Glycoprotein YKL-40: A potential biomarker of disease activity in rheumatoid arthritis during intensive treatment with csDMARDs and infliximab. Evidence from the randomised controlled NEO-RACo trial

RESEARCH ARTICLE Glycop o ein YKL-40: A po en ial bioma ke o disease ac i i y in heuma oid a h i is du ing in ensi e ea men wi h csDMARDs and in liximab. E idence om he andomised con olled NEO-RACo ial Tuija Va ¨a ¨na ¨nen 1 , Ka iina Vuol eenaho 1 *, Hannu Kau iainen 2,3,4,5 , Riina Nieminen 1 , Timo Mo ¨ o ¨nen 6 , Pekka Hannonen 7 , Ma kku Ko pela 8 , Ma kku J. Kauppi 9,10 , Ka i Laiho 9 , Oili Kaipiainen-Seppa ¨nen 11 , Rii a Luosuja ¨ i 12 , Tea Uusi alo 13 , Toini Uu ela 14 , Ma ja a Lei isalo-Repo 12 , Ee a Moilanen 1 , on behal o he NEO-RACo S udy G oup ¶ 1The Immunopha macology Resea ch G oup, Facul y o Medicine and Li e Sciences, Uni e si y o Tampe e and Tampe e Uni e si y Hospi al, Tampe e, Finland, 2Depa men o Gene al P ac ice, Uni e si y o Helsinki, Helsinki, Finland, 3Uni o P ima y Heal h Ca e, Helsinki Uni e si y Hospi al, Helsinki, Finland, 4Uni o P ima y Heal h Ca e, Kuopio Uni e si y Hospi al, Kuopio, Finland, 5MedCa e L d, A ¨a ¨nekoski, Finland, 6Depa men o Rheuma ology, Uni e si y o Tu ku and Tu ku Uni e si y Hospi al, Tu ku, Finland, 7Depa men o Medicine, Jy a ¨skyla ¨Cen al Hospi al, Jy a ¨skyla ¨, Finland, 8Depa men o In e nal Medicine, Cen e o Rheuma ic Diseases, Tampe e Uni e si y Hospi al, Tampe e, Finland, 9Depa men o Medicine, Pa ¨ija ¨ -Ha ¨me Cen al Hospi al, Lah i, Finland, 10 Facul y o Medicine and Li e Sciences, Uni e si y o Tampe e, Tampe e, Finland, 11 Depa men o Medicine, Kuopio Uni e si y Hospi al, Kuopio, Finland, 12 Rheuma ology, Uni e si y o Helsinki and Helsinki Uni e si y Hospi al, Helsinki, Finland, 13 Depa men o Medicine, Ha ¨meenlinna Cen al Hospi al, Ha ¨meenlinna, Finland, 14 Depa men o Medicine, Lapland Cen al Hospi al, Ro aniemi, Finland ¶ All he au ho s named in he au ho lis a e membe s o he NEO-RACo S udy G oup. O he membe s o he NEO-RACo S udy G oup a e p o ided in he Acknowledgmen s. *ka iina. uol eenaho@u a. i Abs ac Objec i e YKL-40, a chi inase-like glycop o ein associa ed wi h in lamma ion and issue emodeling, is p oduced by join issues and ecognized as a candida e au o-an igen in heuma oid a h i is (RA). In he p esen s udy, we in es iga ed YKL-40 as a po en ial bioma ke o dis- ease ac i i y in pa ien s wi h ea ly RA a baseline and du ing in ensi e ea men aiming o ea ly emission. Me hods Nine y-nine pa ien s wi h ea ly DMARD-naï e RA pa icipa ed in he NEO-RACo s udy. Fo he i s ou weeks, he pa ien s we e ea ed wi h he combina ion o sulphasalazine, me h- o exa e, hyd oxychlo oquine and low dose p ednisolone (FIN-RACo DMARD combina ion), and subsequen ly andomized o ecei e placebo o in liximab added on he ea men o u he 22 weeks. Disease ac i i y was e alua ed using he 28-join disease ac i i y sco e and plasma YKL-40 concen a ions we e measu ed by immunoassay. PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0183294 Augus 25, 2017 1 / 15 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 OPEN ACCESS Ci a ion: Va¨a¨na¨nen T, Vuol eenaho K, Kau iainen H, Nieminen R, Mo¨ o¨nen T, Hannonen P, e al. (2017) Glycop o ein YKL-40: A po en ial bioma ke o disease ac i i y in heuma oid a h i is du ing in ensi e ea men wi h csDMARDs and in liximab. E idence om he andomised con olled NEO- RACo ial. PLoS ONE 12(8): e0183294. h ps://doi. o g/10.1371/jou nal.pone.0183294 Edi o : Michael Nu mohamed, VU Uni e si y Medical Cen e , NETHERLANDS Recei ed: Oc obe 28, 2016 Accep ed: Augus 2, 2017 Published: Augus 25, 2017 Copy igh : ©2017 Va¨a¨na¨nen e al. This is an open access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal au ho and sou ce a e c edi ed. Da a A ailabili y S a emen : All ele an da a a e wi hin he pape and i s Suppo ing In o ma ion iles. Funding: This s udy was inancially suppo ed by he Compe i i e Resea ch Funding o Tampe e Uni e si y Hospi al (KV and EM), he Helsinki Uni e si y Cen al Hospi al Resea ch Funds (ML-R) and Finska La¨ka esa¨llskape (ML-R). A baseline an un es ic ed g an was p o ided by Sche ing- Resul s A he baseline, plasma YKL-40 concen a ion was 57 ±37 (mean ±SD) ng/ml. YKL-40 was signi ican ly associa ed wi h he disease ac i i y sco e, in e leukin-6 and e y h ocy e sedi- men a ion a e bo h a he baseline and du ing he 26 weeks’ ea men . The csDMARD combina ion dec eased YKL-40 le els al eady du ing he i s ou weeks o ea men , and he e was no u he educ ion when he umou nec osis ac o -αan agonis in liximab was added on he combina ion ea men . Conclusions High YKL-40 le els we e ound o be associa ed wi h disease ac i i y in ea ly DMARD-naï e RA and du ing in ensi e ea - o- a ge he apy. The p esen esul s sugges YKL-40 as a use ul bioma ke o disease ac i i y in RA o be used o s ee ea men owa ds emission. In oduc ion YKL-40 is an in lamma ion-associa ed glycop o ein wi h a molecula weigh o 40 kDa. I belongs o he amily o chi inase like p o eins bu lacks he enzyma ic ac i i y o ue chi i- nases. YKL-40 is known also by names chi inase-3-like p o ein 1 (Chi3-l1), b eas eg ession p o ein 39 (BRP-39), human ca ilage glycop o ein 39, and chond ex. I is exp essed in a ious cell ypes and inc eased le els o YKL-40 ha e been linked o in lamma ion, issue emodeling and cance , bu he exac biological ac i i ies a e ye o be iden i ied [1]. Rheuma oid a h i is (RA) is a ch onic au oimmune disease a ec ing p incipally he join s, bu he mechanisms ha igge he au oimmune esponses leading e en ually o join des uc- ion a e no ully known. A ea ly phases o he p ocess, exogenous and au ologous an igens a e p esen ed o T-cells by an igen p esen ing cells and in e es ingly, YKL-40 is ecognized as a candida e au oan igen [2–6]. Ci cula ing YKL-40 le els ha e been shown o be highe in RA pa ien s as compa ed o heal hy con ols [7–13]. Also, he YKL-40 concen a ions in syno ial luid (SF) a e highe han hose measu ed in plasma indica ing signi ican in a-a icula p o- duc ion [7,11]. Wi hin RA join s, YKL-40 has been ecognized as a majo sec e o y p o ein o a icula chond ocy es [14]. Syno ial cells, mac ophages and neu ophils in il a ing in o he RA syno ium also p oduce YKL-40 [1,14–16], and jus ecen ly, splenic T-cells ha e been added o he lis o YKL-40 p oducing cells in RA [17]. In he ea men o RA, he cu en ea - o- a ge app oach aims o ea ly emission o maximally low disease ac i i y. Biological disease modi ying an i- heuma ic d ugs (bDMARDs), including umou nec osis ac o -α(TNF-α) inhibi o s, a e ecommended o be commenced wi h con en ional sys emic disease-modi ying an i heuma ic d ugs (csDMARDs) i he ea men a ge is no eached wi h csDMARDs alone [18]. In RA, assessmen o disease ac i i y is based on composi e indices, such as he 28-join disease ac i i y sco e (DAS28) [19] e alua ing he coun o ende and swollen join s, in lamma ion and pa ien ´s assessmen o he disease ac i i y. We ha e epo ed p e iously based on he cu en NEO-RACo ial [20– 23] excellen sus ained clinical esul s wi h ea - o- a ge app oach in pa ien s wi h ea ly, DMARD-naï e RA by using he in ensi ied Finnish Rheuma oid A h i is combina ion (FIN- RACo) ea men . This consis s o a combina ion o h ee csDMARDs (i.e. sulphasalazine, me ho exa e and hyd oxychlo oquine) and low dose glucoco icoid (GC) ollowing a p ede- ined p o ocol supplemen ed wi h ac i e ea men o in lamed join s wi h in a-a icula GC YKL-40 as a bioma ke o disease ac i i y in heuma oid a h i is PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0183294 Augus 25, 2017 2 / 15 Plough Finland, which was used o he pu chase o in liximab. Sche ing-Plough Finland also p o ided suppo o in es iga o mee ings. The unde s did no ha e any ole in he s udy design, da a collec ion and analysis, p epa a ion o he manusc ip , o decision o publish. Compe ing in e es s: Au ho s ha e epo ed he ollowing compe ing in e es s: KV has ecei ed hono a ia om GSK. HK has ecei ed hono a ia om Abb ie and P ize . TM has ecei ed hono a ia om Abb ie, Amgen, BMS, Eli Lilly, MSD, No a is, P ize , and Roche. PH has ecei ed hono a ia om Abb ie, As aZeneca, BMS, MSD, P ize , P o essio Finland, and Roche. MK has ecei ed hono a ia o consul ing ees om Abbo , BMS, Celgene, GSK, Hospi a, MSD, No a is, P ize , Roche, UCB Pha ma. MJK has ecei ed hono a ia om Abbo , BMS, GSK, MSD, P ize , Roche, UCB Pha ma. OK- S has ecei ed hono a ia om Abb ie, Medac, MSD, No a is, P ize , Roche, and UCB Pha ma. M-LR has ecei ed hono a ia om Abb ie, BMS, MSD, P ize , Regene on and Roche. EM has ecei ed hono a ia om GSK, BMS and P ize . No non- inancial con lic s o in e es exis o any o he au ho s. A baseline an un es ic ed g an was p o ided by Sche ing-Plough Finland, which was used o he pu chase o in liximab. Sche ing- Plough Finland also p o ided suppo o in es iga o mee ings. The unde s had no in ol emen in he s udy design, in he collec ion, analysis and in e p e a ion o he da a in he w i ing o he epo , o in he decision o submi he pape o publica ion. This does no al e ou adhe ence o PLOS ONE policies on sha ing da a and ma e ials. injec ions [24]. A 2 yea s, DAS28 emission was achie ed in 82% o he pa ien s [20]. Adding in liximab o he FIN-RACo combina ion ea men o he i s 6 mon hs in a andomized, double-blind and pa allel-g oup manne , induced emission mo e apidly, bu di e ences be ween he ea men g oups we e no s a is ical signi ican a 2 o 5 yea s ollow-up [20,21]. In he sea ch o no el bioma ke s, we hypo hesized ha YKL-40, an in lamma o y ac o p oduced mainly by in a-a icula issues, could e lec disease ac i i y and in lamma ion in RA pa ien s. In he p esen s udy, we es ed his hypo hesis by measu ing he YKL-40 plasma le els in DMARD-naï e pa ien s a he baseline and du ing in ensi e an i- heuma ic ea men in he NEO-RACo s udy. Me hods S udy design, pa ien s, ou comes and ollow-up Nine y-nine (99) pa ien s wi h ea ly ac i e RA ul illing he classi ica ion c i e ia posi ioned by ACR [25] we e ec ui ed in o his in es iga o ini ia ed mul icen e s udy be ween Ma ch 2003 and Ap il 2005. NEO-RACo s udy is a p ospec i e 5 yea ial, wi h ex ension o 10 yea s. Fol- low-up was comple ed in 2015 a e which seconda y endpoin analyses we e pe o med. Pa ien s we e ea ed wi h an in ensi ied FIN-RACo combina ion including sulphasalazine (up o a maximum o 2 g/day), me ho exa e (up o a maximum o 25 mg/week), hyd oxychlo o- quine (35 mg/kg/week), and p ednisolone (7.5 mg/day) and double blindly andomized o ecei e in alloca ion a io 1:1 in usions o ei he placebo (sol en wi hou ac i e d ug) o in lix- imab (3 mg/kg), a weeks 4, 6, 10, 18 and 26. The pa ien s we e s a i ied acco ding o se oposi- i i y ( heuma oid ac o de ined in a local acc edi ed labo a o y). The andomisa ion was pe o med cen ally by an ex e nal labo a o y no pa icipa ing in he clinical ial in blocks o 10 pa ien s. As pa o he p o ocol, all in lamed join s we e ac i ely ea ed wi h in a-a icula injec ions o glucoco icoids. A e 24 mon hs, i he pa ien was in emission, p ednisolone and DMARDs we e g adually ape ed o wi h a p ede ined p o ocol o in he case o non- emission medica ion was modi ied acco ding o he judgmen o he ea ing heuma ologis aiming a s ic emission. The DAS28 sco e (ESR) was used o e alua e disease ac i i y [19]. The modi ied Sha p- an de Heijde sco e was used o e alua e adiologic sco e [26]. The pa ien selec ion c i e ia as well as he ea men p o ocol, ou comes and ollow-up ha e been desc ibed ea lie [20–23]. The s udy was conduc ed acco ding o he Decla a ion o Helsinki. All pa ien s ga e in o med w i en consen . The s udy p o ocol was app o ed by he e hics commi ee o he Hospi al Dis ic o Helsinki and Uusimaa (s a emen numbe 676/E5/02, da e Janua y 14, 2002) and egis e ed a Helsinki Uni e si y Cen al Hospi al HUCH S udy Regis e (da e Ap il 29, 2003). The s udy was egis e ed a he Clinical T ials Da abase (h ps://clinical ials.go / c 2/show/NCT00908089? e m=NCT00908089& ank=1) in 2009, soon a e he e ision o he Decla a ion o Helsinki in 2008, which p omo ed ial egis a ion in a publicly accessible da a- base. The au ho s con i m ha all ongoing and ela ed ials o his d ug/in e en ion a e egis e ed. Plasma samples and ELISA Plasma samples we e collec ed a weeks 0, 4, 10, 18, 26 (all ime poin s we e a ailable om 88 pa ien s, 60 emales, Fig 1) and kep a -80˚C un il assayed. The concen a ions o YKL-40 and ma ix me allop o einase-3 (MMP-3) (R&D Sys ems, Minneapolis, MN, USA), and in e leu- kin-6 (IL-6) (eBioScience Inc., San Diego, CA, USA) we e measu ed by ELISA ollowing he ins uc ions o he manu ac u e . YKL-40 as a bioma ke o disease ac i i y in heuma oid a h i is PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0183294 Augus 25, 2017 3 / 15 S a is ics Resul s a e exp essed as medians wi h in e qua ile anges (IQRs) o means wi h s anda d de ia ions (SDs) o 95% con idence in e als (95% CIs). S a is ical compa isons be ween he g oups we e ca ied ou by using - es , boo s ap ype es (5000 eplica ions) o chi-squa e es , when app op ia e. Analysis o epea ed da a was ca ied ou by using gene alizing es ima - ing equa ions (GEE) wi h he uns uc u ed co ela ion s uc u e. Gene alized es ima ing equa- ions we e de eloped as an ex ension o he gene al linea model (e.g. OLS eg ession analysis) o analyze longi udinal and o he co ela ed da a. In longi udinal GEE analyses, DAS28 and i s componen s we e used as ou come a iables, YKL-40 and ime as co a ia es and he model was adjus ed wi h age, gende , heuma oid ac o posi i i y and ea men . In he case o iola- ion o he assump ions (e.g. non-no mali y), a boo s ap- ype es was used (5 000 eplica- ions). A ea unde he cu e (AUC) was calcula ed wi h he apezoidal me hod o e he ime Fig 1. CONSORT low diag am o he NEO-RACo s udy. h ps://doi.o g/10.1371/jou nal.pone.0183294.g001 YKL-40 as a bioma ke o disease ac i i y in heuma oid a h i is PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0183294 Augus 25, 2017 4 / 15 poin s om 0 o 26 weeks. A possible non-linea ela ionships be ween YKL40(AUC) and DAS28(AUC) and IL-6(AUC) was assessed by using eg ession models wi h quad a ic e m. Co ela ion coe icien s we e calcula ed by he Spea man me hod. The no mali y o he a i- ables was es ed by using he Shapi o-Wilk W es . STATA so wa e ( e sion 13.1, S a aCo p, LP, Texas, USA) was used in he s a is ical analysis. Resul s A he baseline, YKL-40 concen a ion in he DMARD-naï e RA pa ien s was 57 ±37 ng/ml (mean ±SD, n = 88) and he e was no di e ences be ween emales and males (55 ±37 ng/ml, n = 60 s 61 ±38 ng/ml, p = 0.43). Cha ac e is ics o he pa ien s a e shown in Table 1. The le - els o YKL-40 unde baseline condi ions co ela ed posi i ely wi h he disease ac i i y mea- su ed as DAS28 sco e ( = 0.41, p<0.001, Fig 2A). When he DAS28 componen s we e assessed sepa a ely, he baseline YKL-40 co ela ed posi i ely wi h he coun o ende join s ( = 0.28, p = 0.004, Fig 2B) and ESR ( = 0.50, p<0.001, Fig 2C), bu he e was no co ela ion wi h he coun o swollen join s o pa ien ´s global assessmen o gene al heal h. P oin lamma- o y cy okine IL-6 as well as MMP-3 showed a posi i e co ela ion wi h YKL-40 a he baseline ( = 0.30, p = 0.004, Fig 2D and = 0.39, p<0.001, Fig 2E, espec i ely). The pa ien s ecei ed a combina ion o sulphasalazine, me ho exa e, hyd oxychlo oquine and low dose p ednisolone o he i s ou weeks ollowing he in ensi ied FIN-RACo combi- na ion ea men s a egy [24]. Subsequen ly, he pa ien s we e andomized o ecei e in lixi- mab o placebo, which was added on he csDMARD combina ion ea men o he u he 22 weeks. Table 1. The baseline cha ac e is ics o he pa ien s. Cha ac e is ic The ini ial andomiza ion g oup P- alue All FIN-RACo + In liximab (n = 44) FIN-RACo + Placebo (n = 44) (n = 88) Demog aphic da a a baseline Female, no (%) 32 (73) 28 (64) 0.41 60 (68) Age (yea s), mean ±SD yea s 47 ±9 44 ±11 0.22 46 ±10) Du a ion o symp oms (mon hs), median (IQR) 4 (2, 5) 4 (3, 6) 0.54 4 (2, 6) Rheuma oid ac o p esen (%) 35 (80) 32 (73) 0.56 67 (76) Measu es o disease ac i i y a baseline Swollen join coun , mean ±SD 14 ±5 16 ±8 0.33 15 ±6 Tende join coun , mean ±SD 18 ±9 21 ±11 0.10 20 ±10 E y h ocy e sedimen a ion a e (mm/h), mean ±SD 33 ±20 33 ±23 0.84 33 ±21 Pa ien ’s global assessmen (VAS, mm), mean ±SD 49 ±25 47 ±28 0.71 48 ±26 Pain (VAS, mm), mean ±SD 52 ±28 51 ±28 0.79 52 ±27 Physician’s global assessmen (VAS, mm), mean ±SD 46 ±20 53 ±20 0.12 50 ±20 DAS28, mean ±SD 5.4 ±1.0 5.6 ±1.4 0.66 5.5 ±1.2 Physical unc ion (HAQ), mean ±SD 1.0 ±0.6 0.9 ±0.7 0.47 0.9 ±0.7 Radiog aphy a baseline E osion sco e*, mean ±SD 2.5 ±7.1 1.9 ±4.4 0.66 2.2 ±5.9 Na owing sco e*, mean ±SD 0.5 ±1.6 0.3 ±0.7 0.48 0.4 ±1.2 To al sco e, mean ±SD 3.0 ±8.3 2.1 ±4.7 0.62 2.6 ±6.7 S a is ical compa ison be ween he g oups was pe o med by - es , boo s ap- ype - es (5000 eplica ions), o chi-squa e es , when app op ia e. *Radiologic sco e by modi ied Sha p- an de Heijde sco e. h ps://doi.o g/10.1371/jou nal.pone.0183294. 001 YKL-40 as a bioma ke o disease ac i i y in heuma oid a h i is PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0183294 Augus 25, 2017 5 / 15 Du ing he i s ou weeks o ea men wi h he csDMARD combina ion, YKL-40 le els dec eased signi ican ly (p<0.001, Fig 3). This dec ease in YKL-40 le els showed a posi i e co - ela ion o he change in IL-6 ( = 0.38, 95% CI: 0.18, 0.54) as well as he change in MMP-3 le - els ( = 0.42, 95% CI: 0.23, 0.58) om baseline o ou weeks; whe eas no co ela ion o he change in DAS28 was ound. Du ing he ollowing 22 weeks o ea men wi h csDMARD ±in liximab, YKL-40 le els showed a mino u he dec ease (g oups combined: -5.0 ng/ml, 95%CI: -9.6 o -0.5, p = 0.031). Howe e , he e was no di e ence in YKL-40 le els be ween placebo and in liximab ea ed g oups o e ime (Fig 3). The esul oge he sugges ha dec ease in YKL-40 le els is ela ed o he an i heuma ic e ec o DMARDs, no o in liximab ea men pe se. To assess he associa ions o YKL-40 wi h disease ac i i y o e ime (du ing he 26 weeks o ea men ), a ea unde cu e (AUC) analysis adjus ed o age, gende and heuma oid ac o (RF) posi i i y was used. A signi ican co ela ion be ween DAS28 AUC 0-26 weeks and YKL-40 AUC 0-26 weeks ( = 0.41, p<0.001) was ound as shown in Fig 4A. Simila ly, du ing he 26 Fig 2. YKL-40 was associa ed wi h disease ac i i y, IL-6 and MMP-3 in he DMARD-naï e RA pa ien s. The igu e shows sca e plo s o YKL-40 le els wi h disease ac i i y measu ed as DAS28 (A), ende join coun (B), e y h ocy e sedimen a ion a e (C), IL-6 (D), and MMP-3 (E) a he baseline. Co ela ion coe icien s we e calcula ed by he Spea man me hod. YKL-40 showed posi i e co ela ion o DAS28, ende join coun , ESR, IL-6 and MMP- 3 a he baseline. h ps://doi.o g/10.1371/jou nal.pone.0183294.g002 YKL-40 as a bioma ke o disease ac i i y in heuma oid a h i is PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0183294 Augus 25, 2017 6 / 15 weeks’ ea men , YKL-40 AUC 0-26 weeks showed a signi ican posi i e co ela ion o ESR AUC 0-26 weeks ( = 0.56, p<0.001), PGA AUC 0-26 weeks ( = 0.23, p = 0.035) as well as o IL-6 AUC 0-26 weeks ( = 0.31, p = 0.004, Fig 4B). These esul s we e con i med in longi udinal gene - alized es ima ing equa ion model (GEE, adjus ed o ea men g oup as well as o age, gende and heuma oid ac o ) whe e YKL-40 showed a signi ican independen associa ion wi h Fig 3. YKL-40 le els dec eased du ing he in ensi e an i- heuma ic ea men wi h a combina ion o csDMARDs. The igu e shows he mean change in plasma YKL-40 le els in 88 pa ien s in he NEO-RACo -s udy ea ed wi h a combina ion o sulphasalazine, me ho exa e, hyd oxychlo oquine and low dose p ednisolone o he i s ou weeks, and he ea e andomized o ecei e ei he placebo in usions (Fin-RACo + Pla) o in liximab in usions (Fin-RACo + INFL) added on he csDMARD combina ion o ano he 22 weeks. YKL-40 le els dec eased signi ican ly (p<0.001) du ing he i s ou weeks o ea men wi h csDMARDs; when in liximab (o placebo) was added o he ea men he e was a mino u he dec ease (g oups combined, p = 0.031) bu no di e ence be ween placebo and in liximab ea men g oups was ound. The change is p esen ed as ng/ml, mean ±95% CI, n = 88. h ps://doi.o g/10.1371/jou nal.pone.0183294.g003 YKL-40 as a bioma ke o disease ac i i y in heuma oid a h i is PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0183294 Augus 25, 2017 7 / 15 Fig 4. YKL-40 was associa ed wi h disease ac i i y and in lamma ion du ing in ensi e ea men wi h DMARDs. The igu e shows quad a ic ela ionships o YKL-40 wi h he disease ac i i y measu ed as DAS28 (A) and wi h he in lamma o y cy okine IL-6 (B) du ing he 26 weeks o ea men . The pa ien s we e ea ed o he i s ou weeks wi h a combina ion o sulphasalazine, me ho exa e, hyd oxychlo oquine and low dose p ednisolone, and he ea e andomized o ecei e ei he placebo in usions o in liximab in usions added o YKL-40 as a bioma ke o disease ac i i y in heuma oid a h i is PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0183294 Augus 25, 2017 8 / 15 DAS28, and associa ed mos po en ly wi h ESR and PGA when he indi idual measu es o his a iable we e analyzed (Table 2). Discussion In he p esen s udy, we ound ha high YKL-40 le els we e associa ed wi h mo e se e e in lamma ion and wi h disease ac i i y in DMARD naï e pa ien s wi h ea ly RA, as well as in he same pa ien s du ing 26 weeks’ in ensi e an i- heuma ic ea men . YKL-40 is hus p o- posed as a po en ial bioma ke o disease ac i i y and in lamma ion no only in ea ly ea men naï e RA, bu also du ing ac i e d ug ea men aiming o emission. YKL-40 is p esen in RA join s and i s concen a ions in syno ial luid a e highe han hose in se um indica ing in a-a icula syn hesis o he glycop o ein [7,11]. Possible in a- a icula cell ypes p oducing YKL-40 a e ib oblas -like syno ial cells, mac ophages, neu o- phils and chond ocy es [1,11,27]. Hypome hyla ed DNA loci o he YKL-40 gene we e ecen ly de ec ed in syno ial cells om RA pa ien s and his epigene ic change was p esen ed o lead o he o e exp ession o he gene [28]. Also, pe iphe al blood de i ed CD16 posi i e mononu- clea cells wi h a mac ophage pheno ype in il a ing in o he syno ium a e ano he po en ial cellula sou ce o YKL-40 in RA [15]. In addi ion, lymphocy es may p oduce YKL-40: Recen s udy by Tanaka e al. using glucose-6-phospa ase isome ase immunized mice as a model o RA, showed o e exp ession o YKL-40 in splenic egula o y T cells [17]. The cellula e ec s o YKL-40 a e no , howe e , known e y well. I may be ha YKL-40 i sel is no he unc ional e ec o molecule bu a p ecu so . To suppo ha assump ion, YKL-40 de i ed pep ides ha e been shown o induce T-cell p oli e a ion in RA pa ien s [2]. An igen p esen ing cells (APCs) om RA syno ial luid we e epo ed o p ocess YKL-40 and p esen he decay pep ides o T-cells [6]. This was associa ed wi h he RA- ela ed HLA-DR4 pheno ype, and especially he pep ide 259–271 was ound o induce T-cell p oli e a ion [2,6,29]. Also, he immune esponse was shi ed owa ds he p oin lamma o y Th1 ype eac- ion by YKL-40 in pa ien s wi h RA, bu no in heal hy con ols [5]. Fu he , T-lymphocy es wi h RA-associa ed HLA-DR4 alleles we e epo ed o p oduce high amoun s o in e e on-γ and TNF-αwhen e-exposed o YKL-40 de i ed pep ides, while he p oduc ion o hese cy okines in T-cells wi h HLA-alleles non-associa ed o RA was e y low [29]. Mo eo e , he ea men o ano he 22 weeks. Measu emen s we e ca ied ou a weeks 0, 4, 10, 18, and 26 du ing he ea men and a ea unde he cu e analysis o e ime (AUC 0-26 weeks ) adjus ed o age, gende and RF posi i i y was used. G ay-shaded a eas ep esen 95% con idence in e als a ound he mean. YKL-40 AUC 0-26 weeks showed a s a is ically signi ican co ela ion o DAS28 AUC 0-26 weeks and IL-6 AUC 0-26 weeks . h ps://doi.o g/10.1371/jou nal.pone.0183294.g004 Table 2. Gene alized es ima ing equa ions models o he e ec o YKL40, ime and in e ac ion in measu es o disease ac i i y om baseline o 26 weeks. DAS28 TJC SJC ESR PGA B (95% CI) p- alue B (95% CI) p- alue B (95% CI) p- alue B (95% CI) p- alue B (95% CI) p- alue YKL-40 0.03 (0.02 o 0.04) <0.001 0.08 (0.04 o 0.12) <0.001 0.07 (0.03 o 0.10) <0.001 0.32 (0.20 o 0.44) <0.001 0.29 (0.15 o 0.43) <0.001 Time -0.10 (-0.12 o -0.08) <0.001 -0.17 (-0.25 o -0.09) <0.001 -0.16 (-0.22 o -0.10) <0.001 -0.31 (-0.52 o -0.10) 0.004 -0.84 (-1.17 o -0.51) <0.001 TimexYKL-40 0.24 <0.001 0.004 0.017 0.073 Models we e adjus ed using age, gende , heuma oid ac o posi i i y and ea men . DAS28—disease ac i i y sco e in 28 join s, TJC— ende join coun , SJC—swollen join coun , ESR—e y h ocy e sedimen a ion a e, PGA—pa ien ´s global assessmen . h ps://doi.o g/10.1371/jou nal.pone.0183294. 002 YKL-40 as a bioma ke o disease ac i i y in heuma oid a h i is PLOS ONE | h ps://doi.o g/10.1371/jou nal.pone.0183294 Augus 25, 2017 9 / 15