Prediction of stable SSc-ILD depends on definition of ILD progression

<bold>Background:</bold> Identifying long-term stable SSc-ILD patients is important in clinical practice to avoid overtreatment. <bold>Objectives:</bold> To determine factors associated with stable SSc-ILD. <bold>Methods:</bold> We included SSc patients from two expert centers with ILD on HRCT, available consecutive annual lung function and comprehensive serial clinical and imaging assessments. Long-term stable ILD was defined as no progression observed over three years (defined by FVC, PPF guideline and PF-ILD INBUILD). Logistic regression was applied, adjusting for known risk factors for ILD progression, including treatment, to identify factors predictive of stable ILD. <bold>Results:</bold> In total, 231 SSc-ILD patients were included. We identified 75 (32%), 133 (58%) and 105 (45%) patients with stable ILD defined by no FVC decline ≥5%, no PPF guideline criteria and no INBUILD criteria respectively. Multivariable logistic regression models predicting long-term, stable ILD varied based on the applied definition. Stable ILD using no FVC ≥5% decline was predicted by lower baseline FVC (Figure 1A). Stable ILD defined by no PPF and no PF-ILD was predicted by the absence of dcSSc and of ground glass opacities on HRCT (Figure 1B and C). <bold>Conclusion:</bold> Long-term stable ILD in SSc occurs, but the frequency varies based on which definition is applied. Prediction of stable patients is challenging, highlighting the necessity of comprehensive disease assessment and monitoring over time.