Late Breaking Abstract - Responder phenotyping using functional respiratory imaging (FRI) in IPF patients treated with anti-CGTG monoclonal antibody FG3019
2017
Aims and objectives To identify baseline FRI parameters that predict changes in FVC and fibrosis after 48w of treatment with FG3019 Methods: In this open label study, 66 IPF patients (67.9±7.0 years) completed 48 weeks of treatment with an anti-CTGF monoclonal antibody. Pulmonary function was assessed every 12 weeks and HRCT scans (TLC and RV) were taken at baseline and after 24 and 48w. Machine learning algorithms were used on the baseline FRI parameters to obtain accurate predictions of changes in FVC and fibrosis after treatment (responder phenotyping). Results: FVC responders, as defined by stable or improving FVC, could be predicted with an accuracy of 87% by assessing lobe volumes and specific airway radii (siraw) of all zones measured at TLC together with siraw in the lower lobes measured at RV (Figure 1). Fibrosis responder, as defined by stable or improving fibrosis, could be predicted with an accuracy of 76,2% by combining the specific image based airway volumes (siVaw) and the specific image based surface area (siSaw) of the lower lobes measured at TLC. Figure 1: Change in FVC predicting features Conclusion: IPF is a highly heterogeneous disease with a heterogeneous response to treatment. Quantitative imaging can adequately describe the heterogeneity and has the potential to provide accurate responder phenotyping.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
0
References
0
Citations
NaN
KQI