Kidney Outcomes With Rituximab Versus Cyclophosphamide in Crescentic ANCA-Associated Glomerulonephritis: Contextualizing Observational Findings

To the Editor:

We read with great interest the study by Uzzo et al.¹ reporting kidney outcomes according to induction immunosuppression and histopathological class in antineutrophil cytoplasmic antibodies-associated glomerulonephritis.¹ The observation of poorer kidney recovery and higher risk of kidney failure with rituximab (RTX) monotherapy compared with cyclophosphamide (or combination) in the crescentic subgroup, is clinically relevant and deserves careful consideration.

First, beyond the retrospective nature of the study, this finding relies on a very small subgroup (n = 12 crescentic patients treated with RTX), which may result in unstable effect estimates.

Second, several baseline characteristics and treatment-related variables were unevenly distributed between groups and were insufficiently described in univariable analyses. Clinical phenotype was not accounted for, despite a higher (although nonsignificant) proportion of microscopic polyangiitis in the RTX group (64% vs. 53%), a phenotype frequently associated with greater chronic renal damage. Plasma exchange (44% vs. 60%) and intravenous glucocorticoid pulses (75% vs. 87%) were less frequently used in RTX-treated patients. Maintenance strategies also differed, with more frequent glucocorticoid-only maintenance in the RTX group (50% vs. 8%).

In addition, the study period spans diagnoses from 1997 onward, whereas RTX was introduced into routine remission-induction practice after 2010.² As treatment strategies have evolved over time, more recently treated patients (who were more likely to receive RTX) also tended to be exposed to lower cumulative glucocorticoid doses, a factor that may have influenced outcomes. Indeed, in the plasma exchange and glucocorticoid vasculitis assessment (PEXIVAS trial)³ patients treated with RTX in the reduced-dose glucocorticoid arm showed a trend for lower response compared with those receiving standard-dose glucocorticoids.

Finally, and as clearly stated by the authors, lack of adjustment for histologic chronicity markers and limited covariate adjustment left substantial residual confounding.

Importantly, these findings contrast with several large observational studies in patients with severe renal involvement. Cohorts restricted to advanced kidney dysfunction (eGFR < 30 or < 15 ml/min per 1.73 m²) have consistently shown comparable renal outcomes between RTX- and cyclophosphamide-based induction strategies, with renal prognosis primarily driven by baseline kidney function and histologic chronicity rather than induction agent choice.4, 5, 6, 7

Taken together, the signal observed in the crescentic subgroup should be viewed as hypothesis-generating. Although it highlights the complexity of treatment decisions in severe ANCA-GN, these findings should be interpreted with caution and do not allow firm conclusions regarding the comparative effectiveness of RTX and cyclophosphamide in this setting.

Disclosure

The authors declare no competing interests.

Authors Contributions

BB prepared the first draft of the manuscript. JFA revised the manuscript. All authors approved the final manuscript.

Declaration of AI and AI-Assisted Technologies in the Writing Process

During the preparation of this work the authors used ChatGPT 5 model to generate suggestions for improvements in writing style. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.