Objective. Current studies of serum sclerostin levels in AS and RA patients are inconsistent. This meta-analysis was performed to identify the association of serum sclerostin level with AS and RA patients. Methods. Embase, PubMed, MEDLINE, and Cochrane Library databases (up to 25 January 2017) were used to collect all relevant published articles. Studies were pooled and standard mean difference (SMD) with 95% confidence interval (CI) was calculated. All data analyses were performed using RevMan 5.3. Results. Totally eight studies of AS including 420 AS patients and 317 healthy controls (HC) and three studies of RA including 145 RA patients and 127 HC were finally included in this meta-analysis. The results revealed that the serum sclerostin levels in both AS patients (SMD=-0.14; 95% CI=[-0.39,0.11]; P=0.28) and RA patients (SMD=-0.10; 95% CI=[-0.34,0.15]; P=0.43) were not significantly different from those in HC. Conclusion. The difference of serum sclerostin levels in AS and RA patients was not significantly different from HC, indicating that the sclerostin may not associate with the development of AS and RA.
National Natural Science Foundation of China81672086Science and Technology Plan of Taizhou City162yw011. Introduction
Ankylosing spondylitis (AS) and rheumatoid arthritis (RA) are both chronic systemic autoimmune diseases. AS is a chronic inflammatory disease that mainly involves axial skeleton and sacroiliac joint, which is characterized by entheses inflammation, resulting in uncontrolled osteoproliferation that usually leads to fusion and rigidity of the affected spine. RA is characterized by persistent inflammation of synovium, which eventually gives rise to joint destruction and deformation [1]. Progress had been greatly made in the long-term research; however, the pathogenesis of AS and RA is still unclear. Recently, several studies had revealed that Wnt signaling pathway inhibitor sclerostin plays a significant role in the development of AS and RA [2, 3].
Sclerostin is encoded by SOST gene and mainly expressed and secreted by osteocytes and other terminally differentiated cells embedded within mineralized matrix, such as osteocytes, chondrocytes, and cementocytes [4]. Sclerostin emerges as a natural inhibitor regulating the Wnt/β-catenin pathway, which had been considered as a crucial modulating pathway for bone formation [5]. Substantially, the canonical signaling was activated by the binding of Wnt ligands to the Frizzled receptor and the coreceptors low-density lipoprotein receptor-related proteins 5 and 6 (LRP-5 and LRP-6), thus maintaining the structural stability of β-catenin, which acts as a prominent component in the signaling pathway. Subsequently, β-catenin increases in cytoplasm and translocates into nucleus to modulate target genes transcription [6]. Sclerostin is capable of binding to LRP-5/LRP-6, which prevents Wnt proteins from reaching LRP-5/LRP-6, resulting in the inhibition of the canonical Wnt signaling pathway [7, 8].
The serum sclerostin levels had been suggested to implicate the pathogenesis of AS and RA in several studies; however, the results were inconsistent [9]. Therefore, the objective of our study is to comprehensively estimate the role of serum sclerostin in the development of AS through a meta-analysis.
2. Materials and Methods2.1. Publication Selection
This meta-analysis was performed using PubMed, Embase, MEDLINE, and Cochrane Library databases to identify all relative publications involved in serum sclerostin level in AS. The search keywords were as follows: “sclerostin,” “SOST,” “ankylosing spondylitis,” “spondyloarthritis,” “axial spondyloarthritis,” “peripheral spondyloarthritis,” “radiographic axial spondyloarthritis,” “non-radiographic axial spondyloarthritis,” “Bechterew’s disease,” “rheumatoid Arthritis,” and “RA.” Studies meeting the criteria as follows were included: (1) they were case-control studies or section-control studies; (2) study subjects were human AS patients according to the modified New York criteria [10] or ASAS diagnosis and classification criteria [11]; (3) studies provided the mean and the standard deviation (mean ± SD) or mean and the standard error (mean ± SE) of the serum sclerostin levels in any AS patients and HC. If there were duplicate publications, the one with the largest samples was selected and any meeting or conference abstracts were excluded. All analyses were based on previous published studies; thus no ethical approval and patient consent were required.
2.2. Data Extraction and Quality Assessment
Two researchers independently extracted data from all identified records according to the following criteria: first author’s name, publication year, country of study, patient ethnicity, study type, number, age, mean ± SD, source of control, measurement, and P value of the estimated effects. When original important data were uncertain in identified articles, we mailed the corresponding author to obtain further details. Any discrepancy on data extraction was discussed by the two authors.
Methodological quality of each of the articles was also assessed and scored independently by the two researchers using the Newcastle-Ottawa quality assessment scale (NOS) for case-control study and Agency for Healthcare Research and Quality (AHRQ) for cross-control study. NOS is composed of eight questions with nine possible points: (1) participants selection, 0–4; (2) subjects comparability, 0–2; and (3) ascertainment for the exposure, 0–3. AHRQ consists of 11 items. An item would be scored “0” when it was answered with “NO” or “UNCLEAR”; if it was answered with “YES,” then the item would be scored “1”.
2.3. Statistical Analysis
To evaluate the overall serum sclerostin levels, we calculated the standardized difference (SMD) for every study with 95% confidence intervals (CIs) due to the fact that the units of concentration of serum sclerostin were different. The mean ± SD was extracted and calculated in any included publications. While the original data were mean ± SEM, we transformed them to mean ± SD. I2 statistic was calculated to assess heterogeneity for the outcomes. A value of 25%−50% indicates a low degree of heterogeneity, a value of 50%–75% indicates a moderate degree, and a value of >75% indicates a high degree. When I2 value was >50%, a random-effects model was used to pool the data; otherwise, a fixed-effects model was selected. The funnel plot was applied to estimate the publication bias. Sensitivity analysis was used to investigate the source of heterogeneity. P<0.05 was considered statistically significant. All statistical analyses were carried out using RevMan 5.3 (the Cochrane Collaboration).
3. Results3.1. Publication Search
Initially altogether 249 articles were acquired. Among them, 57 articles were searched from PubMed, 156 from Embase, 31 from MEDLINE, and 5 from Cochrane Library. After reviewing the abstracts and full text, 239 articles were excluded due to their duplicate publication, unmatched purposes, review, conference abstracts, and low quality (Figure 1), and 10 articles which consisted of AS (n=7) and RA (n=3) were eventually included in this meta-analysis.
Study selection flow chart.
3.2. Characteristics of This Study
Finally seven studies [9, 12–17] including 420 AS patients and 317 HC and three studies [18–20] including 145 RA patients and 127 HC were in accordance with the inclusion criteria. The basic features of the included studies were shown in Table 1. The testing method of serum sclerostin levels in all the ten studies was enzyme-linked immunosorbent assay (ELISA). The results of methodological quality assessment using NOS for case-control studies and AHRQ for cross-sectional studies were shown in Table 1, in which the scores of the included articles were between 6 and 9.
Characteristics of included studies in the present meta-analysis.
Case
Control
References
Year
Region
Study type
N
Sex (M/F)
Mean (pg/ml)/(pmol/L)
SD (pg/ml)/(pmol/L)
N
Sex (M/F)
Mean (pg/ml)/(pmol/L)
SD (pg/ml)/(pmol/L)
P
Criteria for diseases
Source for control
Measurement
Quality score
Saad et al. [12]
2012
Brazil
Case-control
30
24/6
60.5
32.7
30
24/6
96.7
52.9
0.002
New York (1984)
Hospital
ELISA
6
Sakellariou et al. [13]
2016
Greece
Cross-sectional
57
53/4
357.4
280.85
34
32/2
276.4
103.79
0.611
New York (1984)
NA
ELISA
7
Appel et al. [14]
2009
German
Case-control
46
30/16
240.78
1078.35
50
33/17
522.26
1100.3
<0.01
NA
NA
ELISA
7
Korkosz et al. [9]
2013
Poland
Cross-sectional
78
290
639.4
23
120
431.62
0.09
New York (1984)
NA
ELISA
6
Rossini et al. [15]
2016
Italy
Case-control
71
59/12
25.2
79.21
71
59/12
38
144.93
<0.01
New York (1984)
Hospital
ELISA
8
Tuylu et al. [16]
2014
Turkey
Case-control
94
65/29
137.97
113.88
68
48/20
151
158
0.6
New York (1984)
NA
ELISA
6
Ustun et al. [17]
2014
Turkey
Cross-sectional
44
34/10
427.69
368.1
41
31/10
656.32
643.51
0.037
New York (1984)
NA
ELISA
7
Paccou et al. [18]
2014
France
Cross-sectional
75
19/56
440
190
75
19/56
450
200
0.96
ACR (2010)
Hospital
ELISA
8
Mehaney et al. [19]
2015
Egypt
Cross-sectional
40
28/12
400
200
40
28/12
500
200
0.14
ACR (2010)
Hospital
ELISA
8
Świerkot et al. [20]
2015
Poland
Clinical study
30
30/0
400
260
12
12/0
350
170
>0.05
ACR (1987)
NA
ELISA
—
NA: not available; SD: standard deviation.
3.3. Meta-Analysis in AS
Among the seven studies, the heterogeneity was statistically significant (P=0.01; I2 = 64%), and random-effects model was used, which showed that serum sclerostin levels in patients with AS were not statistically different compared with those in HC (SMD = −0.14; 95% CI = [-0.39,0.11]; P=0.28) (Figure 2). The shape of the funnel plot, which was recommended for estimating the bias stated in Cochrane Handbook, looks to be symmetrical, indicating that potential publication bias might slightly affect the present meta-analysis (Figure 3). In order to investigate the source of heterogeneity, the sensitivity analysis was performed and it was found that the source of heterogeneity was mainly from the study of Carla GS Saad et al., which only recruited AS patients with a Bath AS Disease Activity Index (BASDAI) ≥ 4 and/or refractory high inflammatory parameters. This was a big difference from the other included studies. After excluding the data extracted from the study, the value of I2 was reduced to 47% (P=0.09), which was considered to be acceptable, and the pooled SMD = −0.06; 95% CI = [-0.28,0.16]; P=0.61 (Figure 4).
Forest plot of serum sclerostin levels for AS patients versus healthy controls.
The shape of funnel plot.
Forest plot of serum sclerostin levels for AS patients versus healthy controls, which excluded the study of Saad et al.
3.4. Meta-Analysis in RA
Among the three studies of serum sclerostin in RA, the heterogeneity was not significant (P=0.41; I2=0), and fixed-effects model was used, and it demonstrated that serum sclerostin levels between RA patients and HC were not significantly different (SMD = −0.10; 95% CI = [-0.34,0.15]; P=0.43) (Figure 5). The shape of the funnel plot was not shown due to the fact that the sample is very small.
Forest plot of serum sclerostin levels for RA patients versus healthy controls.
4. Discussion
Sclerostin had been considered to work as a suppressor during bone formation, which was backed up by the observation that the differentiation and proliferation of human and mouse osteoblastic cells were suppressed as exogenous sclerostin was added to the cultures [21–23]. Furthermore, the investigation of sclerostin knockout mice demonstrates striking increases in bone formation, bone mineral density, and bone strength [24]. The accumulating evidences showed an adverse impact of sclerostin during bone formation. The exact mechanism by which sclerostin affects the development of AS remains unclear. It is generally agreed that sclerostin inhibits the development of bone formation through the Wnt pathway [5, 25]. Sclerostin binds to LRP-5/LRP-6 and prevents Wnt proteins from reaching LRP-5/LRP-6 and gives rise to inhibiting the canonical Wnt signaling pathway [7, 8]. During the development, AS is characterized by excessive bone formation, like syndesmophytes and enthesiophytes [26]. Therefore, the declined sclerostin may contribute to the binding of Wnt proteins and LRP-5/LRP-6, promoting the Wnt signaling.
In this present study, we retrieved seven articles to estimate the serum sclerostin levels in AS patients by meta-analysis. These results demonstrated no difference in serum sclerostin levels between AS patients and HC, suggesting that serum sclerostin levels may be irrelevant to the pathogenesis of AS in patients. Dickkopf-1 (Dkk-1), another Wnt signal pathway antagonist, had been reported to increase in AS patients in a meta-analysis [27]. A recent study suggested that AS progress includes the cycles of bone resorption and formation [28]. Sclerostin and Dkk-1 may also serve as promoters in bone resorption. In addition to regulation of bone homeostasis, Wnt signaling pathway had been suggested to affect the T cell populations and behavior, which plays a center role in AS. Wnt signaling promotes differentiation of regulatory T cells with FOXP3 and inhibited differentiation of proinflammatory T cells such as T helper 1 (Th1) and Th17 cells [29, 30], indicating that sclerostin may emerge as a double-edged sword during bone formation.
Considering the suppression of sclerostin in bone formation, the role of sclerostin was also investigated in RA patients. Vis et al. had firstly shown that serum sclerostin level related to the disease activity and radiographic joint damage in RA patients; however, we only get the abstract of their study. Sclerostin inhibition is considered as a powerful tool to enhance bone repair in inflammatory arthritis in mice reported by Chen et al., indicating that sclerostin plays a pivotal role in the development of RA [31]. Recently, several studies [18–20] had demonstrated that serum sclerostin levels between RA patients and controls were not significantly different, and Mehaney et al. had further revealed that there was no significant correlation between serum sclerostin level and disease activity and bone mineral density. In our meta-analysis, we evaluated the association of serum sclerostin levels with RA, and there was no significant difference between RA patients and HC, indicating that sclerostin may not implicate the development of human RA.
As far as we know, this is the first comprehensive meta-analysis of serum sclerostin levels in AS and RA patients. However, several limitations should be considered in this study. First, there are very few studies after the first sorting, which account for just 4%, and the number of patients is relatively small; thus the limited size might affect the conclusion. Second, the articles, which only support median and range, were excluded. The method of transformation had been reported by Hozo et al. [32]; however, the result was not accurate as transformed and even presented the opposite results [33, 34]. Third, the information about factors that affected serum sclerostin was not given in the included studies, like age, sex, and ethnicity. Therefore, we will not succeed in further analysis between serum sclerostin and AS, which may influence the reliability of our study. Finally, the sclerostin enzyme-linked immunosorbent assay (ELISA) kit, which is different among the included papers, should be considered. In the group of included papers of AS, Saad et al., Rossini et al., and Tuylu et al. used Biomedica kit, Appel et al. and Korkosz et al. used R&D Systems kit, and Sakellariou et al. and Ustun et al. used BD Biosciences kit. In the group of included papers of RA, they all used TECOmedical kit. The serum sclerostin concentrations were higher when measured with the Biomedica kit as compared with TECOmedical kit and R&D Systems kit [35, 36]. This may be used to explain why the value of I2 was, respectively, high and low in the meta-analysis of serum sclerostin in AS patients and RA patients. To our knowledge, no data have been published comparing the BD Biosciences kit and R&D systems kit, Biomedica kit, or TECOmedical kit. The variability in values generated from these sclerostin ELISA kits raises questions regarding the accuracy and specificity of the assays.
In conclusion, our meta-analysis demonstrated that serum sclerostin levels in AS and RA patients were not significantly different from those in HC. These results suggest that sclerostin may not be associated with the development of AS and RA in patients. Nonetheless, determining the underlying mechanisms of sclerostin in AS and RA patients still awaits further analysis using larger samples.
Conflicts of Interest
The authors declare that there are no conflicts of interest.
Acknowledgments
This work was supported by grants from the National Natural Science Foundation of China (81672086) and the Science and Technology Plan of Taizhou City (162yw01).
FiresteinG. S.Evolving concepts of rheumatoid arthritis2003423693735636110.1038/nature016612-s2.0-0038011833XieW.ZhouL.LiS.HuiT.ChenD.Wnt/beta-catenin signaling plays a key role in the development of spondyloarthritis20161364253110.1111/nyas.12968SerorR.BoudaoudS.PavyS.NocturneG.SchaeverbekeT.SarauxA.ChansonP.GottenbergJ.-E.Devauchelle-PensecV.TobónG. J.MarietteX.Miceli-RichardC.Increased Dickkopf-1 in recent-onset rheumatoid arthritis is a new biomarker of structural severity. data from the ESPOIR Cohort201661842110.1038/srep184212-s2.0-84955446241MoesterM. J.PapapoulosS. E.LöwikC. W.van BezooijenR. L.Sclerostin: current knowledge and future perspectives20108729910710.1007/s00223-010-9372-12-s2.0-79952118262BaronR.RawadiG.Minireview: targeting the Wnt/β-catenin pathway to regulate bone formation in the adult skeleton200714862635264310.1210/en.2007-02702-s2.0-34250823973RaoT. P.KühlM.An updated overview on wnt signaling pathways: a prelude for more2010106121798180610.1161/CIRCRESAHA.110.2198402-s2.0-77954210220LiX.ZhangY.KangH.LiuW.LiuP.ZhangJ.HarrisS. E.WuD.Sclerostin binds to LRP5/6 and antagonizes canonical Wnt signaling20052802019883198872-s2.0-2124448092410.1074/jbc.M413274200SemënovM.TamaiK.HeX.SOST is a ligand for LRP5/LRP6 and a Wnt signaling inhibitor20052802926770267752-s2.0-2284444593410.1074/jbc.M504308200KorkoszM.Ga̧sowskiJ.LeszczyńskiP.Pawlak-BuśK.JekaS.KucharskaE.GrodzickiT.High disease activity in ankylosing spondylitis is associated with increased serum sclerostin level and decreased wingless protein-3a signaling but is not linked with greater structural damage201314, article 9910.1186/1471-2474-14-992-s2.0-84875031391van der LindenS.ValkenburgH. A.CatsA.Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria198427436136810.1002/art.17802704012-s2.0-0021272107RudwaleitM.Van Der HeijdeD.LandewéR.ListingJ.AkkocN.BrandtJ.BraunJ.ChouC. T.Collantes-EstevezE.DougadosM.HuangF.GuJ.KhanM. A.KirazliY.MaksymowychW. P.MielantsH.SørensenI. J.OzgocmenS.RoussouE.Valle-OñateR.WeberU.WeiJ.SieperJ.The development of Assessment of SpondyloArthritis international Society classification criteria for axial spondyloarthritis (part II): validation and final selection200968677778310.1136/ard.2009.1082332-s2.0-67449128733SaadC. G. S.RibeiroA. C. M.MoraesJ. C. B.TakayamaL.GoncalvesC. R.RodriguesM. B.de OliveiraR. M.SilvaC. A.BonfaE.PereiraR. M. R.Low sclerostin levels: a predictive marker of persistent inflammation in ankylosing spondylitis during anti-tumor necrosis factor therapy?2012145, article R21610.1186/ar40552-s2.0-84867325736SakellariouG. T.IliopoulosA.KonstaM.KenanidisE.PotoupnisM.TsiridisE.GavanaE.SayeghF. E.Serum levels of Dkk-1, sclerostin and VEGF in patients with ankylosing spondylitis and their association with smoking, and clinical, inflammatory and radiographic parameters201610.1016/j.jbspin.2016.05.0082-s2.0-84979650146AppelH.Ruiz-HeilandG.ListingJ.ZwerinaJ.HerrmannM.MuellerR.HaibelH.BaraliakosX.HempfingA.RudwaleitM.SieperJ.SchettG.Altered skeletal expression of sclerostin and its link to radiographic progression in ankylosing spondylitis20096011325732622-s2.0-7035055848810.1002/art.24888RossiniM.ViapianaO.IdolazziL.GhellereF.FracassiE.TropliniS.PovinoM. R.KunnathullyV.AdamiS.GattiD.Higher level of Dickkopf-1 is associated with low bone mineral density and higher prevalence of vertebral fractures in patients with ankylosing spondylitis201698543844510.1007/s00223-015-0093-32-s2.0-84949521640TuyluT.SariI.SolmazD.KozaciD. L.AkarS.GunayN.OnenF.AkkocN.Fetuin-A is related to syndesmophytes in patients with ankylosing spondylitis: a case control study2014691068869310.6061/clinics/2014(10)072-s2.0-84920490140UstunN.TokF.KalyoncuS.MotorR.YukselA. E.YagizH.Guler.TurhanogluA. D.Sclerostin and Dkk-1 in patients with ankylosing spondylitis2014392146151PaccouJ.MentaverriR.RenardC.LiabeufS.FardelloneP.MassyZ. A.BrazierM.KamelS.The relationships between serum sclerostin, bone mineral density, and vascular calcification in rheumatoid arthritis20149912474047482-s2.0-8491663065210.1210/jc.2014-2327MehaneyD. A.EissaM.AnwarS.Fakhr El-DinS.Serum sclerostin level among egyptian rheumatoid arthritis patients: relation to disease activity, bone mineral density and radiological grading2015403268274ŚwierkotJ.GruszeckaK.MatuszewskaA.WilandP.Assessment of the effect of methotrexate therapy on bone metabolism in patients with rheumatoid arthritis201563539740410.1007/s00005-015-0338-x2-s2.0-84942194280van BezooijenR. L.RoelenB. A. J.VisserA.van der Wee-PalsL.De WiltE.KarperienM.HamersmaH.PapapoulosS. E.Ten DijkeP.LöwikC. W. G. M.Sclerostin is an osteocyte-expressed negative regulator of bone formation, but not a classical BMP antagonist2004199680581410.1084/jem.200314542-s2.0-12144286871LootsG. G.KneisselM.KellerH.BaptistM.ChangJ.ColletteN. M.OvcharenkoD.Plajzer-FrickI.RubinE. M.Genomic deletion of a long-range bone enhancer misregulates sclerostin in Van Buchem disease200515792893510.1101/gr.34371052-s2.0-22244474640SutherlandM. K.GeogheganJ. C.YuC.TurcottE.SkonierJ. E.WinklerD. G.LathamJ. A.Sclerostin promotes the apoptosis of human osteoblastic cells: a novel regulation of bone formation200435482883510.1016/j.bone.2004.05.0232-s2.0-5344234448LiX.OminskyM. S.NiuQ.-T.SunN.DaughertyB.D'AgostinD.KuraharaC.GaoY.CaoJ.GongJ.AsuncionF.BarreroM.WarmingtonK.DwyerD.StolinaM.MoronyS.SarosiI.KostenuikP. J.LaceyD. L.SimonetW. S.HuaZ. K.PasztyC.Targeted deletion of the sclerostin gene in mice results in increased bone formation and bone strength200823686086910.1359/jbmr.0802162-s2.0-44449099165GoldringS. R.GoldringM. B.Eating bone or adding it: the wnt pathway decides200713213313410.1038/nm0207-1332-s2.0-33846972636van EchteldI.CiezaA.BoonenA.StuckiG.ZochlingJ.BraunJ.van der HeijdeD.Identification of the most common problems by patients with ankylosing spondylitis using the international classification of functioning, disability and healt2006331224752483ZhangL.OuyangH.XieZ.Liang. Z. H.WuX. W.Serum DKK-1 level in the development of ankylosing spondylitis and rheumatic arthritis: a meta-analysis201648e22810.1038/emm.2016.12CortesA.MaksymowychW. P.WordsworthB. P.InmanR. D.DanoyP.RahmanP.StoneM.CorrM.GenslerL.GladmanD.MorganA.Marzo-OrtegaH.WardM. M.LearchT. J.ReveilleJ. D.BrownM.WeismanM. H.Association study of genes related to bone formation and resorption and the extent of radiographic change in ankylosing spondylitis2015747138713932-s2.0-8493490603810.1136/annrheumdis-2013-204835MuralidharanS.HanleyP. J.LiuE.ChakrabortyR.BollardC.ShpallE.RooneyC.SavoldoB.RodgersJ.DottiG.Activation of Wnt signaling arrests effector differentiation in human peripheral and cord blood-derived T lymphocytes201118710522152322-s2.0-8145515495510.4049/jimmunol.1101585DingY.ShenS.LinoA. C.Curotto de LafailleM. A.LafailleJ. J.Beta-catenin stabilization extends regulatory T cell survival and induces anergy in nonregulatory T cells200814216216910.1038/nm17072-s2.0-38949168084ChenX.-X.BaumW.DwyerD.StockM.SchwabeK.KeH.-Z.StolinaM.SchettG.BozecA.Sclerostin inhibition reverses systemic, periarticular and local bone loss in arthritis20137210173217362-s2.0-8488381851610.1136/annrheumdis-2013-203345HozoS. P.DjulbegovicB.HozoI.Estimating the mean and variance from the median, range, and the size of a sample20055, article 1310.1186/1471-2288-5-132-s2.0-21544449115TaylanA.SariI.AkinciB.BilgeS.KozaciD.AkarS.ColakA.YalcinH.GunayN.AkkocN.Biomarkers and cytokines of bone turnover: extensive evaluation in a cohort of patients with ankylosing spondylitis201213, article 19110.1186/1471-2474-13-1912-s2.0-84866791414KlingbergE.NurkkalaM.CarlstenH.Forsblad-d'EliaH.Biomarkers of bone metabolism in ankylosing spondylitis in relation to osteoproliferation and osteoporosis20144171349135610.3899/jrheum.131199PiecI.WashbourneC.TangJ.FisherE.GreevesJ.JacksonS.FraserW. D.How accurate is your sclerostin measurement? comparison between three commercially available sclerostin ELISA kits201698654655510.1007/s00223-015-0105-32-s2.0-84953398804CostaA. G.CremersS.BilezikianJ. P.Sclerostin measurement in human disease: validity and current limitations201796242810.1016/j.bone.2016.10.012