Examining Rituximab in Refractory Lupus Nephritis
Key Takeaways
- Among patients with relapsing or refractory lupus nephritis receiving rituximab-based rescue or add-on therapy, overall renal response was observed in approximately two-thirds.
- Pooled complete renal response estimates were higher and partial response estimates lower at later assessment windows, although the pools do not track the same patients over time.
- Pooled proteinuria and serum creatinine changes were both negative after rituximab-based therapy, without establishing a comparative treatment effect.
Wang and colleagues’ registered systematic review and meta-analysis of rituximab-based therapy searched PubMed, Embase, and Web of Science from inception and pooled single-arm data using random-effects models. The synthesis included 19 studies and 445 patients with relapsing, refractory, or relapsing/refractory lupus nephritis. Most studies were observational; two randomized trials contributed only rituximab-arm data. Eligible full-length human studies could be prospective, retrospective, registry-based, or randomized; pediatric-only studies and newly diagnosed disease without separable target-population data were excluded. Rituximab was used as rescue or add-on treatment, sometimes with cyclophosphamide- or mycophenolate mofetil-based therapy. Investigators pooled complete renal response (CRR), partial renal response (PRR), and overall renal response (ORR) at available follow-up windows, using each study’s response criteria and permitting near-final-window observations. They standardized within-treatment mean changes in proteinuria to grams per day equivalent (g/day-equivalent) and serum creatinine to micromoles per liter (μmol/L).
Pooled ORR estimates were 0.67 (95% confidence interval [CI] 0.52–0.80) at 6 months, 0.70 (95% CI 0.51–0.86) at 12 months, and 0.66 (95% CI 0.43–0.86) at 24 months. Complete-response pools were higher at later windows, while partial-response pools were lower. Different studies contributed to different time points and endpoints, so these pooled estimates do not describe how individual patients’ responses evolved.
Across rituximab-treated groups, pooled proteinuria mean change was −2.69 g/day-equivalent (95% CI −3.23 to −2.16). Pooled serum creatinine mean absolute change was −11.79 μmol/L (95% CI −19.86 to −3.72). Both estimates reflect within-treatment changes, not differences from an alternative therapy.
Largely observational, single-arm evidence, including only treated arms from the randomized trials, cannot establish causality or incremental efficacy relative to conventional immunosuppression or newer agents. Renal-response definitions, disease phenotypes, prior and concomitant therapies, and follow-up varied, with moderate to substantial heterogeneity. Differences by nephritis type, treatment method, or combination regimen were exploratory, sometimes rested on very few studies, and were mostly not significant between subgroups. Proteinuria reductions remained directionally stable in sensitivity analyses; in one leave-one-out analysis, the creatinine change lost statistical significance despite retaining its direction. Long-term renal and safety outcomes were not consistently available for pooling, limiting conclusions about sustained outcomes or harms.
Wang and colleagues associated rituximab-based escalation in selected patients with difficult-to-treat lupus nephritis with renal responses and lower proteinuria and serum creatinine. The authors called for prospective comparative studies to clarify rituximab’s role in this setting.
Clinician Questions
Do rituximab response estimates for refractory lupus nephritis extend to patients previously treated with belimumab?
Patients with relapsing or refractory lupus nephritis in Wang and colleagues’ rituximab synthesis were predominantly treated after conventional immunosuppression, commonly cyclophosphamide and/or mycophenolate mofetil. The analysis does not establish renal outcomes for rituximab after newer regimens such as belimumab.
How was a urine protein-to-creatinine ratio incorporated into the rituximab proteinuria analysis for relapsing or refractory lupus nephritis?
The pooled proteinuria analysis of rituximab-treated relapsing or refractory lupus nephritis used daily or 24-hour urine protein excretion from eight studies and converted protein-to-creatinine ratio data from one study into g/day-equivalent. Excluding the converted study left the reduction similar in direction and magnitude.
Recommended Reading
- For more on proteinuria and renal outcomes in lupus nephritis: Low Proteinuria Linked to Fewer Events in Lupus Nephritis