A proteomic profile correlates with salivary gland inflammation and disease-associated antibodies in Sjögren's disease.

OBJECTIVES

To identify serum and salivary gland proteomic biomarkers in primary Sjögren's disease (SjD), evaluate their correlation with tissue inflammation, and relate findings to clinical, autoantibody, and genetic profiles.

METHODS

Paired salivary gland extracts and serum samples from patients with SjD (n = 81) and sicca controls (n = 19), and serum from larger Swedish (SjD n = 456; controls n = 141) and Norwegian (SjD n = 233; controls n = 137) cohorts, were analysed using proximity extension assay technology. Patients were stratified by antinuclear antibodies (ANA), anti-Ro/SSA, anti-La/SSB autoantibodies, and extraglandular manifestations. Human leukocute antigen (HLA) alleles were integrated to assess genetic associations.

RESULTS

Proteomic analysis revealed more pronounced alterations in the salivary gland extracts compared to serum. In serum, 27 proteins were associated with SjD in the Swedish discovery cohort, of which 16 proteins were validated in the Norwegian replication cohort. Several proteins including both described biomarkers C-X-C motif chemokine ligand (CXCL)10 and CXCL13 and more novel lysosome-associated membrane glycoprotein 3 and cytotoxic and regulatory T cell molecule correlated with extraglandular manifestations, particularly pulmonary involvement. Levels of the 16 validated proteins increased in a stepwise fashion across serological subgroups, from lowest in ANA negative to highest in anti-Ro/SSA and anti-La/SSB double-positive patients and were linked to risk alleles HLA-DRB1*03 and DRB1*15. A composite 'Sjögren protein score' based on the 16 validated proteins correlated with salivary gland focus score and successfully distinguished SjD from healthy controls (area under the curve 0.888; 89% specificity, 76% sensitivity).

CONCLUSIONS

We define a validated serum proteomic signature that reflects local glandular inflammation and correlates with genetic and serological features. The 'Sjögren protein score' holds promise as a noninvasive biomarker for the diagnosis of SjD.