Background

Systemic sclerosis (SSc) is a rare connective tissue disease associated with rapid evolving interstitial lung disease (SSc-ILD). Imaging-based biomarkers associated with the evolution of the lung disease are highly needed to help in the prediction outcome.

Methods

We evaluated the potential of the AI-based quantification tool on chest CT imaging icolung (icometrix, Belgium). To do so, we retrospectively correlated icolung outcomes with pulmonary function tests and evolution over time.

Results

We evaluated a group of 75 patients suffering from SSc-ILD ranging from limited to extended diseases out of which 31 were presenting progressive pulmonary fibrosis (PPF) associated with SSc. The patients presenting PPF were exhibiting more extended lesions out of the Total Lung Volume (TLV in %): 3.21% (0.36-6.97) versus 0.6% (0.1-4.11) respectively whereas pulmonary functional test were showing a typical reduction in FVC (%pred)(73 (62-91), 90 (73-102)(p<0.05) respectively). We identified a trend to a higher severity score in PPF of 4 (1-6) versus 1 (0-4.5) respectively (p=0.52). Percentage of ILD quantification was also correlated to FVC and DLCO modifications over time (r=-0.39, p<0.01; r=-0.40, p<0.01 respectively).

Conclusion

AI-based automatized quantification of chest CT images in SSc-ILD is correlated with physiological parameters and could help in disease evaluation. Further clinical multicentric validation is necessary in order to confirm its potential in the prediction of patient?s outcome and in treatment management.