Identifying heterogeneous subgroups of systemic autoimmune diseases by applying a joint dimension reduction and clustering approach to immunomarkers

Abstract

Background: The high complexity of systemic autoimmune diseases (SADs) has hindered precise management. This study aims to investigate heterogeneity in SADs.

Methods: We applied a joint cluster analysis, which jointed multiple correspondence analysis and k-means, to immunomarkers and measured the heterogeneity of clusters by examining differences in immunomarkers and clinical manifestations. The electronic health records of patients who received an antinuclear antibody test and were diagnosed with SADs, namely systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), and Sjögren's syndrome (SS), were retrieved between 2001 and 2016 from hospitals in Taiwan.

Results: With distinctive patterns of immunomarkers, a total of 11,923 patients with the three SADs were grouped into six clusters. None of the clusters was composed only of a single SAD, and these clusters demonstrated considerable differences in clinical manifestation. Both patients with SLE and SS had a more dispersed distribution in the six clusters. Among patients with SLE, the occurrence of renal compromise was higher in Clusters 3 and 6 (52% and 51%) than in the other clusters (p < 0.001). Cluster 3 also had a high proportion of patients with discoid lupus (60%) than did Cluster 6 (39%; p < 0.001). Patients with SS in Cluster 3 were the most distinctive because of the high occurrence of immunity disorders (63%) and other and unspecified benign neoplasm (58%) with statistical significance compared with the other clusters (all p < 0.05).

Conclusions: The immunomarker-driven clustering method could recognise more clinically relevant subgroups of the SADs and would provide a more precise diagnosis basis.

Keywords: Autoimmune diseases; Cluster analysis; Disease heterogeneity; Immune markers.

Fig. 1

Flowchart of patient inclusion rules. (RA, rheumatoid arthritis; SLE, systemic lupus erythematosus; SS, Sjögren’s syndrome)

Fig. 2

Diverse clustering of the patients with SADs based on immunomarkers. (A) Clusters of patients obtained using MCA k-means with abnormal immunomarkers contributing to the lowest two principal components (PCs). Coloured points with diseases labelled in shapes represent the patients, and immunomarkers with abnormal results are shown as black crosses (x) along with their names. The relative location between a cluster’s centroid and an immunomarker suggests the tendency, compared with other clusters, of patients in a specific cluster to test abnormal for immunomarkers located nearby. For example, Cluster 3 is located near C3, C4, and anti-dsDNA, suggesting that patients in this cluster are more likely to test abnormal for these immunomarkers. (B) SADs are grouped into six clusters. The patients with the SADs (RA, SLE, and SS) are grouped into six clusters based on the pattern of immunomarkers. Each cluster is composed of multiple SADs. For example, Cluster 2 comprises SLE, SS, and a few RA cases. The heterogeneity indicates that the patients exhibit similar immunomarker patterns even if diagnosed with different SADs. (C) Heterogeneity of SADs. RA is predominantly distributed only in Cluster 1. High heterogeneity is noted for SLE and SS – patients with SS can exhibit the immunomarker patterns of Clusters 1, 2, 4, and 5; patients with SLE are also observed in Clusters 2 and 3

Fig. 3

Immunological characteristics of clusters based on immunomarkers. (A) and (B) The proportion of abnormal results in the 10 immunomarkers in the clusters

Fig. 4

Heatmaps illustrate the rates of (A) Clinical Classification Software (CCS) diagnosis groups and (B) common clinical manifestation occurrences of the systemic autoimmune diseases between clusters in each of the individual diseases. We classified all ICD codes (not limited to SADs related codes) into CCS single-level diagnosis groups to identify clinically meaningful manifestations. As shown in Fig. 4B, we collected, from other studies, manifestations that were commonly observed in patients with the SADs