Impact of Dose Calculation Algorithm On Airway Dosimetry: Comparison of AAA and Acuros
P Kinkopf1*, A Modiri1, K Yu2, Y Yan3, P Mohindra1, R Timmerman3, A Sawant1, E Vicente1, (1) University of Maryland School of Medicine, Baltimore, MD, (2) Broncus Medical, Inc., San Jose, CA, (3) UT Southwestern Medical Center, Dallas, TX.
Presentations
PO-GePV-T-403 (Sunday, 7/25/2021) [Eastern Time (GMT-4)]
Purpose: Minimizing post-radiotherapy (RT) respiratory function loss entails better understanding of RT-induced airway injury. Previously, we showed the importance of including individual airways as organs at risk (OARs) to reduce lung stereotactic ablative radiotherapy (SAbR) toxicity after treatment. Estimating and fulfilling airway dose constraints during RT plans require accurate dose calculations to these structures. Therefore, here we extend the investigations on common dose calculation algorithms to include the study of airway dosimetric accuracy.
Methods: We retrospectively evaluated the anisotropic analytical algorithm (AAA) and Acuros algorithm by using a high-resolution breath-hold CT (BHCT), simulation 4DCT, and prescribed RT plan from four lung SAbR patients. A virtual bronchoscopy software was used to autosegment 856 airways on the BHCT images. Planning target volumes (PTVs) and standard OARs were contoured on the average image of the 4DCT phases (AVG). Conformal RT (CRT) and intensity-modulated RT (IMRT) plans were recreated on the BHCT (with OARs warped to this image and breath-hold PTV contoured on this image) and on the AVG (with airways warped to this image), for a total of four plan types/patient. Dose calculations were performed using AAA and Acuros, and the differences in mean and maximum dose (Dmax) were calculated.
Results: Dose differences in PTVs/OARs were comparable to those previously reported. Median differences in airway Dmax ranged from -0.03 to +0.22Gy with respect to airway diameter without clear dependence, and from -0.52 to +0.21Gy with respect to Dmax for all plan types. AAA underestimations (negative values) were typically observed in BHCT plans, overestimations in AVG plans. Translating these results to airway injury risk, these dose differences corresponded to variations in airway collapse ≤±0.7%.
Conclusion: These results suggest that airway dosimetric differences between AAA and Acuros are not clinically relevant and that using either algorithm is appropriate when including patient airways as individual OARs.
Funding Support, Disclosures, and Conflict of Interest: This work is supported by the National Institutes of Health (R01 CA202761)
ePosters
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email