Pre-treatment serum albumin predicts relapse in idiopathic inflammatory myopathies: a retrospective cohort study with cytokine profiling

Abstract Objective To identify predictors of relapse in patients with idiopathic inflammatory myopathies (IIMs) and investigate the biological mechanisms underlying these associations using cytokine profiling. Methods We conducted a retrospective study of 99 patients who achieved remission after initial treatment for IIMs or myositis-specific antibody-positive interstitial lung disease at Nagasaki University Hospital. Among 197 diagnosed patients, we included those who achieved clinical improvement and were followed up for ≥3 months, excluding early deaths and those with insufficient follow-up. Relapse was defined as disease worsening requiring treatment intensification. Risk factors were analyzed using the Cox regression analysis. Serum cytokine profiles were compared between patients with low and high pre-treatment albumin levels. Results Forty-one patients (41%) relapsed. Multivariate analysis revealed that pretreatment albumin level remained the only independent predictor of overall relapse (HR, 0.60; 95% CI: 0.40–0.92, P = 0.019). For ILD relapse, both albumin (HR: 0.51, 95% CI: 0.30–0.88, P = 0.015) and female sex (HR: 0.70, 95% CI: 0.52–0.95, P = 0.022) remained independent predictors. In antibody-specific analyses, the V-neck sign strongly predicted ILD relapse in anti-ARS-positive patients (HR, 5.07; 95% CI: 1.64–15.75, P = 0.006). After false discovery rate correction, 14 cytokines remained significantly elevated in patients with low albumin levels, including IL-6 (7.5-fold, q < 0.001), IP-10 (5.3-fold, q < 0.001) and MCP-1 (2.5-fold, q < 0.001). Conclusion Pretreatment albumin levels independently predict relapse in IIMs, and cytokine profiling demonstrates that low albumin levels reflect active systemic inflammation. These findings support the use of albumin as a practical biomarker for risk stratification in clinical practice.