Click here to

Session: Therapy General ePoster Viewing [Return to Session]

A Novel Two-Step Optimization Method Incorporating Knowledge-Based Planning for Head-And-Neck Radiotherapy

H Liu1*, B Sintay2, D Wiant3, (1) Cone Health Cancer Center, Kernersville, NC, (2) Cone Health, Greensboro, NC, (3) Cone Health Cancer Center, Greensboro, NC


PO-GePV-T-258 (Sunday, 7/10/2022)   [Eastern Time (GMT-4)]

ePoster Forums

Purpose: There has been much research interest in automated head-and-neck (HN) planning, with the intention of reducing planning time and variability while improving quality. However, clinical uses are still limited, and institution dependent, due to the complexity of HN treatment. The purpose of this work is to investigate whether the use of a novel semi-automated two-step optimization method (RP) can improve the quality and efficiency of HN planning.

Methods: 40 patients (2 and 3 prescription isodose levels) were retrospectively studied. Plans were generated by RP which incorporates a knowledge-based planning solution. Comparisons were performed for selected dose-volume indices between the clinical plan (CP) used previously for treatment and RP. Blind reviews were carried out by 13 clinicians to determine the preference between the CP and RP, as well as clinical suitability.

Results: No significant dosimetric differences are observed between the CP and RP for OARs away from the target. For the majority of patients studied, RP has slightly better, or similar conformity, for the high-dose PTV, and better conformity to low-dose PTV and 45Gy isodose lines compared to the CP. The only statistically significant difference observed for the serial organs is a maximum dose reduction to the spinal cord with RP. Except for left parotid gland (Dmean and V30 for 2Rx and 3Rx groups) and oral cavity (Dmean for 3Rx group), the RP has statistically significant dose reductions for all parallel organs compare to the CP. 83.4%/88.4% (2Rx/3Rx) of the RP were either preferred or marked as no preference with the CP. Looking on a per patient level, all RP received the majority vote as the preferred choice.

Conclusion: Our novel RP method allows plans to be created within a one-hour time frame while offering significant improvements in plan quality and less inter-planner variability as compared to traditional planning.

Funding Support, Disclosures, and Conflict of Interest: This research was supported by a grant from Varian Medical Systems, Palo Alto, CA.


Not Applicable / None Entered.


Not Applicable / None Entered.

Contact Email