OP0256 IN MEDIUM TO LONG-TERM JUVENILE DERMATOMYOSITIS, PROINFLAMMATORY AND PROFIBROTIC CYTOKINES ARE DIFFERENTLY ASSOCIATED WITH PULMONARY INVOLVEMENT IN ACTIVE VS INACTIVE DISEASE

Abstract

Background Knowledge about associations between cytokines and lung involvement in juvenile dermatomyositis (JDM) is scarce. Objectives To examine associations between cytokines and pulmonary involvement in JDM assessed after medium to long-term follow-up. Methods 58 JDM patients examined median 17y after disease onset were stratified in active and inactive disease by the updated PRINTO criteria (Ref 1). Serum levels of cytokines were analyzed by Luminex or ELISA. Pulmonary function tests (PFT) included forced vital capacity (FVC), total lung capacity (TLC) and diffusing capacity for carbon monoxide (DLCO) (hgb adjusted). PFT variables are expressed as% of predicted. High resolution computed tomography (HRCT) scans were scored for interstitial lung disease (ILD), calcinosis in chest wall and airway disease. Associations between lung involvement and cytokines are presented as Spearman’s correlation coefficient (rs). Results No significant differences were found between active and inactive patients in tbl.?1. Associations between cytokines and PFT/HRCT findings Numbers are rs, all p’s In patients total, TGFb1 and PDGF correlated with TLC and FVC (- 0.38 - -0.30). For TGFb1 correlations with TLC and FVC were - 0.62 and -0.59 in the inactive group, whereas PDGF correlated with FVC in the active group (- 0.48). IP-10 correlated with TLC in active patients only (- 0.35). In patients total and active patients, eotaxin correlated with DLCO% (0.39 and 0.52). In inactive patients, MIP1b correlated with DLCO% (0.47). In patients total, IP-10 correlated with any HRCT finding and airways disease (0.37 and 0.44); which were also present in active patients (0.34 and 0.61). In patients total, MCP-1 correlated with any HRCT finding and calcinosis (0.34 and 0.40); in active patients MCP-1 and calcinosis correlated (0.34). Eotaxin correlated with ILD in both patients total and active patients (0.37 and 0.38). The following cytokines correlated with HRCT findings in inactive patients only; IL-17 with any HRCT findings and ILD (0.61 and 0.52), MIP1a with any HRCT findings (0.63) and IL-5 with ILD (-0.42). Conclusion In JDM, lung involvement was associated with mainly proinflammatory cytokines in active and profibrotic/proinflammatory cytokines in inactive patients. The association between higher eotaxin and better gas diffusion was interesting, since we have previously demonstrated an association between higher eotaxin and cardiac dysfunction (Ref 2).