The Fidelity of Rheumatoid Arthritis Multivariate Diagnostic Biomarkers Using Discriminant Analysis and Binary...

Area of Science:

• Immunology
• Biochemistry
• Rheumatology

Background:

• Rheumatoid arthritis (RA) is a chronic autoimmune disease causing joint inflammation and irreversible damage.
• Cytokines are implicated in RA pathogenesis and may serve as diagnostic biomarkers.
• Early and accurate diagnosis of RA is crucial for effective management and preventing joint destruction.

Purpose of the Study:

• To identify cytokine profiles that can differentiate RA patients from healthy controls.
• To evaluate the diagnostic performance of binary logistic regression (BLR) and discriminant analysis (DA).
• To explore the utility of gender-specific cytokine panels in RA diagnosis.

Main Methods:

• Serum samples from 78 RA patients and age/sex-matched controls were analyzed for 27 cytokines, chemokines, and growth factors.
• Binary logistic regression (BLR) and discriminant analysis (DA) were employed for statistical modeling.
• Disease activity was assessed using the 28-joint disease activity score (DAS-28).

Main Results:

• Multiple cytokine profiles effectively distinguished RA patients from controls.
• IL-17, IL-4, and RANTES were significant predictors in combined gender models.
• Gender-specific models, particularly for women, showed marginally improved detection fidelity using specific cytokine combinations.

Conclusions:

• Cytokine panels show promise as diagnostic biomarkers for rheumatoid arthritis.
• Binary logistic regression demonstrated higher accuracy than discriminant analysis in this cohort.
• Gender-specific analysis may offer incremental benefits for RA diagnostic accuracy, warranting further investigation.