To the Editor: We read with great interest the recent article by Mullol et al. examining early and sustained improvements in olfactory function with tezepelumab in patients with severe uncontrolled chronic rhinosinusitis with nasal polyps (CRSwNP) 1. The authors should be commended for conducting a rigorous Phase 3 randomized controlled trial and for emphasizing olfactory outcomes, which represent one of the most disabling and safety-relevant symptoms for patients with CRSwNP. The demonstration of rapid onset and sustained improvement in smell-related outcomes provides clinically meaningful insight into the therapeutic potential of upstream epithelial cytokine blockade in this population. One important consideration relates to the clinical interpretability of the reported changes in olfactory measures. Although statistically significant between-group differences were observed in validated instruments such as the University of Pennsylvania Smell Identification Test (UPSIT) and symptom-based measures, mean score differences do not necessarily reflect meaningful recovery at the individual patient level. Given that minimal clinically important difference (MCID) thresholds were defined, reporting responder-based outcomes—such as the proportion of patients achieving MCID, transitioning from anosmia to microsmia or normosmia, or achieving normalization of olfactory function—would provide additional context regarding the magnitude and clinical relevance of treatment benefit 2. Such analyses would help clarify whether early separation between treatment groups translates into meaningful restoration of olfactory function rather than modest statistical improvement. An additional methodological consideration involves the handling of rescue interventions and missing data. Replacement of post-rescue or postsurgical values with worst-case scores, combined with multiple imputation under a missing-at-random assumption, represents a conservative and commonly accepted approach. However, differential rates or timing of rescue therapy between treatment arms could potentially influence treatment effect estimates. Additional sensitivity analyses using alternative estimands or reporting the frequency, timing, and distribution of rescue interventions would strengthen confidence in the robustness of the observed olfactory improvements 3. These analyses would help distinguish true treatment effects from potential artifacts introduced by rescue-related data handling. Generalizability of the findings also warrants further discussion. The study population consisted of patients with severe, uncontrolled CRSwNP with prior systemic corticosteroid exposure and/or prior sinus surgery, representing a population enriched for advanced disease 4. While appropriate for evaluating efficacy in refractory disease, this may limit extrapolation to patients with less severe CRSwNP or earlier disease stages. Additional subgroup analyses stratified by baseline olfactory dysfunction severity, disease duration, inflammatory biomarkers, or prior biologic exposure could help identify patient populations most likely to benefit from tezepelumab and inform individualized treatment strategies. Finally, the multidimensional nature of olfaction should be considered when interpreting UPSIT findings. UPSIT primarily evaluates odor identification rather than olfactory threshold or discrimination 5. Incorporating complementary olfactory testing modalities in future studies could provide a more comprehensive assessment of sensory recovery and help clarify whether tezepelumab restores fundamental olfactory sensitivity or primarily improves higher-order odor recognition. In summary, this study provides important evidence supporting the beneficial effects of tezepelumab on olfactory dysfunction in severe CRSwNP. Further analyses incorporating responder-based outcomes, sensitivity analyses addressing rescue interventions, expanded subgroup characterization, and multidimensional olfactory assessment would enhance understanding of the clinical impact and optimal positioning of tezepelumab in CRSwNP management. We congratulate the authors on their valuable contribution and believe that these considerations may help guide future investigations and clinical applications. The authors have nothing to report. The authors declare no conflicts of interest.
Liu et al. (Mon,) studied this question.