Click here to

Session: Biologically- and Functionally-Guided Radiation Therapy [Return to Session]

Predicting Lung Ventilation Loss Caused by Local and Distant Radiation-Induced Damage

E Vicente1*, A Modiri11, J Kipritidis2, K Yu3, Y Yan4, D R Owen5, M M Matuszak5, P Mohindra1, R Timmerman4, A Sawant1, (1) University of Maryland School of Medicine, Baltimore, MD, (2) Northern Sydney Local Health District, Sydney, Australia, (3) Broncus Medical, Inc., San Jose, CA, USA, (4) UT Southwestern Medical Center, Dallas, TX, USA, (5) University of Michigan, Ann Arbor, MI, USA


TU-C-TRACK 4-4 (Tuesday, 7/27/2021) 1:00 PM - 2:00 PM [Eastern Time (GMT-4)]

Purpose: This work challenges the conventional assumption of “locoregional radiation damage” to lung parenchyma, implicit in current functional lung avoidance (FLA) radiotherapy (RT). Here, we investigate the impact of including “distant damage” by accounting for the radiation-induced injury to airways, branching serial structures distant from the point where the damage is observed. Consequently, we propose a method to predict post-RT ventilation loss that considers not only the locoregional parenchymal damage but also the airflow disruption caused by injury to airways.

Methods: Breath-hold CTs (BHCTs) and simulation 4DCTs were acquired from seven lung RT patients. Airways were auto-segmented (12 generations, ~235 airways/patient) on the BHCTs using a virtual bronchoscopy software. 4DCT-based ventilation maps were used to estimate pre-RT ventilation. Conventional IMRT plans were recreated to obtain dose distribution. Post-RT ventilation loss was estimated studying three contributions: (i) Using the pre-RT ventilation and a normal tissue complication probability model we estimated loss due to parenchymal damage (ΔVˡᵒˢˢ⁽ᴾ⁾). (ii) Connecting each terminal airway with a lung sub-lobar volume (SLV), we estimated loss due to airway damage (ΔVˡᵒˢˢ⁽ᴬ⁾) by calculating the contribution of pre-RT ventilation in not-ventilated SLVs. A SLV was considered not ventilated if only one of the upstream airways was collapsed post-RT, using our probability of airway collapse model. (iii) Finally, the loss due to both effects (ΔVˡᵒˢˢ⁽ᴾ⁺ᴬ⁾) was calculated by subtracting from the total pre-RT ventilation the contribution of the post-RT ventilation (due to parenchyma damage) in the post-RT ventilated SLVs.

Results: Preliminary results showed that average ΔVˡᵒˢˢ⁽ᴾ⁾ was 13%, ΔVˡᵒˢˢ⁽ᴬ⁾ was 22%, and the combined effect, ΔVˡᵒˢˢ⁽ᴾ⁺ᴬ⁾, was 25%.

Conclusion: These results reveal that airway damage is not only important but may be a more critical factor than locoregional parenchymal damage. Ignoring this distant effect and focusing primarily on parenchymal damage can lead to spurious choices in current FLA strategies.

Funding Support, Disclosures, and Conflict of Interest: This work is supported by the National Institutes of Health (R01 CA202761)



    Not Applicable / None Entered.


    Not Applicable / None Entered.

    Contact Email