B79-16 Endothelial Loss of GDF15 Protects Against Bleomycin-Induced Fibrosis in Mice

Abstract Rationale GDF15 is elevated across a spectrum of cardiopulmonary diseases. Plasma levels of GDF15 correlate with mortality and inversely correlate with parameters of normal function (e.g. diffusing capacity). We sought to clarify the significance and role of GDF15 in pulmonary fibrosis, which is often complicated by secondary pulmonary hypertension. This inquiry may guide future studies into the molecular activity and biomarker potential of GDF15. Methods Subjects were recruited from the Simmons Center for ILD at the University of Pittsburgh Medical Center (UPMC) between 1/2003 through 10/2016. Specific diagnostics (e.g. ILD subtype) were performed according to standard criteria and current ATS guidelines. Control subjects without any history of lung disease were randomly recruited at UPMC. Plasma GDF15 levels were measured by quantitative ELISA. Control versus disease was analyzed with a two-sample independent t-test. Pulmonary function testing was acquired and separated into disparate severity groupings by standard criteria according to forced expiratory volume in one second (FEV1). GDF15 levels were analyzed across five groups via a Kruskal-Wallis with a post-hoc Dunn’s test. Sample control and diseased lung tissue sections were stained with antibodies against smooth muscle actin (SMA), platelet endothelial cell adhesion molecule (PECAM-1), and GDF15; histologic sections were analyzed for medial thickening to assess for vascular remodeling. Results The median age was 51 and 61 years, and the constitution was 48% and 42% female, for the control and ILD groups, respectively. The 299 case subjects (disease attributed to IPF, hypersensitivity pneumonitis, autoimmune ILD, anti-synthetase syndrome (AS), dermatomyositis/polymyositis, rheumatoid arthritis-associated ILD, or systemic sclerosis) collectively had a mean plasma GDF15 of 1692 pg/mL (standard deviation SD 1107 pg/mL), compared to controls (n = 72) with a mean GDF15 of 446 pg/mL (SD 366 pg/mL), p 0.001. There was no significant difference in GDF15 levels when compared across fibrosis severity (as represented by FEV1), p = 0.41. Increased GDF15 was significantly correlated with surrogate clinical measures of increased right heart strain, including elevated levels of brain natriuretic peptide, pulmonary arterial systolic pressure, and tricuspid regurgitant velocity. Tissue from patients with pulmonary fibrosis demonstrated evidence of increased vascular remodeling (SMA positivity 2.12% of region of interest) when compared with controls (0.88%). Conclusions These data suggest that GDF15 may act as a biomarker for pulmonary fibrosis complicated by vascular sequelae, such as pulmonary hypertension or vascular remodeling. Furthermore, these data suggest that GDF15 may have an active role in the evolution of vascular sequelae in pulmonary fibrotic disease. This abstract is funded by: 5T32HL007563