Stratification of Patients With Sjögren’s Syndrome and Patients With Systemic Lupus Erythematosus According to Two Shared Immune Cell Signatures, With Potential Therapeutic Implications

Abstract

Objective Similarities in the clinical and laboratory features of primary Sjögren’s syndrome (SS) and systemic lupus erythematosus (SLE) have led to attempts to treat patients with primary SS or SLE with similar biologic therapeutics. However, the results of many clinical trials are disappointing, and no biologic treatments are licensed for use in primary SS, while only a few biologic agents are available to treat SLE patients whose disease has remained refractory to other treatments. With the aim of improving treatment selections, this study was undertaken to identify distinct immunologic signatures in patients with primary SS and patients with SLE, using a stratification approach based on immune cell endotypes. Methods Immunophentyping of 29 immune cell subsets was performed using flow cytometry in peripheral blood from patients with primary SS (n = 45), patients with SLE (n = 29), and patients with secondary SS associated with SLE (SLE/SS) (n = 14), all of whom were considered to have low disease activity or be in clinical remission, and sex‐matched healthy controls (n = 31). Data were analyzed using supervised machine learning (balanced random forest, sparse partial least squares discriminant analysis), logistic regression, and multiple t‐tests. Patients were stratified by K‐means clustering and clinical trajectory analysis. Results Patients with primary SS and patients with SLE had a similar immunologic architecture despite having different clinical presentations and prognoses. Stratification of the combined primary SS, SLE, and SLE/SS patient cohorts by K‐means cluster analysis revealed 2 endotypes, characterized by distinct immune cell profiles spanning the diagnoses. A signature of 8 T cell subsets that distinctly differentiated the 2 endotypes with high accuracy (area under the curve 0.9979) was identified in logistic regression and machine learning models. In clinical trajectory analyses, the change in damage scores and disease activity levels from baseline to 5 years differed between the 2 endotypes. Conclusion These findings identify an immune cell toolkit that may be useful for differentiating, with high accuracy, the immunologic profiles of patients with primary SS and patients with SLE as a way to achieve targeted therapeutic approaches.

Multi-Omic Biomarkers for Patient Stratification in Sjogren’s Syndrome—A Review of the Literature

“…Martin-Gutierrez et al [ 33 ] identified an immune signature derived from the analysis of 29 different cell subsets including B and T cells, which was driven by five distinct cell subsets: transitional Bm2′ cells, late memory Bm5 cells, IgD-CD27-B cells, and CD8 + naive and CD8 + Tem which differentiated between pSS patients and matched HCs. Moreover, they identified a shared immunological profile across three disease phenotypes: systemic lupus erythematosus (SLE), pSS and SLE associated with SS.…”

“…The main players seem to be activated CD8 + T cells, terminally differentiated plasma cells, and activated epithelial cells in biopsies, whereas the peripheral blood signatures comprised high numbers of activated CD4 + , CD8 + T cells. Although these signatures were not perfectly validated across various studies [ 30 , 33 ], some correlated with serological and clinical parameters, suggesting a potential clinical utility.…”

Machine Learning Techniques for Personalised Medicine Approaches in Immune-Mediated Chronic Inflammatory Diseases: Applications and Challenges

“…In the recent study of primary Sjögren’s syndrome (pSS) and SLE ( Martin-Gutierrez et al, 2021 ), researchers applied supervised ML models to identify shared immunological characteristics between pSS and SLE. These two diseases share some clinical and laboratory features, despite differences in disease pathogenesis and overall clinical presentation, leading to a distinct diagnostic label ( Pasoto et al, 2019 ).…”

“…The process of drug repurposing is much conserved by renouncing the standard drug development pipeline approach and investigating similarities between various disease processes potentially targeted by the same therapeutic interventions, so that new effective treatments can be delivered faster to patients. This is a more cost-effective approach which led in recent years to the testing and licensing of similar classes of therapeutic agents across many immune-mediated chronic inflammatory diseases ( March-Vila et al, 2017 ; Balasundaram et al, 2019 ; Gilvary et al, 2020 ; Martin-Gutierrez et al, 2021 ). In a recent study of computational drug repurposing, researchers developed a ML (Gradient Boosting model) approach, Creating A Translational Network for Indication Prediction (CATNIP) which can effectively connect similar drugs by solely analysing the biological and chemical data of the molecule without the knowledge of the current therapeutic disease applications of the drug ( Gilvary et al, 2020 ).…”

“…Clustering approaches within unsupervised learning, including hierarchical clustering, K-means clustering and Gaussian mixture models, are the most popular techniques for assembling data into previously ambiguous bundles. Unsupervised clustering approaches form the decisive component in most patient stratification studies and in identifying disease subtypes ( Mossotto et al, 2017 ; Orange et al, 2018 ; Robinson et al, 2020 ; Martin-Gutierrez et al, 2021 ). Finally, reinforcement learning is scripted to sequentially self-correct from environmental feedback (positive or negative) and therefore improve the overall model function without having labelled data ( Kaelbling et al, 1996 ).…”