Pos0039 High Throughput Multiplex Assay Identifies Clinically Meaningful Clusters in Systemic Sclerosis Patients With Low Cardiovascular Risk and No History of Heart Disease.

Background: People with systemic sclerosis (SSc) often show subclinical evidence of primary heart involvement (pHI)1, which when clinically evident, is a prognostic factor of poor outcome. Objectives: We aimed to profile serum proteins in patients with SSc and no history of pHI and identify clinically meaningful clusters to aid early identification and stratification for pHI. Methods: Serum from 78 patients from CONVAS ('CONnective Tissue Disease and VASculitis Cohort') and ELCASA ('ELectrophysiology and CArdiac imaging in SclerodermA') cohorts with no history of pHI, pulmonary hypertension, diabetes or >2 traditional cardiovascular (CV) risk factors was used to measure 355 protein levels across inflammation, cardiometabolic and cardiovascular II/III Olink panels. Unsupervised hierarchical clustering using scaled normalised protein expression was used to define clusters. Demographic and clinical characteristics were compared between clusters and significant differences reported using Kruskal-Wallis rank sum test for continuous or Pearson's chi-squared test for categorical variables, at the 5% significance threshold. Results: Clinical characteristics of the patient cohorts have been previously published and are summarised in Figure 1A (column 2). Unsupervised hierarchical clustering revealed 3 patient clusters – C1 27/78 (35%), C2 33/78 (42%) and C3 18/78 (23%) with intermediate, low and high protein expression respectively (Figure 1B). Top 5 proteins contributing to the clustering were IL18BP, TNFR1, TNFR2, TNFRSF14, and PAR1. C2 patients were younger (median age 51 years) compared to C1 (61 years) and C3 (58 years) (p Conclusion: Differences in serum protein expression identifies clinically meaningful clusters of SSc patients with elevated acute phase and serum cardiac markers despite low CV risk profile. Future work should assess their utility as prognostic biomarkers of SSc-pHI. REFERENCES: 1 Dumitru RB et. al., RMD Open. 2021 Oct;7(3):e001689. Acknowledgements: NIL. Disclosure of Interests: None declared.