Click here to

Session: Adaptive and Biology-guided Radiation Therapy [Return to Session]

Implementation of a Knowledge-Based Planning Model for Gastrointestinal (GI) Site Specific Tumors in Photon Radiotherapy

A Kassaee1*, A Dimofte2, P Irmen3, J Marcel4, T Zhu5, (1) University of Pennsylvania, Philadelphia, PA, (2) University of Pennsylvania, Philadelphia, PA, (3) Inova Health System, Fairfax, VA, (4) University of Pennsylvania, Philadelphia, PA, (5) University of Pennsylvania, Philadelphia, PA


TU-D1000-IePD-F7-3 (Tuesday, 7/12/2022) 10:00 AM - 10:30 AM [Eastern Time (GMT-4)]

Exhibit Hall | Forum 7

Purpose: To evaluate the dosimetric plan quality improvement of knowledge based planning models for GI site specific tumors. To improve dosimetric workflow by expanding our clinically available GI knowledge based planning models.

Methods: Multiple models were created using Varian’s Rapidplan™ for lymph abdomen and lymph pelvis. 65 and 50 patients respectively were selected for model training. Both models include patients using IMRT and VMAT treatment delivery techniques. Training plans were those that focused on institutional constraints for the abdominal and pelvic organs at risk (OARs). 5 patients were selected outside each model used for validation. Each patient was validated for PTV and CTV coverage (D95% and D2%), as well as dose-volume constraints for the OARs that are relevant to each site. Dose distributions were looked at to verify adequate fall off and conformity.

Results: For all patient validated plans, the model generated clinical acceptable plans meeting all dosimetric constraints set by our institution. Target coverage remained similar across all patient plans compared to that of the clinically delivered plans. Dose to most critical structures either saw improvement or remained consistent with manually optimized plans. Lymph abdomen validated patients had notable differences in the spinal cord D0.03 and right kidney D50% on average showing a reduction of approximately 2.0Gy and 1.5Gy respectively. Lymph pelvis validated patients had a reduction in bladder D50% and rectum D50% sparing on average showing a reduction of approximately 3.0Gy and 1.5Gy respectively.

Conclusion: Knowledge-based planning implementation through RapidPlan has been shown to create clinically acceptable plans and shows improvement in sparing of some clinical relevant OARs.


Not Applicable / None Entered.


Not Applicable / None Entered.

Contact Email