OP0047 IDENTIFICATION OF CLINICAL PHENOTYPES IN PATIENTS WITH AXIAL SPONDYLOARTHRITIS, PERIPHERAL SPONDYLOARTHRITIS AND PSORIATIC ARTHRITIS ACCORDING TO PERIPHERAL MUSCULOSKELETAL MANIFESTATIONS: A CLUSTER ANALYSIS IN THE INTERNATIONAL ASAS-PERSPA STUDY

<jats:sec><jats:title>Background:</jats:title><jats:p>Patients with a diagnosis of Spondyloarthritis (SpA) and Psoriatic Arthritis (PsA) may have predominant axial or peripheral symptoms, and the frequency and distribution of these symptoms may determine the clinical diagnosis by the rheumatologist (“clinical clusters”). Clustering analysis represents an unsupervised exploratory analysis which tries to identify homogeneous groups of cases (“statistical clusters”) without prior information about the membership for any of the cases.</jats:p></jats:sec><jats:sec><jats:title>Objectives:</jats:title><jats:p>To identify “statistical clusters” of peripheral involvement according to the specific location of these symptoms in the whole spectrum of SpA and PsA (without prior information about the diagnosis of the patients), and to evaluate whether these “statistical clusters” are in agreement with the “clinical clusters”.</jats:p></jats:sec><jats:sec><jats:title>Methods:</jats:title><jats:p>Cross-sectional and multicentre study with 24 participating countries. Consecutive patients considered by their treating rheumatologist as suffering from either PsA, axial SpA (axSpA) or peripheral SpA (pSpA) were enrolled. Four different cluster analyses were conducted: the first one using information about the specific location from all the peripheral musculoskeletal manifestations (i.e., peripheral arthritis, enthesitis and dactylitis), and thereafter a cluster analysis for each peripheral manifestation individually. Multiple correspondence analyses and k-means clustering methods were used. Distribution of peripheral manifestations and clinical characteristics were compared across the different clusters.</jats:p></jats:sec><jats:sec><jats:title>Results:</jats:title><jats:p>4465 patients were included in the analysis. Two clusters were found with regard to the <jats:italic>location of all the peripheral manifestations</jats:italic> (Fig. 1). Cluster 1 showed a low prevalence of peripheral manifestations in comparison with cluster 2; however, when peripheral involvement appeared in cluster 1, this was mostly represented by arthritis of hip, knee and ankle, as well as enthesitis of the heel. Patients from cluster 1 showed a higher prevalence of males (63% vs 44%), HLA-B27 positivity (69% vs 38%) and axial involvement (80% vs 52%), as well as more frequent diagnosis of axSpA (66% vs 21%) and more frequently fulfilling the ASAS axSpA criteria (69% vs. 41%). Patients from cluster 2 showed a higher prevalence of psoriasis (63% vs 25%), a more frequent diagnosis of PsA (61% vs 19%), and they fulfilled more frequently the peripheral ASAS (26% vs 11%) and the CASPAR criteria (57% vs 19%).</jats:p><jats:fig id="F1" position="float" orientation="portrait"><jats:label>Figure 1.</jats:label><jats:caption><jats:p>Distribution of the peripheral involvement across clusters</jats:p></jats:caption><jats:graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="EUROAB-2021-OP-2-OP0047_F0001" position="float" orientation="portrait" /></jats:fig><jats:p>Three clusters were found with regard to the <jats:italic>location of the peripheral arthritis.</jats:italic> Clusters 2 and 3 showed a high prevalence of peripheral joint disease, although this was located more predominantly in the lower limbs in cluster 2, and in the upper limbs in cluster 3. Cluster 1 showed a higher prevalence of males, HLA-B27 positivity, axial involvement, a lower presence of psoriasis, a more frequent diagnosis of axSpA and fulfilling the ASAS axSpA criteria in comparison with clusters 2 and 3, respectively. Clusters 2 and 3 showed a higher prevalence of enthesitis and dactylitis in comparison with cluster 1, a more frequent diagnosis of PsA and fulfillment of the CASPAR criteria.</jats:p><jats:p>Information about the <jats:italic>location of enthesitis</jats:italic> exhibited three groups: cluster 1 showed a very low prevalence of enthesitis, while cluster 2 and 3 showed a high prevalence of enthesitis, with a predominant involvement of axial enthesis in cluster 2 and peripheral enthesitis in cluster 3.</jats:p><jats:p>Finally, the <jats:italic>analysis of dactylitis</jats:italic> also exhibited three clusters that showed a very low prevalence of dactylitis, predominantly toes and predominantly fingers involvement, respectively.</jats:p></jats:sec><jats:sec><jats:title>Conclusion:</jats:title><jats:p>These results suggest the presence of heterogeneous patterns of peripheral involvement in SpA and PsA patients without clearly defined groups, confirming the clear overlap of these peripheral manifestations across the different underlying diagnoses.</jats:p></jats:sec><jats:sec><jats:title>Acknowledgements:</jats:title><jats:p>This study was conducted under the umbrella of ASAS with unrestricted grant of Abbvie, Pfizer, Lilly, Novartis, UCB, Janssen and Merck.</jats:p></jats:sec><jats:sec><jats:title>Disclosure of Interests:</jats:title><jats:p>None declared</jats:p></jats:sec>