Background: Leflunomide/hydroxychloroquine (LEF/HCQ) combination therapy (RepurpSS-I trial) recently was shown to successfully reduce disease activity in patients with primary Sjögren’s disease (pSD) [1]. The mechanisms of action by which the drug combination facilitates this effect in the exocrine glands and in the circulation are still unknown. In addition it is unknown how these are associated with systemic molecular changes.

Objectives: To unravel the transcriptomic networks that are associated with disease inhibition by LEF/HCQ therapy in parotid tissue and their association with systemic molecular shifts.

Methods: RNA sequencing was performed on RNA isolated from parotid tissue and circulating cells (Whole blood and MACS-isolated CD19 B-cells). Gene networks of the transcriptomes were identified using weighted gene co-expression network analyses (WGCNA) and correlations of cluster eigen values with clinical and laboratory parameters were assessed. Additionally, we tested whether baseline networks were associated with the response to therapy at the clinical endpoint (24 weeks). Furthermore, we evaluated correlations with QPCR-assessed IFN scores from WB and PBMC. Finally, newly identified soluble IFN-associated proteins were tested as biomarkers.

Results: Four gene clusters were identified in the parotid tissue that were associated with B cell hyperactivity. Three of these were enriched for Ig genes and genes involved in B-cell/lymphocyte activation and RNA transcription. Over time, from baseline to the clinical endpoint in the placebo and LEF/HCQ groups, these clusters showed statistically significant differences (all at least p<0.05), indicating inhibition of B cell progression. The 4th module was enriched for virus/IFN-associated genes and was related to disease activity (r=0.32, p=0.017) showing differences over time in the Responder (p=0.008) but not Non-Responder (p=0.122) group. This parotid IFN gene module was strongly associated with serum levels of IFN-associated proteins, including Galectin-9 (r=0.675, p-adj p<0.0001).

In whole blood and isolated CD19 B-cells IFN modules were identified that exhibited a significant impact on IFN-associated gene expression that was related to clinical response, most clearly in B cells. The B-cell IFN module was significantly downregulated in responders (p=0.039) but not in non-responders (p=0.77), showing a significant change over time in responders but not in non-responders as compared to placebo (p=0.003 vs 0.36). Interestingly, this gene network was able to identify differences in non-responders and responders at baseline (p=0.0052). This B cell transcriptomic gene network was also robustly associated with serum IFN-associated proteins, including Galectin9 (r=0.71, p<0.0001). Additionally, at baseline a B-cell gene network enriched for immunoglobulin genes was significantly different between non-responders and responders (p=0.020) and was significantly inhibited in responders (p=0.039). This was in line with strong inhibition of two B-cell-associated gene modules in WB, of which one was associated with clinical response (immunoglobulin genes) and another with the drug effect (B-cell activation genes).

Conclusion: Our data for the first time (upon LEF/HCQ therapy) reveal targeting of a number of tissue and blood gene networks that are associated with changes in disease activity. These networks reflect genes associated with B-cell hyperactivity and virus/IFN activity and could help our understanding of (lack of) response to leflunomide/hydroxychloroquine combination therapy and Sjögren immunopathology.

REFERENCES: [1] van der Heijden EHM, Blokland SLM, Hillen MR, et al. Leflunomide–hydroxychloroquine combination therapy in patients with primary Sjögren’s syndrome (RepurpSS-I): a placebo-controlled, double-blinded, randomised clinical trial. The Lancet Rheumatology. 2020;2(5):e260-e9.

Acknowledgements: NIL.

Disclosure of Interests: Valentin Baloche: None declared, Helen Leavis: None declared, Cornelis Bekker: None declared, Perrine Soret Translational Data Scientist & Project Leader (Servier), Saba Nayar: None declared, Martina Tarozzi: None declared, Gastone Castellani: None declared, Laurence Laigle Translational Data Scientist & Project Leader (Servier), Joel van Roon: None declared.