Association of Lymphangiogenic Factors With Pulmonary Arterial Hypertension in Systemic Sclerosis

Objective Pulmonary arterial hypertension (PAH) is a major complication in systemic sclerosis (SSc), a disease marked by vascular and lymphatic vessel abnormalities. This study was undertaken to assess the role of the lymphangiogenic factors vascular endothelial growth factor C (VEGF‐C) and angiopoietin 2 (Ang‐2) and the soluble forms of their respective cognate receptors, soluble VEGF receptor 3 (sVEGFR‐3) and soluble TIE‐2, in patients with SSc, and to evaluate their predictive ability as markers for PAH development in SSc. Methods In this cohort study, we used multiplex bead assays to assess serum levels of lymphangiogenic factors in 2 well‐characterized SSc cohorts: an unselected identification cohort of SSc patients from Oslo University Hospital (n = 371), and a PAH‐enriched validation cohort of SSc patients from Zurich University Hospital and Oslo University Hospital (n = 149). As controls for the identification and validation cohorts, we obtained serum samples from 100 healthy individuals and 68 healthy individuals, respectively. Patients in whom SSc‐related PAH was identified by right‐sided heart catheterization (RHC) in both cohorts were studied in prediction analyses. PAH was defined according to the European Society of Cardiology/European Respiratory Society 2015 guidelines for the diagnosis and treatment of PAH. Associations of serum levels of lymphangiogenic factors with the risk of PAH development were assessed in logistic regression and Cox regression analyses. Associations in Cox regression analyses were expressed as the hazard ratio (HR) with 95% confidence interval (95% CI). Results In the identification cohort, SSc patients had lower mean serum levels of VEGF‐C and higher mean serum levels of Ang‐2 compared to healthy controls (for VEGF‐C, mean ± SD 2.1 ± 0.5 ng/ml in patients versus 2.5 ± 0.4 ng/ml in controls; for Ang‐2, mean ± SD 6.1 ± 7.6 ng/ml in patients versus 2.8 ± 1.8 ng/ml in controls; each P < 0.001); these same trends were observed in SSc patients with PAH compared to those without PAH. The association of serum VEGF‐C levels with SSc‐PAH was confirmed in the PAH‐enriched RHC validation cohort. For prediction analyses, we assembled all 251 cases of SSc‐PAH identified by RHC from the identification and validation cohorts. In multivariable Cox regression analyses adjusted for age and sex, the mean serum levels of VEGF‐C and sVEGFR‐3 were predictive of PAH development in patients with SSc (for VEGF‐C, HR 0.53 [95% CI 0.29–0.97],