Background:The course of interstitial lung disease (ILD) varies considerably in patients with systemic sclerosis (SSc), and no biomarkers have been found to consistently predict ILD progression in this population. Treatment may affect how a candidate biomarker correlates with improvement/worsening of SSc-ILD. We hypothesized that specific proteins recovered from bronchoalveolar lavage (BAL) would differentially predict progression of SSc-ILD based on whether a patient was receiving ILD therapy.Objectives:(1) To assess the relationship between 68 unique BAL proteins measured in participants of Scleroderma Lung Study (SLS) I1 and changes in radiographic extent of SSc-ILD; (2) To determine if treatment affects whether a specific protein predicts improvement or worsening of SSc-ILD.Methods:Bronchoscopy was performed on 144 of the 158 participants in SLS I (Cyclophosphamide [CYC] vs. placebo) with 103 BAL samples available for analysis. BAL was lyophilized, concentrated 10X and used in a multiplex protein analysis of 68 distinct cytokines, chemokines and growth factors. Quantitative imaging analysis (QIA) was used to calculate the extent of radiographic fibrosis (QLF) in the whole lung using HRCT of the chest at baseline and 12 months. Multivariable linear regression models were created to determine the key BAL proteins associated with change in QLF scores using a backward selection process adjusting for treatment arm and ILD severity. The bootstrap procedure was employed for internal validation.Results:A number of BAL proteins were significantly associated with change in QLF scores at 12 months; however, the directionality of these associations was often based on the presence/absence of treatment. For example, increased levels of granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin (IL)-1, monocyte chemoattractant protein (MCP)-3, chemokine ligand (CCL)-5, transforming growth factor (TGF)-β, hepatocyte growth factor (HGF), stem cell factor (SCF), IL-4, TGF-α, were associated with worse QLF scores in patients who received placebo; whereas, increased levels of these same proteins were associated with improved QLF scores in patients who received CYC (Figure). Increased levels of Fractalkine were associated with worse in QLF scores, and increased levels of IL-7 were associated with improved QLF scores, regardless of treatment arm. In the multivariable model adjusting for treatment arm and baseline severity of ILD, IL-1, MCP-3, surfactant protein C, IL-7, and CCL-5 were independently associated with change in QLF scores.Figure 1.Example of a specific BAL protein (GM-CSF) that predicts worse QLF scores in patients receiving placebo (Group B, Red dotted line) and improved QLF scores in patients receiving CYC (Group A, Blue solid line). Shaded areas represent 95% confidence intervals.Conclusion:Proteins that mediate both inflammation and fibrosis differentially affected progression of SSc-ILD based on treatment status. Higher levels of certain proteins predicted worsening of ILD in patients receiving placebo, but improvement in patients receiving CYC. Measuring these proteins could help to identify patients who: (1) are at risk for ILD progression, and (2) may preferentially benefit from treatment with immunosuppression.References:[1]Tashkin DP, et al. NEJM 2006.Disclosure of Interests:Elizabeth Volkmann Consultant of: Boehringer Ingelheim, Grant/research support from: Corbus, Forbius, Donald Tashkin: None declared, Mei Leng: None declared, Ning Li: None declared, Grace Kim: None declared, Jonathan Goldin: None declared, Airi Harui: None declared, Michael Roth Grant/research support from: Genentech/Roche
Dynamic expression of transformating growth factor-β1 and caveolin-1 in the lung of Bleomycin-induced interstitial lung disease
