Analysis of clinical characteristics and risk factors of progressive pulmonary fibrosis associated with systemic sclerosis

OBJECTIVES: To investigate the clinical characteristics and associated risk factors of progressive pulmonary fibrosis (PPF) in patients with systemic sclerosis (SSc), and identify serum biomarkers for SSc-PPF. METHODS: A retrospective investigation was conducted on patients with SSc complicated by interstitial lung disease (ILD) who visited the Department of Rheumatology and Immunology at Tianjin Medical University General Hospital between January 2014 and August 2024. SSc-ILD patients with complete follow-up data were divided into two groups: SSc-PPF and SSc-nonPPF, based on the presence of PPF. Clinical characteristics were analyzed between groups, and logistic regression was employed to identify risk factors for SSc-PPF development. RESULTS: Among 84 SSc-ILD patients with follow-up data, 28 were in the SSc-PPF group and 56 in the SSc-nonPPF group. Among the 28 SSc-PPF patients, HRCT patterns were nonspecific interstitial pneumonia (NSIP) in 24 cases and usual interstitial pneumonia (UIP) in 4 cases. Among the 56 SSc-nonPPF patients, 51 showed an NSIP pattern, and 5 showed a UIP pattern. Compared to the SSc-nonPPF group, the SSc-PPF group had higher levels of lactate dehydrogenase (LDH) (P=0.015), creatine kinase isoenzymes (P=0.045), cytokeratin 19 fragment 21-1 (CYFRA21-1) (P=0.002), HRCT score (P=0.037), and the proportion of ground-glass opacities (P=0.035). Furthermore, logistic regression analysis indicated that elevated serum CYFRA21-1 levels [OR (95% CI)=1.615 (1.097, 2.377), P=0.015] constituted an independent risk factor for PPF development in SSc. ROC curve analysis indicates that CYFRA21-1 combined with LDH demonstrates the highest diagnostic value for PPF, with an AUC of 0.731 (95% CI: 0.615-0.846, P=0.000), specificity of 0.929, and sensitivity of 0.429. CONCLUSION: Elevated serum CYFRA21-1 levels constitute an independent risk factor for the development of PPF in SSc patients, offering potential as a biomarker for SSc-PPF. The combination of CYFRA21-1 and LDH can be used to predict PPF.