Defining Treatment Response in Pediatric Asthma Pharmacogenomics: Comments on “Gene Polymorphisms Associated with Treatment Response of Asthma in Chinese Children” [Letter]

Dear editor

I read with great interest the study by Zhu et al investigating genetic polymorphisms associated with asthma susceptibility and response to budesonide/formoterol in Chinese children.¹ Identifying reproducible pharmacogenomic markers of treatment response is an important step toward precision therapy in pediatric asthma. The reported associations between eight single-nucleotide polymorphisms (SNPs) and outcomes during 12 weeks of budesonide/formoterol treatment are therefore intriguing. Before these variants are interpreted as markers that could guide treatment selection, however, two aspects of the treatment-response phenotype deserve clarification.

The first concerns the direction of ΔFEV1%. In the Methods, ΔFEV1% is defined as “FEV1% before inhaler therapy − FEV1% after 12 weeks of inhaler therapy”, while a ΔFEV1% ≥10% is simultaneously defined as clinically significant improvement.¹ These definitions point in opposite directions. If FEV1% increased from 70% to 85%, for example, the stated formula would yield −15 rather than +15. The Results, however, appear to use the opposite convention: a value of +16.97 in TBXT-rs2305089 C-allele carriers is described as improvement, whereas −9.10 in TT carriers is described as deterioration. A similar direction is used for LTA4H-rs2540487.¹

This may simply reflect a reversed equation in the Methods rather than an error in the underlying analysis. Nevertheless, the distinction is consequential because the reported efficacy rates for LTC4S-rs730012 and CYSLTR1-rs320995 are defined by the ΔFEV1% ≥10% threshold.¹ Clarifying whether ΔFEV1% was calculated as post-treatment minus baseline, and confirming that responder/non-responder classification and the efficacy analyses were based on that intended direction, would substantially strengthen confidence in these findings. In pharmacogenomic studies, precise and reproducible phenotype definition is as important as accurate genotyping.²

A second issue is whether the reported associations identify drug-specific response or broader prognosis during treatment. The eight SNPs were not linked to a single common response phenotype. ADRB2-rs1042713 and STIP1-rs4980524 were associated with rescue medication use; SLCO2B1-rs2712807 and FCER2-rs28364072 with respiratory infections; TBXT-rs2305089 and LTA4H-rs2540487 with change in FEV1%; and LTC4S-rs730012 and CYSLTR1-rs320995 with the dichotomized FEV1-based efficacy rate.¹ These are clinically relevant outcomes, but they are not interchangeable measures of pharmacologic response.

This distinction becomes particularly important because all 89 children in the treatment-response analysis received budesonide/formoterol.¹ Without a comparison treatment or a genotype-by-treatment interaction, an association between genotype and outcome among treated children cannot by itself establish that the genotype predicts differential benefit from budesonide/formoterol.³ A variant associated with respiratory infections, for example, may reflect infection susceptibility rather than reduced drug efficacy. Likewise, rescue medication use may reflect baseline disease liability, adherence, inhaler technique, environmental exposure, or treatment response. Such variants may therefore be prognostic markers in treated children without yet being predictive biomarkers for choosing budesonide/formoterol.

This distinction matters clinically because the authors propose that these findings may support “premedication genetic diagnosis” and individualized therapy.¹ That is an important long-term goal, but treatment selection requires evidence that genotype modifies benefit from one therapeutic strategy relative to another, ideally using a prespecified response phenotype and independent replication.^2–4

The study provides valuable hypothesis-generating data for pediatric asthma pharmacogenomics. A useful next step would be to clarify the ΔFEV1 calculation, re-confirm the responder classification, and evaluate the most promising variants against a clearly prespecified asthma-treatment endpoint in an independent cohort. Ultimately, genotype-stratified comparisons across treatment strategies would help distinguish markers of asthma prognosis from markers that truly predict differential treatment benefit. Precision asthma therapy depends not only on identifying the right genotype, but also on defining exactly what clinical response that genotype is expected to predict.

Funding Statement

No funding was received for this work.

Disclosure

The author reports no conflicts of interest in this communication.