Purpose: The purpose of this study is to assess the efficacy of Varian Ethos IOE for SBRT lung metastases treatment plans.
Methods: Six retrospective upper lung cases, planned for SBRT on Varian Eclipse v.16.1 by an experienced dosimetrist and treated with Varian Halcyon v.3.0, were chosen for the present study. In accordance with NRG-BR002 protocol, dose regimen was 50Gy/5fxs. Eclipse plans were exported to Ethos v.2.0, where 3 IMRT (9-field, 12-field, and 7-field with laterality) and 2 VMAT (2-full arc and 2-partial arc with laterality) treatment plans were autogenerated, using priority goals set in Ethos RT intent. All plans were re-normalized, where 95% of the planning target volume (PTV) received 100% of the prescription dose.
Results: On average, it took under 14 minutes to generate 5 Ethos plans for each case. PTV global maximum dose, conformity index, and dose at 2.0cm (D2cm) from PTV ranged between 110.9%-114.6%, 1.03-1.06, and 52.3%-64.7%, for the Ethos plans, respectively; Eclipse reference plan values were 112.6%, 1.07, and 64.2%, respectively. Lungs V1250cGy and V1350cGy ranged between 163.3cc-194.0cc and 146.3cc-175.9cc, for the Ethos plans, respectively; Eclipse reference plan values were 187.2cc and 171.3cc, respectively. Overall, Ethos VMAT plans were dosimetrically hotter and more conformal, while IMRT plans rendered lower D2cm values. Between Ethos VMAT plans, the 2-full arc plans resulted in higher conformity and lower D2cm values, but at the expense of slightly higher lung dose. Eclipse reference plans were least conformal and rendered relatively higher D2cm values. All Ethos plans were deemed clinically acceptable by a radiation oncologist. On average, all plans met dose metrics using the NRG-BR002 protocol.
Conclusion: Varian Ethos IOE efficiently creates SBRT plans for lung metastases that are superior to human-generated plans. Ethos generates more conformal VMAT plans with higher PTV global maximum doses, which is desirable for SBRT.
Inverse Planning, Optimization
TH- External Beam- Photons: Treatment planning using machine learning/Knowledge Based Planning/automation