Author: Ryan Andosca, Peter Boyle, Grace Hyun Kim, Minji Victoria Kim, Michael Vincent Lauria, Michael F. McNitt-Gray, Gabriel Melendez-Corres, Jack Neylon, Brad Stiehl, Pang Yu Teng 👨🔬
Affiliation: David Geffen School of Medicine at UCLA, University of California, Los Angeles, UCLA Department of Radiology, Department of Radiation Oncology, University of California, Los Angeles, UCLA, Cedars-Sinai Medical Center 🌍
Purpose: Endobronchial Valves (EBV) are one of the few treatment options for patients with moderate to severe emphysema. Eligibility is typically assessed from CT image data analysis including Emphysema Score (Relative Area, RA950) and Fissure Integrity Score (FIS). However not all EBV candidates are responders. This study aims to investigate the potential role of estimating lung elasticity based on CT image data in identifying candidates who will benefit from EBV treatment.
Methods: We performed a retrospective study using data from 10 patients with moderate to severe emphysema and intact fissures (high FIS) who underwent EBV treatment in left lung (5 responders, 5 non-responders). Each patient underwent baseline (pre-treatment) CT scanning at both Total Lung Capacity (TLC) and Residual Volume (RV) breath-holds. Lung lobe segmentations were performed on CT image data to automatically identify lobar regions. RA910 and RA950 were extracted for each lobe. Using a well-validated method for physics-based biomechanical modelling applied to TLC and RV CT images, elastography estimations for the lungs were performed on a voxel-by-voxel basis. Summary statistics of elasticity were calculated to obtain elastography values at each lobe. To distinguish between responders and non-responders, we used Spearman correlation between mean elasticity(kPa) and RA950(%), and investigated the differences in mean elasticity.
Results: We found a positive, significant correlation with large effect size between mean elasticity and RA950 for responders (ρ=0.673, p=0.033) and negative correlation with small effect size for non-responders (ρ=-0.134, p=0.713). Responders had higher mean elasticity than non-responders (2.81±0.72 kPa and 2.01±0.46 kPa, respectively), and a higher range of mean elasticity ([2.23,4.09] and [1.57,2.67], respectively). These results indicate that higher kPa may be associated with response to EBV treatment.
Conclusion: This study demonstrated the feasibility of biomechanically-guided lung tissue elasticity estimations as a robust imaging biomarker for the EBV prediction model compared to other standard metrics.