Is antibody clustering predictive of clinical subsets and damage in systemic lupus erythematosus?

Arthritis Rheum. 2005 Dec;52(12):4003-10. doi: 10.1002/art.21414.


Objective: To examine autoantibody clusters and their associations with clinical features and organ damage accrual in patients with systemic lupus erythematosus (SLE).

Methods: The study group comprised 1,357 consecutive patients with SLE who were recruited to participate in a prospective longitudinal cohort study. In the cohort, 92.6% of the patients were women, the mean +/- SD age of the patients was 41.3 +/- 12.7 years, 55.9% were Caucasian, 39.1% were African American, and 5% were Asian. Seven autoantibodies (anti-double-stranded DNA [anti-dsDNA], anti-Sm, anti-Ro, anti-La, anti-RNP, lupus anticoagulant (LAC), and anticardiolipin antibody [aCL]) were selected for cluster analysis using the K-means cluster analysis procedure.

Results: Three distinct autoantibody clusters were identified: cluster 1 (anti-Sm and anti-RNP), cluster 2 (anti-dsDNA, anti-Ro, and anti-La), and cluster 3 (anti-dsDNA, LAC, and aCL). Patients in cluster 1 (n = 451), when compared with patients in clusters 2 (n = 470) and 3 (n = 436), had the lowest incidence of proteinuria (39.7%), anemia (52.8%), lymphopenia (33.9%), and thrombocytopenia (13.7%). The incidence of nephrotic syndrome and leukopenia was also lower in cluster 1 than in cluster 2. Cluster 2 had the highest female-to-male ratio (22:1) and the greatest proportion of Asian patients. Among the 3 clusters, cluster 2 had significantly more patients presenting with secondary Sjögren's syndrome (15.7%). Cluster 3, when compared with the other 2 clusters, consisted of more Caucasian and fewer African American patients and was characterized by the highest incidence of arterial thrombosis (17.4%), venous thrombosis (25.7%), and livedo reticularis (31.4%). By using the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index, the greatest frequency of nephrotic syndrome (8.9%) was observed in patients in cluster 2, whereas cluster 3 patients had the highest percentage of damage due to cerebrovascular accident (12.8%) and venous thrombosis (7.8%). Osteoporotic fracture (11.9%) was also more common in cluster 3 than in cluster 2.

Conclusion: Autoantibody clustering is a valuable tool to differentiate between various subsets of SLE, allowing prediction of subsequent clinical course and organ damage.

Publication types

  • Research Support, N.I.H., Extramural

MeSH terms

  • Adult
  • Antibodies, Anticardiolipin / blood
  • Antibodies, Antinuclear / blood
  • Autoantibodies / blood*
  • Autoantigens / immunology
  • Cluster Analysis
  • DNA / immunology
  • Female
  • Humans
  • Lupus Coagulation Inhibitor / blood
  • Lupus Erythematosus, Systemic / classification
  • Lupus Erythematosus, Systemic / immunology*
  • Lupus Erythematosus, Systemic / pathology*
  • Male
  • Middle Aged
  • Predictive Value of Tests
  • Ribonucleoproteins / immunology
  • Ribonucleoproteins, Small Nuclear / immunology
  • snRNP Core Proteins


  • Antibodies, Anticardiolipin
  • Antibodies, Antinuclear
  • Autoantibodies
  • Autoantigens
  • Lupus Coagulation Inhibitor
  • Ribonucleoproteins
  • Ribonucleoproteins, Small Nuclear
  • SS-A antibodies
  • SS-B antigen
  • snRNP Core Proteins
  • DNA