Almquist and colleagues present a pharmacometric analysis supporting an intensified anifrolumab induction regimen for the Phase 3 IRIS trial in lupus nephritis (LN). 1 The work exemplifies model‐informed drug development, but several assumptions warrant discussion before concluding this is optimal.
The analysis assumes restoring anifrolumab exposure in LN to levels seen in non‐renal systemic lupus erythematosus will restore efficacy. However, LN differs fundamentally from non‐renal disease in tissue pathology and mechanisms of organ injury, so matching systemic exposure may be necessary but is unlikely to be sufficient for renal benefit.
The model is built around urine protein‐creatinine ratio (UPCR), which functions as both a pharmacodynamic endpoint and a determinant of clearance. This positive‐feedback structure amplifies uncertainty, as small errors in either relationship may propagate when extrapolating beyond the observed regimen. Moreover, UPCR cannot distinguish active inflammation from irreversible damage and correlates imperfectly with complete renal response, the accepted clinical endpoint, so optimizing dose on UPCR reduction may not optimize outcomes that matter most to patients.
Simulations show diminishing UPCR gains with additional 900‐mg doses, suggesting the efficacy curve approaches a plateau, yet the manuscript remains an exercise in efficacy rather than therapeutic‐index optimization: extending induction from 3 to 6 months increases cumulative immunosuppressive exposure without an accompanying exposure‐safety model for infection, herpes zoster, or hypogammaglobulinemia.
The discontinuation model also warrants caution, since discontinuation in TULIP‐LN was influenced by protocol‐mandated glucocorticoid tapering and other operational factors, not purely biological treatment effects.
Finally, model identifiability deserves more attention. This framework estimates multiple interacting processes from a modest Phase 2 dataset while fixing several PK parameters from non‐renal models. Alternative parameter combinations may fit the data comparably yet yield different extrapolated predictions; sensitivity analyses or profile likelihoods would strengthen confidence in a uniquely supported optimum.
In summary, the analysis shows that prolonged induction normalizes exposure and accelerates proteinuria reduction, but does not establish that 6 months of intensified dosing optimizes long‐term clinical benefit or therapeutic index. This should be confirmed prospectively.
FUNDING
No funding was received for this work.
CONFLICT OF INTEREST
The author declared no competing interests for this work.
References
- 1. Almquist, J. et al. Anifrolumab dose regimen selection for a phase 3 trial in lupus nephritis. Clin. Pharmacol. Ther. 120, 429–439 (2026). [DOI] [PMC free article] [PubMed] [Google Scholar]