CURRENT OPINION Biomarkers in Axial Spondyloarthritis

REVIEW

CURRENT OPINION Biomarkers in axial spondyloarthritis

Walter P. Maksymowych

Purpose of review To assess the literature for biomarker validation studies that address key unmet needs related to the evaluation and management of patients with axial spondyloarthritis (SpA). This review focused on biomarkers facilitating early diagnosis and reflecting disease activity, structural damage on radiography, and clinical response to major therapies. Recent findings Early diagnosis may be facilitated by measurement of antibodies to the human leukocyte antigen class II- associated invariant chain peptide (anti-CD74) but sensitivity declines with increasing duration of disease. No disease activity biomarkers have demonstrated consistent superiority over standard C-reactive protein (CRP), and future validation should employ multivariate analysis aimed at demonstrating the added value of any associated biomarkers beyond available clinical parameters of disease activity and the use of magnetic resonance imaging inflammation as the primary endpoint. Several biomarkers reflecting inflammation (CRP and calprotectin), angiogenesis (vasoactive endothelial growth factor), and connective tissue turnover (C2 M, C3 M, and citrullinated metalloproteinase degraded fragment of vimentin ) have recently been shown to reflect radiographic progression in multivariate studies adjusted for baseline severity. Future studies should be prospective and demonstrate that predictive capacity adds to the information provided by known predictors such as CRP and baseline modified Stoke AS Spine Score. Calprotectin is a promising predictor of response to major therapies for axial SpA. Summary Several promising biomarkers addressing major unmet clinical needs require further validation in prospective studies. Keywords axial spondyloarthritis, biomarker, magnetic resonance imaging radiography, validation

INTRODUCTION solely on clinical evaluation is, therefore, fraught The identification and validation of soluble bio- with misinterpretation of the true nature of a markers represent a significant opportunity to patient’s symptoms and may lead to inappropriate address several unmet needs in the management treatment with anti-inflammatory agents. The third of patients with axial spondyloarthritis (axSpA). is assessment of prognostic risk, which is necessary The first is diagnosis prior to the unequivocal devel- to determine which patients with undifferentiated opment of radiographic sacroiliitis, especially as arthritis might develop SpA and which patients with the most sensitive imaging modality, magnetic SpA are destined to develop structural joint damage, resonance imaging (MRI) is costly and/or not especially ankylosis in the spine. Radiography lacks widely available. The second is the assessment and sensitivity to change and cannot be used for monitoring of disease activity in routine practice, monitoring in clinical practice. The fourth is assess- which is limited to patient self-reported measures ment of response to therapy and implementation of that lack correlation with MRI parameters of inflammation, physical examination findings, which are University of Alberta, Edmonton, Alberta, Canada primarily evident in peripheral joints, and C-reac- Correspondence to Walter P. Maksymowych, Professor of Medicine, tive protein (CRP), which is insensitive. Moreover, University of Alberta, 562 Heritage Medical Research Building, Edmon- symptoms of inflammatory back pain may be diffi- ton, AB T6G 2S2, Canada. Tel: +1 780 407 1573; fax: +780 407 6055; cult to distinguish from mechanical causes and the e-mail: [email protected] two conditions may occur simultaneously in the Curr Opin Rheumatol 2015, 27:343–348 same patient. Monitoring of disease activity based DOI:10.1097/BOR.0000000000000180

recruited patients with chronic back pain who KEY POINTS do not yet demonstrate definite radiographic CRP is currently the best-validated biomarker for sacroiliitis. disease activity and radiographic progression. Increasing evidence supports biomarkers reflecting DISEASE ACTIVITY BIOMARKERS inflammation and joint tissue turnover as predictors of Clinical validation of disease activity biomarkers structural damage. suffers from the lack of availability of an appropriate Clinical validation should be undertaken in both gold standard. A consensus position document prospective cohorts and clinical trials using multivariate recommends that a major treatment target should longitudinal analyses of sufficient duration to account be clinical remission as defined by the absence of for the slow progression and variability of disease. clinical and inflammatory evidence of significant MRI inflammation is currently the most appropriate inflammatory disease, and that disease activity endpoint for validation of biomarkers reflecting should be measured using clinical signs and symp- disease activity. toms and acute phase reactants [3&]. However, a Predictive capacity of biomarkers for structural major drawback to this approach is that no clinical progression may be more readily apparent when parameters have been shown to reflect the histopa- combined with known predictors, for example thology of inflammation at the site of disease. More- baseline mSASSS. over, there is only a weak correlation between clinical and MRI parameters of inflammation, primarily in early SpA, and inflammation may still be apparent on MRI of the sacroiliac joints and/or spine a personalized treatment strategy. This review aims even in the presence of clinical remission [4–6]. On at summarizing the recent literature that addresses the contrary, MRI parameters of inflammation in clinical validation of diagnostic, disease activity, the sacroiliac joint have been shown to correlate prognostic, and theragnostic biomarkers, and the with histopathological features of inflammation results of biomarker studies that have employed and also predict the development of radiographic these strategies in patients with axial SpA. sacroiliitis [7–9]. A similar relationship has been described between MRI inflammation in the spine and new bone formation [10–13]. Consequently, DIAGNOSTIC BIOMARKERS since MRI inflammation represents our current best Human leukocyte antigen (HLA) B27 is the primary measure of disease activity, clinical validation of biomarker used to identify patients with axial biomarkers reflecting disease activity should be con- SpA with sensitivity of 60–95% and specificity of ducted using MRI as the gold standard. CRP is an 80–90%. One report described a potential role of example of a biomarker that correlates with MRI antibodies to the HLA class II-associated invariant inflammation albeit moderately [14,15]. chain peptide (anti-CD74 antibodies), which were Most of the studies [16–20] assessing disease detected in 85.1% of axSpA (n ¼ 94) versus 7.8% of activity biomarkers have been cross-sectional and non-SpA (n ¼ 51) patients [1&]. Among non-SpA used correlation analysis with acute phase reactants, patients, positivity for anti-CD74 was observed in the Bath AS Disease Activity Index, or the AS Disease two with degenerative spinal disease, one with Activity Score rather than MRI. A major limitation of rheumatoid arthritis, and one with psoriatic arthri- such biomarker analyses is that many biomarkers tis. In a subsequent report analyzing 156 patients correlate with each other, for example CRP with with axial SpA, sensitivity was somewhat less at 69% interleukin-6, and therefore do not indicate what [2&]. However, sensitivity was 97% in patients with is the independent information provided by the duration of symptoms less than 1 year. The rate of biomarker. Instead, studies should employ multi- positivity declined after the first year. In controls, variate longitudinal analysis to determine to what the prevalence of anti-CD74 was 18/40 (45%) in degree biomarkers contribute independently to psoriatic arthritis without axial involvement, 9/80 assessment of disease activity, as assessed by MRI, (11%) in rheumatoid arthritis, 6/40 (15%) in beyond standard clinical and laboratory assess- systemic lupus erythematosus, 1/40 (2.5%) in HIV, ments. There have been no recent studies that have and 1/125 (0.8%) in blood donors. Consequently, employed such a study design. Previous analyses diagnostic potential of this biomarker should be using MRI as gold standard have shown no corre- regarded as preliminary. Addressing the unmet lations between baseline to 12-week change in clinical need will require biomarker analyses in metalloproteinase 3 (MMP3), type II collagen sera obtained from prospective studies that have C-telopeptide, or type I collagen N-telopeptide

344 www.co-rheumatology.com Volume 27 Number 4 July 2015 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. Biomarkers in axial spondyloarthritis Maksymowych and change in MRI inflammation [21], whereas a CRP in multivariate analyses adjusted for the single multivariate analysis revealed an independ- severity of radiographic damage at baseline assessed ent although weak association of baseline to 2-week using the modified Stoke AS Spine Score (mSASSS). change in interleukin6 (IL6) with baseline to In the German SPondyloarthritis Inception Cohort, 24-week change in MRI inflammation [22], casting time-averaged CRP, defined as mean CRP level more doubt as to the role of this mediator in the patho- than 6 mg/l over five time points including baseline physiology of inflammation. and four follow-up visits every 6 months thereafter, In conclusion, there is no disease activity bio- was independently associated with radiographic marker that has proven superiority over CRP. Future progression [26]. Current smoking was also inde- studies should focus on multiplex assays that pro- pendently associated with progression and a sub- vide an opportunity to simultaneously analyze a sequent report demonstrated higher CRP levels large number of biomarkers using MRI inflam- according to the degree of smoking [27]. Acute mation as the primary endpoint. Of particular phase reactants were also significantly associated importance, multivariate analysis should demon- with radiographic progression in posthoc multivari- strate the added value of any associated biomarkers ate analysis of a randomized trial comparing beyond available clinical parameters of disease continuous and on-demand nonsteroidal anti- activity. inflammatory agent treatment [28]. In a prospective study [29&&] over 12 years with radiography conducted every 2 years, the relationship between dis- BIOMARKERS OF STRUCTURAL DAMAGE ease activity measures and radiographic damage was A retrospective cohort analysis of patients with AS investigated using longitudinal, autoregressive showed that baseline clinical parameters, including models with 2-year time lags. CRP was associated B27 and CRP, did not predict radiographic pro- with radiographic progression and an increase in gression in the spine and that only the presence 10 mg/l was associated with an increase in 0.2 of syndesmophytes at baseline was significantly mSASSS units in the subsequent 2 years. This predictive [23]. Longitudinal analysis over 12 years analysis also clearly established a link between dis- in axial SpA also revealed that the predominant ease activity and development of structural damage, factor associated with radiographic progression in especially in patients with shorter duration of symp- the spine, which is primarily associated with long- toms. Radiographic progression was still evident term functional impairment, is the presence of even in patients with normal clinical measures of radiographic new bone at baseline [24&&]. However, disease activity, suggesting either mechanism lead- this analysis also indicated the importance of B27 as ing to new bones that are independent of inflam- a predictor of progression, especially in men. In a mation and/or pathways of inflammation that are retrospective casecontrol study [25&&] of 1,537 not captured by existing clinical measures. As noted cases, almost 500 single nucleotide polymorphisms previously, currently used clinical measures demon- (SNPs) in genes involved in anabolic or catabolic strate weak correlations with MRI parameters of bone pathways were analyzed for associations with inflammation. radiographic severity. B27 was not associated with Limited analyses of biomarkers that might pre- radiographic severity. Two SNPs, one in the receptor dict radiographic progression conducted several activator of nuclear factor k B gene and one in the years ago implicated biomarkers involved in joint prostaglandin-endoperoxide synthase 1 gene, were tissue degradation, such as MMP3 [30], and regula- associated with radiographic severity. An important tors of bone remodeling, such as sclerostin [31] and observation of the 12-year longitudinal evaluation dickkopf-1 [32]. More recent studies [33,34] have of radiographic progression in axial SpA is that assessed biomarkers that promote angiogenesis, periods of steep progression and relative quiescence such as vasoactive endothelial growth factor may alternate, and this process of oscillation may (VEGF), which is also required for effective linking occur both in ‘early AS patients’ as well as in of angiogenesis and osteogenesis. Serum levels are ‘advanced’ patients with ages greatly more than elevated in and correlate with disease activity 60 years. This reinforces the importance of longi- in SpA, including CRP [35,36]. Baseline level of tudinal evaluation for clinical validation of candi- VEGF was associated with 2-year radiographic pro- date biomarkers reflecting structural damages that gression after adjustment for the presence of radio- are modifiable. graphic damage at baseline (syndesmophytes), Cross-sectional data had previously questioned elevated CRP, smoking status, sex, HLA-B27 status, the predictive value of CRP for structural pro- and treatment [37&]. The risk for progression was gression [23]. However, several prospective studies greater when baseline level of VEGF was combined have now demonstrated the predictive capacity of with either baseline CRP or the presence of a

1040-8711 Copyright ß 2015 Wolters Kluwer Health, Inc. All rights reserved. www.co-rheumatology.com 345 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. Spondyloarthropathies syndesmophyte at baseline. Calprotectin is a heter- mSASSS. These data are consistent with several lines odimer consisting of S100A8 and S100A9 (pre- of data supporting a role for metalloproteinases [30], viously called myeloid-related protein 8 and 14, smoking, and ongoing connective tissue turnover respectively), and is expressed during monocyte associated with posttranslational modification of infiltration into inflamed tissues, including macro- proteins as key factors in the development of new phage infiltration in SpA synovitis. Serum levels are bone in axial SpA. Moreover, they emphasize the elevated in axial SpA and are decreased following advantage of biomarker technologies that focus tumor necrosis factor alpha (TNFa) inhibitor on fingerprinting technologies aimed at proteins therapy associated with a rapid decrease in calpro- relevant to disease, rather than just on measurement tectin-positive cells in the inflamed synovium of proteins in their healthy conformation. [38,39]. In a preliminary report, it correlated signifi- In conclusion, increasing evidence supports bio- cantly with VEGF and was associated with radio- markers reflecting inflammation and joint tissue graphic progression after adjusting for VEGF [40]. turnover as predictors of structural damage. Future A new approach has been taken with the devel- studies should be prospective and employ multi- opment of biomarker assays that detect de-novo variate longitudinal analyses to demonstrate that epitopes generated by enzymatic cleavage of joint predictive capacity adds to the information pro- tissue proteins. Two examples that have been vided by known predictors such as CRP and baseline assessed in AS are C2 M and C3 M, which are metal- mSASSS. Moreover, because at least 2 years’ follow- loproteinase-generated neoepitopes of type II and up is required before progression can be reliably type III collagen, respectively. Both were elevated in detected, biomarker validation studies should also AS patients compared with healthy controls and aim at demonstrating that an early change in bio- C3 M correlated significantly with radiographic marker after treatment predicts long-term change in damage score [41]. Used in combination, high progression as recently formulated by the Outcome serum level of C2 M and C3 M could predict 80% Measures in Rheumatology Soluble Biomarker Inter- of patients with radiographic progression and 61% national Working Group [45]. An example of an of those who did not progress over a 2-year follow- appropriately validated biomarker would be C-telo- up. Vimentin is a type III intermediate filament peptide of type I collagen (CTX-I). Early decrease in protein that is expressed by various cells as an CTX-I after treatment with antiosteoporotic agents important part of the cytoskeleton. It is secreted predicts long-term change in bone mineral density by activated macrophages, which account for its (BMD) and fracture risk, early change in CTX-I presence in the extracellular matrix, and it is also reflects dose of treatment, and combined with a well-established target for intracellular citrullina- BMD, age, and/or prior fracture rates, it contributes tion. Citrullinated vimentin fragments have been to improvement in the prediction of fracture reported in synovial tissue from both RA and SpA [46,47]. patients [42]. A biomarker assay that detects a neo- epitope derived from a citrullinated metalloproteinase degraded fragment of vimentin (VICM) has THERAGNOSTIC BIOMARKERS been developed and one report demonstrated that CRP is currently the best available biomarker that VICM was elevated in patients with liver fibrosis, predicts response to anti-TNF therapies and a indicating that VICM could be a marker of local previous report showed that high sensitivity CRP inflammation [43]. A recent report has demon- (hs-CRP) offered no advantage [48]. A recent study strated elevated levels of VICM in patients with [49&] compared hs-CRP, IL-6, pentraxin-3, a-2-mac- AS which correlated with CRP and spinal radio- roglobulin,MMP-3,calprotectin,andVEGFin graphic damage (mSASSS) [44]. VICM was signifi- healthy controls and SpA patients before and after cantly associated with mSASSS progression after 2 2 weeks of infliximab or placebo and results were years in a multivariate model that included baseline validated in an additional AS cohort receiving mSASSS. Patients with both high VICM and high infliximab. Responsivenesswashighestforcalpro- baseline mSASSS had significantly higher odds of tectin and its use would have required a smaller developing radiographic progression (odds ratio sample size to show treatment group differences at 2 (OR) ¼ 13 for mSASSS progression over 2 years; weeks compared with hs-CRP. In a recent proof-of- OR ¼ 32 for new syndesmophyte) compared with concept study [50] with the monoclonal anti-IL17A patients with both low baseline mSASSS and low antibody, secukinumab, baseline level of calprotec- VICM levels. Of the patients with both high VICM tin predicted response to therapy and the change and high mSASSS, 67% had radiographic pro- from baseline to week 6 in calprotectin, but not gression after 2 years, compared with only 13% of CRP, also correlated with clinical response at those with low VICM levels but high baseline week 6.

4. Machado P, Landewe RB, Braun J, et al. MRI inflammation and its relation with CONCLUSION measures of clinical disease activity and different treatment responses in Advances in outcome assessment and the availabil- patients with ankylosing spondylitis treated with a tumour necrosis factor inhibitor. Ann Rheum Dis 2012; 71:2002–2005. ity of major new therapies for axial SpA have under- 5. Weis A, Song IH, Haibel H, et al. Good correlation between changes in lined the unmet clinical need in evaluating patients objective and subjective signs of inflammation in patients with short- but not long duration of axial spondyloarthritis treated with tumor necrosis factor- with this condition. Several promising biomarkers blockers. Arthritis Res Ther 2014; 16:R35. addressing early diagnosis, the assessment of 6. Maksymowych WP, van der Heijde D, et al. Relationship between MRI and clinical remission patients with nonradiographic axial spondyloarthritis after disease activity, prediction of radiographic pro- two years of adalimumab therapy. J Rheumatol 2014; 41:1496. gression, and responsiveness to major therapies 7. Gong Y, Zheng N, Chen S, et al. Ten years experience on needle biopsy in the early diagnosis of sacroiliitis. Arthritis Rheum 2012; 64:1399–1406. have been identified in preliminary studies and 8. Bollow M, Fischer T, Reisshauer H, et al. Quantitative analyses of sacroiliac now require further clinical validation in large biopsies in spondyloarthropathies: T cells and macrophages predominate in early and active sacroiliitis- cellularity correlates with the degree of enhance- prospective cohorts and clinical trials. This needs ment detected by magnetic resonance imaging. Ann Rheum Dis 2000; to be primarily conducted using multivariate longi- 59:135–140. 9. Bennett AN, McGonagle D, O’Connor P, et al. Severity of baseline magnetic tudinal analyses that will inform the clinician to resonance imaging-evident sacroiliitis and HLA-B27 status in early inflam- what degree candidate biomarkers enhance infor- matory back pain predict radiographically evident ankylosing spondylitis at eight years. Arthritis Rheum 2008; 58:3413–3418. mation provided by currently available outcome 10. Appel H, Loddenkemper C, Grozdanovic Z, et al. Correlation of histopatho- parameters. The relatively slow progression and logical findings and magnetic resonance imaging in the spine of patients with ankylosing spondylitis. Arthritis Res Ther 2006; 8:R143. variability of disease over time will require long- 11. Baraliakos X, Listing J, Rudwaleit M, et al. The relationship between inflam- term data and collaborative networks of clinical mation and new bone formation in patients with ankylosing spondylitis. Arthritis Res Ther 2008; 10:R104. investigators together with platform technologies 12. Maksymowych WP, Chiowchanwisawakit P, Clare T, et al. Inflammatory that will facilitate the assessment of a large array of lesions of the spine on magnetic resonance imaging predict the development of new syndesmophytes in ankylosing spondylitis: evidence of a relationship biomarkers simultaneously. between inflammation and new bone formation. Arthritis Rheum 2009; 60:93–102. 13. van der Heijde D, Machado P, Braun J, et al. MRI inflammation at the vertebral Acknowledgements unit only marginally predicts new syndesmophyte formation: a multilevel analysis in patients with ankylosing spondylitis. Ann Rheum Dis 2012; None. 71:369–373. 14. Lambert RGW, Salonen D, Rahman P, et al. Adalimumab significantly reduces both spinal and sacroiliac joint inflammation in patients with ankylosing Financial support and sponsorship spondylitis. Arthritis Rheum 2007; 56:4005–4014. None. 15. Maksymowych WP, Salonen D, Inman RD, et al. Low-dose infliximab (3 mg/ kg) significantly reduces spinal inflammation on magnetic resonance imaging in patients with ankylosing spondylitis: a randomized placebo-controlled Conflicts of interest study. J Rheumatol 2010; 37:1728–1734. 16. Toussirot E, Dumoulin G, Saas P, et al. Increased tartrate-resistant acid There are no conflicts of interest. phosphatase serum levels in ankylosing spondylitis and relationship with the inflammatory process. Ann Rheum Dis 2008; 67:430–431. 17. Bal A, Unlu E, Bahar G, et al. Comparison of serum IL-1 beta, sIL-2R, IL-6, and TNF-alpha levels with disease activity parameters in ankylosing spondylitis. REFERENCES AND RECOMMENDED Clin Rheumatol 2007; 26:211–215. 18. Keller C, Webb A, Davis J. Cytokines in the seronegative spondyloarthro- READING pathies and their modification by TNF blockade: a brief report and literature Papers of particular interest, published within the annual period of review, have review. Ann Rheum Dis 2003; 62:1128–1132. been highlighted as: 19. Toussirot E, Saas P, Deschamps M, et al. Increased production of soluble & of special interest CTLA-4 in patients with spondyloarthropathies correlates with disease activ- && of outstanding interest ity. Arthritis Res Ther 2009; 11:R101. 20. Melis L, Vandooren B, Kruithof E, et al. Systemic levels of IL-23 are strongly 1. Baraliakos X, Baerlecken N, Witte T, et al. High prevalence of anti-CD74 associated with disease activity in rheumatoid arthritis but not spondyloar- & antibodies specific for the HLA class II-associated invariant chain peptide thritis. Ann Rheum Dis 2010; 69:618–623. (CLIP) in patients with axial spondyloarthritis. Ann Rheum Dis 2014; 21. Maksymowych WP, Rahman P, Shojania K, et al. Beneficial effects of 73:1079–1082. adalimumab on biomarkers reflecting structural damage in patients with This study demonstrated diagnostic capacity for antibodies to the HLA class II- ankylosing spondylitis. J Rheumatol 2008; 35:2030–2037. associated invariant chain peptide (CLIP) (anti-CD74 antibodies) in 94 patients 22. Visvanathan S, Wagner C, Marini JC, et al. Inflammatory biomarkers, with axial SpA. The control group was limited to 51 patients (13 with rheumatoid disease activity and spinal disease measures in patients with ankylosing arthritis, 17 with fibromyalgia and 17 with degenerative spinal disease, three with spondylitis after treatment with infliximab. Ann Rheum Dis 2008; 67:511– psoriatic arthritis without axial involvement, and one with polymyalgia rheumatic). 517. 2. Baerlecken NT, Nothdorft S, Stummvoll GH, et al. Autoantibodies against 23. Baraliakos X, Listing J, von der Recke A, et al. The natural course of radio- & CD74 in spondyloarthritis. Ann Rheum Dis 2014; 73:1211–1214. graphic progression in ankylosing spondylitis: evidence for major individual This study demonstrated that sensitivity of anti-CD74 antibodies may be as high as variations in a large proportion of patients. J Rheumatol 2009; 36:997– 97% in SpA patients with less than 1-year’s disease duration although levels 1002. decline over time so that sensitivity in the overall SpA population is only 67%. In 24. Ramiro S, Stolwijk C, van Tubergen A, et al. Evolution of radiographic damage addition, this biomarker was present in 45% of psoriatic arthritis patients without && in ankylosing spondylitis: a 12 year prospective follow-up of the OASIS study. spinal involvement and its presence in patients with mechanical spinal disorders is Ann Rheum Dis 2015; 74:52–59. unknown. It will require further study in sera obtained from prospective studies that In this 12-year prospective cohort analysis, patients with AS were followed have recruited patients with chronic back pain who do not yet demonstrate definite prospectively with 2 yearly radiographs. This longitudinal analysis revealed that radiographic sacroiliitis. at the group level radiographic progression follows an approximately linear course, 3. Smolen JS, Braun J, Dougados M, et al. Treating spondyloarthritis, including whereas observation of progression in individuals revealed that periods of steep & ankylosing spondylitis and psoriatic arthritis, to target: recommendations of an progression and relative quiescence may alternate, and this process of oscillation international task force. Ann Rheum Dis 2014; 73:6–16. may occur both in ‘early AS patients’ as well as in ‘advanced’ patients with ages This study presents recommendations for treat-to-target strategies in axial SpA. A greatly more than 60 years. B27, especially in men, was a significant predictor of major treatment target should be clinical remission as defined by the absence of progression. This analysis also reinforces the importance of longitudinal evaluation clinical and laboratory evidence of significant inflammatory disease activity as for clinical validation of candidate biomarkers reflecting structural damage that are measured by clinical signs and symptoms and acute phase reactants. modifiable.

25. Cortes A, Maksymowych WP, Wordsworth BP, et al. Association study of genes 38. De Rycke L, Baeten D, Foell D, et al. Differential expression and response to && related to bone formation and resorption and the extent of radiographic change in anti-TNFalpha treatment of infiltrating versus resident tissue macrophage ankylosing spondylitis. Ann Rheum Dis 2014; Online First, published on March 20. subsets in autoimmune arthritis. J Pathol 2005; 206:17–27. This retrospective casecontrol analysis of 500 SNPs in 1,537 AS cases revealed 39. Kruithof E, De Rycke L, Vandooren B, et al. OMERACT Special Interest Group that two SNPs, one in RANK and one in PTGS1, were associated with radio- on synovial analysis in clinical trials: identification of synovial biomarkers of graphic progression in AS. B27 was not a predictor. response to experimental treatment in early-phase clinical trials in spondylar- 26. Poddubnyy D, Haibel H, Listing J, et al. Baseline radiographic damage, elevated thritis. Arthritis Rheum 2006; 54:1795–1804. acute-phase reactant levels, and cigarette smoking status predict spinal radiographic 40. Turina MC, Sieper J, Yeremenko N, et al. Calprotectin serum level is an progression in early axial spondylarthritis. Arthritis Rheum 2012; 64:1388–1398. independent marker for radiographic spinal progression in axial spondyloar- 27. Poddubnyy D, Haibel H, Listing J, et al. Cigarette smoking has a dose thritis. Ann Rheum Dis 2014; 73:1746–1748. dependent impact on progression of structural damage in the spine in patients 41. Bay-Jensen AC, Wichuk S, Byrjalsen I, et al. Circulating protein fragments of with axial spondyloarthritis: results from the German SPondyloarthritis Incep- cartilage and connective tissue degradation are diagnostic and prognostic tion Cohort (GESPIC). Ann Rheum Dis 2013; 72:1430–1432. markers of rheumatoid arthritis and ankylosing spondylitis. PLoS One 2013; 28. Kroon F, Landew R, Dougados M, et al. Continuous NSAID use reverts the 8:e54504. effects of inflammation on radiographic progression in patients with ankylos- 42. Tilleman K, van Steendam K, Cantaert T, et al. Synovial detection ing spondylitis. Ann Rheum Dis 2012; 71:1623–1629. and autoantibody reactivity of processed citrullinated isoforms of vimentin 29. Ramiro S, van der Heijde D, van Tubergen A, et al. Higher disease activity in inflammatory arthritides. Rheumatology (Oxford) 2008; 47:597– && leads to more structural damage in the spine in ankylosing spondylitis: 12-year 604. longitudinal data from the OASIS cohort. Ann Rheum Dis 2014; Online First, 43. Vassiliadis E, Oliveira CP, Alvares-da-Silva MR, et al. Circulating levels of published on May 7. citrullinated and MMP-degraded vimentin (VICM) in liver fibrosis related The relationship between disease activity measures and radiographic damage was pathology. Am J Transl Res 2012; 4:403–414. investigated using longitudinal, autoregressive models with 2-year time lags in this 44. Bay-Jensen AC, Karsdal MA, Vassiliadis E, et al. Circulating citrullinated 12-year prospective study with radiographs conducted every 2 years. CRP was vimentin fragments reflect disease burden in ankylosing spondylitis and have independently associated with radiographic progression. prognostic capacity for radiographic progression. Arthritis Rheum 2013; 30. Maksymowych WP, Landewe R, Conner-Spady B, et al. Serum matrix metallo- 65:972–980. proteinase 3 is an independent predictor of structural damage progression in 45. Maksymowych WP, Landewe R, Tak PP, et al. Reappraisal of OMERACT 8 patients with ankylosing spondylitis. Arthritis Rheum 2007; 56:1846–1853. draft validation criteria for a soluble biomarker reflecting structural damage 31. Appel H, Ruiz-Heiland G, Listing J, et al. Altered skeletal expression of endpoints in rheumatoid arthritis, psoriatic arthritis, and spondyloarthritis: the sclerostin and its link to radiographic progression in ankylosing spondylitis. OMERACT 9 v2 criteria. J Rheumatol 2009; 36:1785–1791. Arthritis Rheum 2009; 60:3257–3262. 46. Ravn P, Clemmesen B, Christiansen C. Biochemical markers can predict the 32. Heiland GR, Appel H, Poddubnyy D, et al. High level of functional dickkopf-1 response in bone mass during alendronate treatment in early postmenopausal predicts protection from syndesmophyte formation in patients with ankylosing women. Alendronate Osteoporosis Prevention Study Group. Bone 1999; spondylitis. Ann Rheum Dis 2012; 71:572–574. 24:237–244. 33. Clarkin CE, Gerstenfeld LC. VEGF and bone cell signalling: an essential 47. Garnero P, Hausherr E, Chapuy MC, et al. Markers of bone resorption predict vessel for communication? Cell Biochem Funct 2013; 31:1–11. hip fracture in elderly women: the EPIDOS Prospective Study. J Bone Miner 34. Gerber HP, Vu TH, Ryan AM, et al. VEGF couples hypertrophic cartilage Res 1996; 11:1531–1538. remodeling, ossification and angiogenesis during endochondral bone forma- 48. de Vries MK, van Eijk IC, van der Horst-bruinsma I, et al. Erythrocyte tion. Nat Med 1999; 5:623–628. sedimentation rate, C-reactive protein level, and serum amyloid A 35. Drouart M, Saas P, Billot M, et al. High serum vascular endothelial growth protein for patient selection and monitoring of anti–tumor necrosis factor correlates with disease activity of spondylarthropathies. Clin Exp factor treatment in ankylosing spondylitis. Arthritis Care Res 2009; 61: Immunol 2003; 132:158–162. 1484–1490. 36. Pedersen SJ, Sorensen IJ, Garnero P, et al. ASDAS, BASDAI and different 49. Turina M, Yeremenko N, Paramarta J, et al. Calprotectin (S100A8/9) as serum treatment responses and their relation to biomarkers of inflammation, cartilage & biomarker for clinical response in proof-of-concept trials in axial and peripheral and bone turnover in patients with axial spondyloarthritis treated with TNF spondyloarthritis. Arthritis Res Ther 2014; 16:413. alpha inhibitors. Ann Rheum Dis 2011; 70:1375–1381. Several biomarkers (hs-CRP, IL-6, pentraxin-3, a-2-macroglobulin, MMP-3, cal- 37. Poddubnyy DA, Conrad K, Haibel H, et al. Elevated serum level of the vascular protectin, and VEGF) were assessed at baseline for their ability to predict response & endothelial growth factor predicts radiographic spinal progression in patients to anti-TNF therapy. Responsiveness was highest for calprotectin and its use with axial spondyloarthritis. Ann Rheum Dis 2014; 73:2137–2143. would have required a smaller sample size to show treatment group differences at This preliminary report of 172 patients from the GESPIC cohort of patients with 2 weeks compared with hs-CRP. SpA shows in multivariate analysis that VEGF independently predicts radiographic 50. Baeten D, Baraliakos X, Braun J, et al. Antiinterleukin-17A mono- progression in the spine of patients with axial SpA. The risk is greater when high clonal antibody secukinumab in treatment of ankylosing spondylitis: a baseline level of VEGF combines with high baseline CRP or the presence of randomised, double-blind, placebo-controlled trial. Lancet 2013; 382: syndesmophyte at baseline. 1705–1713.