Using chest computed tomography and unsupervised machine learning for predicting and evaluating response to lumacaftor-ivacaftor in people with cystic fibrosis.

Objectives: Lumacaftor-ivacaftor is a cystic fibrosis transmembrane conductance regulator (CFTR) modulator known to improve clinical status in people with cystic fibrosis (CF). The aim of this study was to assess lung structural changes after 1 year of lumacaftor-ivacaftor treatment and to use unsupervised machine learning to identify morphological phenotypes of lung disease that are associated with response to lumacaftor-ivacaftor.
Methods: Adolescents and adults with CF from a French multicentre real-world prospective observational study evaluating the first year of treatment with lumacaftor-ivacaftor were included if they had pre-therapeutic and follow-up chest computed tomography (CT) scans available. CT scans were visually scored using a modified Bhalla score. A k-means clustering method was performed based on 120 radiomics features extracted from unenhanced pre-therapeutic chest CT scans.
Results: In total, 283 patients were included. The Bhalla score significantly decreased after 1 year of lumacaftor-ivacaftor (-1.40±1.53 points compared with pre-therapeutic CT, p<0.001). This finding was related to a significant decrease in mucus plugging (-0.58±0.88 points, p<0.001), bronchial wall thickening (-0.35±0.62 points, p<0.001) and parenchymal consolidations (-0.24±0.52 points, p<0.001). Cluster analysis identified three morphological clusters. Patients from cluster C were more likely to experience an increase in per cent predicted forced expiratory volume in 1 s (FEV ₁ % pred) ≥5% under lumacaftor-ivacaftor than those in the other clusters (54% of responders versus 32% and 33%; p=0.02).
Conclusion: 1-year treatment with lumacaftor-ivacaftor was associated with a significant visual improvement of bronchial disease on chest CT. Radiomics features on pre-therapeutic CT scans may help to predict lung function response under lumacaftor-ivacaftor.
Competing Interests: Conflict of interest: A. Campredon has nothing to disclose. Conflict of interest: E. Battistella has nothing to disclose. Conflict of interest: C. Martin reports lecture payments or honoraria from Chiesi and Zambon, outside the submitted work. Conflict of interest: I. Durieu has nothing to disclose. Conflict of interest: L. Mely has nothing to disclose. Conflict of interest: C. Marguet reports consulting fees from Gleamer; lecture payments or honoraria from Vertex, Viatis and Zambon; support for attending meetings and/or travel from Zambon; and participation on a Data Safety Monitoring Board or Advisory Board for Zambon and Viatis; outside the submitted work. Conflict of interest: C. Belleguic has nothing to disclose. Conflict of interest: M. Murris-Espin has nothing to disclose. Conflict of interest: R. Chiron has nothing to disclose. Conflict of interest: A. Fanton has nothing to disclose. Conflict of interest: S. Bui reports payment for expert testimony for inhaled antibiotics for Zambon, outside the submitted work. Conflict of interest: M. Reynaud-Gaubert has nothing to disclose. Conflict of interest: P. Reix has nothing to disclose. Conflict of interest: T-N. Hoang-Thi has nothing to disclose. Conflict of interest: M. Vakalopoulou has nothing to disclose. Conflict of interest: M-P. Revel has nothing to disclose. Conflict of interest: J. Da Silva has nothing to disclose. Conflict of interest: P-R. Burgel reports grants or contracts from Vertex and GSK; consulting fees from AstraZeneca, Chiesi, GSK, Insmed, Vertex and Zambon; and lecture payments or honoraria from Pfizer and Novartis; outside the submitted work. Conflict of interest: G. Chassagnon reports lecture payments or honoraria from Chiesi, outside the submitted work.
(Copyright ©The authors 2022. For reproduction rights and permissions contact permissions@ersnet.org.)