Predicting membranous nephropathy remission: a nomogram based on early dynamic biomarkers

Traditional prognostic assessment of membranous nephropathy (MN) relies on baseline static indicators, which do not allow for the dynamic monitoring of early treatment response. This study aims to evaluate whether early changes in anti-PLA2R antibody titers, 24-h urine protein quantification, and serum albumin levels can predict 12-month remission and to develop a nomogram model to facilitate clinical decision-making.

A retrospective analysis was conducted on 144 patients with PLA2R-related MN treated with rituximab between 2022 and 2025. Patients were categorized into remission (n = 86) and non-remission (n = 58) groups based on 12-month outcomes. Dynamic changes in biomarkers were analyzed at 1 and 6 months post-treatment. Risk factors were identified using Lasso-Logistic regression, and a nomogram was constructed and evaluated using receiver operating characteristic (ROC) curves and calibration plots.

At 1 month post-treatment, non-remission patients demonstrated a significantly lower percentage decline in anti-PLA2R antibodies (47.26% vs. 85.74%, p < 0.001), less reduction in 24-h urinary protein (4.5% vs. 42.5%, p < 0.001), and a smaller increase in serum albumin (6.47% vs. 17.06%, p < 0.001). Multivariate analysis identified 1-month urinary protein (OR = 1.374), anti-PLA2R decline (OR = 0.944), and albumin increase (OR = 0.967) as independent predictors of remission. The 1-month nomogram showed excellent discrimination (AUC = 0.905) and strong calibration, with predictive value comparable to 6-month monitoring.

Dynamic changes in anti-PLA2R antibodies, urinary protein, and serum albumin at 1 month are effective predictors of 12-month remission in MN. This nomogram model enables early risk stratification and the optimization of personalized treatment strategies.