Multiple serum biomarkers associate with mortality and interstitial lung disease progression in systemic sclerosis.

Abstract

To investigate the prognostic utility of 28 serum biomarkers in systemic sclerosis (SSc), SSc-associated interstitial lung disease (SSc-ILD) and clinically relevant disease subgroups. Participants with sera, high-resolution CT and lung function within 12 months of baseline were identified from the Australian Scleroderma Cohort Study. Baseline was the time of serum collection. Twenty-seven of the prespecified 28 serum biomarkers were analysed and biomarker associations with mortality and ILD progression were investigated in univariable and multivariable analyses, including within disease subgroups and combined with established risk factors for poorer prognosis in SSc. A total of 407 participants were identified, 252 (61.9%) with SSc-ILD. The median (interquartile range) follow-up after biomarker measurement was 6.31 (3.11-9.22) years. Sixteen biomarkers were associated with increased mortality. High levels of VCAM-1 were most strongly associated with mortality [hazard ratio (HR) 3.55; 95% CI 2.37-5.33; P < 0.001]. Five additional biomarkers had an HR >2: SP-D (2.28, 1.57-3.31; P < 0.001), E-selectin (2.19, 1.53-3.14; P < 0.001), IL-6 (2.15, 1.50-3.09; P < 0.001), MMP-3 (2.05, 1.42-2.95; P < 0.001) and ET-1 (2.03, 1.40-2.92; P < 0.001). Eleven biomarkers were independently associated with mortality following adjustment for sex, age and baseline forced vital capacity (FVC%predicted). Three biomarkers were associated with ILD progression at 1-year follow-up: CXCL4 (odds ratio 2.67, 1.46-4.88; P = 0.001), MMP-1 (2.56, 1.43-4.59; P = 0.002) and ET-1 (2.18, 1.24-3.83; P = 0.007). Multiple biomarkers, especially VCAM-1, E-selectin, SP-D and CXCL4, provide prognostic utility beyond that of established risk factors for patients with SSc.