(Sunday, 7/12/2020) [Eastern Time (GMT-4)]
Room: AAPM ePoster Library
Purpose: To study the effect of daily variation of normal tissue influencer structures on the volume and dose of a tumor target for patients undergoing adaptive radiation therapy treatments.
Methods: Five patients with cervical cancer (36 simulated sessions) underwent virtual adaptive radiotherapy using a commercial, CT-based, adaptive radiation therapy solution. The ITV structure was delineated, while the PTV was calculated by expanding the ITV 0.5 cm laterally and 1.0 cm otherwise. The influencing structures considered here were rectum and bladder. These influencer ROIs are AI-segmented and pre-introduced to the models of the planning system. To evaluate daily variations in these ROIs, geometrical and morphological properties were studied via in-house developed algorithms. Daily baseline changes for volumes, centroids, orientation, and air pocket percent volumes were calculated across all patients for relevant structures. The impact on the original PTV/ITV by daily shape variations in the influencers were studied and estimated by convolution methods.
Results: For all patients, the mean bladder volume was 94.9±33cc and experienced a change of up to 135%. The mean rectal volume was 47.9±14.8cc and experienced a change up 82%. Changes in the bladder centroids were on average 4.1±1.4mm in the lateral direction and < 2mm otherwise. For the rectum, the centroids retained the position laterally within 0.5mm and <2mm otherwise. For the ITV and PTV, the mean bladder-ITV/rectum-ITV volume impact was 17.4±8cc/12.6±15cc, respectively. Whereas for PTV, the mean bladder/rectum volume impact was 45.3±11.8cc/27±16cc, respectively. Dosimetric impact, on the clinical goals (V95%>98% and 95% for ITV and PTV, respectively), were 4.3±7.4% and 9.4±6.5% for ITV and PTV, respectively, due to the daily variations in the rectum and the bladder.
Conclusion: These preliminary results are the first step in the development of IGRT vs ART patient-gained risk/benefit comparisons. Further simulations are essential to build predictive models for efficient decision making.
Funding Support, Disclosures, and Conflict of Interest: This work is partially supported by Varian Medical Systems
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email
