Identification of Novel Classes for Patients With Lupus Nephritis Using Two-step Cluster Model
Abstract Objectives: To identify and reclassify the patients in the LN cohort, and to further analyze the prominent clinical features and clinical significance of each cluster of patients.Methods: This is a cross-sectional study of a cohort of 635 LN patients from the Rheumatology Department of Xiangya Hospital of Central South University. Demographic data, laboratory findings and clinical evaluation system include physician’s global assessment and the SLICC/ACR Damage Index were collected. Using two-step cluster analysis, patients with similar clinical property were identified and compared.Results: Among the 635 LN patients, 599 patients (94.3%) were female. The mean age of the patients were 33.8 ± 10.4 years. Three subgroups were identified by two-step cluster analysis. Cluster 1 included 130 (20.5%) patients, Cluster 2 included 132(20.8%) patients and Cluster 3 included 373 (58.7%) patients. Cluster 3 was the largest group of mild disease activity, patients in this cluster had lower white blood cells, neutrophils, lymphocytes and mean SDI scores compared to those in the other two clusters. Cluster 1 was the smallest group of severe damage, patients in this cluster had multiple positive auto-antibodies, higher SDI scores and lower complement level. Patients of cluster 2 had the highest levels of granulocytes, but the results of other laboratory tests were roughly between the cluster 1 and cluster 3.Conclusions: This study reclassified three groups of LN patients in a large cohort. Our research shows that the multiple positive ANA antibody may be related to the high SDI score of LN patients. Clinicians can identify patients at different stages through cluster analysis to better implement prognosis.