Systemic inflammation and B cell indices predict rituximab responses in membranous nephropathy

Suyan Duan, Yuyu Ye, Qian Zhou, Hujia Hua, Ming Zeng, Chengning Zhang, Yanggang Yuan, Changying Xing, Huijuan Mao, Bo ZhangClinical 2025 · DOI

TL;DR

This study identifies novel biomarkers, specifically the Systemic Immune-Inflammation Index (SIRI) and B cell levels, that can predict treatment response in patients with membranous nephropathy (MN) receiving rituximab (RTX). Integrating these markers into monitoring frameworks offers a more accurate and timely assessment of both short-term efficacy and long-term renal outcomes compared to traditional methods.

What Problem Does This Paper Address?

Membranous nephropathy (MN) is a significant cause of nephrotic syndrome, and while rituximab (RTX) is a common treatment, its effectiveness varies considerably among patients. Traditional markers used to predict RTX response, such as proteinuria and anti-phospholipase A2 receptor (PLA2R) antibody levels, have limitations. They can be slow to change, delayed in their predictive value, and sometimes don't accurately reflect treatment outcomes. This leads to challenges in personalizing RTX treatment, potentially resulting in overtreatment or undertreatment. The paper aims to find more robust and earlier predictive biomarkers for RTX response in MN patients.

Methods

The researchers conducted a prospective study involving 149 MN patients treated with RTX. They collected data at baseline and at 3, 6, 9, and 12 months post-treatment. The study focused on evaluating both traditional markers (proteinuria, serum albumin, anti-PLA2R antibodies) and novel inflammatory indices: neutrophil-to-lymphocyte ratio (NLR), monocyte-to-lymphocyte ratio (MLR), and the Systemic Immune-Inflammation Index (SIRI). B cell levels were also monitored.

Key methodological aspects include:

  • Multivariate Regression Analysis: Used to identify independent predictors of remission.
  • Receiver Operating Characteristic (ROC) Analysis: Employed to assess the predictive accuracy of different models and determine optimal cut-off values for biomarkers.
  • Development of Predictive Models: Two models were developed for predicting 6-month and 12-month remission, comparing traditional markers alone versus models incorporating the novel inflammatory indices and B cell levels.
  • Longitudinal Monitoring: Tracking changes in biomarkers over time to understand their dynamic role in predicting outcomes.

Datasets

The study utilized data from 149 patients diagnosed with membranous nephropathy (MN) who underwent RTX treatment at the First Affiliated Hospital of Nanjing Medical University. This cohort served as the primary dataset for analysis. No new benchmark datasets were introduced.

Key Insights and Ideas

  • Inflammation as a Key Predictor: Persistently elevated inflammatory markers (NLR, MLR, SIRI) were observed in non-responders throughout the study, highlighting the critical role of systemic inflammation in treatment outcomes.
  • SIRI's Dual Role: The Systemic Immune-Inflammation Index (SIRI) emerged as a powerful predictor, not only for short-term (6-month) efficacy but also for long-term (12-month) renal outcomes.
  • B Cells for Early Prediction: B cell levels, particularly a proportion below a certain threshold, proved to be a robust biomarker for predicting early (6-month) therapeutic response.
  • Synergistic Predictive Power: Combining novel markers (SIRI, B cells) with traditional ones significantly enhanced the predictive accuracy for 6-month remission, outperforming models based solely on traditional markers.
  • Shifting Predictive Landscape: While B cell levels and anti-PLA2R antibodies were important for short-term prediction, SIRI and albumin changes became the primary predictors for long-term (12-month) outcomes, suggesting a shift in the underlying disease mechanisms or their reflection in biomarkers over time.

Key Results

  • Short-Term Prediction (6-month remission): At 3 months, a SIRI ≤1.25 and B cell proportion ≤0.2% independently predicted 6-month response. Incorporating these into a model with traditional markers improved the Area Under the Curve (AUC) from 0.81 to 0.86.
  • Long-Term Prediction (12-month remission): At 6 months, SIRI ≤0.9 and albumin change were the key predictors for 12-month prognosis. Anti-PLA2R antibody levels and B cell levels lost their significance for long-term prediction. A model including SIRI achieved a higher AUC (0.82) compared to a model with only proteinuria and albumin changes (0.79).
  • Non-responders Profile: Patients who did not achieve remission showed persistently elevated inflammatory markers (NLR, MLR, SIRI) throughout the observation period.

Limitations and Future Work

  • Limitations:
    • The study acknowledges that while B cell monitoring predicts short-term response, its predictive value diminishes for long-term outcomes.
  • The study did not explore the prognostic value of specific B cell subsets, which might offer more nuanced insights.
  • The paper mentions that NLR and MLR did not show high predictive value in this study, though prior research suggests their importance in other conditions.
  • Future Work:
    • The authors strongly suggest integrating concurrent monitoring of B lymphocyte levels and SIRI values into standardized monitoring frameworks for clinical management of MN patients.
  • Future research should explore the prognostic value of other inflammatory indicators (e.g., platelet-to-lymphocyte ratio) and specific immune cell subsets (e.g., T follicular helper cells, regulatory B cells) and cytokines.
  • Further investigation into the role of B cell subsets and their reconstitution patterns could enhance predictive accuracy.

Summary updated

Abstract

B lymphocyte levels constitute a robust predictive biomarker for assessing short-term therapeutic response in patients with MN receiving RTX therapy. Furthermore, SIRI emerges as a valuable prognostic indicator capable of predicting both short-term efficacy and long-term renal outcomes. These findings suggest that concurrent monitoring of B lymphocyte levels and SIRI values warrants integration into standardized monitoring frameworks within clinical management protocols.
BibTeX
@inproceedings{duan2025systemic,
  author = {Suyan Duan and Yuyu Ye and Qian Zhou and Hujia Hua and Ming Zeng and Chengning Zhang and Yanggang Yuan and Changying Xing and Huijuan Mao and Bo Zhang},
  title = {{Systemic inflammation and B cell indices predict rituximab responses in membranous nephropathy}},
  booktitle = {Clinical},
  year = {2025},
  doi = {10.1093/ckj/sfaf396},
}
Figure