Two Phenotypes Are Identified by Cluster Analysis in Early Inflammatory Back Pain Suggestive of Spondyloarthritis Results From the DESIR Cohort

Objective. To determine whether disease manifestations at baseline would combine according to distinguishable ordered phenotypes in patients with early inflammatory back pain (IBP) suggestive of spondyloarthritis (SpA). Methods. Baseline clinical and demographic characteristics as well as imaging features and biologic data on patients included in the French multicenter Devenir des Spondyloarthropathies Indiffererenciees Recentes cohort were analyzed by multiple correspondence analysis and cluster analysis to identify subgroups of patients based on shared characteristics. Results. Cluster analysis allowed us to classify the 679 patients with no missing data into 2 major groups-one with a predominance of isolated axial manifestations and the other with associated peripheral symptoms. The application of the same analysis to selected subsets of the cohort, such as HLA-B27-positive and -negative patients and patients fulfilling the Assessment of SpondyloArthritis international Society classification criteria for axial SpA, resulted again in an optimal division of the samples into 2 recurrent clusters of patients similar to those observed in the whole cohort. Conclusion. Cluster analysis of SpA manifestations among patients with early IBP highly suggestive of SpA allowed us to clearly identify at baseline 2 different clinical phenotypes-one with predominant axial manifestations and the other with predominant peripheral manifestations. Ongoing follow-up will allow us to determine whether these clusters correspond to different patterns of disease severity.