD compared with other dementing disorders [35,49,61–66] as well as non-AD neurodegenerative diseases [67–72].
1.1. Objective and literature research The objective of this review is to summarize the published literature on YKL-40 as a candidate biomarker for AD and other neurodegenerative diseases. By using the research terms ‘YKL-40’ and ‘chitinase-3-like protein 1,’ we interrogated the PubMed database for studies published in English between 2010 [the year on which the first paper in the field was released [35]] and October 2016 which measured YKL-40 concentrations in AD or other neurodegenerative diseases. A total of 29 articles were identified, of which 21 were focusing on AD. We analyzed for each available study (I) the target population; (II) the diagnostic and/or prognostic significance of cerebrospinal fluid (CSF) YKL-40 concentrations; and (III) the methodology used for data analysis. As far as AD was concerned, we merged the International Working Group-2 (IWG-2) criteria [73,74] and the National Institute on Aging and the Alzheimer’s Association (NIA-AA) diagnostic guidelines whenever possible [75–77]. For example, patients with prodromal AD [74] and MCI due to AD [75], were grouped together as well as asymptomatic-at-risk of AD [73], and preclinical AD [77] for the purpose of analysis. We specifically noted when the diagnosis was solely based on clinical criteria [78,79]. Particularly, the discriminatory ability of YKL-40 to correctly allocate participants to the different diagnostic groups was assessed as follows: ‘excellent’ (area under ROC curve [AUROC] 0.90–1.00), ‘good’ (AUROC 0.80–0.89), ‘fair’ (AUROC 0.70–0.79), ‘poor’ (AUROC 0.60–0.69), or ‘fail’ (i.e. no discriminatory capacity) (AUROC 0.50–0.59) [80].
1.2. YKL-40 and AD dementia stage In the first study evaluating the role of YKL-40 in AD dementia patients, Craig-Shapiro and colleagues designed a two-step analysis. In the first step, they performed an unbiased mass spectrometry (MS)-based approach, namely the 2-D DIGE LCMS/MS, to identify protein biomarkers for AD in a cohort of 24 cognitive healthy subjects with a clinical dementia rating (CDR) scale equal to zero (CDR = 0) and 23 mild demented patients (CDR = 1) which were classified following AD pathology. In this preliminary phase, CSF YKL-40 was recognized as a candidate biomarker of AD. In the second step, CSF YKL-40 was used in a validation cohort of 292 individuals, with a 5year follow-up, a baseline CDR score ranging from 0 to 1, and not classified following AD pathology. Notably, the ratio between YKL-40 and the 42-amino acid-long Aβ peptide (Aβ1-42) in the CSF (YKL-40/Aβ1-42) was reported as a useful biomarker to predict the onset of cognitive decline. In particular, subjects in the upper tertile of CSF YKL-40/Aβ1-42
EXPERT REVIEW OF PROTEOMICS
converted faster than those with lower ratios (lower tertiles) (Table 1). A similar trend was obtained to predict the progression of cognitive impairment from CDR = 0.5 to CDR = 1 [35] (Table 1). On the contrary, plasma measurement of YKL-40 did not show any utility in predicting cognitive decline although plasma YKL-40 concentrations in subjects with a CDR = 0 were significantly lower than in patients with CDR≥0.5 [35]. In contrast, Choi and colleagues disclosed in a different cohort that plasma YKL-40 were higher in mild clinical AD compared with controls and MCI subjects. Interestingly, plasma concentrations of YKL-40 were not elevated in moderate/severe AD, suggesting that plasma YKL-40 increase probably occurred in early AD phases [81]. Finally, a recent meta-analysis reported a large but not significant effect size for YKL-40 plasma/serum concentrations in AD versus controls [82]. However, further conclusions were limited by the lack of additional studies evaluating plasma/serum concentrations of YKL-40 in AD for diagnostic purposes. From the same cohort evaluated by Craig-Shapiro and colleagues, Perrin and colleagues also reported that the combination of the CSF YKL-40, neuronal cell adhesion molecule (NrCAM), and tau showed the highest AUROC in discriminating individuals with a CDR = 0 from individuals with a CDR≥0, highlighting a potential utility of YKL-40 in the classification of dementia [83] (Table 1). In two other studies [66,84], the use of CSF YKL-40 generated remarkable AUROCs when it was used as the only diagnostic biomarker (Table 1). Both studies reported a control group significantly younger than AD, possibly affecting the AUROCs results since several analyses clearly indicated that CSF YKL-40 concentrations positively correlated with age [35,66,84]. Moreover, Wennstrom and colleagues highlighted that the control subjects were non-demented and indicated that they might report subtle cognitive symptoms, although cognitively normal after testing. Conversely, in two independent analyses, CSF YKL-40 did not ameliorate the diagnostic accuracy. In particular, no differences among AD, other dementia, MCI and controls were reported in terms of CSF YKL-40 concentrations not improving AD diagnostic accuracy [61]; moreover, in the study by Janelidze and colleagues, CSF YKL-40 and Aβ1-42/ YKL-40 ratio were shown not to significantly contribute to the accuracy of AD diagnosis when they were compared with established core biomarkers [64] (Table 1). CSF YKL-40 was also tested as potential biomarker in a large multicenter study without improving significantly the diagnostic accuracy between typical and atypical forms of AD and between subgroups of atypical forms of AD [85]. Additionally, four studies reported higher CSF YKL-40 concentrations in AD versus controls [62,65,67,89] while no significant differences were reported in two independent analyses [61,90]. Interestingly, in a longitudinal follow-up study, Kester and colleagues reported conflicting data showing low baseline CSF YKL-40 concentrations in AD patients with a marked cognitive decline during follow-up (mean = 2.0 years), and high CSF concentrations in subjects with MCI progressing to AD. YKL-40 concentration dynamically increased with time in the MCI and AD groups but not in controls. Then, YKL-40 measurements remained stable in a subgroup of AD performing the second spinal tap after 6 months [65]. Notably, as a major methodological limitation,
287
this study did not exclude individuals with neurodegenerative diseases in the control group [65] (Table 1). The association of YKL-40 with CSF core biomarkers of AD was further assessed: YKL-40 correlated with Aβ1-42 and total tau (t-tau) protein in one study [64] and with only tau protein in an independent investigation [82]; no correlation with any core biomarker was found in two studies [66,84], thus stressing that CSF YKL-40 concentrations may present a pathophysiological state marker reaching a plateau level during the clinical and symptomatic phase of the disease. Finally, four studies revealed a negative association of CSF YKL-40 concentrations with the APOE-ε4 allele in AD, although the final reflections were limited since the APOE genotype was not systematically examined [35,62,64,66]. In conclusion, CSF YKL-40 concentrations may have diagnostic value, particularly for AD dementia in comparison to cognitively healthy agematched controls and may predict the progression of cognitive decline from prodromal MCI to AD dementia. The analyses conducted so far, however, stressed the notion that CSF YKL40 concentrations do not to discriminate between typical and atypical AD phenotypes.
1.3. YKL-40 and the prodromal phase of AD (or MCI due to AD stage) Nine studies have scrutinized CSF YKL-40 concentrations in individuals with MCI [49,61–65,88,90,91]. In a cross-sectional study, Antonel and colleagues reported a good diagnostic accuracy of CSF YKL-40 to differentiate between prodromal AD and controls [88] (Table 1). Notably, subjects with subjective memory complaints were included in the controls, thus potentially affecting the results since this population is considered to be at risk for AD. Similarly, Hellwig and colleagues confirmed the potential diagnostic utility of CSF YKL-40 in distinguishing MCI subjects with an underlying AD pathology from other MCI individuals, although other novel CSF biomarkers, such as neurogranin, showed higher diagnostic performance than YKL-40. Nonetheless, it should be noticed that (I) this study lacked of a cognitively normal control group and (II) MCI due to AD patients were merged with AD and compared with a group of patients not showing any AD pathology (Table 1) [63]. Moreover, higher concentrations of CSF YKL-40 might help predict MCI to AD progression and conversion. In a longitudinal study, baseline concentrations of CSF YKL-40 discriminated between MCI converters to AD dementia and stable MCI subjects [67]. However, in this analysis, the combined biomarker ratio YKL-40/Aβ1–42 performed superior than YKL-40 alone (Table 1). These observations were then confirmed by Kester and colleagues in a large sample of MCI subjects [65]. Indeed, higher concentrations of CSF YKL-40 at baseline were predictive of conversion from MCI to AD for the highest and middle tertiles (Table 1). Additionally, similar results were achieved for the conversion of MCI to all types of dementia for the highest and middle tertiles (Table 1). However, there is no general agreement on a potential added value of CSF YKL-40 in the diagnosis of subjects with MCI. In this regard, Janelidze and colleagues reported that CSF YKL-40 was not ‘essential’ in improving the diagnostic accuracy compared with the core biomarkers to discriminate
CHC = 14
Rodrigues 2016 [86] (cross-sectional study) Kester 2015 [65] (longitudinal study)
No control group.
169 middle aged, cognitive normal (CDR 0) subjects (age 45–74). CHC = 30
Hellwig 2015 [63] (cross-sectional study)
Sutphen 2015 [87] (longitudinal study)
CNC = 25
CNC = 43 (including SCI = 25)
CHC = 24
Rosén 2014 [84] (cross-sectional study)
Antonell 2014 [88] (cross-sectional study)
Alcolea 2014 [62] (cross-sectional study)
Magdalinou 2015 [69] (cross-sectional study)
Non-demented controls = 44
Wennstrom 2015 [66] (cross-sectional study)
CNC = 37 (SCI = 31; epileptics = 2, psychiatrics = 2; HC = 2) SCI-MCI = 6 SCI-AD = 3 SCI-VaD = 1
CHC = 21
Controls CHC = 53
Hall 2016 [68] (longitudinal study)
First author (type of study) Janelidze 2016 [64] (longitudinal study)
PreAD§ = 18 ProAD° = 22 iRBD = 12 AD* = 59 aMCI** = 45 naMCI**+SCI = 44; FTD = 22
25 AD°
AD = 49* DLB = 36 PD = 61 AD§ = 39 MCI due to AD§ = 13 MCI not due to AD: 29 OD = 14 (FTD = 7, VaD = 3, DLB = 1, UK = 3) 14 subjects cognitively declined (CDR 0,5–1) at follow-up PSP = 33 MSA = 31 PD = 31 AD = 26 FTD = 16 CBS = 14
MCI-AD** = 36 sMCI** = 17 MCI-OD** = 8
Premanifest HD = 3 HD = 20 AD* = 65
PD = 63
Patients AD* = 74 sMCI** = 62 MCI-AD** = 35 DLB/PDD = 47 VaD = 34 FTD = 33
ProAD vs. CNC AUROC = 0.83 (cutoff = 316.01 ng/ml with 77.3% of sensitivity and 81.4% of specificity). AD vs. FTD No significantly contribute to diagnostic accuracy
All Parkinsonian syndromes vs. HC (AUROCs = 0.98, CI 0.97–0.99), PD vs. MSA,PSP, CBS (AUROC = 0.95, CI 0.88–0.99), PD vs. PSP (AUROC = 0.95, CI 0.87.0.99), PD vs. CBS (AUROC = 0.98, CI 0.97–0.99), PD vs. MSA (AUROC = 0.96, CI 0.91–0.99), PSP vs. MSA (AUROC = 0.84, CI 0.73–0.94) AD vs. CNC AUROC = 0.88 (CI 95% 0.78–0.98)
No significantly contribute to predict cognitive decline onset.
Diagnostic/Prognostic value of CSF YKL-40 YKL-40 AUROC and Aβ42/YKL-40 AUROC did not improve (in comparison to core biomarkers) diagnostic accuracy of AD vs. OD (AUROCs: 0.60 [95% CI 0.52–0.69], 0.78 [95% CI 0.72–0.85]), AD vs. MCI-AD converters (AUROCs: 0.61 [CI 95% 0.50–0.72], 0.72 [95% CI 0.63–0–82]) sMCI vs. MCI-AD converters (AUROCs: 0.69 [95% CI 0.58–0.80], 0.82 [96% CI 0.74–0.91], respectively). PD patients showed increased concentrations of YKL-40 during the follow-up in comparison to baseline values. HD (including premanifest HD) vs. HC with AUROC = 0.80. Higher baseline level of YKL-40 is predictive for conversion to AD in MCI for the highest and middle tertiles (HR = 3.0 95% CI 1.1–7.9, HR = 2.9 95% CI 1.0-8.1, respectively) and for conversion to all types of dementia for the highest and middle tertiles (HR = 3.2 95% CI 1.3–7.9, HR = 2.9 95% CI 1.1–7.2). AD vs. controls AUROC = 0.82 AD vs. DLB AUROC = 0.74 AD vs. PD AUROC = 0.82 AD+ MCI due to AD vs. MCI not due to AD+ OD patients AUROC = 0.74.
Table 1. The utility of YKL-40 in the diagnostic and progression evaluation of neurodegenerative diseases.
(Continued )
Cognitive normal controls significantly younger. AD diagnostic criteria both clinic and based on tau/Aβ42 ratio. FTD population was heterogenous: bvFTD (n = 12), progressive non-fluent aphasia (n = 7), and semantic dementia (n = 3).
Controls were cognitive normal individuals with psychometric tests within 1.5 SD. Patients were older than controls. Subjects with subjective memory complain were included.
The number of progressors is low (14 subjects). Serial spinal taps at 3-year intervals at follow-up, mean 6 years YKL-40 was used in combination with other 8 CSF biomarkers (NFL, sAPPα, sAPPβ, α-synuclein, MCP-1, Aβ42, t-tau, p-tau).
Clinical diagnostic criteria for OD patients were not clearly specified
Non-demented controls were SCI. They are younger than patients.
YKL40 and IL-6 had the higher diagnostic accuracy (AUROC = 0.86) MCI patients are older than controls that included individuals with neurological diseases and SCI.
The increase in YKL-40 CSF concentrations over 2 years negatively correlated with cognitive function
Notes The controls are cognitive normal without subjective complains and psychiatric disorders. The median follow-up for sMCI was 5.8 years (3.0–9.6). FTD consisted of 32 bvFTD subtype and 1 semantic variant.
288 F. BALDACCI ET AL.
CDR 0 (CHC) = 198
CDR 0 (CHC) = 198
CHC = 20
Perrin 2011 [83] (longitudinal study)
Craig-Schapiro 2010 [35] (longitudinal study)
Mattson 2011 [61] (longitudinal study)
AD* = 25 MCI** = 24 (sMCI = 13, MCI-AD converters = 7 MCI-OD converters = 4) OD = 11 (VaD = 7, DLB = 4)
AD* = 94 (CDR 0.5 = 65; CDR 1 = 29) FTD = 9 PSP = 6
PD = 50 MSA = 37 PSP = 32 CBS = 10 AD* = 94 (CDR 0.5 = 65; CDR 1 = 29)
Patients AD* = 96 sMCI** = 81 MCI-AD** = 61 MCI-VaD = 19 MCI-OD = 9
The combination of YKL-40, NrCAM and tau discriminated with the highest AUROC (0.90) CDR 0 from CDR>0. YKL-40/abeta42 ratio was predictive of the onset of cognitive decline (from CDR = 0 to CDR = 0.5) in the upper tertile of CSF YKL-40/ Aβ42 in comparison with lower tertile (HR = 3.35 CI 95% 1.42–7.90, p < 0.05) and of a trend of cognitive impairment progression (from CDR 0.5 to CDR 1) (HR = 2.63 95% CI 1.10–6.32, p = 0.065). No significantly contribute to diagnostic accuracy
Diagnostic/Prognostic value of CSF YKL-40 MCI-AD vs. sMCI AUROC = 0.62 (95% CI 0.53– 0.70). MCI-VaD vs. sMCI AUROC = 0.65 (95% CI 0.52– 0.78) YKL-40/Aβ42: MCI-AD vs. sMCI AUROC = 0.81 (95% CI 0.74–0.88) PD vs. HC (cut-off = 126,368 ng/L with a specificity of 81.5% and a sensitivity of 60.5%).
The clinical follow-up period has a median value of 3 yy. The samples did not differ in age and sex. Cardiovascular comorbidity has been excluded by controls.
Data reported are relative to the validation cohort study. The follow-up period of the validation cohort was up to 5 yy Data reported are relative to the validation cohort study. The follow-up period of the validation cohort was up to 5 yy. CSF YKL-40> FTD> AD> PSP. FTD group included 6 bvFTD and 3 progressive aphasia
YKL-40 concentrations were more elevated in tauopathies than synucleinopathies. PD and MSA patients were younger than CHC, PSP and CBS.
Notes Follow-up of 5,7 yy. The MCI-OD converters were diagnosed as VaD (n = 19), DLB (n = 4), PSP (n = 3), normal pressure hydrocephalus (n = 1).
AD: Alzheimer’s disease; AUROC: area under receiver operating curve; CBS: corticobasal syndrome; CI: confidential interval; PreAD: preclinical AD; ProAD: prodromal AD; bvFTD: behavioral variant of frontotemporal dementia; CDR: cognitive dementia rating; CHC: cognitively healthy controls; CNC: cognitively normal controls; FTD: frontotemporal dementia; HR: hazard ratio; IWG: International Working Group; DLB: Lewy body disease dementia; MCI: mild cognitive impairment; MCI-AD = MCI converters to AD; MCI-OD: MCI converters to OD; MCI-VaD: MCI converters to VaD; aMCI: amnestic MCI; naMCI: non-amnestic MCI; sMCI: stable MCI; monocyte chemoattractant protein-1: MCP-1; NrCAM: neuronal cell adhesion molecule; NFL: neurofilament light chain; OD: other dementia; PD: Parkinson’s disease; PDD: Parkinson’s disease with dementia; PSP: progressive supranuclear palsy; sAPPα, sAPPβ: soluble amyloid precursor protein α and β; SCI: subjective cognitive impairment; SCI-AD: SCI converters to AD; SCI-OD: SCI converters to OD; SCI-VaD: SCI converters to VaD; SD: standard deviation; UK: unknown; VaD: vascular dementia; yy: years *NINCDS-ADRDA criteria (National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association) **Petersen criteria at baseline diagnosis §NIA-AA guidelines (National Institute on Aging–Alzheimer’s Association) °IWG criteria (International Working Group)
CHC = 37
Controls CHC = 65
Olsson 2013 [67] (cross-sectional study)
First author (type of study) Olsson 2013 [49] (longitudinal study)
Table 1. (Continued).
EXPERT REVIEW OF PROTEOMICS 289
290
F. BALDACCI ET AL.
between MCI converters to AD and stable MCI subjects, although the performance of CSF YKL-40 and the combined ratio Aβ1-42/YKL-40 was relatively well (Table 1) [64]. Generally, available studies stated that MCI subjects with AD pathology showed higher CSF YKL-40 concentrations than those in cognitively normal controls. Gispert and colleagues detected higher concentrations of CSF YKL-40 in MCI subjects with AD pathology compared with cognitively normal subjects and even AD dementia, thus emphasizing a role of YKL-40 in the early disease [90]. Similarly, Alcolea and colleagues disclosed more elevated CSF YKL-40 in amnestic MCI (aMCI) than in controls [91]. In a different study, they reported a trend of higher concentrations of CSF YKL-40 in the Aβ+/tau+ aMCI subgroup than in the Aβ-/tau- aMCI subset, thus suggesting an association of the biomarker with AD pathophysiological processes [62]. Moreover, when considering the whole MCI group, Kester and colleagues found higher concentrations of CSF YKL-40 in relation to cognitively normal controls; CSF YKL40 was also higher in MCI converting to AD compared with stable MCI [65]. Interestingly, following a second lumbar puncture (mean time 2.0 years), CSF YKL-40 was reported to be increased in the MCI and AD groups but not in cognitively normal controls. A major limitation of this study is that MCI subjects were older than controls and included individuals with subjective cognitive complaints and epilepsy as well [65] (Table 1). Similar results were obtained in another longitudinal study in which MCI converters to AD with AD pathology presented higher concentrations of CSF YKL-40 than in stable MCI subjects, therefore confirming the key role of YKL40 in AD pathophysiology [67]. In addition, Olsson and colleagues correlated CSF and serum YKL-40 concentrations in a group of 10 patients that were randomly selected from the clinical routine workflow. Since no correlation was found between YKL-40 CSF and serum concentrations, CSF YKL-40 was assumed to be modulated within the CNS rather than be regulated by its concentration in plasma or by the permeability of the brain barrier. Actually, CSF/serum albumin ratio, the indicator of blood–brain barrier integrity, correlated with CSF YKL-40 in stable MCI subjects but not in MCI converters to AD, probably indicating a peripheral origin of YKL-40 in stable MCI. Albeit most of the studies reported the link of CSF YKL-40 to MCI with AD pathology, Mattson and colleagues did not report significant differences between MCI converters to AD dementia and cognitively healthy controls in terms of CSF YKL-40 concentrations. However, it should be considered the relatively small size of the sample and the purely clinical assessment of the conversion to AD that led to disregard the role of in vivo biological markers of AD pathology (Table 1). Notably, the groups did not show any substantial difference in terms of age; finally, cardiovascular comorbidity was considered as exclusion criterion for the control group [61]. Although the APOE-ε4 allele was not systematically examined in studies evaluating the concentrations of CSF YKL-40 in MCI subjects, three analyses reported a negative association [62,64,88]. In summary, the majority of available studies indicate that CSF concentrations of YKL-40 may serve as a predictive marker of progression to dementia in prodromal MCI
subjects and may help discriminate between MCI converters to AD dementia and stable MCI subjects.
1.4. YKL-40 and the preclinical, asymptomatic-at-risk stage of AD (or preclinical stage) Eight studies investigated the role of CSF YKL-40 in preclinical AD or asymptomatic-at-risk for AD subjects [62,83,87,90,92–94]. Antonel and colleagues reported a positive correlation of CSF YKL-40 with t-tau protein in 18 cognitively normal subjects in the preclinical stage of AD (stage 1 according to the NIA-AA guidelines) and a negative correlation of YKL-40 with neuropsychological scores [88]. Similarly, Alcolea and colleagues reported that 10 subjects at the advanced preclinical phase (stages 2–3) and 27 preclinical subjects with suspected non-Alzheimer’s pathophysiology (SNAP) showed higher concentrations of YKL-40 than those in both cognitively normal subjects and early preclinical phases (stage 0–1) [92]. Moreover, the authors reported a correlation of CSF YKL-40 with t-tau protein independently by the concentrations of CSF Aβ. This probably indicates a stronger association with core feasible AD biomarkers than with those of amyloid pathology. In addition, two groups reported a correlation with CSF tau concentrations in 192 subjects and in 20 preclinical AD subjects [90,93]. Interestingly, one study showed a positive correlation in the control group, thus highlighting a potential link between the two biomarkers even before the preclinical pathophysiology stage of AD [90]. However, in spite of the close association between the neurodegenerative process and CSF YKL-40, two independent investigations revealed divergent results. Racine and colleagues examined the link between CSF YKL-40 concentrations at baseline and the progression of Aβ deposition using 11C- Pittsburgh compound B (PiB)-PET in a cohort of 104 cognitively normal subjects in a 2-year longitudinal study and found that the YKL-40/Aβ42 ratio predicted baseline PiB burden. However, this significant result disappeared once CSF measures of amyloid pathology, neural injury, and inflammation were simultaneously included in a multivariate model as predictors [94]. In a further longitudinal study including non-demented middle-aged subjects undergoing repeated lumbar punctures (serial spinal taps at 3-year intervals, mean of 6 years), CSF YKL-40 concentrations correlated with age and were associated with the APOE-ε4 allele [87]; however, YKL-40 did not show properties of a disease progression marker compared with core, feasible biomarkers predictive of AD dementia. On the other hand, the authors highlighted, as a drawback, that the low number of subjects (n = 14) converted to dementia prevented a reliable and conclusive evaluation (Table 1) [87]. Overall, these outcomes emphasize the existence of a positive association between CSF YKL-40 and other CSF AD-related biomarkers, especially t-tau protein, in the preclinical, asymptomatic stage of AD. However, these data do not appear to be in any way definitive or conclusive and need to be confirmed by well-powered and standardized longitudinal analyses.
EXPERT REVIEW OF PROTEOMICS
1.5. YKL-40 in AD and other neurodegenerative diseases At present, the studies evaluating the diagnostic value of CSF YKL-40 in other categories of neurodegenerative diseases and dementia disorders are relatively limited [35,49,61– 66,68,69,71]. In an analysis by Alcolea and colleagues, CSF YKL-40 did not improve the diagnostic accuracy to distinguish AD patients from frontotemporal dementia (FTD). However, their FTD group was characterized by a high degree of heterogeneity [62] (Table 1). In an additional study, Craig-Shapiro and colleagues reported greater concentrations of the biomarker in FTD compared with AD patients in a very-mild/mild stage of cognitive impairment (CDR = 0.5–1) and more elevated concentrations in the AD group compared with those found in progressive supranuclear palsy (PSP) patients (Table 1). These data highlighted the potential usefulness of CSF YKL-40 in the differential diagnosis of AD from other forms of neurodegenerative dementia disorders [35]. However, as reported for Alcolea and colleagues [62], this analysis included a heterogeneous group of FTD patients as well. Recently, an unbiased high-resolution MS-based proteomics study [71] led to disclose a twofold increase of CSF YKL40 concentrations in FTD patients (n = 31) compared with 23 individuals with subjective cognitive impairment. In addition, a diagnostic pathological confirmation allowed all FTD patients to be divided into two subcategories, namely FTD TDP-43 positive (n = 21) and FTD tau positive (n = 10). Interestingly, the FTD tau positive subtype presented significantly higher concentrations of CSF YKL-40 compared with AD (n = 20), DLB (n = 20), and VaD (n = 18). A trend toward increased concentrations of the protein versus the FTD TDP-43 subtype was also observed. Notably, this analysis emphasized that the different results emerged from previous studies exploring CSF YKL-40 concentrations in FTD are potentially affected by the FTD subcategory underlying tau-driven pathophysiological neurodegenerative mechanism. The presence of elevated concentrations of CSF YKL-40 in FTD, similarly to AD patients, was also confirmed in a study by Janelidze and colleagues: when related to core, feasible biomarkers, both CSF YKL-40 and the Aβ1-42/YKL-40 ratio did not significantly improve the ability in discriminating AD dementia from other forms of dementia, although the AUROCs of CSF Aβ1-42/YKL-40 was fair (Table 1) [64]. Notably, the group including the other types of dementia was highly heterogeneous since it encompassed vascular dementia (VaD) and dementia with Lewy bodies (DLB) patients, besides FTD cases; CSF YKL-40 concentrations in VaD patients were comparable to those detected in AD and FTD; in contrast, DLB patients revealed significantly lower concentrations compared with those measured in AD [64]. The potential usefulness of YKL-40 to discriminate between AD and DLB seems to be further confirmed by Wennström and colleagues showing that CSF YKL-40 concentrations were higher in AD thus improving the diagnostic accuracy in distinguishing AD from DLB (Table 1) [66]. Interestingly, PD patients exhibited lower concentrations of CSF YKL-40 compared with AD, hence indicating a potential reduction of the biomarker in α-synucleinopathies [66]. Moreover, a cross-sectional study by Olsson and colleagues reported decreased concentrations of CSF YKL-40 in PD patients (n = 50) versus atypical
291
Parkinsonism cases that included multiple system atrophy (MSA) (n = 37), PSP (n = 32) and corticobasal syndrome (CBS) (n = 10) [67]. It should be noticed that, when patients were stratified and compared according to their probable pathological profile, CSF YKL-40 concentrations were more elevated in tauopathy-positive (PSP and CBS) patients than in those showing positivity to α-synucleinopathy (PD and MSA); these results supported the existence of a strict association between YKL-40 and tau protein-related pathophysiological mechanisms in non-AD neurodegenerative diseases (Table 1). Remarkably, decreased concentrations of CSF YKL-40 were documented in PD patients compared with healthy controls. However, the latter were significantly older than PD patients, thus likely affecting the results. Indeed, a longitudinal study [68] in which PD patients and healthy controls were comparable in terms of age, revealed a significant increase of CSF YKL-40 concentrations in PD patients, after a 2-year follow-up, compared with baseline but not in healthy controls, hence indicating a progressive neurodegenerative process in the PD group. Moreover, YKL-40 was associated to a faster cognitive decline in PD versus healthy controls and positively correlated with α-synuclein, t-tau, and hyperphosphorylated tau (p-tau) proteins (Table 1). The potential usefulness of YKL-40 in the diagnostic framework of Parkinsonian syndromes is presented in a longitudinal study where CSF YKL-40 concentrations were compared among PD patients (n = 31), atypical Parkinsonism cases (PSP n = 33, CBS n = 14, and MSA n = 16), AD (n = 26), FTD (n = 16), and controls (n = 30) [69]. In particular, Magdalinou and colleagues disclosed that CSF YKL-40 – utilized in a multiple variated analysis in combination with other 8 CSF biomarkers (Neurofilament light chain (NFL) protein, sAPPα, sAPPβ, α-synuclein, MCP-1, Aβ42, t-tau, p-tau) – discriminated all Parkinsonian syndromes from healthy controls, PD patients, and the atypical Parkinsonism cases (namely PSP, CBS, MSA) with excellent AUROCs (Table 1). CSF YKL-40 was also able to differentiate among the subgroups of atypical Parkinsonisms – PSP from MSA patients – with a good AUROC (Table 1). Notably, among the Parkinsonian syndromes, CSF YKL-40 concentrations were the highest in the PSP group, while PD group had the lowest, thus indicating an increase of YKL-40 in more aggressive neurodegenerative diseases with an underlying tau proteinopathy. The positive correlation of CSF YKL-40 with t-tau protein documented in DLB and PD patients is potentially indicative of an association between CSF YKL-40 and the pathophysiological processes of α-synucleinopathies. Interestingly, CSF YKL-40 was reported to correlate with CSF α-synuclein, a candidate diagnostic biomarker for PD and DLB [95]. In spite of few evidences, non-AD neurodegenerative diseases – especially FTD subtypes with an underlying tau positive proteinopathy and PSP – are assumed to be associated with increased concentrations of CSF YKL-40. Conversely, Mattson and colleagues did not find significant differences between AD and a group including ‘other dementias’. However, the absence of FTD patients in the ‘other dementias’ group (exclusively consisting of DLB and VaD patients) might have affected the final results (Table 1) [61]. In a another study, Hellwig and colleagues did not find any substantial difference in terms of CSF YKL-40
292
F. BALDACCI ET AL.
concentrations between AD and the ‘other dementias’ group (including FTD cases with different heterogeneous phenotypes that might affect CSF YKL-40 concentrations) (Table 1) [63]. Finally, a non-demented group of 12 patients suffering from idiopathic REM sleep behavior disorder – often associated with neurodegenerative diseases such as α-synucleinopathies [88] – did not show any substantial increase of CSF YKL-40 concentrations versus controls. The absence of significant results might be due to various reasons: the small sample size, the inclusion of individuals with subjective memory impairment in the control group (a condition recognized to be at risk of AD [88]), and the fact that patients with REM sleep behavior disorders are likely to be prodromal of positivity to α-synucleinopathy rather than to tauopathies. In two longitudinal studies, CSF YKL-40 predicted the conversion to other forms of dementia in MCI subjects. In particular, Kester and colleagues reported increased baseline concentrations of CSF YKL-40 in MCI subjects predicting not only AD but also other dementias (mean follow-up of 2.0 years) [65] (Table 1). Olsson and colleagues showed that CSF YKL-40 allowed to distinguish (although poorly) between MCI subjects who converted to VaD and stable MCI, at baseline [49] (Table 1). Only two studies evaluated the CSF concentrations of YKL-40 in HD [72]. HD represents the paradigm of a clinically heterogeneous group of several genetic rare neurodegenerative diseases caused by trinucleotide expansions, CAG repeats, of the huntingtin gene inducing cellular toxic misfolded protein depositions within neurons finally resulting in neurodegeneration and neuroinflammatory (astrogliosis and microgliosis) processes [32]. In a crosssectional study [72], the CSF YKL-40 concentrations were compared among three groups, 27 HD cases, 27 asymptomatic HD gene-expansion carriers, and 14 healthy controls. The results revealed that the symptomatic HD patients had a non-significant trend toward increased CSF YKL-40 concentrations when compared with HD gene-expansion carriers and healthy controls. Notably, CSF YKL-40 concentrations positively correlated with the length of CAG repeats. A further study [86] showed significant higher concentrations of CSF YKL-40 in mutation carriers, and a positive correlation with disease stage and severity. Overall, these studies suggested a potential role of CSF YKL-40 as a marker of glia activation and neuroinflammation in HD (Table 1). However, longitudinal studies investigating the alterations of CSF YKL-40 concentrations in asymptomatic HD gene-expansion carriers are needed to define the effective role of YKL-40 as a potential pathophysiological biomarker of HD. In view of all the data collected from the existing literature, it remains unclear whether YKL-40 exhibits a sufficiently robust performance to allow stratifying different categories of primary progressive neurodegenerative diseases and dementia disorders. Based on the evidence described so far, CSF YKL-40 appears a promising biomarker candidate in differentiating patients with AD dementia and α-synucleinopathies, although further studies are eagerly required.
1.6. YKL-40 and neuroimaging markers The association between CSF concentrations of YKL-40 and in vivo neuroimaging markers of AD has been little investigated so far [90,91,93]. The study by Alcolea and colleagues conducted on cognitively healthy controls, preclinical AD, and aMCI individuals reported a negative correlation between CSF YKL-40 concentrations and brain cortical thickness, especially in the middle and inferior temporal areas [91]. In addition, when the presence of in vivo Aβ pathology was considered, the negative correlation between YKL-40 and cortical thickness remained significant only in those individuals showing in vivo Aβ pathology both in the full sample and in the subsample of preclinical AD subjects. These results suggest that high CSF concentrations of YKL-40 might be related to Aβ pathology and mechanisms leading to cortical thinning. This correlation, however, disappeared after correcting for p-tau, revealing that both YKL-40 and p-tau were associated with cortical thinning in the same cortical areas. The major limitation of the above study is due to the younger age of cognitively healthy controls compared to the preclinical AD and the aMCI subjects, preventing the generalization of the results [91]. Further studies investigated the association between brain atrophy and CSF YKL-40 concentrations using different techniques such as the voxel-based morphometry method [90]. Results showed an inverse U-shaped curve between CSF YKL-40 concentrations and the typical brain pattern of AD atrophy in 28 MCI subjects due to AD and 15 AD patients. In particular, higher CSF YKL-40 concentrations were related to gray matter reduction in the right inferior temporal cortex, in the right angular, and in the supramarginal gyri in both individuals with MCI due to AD and AD dementia. Additionally, a significant association was documented in the right insula and bilaterally in the cerebellum. On the contrary, in the control and preclinical AD subjects, no statistically significant association between CSF YKL-40 concentrations and gray matter volume was found. As in the previous study, the group of cognitively healthy controls significantly differed from patients regarding age [90]. Finally, the interaction between CSF YKL-40 concentrations and Diffusion Tensor Imaging (DTI) indexes of white matter microstructure was recently examined. This investigation, conducted in asymptomatic-at-risk individuals for AD subjects, did not show any direct associations of CSF concentrations of YKL-40 with indexes of white matter microstructure such as fractional anisotropy and mean diffusivity. However, in the same study, a correlation with markers of both thinner and larger axonal degeneration, as indicated by CSF tau protein and neurofilament light chain protein, suggested that in asymptomatic-atrisk individuals for AD, CSF YKL-40 alterations might precede in vivo white matter microstructural changes detected using DTI, although further studies are needed to confirm this hypothesis [93]. In summary, due to the relatively low number of studies and the heterogeneity of the population examined, further studies investigating the association between CSF inflammatory markers and in vivo brain functional/structural alterations are needed. In particular, longitudinal studies conducted in
EXPERT REVIEW OF PROTEOMICS
preclinical and prodromal stages of AD will be essential to track within person trajectories describing the relation between CSF YKL-40 concentrations in association with brain morphology and its white matter tracts. Moreover, cross-sectional and longitudinal studies comparing concentrations of CSF neuroinflammatory biomarkers such as YKL-40 with 11CPK11195-PET are absolutely crucial to shed light the role of neuroinflammation in the onset and progression of AD as well as to study the relation of neuroinflammatory markers to neurodegenerative processes of AD.
1.7. YKL-40 and novel non-core feasible CSF biomarkers Although the limited availability of data, CSF YKL-40 concentrations have been associated with the concentrations of other novel non-core candidate biomarkers of AD. Notably, CSF YKL-40 concentrations are positively related to those of NFL – a marker of damage of large-caliber myelinated axon – in both preclinical AD [93] and PD patients [68]. Furthermore, YKL-40 has a positive association with CSF αsynuclein – a promising biomarker of α-synucleinopathies – in AD, PD, DLB, and other atypical Parkinsonisms [66,68,69]. In support of these observations, growing evidence indicates that α-synucleinopathy and amyloid pathology are intertwined [6,95–97]. Rosén and colleagues identified a positive correlation between CSF YKL-40 and CSF monocyte chemoattractant protein-1 (MCP-1) both in AD patients and in healthy control individuals [84]. Because inflammatory biomarkers are not specific, it is not surprising that the association of CSF YKL-40 with CSF MCP-1 is not limited to AD but can also be found in MCI subjects who converted to non-AD dementing disorders [61], possibly because of common pathophysiological pathways shared by apparently unrelated neurodegenerative diseases [98]. Nonetheless, other authors failed to identify an association of YKL-40 with CSF MCP-1 in AD patients [66]. Interestingly, YKL-40 has a moderately positive correlation with t-tau and p-tau but not with Aβ markers − further confirming its lack of AD-specificity [62]. The concentrations of the astrocyte biomarker glial-fibrillary acidic protein (GFAP) do not correlate with YKL-40, suggesting the non-astrocytic origin of YKL-40 [62]. In this regard, growing evidence indicates that YKL-40 is mainly released from microglial cells [49]. Further studies are needed to investigate the diagnostic and prognostic significance of YKL-40 – either alone or in combination with other neuroinflammatory markers – in different neurodegenerative diseases. Intriguingly, a strong positive association of YKL-40 with CSF superoxide dismutase (SOD1) has been reported, suggesting a potential relationship between neuroinflammation and oxidative stress [70]. In contrast, YKL-40 does not seem to be associated with neurogranin [63] – an emerging biomarker of synaptic disruption in AD [99] – and showed a weak correlation with progranulin in a sample of 229 patients with neurodegenerative diseases including AD, DLB, FTD, PSP, and CBS [100]. In summary, the available evidence indicates that CSF YKL-40 concentrations are mainly related to concentrations of CSF NFL, α-synuclein, and tau-related markers. Because it is unlikely that a single neuroinflammatory marker can reflect the complex
293
pathophysiological processes that underlie the development, clinical onset, and progression of AD, it is hoped that future studies will focus on a combination of different analytes that could reflect different pathophysiological mechanisms.
2. Conclusions There is growing evidence suggesting that CSF YKL-40 might be of diagnostic value in distinguishing AD from healthy controls and it may contribute to predict the progression of cognitive decline from MCI to AD. It should be observed that control subjects were not exclusively represented by ‘true’ cognitively healthy controls in all studies; actually, they also included individuals with subjective cognitive impairment or patients with neurological diseases other than neurodegenerative syndromes, thus introducing a possible bias (Table 1). A recent meta-analysis, covering studies published until July 2014, compared CSF YKL-40 in six different cohorts of AD patients and five cohorts of ‘real’ cognitively normal controls demonstrating a moderate significant effect size for this neuroinflammatory biomarker. Furthermore, the authors reported a possible diagnostic value of plasma YKL-40 concentrations in spite of the limited currently available data [82]. There is accumulating evidence indicating that CSF YKL-40, in combination with other biomarkers, might differentiate Parkinsonian syndromes from healthy controls and PD patients from atypical Parkinsonisms; moreover, it could predict cognitive decline in PD. In contrast, data are contradictory for FTD. This would be clinically defined as a complex disease spectrum overlapping with other neurodegenerative diseases. FTD is typically characterized by different phenotypes showing progressive behavioral, executive, and language impairment, and, in addition, might develop typical symptoms of motor neuron diseases and akinetic extrapyramidal syndromes. However, FTD clinical phenotypes with an underlying tau pathology are likely to be associated with increased CSF YKL40 concentrations. Finally, since CSF YKL-40 is strictly related to unspecific tau protein neurodegeneration processes, it could be useful to differentiate tauopathies from other neurodegenerative diseases. All the studies summarized here highlighted the positive correlation of YKL-40 with age; this result is not surprising given the overexpression of microglia as documented by both post-mortem and PET studies (using the 11C-PK11195 ligand) in brains of healthy aged individuals [101]. Hence, aging should be included as a potential confounding factor in all studies evaluating the association between CSF YKL-40 and cognitive impairment. However, several studies included control groups consistently younger than study patients with clinical disease, thus introducing a potentially significant bias (Table 1). Additionally, as increased CSF YKL-40 concentrations have been reported in subjects with traumatic brain injury [102], the plausible pathophysiological association between CSF YKL-40 and anamnestic data of head trauma in AD should be considered. Actually, assessing whether CSF YKL-40 concentrations are more elevated in AD patients with head trauma history than those without brain trauma would be helpful to disclose different
294
F. BALDACCI ET AL.
Figure 1. Factors contributing to increase YKL-40 concentrations in the CSF. Pathophysiological mechanisms, such as neurodegeneration (t-tau and p-tau), endothelial dysfunction (vascular risk factor and diabetes), and probably BBB disruption determine increments of YKL-40 concentrations in the CSF. Aging, but not sex, is associated with CSF YKL-40 concentrations. Potentially, the increase of CSF YKL-40 is not directly associated with Aβ deposition in the brain. Presently, limited data exist on the contribution of other non-core feasible biomarkers, an antecedent history of traumatic brain injury, and a systemic inflammatory disease comorbidity. Abbreviations: Aβ: amyloid-β; BBB: blood brain barrier; CSF: cerebrospinal fluid; p-tau: hyperphosphorylated tau; t-tau: total tau.
pathophysiological pathways leading to AD. Moreover, given the well-known impact of cardiovascular risk factors and diabetes (frequently documented in AD patients and elderly subjects) on CSF YKL-40 concentrations (Figure 1), an accurate comorbidity and pharmacological assessment is needed. In contrast to aging, CSF YKL-40 concentrations seem not to be apparently affected by sex in AD although this topic necessitates further assessment (Figure 1). Finally, a potential association of YKL-40 with the APOE-ε4 risk factor is not fully elucidated and reported data are conflicting; actually, despite the existence of several genetic variants of AD risk associated with neuroinflammatory processes, no additional genetic associations involving YKL-40 have been demonstrated so far [17]. Certainly, YKL-40 is hypothesized to be an early pathophysiological marker of neuroinflammatory mechanisms – and of course not a specific pathognomonic indicator of AD or other neurodegenerative diseases. Indeed, CSF YKL-40 predicts the conversion of individuals from MCI not only to AD but also to other causes of dementia [49,65]. On the other hand, the currently developed clinical diagnostic criteria for neurodegenerative diseases still remain largely ‘flexible’ in terms of phenotypical characterization as well as diagnostic and classificatory accuracy as a result of the frequent pathological overlapping, as demonstrated by several post-mortem studies [6,96].
3. Expert commentary Currently, neuroinflammation is assumed to be an early pathophysiological mechanism in the evolving natural history of AD and other dementias. Epidemiological data indicated that nonsteroidal anti-inflammatory drugs (NSAIDs) might reduce the risk of AD [103–105] since patients suffering from rheumatic diseases that were treated with NSAIDs for a long time showed a decreased risk of developing AD. However, several trials reported negative results after anti-inflammatory
treatment in AD patients [8]. The critical issue may be that anti-inflammatory therapies should be administered at earlier stages (the asymptomatic preclinical phase) of the disease and be specifically geared toward relevant neuroinflammatory mechanisms occurring at certain time points during the progression of a specific neurodegenerative disease. In this regard, a relatively recent follow-up trial of naproxen initially documented negative results in AD therapy; however, longerterm follow-up results revealed a potential protective role of naproxen in asymptomatic subjects at baseline [106,107]. In addition, epidemiological studies reported a decreased incidence of PD attributable to the chronic consumption of NSAIDs [30], although no anti-inflammatory trials have been executed to explore the possible preventing/slowing effect on PD progress. Hence, since the increased concentration of CSF YKL-40 is a marker of inflammation within the CNS, and it is likely altered in the very early phase of AD (and probably PD), it could be employed as predictive biomarker of positive clinical response in anti-inflammatory trials. The increased concentration of CSF YKL-40 is not an inflammatory marker specifically associated with AD pathophysiology or other neurodegenerative diseases since its plasma and/ or CSF elevated concentrations were also reported in stroke [50–53], atrial fibrillation [108,109], diabetes [110], hypertension [111], and were linked to indicators of vascular risk factors such as carotid arteries stiffness [38] and endothelial dysfunction [112]. Although the origin of YKL-40 in the CSF from the periphery is not fully assessed, it is presumed to be modified by various comorbidities, CSF flow and blood CSF and brain barrier permeability (Figure 1). Interestingly, Olsson and colleagues reported that CSF YKL-40 concentrations were not affected by plasma concentrations and blood–brain barrier integrity in a small subset of patients; however, the subset size was small (n = 10) [49]. On the other hand, neuroinflammation, neurodegenerative processes, and vascular oxidative stress associated with cardiovascular risks factors are all
EXPERT REVIEW OF PROTEOMICS
assumed to contribute to the pathogenesis of AD and other neurodegenerative diseases. As a result, all these pathophysiological mechanisms contribute with a high likelihood to both genesis and progression of AD [113–116], PD [117], and, probably, other idiopathic neurodegenerative diseases [97,98]. Indeed, it is acknowledged that obesity, diabetes, and cholesterol metabolism alterations are strictly linked to AD [118,119] and PD [30,120,121]. Furthermore, cerebrovascular diseases, diabetes mellitus, as well as midlife hypertension and obesity are largely considered established risk factors for AD and dementia [118]. The existing evidence supports the role of CSF YKL-40 as a promising candidate biomarker for AD that might be employed in clinical trials to track the dynamic of glial neuroinflammatory processes in relation to ongoing neurodegeneration. Both cross-sectional and longitudinal studies – including cardiovascular comorbidities as possible confounding factors and recruiting cognitively healthy controls comparable to patients in terms of age as well as homogenous patients with other dementias – are needed to elucidate the potential usefulness of YKL-40 in the whole spectrum of neurodegenerative diseases including AD. CSF YKL-40 is likely not to be precisely driven by a specific pathophysiological mechanism. Indeed, aging, cardiovascular risk factors, systemic inflammatory disease comorbidity, diabetes, and potential pharmacological treatments are confounding factors that would need to be integrated in designing studies detecting CSF YKL-40 in patients with AD and other neurodegenerative diseases. Certainly, CSF YKL-40 is highly correlated with the increase of CSF tau protein concentrations, which is a common feature of AD. Notably, this rise in concentration has been observed in asymptomatic and probably preclinical AD subjects as well, thus suggesting a possible early association between YKL-40 and tau protein with the progression of neurodegenerative processes. However, whether the increased concentrations of CSF YKL-40 are an upstream or downstream effect of CSF tau alterations is not yet clarified.
4. Five-year view Considerable progress has been achieved over the past decades in elucidating AD pathophysiology; however, the early detection and diagnosis of AD and other age-related neurodegenerative diseases still remains a clinical challenge. Therefore, the validation and qualification of biological markers, both diagnostic and prognostic, characterized by a clear performance profile in terms of specificity, sensitivity, as well as positive/negative predictive value is required. Further systematic diagnostic validation studies on candidate biomarkers are needed and beyond that complementary exploratory research, embracing the systems biology paradigm – integrating experimental biology with accurate computational modeling to predict and illustrate the dynamic features of biological systems [122,123] Systems biology is based on the use of high-throughput screening approaches, the ‘omic’ sciences. Such evolving ‘omic’ technological platforms will allow researchers to speed up the biomarker
295
discovery and validation workflow needed to substantially enhance the diagnosis of AD. Since the ‘omic’ sciences enable to comprehensively describe the human body at a genetic and biomolecular level, they represent the leading disciplines required for detecting pathophysiological biomarker candidates for neurodegenerative diseases and their genetic and biological subsets – such as the neuroinflammatory marker YKL-40 – assumed to play a supportive role for early detection, diagnosis, prognosis, and therapy development in AD [122,123]. YKL-40 and other innovative biological candidate markers of growing interest – such as neurogranin, a marker of early synaptic dysfunction and loss, or NFL, a marker of disintegration of large-caliber myelinated axons – are supposed to enable early patient stratification and selection for mechanistic biomarker guided targeted therapies [124]. In the upcoming years, this evolving array of pathophysiological biomarkers is supposed to constitute an enhanced basis to accelerate the development of effective customized therapeutic approaches – namely, ‘molecularly’ targeted therapies – for the precise treatment of molecular pathophysiological pathways associated with AD. General and specialized clinicians of the future will have, therefore, the practical opportunity to deliver targeted and tailored interventions, i.e. adapted to the specific biological profiles of AD patients, according to the evolving precision medicine paradigm [124]. The precision medicine-based strategy of ‘customized,’ mechanism-based care – formerly applied in oncology within the therapeutic approach for treating patients with specific cancer profiles – is now beginning to face the challenge of the clinical and biological complexity and multi-factorial heterogeneity of AD and other neurodegenerative diseases. Toward this objective, precision medicine is grounded on a fundamentally different approach than the traditional drug development practice historically founded on providing a ‘one-size-fits-all’ therapy, very rarely adequate and successful enough for reaching a significant treatment effect across all biologically heterogeneous groups of clinical phenotypes [97]. Precision medicine applied to the field of AD and neurodegeneration aims at targeting the genetic risk and the molecular stages of disease, particularly at the earliest preclinical asymptomatic stage [125], when the disease is still potentially reversible, tailored to delay, stop, and, if possible, prevent the progression to clinical symptoms. After decades of clinical trial failures, the recognition of the precision medicine paradigm for AD research and drug development is gaining momentum; thus, the time seemed to be mature to initiate the Alzheimer’s Precision Medicine Initiative (APMI), an international collaboration of leading interdisciplinary clinicians and scientists devoted toward the implementation of precision medicine in Neurology, Psychiatry, and Neuroscience. The realization of precision medicine in AD and other neurodegenerative diseases will result in better treatment responses and optimized safety profiles through biomarker-guided early preclinical disease stage clinical trials [124,126]. YKL-40 substantially adds to the growing array of biomarkers reflecting distinct molecular brain mechanisms prospectively valuable for differentiated stratification of individuals for
296
F. BALDACCI ET AL.
biomarker guided targeted therapies emerging from the precision medicine paradigm.
Key issues ● CSF YKL-40 − a glycoprotein expressed in activated microglia − is a marker of neuroinflammation, a key pathophysiological mechanism in AD and neurodegenerative diseases. ● In the last six years, the CSF concentrations of YKL-40 have been investigated in subjects within the AD spectrum, and, more recently, in subjects with other dementia disorders as well as in non-AD neurodegenerative diseases. ● CSF YKL-40 concentrations may discriminate AD dementia in comparison to cognitively healthy age-matched controls, may predict the progression of cognitive decline from MCI to AD dementia and may help to distinguish MCI converters to AD dementia from stable MCI subjects. ● There is a positive association between CSF YKL-40 and CSF NFL, α-synuclein, and tau-related markers in AD and nonAD neurodegenerative diseases. ● CSF YKL-40 might be a promising biomarker in differentiating neurodegenerative diseases associated with protein tau protein related pathophysiological processes from neurodegenerative diseases related to α-synuclein mechanisms. ● Aging, cardiovascular risk factors, systemic inflammatory disease comorbidity, diabetes, and potential pharmacological treatments are possible confounding factors in studies evaluating CSF YKL-40 concentrations in patients with AD and other neurodegenerative diseases. ● YKL-40 adds to the growing array of biomarkers reflecting distinct molecular brain mechanisms (neuroinflammation) prospectively valuable for differentiated stratification of individuals for biomarker guided targeted (anti-inflammatory) therapies emerging from the precision medicine paradigm.
Funding This work was supported by the AXA Research Fund, the “Fondation Université Pierre et Marie Curie” and the “Fondation pour La Recherche sur Alzheimer”, Paris, France. Ce travail a bénéficié d’une aide de l’Etat “Investissements d’avenir” ANR-10-IAIHU-06 (Harald Hampel). The research leading to these results has received funding from the program “Investissements d’avenir” ANR-10-IAIHU-06 (Agence Nationale de la Recherche-10-IA Agence Institut Hospitalo-Universitaire-6) (Harald Hampel).
Declaration of interest S Lista has received lecture honoraria from Roche. U Bonuccelli has received fees for consultation from GSK and Eisai and for speeches from Novartis, GSK, and Lundbeck. H Hampel declares no competing financial interests related to the present article. He serves as Senior Associate Editor for the journal Alzheimer’s & Dementia ; he has been a scientific consultant and/or speaker and/or attended scientific advisory boards of Axovant, Anavex, Eli Lilly and company, GE Healthcare, Cytox, Jung Diagnostics, Roche, Biogen Idec, Takeda-Zinfandel, Oryzon Genomics, Qynapse; and receives research support from the Association for Alzheimer Research (Paris), Pierre and Marie Curie University (Paris), Pfizer & Avid (paid to institution); and has patents as co-inventor, but
®
received no royalties: A patent in vitro multiparameter determination method for the Diagnosis and early diagnosis of neurodegenerative disorders. Patent number: 8916388 Issued. A patent in vitro procedure for diagnosis and early diagnosis of neurodegenerative diseases. Patent number: 8298784 Issued. A patent Neurodegenerative Markers for Psychiatric Conditions. Publication number: 20120196300 Issued. A patent IN VITRO MULTIPARAMETER DETERMINATION METHOD FOR THE DIAGNOSIS AND EARLY DIAGNOSIS OF NEURODEGENERATIVE DISORDERS. Publication number: 20100062463 Issued. A patent IN VITRO METHOD FOR THE DIAGNOSIS AND EARLY DIAGNOSIS OF NEURODEGENERATIVE DISORDERS. Publication number: 20100035286 Issued. A patent In vitro Procedure for Diagnosis and Early Diagnosis of Neurodegenerative Diseases. Publication number: 20090263822 Issued. A patent in vitro method for the diagnosis of neurodegenerative diseases. Patent number: 7547553 Issued. A patent CSF Diagnostic in Vitro Method for Diagnosis of Dementias and Neuroinflammatory Diseases. Publication number: 20080206797 Issued. A patent in vitro Method For the Diagnosis of Neurodegenerative Diseases. Publication number: 20080199966 Issued. A patent Neurodegenerative Markers for Psychiatric Conditions. Publication number: 20080131921 Issued. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
References Papers of special note have been highlighted as either of interest (•) or of considerable interest (••) to readers. 1. Ransohoff RM. How neuroinflammation contributes to neurodegeneration. Science. 2016;353(6301):777–783. 2. Andreasson KI, Bachstetter AD, Colonna M, et al. Targeting innate immunity for neurodegenerative disorders of the central nervous system. J Neurochem. 2016;138(5):653–693. 3. Scheltens P, Blennow K, Breteler MMB, et al. Alzheimer’s disease. Lancet. 2016;388:505–517. 4. Hardy J. The amyloid hypothesis for Alzheimer’s disease: a critical reappraisal. J Neurochem. 2009;110(4):1129–1134. 5. Hashimoto M, Rockenstein E, Crews L, et al. Role of protein aggregation in mitochondrial dysfunction and neurodegeneration in Alzheimer’s and Parkinson’s diseases. Neuromolecular Med. 2003;4(1–2):21–36. 6. Hyman BT, Phelps CH, Beach TG, et al. National Institute on AgingAlzheimer’s Association guidelines for the neuropathologic assessment of Alzheimer’s disease. Alzheimers Dement. 2012;8(1):1–13. 7. Johnson LV, Leitner WP, Rivest AJ, et al. The Alzheimer’s A beta -peptide is deposited at sites of complement activation in pathologic deposits associated with aging and age-related macular degeneration. Proc Natl Acad Sci U S A. 2002;99(18):11830–11835. 8. Heneka MT, Carson MJ, El Khoury J, et al. Neuroinflammation in Alzheimer’s disease. Lancet Neurol. 2015;14(4):388–405. •• This review evaluates in depth the interactions between the immunological mechanisms and the pathophysiology of AD highlighting their genetic and environmental interrelations. 9. Heppner FL, Ransohoff RM, Becher B. Immune attack: the role of inflammation in Alzheimer disease. Nat Rev Neurosci. 2015;16 (6):358–372. 10. Heneka MT, Kummer MP, Latz E. Innate immune activation in neurodegenerative disease. Nat Rev Immunol. 2014;14(7):463–477. 11. Glass CK, Saijo K, Winner B, et al. Mechanisms underlying inflammation in neurodegeneration. Cell. 2010;140(6):918–934. • This review provides a comprehensive description of the contribution of the immunological system that is supposed to trigger the pathophysiological mechanisms related to neurodegenerative diseases. 12. Calsolaro V, Edison P. Neuroinflammation in Alzheimer’s disease: current evidence and future directions. Alzheimers Dement. 2016;12(6):719–732.
EXPERT REVIEW OF PROTEOMICS
13. Minter MR, Taylor JM, Crack PJ. The contribution of neuroinflammation to amyloid toxicity in Alzheimer’s disease. J Neurochem. 2016;136(3):457–474. 14. Wyss-Coray T, Rogers J. Inflammation in Alzheimer disease-a brief review of the basic science and clinical literature. Cold Spring Harb Perspect Med. 2012;2(1):a006346. 15. Mosher KI, Wyss-Coray T. Microglial dysfunction in brain aging and Alzheimer’s disease. Biochem Pharmacol. 2014;88(4):594–604. 16. Bettens K, Sleegers K, van Broeckhoven C. Genetic insights in Alzheimer’s disease. Lancet Neurol. 2013;12(1):92–104. 17. van Cauwenberghe C, van Broeckhoven C, Sleegers K. The genetic landscape of Alzheimer disease: clinical implications and perspectives. Genet Med. 2016;18(5):421–430. • This is a complete and updated review focused on the genetic risk related to AD examining the potential contribution of genes involved in the regulation of the inflammation and in its pathogenesis. 18. Perry VH, Teeling J. Microglia and macrophages of the central nervous system: the contribution of microglia priming and systemic inflammation to chronic neurodegeneration. Semin Immunopathol. 2013;35(5):601–612. 19. Li C, Zhao R, Gao K, et al. Astrocytes: implications for neuroinflammatory pathogenesis of Alzheimer’s disease. Curr Alzheimer Res. 2011;8(1):67–80. 20. Veerhuis R, Nielsen HM, Tenner AJ. Complement in the brain. Mol Immunol. 2011;48(14):1592–1603. 21. Ishizuka K, Kimura T, Igata-Yi R, et al. Identification of monocyte chemoattractant protein-1 in senile plaques and reactive microglia of Alzheimer’s disease. Psychiatry Clin Neurosci. 1997;51(3):135–138. 22. Tarkowski E, Andreasen N, Tarkowski A, et al. Intrathecal inflammation precedes development of Alzheimer’s disease. J Neurol Neurosurg Psychiatry. 2003;74(9):1200–1205. 23. Burguillos MA, Deierborg T, Kavanagh E, et al. Caspase signalling controls microglia activation and neurotoxicity. Nature. 2011;472 (7343):319–324. 24. Vodovotz Y, Lucia MS, Flanders KC, et al. Inducible nitric oxide synthase in tangle-bearing neurons of patients with Alzheimer’s disease. J Exp Med. 1996;184(4):1425–1433. 25. Okello A, Edison P, Archer HA, et al. Microglial activation and amyloid deposition in mild cognitive impairment: a PET study. Neurology. 2009;72(1):56–62. 26. Wiley CA, Lopresti BJ, Venneti S, et al. Carbon 11-labeled Pittsburgh compound B and carbon 11-labeled (R)-PK11195 positron emission tomographic imaging in Alzheimer disease. Arch Neurol. 2009;66 (1):60–67. 27. Varley J, Brooks DJ, Edison P. Imaging neuroinflammation in Alzheimer’s disease and other dementias: recent advances and future directions. Alzheimers Dement. 2015;11(9):1110–1120. 28. Edison P, Archer HA, Gerhard A, et al. Microglia, amyloid, and cognition in Alzheimer’s disease: an [11C](R)PK11195-PET and [11C]PIB-PET study. Neurobiol Dis. 2008;32(3):412–419. 29. Rayaprolu S, Mullen B, Baker M, et al. TREM2 in neurodegeneration: evidence for association of the p.R47H variant with frontotemporal dementia and Parkinson’s disease. Mol Neurodegener. 2013;8:19. 30. Noyce AJ, Bestwick JP, Silveira-Moriyama L, et al. Meta-analysis of early nonmotor features and risk factors for Parkinson disease. Ann Neurol. 2012;72(6):893–901. 31. Gerhard A. TSPO imaging in parkinsonian disorders. Clin Transl Imaging. 2016;4:183–190. 32. Crotti A, Glass CK. The choreography of neuroinflammation in Huntington’s disease. Trends Immunol. 2015;36(6):364–373. 33. Corcia P, Tauber C, Vercoullie J, et al. Molecular imaging of microglial activation in amyotrophic lateral sclerosis. PLoS One. 2012;7 (12):e52941. 34. Zhang J. Mapping neuroinflammation in frontotemporal dementia with molecular PET imaging. J Neuroinflammation. 2015;12:108. 35. Craig-Schapiro R, Perrin RJ, Roe CM, et al. YKL-40: a novel prognostic fluid biomarker for preclinical Alzheimer’s disease. Biol Psychiatry. 2010;68(10):903–912.
297
•• In this original article, YKL-40 is identified – for the first time – as a potential candidate cerebrospinal fluid biomarker for AD by using two-dimensional difference gel electrophoresis and liquid chromatography tandem mass spectrometry. 36. Rehli M, Krause SW, Andreesen R. Molecular characterization of the gene for human cartilage gp-39 (CHI3L1), a member of the chitinase protein family and marker for late stages of macrophage differentiation. Genomics. 1997;43(2):221–225. 37. Hauschka PV, Mann KG, Price P, et al. Report of the Ad Hoc committee on nomenclature and standards for bone proteins and growth factors. J Bone Miner Res. 1986;1(5):485–486. 38. Rathcke CN, Vestergaard H. YKL-40–an emerging biomarker in cardiovascular disease and diabetes. Cardiovasc Diabetol. 2009;8:61. 39. Johansen JS, Jensen BV, Roslind A, et al. Serum YKL-40, a new prognostic biomarker in cancer patients? Cancer Epidemiol Biomark Prev. 2006;15(2):194–202. 40. Henrissat B, Bairoch A. New families in the classification of glycosyl hydrolases based on amino acid sequence similarities. Biochem J. 1993;293(3):781–788. 41. Prakash M, Bodas M, Prakash D, et al. Diverse pathological implications of YKL-40: answers may lie in “outside-in” signaling. Cell Signal. 2013;25(7):1567–1573. 42. Lee CG, Da Silva CA, Dela Cruz CS, et al. Role of chitin and chitinase/chitinase-like proteins in inflammation, tissue remodeling, and injury. Annu Rev Physiol. 2011;73:479–501. 43. Malinda KM, Ponce L, Kleinman HK, et al. Gp38k, a protein synthesized by vascular smooth muscle cells, stimulates directional migration of human umbilical vein endothelial cells. Exp Cell Res. 1999;250(1):168–173. 44. Fusetti F, Pijning T, Kalk KH, et al. Crystal structure and carbohydrate-binding properties of the human cartilage glycoprotein-39. J Biol Chem. 2003;278(39):37753–37760. 45. Rehli M, Niller -H-H, Ammon C, et al. Transcriptional regulation of CHI3L1, a marker gene for late stages of macrophage differentiation. J Biol Chem. 2003;278(45):44058–44067. 46. Johansen JS, Jensen BV, Roslind A, et al. Is YKL-40 a new therapeutic target in cancer? Expert Opin Ther Targets. 2007;11 (2):219–234. 47. Bonneh-Barkay D, Wang G, Starkey A, et al. In vivo CHI3L1 (YKL-40) expression in astrocytes in acute and chronic neurological diseases. J Neuroinflammation. 2010;7:34. 48. Bonneh-Barkay D, Bissel SJ, Kofler J, et al. Astrocyte and macrophage regulation of YKL-40 expression and cellular response in neuroinflammation. Brain Pathol. 2012;22(4):530–546. 49. Olsson B, Hertze J, Lautner R, et al. Microglial markers are elevated in the prodromal phase of Alzheimer’s disease and vascular dementia. J Alzheimers Dis. 2013;33(1):45–53. 50. Hjalmarsson C, Bjerke M, Andersson B, et al. Neuronal and gliarelated biomarkers in cerebrospinal fluid of patients with acute ischemic stroke. J Cent Nerv Syst Dis. 2014;6:51–58. 51. Park HY, Jun C-D, Jeon S-J, et al. Serum YKL-40 levels correlate with infarct volume, stroke severity, and functional outcome in acute ischemic stroke patients. PLoS One. 2012;7(12):e51722. 52. Rathcke CN, Thomsen SB, Linneberg A, et al. Variations of CHI3L1, levels of the encoded glycoprotein YKL-40 and prediction of fatal and non-fatal ischemic stroke. PLoS One. 2012;7(8):e43498. 53. Kjaergaard AD, Johansen JS, Bojesen SE, et al. Elevated plasma YKL40, lipids and lipoproteins, and ischemic vascular disease in the general population. Stroke J Cereb Circ. 2015;46(2):329–335. 54. Malmeström C, Axelsson M, Lycke J, et al. CSF levels of YKL-40 are increased in MS and replaces with immunosuppressive treatment. J Neuroimmunol. 2014;269(1–2):87–89. 55. Martínez MAM, Olsson B, Bau L, et al. Glial and neuronal markers in cerebrospinal fluid predict progression in multiple sclerosis. Mult Scler. 2015;21(5):550–561. 56. Burman J, Raininko R, Blennow K, et al. YKL-40 is a CSF biomarker of intrathecal inflammation in secondary progressive multiple sclerosis. J Neuroimmunol. 2016;292:52–57.
298
F. BALDACCI ET AL.
57. Hinsinger G, Galéotti N, Nabholz N, et al. Chitinase 3-like proteins as diagnostic and prognostic biomarkers of multiple sclerosis. Mult Scler. 2015;21(10):1251–1261. 58. Matute-Blanch C, Río J, Villar LM, et al. Chitinase 3-like 1 is associated with the response to interferon-beta treatment in multiple sclerosis. J Neuroimmunol. 2017;303:62–65. 59. Cantó E, Tintoré M, Villar LM, et al. Chitinase 3-like 1: prognostic biomarker in clinically isolated syndromes. Brain. 2015;138(Pt 4):918–931. 60. Comabella M, Fernández M, Martin R, et al. Cerebrospinal fluid chitinase 3-like 1 levels are associated with conversion to multiple sclerosis. Brain. 2010;133(Pt 4):1082–1093. 61. Mattsson N, Tabatabaei S, Johansson P, et al. Cerebrospinal fluid microglial markers in Alzheimer’s disease: elevated chitotriosidase activity but lack of diagnostic utility. Neuromolecular Med. 2011;13 (2):151–159. 62. Alcolea D, Carmona-Iragui M, Suárez-Calvet M, et al. Relationship between β-Secretase, inflammation and core cerebrospinal fluid biomarkers for Alzheimer’s disease. J Alzheimers Dis. 2014;42 (1):157–167. 63. Hellwig K, Kvartsberg H, Portelius E, et al. Neurogranin and YKL40: independent markers of synaptic degeneration and neuroinflammation in Alzheimer’s disease. Alzheimers Res Ther. 2015;7 (1):74. 64. Janelidze S, Hertze J, Zetterberg H, et al. Cerebrospinal fluid neurogranin and YKL-40 as biomarkers of Alzheimer’s disease. Ann Clin Transl Neurol. 2016;3(1):12–20. •• This original study clearly demonstrates that high CSF concentrations of YKL-40 are related both to AD and frontotemporal dementia as well as to vascular dementia. 65. Kester MI, Teunissen CE, Sutphen C, et al. Cerebrospinal fluid VILIP1 and YKL-40, candidate biomarkers to diagnose, predict and monitor Alzheimer’s disease in a memory clinic cohort. Alzheimers Res Ther. 2015;7(1):59. 66. Wennström M, Surova Y, Hall S, et al. The inflammatory marker YKL40 is elevated in cerebrospinal fluid from patients with Alzheimer’s but not parkinson’s disease or dementia with lewy bodies. PLoS One. 2015;10(8):e0135458. 67. Olsson B, Constantinescu R, Holmberg B, et al. The glial marker YKL-40 is decreased in synucleinopathies. Mov Disord. 2013;28 (13):1882–1885. 68. Hall S, Surova Y, Öhrfelt A, et al. Longitudinal measurements of cerebrospinal fluid biomarkers in Parkinson’s disease. Mov Disord. 2016;31(6):898–905. 69. Magdalinou NK, Paterson RW, Schott JM, et al. A panel of nine cerebrospinal fluid biomarkers may identify patients with atypical Parkinsonian syndromes. J Neurol Neurosurg Psychiatry. 2015;86 (11):1240–1247. • This original article highlights the potential usefulness of YKL40 in the diagnostic workout of Parkinsonian syndromes. Particularly, YKL-40 contributes to discriminate patients with Parkinson disease from atypical Parkinsonism cases. 70. Winer L, Srinivasan D, Chun S, et al. SOD1 in cerebral spinal fluid as a pharmacodynamic marker for antisense oligonucleotide therapy. JAMA Neurol. 2013;70(2):201–207. 71. Teunissen CE, Elias N, Koel-Simmelink MJA, et al. Novel diagnostic cerebrospinal fluid biomarkers for pathologic subtypes of frontotemporal dementia identified by proteomics. Alzheimers Dement (Amst.). 2016;2:86–94. • This original article shows an increase of CSF YKL-40 concentrations in FTD patients versus individuals with subjective cognitive impairment, based on unbiased proteomic technologies. In addition, YKL-40 is confirmed to be more associated with tau protein than TDP-43 neurodegeneration pathways after diagnostic pathological confirmation. 72. Vinther-Jensen T, Budtz-Jørgensen E, Simonsen AH, et al. YKL-40 in cerebrospinal fluid in Huntington’s disease–a role in pathology or a nonspecific response to inflammation? Parkinsonism Relat Disord. 2014;20(11):1301–1303.
73. Dubois B, Feldman HH, Jacova C, et al. Advancing research diagnostic criteria for Alzheimer’s disease: the IWG-2 criteria. Lancet Neurol. 2014;13(6):614–629. 74. Dubois B, Feldman HH, Jacova C, et al. Revising the definition of Alzheimer’s disease: a new lexicon. Lancet Neurol. 2010;9(11):1118– 1127. 75. Albert MS, DeKosky ST, Dickson D, et al. The diagnosis of mild cognitive impairment due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7(3):270–279. 76. McKhann GM, Knopman DS, Chertkow H, et al. The diagnosis of dementia due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7(3):263–269. 77. Sperling RA, Aisen PS, Beckett LA, et al. Toward defining the preclinical stages of Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7(3):280–292. 78. McKhann G, Drachman D, Folstein M, et al. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology. 1984;34(7):939–944. 79. Winblad B, Palmer K, Kivipelto M, et al. Mild cognitive impairment– beyond controversies, towards a consensus: report of the International Working Group on Mild Cognitive Impairment. J Intern Med. 2004;256(3):240–246. 80. Xia J, Broadhurst DI, Wilson M, et al. Translational biomarker discovery in clinical metabolomics: an introductory tutorial. Metabolomics. 2013;9(2):280–299. 81. Choi J, Lee H-W, Suk K. Plasma level of chitinase 3-like 1 protein increases in patients with early Alzheimer’s disease. J Neurol. 2011;258(12):2181–2185. 82. Olsson B, Lautner R, Andreasson U, et al. CSF and blood biomarkers for the diagnosis of Alzheimer’s disease: a systematic review and meta-analysis. Lancet Neurol. 2016;15(7):673–684. •• This systematic review and meta-analysis includes several CSF and blood biomarkers studies on AD up to July 2014. Notably, the authors highlight that YKL-40 is moderately associated with AD. 83. Perrin RJ, Craig-Schapiro R, Malone JP, et al. Identification and validation of novel cerebrospinal fluid biomarkers for staging early Alzheimer’s disease. PLoS One. 2011;6(1):e16032. 84. Rosén C, Andersson C-H, Andreasson U, et al. Increased levels of chitotriosidase and YKL-40 in cerebrospinal fluid from patients with Alzheimer’s Disease. Dement Geriatr Cogn Disord Extra. 2014;4 (2):297–304. 85. Paterson RW, Toombs J, Slattery CF, et al. Dissecting IWG-2 typical and atypical Alzheimer’s disease: insights from cerebrospinal fluid analysis. J Neurol. 2015;262(12):2722–2730. 86. Rodrigues FB, Byrne LM, McColgan P, et al. Cerebrospinal fluid inflammatory biomarkers reflect clinical severity in Huntington’s disease. PLoS One. 2016;11(9):e0163479. 87. Sutphen CL, Jasielec MS, Shah AR, et al. Longitudinal cerebrospinal fluid biomarker changes in preclinical Alzheimer disease during middle age. JAMA Neurol. 2015;72(9):1029–1042. 88. Antonell A, Mansilla A, Rami L, et al. Cerebrospinal fluid level of YKL-40 protein in preclinical and prodromal Alzheimer’s disease. J Alzheimers Dis. 2014;42(3):901–908. 89. Wildsmith KR, Schauer SP, Smith AM, et al. Identification of longitudinally dynamic biomarkers in Alzheimer’s disease cerebrospinal fluid by targeted proteomics. Mol Neurodegener. 2014;9:22. 90. Gispert JD, Monté GC, Falcon C, et al. CSF YKL-40 and pTau181 are related to different cerebral morphometric patterns in early AD. Neurobiol Aging. 2016;38:47–55. 91. Alcolea D, Vilaplana E, Pegueroles J, et al. Relationship between cortical thickness and cerebrospinal fluid YKL-40 in predementia stages of Alzheimer’s disease. Neurobiol Aging. 2015;36(6):2018–2023.
EXPERT REVIEW OF PROTEOMICS
92. Alcolea D, Martínez-Lage P, Sánchez-Juan P, et al. Amyloid precursor protein metabolism and inflammation markers in preclinical Alzheimer disease. Neurology. 2015;85(7):626–633. 93. Melah KE, Lu SY-F, Hoscheidt SM, et al. Cerebrospinal fluid markers of Alzheimer’s disease pathology and microglial activation are associated with altered white matter microstructure in asymptomatic adults at risk for Alzheimer’s disease. J Alzheimers Dis. 2016;50(3):873–886. 94. Racine AM, Koscik RL, Nicholas CR, et al. Cerebrospinal fluid ratios with Aβ42 predict preclinical brain β-amyloid accumulation. Alzheimers Dement (Amst.). 2016;2:27–38. 95. Magdalinou N, Lees AJ, Zetterberg H. Cerebrospinal fluid biomarkers in parkinsonian conditions: an update and future directions. J Neurol Neurosurg Psychiatry. 2014;85(10):1065–1075. 96. McCann H, Stevens CH, Cartwright H, et al. α-Synucleinopathy phenotypes. Parkinsonism Relat Disord. 2014;20(Suppl 1):S62–67. 97. Reitz C. Toward precision medicine in Alzheimer’s disease. Ann Transl Med. 2016;4(6):107. 98. Baldacci F, Lista S, Garaci F, et al. Biomarker-guided classification scheme of neurodegenerative diseases. J Sport Health Sci. 2016;5 (4):383–387. • This article points the finger on the necessity of an integrated vision focused on the common and overlapping pathophysiological mechanisms of the neurodegenerative diseases. 99. Lista S, Hampel H. Synaptic degeneration and neurogranin in the pathophysiology of Alzheimer’s disease. Expert Rev Neurother. 2016;17(1):47–57. 100. Morenas-Rodríguez E, Cervera-Carles L, Vilaplana E, et al. Progranulin protein levels in cerebrospinal fluid in primary neurodegenerative dementias. JAD. 2016;50(2):539–546. 101. Schuitemaker A, van der Doef TF, Boellaard R, et al. Microglial activation in healthy aging. Neurobiol Aging. 2012;33(6):1067– 1072. 102. Bonneh-Barkay D, Zagadailov P, Zou H, et al. YKL-40 expression in traumatic brain injury: an initial analysis. J Neurotrauma. 2010;27 (7):1215–1223. 103. Pasinetti GM. From epidemiology to therapeutic trials with antiinflammatory drugs in Alzheimer’s disease: the role of NSAIDs and cyclooxygenase in beta-amyloidosis and clinical dementia. J Alzheimers Dis. 2002;4(5):435–445. 104. Etminan M, Gill S, Samii A. Effect of non-steroidal anti-inflammatory drugs on risk of Alzheimer’s disease: systematic review and metaanalysis of observational studies. BMJ. 2003;327(7407):128. 105. Hoozemans JJM, Veerhuis R, Rozemuller AJM, et al. Non-steroidal anti-inflammatory drugs and cyclooxygenase in Alzheimer’s disease. Curr Drug Targets. 2003;4(6):461–468. 106. ADAPT Research Group; Martin BK, Szekely C, Szekely C, Brandt J, et al. Cognitive function over time in the Alzheimer’s Disease Antiinflammatory Prevention Trial (ADAPT): results of a randomized, controlled trial of naproxen and celecoxib. Arch Neurol. 2008;65 (7):896–905. 107. Breitner JC, Baker LD, Montine TJ, et al. Extended results of the Alzheimer’s disease anti-inflammatory prevention trial. Alzheimers Dement. 2011;7(4):402–411. 108. Henningsen KM, Therkelsen SK, Johansen JS, et al. Plasma YKL-40, a new biomarker for atrial fibrillation? Europace. 2009;11(8):1032–1036. 109. Marott SCW, Benn M, Johansen JS, et al. YKL-40 levels and atrial fibrillation in the general population. Int J Cardiol. 2013;167 (4):1354–1359.
299
110. Pickup JC. Inflammation and activated innate immunity in the pathogenesis of type 2 diabetes. Diabetes Care. 2004;27(3):813–823. 111. Ma W, Wang X, Du Y, et al. Association between human cartilage glycoprotein 39 (YKL-40) and arterial stiffness in essential hypertension. BMC Cardiovasc Disord. 2012;12:35. 112. Rathcke CN, Vestergaard H. YKL-40, a new inflammatory marker with relation to insulin resistance and with a role in endothelial dysfunction and atherosclerosis. Inflamm Res. 2006;55(6):221–227. 113. Iadecola C. The overlap between neurodegenerative and vascular factors in the pathogenesis of dementia. Acta Neuropathol. 2010;120(3):287–296. 114. Grammas P. Neurovascular dysfunction, inflammation and endothelial activation: implications for the pathogenesis of Alzheimer’s disease. J Neuroinflammation. 2011;8:26. 115. Zlokovic BV. Neurovascular pathways to neurodegeneration in Alzheimer’s disease and other disorders. Nat Rev Neurosci. 2011;12(12):723–738. 116. Erickson MA, Banks WA. Blood-brain barrier dysfunction as a cause and consequence of Alzheimer’s disease. J Cereb Blood Flow. 2013;33(10):1500–1513. 117. Kalra S, Grosset DG, Benamer HTS. Differentiating vascular parkinsonism from idiopathic Parkinson’s disease: a systematic review. Mov Disord. 2010;25(2):149–156. 118. Barnes DE, Yaffe K. The projected effect of risk factor reduction on Alzheimer’s disease prevalence. Lancet Neurol. 2011;10(9):819–828. 119. Reitz C, Mayeux R. Alzheimer disease: epidemiology, diagnostic criteria, risk factors and biomarkers. Biochem Pharmacol. 2014;88 (4):640–651. 120. Cereda E, Barichella M, Pedrolli C, et al. Diabetes and risk of Parkinson’s disease: a systematic review and meta-analysis. Diabetes Care. 2011;34(12):2614–2623. 121. Giuntini M, Baldacci F, Del Prete E, et al. Diabetes is associated with postural and cognitive domains in Parkinson’s disease. Results from a single-center study. Parkinsonism Relat Disord. 2014;20(6):671– 672. 122. Lista S, Khachaturian ZS, Rujescu D, et al. Application of systems theory in longitudinal studies on the origin and progression of Alzheimer’s disease. Methods Mol Biol. 2016;1303:49–67. 123. Hampel H, Lista S, Khachaturian ZS. Development of biomarkers to chart all Alzheimer’s disease stages: the royal road to cutting the therapeutic Gordian Knot. Alzheimers Dement. 2012;8(4):312–336. 124. Hampel H, O’Bryant SE, Castrillo JI, et al. PRECISION MEDICINE - The golden gate for detection, treatment and prevention of Alzheimer’s disease. J Prev Alz Dis. 2016;3(4):243–259. DOI:10.14283/ jpad.2016.112. •• This article summarizes the state of precision medicine in AD discussing the potential benefits for this clinically and pathologically heterogeneous neurodegenerative disease. 125. Dubois B, Hampel H, Feldman HH, et al. Preclinical Alzheimer’s disease: definition, natural history, and diagnostic criteria. Alzheimers Dement. 2016;12(3):292–323. 126. Hampel H, O’Bryant SE, Durrleman S, et al.; for the Alzheimer Precision Medicine Initiative (APMI). A precision medicine initiative for Alzheimer’s disease – The road ahead to biomarker-guided integrative disease modeling. Climacteric. 2017;20(2):107–118. doi:10.1080/13697137.2017.1287866 •• This manuscript introduces the concept of integrative disease modeling (IDM) in precision medicine and the Alzheimer Precision Medicine Initiative (APMI) and its cohort Program.
