Longitudinal Changes in Quantitative Imaging Biomarkers are Associated with Increased Risk of Mortality in Connective Tissue Disease-Related Interstitial Lung Disease.

<bold>Background:</bold> The role of quantitative CT (qCT) metrics in fibrotic lung disease due to connective tissue disease (CTD) is underexplored compared to idiopathic pulmonary fibrosis (IPF). <bold>Aims:</bold> To assess the prognostic utility of short-term changes in deep learning-based qCT metrics in an all-comers CTD population. <bold>Methods:</bold> CTD participants from OSIC (n=68) with follow up CT imaging were included. Baseline and 3 month volumetric assessments of airway volume, pulmonary vessel volume, fibrosis and ground glass opacification were quantified. Cox regression, including baseline and 3 month change variables, was conducted to explore relationships between these metrics and all-cause mortality. Progressive disease was 10% or more decline in %ppFVC or death within 12 months. <bold>Results:</bold> Adjusting for %ppFVC, baseline and 3-month interval change in fibrosis extent (HR 1.07, p<0.0001, 95%CI 1.03-1.10, HR1.28, p=0.02, 95%CI 1.04-1.58, C-index:0.80) and pulmonary vascular volume (HR 4.84, p=0.01, 95%CI 1.46-15.97, HR1.12, p=0.007, 95%CI 1.03-1.23, C-index:0.73) and 3-month interval change in airway volume (HR1.02, p<0.0001, 95%CI 1.01-1.03, C index:0.76) were associated with increasing risk of mortality. A 1% increase fibrosis extent over 3 months was associated with a 1.6-fold increased risk of progressive disease. 3-month change in airway or pulmonary vascular volume were not associated with progressive disease. <bold>Conclusions:</bold> In CTD-related fibrotic lung disease, short-term changes in qCT metrics provide additional prognostic information over baseline measures of disease severity.