Naman Jain MD, DrNB
Consultant Rheumatologist, Ramkrishna Care Hospitals, Raipur, Chhattisgarh
Santiago-Lamelas et al. identified a concise seven-gene transcriptomic signature capable of predicting therapeutic response to tumour necrosis factor inhibitors (TNFi) in patients with rheumatoid arthritis (RA). Using RNA sequencing and integrative bioinformatics across five cohorts (n = 279; 169 responders, 110 non-responders), the study analysed over 53,000 gene expression profiles to isolate the most discriminative transcripts.
Using machine learning and logistic regression, seven genes—KCNK17, MRPL24, DNTTIP1, GLS2, GTPBP2, IL18R1, and COMTD1—were identified as distinguishing responders from non-responders with high predictive accuracy (AUC 0.84–0.95). External validation across three independent datasets confirmed the model’s robustness (AUC up to 0.94). Notably, the transcriptomic signature outperformed conventional serological biomarkers such as anti-cyclic citrullinated peptide (anti-CCP) antibodies and rheumatoid factor.
Functionally, several genes within the signature are involved in metabolic and immunoregulatory pathways: GLS2 and MRPL24 regulate mitochondrial metabolism, IL18R1 mediates cytokine signalling, and COMTD1 and GTPBP2 are linked to ferroptosis and RNA splicing regulation. This data-driven approach demonstrates how transcriptomic integration with computational modelling can inform biologic therapy selection, potentially reducing patient exposure to ineffective treatments and lowering healthcare costs.
Overall, this work represents a pivotal step toward implementing molecular predictors in clinical rheumatology, marking progress towards precision and personalised medicine in RA management.
Pontarini et al. conducted bulk RNA sequencing of labial salivary glands from 55 patients with Sjögren’s disease (SjD) and 44 non-specific sicca controls (QMUL cohort), with independent validation in the TRACTISS clinical trial cohort (n = 29). The study revealed two parallel immune activation pathways underpinning disease heterogeneity. Transcriptomic profiling distinctly segregated SjD from control sicca samples, identifying 1,749 upregulated and 698 downregulated genes. Upregulated genes were enriched for T and B-cell activation signatures, HLA-DR/DQ loci, and type I interferon pathways.
Stratification by glandular histopathology delineated two immunopathologic patterns:
(1) ELS-positive glands enriched for germinal centre-related pathways involving B- and T-cell receptor signalling, IL-21, and IFN-γ;
(2) ELS-negative glands showing predominant type I IFN and antiviral transcriptional signatures, implicating extrafollicular immune activation.
Rheumatoid factor (RF) positivity—more so than anti-Ro/SSA—emerged as the dominant driver of transcriptomic variability, highlighting IFN/RIG-I pathway activation as a pivotal mechanism in autoreactive B-cell expansion. This transcriptional pattern showed strong correlations with serum RF titres and IgG levels and was validated in the TRACTISS cohort. This comprehensive study provides the first transcriptomic evidence that both follicular and extrafollicular B-cell activation, including antiviral sensing pathways, co-exist in SjD pathogenesis. The findings refine current understanding of disease endotypes and underscore the potential of molecular stratification to predict lymphoma risk and optimise therapeutic decision-making.
