Purpose: Initial planning techniques and plan quality with the novel Ethos CBCT-guided ring-gantry adaptive radiotherapy system are evaluated for head and neck(H&N) patients.
Methods: Clinical data(planning CT/structures) are used to replan previously treated H&N patients(n=9). The Ethos Intelligent Optimization Engine generates 5 plans(7/9/12 fixed-field IMRT and 2/3 full-arc VMAT) based on clinical goal prioritization (rather than cost function parameter inputs). Four priority levels exist (P4-P1) with fine-tuning provided by ordering clinical goals within each level. Planning target volume(PTV) goals and critical organ-at-risk(OAR) Dmax are assigned higher priority(P1-P2), while other OAR goals are assigned lower priority(P3-P4). Within this paradigm, different planning techniques are assessed: utilization of physician goals; PTV coverage preference ignoring overlapping OARs; clinical goals applied to OARs cropped from PTVs. Generic helper structures for isodose shaping and hot-spot control are evaluated. Dosimetric comparison of Ethos(AcurosXB) and clinical(Collapsed-cone-convolution superposition(n=8); Monte Carlo(n=1)) plans is based on NRG guidelines.
Results: The first two planning techniques result in under-covered PTVs, unacceptable PTV heterogeneity, and high dose to OARs intersecting PTVs. The third technique, with the use of helper structures, provides the most comparable results to clinical plans. These Ethos versus clinical plan PTV average D99%, D95%, and Dmax are within the NRG guidelines: 103.6% vs. 100.5%, 102.2 vs. 100.9%, and 108.8% vs. 105.5%, respectively. Average global hot-spot is 110.5% in ethos(occasionally outside of PTVs) and 105.6% in clinical plans (potentially due to dose calculation algorithm differences). Most OAR doses are within NRG guidelines for both plans(with lower doses in Ethos) except for larynx, pharynx and submandibular glands (with higher doses in Ethos) due to PTV proximity.
Conclusion: Ethos generates clinically comparable initial H&N plans with better PTV coverage and OAR sparing. However, care must be taken to precisely evaluate the magnitude and location of hot-spots in the produced plans to ensure clinical acceptability.
Funding Support, Disclosures, and Conflict of Interest: The research is supported by Varian Medical Systems, Inc.