Related Experiment Video
Updated: Apr 18, 2026
Imaging Features of Systemic Sclerosis-Associated Interstitial Lung Disease
Published on: June 16, 2020
Cluster analysis and risk prediction model construction for antisynthase syndrome-associated interstitial lung
Jin Zhang1, Wenfeng Gao2, Baoting Chao3
1Department of Rheumatology and Immunology, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Jinan, Shandong, China.
Objective:
Interstitial lung disease (ILD) is a common and serious complication of anti-synthase syndrome (ASS), exhibiting high heterogeneity. This study aimed to stratify patients with ASS-associated ILD (ASS-ILD) based on their clinical features, high-resolution computed tomography (HRCT) findings, and specific antibody profiles. We also aimed to build a predictive model for severe ASS-ILD to help identify it more accurately.
Methods:
We retrospectively analyzed the clinical and laboratory data of 100 patients with ASS-ILD. Unsupervised clustering of clinical characteristics was done using multiple correspondence analysis (MCA) and hierarchical cluster analysis. Univariate and multivariate logistic regression were used to find independent risk factors for severe ASS-ILD, and a risk prediction model was built.
Results:
We identified three patient groups: Cluster 1 (n = 46), the joint-predominant group, featured joint pain as the primary clinical manifestation. HRCT typically showed a non-specific interstitial pneumonia (NSIP) pattern, anti-Jo-1 antibodies were most common, and lung damage was relatively mild. Cluster 2 (n = 41), the fever and Raynaud's phenomenon group, had significantly higher anti-EJ antibody levels than the other groups. Cluster 3 (n = 13), the severe lung group, had much higher Warrick scores than the others. This group also had a higher frequency of a usual interstitial pneumonia (UIP) pattern on HRCT, more cases positive for anti-PL-7 or anti-PL-12 antibodies, more severe lung diffusion problems, and more heart involvement. We created and tested a combined biomarker index based on complement C3 level, rash, and age. This index demonstrated superior performance in distinguishing severe from non-severe ILD in ASS patients, independent of clinical variables such as gender, age, and smoking history.
Conclusion:
Identifying ASS-ILD heterogeneity requires comprehensive consideration of HRCT features, antibody profiles, and clinical manifestations. The combined biomarker index (based on complement C3, rash, and age) offers an objective tool for ASS-ILD diagnosis and risk stratification.
