Kidney Week

Abstract: SA-PO0859

Assessing Proteinuria as a Mediator of Treatment Effects of Calcineurin Inhibitors (CNI) vs. Rituximab (RTX) for Membranous Nephropathy (MN)

Session Information

• Glomerular Diseases: Management, Evolving Strategies, and Practice-Changing Advances
  October 24, 2026 | Location: Exhibit Hall A, Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Glomerular Diseases

• 1402 Glomerular Diseases: Clinical, Outcomes, and Therapeutics

Authors

• Liu, Xiaohan, University of Pennsylvania Department of Biostatistics Epidemiology & Informatics, Philadelphia, Pennsylvania, United States

Xiaohan Liu, MS

• Gerety, Meghan, University of Pennsylvania Department of Biostatistics Epidemiology & Informatics, Philadelphia, Pennsylvania, United States

Meghan Gerety, MS

• Wyatt, Nicole Elizabeth, The University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, North Carolina, United States

Nicole Elizabeth Wyatt, DO

• Smith, Abigail R., Northwestern University Feinberg School of Medicine, Chicago, Illinois, United States

Abigail R. Smith, PhD

• Mariani, Laura H., University of Michigan Medical School, Ann Arbor, Michigan, United States

Laura H. Mariani, MD, MS, FASN

• Zee, Jarcy, University of Pennsylvania Department of Biostatistics Epidemiology & Informatics, Philadelphia, Pennsylvania, United States

Jarcy Zee, PhD, FASN

Background

The pathways by which treatments affect long-term kidney outcomes in MN are not yet completely characterized, particularly the role of proteinuria reduction. This study applied modern statistical methods to decompose the total effect of CNI versus RTX on a long-term disease progression outcome into indirect effects mediated by proteinuria and direct effects.

Methods

MN participants enrolled in CureGN initiating RTX or CNI without previous immunosuppressants (IST) 6 months before treatment start were eligible. Propensity scores (PS) based on demographics, labs, pathology, and comorbidities at start were used for PS-weighting. Individuals were censored upon initiation of another IST and censoring weights were applied. Longitudinal measures of urine-protein-to-creatinine-ratio were evaluated as a mediator of the effect of CNI vs. RTX on time from start to 40% eGFR decline or kidney failure. Survival indirect effects (SIE) and survival direct effects (SDE) were estimated using Aalen’s method for mediation analysis.

Results

After weighting, average effective sample sizes of N=136 RTX and N=120 CNI were analyzed. The estimated effect of CNI vs. RTX on disease progression through proteinuria remained small over follow-up (SIE range=0.98–1.00 over 5 years), while the effect on disease progression through mechanisms beyond proteinuria became more pronounced (SDE=1.00 at 1 year and 0.81 at 5 years). In sensitivity analyses that censored study participants with 2 or more years of continuous treatment, SDE was substantially attenuated (range=0.92-1.00) while SIE remained stable (range=0.96–1.00).

Conclusion

Proteinuria mediated a small proportion of the long-term treatment effect of CNI vs. RTX. Sensitivity analysis suggests CNI-associated nephrotoxicity from prolonged use contributes substantially to inferior long-term kidney outcomes compared with RTX. Treatment duration is an important factor in the amount of mediation by proteinuria in this comparison.

Estimated effects over follow-up for PP/sensitivity analyses.