Several mechanisms underlying this variability have been proposed, related both to individual biology and to environment or behaviour (
CFTR modulators have transformed cystic fibrosis (CF) treatment, but individual responses vary even among patients with identical CFTR genotypes. This underscores the need for predictive biomarkers to optimise therapeutic selection.
We evaluated 24 paediatric patients homozygous for F508del-CFTR, assessing lung function (FEV1pp) and sweat chloride (SC) before and after CFTR modulator therapy. Whole-gene sequencing was utilised to identify CFTR and pharmacogene variants. Patient-derived human nasal epithelial cells (HNECs) were expanded and differentiated at the air-liquid interface to assess CFTR function via ion transport (Isc).
Clinical responses varied widely. Twelve participants changed modulators during the study. Sequencing identified 231 additional CFTR variants and pharmacogene polymorphisms, but none correlated with response variability. However, a significant linear relationship emerged between Isc and FEV1pp improvement in patients with baseline FEV1pp<90 (R² = 0.652, p =0.001) and SC reduction (R² = 0.535, p=0.004). Receiver operating characteristic analysis demonstrated high predictive accuracy for SC reduction (area under the curve (AUC)=0.88) and combined FEV1pp/SC response in patients with baseline FEV1pp<90 (AUC=1.00). Exploratory analysis confirmed that Isc predicts FEV1pp changes, modulated by baseline lung function and CFTR modulator type.
Patient-derived differentiated HNEC cultures serve as a robust predictive tool for CFTR modulator response in paediatric CF patients. Their integration into clinical practice can enhance personalised treatment strategies, minimising ineffective therapy use and improving CF patient outcomes with precision medicine.
Personalised management of cystic fibrosis (CF) lung disease in the era of CF transmembrane conductance regulator (CFTR) modulator therapy will require novel approaches to assess treatment response and monitor longitudinal progression.
We evaluated the heterogeneity of short-term response to elexacaftor-tezacaftor-ivacaftor (ETI) using 129Xe MRI, multiple breath washout (MBW) and spirometry. For CFTR modulator-naïve people ≥6 years old starting commercial ETI, we measured same-day 129XeMRI ventilation defect per cent (VDP), lung clearance index (LCI), forced expiratory volume in 1 s (FEV1) and CF Questionnaire-Revised respiratory-domain (CFQ-R(R)) at pre-ETI baseline and up to 4 months post-ETI. Baseline versus follow-up measures were compared using Wilcoxon-signed-rank tests with r effect size coefficients. Abnormal follow-up measures determined using z-scores and upper/lower limits of normal; correlations evaluated using Spearman coefficients.
VDP, FEV1, LCI and CFQ-R(R) significantly improved in the total group (n=40; all p<0.001, r=0.69, r=0.79, r=0.70, r=0.77) and both ≥12-year-old subgroups (n=15 FEV1<80%, p=0.005/r=0.69, p<0.001/r=0.87, p=0.004/r=0.70, p<0.001/r=0.88; n=17 FEV1≥80%, p=0.003/r=0.54, p=0.003/r=0.74, p=0.01/r=0.56, p=0.006/r=0.73). VDP (p=0.008/r=0.89), LCI (p=0.03/r=0.83) and percent-predicted FEV1 (p=0.03/r=0.83) significantly improved in the 6–11-year-old subgroup, with VDP having the greatest effect size in children. VDP remained abnormal at follow-up in 21 participants with normal FEV1, whereas LCI was abnormal in 12. Baseline VDP (=0.39/p=0.01) and LCI (=0.43/p=0.008) were significantly correlated with the change in CFQ-R(R).
Though all outcome measures showed a positive response, response to ETI was heterogeneous between individuals and between 129XeMRI, MBW and spirometry. Leveraging the rich information provided by multimodal tools will be critical to understanding the nature of CF lung disease and measuring response to future therapies in the era of CFTR modulators.